The Linguistic Analysis of Word and Sentence Structures

pexels-allie-reefer-9969267

juliedoner — Tue, 14 Nov 2023 18:01:59 +0000

pexels-snche-7507105

juliedoner — Tue, 14 Nov 2023 18:05:49 +0000

pexels-thirdman-5649892

juliedoner — Tue, 14 Nov 2023 18:09:17 +0000

pexels-kastograph-)-17199829

juliedoner — Tue, 14 Nov 2023 18:16:22 +0000

Swahili-Region

juliedoner — Wed, 15 Nov 2023 05:29:35 +0000

A map of part of Africa. Parts of Kenya, Tanzania, Uganda and surrounding areas are in purple, indicating where Swahili is spoken in East African

Swahili-Region

juliedoner — Wed, 15 Nov 2023 05:32:37 +0000

A map of part of Africa. Parts of Kenya, Tanzania, Uganda and surrounding areas are in purple, indicating where Swahili is spoken in East African

Screen Shot 2023-11-07 at 9.58.38 AM

juliedoner — Wed, 15 Nov 2023 19:17:23 +0000

pexels-arminas-raudys-808941

juliedoner — Wed, 22 Nov 2023 19:48:08 +0000

pexels-alina-matveycheva-15485496

juliedoner — Wed, 22 Nov 2023 20:10:19 +0000

pexels-charles-parker-5859246

juliedoner — Wed, 22 Nov 2023 20:19:10 +0000

pexels-juan-vargas-5432188

juliedoner — Wed, 22 Nov 2023 20:27:30 +0000

IMG_1606

juliedoner — Tue, 28 Nov 2023 23:37:03 +0000

640px-Grand_Beach_and_Provincial_Park_in_Lake_Winnipeg_in_Fall_2008_Manitoba_Canada_(3)

juliedoner — Wed, 29 Nov 2023 01:41:15 +0000

Noam_Chomsky,_2004

juliedoner — Wed, 29 Nov 2023 04:48:45 +0000

White and Green Bike Leaning on Wall

juliedoner — Thu, 30 Nov 2023 05:54:50 +0000

4167105376_0fd30f161e_c

juliedoner — Tue, 05 Dec 2023 10:47:28 +0000

IMG_1619

juliedoner — Tue, 05 Dec 2023 18:23:43 +0000

IMG_1612

juliedoner — Tue, 05 Dec 2023 18:33:29 +0000

Carbon_atom_(Bohr_model)

juliedoner — Wed, 20 Dec 2023 00:21:13 +0000

put [the dog food in the bowl] on the floor

juliedoner — Fri, 19 Apr 2024 15:36:53 +0000

Check Yourself Questions

markc — Wed, 29 Nov 2023 18:24:51 +0000

This is a printable version of the interactive Check Yourself questions and answers from the end of each subsection.

Part I. Foundations

1.1 What is morphology?

Questions

Question 1.1-1 The following passage is taken from The Tale of Peter Rabbit by Beatrix Potter, which is in the public domain. Click on all of the polymorphemic words in the passage. There are some tricky words here, so don't be worried if you get stumped! After you have attempted the question, click on the blue arrow to read an explanation of the answer. We will spend some more time discussing what to do with tricky words in Chapter 8. Then old Mrs. Rabbit took a basket and her umbrella, and went through the wood to the baker's. She bought a loaf of brown bread and five currant buns. Flopsy, Mopsy, and Cotton-tail, who were good little bunnies, went down the lane to gather blackberries: But Peter, who was very naughty, ran straight away to Mr. McGregor's garden, and squeezed under the gate!

Answers

Question 1-1.1 Sometimes it is very tricky to segment a word into morphemes, and this passage illustrates some of the difficulties. Irregular past tense verbs like took, went, bought, were, and ran are arguably composed of a base word and an abstract past tense marker. If you have studied a Romance language, you might have picked umbrella as a polymorphemic word, because it comes from the Latin morphemes umbra meaning 'shade' and -ella meaning 'small thing.' However, this word was borrowed as a whole into English, and most English speakers probably have this word stored in their mental dictionary as one piece. Likewise, garden was also polymorphemic in Latin but was borrowed as a whole into English. For proper names, like Flopsy, Mopsy, Cotton-tail, and McGregor, you could argue either way. Historically, these names come from words with more than one meaning component, but it is hard to say whether people still treat them that way. However, once you add the possessive marker, McGregor's is definitely polymorphemic. baker's has three morphemes: bake + -er + -'s buns has two morphemes: bun + -s Technically, currant buns is one compound word with three morphemes, currant, bun, and -s. Some compound words are written with a space! We will learn to identify compound words in Chapter 10. bunnies has two morphemes: bunny + -s Note that a morpheme might change its spelling depending on context but it is still the same morpheme! blackberries has three morphemes: black + berry + -s naughty has two morphemes: naught + -y Since the polymorphemic word naughty is more common than naught, this one is very easy to miss. under only has a single morpheme! Even though under looks like it could be composed of un- + der, it doesn't share the meaning component of un-, meaning this similarity is a coincidence, and der does not exist as a morpheme at all. squeezed has two morphemes: squeeze + -ed

Proto-Algonquian

juliedoner — Fri, 19 Apr 2024 21:13:36 +0000

Cover Image

juliedoner — Mon, 06 May 2024 18:57:06 +0000

UML_OER_7_Doner4s

juliedoner — Mon, 06 May 2024 20:53:55 +0000

IMG_3097

juliedoner — Sat, 11 May 2024 14:01:55 +0000

IMG_3099

juliedoner — Sat, 11 May 2024 14:02:49 +0000

Acknowledgements

juliedoner — Sat, 06 Jul 2024 20:21:12 +0000

OER working group including Glenn and Janice audiences of OE Global and OE Conference Nicole Connor Heather Twitterlings, including Jena Barchas-Lichtenstein Rebus Iwona Kristy from Campus Manitoba for glosses

Khashayar Hamidzadeh for feedback on accessibility for blind users

Clarissa Forbes

people from the centre for teaching and learning

Language consultants

Language	Consultant(s)
Hindi
Punjabi	one anonymous consultant
Russian
Tagalog
Urdu
Yoruba	Tolani Akinlade

Anatomy of a journal article

juliedoner — Thu, 01 Aug 2024 03:28:27 +0000

LSA_logo_white

juliedoner — Mon, 09 Sep 2024 23:39:09 +0000

Screen Shot 2024-09-16 at 4.08.57 PM

juliedoner — Mon, 16 Sep 2024 21:09:37 +0000

Screen Shot 2024-09-16 at 4.09.05 PM

juliedoner — Mon, 16 Sep 2024 21:09:39 +0000

YorubaPedagogicalGrammarExample

markc — Wed, 09 Oct 2024 20:56:34 +0000

Picture1

juliedoner — Thu, 14 Nov 2024 23:59:47 +0000

IMG_5081

juliedoner — Fri, 15 Nov 2024 00:09:52 +0000

IMG_5082

juliedoner — Fri, 15 Nov 2024 00:09:57 +0000

Screen Shot 2023-02-02 at 5.24.40 PM

juliedoner — Wed, 26 Feb 2025 19:26:31 +0000

aps_mss_b_b61_u5_1_29

curatolm — Wed, 26 Feb 2025 19:43:21 +0000

George Hunt

curatolm — Wed, 05 Mar 2025 18:17:42 +0000

Franz Boas

curatolm — Wed, 05 Mar 2025 18:56:21 +0000

A Chief's daughter - Nakoaktok

curatolm — Wed, 05 Mar 2025 19:11:41 +0000

Appendix

juliedoner — Wed, 11 Oct 2023 18:02:56 +0000

This is where you can add appendices or other back matter.

(TEST) Colours + Glosses/Examples

markc — Wed, 25 Oct 2023 02:23:49 +0000

Caption - linguistic gloss sample 1.

(1)	a.	John	goes	to	the	store₁
		subject₂	verb	preposition Small caps	article	noun
		"John goes to the store."

Caption - linguistic gloss sample 2.

(2)	a.	Jean	parle	français
		Jean	speaks	French
		"John speaks French."

Caption - linguistic gloss sample 3.

(3)	a.	Jean	aime	la	linguistique
		Jean	likes	the	linguistics
		"John likes linguistics."
	b.	Jean	l'	aime
		Jean	it	likes
		"John likes it."

(1)	a.	okunrin	ti	jo.
		boy	PAST	dance
		"The boy danced."
	b.	okunrin	O	jo.
		boy	PRES	dance
		"The boy is dancing."
	c.	okunrin	ma	jo.
		boy	FUTURE	dance
		"The boy will dance."

(2)	a.	okunrin	ti	jo.
		boy	PAST	dance
		"The boy danced."
	b.	okunrin	O	jo.
		boy	PRES	dance
		"The boy is dancing."
	c.	okunrin	ma	jo.
		boy	FUTURE	dance
		"The boy will dance."

Gila: gila; now

aburun: abur-u-n; they-OBL-GEN₁

ferma: ferma; farm

hamišaluǧ: hamišaluǧ; forever

güǧüna: güǧüna; behind

amuq’dač.₂: amuq’-da-č; stay-FUT-NEG

Now their farm will not stay behind forever.

(1)	a.	Il		voit	le	gâteau.
		he		sees	the	cake
		'He sees the cake.'
	b.	Il	le	voit.
		he	it	sees
		'He sees it.'

(2)	a.	J'	aime	le	pizza.
		I	like	the	pizza
		'I like pizza.'

(1)	a.	l'	homme	que	j'	ai	vu
		the	man	that	I	have	seen
		The man that I saw'

(1)	a.	l'	homme	que	j'	ai	vu
		the	man	that	I	have	seen
		The man that I saw'

		y/n
(1)	a.	iudasfghaiu	isuadfhp	puisdafgh	izudhg	iaudgbf;
		'This is an example sign language gloss.'

			y/n
				neg
(2)	a.	iudasfghaiu	isuadfhp	puisdafgh	izudhg	iaudgbf;
		'This is another example."

[latex] \setcounter{ExNo}{1} \ex. \a. Greta, I like, but not Talia. \b. Olives with blue cheese, I hate. \c. \textit{Anna Karenina} was okay, but \textit{War and Peace}, I loved. [/latex]

Learning Objectives [4.87:1]

In this section, you will learn... [>9.78:1]

About different types of affixes.
How to distinguish between different types of them.

Examples of Affixes [13.72:1]

Examples of affixes include: [>4.59:1]

Inflectional suffixes (ex. past tense -ed).
Derivational suffixes (ex. agentive -er).
etc.

Exercise on Affixes [4.77:1]

Are the following undlined affixes inflectional or derivational? Make sure to look at their use in the sentences! [>9.7:1]

Some gardeners worked outside.
I was looking for some ways to clarify concepts with my instructor.

Key Takeaways [10.54:1]

In this section, you have learned... [>5.28:1]

The differences between inflectional and derivation morphology.
How the two types of morphology work together structurally to build words.
How ambiguity in meaning can occur via different orders of affixation.

As can be seen in (1), objects in English do not undergo leftwards movement when pronominalized in English but they do in French.

Glossary

juliedoner — Fri, 03 Nov 2023 07:47:44 +0000

Making a PSR grammar for a language you don't speak

juliedoner — Wed, 08 Nov 2023 17:33:25 +0000

One thing that linguists often do is look at languages that they don't know much about and look for the word order patterns in that language. They can then describe the patterns with phrase structure rules. The first step in figuring out the phrase structure rules for another language is to determine the constituency in your data. You can use the meaning in the English translation to determine what modifies what in the other language. Let's look at some examples together. The following examples are from Turkish. Remember that in examples from other languages, the first line is written naturally in the other language. The second line tells you what each word in the other language means. The third line tells you what the sentence means in English.

The abbreviations used in these examples are listed in the table below.

Abbreviation	Meaning
1	First person (I/me)
3	Third person (it/she/he)
OBJ	Object case marker
PASS	Passive voice marker
POSS	Possessive marker
PST	Past tense
SG	Singular number

Table 17.x: Turkish glossing abbreviations

Step 1: Identify the order of subject, object, and verb

The first constituents you should look for are the subject and the verb phrase. The subject will typically be the NP that performs the action, while the VP will include the verb and the object NP. The object NP is the NP at which the action is directed. The VP always includes the verb and the object (if there is an object). It never includes the subject. In sentence (1), the word order is roughly Toprak Deniz see, as we can see from the second line. But how do we know whether Toprak saw Deniz or Deniz saw Toprak? Well, we look at the English translation in the third line. The sentence means that Toprak is the one doing the seeing, not Deniz, so Toprak is the subject. Likewise, Deniz is the one being seen, so Deniz is the object.

(1)	Toprak	Deniz’-i	gör-dü.
	Toprak	Deniz-OBJ	see–PST
	'Toprak saw Deniz.'

From this, we can hypothesize that Turkish is an SOV language, which means that sentences in Turkish follow a subject - object - verb word order by default. Turkish also has object case markers which can help us find the object (see Section 5.7 for a refresher on what case is). Not all languages use case markers, and sometimes case markers are used in unexpected ways, so we cannot always depend on case markers to help us determine which NP is the subject or the object. However, it can be a useful piece of secondary evidence! You can draw boxes around the constituents you identify so you don't get confused! In example (1) we might draw a box around Deniz’i gördü 'saw Deniz' to remind ourselves that this is a constituent, the VP.

Working Hypotheses

This approach works on the assumption that the English translation in the third line has an equivalent structure as the original language, which is not always the case. Linguists will try to translate the sentence as close as possible, but sometimes the language will have a structure that does not have an equivalent in English. For example, here is what a passive sentence looks like in Turkish, although, unlike in English, it is unusual (but still possible!) to include the subject in a passive like we did here.

(2)	Deniz	Toprak	taraf-ın-dan	gör-ül-dü.
	Deniz	Toprak	side-3SG-POSS	see-PASS-PST
	'Deniz was seen by Toprak.' Or literally, 'Deniz was seen from Toprak's side.'

Because it’s passive, the one doing the seeing is not in subject position, and the one being seen is not in object position. It is important to keep in mind the difference between the structural subject and the thematic subject. The structural subject is the NP that appears in the subject position of the clause (the daughter of S). The thematic subject is the NP with the agent theta role. In most sentences, both of these pick out the same NP, the subject. However, sometimes they aren’t the same, such as in passive sentences. When you’re working on an unfamiliar language, you won’t always know if there’s something happening that is adding extra complication to your data, like if it’s a passive sentence. But we have to start somewhere! So what we do is we make a hypothesis based on the data we have. We try to make it the simplest hypothesis possible. After that, we collect more data to check our first hypothesis. If the passive sentence was the only sentence of Turkish we had, we might conclude that Turkish has OSV word order instead of SOV (depending on how seriously we took the structure of the English translation). However, once we collected more data, we’d probably notice that the passive sentence has a different word order than the others. We might also notice that the passive sentence has morphological differences. For example, it has the passive marker on the verb but it doesn’t have the object case marker on the object. We would use these clues to help us revise our initial hypothesis and conclude that sentence (2) has a change in argument structure and that Turkish really is SOV.

Step 2: Adding the modifiers to the constituents

After we identify the order between subject, object, and verb, we can start to fill out some of the other constituents, by identifying what the modifiers are modifying. For example, in sentence (3), we have the adjective siyah ‘black’ occurring between two nouns. Some languages (like English) put adjectives on the left of the noun it modifies, while others (such as French) put adjectives on the right. How do we know, then, whether Turkish is like English, and this sentence is about a black table, or whether Turkish is like French, and this sentence is about a black book?

(3)	Kitap	siyah	masa-ya	dokun-du.
	book	black	table-OBJ	touch-PST
	'The book touched the black table.'

Again, we can tell by looking at the English translation in the third line. In the translation, black modifies table, so we can know that siyah modifies masaya, and therefore that siyah masaya forms a constituent. You may want to put a box around this constituent too, so we remember what we figured out.

Turkish has a phonological assimilation rule called vowel harmony which can change the pronunciation of vowels to match the backness feature of other vowels in the word. This is why the past tense marker is sometimes -dü and sometimes -du. This difference doesn’t affect the syntax, so you don’t need to worry about that difference here.

Step 3: Looking for patterns

After you've identified the constituents, you need to go through each constituent one by one and identify what goes into that constituent. At this step, the sort of questions you should ask are:

What is optional in this kind of constituent?
What is obligatory in this kind of constituent?
Which elements can be repeated, so that they need a plus sign?
What is the relative order of the elements inside a constituent?

You can identify that something is optional in a given constituent by noticing that it isn’t there each time that kind of constituent appears in your data. For example, the S rule will show the order between the subject NP and the VP. However, in some languages, including Turkish, the subject is optional in some contexts. This is shown in sentences (4) and (5). In sentence (4), there is a subject pronoun ben ‘I’. But in sentence (5), which has the exact same meaning, there is no subject pronoun! You can still tell who the subject is, but only because of the agreement marker on the verb.

(4)	Ben	Deniz’i	gör-dü-m
	I	Deniz-OBJ	see-PST-1SG
	'I saw Deniz.'

(5)	Deniz’i	gör-dü-m
	Deniz-OBJ	see-PST-1SG
	'I saw Deniz.'

Because of this, when we write the S rule for Turkish, we will put the subject NP in brackets to show that it is optional.

(6)	S → (NP) VP

When we’re working with a small data set, like you will be doing on most of your homework in this class, you won’t usually be able to tell for sure whether something is obligatory. However, if it appears in all of the relevant places in your data set, you should assume it is obligatory until you find evidence to the contrary. So, for example, all of the sentences we’ve looked at in this section are transitive, and have an object. So we would not put the object NP in parentheses. (Turkish does have intransitive verbs, though, so if our data set was a bit bigger, this would be different.) Remember, we are writing our PSRs as a model that describes the data we have, not the data we expect to exist. Once we gather that further data, we can revise our hypotheses. When you are listing the members of a constituent in your PSRs, be careful that you list the constituents of just the next layer of structure. For example, let’s look at sentence (3) again.

(7)	Kitap	siyah	masa-ya	dokun-du.
	book	black	table-OBJ	touch-PST
	'The book touched the black table.'

In this sentence, the VP is siyah masaya dokundu ‘touched the black table.’ Many students will look at this sentence and conclude that the VP rule for Turkish is an AdjP (for siyah), followed by an NP (for masaya), followed by a V (for dokundu), as shown in (8). But this isn’t quite right!

(8)	Incorrect rules
	VP → (AdjP) NP V
	NP → (AdjP) N

Earlier we decided that siyah masaya ‘black table’ was a constituent. The adjective doesn’t belong directly in the VP; instead it belongs inside of the NP meaning black table, and the whole NP is inside the VP, as shown in the tree in Figure A1.4. [caption id="attachment_3745" align="alignnone" width="298"] Figure A1.4: Tree of example (7)[/caption]The incorrect rules in (8) put the same AdjP in two different places—both inside the VP and inside the NP. Instead, it should only belong to the NP, as in (9). Siyah does not modify the verb. It only modifies the noun, so it belongs to the NP.

(9)	Correct rules
	VP → NP V
	NP → (AdjP) N

Step 4: Putting it all together

In your last step, you should collect all of your PSRs in one list and then double check them. Here is the list of PSRs for the Turkish data from this section.

(10)	Turkish PSRs
	S → (NP) VP
	VP → NP V
	NP → (AdjP) N
	AdjP → Adj

The first thing you should do to double check your answers is to compare all of the data in your data set to the final version of your rules. Sometimes when you revise a rule, you accidentally make it inconsistent with some data you looked at earlier in your process. I also recommend you draw a tree of one or more sentences in your data using your PSRs, to check to see if following your PSRs strictly makes the word order come out right with no pieces missing. If you only draw one sentence, choose the most complicated one. That is the one that is most likely to have a mistake! If you have to draw something that isn’t listed in your rules, then either you’ve drawn it incorrectly or there’s a mistake in your rules. Special thanks to Çağrı Bilgin for providing the Turkish data in this section.

Draw a syntax tree step-by-step

juliedoner — Wed, 08 Nov 2023 17:41:36 +0000

Here are some step-by-step instructions on how to analyze sentences and then draw a tree of them. Notice that before you start drawing, you should analyze the structure of the sentence! You are much more likely to draw it correctly that way.

Step One: Identify the main verb.

Don’t get distracted by auxiliary verbs! In the sentences in (1), the main verbs are bolded. Notice that in (1a), have is a main verb, while in (1d), (1e), and (1f), it is an auxiliary verb.

(1)	a.	I have a green spaceship.
	b.	I think that aliens exist.
	c.	That aliens exist surprised my sister.
	d.	You have seen an alien with a telescope.
	e.	Have you seen an alien?
	f.	What have you seen?

Step two. Find the subject of each verb.

The subject will usually be right before the verb in English (or the auxiliary verb), it will trigger subject-verb agreement, and it is often the one doing the action (but not always). You can stick a question word (who or what) before the verb, and the answer to the question will be the subject. For example, who has a green spaceship? I do. Or What surprised my sister? That aliens exist.

(2)	a.	I have a green spaceship.	The subject of have is I.
	b.	I think that aliens exist.	The subject of think is I, and the subject of exist is aliens.
	c.	That aliens exist surprised my sister.	The subject of exist is aliens, and the subject of surprised is that aliens exist.
	d.	You have seen an alien with a telescope.	The subject of (have) seen is you.
	e.	Have you seen an alien?	The subject of (have) seen is you.
	f.	What have you seen?	The subject of (have) seen is you.

Step three. Identify the clauses of the sentence.

There will be one clause for each main verb (but not auxiliary!). I have put brackets around each clause in (3). Notice that main clauses include the embedded clauses. Embedded clauses can be the object of the verb, as in (3b), or the subject of the verb, as in (3c).

(3)	a.	[I have a green spaceship].
	b.	[I think [[that aliens exist].
	c.	[[That aliens exist] surprised my sister].
	d.	[You have seen an alien with a telescope].
	e.	[Have you seen an alien]?
	f.	[What have you seen]?

Step four. Classify each clause.

For each clause, identify whether it is question or a statement. If it is question, identify whether it is a yes-no question or a content question. If it is a content question, figure out where the question word started before movement. You can figure out where the question word started out by saying the echo question: You have seen WHAT? or by answering the question and looking at where the answer goes: What have you seen? I have seen an alien.

(4)	a.	[I have a green spaceship].	This is a statement.
	b.	[I think [[that aliens exist].	The main clause is a statement and the embedded object clause is a statement.
	c.	[[That aliens exist] surprised my sister].	The main clause is a statement and the embedded subject clause is a statement
	d.	[You have seen an alien with a telescope].	This is a statement.
	e.	[Have you seen an alien]?	This is a yes-no question.
	f.	[What have you seen]?	This is a content question. The wh-phrase started in object position.

Step five. Identify the modifiers.

Identify the modifiers, such as adjectives and prepositional phrases. Figure out what they modify. Watch for ambiguity! Maybe it is possible that they can attach to more than one thing. In our set of six sentences, there are two modifiers, green in the (a) sentence and with a telescope in the (d) sentence. Green modifies spaceship. But with a telescope is ambiguous! It can modify an alien or seen. You can use constituency tests to see the two different meanings. For example, in (5), the two different meanings are illustrated with the movement test.

(5)	a.	It is [an alien with a telescope] that you have seen.
	b.	It is [an alien] that you have seen [with a telescope].

In (5a), an alien with a telescope is a constituent. This means that with a telescope modifies alien and is sister to alien (according to the Principle of Modification). This constituency structure derives the meaning that the alien has the telescope. In (5b), an alien is a constituent that does not include with a telescope. This means that with a telescope does not modify alien; instead it modifies the verb and is sister to the verb. This constituency structure derives the meanings that the telescope was used for the act of seeing.

Step six. Draw the spine of the clause.

Start at the top of the page with an S. Draw two to four branches down from the S—one for the subject, one each for the auxiliary or negation if they are there, and one for the VP. Now draw one or two branches down from the VP—one for the V head and a second one for the object if there is an object. [caption id="attachment_3751" align="alignnone" width="300"] Figure A1.22: The spine for a declarative main clause[/caption]If it is a question, also put a CP on top of the TP. [caption id="attachment_3746" align="alignnone" width="300"] Figure A1.23: The spine for a main clause question[/caption]If there is an embedded clause in your sentence, check again if it is a subject clause or an object clause. If it is a subject clause, put it under the far left branch under the S, instead of a subject NP. If it is an object clause, put it under the right branch under VP, instead of an object. Draw the spine of the embedded clause, starting at CP, with the C head on the left as either that or null. Under the CP, draw an S that contains an NP and a VP, just like you did with the main clause. [caption id="attachment_3753" align="alignnone" width="300"] Figure A1.24: The spine for a sentence with an embedded subject clause[/caption] [caption id="attachment_3752" align="alignnone" width="300"] Figure A1.25: The spine for a sentence with an embedded object clause[/caption]

Step seven. Fill in the subjects, objects, and modifiers.

Draw the subject and object NPs in the subject and object positions. If there are no modifiers, an NP will usually be two branches, one for the determiner and one for the head noun. If there are modifiers, add as many extra branches as you need. [caption id="attachment_3738" align="alignnone" width="300"] Figure A1.26: Tree diagram of "I have a green spaceship"[/caption] [caption id="attachment_3743" align="alignnone" width="257"] Figure A1.27: Tree diagram of "I think that aliens exist"[/caption][caption id="attachment_3754" align="alignnone" width="300"] Figure A1.28: Tree diagram of "That aliens exist surprised my sister."[/caption]If there are any modifiers for the verb (such as a PP or an adverb), add branches to the VP as needed. In English, adverbs can go to the left or right of the verb head, but PPs always go to the right. Remember again to watch out for ambiguity! Watch out especially for constituents at the end of the sentence. They might attach to the VP or they might attach to the object. If there are two clauses, they might also attach to either the main clause or the embedded clause! Use your constituency tests to figure it out or use the meaning to determine which part of the sentence it modifies. [caption id="attachment_3766" align="alignnone" width="398"] Figure A1.29: Tree diagram of "You have seen an alien with a telescope" where "with a telescope" modifies the verb. This structure means that you used the telescope to see the alien.[/caption][caption id="attachment_3765" align="alignnone" width="417"] Figure A1.30: Tree diagram of "You have seen an alien with a telescope" where "with a telescope" modifies "alien." This structure means that the alien has a telescope.[/caption]If one of the NPs or PPs is particularly complex, like in (6), break it down into smaller pieces, and do one piece at a time.

(6)	The outlandish lady with the purple purse from Wal-Mart on her elegant arm

First, identify the head. Is this NP as a whole identifying a lady, a purse or an arm? It is identifying a lady, so lady is the head of the phrase. Which adjectives and PPs modify lady? The adjective outlandish and the PP with the purple purse from Wal-Mart on her elegant arm both modify lady. So we can draw an NP with four branches: one for the determiner the, one for the adjective outlandish, one for the head N lady, and one for the very long PP. Next you can repeat the steps with the PP. The PP with the purple purse from Wal-Mart on her elegant arm begins with the P with and contains an NP headed by purse. So under the PP, you can draw two lines, one for the P with and one for the long NP. Next, ask what modifies purse? Keep repeating these steps until you have identified all of the constituents and what each one modifies. [caption id="attachment_3695" align="alignnone" width="562"] Figure A1.31: The tree diagram for example (6).[/caption]

Step eight. If it’s a question, add the movement.

If there’s subject-auxiliary inversion, put angle brackets around the auxiliary under S, and re-write the auxiliary under the C head. Then draw an arrow from the Aux under S to the C head. [caption id="attachment_3737" align="alignnone" width="300"] [VP seen [NP an alien]]]]" width="300" height="274" /> Figure A1.32: Tree diagram of "Have you seen an alien?"[/caption]If there’s wh-movement, put the wh-phrase where it started out in the sentence, as you determined in Step 4, with angle brackets around it. Then add an extra branch to the left under the CP, rewrite the label of the wh-phrase, and draw the internal structure of the wh-phrase underneath. Then draw an arrow from where the wh-phrase started to where it ended up. [caption id="attachment_3763" align="alignnone" width="300"] [VP seen ]]]" width="300" height="287" /> Figure A1.33: Tree diagram of "What have you seen?"[/caption]

Step nine. Double check your answer!

Here are some things you should check:

Does every phrase have a head? Does every head (except Det) have a phrase?
If you read the ends of the branches from left to right, do all the words come out in the right order?
Do any branches cross? (They shouldn’t!)
Are your subject clauses in subject position and your object clauses in object position?
Are your subjects under TP, your objects under VP, and your modifiers sister to what they modify?
Are all your words labeled for part of speech?
Is movement clearly indicated with an arrow? Is it clear from your diagram where the movement started and where it ended?

Sometimes you will see trees with triangles in them. For example, there is a triangle in Figure A1.19 of Section A1.6 or in Figure A1.34 below. [caption id="attachment_3767" align="alignnone" width="340"] Figure A1.34 Example of a tree with a triangle[/caption]Triangles are a way to abbreviate the structure. Instead of drawing all the pieces of a constituent, you can write the entire constituent underneath a triangle (instead of branches). They are usually used in papers if part of the structure is not relevant. In your homework and tests, if you are asked to draw a tree and you use a triangle, you might not get full marks for the part of the tree inside the triangle--check with the grading scheme for your course to see if triangles are permitted and how they will be graded! That being said, if you are feeling very overwhelmed, you could use a triangle and you may at least get part marks. Or, even better, you can use a triangle in your rough work to help you break up the sentence into smaller, more manageable parts. Just remember to come back and finish drawing the bits you placed in triangles.

[h5p id="128"]

Syntactic categories

juliedoner — Wed, 15 Nov 2023 05:32:37 +0000

Learning Objectives

By the end of this chapter, you should be able to,

understand how languages can be described in terms of abstract descriptive rules.
use [pb_glossary id="750"]morphology[/pb_glossary] and [pb_glossary id="751"]distribution[/pb_glossary] to determine word categories
identify characteristic properties of core parts of speech (e.g., nouns, verbs, adjectives, adverbs, adpositions, etc)

To build a theory of syntax, we need to look at data. We'll start by considering the following sentences of English. [latex] \ex. \a. The mailman slept. \b. A storm rages. \c. The student cried. \d. A dog barked. \e. The senators lie. [/latex] Here's an obvious observation about these sentences: All the sentences in (1) are different. For instance, they use different words; they mean different things; some of them are in different tenses. Ask any English speaker, and they would agree: these are different sentences. Here's a not-so-obvious observation about these sentences: All the sentences in (1) are the same. Why? Well, they all can be described in the same way. They consist of a [pb_glossary id="752"]determiner[/pb_glossary] followed by a noun followed by a verb. Stated schematically, I can say that,

Sentence_English → Determiner Noun Verb

"A sentence in English consists of a determiner, followed by a noun, followed by a verb."

What have we done? We've translated English into a system of variables. For whichever values (that is, words) that we choose for each of the variables (Determiner, Noun, Verb) we will generate a sentence of English.

This means that I, as an English learner/speaker, don't have to memorize every combination of determiner, noun, and verb in English. All I have to know is this rule, and I can generate an infinite number of sentences (...as long I know the possible determiners, nouns, and verbs in English).

The goal for every generative syntactician is the set of rules that describe a human language --- and nothing else. That is, we want to find the rules that can accurately describe what is, and is not, a sentence in a language.

In general, syntactic theories (of which there are many) differ on what the rules should look like and what the background assumptions should be (that is, what is given, and what must be stated as part of the rule). But the goal is always the same: a precise and accurate description of language. This is what we are aiming for in this class.

So far we are able to capture only a limited set of data. Obviously it is possible to say many more things in English than sentences with [pb_glossary id="755"]intransitive[/pb_glossary] verbs. Let's look at some more examples.

[latex] \setcounter{ExNo}{1} \ex. \a. The happy mailman slept. \b. A terrible storm rages. \c. The new student cried. \d. A stubborn dog barked. \e. The stupid senators lie. [/latex]

Our rule for Sentence_English doesn't allow us to generate the sentences in (2), because the rule doesn't know what to do with words like happy, terrible, new, stubborn, and stupid. So let's make another rule.

Sentence_English → Determiner Adjective Noun Verb "A sentence in English consists of a Determiner, followed by an Adjective, followed by a Noun, followed by a Verb."

By itself, this rule generates all the sentences in (2), but now it over-generates, because according to the rule, every sentence in English must consist of the sequence Determiner-Adjective-Noun-Verb. So according to this new rule, the sentences in (1) are not possible sentences in English because they don't have adjectives.

So we conclude that we need two rules for English. Now our [pb_glossary id="753"]grammar[/pb_glossary] fragment of English is the following

A grammar fragment of English

Rule 1: Sentence_English → Determiner Noun Verb.

"An English sentence consists of a determiner followed by a noun followed by a verb."

Rule 2: Sentence_English → Determiner Adjective Noun Verb.

"An English sentence consists of a determiner followed by an adjective followed by a noun followed by a verb."

This is simply a logical step. Rule 1 and Rule 2 both correctly generate the sentences in (1) and (2), and they don't generate any sentences that aren't in English, like *Mailman happy slept the. Our rules predict that this sentence shouldn't be an English sentence, because our rules do not allow this sequence of elements.

Together, Rules 1 and 2 are a grammar fragment. They just describe some of what is possible in English. To continue the process and develop a complete grammar, we'd look at more data, and determine the rules that describe the data. We would do this until we developed a set of rules that could generate all the possible sentences in English, and wouldn't generate any sentences that weren't in English. We would then have a complete grammar of English.

However, there are a couple of things we seem to be missing in this grammar fragment. Foremost, these rules don't really capture the relationships between the words. For instance, intuitively, we have the sense that the happy mailman "goes together," like it's a unit, more than mailman slept. For instance, I can replace the happy mailman with he and the sentence means the same thing: He slept. I cannot replace mailman slept with anything to get an equivalent sentence: *The happy he. (This is called substitution; we'll talk about it in the next chapter.) Thus, if we want our rules to truly represent English, then we want to represent that the happy mailman goes together to the exclusion of slept.

We can actually re-write our grammar fragment to capture this grouping. We'll use the term "Determiner Phrase" (abbreviated DP) to refer to the sequence Determiner-Adjective-Noun or Determiner-Noun.

A revised grammar fragment of English Rule 1: Sentence_English → DP Verb "A sentence in English consists of a determiner phrase followed by a verb." Rule 2: DP → Determiner Noun "A determiner phrase consists of a determiner followed by a noun." Rule 3: DP → Determiner Adjective Noun "A determiner phrase consists of a determiner followed by an adjective followed by a noun."

By adding the term Determiner Phrase, we can now correctly represent that the words the happy mailman go together. Putting a determiner phrase together with a verb makes a sentence in English. I can choose either one of the DP rules, Rule 2 or Rule 3, and put it into Rule 1 to make a sentence. Importantly, our revised grammar fragment does all the things that our first attempt did, that is, it gets all the words in the right order, and does not generate sentences of English that don't exist.

This level of abstraction raises two issues.

Issue 1: What are the chunks? How do we divide up the sentence? How do we know what goes with what? For instance, how do know that there isn't a verb phrase as well? And if so, how do we determine what gets included in the determiner phrase, and what is in the verb phrase.

This topic is addressed in the chapter titled Constituency and so we'll put it aside for now.

Issue 2: What do we call the chunks? Assuming that we can divide up every sentence into chunks, what do we call each of the chunks? Stated differently, What are the categories for the variables? For example, why did I choose to call the happy mailman a Determiner Phrase and not a Noun Phrase, or an Adjective Phrase?

The second question is asking about [pb_glossary id="754"]syntactic categories[/pb_glossary]. A noun has the syntactic category of N. A determiner has syntactic category D. A verb has the syntactic category of V, an adjective has the syntactic category of A, etc. A noun phrase is a group of words that "acts like" a noun. A determiner phrase is a group of words that "acts like" a determiner, etc.

The names are essentially arbitrary. We're simply saying that there are classes of things that behave similarly. Table, dog, teacher, brick, honesty, ..., all behave in a similar fashion (in some respects). Lick, sit, own, jump, describe, believe, ..., all behave in a similar fashion (in some respects). When we say that something is a "noun" and has category N all we're saying is that that thing "acts like" the other noun-words. Likewise, when we say that something is a verb, we're simply saying that that thing "acts like" the other verb-words.

What does it even mean to "act like noun" or "act like a verb?" In what way are "nouns" differentiated from "verbs" or "adjectives?" In defining syntactic categories, it is tempting to rely on what you were probably taught in school: nouns are "people, places, or things" and verbs are "actions," adjectives "describe," etc. That is, we can attempt to define syntactic category using meaning.

This turns out be wrong for a number of reasons. For instance, we can all agree that idea is a noun, but it's debatable whether it's a person, place, or thing. Or what about the (uncontroversial) noun destruction. That's an action! The destruction of the city by Godzilla describes an event---it's clearly not a person, a place, or a thing.

Likewise, not all verbs are actions. Tasha owns a car contains the verbs owns. Is owning an action? Only under a very loose definition of "action." What about the verb seems in Abigail seems tired. Is seem an action? Doubtful... Perhaps most problematic for the idea that we can use meaning to define syntactic category comes when we look at languages other than English. For instance, if you've studied a Romance language, you've learned that certain concepts which are expressed in English using adjectives are expressed in Romance using verbs and nouns. This happens to be a robust strategy found across the world's languages, occurring beyond Romance languages, like, e.g., Swahili. [latex] \setcounter{ExNo}{2} \ex. \ag. Yo tengo hambre\\ 1\textsc{sg} have.1\textsc{sg} hunger\\ \trans `I'm hungry.' (Literally: `I have hunger.') \hfill Spanish \bg. J'-ai faim\\ 1\textsc{sg}-have.\textsc{1sg} hunger\\ \trans `I'm hungry' (Literally: `I have hunger.') \hfill French \cg. nina njaa\\ have.1\textsc{sg} hunger\\ \trans `I'm hungry' (Literally: `I have hunger.') \hfill Swahili [/latex] If we (as English speakers) were to use meaning to define word category, we might be tempted to call words like hambre, faim, and njaa adjectives meaning "hungry." This would clearly be wrong, because in fact they're nouns, as the literal translations suggest. So we simply cannot use meaning to determine syntactic category; it doesn't help us to define categories within a language, and it doesn't help us define across categories across languages. So how actually do I know that idea, table, hambre, faim, and njaa are nouns if I can't use meaning? Or stated differently, how can I uniquely define a class of words in any particular language? There are two kinds of reliable evidence we can use to classify syntactic categories.

Morphological properties. What are the pieces of [pb_glossary id="750"]morphology[/pb_glossary] that distinguish a particular word class?

For instance, in English only nouns can appear with plural morpheme (typically -s). Again in English, only verbs can appear with tense morphology (like past tense -ed). Morphological properties are going to be specific to a language. The morphological properties of nouns in Swahili are different from the morphological properties of nouns in English.

Distributional properties. Where do members of a particular noun appear with respect to other categories.

For instance, a determiner in English must always be the first thing in a determiner phrase. A verb in English must always be preceded by a determiner phrase. Distributional properties, too, are going to be specific to a language. In Swahili, determiners generally follow nouns, whereas in English, they strictly precede nouns.

Morphology refers to pieces of words. In any particular language, every word class may have its own specific morphological pattern. In English, only nouns can take a plural [pb_glossary id="822"]suffix[/pb_glossary]: ideas, tables, children, etc. Adjectives cannot inflect for plural in English: *reds, *sads, etc. Only verbs in English can inflect for (be marked with) tense and aspect information. walked, walking, *childed, childing. Only adjectives in English can inflect for the comparative suffix: sadder, *childer. Importantly, morphological properties are going to be language specific. For instance, in Estonian, nouns inflect for [pb_glossary id="823"]number[/pb_glossary] (singular or plural), just like in English and also for [pb_glossary id="838"]case[/pb_glossary] (which indicates the role the noun plays in the sentence).

Partial declension of the word *vend* "brother" in Estonian
	Singular	Plural
Nominative	vend	venna
Genitive	venna	venda
Partitive	venda	vendi
Illative	vennasse	vendasse

Thus, in Estonian, if a word inflects for both number and case in the same way, then we have evidence that it's a noun. Just like in English, if a word inflects for tense and aspect, then we conclude that it's a verb. If a word in English can take the comparative morpheme -er, then we conclude that it's an adjective. The second piece of evidence we can use to determine syntactic category is distributional evidence. Every word class goes in a particular position with respect to other word classes. In English, the only thing that can appear in the underlined space is a noun: The ____ is... . That is, only nouns can appear between a determiner and [pb_glossary id="824"]copula[/pb_glossary]. Or stated differently: only certain words can appear between a determiner and copula in English, and we call those words "nouns." In French, the only thing that can appear in the underlined space in a comparative construction is an adjective. [latex] \setcounter{ExNo}{3} \exg. Marie est plus {\underline{\hspace{50pt}}} que Anne\\ Marie is more {\underline{\hspace{50pt}}} than Anne\\ \trans `Marie is {\underline{\hspace{50pt}}} than Anne \hfill French [/latex] Or stated slightly differently, the only kind of word that can appear in the configuration above is a word like grande "tall," or intelligente, "smart," and we give those words the label "adjective." (Notice that adjectives in French also have a morphological distinction: they inflect for masculine and feminine gender, in addition to singular and plural.) Again, distributional properties of syntactic categories are going to be language-specific. We can't make general statements like "adjectives always precede nouns" because there are many languages in which they come after nouns. Similarly we can't say something like "verbs always come after subjects" because there are many languages in which the verb comes after the object, or even before the subject. To distinguish syntactic categories for any particular language, you have to look at the properties of that particular language.

[embed]https://www.youtube.com/embed/6I56Q1iHzJQ[/embed] Parts of speech (length: 2m 46s)

Where we're going

Now that we can identity parts of speech (using distribution and morphology), we can use that information to write our abstract rules about language, as demonstrated earlier. That is, we can translate any language into a system of variables.

We can also determine the category of groups of words. For instance, suppose I wanted to know what category the group of words happy mailman is. From distributional evidence, we would conclude that it behaves like a noun, not like an adjective (or a verb, adverb, etc). For instance, we can't use this phrase to modify another noun (5a), and it can go in the location that only nouns can go (5b).

[latex] \setcounter{ExNo}{4} \ex. \a. *The happy mailman table \b. The {\underline{happy mailman}} is {\ldots} [/latex] We're going to use this information to help us determine how to "chunk up" the sentences we find, and in turn how to write the rules that are in our grammar fragment.

Things to remember

Language description in terms of abstract rules
Using data to create a grammar fragment
How to use morphological and distributional properties of words to determine syntactic categories.

Advanced

Changing categories

One of the complications that frequently crops up when we're considering category is that words can change category. Consider the English word cut. If I asked you what category cut was you could say, "It's a verb." Or you could say, "It's a noun." Both answers would be correct. [latex] \setcounter{ExNo}{5} \ex. \a. Sarah cut the bread. \hfil Verbal \textit{cut} \b. It was a deep cut. \hfil Nominal \textit{cut} [/latex] One question that this raises is, for any particular word, which category (if any) is the "true" form? That is, is the word cut fundamentally a verb which can be turned into a noun? Or is it a noun which can be turned into a verb? A third option is that it's fundamentally neither a noun nor a verb! It's simply some abstract semantic information which can be used either as a noun or a verb. It's also interesting to think about the limit of possibilities For cut, there is an upper bound on the category. It can be a verb, it can be a noun, and it can also be an adjective (technically a participial form): the cut bread. That's probably it. But other words can definitely have more categories. Consider the word down: [latex] \setcounter{ExNo}{6} \ex. \a. The ball is down the hole. \hfill Preposition \b. John sat down. \hfill Verbal particle \c. John downed the soda. \hfill Verb \d. The Chiefs fumbled on the first down. \hfill Noun \e. Don't talk to Mary, she's feeling down right now. \hfill Adjective [/latex] Given the abundance of forms for down, why are some "alternations" not allowed? For instance, unlike cut, the noun chop doesn't naturally have similar noun form: Mary chopped the wood. ??It was a deep chop. Why not? Why are some words restricted in "changing" categories, and others less so?

Morphological complexity in word formation

When a word is morphologically ambiguous between different categories, we call it conversion. So the difference between cut as a verb and cut as a noun is descriptively an instance of conversion. Sometimes, in fact often, there are morphological reflexes of changing category. Consider the (famous) example of the verb destroy, which has the noun form destruction. Unlike cut, the nominalization of destroy involves at least addition of the suffix -tion, plus a change in the stem of the verb from destroy to destruc- One question we can ask about such processes is how "productive" they are. By this, we mean, "How many words exhibit a similar change?" For instance, what other verbs form nouns with -tion? If we find that it's a lot of verbs, we might say that this kind of nominalization is generally productive. If it only happens with a few idiosyncratic verbs, then we might say that it isn't fully productive. English has a few fully productive category changing processes. Consider -ness affixation. Basically any adjective can be suffixed with -ness to form a noun.

Adjectival *-ness* nominalizations in English
Adjective	Noun
happy	happiness
tough	toughness
mean	meanness
red	redness
sad	sadness

Interestingly, such morphological affixation "stacks," meaning that I can add multiple affixes to a word, each one changing the category. Consider the following productive affixation processes:[footnote]Don't be fooled by the orthography! For historical (and phonological) reasons, we sometimes use a slightly different spelling for some affixes, but it's still the same affix.[/footnote]

Deverbal derivations in English
Verb	Adjective (-able)	Noun (-ity)
break	breakable	breakability
find	findable	findability
drink	drinkable	drinkability

Other languages have a much richer process of category alternations. In Dumi, a Kirati (Tibeto-Burman) language spoken in Eastern Nepal, adjectives can be made from verbs by putting -sa at the end.

Adjective formation in Dumi
Verb	Gloss	Adjective	Gloss
rek	'sharpen'	reksa	'sharp'
sur	'wash'	sursa	'washed'
tuk	'keep'	tuksa	'kept'

In K'iche', a Mayan language spoken in Guatemala, verbs can be suffixed with -ib'al to forms nouns meaning "location for VERBing."

Locative deverbal nouns in K'iche'
Verb	Gloss	Noun	Gloss
atin	'bathe'	atinib'al	'bath, place for bathing'
war	'sleep'	warib'al	'bedroom, dormitory'
el	'leave'	elib'al	'(an) exit'

Overt realizations of a category changing affixes introduce a number of interesting theoretical questions. For instance, what actually is a category? Is it just the affix that attaches last? That is, since -ness always "makes" a noun, can we just say that -ness is the category noun? This is schematized below. [latex] \setcounter{ExNo}{7} \ex. sad\textsubscript{A}+ness\textsubscript{N} [/latex] But if that's true, does that mean when we have conversion (e.g., cut_Noun versus cut_Verb), do we have to say that there's a null affix, corresponding to a noun or verb? [latex] \setcounter{ExNo}{8} \ex. \a. cut\textsubscript{V}+$\emptyset$\textsubscript{N} \hfil or, \b. cut\textsubscript{N}+$\emptyset$\textsubscript{V} [/latex] Another question concerns the stacking noted earlier. Take the word breakability. This consists of three distinct morphemes, each associated with a different category. Break is a verb; -able (here spelled -abil) attaches to verbs and makes adjectives; -ity attaches to adjectives and makes nouns. It's clear that there is an order in how the elements stack. -Able has to attach to break before -ity can. What about when there is a suffix and a prefix? Is there also an ordering? Consider the word unbreakable. This also consists of three morphemes: un-break-able. At first glance, it might not seem possible to determine the order of attachment of un- and -able. But in fact, it must be the case that -able attaches first, to form the word breakable, and then un- attaches after that, to form the word unbreakable. If it happened in the opposite order, we would first make the word *unbreak---but this isn't a word! More importantly, the idea that there is a determined order of "introduction" for each piece of a word---and that order might not be the same as the linear order---will be a central observation when we start looking not at individual words, but at groupings of words, otherwise called constituents. What is remarkable is that the same processes that underlie our understanding of syntax also underlies our understanding of morphology (for the most part). These topics are discussed more in depth in LING 527/727.

Constituency

juliedoner — Wed, 15 Nov 2023 05:32:37 +0000

Learning Objectives

By the end of this chapter, you should be able to,

identify and use seven constituency tests for determining structure
understand how to interpret constituency tests to expand grammar fragments
understand the limitations of constituency tests, and what to do with false positives and negatives

The program of [pb_glossary id="840"]generative[/pb_glossary] syntax is to find the set of rules that describes a language (really that describes every language) and that doesn't describe things that aren't in a language. In chapter one, we abstracted away from the words themselves to talk about syntactic categories, which allowed us to write descriptive rules about a language. Our English grammar fragment is the following right now.

Grammar fragment for English

Sentence_English → Determiner Phrase Verb

Determiner Phrase → Determiner Adjective Noun Determiner Phrase → Determiner Noun

We are thus able to talk abstractly about what is possible in English syntax. This way of describing language raised two issues, repeated here.

Issue 1: What are the chunks? How do we divide up the sentence? How do we know what goes with what? That is, how do I know that the sequence "Determiner Adjective Noun" forms a group?

Issue 2: What do we call the chunks? Assuming that we can divide up every sentence into chunks, what do we call each of the chunks? Stated differently, What are the categories for the variables?

In response to Issue 2, we developed ways to identify syntactic categories: morphology and distribution. Thus, we can now empirically distinguish between parts of speech in any particular language because each part of speech will have specific morphological and distributional properties. I can determine when something is a noun, a verb, an adjective, etc through not only how it looks (morphology) but also where it appears with respect to other words (distribution).

Importantly, we also noticed that group of words also have syntactic categories. When I put happy and mailman together, I get something that distributes like a noun. That is, happy mailman goes in all the same places that mailman goes, like in between determiners and copulas: the happy mailman is... . (Alternatively, the group of words happy mailman doesn't act like an adjective: it doesn't do in the places that adjectives go, like in front of a noun: *the happy mailman carpenter.) We call a string of words that forms a group a [pb_glossary id="196"]constituent[/pb_glossary]. A constituent, by definition, has a syntactic category. This is because a constituent has a syntactic distribution. As I just stated, happy mailman has the distribution of a noun. That is, the entire string of words "acts like" a noun, not an adjective, not a verb, etc. Thus, there is a sense---an intuition at this point---that happy mailman is "noun-y." The tests that are introduced below will provide empirical evidence that this is so.

What constituency means, practically

When I claim that a string of words is a constituent, I'm claiming that all of those words are grouped together under one label. Note that a single word, is, by definition, a constituent. A sentence is also, by definition, a constituent. What I'm ultimately trying to do when I figure out constituency is I'm trying figure out how information is packaged. How does any language put its pieces together to make a meaningful utterance? Precisely, in any given language, I want to know which words go with which words. This allows me to write the simplest possible rules. For instance, with our DP rule, I can capture the idea that The mailman slept and The happy mailman slept both involve a subject and verb. Identifying The mailman and The happy mailman as the same "thing" simplifies my understanding---and representation---of the English language. I can reduce sentences to grammatical roles like Subject-Verb, or Subject-Verb-Object by identifying that the the term "subject" may involve something that consists of more than one word, like the happy mailman.

It is worth noting that in English, it is difficult if not impossible to use morphology to determine constituency, so all the tests below involve distributional evidence. But morphology is actually relevant and important in other languages in determining constituency. For instance, in Swahili, adjectives must "agree" with the nouns that they modify. This typically means that the prefix of an adjective must match the prefix on a noun. [latex] \ex. \ag. {\underline{m}}toto {\underline{m}}zuri\\ child good\\ \trans `good/beautiful child' \bg. {\underline{wa}}toto {\underline{wa}}zuri\\ children good\\ \trans `good/beautiful children' [/latex] The same point can be made in Romance languages. In Spanish, adjectives modifying masculine singular nouns appear with particular distinctive morphology -o, and adjectives modifying feminine singular nouns appear with different morphology -a. It is thus possible to identify groups of words using morphology in languages which have "richer" inflectional morphology than English. However, such morphological constituency diagnostics come with their own (considerable) complications; we will not be able to address them fully in this class.

Constituency tests

1. Substitution tests

[pb_glossary id="756"]Substitution[/pb_glossary] tests are a family of tests that can be used to determine both constituency and category label. The premise behind these tests is that if you can replace the string of words in questions with something whose category you know, then the string of words must share a category with the thing you've replaced it with. For instance, if I can replace, or substitute, the tired doctor with something whose category I know, then I can conclude that the tired doctor has the same category as whatever I replaced it with.

Applying the test: The tired doctor slept. → That's right, she slept. Step-by-step:

Identify the string of words whose constituency you want to test.
Identify an appropriate substitution word, whose category you know.
Replace the string of words (and nothing else) with the chosen word.
Check the grammaticality. If it's grammatical and it means exactly the same thing, then you can conclude it's a piece of evidence in favor of treating the string as a constituent.

[embed]https://youtu.be/U6UnaJSHXcY[/embed] Substitution test demonstration (length: 1m 29s)

There are two tricky things about substitution tests. First, you need to find an "appropriate" substitution word. What is and is not appropriate changes depending on what the string of words is. For instance, it wouldn't makes sense to chose the pronoun it to replace the tired doctor, because it refers to an inanimate object (most of the time). In order to perform a substitution test, you have to make a hypothesis about what constituent you think the string of words might be. Below is a list of [pb_glossary id="209"]pro-forms[/pb_glossary] that are used to test various constituents.

Various pro-forms

Substitution of Determiner Phrases ⇒ Use a pronoun (he, she, it, they, etc). (Make sure you use the correct form of the pronoun! Don't use he when him works better, etc.)
- Example: Carol saw the trees on the hill → Bill saw them, too.
Substitution of Verb Phrases ⇒ Use do so, changed to reflect the appropriate tense if necessary.
- Example: Carol saw the trees on the hill → Bill did so, as well.
Substitution of Locative Prepositional Phrases ⇒ Use there. (This won't work for all prepositional phrases, only ones which refer to a location.)
- Carol sat on a bench → Bill sat there, too
Substitution of clauses ⇒ Use so. (Again, this won't work for all clauses, only certain ones.)
- Example: Carol thinks that Isabelle left → Sam thinks so, too

Thus, if you think it's a DP, use an appropriate pronoun. If you think it's a verb, use do so. Note that there's nothing wrong with applying more than one test. If you're not sure which substitution test to do, just do them all! Do any of them work? If so, you've learned something about the category of the constituent! But note that not all categories permit substitution. There isn't really a substitute for adjectives in English. Likewise, in many languages, there is no substitute for verb phrases. If there is no substitute for that category, then you simply cannot apply the test. (If that comes up on an assignment, just right "N/A.") The second tricky thing about substitution tests is that you often need to set up a context. For instance, the sentence Bill did so, too doesn't make sense out of context. I can't walk into a room and say this sentence out of the blue. There needs to be an [pb_glossary id="759"]antecedent[/pb_glossary] for the pro-form (do so). The same is true of any of the pro-forms above. I can't just walk into a room and say She slept, unless you know who I'm talking about. That's why when I demonstrate substitution, I typically put the sentences into a small discourse or exchange. "The tired doctor slept. That's right! She slept." The actual test is the sentence She slept, but the preceding discourse provides an antecedent for she that makes this utterance felicitous.

2. Sentence Fragments

The second constituency test we can use is called a [pb_glossary id="758"]sentence fragment[/pb_glossary] test. When you apply this test, you're asking whether the string of words in question can stand on its own. The best way to set this up is to see whether the string of words can be the response to a question.

Applying the test: Who slept? → The tired doctor. Step-by-step:

Identify the string of words whose constituency you want to test.
Write down that string of words separate from the sentence.
Try to come up with a question which could have the string as a response.

[embed]https://youtu.be/vNFLrZXgf14[/embed] Fragment test (length: 1m 12s) credit: Emma Scott [embed]https://youtu.be/rFgoTu5R_LY[/embed] Fragment test (length: 1m 39s)

Note that the fragment test alone doesn't tell you what the category of the constituent is. In contrast, the substitution test, when done correctly, does tell you about the category of the constituent. If I successfully substitute a string of words with a pronoun, then I can conclude that the string of words has the same category as the pronoun. (They're both DPs, as we'll talk about in chapter 5). If I substitute a string of words successfully with do so, then the string of words in question is a Verb Phrase, etc. This is, essentially, marshaling the fact that the classes of words share a distribution, as discussed in the previous chapter.

If you think about the sentence fragment test, in many ways, it's really just a substitution test! If I can ask the question Who slept?, and answer The tired doctor, then in a sense who has substituted for the tired doctor.

3. Movement/Displacement tests

The third constituency test we can use is also a family of tests, called movement tests (or also displacement tests). In this class we'll cover four tests. In these tests, you're taking your original sentence, and "transforming" it by moving things around. The idea behind the movement tests is that you can only move a constituent. When first learning these tests, the easiest thing to do is to memorize a "template," and then plug the sentence into the template. The general idea is that you want to label one part of the sentence as "A"---this is the string of words you're wondering about---and the other part of the sentence as "B." Then you want to put A somewhere else, leaving B alone.

a) It-clefting.

The template for it-clefting is, "It was __A__ that __B__."

Applying the test: The tired doctor slept → It was the tired doctor that slept Step-by-step:

Identify the string of words that you want to test.
Label the string of words you want to test "A," and everything else in the sentence "B."
Fill in the template, It was ___A___ that ___B___. (You may need to change the tense of was.)
Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/7T4A1gin5Io[/embed] It-clefting demonstration (length: 1m 17s)

IMPORTANT! If you find that the test works and you get a grammatical sentence, then you can conclude that A is a constituent, but you cannot conclude that B is a constituent. For example, suppose I apply the it-clefting to the following string of words.

The doctor treated a sick patient yesterday → It was a sick patient that the doctor treated yesterday.

The test indicates that a sick patient is a constituent, but it does not indicate that the doctor treated yesterday is a constituent. You can only make conclusions about the part that you've moved, not about the part that is left behind. This goes for all of the movement/displacement tests.

b) Pseudo-clefting

The template for pseudo-clefting is, "__A__ is who __B__".

Applying the test: The tired doctor slept → The tired doctor was who slept

Identify the string of words you want to test
Label the string of words you want to test "A," and the rest of the sentence "B".
Fill in the template, A was who/what B.
Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/eYG2p0CRCvQ[/embed] Pseudo-clefting demonstration (legnth: 1m 15s)

c) All-clefting

There are number of different kinds of clefts---some are language specific. The final cleft that we'll use is an all-cleft. The template for all-clefting is, "__A___ was all (that) __B__".

Applying the test: The tired doctor slept → The tired doctor was all who slept

Identify the string of words you want to test
Label the string of words you want to test "A," and the rest of the sentence "
Fill in the template, A was all who/that B.
Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/LwGpznHQmaE[/embed] All-clefting demonstration (length: 2m 14s)

d) Topicalization

The final movement/displacement test we'll use is topicalization. This is, in some ways, the easiest movement test, because all you're doing is (potentially) rearranging the sentence. You're not adding any new words in. The template for topicalization is, "__A__ , __B___". The following are all examples of topicalization. If you're a native English speaker, you'll want to read these with a slight pause at the comma. [latex] \setcounter{ExNo}{1} \ex. \a. Greta, I like, but not Talia. \b. Olives with blue cheese, I hate. \c. \textit{Anna Karenina} was okay, but \textit{War and Peace}, I loved. [/latex] Note that it's possible to topicalize a subject, but this might not result in a new word order, just a different intonational contour.

Applying the test: Maisha met the tired doctor → The tired doctor, Maisha met.

Identify the string of words you want to test
Label the string of words you want to test "A," and the rest of the sentence "
Fill in the template, __A__, __B__.
Check the grammaticality of the resulting sentence.

[embed]https://www.youtube.com/embed/6s7wCevQz8I[/embed] Topicalization (length: 1m 50s)

Like substitution tests, you want to make sure with topicalization that you've set up the context adequately. Often the best topicalization tests involves contrastive topics. You can contrast something with something else. This is what is shown in specifically in (2c), where the topicalized phrase War and Peace is being contrasted with Anna Karenina.

Keep in mind that while I've shown you four movement/displacement tests, there are many, many more. Any way you can "rearrange" a sentence, while still keeping the meaning, is potentially a constituency test.

4. Coordination

The final constituency test we'll cover is one that works in every language, but it is also the one that is easiest to give you false positives. There are a lot of ways to do the coordination test wrong.

Applying the test: The tired doctor slept → The tired doctor and Sarah slept.

Identify the string of words you want to test
Place a coordinator directly after that string of words.
After the coordinator, add something that you know is a constituent, and that you think is of the same category as the phrase you're testing. For instance, I chose "Sarah" above because I think the tired doctor might be a DP.
Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/5Uc4UDw6ilc[/embed] Coordination test (length: 2m 48s)

It's very important to keep in mind that constituency tests are not infallible. We often get false positives and false negatives. That's why we have so many tests, so that we can confirm our results! We don't ever really know why a test fails. Therefore, we cannot rely on negative results, only positive results. For instance, it-clefting and pseudo-clefting give different results when testing verbs (really, verb phrases):

Nekesa grew tomatoes → *It was grow tomatoes that Nekesa did.
Nekesa grew tomatoes → Grow tomatoes is what Nekesa did.

What do we conclude when have such conflicting evidence? Well, hopefully our other tests will help us decide whether a piece is or isn't a constituent. In the end, there may be additional factors about certain tests and/or configurations that influence the grammaticality of the test. Something about it-clefting in English doesn't work with Verb Phrases. A deeper investigation into the syntax and semantics of clefts may provide an answer, but it is, unfortunately, beyond what we can cover in this class.

Where we're going

We want to know how to write the rules that describe a language. Our rules need to be informative, and they need to accurately and precisely reflect the language in question. The constituency tests inform us about how our rules should look. If I discover that happy mailman is a constituent, then this should be reflected in our rules. More importantly, if I discover that happy mailman has the distribution of a noun, then I can conclude that this entire string of words is functionally a noun. This is a major step. By claiming that there are "noun phrases" not just "nouns" we are claiming that the category of a single word can determine the category of an entire string of words. It surely cannot be accidental that happy mailman both distributes like a noun, and also has a noun in it. Similarly, the phrase very happy distributives like an adjective and has an adjective in it.

[embed]https://www.youtube.com/embed/3zFRrtsv3NM[/embed] Constituency and category (length: 1m 12s)

Looking forward a bit, our constituency tests are ultimately telling us how our trees should look. Applying constituency tests to the sentence Mary saw a dog would indicate that a dog is a constituent, saw a dog is a constituent, and there are no other constituents (except the entire sentence, Mary saw a dog). To represent these groupings based solely on constituency tests and ignoring categories, my tree would have the following structure. Each ``junction'' where two lines meet indicates that everything under it is a constituent, or one grouping of words.[latex] \begin{tikzpicture}[mytree] \node(top) {} child {node {Mary}} child {node {} child {node {saw}} child {node {} child {node {a}} child {node {dog}} } }; \end{tikzpicture}[/latex]Thus, even without labels, I can correctly represent with this tree that a dog is a constituent, and that saw a dog is a constituent. And indeed, sometimes this kind of tree is the best we can do. Our constituency tests may indicate a grouping of words for which we do not know the category. That's fine! As we noted in the last chapter, the term "category" is really just a way of saying "this is bunch of things that act alike." The actual name of the category itself is (relatively) unimportant in the long run, as long as we can correctly characterize what distinguishes (morphologically or distributionally) a group of words or word.

Things to remember

How to apply and interpret seven constituency tests
What to do when you have mixed results in your constituency tests
How constituency tests inform our syntactic representation (trees)

Advanced

It is easy to overlook how crucial---and absolutely necessary---constituency is. It is one of the most fundamental aspects of Human Language, and it is acquired by all babies at a fairly early age. The affects of constituency can be observed across a wide range of phenomena. Consider the examples (adapted from Chomsky 2021.[footnote]Chomsky, Noam. (2021). Minimalism: Where Are We Now, and Where Can We Hope To Go. Gengo Kenkyu 160: 1-41.[/footnote]). [latex] \setcounter{ExNo}{2} \ex. \a. The robbery at the pet-stores \textbf{was} criminal. \b. The robberies at the pet-store \textbf{were} criminal. [/latex] It is constituency that ultimately determines the form of the verb. We interpret the phrase The robbery at the pet stores as an entire unit, whose head is robbery. Thus, the verb indicates a singular subject: was. In contrast, when the head is plural, robberies, the verb indicates this, too: were. What is crucial here is that the form of the verb is not determined by linear order. The verb doesn't look to its immediate left to see whether that word is singular or plural. If it did, we would expect *The robbery at the pet-stores were criminal and *The robberies at the pet-store was criminal, where the choice between was and were is determined by pet-store(s). That is, we have to "see" the entire subject as a constituent in order to determine the right agreement on the verb.

Language-specific constituency tests.

We've explore above constituency tests in English. When we look cross-linguistically, we find that the same principles apply. Can you substitute the string of words with a word whose category you know? Can you displace the string of words to a different part of the sentence? Can you coordinate the string of words with something whose category you know? What is important to keep in mind is that each of these processes may ultimately look different depending on the language. For instance, in Swahili, the replacement test looks different because pronouns work different in that language. [latex] \setcounter{ExNo}{3} \ex. \ag. Juma ali-ona mtoto mrefu\\ Juma \textsc{past}-see child tall\\ \trans `Juma saw a tall child.' \cg. Juma ali-mu-ona\\ Juma \textsc{past}-see-3\textsc{sg.obj}-see\\ \trans `Juma saw him/her.' [/latex] In Swahili, object pronouns are attached directly to verb. Mu- doesn't linearly replace the string mtoto mrefu---that is, it doesn't go in the same place. But we can still use this as a constituency test, keeping in mind the independent observation that object pronouns simply have to go in a particular position. (Indeed, we could make the same point with Romance languages, in which object pronouns similarly go in front of the verb.)

Nonconfigurationality

Still, while the notion of constituency has crosslinguistic validity, the crosslinguistic picture on constituency tests is fairly complicated. We make a broad (and far from absolute) distinction between languages which are configurational, like English, and those which are nonconfigurational. In a configurational, the elements in any one sentence, have a particular order that they need to be in. For instance, in English, as a rule, adjectives comes before nouns: red coat not *coat red. A nonconfigurational language is described as having "free" word order. Consider the famous case of Warlpiri, a Pama-Nyungan language spoken in Central Australia. [footnote]Hale, Ken. 1983. Warlpiri and the Grammar of Non-configurational Languages. Natural Language and Linguistic Theory 1(1): 5-47[/footnote] [latex] \setcounter{ExNo}{4} \ex. \ag. Ngarrka-ngku ka wawirri panti-rni\\ man-\textsc{erg} \textsc{aux} kangaroo spear-\textsc{nonpast}\\ \trans `The man is spearing the kangaroo.' \bg. Wawirri ka panti-rni ngarrka-ngku\\ kangaroo \textsc{aux} spear-\textsc{nonpast} kangaroo-\textsc{erg}\\ \trans `The man is spearing the kangaroo.' \cg. Panti-rni ka ngarrka-ngku wawirri\\ spear-\textsc{nonpast} \textsc{aux} man-\textsc{erg} kangaroo\\ \trans `The man is spearing the kangaroo.' [/latex] All three orders of words are perfectly acceptable, and all three mean exactly the same thing. (The only rule is that the auxiliary element ka has to be the second thing in the sentence.) The issue that nonconfigurational languages raise is that they appear to disregard constituency. For instance, nouns and their modifiers need not be next to each other. [latex] \setcounter{ExNo}{5} \ex. \ag. \textbf{wawirri} \textbf{yalumpu} kapi-rna panti-rni\\ kangaroo that \textsc{aux} spear-\textsc{nonpast}\\ \trans `I will spear that kangaroo.' \bg. \textbf{wawirri} kapi-rna panti-rni \textbf{yalumpu}\\ kangaroo \textsc{aux} spear-\textsc{nonpast} that\\ \trans `I will spear that kangaroo [/latex] In English, we strictly cannot separate kangaroo from that, a fact which we use to determine that kangaroo and that form a constituent. So if Warlpiri allows us to do such separation, does it have constituency? The answer is yes---but it's complicated. As we'll learn later in class, the order of words we pronounce is not always the order that the words started in. That is, things move, and therefore, we sometimes have discontinuous constituents because the parts of the constituent have been separated by movement. Of course, we need to show that movement has taken place, and we have a suite a diagnostic tests for this; we'll get to them later in the class. The takeaway here is that nonconfigurationality, in particular seemingly discontinuous constituents, ultimately are not an argument against constituency in general.

Phrase-structure grammars

juliedoner — Wed, 15 Nov 2023 05:32:39 +0000

Learning Objectives

By the end of this chapter you should,

be able to build a phrase structure grammar by looking at data,
recognize and interpret constituency tests in another language,
use a phrase structure grammar to build trees,
have a basic grasp of Swahili syntax, and
understand the principles behind a data-driven approach to syntactic theory

Building a Grammar

We now to turn to the question of building an accurate phrase structure grammar. We want our phrase structure rules to be empirically correct, and so they need to correctly describe all and only the phrases that exist in a language. To emphasize a central point of the last chapter, we want our phrase structure grammar for language X to precisely reflect how language X organizes its information. In this chapter, we'll work through how we might accomplish this with Swahili.

Swahili (or Kiswahili as it is pronounced in Swahili) is a Bantu language spoken throughout East Africa by upwards of 100 million people. It is a national language of Kenya, Tanzania, and Uganda, and the Democratic Republic of Congo, and it is also spoken in Rwanda, Burundi, Malawi, Somalia, Zambia, and Mozambique. [caption id="attachment_26" align="aligncenter" width="244"] Map of region where Swahili is spoken[/caption] You can learn to speak Swahili here at KU! Courses are offered every semester and over the summer through the African and African American Studies Department.

Building a Noun phrase rule

Let's try to figure out the phrase structure rules for Swahili nouns. To do this we simply look at (a lot of) data, applying constituency tests. We start small, and then build up. For the sake of simplicity, we will not take the time to confirm all the constituency tests for each phrase that we build. But if we were being diligent, then we would do exactly that.

The following is our first set of data.[footnote]The verb changes from ulianguka to ilianguka in the last sentence because there is a plural subject.[/footnote]

[latex] \ex. \ag. mti ulianguka\\ tree fall.\textsc{past}\\ \trans `A tree fell.' \bg. mti mkubwa ulianguka\\ tree big fall.\textsc{past}\\ \trans `A big tree fell.' \cg. Ndio, huu ulianguka\\ Yes, \textsc{dem} fall.\textsc{past}\\ \trans `Yes, that fell.' \hfill [that=A big tree] \dg. mti mkubwa na mti mdogo ilianguka\\ tree big and tree small fall.\textsc{past}\\ \trans `A big tree and a small tree fell.' [/latex] These sentences provide a wealth of information. First, it looks like in Swahili (just like in English) subjects come before the verb. But based on (1a) and (1d), it looks like adjectives go after nouns in Swahili. The third and fourth sentences are constituency tests ([pb_glossary id="756"]substitution[/pb_glossary] and [pb_glossary id="801"]coordination[/pb_glossary]), showing that the sequence noun-adjective is a constituent in Swahili. With the information from (1), we can write some rules. Let's assume that the category of the noun-adjective sequence is a Noun Phrase (NP). We'll revisit that question in the next chapter.

Swahili grammar fragment (in progress)

S_S → NP V
NP → N
NP → N A

Now our grammar fragment for Swahili allows us to draw the following trees---that is, it represents Swahili's structure in the following ways. [latex] \setcounter{ExNo}{1} \ex. \ag. mti ulianguka\\ tree fall.\textsc{past}\\ \trans `A tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} \ex. \ag. mti mkubwa ulianguka\\ tree big fall.\textsc{past}\\ \trans `A big tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} child {node {A \\ mkubwa \\ big}} } child[sibling distance = 10em] {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]

There is one other phrase structure rule that we assume to be given. It's the coordination rule, shown below.

XP → XP conj XP

Here, "XP" means any phrase---"X" is a variable over categories. The important things to note about this rule is that it only allows coordination of the same category. That is, XP doesn't consist of XP conj YP. Practically, this means that a Noun Phrase can only be coordinated with another Noun Phrase. A Verb Phrase can only be coordinated with another Verb Phrase, etc. The category "conj" is short for [pb_glossary id="771"]conjunction[/pb_glossary]; there are many different conjunctions: and, but, nor, yet, ...

[embed]https://youtu.be/raks-QFct7A[/embed] Coordination (length: 1m 22s)

We have two rules for NP. There is a convention for collapsing this into one rule, which is to use parentheses. So the two NP rules above can be re-written as the single rule below.

Swahili grammar fragment (in progress)

S_S → NP V
NP → N (A)

The second rule can be read, "A Noun Phrase consists of a noun, which is optionally followed by an adjective." To be clear, the second rule contains the information of both NP rules from our first grammar fragment, it's just a conventionalized representation. Let's look at more sentences. [latex] \setcounter{ExNo}{3} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \bg. mti mkubwa mzuri mweusi ulianguka\\ tree big good black fall.\textsc{past}\\ \trans `A big beautiful black tree fell.' [/latex] We learn here that, in fact, nouns in Swahili can be followed two or three adjectives. Indeed, we could keep going! Noun Phrases in Swahili can involve an infinitive number of adjectives (up to cognitive limitations).[footnote]In fact, there actually aren't very many words whose category is adjective in Swahili (as compared to, e.g., English). Moreover, using more than one in a sentence is fairly degraded. We're putting aside some issues here for the sake of exposition.[/footnote] Again, we have a convention to express this in a single rule.

Swahili grammar fragment (in progress)

S_S → NP V
NP → N (A)+

The second rule can be read, "A Noun Phrase consists of a noun which is optionally followed by any number of adjectives." Note that we are working under the assumption that "fewer rules is better." That is, eventually, we want the phrase structure grammar that is the simplest, where "simplest" here means "uses the fewest amount of rules." [latex] \setcounter{ExNo}{4} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} child {node {A \\ mkubwa \\ big}} child {node {A \\ mzuri \\ good}} } child[sibling distance = 14em] {node {V \\ ulianguka \\ fall.\textsc{past}}}; \end{tikzpicture} [/latex] There's a more important message that our NP rule illustrates. Notice that whenever we put two things together---whether we're making a compound or making a phrase---the result always has the category of one of the pieces we've put together. When we put together an adjective and a noun, we name the resulting constituent a Noun Phrase---not a Verb Phrase, not a Prepositional Phrase, etc. This is because the resulting constituent is still "noun-y," not "verb-y" or "preposition-y." As a rule, a phrase will always be named after one of the [pb_glossary id="362"]heads[/pb_glossary] that it contains. Thus, every phrase, by definition has a head. That is to say, a Noun Phrase contains a noun, a Verb Phrase contains a verb, an Adjective Phrase contains an adjective, etc. And vice versa! We will assume that the opposite is true as well. Every noun is in a Noun Phrase, every verb is in a Verb Phrase, and every adjective is in an Adjective Phrase. We will refer to these two properties of heads as the Headedness Principle.

The Headedness Principle Every phrase has exactly one head and every head is in its phrase.

Though we will see more motivation for this later, for now it suffices to recognize that adopting the Headedness Principle greatly reduces the number of rules we need. If we simply assume that all nouns, verbs, adjectives are in phrases, then we don't need to list two rules for each category. However, once we adopt the Headedness Principle, then we have to adjust our rules. It is no longer possible to have a rule like NP → N (A)+ because there is a head, A, that is not in its phrase. Same for V inside S_S. To fix this issue, we need a few more rules, specifically, we need at least one rule for every head.

Swahili grammar fragment (in progress)

S_S → NP VP
VP → V
NP → N (AP)+
AP → A

If you're paying close attention, you have noticed exactly one exception to the Headedness Principle: S_S does have a head! We'll return to this issue in the next chapter. For now we'll assume that the top of the tree (sometimes call the root) is "special."

With the Headedness Principle and these new rules, we can re-draw the tree from (5) as in (6). [latex] \setcounter{ExNo}{5} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [sibling distance = 10em] child {node {NP} [sibling distance = 6em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} } child {node {AP} child {node {A \\ mzuri \\ good}} } } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex] One immediate benefit of adding in these extra layers is that now we have a place to put adverbs---which also need to be in a phrase! The example in (7a) demonstrates where adverbs go with respect to adjectives, and the example in (7b) demonstrates through coordination that adjectives and adverbs form a constituent [latex] \setcounter{ExNo}{6} \ex. \ag. mti mkubwa sana ulianguka\\ tree big very fall.\textsc{past}\\ \trans `A very big tree fell.' \bg. mti mkubwa sana na mweusi sana ulianguka\\ tree big very and black very fall.\textsc{past}\\ \trans `A very big and very black tree fell.' [/latex]

Swahili grammar fragment (in progress)

S_S → NP VP
VP → V
NP → N (AP)+
AP → A (AdvP)
AdvP → Adv

[latex] \setcounter{ExNo}{7} \ex. \ag. mti mkubwa sana ulianguka\\ tree big very fall.\textsc{past}\\ \trans `A very big tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [sibling distance = 10em] child {node {NP} [sibling distance = 6em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} child {node {AdvP} child {node {Adv \\ sana \\ very}} } } } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]

Let's keep working on Noun Phrases. Here is some more data.

[latex] \setcounter{ExNo}{8} \exg. mti kwenye kilima ulianguka\\ tree on hill fall.\textsc{past}\\ \trans `A tree on a hill fell.' %\bg. mti karibu-na mto ulianguka\\ %tree near river fall.\textsc{past}\\ %\trans `A tree near a river fell.' [/latex] From these sentences we learn something about [pb_glossary id="839"]adpositional[/pb_glossary] phrases. Swahili appears to have the category preposition; the adposition comes before the noun. (It is a preposition as opposed to a postposition; postpositions come after nouns.) We should add a rule like PP → P NP, but where do we put it? Notice that we have more than one thing that can come after the noun, a prepositional phrase and an adjectival phrase. Which order should they go in? In order to answer this question, we need to look at examples that includes both an adjective and a prepositional phrase. (This is the first case in which we're actually using data to clarify our model. Previously, we had just been taking data and modeling. But now we're specifically seeking our particular data to determine what our model should look like.) [latex] \setcounter{ExNo}{9} \ex. \ag. mti mkubwa kwenye kilima ulianguka\\ tree big on hill fall.\textsc{past}\\ \trans `A big tree on a hill fell.' [/latex] From this example, we conclude that adjectives precede prepositional phrases.

Swahili grammar fragment (in progress)

S_S → NP VP
VP → V
NP → N (AP)+ (PP)
AP → A (AdvP)
AdvP → Adv
PP → P NP

So now we can draw the following tree using our phrase structure grammar for Swahili. [latex] \setcounter{ExNo}{10} \ex. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [level distance = 5em, sibling distance = 14em] child {node {NP} [level distance = 3em, sibling distance = 8em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} } child {node {PP} child {node {P \\ kwenye \\ on}} child {node {NP} child {node {N \\ kilima \\ hill}} } } } child {node {VP} [level distance = 3em, sibling distance = 6em] child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]

Building a Verb phrase rule

We could keep going with Noun Phrases by looking at yet more data, but let's turn to the Verb Phrase. This is often the most complicated phrase to describe, because the verb is, in a sense, the "glue" that holds a sentence together. As always, we start by looking at data.

[latex] \setcounter{ExNo}{11} \ex. \ag. Juma aliona mti\\ Juma see.\textsc{past} tree\\ \trans `Juma saw a tree.' \bg. Juma aliona mti na aliona kilima\\ Juma see.\textsc{past} tree and see.\textsc{past} hill\\ \trans `Juma saw a tree and saw a hill.' [/latex] Again, these two sentences provide an enormous amount of information. We'll concentrate on two things. First, just like in English, the [pb_glossary id="770"]object[/pb_glossary] of the verb comes after the verb. (Such languages are called verb-medial languages; we'll return to cross-linguistic variation in chapter 6.) The second point is that the verb and object form a constituent to the exclusion of the subject. This fact is demonstrated through the coordination test in (12b). We can describe both of these properties by making the minimal adjustment to our VP rule: we add in an optional NP!

Swahili grammar fragment (in progress)

S_S → NP VP
VP → V (NP)
NP → N (AP)+ (PP)
AP → A (AdvP)
AdvP → Adv
PP → P NP

Now our grammar can handle both [pb_glossary id="769"]transitive[/pb_glossary] and [pb_glossary id="755"]intransitive[/pb_glossary] verbs.

[latex] \setcounter{ExNo}{12} \begin{minipage}{.49\textwidth} \ex. \ag. Juma alicheka\\ Juma laugh.\textsc{past}\\ \trans `Juma laughed.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ Juma}} } child {node {VP} child {node {V \\ alicheka \\ laugh.\textsc{past}}} }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \ex. \ag. Juma aliona mti\\ Juma see.\textsc{past} tree\\ \trans `Juma saw a tree.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ Juma}} } child {node {VP} child {node {V \\ aliona \\ see.\textsc{past}}} child {node {NP} child {node {N \\ mti \\ tree}} } }; \end{tikzpicture} \end{minipage} [/latex] Our grammar fragment can continue indefinitely until we have developed a full descriptive grammar of Swahili. For instance, we can develop rules for adverbs in VP (15a), for [pb_glossary id="802"]double-object constructions[/pb_glossary] (15b) (see chapter 10) and for questions (15c). [latex] \setcounter{ExNo}{14} \ex. \ag. Juma alicheka sana\\ Juma laugh.\textsc{past} very\\ \trans `Juma laughed a lot.' \bg. Juma alimnunulia Maria kitabu\\ Juma buy.\textsc{appl}.\textsc{past} Maria book\\ \trans `Juma bought Maria a book.' \cg. Je Juma alicheka?\\ \textsc{ques} Juma laugh.\textsc{past}\\ \trans `Did Juma laugh? [/latex] We won't bother going too in-depth in this endeavor, because in the next chapter we'll refine our theory slightly. But it is important at this point to re-emphasize what exactly we're trying to do when we build a phrase structure grammar. We're attempting to build an accurate description of a language. The way that we're doing this is by applying constituency tests. Note that we're also (implicitly) applying tests to determine category as well. I stipulated above that words like ulianguka and alicheka were verbs. This is not because when I translate them in English, the English word is a verb. It is because, in Swahili, the words have properties of being verbs. In particular, they inflect for person and number marking in their prefixes (in the examples above, the first vowel of the verb is always the subject agreement), and they also inflect for tense marking. (Past tense in Swahili is expressed by the prefix li-.) The words are also distributionally verbs: they occur in between subjects and objects. This basic approach, using constituency tests and identifying categories will not change as we move forward, even as we update our theoretical model. Indeed, these two things are the very heart of all syntactic research. We use empirical evidence to motivate a theoretical model. Our model will eventually be more sophisticated, but only as a response to the data that we look at.

[embed]https://youtu.be/Gjtv4zBU0bU[/embed] Tree drawing with phrase structure grammars (length: 1m 41s)

Where we're going

On the approach taken in this chapter, we assume that all languages have a distinct phrase structure grammar. That is, for any one language, the rules are idiosyncratic to that language. We just look at the data and determine what those rules are. And if language X and language Y in fact share the exact same rules, then they're the same language! (Like if x=2 and y=2 then x=y.) As we'll point out in the next chapter, though, the idea that every language is different is actually somewhat misleading. Languages turn out to be much more similar than they are different. For instance, it is robustly attested across languages that the object and verb form a constituent to the exclusion of the subject. It is also robustly attested across languages that if adjectives go after nouns in language X, then the object goes after the verb in language X as well. On the approach developed in this chapter, these similarities and patterns go unexplained. That is, given our current model, it is essentially entirely arbitrary how a language can put its pieces together, or better, how a language organizes its information. But when we truly look at a lot of cross-linguistic data, we find that this doesn't seem right. Ideally, our theoretical model isn't just adequate for, e.g., Swahili, but for all languages. This is what we're going to work towards in the next chapter. We'll look at more data, and attempt to build a theory of syntax that applies to any human language.

Things to remember

the Headedness Principle
the coordination rule
how to use constituency tests to infer structure
how to draw trees using phrase structure grammars

Advanced

Complexity in Phrase Structure Grammars

How complex can a phrase-structure grammar get? The answer is: very. Consider the following rules, taken from one of the standard syntax textbooks.[footnote]Carnie, Andrew. 2006. Syntax: A Generative Introduction, Second edition. Malden, MA: Blackwell.[/footnote]

CP → (C) TP
TP → {NP/CP} (T) VP
VP → (AdvP+) V (NP) ({NP/CP}) (AdvP+) (PP+) (AdvP+)
NP → (D) (AdjP+) N (PP+) (CP)
PP → P (NP)
AdjP → (AdvP) Adj
XP → XP conj XP
X → X conj X

One notational addition in these rules are the braces: "{.../...}". These are used to mean "either/or (but not both at once)." So the rule for TP, either an NP (what we have called DP in this textbook) or a CP can preceded T, i.e., be a subject. This accounts for sentences like The noise surprised Malika and That Jessie came home late surprised Malika. The latter sentence has a CP subject: that Jessie came home late is the subject of surprised. Notational differences aside, it is important to note in these rules that basically anything goes---as long as there is empirical motivation. So the reason that the VP is so long is because VP can have a lot of stuff in them. This is empirically necessary. But note that there really is not upper limit how how complex a rule can be. This is not a good outcome. Because it predicts that Human Languages can be random. That is, it reduces every Human Language to an unpredictable system of rules describing each phrase. This turns out to be incorrect. As we'll discuss in the next chapter, Human Languages are actually extremely systematic, meaning that, all things considered, the rules that describe a language tend to look very similar to each other. There is very little "randomness" in any given language. But the complex Phrase Structure Grammars serve a purpose. First, empirically, they show that phrases are complex creatures, often consisting of multiple distinct pieces. The Phrase Structure Grammars are complex because Human Language is complex (which is different from saying that Human Language is random/unpredictable). Second, the rules above serve a theoretical purpose.

Rule or transformation?

Sometimes, it's not clear when looking at data whether we should treat the given phrase as the result of a Phrase Structure Rule, or the result of a transformation. The term "transformation" is used to describe any changes that apply to a sentence. For instance, we typically describe topicalization as being derived via transformation. Recall that with topicalization, some constituent is moved to the beginning of the sentence. [latex] \setcounter{ExNo}{15} \ex. \a. Juma hugged the fruit seller. \b. The fruit seller, Juma hugged. [/latex] Should we make a Phrase Structure Rule to describe (16b)? The answer is usually taken to be "no." Why not? Because we can describe this kind of a sentence via a "rule of transformation," that is, a rule that describes a generalized process for turning one kind of sentence into another kind of sentence. In the case of topicalization, the rule is simple: place whatever you want to topicalize at the beginning of the sentence. Rules of transformation serve a dual purpose. On the one hand, they underscore the overall goal of reducing Human Language to a set of abstract rules that can apply to an infinite number of sentences. With a rule that describes topicalization, I can topicalize anything I want. On the other hand, rules of transformation associate particular word orders with particular meanings. This is crucial. The order of DP DP V in English is uniquely associated with the meaning of topicalization. This differentiates it then from the more basic word orders that our Phrase Structure Rules generally describe. For instance, the order V DP (a basic rule for VPs in English) doesn't "mean" a particular thing; this rule applies generally to all transitive constructions: build the house; hold a meeting; sing a tune; drop a single; ... . This is the "basic" order, and it doesn't have an inherent "meaning" associated with that order. Topicalization is different, though, because there is a meaning associated with that order. So we want a rule that applies just in that case. It is sometimes hard to know for any given word order whether it should be derived by a transformation, or whether it should be part of the basic word orderof the language. For instance, consider adjective orders in English. While adjectives are always allowed in front of a noun, sometimes adjectives can appear after a noun.[footnote]Cinque, Guglielmo. 2010. The Syntax of Adjectives: A Comparative Study. Cambridge, MA: MIT Press. [/footnote] [latex] \setcounter{ExNo}{16} \ex. \a. The only visible stars are Aldebaran and Sirius. \b. The only stars visible are Aldebaran and Sirius. [/latex] Should our DP rule include both orders of N and AP? This idea is debated, but one hint that maybe a rule of transformation can derive one of the order comes from meaning. Remember that if a particular order of words is associated with a particular meaning, then it might be better to try to derive that order via a rule of transformation. In (17), the two orders actually do have distinct meanings: the visible stars means something slightly different than the stars visible. To use the latter order (the stars visible), you have to talking about about stars that you can see right now. For instance, on cloudy night, when no one can see any stars, you can still talk about which stars are generally visible with (17a). This doesn't quite work for (17b). If no stars are visible, it sounds funny to talk about the stars visible. This difference in meaning suggests that more than a mere change in word order is going on. Rather, it suggests that there is a rule that transforms the order from AP N to N AP to map to a particular meaning. The broader takeaway is the following: As we continue moving forward in describing language, we should be paying attention to not only the patterns of word order (and category and constituency), but also whether there are patterns that associate particular words orders with particular meanings. If we find such patterns, then we can create abstractions over those patters: rules! Ultimately, we won't adopt any rules of transformation---but we will have movement, which accomplished the same thing.

X-bar syntax

juliedoner — Wed, 15 Nov 2023 05:32:39 +0000

Learning Objectives

By the end of this chapter, you should be able to,

identify intermediate levels ("bar levels") using constituency tests
understand the motivation for the X-bar schema
draw a tree using the X-bar schema,
identify the distinct positions within the X-bar schema,
diagnose the difference between adjuncts and complements

By the end of this chapter, we will have technically changed theories. We're going from Phrase Structure Grammars to X-bar Syntax. But the trick is to recognize that, even though we're going to be using the term "X-bar syntax," we aren't actually changing theories at all. The same principles, formalizations, and theoretical tools will be used in X-bar syntax that we've already been using. Our trees are just going to look more complicated. The reason that our trees will look more complicated is that we're going to discover in this chapter that there are lot of "hidden" layers. Just like when we discussed functional categories we learned to infer the presence of structure, even when we can't directly see it, we'll use the same tools to infer the presence of "bar-levels." From this, we'll make a theoretical leap. As a hypothesis about syntax, we'll adopt a uniform structural representation for all phrases. This is "X-bar syntax." It's the hypothesis that all phrases share a structure, i.e., every phrase looks identical. The hypothesis makes a number of testable predictions, which we'll explore in terms of the complement/adjunct distinction.

Deconstructing VP

Let's start by looking again at English VPs. Assume the (very simplistic) rule for VPs below.

Simplified VP rule for English (to be revised)

VP → V (DP) (PP)+

This rule allows us to minimally distinguish between [pb_glossary id="769"]transitive[/pb_glossary] and [pb_glossary id="755"]intransitive[/pb_glossary] verbs, each of which could have a modifying prepositional phrase. For instance, using this VP rule, here's the structure for The doctor will sing a song in the park. [latex] \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 6em] child {node {DP} [sibling distance = 5em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T \\ will}} child {node {VP} [sibling distance = 5em] child {node {V \\ sing}} child {node {DP} [sibling distance = 3em] child {node {D \\ a}} child {node {NP} child {node {N \\ song}} } } child {node {PP} [sibling distance = 3em] child {node {P \\ in}} child {node {DP} child {node {D \\ the}} child {node {NP} child {node {N \\ park}} } } } } }; \end{tikzpicture} [/latex] This tree correctly represents that in the park is a constituent. It also correctly represents that sing a song in the park is a constituent. However, it still is getting the constituency wrong. Look at the substitution test below. What does it show is a constituent? [latex] \setcounter{ExNo}{1} \ex. \a. The doctor will sing a song in the park. \b. and Mary will do so in the office. [/latex] What's the problem here? The test demonstrates that sing a song is a constituent, because I am able to replace that string with do so. The tree in (1) gets this wrong. The string sing a song isn't a constituent according to this tree because there is no node that contains the words sing a song and nothing else. Is our constituency test just giving us a false positive? Let's confirm. Below I'm applying the other constituency tests. [latex] \setcounter{ExNo}{2} \ex. Fragments\\ What will the doctor do in the park? {\underline{Sing a song}} \ex. \a. \textit{It}-Clefting\\ {}*It was {\underline{sing a song}} that the doctor will do in the park. \b. Pseudo-clefting\\ {\underline{Sing a song}} is what the doctor will do in the park. \c. \textit{All}-clefting\\ {\underline{Sing a song}} is all the doctor will do in the park. \d. Topicalization\\ {}*{\underline{Sing a song}}, the doctor will do in the park. \ex. Coordination\\ The doctor will [ sing a song ] and [ dance ] in the park [/latex]

So our constituency tests are telling us that sing a song is a constituent---and that it's a VP. Recall that substitution and coordination allow us to diagnose the category of the constituent. But here's the thing: the string sing a song in the park is also a VP. How do I know? It passes all the same constituency tests, and we can detect its category through substitution and coordination.

[latex] \setcounter{ExNo}{5} \ex. Substitution\\ The doctor will {\underline{sing a song in the park}}, and the nurse will {\underline{do so}}, too \ex. Fragments\\ What will the doctor do? {\underline{Sing a song in the park}} \ex. \a. \textit{It}-Clefting\\ {}*It was {\underline{sing a song in the park}} that the doctor will do. \b. Pseudo-clefting\\ {\underline{Sing a song in the park}} is what the doctor will do. \c. \textit{All}-clefting\\ {\underline{Sing a song in the park}} is all the doctor will do. \d. Topicalization\\ {}*{\underline{Sing a song in the park}}, the doctor will do. \ex. Coordination\\ The doctor will [ sing a song in the park ] and [ dance at home ] [/latex] So based on this evidence, it looks like our tree should actually have the following representation. [latex] \setcounter{ExNo}{9} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance=6em] child {node {DP} [sibling distance = 5em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T \\ will}} child {node {VP} child {node {VP} [sibling distance = 5em] child {node {V \\ sing}} child {node {DP} [sibling distance = 3em] child {node {D \\ a}} child {node {NP} child {node {N \\ song}} } } } child {node {PP} [sibling distance = 3em] child {node {P \\ in}} child {node {DP} child {node {D \\ the}} child {node {NP} child {node {N \\ park}} } } } } }; \end{tikzpicture} [/latex] This version of the tree correctly captures our empirical result: we have two distinct VP constituents. (It also correctly predicts that a song in the park is not a constituent, which you can confirm on your own.) The "trick" is that we've added a VP in-between the head and the highest VP. This projection is empirically motivated, though. It's there because our tests tell us it's there. (Note that technically, the highest VP violates the Headedness Principle.) We call this intermediate projection the "bar" level, and write it either with a line over the top (V) or an apostrophe after (V'). It's pronounced "V bar" (not "V prime"). This is just a notation for saying "I'm not the highest phrase in this phrase." To review: we've used our constituency tests to determine that there is a "hidden" level in verb phrases.

Let's look at more data.

[latex] \setcounter{ExNo}{10} \ex. \a. John will sing a song on a bike in the park and Mary will do so at the office. \b. John will sing a song on a bike and dance a jig at the park at different times. [/latex] These data are just like the data above, except that I've added in more prepositional phrases. So now sing a song on a bike can't be the highest VP in the structure. The highest VP is the entire constituent sing a song on a bike in the park. This is the topmost VP. Now in fact, we need two intermediate projections, because the constituents sing a song and sing a song on a bike and sing a song on a bike in the park are all VPs. [latex] \setcounter{ExNo}{11} \ex. [ John will [$_{VP}$ [$_{V'}$ [$_{V'}$ sing a song ] on a bike ] in the park ] ] [/latex] These data actually suggest that I can simplify my representation quite a bit, because they suggest that I can make do without the "+" symbol. All I need are three rules like the following. Note that three rules is actually simpler than one rule with a plus sign because now we can get by with less formalism. By analogy, 2 + 2 + 2 is simpler than 2 x 3 because it provides the same information with a less complex operator.

Simplified VP rule for English (revised)

VP → V'
V' → V' PP
V' → V (DP)

I've done a very tricky (but ultimately useful) thing here. I've explicitly added [pb_glossary id="760"]recursion[/pb_glossary] into my English phrase structure grammar. There's a rule that says V' (i.e., an intermediate projection between the head and the topmost VP) can consist of itself and a prepositional phrase. Why does this work? Consider the tree for John will sing a song on a bike in the park which uses the new rule above. [latex] \setcounter{ExNo}{12} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C}} child {node {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ John}} } } child {node {T \\ will}} child {node {VP} child {node {V$'$} child {node {V'} child {node {V'} child {node {V \\ sing}} child {node {DP \\ a song}} } child {node {PP \\ on a bike}} } child {node {PP \\ in the park}} } } }; \end{tikzpicture} [/latex] When I build a VP now, I have an option: I can either add in an infinite number of prepositional phrases by adding in a new V' level, or I can just go directly to the head V. I've essentially created a way to make an infinitely recursive loop---with an escape hatch. Most importantly, all of these intermediate projections correlate with distinct, identifiable constituents. That is, our structure correctly represents how English packages its information.

Deconstructing TP

We can do the exact same thing with TP. Let's assume our rule for TP developed in the last chapter.

TP rule for English (to be revised)

TP → DP T VP

Now consider the following data, which suggests that there is a bar-level---an intermediate projection---in TP. [latex] \setcounter{ExNo}{13} \ex. \a. John will eat sushi and will drink sake. \b. Mary has forgiven Jed and might forgive Brad. \c. Conrad can sing and could dance. [/latex] What is this data showing? It shows that I can coordinate will eat sushi with something else. That is, there appears to be a constituent that include T and VP, but does not include the subject. This is not possible, according to our rule above. To fix this issue, we can re-write our rule, parallel to what we did with VP.

TP rule for English (revised)

TP → DP T'
T' → T VP

Adding in this extra level correctly captures what our constituency test is telling us. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 7em] child {node {DP} [sibling distance = 5em] child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ John}} } } child {node {T$'$} [sibling distance = 4em] child {node {T \\ will}} child {node {VP} [sibling distance = 2em] child {coordinate(left1)} child {node {eat sushi}edge from parent[draw=none]} child {coordinate(right1)} } } }; \draw (left1)--(right1); \end{tikzpicture} [/latex] This tree now correctly captures the fact that T forms a constituent with VP, to the exclusion of the subject. That constituent is the bar level, T'. What if I add in more words. Here I'm adding a modifier of TP. [latex] \setcounter{ExNo}{14} \ex. \a. John definitely will eat sushi and probably will drink sake. \b. John definitely will eat sushi and drink sake. [/latex] This data now suggests that the modifier gets grouped together with T and the VP in the first example (because both constituents have a similar modifier), but in the second example shows us that T and the VP can form a constituent that is separate from this modifier (because definitely is describing both the eating and drinking events). How are we going to make this work? If Adverbs[footnote]Note that the term "adverb" is a bit misleading: this aren't "adding" to the "verb".[/footnote] like definitely, probably, etc may sometimes be distinct from the VP, and sometimes not, we need to adjust our rules about where adverbs go in TP. Just like with VP, we can fix this issue concisely by adding a recursive T' level!

TP rule for English (revised)

TP → DP T'
T' → AdvP T'
T' → T VP

This set of rules allows us to add as many modifiers of TP as we want, and we will always find that each additional modifier "creates" a subconstituent within TP. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node {DP} [sibling distance = 3em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T$'$} [sibling distance = 6em] child {node {AdvP} child {node {Adv \\ definitely}} } child {node {T$'$} [sibling distance = 4em] child {node {T \\ will}} child {node {VP} [sibling distance = 2em] child {coordinate(left1)} child {node {eat sushi}edge from parent[draw=none]} child {coordinate(right1)} } } } }; \draw (left1)--(right1); \end{tikzpicture} [/latex]

Deconstructing NP

We'll do the exact same thing with NP. Keep in mind that we've adopted the [pb_glossary id="827"]DP-hypothesis[/pb_glossary], so all NPs are inside of DPs.

Simplified NP rule for English (to be revised)

NP → (AP)+ N

With this rule in mind, consider the patterns of one substitution shown below. [latex] \setcounter{ExNo}{15} \ex. \a. Tricia read the short book, not the long one. \b. Tricia read the short blue book, not the long one. [/latex] We observed in the last chapter that one substituted for NP, but these data suggest that that isn't quite accurate. One must be substituting for something between NP and N. Consider (16b). One is replacing the constituent blue book. This includes an adjective, so it's not just the head N, but it excludes the higher adjective short/long, so this cannot be the entire NP. We therefore find evidence in NP for an intermediate projection, and we re-write our rules accordingly. Now we can add in an infinite number of prenominal adjectives by adding in more intermediate projections.

Simplified NP rule for English (revised)

NP → N'
N' → AdjP N'
N' → N

These rules correspond to the following tree. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ the}} child {node {NP} child {node {N'} child {node {AdjP} child {node {Adj \\ short}} } child {node {N$'$} child {node {AdjP} child {node {Adj \\ blue}} } child {node {N$'$} child {node {N \\ book}} } } } }; \end{tikzpicture} [/latex]

Generalizing across categories

When we look across all of our rules, we can make a few generalizations:

There's always a rule of the form XP → X', where the phrase level goes to the intermediate level. (Sometimes this rule has an element in front of X'.)
There's always a rule of the form X' → X, where the bar-level consists of the head. (Sometimes the head is followed by something on the right.)
There's always a recursive rule of the form X' →YP X' or X' → X' YP, where YP is a modifier of some sort.

If this pattern exists for every single category, it suggests that we can generalize our schema for phrase structure. We hypothesize that every phrase has the following rules. The "X, Y, Z, W" below are variables meaning "any category."

Generalized phrase structures rules (final version)

XP → (YP) X'
X' → X' ZP
X' → ZP X'
X' → X (WP)

If this is right, it means that all phrases are constructed based on these rules. In the rules above X, Y, and W are variables over categories.

What about the Headedness Principle? Have we entirely abandoned it? No! The Headedness Principle, which says that every phrase has a head and every head is in its phrase is exactly the same. But we've redefined what "phrase" means. Now a phrase minimally involves at least one X'-level. A phrase consists of all of the rules above: it is the X-bar schema. The term "phrase" is now a structural notion. The X-bar schema is a hypothesis. By looking at a lot of data, we identify that there are intermediate projections in every phrase that we look at. The subsequent phrase structure rules all end up looking very similar---suspiciously similar in fact. The X-bar schema is a way to make sense of this information. It proposes that the similarities reflect a "deep" uniformity across syntax. Our trees end up looking more complex, because now everything needs at least one intermediate projection. But the complexity of the representations comes with an overall simplification of the theory: it isn't necessary to learn a bunch of arbitrary rules for syntax. You only need the four above, because they apply to every phrasal category. [latex] \setcounter{ExNo}{16} \ex. The tired doctor cured the sick patient yesterday. Old way: \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 8em] child {node {DP} [sibling distance = 4em] child {node {D \\ the}} child {node {NP} child {node {AP} child {node {A \\ tired}} } child {node {N \\ doctor}} } } child {node {T \\ $[$past$]$}} child {node {VP} [sibling distance = 6em] child {node {V \\ cured}} child {node {DP} [sibling distance = 4em] child {node {D \\ the}} child {node {NP} child {node {AP} child {node {A \\ sick}} } child {node {N \\ patient}} } } child {node {AdvP} child {node {Adv \\ yesterday}} } } }; \end{tikzpicture} [/latex] [latex] New way: \begin{tikzpicture}[mytree, scale=.9] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance=8em] child {node {DP} [sibling distance=3em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {AP} child {node {A$'$} child {node {A \\ tired}} } } child {node {N$'$} child {node {N \\ doctor}} } } } } } child {node {T$'$} [sibling distance=6em] child {node {T \\ $[$past$]$}} child {node {VP} [sibling distance=4em] child {node {V$'$} [sibling distance = 6em] child {node {V$'$} [sibling distance = 4em] child {node {V \\ cured}} child {node {DP} [sibling distance = 3em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {AP} child {node {A$'$} child {node {A \\ sick}} } } child {node {N$'$} child {node {N \\ patient}} } } } } } } child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ yesterday}} } } } } } } }; \end{tikzpicture} [/latex]

[embed]https://youtu.be/BrTh1zqifdA[/embed] Trees in the X' schema (length: 1m 28s)

Note we also still need special rules for coordination, which look basically the same as before. Rules for coordination (final version)

XP → XP conj XP
X' → X' conj X'
X → X conj X

These rules allow you to coordinate heads (X), bar-levels (X') and phrases (XP), but you can't mix-and-match. You cannot coordinate a bar level with a head, for instance.

Positions in X'-syntax

We define four positions within the X'-schema: specifier, head, complement and adjunct. The positions are indicated in the trees below. [latex] \begin{tikzpicture}[mytree] \node(top) {XP} child {node(spec1) {YP}} child {node {X$'$} child {node {X$'$} child {node(head1) {X}} child {node(comp1) {WP}} } child {node(adj1) {ZP}} }; \node(head) at (-4,-5) {``head"}; \node(spec) at (-5,0) {``specifier"}; \node(comp) at (5,-5) {``complement"}; \node(adj) at (5,-2) {``adjunct"}; \draw[->, dashed] (head)--(head1); \draw[->, dashed] (spec)--(spec1); \draw[->, dashed] (comp)--(comp1); \draw[->, dashed] (adj)--(adj1); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {XP} child {node(spec1) {YP}} child {node {X$'$} child {node(adj1) {ZP}} child {node {X$'$} child {node(head1) {X}} child {node(comp1) {WP}} } }; \node(head) at (-4,-5) {``head"}; \node(spec) at (-5,0) {``specifier"}; \node(comp) at (5,-5) {``complement"}; \node(adj) at (-5,-2) {``adjunct"}; \draw[->, dashed] (head)--(head1); \draw[->, dashed] (spec)--(spec1); \draw[->, dashed] (comp)--(comp1); \draw[->, dashed] (adj)--(adj1); \end{tikzpicture} [/latex]

A head is the only non-phrasal [pb_glossary id="811"]terminal node[/pb_glossary] in a phrase.
A specifier is defined as, "The phrase that is [pb_glossary id="812"]daughter[/pb_glossary] of XP and [pb_glossary id="813"]sister[/pb_glossary] to X." Every phrase that meets this description is a specifier.
A complement is defined as, "The phrase that is sister to X and daughter to X." Every phrase that meets this description is a complement.[footnote]FYI Complement is spelled with an 'e'. It is not compliment.[/footnote]
An adjunct is defined as, "The phrase that is daughter to X and sister to X." Every phrase that meets this description is an adjunct.

Note that the definitions above do not make reference to linear order. It doesn't matter whether an adjunct is to the left or the right. All that matters is that its sister and mother are bar-levels. Likewise, complements can be on the left or the right of a head, as long as it is sister to the head. And a specifier can be on the left right as well, as long its mother is a phrase. Every language makes particular choices however about where specifiers and complements appear. We'll return to this momentarily. For now, we note that in English, specifiers are always on the left, and complements are always on the right.

It is conventional to use the "genealogical" terms mother, sister, and daughter when referring to the relationships between elements in a tree. The terms are meant to be transparent: A "mother" is something that has "daughters." If two things have the same mother, then they're "sisters." In the tree below, DP is the mother of D'. D' is the mother of D and NP. D and NP are sisters, and are both daughters of D'. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP}} }; \end{tikzpicture} [/latex] The analogy basically stops there. There is no such thing as an "aunt" in syntax, or a "niece."

We've seen a lot of complements. TP is a complement to CP. NP is a complement to D. DP can be a complement to V when it's an object of the verb. The only specifier we've seen is the subject. It's the specifier of TP. There is another place where a specifier is necessary: DP. Possessive structure. The other place where we observe specifiers is in possessive DPs. [latex] \setcounter{ExNo}{17} \ex. the mailman's bag \begin{tikzpicture}[mytree] \node(top) {DP} [sibling distance =7em] child {node {DP} [sibling distance =4em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ mailman}} } } } } child {node {D$'$} [sibling distance =4em] child {node {D \\ 's}} child {node {NP} child {node {N$'$} child {node {N \\ bag}} } } }; \end{tikzpicture} [/latex] We treat the possessive marker -s as a realization of D. This is because possessives are in complementary distribution with determiners: *the mailman's the bag. And because the entire DP acts as a constituent referring to a bag, not a mailman, the head of the highest DP must be the possessive D.

[embed]https://youtu.be/XQmS-ue8oko[/embed] Possessive DPs (length: 1m 51s)

Complements vs. Adjuncts

The X-bar schema illustrated above makes a prediction that some phrases that are next to a head are complements while some are adjuncts. They may look identical on the surface, but they derive from different structures.

For instance, consider the phrase the state of Kansas. In the X-bar schema, this constituent (and it definitely is a constituent) could hypothetically correspond to either tree below. The prepositional phrase of Kansas could either be a complement to state (as in the tree on the left) or an adjunct to state (as in the tree on the right).

[latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \draw (left)--(right); \hspace{200pt} \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N \\ state}} } child {node {PP} [sibling distance=2.5em] child {coordinate(left1)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right2)} } } } }; \draw (left1)--(right2); \end{tikzpicture} [/latex] The two trees actually make distinct empirical predictions. We have three tests to distinguish between complements and adjuncts. First, I just want to briefly mention that it is easy to confuse adjuncts and specifiers. When drawing your trees, remember that for every adjunct, you should have "plus one" bar-level, because adjuncts need both a bar-level which is a sister, and bar-level which is a mother. So you have to add a layer whenever you add an adjunct.

[embed]https://youtu.be/nOe_4igGwfA[/embed] Adjuncts in X'-syntax (length: 2m 15s)

Substitution

The constituency test of substitution has a different use. As we observed above, one substitution and do so substitution both target the bar-level. Since complements are under all bar-levels in a phrase, if you try to do substitution, a complement cannot be "left behind." An adjunct can, though, because there's always a bar-level that doesn't include the adjunct.

Visually, consider the two trees again. If the PP is a complement to N, then when you substitute with one everything under N' has to go too. In contrast, if the PP is an adjunct, it's possible to substitute with one and leave behind the PP. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node(nbar) {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \node(text) at (-3,-5) {\textit{one} substitutes \\ here}; \draw (left)--(right); \draw[->, dashed] (text)--(nbar); \hspace{250pt} \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node(nbar2) {N$'$} child {node(nbar1) {N$'$} child {node {N \\ state}} } child {node {PP} [sibling distance=2.5em] child {coordinate(left1)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right2)} } } } }; \node(text2) at (-3,-5) {\textit{one} substitutes \\ either place}; \draw (left1)--(right2); \draw[->, dashed] (text2)--(nbar1); \draw[->, dashed] (text2)--(nbar2); \end{tikzpicture} [/latex] So to use the substitution test to determine if something is a complement or adjunct, you simply replace the head of the phrase, and see if you can leave behind the constituent in question. Compare the sentences below. [latex] \setcounter{ExNo}{18} \ex. \a. the state of Kansas, and\\ {}*the one of Maine \b. the state near Kansas, and\\ {} the one near Maine. [/latex] As illustrated above, one-substitution crucially yields an ungrammatical result in (19a), but not in (19b). Why is this? Again, note here that one-substitution targets bar-levels; you can only replace the material that exists under a bar-level. As such, the reason for the grammaticality distinction in the examples above is that the NPs, the state of Kansas and the state near Kansas, crucially differ in the number of bar-levels they contain. That is, unlike in the NP, [ _NP state of Kansas ], where one-substitution can replace both N and PP (20a), but not N alone (20b), with the NP, [ _NP state near Kansas ], one-substitution can either replace the larger constituent containing N and PP (21a), or N alone (21b), meaning that the NP in (19) must crucially contain two bar-levels, and the NP in (20), only one. [latex] \setcounter{ExNo}{19} \ex. I visited the state of Kansas. \\ \a. I visited the \textit{one}. \hfill ( targets: [ \textsubscript{N'} [\textsubscript{N} state ][ \textsubscript{PP} of Kansas ]] ) \\\b. *I visited the \textit{one} of Kansas. \hfill ( targets: [ \textsubscript{N'} [ \textsubscript{N} state ]] )\\ \vspace{3mm} \ex. I visited the state near Kansas\\ \a. I visited the \textit{one}. \hfill ( targets: [ \textsubscript{N'} [\textsubscript{N} state ][ \textsubscript{PP} near Kansas ]] )\\\b. I visited the \textit{one} near Kansas. \hfill ( targets: [ \textsubscript{N'} [ \textsubscript{N} state ]] )\\ [/latex] Since we have defined a complement as being the sister of the head that selects it, and the daughter of the bar-level that contains that same head (i.e. a head and its complement share a bar-level parent), while conversely defining an adjunct as being both the sister to a bar-level and the daughter to a bar-level, the fact that one-substitution can target two different constituents in (21) means that the NP, the state near Kansas, necessarily contains two bar-levels, and, therefore, must contain an adjunct. That is, the PP, near Kansas, must be an adjunct because there are two different targets (i.e. bar-levels) for one-substitution in (20); one can replace the lower bar-level to the exclusion of PP, i.e. [ _N' [ _N state ]], or it can replace the higher bar-level that includes both the lower N' and PP, i.e. [ _N' [_N' [ _N state][ _PP near Kansas ]].

Step-by-step walkthrough of the substitution test

Identify the head in question and the phrase which you want to test
Identify the appropriate substitution word for the head.
Replace the head with the substitution word, but don't replace the phrase you're testing. (Note that technically, you're substituting at the lowest bar-level, which contains the head. The lowest bar-level could, in principle also have a complement.)
Check the grammaticality.

[embed]https://youtu.be/hZvQgFIutiw[/embed] Substitution test for complements vs. adjuncts (length: 1m 6s)

[embed]https://youtu.be/4S4euaW9NDM[/embed] Substitution test for complements vs. adjuncts (length: 1m 9s) Credit Damian White Lightning

Reordering

Complements must appear as sisters to the head, and so they have a fixed position with respect to other phrases. But adjuncts can, theoretically, be reordered. This is because all that's necessary is that they have a mother that's a bar-level and a sister that's a bar-level.

For instance, consider the following trees which each have two prepositional phrases. The tree on the left treats PP₁ as an complement and PP₂ as an adjunct. If we switched their positions, then we switch their roles. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N}} child {node {PP$_1$}} } child {node {PP$_2$}} } } }; \hspace{250pt} \node(top) {DP} child {node {D$'$} child {node {D}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N$'$} child {node {N}} } child {node {PP$_1$}} } child {node {PP$_2$}} } } }; \end{tikzpicture} [/latex] In contrast, in the tree on the right, both the PPs are adjuncts. It shouldn't matter whether we switch their positions, because they both will still be adjuncts.

What this means is that if a constituent is a complement, it cannot be reordered with another constituent. If a constituent is an adjunct, it can be.

The difference can be shown empirically in the following examples. You can't reorder of Kansas and near Maine, suggesting that of Kansas is a complement to state. In contrast, you can reorder by the ocean and near Maine, so we conclude that by the ocean is an adjunct to state. [latex] \setcounter{ExNo}{20} \ex. \a. the state of Kansas near Maine \b. *The state near Maine of Kansas \ex. \a. the state by the ocean near Maine \b. the state near Maine by the ocean [/latex]

Step-by-step walkthrough of the reordering test

Identify the head and the phrase which you want to test.
Choose an appropriate phrase that can appear with the head and phrase above.
Reorder or the two phrases with respect to the head, and check the grammaticality.

[embed]https://youtu.be/FY7XaY7VKxY[/embed] Reordering test (length: 1m 2s)

[embed]https://youtu.be/wX7EHFCTJTk[/embed] Reordering test (length: 1m 6s) Credit: Damian White Lightning

Sometimes we can't reorder phrases for semantic reasons, independent (presumably) of syntax. There happens to be a fixed order for modification that is fairly robust cross-linguistically. For instance, temporal modifiers (things that deal with "time") tend to be merged higher (or "outside") of manner modifiers (things that describe "how" the event was accomplished). Likewise, there is a fairly fixed ordering of adjectives in English (and many languages).

[latex] \setcounter{ExNo}{21} \ex. \a. Sammie cut Vaughn's hair {\underline{with scissors}} {\underline{often}} \b. ??Sammie cut Vaughn's hair {\underline{often}} {\underline{with scissors}} \ex. \a. the big brown ball \b. ??the brown big ball [/latex]

Linear ordering

As a corollary to the last test, if you can't reorder the phrases, the phrase nearest the head must be a complement. In fact, we predict that complements are always strictly adjacent to the head. Therefore, it should be impossible to insert anything between a head and complement. (Note that this test is irrelevant (N/A) if you find that you can reorder the phrase in question.)

Again, the difference is shown empirically below. [latex] \setcounter{ExNo}{22} \ex. \a. *the state that I visited of Kansas \b. the state that I visited near Maine [/latex]

Step-by-step walkthrough of linear ordering test

Identify the head and the phrase you want to know about.
Identify some other constituent that can also directly follow the head.
Put the phrase you want to know about after the new phrase, both after the head, and check the grammaticality.

[embed]https://youtu.be/hcClPu5znmQ[/embed] Linear ordering test (length: 1m 10s)

The reason that this test works is that the relationship between the head and complement is fixed: they must be sisters. But the relationship between a head and an adjunct is not fixed; there can be any number of adjuncts be between the head and the adjunct. Based on the three tests above, we find that of Kansas is a complement to state. So the following is the correct representation: [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \draw (left)--(right); \end{tikzpicture} [/latex]

Parameters

The order of the specifiers and complements is fixed for any one language. What does that mean? For instance, in English, the complement always comes after the head. That is, English is head-initial. With specifiers, the specifier always comes before the head. English is specifier-initial.

However, logically for each position, there are four ways things could be.

Specifier-initial : The specifier is to the left of its head (meaning to the left of the bar-level).
Specifier-final : The specifier is to the right of its head (meaning to the right of the bar-level).
Head-initial : The head is the left of its complement.
Head-final : The head is to the right of its complement.

[latex] \begin{minipage}{.49\textwidth} Specifier-initial: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {YP}} child {node {X$'$} child {node {X}} }; \end{tikzpicture} \end{minipage} \hspace{50pt} \begin{minipage}{.49\textwidth} Specifier-final: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {X}} } child {node {YP}}; \end{tikzpicture} \end{minipage} [/latex] [latex] \begin{minipage}{.49\textwidth} Head-initial: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {X}} child {node {YP}} }; \end{tikzpicture} \end{minipage} \hspace{50pt} \begin{minipage}{.49\textwidth} Head-final: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {YP}} child {node {X}} }; \end{tikzpicture} \end{minipage} [/latex]

It turns out that languages make different choices as to how they order their heads and specifiers (and adjuncts). This is called parametric variation. English is specifier-initial and head-initial. Thus, in "neutral" contexts, it will have the order of Subject-Verb-Object, or SVO.

Kashaya (a Pomoan language spoken in Northern California) has set its parameters to specifier-initial and head-final. So it's basic word order is "Subject-Object-Verb," or SOV.

[latex] \setcounter{ExNo}{23} Kashaya: \exg. Conrad Anita-to cadu\\ Conrad Anita-\textsc{acc} saw\\ \trans `Conrad saw Anita.' [/latex] Malagasy (an Austronesian language spoken on Madagascar) has set its parameters to specifier-final and head-initial. It's basic word order is "Verb-Object-Subject," or VOS. [latex] \setcounter{ExNo}{24} Malagasy: \exg. nahita ny mpianatra ny vehivavay\\ saw the student the woman\\ \trans `The woman saw the student.'' [/latex] Lastly, Hixkaryana (a Carib language spoken in Brazil) has set its parameters to specifier-final and head-final. So it's basic word order is "Object-Verb-Subject," or OVS. [latex] \setcounter{ExNo}{25} Hixkaryana: \exg. kana yan{\"i}mno b{\"i}yekomo\\ fish caught boy\\ \trans `The boy caught a fish.' [/latex]

If you're counting, you'll notice that I've said nothing about whether the two other logically possible words orders of Subject, Verb, and Object exists: VSO and OSV. These cannot be straightforwardly derived by simply stipulating specifier and head directions. Something more (i.e., movement) must be added to the theory.

Nonetheless, K'iche' (a Mayan language spoken in Guatemala) has a basic word order of "Verb-Subject-Object," or VSO. [latex] \setcounter{ExNo}{26} \exg. xril la ali Marij la a Xwaan\\ saw \textsc{det} \textsc{fem} Mary \textsc{det} \textsc{masc} John\\ \trans `Mary saw John.' [/latex] OSV is incredibly rare. This points to a very interesting distributional fact: while all possible orders exist, they don't all exist with the same frequency. SOV is the predominant word-order cross-linguistically, accounting for about 40% the world's languages. Next is SVO, accounting for about another 35%. The other categories account for the remaining 25%.

Interestingly, languages tend to be largely consistent in their head and specifier settings. That is, English is generally head-initial and specifier-initial. Kashaya is generally head-final and specifier-initial. Same with Malagasy and Hixkaryana.

It was proposed that the theory of syntax, coupled with a universal set of parameters can account for the entirety of the world's languages. That is, suppose that there is a finite number of "switches," and each language has a unique cumulative setting for each switch. This was (and still is) a popular idea. Even if it's not entirely correct, it's still a very handy way to think about language variation.

[embed]https://youtu.be/u2iza_JDWZY[/embed] Parametric Variation (length: 2m 39s) Credit: Sydney Pritchard

Recap and a look forward

This ends the first part of this book. We've built a theory of syntax: X-bar theory. We did so by looking at data, and thinking about how that data tells us. In the end, we decided that the X-bar schema was an efficient way to capture observed patterns in syntactic structures. It is worth reflecting again on what the benefit of the X-bar schema is. It is no longer necessary for us (as researchers) to try to figure out the "VP" rule, or the "NP" rule, because in fact, they're the same rule! It's just the X-bar tree. Even more, consider the task of language acquisition. Every baby is confronted with a huge challenge: acquiring the intricacies of language. That challenge is greatly simplified if the baby intrinsically knows that everything they're hearing conforms to X-bar syntax. The baby doesn't have to learn about all the different things that can go in VP, nor all the different things that can go in NP. For the most part, the baby doesn't have to figure out the constituency of the words they are hearing, because the constituency is essentially dictated by the X-bar schema (in most cases). This greatly simplifies what a baby needs to learn to acquire a language --- any language. All they need to figure out (besides the words) are the parameters. Is the language head-initial or head-final? Is the language specifier-initial or specifier-final? From here on out, we're going to stick with the X-bar schema; we won't make any more changes to the structure that we use. Going forward, we're going to start talking about manipulating structure. What happens when things start to move around?

Key Takeaways

Intermediate levels are found across all phrases.
The X-bar schema is empirically motivated by looking at constituency
The X-bar schema defines four positions: head, complement, specifier, adjunct.
Complements and adjuncts can be empirically distinguished through three tests.

Advanced

The X-bar schema only permits maximally binary branching trees. That is, according to the X-bar schema, a node can have at most two daughters. It can have fewer than two daughters, i.e., one or no daughters. But it cannot have three daughters. In light of this constraint, we should revisit two of our phrase structure rules which seem to require a node to have three daughters. Recall our rules for coordination:

XP --> XP conj XP
X' --> X' conj X'
X --> X conj X

It's not immediately clear how to map these rules into X'-syntax. Is there any way to demonstrate constituency that is consistent with binary branching structures? It turns out that there are a few tests that suggest an "asymmetric" constituent structure. Consider the following minimal, which involve (apparent) rightward movement of part of a coordinated element. [latex] \setcounter{ExNo}{26} \ex. \a. Patrick ate an apple yesterday, and also a banana. \b. *Patrick ate an apple and, yesterday, also a banana. [/latex] This data show that the conjunction and groups together with the second conjunct, not the first. Additional data comes from what is called Left-Branch Extraction. The basic observation is that sometimes you can move (something inside of) the leftmost part of a coordinate, but nothing in the rightmost part. An illustrative example comes from Japanese.[footnote]Bošković, Željko. "On the coordinate structure constraint, across-the-board-movement, phases, and labeling." Recent developments in phase theory (2020): 133-182.[/footnote] (The underlines indicate where the first word has moved from.) [latex] \setcounter{ExNo}{27} \ex. \ag. ringo-o Taro-wa [{\underline{\hspace{10pt}}} san ko] to [banana-o ni hon] tabeta\\ apple-\textsc{acc} Taro-\textsc{top} {} three \textsc{cl} and banana-\textsc{acc} two \textsc{cl} ate\\ \trans `Taro ate three apples and two bananas.' \bg. *banana-o Taro-wa [ringo-o san ko] to [{\underline{\hspace{10pt}}} ni hon] tabeta\\ apple-\textsc{acc} Taro-\textsc{top} apple-\textsc{acc} three \textsc{cl} and {} two \textsc{cl} ate\\ \trans `Taro ate three apples and two bananas.' [/latex] While an analysis of Left Branch Extraction is beyond the scope of this class, what we probably should conclude is that in coordination, the head of the coordination, i.e., and forms a constituent with the two element, to the exclusion of the first. It's possible therefore to translate our phrase structure rules for coordination above into the following X' compliant trees. [latex] \begin{tikzpicture}[mytree] \node(top){ConjP} child {node {XP}} child {node {Conj$'$} child {node {conj \\ \textit{and}}} child {node {XP}} }; \end{tikzpicture} [/latex] This tree treats the first coordinate as a specifier, and the second as a complement. And crucially, it groups the head of ConjP together with the second coordinate.

Head-movement

juliedoner — Wed, 15 Nov 2023 05:32:40 +0000

Learning Objectives

By the end of this chapter, you should be able to,

diagnose movement of T and V using adverb placement,
represent movement of heads in trees
recognize and represent do-support in English trees

Morphological Pieces in syntax

Before turning to the syntactic process of head-movement, we have to lay some groundwork. Consider a word like horses. English speakers will recognize that this word is made up of two pieces: the noun horse and the [pb_glossary id="798"]plural[/pb_glossary] suffix -s. We call each piece of a word a [pb_glossary id="799"]morpheme[/pb_glossary]. Morphemes are the smallest units of sound-meaning correspondence. The morpheme horse cannot be divided up into any smaller unit. Hor doesn't mean anything; orse doesn't mean anything. Similarly, there's no way to further sub-divide the suffix -s into anything smaller. [pb_glossary id="750"]Morphology[/pb_glossary] is the study of word-formation. What are the (perhaps language-specific) rules for putting morphemes together to form words? How do languages "build words?" There is a long debate about "where" word-building happens. Many scholars think that there is an independent, autonomous part of the grammar that governs how words are built. Once you've built a word, you can put it into your syntactic representation. On the other hand, many scholars believe that word building is just another component of syntax. You don't build words before you put them into the structure, you build words in the tree. Why would people think this? Well, one pretty strong piece of evidence that syntax and morphology are intrinsically connected is that morphology often reflects syntactic properties. For instance, as we will discuss at length below, Tense information is very often found bundled together with the verb; in English, the past tense walked contains two morphemes, walk and -ed. We know though that English also has a Tense Phrase. In the sentence Tamisha walked to the store, the head T will bear the feature [past] but that information is realized (i.e., pronounced) as the suffix -ed on the verb. Thus, there is, at least at some level, a correspondence between the head T and the head V. For researchers that believe that word-building is done in the syntax, the feature [past] in English is pronounced (at least sometimes) as -ed. And moreover, there must be a way to get T and V together---that is, there must be a way to move either T or V---so that they form a word. This is the view that we will adopt. Unfortunately, in this class we cannot delve deeply into the very interesting (and surprisingly contentious) debate concerning how words are built. If you're interested, I recommend taking LING 527.

Tense Morphology

Let's revisit an issue we put aside from chapter 5. There, we noted that, even though there is evidence for a Tense Phrase, sometimes, the expression of Tense is as part of the verb. In Swahili, we noted that tense is a verbal prefix.

Past tense	/li-/	a-li-anguka, 'she/he fell'
Present tense	/na-/	a-na-anguka, 'she/he falls'
Future tense	/ta-/	a-ta-anguka, 'she/he will fall'

Marking tense on the verb is robustly attested across languages. [latex] \begin{flushleft} Kannada (Dravidian) \end{flushleft} \exg. s'aNkaraswami nenne cenna:gi ha:d-id-a\\ Shankaraswamy yesterday well \textbf{sing-\textsc{past}-3\textsc{sm}}\\ \trans `Shankaraswamy sang well yesterday.' \hfill (Sridhar, 1990:220) Kurdish (Indo-Iranian) \exg. Azad-i: Perwin ma:chi: kir\\ Azad-\textsc{erg} Perwin kiss do.\textsc{past}\\ \trans `Azad kissed Perwin.' \hfill (Sarmad Majeed, \textit{pres comm}) [/latex] As we noted above, if T and V are syntactically introduced in different locations, then there must a way to move one to the other so that they're pronounced as a single word. Using an English example, suppose that the expression of past tense (the feature [past]) in English is -ed. Then it appears under the head T, and the verb walk appears under the head V. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$past$]$ \\ -ed}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] So do we move V up to T? Or do we move T down to V? It turns out that languages make different choices. We can diagnose the difference by using adverb placement. Consider first the fact that some adverbs can be placed in-between T and V: [latex] \setcounter{ExNo}{3} \ex. Sarah will often walk home. [/latex] (4) tells us that that when T and V are separate words, then an adverb can be placed between them. This must correspond to VP-adjunction on the left. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ will}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] We can use this information---the fact that some adverbs can left-adjoin to VP to diagnose how T and V get together. In English, we find that when T is a suffix, the V+T word appears to the right of the adverb. [latex] \setcounter{ExNo}{4} \ex. \a. Sarah often walked home. \b. *Sarah walked often home. [/latex] This tells us that T must move down to the V. Otherwise, how would the T information (-ed) get on the other side of the adverb? In contrast, compare what happens with adverb placement in French. Despite the fact that English and French are both canonical [pb_glossary id="828"]SVO[/pb_glossary] languages, only English allows you to put adverbs in between the subject and the verb. In French, you have to put the adverb between the verb and object. [latex] \setcounter{ExNo}{5} \begin{flushleft} French \end{flushleft} \ex. \ag. *Sarah souvent mange des fruits\\ Sarah often eat.\textsc{pres} \textsc{indef} fruit\\ \trans Intended: `Sarah often eats fruit.'\\\bg. Sarah mange souvent des fruits\\ Sarah eat.\textsc{pres} often \textsc{indef} fruit\\ \trans `Sarah often eats fruit.' [/latex] In French, the inflected verb appears to the left of the VP-adjoined adverb. It must therefore be the case that in French, the verb moves up to T. This test allows us to infer movement. Given that we can independently diagnose the location of Tense by where will occurs, we can then consider what happens when we syntactically "separate" T and V. When T and V must form a word, where does that word appear with respect to the intervening material? It appears that some languages are Tense-lowering languages, and some languages are verb-raising languages.

Tense-Lowering and Verb-Raising

Again, the placement of the adverb tells us something about how V and T get together. In particular, because adverbs appear in front of the word V+T in English, we must conclude that T moves down to V in English. How else could we get the adverb in front of Tense? [latex] \setcounter{ExNo}{5} \ex. Sarah often eats fruit. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V+T \\ walk + $[$past$]$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (T.south)--++(south:15em)-|(V.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] In contrast, because adverbs appear after the verb in French, we must conclude that V moves up to T in French. Again, how else could we get the adverb to appear after the verb----between the verb and object? [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ mange + $[$pres$]$ \\ `eats'}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ souvent \\ `often'}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {$t_{V}$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {des fruits \\ `fruits'}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (V.south)--++(south:6em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This kind of movement is called head-movement, because it involves movement of heads. In these trees, we can represent movement with arrows. The arrows in the trees have no theoretical significance. They are purely so that we can read the trees easier. It is perfectly acceptable to write the trees without the arrows. This is because the traces in the trees tell you what has moved, and where. In the English tree, I can see that the head T is not present in TP in the tree above, instead the trace t_T is there. The head T can be found in VP. Likewise, when V raises in French, the trace t_Vtakes the place of the head of VP. Traces are theoretically significant. They are absolutely crucial in our representations. If we left out traces, then there would be no head of TP or VP when T or V moved away. Keep in mind that T-lowering and V-raising are "necessity" movements. The problem with leaving T in TP in a past tense sentence in English is that the [pb_glossary id="772"]feature[/pb_glossary] [past] simply isn't a word on its own. It's a morpheme that can only be pronounced when it's attached to another word. So [past] must move somewhere, because it isn't possible to pronounce it on its own. The problem of getting T and V together goes away when T is just a word, not an affix. Thus, in English, future tense is expressed with the word will. There is no movement of T to V, and hence the adverb is pronounced in between the two, precisely where we expect it to be! [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ will}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] In contrast, in French, future tense is still a suffix, and V still raises to T, as the adverb placement tells us. [latex] \setcounter{ExNo}{6} \ex. \ag. Sarah mang-er-a souvent des fruits\\ Sarah eat-\textsc{fut}-\textsc{3sg} often \textsc{indef} fruits\\ \trans `Sarah will often eat fruit.' \bg. *Sarah souvent mang-er-a des fruits\\ Sarah often eat-\textsc{fut}-\textsc{3sg} \textsc{indef} fruits\\ \trans [Intended: `Sarah will often eat fruit.'] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ mange + $[$future$]$ \\ `will eat'}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ souvant \\ `often'}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {$t_{V}$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {des fruits}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (V.south)--++(south:6em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex]

Key Takeaways

To create morphologically complex words from syntactic heads, movement must take place.
In English T lowers to V; in French V raises to T

Note that the English way of doing things is much rarer cross-linguistically than the French way. As a rule, when V and T form a word together, it's because V raises to T. English is exceptional in this regard.

Tense-to-Complementizer movement

Another clear instance of head-movement occurs in questions in English. [latex] \setcounter{ExNo}{7} \ex. \a. Will Sarah eat fruit? \b. Should Becky drink milk? \c. Could the Celtics win? [/latex] Again, to get this word order, something needs to move, because normally the subject appears in front of T. In this case, T moves to the "question C," represented as C_ques. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C$_{ques}$ + T \\ $\emptyset$ + will}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } }; \draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This sort of movement is called T-to-C movement, or descriptively, "subject-auxiliary inversion." This sort of movement is also head-movement because it again seems to only affect words, not phrases. The head T displaces to a pre-sentential position. T-to-C movement can also be thought of an a "necessity" movement. We can frame this as saying that C_ques is an affix that needs to form a word with something. In English, it does so by "pulling" T to itself, like T does to V in French.

[embed]https://youtu.be/KQhBNhpS8ak[/embed] Head-movement (length: 1m 12s)

Do-support

In English, things get tricky when there are conflicting requirements. For instance, we saw that T can lower to V, and it can also raise to C. So what happens when those two movements come into conflict? What happens when there's a question, but T doesn't contain a full word, only a feature like [pres]? Let us consider the following example. [latex] \setcounter{ExNo}{8} \ex. Does Sarah eat fruit? [/latex] Something very particular happens here: in English, we insert a vacuous element do. What does do mean in this sentence? Nothing. Sarah isn't "doing eating." (For instance, we get the same do with non-action verbs: Does Sarah own a house?) Do is only there to fix a problem, namely, there is no way to pronounce the string of morphemes created when T moves to C. Crucially, C doesn't have a phonological form, and T is just a feature [pres], or at best a [pb_glossary id="822"]suffix[/pb_glossary], -s, that can only be realized on another word (as we have seen with T-to-V movement in English). [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C+T \\ $\emptyset$+[pres]}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] Because do doesn't mean anything, we will assume that it is merely a "trick" of phonology, or more specifically, it is a repair strategy. It repairs something that would be ineffable (meaning "can't be said"). When do-support is needed, we will therefore leave it out of the tree. There's another place where do-support applies in English: negation. [latex] \setcounter{ExNo}{9} \ex. \a. Sarah does not eat fruit. \b. *Sarah eats not fruit. \c. *Sarah not eats fruit. [/latex] Just like the adverb often, the negative element not appears between T and V: [latex] \setcounter{ExNo}{10} \ex. Sarah will not eat fruit. [/latex] But unlike adverbs, not appears to block T lowering to V. So just like when T moves to C, a feature like [pres] cannot be pronounced without some "help." We insert do to repair the situation. Again, we will not represent the word do in our tree, because we are assuming that do-support is a "post-syntactic" operation; it happens after syntax. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$pres$]$}} child {node {NegP} child {node {Neg$'$} child {node {Neg \\ not}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } }; %\draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] But why does an intervening Neg head block T to V lowering? This is explained by appealing to a general constraint on head-movement called, fittingly, the Head-Movement Constraint or HMC. Below is the "informal" version of this constraint. We'll be able to state a "formal" version once we have covered [pb_glossary id="776"]c-command[/pb_glossary] (introduced in chapter 9).

The Head-Movement Constraint (informal version) Head movement cannot "skip over" heads.

What this constraint says is that, when moving one head to another head, you can't pass an intervening head. So the thing that goes wrong in the tree above is that T can't get past Neg.

Note that the Head-Movement Constraint is formulated in such a way that intervening phrases are not a problem. So heads can skip over any and all phrases that are in-between two positions. That's why it can skip over adverbs like often. AdvP comes between T and V, but head-movement simply doesn't "care" about phrases, it only cares about heads.

Auxiliaries

English head-movement is particularly complex. We've seen above that, when necessary and possible, T lowers to V. This is an over-simplification. There are two instances where V raises to T in English: the auxiliary verbs BE and HAVE. Consider what happens when the auxiliaries are present in a question. [latex] \setcounter{ExNo}{11} \ex. \a. Is Sarah eating fruit? \b. *Does Sarah be eating fruit? \ex. \a. Has Sarah eaten fruit? \b. *Does Sarah have eaten fruit? [/latex] With auxiliary verbs, do-support doesn't apply. This makes sense since we've been treating auxiliary verbs like is and have as T-elements. So in a question, we expect that they'll move to C. However, BE and HAVE are clearly verbs---they are of category V. They bear all the morphological indicators of verbs (inflect for person, number, and tense) and they distribute like verbs (appear in places where verbs appear). We therefore conclude that sentences with auxiliaries verbs have two distinct VPs. One VP is headed by the main verb, eaten, and the other is headed by the auxiliary. We moreover conclude that auxiliary verbs raise to T. [latex] \setcounter{ExNo}{13} \ex. Sarah is eating fruit. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eating}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } }; \draw[->, dashed] (aux.south)--++(south:1em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \setcounter{ExNo}{14} \ex. Is Sarah eating fruit? \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} [sibling distance=7em] child {node(C) {C+V+T \\ $\emptyset$+\textsc{be}+$[$pres$]$}} child {node {TP} [sibling distance=6em] child {node {DP} [sibling distance=1em, level distance=2em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eating}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } }; \draw[->, dashed] (aux.south)--++(south:1em)-|(T.south); \draw[->, dashed] (T.west) -| (C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This directly predicts that auxiliaries should appear in front of adverbs like often. [latex] \setcounter{ExNo}{15} \ex. \a. Sarah is often eating fruit. \b. Sarah has often eaten fruit. [/latex]

In this class, we aren't going to deal with the various forms of the main verb. Why the [pb_glossary id="829"] progressive [/pb_glossary] aspect involves suffixing the main verb with -ing in, and why the [pb_glossary id="830"] perfect[/pb_glossary] aspect involves suffixing the main verb with -en are questions that we will not try to address in this class. We will simply assume that the morphological form of the verb that is used in any particular aspect is stipulated. Though note that we could easily arrange for eating to be the result of two distinct heads (-ing and eat) coming together through head-movement. However, in order to keep our trees as small as possible, we won't adopt this idea here.

One of the trickiest problems with English is the fact that auxiliary verbs appear to be able to ignore the [pb_glossary id="831"]head-movement constraint[/pb_glossary]. V-to-T movement is not blocked in the presence of negation. [latex] \setcounter{ExNo}{16} \ex. \a. Sarah is not eating fruit. \b. Sarah has not eaten fruit. [/latex] I know of no (good) explanation for why the HMC doesn't apply here. But that doesn't mean that the HMC isn't a valid generalization. Its effects are robustly attested across languages.

Non-auxiliary be and have

As we just saw auxiliary be and have require move to T. What about when be and have are the only verbs in the sentence? [latex] \setcounter{ExNo}{17} \ex. \a. Sarah is tired. \b. Sarah has a cold. [/latex] Does V still move to T in these sentences? Or, because they're main verbs, maybe T moves to V. This is an empirical question, meaning that we simply have to look at the relevant data. That data is adverb placement, questions, and negation. [latex] \setcounter{ExNo}{18} \ex. \a. Sarah is often tired. \b. Is Sarah tired? \c. Sarah is not tired. \ex. \a. Sarah often has a cold\\(cf, *Sarah has often a cold.) \b. Does Sarah have cold?\\(cf, *Has Sarah a cold?) \c. Sarah does not have a cold\\(cf, *Sarah has not a cold.) [/latex] Looking at the data, we see that be as a main verb patterns just like be as an auxiliary. An adverb can come between be and its complement (AdjP), be raises to C in questions, and be appears in front of not in negative sentences. This behavior makes sense if be always raises to T, whether it's an auxiliary or not. In contrast, when have is used as a main verb, it's behavior is different from have as an auxiliary. Adverbs must appear in front of have, and do-support applies in questions and negative sentences. This behavior makes sense if, with non-auxiliary have, T lowers to V. That is, main-verb have looks just like any other "real" verb, eat, walk, buy, smile, ... [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw[->, dashed] (aux.south)--++(south:.5em)-|(T.south); \draw (left1)--(right1); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} % [sibling distance=7em] child {node(C) {C \\ $\emptyset$}} child {node {TP} % [sibling distance=6em] child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(aux) {V+T \\ \textsc{have}+$[$pres$]$}} child {node {DP} child {node {D$'$} child {node {D \\ a}} child {node {NP} child {node {N$'$} child {node {N \\ cold}} } } } } } } } } }; \draw[->, dashed] (T.south) |- (aux); \draw (left1)--(right1); \end{tikzpicture} [/latex]

Summary of head-movement

In most languages in which tense and the verb are pronounced as one word, V moves to T.
In English,
- For main verbs (including have), T lower to V, unless its blocked by negation or T moves to C.
- For auxiliary verbs, V raises to T in all cases.
- For main verb be, V raises to T in all cases.

The facts of English head-movement are particularly complicated. It is probably best if you just memorize these facts at the beginning. Keep in mind though that you can always reproduce the facts with the data above.

Where we're going: Levels in Syntax

Drawing a tree with movement requires two steps.

Draw the tree without movement first.
Re-draw the tree with movement. Do not erase the first tree.

Thus, to draw the tree for the sentences we've been looking at, it is helpful to draw to distinct trees. [latex] {\underline{\textbf{Before movement (D-structure)}}} \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {V \\ \textsc{be}}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw (right1)--(left1); \end{tikzpicture} [/latex] [latex] {\underline{\textbf{After movement (S-structure)}}} \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw[->, dashed] (aux.south)--++(south:.5em)-|(T.south); \draw (right1)--(left1); \end{tikzpicture} [/latex] There are two "shortcuts" that you might try to do when first drawing trees, and they will invariably get you into trouble. The first shortcut is that you'll want to draw just one tree, putting the movement in as you go. This is a bad way to do it, because sometimes---in fact, often----there is more than one movement. For instance, questions that involve the verb be require movement of V to T, and then T to C. (See tree in above). The second way you might get into trouble is if you start erasing. Here again, you're bound to get confused once multiple movements start happening. Actually, drawing two trees serves a theoretical purpose as well. Once we've introduced the idea of movement, we can think of our syntactic "object" as existing in levels. There's one level which exists "pre-movement." We call this level the Deep-Structure or D-structure for short. Then there's the level after movement. We call this level the Surface-Structure, or S-structure for short. These levels have theoretical significance. Looking forward, we'll find that certain constraints on syntax make reference only to the D-structure, while certain other constraints are sensitive to S-structure.

Advanced

Agglutination

English is a morphologically "poor" language: it tends to use separate words, rather than affixes, to express new meanings. But many languages are more expressive, allowing morphemes to "pile up" on words. The assumption we make is that each new morpheme on a word gets there through Head Movement. Thus, when we see a morphologically complex word with a lot of affixes, this word was "built" in the syntax by iterated instances of Head Movement. Consider a morphologically complex word from the Bantu language Logoori. [latex] \exg. gu-duy-ek-an-i\\ \textsc{sm}-hit-\textsc{ac}-\textsc{EvPl}-\textsc{past}\\ \trans `It was hit repeatedly.' [/latex] The verb guduyekani consists of five different morphemes. There's the root -duy- and then the affixes gu-, -ek, -an, and -i. In the view of syntax that we're adopting in this class, each of these morphemes is added onto the root -duy as is makes its way up the tree. This is called agglutination. That is, each morpheme is the realization of a head in the tree which -duy- moves to. For instance, the suffix -an is an Event Pluralizer, and provides the meaning "repeatedly." This head, call it EvPl, gets attached to -duy- when the verb moves to that phrase. The suffix -ek is an Anticausative, and detransitivizes the verb: the Patient becomes the subject. The prefix gu- and suffix -i are both considered functional morphemes. We typically assume that "Subject Marker" (SM), is associated with the head T, because subject marking in general varies with specifications of T. (For instance, in English, Spanish, French, etc, nonfinite verbs do not agree with subjects; only finite verbs do.) And we (that is, people who study Bantu languages), typically assume that the final vowel -i is associated with the head C. The D-structure and S-structure trees are exemplified below. Note that we're ignoring where the subject is for now. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ \textit{-i}}} child {node {TP} child {node {T$'$} child {node(T) {T \\ \textit{gu-}}} child {node {EvPlP} child {node(Evpl) {EvPl$'$} child {node {EvPl \\ \textit{-an}}} child {node {ACP} child {node(AC) {AC$'$} child {node {AC \\ \textit{-ek}}} child {node {VP} child {node {V$'$} child {node(V) {V \\ \textit{-duy-} \\ hit}} } } } } } } } } }; \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} [sibling distance = 8em] child {node(C) {T+V+AC+EvPl+C \\ \textit{gu-duy-ek-an-i}}} child {node {TP} [sibling distance = 6em] child {node {T$'$} child {node(T) {$t_{T}$}} child {node {EvPlP} child {node {EvPl$'$} child {node(Evpl) {$t_{EvPl}$}} child {node {ACP} child {node {AC$'$} child {node(AC) {$t_{AC}$}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} } } } } } } } } }; \draw[->, dashed, rounded corners] (V.west)-|(AC.south); \draw[->, dashed, rounded corners] (AC.west)-|(Evpl.south); \draw[->, dashed, rounded corners] (Evpl.west)-|(T.south); \draw[->, dashed, rounded corners] ([yshift=-5pt]T.west)-|(C.south); \end{tikzpicture} [/latex] By allowing the head -duy- to move up the tree, it can "collect" the heads along the way, creating the word in the syntax. Note, though, whether something ends up being a prefix or affix is not syntactic. The fact that gu- gets pronounced as a prefix and -i as a suffix is just an unpredictable property of these affixes.

The Mirror Principle

While Head Movement doesn't determine what ends up being a prefix of suffix, it can determine the relative order of multiple prefixes and suffixes. For instance, in the trees above, I have placed the Anticausative head -ek lower in the tree than the Event Pluralizer -an. This is because in the word guduyekani, the suffix -ek is closer to root -duy-. Explicitly, the first head that -duy- moves to is -ek, so that gets attached to the root first. Then everything moves to -an, which will attach outside of -ek. Thus, the order of the morphemes mirrors the order of heads in the syntax.

The Mirror Principle[footnote]Baker, Mark (1985). "The mirror principle and morphosyntactic explanation". Linguistic Inquiry. 16: 373–415 Note that the Mirror Principle is often interpreted as a "rule" that languages have to follow. This is the wrong interpretation. The Mirror Principle is an observation about language. But there are certainly exceptions that appear to violate Mirror Principle orderings.[/footnote] Morphological order reflects syntactic order, and vice versa.

Intuitively, the Mirror Principle says that things that are lower in the tree are closer to the root, and vice versa: things that are closer to the root are lower in the tree. The Mirror Principle is both widely adopted and widely questioned. On the one hand, it is extremely successful in explaining the relative ordering of morphemes in morphological complex words. On the other hand, there are many exceptions to the Mirror Principle, some of which defy good explanation. Putting the empirical picture aside, it is worth noting that this topic is heavily debated. Adopting the Mirror Principle, as we do in this class, forces us to assume a tight relationship between the fields of syntax and morphology. This is a controversial assumption. For more on this debate, and alternative views, consider taking LING 527/727.

Selection

juliedoner — Wed, 15 Nov 2023 05:32:40 +0000

Learning Objectives

By the end of this chapter, you should,

be able to identify c-selectional and s-selectional properties of predicates,
be able to recognize and label thematic roles
understand the basis for UTAH and its implications

We've concerned ourselves so far with making abstract and broad rules for language. In this chapter, we're concerned with more specific things. In particular, what explains why certain words occur in certain environments. As we'll discover below, a word's syntactic environment (what it appears with) is highly idiosyncratic---within bounds. It's still the case that the category 'verb' has a particular syntactic distribution, as does noun, complementizer, etc. First a terminological note. We'll be discussing verbs heavily in this chapter, but sometimes I'll use the term predicate. Predicates are, loosely, what subjects are "about." This notion of predicate groups together verbs, adjectives, and even prepositional phrases. So in the terminology defined below, verbs can "select," adjectives can "select," prepositions can "select," etc.

C(ategory)-selection

Right now, our theory will generate an infinite number of sentences in any human language. But we've done very little to constrain our theory. For instance, nothing we've said so far will explain why (1) and (2) are not possible sentences in English.

[latex] \ex. \a. * Sarah ran Meagan \b. * Amy will kiss \ex. \a. \# The tree ran. \b. \# Chicago will kiss John. [/latex] The examples in (1) seem to be wrong for syntactic reasons (marked ungrammatical with "*"). The verb run doesn't take an object and kiss must take an object. On the other hand, (2) seem to be wrong for semantic reasons (marked infelicitous with "#"). The verbs run and kiss need an [pb_glossary id="815"]animate[/pb_glossary] subject. There's nothing in our theory so far that directly accounts for these patterns. That is, there is nothing built into our theory that makes sure that a verb like run appears without an object and with an animate subject.

Here's another example: Some verbs allow DP and CP complements, while others don't.

[latex] \setcounter{ExNo}{2} \ex. \a. John fears [$_{DP}$ the statue ] \b. John fears [$_{CP}$ that Bill left ] \ex. \a. John kicked [$_{DP}$ the statue ] \b. *John kicked [$_{CP}$ that Bill left ] [/latex]

There are some patterns that we've left implicit up until now, but if we want a coherent theory, we need to make everything as explicit as possible. What the above data teach us is that verbs make different requirements on what the rest of the structure can be. A verb like run dictates that there is only one other DP argument, and that argument is animate. A verb like kiss dictates that there are two other DP arguments, one of which is animate.

The idea that we'll flesh out this week is that on top of our X'-schema, there's a certain amount of information that is associated with each head, in particular, what it needs to combine with. We call this information the selectional properties of the head. [pb_glossary id="794"]Selection[/pb_glossary] comes in two "flavors."

Category selection (c-selection): The categorial requirements of a head.
Semantic selection (s-selection): The semantic requirements of a head.

For example, we say about the verb run that it c-selects for a DP subject and it s-selects for an animate subject. This means that it must have a DP as a subject, and that DP must refer to an animate individual. We say about the verbs kiss and kick that they c-select for a DP subject and a DP object. Both verbs require a DP subject and object, and in both cases, only the subject DP needs to be animate. Fear c-selects for a DP subject and either a DP or a CP object, etc.

As a rule, c-selection is concerned with the syntactic category of the arguments of a predicate, like whether the object must be a CP or a DP. C-selection is also concerned with the number of arguments that a predicate has. For instance, eat and consume have different c-selectional properties: it's possible for eat to only occur with a subject, while consume must occur with both a subject and an object.[footnote]Putting an asterisk outside of parenthesis means "ungrammaticality without this element." Putting asterisks inside of a parenthesis (*X) means, "ungrammatical with this element."[/footnote]

[latex] \setcounter{ExNo}{4} \ex. \a. Brenda ate (pizza). \b. Brenda consumed *(pizza). [/latex]

In short: each word (really [pb_glossary id="799"]morpheme[/pb_glossary]) dictates the syntax around it, both in terms of the category of arguments, and also the number of arguments.

Notice that c-selection very often (though not always) involves a head selecting its complement. A general rule of thumb is that all complements are selected.

Another rule of thumb is that all adjuncts are not c-selected. That is, c-selection dictates the properties of the head that must be met. Since adjuncts are typically optional, they are not c-selected. [latex] \setcounter{ExNo}{5} \ex. \a. April met Leah in the park. \b. *Met Leah in the park. \hfill (must have the subject!) \c. *April met in the park. \hfill (must have the object!) \d. April met Leah. \hfill (doesn't need the adjunct!) [/latex]

S(emantic)-selection

Every predicate (really, every head) dictates what it can, and cannot combine with. So a verb like run cannot combine with a DP object, but a verb like kiss must combine with both a DP subject and a DP object.

However, sometimes things go wrong even when the c-selectional criteria are met. For instance, the following sentences sound incorrect, even though all the c-selectional requirements are met.

[latex] \setcounter{ExNo}{6} \ex. \a. \# The tree ran. \b. \#The wall will punch John. [/latex]

These sentences sounds off for semantic reasons. Trees don't run and walls don't punch. This suggests that in addition to the categorical information, verbs also dictate the "semantic features" of what they combine with. For instance, run and punch both need [pb_glossary id="815"]animate[/pb_glossary] DP subjects. We call this requirement Semantic selection --- it's the semantic counterpart to c-selection.

Semantic selection (s-selection) : The semantic requirements of a head.

When we talk about s-selection with respect to verbs, we're typically (though not always) talking about thematic roles.

Thematic role (theta-role or Θ-role) : The semantic role assigned to each argument in a sentence.

Theta-roles are ways to classify how each argument is meaningfully related to the predicate. For instance, the subject of both run and punch is a "doer." It's performing an action, thus needs to be animate. We call this role Agent. So the subject of both run and punch is an Agent, or has the thematic role of Agent.

On the other hand, the object of punch isn't performing an action, it's affected by the action. We call this role Patient. So the object of punch has the thematic role of Patient. Every [pb_glossary id="841"]argument[/pb_glossary] of a verb has a thematic role.

Examples of thematic roles.

Agent : Someone who purposefully does something.
- John broke the vase
Patient : Someone/thing that is affected, or changes
- John broke the vase
Experiencer : Someone who feels or uses their mind
- John loves Mary
Cause : Someone/thing that causes an action
- The wave sank the ship
Goal : An endpoint, location
- John went to Paris
Recipient : Someone who receives something
- Mary gave John a book
Location : a place
- John is sitting in the park
Instrument : Something that is used
- John chopped the tree with an axe
Theme : The catch-all, if it doesn't fit into any other category.
- John assisted the campaign

Tips in identifying theta-roles

Agents, and Experiencers must be animate (i.e., alive)
Causes and Agents are very similar. A Cause tends to lack a notion of "intentionality" or "purpose."
A Recipient generally involves a "transfer of possession" of some sort, which differentiates it from a Goal.
Theta-roles can vary depending on context.
When in doubt use Theme.

[embed]https://youtu.be/DKthafDLU7M[/embed] Thematic roles (length: 1m 53s) credit: Emma Scott

The point here is that, just like c-selection, different verbs have different requirements on what they s-select for. Some verbs need Agents, some need Patients, some need Experiencers, etc. Notice that, just like with c-selection, order matters. It's not just that punch needs to combine with an Agent and a Patient. The Agent must be the subject, and the Patient must be the object. We call this ordering of thematic roles the thematic structure of a verb.

Thematic structure : The "order" that the s-selectional requirements of the verb are expressed in.

So we say about the thematic structure of run that, "run has an Agent subject." And we say about the thematic structure of punch that, "punch has an Agent subject and a Patient object."

It's important to keep in mind that s-selection is a bit broader than just distribution of thematic roles. For instance, the following sentences sound off, but it's not because the thematic structure is wrong (and it's not because the c-selectional requirements haven't been met).

[latex] \setcounter{ExNo}{7} \ex. \a. \#The doctor gathered in the park \b. \#The soldier surrounded the castle. \c. \#Jesse multiplied the number two. [/latex] The reason that (8) sound wrong is that gather and surround need a plural subject, while multiply needs a plural object. This is, again, an s-selectional requirement of these verbs, not a c-selectional requirement, because the ungrammaticality stems from a semantic "mistake," not a problem with the category of the arguments.

Uniformity of Theta assignment hypothesis

If we spent a long time looking at thematic structure, we'd come up with a pretty striking correspondence between thematic structure and syntactic structure: Particular thematics roles are correlated with particular syntactic positions in a language. For instance, Agents are almost always subjects. Patients are almost always objects. Instruments are almost always in prepositional phrases or otherwise "[pb_glossary id="832"]oblique[/pb_glossary]." This old observation led to a hypothesis: the Uniformity of Theta Assignment Hypothesis or UTAH.

Uniformity of Theta Assignment Hypothesis (UTAH)

Each theta-role always appears in the same syntactic position.

We're going to assume for this class that UTAH is basically right --- although we'll revisit the question in a few chapters. In short, we're going to assume that Agents always appear in the subject position (for now assumed to be specifier of TP). Patients always appear in the object position (complement to V). Instruments always appear in a prepositional phrase, etc.

[embed]https://youtu.be/cxnVszdv4MY[/embed] Thematic structure and UTAH (length: 2m 39s)

Finally, we say that auxiliary verbs do not select for subjects. Their only requirement is that they select for a VP complement. Why? Well, the properties of a subject are never determined by the auxiliary, but rather by lower, main verb. In the examples below, the verbs start and cry determine what is an appropriate subject. [latex] \setcounter{ExNo}{8} \ex. \a. The computer/\#the child is starting. \b. \#The computer/the child is crying. [/latex] The observation that subjects are selected by the main verb, not an auxiliary, is actually extremely important, and it's something that we'll return to in chapter 10.

Where we're going

One purpose of determining selectional properties is to discover classes of words. For instance, we can talk about the class of verbs that only select for a single DP: the intransitive verbs. This are distinguished from the class of verbs that only select for two DPs: the transitive verbs.

We can likewise talk about the class of verbs that take CP complements. It surely cannot be a coincidence that in no language does the verb kick combine with a CP, but basically every language allows say to combine with a CP. In the end, we'd like to come up with an explanation for why certain classes of verbs exist. But sometimes, this is impossible. For instance, eat and consume both mean, roughly, the same thing, but only one obligatorily c-selects for a DP object.

The second reason why it's important recognize that heads dictate the syntax around them, is that we can use that information to diagnose movement. For instance, recall the [pb_glossary id="833"]topicalization[/pb_glossary] constituency test. [latex] \setcounter{ExNo}{9} \ex. Susan, Rachel hugged. [/latex] Intuitively, we think that Susan has moved to the front of the sentence. Why? Because she's who Rachel is hugging. Susan is the object of the verb. Implicit in this intuition is selection. We know (again, perhaps intuitively) that Susan is a Patient in this sentence, because hug is the kind of verb which dictates that its object is a Patient. Or stated differed, hug is the kind of verb that needs an object (due to c-selection), and so we understand that the DP Susan fills that syntactic role. This is our first taste of phrasal movement, meaning, naturally, the movement of phrases. (As opposed to head-movement, which is the movement of heads, discussed in the previous chapter.) Because we know that heads dictate what they must and must not combine with, we conclude that the DP Susan must have combined with hug to satisfy the verb's selectional requirement, and then moved to the front of the sentence.

Advanced

Binding theory

juliedoner — Wed, 15 Nov 2023 05:32:41 +0000

Learning Objectives

By the end of this chapter, you should be able to,

understand c-command
recognize the difference between R-expressions, pronouns, and anaphors
understand their distributional differences
restate the three binding conditions (A, B, and C)
understand how the distribution of referring expressions is constrained by binding theory

In this chapter we'll change gears a little. We are now going to be concerned for the first time with what are apparently non-local dependencies. By this I mean that we are going to be looking at things that interact over a syntactic distance; they aren't next to each other in the tree. We've sort of been hinting at this (e.g., with [pb_glossary id="816"]head-movement[/pb_glossary] in chapter 7) but now we'll delve into some real cases. Besides introducing Binding Theory (which is a perennially interesting topic) in this chapter, the ultimate goal here is to introduce the notion of c-command. This is, arguably, the most important thing ever discovered in syntax, and it underlies almost everything in current theory. C-command (short for "constituent command") is a structural relation. It is defined in the following way.

C-Command A c-commands B iff (if and only if),

A does not [pb_glossary id="777"]dominate[/pb_glossary] B.
B does not dominate A.
Every node that dominates A dominates B.

In prose, X c-commands its sister and everything under its sister. The following diagram illustrates c-command, as well as the notion of domination. [latex] \begin{tikzpicture} \node(top) {XP} child {node(yp) {YP}} child {node {X$'$} child {node {X}} child {node(wp) {WP}} }; \draw[->, dashed, blue] (top)..controls +(east:2) and +(east:2)..(wp) node[midway, right]{\footnotesize{domination}}; \draw[->, dashed, blue] (yp)..controls +(south west:2) and +(south:2)..(wp) node[midway, left]{\footnotesize{c-command}}; \end{tikzpicture} \begin{minipage}{.49\textwidth} \textbf{C-command relations} \begin{itemize} \item YP c-commands X$'$ \item YP c-commands X \item YP c-commands WP \item X$'$ c-commands YP \item X c-commands WP \item WP c-commands X \end{itemize} \end{minipage} \begin{minipage}{.49\textwidth} \textbf{Domination relations} \begin{itemize} \item XP dominates YP, X$'$, X, and WP \item X$'$ dominates X and WP \item[] \item[] \end{itemize} \end{minipage} [/latex] It turns out that c-command matters whenever we talk about non-local dependencies. In this context, "non-local" just means whenever two things interact which aren't in the same phrase. Binding Theory is a very clear example of the importance of c-command, but it's influence can be found in a variety of syntactic processes.

[embed]https://youtu.be/bFCy_cHKzac[/embed] C-command (length:1m 34s) Credit: Megan Spurgeon

Things that refer

Gendered language and referring expressions

In this unit, because we are talking about individuals and their properties, we are using a number of terms that are inherently gendered. It is important to note that our use of gendered terms here is purely a linguistic fact, and does not (or should not) reflect our opinions about gender. In working through the following data, please assume that the individuals in question self-identify as the gender that is indicated. That is, assume that Marta below self-identifies as female, and so we use the pronouns 'she/her' when referring to her. If Marta identified as non-binary, and chose the pronouns 'they/them/their,' the linguistic facts about the distribution of these pronouns would not change: these pronouns---that is, all pronouns---are subject to Condition B of Binding Theory.

In German, there are three linguistic expressions for referring to an individual. Each one is underlined in the following examples. [latex] \ex. \ag. {\underline{Marta}} singt\\ Marta sing.\textsc{pres}\\ \trans `Marta sings.' \bg. {\underline{Sie}} tanzt.\\ she dance.\textsc{pres}\\ \trans `She dances.' \cg. Marta liebt {\underline{sich}} sehr.\\ Marta love.\textsc{pres} \textsc{refl} very\\ \trans `Marta loves herself very much.' [/latex] The first way we can refer is to simply use an [pb_glossary id="773"]R-expression[/pb_glossary] which is short for "referential expression." R-expressions uniquely refer. Marta is an example of an R-expression because there is a unique individual named "Marta" that I'm referring to when I say that word. The reference of Marta never changes. Whenever I use the word Marta, I refer to the individual named by that word. (It is a "rigid designator.") In contrast in (2) we see a [pb_glossary id="775"]pronoun[/pb_glossary], sie. Pronouns can also refer to individuals, but crucially, their reference can change depending context. For instance, sie in (2) likely refers to Marta, but in another context, it could refer to Elise, or Helga, etc. Thus, pronouns and R-expressions share the property of being able to refer, but they differ in that R-expressions have strict reference, while pronouns do not. Finally in (3), we see a third way to refer to an individual in German: the [pb_glossary id="774"]reflexive anaphor[/pb_glossary] sich. Again, this word refers to an individual. It is used to indicate that Marta loves Marta (not Elise, not Helga, etc). Like pronouns, sich doesn't always refer to Marta. So in the sentence, Helga liebt sich sehr, then sich refers to Helga. But unlike pronouns, sich's reference isn't totally free. While sie in (2) could, in principle be anyone (who identifies as female), sich in (3) can only be Marta. It cannot be Elise, or Helga, etc.

Summary of referential terms:

R-expressions R-expressions strictly refer, meaning that they always refers to the same individual across all contexts.
Pronouns Pronouns have variable reference, meaning that the individual that they refer to can change depending on context.
Anaphors Anaphors also have variable reference, because they do not always refer to the same individual. But unlike pronouns, they are not free to refer to anyone. Their reference is fixed in any one context.

This chapter is concerned with the distribution of referential expressions. Even though all three things above can be used to refer to individuals, they all are governed by slightly different rules---and it turns out that these are fundamentally syntactic nature. That is, we will discover that the rules governing when you can and cannot say an R-expression, a pronoun, and a reflexive anaphor are defined purely in terms of syntactic structure.

Constraint on R-Expressions: Condition C

Turning back to German R-expressions (e.g, Marta), pronouns (e.g., sie), and anaphors (e.g., sich), we noted above that all three categories have pretty much the same function: they refer. That is, they pick out an individual or thing. We 'll start with the distribution of R-expressions. Consider, [latex] \setcounter{ExNo}{1} \ex. \ag. Marta sagt, da{\ss} sie singt\\ Marta say.\textsc{pres} that she sing.\textsc{pres}\\ \trans `Marta says that she sings.' \cg. Sie sagt, da{\ss} Marta singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `She says that Marta sings.' [/latex] In these sentences, I've used two different ways to refer to Marta: an R-expression and a pronoun. In principle, both sentences should be able to mean the same thing, because, as we noted above Marta always mean "Marta," and sie can mean "Marta" as well. These sentences do not mean the same thing, though. (2a) means, "Marta says that someone who identifies as female sings." That "someone" can be anyone, including Marta. So it can mean, "Marta says that Marta sings." Indeed, that is probably the preferred meaning. (2b) on the other hand means, "Someone who identifies as female says that Marta sings." What this sentence absolutely cannot mean, though, is "Marta says that Marta sings." We have a way of representing the possible meanings, using indices. I put an index as a subscript on any referring expression. The index is a "pointer." Anything with the same index points to the same thing. So in (3a), since Marta has the subscript "i", sie can have the same subscript, i.e., point to the same individual. In (3b), the two expressions cannot be co-indexed. This is what (3c) demonstrates. The two expressions bear different indices, i.e., refer to different individuals. [latex] \setcounter{ExNo}{2} \ex. \ag. Marta$_i$ sagt, da{\ss} sie$_i$ singt\\ Marta say.\textsc{pres} that she sing.\textsc{pres}\\ \trans `Marta$_i$ says that she$_i$ sings.' \bg. *Sie$_i$ sagt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `*She$_i$ says that Marta$_i$ sings.' \cg. Sie$_j$ sagt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `She$_j$ says that Marta$_i$ sings.' [/latex]

[embed]https://youtu.be/N00FWdPu0f4[/embed] Indices (length: 1m 12s)

Here's a simple hypothesis about what's going on in German.

German Reference Hypothesis #1 (to be rejected) An R-expression cannot be preceded by a co-referential pronoun.

This hypothesis says that the linear order is the problem. It predicts that no matter how long we make the sentence, we still see a similar result. And the data in (4) appear to bear out this prediction! [latex] \setcounter{ExNo}{3} \exg. *Sie$_i$ sagt, da{\ss} Karl$_j$ denkt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Karl think.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `*She$_i$ says that Karl$_j$ thinks that Marta$_i$ sings. [/latex] So Hypothesis #1 looks promising! But here are some counter-examples. [latex] \setcounter{ExNo}{4} \ex. \ag. Ihre$_i$ Schwester sagt da{\ss} Marta$_i$ singt\\ her sister say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `Her sister says that Marta sings.' \bg. Da{\ss} sie$_i$ singt, gef\"allt Marta$_i$\\ that she sing.\textsc{pres} please.\textsc{pres} Marta\\ \trans `That she$_i$ sings pleases Marta$_i$.' [/latex] So it's clear that linear order is not enough. But if linear order isn't sufficient, then there must be something structural. Let's look at the trees for a grammatical and ungrammatical sentence. Note that just like in English, possessive pronouns like ihre 'her' are analyzed as combinations of a pronoun sie 'she' and the possessive determiner 's.[footnote]These trees slightly simplify German's syntax. German has a special property called "V2." In German, the finite verb is always the second thing in the sentence in main clauses. This is derived by V-to-C movement. In embedded clauses, however, we discover that German is actually verb-final (SOV) when the verb does not move to C. [/footnote] Ungrammatical Tree (4) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance = 2ex] child {coordinate(left)} child {node {sie}edge from parent[draw=none]} child {coordinate(right)} } child {node {T$'$} child {node(T) {T+V \\ $[$pres$]$+sagt}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {CP} child {node {C$'$} child {node {C \\ da\ss}} child {node {TP} child {node {DP} [sibling distance=3ex] child {coordinate(left1)} child {node {Maria} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T2) {T+V \\ $[$pres$]$+singt}} child {node {VP} child {node {V$'$} child {node(V2) {$t_{V}$}} } } } } } } } } } } }; \draw[->, dashed] (V2.west)--++(west:1ex)-|(T2); \draw[->, dashed] (V.west)--++(west:1ex)-|(T); \draw (left)--(right); \draw (left1)--(right1); \end{tikzpicture} [/latex] Grammatical Tree (5a) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node {DP} [sibling distance = 3em] child {node {DP} [sibling distance = 2ex] child {coordinate(left)} child {node {sie}edge from parent[draw=none]} child {coordinate(right)} } child {node {D$'$} child {node {D \\ 's}} child {node {NP} [sibling distance = 3ex] child {coordinate(left3)} child {node {Schwester}edge from parent[draw=none]} child {coordinate(right3)} } } } child {node {T$'$} [sibling distance = 5em] child {node(T) {T+V \\ $[$pres$]$+sagt}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {CP} child {node {C$'$} child {node {C \\ da\ss}} child {node {TP} child {node {DP} [sibling distance=3ex] child {coordinate(left1)} child {node {Maria} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T2) {T+V \\ $[$pres$]$+singt}} child {node {VP} child {node {V$'$} child {node(V2) {$t_{V}$}} } } } } } } } } } } }; \draw[->, dashed] (V2.west)--++(west:1ex)-|(T2); \draw[->, dashed] (V.west)--++(west:1ex)-|(T); \draw (left)--(right); \draw (left1)--(right1); \draw (left3)--(right3); \end{tikzpicture} [/latex] In these trees, we observe that in the ungrammatical case, the pronoun sie c-commands the R-expression. In the grammatical case, the pronoun doesn't. That is, when we look at what sie c-commands in each tree, we find a clear distinction between them. Specifically, we see that when sie c-commands Maria - as in (4), where sie is a sister to T', and therefore c-commands T' and everything beneath it (including the R-expression Maria) - the pronoun cannot also refer to Maria. Conversely, we see that when sie doesn't c-command Maria - as in (5a), where sie instead c-commands D' (which crucially does not include Maria) - then the pronoun can refer to Maria. This suggests a different hypothesis:

German Reference Hypothesis #2 (correct): An R-expression cannot be c-commanded by a co-referential pronoun.

Hypothesis #2 doesn't care about the linear order, i.e., the order in which we pronounce the words. Instead Hypothesis #2 cares about the hierarchical relationship between the referring expressions. It says that an R-expression cannot be c-commanded by a pronoun if they refer to the same individual. Referring to same individual is called [pb_glossary id="778"]coreference[/pb_glossary]. We have a specific term for this relationship involving c-command and coreference: binding.

Binding: X binds Y iff [if and only if],

X c-commands Y
X and Y are co-indexed (i.e., X and Y refer to the same thing).

Hypothesis #2 appears to be empirically correct for German. If we were to look at a lot of data (and control for a lot of factors), we would find that, as a rule, R-expressions cannot be c-commanded by a co-referential pronoun. Indeed, an R-expression cannot be c-commanded by anything co-refential. This restriction on R-expressions, doesn't just hold for German. It appears to be a general property of language --- basically universal. It is called [pb_glossary id="779"]Condition C[/pb_glossary] (also Principle C) of Binding Theory.

Condition C An R-expression cannot be bound, or, An R-expression must be free. ("Free" means "not bound.")

A Condition C "violation" involves two things

The R-expression being c-commanded by a co-referential expression; and
both expressions referring to the same individual.

If both of these things happen, then Condition C is violated, and the sentence is ungrammatical. For Binding Theory, we calculate c-command from the phrase level. That is, we always calculate binding by considering the relationship between referential DPs. [latex] {\ovalbox{ \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.DP$_i$ [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] \hspace{150pt} \Tree [.DP [.{D$'$} {D \\ $\emptyset$} [.NP$_i$ [.{N$'$} {N \\ John} ] ] ] ] \node(upleft) at (-1,0) {}; \node(upright) at (1,0) {}; \node(downleft) at (-1,-5.5) {}; \node(downright) at (1,-5.5) {}; \draw[thick] (upleft) -- (downright); \draw[thick] (upright) -- (downleft); \hspace{150pt} \Tree [.DP [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N$_i$ \\ John} ] ] ] ] \node(upleft) at (-1,0) {}; \node(upright) at (1,0) {}; \node(downleft) at (-1,-5.5) {}; \node(downright) at (1,-5.5) {}; \draw[thick] (upleft) -- (downright); \draw[thick] (upright) -- (downleft); \end{tikzpicture} }} [/latex] This must be the case when we consider, say, possessives. The top DP refers to something that "encompasses" the embedded DP. The only constituent that contains both phrases is the highest DP. The phrase John's dog refers to a dog, not John. [latex] \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.DP$_i$ [.DP$_j$ [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] [.{D$'$} {D \\ 's} [.NP [.{N$'$} {N \\ dog} ] ] ] ] \end{tikzpicture} [/latex]

How to talk about binding

Below are various ways students have talked about binding and binding theory. They are all wrong. [latex] \Tree [.{\ldots} XP$_i$ [.{\ldots} {\ldots} YP$_i$ ] ] [/latex]

"YP is bound to/with XP" → should be "YP is bound by XP''
"XP and YP are bound together." → should be "XP binds YP."
"YP is coindexed." → should be "YP is coindexed with XP."
"Y is binded/bounded" → should be "Y is bound."
"XP and YP c-command" → should be "XP c-commands YP."
"YP is bound by a c-commanding XP" is redundant. Being bound entails being c-commanded.

Things you can say about this binding relationship.

"XP binds YP."
"YP is bound by XP."
"XP is a binder for YP, is YP's binder."
"XP c-commands YP."
"XP and YP are coindexed," or "XP and YP have the same index."

Before turning in assignments, you should check to make sure you're not making one of these errors.

Constraint on Anaphors: Condition A

Condition C tells us about the distribution of R-expressions. We now turn to anaphors, like German sich. Here are some examples. Pay careful attention to the indices.

[latex] \setcounter{ExNo}{5} \ex. \ag. Marta$_i$ liebt sich$_i$\\ Marta love.\textsc{pres} \textsc{refl}\\ \trans `Marta$_i$ loves herself$_i$.' %\bg. *Marta$_i$ sagt da{\ss} Helga$_j$ sich$_i$ liebt \\ %Marta say.\textsc{pres} that Helga \textsc{refl} love.\textsc{pres}\\ %\trans `Marta$_i$ says that Helga loves herself$_i$.' \cg. *Martas$_i$ Bruder liebt sich$_i$\\ Marta.\textsc{gen} brother love.\textsc{pres} \textsc{refl}\\ \trans `*Marta$_i$'s brother loves herself$_i$.' [/latex]

Using the same reasoning as above, we might come to the conclusion that anaphora are governed by rules that are precisely the opposite of R-expressions.

German Anaphor Hypothesis #1: (to be revised) An anaphor must be bound.

That is, unlike an R-expression, an anaphor must be c-commanded by a co-referring expression. This looks pretty good, but it isn't quite enough. Consider the following sentence. Why does it appear to falsify the hypothesis above?

[latex] \setcounter{ExNo}{6} \exg. *Marta$_i$ sagt da{\ss} Helga$_j$ sich$_i$ liebt\\ Marta say.\textsc{pres} that Helga \textsc{refl} love.\textsc{pres}\\ \trans `Marta$_i$ says that Helga loves herself$_i$.' [/latex] The problem here is that sich is bound! Marta is coindexed with anaphor, and Marta definitely c-commands the anaphor, since Marta is a the subject of the entire sentence. It's possible to keep the idea that anaphors must be bound, but we have to add another clause in our hypothesis. The generalization is that anaphors have to be bound "close enough". That is, there is some minimal syntactic unit inside of which an anaphor must be bound. We call this minimal syntactic unit the [pb_glossary id="780"]binding domain[/pb_glossary]. For the sake of this class, we'll say that its the closest TP that dominates the anaphor.

The constraint on anaphors is Condition A of the Binding Theory.

Condition A

An anaphor must be bound in its binding domain (where the binding domain is the closest dominating TP).

For instance, consider the first sentence again,

[latex] \setcounter{ExNo}{7} \exg. Marta liebt sich sehr\\ Marta love.\textsc{pres} \textsc{refl} very\\ \trans `Marta loves herself very much.' [/latex] Here the binding domain for sich is the highest TP---which happens to be the only TP. And because Marta is also inside of this TP, sich is bound, satisfying Condition A. In (7) on the other hand, the binding domain for sich is the lowest TP. Since Marta is not in this TP, sich is free (unbound) and so violates Condition A. If there are two anaphors in a sentence, they might each have their own binding domain. [latex] \setcounter{ExNo}{8} \exg. Marta$_i$ erkl\"art sich$_i$, da{\ss} Helga$_j$ sich$_j$ liebt \\ Marta explain.\textsc{pres} \textsc{refl} that Helga \textsc{refl} love.\textsc{pres}\\ \trans `Marta$_i$ explains to herself$_i$ that Helga$_j$ loves herself$_j$.' [/latex] Again, Condition A of Binding Theory is universal---although there are some additional complications. First, there may be further constraints on an anaphor in a language. For instance, in English, anaphors must match in number and gender. Assuming that Sue identifies as female, it is not possible to say *Sue loves himself. This not because the anaphor isn't bound, but because the anaphor doesn't match the gender. An additional complication is that the things that look like anaphors in a language often have other (perhaps related) uses. For instance, reflexive pronouns in English are often used colloquially for emphasis. [latex] \setcounter{ExNo}{9} \ex. Gary likes to put his feet up after a long day of work. As for myself, I like to hit the gym. [/latex]

Constraint on pronouns: Condition B

Finally, we can look at pronouns. These are actually the most complex. In some ways they're like R-expressions. Consider the following.

[latex] \setcounter{ExNo}{10} \ex. \ag. *Marta$_i$ liebt sie$_i$\\ Marta love.\textsc{pres} her\\ \trans `*Marta$_i$ loves her$_i$.' \bg. Martas$_i$ Bruder liebt sie$_i$\\ Marta.\textsc{gen} brother love.\textsc{pres} her\\ \trans `Marta$_i$'s brother loves her$_i$.' [/latex] This is similar to Condition C because it looks like the pronoun cannot be bound by a co-referring expression. In (10a), Marta c-commands sie, and they're co-indexed, so the pronoun is bound. The result is ungrammatical. In contrast, in (10b), the pronoun is free, and the result is grammatical. Based on this data, we therefore posit the following hypothesis:

German Pronoun Hypothesis #1 (to be rejected):

Pronouns are subject to Condition C.

However, if pronouns were just like R-expressions, we would expect the sentence in (11) to be ungrammatical. (For comparison, you can reverse the pronoun and the R-expression, in which case it's a Condition C violation.)

[latex] \setcounter{ExNo}{11} \exg. Marta$_i$ sagt da{\ss} Helga sie liebt\\ Marta say.\textsc{pres} that Helga her$_i$ love.\textsc{pres}\\ \trans `Marta$_i$ says that Helga loves her$_i$.' [/latex] In fact, pronouns are also a bit like Condition A, in that they make reference to a binding domain (again assumed to be the closest TP). Here's Condition B of Binding Theory.

Condition B

A pronoun cannot be bound in its binding domain,

or,

A pronoun must be [pb_glossary id="817"]free[/pb_glossary] in its binding domain.

Summary of Binding Theory

Condition A: An anaphor must be bound in its binding domain.
Condition B: A pronoun must be free (=cannot be bound) in its binding domain.
Condition C: An R-expression must be free (=cannot be bound).

We will assume for the purpose of this class that the binding domain for all referential expressions is the closest TP.

A helpful (?) mnemonic:

Condition A covers anaphors.
Condition B covers bronouns.
Condition C is for the other thing: R-expressions.

Using Binding Theory

Because Binding Theory is assumed to be universal, that is, we can detect Conditions A, B, and C, in every language, is extremely helpful in diagnosing structure.

For instance, consider the case of [pb_glossary id="781"]pro-drop[/pb_glossary] languages. These are languages in which a pronominal subject and/or object does not need to be explicitly expressed. Swahili is pro-drop language.

[latex] \setcounter{ExNo}{12} \ex. \ag. a-li-cheka\\ 1\textsc{sm}-\textsc{past}-laugh\\ \trans `She/he laughed.' \bg. a-li-ona mti mkubwa\\ 1\textsc{sm}-\textsc{past}-see tree big\\ \trans `She/he saw a big tree.' [/latex] It's possible to think of pro-drop languages in two different ways.

Hypothesis #1: Pro-drop languages don't have pronouns. That is, there just isn't anything in the structure to represent he, she, they, etc.

Hypothesis #2: Pro-drop languages have silent pronouns. That is, there's a DP pronoun that doesn't have phonological content.

The two hypotheses are actually widely debated, with many different variations on these hypotheses. The question you can consider, What predictions does each hypothesis make with respect to Binding Theory? For instance, consider the following sentence and its English translation. (Note that Juma is typically a name for a male individual.) [latex] \setcounter{ExNo}{13} \exg. a-li-sema kwamba Juma a-li-cheka\\ 1\textsc{sm}-\textsc{past}-say that Juam 1\textsc{sm}-\textsc{past}-laugh\\ \trans `She$_j$/he$_j$ said that Juma$_i$ laughed.'\\ CANNOT MEAN: `He$_i$ said that Juma$_i$ laughed.' [/latex] The meaning that is attributed to the sentence in (13) is consistent with only one of the hypotheses above concerning pro-drop languages. Hypothesis 2 predicts that the meaning He_i said that Juma_i laughed does not exist because there is a silent pronoun binding the R-expression, thereby violating Condition C. Hypothesis 1 does not predict the unavailability of this meaning. A different explanation would have to be developed. So Binding Theory can be used to determine structure. This is true even for well-studied languages like English. Consider double object constructions. [latex] \setcounter{ExNo}{14} \ex. Mary showed Trisha a painting. [/latex] In such example, it looks like the verb show is selecting for two objects, Trisha and a painting. This suggests that double object construction should have a "flat" structure, directly contradicting our X'-schema. [latex] \begin{tikzpicture}[mytree] \node(top) {VP} child {node {V$'$} child {node {V \\ show}} child {node {DP \\ Trisha}} child {node {DP \\ a painting}} }; \end{tikzpicture} [/latex] This tree makes very clear predictions though. It says that Trisha c-commands a painting and that a painting c-commands Trisha. They mutually c-command each other. The following data suggest however that this cannot be correct. [latex] \setcounter{ExNo}{15} \ex. \a. *Mary showed her\textsubscript{i} a painting of Susan\textsubscript{i} \b. Mary showed Susan\textsubscript{i}'s friend her\textsubscript{i} (in the photograph) [/latex] This data suggests that the indirect object, the first object, c-commands the direct object, the second object, but not vice versa. We'll return to how we can make this syntactically work later. For now, the takeaway is that we can use Binding Theory to determine what the syntax looks like.

Things to remember

The difference between R-expressions, pronouns, and anaphors
The difference between Conditions A, B, and C of Binding Theory
How to use Binding Theory to determine syntactic structure
C-command

Where we're going: C-command elsewhere in syntax

We've demonstrated here the importance of c-command with respect to Binding Theory. But c-command is evident everywhere is syntax. Whenever we have a non-local dependency, we find that c-command mitigates that dependency in some way. For instance, consider again Head Movement from chapter 7. It is not clear that Head Movement also obeys c-command: you can only move to the closest c-commanding head. For instance, recall Head Movement, where a head is allowed to move up the tree ``past'' other phrases. [latex] \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.CP [.{C$'$} {C \\ $\emptyset$} [.TP [.DP [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] [.{T$'$} \node(T){V+T \\ be+$[$pres$]$}; [.VP [.{V$'$} [.AdvP [.{Adv$'$} {Adv \\ sometimes} ] ] [.{V$'$} \node(V){V \\ $t_{be}$}; [.AdjP [.{Adj$'$} {Adj \\ hungry} ] ] ] ] ] ] ] ] ] %\draw[->, dashed] (V) ..controls +(south:4) and +(south:6)..(T); \draw[->, dashed] (V)--++(south:6em)-|(T); \end{tikzpicture} [/latex] The reason that we never see V more to, say, sometimes, is because the head Adv doesn't c-command V. That is, you can only head-move to a c-commanding head. We'll discover soon that there are other types of movement, and those two are constrained by c-command. As a general rule, you can only move to a position that c-commands your starting position. (The exception being head-lowering. ADD MORE)

Advanced

Pronouns are "special." We've discussed above the cases if pronouns that refer, that is, that point to an individuals (or individuals) in the discourse. But pronouns can sometimes have other functions. Consider the following pair of sentences.[footnote]Many people nowadays prefer the non-gendered "singular they" instead of she in sentences like this. There are prescriptive rules against this kind of singular they, it has been standard usage for hundreds of years.[/footnote] [latex] setcounter{ExNo}{16} \ex. \a. Beth thinks that she's the smartest. \b. Every student thinks that she's the smartest. [/latex] The sentences in (17) actually involve two different uses of the pronoun she. In (17a), the most natural reading is that she refers back to Beth.This is the use that we're familiar with; the pronoun is being used to refer to an individual. But what about (17b). Does she refer to a particular individual? On one reading, the answer is no. There's a reading of (17b) in she refers to different people. Consider the context in which we are talking about the students Sarah, Nekesa, Maria, and Althea (and each of these individuals identifies as female). We can understand (17b) to mean that "Sarah thinks that Sarah is the smartest" and "Nekesa thinks that Nekesa is the smartest" and "Maria thinks that Maria is the smartest" and "Althea thinks that Althea is the smartest." That is, as we run through the list of students, she refers to each of the students. (Note that she doesn't refer to every student: this sentence cannot mean "Every student thinks that every student is the smartest.") This reading doesn't treats she as a referring expressing because the pronoun doesn't pick out a particular individual in the discourse. Instead, she is a variable; she's reference varies depending on the associated quantifier. In this case, the quantifier is every. It turns out that variables have very different properties than referring pronouns: variables must be bound. For instance, the co-varying interpretation of she goes away if there is no c-command. [latex] \setcounter{ExNo}{17} \ex. *Every student$_i$'s father thinks that she$_i$'s the smartest [/latex] In (18), the variable interpretation doesn't seem to work; we can't let she mean "each of the students" in this example. (Note the use of indices here: putting an index on every student and she indicates that we should make those two DP's co-refer.) Variables can be bound "long-distance" as well: [latex] \setcounter{ExNo}{18} \ex. Every student$_i$ thinks that Bill said that she$_i$'s the smartest. [/latex] The takeaway from this is that pronouns in English (and many other languages) have two distinct functions; each function has a distinct distribution.

Referential pronouns refer to a discourse salient individual and are subject to Condition B of Binding Theory. Variable pronouns have variable reference and must be bound by a quantified expression.

Phrasal movement: A-movement

juliedoner — Wed, 15 Nov 2023 05:32:41 +0000

Learning Objectives

By the end of this chapter, you should be able to,

diagnose unaccusative and unergative verbs
draw trees representing the unaccusative/unergative distinction
understand the motivation for a low position of transitive/unergative subjects
understand the motivation for Voice

We have seen empirical evidence that the subject of the sentence (in the languages we've looked at) is in spec-TP. More specifically, we've seen evidence that spec-TP is filled at [pb_glossary id="782"]S-structure[/pb_glossary]. In this chapter, we're going to discover that the subject doesn't start there; it moves there. Again more specifically, at [pb_glossary id="834"]D-structure[/pb_glossary], spec-TP is empty. Subjects start lower in the tree and move to spec-TP. This is phrasal movement, that is, the movement of phrases (rather than heads, discussed in chapter 7). Phrasal movement is divided into two distinct domains: A-movement and A'-movement. This chapter will deal exclusively with A-movement. A'-movement is addressed in chapter 10. We'll approach phrasal movement in steps. First we'll show that some subjects can move to spec-TP. Then we'll show that all subjects move to spec-TP. We'll then deal with the question of where the subject starts from.

A quick refresher of what what was covered in chapter 8.

Verbs c-select their arguments. That is, they dictate how many, and what kind of, phrases they are required to appear with.
Verbs s-select their arguments. That is, they dictate the semantic properties, including thematic roles, of their arguments.
Every argument has one (and only one) theta-role .
Every theta-role always appears in the same "syntactic configuration" (=[pb_glossary id="784"]UTAH[/pb_glossary]).

Unaccusatives and unergatives

Consider the verbs in (1a) and (1b). [latex] \ex. \a. John sang. \b. The vase broke. [/latex] The question we will consider is whether (1a) and (1b) have the same tree. That is, if they're both intransitive verbs, do they both simply have a subject in the specifier of TP? Possibly. Call this Hypothesis #1.

Intransitive Verb Hypothesis #1 All subjects of intransitive verbs start in spec-TP.

This hypothesis is what we have been assuming: there is a dedicated subject position, and all subjects sit there. However, there's an alternative hypothesis which follows from our observation that thematic roles are associated with particular positions in the syntax (i.e., [pb_glossary id="784"]UTAH[/pb_glossary]). Notice that the parallelism between the intransitive verbs breaks down when we look at the transitive verbs. With sing, John is the subject of the transitive verb and the subject of the intransitive verb. But with break, the vase is the object of the transitive verb and subject of the intransitive verb. [latex] \setcounter{ExNo}{1} \ex. \a. \textbf{John} sang a song. \b. \textbf{John} sang. \ex. \a. John broke \textbf{the vase}. \b. \textbf{The vase} broke. [/latex] Thus, the alternative hypothesis is that, because the vase is a Patient in both the transitive and intransitive verb frames (it undergoes a change), it starts life as the object of break, and then moves to the subject position. This follows from UTAH, because UTAH holds that the Patient thematic role is always syntactically represented as the complement to V. Call this Hypothesis #2.

Intransitive Verb Hypothesis #2 Some subjects of intransitive verbs start as complements to V.

[latex] \textbf{Hypothesis \#1 D-structures} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ John}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ sing}} } } }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ the vase}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ break}} } } }; \end{tikzpicture} \end{minipage} \vspace{10pt} \textbf{Hypothesis \#2 D-structures} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ John}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ sing}} } } }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ break}} child {node {DP \\ the vase}} } } }; \end{tikzpicture} \end{minipage} [/latex] The idea behind hypothesis #2 is that when there is no Agent, the object of break becomes the subject by moving from comp-VP to spec-TP. [latex] \textbf{Hypothesis \#2 S-structure} \begin{tikzpicture}[mytree] \node(top) {TP} child {node(subj) {DP \\ the vase}} child {node {T$'$} child {node(T) {$t_T$}} child {node {VP} child {node {V$'$} child {node(V) {V+T \\ break + $[$past$]$}} child {node(obj) {$t_{DP}$}} } } }; \draw[->, dashed] (T.west)--++(west:.2em)|-(V.west); \draw[->] (obj) --++ (south:3em) -| (subj); \end{tikzpicture} [/latex]

The following diagnostics allow us to determine which of the two hypothesis is correct. In fact, the data suggest that hypothesis #2 is right. We distinguish two kinds of intransitive verbs.

Unergatives are intransitive verbs that lack an object, i.e., only have an [pb_glossary id="836"]external argument[/pb_glossary].

Unaccusatives are intransitives that lack a subject, i.e., only have an [pb_glossary id="837"]internal argument[/pb_glossary].

Note that when we say "lack a subject," we mean that they lack a subject at one level of representation: D-structure.

Unergatives and unaccusatives differ empirically on a number of levels. All the following diagnostics highlight the crucial difference between unergatives and unaccusatives: the former lack an internal argument, while the latter lack an external argument. That is, you can use these as tests to determine whether a verb is unaccusative or unergative.

Resultatives test.

A resultative phrase is a phrase that describes a "result state." For instance, in Keisha broke the vase into pieces, the phrase into pieces is a resultative, because it describes the result state of the vase after Keisha broke it.

Unaccusatives permit modification of the object with a "resultative" phrase, i.e., something that describes the result state of the object.

[latex] \setcounter{ExNo}{3} \ex. \a. The vase broke into pieces. \b. The water froze solid. \ex. \a. *John sang into pieces. \b. *Mary laughed solid. [/latex] The idea behind this test is that resultative phrases are in some kind of relationship with the object of the sentence. When I say, Keisha broke the vase into pieces, I'm describing the state that the vase was in after Keisha broke it. More abstractly, I'm saying something like "Keisha CAUSED the vase BE in pieces," where "CAUSE" and "BE" are abstract verbal notions. The intransitive sentence The vase broke into pieces is simply missing the CAUSE abstract verb, but the predication "the vase BE in pieces" is still present. Unergatives don't allow resultatives because the intransitive subject isn't in the same relationship with the resultative phrase: the predication "John BE in pieces" or "Mary BE solid" does not work in (5) because John and Mary were never objects of the verb (and therefore cannot be modified by the resultative).

[embed]https://youtu.be/sx29Gi_d9z8[/embed] Resultatives test (length 59s) credit: Malena Schoeni

Participial modification.

You can use unaccusatives as participial modifiers.

[latex] \setcounter{ExNo}{5} \ex. \a. The vase broke $\rightarrow$ the broken vase \b. The water froze $\rightarrow$ the frozen water \ex. \a. The man sang $\rightarrow$ *the sung man \b. the woman laughed $\rightarrow$ *the laughed woman [/latex] The idea behind the participial modification test is that the participial form of the verb still subcategorizes for an object position, but excludes a subject position.

[embed]https://youtu.be/vRhTmmUiOJU[/embed] Participial modification (length: 45s) credit: Corey McCulloch

X's way/X-self.

You can add either X's way or X-self to unergatives, but not unaccusatives.

[latex] \setcounter{ExNo}{7} \ex. \a. *The vase broke its way home \b. *The water froze itself silly. \ex. \a. John sang his way through the mall. \b. Mary laughed herself silly. [/latex] The idea behind this test is that with unergative verbs, there is no object, and so something else can be placed in that position. That is, you can "give" an unergative verb an object because comp-VP is, in a sense, "free." You can't give unaccusative verbs an object because comp-V is already occupied; there's nowhere to put an extra object.

[embed]https://youtu.be/020jEqhxDPY[/embed] X's way (length: 45s) credit: Corey McCulloch

Cognate objects.

Unergatives allow an object that expresses the same thing as the verb --- a cognate object, which is a noun that describes the same action as the verb.

[latex] \setcounter{ExNo}{9} \ex. \a. *The vase broke a big break. \b. *The water froze a big freeze. \ex. \a. John sang a song \b. Mary laughed a hearty laugh. [/latex] This test has the same purpose as the X's way/X-self test: it's diagnosing whether the object position (comp-VP) is empty. If it is, you can put a cognate object in the structure.

[embed]https://youtu.be/GWhH_Bt8YnY[/embed] Unaccusative syntax (length: 1m 31s)

Crosslinguistic evidence for the ergative/unaccusative distinction

The so-called Unaccusative Hypothesis has quite a bit of evidence in its favor, and a lot of that evidence come from crosslinguistic patterns.

Auxiliary selection.

In many languages, particularly Indo-European languages like Romance and Germanic, the auxiliary used to express past tense/perfect aspect is (at least superficially) sensitive to the distinction between unergatives and unaccusatives. In many languages, the auxiliary verb be is used for unaccusatives, and have is used for unergatives. [latex] \setcounter{ExNo}{12} German \ex. \ag. Die Vase ist gebrochen\\ the vase is broken\\ \trans `The vase broke.' \bg. Maria hat gesungen\\ Maria has sung\\ \trans `Mary sang.' [/latex]

Morphological distinctions.

In many languages, there are morphological indicators that distinguish unaccusatives and unergatives. For instance, in Romance languages, many unaccusatives obligatorily occur with the so-called reflexive clitic se/si. Unergative verbs never appear with reflexive clitic. [latex] \setcounter{ExNo}{13} \ex. \ag. La vase se casse\\ the vase \textsc{refl} break.\textsc{pres}\\ \trans `The vase breaks.' \bg. Marie (*se) chante\\ Marie \textsc{refl} sing.\textsc{pres}\\ \trans `Marie sings.' [/latex] In Logoori, a Bantu language of Kenya, many unaccusative verbs appear with the suffix -ik. Unergative verbs may never appear with this suffix. [latex] \ex. \ag. Sira a-han-i muriang'o\\ Sira \textsc{sm}-close-\textsc{fv} door\\ \trans `Sira closed the door.' \hfill Transitive \bg. muriang'o gu-han\textbf{-ik}-i\\ door \textsc{sm}-close-\textsc{ac}-\textsc{fv}\\ \trans `The door closed.' \hfill Unaccusative \ex. \ag. Sira y-iimb-i ruembo\\ Sira \textsc{sm}-sing-\textsc{fv} song\\ \trans `Sira sang a song.' \hfill Transitive \bg. Sira y-iimb-i\\ Sira \textsc{sm}-sing-\textsc{fv}\\ \trans `Sira sang.' \hfill Unergative [/latex]

To be clear, we have to be somewhat cautious about our diagnostics and evidence. While the data listed above can pick out unaccusatives, it is not guaranteed that an unaccusative verb must display all the properties listed above. For instance, the English verb arrive does not permit cognate objects, but also does not permit participial modification. [latex] \ex. \a. *Carol arrived a big arrival. \b. ??the arrived man. [/latex] Despite this divergent evidence, we generally believe that arrive is unaccusative. This is in part based on cross-linguistic evidence, and also in part based on factors we do not have time to consider here.

On passivization

We've shown above that sometimes, the subject moves to spec-TP. Indeed, there is pretty clear evidence that not all subjects start in spec-TP just from [pb_glossary id="785"]passivization[/pb_glossary]. Passivization is a syntactic process that turns a transitive verb into an intransitive verb. [latex] \setcounter{ExNo}{14} \ex. \a. John broke the vase. \hfill ``active'' \b. The vase was broken. \hfill ``passive'' [/latex]

The subject in (15b) starts as the object of break, and then gets "promoted" to the subject position. This is just like the unaccusative syntax we observed earlier: the D-structure object becomes the S-structure subject. However, passives and unaccusatives differ in a few crucial ways. Most notably, passives allow the external argument to be expressed as an [pb_glossary id="832"]oblique[/pb_glossary] argument (in a prepositional phrase). Unaccusatives do not permit this.

[latex] \setcounter{ExNo}{15} \ex. \a. The vase was broken by Tabitha \hfill passive \b. *The vase broke by Tabitha \hfill unaccusative [/latex]

This suggests that even though both passives and unaccusatives involve movement of an object to a subject, they mean slightly distinct things. The passive alternation in English is complex, and so we're going to put it aside here.

Generalized movement to spec-TP

We now know that some subjects don't start in the specifier of TP, they move there. Specifically, unaccusatives (and passives) demonstrate that not all subjects start in D-structure in the specifier of TP.

Now let's ask whether anything starts in spec-TP.That is, we used to have a uniform theory: subjects starts in spec-TP. Now we have a "split" theory: some subjects start in spec-TP, some move there. Now we're going to explore whether a uniform theory is still possible: maybe all subjects move to spec-TP.

Though we'll go over empirical evidence for this momentarily, note that purely on theoretical grounds, having subjects of transitive and unergative verbs start in spec-TP is fishy. In all other cases, selection is local, meaning that "a head dictates the syntax around it" (as discussed in chapter 8). The problem appears to be that subjects (of unergative and transitive verbs) aren't next to what is selecting them. That is, if sing is unergative and it selects for its subject, why is the subject so far away from the verb in examples like (17)? [latex] \setcounter{ExNo}{16} \ex. The man has been singing. [/latex] In the following sections, we'll again be weighing two hypotheses. The first hypothesis is the null hypothesis: subjects of transitive and unergative verbs start in spec-TP, just like we've been doing. The second hypothesis is that subjects start somewhere closer to the verb. (We'll be more precise about where shortly.) That is, nothing starts in spec-TP. Everything that we pronounce in spec-TP has moved there. The two hypotheses are sketched below. [latex] \noindent\textbf{Hypothesis \#1 ($=$the ``old'' way)}\\ \begin{tikzpicture} \Tree [.TP [.DP ] [.\textoverline{T} T [.{\ldots} \edge[roof]; {\ldots} ] ] ] \end{tikzpicture} \noindent\textbf{Hypothesis \#2 ($=$the ``new'' way)}\\ \begin{tikzpicture} \Tree [.TP [.\node(dp){DP}; ] [.\textoverline{T} T [.{\ldots} \edge[roof]; {\ldots} ] ] ] \node(t) at (.6,-3.2) {$t_{DP}$}; \draw[->, rounded corners] (t) -- ++(west:.75) -| (dp); \end{tikzpicture} [/latex]

These hypotheses make different predictions.

Hypothesis #1 predicts that a transitive/unergative subject can never be pronounced in the verb phrase, and that there is no evidence for movement to spec-TP.

Hypothesis #2 predicts that a subject can be pronounced in the verb phrase (if it hasn't moved), and that there is evidence for movement to spec-TP.

So we're looking for two kinds of data.

Is there evidence for a low position of the subject?
Is there evidence for movement to spec-TP?

There's lots of evidence for 1., and a bit of evidence for 2. Let's start with arguments that the subject can be pronounced lower than spec-TP.

Evidence for a subject low position

Small clauses.

Some clauses are "smaller" than others. The bracketed constituents are missing a tense phrase (TP).

[latex] \setcounter{ExNo}{17} \ex. \a. I saw [ Mary hug a pedestrian ] \b. I heard [ John sing ] \c. I made [ Carrie build a house ] [/latex] The subjects of hug, sing and build are not in spec-TP because there is no TP above hug, sing and build. (You can tell because hug, sing, and build do not---and cannot---appear with tense morphology: *I saw Mary will hug a pedestrian.) How could Mary, John and Carrie even get into the tree if all subjects start in spec-TP? These data suggest that Hypothesis #2 is correct, because Hypothesis #1 predicts subjects are only found in spec-TP; if there's no spec-TP, there can be no subject.

Existential constructions.

Consider the following alternation. There in (19b) is called an [pb_glossary id="787"]expletive[/pb_glossary] subject. Expletive subjects are semantically vacuous. [latex] \setcounter{ExNo}{18} \ex. \a. Three woman are climbing a mountain. \b. There are three women climbing a mountain. [/latex] Again, these example suggest that, at least sometimes, the subject of the sentence can be pronounced not in spec-TP. In the case of expletive there, something else sits in spec-TP, namely there, and so the subject three women must be somewhere else---somewhere lower.

Should we also consider the possibility that there (and expletive it) start lower in the clause? Maybe, but the rule of thumb is that it and there are not semantically selected by the verb. Expletive subjects are inserted to satisfy the requirement that every clause needs a subject, and so expletive subjects do not start in the verb phrase, they start in spec-TP.

VSO languages.

Many languages have a word order that suggests that the subject stays "low" in the clause. Consider Irish, which has canonical Verb-Subject-Object ordering for declaratives sentences.

[latex] \setcounter{ExNo}{19} \exg. Chonaic m\'e Eoghan\\ saw I Owen\\ \trans `I saw Owen.'\label{irish1} [/latex] These data make sense if V moves across the subject to T, and the subject stays low. Thus, they are most consistent with hypothesis #2. They are only consistent with hypothesis #1 with additional stipulations.

Locative inversion.

Here's some data from Zulu (from Buell 2007). In Zulu, locative phrases can be in spec-TP, triggering agreement on the verb; we see agreement between houses (noun class 10) and the subject prefix on stay in (21b).

[latex] \setcounter{ExNo}{20} \ex. \ag. Abantu abadala ba-hlala ku lezi zindlu. \\ 2people 2old 2-stay at 10these 10houses \\ \trans `Old people live in these houses.' \bg. Lezi zindlu zi-hlala abantu abadala. \\ 10these 10houses 10-stay 2people 2old \\ \trans `Old people live in these houses.' [/latex] Herero (Bantu) shows the same thing. (Data from Buell 2007 as well.) [latex] \setcounter{ExNo}{21} \ex. \ag. \`om\'uk\'az\'end\'u \'u-t\'er\'ek\`a \`ony\`am\`a p\`ong\`and\'a\\ 1woman 1-cook 9meat 16at.home \\ \trans `The woman cooks the meat at home.' \bg. p\`ong\`and\'a p\'e-t\'er\'ek\`a \`om\'uk\'az\'end\'u \`ony\`am\`a\\ 16at.home 16-cook 1woman 9meat\\ \trans `The woman cooks the meat at home.' [/latex] Notice in this pair that the English translation is the same for both sentences!

Locative inversion data are also consistent with Hypothesis #2, but are not consistent with Hypothesis #1. Under Hypothesis #2, what is happening with locative inversion is that something other than the "logical" subject gets to move to spec-TP, namely, the locative phrase. With Hypothesis #1, the word order doesn't seem possible: if the subject is always in spec-TP, how could we possible derive the word order where it appears after the verb?

Evidence for movement to spec-TP.

Quantifer float.

Some quantifiers can appear separated from the noun they combine with. This is called [pb_glossary id="788"]quantifier float[/pb_glossary].

[latex] \setcounter{ExNo}{22} \ex. \a. \textbf{All the children} will visit Santa. \hfill {\small{(the normal position)}} \b. \textbf{The children} will \textbf{all} visit Santa. \hfill {\small{(stranded! where?)}} \c. * \textbf{The children} will visit \textbf{all} Santa. \hfill {\small{(can't go here{\ldots})}} \d. * \textbf{The children} will visit Santa \textbf{all}. \hfill {\small{(can't go here{\ldots})}} [/latex]

The first two examples show us where all can go:

It can appear in front of the DP.
It can appear in front of the verb.

These data make sense if quantifiers like all can "float" in positions that the subject has occupied. Phrased differently, the data make sense if all can be "left behind" when the subject moves to spec-TP. All appears wherever the DP used to be. The ungrammatical sentences confirm this: we can't put all after the verb because the subject was never after the verb.

Voice aka "Little V"

In sum, the empirical evidence suggests that even subjects of transitive and unergative verbs start lower in the structure. The question we'll address here is where precisely the subject begins life. We said in chapter 8 that a verb like throw selects its subject. If I asked you what thematic role throw assigns to its subject, you'd probably say Agent. But consider the verb phrases in (24). All of these involve the verb throw, but the object varies. [latex] \setcounter{ExNo}{23} \ex. \a. throw a baseball \b. throw support behind someone \c. throw a match \d. throw a party \e. throw a fit \f. throw my back out \hfill (adapted from Marantz, 1984) \g. \ldots [/latex] Interestingly, we see that the thematic role of the subject depends not on the verb alone, but rather the combination of the verb and object. Throwing a baseball is not the same act as throwing a fit, or throwing my back out. In each case, the subject has a distinct thematic role, depending on both the verb and the object. The same thing can be demonstrated with the verb kill. Again, if I asked what thematic role kill assigns to the subject, you would probably say Agent. But now consider what happens when we vary the object. [latex] \setcounter{ExNo}{24} \ex. \a. kill a cockroach \b. kill a conversation \c. kill some time \d. kill an audience \e. \ldots [/latex] Again, it looks like it isn't just the verb, but the combination of verb and object that determines the semantic role of the subject. For instance, My feet are killing me, certainly cannot be attributing an Agent thematic role to my feet. Likewise, Stephen kills time by playing Switch does not attribute to Stephen the same thematic as in, Stephen killed a cockroach. The generalization is that selection of the subject depends on the entire VP, not just V. That is, the combination of verb and object together---the VP---determines the thematic properties of the subject. For our theory then, we then need a way to relate the VP to the subject. We will adopt the functional projection Voice to perform this role. [latex] \begin{tikzpicture}[mytree] \node(top) {TP} child {node {T$'$} child {node {T}} child {node {VoiceP} child {node(S2) {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} child {node {DP}} } } } } }; \node(S1) at (-3,-5) {the subject \\ at D-structure}; \draw[->, dashed] (S1) -- (S2); \end{tikzpicture} [/latex] Like other functional heads, Voice's job is to provide a relation between two phrases. Voice maps the subject to the VP---or more specifically, it relates the subject to the event described by the VP. It allows the VP to dictate the thematic properties of the subject. More generally, Voice controls the transitivity of the verb. We will assume for simplicity in this class that transitive and unergative verbs require Voice, but unaccusatives do not. This is, however, a significant simplification of the issues. [latex] \begin{minipage}{.33\textwidth} \textbf{Unaccusative schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VP} child {node {V$'$} child {node {V}} child {node {DP}} }; \end{tikzpicture} \end{minipage} \hspace{10pt} \begin{minipage}{.33\textwidth} \textbf{Unergative schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VoiceP} child {node {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} } } }; \end{tikzpicture} \end{minipage} \hspace{10pt} \begin{minipage}{.33\textwidth} \textbf{Transitive schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VoiceP} child {node {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} child {node {DP}} } } }; \end{tikzpicture} \end{minipage} [/latex]

Note that in many works, Voice is called v, pronounced "little v." We will not adopt this terminology, but be aware that Voice and v are often treated interchangeably.

Adding Voice requires a small modification to our T-lowering rule. We'll assume that instead of T lowering to V, it lowers to Voice, and that V raises to Voice. They "meet in the middle." (Nothing changes for unaccusatives; T still lowers to V.) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Britta}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T\\ build+$\emptyset$+$[$past$]$}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {DP \\ a house}} } } } } } } }; \draw[->] (Slow.west)-|(S); \draw[->, dashed] (T)|-(Voice.west); \draw[->, dashed] (V.west)-|(Voice); \end{tikzpicture} [/latex] This is basically a stipulation. For exposition's sake, we will not bother to justify the movement of V to Voice. It is probably necessary for theory-internal reasons (in English), though if V didn't move to Voice, nothing in our theory would be significantly affected.

[embed]https://youtu.be/GzaiEoRBh_k[/embed] Voice in trees (length: 2m 27s)

Key Takeaways

Voice always introduces the external argument. That is, when it's in the tree, it serves the function of bringing in the argument that does not start inside of the VP (as the complement to V). You need Voice above the VP if,

the verb is transitive, that is, has both a subject and an object, or
the verb is unergative.

The only times you don't have Voice above the VP is if,

the verb is unaccusative, that is, has a single argument which starts as the object of the verb, or
the verb is an auxiliary verb.

In other words, unaccusatives and auxiliary verbs do not select for an external argument, and so do not need Voice.

The Unergative/unaccusative distinction beyond verbs

We have focused above on detecting unaccusative and unergative verbal predicates. It has been argued, however, that the distinction extends to other categories, notably, adjectives. For instance Hans Bennis argues based on data from Dutch that the adjective duidelijk 'clear' is unaccusative, while the adjective trouw 'loyal' is unergative.[footnote]Technically, Bennis uses the term "ergative" to mean "unaccusative." This is terminological difference, and does not matter for his analysis.[/footnote] To prove this distinction requires a separate set of diagnostics from what we've introduced above, so we will put aside the alternation for adjectives here. However, we still need a place to generate the subject in a sentence like Mary is mean. Since mean selects for a subject, and assigns it a thematic role, Mary must be selected by mean. We will assume that it is introduced in the specifier of AP, and subsequently moves to spec-TP. This conclusion is consistent with everything we've said above: spec-TP is always empty at D-structure. [latex] \begin{tikzpicture}[mytree] \node(top){CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Mary}} child {node {T$'$} child {node(T) {T+V \\ $[$pres$]$+is}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {AdjP} child {node(tS) {$t_{DP}$}} child {node {Adj$'$} child {node {Adj \\ mean}} } } } } } } }; \draw[->, dashed] (V.west)-|(T); \draw[->] (tS.west)-|(S); \end{tikzpicture} [/latex]

Morphological evidence for Voice

The reasoning for Voice so far has been largely theory internal. When we look cross-linguistically, however, we find a great deal of evidence for a Voice head. For instance, consider the morphology of the following alternations in Kunuz Nubian (data from Abdel-Hafiz 1988).

Transitive and intransitive verbs in Kunuz Nubian
Intransitive	Transitive
dab 'disappear'	dabir 'disappear'
bokki 'hide'	bokkir 'hide'
wacci 'crack'	waccir 'crack'
bassi 'leak'	bassir 'leak'

What is the pattern? To make a transitive verb from an intransitive verb in Kunuz Nubian, you add the suffix -ir (with possibly vowel deletion with verb-final roots). Why is this meaningful? It shows that the difference between transitive and intransitive is not just the addition of a subject, but you actually add something to the verb.

Swahili also has a suffix -ish or -esh that does something similar.[footnote]The choice of vowel depends on the preceding vowels in the word. It is a case of "vowel harmony."[/footnote]

Transitive and intransitive verbs in Swahili
Intransitive	Transitive
-rudi 'return'	-rudisha 'return sth, bring sth. back'
-chelewa 'be late'	-chelewesha 'make late, delay'
-fa 'die'	-fisha 'destroy, cause to die'
-enda 'go, move'	-endesha 'make go, drive'
-weza 'be able'	-wezesha 'enable'

If there were no Voice head, then we couldn't explain why there is a systematic distinction between transitive/intransitive pairs. If we simply claim that the causative affix in Kunuz Nubian and Swahili are just how you pronounce Voice on transitive verbs, then the issue goes away. The following examples illustrate the phrasal movements and Voice. (Recall that in Swahili, V moves to T, so the head-movement goes up the tree, stopping at Voice on the way.) [latex] \setcounter{ExNo}{25} \exg. Sira a-me-chelewa\\ Sira 1\textsc{sm}-\textsc{perf}-be.late\\ \trans `Sira is late.' \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance =7em] child {node(S) {DP \\ Sira}} child {node {T$'$} child {node(T) {T+V \\ \textit{a-me-chelewa} \\ $[$perf$]$+be.late}} child {node {VP} child {node(V) {$t_{V}$}} child {node(t) {$t_{DP}$}} } } } }; \draw[->] (t)--++(south:3em)-|(S); \draw[->, dashed] (V.south)--++(south:1em)-|(T.south); \end{tikzpicture} [/latex] [latex] \setcounter{ExNo}{26} \exg. Juma a-me-chelew-esha Sira\\ Juma 1\textsc{sm}-\textsc{perf}-be.late-\textsc{caus} Sira\\ \trans `Juma made Sira late.' \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em, level distance =3em] child {node(S) {DP \\ Juma}} child {node {T$'$} child {node(T) {T+V+Voice \\ \textit{a-me-chelew-esha} \\ $[$perf$]$ + be.late + $\emptyset$}} child {node {VoiceP} child {node(t) {T$_{DP}$}} child {node {Voice$'$} child {node(Voice) {$t_{Voice}$}} child {node {VP} child {node(V) {$t_{V}$}} child {node {DP \\ Sira}} } } } } } }; \draw[->, dashed] (V)--++(south:1em)-|(Voice); \draw[->, dashed] (Voice.west)--++(west:1em)-|(T.south); \draw[->] (t)--++(south:1em)-|(S.south); \end{tikzpicture} [/latex]

Double-Object Constructions

We've been dealing exclusively with intransitive and transitive verbs, but we can also sketch a solution to transitive verbs, like give, which selects for two objects in (28), Margit and a book. [latex] \setcounter{ExNo}{27} \ex. Stacy gave Margit a book [/latex] This is called a double-object construction because the verb give appears to have two objects. We can say that the indirect object (Margit) is introduced as a specifier to VP, since this position in completely open! [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Stacy}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} [sibling distance = 6em] child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V+T \\ $\emptyset$+give+$[$past$]$}} child {node {VP} child {node {DP \\ Margit}} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {DP \\ a book}} } } } } } } }; \draw[->, dashed] (V)--++(south:1em)-|(Voice); \draw[->, dashed] (T)|-(Voice.west); \draw[->] (Slow.west)-|(S); \end{tikzpicture} [/latex]

Where we're going

In this chapter we introduced the idea that all subjects start lower than TP, and move there (at least in English and the languages we've looked at). This is phrasal movement, meaning the movement of phrases. Recall that the impetus for head-movement (the movement of heads) is, basically, word-formation. The suffix -ed in English just can't stand on its own; it has to be part of a word, and so either T moves to V, or V moves to T. So what is the impetus for phrasal movement? The answer to this question is much more complex. For the movements we've looked at in this chapter, we might say that the impetus for movement to spec-TP is simply that "every clause needs a subject." This idea can be syntactically encoded in the following constraint.

The Subject Condition Spec-TP must be filled at S-structure.

It's a stipulation, but it is "surface true," meaning that it appears to be a good generalization about English, and many other languages as well. There is something "special" about spec-TP in that it must be filled in the surface structure. Indeed, the idea that spec-TP "wants" to have something in it will lead us to consider other cases of movement to the subject position in the next chapter. As we'll discover in chapter 10, there are some movements that cannot be explained in the same way. It is an ongoing research question as to why things move.

Key Takeaways

identifying and representing unaccusative and unergative predicates
diagnostics for low position of (unergative/transitive) subjects
evidence for Voice
drawing trees representing movement to spec-TP

Exercises

VSO tree drawing

Here's some data from Niuean (from Woolford 1991). What do you observe about this data that bears on the question of whether the subject is low or high? (Notice the placement of the adverb!)

[latex] \setcounter{ExNo}{28} \ex. \ag. to lagomatai he ekekafo a ia\\ \textsc{fut} help \textsc{erg} doctor \textsc{abs} him\\ \trans `The doctor will help him.' \bg. Ne tutuli tumau he pusi e lapiti\\ \textsc{past} chase always \textsc{erg} cat \textsc{abs} rabbit\\ \trans `The cat always chased the rabbit.' [/latex]

Straight to spec-TP?

Consider again the data illustrating quantifier float in English, this time with the addition of the modal might. Think about the different positions that all can "float" (i.e. be left behind), and what this tells us about movement. How does the subject get to spec-TP? [latex] \setcounter{ExNo}{29} \ex. \a. \textbf{The children all} might have been visiting Santa. \b. \textbf{The children} might \textbf{all} have been visiting Santa. \c. \textbf{The children} might have \textbf{all} been visiting Santa. \d. \textbf{The children} might have been \textbf{all} visiting Santa. [/latex]

Raising versus Control

juliedoner — Wed, 15 Nov 2023 05:32:41 +0000

Learning Objectives

By the end of this chapter, you should,

be able to identify raising and control predicates using the diagnostics listed below
understand the structural difference between raising/control verbs, and be able to structural represent that difference
appreciate the arguments and evidence for PRO, and how it differs from pro.

We now know that spec-TP is empty at D-structure, and that subjects raise to spec-TP from where they are selected. This chapter addresses further issues in phrasal movement. We will mainly be concerned with the similarities and differences between the following sentences.

[latex] \ex. \a. Khufu seems to run fast. \b. Khufu wants to run fast. [/latex] The sentences of course mean different things. But they consist of the same linear sequence of words. A subject (Khufu), a finite verb (seems/wants), a nonfinite verb (to run) and an adverb (fast). The sole difference is the finite verb. In (1a), the verb is seems and in (1b) the verb is wants. Despite the surface similarities, we'll discover in this chapter that the two sentences have wildly different syntactic structures, all of which are related to the following core difference:

Want selects for a subject, seem does not.

That is, want c-selects for a subject and assigns it a thematic role. Khufu in (1b) is an argument of want and gets the thematic role Experiencer. Seem on the other does not select for a subject and does not assign it a thematic role. Instead, in (1a), Khufu is selected by the infinitive and "raises" into the main clause from the lower clause. The difference between the two verbs is sketched below. We'll provide a more precise analysis later. The important thing to keep in mind is the starting place of the subject. For verbs like seem, the subject starts inside of the lower clause. For verbs like want, the subject starts outside of the lower clause. [latex] \setcounter{ExNo}{1} \tikzstyle{every picture}+=[remember picture, inner sep=0pt, baseline, anchor=base]% \ex.{}[\textsubscript{TP} \tikz\node(high){Khufu}; [ seems [ to [ \tikz\node(t){$t$}; run fast ] ] ] ] \begin{tikzpicture}[overlay] \draw[->] (t.south) -- ++(south:1.5ex) -| (high.south); \end{tikzpicture} \tikzstyle{every picture}+=[remember picture, inner sep=0pt, baseline, anchor=base]% \ex. {}[\textsubscript{TP} \tikz\node(high2){Khufu}; [ \tikz\node(t2){$t$}; [ wants [ to [ run fast ] ] ] ] ] \begin{tikzpicture}[overlay] \draw[->] (t2.south) -- ++(south:1.5ex) -| (high2.south); \end{tikzpicture} [/latex] We call predicates like seem [pb_glossary id="790"]raising[/pb_glossary] verbs, and we call predicates like want [pb_glossary id="791"]control[/pb_glossary] verbs. All of the following diagnostics illustrate this same difference: that the starting place of the subject of a raising verb is different than the starting place of the subject of a control verb. Note that while the data is given in English, these diagnostics have cross-linguistic validity. All else equal, you can use these diagnostics to determine whether a verb embedding a nonfinite clause belongs to the class of raising or control verbs.

Raising vs. control

All of the following diagnostics allow you to determine if something is a raising or a control verb. To apply the diagnostic, you mechanically use the data given.

Expletives subjects I.

Raising verbs often permit an alternation between a referential subject and an [pb_glossary id="787"]expletive[/pb_glossary] subject. Control verbs don't. [latex] \setcounter{ExNo}{3} \ex. \a. It seems that Khufu runs fast. \b. *It wants that Khufu runs fast. [/latex] How do we know that it in (4a) is an expletive? Well, it doesn't appear to have semantic content. It doesn't refer to anything. For instance, you can't ask the question, *What seems that Khufu runs fast? These data support the idea that seem doesn't s-select for a subject, because that position can be filled by a semantically "vacuous" element. Note that many languages don't (obviously) have expletive subjects. For instance, in pro-drop languages like Spanish and Swahili, the subject position is obligatorily empty. [latex] \setcounter{ExNo}{4} \exg. (*eso) parece que Mar\'ia vio un \'arbol\\ it seem.\textsc{pres} that Mar\'ia see.\textsc{past} a tree\\ \trans `It seems that Maria saw a tree.' \exg. (*hii) i-na-onekana kwamba Beni a-na-penda magari\\ this 9\textsc{sm}-\textsc{pres}-seem that Beni 1\textsc{sm}-\textsc{pres}-like cars\\ \trans `It seems that Beni likes cars.' [/latex] Again, this supports the idea that such sentences involve a semantically empty subject, since it cannot be referred to using an overt pronoun.

How to apply the expletive subject test

Start with you original sentence → Khufu seems to run fast
Replace the subject with an expletive, and make the lower clause finite → It seems that Khufu runs fast.
Check the grammaticality. If it's grammatical, the verb is a raising verb.

[embed]https://youtu.be/yNBizEZwPoQ[/embed] Expletives I (length: 55s) credit: Keira Dobbs

It is worth asking why there are expletive subjects at all. They don't appear to serve a semantic purpose---that is, they don't mean anything. So why do expletive subjects exist? This is a complex question, and there are many ways to answer it. But one thing we can say is that, though expletive subjects do not serve a semantic function, they do serve a syntactic function, namely, the [pb_glossary id="795"]Subject Condition[/pb_glossary]. If there is nothing else that can filled spec-TP, then an expletive subject is inserted. The following are examples of other types of expletive subjects in English. [latex] \setcounter{ExNo}{6} \ex. \a. It's raining. \b. It's hot outside. \c. It was a dark and stormy night. [/latex]

Idiom chunks.

Some phrases are "non-compositional" in that the meaning isn't predictable from the parts. These are called [pb_glossary id="789"]idioms[/pb_glossary] or idiom chunks.

[latex] \setcounter{ExNo}{7} \ex. \a. The cat is out of the bag. $\longrightarrow$ ``The secret is known." \b. The shit hit the fan. $\longrightarrow$ ``There's a problem." [/latex]

We find a distinction between raising and control verbs with respect to idiom chunks.

[latex] \setcounter{ExNo}{8} \ex. \a. The cat seems to be out of the bag. \b. ??The cat wants to be out of the bag. \ex. \a. The shit seems to have hit the fan. \b. ??The shit wants to have hit the fan. [/latex] What we observe is that with raising verbs, the idiom meaning is retained. With control verbs it isn't. The a. examples sound acceptable because we can interpret the entire idiom chunk together. The control verbs lose the idiomatic meaning because the subject is actually a subject of the higher verb. Note that this diagnostic is concerned with interpretation. If we get the non-literal (idiomatic) reading after inserting the verb – as we do in (9a) and (10a) – then we conclude that the subject must have raised from the lower clause, as this is where the rest of the material from the idiomatic expression resides. This same idea explains why the idiomatic meaning is not retained with control verbs; the subject the cat in (9b) can only be interpreted as an actual animal that wants out of a bag (literal reading) because it was never in the lower clause, and therefore cannot be interpreted with the idiomatic material.

Applying the idiom chunk test

Choose your favorite idiom → The shit hit the fan.
Take the predicate you want to test (seem) and put it in the idiom → The shit seems to have hit the fan
Check whether the idiom meaning is retained. You're not looking for (un)grammaticality, rather you're looking for whether the idiom meaning survives. If the idiom meaning survives, then it's a raising verb.

[embed]https://youtu.be/VpVjkb-VTc4[/embed] Idiom test (length: 2m 49s) credit: Alesha Pisciotta [embed]https://youtu.be/6uB_ERNvhxA[/embed] Idiom test (length: 1m 1s) credit: Keira Dobbs

Expletive subjects II.

English has another expletive subject, there. It's only possible to use expletive there with a particular class of verbs, namely, verbs which assert existence, like the copula be.

[latex] \setcounter{ExNo}{10} \ex. \a. There is a book on the table. \b. There are three women in the room. [/latex] Again we see a distinction between raising and control verbs with respect to expletive there. [latex] \setcounter{ExNo}{11} \ex. \a. There seems to be a book on the table. \b. *There wants to be a book on the table. [/latex] The explanation behind this data goes as follows. Expletive there must be selected for by a particular predicate. Its appearance in (12a) follows because it has been selected for by the infinitive be in the lower clause. The reason it cannot appear in (12b) is that want cannot select for there.

Applying the expletive subject II test

Take an existential sentence → There is a book on the table.
Put the predicate you're testing (seem) into the sentence → There seems to be a book on the table.
Check the grammaticality. If it's grammatical, then it's a raising verb.

[embed]https://youtu.be/NDjYdycsRao[/embed] Expletive subjects II (length: 1m 22s) Credit: Megan Spurgeon

Scope ambiguity.

This one is a bit more complex, as it presupposes a bit of semantics. This test asks us to consider how quantified expressions are interpreted in Raising and Control structures. A quantified expression is any expression that asserts a "quantity" of a set of individuals.[footnote]In fact, quantifiers can apply to really any number of things beyond individuals. For instance, there are quantifiers over times (sometimes) and even over "worlds" (believe). For more discussion of quantifiers, I recommend taking Semantics and Syntax II. [/footnote] As example of quantified expressions, consider the sentence Every student met some professor. This sentence is ambiguous with respect to how the quantified expressions every student and some professor interact. On one reading (called the "surface scope" reading), we are talking about the list of students. As long as each student has met a professor, then the sentence is true. On this reading, the professors could be different. Timmy met Prof. Smith. Susie met Prof. Jones. Rachel met Prof. Scott. Etc.

On an alternate reading of this sentence, we are talking about a particular professor. On this reading, the sentence is true if every student met the same professor. So Timmy met Prof. Smith and Susie met Prof. Smith and Rachel met Prof. Smith. The difference between the two readings is the scope of some professor. On the first reading, some professor scope "under" the every student. On the second reading some professor scopes "above" every professor.

Now consider the following sentence with a Raising predicate, be likely.

[latex] \setcounter{ExNo}{12} \ex. Someone from Florida is likely to win the lottery. \a. $\approx$ ``There is a particular person from Florida who is likely to win the lottery (her name is Cheryl).'' \b. $\approx$ ``It's likely is that someone or other in Florida will win the lottery (but we don't know who).'' [/latex] Notice that reading in (13b) is paraphrased as if someone were in an embedded clause. This is the reading we get when we interpret the quantified expression someone lower than where it's pronounced. Crucially, this same ambiguity doesn't seem to exist with want. [latex] \setcounter{ExNo}{13} \ex. Someone from Florida wants to win the lottery. \a. $\approx$ ``There is a particular person from Florida who wants to win the lottery (her name is Cheryl).'' \b. $\approx$ ``??What is wanted is that someone or other in Florida win the lottery (but we don't know who).'' [/latex] The sentence Someone from Florida wants to win the lottery is not ambiguous in the same way. It does not have a reading in which someone is interpreted lower than the position that it is pronounced.

Applying the scope ambiguity test

Choose a sentence with an indefinite subject → Someone from Florida won the lottery
Put the predicate you're testing in the sentence → Someone from Florida seems to have won the lottery
Determine if the resulting meaning is equivalent to having an expletive subject (see expletive subject I test above) → ≈ It seems like someone from Florida won the lottery." If it is, then it's a raising verb.

Truth changing passive.

If you passivize the infinitive under a raising verb, it basically means the same as the active version. If you passivize the infinitive under a control structure, it means something very different.

[latex] \setcounter{ExNo}{14} \ex. \a. The doctor seemed to examine Mary $\approx$ Mary seemed to be examined by the doctor. \b. The doctor wanted to examine Mary $\not\approx$ Mary wanted to be examined by the doctor [/latex] Here again the explanation is that the subject of seem "inherits" whatever thematic role is assigned to the subject of the passivized verb. But the subject of want gets its own thematic role.

Applying the passivization test

Choose a sentence with a transitive verb → The doctor examined Mary.
Put the predicate you're testing (seem) in the sentence → The doctor seems to have examined Mary
Passivize your original sentence → Mary was examined by the doctor
Put the predicate you're testing (seem) in the sentence → Mary seems to have been examined by the doctor
Check whether the two sentences you've created meaning the same thing → The doctor seems to have examined Mary = Mary seems to have been examined by the doctor. If the sentences are interchangeable, then it's a raising verb.

[embed]https://youtu.be/WH82x4xtOos[/embed] Passivization (Alesha Pisciotta; length: 3m 20s)

The final two tests below pick out only a subset of control verbs. Thus, if they pass, then you can confidently conclude that you have a control predicate. If they fail, you can conclude nothing.

For-CPs

Some control predicates allow both subject positions to be explicitly filled. The subject of the infinitive appears in front of the complementizer for.

[latex] \setcounter{ExNo}{15} \ex. \a. *John seemed for Mary to read the book. \b. John wished for Mary to leave. [/latex]

Applying the for-CP test

Make a sentence with the predicate you're testing (wish) → Chi wishes to leave
Add a subject in the lower clause after for → Chi wishes for Drey to leave.
Check the grammaticality. If the sentence is grammatical, then it's a control verb.
If the sentence is not grammatical, then you can conclude nothing.

[embed]https://youtu.be/9QzYUes4TkQ[/embed] For-CPs (length: 1m 26s) credit: Morgan Robertson

Partial control.

Some control predicates allow a "mismatch" in semantic features. Sometimes the lower subject can refer to "more" than the higher subject. For instance, a verb like gather requires a plural subject.

[latex] \setcounter{ExNo}{16} \ex. \a. *The student gathered in the park. \b. The students gathered in the park. [/latex] With certain control predicates, the subject of the higher verb can be singular, even if the infinitive selects for a plural subject. This type of a "mismatch" simply isn't permitted with any raising verb. [latex] \setcounter{ExNo}{17} \ex. \a. *The student seemed to gather in the park. \b. The student wanted to gather in the park. [/latex]

Applying the partial control test

Make a sentence with a verb that requires a plural subject → The students gathered in the park.
Put the predicate you're testing (want) into the sentence, but make the subject singular → The student wants to gather in the park
Check the grammaticality

[embed]https://youtu.be/3MdDn68Sw2Q[/embed] Partial control (length: 1m 51s) credit: Morgan Robertson

Analyzing raising structures

There are a number of different predicates that share all the raising properties. Note that the term "predicate" includes verbs, adjectives, and even prepositions. The following are standardly recognized raising predicates; this is not an exhaustive list.

seem (Khufu seems to run fast.)
appear (Khufu appears to run fast.)
be likely (Khufu is likely to run fast.)
be about (Khufu is about to run fast.)
...

The analysis of raising is fairly straightforward. The subject moves from the lower clause into spec-TP of the higher clause. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ $[$pres$]$}} child {node {VP} child {node {V$'$} child {node {V \\ seem}} child {node {TP} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {Voice$'$} child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node {V \\ run}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=2em] child {coordinate(left1)} child {node(high) {Khufu} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(V) {V+T \\ seem+$[$pres$]$}} child {node {TP} child {node(mid) {$t_{DP}$}} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node(low) {$t_{DP}$}} child {node {Voice$'$} child {node(voice2) {Voice$+$V \\ $\emptyset$+run}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node(v2) {$t_V$}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } }; \draw[->, rounded corners=2pt] (low)--++(west:2)-|(mid); \draw[->, rounded corners=2pt] (mid)--++(west:2)-|(high); \draw[>-, round corners=2pt, dashed] (V)--++(west:1)-| ++(west:.5)--(T);\draw (left1)--(right1); \draw[->, round corners-2pt, dashed] (v2)--++(west:1)-|(voice2); \draw (left2)--(right2); \end{tikzpicture} [/latex] Let's break down the important things to note about this analysis.

There is [pb_glossary id="793"]cyclic movement[/pb_glossary]. The subject moves through the specifier of TP in the lower clause. The reason that this movement is necessary is that it satisfies the Subject Condition: Spec-TP must be filled at S-structure.
The TP dominating the infinitive doesn't have a CP. (This is the only time in this class we'll see a TP without a CP.) The reason that there is no CP here is that we cannot detect a CP here: we cannot use any kind of non-finite complementizer (e.g. for) with raising predicates.
There is no VoiceP in the main clause. This is because there is no external argument: seem doesn't select for a subject.

In a sense, raising verbs present a particularly interesting and complicated study in long-distance dependencies in syntax. Building on the idea in the previous chapter that phrases can move to the subject position, raising verbs illustrate that this movement can cross clauses sometimes.

Be careful! The following three things are not the same. Do not confuse them.

Raising verb: A raising verb is a verb which raises a subject out of a lower clause. It involves phrasal movement. Examples of raising verbs in English are seem and appear.
Verb raising: Verb raising is when the head V raises to T (or some other higher functional projection). It is head-movement. In English, only the auxiliaries be and have undergo raising.
Raising to subject: Raising to subject describes when a phrase moves to the subject position. This can be clause-internal---movement from the verb phrase to spec-TP, or cross-clausal: movement across a raising verb.

Analyzing control structures

It is temping to simply treat control verbs like want analogous to raising verbs, except that the subject starts in the main clause. This is shown in the tree below. (This is the D-structure.)

The selectional requirements of run haven't been met. (Violates [pb_glossary id="794"]selection[/pb_glossary].)
Spec-TP of the lower clause isn't filled. (Violates the [pb_glossary id="795"]Subject Condition[/pb_glossary].)

Generally, if we allow trees like that above, then we need to stipulate that sometimes verbs don't select, and sometimes spec-TP doesn't have a specifier. This is a bitter pill because it really weakens the overall theory we've built.

Instead, to satisfy all of our constraints and keep the theoretical principles we've established, we are forced to posit a silent argument. We need something to be the subject of the lower clause. We call this silent argument PRO, pronounced "big pro." PRO is a pronoun, but it's a special pronoun. It only appears in particular configurations in order to satisfy grammatical constraints.

[latex] \begin{tikzpicture}[mytree, scale=.8] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ $[$pres$]$}} child {node {VoiceP} [sibling distance = 7em] child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {Voice$'$} [sibling distance = 6em] child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V \\ want}} child {node {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node {DP} child {node {D$'$} child {node {D \\ PRO}} } } child {node {Voice$'$} child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node {V \\ run}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree, scale=.8] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node(high) {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {T$'$} child {node(T) {$t_T$}} child {node {VoiceP} child {node(low) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T \\ want+$\emptyset$+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(V) {$t_V$}} child {node {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} child {node {D$'$} child {node(PROhigh) {D \\ PRO}} } } child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node(PROlow) {$t_{DP}$}} child {node {Voice$'$} child {node(voice2) {Voice$+$V \\ $\emptyset$+run}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node(v2) {$t_V$}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \draw[->, rounded corners=2pt] (PROlow)--++(south:2)-|(PROhigh); \draw[->, rounded corners=2pt] (low)--++(west:2)-|(high); \draw[->, rounded corners=2pt, dashed] (v2)--++(west:1)-|(voice2); \draw[->, dashed] (T)|-(Voice); \draw[->, dashed] (V)-|(Voice); \end{tikzpicture} [/latex] PRO solves our problems because it's a DP, and it is merged in spec-VoiceP and moves to spec-TP, just like any other subject. It's "special" because it only ever appears in nonfinite clauses.

The idea behind PRO is that it acts kind of like an anaphor. Its reference must be fixed by some c-commanding expression, i.e., it has to be [pb_glossary id="797"]bound[/pb_glossary]. In all the previous examples, PRO is "controlled" by the subject.

Note though that PRO isn't actually an anaphor. PRO obeys its own rules, and has its own distribution. For instance, PRO can be "arbitrary," as in the following sentences.

[latex] \setcounter{ExNo}{18} \ex. \a. It's important to exercise. \b. To exercise is important. [/latex] The verb exercise needs a subject, so PRO must be present in these structures. But it gets an "arbitrary" reference, meaning something like "we all" or "someone generic."

Confusingly, PRO (or "big PRO") is different from pro (or "little pro").

pro is a pronoun that appears in pro-drop languages (like Swahili or Spanish). It's just a regular pronoun, but it doesn't have phonological content.
PRO only appears in one very particular position: spec-TP of non-finite clauses. It's a pronoun, but it's a special kind of pronoun.

English doesn't have pro, but it has PRO. Swahili has both pro and PRO (but they're in complementary distribution).

Evidence for PRO

You may be skeptical. Postulating the existence of something that is never pronounced and only appears in very particular circumstances should make you wary. This is good! You are acting like a scientist. PRO (or in general, the phenomenon of control of PRO) is a hypothesis. Our theory predicts that such an element should exist --- that is, that there should be something "silent" in the syntax. So we'd like to find empirical evidence for PRO.

In fact, we find a lot of empirical evidence for PRO! This is hypothesis-driven research at its finest. Based on independent principles, our theory forces us to postulate something that seems odd. But then we find evidence for it. Here's one piece of evidence from English, which utilizes what you already know about Binding Theory.

Binding theory and PRO.

The following sentences contain anaphors. Based on on Binding Theory ([pb_glossary id="835"]Condition A[/pb_glossary]), an anaphor must be bound. [latex] \setcounter{ExNo}{19} \ex. \a. It's important to trust oneself/yourself. \b. It's useless to nominate oneself/yourself for president. [/latex] If PRO didn't exist, how would the anaphor satisfy Condition A?

Where we're going

The difference between raising and control is widely debated in the current theoretical landscape. The complexities of control in particular continue to be challenge to theoretical models of syntax. (I recommend Syntax II, LING 526, for a deeper investigation into the wonders of PRO.) There is a larger issue though. Often, we come across structures in languages which we don't know how to analyze straight-away. Take for instance the sentence in (21). [latex] \setcounter{ExNo}{20} \ex. It took John an hour to get to work. [/latex] What is the relationship between John and the predicate get to work? Does John move to its surface position? Or is it generated there? This is a question of raising vs. control---which is really the question of Where do arguments start? At this point, we've introduced the main issues with one kind of phrasal movement: A-movement. We will continue now with the other kind of phrasal movement: A'-movement. As we'll see, the two movements "look" different and also "act" different. In my opinion, A-movement is the "trickier" of the two kinds of phrasal movement. A'-movement is quite straightforward to diagnose---usually.

Exercises

Phrasal movement: A'-movement

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Learning Objectives

By the end of this chapter, you should be able to,

recognize and represent A'-movement
determine instances of A'-movement
understand the difference between A- and A'-movement

In the last two chapters, we've been exploring the movement of phrases. This involved substantial discussion of argument structure, and in particular, the position of subjects.

In this chapter, we'll explore the second side of phrasal movement. The cases we'll look at below are in some ways more "obvious" examples of movement. We will clearly see that some constituent is being pronounced in a place where we do not expect it to be pronounced. Some examples of the movement that we'll be looking at here are given below.

[latex] \ex. \a. Who did Tamara sue? \hfill \textit{Wh-}question \b. What will Emma buy? \hfill \textit{Wh-}question \c. This book, I read. \hfill Topicalization [/latex]

As we can see in the above examples, some pre-sentential element corresponds to some "gap" later in the sentence. In all three cases, the first noun phrase corresponds to the object, which we expect to be pronounced after the verb. It is more obvious in (1) that something has moved, because not only is there something in a position that it shouldn't be, but there's also something missing after the verb where the object should be. We generally refer to this empty space as a "gap."[footnote]In this book, the terms "gap" and "trace" are functionally interchangeable. Though it should be noted that "gap" is not a formal term.[/footnote]

This kind of movement is clearly phrasal movement, as it can involve moving more than just one word (Which book will Emma buy?). Crucially however, this type of movement does not affect argument structure; who in (1a) is the object the verb, wherever we move it. Or stated differently, this type of movement can doesn't change the grammatical role of the element that has moved. The movements in the previous chapter are different: all the phrasal movements we've previously considered "created" subjects. For instance, moving the object of unaccusative V turned this object into a subject.

One of the main findings of the last few decades is that phrasal movement comes in two varieties. We looked in the last two chapters at A-movement. The examples above are called A'-movement (pronounced "A-bar movement").

A-movement : Movement to an argument position. A'-movement : Movement to a non-argument position.

The two kinds of movement have distinct, identifiable characteristics. Generally, A-movement "creates" subjects (and objects) by moving them to the "subject (and object) positions." For example, the A-movement cases we've seen have all involved movement to spec-TP, which is the "subject position" in English (and many many languages). In contrast, A'-movement does not "create" subjects and objects, it merely moves them around. In the example in (1c) above, the book is an object of read, but it's been moved to the front of the sentence. The movement to the front of the sentence didn't affect the "objecthood" of the book; it was for other reasons. So this is the core distinction between A- and A'-movement. That said, both kinds of movement also have a "signature," i.e., a core set of properties that we can use to diagnose whether something has moved via A- or A'-movement.

The landing site of A'-movement

As the above examples demonstrate, things can be moved to a pre-sentential position (meaning a position in front of the sentence). What is that position?The only phrase above TP is CP, therefore, this movement must target (i.e., land in) the specifier of CP. So [pb_glossary id="818"]wh-movement[/pb_glossary] lands in spec-CP.

[latex] \begin{tikzpicture}[mytree] \node(top) {CP} [sibling distance = 5em] child {node(wh) {DP \\ who}} child {node {C$'$} child {node {C$_{q}$+T \\ $\emptyset$+$[$past$]$}} child {node {TP} [sibling distance = 3em] child {coordinate(left)} child {node(bot) {Tarmara sue $t$} edge from parent[draw=none]} child {coordinate(right)} } }; \draw (left)--(right); \draw[->, rounded corners, red] ([xshift=2.7em]bot.south)--++(south:2ex) -| (wh.south); \end{tikzpicture} [/latex]

In general, the strategy for wh-questions in English is: "Move the wh-word to the front of the sentence, and move T to C.'' Notice that since this is a [pb_glossary id="819"]question C[/pb_glossary] (C_q), we expect to see T-to-C movement, just like with [pb_glossary id="820"]polar questions[/pb_glossary] (i.e., Did John see Mary?).

One tricky thing about wh-question formation in English involves subjects (again!). Consider the subject wh-questions below. [latex] \setcounter{ExNo}{1} \ex. \a. Who will talk to Brian? \b. What has happened? [/latex] Although it looks like there's no movement, such examples are in fact perfectly compatible with the description above. Some of the movements here are "string vacuous," meaning that they do not affect the linear order of the constituents. In the tree below, A-movement is represented in black, while A'-movement is represented in red. The idea here is that since we expect T-to-C movement to apply in question formation, the subject wh-word must move to spec-CP to yield the observed linear order. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node(whtop) {DP \\ who}} child {node {C$'$} child {node(C) {C$_{q}$+T \\ $\emptyset$+will}} child {node {TP} child {node(S) {$t_{DP}$}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Sbottom) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V \\ $\emptyset$+talk}} child {node {VP} child {node {V$'$} child {node(V) {$t_V$}} child {node {PP} [sibling distance=2em] child {coordinate(left)} child {node {to Brian} edge from parent[draw=none]} child {coordinate(right)} } } } } } } } }; \draw (left)--(right); \draw[->, rounded corners] (Sbottom)--++(south:1em)-|(S); \draw[->, red, rounded corners] (S.west)--++(west:2em)-|(whtop); \draw[->, dashed] (T.west)--++(west:2em)-|(C); \draw[->, dashed] (V.west)-|(Voice); \end{tikzpicture} [/latex] As usual, things get more complicated when do-support comes into the picture. The problem can be illustrated in the following sentence. [latex] \setcounter{ExNo}{2} \ex. Who called? [/latex] We expect to see something like *Who did call? This is because T moves to C in questions, and therefore shouldn't be able to appear on the verb. There are a couple of solutions to the puzzle --- but they all involve simply outright stipulating the solution.

For our purposes, we'll assume that T-to-C movement applies, and then a "magical" (=morphological) process allows the complex to come together with V after syntax.

Many sentences will now have at least two phrasal movements. There will almost always be something moving to spec-TP (unless there is an expletive subject), and then there might be something moving to spec-CP. There is a general and robustly attested "ordering" of movements. A-movement always precedes A'-movement. Practically, this means that you should move to spec-TP before you move anything to spec-CP.

[embed]https://www.youtube.com/embed/7Ucj7OZ8STs[/embed] Subject wh-question (length: 3m 5s)

Pied-Piping

Often, more than just the wh-element moves. This is called pied-piping.

[latex] \setcounter{ExNo}{3} \ex. \a. Which picture did John see? \b. How big is the box? \c. Whose sister did John visit? [/latex]

In most cases, there's a very intuitive explanation for why pied-piping happens: you simply can't separate some things. For instance, can we ever separate a determiner and it's corresponding NP? Likewise, how could we get whose to move, since it's not even a constituent?

[latex] \begin{tikzpicture}[mytree] \node(top) {CP} [sibling distance = 8em] child {node(wh) {DP} [sibling distance = 5em] child {node {D$'$} child {node {D \\ which}} child {node {NP} child {node {N$'$} child {node {N \\ picture}} } } } } child {node {C$'$} [sibling distance = 5em] child {node(C) {C$_{q}$+T \\ $\emptyset$+$[$past$]$}} child {node {TP} child {node(S) {DP} [sibling distance = 1em] child {coordinate(left1)} child {node {John}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V \\ $\emptyset$+see}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node(O) {$t_{DP}$}} } } } } } } }; \draw (left1)--(right1); \draw[->, red, rounded corners] (O)--++(south:2em)--++(west:27em)|-(wh.west); \draw[->, dashed, rounded corners] (V)--++(west:1em)-|(Voice.south); \draw[->, rounded corners] (Slow)-|([yshift=-3em]S.south); \draw[->, dashed, rounded corners] (T)--++(south:2em)-|(C.south); \end{tikzpicture} [/latex]

Evidence for movement

As always, we want to support our analysis with data. It's not enough to say, "Look, the wh-word is at the front of the sentence, it must have moved there." We want to show empirically that movement has occurred.

Binding.

Binding Theory and phrasal movement interact. The following sentences illustrate that movement must have taken place.

[latex] \setcounter{ExNo}{4} \ex. \a. Which picture of himself$_{i}$ did John$_{i}$ see? \b. Which picture of John$_{i}$ did he$_{i}$ see? [/latex] In (5a), we have an anaphor. Recall that anaphors are subject to [pb_glossary id="835"]Condition A[/pb_glossary] of Binding Theory. In order for that to happen, at some point himself must have been bound by John. That must have happened before movement. On the other hand, in (5b), we have an R-expression, which is subject to [pb_glossary id="779"]Condition C[/pb_glossary] of Binding Theory. Here, the movement appears to "rescue" the sentence, since if picture of John were pronounced as the object of the verb, it would be ungrammatical: *He_i saw the picture of John_i.

Languages with overt C.

Some languages actually pronounce the complementizer, like Irish:[footnote]There are a few different complementizers in Irish, which interact with A'-movement in interesting ways. The superscript "L" indicates one type of complementizer. [/footnote]

[latex] \setcounter{ExNo}{5} \exg. Cad a\textsuperscript{L} t\'a sa seomra\\ what \textsc{C} is in.the room\\ \trans `What is in the room? [/latex] After the wh-word cad, the complementizer a^Lappears, showing that the wh-word comes before (=above) C.

Wanna contraction.

There are strong intuitions about when it's possible to contract want+to into wanna in English.

[latex] \setcounter{ExNo}{6} \ex. \a. Does John wanna talk to Mary? \b. Who does John wanna talk to? \c. *Who does John wanna talk to Mary? [/latex] This pattern suggests that you can contract wanna in (7b) because the wh-word who started at the end of the sentence. In (7c) on the other hand, the gap is the subject position of the nonfinite clause: Who does John want __ to talk to Mary? We can explain this ungrammaticality if we simply say that you can't contract across traces.

Other instances of a'-movement

Embedded questions

Questions can be "embedded" under certain verbs: [latex] \setcounter{ExNo}{7} \ex. \a. Bill knows [ who saw Mary ] \b. Susan wondered [ how Mary fixed the car ] \c. Sarah decided [ which book she should buy ] [/latex]

Embedded questions are formed exactly like regular questions, except that there's no T-to-C movement in the lower clause.

Relative clauses.

Another place where we see A'-movement, in particular wh-movement, is relative clauses. [latex] \setcounter{ExNo}{8} \ex. \a. the woman [ who I saw {\underline{\hspace{10pt}}} ] is smiling \b. the man [ who {\underline{\hspace{10pt}}} laughs ] arrived. \c. I licked the fork [ which I eat with {\underline{\hspace{10pt}}} ] [/latex] Relative clauses (in English) involve the following properties:

There's a gap (a trace) inside of a CP
The CP is adjoined rightward to NP
There is A'-movement
There is no T-to-C movement inside of the relative clause.

The following is the representation of The woman who I saw was happy.

[latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node(S) {DP} [sibling distance = 5em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N \\ woman}} } child {node {CP} child {node {DP} child {node {D$'$} child {node(ophigh) {D \\ who}} } } child {node {C$'$} child {node {C$_{RC}$ \\ $\emptyset$}} child {node {TP} child {node(S2) {DP \\ I}} child {node {T$'$} child {node(T2) {$t_{T}$}} child {node {VoiceP} child {node(S2low) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T \\ see+$\emptyset$+$[$past$]$}} child {node {VP} child {node(V2) {$t_V$}} child {node(t) {$t_{DP}$}} } } } } } } } } } } } child {node {T$'$} [sibling distance = 5em] child {node(T) {T+V \\ $[$pres$]$+\textsc{be}}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(voi) {Voice+V \\ $\emptyset$+smiling}} child {node {VP} child {node {V$'$} child {node(V2a) {$t_V$}} } } } } } } } } }; \draw[->, dashed] (V2a.west)-|(voi.south); \draw[->, dashed] (V)-|(T); \draw[->, dashed] (T2)--++ (south:1em)|-(Voice.west); \draw[->] (S2low) --++(west:3em)-|(S2); \draw[->, red, rounded corners] (t)--++(south:2em)-|(ophigh); \draw[->, dashed] (V2.west)-|(Voice.south); \draw[->] (Slow.west)..controls +(west:2) and +(east:1)..(S); \end{tikzpicture} [/latex]

This tree looks much more complicated than it actually is. The trick here is realizing that there are two clauses. If I take away the constituent who I saw, I would just have the sentence, The woman is smiling, and we know how to draw that tree: the subject the woman is selected by smile and so starts in spec-VoiceP and then moves to spec-TP. The relative clause is entirely independent of this process. You build the relative clause and attach it to the NP as a modifier.

In English, there is a relatively large amount of variation in how you pronounce a relative clause. The following sentences all have identical structures.

[latex] \setcounter{ExNo}{9} \ex. \a. The woman [ who I saw {\underline{\hspace{10pt}}} ] is smiling. \b. The woman [ that I saw {\underline{\hspace{10pt}}} ] is smling. \c. The woman [ I saw {\underline{\hspace{10pt}}} ] is smiling. [/latex]

In all three cases, there has been wh-movement of who to spec-CP of the relative clause. In the first case, we pronounce who. In the second case, we don't pronounce who, but we pronounce the complementizer that. And in the third case, we pronounce neither. Since this doesn't appear to affect the meaning at all, we assume that the structure is identical in all cases.

In English, there appears to be a rule of pronunciation that says that you cannot pronounce both a wh-word and a complementizer at the same time. This has been called the Doubly-Filled Comp Filter. (Note that there is in fact dialectal variation with respect the Doubly Filled Comp filter. In some dialects, it's actually acceptable to pronounce both who and that.)

Relative clauses should not be confused with complex-NPs. The example in (11a) is a relative clause. The example in (11b) is a complex-NP. [latex] \setcounter{ExNo}{10} \ex. \a. The lie [ that Mary told {\underline{\hspace{10pt}}} ] was bad. \b. The lie [ that Mary went to Moscow ] was bad. [/latex] Relative clauses have gaps, which we analyze as movement, while complex-NPs do not.

Cross-linguistic variation in A'-movement

Not all languages display wh-movement to spec-CP. For instance, Swahili has wh-in situ.

[latex] \setcounter{ExNo}{11} \exg. Duma a-li-ona nani\\ Duma \textsc{3sg}-\textsc{past}-see who\\ \trans `Who did Duma see?' [/latex] The typological variation in (non-)movement of wh-expressions is fascinating, and it has inspired a huge amount of literature. It is a major focus of Syntax II, LING 526.

Differences between A- and A'-movement

Generally, syntax can be viewed as the investigation of silence. Many of the core syntactic problems arise in places where we expect to see something pronounced, but instead we see nothing. It is the job of the syntactician to discover why there's silence.

Option #1 : There's a silent element. For instance, pro and PRO are silent elements. They're DPs that aren't pronounced.
Option #2 : There's movement.
- Option #2a : It's head-movement. We recognize this movement because it forms new words.
- Option #2b : It's phrasal-movement.
  - Option #2b.i : It's A-movement.
  - Option #2b.ii : It's A'-movement.
Option #2: There's [pb_glossary id="821"]ellipsis[/pb_glossary]. In this case the silence arises through "deletion." We won't address ellipsis phenomena in this class. See LING 526.

For instance, in clefts there's clearly a gap, and if we want to analyze this structure, we have to figure out how this gap was formed. Is there a silent element? Is there movement? If there's movement, what kind of movement?

[latex] \setcounter{ExNo}{12} \ex. It was Mary that John saw {\underline{\hspace{10pt}}}. [/latex]

Clearly, this cannot be a case of head-movement; the gap does not contribute to making a new word. So this is either a case of a silent element, or phrasal movement.

In this chapter, we'll discover that different kinds of phrasal movements have a "signature." That is, it is possible to identify through diagnostic tests whether something a) has moved at all, and b) whether it has undergone A or A'-movement.

A'-movement diagnostics

Perhaps the defining characteristic of A'-movement is that it is unbounded: the moved element can be any number of clauses away from the gap.

[latex] \setcounter{ExNo}{13} \ex. \a. Who did John see {\underline{\hspace{10pt}}} \b. Who does Mary think [$_{CP}$ that John saw {\underline{\hspace{10pt}}} ] \c. Who did Bill say [$_{CP}$ that Mary thinks [$_{CP}$ that John saw {\underline{\hspace{10pt}}} ] ]? \d. Who did Carol hear [$_{CP}$ that Bill said [$_{CP}$ that Mary thinks [$_{C}$ that John saw {\underline{\hspace{10pt}}} ] ] ]? [/latex]

The term "unbounded" here has nothing to do with Binding Theory. As it applies to movement, "boundedness" refers to the ability to move across clausal domains.

Descriptively, A'-movement can cross an infinite amount (up to cognitive limitations) of CPs. This makes it different from A-movement, which cannot cross any CPs. That is, I cannot raise to subject across a CP:

[latex] \setcounter{ExNo}{14} \ex. *John thinks that {\underline{\hspace{10pt}}} saw Mary. [/latex] However, A'-movement is not an "anything goes" movement. That is, there are some movements that are impossible. For instance, it's not possible to move to a position that doesn't c-command the trace. That is, you cannot move into spec-CP of a CP in the subject position. [latex] \setcounter{ExNo}{15} \ex. \a. {} [$_{CP}$ That John came home early ] frightened Mary. \b. * [$_{CP}$ Which student (that) John came home early ] frightened $t_{DP}$ [/latex] One way to understand this is that traces are kind of like anaphors in that they are subject to a version of Condition A: they have to be c-commanded by a co-indexed element. More interestingly, even though A'-movement can cross CPs, it cannot cross all CPs. There are some clauses that you simply cannot move out of. These are called islands. It's a metaphor: the constituent is an island if nothing can get "off of" (i.e., move out of) the island. In addition to unboundedness, islands are part of the "signature" of A'-movement. That is, as a rule of thumb, A'-movement "respects" islands, meaning that all A'-movement are sensitive to islandhood diagnostics. There are many islands; we'll cover four in this class.

Complex-NP Constraint.

Recall Complex-NPs, which are CP complements to N. (They don't have a gap). You cannot move out of CP that is complement to NP.

[latex] \setcounter{ExNo}{16} \ex. \a. Sarah heard the lie [$_{CP}$ that Susan sued Bill ] \b. *Who did Sarah heard the lie [$_{CP}$ that Susan sued {\underline{\hspace{10pt}}} ] [/latex]

Adjunct island.

You cannot do A'-movement out of a clause that is adjoined. Adjoined clauses are typically headed by complementizers like because, if, when, after, before.

[latex] \setcounter{ExNo}{17} \ex. \a. John went home [$_{CP}$ after he broke the vase ] \b. *What did John go home [$_{CP}$ after he broke {\underline{\hspace{10pt}}} ] \ex. \a. Sarah ducked [$_{CP}$ because Ben kicked the ball ] \b. *Who did Sarah duck [$_{CP}$ because {\underline{\hspace{10pt}}} kicked the ball ] [/latex]

Wh-island constraint.

You cannot do A'-movement out of something that already has a wh-movement. For instance, you cannot do wh-movement out an embedded question. [latex] \setcounter{ExNo}{19} \ex. \a. John knows [$_{CP}$ who broke the vase ] \b. *What does John know [$_{CP}$ who broke {\underline{\hspace{10pt}}} ] [/latex] Compare this with (21), which doesn't have an embedded question. [latex] \setcounter{ExNo}{20} \ex. \a. John knows [$_{CP}$ that Bill broke the vase ] \b. What does John know [$_{CP}$ that Bill broke {\underline{\hspace{10pt}}} ] [/latex]

Subject islands.

You cannot move out of a constituent that is in the subject position.

[latex] \setcounter{ExNo}{21} \ex. \a. John read [$_{DP}$ a book by Tolstoy ] \b. Who did you [$_{DP}$ read book by {\underline{\hspace{10pt}}} ]? \ex. \a. {} [$_{DP}$ A book by Tolstoy ] is on the table. \b. *Who is [$_{DP}$ a book by {\underline{\hspace{10pt}}} ] on the table? [/latex] The purpose of identifying a particular signature for A'-movement is that it helps us identify kinds of movement when we encounter new data. For instance, the following sentences all contain gaps/unpronounced elements. [latex] \setcounter{ExNo}{23} \ex. \textbf{Raising-to-Object/ECM}\\ John made Carol out {\underline{\hspace{10pt}}} to be a liar. \ex. \textbf{\textit{Tough}-construction}\\ This book was tough to read {\underline{\hspace{10pt}}} . \ex. \textbf{Pseudo-clefts}\\ Mary was who I talked to {\underline{\hspace{10pt}}} . [/latex] If we want to understand the syntax of such sentences, then we need to figure out how that gap came into existence. We can apply the tests above (including a test for unbounded dependendencies) to determine if there's movement, and what kind of movement it is.

A-movement diagnostics

We've focused here on a few island constraints, which diagnose A'-movement. There are ways to diagnose A-movement as well. The clearest way to diagnose A-movement is to use binding. A-movement creates new binding dependencies; A'-movement doesn't.

[latex] \setcounter{ExNo}{26} \ex. \a. Mary$_i$ seems to herself$_i$ {\underline{\hspace{10pt}}} to be tired. \b. *Who$_i$ does it seem to herself$_i$ that John likes {\underline{\hspace{10pt}}} ? \c. *Mary$_i$, it seems to herself$_i$ that John like likes {\underline{\hspace{10pt}}} . [/latex] When you raise to subject across seem, the resulting movement can feed Condition A: the subject can now antecede an anaphor. A'-movement does not permit this. If I question or topicalize some constituent, it cannot antecede an anaphor.

Exercises

Gloss Template HTML and CSS

lacroixk — Thu, 13 Jun 2024 18:51:54 +0000

These HTML templates and the accompanying CSS stylesheet facilitate the presentation of glosses in online linguistic documentation. They offer a clean, organized structure using basic HTML tables with standard row and data cell tags, a structure which ensures accessibility for screen readers and keyboard navigation without needing the header cells or captions required for accessibility in standard use of HTML tables. The templates are easily customizable to fit various linguistic projects, and the stylesheet provides a cohesive and consistent gloss design across the templates.

Gloss Template

(1)	a.	John	goes	to	the	store₁
	b.	subject₂	verb	preposition Small caps	article	noun
	c.	"John goes to the store."

Gloss Template HTML






(1)
a.
John
goes
to
the
store₁


No Data
b.
subject₂
verb
preposition
Small caps

article
noun


No Data
c.
"John goes to the store."

Sign Language Gloss Template

		re
(37)	a.	INDEX₂	BITE-BAR		LONG!
		'Bite the bar for a long time!'
			hn

		re
	b.	PLEASE	INDEX₂	BITE!
		'Please, bite it!'

Sign Language Gloss Template HTML







re


(37)
a.
INDEX₂
BITE-BAR
LONG!




'Bite the bar for a long time!'





hn














re




b.
PLEASE
INDEX₂
BITE!





'Please, bite it!'

CSS Master Stylesheet for both templates

/* CSS Document */ /* Container for all elements */ .glosscontainer { width: 50%; /* Adjust this width up to 100%, whatever your preference */ margin-left: auto; /* Delete if you don't want the tables centered */ margin-right: auto; /* Delete if you don't want the tables centered */ } /* Table styling */ .interlinear { border-collapse: collapse; width: 100%; } .interlinear td { border: 1px solid #ededed; /* Delete '1px solid #' and replace with 'none' when glosses are all complete. */ padding: 8px; } /* Table top specific styling */ .list1 { width: 3%; text-align: center; vertical-align: middle; } .list2 { width: 2.5%; } .spacer { width: 12%; text-align: right; /* Change to 'center' in .spacer if needed */ vertical-align: middle; } /* Inline container styling */ .inline-container { display: flex; flex-wrap: nowrap; /* Prevents the items from wrapping */ align-items: center; /* Vertically centers the content in the container */ justify-content: center; /* Horizontally centers the content in the container */ } .inline-container p { margin-right: 10px; /* Optional: Adds some space between the paragraphs */ } /* Sets italics for the bottom row */ .interlinear tr:last-child { font-style: italic; } /* Small caps styling */ .small-caps { font-variant: small-caps; padding-top: 0; margin: 0; } /* For sign language glosses ONLY*/ .interlinear-SL { border-collapse: collapse; width: 50%; /* Keeps table at a optimal width - can be adjusted */ } td.cell-border { border-top: solid 4px; /* Creates the border above cells */ } td.spacer-left { text-align: left; /* Controls text alignment in td cell */ vertical-align: middle; } td.spacer-center { text-align: center; /* Controls text alignment in td cell */ vertical-align: middle; } td.spacer-right { text-align: right; /* Controls text alignment in td cell */ vertical-align: middle; } .interlinear-SL td { padding-top: 5px; /* Controls spacing between cells to prevent collapse */ padding-bottom: 5px; } .non-manual { text-align: right; /* Alignment for non-manual 'hn' and 're' elements */ }

List of abbreviations

juliedoner — Wed, 17 Jul 2024 22:44:47 +0000

Parts of speech abbreviations

Abbreviation	Meaning
Adj	adjective
Adv	adverb
C	complementizer
Conj	conjunction
D	determiner
N	noun
P	preposition
T	tense
V	verb

Glossing abbreviations

Abbreviation	Meaning
1	first person
2	second person
3	third person
ABS	absolutive [pb_glossary id="700"]case[/pb_glossary]
ACC	accusative [pb_glossary id="700"]case[/pb_glossary]
CN	[pb_glossary id="3950"]common noun[/pb_glossary] marker
DUR	durative [pb_glossary id="3551"]aspect[/pb_glossary]
ERG	ergative [pb_glossary id="700"]case[/pb_glossary]
F	feminine noun class
FUT	future [pb_glossary id="3550"]tense[/pb_glossary]
HAB	habitual [pb_glossary id="3551"]aspect[/pb_glossary]
M	masculine noun class
NOM	nominative [pb_glossary id="700"]case[/pb_glossary]
OBJ	object
PFV	perfective [pb_glossary id="3551"]aspect[/pb_glossary]
PL	plural number
PN	[pb_glossary id="3949"]proper noun[/pb_glossary] marker
PolQP	polarity question particle
PRS	present [pb_glossary id="3550"]tense[/pb_glossary]
PST	past [pb_glossary id="3550"]tense[/pb_glossary]
REAL	realis mood
SG	singular number
SUBJ	subjunctive mood
VTI	transitive verb with inanimate goal

Other abbreviations

Abbreviation	Meaning
AI	artificial intelligence
CJL	Canadian Journal of Linguistics
IPA	International Phonetic Alphabet
OED	Oxford English Dictionary
UG	Universal Grammar

Image descriptions for figures and diagrams

juliedoner — Wed, 11 Sep 2024 20:13:14 +0000

List of Tables

juliedoner — Mon, 02 Dec 2024 18:14:06 +0000

Lists of tables and figures

juliedoner — Mon, 02 Dec 2024 18:11:03 +0000

Tables

Chapter 1. What are morphology and syntax?

Section 1.1. What is morphology?

Table 1. Words with un-

Chapter 2. The scientific method in morphosyntax

Section 2.3. Prescriptivism and descriptivism

Table 1. Some contractions of standardized English Table 2. Examples of prescriptive rules in English Table 3. Change in the comparative and superlative forms of fun

Section 2.4. The lexicon and dictionaries

Table 1. The differences between mental lexicons and dictionaries

Section 2.5. Becoming a linguist: Identifying academic sources

Table 1. Kinds of publications

Chapter 4. Linguistic data in morphosyntax

Section 4.2. Methods of data collection

Table 1. Classification of data collection methods

Section 4.7. Becoming a linguist: Glossing spoken language data

Table 1. Acceptability markings

Section 4.8. Becoming a linguist: Glossing signed language data

Table 1. Common symbols in signed language glosses (adapted from Baker et al. 2016: 338-339) Table 2. Common abbreviations for non-manual markers in ASL (adapted from Baker et al. 2016: 339-340)

Chapter 4 practice exercises

Table 1. Mini Yoruba dictionary Table 2. Yoruba pronoun paradigm

Chapter 6. Kinds of morphemes and morphological processes

Section 6.10. Guided reading: Is n’t a clitic or an affix?

Table 1: Non-standard transciption used in Zwicky and Pullum (1983)

Chapter 7. Predicates, arguments, and clauses

Section 7.2. Arguments

Table 1. Kinds of transitivity Table 2. Ditransitive constructions

Section 7.4. Statements, questions, and commands

Table 1. Combinations of main and embedded clause functions

Chapter 8. Parts of speech

Section 8.1. Determining part of speech

Table 1: Traditional semantic definitions of parts of speech

Section 8.2. Nouns

Table 1. English derivational morphemes Table 2. English personal pronouns Table 3. The meanings of some common cases. Table 4. Niuean personal pronouns (Seiter 1980: 49) Table 5. Inflectional paradigms of different gender singular nouns in Russian Table 7. Deverbal nouns (V → N) (adapted from Haspelmath and Sims 2010: 87) Table 8. Deadjectival nouns (A → N) (adapted from Haspelmath and Sims 2010: 87) Table 9. Denominal nouns (N → N) (adapted from Haspelmath and Sims 2010: 87)

Section 8.4. Adjectives and adverbs

Table 1. The inflectional paradigms of some English adjectives Table 2. Deverbal adjectives (V → A) (adapted from Haspelmath and Sims 2010: 89) Table 3. Denominal adjectives (N → A) (adapted from Haspelmath and Sims 2010: 89) Table 4. Deadjectival adjectives (A → A) (adapted from Haspelmath and Sims 2010: 89)

Section 8.5. Functional parts of speech

Table 1. Subclasses of the determiner category Table 2. Examples of some gender-neutral neopronouns in English Table 3. Some possible parts of speech of question words in English Table 4. Subclasses of the tense category

Chapter 9.Words

Section 9.1. Word formation processes

Table 1. Examples of derivation in English Table 2. The productivity of -ish and -ness

Section 9.3. Packaging words and morphemes

Figures

Chapter 1. What are morphology and syntax?

Section 1.1. What is morphology?

Figure 1. An untidy workspace. Photo by SN.CHE. Used under Pexels license. Figure 2. A selection of pumpkins. Photo by Allie Reefer. Used under Pexels license. Figure 3. Language play. Photo by Thirdman. Used under Pexels license.

Section 1.3. Morphosyntax as a subfield of linguistics

Figure 1: A toddler kicking a ball in a field. Photo by Paul Ney. Used under CC BY-NC-SA license.

Chapter 2. The scientific method in morphosyntax

Section 2.2. Evidence and hypotheses

Figure 1. A decorative ghost. Photo by Juan Vargas. Used under Pexels license. Figure 2. Four macarons. Photo by Arminas Raudys. Used under Pexels license.

Section 2.4. The lexicon and dictionaries

Figure 1: The Dictionary of Proto-Algonquian by David Pentland. This dictionary includes a lot of etymological information about numerous Algonquian languages, which would not be included in the mental lexicons of speakers. Photo by Julie Doner. Reproduced with permission.

Section 2.6. Becoming a linguist: Scientific writing in morphosyntax

Figure 1: A model of a carbon atom. Photo by SrKellyOP. Used under CC0 1.0 license.

Chapter 3. Theories of grammar and language acquisition

Section 3.1. Becoming a linguist: Scientific writing in morphosyntax

Figure 1. A speed limit sign. Photo by Shahnoor Habib Munmun. Used under CC BY 3.0 license. Figure 2. A bookshelf at a university library in the Indigenous languages section. This section contains reference and pedagogical grammars, as well as dictionaries, texts in Indigenous languages, and academic books on specific topics in Indigenous linguistics. What kinds of books can you spot in this picture? Photo by Julie Doner. Figure 3: A page from a pedagogical grammar of Yoruba (Mosadomi 2011). Yoruba is a member of the Niger-Congo language family and is spoken primarily in Nigeria with over 43 million native speakers (Eberhard et al. 2023). Figure 4: A picture of page 35 of Bloomfield (1957), a classic reference grammar of Nishnaabemwin (also known as Eastern Ojibwa). Nishnaabemwin belongs to the Algonquian language family and is spoken by about 220 people in Canada, according to the 2021 census. It is spoken in and around Lake Huron, especially Manitoulin Island (Eberhard et al. 2023). Photo by Julie Doner.

Section 3.2. Generative grammar

Figure 1: Noam Chomsky in 2004. Photo by Duncan Rawlinson. Used under CC BY-NC 2.0 license. Figure 2: A bicycle. Photo by LUM3N. Used under Pixabay Content License.

Chapter 4. Linguistic data in morphosyntax

Section 4.8. Becoming a linguist: Glossing signed language data

Figure 1. Differences between ASL and STS in handshape meanings (Anderson et al. 2022 Section 3.8). Figure 2. Different orientations of the right hand as seen from the signer’s point of view (Anderson et al. 2022 Section 3.8).

Chapter 6. Kinds of morphemes and morphological processes

Section 6.2. Affixes

Figure 1. A stack of five books. Photo by Julie Doner. Figure 2. A stack of books with one book inside of another book. Photo by Julie Doner.

Section 6.8. Becoming a linguist: Anatomy of an academic article

Figure 1. The first page of an academic journal article, Bjorkman (2022), published by the Canadian Journal of Linguistics. Used under CC BY 4.0 license.

Chapter 8. Parts of speech

Section 8.2. Nouns

Figure 1: A schematic representation of nominative-accusative case alignment. Figure 2: A schematic representation of ergative-absolutive case alignment.

Morphology

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Concepts and skills you will need for LING 527/727

How to determine word category
Constituency and how it is represented
- Trees and bracketing structures
Headedness
Functional vs. lexical categories
Head-movement constraint

Morphology and Morphological theory

Morphology is the study of words. Morphologists look at the pieces that make up words (e.g., how many pieces of meaning are in the word sang?), what processes govern how words are made (e.g., the Head Movement Constraint), and the relationship between morphology and other aspects of grammar, like syntax and phonology. Because morphology is such a broad field, it is (arguably) the most contentious field. There is a general disagreement about what the field of morphology is even concerned with. Once central topic of disagreement concerns how much the fields of morphology and syntax are related. Or put differently: do the same "rules" that govern syntax also govern morphology? We've encountered quite a bit of morphology in this book already, because I happen to subscribe to the belief that morphology and syntax are indeed quite related fields. In many ways, the connections between syntax and morphology are clear. For instance, we postulated an independent phrase TP, where tense information is introduced. But we also noted that tense is often expressed together with the verb, like walk-ed in English or ali-anguka 'fell' in Swahili. So if you think that TP is distinct from VP, then you need an explanation for why they are sometimes pronounced in the same word. Our explanation in the book was head-movement: V moves to T or T moves to V. This makes very specific predictions, because it puts word-formation into the syntax. An alternative theory would claim that words like walked and alianguka are not "made" in the syntax, but rather are simply inserted into the trees as is. Other similarities are concerned with constituency and structure. Recall from Chapter 3 that we started out discussion of brackets and trees by first looking at compound words like swan-boat. Compounding is a morphological phenomenon (or a morphological process, again depending on your theoretical assumptions) which happens to precisely parallel the key aspects of syntax phrase structure: headedness and constituency. Morphological theory is concerned with details and idiosyncracy. It is fairly easy to create a theory that captures the general pattern, but it is much more difficult to create a theory that captures all the data, including all the exceptions to the pattern. Just taking English past tense as the canonical example, we can generally say that English past tense is formed by suffixing -ed to the end of the verb, e.g., walked. But there are number of exceptions in English. The partially [pb_glossary id="842"]suppletive[/pb_glossary] verbs, like teach ~ taught, bring~brought, seek~sought, fight~fought, etc are one issue. In what way are the non-past and past forms related to each other? Moreover, note that all of these verbs have similar past tense forms, ending in -ought. Is there any sense that these forms are a (synchronic) natural class? And then what do we do with the fully suppletive forms like go~went? Do we say that there are two pieces of meaning in went (go+past)? Some morphologists say yes, and some say no. The study of morphology really gets interesting when we look at languages beyond English, and really, Indo-European in general. For instance, in Swahili, verbs may have a range of suffixes, each a distinct morpheme which supplies a particular piece of information. [latex] \exg. Juma a-li-nunu-li-w-a kitabu\\ Juma 1\textsc{sm}-\textsc{past}-buy-\textsc{appl}-\textsc{pass}-\textsc{fv} book\\ \trans `Juma was bought a book.' (I.e., `Someone bought a book for Juma.') [/latex] Swahili packs more information into its verb than the corresponding English sentence, indicating with -li that the buying was "for" someone (an applicative suffix). The passive -w indicates that the verb is a passive form; Juma is the recipient of the action. Native American languages are also offer important lessons in morphology, with many being extremely complex. Consider Southern Pomo, which can pack an extraordinary amount of information into the verb.[footnote]This example is from Neil Walker (2013) A Grammar of Southern Pomo: An Indigenous Language of California. PhD thesis. University of California, Santa Barbara. p. 401[/footnote] [latex] \exg. \v{c}ahnu k'o\textipa{?}di \v{c}anho-den-t\textsuperscript{h}-i:ba=\textipa{?}wa=\textipa{?}a\\ speech good speak-\textsc{dir}-\textsc{neg}-\textsc{cond}=\textsc{cop.evid}=1\textsc{sg.act}\\ \trans `I can't talk well.' [/latex]

Fieldwork

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Concepts and skills you'll need for Fieldmethods (LING 527/727)

Identifying word category
Constituency
Functional vs. lexical categories
Binding Theory

Where does linguistic data come from? At this point in your linguistics career at KU, you've almost certainly encountered some data sets that you've been asked to analyze. These data sets are almost always neatly tailored to fit the needs of the class and the assignment. Language is messy, though, and doesn't just give you data sets. So where do we (the professional linguists) get our beautiful data from? The answer is fieldwork! Fieldwork is the exploration of new (to the researcher) languages. Fieldworkers work with native speakers to construct a description of a language, or at least a part of the language. As the fieldworker learns more about the language, he or she might eventually be able to identify an interesting pattern---and thus a data set is born! You did a very brief imitation of fieldwork as you worked through the Swahili examples to build up the Phrase Structure Grammar. Fieldworkers elicit data from speakers. With the elicited data, a fieldworker can refine their theoretical model, or they can simply develop an accurate description of the language in question. In my opinion, fieldwork is one of the most challenging and most rewarding of the linguistic disciplines. Good fieldworkers must have a working knowledge of phonetics, phonology, morphology, syntax, and semantics. Indeed, it is often that case that a fieldworker has to use skills from all of these disciplines within the span of 5 or 10 minutes. When you start working with a language that you don't know anything about, then everything is new! You don't know what the sounds are, you don't know what the word order is, you don't know anything. Slowly, a fieldworker builds up an understanding of the language. Fieldworkers have a variety of tools at their command. These tools are called fieldmethods. The most important fieldmethods are those that teach you how to elicit the data you want. Different tools do different things. We have tools to elicit tone. We have elicit modals. Etc. From syntax, the most important tools are those that help us categorize things. So being able to identify parts of speech is crucial. And being cable to determine constituency is as well. It's also important to have a sense of the distinction between functional vs. lexical categories. It may come as a surprise that Binding Theory is also a crucial skill for a fieldworker. The simple reason is that the Binding Conditions are great ways to determine hierarchical structure. We can use Conditions A, B, C as diagnostics for which elements are above other elements, and which elements have moved. As always, though, it is important to keep in mind that, going into a language you're unfamiliar with, you don't know how anything works! You'll need to figure out the tests for that language. Finally, perhaps the best thing about fieldwork is being able to work with a native speaker of language that you don't know. Fieldworkers don't just learn about the language, but they also learn about the culture and society of the speakers of that language.

Being a syntactician and being a scientist

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000

What is "syntax"?

Naively, syntax is the study of word order. More accurately, syntax is the study of how languages package information. When we conduct syntactic investigation of a particular language, we are learning how that language chooses to organize the pieces of the sentence, that is, how it organizes the information that is being communicated. By looking at more than one language, we can make generalizations about Human Language, i.e., the unique capacity for humans to communicate.

Why is syntax a science?

Syntactic theory is scientific because it is, fundamentally, hypothesis driven research. By looking at one language or many languages, we come up with a hypothesis about how that language(s) work(s). We then test that hypothesis by looking at one or more languages. It is absolutely vital that we treat syntax as a scientific discipline. It is tempting to succumb to the false belief that "I speak a language, therefore I understand how it works." This is equivalent to saying "I have a body, so I understand how it works." It is likewise critical that we avoid relying on our "impressions" about a language, the prescriptive rules we were taught in English/Spanish/German/Swahili/Hindi/... about "proper" writing, or our judgements about what is "good" and "bad" because these things are not scientifically rigorous and they do not reflect language as it is used. We cannot use those things to make hypotheses because in fact, they do not represent the actual data.

Suppose a biologist told you that they had found the cure for the common cold. "Great news! How do you know?" you would ask. If the biologist responds, "Well, I just have a really good feeling about this vaccine," you would be rightfully doubtful. Ideally, you want the biologist to have run clinical trials which illustrated the effectiveness of the vaccine. The same is true of syntax (and all linguistics). We do not base our theory on our intuitions or our "feelings" or the rules that others impose. We base them on data. For this reason, much of this course will focus on developing diagnostic tests that we can use to identify properties of language.

[embed]https://youtu.be/uyFGGTkTqC4[/embed] Syntax as a science (Andrew Collins; length: 2m 51s)

Why this theory?

There are many theories of syntax. This book teaches (a version of) X'-syntax (read "X-bar syntax").[footnote]Indeed, the theory taught here is a mere shadow of the extensive theoretical apparatus that was worked out and utilized in the '80s and '90s. That theory was called Government and Binding and it existed within the over-arching Principles and Parameters framework.[/footnote] However, very few working theoretical syntacticians use X'-theory anymore. In fact, in the second level of syntax at KU (Syntax II LING 526/726), we transition to Minimalist Syntax, which is the syntactic theory that is most widely adopted now. So it is fair to ask, why don't we just start with Minimalism, and skip X'-syntax? What's the point in learning a theory that practically no one uses anymore? In fact there are many reasons why it's best to start with X'-syntax. Foremost, utility. While the theory itself is no longer used by working theoreticians, the core principles behind the theory are useful in every other branch of linguistics that interfaces with syntax, including semantics, morphology, psycholinguistics, neurolinguistics, first and second language acquisition, and others. If you know X'-theory, then you can read those papers, and understand the conclusions of those researchers. Moreover, in related non-theoretical fields, particularly experimental fields, you are quite likely to encounter people still relying on trees that use X'-syntax. Since those fields are non-theoretical, the differences between X'-syntax and Minimalism are usually irrelevant. Second, though Minimalism is the theory that is adopted by most working (generative) syntacticians, it is important to realize that Minimalism grew out of X'-syntax and in fact presupposes all of the empirical groundwork that built X'-theory. It is this empirical groundwork that we are covering in this class. For instance, in Syntax II, we take for granted the following ideas that we will empirically motivate in this class.

Constituency (in chapter 3)
Head-movement (in chapter 7)
Selection (in chapter 8)
C-command (in chapter 9)
Phrasal movement (in chapter 10 and chapter 12)
Raising and Control (in chapter 11)
...among many other things...

Thus, in order to learn and understand Minimalist syntax, you have to have an understanding of all of the above topics. That is what this class teaches. And finally, it many ways, X'-syntax is a better theory than Minimalism. It is more powerful than Minimalism: it offers clear explanations for almost all of the facts presented in this work. However, in the mid-'90s, the field of syntax decided that the explanations that were offered by X'-syntax just weren't very good explanations. When we transitioned to Minimalism in the mid-to-late '90s, we lost much of that empirical coverage: all the explanations we had in X'-theory were no longer possible, but we didn't have any replacements for those explanations! A great deal of research since then has been to try to capture the facts that X'-theory explained. We have been partially successful. Some things have become clearer with time, others are still mysterious. In the end, the purpose of this class is to teach you to think like a syntactician. This means relying on data to formulate hypotheses, and testing those hypotheses by looking at more data. This scientific orientation transcends theoretical models. If you leave this class with a scientific mindset for syntax, but with a healthy dose of skepticism about the theory taught, then I will consider the class a success. (I look forward to seeing you in Syntax II!)

How to use this book

Foremost, you need to read it. This should go without saying, but of course many people will not read the book. (Indeed, many people will not even have made it this far in the introduction.) It is also not enough to simply look at the videos that appear in each chapter. Everything that is assigned should be read thoroughly, with the intention of practical application. We will mostly be going over diagnostic tools for identifying syntactic phenomena. You the student should endeavor to learn those tools so that you can apply them to new data. It is the same as learning the Pythagorean theorem, or how to calculate the slope of a line. The tools taught here are meant to be applied. Finally, as an Open Education Resource, this book is a "living document." It can be edited at any time. I encourage you to contact the author (Dr. John Gluckman: johnglu@ku.edu) if you have questions or think that something should be changed or added. Your comments are crucial for making this a document that later classes will benefit from fully.

Drawing Trees

You will end up drawing many trees in this class. I encourage you to use the web application here (https://dprebyl.github.io/syntree/), developed by Drake Prebyl, to draw all your trees. The application offers an easy-to-use interface for drawing very beautiful structures, with lots of ways to customize the representations.

Brackets and Trees

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000

Learning Objectives

By the end of this chapter, you should be able to,

understand and utilize three different ways to represent syntactic information
appreciate the relationship between constituency and representation (phrase structure rules, brackets, and trees)
understand the relationship between meaning and structure, and what ambiguity tells us about that relationship

Starting Small

The constituency tests that we performed in the last chapter allowed us to see that every sentence is made up of smaller units of the sentence. Importantly, each constituent forms its own "semantic" unit as well. This is sometimes difficult to understand, and even harder to demonstrate, so we'll start first by looking at a somewhat "smaller" problem, and work up to full sentences.

English (like all Germanic languages) has a productive process through which new words can be formed that involves putting two words together. The resulting words are called [pb_glossary id="761"]compounds[/pb_glossary].

Compounding in English
rice	pot	→	rice-pot
swan	boat	→	swan-boat
phone	case	→	phone-case

This process is, theoretically, infinitely [pb_glossary id="760"]recursive[/pb_glossary], meaning that we can continue to make new words from existing words.

More compounding in English
rice-pot	rack	→	rice-pot-rack
swan-boat	jacket	→	swan-boat-jacket
phone-case	store	→	phone-case-store

We can think of each piece of a compound as a constituent. Rice-pot has two pieces of meaning, rice and pot. When we put them together, we make something whose meaning is the combination of both. This idea is represented in terms of [pb_glossary id="762"]bracketing structure[/pb_glossary].

[latex] \ex. \a. \textit{rice-pot} : [ [ rice ] [ pot ] ] \b.\textit{swan-boat} : [ [ swan ] [ boat ] ] [/latex]

Inside each pair of brackets is one meaning "unit." So rice-pot involves three pairs of brackets: one around rice, one around pot and one around the compound rice-pot. Same with swan-boat.

This way of representation allows us to iteratively represent more complex compounds, as in (2).

[latex] \setcounter{ExNo}{1} \ex. \a. \textit{rice-pot-rack} : [ [ [ rice ] [ pot ] ] [ rack ] ] \b. \textit{swan-boat-jacket} : [ [ [ swan ] [ boat ] ] [ jacket ] ] [/latex] When we build larger compounds, we are creating a meaningful unit, and then adding to it. So first I create the word rice-pot, and it has a meaning like, "a pot for rice." Then I create the word rice-pot-rack, which has the meaning "a rack for a pot for rice." In other words, the meaning of the entire word depends on the meaning of its parts.

Of course, the more complex our compounds become, the more difficult it is to read the bracketing structure. So another way to express the exact same information is by using a tree, as in (3).

[latex] \setcounter{ExNo}{2} \ex. \begin{tikzpicture}[mytree] \node(top) {rice-pot} child {node {rice}} child {node {pot}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {rice-pot-rack} child {node {rice-pot} child {node {rice}} child {node {pot}} } child {node {rack}}; \end{tikzpicture} \ex. \begin{tikzpicture}[mytree] \node(top) {swan-boat} child {node {swan}} child {node {boat}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {swan-boat-jacket} child {node {swan-boat} child {node {swan}} child {node {boat}} } child {node {jacket}}; \end{tikzpicture} [/latex] Each [pb_glossary id="763"]node[/pb_glossary] in the tree corresponds to one pair of brackets. Thus, trees and brackets provide the exact same amount of information, it's just that trees do it in a visually more appealing way. But it's always possible to state a bracketing structure as a tree, and vice versa. I can actually make my representations slightly more informative by adding in category information. The following trees and bracketing structures give us all the preceding information, and they additionally tell us what the category of each constituent is. [latex] \setcounter{ExNo}{4} \ex. \a. \begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ rice}} child {node {N \\ pot}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {N} child {node {N} child {node {N \\ rice}} child {node {N \\ pot}} } child {node {N \\ rack}}; \end{tikzpicture} \b. [$_N$ [$_N$ rice ] [$_N$ pot ] ] \hfill [$_N$ [$_N$ [$_N$ rice ] [$_N$ pot ] ] [$_N$ rack ] ] \ex. \a.\begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ swan}} child {node {N \\ boat}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {N} child {node {N} child {node {N \\ swan}} child {node {N \\ boat}} } child {node {jacket}}; \end{tikzpicture} \b. [$_N$ [$_N$ swan ] [$_N$ boat ] ] \hfill [$_N$ [$_N$ [$_N$ swan ] [$_N$ boat ] ] [$_N$ jacket ] ] [/latex] Note that it's not necessary to explicitly state the compound words that are formed at each junction, because I can just look lower in the tree to figure out what that word is. The reason it's good to list the category, though, is that you can combine more than one category. We therefore want to know which category "projects," that is, names the resulting word. [latex] \setcounter{ExNo}{6} \ex. \textit{black-bird} : \begin{tikzpicture}[mytree] \node(top) {N} child {node {A \\ black}} child {node {N \\ bird}}; \end{tikzpicture} \hspace{50pt}[$_N$ [$_A$ black ] [$_N$ bird ] ] \ex. \textit{sweet-talk} : \begin{tikzpicture}[mytree] \node(top) {V} child {node {A \\ sweet}} child {node {V \\ talk}}; \end{tikzpicture} \hspace{50pt}[$_V$ [$_A$ sweet ] [$_V$ talk ] ] [/latex]

[embed]https://youtu.be/FZ1l70p2_RA[/embed] Brackets and trees (length: 3m 49s)

Scaling up to sentences

Just like each word in a compound is a unit of meaning, which combine to create bigger meanings, sentences also behave the same way. Each constituent contributes a piece of meaning. The constituents combine to create larger meanings, until the sentence is complete. Our constituency tests are telling us how the pieces of meaning come together. Just like we know that the compound sweet potato chickpea salad is made up of two smaller compounds sweet potato and chickpea salad, sentences can be broken up into smaller chunks. For instance, we found that in the sentence The tired doctor slept, the string of words the tired doctor is a constituent. Thus, this piece of meaning is enclosed in brackets or corresponds to a single node in a tree. [latex] \setcounter{ExNo}{8} \ex. \begin{tikzpicture}[mytree] \node(top) {the tired doctor slept} child {node {the tired doctor} child {node {the}} child {node {tired}} child {node {doctor}} } child {node {slept}}; \end{tikzpicture} {}\\ {}\\ {} [ [ the tired doctor ] slept ] [/latex] And just like with compounds, we can also add in information about categories. The constituent the tired doctor is a Determiner Phrase (for reasons that we'll discuss in chapter 5), and the whole thing is a Sentence_English (abbreviated S_E). [latex] \setcounter{ExNo}{9} \ex. \begin{tikzpicture}[mytree] \node(top) {S$_E$} child {node {DP} child {node {D \\ the}} child {node {A \\ tired}} child {node {N \\ doctor}} } child {node {V \\ slept}}; \end{tikzpicture} {}\\ {}\\ {} [$_S$ [$_{DP}$ [$_D$ the ] [$_A$ tired ] [$_N$ doctor ] ] [$_V$ slept ] ] [/latex]

[embed]https://youtu.be/BDV3IKuDAgE[/embed] Constituency, brackets, and trees (length: 2m 59s)

Note that the combination of words in a compound word results in a category label (N, A, V, D, ...). This is because when you make a compound, you're making another word. When you create constituents that are bigger than words, we called these phrases. Putting the and doctor together doesn't make a new word, it makes a phrase---a DP.

Here comes the crucial leap to phrase structure grammars. The rules of a phrase structure grammar are the possible structures in a language. That is, the phrase structure grammars tell you what the possible bracketing structures and trees are in a language. Consider again our rules for English from chapter 2 (now with abbreviations).

A revised grammar fragment of English Rule 1: S_E → DP V Rule 2: DP → D N Rule 3: DP → D A N

These phrase structure rules tell you how information is packaged in English. That is, they tell that the phrase [_D_P ... ] precedes V. And they tell you that inside of DP you can have either [_D_P D N ] or [_D_P D A N ]. More generally, the three phrase structure rules say the following are the (only) possible trees and bracketing structures in English. [latex] \setcounter{ExNo}{10} \ex. {}[$_{S_E}$ [$_{DP}$ D N ] V ] \begin{tikzpicture}[mytree] \node(top) {S$_E$} child {node {DP} child {node {D}} child {node {N}} } child {node {V}}; \end{tikzpicture} \ex. {}[$_{S_E}$ [$_{DP}$ D A N ] V ] \begin{tikzpicture}[mytree] \node(top) {S$_E$} child {node {DP} child {node {D}} child {node {A}} child {node {N}} } child {node {V}}; \end{tikzpicture} [/latex] Phrase structure grammars are thus "instructions" for what trees/bracketing structures can be drawn in a language. Our fragment for English above is quite limited; it only generates a few trees/structures. We will expand on this fragment in the following chapter.

[embed]https://www.youtube.com/embed/GduRxnZVyWY[/embed] Representations (length: 44s)

Structural Ambiguity

What does the following compound mean?

[latex] \setcounter{ExNo}{12} \ex. baby-tree-planter [/latex]

Many of you will say that it means, "someone who plants baby trees." And many of you will say that it means "a tree-planter who is a baby." Both of these are right; it's possible to get both readings of the compound.

If we wanted to represent each meaning, we would choose different structures. The "baby-trees" meaning would have the following representations. [latex] \setcounter{ExNo}{13} \ex. ``A planter of baby trees'' : \begin{tikzpicture}[mytree] \node(top) {N} child {node {N} child {node {N \\ baby}} child {node {N \\ tree}} } child {node {N \\ planter}}; \end{tikzpicture} [/latex] The reason is that we want "baby tree" to be a unit of meaning, i.e., a constituent, because that meaning describes the kind of planter it is. On the other hand, the "tree-planting-baby" meaning would have the following representation [latex] \setcounter{ExNo}{14} \ex. ``A tree-planter who is a baby'' : \begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ baby}} child {node {N} child {node {N \\ tree}} child {node {N \\ planter}}}; \end{tikzpicture} [/latex] The reason is that we want "tree-planter" to be a unit of meaning first, because ultimately we're describing a kind of tree-planter. This kind of ambiguity is called [pb_glossary id="765"]structural ambiguity[/pb_glossary], because we can represent the ambiguity structurally.

There is another kind of ambiguity called lexical ambiguity. The word bank is lexically ambiguous, because it means two different things (a financial institution and the side of a river). But it's not structurally ambiguous because the two meanings do not correspond to different structures.

The take-away point here is that different meanings correspond to different structures. More abstractly, the different meanings are the result of packaging the information in different ways.

The same is true when we scale up to full trees. Consider the following (very famous) sentence in English.

[latex] \setcounter{ExNo}{15} \ex. Mary saw the man with the binoculars. [/latex]

This sentence is ambiguous as well. On the one hand, it can mean that Mary saw a man who was holding binoculars. On the other hand, it can mean that Mary used binoculars to see the man. In the first meaning with the binoculars describes a property of the man. On the second meaning, with the binoculars describes how Mary is "seeing."

Just like baby-tree-planter, we observe different meanings. We can ask then whether those meanings correspond to different structures. The way to answer this question is to apply constituency tests, because constituency tests tell us how the parts of a sentence fit together. For instance, let's apply the [pb_glossary id="756"]substitution[/pb_glossary] test. Specifically what I'm doing in this test is trying to see if with the binoculars must connect with the other words in a particular way, in particular, if it must [pb_glossary id="757"]modify[/pb_glossary] man.

[latex] \setcounter{ExNo}{16} \ex. Mary saw {\underline{the man with the binoculars}}. $\longrightarrow$ That's right, and John saw {\underline{him}}, too. \ex. Mary saw {\underline{the man}} with the binoculars. $\longrightarrow$ That's right, and John saw {\underline{him}} with the telescope. [/latex] Notice first that both of these sentences are grammatical. At first, you might be tempted to think that we've done our test wrong, since the results of (17) and (18) suggest that there are two ways to do the constituency of this sentence. But in fact, if we applied all of our tests, we'd find that both underlined strings of words are possible constituents. What is crucial is that the constituent parsings in (17) and (18) correspond to a different meanings. This is the hallmark of a structural ambiguity: different meanings correlate with different structures. Explicitly, what (17) and (18) shows is that once we've applied a test, then the ambiguity goes away. When we force [ man with the binoculars ] to be a constituent by replacing that string with him, the sentence can only mean that the man has the binoculars. In contrast, when we force [ the man ] to be a constituent by replacing just that string with him, then the sentence can only mean that Mary was using binoculars. This is because by applying the test, we're forcing the sentence to have a certain structure, and therefore to have a certain meaning. Again, the take-away message here is: different meanings have different structures---and vice versa.

[embed]https://youtu.be/YvWiP7bIatI[/embed] Structural ambiguity (length: 1m 27s)

Before moving on, it's important to recognize that trees are not syntax. Very often, students in an introductory syntax class come away with the impression that trees are the most important thing for a syntactician or that syntacticians "study" trees. This is incorrect. There are many, many syntax papers, books, and analyses that don't contain a single tree, and in fact there are many theories of syntax that largely eschew trees. Trees are simply representations of information. We use trees to visualize how information is packaged, but syntacticians don't study trees. They are just tools, and like all tools, they're good at doing certain things, and bad at doing other things. However, if we draw a tree, then we must be precise. Just like y=2x²+5 is not the same as y=5x²+2, each structure that we create in syntax represents different information. Indeed, the math analogy is telling: we can represent y=5x²+2 in different ways, say, by listing all the possible values for x and their corresponding values for y, or by graphing the equations on a two dimensional plane. This is just like syntax: there are multiple ways to represent information (trees, brackets, listing sentences, PSGs). We can use any of them, but some are better in certain contexts. Still, we the instructors are going to hold you to a high standard in your trees. Not because trees are important per se, but because a precise representation of language is important.

Where we're going

With these representational tools, brackets and trees, we're going to try to build a functional description of a language using phrase structure rules. What is important is that our description---that is the phrase structure grammar---accurately represents the language in question. To do this we simply use the tools we have: constituency and category diagnostics.

Things to remember

The relationship between brackets, trees, and phrase structure grammar
How constituency and structures are related
Structural ambiguity derives from different structural representations.

Functional categories

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000

Learning Objectives

By the end of this chapter, you should,

recognize the distinction between functional and lexical categories
be able to use constituency tests to infer the presence of functional structure
understand the motivation for projecting D, C, and T, even when they're silent.
be able to draw the trees for DP, TP, and CP across languages.

In this chapter we're going to investigate some of the "hidden" parts of syntax. We have until now been looking at what we can see. We can see (and diagnose the category of) nouns, verbs, and adjectives, etc. And we can see constituency. But there are very many categories and constituents that we cannot always see directly. Much of syntactic research involves inferring the presence of structure, things we cannot see.

This is, for many people, an inherent flaw in syntactic theory. Much of syntactic theory involves talking about "invisible" things. However, the reason that we propose such invisible objects is that we can detect the presence of them, even when we can't see them. By analogy, we will never directly observe a black hole. It is "invisible" by virtue of the fact that light cannot escape its gravitational pull. However, the evidence suggests that black holes exist, and we can directly observe that evidence. That is, the black hole affects space around it, and we can see those effects. This is the same thing that syntacticians do. (And semanticists too!) Sometimes we infer the presence of an element because we see its effects---even when we do not always see the element itself.

This chapter focuses on [pb_glossary id="803"]functional[/pb_glossary] phrases. The categories we've been looking at so far have been [pb_glossary id="804"]lexical[/pb_glossary] categories. Lexical words are, loosely, words with definable meanings. Dog, table, honesty, sing, walk, sadly are all lexical words because the meanings for these words are relatively easy to define. In contrast words like the, a, should, of are functional words of different categories, because it's not so easy to define these words. Some people have suggested that the "role" of functional words is not to provide meaning, but rather to link meanings together. For instance, in box of matches the functional words of (a preposition) serves to relate the independent meanings of box and matches. Of by itself doesn't really carry "weight;" it doesn't really mean anything. Rather it simply relates the meanings of box and matches. The same thing with the copula is. In Mary is a doctor, the copula is simply equating the meanings of Mary and a doctor. But it doesn't mean something in the same way that Mary and a doctor mean something. The lexical/functional distinction also broadly corresponds with open class vs. closed class of words. An open class of words is one which can easily add new words. For instance, nouns are an open class because I can very easily invent a new noun---we do this all the time! A closed class of words is one which does not easily get new words added. It's much harder to invent a new tense, or a new determiner. It has also been noticed that the lexical/functional distinction is sometimes apparent in the phonology and morphology. Functional elements are often not even separate words: they're affixed onto lexical words. This is very true of tense, which is often expressed on the verb (e.g., walked). And often functional elements are "minimal" words, like the determiners a and the. We'll focus in this chapter on three functional projections D (for Determiner), T (for Tense), and C (for Complementizer).

[embed] https://youtu.be/RY-neLbjBzQ[/embed] Functional vs. Lexical words (Andrew Collins; length: 2m 52s)

Determiners

The first functional category we'll talk about we've already seen: Determiners (D). Determiners in English include a and the. The demonstratives this, that, these and those are also category D (in this class). And we'll also assume that [pb_glossary id="825"]quantifiers[/pb_glossary] like every, some, few, many, etc are determiners, too. So far, I've stipulated to you that constituents like the doctor are DPs, Determiner Phrases, and not, say, NPs, Noun Phrases. It is time to back up this assertion. Logically, there are two choices: we could make DP be a part of NP, or we could make NP be a part of DP. The two choices are shown below. [latex] \begin{minipage}{.49\textwidth} \ex. \begin{tikzpicture}[mytree] \node(top) {NP} child {node {DP} child {node {D \\ the}} } child {node {N \\ dog}}; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textiwidth} \ex. \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ the}} child {node {NP} child {node {N \\ dog}} }; \end{tikzpicture} \end{minipage} [/latex] In this class, we're going to choose the analysis on the right, in (2). That is, we're adopting the DP-hypothesis. The DP-hypothesis is the proposal that all noun phrases are encased in a "DP shell." Below I provide the DP hypothesis in prose, and as a phrase structure rule.

The DP Hypothesis (Abney 1987) The head of a Determiner-X-Noun-Y sequences is D. (Where X and Y are potential noun modifiers.)

DP → D (XP) (NP) (YP)

What this hypothesis says is that constituents like the doctor, the happy doctor, the happy silly doctor are not NPs, but rather they're DPs. It also says that pronouns like I, we, you, us etc, are of category D, and are just DPs, with no associated NP. Evidence in support of the DP hypothesis comes from a few different sources. The first is distributional evidence. As we've observed before, pronouns and constituents like the doctor share a distribution, i.e., they occur in all the same places. We therefore conclude that they are of the same category. (This was the basis for our [pb_glossary id="756"]substitution[/pb_glossary] test.) So what category are pronouns? They are of category D! Why? Well, for one thing, they distribute like determiners. I can say We linguists sang, where we appears in the position reserved for a determiner: in front of a noun. The structures for we sang and we linguists sang are shown below. [latex] \setcounter{ExNo}{2} \begin{minipage}{.49\textwidth} \ex. \begin{tikzpicture}[mytree] \node(top) {S$_E$} child {node {DP} child {node {D \\ we}} } child {node {VP} child {node {V \\ sang}} }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \ex. \begin{tikzpicture}[mytree] \node(top) {S$_E$} child {node {DP} child {node {D \\ we}} child {node {NP} child {node {N \\ linguists}} } } child {node {VP} child {node {V \\ sang}} }; \end{tikzpicture} \end{minipage} [/latex] Again, the reasoning is: If she is a DP, and it shares a distribution with the doctor, then the doctor must be a DP as well. Note that, despite the name "pro-noun," pronouns do not actually substitute for nouns! In English, we use the word one to substitute for a noun --- or really an entire noun phrase, which can include an adjective. But one cannot replace the sequence Det-(Adj-)Noun. [latex] \setcounter{ExNo}{4} \ex. \a. I like this {\underline{book}}, but Margit likes that {\underline{one}} \b. I like this {\underline{long book}}, but Margit likes that {\underline{one}} \c. I like {\underline{this book}}, *but Margit likes {\underline{one}}, too. [/latex] Moreover, in other languages, we clearly see that pronouns share a morphological shape with determiners. This is true of every Romance language (like French) and also many Germanic languages, like Northern Norwegian. [latex] \setcounter{ExNo}{5} \begin{flushleft} French \end{flushleft} \ex. \ag. Marie voit \textbf{la} voiture\\ Marie see.\textsc{pres} D.\textsc{fem.sg} car\\ \trans `Marie sees the car.' \bg. Marie \textbf{la} voit\\ Marie it.\textsc{fem.sg} see.\textsc{pres}\\ \trans `Marie sees it.' Northern Norwegian \ex. \ag. Ho Marit s\r{a} han \O ystein\\ she Marit saw he \O ystein\\ \trans `Marit saw \O ystein \bg. Han \O ystein s\r{a} ho Marit.\\ he \O ystein saw she Marit\\ \O ystein saw Marit. \hfill (Matushansky, 2009, p. 7) [/latex] It is a relevant question now to ask whether all NPs are DPs. For instance, in the sentence Birds sing is birds an NP? Or is there a "silent" D present? Based on the previous reasoning, we would conclude that, Yes, there is a silent D. This is because I can substitute the bare noun birds with a pronoun: Birds sing. Yes that's right, they sing. Thus, again, if pronouns are DPs, and "bare" NPs share a distribution with pronouns, then bare nouns must be DPs as well. Another reason to think that there really is a Determiner present comes from the basic paradigm of determiners in English. There is a morphological form for singular and plural definite determiners, and one for singular indefinite determiners, but not one for plural indefinite determiners. Given that we would expect there to be an indefinite plural form based on this paradigm, the fact that there isn't just looks accidental (i.e. given this paradigm, we would expect that the plural indefinite determiner is present, but crucially unpronounced).

Definite and indefinite determiners in English
	Definite	Indefinite
Singular	the bird	a bird
Plural	the birds	∅ birds

Moreover, clearly determiners are adding some meaning, so we need to have some way to introduce whatever meaning determiners contribute. How could we do that if there were no determiner with plural indefinites? Indeed, there are many languages which do have plural indefinite determiners, like French, and those DPs mean the same thing as English DPs.[footnote]A note about glossing conventions: we use parentheses in combination with an asterisk in two different ways. Written with the asterisk on the outside, *(XP) indicates that the sentence is ungrammatical without XP. Written with the asterisk on the inside, (*XP) indicates that the sentence is ungrammatical with XP.[/footnote] [latex] \setcounter{ExNo}{6} \exg. Marie a v\^u *(des) oiseaux dans le parc\\ Marie has seen \textsc{indef.pl} bird.\textsc{pl} in the park\\ \trans `Marie saw birds in the park.' [/latex] We therefore will assume that "bare" nouns in English and any language have a silent determiner. For a language like Swahili, which does not have a word for the or a, we will assume that they are DPs as well. [latex] English: \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ children}} }; \end{tikzpicture} Swahili: \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ mtoto \\ `child'}} }; \end{tikzpicture} [/latex] Finally, based on the same distributional facts, proper names are also DPs with a silent D. They must be DPs because they are interchangeable with pronouns and constituents like the doctor. [latex] Adam: \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Adam}} }; \end{tikzpicture} Kansas: \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Kansas}} }; \end{tikzpicture} \end{multicols} [/latex] Interestingly, we in fact find that in many languages proper names do appear with Determiners. This is famously true of Greek. [latex] \ex. \ag. O Yanis plithike\\ the John wash.3\textsc{sg}\\ \trans `John washed.' \bg. i Maria ekapse ti soup\\ the Mary burn.3\textsc{sg} the soup\\ \trans `Mary burned the soup.' [/latex]

[embed]https://youtu.be/ASiPyfI7LMY[/embed] DP-Hypothesis (length: 1m 22s) Credit: Kade Traffas

[embed]https://youtu.be/yH6Nr-9VwlU[/embed] Trees with DPs (length: 57s)

Key Takeaways about Determiners

All NPs are inside of a DP (= the DP Hypothesis).
Sometimes D is silent.
Proper names are DPs with a silent D..
Pronouns have category D.

Tense

In this section, we'll consider the category T, for Tense.[footnote]In many theories, this category is simplified as "I" for "Inflection." T and I are interchangeable for all intents and purposes. The reason that some prefer the term I is that this category is associated with subject agreement cross-linguistically, which is inflectional material.[/footnote] In Swahili, we saw evidence that the subject does not form a constituent with the verb. That evidence is repeated below. (8) shows that we can coordinate the verb and object, to the exclusion of the subject.

[latex] \setcounter{ExNo}{7} \ex. \ag. Juma aliona mti\\ Juma see.\textsc{past} tree\\ \trans `Juma saw a tree.' \bg. Juma aliona mti na aliona kilima\\ Juma see.\textsc{past} tree and see.\textsc{past} hill\\ \trans `Juma saw a tree and saw a hill.' [/latex]

So where is the subject? This question in fact has a number of different answers depending on which theory you choose. Our idea about the position of the subject will change considerably over the course in this class. This is not because I wish to "trick" you, or to teach you something incorrect. Throughout this book, our understanding of the position of the subject will evolve as we learn more about what is, and is not possible, in Human Language. (We also need to introduce some concepts like "movement," which we don't yet have the tools for. We'll get there in chapter 7.)

We're going to switch to English for a moment now, and then return to Swahili presently. Look at the following sentences. [latex] \setcounter{ExNo}{8} \ex. \a. Adam {\underline{will}} cook curry. \b. Charthel {\underline{must}} walk home. \c. Cammie {\underline{might}} paint a portrait. [/latex] These elements don't seem to belong to any of the categories we're familiar with. They aren't verbs because they don't inflect for tense/aspect/person and they can't stand on their own in a sentence: *Adam will curry. We must conclude that they are a distinct and unique class of words. Notice also that will is the marker of future tense in English. We therefore hypothesize that these words have the category Tense, abbreviated T.

The following sentences also reveal something important about will, must, and might.

[latex] \setcounter{ExNo}{9} \ex. \a. Adam will cook curry and bake a cake. \b. Carthel must walk home and do the dishes. \c. Cammie might paint a portrait and hang it on the wall. [/latex]

The takeaway from this data is that the verb phrase forms a constituent which is distinct from tense. In sum, the data indicate that,

VPs form a constituent which is distinct from Tense, and
Tense changes the category of the phrase. The constituent will cook curry is not a VP, because it doesn't have the distribution of a VP. For instance, we cannot simply put a plain verb like walk in the same place as will cook curry: *Adam walk.

We conclude therefore that TP is a distinct phrase from VP, and that VP is inside of TP. Thus, the Tense Phrase (TP) rule for English looks like the following.

Tense Phrase in English

TP → DP T VP

Our TP rule says that the subject is in TP. This is something of a simplification for now, but it's a move that allows us to get rid of a problematic category: S. You may have noted that S is the sole category that appears to violate the Headedness Principle, since it doesn't have a head that defines the phrase. With the addition of the TP rule, we've fixed this issue. Now all of our phrases obey Headedness! [latex] \setcounter{ExNo}{10} \ex. \a. Adam will cook curry. \b. \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Adam}} } } child {node{T \\ will}} child {node {VP} child {node {V \\ cook}} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ curry}} } } }; \end{tikzpicture} [/latex] So what about the sentences in (12), in which will, must, might are absent? Do these sentences lack a Tense Phrase? [latex] \setcounter{ExNo}{11} \ex. \a. Adam cooked curry. \b. Carthel walked home. \c. Cammie painted a portrait [/latex] Of course these sentences have tense! They're past tense. Here, the categories of T and V are expressed as a single word: cooked is a word that is made up of two parts: the verb /cook/ and the past tense marker /-ed/. We will assume for the time being that there is some "morphological magic" that puts words together. (We will return to this issue in chapter 7.) [latex] \setcounter{ExNo}{12} \ex. \a. Adam cooked curry. \b. \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Adam}} } } child {node{T \\ $[$past$]$}} child {node {VP} child {node {V \\ cook}} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ curry}} } } }; \end{tikzpicture} [/latex]

The tree above introduces a new idea: a feature. Features are things like [past], [present], [singular], [plural], etc. Features play a prominent role in theories of [pb_glossary id="750"]morphology[/pb_glossary], where we have to talk about pieces of words. We typically say that the word walked consists of a stem /walk/ and a feature [past]. The pronunciation of [past] is -ed. In this class, we will use features to represent different "settings" of tense.

What else is Tense? In English, the auxiliary words be and have can also instantiate the category Tense (for now), as in the following sentences.

[latex] \setcounter{ExNo}{13} \ex. \a. Adam is eating curry. \b. Carthel has gone home. \c. Cammie has hit a home-run. [/latex] Keep in mind that our terminology is somewhat arbitrary. When we put words into categories, we're really just saying, "Here's a group of things that have the same properties." A noun (phrase) acts like a "noun." A verb (phrase) acts like a "verb." It's a separate question whether the words will, might, can, is, and has form a semantically coherent group (and whether the semantics is that of tense).

Ways to pronounce Tense in English (so far) Besides tense features, the following are some ways to pronounce T. [latex] \setcounter{ExNo}{14} \ex. Adam \{ might, should, could, can will, ... \} cook curry. [/latex] The verb be also patterns as a tense element (when it is inflected). [latex] \setcounter{ExNo}{15} \ex. Mary is coming to school and singing a song. [/latex] And lastly, the word have is sometimes a tense element---whenever it is used as an auxiliary with some other verb. [latex] \setcounter{ExNo}{16} \ex. \a. Adam has eaten sushi. \hfill \textit{has} = tense \b. Adam has a dog. \hfill \textit{has} $\neq$ tense [/latex] The difference here is really the distinction between functional and lexical elements. In (17a), has is functional, while in (17b) it's lexical, denoting the meaning "possesses."

Returning to our Swahili examples, we notice that, just like English past tense, Swahili expresses tense as part of the verb. Tense is a verbal [pb_glossary id="826"]prefix[/pb_glossary]. Every finite verb in Swahili consists minimally a subject agreement prefix, followed by a tense marker, followed by a verb. (There are other things that can go in there as well; we'll cover them later.) [latex] \setcounter{ExNo}{17} \ex. \ag. mti u-li-anguka\\ tree 3\textsc{sm}-\textsc{past}-fall\\ \trans `The tree fell.' \bg. mti u-na-anguka\\ tree 3\textsc{sm}-\textsc{pres}-fall\\ \trans `The tree falls.' \cg. mti u-ta-anguka\\ tree 3\textsc{sm}-\textsc{fut}-fall\\ \trans `The tree will fall.' [/latex]

How are words formed? This question gets to heart of a broad and fascinating subfield in linguistics called morphology or morphological theory. Many researchers adopt the idea that word-formation is inextricably tied up with syntactic structure. Words are formed in the syntax, through syntactic processes. Under this theory, words are really a product of the syntax; they are the result of building structure.

On the other hand, many researchers think that words are formed in the lexicon, before syntactic computation. Words are first created, and then inserted into a structure already forms. The debate is one of the most heated in linguistics, and the two camps are quite distinct, and have argued vehemently against the other. If you're interested, consider taking Morphology LING 527, which explores these issues in-depth.

Key Takeaways about Tense

Tense is a projection above the verb phrase.
Tense can either be a word (might, will, can, ...) or just a feature ([past])
The subject is inside of TP.

Complementizers

The complementizer phrase (CP) is the phrase that "houses sentences." The function of C, broadly, is to specify whether the sentence is declarative (a statement), interrogative (a question), imperative (a demand), or subjunctive (a non-actual event). (Altogether, these are different Moods.)

C's function is also to relate an embedded proposition to the main clause. So the clearest place to find complementizer phrases is in front of embedded clauses, as shown in the following Swahili examples.

[latex] \setcounter{ExNo}{18} \ex. \ag. Juma alisema [\textsubscript{CP} kwamba kuna joto ]\\ Juma say.\textsc{past} {} C there.is heat\\ \trans `Juma said that it's hot.' \bg. Jemal anafikiri [\textsubscript{CP} kwamba kuna upepo ]\\ Jemal think.\textsc{pres} {} C there.is wind\\ \trans `Jemal thinks that it's windy.' \cg. Mariamu anaamini [\textsubscript{CP} $\emptyset$ kuna jua ]\\ Mariam believe.\textsc{pres} {} C there.is sun\\ \trans `Mariam believes that it's sunny.' [/latex] Notice that sometimes we don't pronounce the complementizer, like in (19c)---but that doesn't mean it's not there. Recall that we can have silent determiner, and silent tense elements, so it isn't really a concession to posit a silent complementizer.

English has a few complementizers, notably that, whether, and if. Other languages can have more or fewer. In fact, Swahili has at least two others, whose meanings are difficult to determine (for a linguist---every Swahili speaker knows how to use them properly!).

As stated earlier, the function of a complementizer is to specify the "kind" of sentence that is being uttered. Thus, many languages have complementizers whose function is to signify that the sentence is a question. In Swahili, je at the beginning of a sentence signifies that it's a Yes/No question (20). In Malayalam, the question particle is attached to the verb. [latex] \setcounter{ExNo}{19} \begin{flushleft} Swahili \end{flushleft} \exg. Je Juma alicheka?\\ C.\textsc{ques} Juma laugh.\textsc{past}\\ \trans `Did Juma laugh?' Malayalam (Dravidian) \exg. Mary wannu-oo?\\ Mary came-C.\textsc{ques}\\ \trans `Did Mary come?' Polish (Slavic) \exg. Czy Marta lubi koty?\\ C.\textsc{ques} Martha like.3\textsc{sg} cats\\ \trans `Does Martha like cats?' Tetun (Austronesian) \exg. \'O la b\'a sekola k\'a?\\ you not go school C.\textsc{ques}\\ \trans `Didn't you go to school?' \hfill (Bailey, 2013) [/latex] Thus, unembedded clauses can have complementizers as well.[footnote]And in fact, many languages allow complementizers to be used on unembedded sentences in non-question contexts. These often have "exclamative" or [pb_glossary id="805"]evidential[/pb_glossary] meanings.[/footnote]

So where do we need a complementizer? In fact, there is a strong consensus that every TP has a CP on top it. (There's exactly one exception to this generalization, discussed in chapter 11.) Thus, a simple sentence like Mary ate churros or the Swahili sentence below both have a silent C over TP.

[latex] \setcounter{ExNo}{24} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Mary}} } } child {node {T \\ $[$past$]$}} child {node {VP} child {node {V \\ ate}} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ churros}} } } } }; \end{tikzpicture} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ Juma}} } } child {node {T \\ $[$past$]$}} child {node {VP} child {node {V \\ alikula \\ ate}} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ ugali}} } } } }; \end{tikzpicture} [/latex]

This should probably sound surprising to you, since the following sentences are absolutely not possible in English nor Swahili. That is, we can never pronounce the silent C at the top of the sentence.

[latex] \setcounter{ExNo}{26} \ex. \a. *That the hiker bought a pencil. \b. *Whether the river flooded. \c. *If the hideout is in a cave. \ex. \ag. *Kwamba kuna joto\\ C there.is heat\\ \trans [Intended: `*That there is heat.'] [/latex] Still there is very good evidence that that all TPs have CPs on top of them. Consider this (somewhat complex) argument from English.

A complementizer must be present when forming a question. The presence of C_ques is why we have [pb_glossary id="806"]subject-auxiliary inversion[/pb_glossary] in English. The question-complementizer forces the auxiliary to displace to a different position (for reasons we'll discuss in chapter 7).
- [ C_ques [ John will see Mary ] ] → [ C_ques+will [ John will see Mary ] ]
Recall that we cannot coordinate different types of phrases. So I cannot coordinate an NP and PP: *the dog and on the porch. I cannot coordinate a VP and an AdvP *sing a song and loudly.
Now consider the following sentence. What is being coordinated using but?
- The Lakers made it to the finals, but will Lebron be healthy enough to play?

The answer is that what's being coordinated is a CP. But that's only possible if the sentence [ The Lakers made it to the finals ] is a CP.

The way we think of C is that it "types" a clause. It distinguishes between declarative statements (assertions of fact), interrogative statements (questions), demands (imperatives), among others. That is, depending on which C you choose, you'll get either an assertion, or a question, or an imperative, or something else. Thus, every clause needs something to tell it what kind of a clause it is.

There are non-finite complementizers as well. In English, the non-finite complementizer is for. It heads clauses with infinitives.

[latex] \setcounter{ExNo}{28} \ex. \a. The shark waited [\textsubscript{CP} for the fish to sleep. ] \b. Carol prayed [\textsubscript{CP} for the Lakers to beat the Celtics ] \c. The doctor hoped [\textsubscript{CP} for the swelling to reduce. ] [/latex] In Romance languages, the non-finite complementizer is de/di. [latex] \setcounter{ExNo}{29} \begin{flushleft} French \end{flushleft} \exg. Il est important de faire tes devoirs\\ it is important C to.do your homework\\ \trans `It's important to do your homework.' [/latex]

[embed]https://youtu.be/uLl4ZwkYVpo[/embed] Trees with silent functional categories (length: 1m 21s)

Students are often confused by sentences like the following. [latex] \setcounter{ExNo}{30} \ex. \a. That Erin likes sushi surprised me. \b. For Erin to eat sushi would surprise me. [/latex] Students mistakenly think that the complementizers at the start of the sentences (that and for) are the highest complementizers in the sentence. This is incorrect. The trick here is that the these CPs are subjects. So these sentences have the following structures. [latex] \setcounter{ExNo}{31} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} child {node {CP} [sibling distance = 4em] child {coordinate(left)} child {node {that Erin likes sushi} edge from parent[draw=none]} child {coordinate(right)} } child {node {T \\ $[$past$]$}} child {node {VP} child {node {V \\ surprised}} child {node {DP} child {node {D \\ me}} } } }; \draw (left)--(right); \end{tikzpicture} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} child {node {CP} [sibling distance = 4em] child {coordinate(left)} child {node {for Erin to eat sushi} edge from parent[draw=none]} child {coordinate(right)} } child {node {T \\ would}} child {node {VP} child {node {V \\ surprise}} child {node {DP} child {node {D \\ me}} } } }; \draw (left)--(right); \end{tikzpicture} [/latex]

[embed]https://youtu.be/LNqdRc_c-9s[/embed] Trees with clausal (CP) subjects (length: 2m)

Key Takeaways about Complementizers

Every TP is in a CP (because C determines the "type" of sentence it is).
C is often silent, but in many languages it can also be pronounced as a separate word, or as part of the verb

Where we're going

We've introduced functional categories. In doing so, we've taken a first look at how we can use syntactic tests to infer the presence of structure. Sometimes we don't pronounce everything in the tree---but that doesn't mean it's not there. We've seen evidence for null D, null T and null C. There is real empirical motivation for postulating null elements in syntax. Inferring the presence of hidden things is the bread-and-butter of syntax. There is no human language on earth which uniformly transparently tells us "this is a noun phrase," or "now I'm pronouncing the verb phrase," etc. Instead, we as syntacticians must deduce the presence of such constituents. If we're doing it correctly, then we're using empirical data to formulate a hypothesis. Hopefully that hypothesis is testable. Before moving on, I want to emphasize one other thing about the functional heads we've looked at. There is a certain parallelism between D, T, and C. In all three cases, the functional head combines directly with some constituent, and in fact, the functional element is in a sense dependent on that constituent. Tense "needs" a verb phrase because the meaning contribution of tense is to situate the event of the verb relative to a time. Determiners "need" an NP because they "determine" the NP. And C needs a TP because it situates a type of clause. D, T, C are meaningless without NP, VP, and TP, respectively. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D}} child {node {NP}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {T}} child {node {VP}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C}} child {node {TP}}; \end{tikzpicture} [/latex] This parallelism is not accidental. As we'll talk about in the next chapter, there are certain regularities across categories which allow us to make broad generalizations about language. Indeed, the similarities across structures will be one motivation for adopting the X-bar schema introduced in the chapter.

Advanced

DP: Gender and Number

There are potentially dozens, perhaps even hundreds, of functional categories. For many syntacticians every piece of inflectional morphology corresponds to a distinct functional head. For instance, consider languages which have a productive noun class system. These are languages in which every noun falls into a particular class, as indicated by the morphology that appears on the noun and possibly by how the noun agrees with other elements in the sentence. All Romance languages have such a system, typically called "gender." In Spanish, it is almost always the case that nouns ending in -o are in the "masculine" class, and nouns ending in -a are in the "feminine" class.[footnote]It is important to keep in mind though that there is nothing inherently "gendered" about this kind of noun class system; chairs, windows, trees, etc, are not inherently masculine feminine or otherwise.[/footnote] Regardless of the gender of noun in Spanish, plurals are almost always formed by adding an -s at the end of the noun. So each noun can potentially have two pieces of inflectional morphology, which correspond to two distinct heads in the syntax, schematized as below. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} [sibling distance = 7em] child {node {D \\ las}} child {node {NumP} [sibling distance = 5em] child {node {GenP} [sibling distance = 3em] child {node {NP} child {node {N \\ \textit{puert-} \\ `door'}} } child {node {Gen \\ \textit{-a}}} } child {node {Num \\ \textit{-s}}} }; \end{tikzpicture} [/latex] Note that the order that morphemes appear in the word tells us the order that they appear in the tree: closer to the root means lower in the tree. Thus, the gender information, which always attaches directly to the root of the word (in this case, puert- 'door') is under the head that expresses plurality.

Functional Hierarchies and Cartography

The sequence of functional heads in Spanish D > Num > Gen > N is called a functional hierarchy. This term refers to the hierarchical ordering of functional elements in a tree. In Spanish (and in fact, as a crosslinguistic rule), information concerning noun class (i.e., gender) always appears lower than information concerning number (i.e., singular or plural).[footnote]In linguistics, we use the term "number" to refer to (non-)plurality, not to the numbers 1, 2, 3, 4, ... .[/footnote] The idea of a functional sequence of heads is most famously found in the verbal domain. Guglielmo Cinque, studying the acceptable sequences of adverbial elements, has proposed the following universal sequence of functional projections.[footnote]Cinque, Guglielmo. 2006. Restructuring and Functional Heads: The Cartography of Syntactic Structures. Oxford: Oxford University Press.[/footnote]

MoodP_{speech act} > MoodP_evaluative > MoodP_evidential > ModP_epistemic > TP_Past > TP_future > MoodP_irrealis > TP_anterior > ModP_alethic > AspP_habitual > AspP_{repetitive(I)} > AspP_{frequentative(I)} > ModP_volition > AspP_{celerative(I)} > AspP_terminative > AspP_continuative > AspP_perfect > AspP_{retrospective} > AspP_proximative > AspP_durative >AspP_progressive > AspP_prospective > AspP_inceptive(I) > ModP_obligation > ModP_ability > AspP_{frustrative/success} > ModP_permission > AspP_conative > AspP_{completive(I)} > VoiceP> AspP_{repetitive(II)} > AspP_{frequentative(II)} > AspP_{celerative(II)} > AspP_{inceptive(II)} > AspP_{completive(II)} > V

In this approach, each kind of adverbs appears in a dedicated position in the structure. This approach is called cartographic, because it "maps" the functional projections. For instance, evidential adverbs like clearly (Sasha clearly lied) always appear above epistemic adverbs like probably (Sasha probably lied). Thus, Sasha clearly probably lied but not *Sasha probably clearly lied. Cartographic approaches to functional hierarchies clearly "explode" the trees that build. For theorists, the functional hierarchy describes every tree. That is, the sequence of adverbial heads above is always projected in every tree, whether there is an adverb there are not. For other theorists, the projections are on a "use-them-when-you-need-them" basis. You only need a phrase for AspP_{frequentative(I)} if you actually have an adverb like often in your tree. Obviously in this class we won't assume the extreme version of the cartographic approach: you do not need to draw every single phrase when you're making your trees. But we are implicitly assumed a more moderate approach. For instance, it will become clear later that, when drawing trees DPs in languages like Spanish, we'll want to always have heads for Gen and Num, even when those heads aren't necessarily pronounced. So the singular la puerta will have the following structure, where we assume a null exponent for Num. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} [sibling distance = 7em] child {node {D \\ la}} child {node {NumP} [sibling distance = 5em] child {node {GenP} [sibling distance = 3em] child {node {NP} child {node {N \\ \textit{puert-} \\ `door'}} } child {node {Gen \\ \textit{-a}}} } child {node {Num \\ $\emptyset$}} }; \end{tikzpicture} [/latex] The reason for this is that we still interpret singular number, even if we don't pronounce it.

Where we were, where we're going

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000

We have attempted to find a formal theory for relationships between words, starting from the observation that we can create generalized rules that describe sentences. Recall these sentences from Chapter 1.

[latex] \ex. \a. The mailman slept. \b. The storm rages. \c. A student cried. \d. The senator lies. [/latex]

Using the sentences in (1), we were able to define categories and then define rules over categories. From this starting point, we continued to abstract further. Generalizing over Phrase Structure Rules led to X'-theory. We further refined this model with theories of binding and movement.

Major themes

Throughout this book, a few topics kept re-occurring.

Hypothesis testing. Each step along the way required evidence. The theoretical assumptions are built on empirical observation. This is just good science --- it shouldn't be done any other way.
How much is syntax? That is, where does syntax end and morphology and semantics begin. The answer is not always clear. There are some things, like selection, which appear to be semantic in nature. But they clearly matter for syntax. Likewise, do-support seems to be a morphological "reaction" to syntax.
There are levels in syntax. That is, sometimes things are pronounced in a different place from where they start. Put another way, we need a theory that can account for silence created by movement.
Local vs. non-local dependencies. In general, we created a theory that emphasized local interaction: For instance, selection, we discovered had to happen locally. But then there are some things that appear to interact at a distance, e.g., Binding Theory and Movement.

In one sense, the end goal of (generative) syntax is to find the theory that needs the least amount of stuff. What is absolutely necessary, and what can be framed as simply a product of more processes? As I see it, there are two major things that any syntactic theory needs.

A theory of constituent structure. How do you put pieces together? How do you partition up the words into phrases, and phrases into bigger phrases?
A theory that encodes non-local dependencies. How do we relate pieces of structure across distance? For instance, how do we encode the relationship between an antecedent and an anaphor? And how do we encode the relationship between a trace and moved element? Ideally, both of binding theory and movement (of all types) can be reduced to a single mechanism.

In Syntax II, we'll try to design a theory that reduces to our grammar to just the minimal amount of operations. This new approach is called Minimalism.

The Minimalist Program

What is Minimalist syntax? As the name implies, Minimalism (or the Minimalist Program) is an attempt to discover what the roots of grammar are. As our syntactic theory developed in the mid-eighties, we developed a system of modules, or sub-components of the theory, that interacted to derive the empirical data that we were starting to discover. As a theory, Government and Binding (or generally the Principles and Parameters Framework) were extraordinarily in successful in providing explanations for phenomenon. The problem was that the explanations themselves weren't very plausible. We had contrived a theory that incorporated an enormous of "stuff," and we assumed all human children were born with all this stuff. It was a bit like saying that we're born with a programming language that had a ton of built in functions. Like maybe we're born with knowledge that Binding Theory exists. We're all born with the knowledge phrases can move---but also respect islands. But why? Why do we know this? Why is all of this knowledge universal across languages and speakers? Why did things end up this way, and not some other way? The goal of Minimalism is to ask which things we need and which can be derived. The Minimalist Program is an attempt to ask which things truly make up the core knowledge that we're born with, and which things are consequences of knowing this information. It is worth emphasizing that Minimalism is a program. It is a set of assumptions which guide our theoretical research. But as a theory, it is, in many ways, less successful than X'-theory (and the theories that used X'-theory).

Why this theory

juliedoner

Why this theory?

juliedoner — Wed, 22 Nov 2023 10:02:56 +0000

Over the next few sections, we will be learning a theory of syntax called Phrase Structure Rule grammars, which was in use from the 1950s-1980s. You may be wondering why we are learning an out-of-date theory. There are a few reasons why.

Theories change but the main concepts stay the same

No matter which theory we learn together, it will eventually become outdated. The most current theory also has dozens of variants and unresolved issues, so there isn't one 'correct' theory that I should definitely teach you. Instead, the most important thing is that you learn the core concepts of syntax and how to think like a syntactician, most of which stays the same across all theories.

Concrete first, then abstract

Phrase Structure Rule grammars are the oldest framework in generative grammar, as well as the most concrete. Over time, the theory got more abstract. It is easier to learn the more concrete kind of theory and then use that as a basis for understanding the more abstract frameworks than to start with the abstract frameworks from the beginning. As you will see, Phrase Structure Rule grammars explicitly encode the patterns found in the languages being studied, whereas the theories that came after used variables at first, and then stopped encoded the patterns within syntax at all. I find most students are better at analyzing sentence structure when they have experience working with Phrase Structure Rule grammars.

To understand why

The true goal in theoretical syntax is not just to draw syntax trees of particular sentences according to one framework. Instead, we want to understand what the best way to do that is and why. When we go back and learn the older ways of doing things, we can also learn what the weaknesses of those older frameworks are compared to the newer frameworks, and why we changed and adapted the theory. This helps us understand why the current framework is the way it is.

To learn flexible thinking

If you keep taking syntax classes, you will eventually be asked to read academic articles in theoretical syntax. Some of these papers will be older and some will be newer, but they will almost never have the exact same theoretical assumptions that you have. You will need to be able to 'translate' from one framework to another as you read different people's ideas. This is important not only for you to understand their ideas, but also for you to be able to critique their ideas and decide what are the merits and weaknesses of their way of thinking about things. One day, you might even end up attending a 20-minute conference presentation where someone presents a modification of syntactic theory, and in just that short period of time, you might need to understand why they are modifying it and decide whether you think that's a good idea! By learning about more than one historical framework of syntax, we are building the foundation for the flexibility of thinking required to keep up with new ideas. Even if you don't continue on in syntax, learning how to adapt to different kinds of thinking and different models is one of the most important transferable skills that you will learn in your university education.

4.8. Becoming a linguist: Glossing signed language data

juliedoner — Tue, 16 Jan 2024 02:46:44 +0000

Signed language data is glossed quite a bit differently than spoken language data.

Describing the form of signs

In signed languages, the basic independent meaningful unit, the equivalent of a spoken language word, is generally an individual sign. Signs are typically articulated with the hands and arms, and so are called manual markers, from the Latin word for hand.

Manual markers

Manual markers are best described by breaking them down into four parameters: handshape, orientation, location, and movement. We will only briefly describe manual markers here. ADD A VIDEO EXAMPLE HERE AND DESCRIBE IT THROUGHOUT THIS SECTION

Handshape

The handshape of a sign is how the fingers are configured. Handshape can be desribed in prose, by describing the locations of the different fingers or by including an icon depicting the shape of the hand. Since many handshapes are used to represent numbers or letters from the writing system of the ambient spoken language, a common shorthand for describing handshapes is to use numbers and letters. Thus, since the first handshape in Figure 1 is used to represent the the English letter in ASL, so it can be called the “S handshape”. However, each signed language has different handshapes for different letters and numbers, as shown in Figure 1, and so this shorthand is ambiguous unless it is clear which signed language is being used. [caption id="" align="alignnone" width="2268"] Figure 1. Differences between ASL and STS in handshape meanings (Anderson et al. 2022 Section 3.8).[/caption]
Orientation
The hand, even while maintaining the same handshape, can be rotated, which is known as the orientation of the sign. The orientation of the hand is divided into two components: the palm orientation (which way the palm is facing, shown in blue in Figure 2) and the finger orientation (the direction the bones inside the hand are pointing, which is where the fingers would point when straightened, shown in orange in Figure 2). Note that if the fingers are bent, the finger orientation may be a bit confusing, but you can determine orientation by straightening out the fingers to see where the fingertips point. [caption id="" align="alignnone" width="2086"] Figure 2. Different orientations of the right hand as seen from the signer’s point of view (Anderson et al. 2022 Section 3.8).[/caption]
Location
The location of a sign is where in space or on the body it is articulated. Signs can be articulated in a variety of locations. The default location is neutral signing space, the area just in front of the signer’s torso, but locations can be nearly anywhere on the body. They tend to be around some specific part of the head, but other body parts are also possible locations.
Movement
All signs have a handshape, an orientation, and a location, and many signs also have movement, which is divided into two types: path and local movement. Path movement involves articulation at the elbow and/or shoulder and moves the sign from one location in the signing space to another. Local movement involves articulation below the elbow and changes the sign's handshape and/or orientation.
Non-manual markers
Signed languages do not only involve movement have the hands and arms. The rest of the body, especially the torso and the parts of the face, are also involved and are known as the non-manual markers. Non-manual markers may continue longer than a single sign, sometimes marking an entire phrase. Some examples of movements involved in non-manual markings include eye gaze changes, eyelid narrowing and opening, eyebrow raising and lowering, torso leaning and rotation, head tilting and rotation, cheek puffing, lip rounding and spreading, teeth baring, and more. Nearly any other body part can be a non-manual articulator, even the feet and buttocks in some signed languages, such as Adamorobe Sign Language in Ghana (Nyst 2007) and Kata Kolok in Indonesia (Marsaja 2008).
Encoding the form of signs
There is no commonly accepted transcription system for describing the form of signs. Several notation systems for signed languages have been constructed (see Hochgesang 2014 for an overview), but no single consistent standard has emerged, and the systems that do exist can be difficult to work with. In morphology and syntax, the form of a sign is not always relevant, so it may not be described at all. Instead, the sign is translated into a word in the ambient spoken language and is referred to by that word, which is typically written in all caps. When the form of a sign is relevant, typically texts will include videos, diagrams, and/or pictures of the sign, as well as describing the sign in words. If there is movement in the sign and video cannot be used, a series of pictures or superimposed arrows are used to indicate the movement.
Two-line signed language glosses
Signed languages are typically glossed with two-line glosses, which correspond to the second and third line of three-line spoken language glosses. A line providing the form of the signed language data is not included, due to the lack of a convention for transcribing signed languages. However, pictures, diagrams, or a video are often included in addition to the two-line gloss. The final line in the gloss provides a natural translation into the meta language, just as with spoken languages. In signed language glosses, all signs are glossed in all caps, both lexical and functional, unlike spoken language glosses, where only the functional words are in all caps. This is illustrated in (1), where the possessive marker, POSS, as well as the lexical signs BROTHER, OFTEN, and TRAVEL, are all in caps. Sources will vary as to whether small caps or regular caps are used.

(1) German Sign Language (DGS)

POSS₁ BROTHER OFTEN TRAVEL

'My brother often travels.'

the 1 should be subscript, but I can only make it go superscript - Hey, Julie, this should be fixed now with the CSS update - KL

(Baker et al. 2016: 95)

Indices
Signed languages tend to use space to track reference through indices (singular: index). An index is a spot in the signing space that is assigned to a particular referent. For first and second person reference, signers will point at themself or at the addressee. For third person reference, signers may point directly at the referent (called absolute reference) or at an arbitrary spot in the signing space that they have assigned to that referent by describing the referent and then pointing at that location. Indices are usually indicated with the word INDEX or the abbreviation IX, followed by a subscript that indicates which referent is meant. For example, in (2), the signer signs the word for Ottawa and then points in the location of Ottawa

(2) Inuit Sign Language (IUR)

OTTAWA INDEX-LOC_Ottawa CALL-ON-PHONE LONG AGO

'Long ago, I phoned Ottawa.' (referring to a shop in Ottawa)

I have the first line all in one cell. Does that make sense coding-wise? - I believe it should be fine in this instance, as screen readers can occasionally have difficulty interpreting text in all caps so the accessibility might be difficult for these examples. We'd need to test them to see how they are being read out. - KL OTTAWA SHOULD BE SUBSCRIPT - This should be fixed now as well - KL
(Schuit et el. 2011: 21)
For first, second, and arbitrary third person references, a subscript number is used to gloss distinct index locations. A 1 indicates that the index is pointed at the signer, a 2 that the index is pointed at the addressee, and a 3 that the index is pointed at some other location. Since more than one third person referent may be used in the same conversation, the third person referents are also marked with a letter to keep track of the distinct third person referents. Figure 3. REPLICATE SCHUIT ET AL FIGURE 3?

(3) Inuit Sign Language (IUR)

USE-ICE-AUGER₁ INDEX_3a USE-ICE-AUGER_3a INDEX_3a

'Long ago, I phoned Ottawa.' (referring to a shop in Ottawa)

ALL THE 1/3a SHOULD BE SUBSCRIPT - Fixed; this was a combo of the CSS update and the tag need to be added around the 1/3a. - KL How to get it to not wrap text? - The text should no longer be wrapping with the CSS update, just give it a test and let me know but it seems to be working on my end. - KL
(Schuit et el. 2011: 22)
Verbs are sometimes signed moving from one index location to another, which is often analyzed as subject and object agreement. These are glossed with a subscript number at the beginning of the sign to indicate its starting location and a subscript number at the end of the sign to indicate its endpoint position. For example, in (4), the verb FORCE is assigned moving from the signer to the addressee, while the sign GIVE is signed moving from the addressee towards the signer.

(4) American Sign Language (ASL)

1INDEX 1FORCE2 2GIVE1 MONEY

'I'll force you to give me the money.'

(Sandler and Lillo-Martin 2006: 36)

Classifiers
Another common morphosyntactic marker used in signed languages which requires special mention are the classifiers. Classifiers in signed languages are used to describe objects and actions as part of depicting signs. Depicting signs take advantage of the visual-spatial medium of signed languages to vibrantly describe situations through schematic representations. Particular handshapes are used to indicate different types of objects and entities, and they are placed and moved throughout the signing space in an iconic manner to depict their relative placement and/or their movement through space. You can watch this video to see some examples of classifiers and depicting signs being used and described. REPLACE VIDEO WITH ONE WE MAKE AND HAVE CONTROL OVER? In depicting signs, different handshapes are used as classifier signs for different kinds of objects or different aspects of the same object. Each of these handshapes is not specific to the depicting sign or to a specific object. Instead, a particular classifier handshape is used for a whole class of objects. These depicting signs may indicate the whole entity and how it moves through space, how people hold or handle the entity, a particular body part of the entity, or the size and shape of the entity. Classifiers are typically indicated in glosses with the abbreviation CL, followed by a colon and some indication of the handshape. The handshape may be indicated with a small image of the handshape (5a), with the name of the handshape according to a particular signed language alphabet (5b), by the name of the object (5c), or by a description of the class of objects (5d).

(5) American Sign Language (ASL)

a. TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:👌3a

'Did you give Sue a flower two days ago?'

(5) b. TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:F3a

'Did you give Sue a flower two days ago?'

(5) c. TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:FLOWER3a

'Did you give Sue a flower two days ago?'

(5) d. TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:LONG-THIN-OBJECT3a

'Did you give Sue a flower two days ago?'

(Baker et al. 2016: 338)
Since depicting signs make use of space and iconic movements, they are difficult to gloss. Instead, they are often shown through images and video.
Other symbols
Some other symbols are common in signed language glosses. These are listed in the table below.
Table 1. Common symbols in signed language glosses (adapted from Baker et al. 2016: 338-339)

Symbol Description Meaning

SIGN++ two following plus symbols Indicates that the sign is repeated in a morphological process called reduplication.

SIGN - - - - - five following hyphens Indicates that part of a sign is held in place while the other hand continues signing.

SIGN^SIGN two signs joined by a caret Indicates the formation of a compound sign.

SIGN-SIGN two signs joined by a hyphen Indicates that multiple words in the meta language are required to translate a single sign in the object language.

s-i-g-n lower case letters joined by hyphens Indicates a fingerspelled word.

"xxx" lower case letters enclosed in quotation marks Indicates a gesture.

WOULD BE GOOD TO HAVE A VIDEO (OR SET OF VIDEOS) PAIRED WITH GLOSSES WHICH ILLUSTRATE THESE FEATURES
Glossing non-manual markers
Since non-manual markers occur simultaneously to the manual markers and might last for several signs, they are usually indicated above the gloss with a line to indicate how long the non-manual marker is articulated. At the end of the line, an abbreviation is included to indicate which non-manual marker was used. These might indicate the gesture itself, such as hn for 'head nod' which indicates affirmation or an imperative sentence, or the meaning of the gesture, such as y/n for yes-no questions, which is often indicated by raising the eyebrows. Some examples of non-manual markers are shown below in (x), where a backward head tilt, a marker of negation in some Eastern Mediterranean signed languages, is indicated with the abbreviation bht. Note that the length of the line matters! In (xa), only the sign HURRY is negated. The first part of the sentence, about going to work, is not negated. In (xb), on the other hand, the whole sentence is negated. The length of the line, then, can change the meaning of the sentence. Greek Sign Language (ENG) (47a and b) Baker et al. p. 139 - I have added a new template for the sign language glosses built using the French Sign Language gloss for reference. You can find the template and code for it in the back matter: Gloss Template HTML and CSS – The Linguistic Analysis of Word and Sentence Structures. It still needs to be tested, but the CSS for this textbook has already been updated with the styles for the new sign language gloss template. - KL PUT BOTH THESE IMAGES IN GLOSS TABLES Multiple non-manual markers may be articulated simultaneously, in which case they may be stacked on top of each other, as shown below in (y), where the eyebrows are raised for the entire sentence, while the head nod extend only over the final two signs. FRENCH SIGN LANGUAGE (LSF) b example only, Baker et al p. 135)

(5) American Sign Language (ASL)

a. TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:👌3a

'Did you give Sue a flower two days ago?'

(13) French Sign Language (LSF)

hn

re

PLEASE INDEX2 BITE

'Please, bite it.'

Some common abbreviations for non-manual markers in ASL are listed below in Table 2. Note that pairings between gestures and meanings are likely to vary from language to language, and even within ASL. The gesture and meaning pairs described below are some of the more common ones used in ASL.
Table 2. Common abbreviations for non-manual markers in ASL (adapted from Baker et al. 2016: 339-340)

Abbreviation Gesture Meaning

aff repeated head nods affirmation

bl body lean in a particular direction

hn a single head nod affirmation or imperative marker

neg headshake and negative facial expression negation

re raised eyebrows marks subordinate clauses

t raised eyebrows and forward tilt of the head topicalized constituent

wh lowered eyebrows content (wh-) question

y/n raised eyebrows and head forward yes-no question

/xxx/ various mouth gestures

)( sucked in cheeks 'small'

() blown out cheets 'big'

Key takeaways

Type your learning objectives here.

First

Second

Check yourself!

References and further resources

Attribution
Portions of this section are adapted from the following CC BY NC SA source:
↪️ Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Chapter 3: Phonetics. Essentials of Linguistics, 2nd edition. Pressbooks. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/part/3-phonetics/

For a general audience

Melissa Cayton Coda THAT! 2021. Advanced ASL: Classifiers "How to describe things?! Learn American Sign Language! YouTube. https://www.youtube.com/watch?v=gifQclSut_k

Academic sources

📑 Baker, Anne, Beppie van den Bogaerde, Roland Pfau, and Trude Schermer. 2016. Appendix 1: Notation conventions. The Linguistics of Sign Languages: An Introduction. Amsterdam: John Benjamins. 337–341.

Hochgesang, Julie A. 2014. Using design principles to consider representation of the hand in some notation systems. Sign Language Studies 14(4): 488–542.

Marsaja, I. Gede. 2008. Desa Kolok: A deaf village and its sign language in Bali, Indonesia. Nijmegen: Ishara Press.

Nyst, Victoria. 2007. A descriptive analysis of Adamorobe Sign Language (Ghana). University of Amsterdam PhD dissertation.

Sandler, Wendy, and Diane Lillo-Martin. 2006. Sign language and linguistic universals. Cambridge: Cambridge University Press.

Schuit, Joke, Anne Baker, and Roland Pfau. 2011. Inuit Sign Language: A contribution to sign language typology. Linguistics in Amsterdam 4 (1).

(1)	German Sign Language (DGS)
	POSS₁ BROTHER OFTEN TRAVEL
	'My brother often travels.'

(2)	Inuit Sign Language (IUR)
	OTTAWA INDEX-LOC_Ottawa CALL-ON-PHONE LONG AGO
	'Long ago, I phoned Ottawa.' (referring to a shop in Ottawa)

(3)	Inuit Sign Language (IUR)
	USE-ICE-AUGER₁ INDEX_3a USE-ICE-AUGER_3a INDEX_3a
	'Long ago, I phoned Ottawa.' (referring to a shop in Ottawa)

Tagalog Example

markc — Wed, 17 Jan 2024 21:15:29 +0000

(1) (a) kumaʔin ng mangga ang babaʔe [Tagalog; CREDIT?]

eat.PAST? NG mango ANG girl

'The girl ate a mango.'

(b) kumakaʔin ng mangga ang babaʔe

eat.PRESENT? NG dance ANG girl

'The girl is eating a mango.'

(c) kakaʔin ng mangga ang babaʔe

eat.FUTURE? NG dance ANG girl

'The girl will eat a mango.'

drafts and bits and bobs

juliedoner — Sun, 21 Jan 2024 06:32:03 +0000

(2) (a) kʊɽiː nət͡ʃt͡ʃiː [Punjabi][footnote]The Punjabi data in these examples was provided by an anonymous consultant.[/footnote]

girl.SG.FEM dance.PAST.SG.FEM

'The girl danced.'

(b) mʊndaː nət͡ʃt͡ʃɪaː

boy.SG.MASC dance.PAST.SG.MASC

'The boy danced.'

(c) kʊɽiː nət͡ʃt͡ʃeːgiː

girl.SG.FEM dance.FUTURE.SG.FEM

'The girl will dance.'

(d) mʊndaː nət͡ʃt͡ʃeːgaː

boy.SG.MASC dance.FUTURE.SG.MASC

'The boy will dance.'

11.2 Hierarchical structure with affixes

juliedoner — Wed, 14 Feb 2024 05:54:37 +0000
In the same way that compound words have hierarchical structure, affixes are also arranged in a hierarchical structure in relation to each other and to the roots that host them. Hierarchical structure with affixes sometimes encodes differences in meaning, similar to compounds, but it can also be the result of grammatical properties.
Scope

The mirror principle

Check yourself!

References and further resources

For linguistics students
🧠 Aze Linguistics. 2019. Morphology: Linearity, hence hierarchy. YouTube. https://www.youtube.com/watch?v=RXG0JNU_UBY
Academic sources
Baker, Mark. 1985. The mirror principle and morphosyntactic explanation. Linguistic Inquiry 16(3): 373–415.

11.3 Ordering patterns

juliedoner — Wed, 14 Feb 2024 05:54:58 +0000
In this section, we will look at some cross-linguistic morpheme ordering tendencies.
Derivation, compounding, and inflection

Derivation is closer to the base than inflection
As we discussed in Chapter 9, derivational affixes tends to appear closer to the base than inflectional affixes. For example, consider, the forms in (1). In (1a), the denominal verb criticizing takes the derivational suffix -ize before the progressive suffix -ing can be affixed. In contrast, in (1b), we cannot make the noun plural by adding an inflectional suffix before adding the derivational suffix -ize, even if the intended meaning is that there are multiple critics or multiple criticisms involved. Likewise, in (1c), the verb push is transformed into a noun with the agentive derivational suffix -er. Only after the derivational morpheme is affixed can the plural suffix -s can be added. The reverse order is ungrammatical, as shown in (1d).

(1) a. critic-iz-ing

b. *critic-s-ize

c. push-er-s

d. *push-es-er

In languages which predominately use prefixes, the reverse order occurs, showing that it is closeness to the stem rather than linear order that matters. (2) example with prefixes
Compounding is closer to the base than inflection
The roots that participate in compounding tend to be bare, uninflected roots. Afterwards, the resultant compound may inflect. For example, in (3a-b), it dish in dishwasher must be in singular form, even though dishwashers normally wash multiple dishes at once. Likewise, in (3c-d), the pop in popcorn cannot inflect for past tense, even if the popcorn has already been popped.

(3) a. dish-washer

b. *dish-es-washer

c. pop-corn

d. *popp-ed-corn

There are a few notable exceptions to this pattern. For example, the plural of passerby is prescriptively passersby and in-law compounds such as brother-in-law are prescriptively pluralized by affixing the plural to the kinship term, creating brothers-in-law.
Derivation and compounding appear in either order
It is difficult to find unambiguous examples of derivation and compounding which demonstrate which order they appear in, but it seems that they appear in either order. An example where derivation appears to occur before compounding is the word personal protective equipment. In this example, the verb equip takes the noun-forming derivational suffix -ment, the verb protect takes the adjective-forming derivational suffix -ive, and the noun person takes the adjective-forming derivational suffix -al. Then, protective and equipment combine, and finally, personal and protective equipment combine to form a complex word. (4) tree An example where compounding appears to occur before derivation is in the word rubbernecker, meaning 'a driver who is distracted by a traffic accident'. Here, we can tell that rubber and neck combine to create a compound verb to rubberneck first, since necker isn't a word (at least not with this meaning). (5) tree
Class I and Class II affixes

Native and borrowed affixes
Non-native affixes tend to go closer to the stem than native ones

Determining hierarchical structure

juliedoner — Wed, 14 Feb 2024 05:55:48 +0000

Making tree diagrams electronically

juliedoner — Wed, 14 Feb 2024 05:56:11 +0000

Chapter 11 practice exercises

juliedoner — Wed, 14 Feb 2024 05:56:41 +0000

Chapter 6 practice exercises

juliedoner — Thu, 23 May 2024 05:36:33 +0000

Data analysis

give data with one or two morphological processes, they have to name what kind of process it is

simple morphology problem sets

English allomorphic pattern and they have to say whether it is phonological/weak suppletion/strong suppletion and what kind of conditioning

What kind of morphological pattern is used to mark whether the following words are nouns or verbs?

(1) noun form verb form

a. brood breed

b. doom deem

c. food feed

Communication and study skills
read Zwicky paper on n't. What are the tests for telling about clitics from inflection? Test them on genitive 's. Do you think morphology marked via prosodic changes in a word should be classified as a kind of affix or a kind of non-concatenative morphology? Or do you think that this is not a meaningful distinction?
Research and application
WALS maps 26 - prefixing vs suffixing and 27 - reduplication English is classified as "no productive reduplication". Is English unusual in this regard or is this common? Why isn't there a category for partial reduplication only? run one of the data sets from part 1 through chatgpt. what did it get right? what did it get wrong? look up a suppletive form in the OED and read the etymological description. Where did the suppletion come from?
Language journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise x.
What kinds of morphemes does your language use? Look for examples of different kinds of affixes, suprafixes, non-concatenative morphology, reduplication, and clitics.

9.5. Case study: Inuit words for snow

juliedoner — Wed, 05 Jun 2024 06:46:16 +0000

Content warning: This case study is about commonly believed false information about how many words for 'snow' there are in the languages spoken by the Inuit. Much historical discussion of this false information, including some of the references listed here, refers to the Inuit people by a word that is now considered a slur. It is also reasonable to assume that the perpetuation of this false information was fueled at least in part by tendencies to exoticize Indigenous peoples. In this case study, we will investigate the popularity of the misinformation about Inuit words for 'snow', showing how it is based on a misunderstanding of wordhood, the othering of Inuit peoples, and poor citation practices.
describe the myth
The problems with counting Inuit words for snow

What is the Inuit language?
dialect continuum stretching from Greenland to Siberia (how far?), multiple languages (list them), multiple dialects they don't all use the same words
How do we count words?
sounds easy -- just grab a dictionary and start counting which dictionary? meterological dictionary or ...? what are we counting? distinct word forms? lexemes? words in the same word family? do we count all the words? technical/specialized terminology? archaic words? slang words? PLUS they're polysynthetic --- essentially infinite words on any topic
What counts as a word for snow?
discuss some examples, such as "house-building materials" compare: how many words for snow does English have?
Why is it even interesting?
Imagine an Inuit person came to your city, explored around, entered a beauty salon, browsed the shelf of nail polish colours, and returned home to report that your community has "X words for green!"

The Sapir-Whorf Hypothesis

How it happened

Spotlight on citation practices

Key takeaways

First

Second

Check yourself!

References and further resources

For a general audience

Pullum, Geoffrey. 1991. The great Eskimo vocabulary hoax. The Great Eskimo vocabulary hoax and other irreverent essays on the study of language. Chicago: The University of Chicago Press. 159-171.

Kaplan, Lawrence. 2003. Inuit Snow Terms: How Many and What Does It Mean? In: Building Capacity in Arctic Societies: Dynamics and shifting perspectives. Proceedings from the 2nd IPSSAS Seminar. Iqaluit, Nunavut, Canada: May 26-June 6, 2003, ed. by François Trudel. Montreal: CIÉRA -- Faculté des sciences sociales Université Laval. Available at https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php#Note%201.

Academic sources

Martin, Laura. 1986. "Eskimo words for snow": A case study in the genesis and decay of an anthropological example. American Anthropologist 88 (2): 443-444.

The colonial history of linguistics

juliedoner — Tue, 25 Jun 2024 02:51:22 +0000
a lot of social science research throughout history had as its goal to justify racial hierarchies and eugenics Boas (1912) is credited in debunking it (in anthropology) but Haitian anthropologist Anténor Firmin debunked it first (often overlooked), (Fluehr-Lobban 2000) Charity Hudley et al. 2020: e214
. Over the past century anthropologists

have sought to debunk the biological concept of race (e.g. Boas 1982 [1940], Marks

2017)

Missionary linguistics

from Jena's dissertation "Historical studies have explored both Catholic and Protestant missionary linguistics in Mexico (Hanks 2010; Heath 1972; Warren 1983), North America (McKevitt 1990; Tomalin 2008), Europe (Salmon 1985), Africa (Irvine 2008; Pennycook and Makoni 2005), and Australia(McGregor 2008). Wonderly and Nida (1963), Grey (2000), and Errington (2001) have also considered missionary linguistics in worldwide longitudinal perspective."

Suggestions from the twitterlings thread:

Errington, Joseph 2001 Colonial Linguistics. Annual Review of Anthropology 30:19-39.

Wonderly, William L., and Eugene A. Nida 1963 Linguistics and Christian Missions. Anthropological Linguistics 5(1):104-144.

Gray, Edward G. 2000 Missionary Linguistics and the Description of 'Exotic' Languages. In History of the Language Sciences: An International Handbook on the Evolution of the Study of Language from the Beginnings to the Present, Volume 1. S. Auroux, E.F.K. Koerner, H. Niederehe, and K. Versteegh, eds. Pp. 929-937. Berlin: Walter de Gruyter.

Pennycook, Alastair, and Sinfree Makoni 2005 The Modern Mission: The Language Effects of Christianity. Journal of Language, Identity, and Education 4(2):137-155.

And a bunch of Wesley Leonard’s excellent papers, stuff like “Indigenous Peoples, Ethics, and Linguistic Data” (2022) and “Refusing ‘Endangered Languages’ Narratives” (2023), among others.

From the new Oxford volumes, cite liberally!!! to point readers toward those volumes but e.g. Montoya “Manifestations of Colonialism in Linguistics and Opportunities for Decolonization Through Refusal” and Thomas “Racialization, Language Science, and Nineteenth-Century Anthropometrics”

also Epps and Ladley 2009 is a personal fave (syntax, souls or speakers?)

Four models of community engagement
Charity Hudley et al. 2020: e201
"Further difficulties arise when researchers

examine linguistic systems that they themselves do not participate in, because

they may not be aware of or may overlook important cultural and sociopolitical distinctions,

dimensions, or insights that bear on the research question at hand—not to mention

knowledge of cultural norms that are crucial both in building trust and partnerships

with members of oppressed groups and in conducting ethically responsible research.

Too often, those whom Speas (2009) and others refer to as ‘outsider linguists’ may be

overly focused on disciplinary norms and audiences and may fail to recognize the limits

of their knowledge and its relevance for racially minoritized communities; the resulting

situation creates ethical as well as theoretical and methodological problems,

despite the good intentions those researchers may have.
Charity Hudley et al. 2020: e221
white scholars may feel no obligation to advance social change through their research,

unfairly leaving this task to their colleagues of color in ways that can negatively impact

these scholars’ advancement, due to traditional systems of evaluation and reward (see

Structural racism in the linguistics of the recent past
Charity Hudley et al. 2020: e201
As sociological research has demonstrated, institutional whiteness is a structuring

force in academia, informing the development of theories, methods, and models

in ways that reproduce racism and white supremacy as structural social processes rather

than simply acts of individual ill will (e.g. Zuberi & Bonilla-Silva 2008).

Charity Hudley et al. 2020: e209

In sum, many scholars with a deep interest in language and race center their own

scholarship and professional activity not in linguistics departments, but rather in departments

of communication, English, modern languages, or education, where they

often receive a warmer reception, more support to pursue research on race, and more

opportunities to take part in professional initiatives that advance racial justice. These

disciplines, though often devalued within linguistics (Baker 2010, Mallinson & Charity

Hudley 2018), collectively provide a model for linking theoretical racial knowledge to

the real-world contexts in which language users go about their day talking, writing,

learning, and expressing themselves.

...

Indeed, some of the most significant theoretical work

on race has been conducted not in traditional disciplines (such as those we discuss

above) but rather in interdisciplinary fields that were established as deliberate alternatives

to the ‘quotidian white supremacy’
...
The departments

and programs housing these fields—though typically institutionally marginalized and

underresourced compared to predominantly white disciplines—are invaluable to their

campuses, both for creating innovative, engaging, and inclusive curricula for students

and for shouldering a disproportionate amount of the work of training, mentoring, hiring,

and advocating for scholars of color.

we are behind other closely related fields such as psychology, sociology, and anthropology Charity Hudley et al. 2020: e212
This disci -

plinary failure to recognize the racism within linguistics is largely due to the deeply

entrenched societal ideology that positions racism as intentional and individual, rather

than structural and often below the level of awareness of those who enact it (Hill 2008,

Hodges 2016). Because of the structural nature of racism, even well-intentioned white

people in ‘nice’ disciplines that espouse progressive goals inevitably contribute to its reproduction

(Ku bota 2002), especially when such stances are aligned with the widespread

racial ideology of colorblindness, or the belief that race is sociopolitically irrelevant and

that to discuss race is itself racist
Charity Hudley et al. 2020: e213
linguists of color (including Davis 2017, Green 2002, Leonard 2011,

Meek 2011, Morgan 1994, 2009, Rickford 2006, and Zentella 2018, among others)

have identified persistent scholarly practices in contemporary linguistic research that

diminish and distort the language of racially minoritized groups (see also work by white

linguists such as Walters 1996, Wolfram 2007).
...
the imposition of Eurocentric standards on non-

Indo-European languages and racially minoritized linguistic varieties
...
the devaluation of

the intellectual contributions and social agency of language users from marginalized and

racially minoritized groups. This is most evident in the long-standing pattern of taking linguistic

data from communities without providing adequate ‘service in return’, in the words

of Rickford (1997:182)—and using those data for the linguist’s own purposes, with little

or no consultation of community members’ own perspectives
Charity Hudley et al. 2020: e214
a deficit approach does not merely identify

but actively constructs differences across racial groups, contributing to biased

analyses.
Charity Hudley et al. 2020: e216
In addition, many existing psychological models related to race have taken a deficit

perspective, in which racialized groups are primarily studied as the objects of discrimination

and racism without a corresponding focus on community or individual responses,

alternative models, equity measures, and so forth.
Charity Hudley et al. 2020: e220
the fact that the vast majority of the sociolinguists who have historically carried

out research on African-American English have been white has contributed directly to

circumscribing what research questions have been asked and what findings have been

generated—as Wolfram puts it, the ‘academic exclusivity’ that has surrounded the study

of African-American English is a major factor that has contributed to the perpetuation

of sociolinguistic ‘folklore’ about the variety (2007:307; see also Spears 1999).
Charity Hudley et al. 2020: e221
compared to many other fields, linguistics

remains predominantly white, even twenty years after Rickford exposed this shameful

fact as ‘an academic limitation for our field as well as a socio-political embarrassment’

(1997:171). It may be more comfortable to convince ourselves that linguistics just isn’t

for everyone, but to do so is to abdicate our professional ethical responsibility to make

the discipline an equitable and inclusive place for students and scholars of all backgrounds,

and particularly for those whose communities provide a disproportionate

amount of the data that advance linguistic knowledge. Alienating undergraduate and

graduate curricula; narrow definitions of excellence in admissions, hiring, and promotion

decisions; and the constant policing both of what counts as linguistics and what counts

as research collectively create a hostile climate that drives away many linguists of color,

whether emergent scholars or seasoned researchers.
Charity Hudley et al. 2020: e224
Ideological divisions that

play out along differentially racialized cross-disciplinary and subdisciplinary lines

therefore stifle deep discussion and research around race and racism within linguistics

while also systemically marginalizing linguists from racialized groups to the detriment

of the discipline and the profession.

citations from Charity Hudley et al 2020

(Bolton & Hutton 2000, - on orientalism in linguistics

Errington 2001, - colonial linguistics

Leonard 2018); - decolonialism in lang doc

Zuberi and Bonilla-Silva 2008 - about how our methodologies have racist origins

References and further resources
The references in this section that have at least one Indigenous author are indicated with an ➡️.
For a general audience

➡️ 🔍 Bruchac, Margaret. 2018. Finding our dances: George Hunt and Franz Boas. In Savage Kin. Tucson: University of Arizona Press. 20–47.

Academic sources
Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz. 2020. Toward racial justice in linguistics: Interdisciplinary insights into theorizing race in the discipline and diversifying the profession. Language 96 (4): e200-e235.
Czaykowska-Higgins, Ewa. 2009. Changing fieldwork roles in community-based language research. [Audiorecording]. https://scholarspace.manoa.hawaii.edu/items/8a77aea9-e3a0-4f3a-a9ae-1d2a6de5a747

📚 Czaykowska-Higgins, Ewa. 2009. Research models, community engagement, and linguistic fieldwork: Reflections on working within Canadian Indigenous communities. Language Documentation and Conservation 3(10): 15–50.

➡️ Leonard, Wesley Y. 1998. Reflections on (de)colonialism in language documentation. Language Documentation & Conservation Special Publication No. 15: Reflections on Language Documentation 20 Years after Himmelman 1998, ed. Bradley McDonnell, Andrea L. Berez-Kroeker & Gary Holton. 55-65.

➡️ Leonard, Wesley Y., and Erin Haynes. 2010. Making "collaboration" collaborative: An examination of perspectives that frame linguistic field research. Language Documentation & Conservation 4: 268-293.

Rice, Keren. 2006. Ethical issues in linguistic fieldwork: An overview. Journal of Academic Ethics 4: 123–155.

➡️ Smith, Linda Tuhiwai. 2021. Decolonizing Methodologies: Research and Indigenous Peoples. 3rd edition. London: Zed books.

➡️ Tsikewa, Adrienne. 2021. Reimagining the current praxis of field linguistics training: Decolonial considerations. Language 97 (4): e293-319.

➡️ Woods, Lesley. 2023. Something's gotta change: Redefining collaborative linguistic research. Canberra: Australian National University Press.

Moving forward in a good way

juliedoner — Tue, 25 Jun 2024 02:51:47 +0000

Moving forward in a good way
fieldwork is community work The Decolonizing Linguistics book (Charity Hudley, Mallinson, and Bucholtz 2024) recognizes six tenets for their model of decolonization:

To decolonize is to humanize.

Remember and respect that both scholars and the language communities we research are made of people with their own desires, values, and needs. The goals of researchers should not come after those of the community. Researchers and community members should connect over their shared humanity.

To decolonize is to respect and respond to the local and the particular.

Colonialism comes in many different forms, and the experiences of marginalized peoples differ. A one-size-fits-all approach is not going to work.

To decolonize is to critically reflect on the past and present.

To decolonize is to return and recentre.

Return intellectual property and artefacts, pay reparations, and return control to marginalized communities.

To decolonize is to critically examine and transform our approaches.

To decolonize is to embrace refusal.

Ultimately, TEK [Traditional Ecological Knowledge] is related to Indigenous rights. Indigenous Knowledge [IK] cannot be separated from the people. This means that to protect IK or TEK, the people themselves and their ways of life must be protected. Instead, the focus has been on documenting the knowledge of Aboriginal people before they disappear. The focus on recording TEK has been driven by the idea that TEK is disappearing and it must be documented before it all disappears (Johnson 1992: 18). There is much discussion about the best way to protect Indigenous Knowledge. Perhaps it is not especially helpful to fall all over oneself to document TEK before Indigenous people disappear. Perhaps energies could be better spent assisting Aboriginal people to realize self-determination by protecting their rights so they do not disappear. This is a delicate and sensitive topic for Aboriginal and non-Aboriginal people alike because many Indigenous people have come to believe as well that their knowledge is disappearing. This is a dangerous ideology! I do not want my grandchildren to be learning about the knowledge and lives of their ancestors from published accounts or in museums or from non-Aboriginal academics. Our knowledge cannot be placed in a book or library; it does not work that way. I get a most uncomfortable feeling that this situation resembles far too much the movement to collect the bones of my ancestors and all their "artifacts" for museums, universities, and the life. This movement to collect "Indian artifacts" was based on the "vanishing race ideology" (Francis 1992, Cole 1985).
McGregor 2004: 398-399

Decolonial methods in data collection

A lot of linguistic theory has been built on Indigenous languages because they are typologically distinct from European languages and other better studied languages

Where did the data come from?

History of missionary linguistics work

Random anthropologists and philologists like Boas (historically language consultants being called informants, which has connotations of them being insiders who give them tips, not giving them respect, lasted until the 60s) Us vs. them

Now, linguist-focused methodology – personal relationship, consultant – still getting the data you want to get, maybe giving back some recordings but not giving them resources in the style that is useful for them – not really doing it for them, doing it on them How do i get the language to answer this theoretical question? How do I write a descriptive grammar for other linguists to read

More recent – how to make the relationships less extractive/colonial Your consultant is your research partner – what questions does the consultant want to ask? What questions does the community want to ask? What can the linguist do for the community?

Who owns the data? Linguists selling data to community that was freely shared with the linguist

In order for Indigenous people to do work on their own language in a way that is respected in the academic community, they need to conform to the academic way of doing things

Differences between Indigenous communities and other language communities

The community might have their own linguists

There isn’t a deadline on language documentation / language isn’t endangered

There isn’t as much limitations of resources

Cultural link, valuing the expansion of knowledge is a Western value that is not necessarily shared

History of racism, viewing people as specimen to study (including the residential school systems, where language was taken away from them)

Language endangerment is a direct result of colonialism

Connect back to reciprocity and Calls to Action/Calls for Action

Cite Keren Rice 2009, references of paper Clarissa sent, czaykowska-higgins

Charity Hudley et al. 2020: e215
a careful, critical, and decolonial version of the ethnographic method of participant-observation

is likely to yield more deeply collaborative research relationships across

racial difference than traditional elicitation-based methodologies (Dobrin & Schwartz

2016). Meanwhile, local understandings of Indigeneity as both a political and a cultural

category are necessary for navigating research in communities as well as for making

sense of linguistic data, including views of speakerhood that may differ from those of

linguists (e.g. Ahlers 2006, Davis 2016).
...
Scholars of language who rely on archival and historical

sources should also consider how the dynamics of colonialism and racism may

have influenced the nature of the data (cf. Makoni & Pennycook 2006).
"avoid empirical fetishism" Charity Hudley et al. 2020: e218 race and language are part of culture and ignoring that further marginalizes already marginalized people Charity Hudley et al. 2020: e220
One important consideration for any linguistic study, whether community-based or

lab-based, is the importance of avoiding simplistic characterizations of the race of participants

and instead clearly describing participants’ background and providing information

about the social and racial history of their community as part of the analysis.

References and further resources
The references in this section that have at least one Indigenous author are indicated with an ➡️.
For a general audience

Charity Hudley, Anne and Christine Mallinson. 2019. Statement on race. Linguistic Society of America. https://www.lsadc.org/statement_on_race#:~:text=There%20is%20no%20linguistic%20justice,scholars%20and%20communities%20of%20color.

➡️ 🦬 Ferland, Nicki, Anny Chen, and Gerardo Villagrán Becerra. n.d. Working in good ways: A framework and resources for Indigenous community engagement. Community Engaged Learning, University of Manitoba. https://umanitoba.ca/community-engaged-learning/working-in-good-ways

🦬 Thompson, Robert Moshe. 5 Dec 2022. What it takes to research an Indigenous language. The Manitoban. https://themanitoban.com/2022/11/what-it-takes-to-research-an-indigenous-language/44149

United Nations. 2007. United Nations declaration on the rights of Indigenous peoples. https://social.desa.un.org/issues/indigenous-peoples/united-nations-declaration-on-the-rights-of-indigenous-peoples

Academic sources

Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz, eds. 2024a. Decolonizing Linguistics. Oxford: Oxford UP. https://global.oup.com/academic/product/decolonizing-linguistics-9780197755259?cc=ca&lang=en&#

Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz, eds. 2024b. Inclusion in Linguistics. Oxford: Oxford UP. https://global.oup.com/academic/product/inclusion-in-linguistics-9780197755303?cc=ca&lang=en&

Czaykowska-Higgins, Ewa. 2009. Changing fieldwork roles in community-based language research. [Audiorecording]. https://scholarspace.manoa.hawaii.edu/items/8a77aea9-e3a0-4f3a-a9ae-1d2a6de5a747

📚 Czaykowska-Higgins, Ewa. 2009. Research models, community engagement, and linguistic fieldwork: Reflections on working within Canadian Indigenous communities. Language Documentation and Conservation 3(10): 15–50.
Kievit, Joyce Ann. 2003. A discussion of scholarly responsibilities to Indigenous communities. American Indian Quartely 27 (1/2): 3-45.
➡️ Leonard, Wesley Y. 1998. Reflections on (de)colonialism in language documentation. Language Documentation & Conservation Special Publication No. 15: Reflections on Language Documentation 20 Years after Himmelman 1998, ed. Bradley McDonnell, Andrea L. Berez-Kroeker & Gary Holton. 55-65.

➡️ Leonard, Wesley Y. 2021. Language reclamation through relational language work. International Conference on Language Documentation and Conservation. https://www.youtube.com/watch?v=2Oqye0gvCdA&list=PLTip382WwySK7FMxr_c19Kiw6VHxhcGNC&index=116

➡️ Leonard, Wesley Y., and Erin Haynes. 2010. Making "collaboration" collaborative: An examination of perspectives that frame linguistic field research. Language Documentation & Conservation 4: 268-293.

➡️McGregor, Deborah. 2004. Coming full circle: Indigenous knowledge, environment, and our future. American Indian Quarterly 28 (3/4): 385-410.

Rice, Keren. 2006. Ethical issues in linguistic fieldwork: An overview. Journal of Academic Ethics 4: 123–155.

➡️ Smith, Linda Tuhiwai. 2021. Decolonizing Methodologies: Research and Indigenous Peoples. 3rd edition. London: Zed books.

➡️ Starblanket, Tamara. 2017. 'Kill the Indian in the child': Genocide in international law. InIndigenous peoples as subjects of international law, ed. Irene Watson. Abingdon, Oxon: Routledge. 171-200.

➡️ Tsikewa, Adrienne. 2021. Reimagining the current praxis of field linguistics training: Decolonial considerations. Language 97 (4): e293-319.

➡️ Whalen, D.H., Margaret Moss, and Daryl Baldwin. 2016. Healing through language: Positive physical health effects of indigenous language use. F1000 Research 5:852.

➡️ 🔍 Wilson, Shawn. 2008. Research is ceremony: Indigenous research methods. Halifax: Fernwood Publishing.

➡️ Woods, Lesley. 2023. Something's gotta change: Redefining collaborative linguistic research. Canberra: Australian National University Press.

➡️ 📑 Younging, Gregory. 2018. Elements of Indigenous Style. Brush Education Inc.

1.1. What is morphology?

juliedoner — Fri, 20 Oct 2023 02:37:22 +0000
This textbook is about the structures of words and sentences. In the field of linguistics, the study of the structure of words is called [pb_glossary id="160"]morphology[/pb_glossary].
Technically, the word morphology means the study of form or shape. In the context of linguistics, this means the form and shape of words, but other fields of study use the word morphology as well. For example, in biology, morphology means the study of the form or shape of living things. For this reason, when you do Internet searches for sources in morphology, you will likely get better results if you include a word like "linguistics" or "language" in your search terms alongside morphology.
To study the structure of words, we look at their components. Let's take the word untidy, meaning 'not organized or clean,' as an example. Even though untidy is a single word, we can break it up into two pieces, un- meaning 'not,' and tidy meaning 'organized or clean.' These two components are called [pb_glossary id="161"]morphemes[/pb_glossary]. We define morphemes as the smallest meaningful units in language. When we break up a word into its morphemes, we say it has been [pb_glossary id="208"]segmented[/pb_glossary]. [caption id="attachment_539" align="aligncenter" width="225"] Figure 1: An untidy workspace. Photo by SN.CHE. Used under Pexels license.[/caption] Some words are [pb_glossary id="197"]simplex[/pb_glossary], meaning they only contain a single morpheme. The word pumpkin, a kind of squash, is a simplex word. Other words are [pb_glossary id="198"]complex[/pb_glossary], meaning they contain more than one morpheme. The word untidy, as we have already seen, is complex. Another word for complex is [pb_glossary id="201"]polymorphemic[/pb_glossary]. [caption id="attachment_537" align="aligncenter" width="225"] Figure 2: A selection of pumpkins. Photo by Allie Reefer. Used under Pexels license.[/caption] Note that we can break up the word untidy into even smaller pieces. For example, we can break it up into consonants and vowels, which we could write in the [pb_glossary id="192"]International Phonetic Alphabet (IPA)[/pb_glossary] as /ʌ n t ɑi ɾ i/. Likewise, we can break up the word pumpkin to get /p ʌ m p k ɪ n/. Although individual consonants and vowels do function as units in [pb_glossary id="194"]phonology[/pb_glossary], they are not meaningful units. Each of these individual consonants and vowels do not mean anything on their own, which is why they are not morphemes. We can also break words up into syllables. The word untidy has three syllables, /ʌn/, /tɑi/, and /ɾi/. The sequence /ʌn/ happens to be a morpheme in this word, but the other two syllables are not. A morpheme may be one syllable long, like un-; a morpheme may be longer than a syllable, like tidy; or a morpheme may even be shorter than a syllable, like the plural marker -s in the word cats! Just like consonants and vowels, syllables are structural units of phonology that do not encode meaning. Another possible definition of morphology is the systematic co-variation of form and meaning. In this definition, systematic means a pattern or system that is predictable or rule-governed. Co-variation means that when some element varies (in this case form), another element (meaning) varies in a predictable way at the same time. Form means the sounds or signs used to produce a word. In other words, morphology is the study of how, if you change the shape of a word, its meaning changes, too; and if you change the meaning of a word, its shape changes, as well. Let's consider again the morpheme un- in the word untidy. This morpheme shows up in lots of other words, as shown in Table 1. This is an example of systematic co-variation of form and meaning. In the first column, we have a set of words and their definitions. When we add un- to the beginning of each of these words, the meaning changes to mean the opposite. Adding un- is a variation in form, while adding 'not' to the definition is a variation in meaning. These two variations are related since the same change in form always leads to the same change in meaning, and vice versa.

Table 1. Words with un-

tidy 'clean, organized' untidy 'not clean, not organized'

happy 'in a pleasurable state of mind' unhappy 'not in a pleasurable state of mind'

like 'similar to' unlike 'not similar to'

fair 'equal, just' unfair 'not equal, not just'

On the other hand, the word pumpkin is simplex, meaning it does not have any components that exhibit a systematic co-variation in form and meaning. By coincidence, the word pumpkin is made up of two components that do show up in other words, pump and kin. On its own, pump means 'use pressure or suction to force air or liquid to move from one contained area into another', which has nothing to do with squash vegetables like pumpkins. Although the words pumpkin and pump share form, they do not share meaning. Their similarity is a coincidence. Likewise, the word kin, on its own, means 'family'. Again, this has nothing to do with squash vegetables and its similarity in form to pumpkin is a coincidence. Neither pump nor kin are morphemes in the word pumpkin. Some of the things we study in morphology include the following:

We look at related words to identify the co-variations in meaning and form and identify morphemes.

We classify morphemes based on their structural, functional, and meaning properties.

We determine whether a given string of morphemes is a word or a phrase.

We look at ways words are built or created.

We classify words according to the words they can appear alongside, their meaning, and their structure.

We look for common patterns in word structure and word formation across the languages of the world.

We study how the structure of words interacts with other components of language, such as [pb_glossary id="194"]phonology[/pb_glossary].

Morphology matters because a lot of language play and creativity happens with morphology, when we build new words and modify old ones. (For a fun example, check out this article explaining the grammar of doge!) We have intuitions about the words in our native languages and other languages we speak well. We have opinions about which words are good or come out sounding more awkward, which are funny and creative, which are confusing, and which are plain and boring. But how do we know these things? And which of these patterns are particular to specific languages, and which seem to show up in all languages? [caption id="attachment_541" align="aligncenter" width="200"] Figure 3: Language play. Photo by Thirdman. Used under Pexels license.[/caption]

Key takeaways

Morphology is the study of the structure of words or the systematic co-variation of form and meaning.

Words consist of morphemes. Morphemes are the smallest unit of meaning.

A simplex word only contains one morpheme, but a complex word contains multiple morphemes.

Check yourself!
[h5p id="8"]
References and further resources

Comedy and satirical linguistics

🎉 Wishnetsky, Susan. 2006. Morphemes: A new threat to society. Lingua Pranca. https://specgram.com/LP/26.coma.morpheme.html

For a general audience

🧠 CrashCourse. 2020. Morphology: Crash Course Linguistics #2. https://www.youtube.com/watch?v=93sK4jTGrss

🎉 McCulloch, Gretchen. 2014. A linguist explains the grammar of doge. Wow. The Toast. https://the-toast.net/2014/02/06/linguist-explains-grammar-doge-wow

For linguistics students

🔍 Anderson, Catherine and Bronwyn Bjorkman. 2020. Why we do morphology. Word to the Whys. Teaching in Linguistics Community of Practice. Podcast. https://sites.google.com/view/word-to-the-whys-podcast/home/morphology

1.2. What is syntax?

juliedoner — Fri, 20 Oct 2023 02:38:42 +0000
[pb_glossary id="215"]Syntax[/pb_glossary] is the study of sentence and phrase structure. In other words, syntax is the study of how we combine words and encode the relationships between them. A [pb_glossary id="240"]phrase[/pb_glossary] is any grammatically well-formed string of words. Some examples of English phrases are shown in (1).

(1) a. a pumpkin pie

b. down the stairs

c. very hungry

d. after the truly serious and unfortunate events of last night

e. rode their bikes to school every day

A [pb_glossary id="241"]sentence[/pb_glossary], on the other hand, is a phrase that forms a complete thought. Some examples of English sentences are shown in (2).

(2) a. A pumpkin pie rested on the windowsill.

b. I rolled a ball down the stairs.

c. Eliud Kipchoge was very hungry after winning the marathon.

d. I decided to talk to my therapist after the truly serious and unfortunate events of last night.

e. My children rode their bikes to school every day until it got too cold.

Both phrases and sentences consist of strings of words, but is any string of words allowed? Consider the strings of words in (3). Compared to (1), they are scrambled and do not make much sense. In linguistics, we put an asterisk (*) at the beginning of an example to indicate that it is not well-formed. From the examples in (3), we can see that the order of the words matter.

(3) a. *pumpkin a pie

b. *the stairs down

c. *hungry very

d. *after truly the serious unfortunate and night events last of

e. *their rode every day to bikes school

Interestingly, we are able to identify if a sentence or phrase is well-formed even if we don't understand all of the words used in the sentence. In (4), I give an example with [pb_glossary id="247"]nonce[/pb_glossary] (made-up) words. Even though you can't understand the sentence (because I just made up most of the words in it!), you can still tell that the words are in the right order and the sentence is well-formed.

(4) The osathies pishdupped the sudfy daigus.

Language profile: English

English is one of the most spoken languages of the world, with nearly 380 million first-language speakers and over 1 billion second-language speakers (Eberhard et al. 2023). It originated in England when people from northwestern Germany (the Angles, Saxons, and Jutes) invaded England in the fifth century, which had previously been inhabited by Celtic peoples. English belongs to the Germanic branch of the Indo-European language family. But it is also heavily influenced by some other languages. In 1066, the Norman French invaded England. They spoke a dialect of French and brought many French words into English, including court, crown, village, diamond, feast, and cream. Their influence also caused major changes in the grammatical structure of English. The variety of English spoken before the Norman invasion is known as [pb_glossary id="1010"]Old English[/pb_glossary], while the variety of English spoken afterwards is known as [pb_glossary id="1011"]Middle English[/pb_glossary]. We now speak a variety known as [pb_glossary id="1012"]Modern English[/pb_glossary], which is normally considered to have been spoken since around the year 1500 (Denning et al. 2007). Because the British Empire took control of so many countries worldwide, English is spoken in dozens of countries, each of which has its own dialects and varieties of English—often more than one! All of these dialects have unique linguistic properties and a rich cultural heritage. You can listen to audio recordings of many different dialects and accents at the International Dialects of English Archive.

Word order patterns vary from language to language. In English, we put [pb_glossary id="249"]prepositions[/pb_glossary] (words like in, at, on, and to) before the noun they go with. But in many of the world's languages, the same kind of word goes after the noun. In those languages, they are called [pb_glossary id="250"]postpositions[/pb_glossary] instead. Japanese and Turkish are both examples of languages with postpositions, as shown in example (5).

(5) a. Japanese

nihon ni

Japan in

'in Japan'

(5) b. Turkish

vapur ile

boat with

'with a boat'

(Pereltsvaig 2021: 98, 100)
In the Japanese example in (5a), the postposition ni meaning 'in' comes after the word nihon meaning 'Japan' to form the phrase 'in Japan.' In the Turkish example in (5b), the postposition ile meaning 'with' comes after the word vapur meaning 'boat' to form the phrase 'with a boat.'

Formatting linguistic data

Linguists use very specific conventions for formatting and discussing linguistic data. We will learn together about how to format and discuss examples in Chapter 4. However, as you read this textbook and other linguistics resources, you should pay attention to how examples are formatted and discussed, so that you can learn how to format and discuss data in your own work. Look for patterns and conventions, what works well, and what could be clearer.

Did you know? Adpositions

If you do not want to distinguish between postpositions and prepositions, you can use the word [pb_glossary id="253"]adposition[/pb_glossary]. According to Dryer (2013), about half of the world's languages use prepositions and about half of the world's languages use postpositions. A minority of languages do something else, such as not using adpositions at all, putting adpositions in the middle of words, or allowing adpositions to appear on either side of the noun.

Some of the things we study in syntax include the following:

We look at how words are grouped into sentences and phrases.

We classify phrases and sentences based on their structural, functional, and meaning properties.

We determine the possible word orders of different kinds of phrases in particular languages.

We determine whether words or phrases have relationships with other words or phrases in a sentence and how that affects their placement or their morphological form.

We look for similarities and differences in phrase and sentence structure across languages.

We create models to describe how each of the components of syntax interact to form phrases and sentences.

We study how the structure of phrases and sentences interacts with other components of language, such as [pb_glossary id="264"]semantics[/pb_glossary].

One of the things that sets human language apart from other types of communication is how we are able to create new, specific meanings by combining words in a structured way. We do not merely string words together and guess at or infer how the different words are related. Their relationships are explicitly encoded through syntax—which is why we can say unlikely or even illogical things and have the meaning come through clearly. This property is rather curious and unique. Emojis do not have it. When scientists have taught primates to sign, there is little evidence for syntactic structure in their communication. But without it, would we be able to have philosophical treatises or science fiction novels?

Key takeaways

Syntax is the study of phrase and sentence structure.

A phrase is any well-formed string of words, while a sentence forms a complete thought.

In linguistics, we put an asterisk (*) in front of phrases and sentences that are not well-formed.

Different languages have different word order patterns. For example, some languages have prepositions that go before the noun while other languages have postpositions that come after the noun.

Check yourself!
[h5p id="10"] Consult the following data in answering the check yourself questions:

(6) Boumaa Fijian

au na talai Elia i 'Orovou

I FUTURE send Elia to 'Orovou

'I'll send Elia to 'Orovou.'

(Dixon 1988: 216, as cited by Dryer: 2013)

(7) Lezgian

duxturrin patariw fena

doctors to went

'She went to doctors.'

(Haspelmath 1993: 218, as cited by Dryer: 2013)

References and further resources

For a general audience

🧠 CrashCourse. 2020. Syntax - Morphosyntax: Crash Course Linguistics #3. https://www.youtube.com/watch?v=B1r1grQiLdk

For linguistics students

Denning, Keith, Brett Kessler, and William R. Leben. 2007. English Vocabulary Elements. 2nd edition. Oxford: Oxford University Press.

🔍 Doner, Julie and Tim Mills. 2020. Why we do syntax. Word to the Whys. Teaching in Linguistics Community of Practice. Podcast. https://sites.google.com/view/word-to-the-whys-podcast/home/syntax

Pereltsvaig, Asya. 2021. Languages of the World: An Introduction. 3rd edition. Cambridge: Cambridge University Press.

Academic sources

Dryer, Matthew. 2013. Order of adposition and noun phrase. In WALS Online (v2020.3), ed. Matthew Dryer and Martin Haspelmath. Zenodo. http://wals.info/chapter/85

📑 Eberhard, David M., Gary F. Simons, and Charles D. Fennig, eds. 2023. Ethnologue: Languages of the World. 26th edition. Dallas, TX: SIL International. http://www.ethnologue.com.uml.idm.oclc.org

🔍 Meier, Cameron (executive editor). 2023. International Dialects of English Archive. https://www.dialectsarchive.com

1.3. Morphosyntax as a subfield of linguistics

juliedoner — Fri, 20 Oct 2023 02:39:11 +0000
Both [pb_glossary id="160"]morphology[/pb_glossary] and [pb_glossary id="215"]syntax[/pb_glossary] are related subfields of [pb_glossary id="216"]linguistics[/pb_glossary]. Linguistics can be defined as the scientific study of language. We will go deeper into the scientific aspects of [pb_glossary id="434"]morphosyntax[/pb_glossary] in Chapter 2.
The study of natural human language
What do we mean when we say language? Linguists primarily study natural human language. A [pb_glossary id="430"]natural[/pb_glossary] language is a language that developed within a community without being planned or designed. Instead, a natural language evolves gradually over time based on how the community uses it. In contrast, constructed languages are designed intentionally. [pb_glossary id="450"]Constructed languages [/pb_glossary] include languages designed for art, media, or fun, like Klingon for the Star Trek series, or for a communicative purpose, such as Esperanto, which was designed with the intention that it would be an international language that would promote peace. Constructed languages and non-human communication systems are valid and interesting objects of study. However, depending on the kinds of questions you are asking, they may or may not be able to provide the evidence you need. For example, you may be studying the properties that languages around the world have in common. A constructed language, however, may have been constructed on purpose to sound alien or be unlike human language. Even if it wasn't designed to be different on purpose, we cannot know whether the conscious choices of a language designer will result in the same kinds of patterns as subconscious language acquisition unless we study them separately and then compare. Some linguists hypothesize a genetic or biological component that underlies the human capacity for language. If this is the case, then we would expect there to be differences between human communication and the communicative practices of other species.
The place of morphosyntax within linguistics
Within the field of linguistics as a whole, morphology and syntax are the study of the mediation of form and meaning. The form of language -- the possible speech sounds and signs -- is determined at the level of phonetics and phonology. The meaning of language is determined at the level of semantics and pragmatics. In morphology and syntax, we study how to encode particular meanings with particular forms. Let's use the example in (1) to illustrate the relationship between form and meaning.

(1) The toddler kicked the balls across the field.

The form of the sentence in (1), if it is spoken, sounds similar to (2), when represented in an [pb_glossary id="192"]IPA[/pb_glossary] transcription, and its meaning describes something like the image shown in Figure 1.

(2) /ðətɑɾlɚkɪktðəbɑlsəkɹɑsðəfild/

[caption id="attachment_1287" align="aligncenter" width="300"] Figure 1: A toddler kicking a ball in a field. Photo by Paul Ney. Used under CC BY-NC-SA license.[/caption] But how do we get from the form in (2) to the meaning in Figure 1? Well, we divide up the string of sounds in (2) into morphemes and words, each of which has its own meaning. But that, alone, is not enough. The order of the words and morphemes and how they are combined also matter. Look at the examples in (3). Notice how they all have the same words and morphemes as (1), but combined in different ways. The different combinations produce different meanings, some of which make no sense.

(3) a. The balls kicked the toddler across the field.

b. The field kicked the balls across the toddler.

c. The kicker toddled the balls across the field.

d. The toddlers kicked the ball across the field.

In morphology and syntax, we are assigning a structure to a string of sounds or signs and using that structure to determine its meaning. This structure indicates the relationships between the morphemes and words. For example, the -s plural marker on balls in (1) indicates that there is more than one ball. On its own, -s just means more than one something, but we don't know what that something is until it is part of a structure. Likewise, the morpheme kick indicates that an action of kicking is involved, but until kick is combined with the past tense marker -ed, we don't know when it happened, and unless it is combined with a subject, object, and other modifiers, we don't know who did the kicking, what was kicked, or where the kicking happened.
The subfields of linguistics
Linguistics is very broad and covers every aspect of language structure and use. The lines are fuzzy, but we could roughly divide linguistics into subfields that focus on the structure of language and subfields that focus on how the structure of language intersects with other aspects of the human experience. All of these areas of research are important for advancing our understanding of how language works. There are also interactions between all of the different subfields. For example, you could study how to train a computer to produce artificial speech with the correct pronunciation (phonetics, phonology, and computational linguistics), how children acquire morphology (morphology and first language acquisition), or how a word gains a special meaning within a particular community and then spreads outside of the community (lexical semantics, historical linguistics, and sociolinguistics). Morphology and syntax, as well, can be studied in combination with any other subfield. Here are some examples:
Semantics
[pb_glossary id="264"]Semantics[/pb_glossary] is the study of meaning. One important idea in semantics is [pb_glossary id="479"]compositionality[/pb_glossary], which is the observation that the meaning of a sentence is derived by the meaning of its parts and the way that they are combined. This is very closely related to the study of morphosyntax since morphosyntax is what determines how they are combined.
Phonetics and phonology
The subfields of [pb_glossary id="455"]phonetics[/pb_glossary] and [pb_glossary id="194"]phonology[/pb_glossary] intersect with morphology through the phenomenon of [pb_glossary id="480"]allomorphy[/pb_glossary]. A morpheme may be pronounced in different ways depending on its phonological context. We will learn more about this in Chapter 12. The subfields of [pb_glossary id="455"]phonetics[/pb_glossary] and [pb_glossary id="194"]phonology[/pb_glossary] also intersect with syntax through [pb_glossary id="482"]prosody[/pb_glossary]. Many syntactic structures in various languages affect the prosody of the sentence. For example, in English, questions have a special rising intonation.
Sociolinguistics
In [pb_glossary id="463"]sociolinguistics[/pb_glossary], different dialects and language varieties are studied. Any language variety is a valuable object of study within the subfields of morphology and syntax. Some critics of morphosyntax point out that we do not adequately account for variation in language.
Psycholinguistics
In the subfields of [pb_glossary id="471"]psycholinguistics[/pb_glossary] and [pb_glossary id="484"]acquisition[/pb_glossary], linguists study the processing and acquisition of morphosyntactic patterns. Some critics of morphosyntactic theory argue that there is a mismatch between morphosyntactic models of grammar and the ways we seem to process language.
Anthropological linguistics
[pb_glossary id="465"]Anthropological linguistics[/pb_glossary] is the study of the interaction between language and culture. Different languages encode different aspects of the world directly into the grammatical structure of language. For example, some languages have grammatical categories that can be used to indicate formality or respect, while others do not.
Discourse analysis
In [pb_glossary id="459"]discourse analysis[/pb_glossary], linguists study how different sentences or utterances are combined into wider texts or conversations. Sometimes this involves the use of morphological or syntactic [pb_glossary id="485"]discourse markers[/pb_glossary], like oh, well, so, y'know, etc... Some syntacticians work on integrating discourse markers into syntactic structure, while morphologists might analyze their morphological structure.
Computational linguistics
A [pb_glossary id="477"]computational linguist[/pb_glossary] might study how to program a computer to produce and interpret language in a human-like way. Although the ability of computers to manipulate language has improved by leaps and bounds, there are still some areas of difference, if you look closely.
Crip linguistics
[pb_glossary id="467"]Crip linguistics[/pb_glossary] is not so much a subfield, but more a critical reflection on linguistics as a whole from the perspective of disability studies, challenging us to not consider any attempts at communicating to be deficient or defective. The term crip linguistics was coined by deaf linguists Octavian Robinson and Jon Henner (e.g., Henner and Robinson 2023) based on the reclaimed usage of crip in critical disability studies (e.g., McRuer 2006, Kafer 2013). In morphosyntax, we may investigate how various communicative behaviours exhibit morphological or syntactic patterning.
Which subfields of linguistics are missing here? How are they also connected to morphology and syntax?

Key takeaways

Linguistics can be defined as the scientific study of language.

The primary object of study in linguistics is natural human language.

In both morphology and syntax, we study how to encode particular meanings with particular forms.

Morphology and syntax interact with all of the other subfields of linguistics.

Check yourself!
[h5p id="11"]
References and further resources

For a general audience

🎉🔍 Okrent, Arika. 2010. In the land of invented languages: Adventures in linguistic creativity, madness, and genius. Toronto: Random House.

Academic sources

📚 Henner, Jon, and Octavian Robinson. 2023. Unsettling languages, unruly bodyminds: A Crip Linguistics manifesto. Journal of Critical Study of Communication and Disability 1 (1): 7–37. https://criticalstudycommunicationdisability.org/index.php/jcscd/article/view/4

Kafer, Alison. 2013. Feminist, queer, crip. Bloomington: Indiana University Press.

McRuer, Robert. 2006. Crip Theory: Cultural Signs of Queerness and Disability. New York: New York University Press.

2.7. Becoming a linguist: Citation in morphology and syntax

juliedoner — Fri, 20 Oct 2023 02:39:25 +0000
An important part of academic writing that is distinct from non-academic writing is the practice of [pb_glossary id="361"]citation[/pb_glossary]. Citations are the formulaic ways that academic authors attribute the original sources of the information, examples, and quotations in their work. Different fields use different citation styles. Some common ones you may have encountered in the past are APA (the American Psychological Association style) or MLA (the Modern Language Association style). Because linguistics is interdisciplinary, it doesn't really have a set style. You may find linguistics papers in either APA or MLA style, but most linguistics journals have their own style guide, which is usually similar to but more streamlined than APA. Because of this, most linguistics instructors aren't picky about which citation style students use in their papers as long as they include all of the necessary information in a clear and consistent format.
Why is citation important?
Citation is crucial to academic writing for many reasons. First of all, academic writing should not be done in isolation. When you do research, you do not want to start from scratch. If you can show that your ideas are related to and build on other people's ideas, it gives your work more validity. In academic writing, you should be entering into a community of scholarship, contributing to the body of work, and combining new ideas and old ideas. Proper citation makes it clear which ideas you have taken from others to build upon and which are your own ideas. Secondly, academic writing requires accountability to the community. As we are discovering and creating new knowledge, we are responsible to the community to be as accurate as possible. Scientific research, in particular, should be [pb_glossary id="386"]replicable[/pb_glossary], which means that others should be able to repeat our research methods and get similar results. Providing full documentation of our sources allows our colleagues to double-check our work and minimizes the chances that errors will spread. Thirdly, not providing proper citation is dishonest. When we cite others' research, we give them credit for the work they've done and show respect for their knowledge and effort.
In-text citations
In-text citations are included in the main text of a paper and indicate where each individual piece of information and quotation comes from. This allows you to show the interaction between different people's ideas with each other and with your own ideas. Some other fields use footnotes, but linguistics exclusively uses parenthetical citations. In-text citations indicate the author's surname (also known as family name) and the year of publication. Each in-text citation should have a corresponding entry in the bibliography or references section. Below you will find some examples of in-text citations, marked in bold, from Culicover (2014). Don't worry too much about what the examples mean; pay attention to the structure of the citation! The default way to include an in-text citation is to include the author's surname and the year of publication in brackets after the relevant piece of information. In the following example, the original source has two authors, Pearl and Sprouse.
There are Bayesian approaches, which essentially argue that a structure can be judged unacceptable if there is an alternative structure that is significantly more likely, other things being equal (Pearl and Sprouse 2013).
Culicover (2014: 486)

The author's name is also frequently included in the structure of the sentence, in which case only the year is in parentheses.
The distinction between grammaticality and acceptability was highlighted by Miller and Chomsky (1963), who observed that a sentence can be well-formed in the sense that it follows the rules of linear ordering, phrase structure, and morphological form, but is nevertheless unacceptable.
Culicover (2014: 466)

When should you incorporate the authors into your sentence and when should they be in parentheses? Most of the time, they should be included in parentheses. You want the source to fade into the background, and have the focus be on the information you're sharing. However, sometimes who said the information is important, for example if they were the first person to come up with a certain idea or if you're comparing their ideas with someone else's. In that case, it makes sense to incorporate the author into the sentence structure to give them more prominence.
You can also list multiple sources, as in the following example. In some style guides, they are separated by semi-colons (;), like in this example. In other style guides, they are separated by commas (,). A style guide might also instruct you to list them chronologically or alphabetically.
The unacceptability of center embedding has been generally attributed to processing complexity and not to grammar (Gibson 1998; Lewis 1997).
Culicover (2014: 466)

In-text citations should make it clear which pieces of information comes from which source, as well as which pieces of information are original to you. In the following example, you can see that the pattern was first described by Postal, and then the pattern was explained in two different papers by Wasow.
Wasow (1972; 1979) proposed traces as a way to solve a puzzling binding fact noted by Postal (1971).
Culicover (2014: 475)

If a whole paragraph or chunk of your paper is from the same source, you don't need to repeat the in-text citation after every sentence, as long as it's clear that the whole section comes from the same source. Citations need to show up more frequently when you're incorporating information from many different sources. That being said, synthesizing ideas from many different sources shows an advanced understanding of the subject matter and critical thinking and is something that you should strive for in assignments involving research.
When giving a direct quote or a very specific piece of information, it is good to include the page number. The page number is indicated after the year, separated with a colon (:). In the following example, there is a direct quote, as indicated by the quotation marks. The citation indicates that the quote is taken from page 161 of Bloomfield (1933). This example also illustrates that when you refer to the same source multiple times close together, you do not need to repeat the full in-text citation as long as it is clear which source you mean.
This view has its direct antecedents in the theory of immediate constituents (IC), in which the function of syntax is to mediate between the observed form of a sentence and its meaning: "we could not understand the form of a language if we merely reduced all the complex forms to their ultimate constituents" (Bloomfield 1933: 161). Bloomfield argues that in order to account for the meaning of a sentence, it is necessary to recognize how individual constituents (e.g., words and morphemes), constitute more complex forms, which themselves constitute more complex forms.
Culicover (2014: 467)

This final example illustrates two conventions. First of all, if there are multiple sources that give the information you are citing, you could list just a few of them as examples (the abbreviation e.g. means 'for example,' taken from the Latin phrase exempli gratia). Secondly, when there are more than two or three authors, instead of listing all of the authors, you should list just the first one and include the abbreviation et al. This comes from the Latin phrase et alia which means 'and others.' The full bibliography entry for Fillmore et al. (1988) shows that the paper has three authors: Charles Fillmore, Paul Kay, and Mary Catherine O'Connor.
Such a view has been explicitly formalized in Construction Grammar (Kay 2002), and has been widely adopted (see, e.g., Fillmore et al. 1988; Kay and Fillmore 1999; Goldberg 1995; 2006; Culicover and Jackendoff 2005; Sag 2012).
Culicover (2014: 485)

Common abbreviations in in-text citations

There are many abbreviations that may be used in in-text citations.

a.o. for among others. Use: If you consulted many, many sources that say something similar, you can list just a few, with a.o. on the end. Example: Syntax textbooks in the generative tradition often use many tree diagrams (Adger 2003, Carnie 2020, Cowper 1992, Larson 2009, a.o.).

cf. from the Latin confer 'compare'. Use: If you want to list several sources that are relevant to what you just wrote but say slightly different things (maybe you agree with them, maybe you don't, but they're worth looking at), you can introduce the citation with c.f. Example: There are many arguments for and against the theory of Universal Grammar (cf. Boxell 2016, Dąbrowska 2015).

e.g. from the latin exempli gratia 'for example'. Use: If there are multiple sources that say the same or similar things, and you just want to list one or a few of them. Example: Syntax textbooks in the generative tradition often use many tree diagrams (e.g., Carnie 2020).

et al. from the Latin et alia meaning 'and others'. Use: If your source has more than 2-3 authors, use et al. instead of listing them all. Some style guides will have specific instructions for how many authors to list and when to use et al. Example: Charity Hudley et al. (2024) has advanced the discussion on justice in linguistics research.

n.d. for 'no date'. Use: If your example does not have a date indicated. Be careful! If there is no date listed, then that is a clue that your source is not an academic source. On the other hand, sometimes my students cite something as n.d. when the date is available on the source, just tricky to find. Example: Most languages spoken in Europe belong to the Indo-European language family. One notable example is Basque, a language with no known living relatives (Etxepare Euskala Institutua, n.d.).

p.c. from 'personal communication.' Use: If you are giving credit to someone who told you directly (not in a published format) the information you are citing, you should give them credit with p.c. For example, if a classmate gives you an idea that you use in your assignment, you could give them credit with a p.c. citation. This can be used when the information was told to you directly in person, over email, over text, etc. You can also use p.c. if a linguist shares data with you that they had collected but not yet published. Example: The language Niuean has verb-initial word order (Diane Massam, p.c.).

sic from the Latin 'thus, so'. Use: If you're quoting something with an error in it, use sic to indicate that the error comes from the original text. Example: "There are several subsystems at work in [sic] when we use language" (Carnie 2021: 4).

The references list
The purpose of the bibliography or references list is to provide full citation details for every source in the paper, giving enough information for your reader to find your sources as quickly and efficiently as possible. The bibliography is usually done in a highly formulaic way so that it can be compact, not taking a lot of space, and so that, once you know the pattern, it is very quick and easy to navigate. When making a bibliography entry, the first thing you have to do is identify what kind of source you are dealing with. Templates for the main kinds of sources are provided below. As you look at templates here and in style guides, pay attention to capitalization, italics, and punctuation.
A book
You use the book citation style if the entire book is written by the same author. If each chapter is written by a different author, you should use the style listed below for a book chapter in an edited volume. The template for a book is as follows:
Author's surname, Author's given name. Year. Title of book in italics. City of publication: Publishing company.

Chomsky, Noam. 1995. The Minimalist Program. Cambridge, MA: MIT Press.

A book chapter in an edited volume
If a book has a collection of book chapters, each written by a separate author, you should cite the chapter you use, not the book as a whole. If you use more than one chapter, you should cite each of them separately. The template for a book chapter in an edited volume is as follows:
Author's surname, Author's given name. Year. Title of book chapter. In Title of Whole Book in Italics, ed. list of editors, page numbers of chapter. City of publication: Publishing company.
In the above example, the abbreviation "ed." stands for edited by and is used to introduce the editors.

Cardinaletti, Anna. 2004. Towards a cartography of subject positions. In The structure of CP and IP, ed. Luigi Rizzi, 124–174. Oxford: Oxford University Press.

Chomsky, Noam. 2008. On phases. In Foundational issues in linguistic theory: Essays in honor of Jean-Roger Vergnaud, ed. Robert Freidin, Carlos P. Otero, and Maria Luisa Zubizarreta, 133–166. Cambridge, MA: MIT Press.

An academic article
An academic article is a paper published in an academic journal. The template for an academic article is as follows:
Author's surname, Author's given name. Year. Title of article. Title of Journal in Italics. Volume number (Issue number): Page numbers.
Not all journals use issue numbers. If you can provide it, it can be very helpful in looking up the article.

Barbosa, Pilar. 2011. Pro-drop and theories of pro in the Minimalist Program Part 1: Consistent null subject languages and the Pronominal-Agr Hypothesis. Language and Linguistics Compass 5 (8): 551–570.

Lillo-Martin, Diane. 1986. Two kinds of null arguments in American Sign Language. Natural Language and Linguistic Theory 4: 415–444.

A dissertation
A dissertation is a research project completed in order to get a PhD degree, typically the length of a book. The template for a dissertation is as folllows:
Author's surname, Author's given name. Year. Title of dissertation. Doctoral dissertation, Name of Institution.

Jouitteau, Mélanie. 2005. La syntaxe comparée du breton [Comparative syntax of Breton]. Doctoral dissertation, Université de Nantes.

General patterns in bibliography entries
Here are some general formatting patterns worth noting for bibliographies in linguistics:

In English, we typically put the given name first and the surname last. However, if we are alphabetizing by last name, we typically reverse the order and put in a comma to indicate that the order has been reversed.

My given name is Julianne. This is the name my parents assigned to me when I was born. My surname is Doner. This is the name that was passed on to me from my parents and that I share with my siblings and parents. I typically write my name as "Julianne Doner." But if I am alphabetizing by last name, I would list it as "Doner, Julianne" with a comma.

Although many citation styles mandate only including the initial of the given name, when given the choice, many linguists opt for listing the full given name. There are several linguists with the same surname, so it is helpful!

Unlike APA style, we do not usually put the year in parentheses.

Titles of book chapters and articles are not in italics. Some style guides may put them in quotation marks, but we do not typically do so in linguistics.

Titles of books and journals are in italics.

Titles of one-time publications, including chapters, articles, and books, use sentence capitalization. This means only the first word of the title (and subtitle, if applicable) is capitalized, as well as proper names.

Titles of repeating publications, like journals, use title capitalization. This means that all the major words of the title are capitalized.

If the source is in a different language than the rest of the paper, it is useful to provide a translation of the source's title in square brackets, as shown for the Jouitteau example above.

Putting it all together
Here is how to format your bibliography list as a whole:

Each entry should have a hanging indent. A hanging indent is when the first line is not indented, but the rest of the lines are indented.

The bibliography entries should be listed alphabetically by author's surname.

If there are multiple sources from the same author, those sources should be listed chronologically, as shown with the multiple sources by Chomsky.

Very occasionally, you may find multiple sources from the same author and year. In that case, assign a letter of the alphabet (starting with 'a') to each one, which you put immediately after the year, as with the two papers by Barbosa in the example below. Your in-text citations should also include that letter.

Barbosa, Pilar. 2011a. Pro-drop and theories of pro in the Minimalist Program Part 1: Consistent null subject languages and the Pronominal-Agr Hypothesis. Language and Linguistics Compass 5 (8): 551–570.

Barbosa, Pilar. 2011b. Pro-drop and theories of pro in the Minimalist Program Part 2: Pronoun deletion analyses of null subjects and partial, discourse, and semi pro-drop. Language and Linguistics Compass 5 (8): 571–587.

Chomsky, Noam. 1982. Some Concepts and Consequences of the Theory of Government and Binding. Cambridge: MIT Press.

Chomsky, Noam. 1995. The Minimalist Program. Cambridge: MIT Press.

Chomsky, Noam. 2008. On phases. In Foundational issues in linguistic theory: Essays in honor of Jean-Roger Vergnaud, ed. Robert Freidin, Carlos P. Otero, and Maria Luisa Zubizarreta, 133–166. Cambridge, MA: MIT Press.

Jouitteau, Mélanie. 2005. La syntaxe comparée du breton [Comparative syntax of Breton]. Doctoral dissertation, Université de Nantes.

Lillo-Martin, Diane. 1986. Two kinds of null arguments in American Sign Language. Natural Language and Linguistic Theory 4: 415–444.

Schuit, Joke, Anne Baker, and Roland Pfau. 2011. Inuit Sign Language: A contribution to sign language typology. Linguistics in Amsterdam (ACLC Working Papers) 4.

Key takeaways

Citations are a formalized way to give credit to the sources of your work and should be used in all academic writing.

Each piece of information that you get from another source should be followed by an in-text citation that lists the author's surname and the year of the source in parentheses.

Academic papers should have a references section at the end of the paper that lists the full bibliographic information for all of its sources, detailed enough so that anyone else should be able to find the same sources. Each in-text citation in your paper should have a corresponding entry in the references list.

The field of linguistics doesn't have a single standardized citation format. Most linguistics professors don't care which format you use, as long as you are internally consistent and include all of the relevant information. In this section, we looked at the citation guidelines from the Canadian Journal of Linguistics.

Check yourself!
[h5p id="9"]
References and further resources

Reference materials

📑 Canadian Journal of Linguistics/Revue canadienne de linguistique. Style guide/Protocole de rédaction. https://www-cambridge-org.uml.idm.oclc.org/core/journals/canadian-journal-of-linguistics-revue-canadienne-de-linguistique/style-guide-protocole-de-redaction

📑 Unified style sheet for linguistics. Wayne State University. https://clas.wayne.edu/linguistics/resources/style

Sources for examples

Culicover, Peter. 2014. The history of syntax. In The Routledge Handbook of Syntax, ed. Andrew Carnie, Dan Siddiqi, and Yosuke Sato, 465–489. London: Routledge.

Dąbrowska, Ewa. 2015. What exactly is Universal Grammar, and has anyone seen it? Frontiers in Psychology 6.

2.1. Linguistics as science

juliedoner — Fri, 20 Oct 2023 02:42:23 +0000
In Section 1.3, we defined linguistics as the scientific study of language. In this section, we will learn why we consider linguistics to be a science. Often, when people think of science, they think of the physical sciences, which is the study of the physical properties of our universe. Although some subfields of linguistics, such as phonetics, include the study of physical properties, linguistics generally belongs to the category of social sciences. In the [pb_glossary id="507"]social sciences[/pb_glossary], we study the properties of societies. We can also categorize linguistics as a [pb_glossary id="510"]cognitive science[/pb_glossary], which is the study of the mind. Linguistics is the branch of cognitive science that studies what language can tell us about how the human mind works.
The scientific method
All of the types of science -- physical, social, and cognitive sciences -- are united by their methodology, which is known as the [pb_glossary id="184"]scientific method[/pb_glossary]. The scientific method consists of four parts:

Observe data.

Formulate a [pb_glossary id="496"]hypothesis[/pb_glossary].

Test the hypothesis against more data.

Revise the hypothesis.

The scientific method can be treated as a cycle. After you revise your hypothesis in Step 4, you continue to collect more data, and test and revise your hypothesis, getting more and more precise and accurate with each revision. The second time through the cycle, and every subsequent time, you observe the data you collected from testing your previous hypothesis to formulate a new, revised hypothesis.
In this textbook, we will be focusing on methodologies that use the scientific method. However, there are also other kinds of methodologies used within linguistics that do not use the scientific method and are equally valid. For example, in historical linguistics, the comparative method is used to reconstruct languages from before we have a written record. Although the comparative method uses rigorous and systematic methods, it is not possible to test the hypotheses developed in this method, because we have no record of the languages being reconstructed. Likewise, some other subfields, like discourse analysis and linguistic anthropology, use qualitative methods akin to textual analysis. The methodologies that you choose to use depend on the kinds of questions you are asking and the resources available. Every methodology has its strengths and weaknesses. Our understanding of language will be strongest if we, as a community of linguists, investigate it from multiple perspectives, using many different methodologies.

An example of the scientific method in morphosyntax
Let's look at an example as applied to morphosyntax.
Step 1: Observe data
The first step is to observe some data. We will start with the data in (1).

(1) a. We wrote a story.

b. You gave us a gift.

Sentence (1a) includes the pronoun we, which means something like 'a group of people including the speaker.' Sentence (1b) includes the pronoun us, which also means something like 'a group of people including the speaker.' So an observation we could make about this data is that we and us are very similar in meaning.
Step 2: Formulate a hypothesis
The next step is to make a hypothesis. Since we observed that we and us are very similar, one possible hypothesis could be that they are perfect [pb_glossary id="517"]synonyms[/pb_glossary].
Hypothesis 1: We and us are perfect synonyms.
This hypothesis makes a prediction. If we and us are synonyms, then we might expect them to be interchangeable. That is, we might expect us to be able to switch one for the other without changing the meaning of the sentence.
Step 3: Test the hypothesis
So let's test out the predictions of our hypothesis! We will take the examples from (1), but switch we for us and vice versa and see what happens.

(2) a. *Us wrote a story.

b. *You gave we a gift.

When we switch us and we, the result, shown in (2), doesn't sound quite right, so we mark them with an asterisk symbol (*) to show that the sentences are not well-formed. What have we figured out? Contrary to our first hypothesis, it doesn't seem like we and us are exactly the same. They are different in some way, which is why switching one for the other in (2) results in ill-formed sentences.
Step 4: Revise the hypothesis
The difference between the sentences in (1) and those in (2) is not a difference in meaning, but a difference in grammaticality. Perhaps, then, the difference between we and us is not a difference in meaning, but a difference in grammatical function.
Hypothesis 2: We and us have the same meaning but different grammatical function.
This hypothesis also makes a prediction. According to this hypothesis, if we change the structure of the sentence but keep the same meaning, it might affect whether we need to use we or us.
Step 5: Test the revised hypothesis
One way to change the structure of these sentences is to put them in [pb_glossary id="523"]passive voice[/pb_glossary]. If we take the sentences in (1) and change them to the passive voice, keeping the same pronoun, we get the sentences shown in (3). If, on the other hand, we make them passive but switch the pronouns, we get the sentences in (4).

(3) a. *A story was written by we.

b. *Us were given a gift.

(4) a. A story was written by us.

b. We were given a gift.

This time, the predictions we made in Hypothesis 2 are consistent with our data! Sentences (1a) and (4a) mean the same thing but have different structure, but (1a) must have we and cannot have us, but sentence (4a) must have us and cannot have we. Likewise, sentences (1b) and (4b) also mean the same thing but have different structures and also use different pronouns.
And so on...
Even though our Hypothesis 2 ended up passing our test, that doesn't mean we're done. There are two main ways we might continue to refine our hypothesis. First, notice that our Hypothesis 2 is rather vague. We have proposed that there is a difference in structure, but what exactly is this difference? If we were to continue on with this example, we might try to figure out how to describe the structural difference between us and we. Second, notice that we've only looked at eight sentences. You could even say we've only looked at two sentences and variations on those two sentences! That's not very much data! Another way to continue revising and refining our hypothesis would be to test it against even more data, and different kinds of data, to see if our hypothesis continues to stand.

Key takeaways

Linguistics is often classified as a science. It is sometimes classified as a social science and sometimes a cognitive science.

In morphosyntax, we often use the scientific method to investigate the properties of language.

The four steps of the scientific method are:

Observe data.

Formulate a hypothesis.

Test the hypothesis against more data.

Revise the hypothesis.

The scientific method can be considered an iterative cycle. We repeat it, revising our hypotheses and getting more specific with each cycle.

Check yourself!
[h5p id="16"]

2.3. Prescriptivism and descriptivism

juliedoner — Fri, 20 Oct 2023 02:42:58 +0000

How you language is beautiful. Don't let anyone tell you your language is wrong. Your languaging is the story of your life.
-Jon Henner

A lot of people, when they think about grammar, they think about a set of rules that they have to follow. Many people think they are "bad" at grammar. Others like to correct others' grammar. Some people may have grammar pet peeves that get them especially annoyed. This is known as [pb_glossary id="497"]prescriptive grammar[/pb_glossary]—creating and enforcing rules that tell people how they should use language. Linguists think about grammar in a very different way. Because we approach grammar scientifically, we do not make value judgments about the ways people use language. Instead, we observe it and try to describe it. This is known as [pb_glossary id="498"]descriptive grammar[/pb_glossary].
Language variation and identity
Language is always changing, and language is not uniform. Everybody uses language in slightly different ways, and sometimes in very different ways. We use these differences in the ways we use language to perform our identity—to let other people know how we fit into our society. The ways we use language can indicate to others where we're from, how old we are, whether we are cool or smart, our gender identity, and every other aspect of our identities that you can think of. The study of how our language varies is a major part of the field of [pb_glossary id="463"]sociolinguistics[/pb_glossary]. Unfortunately, not all identities are equally valued by society. The value society places on certain kinds of people often gets transferred to the language use associated with those people. Thus, the varieties of language used most often by the most powerful and prestigious people get viewed as "correct" or "standard" while the varieties of language used by the less powerful and prestigious people get viewed more negatively. However, when studied objectively, all varieties of language follow systematic rules. All varieties allow for clear and creative communication. They are all equally valid, and none of them are better or worse than the others. These different varieties do communicate different things, though. They communicate your identity, sometimes intentionally and sometimes unintentionally! They communicate whether you are rich or poor, whether you are well-educated or not, whether you fit in with a particular crowd or not. Likewise, the people you interact with perceive you in certain ways because of your language, sometimes intentionally and sometimes unintentionally. The most proficient language users aren't the ones who always use the most prestigious variety of language; they're the ones who are able to use many different varieties well and are able to choose which varieties are most effective for different contexts. It is important to remember, though, that not everyone has equal access to the prestigious varieties of language. Children do not get to choose where they are born and grow up, the language variety used by their peers or parents, or the quality of their education. The varieties of language you use has less to do with your intelligence or attention to detail, and more to do with the communities you grew up in. It is possible to learn another variety of language, of course, just as it is possible to learn another language. But children who grow up being exposed to the prescriptively correct and prestigious forms will not have to put in as much effort to produce the prescriptively correct forms in formal situations. Not exposing children to culturally significant varieties of language can also cause harm, though, as children need the tools to be able to explore and express their identities. Negative language attitudes are pervasive in our schools and in the media, and so it is natural to have internalized some of these attitudes yourself. However, it's worth thinking about where those attitudes came from and the power structures they uphold.
Does descriptivism mean I can say whatever I want? When we say that all varieties of language are valid, what we mean is that all varieties are systematic and rule-governed. It does not mean that our language has no social consequences. For example, a slur may be constructed in a way that follows morphological rules, but it is still rude to say and can cause harm to you, your community, and the people you say it to.

Grammar checking computer programs

It is likely that you have seen ads for computer programs that check your writing. Many word processors have basic spelling and grammar checkers built in. You can also buy subscriptions to more powerful and sophisticated ones, some of which are even powered by artificial intelligence. I encourage you, as a linguist and as a student, to critically engage with these grammar checking programs. Most likely, these programs blindly enforce prescriptivist rules without a deeper understanding of the patterns of language, including prescriptivist rules with classist and racist origins. If you do use a grammar checking program, think about every suggestion it gives you before adopting it. You might even consider looking at the settings of your program to see if you can customize some of your settings. For example, I have turned off the setting that marks passive voice as an error.

Case study on ain't
Let's look more closely at one non-prestigious grammatical construction, ain't. Ain't is widely used in several dialects of the United States, including Appalachian English and African American Vernacular English. The following quote describes its use in Appalachia. According to this description, using ain't is an important part of showing that you belong to the community.
Isn’t, aren’t, hasn’t, haven’t are strictly avoided. Ain’t or hain’t, depending on the need for emphasis, is used universally in the place of these “furrin” contractions. The bright mountain boy eager to improve his English but determined to avoid the use of ain’t, is constrained to use he’s not and they’ve not in lieu of he isn’t and they haven’t if he wishes to escape the damnatory label of being “stuck up.” -Cratis D. Williams (1962: 16), as cited in Montgomery (2014: 31).
Outside of Appalachia, however, ain't is considered prescriptively incorrect. But if you look more closely, you'll realize that it actually fills a gap in standardized English. Ain't is an example of a contraction, which is a shortened form of a common phrase into a single word. We use several contractions in standardized English, as shown in Table 1, but there is a gap! We don't have a contracted form of am not.

Table 1. Some contractions of standardized English

Full phrase Contracted form

am not ?

are not aren't

is not isn't

It is not surprising that some dialects of English have developed a contraction that fills that gap. Some dialects use ain't, while others, such as Irish English, use amn't.
Where do prescriptive rules come from?
You may have been taught many prescriptivist rules about "correct" language in school. If you studied English formally, you may have been taught some of the following rules. In this section, we will discuss where these rules came from.

Table 2. Examples of prescriptive rules in English

Rule Prescriptively incorrect example Prescriptively correct example

Do not split infinitives. I want to quickly eat so I’m not late. I want to eat quickly so I’m not late.

Do not end a sentence with a preposition. That’s the person I sat next to. That’s the person next to whom I sat.

Avoid the passive voice. This book has been read by many students. Many students have read this book.

Say and I not and me. You walked past my brother and me. You walked past my brother and I.

Do not use double negatives. I didn't say nothing! I didn't say anything!

Imposing rules from other languages
Some prescriptivist rules are imported from other languages. For example, through the Enlightenment and Romantic periods, Greek and Latin-speaking scholars imposed Greek and Latin spelling and grammar rules on English. Scholars (sometimes correctly, and sometimes incorrectly) inserted silent letters into words to reflect their Greek or Latin origins. This is where the silent in words like debt and doubt come from. But they didn't stop at spelling reform! In Greek and Latin, infinitives are one word, and so a word cannot intervene in the middle of an infinitive construction. But in English, an infinitive is two words (to + a verb). These scholars decided that, if a word cannot go in the middle of an infinitive construction in Greek and Latin, we shouldn't be able to do so in English, either. Another rule that has been imported is the prohibition against ending a sentence with a preposition. This is actually a pretty unique property of English grammar that is not possible in many other languages. Again, scholars began teaching that you shouldn't end a sentence with a preposition to make English more like Latin and Greek. Arguably, these scholars imposed Latin and Greek rules on English because they wanted to show off their higher education. At the very least, these rules imply that Latin and Greek are better than English. The prestige of Latin and Greek is not as strong as it used to be, but these rules get repeated by teachers and editors who may or may not be aware of their origins.
Language change
Language is always changing, and young people of every generation coin new words and even develop new grammatical patterns. However, sometimes the new grammatical forms aren't accepted right away, and the new form is labelled as "incorrect." One example of this is the comparative form of the word fun, as shown in Table 3. Younger speakers produce words such as funner and funnest, especially in informal contexts, whereas older speakers only produce more fun and most fun.

Table 3. Change in the comparative and superlative forms of fun

Older variant Newer variant

Comparative form more fun funner

Superlative form most fun funnest

Stylistics and rhetoric
Another source of prescriptivist rules is from stylistics and rhetoric, or the study of how texts are structured in order to be the most effective and persuasive. In stylistics and rhetoric, literary scholars may give you advice about what kinds of sentences and words to use because of the effect they have on the reader. For example, one piece of advice that is often given is to avoid the passive voice. One justification for avoiding the passive voice is because it removes responsibility and accountability from the person who performed the action. This critique is often directed at the news, for example, for statements like those in (1). In (1b), the actions of the police officer are downplayed.

(1) a. A police officer shot a bystander.

b. A bystander was shot by a police officer.

However, sometimes the actor is not known, or if it is known, is not important. In these cases, the passive voice is definitely appropriate. Furthermore, it is possible to downplay the actor without using the passive voice. Consider this quote from NASA administrator Bill Nelson, shown in (2).

(2) We have had mistakes done in the past.

(Watts 2024)

The sentence in (2) makes it sound like mistakes just happened, and does not lay the blame on anyone, but it doesn't actually use the passive voice. Therefore, instead of a blanket rule to avoid the passive voice, a more appropriate recommendation is to be aware of whether you are downplaying the actor of a sentence and whether doing so is appropriate for the context. Another justification given for avoiding the passive voice is the argument that the active voice gives a more vibrant and active feel to your writing. While this might sometimes be the case, it is possible to write dull sentences in active voice, as in (3a), and vibrant sentences in passive voice, as in (3b). Furthermore, a vibrant and active feel is not always the one you will be aiming for!

(3) a. The wind bent the tree.

b. The tree was bent and ravaged by the constant windstorms.

Although rules coming from stylistics or rhetoric might be reasonable advice in some cases, these are just guidelines given for a particular purpose rather than rules about how language is structured. In syntax, we are more concerned about the possible structures than about which structure is the most persuasive or clear. Although some styles of writing may be clearer or more appropriate in a given context, that doesn't make the other ways of writing "wrong." Unclear, ambiguous, or misleading sentences that are nonetheless grammatical still provide us with valuable information about how language is structured and processed.
Hypercorrection
Another source of prescriptivist rules is from something called [pb_glossary id="1106"]hypercorrection[/pb_glossary], which is when a rule is over-applied and extended to new contexts. The rule about and I originated as a hypercorrection. The difference between I and me, just like the difference between we and us, is about whether the pronoun appears in subject or object position. I generally appears in subject position, while me appears in object position.

(4) a. I ate dinner.

b. *Me ate dinner.

c. Dinner was eaten by me.

d. *Dinner was eaten by I.

Originally, the pattern was the same whether or not the pronoun came after and. But, in English, me is the default form that shows up when it is neither subject nor object, and for some speakers it started showing up after and in subject position, like in (5b). Other people started correcting that usage, and then people started saying and I in every context, even in object position, like (5d). In these questions, the % indicates that the sentence is accepted by some speakers but not all.

(5) a. My friend and I ate dinner.

b. %My friend and me ate dinner.

c. Dinner was eaten by my friend and me.

d. %Dinner was eaten by my friend and I.

This is an example of a hypercorrection. This hypercorrection has been so pervasive that it has arguably changed some people's grammars.
Linguistic imperialism and discrimination
Finally, some prescriptivist rules arise due to contact between two different language communities with a power imbalance. In these situations, the grammar of the language community with less power is often labelled as incorrect or even illogical. Often, this happens alongside other forms of racism and/or colonialism. For example, because the British Empire colonized so many countries around the world, there are many different global dialects of English. Many of these dialects can be considered mixed languages, which combine properties of two different languages. However, although these dialects are expressive and creative, follow systematic rules, and are valid language varieties, they are often described as "broken English." For example, Nigerian Pidgin English is spoken by over half of the population of Nigeria (Faraclas 2008), including university graduates, professors, lawyers, and journalists, but it still has the reputation of being the language of the non-literate (Akande and Salami 2010). These language attitudes can have significant impact on society. For example, in the infamous acquittal of George Zimmerman which launched the Black Lives Matter movement, the six hours of testimony by the star witness Rachel Jeantel were disregarded by the jury. Public remarks by a juror indicate that they disregarded her testimony because of her speech (Rickford and King 2016).
A published academic article written in non-standard English? Is it even possible? Check out Young (2010) for an overview of how descriptivism can work in everyday life, not just linguistics—written in non-standard English.

What should I do when I notice discriminatory bias in myself?

First of all, recognize that the emotions you are feeling are real and are the result of a number of complex factors, many of which you did not choose, such as the environment you grew up in. We may be tempted to deny or ignore unpleasant thoughts, including our biases, but doing so does not get rid of them. Instead, it just keeps them in our subconscious, where we can't do anything about them. Instead, when you notice a discriminatory bias in yourself, do two things: 1. Acknowledge that your feelings and/or thoughts have a discriminatory bias. 2. Intentionally choose actions, reactions, and words that counteract the bias. Later, when you are in a safe space with the time and emotional energy to do so, spend some time thinking about where these thoughts and feelings came from. You should do so in private, either alone or with trusted friends. However, if you do so with friends, keep in mind that such discussions have a high emotional cost for your friends who face discrimination on a regular basis, which you may or may not be aware of. Always check that your friends are able and willing to participate in such discussions.

Key takeaways

We use language to perform our identity. Just like there are not good and bad identities, there are not good and bad forms of language.

Not all identities are equally valued by society. The value society places on certain kinds of people gets transferred to the language use associated with those people. The language used by more powerful and prestigious people is often considered more "correct" than other varieties, even though all varieties are systematic and rule-governed.

In linguistics, we want to describe how language is used, called the descriptive approach, rather than judging how language should be used, called the prescriptive approach.

Check yourself!
[h5p id="17"]
References and further resources

Comedy and satirical linguistics

🎉 Phlogiston, Phineas Q. 2008. Cartoon theories of linguistics part XI - Prescriptivism vs. descriptivisim. Speculative Grammarian CLIV(3). https://specgram.com/CLIV.3/04.phlogiston.cartoon.xi.html

🎉 Starkey, Ryan. 2019. Linguistic prescriptivists make terrible zoologists. Starkey Comics. https://starkeycomics.com/2019/04/18/linguistic-prescriptivists-make-terrible-zoologists

For a general audience

🔍 Bee, Zoe. 2021. Grammarly is garbage, and here's why. https://www.youtube.com/watch?v=Q5rB9jDbTPU

For linguistics students

🔍 Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz. 2022. Talking college: Making space for Black language practices in higher education. New York: Teachers College Press.

🔍 Curzan, Anne. 2014. Fixing English: Prescriptivism and language history. Cambridge: Cambridge University Press.

Sources for examples

Watts, Marina. August 25, 2024. Nasa makes decision to bring home astronauts who have been stuck in space for 80 days. People. https://people.com/nasa-makes-decision-to-bring-home-astronauts-stuck-in-space-for-80-days-8701062

Academic sources

Akande, Akinmade T. and L. Oladipo Salami. 2010. Use and attitudes towards Nigerian Pidgin English among Nigerian university students. In Marginal dialects: Scotland, Ireland and beyond, ed. Robert McColl Millar, 70–89. Aberdeen: Forum for Research on the Languages of Scotland and Ireland.

Faraclas, Nicholas. 2008. Nigerian Pidgin English: Morphology and syntax. In Varieties of English 4: Africa, South and Southeast Asia, ed. Rajend Mesthrie, 340–367. Berlin: Mouton de Gruyter.

Montgomery, Michael. 2014. Hain't we got a right to use ain't and auxiliary contraction?: Toward a history of negation variants in Appalachian English. Southern Journal of Linguistics 38 (1): 31–68.

🔍 Rickford, John, and Sharese King. 2016. Language and linguistics on trial: Hearing Rachel Jeantel (and other vernacular speakers) in the courtroom and beyond. Language 92 (4): 948–988. https://www.linguisticsociety.org/sites/default/files/Rickford_92_4.pdf

Young, Vershawn Ashanti. 2010. Should writers use they own English? Iowa Journal of Cultural Studies 12 (1): 110-117.

3.4. Poverty of the Stimulus and Universal Grammar

juliedoner — Tue, 31 Oct 2023 17:37:02 +0000
Look at this sequence and guess what the next number in it should be:

(1) 1, 2, 3, 4, 5, ?

Most people would guess that it is ‘6,’ assuming the formula in (2a). That is a very reasonable hypothesis, but the fact of the matter is, the next number in the sequence could also be ‘126,’ if you use the formula in (2b).

(2) a. x = y

b. [(x-5) * (x-4) * (x-3) * (x-2) * (x-1)] + x = y

Without the 6th element in the sequence, there is no way to tell which formula is correct. Even if we have the 6th element, though, this problem doesn’t go away. We could easily modify the formula in (2b) so that the 6th element is ‘6’ but the 7th element -- or any other term -- diverges from the formula in (2b). This illustrates the fact that an infinite rule-governed system is unlearnable. When a system is infinite, we can never be sure that the next element in the system will follow the pattern we think we have found. The problem of infinite systems is relevant to language because language is [pb_glossary id="164"]productive[/pb_glossary], allowing us to combine the pieces of language in unlimited ways. This means that language is also an infinite system. Anytime you make a sentence, you can take that sentence and transform it into another sentence by adding “I believe that” or “Javier says that” or “Amita thinks that”, etc…, to the beginning, as shown in (3).

(3) a. Language is infinite.

b. I believe that language is infinite.

c. Javier says that I believe that language is infinite.

d. Amita thinks that Javier says that I believe that language is infinite.

There is no limit to the amount of times you can do this, in principle, although you may eventually run out of breath or forget where you are in the sentence. But those are physical limitations. The grammar itself doesn’t have any limits. In addition to that, using [pb_glossary id="166"]embedded clause[/pb_glossary]s like we did in (3) is only one way we could add to sentences infinitely. This kind of infinite building is in every part of the grammar, including in [pb_glossary id="167"]possessive[/pb_glossary]s (4) or with [pb_glossary id="168"]prepositional phrases[/pb_glossary] (5).[footnote]This example is inspired by the children's song And the Green Grass Grew All Around.[/footnote]

(4) a. Zofia’s idea

b. Zofia’s brother’s idea

c. Zofia’s brother’s friend’s idea

d. Zofia’s brother’s friend’s mother’s idea

(5) a. the hole in the ground

b. the tree in the hole in the ground

c. the limb on the tree in the hole in the ground

d. the branch on the limb on the tree in the hole in the ground

This leaves us with a problem. Language is infinite, but infinite systems are unlearnable. Somehow, though, most people seem to learn language. This is known as the [pb_glossary id="179"]poverty of the stimulus argument[/pb_glossary] since the language we receive in our environment (the stimulus) is insufficient for us to acquire language. In other words, it is impoverished.

Foundational assumptions: Grammar in the abstract vs. language in use

The argument that language is infinite because you can, in principle, always add more words to a sentence makes a pretty big assumption. The assumption is that there is an abstract notion of language or grammar independent of language in use. In other words, even if no one ever actually produces an infinitely long sentence, we are assuming that there is an abstract grammatical system that can produce an infinitely long sentence. The grammar has this possibility in its toolbox, even if no one uses it. This is similar to the distinction between [pb_glossary id="169"]competence[/pb_glossary] and [pb_glossary id="170"]performance[/pb_glossary] from Section 2.5. Not all linguists would agree with this assumption. They might argue that, if language is restricted by physical limitations, than these physical limitations are also part of grammar. In other words, there is no such thing as "abstract grammar" outside of language in use. Including physical limitations in your grammar does add some complications, though, as physical limitations vary. Grammar rules in such a system would have to be a bit fuzzy instead of [pb_glossary id="178"]categorical[/pb_glossary]. What do you think? Should our physical limitations be accounted for within grammar or outside of grammar? Are physical limitations a counter-argument against the infinity of language? What sort of evidence would convince you to change your mind?

Noam Chomsky’s solution to this problem is the idea that some parts of language are innate. That is, some components of language are built into human DNA. In fact, he argued that there are three components to the human capacity for language: general cognitive capacities, experience, and a language-specific genetic component.

General cognitive capacity. There are many parts of human cognition that can help us learn language but aren’t specific to language. These are general cognitive capacities that we already have for other skills, but that we can also use to help us learn language. They include the ability to analyze data and spot patterns, developmental constraints, and principles of efficient computation.

Experience. We know that experience plays a role in language acquisition. If you are surrounded by Japanese speakers as a child, you will learn Japanese. But if you are surrounded by Swahili speakers, you will learn Swahili. The language in your environment is called the input. You will learn a different language—both different grammatical rules and different vocabulary—depending on what language you’re exposed to. This is where all the differences between languages come from.

Language-specific genetic component. This genetic component is a capacity for language that we are born with, but unlike general cognitive capacities, it is specialized for language. Chomsky named it [pb_glossary id="180"]Universal Grammar[/pb_glossary], which is often abbreviated as UG. Universal Grammar should be uniform for all humans, since we have approximately the same genetics. It allows us to take our input and analyze it by instinct as an infant. The timing of language acquisition might also be constrained by genetics.

Besides the poverty of the stimulus argument, there are a few other pieces of evidence for Universal Grammar:

Even children who are not exposed to a rich language environment tend to develop language of some kind. For example, deaf children in a hearing family who do not otherwise have access to a signed language community tend to develop a homesign system.

There is evidence from acquisition that children need to be exposed to language within a certain timeline known as the Critical Period in order to be successful at acquiring language.

There is a dissociation between language development and cognitive development. Some people (those with Specific Language Impairment) show deficient language skills but normal cognitive development, while others (those with Williams Syndrome) seem to show delayed cognitive development but normal language skills.

Specific Language Impairment has even been traced to a particular gene called FOXP2!

Certain parts of the brain are specialized for language use, such as Broca’s and Wernicke’s Areas.

In summary, according to the poverty of the stimulus argument, language is infinite, and although infinite systems should be unlearnable, we still manage to learn language. The solution to this is Universal Grammar, which is the hypothesis that there is an innate, genetically-endowed cognitive capacity that is specific to language. On the other hand, the theory of Universal Grammar and the poverty of the stimulus argument are not accepted by all linguists. However, it is a foundational assumption in the framework of Generative Grammar, which is the approach taken in the textbook. In the next section, we will learn a bit more about the debate between those who adopt UG and who don't.
Check yourself!
[h5p id="13"]
References and further resources

Comedy and satirical linguistics

🎉 Wug Life. 2 Oct 2017. This Meowth can speak human language! Tumblr. https://wuglife.tumblr.com/post/165979527843/hahaha-oh-man-this-is-beautiful-a-quick

Academic sources

🔍 Boxell, Oliver. 2016. The place of Universal Grammar in the study of language and mind: A Response to Dąbrowska (2015). Open Linguistics 2: 352–372.

🔍 Dąbrowska, Ewa. 2015. What exactly is Universal Grammar, and has anyone seen it? Frontiers in Psychology 6: 852.

3.5. Debating Universal Grammar

juliedoner — Wed, 01 Nov 2023 00:25:44 +0000
Since Chomsky proposed [pb_glossary id="180"]Universal Grammar[/pb_glossary] in the 1960s, it has split the field of linguistics in two---those who adopt his hypothesis and those who don’t. Unfortunately, both sides of the debate are really bad at communicating and collaborating with each other. Both sides tend to argue against and make fun of (over)simplified versions of the other side. For example, here is a pair of excerpts from an op-ed and a blog post in the wake of Vyvyan Evans’ 2014 book The Language Myth.
"I was, of course, expecting a negative reaction from Chomsky’s disciples. But, given the practice in normal science, I was expecting a debate, and in particular, a challenge, with argument and evidence, to the substantive criticisms I levelled against the Chomskyan enterprise. This, it seemed to me, would have been a profitable way to take the field forwards. But things, at least, amongst some of Chomsky’s die-hard followers, haven’t panned out like that. In various fora, the reaction has sometimes appeared to border on the hysterical. For instance, in various public on-line venues I have been described as an 'idiot' and a 'quack', and The Language Myth has been repeatedly derided as 'junk' and 'junk science'; there has been a public debunking exercise led by a 'panel' of experts, embroidered with scorn and disdain, conducted in venues ranging from Facebook to Reddit. One prominent, and influential Chomskyan supporter, Professor Norbert Hornstein, has devoted, by now, an extraordinary number of posts to what amounts to a slim volume aimed at the general public, in his Faculty of Language Blogspot. He has repeatedly railed against The Language Myth, even issuing a call to (intellectual) arms: 'criticize this in all venues, especially where non-linguists gather. Consider it part of your linguistic public service' (December 2014). -Evans 2015 (Source)
Hornstein responds in the same vein:
I was going to write something long and pointed in response to this new piece by Vyvyan Evans (VE). I was going to analyze the article noting its inflated combination of self-pity (those 'Chomsky disciples' are saying that my work is junk) and self-aggrandizement (I am the leader of a Kuhnian paradigm shift that will overthrow the Chomskyan orthodoxy that, prior to the heroic efforts of people like me, who are now being viciously pilloried by the intellectually enervated Generative establishment, has had a stranglehold on linguistic pedagogy). I was going to note that VE has yet to address a single criticism leveled against his work, preferring instead to personalize the disagreement in a two fold manner; first by noting his personal travails in fighting the Chomskyan dragon and second by citing the approval of various authorities and luminaries that think him and his work wonderful. However, despite the obvious temptations that VE’s public demeanor make almost irresistible, his latest piece has convinced me that nothing much will be gained by doing this (fun as it may be). -Hornstein 2015 (Source)
Besides being rude, this kind of exchange is actually harmful to the advancement of science. First of all, it is important to remember that just because someone adopts a hypothesis doesn’t mean they think it is true. Often, the only way to test a hypothesis is to assume it is true, and see how far that will take you. Sometimes researchers will follow this line of research because they’re pretty sure it is true, but sometimes researchers will adopt a hypothesis because they think it is wrong. In either case, they would be applying the hypothesis to different sets of data to try and find one where the hypothesis fails to predict the data. The [pb_glossary id="184"]Scientific Method[/pb_glossary] is built on the assumption that your first hypothesis (and probably your second, and your third) will not be correct, but that each time we formulate a hypothesis and test it against new data, our hypothesis will become refined and hopefully closer to the truth. Second of all, science is usually strongest when there are scientists working on the same problem from multiple perspectives. For example, if we assume that the Universal Grammar hypothesis is false, we might miss hard-to-spot patterns and similarities among the languages of the world. On the other hand, if we assume Universal Grammar is true, we might force languages into boxes where they don’t quite fit. Because scientists are humans who generally become emotionally attached to their research, having skeptics look over your work from the opposite perspective helps to reduce errors. Thirdly, by not taking each other seriously, we are missing out on half of the available linguistic research! All of our data is relevant to other theories, and even some of our theories can be reformulated to fit into other approaches. In fact, sometimes researchers from different approaches converge on the same solution, which provides a strong case that they are both on the right track. For example, the trend right now in [pb_glossary id="185"]Chomskyan[/pb_glossary] syntax is to try to minimize the Universal Grammar component and explain as much as possible through general cognitive capacities, in a framework known as [pb_glossary id="186"]Minimalism[/pb_glossary]. It is possible that this line of research might lead to the conclusion that Universal Grammar can be attributed entirely to general cognitive capacities. In sum, Universal Grammar is controversial and has resulted in a lot of contentious debate. However, the field of linguistics benefits from researchers from both sides working together and taking each other’s research seriously.

Universal Grammar, recursion, and Pirahã

In the last section, we learned that Universal Grammar is the language-specific genetic component of the human language ability. In the last few decades, Chomsky begun arguing that, from an evolutionary perspective, it would make sense if this language-specific genetic component only consists of a single mutation that provided an immediate advantage to the human species. In 2002, Chomsky published a paper with two evolutionary biologists named Marc Hauser[footnote] In 2010, Marc Hauser was found guilty of falsifying data and scientific misconduct. However, the paper mentioned here is still cited and used, likely due to the prestige of the other two authors.[/footnote] and Tecumseh Fitch, in which they argue that this single component is the capacity for [pb_glossary id="187"]recursion[/pb_glossary]. Recursion is a property of grammar that allows us to iteratively apply a rule to its own output. This is what allows us to put possessives inside of other possessives, or prepositional phrases inside of other prepositional phrases, like the examples that illustrated the infinity of language from last section. We will learn more about recursion in Chapter 16. A few years later, in 2005, a fieldworker named Daniel Everett published a paper claiming that the language he was working on did not have recursion. The language's name was Pirahã and it is spoken by a tribe in the Amazon rainforest that is still relatively isolated. Some researchers took Everett's claims as evidence that UG does not exist and that Chomsky is wrong. Many others, though, were not convinced. Generative grammarians have criticized many elements of Everett's work. Some have looked through previous work on Pirahã by Everett himself and found data that they think contradicts his 2005 paper. Others have pointed out that Chomsky never claimed that all languages must have recursion, simply that recursion is a human capability. The Pirahã people are still able to learn Brazilian Portuguese, which definitely has recursion, which means that the Pirahã people are capable of recursion even if they don't use it. Similarly, others have pointed out that there are lots of different kinds of recursion, but not all languages use all of them. For example, German allows recursive noun-noun [pb_glossary id="189"]compound[/pb_glossary]s but other languages do not (see Chapter 10). Still others have pointed out that Everett's work is problematic because it primitivizes and exocticizes the Pirahã people and their culture. Everett's work and the Pirahã people still get attention in popular linguistics books and blogs and on mainstream media. The impact on research, on the other hand, was not so large. Because the Pirahã people are so isolated, it is difficult to independently test Everett's claims. After a few years debating the merits and weaknesses of Everett's work, many linguists have moved on to other things.

Check yourself!
[h5p id="14"]
References and further resources

For linguistics students

Hornstein, Norbert. 2015. My (HOPEFULLY) last ever post on Vyvyan Evans and his endless dodging of the central issues. Faculty of Language. http://facultyoflanguage.blogspot.com/2015/05/my-hopefully-last-ever-post-on-vyvyan.html

Evans, Vyvyan. 2015. The structure of scientific revolutions: Reflections on radical fundamentalism in language science. Psychology Today. https://www.psychologytoday.com/ca/blog/language-in-the-mind/201504/the-structure-scientific-revolutions

Academic sources

Hauser, Marc, Noam Chomsky, and Tecumseh Fitch. 2002. The faculty of language: what is it, who has it, and how did it evolve? Science, 298 (5598), 1569–1579.

Everett, Daniel. 2005. Cultural constraints on grammar and cognition in Pirahã: Another look at the design features of human language. Current Anthropology 46(4): 621–646.

Nevins, Andrew, David Pesetsky, & Cilene Rodrigues. 2009. Pirahã exceptionality: A reassessment. Language 85(2): 355–404.

1.4. Becoming a linguist: Study tips for linguistics classes

juliedoner — Fri, 03 Nov 2023 19:17:27 +0000

Reading to learn
Reading is often a major part of the course material in university classes. I have two tips for how to best use your course readings to learn the material. First, read before class. If you read before class, then when your instructor explains the material, you will be hearing it the second time, and it will make more sense. If it feels like class is moving too fast, spending more time with the reading before class might help you keep up. Doing the readings before class will also help you predict where the instructor is going, take better notes, make new connections, and ask better questions. You may find it helpful to also revisit the reading after class, but you can do so strategically, revisiting just the tricky bits or parts you may have misunderstood the first time through. Second, read with questions in mind. You should be asking three kinds of questions:

When you don't understand something or find something confusing, write out a question that puts your confusion into words. This will help you to better identify what you understand well or need to study more. It will also help you figure out how to fill in the gaps in your learning, whether by listening for it in lecture, by asking about in office hours, or by revisiting the reading.

Ask questions about problems or inconsistencies you find. This will help you to become a more critical thinker.

Ask questions about how the course material connects to other units in the same course, to other courses, and to your daily life. This will help you remember the material long-term and make it more meaningful to you.

Problem solving skills
Doing linguistics involves a lot of problem solving and data analysis. The goal we are working towards is that you will be able to apply your problem-solving skills to novel data. That is, that you will have the skills to identify patterns in data that you have never seen before. When you watch your instructor demonstrate problem solving in class, it can be very easy to trick yourself into thinking that you know how to solve that kind of problem, as well. As you watch your instructor, you will likely feel that every step makes sense. But that doesn't mean that you will be able to figure out what the next step should be when your instructor is not there. The best way to improve your problem-solving and analysis skills is to get lots of practice! You do not want your first time attempting any kind of problem to be on a test or an assignment worth lots of points. You want to practice throughout the semester, so that when the test or assignment comes, you will be ready! When practice exercises are assigned to you as homework, first attempt them on your own. If you get stuck, note where you get stuck and then get help. If your instructor is going to go through the answers to your homework in class, do the homework before that class and pay attention to what the instructor did differently than you. When I teach, I try to explain how to avoid as many false paths as I can think of, but my students are always more creative than me and find more! Although I recommend first attempting the exercises on your own, it can be very helpful to compare your answers with your classmates. If you're both stuck, figuring out together how to get unstuck will be easier than doing so on your own, and you're more likely to remember it than if I tell you what to do. Before you work with your classmates, though, be sure to check your course syllabus about what group work is or is not allowed in your class.
Tackling assignments
I recommend breaking up larger assignments into multiple steps. This is especially helpful if the assignment feels scary or overwhelming. If you break up an assignment into several smaller steps, it can feel much more manageable. If it still feels overwhelming, break it into even smaller pieces. On a day when you are really struggling, writing just one word might be a reasonable goal! Breaking an assignment into several steps is also helpful because it gives your brain more time to make connections and find creative solutions, it allows you to identify potential problems and get help early, and it gives you more time to identify and correct your mistakes. For problem sets especially, it is often very helpful to have the problem stewing in the back of your mind for several days. It helps you make connections that aren’t immediately obvious. You should first look at the assignment soon after it's assigned or posted. Just read it without planning on doing it! That will allow you to identify how much time and which resources you will need and if you have any questions. You should also separate your thinking and analysis time from the decision process of how to present or write up your answer. By separating these two very different tasks, you will likely have a much better result. If you have a full or partial draft ready early, before the due date, you can use that opportunity to get feedback from the instructor or your classmates. You will also be able to catch more of your own mistakes if you look at it with a fresh eye on a different day than the day you wrote it up. When you get feedback or a grade on your assignment, separate receiving the feedback from applying it. First give yourself time to process your emotions. Later, you can revisit the feedback and decide how to revise your work or how to improve your next assignment.
Reading to write
As well as using them to learn the course material, you should also be using the course readings to learn how to be a better writer and communicator. As you go through your university degree, you will be exposed to and asked to produce kinds of writing that might be new to you. In fact, this doesn't stop when you graduate! You may be asked to produce writing of different genres, for different communities and audiences, and with different goals. You will be taking many different courses in different fields throughout your degree, which have different genres, cultures, conventions, and more. For example, in some fields, such as English literature or syntactic theory, the convention is to present evidence and analysis together in every paragraph as you go, but in other fields, such as quantitative sociolinguistics or chemistry, the convention is to present your analysis all together in a special discussion section. If you also read and write in multiple language varieties, you may notice that different language communities prefer different strategies and rhetorical structures, as well. English academic writing usually uses explicit and direct argumentation, while other language communities use more inductive and deductive argumentation. So how should you, as a student, learn to navigate all of these different conventions and genres? First of all, it's okay and normal if you're not perfect the first time you try. Learning to be a better writer is a lifelong skill. In this textbook, I am trying to explicitly explain some of the conventions we use in morphology and syntax writing. But the best way to learn how to write in different genres is to read in those genres. The good news is that you are already reading in these genres through your course-assigned readings. The only thing you need to change is to start paying attention to how your readings are structured. Here are some questions you can ask as you read the course readings:

How is the reading structured? Do they tell it like a story? What ‘moves’ do they make?

Is it an effective structure? Not all academic writing is well-written and sometimes bad writing can help you learn more! If it is effective, how can you use the same strategies in your own writing? If it is not effective, how would you improve it?

How does the author group related ideas? How does the author move from one topic to another?

How does the author guide the reader? What signposts do they give the reader to help the reader navigate the reading?

When you don't get the grades you want
Most students hit a point in their education (or really several points) where things get hard. This does not mean that you have "reached your limit" or that you're not smart enough, or anything like that. Learning does not happen in a straight line, and so these bumps happen. For me, looking back, it happened between 11th and 12th grade, between 3rd and 4th year university, and also in my MA degree. But if you approach the learning bump as part of the journey or as a problem to be solved, instead of a scary thing to run away from, that's what's really going to help you the most in life. A lot of time, all you need to do to get past the bump is to not give up. Eventually things will click, and after you will look back and say, "Oh I wonder why I thought that was so hard!" Sometimes, it is also worth it to evaluate your study skills, for example changing when or how you're doing the textbook reading. If this is the first time you've struggled in a course, then you might be learning some new study skills for the first time (on top of learning the course material). That's good! But it is also hard. But the good news is that you can do hard things!
Study smarter, not harder
If it's time to re-evaluate your study skills, here are some tips to consider. One mistake a lot of students make is to study more but in an inefficient way. Instead of spending equal time on everything, figure out which parts need the most attention and focus on those.

Identify what needs to be memorized and what needs to just be understood. Very little material needs to be memorized in linguistics. Instead, if you understand the reasoning behind the way we do things, it will be easier to remember, and it will also be easier to determine which tools are needed for a given problem and to apply your knowledge to new data.

Try to study in active ways (like identifying key points and writing then down) instead of passive ways (listening to the lecture over and over).

Try to study in little chunks with lots of breaks, maybe spread out over lots of days, instead of all at once. This will result in the material going into your long-term memory instead of your short-term memory, so that you don't have to study it again the next time it turns up in your degree.

Try to identify specifically what you understand well and where you're stuck.

Use more than one format and more than one sense. Read the material and listen to the material. Draw a picture, diagram, or chart that helps you organize the material in your mind. Read the material out loud. Different people may have different preferred learning styles, but everyone learns better when you combine multiple styles.

If you're starting to feel tense or anxious, take a break, maybe go for a walk, have a glass of water, and come back when you're feeling calm.

What to do if you're falling behind
For a variety of reasons, you might fall behind in one or more of your courses. Don't panic! The first thing you should do, as soon as you realize you've fallen behind, is reach out to your instructor. The earlier you ask for help, the more the instructor can do for you. Your instructor can help you determine if it's realistic for you to catch up, give you extensions on assignments, and help you decide which assignments to prioritize. If you wait until the end of term to ask for help, though, there is often very little your instructor can do at that point. Depending on your exact circumstances, it is often better to focus on the upcoming coursework rather than trying to complete all of the missed coursework and follow the new material at the same time. Here are some things you should do to help you make a plan on how to catch up:

Consider why you got behind in the first place. Was it a temporary disruption to your schedule that is no longer an issue, or is it possibly going to continue disrupting your ability to complete your coursework? If the latter, make a plan to mitigate its disruption and/or reduce your commitments.

Check the syllabus for the late submission policies for your course. If you've fallen behind for exceptional circumstances beyond your control, such as ill physical or mental health, there may also be additional accommodations available to you not listed in the syllabus.

Look up the drop deadlines for your institution and determine what the penalties are for dropping at different dates. If you've fallen behind on multiple courses, it might make sense for you to drop one so you have more time to focus on the remainder.

Check the syllabus to see how much each assignment is worth.

Figure out how much work you need to catch up on and if it's realistic for you to be able to catch up. Calculate the weight of the assignments you missed and whether it's mathematically possible for you to pass and/or get your ideal grade.

When making your plan for what to do next, prioritize the upcoming assignments that are worth more.

The linguistics department at your university probably has an undergraduate linguistics group or club. These groups may offer social gatherings, study events, and even tutoring. Getting connected with other students studying linguistics can help your studies!

First generation students

Are you the first in your immediate family to attend university? Will you be the first to finish? Will you be the first to go to graduate school, if you go? Are you the first in your family to attend university in the country where you're studying? If you answered yes to any of these questions, you are a first generation student. Although first generation students make up a sizeable population of the university, many first generation students feel lost and alone. There may be a lot of background knowledge that some of your classmates and instructors take for granted, but you don't know. Maybe you don't even know you don't know it! There are lots of different kinds of first generation students, with different backgrounds and experiences. Many first generation students simultaneously also face other boundaries to university education, including financial boundaries and identities targeted by discrimination. If any of this sounds familiar, don't despair. This Becoming a Linguist series is written with you in mind! Although I hope this textbook will help, there are two important things you should do to help overcome these barriers. First, ask questions and be curious! Ask questions in class. Ask questions after class. Email your instructor. Post on discussion boards. Go to office hours. Your instructors' job is to help you learn, so ask them for help! Second, form a community around you. Here are some ways you can build community:

Make friends with your classmates. Things are less scary when you're not alone.

Join clubs. Look for undergraduate linguistics clubs and clubs where other people who share your identities meet up.

Look for mentorship: by peers who are a little bit ahead of you in the program, by recent graduates, by people in the career you're interested in, or by faculty members.

Join the Linguistic Society of America's First Generation Access and Equity Committee or attend some of their events.

A third thing you can do is explore some resources about how to succeed in university. Some suggestions are listed below in the References and Further Resources section, and even more are listed on the LSA's website. [caption id="attachment_4977" align="aligncenter" width="263"] Figure 1. The logo for the LSA's First Generation Access and Equity Committee, designed by Iyad Ghanim.[/caption]

Key takeaways

Complete the assigned readings before class.

Read with questions in mind.

Get lots of practice with problem-solving and analysis.

Break up assignments into multiple steps over multiple days.

Pay attention to the structure of assigned readings to help you learn to write better.

Don't give up if you aren't getting the grades you want!

Try to study in active ways, in short sessions spread out over many days.

If you're falling behind, talk to your instructor and make a plan.

Check yourself!
[h5p id="12"]
References and further resources

For students

Charity Hudley, Anne, Cheryl Dickter, and Hannah Franz. 2017. The Indispensable guide to undergraduate research: Success in and beyond college. New York: Teachers College Press.

First Generation Access and Equity Committee. n.d. Resources for First-Gen Students in Linguistics. Linguistic Society of America. https://www.lsadc.org/af_membercommittee.asp?committeeid=2&sp=1

Macaulay, Monica. 2011. Surviving linguistics: A guide for graduate students, 2nd edition. Somerville: Cascadilla Press.

🔍 Receiving and giving effective feedback. Centre for Teaching Excellence, University of Waterloo. https://uwaterloo.ca/centre-for-teaching-excellence/catalogs/tip-sheets/receiving-and-giving-effective-feedback

Chapter 1 practice exercises

juliedoner — Fri, 03 Nov 2023 19:30:23 +0000

Data analysis

Exercise 1. Simplex or complex?
Beginner-Intermediate [1.1] Are the following English words simplex or complex? If they are complex, segment them into morphemes using hyphens.

a. sofa

b. beautiful

c. bookshelf

d. pastries

e. recliner chair

f. reassessment

g. sunny

h. spilled

i. strawberry pie filling

j. distressing

Exercise 2. English analysis
Intermediate [1.1, 1.2] Consider the following nursery rhyme.
Little Bo-Peep has lost her sheep, And doesn't know where to find them; Leave them alone, and they'll come home, Wagging their tails behind them.
Brainstorm a list of things that someone would need to know about English in order to understand this nursery rhyme. Which of the things you listed pertain to morphology or syntax?

Communication and study skills

Exercise 3. Personal skills
Beginner [1.4] What background skills, experience, and knowledge do you bring to this course that might help you succeed? For example, do you have experience with multiple languages or dialects? Do you have experience with pattern recognition and problem solving? Are you skilled at creative language play? Do you have good time management and organizational skills?
Exercise 4. Cultural reflection
Intermediate [1.4] What is a traditional value from your culture? How will you apply this value during this course? For example, you could apply the Inuit value of working together for a common cause by establishing a study group or the Métis value of courage by participating in class in ways that stretch your comfort zone.
Exercise 5. Academic integrity
Intermediate [1.5] Which of the following would count as an academic offence? Explain what you should do instead.

Using lots of quotations in your paper, with in-text citations and a bibliography.

Listing the headings or topic sentences of a paper word for word to summarize a text, without quotation marks.

Providing a paraphrase without quotation marks.

Working together with a classmate when group work is not allowed.

Using a word-for-word definition from wikipedia without citing it.

Submitting an answer composed by Artificial Intelligence without permission to do so.

Having a friend proofread your essay for grammar errors.

Listing the same ideas in the same order, but in your own words.

Visiting the university writing centre for advice.

Copying another student's answer on homework or a quiz.

Buying an essay from an essay-writing service.

Checking the spelling of a word in a dictionary without citing it.

Forgetting to include a references section in your paper.

Hiring a tutor who tells you the answers if you can't figure them out on your own.

Research and application

Exercise 6. WALS data analysis
Beginner [1.2] Look at map 85A on the online language database WALS and answer the following questions.
To calculate percentages, divide the number of languages with prepositions or postpositions by the total number of languages and then multiply by 100.

a. Based on the sample in this map, what percentage of the languages of the world use prepositions?

b. Based on the sample in this map, what percentage of the languages of the world use postpositions?

c. Are languages that use prepositions and languages that use postpositions spread randomly throughout the world, or do they seemed to be grouped according to some sort of pattern?

d. Scroll down to the list of languages and click on the name of a language from the list that you've never heard of before.

What is the language called?

Does this language use prepositions, postpositions, or something else?

What language family does it belong to? (Hint: Look at the top of the page, just below the menu.)

Where is it spoken? (Hint: Look underneath the map on the right side of the page.)

e. Write up your answer to all of the above questions in paragraph format, using in-text citations as appropriate. (Hint: Cite the information you got from this specific map as Dryer (2013) and the information coming from the WALS database as a whole as Dryer and Haspelmath (2013). This is according to the instructions for citing WALS on the WALS homepage.)

Language Journal
For your language journal, you should pick a language that you don't know much about. At the end of each chapter, there will be a few questions for you to research about your language. By the end of the semester, you will have created a profile of the morphosyntax of your language!
In the spirit of decolonization, I encourage you to pick a language that disrupts the status quo in some way, which can begin to provide you with a deeper understanding of language diversity and the unevenness of the language landscape of the world and in the field of linguistics, such as:

A language that the student has never heard of before (which you can discover by searching a language database like WALS, Ethnologue, or Glottolog).

A signed or tactile language.

A language that is negatively affected by colonialism or globalization, especially if the student is or plans to be involved in community-led revitalization or documentation efforts.

A non-standard variety of a language (e.g., AAVE or Michif French).

Exercise 7. Language profile
Beginner-Intermediate [1.2] Research the basic background facts of your language. Where is it spoken? Approximately how many speakers are there? What language family does it belong to?
Exercise 8. Prepositional vs. postpositional
Beginner [1.2] Does your language use prepositions or postpositions? Include an example that supports your answer.

2.2. Evidence and hypotheses

juliedoner — Tue, 14 Nov 2023 08:38:47 +0000
As we use the scientific method to study language, there are a few things we need to keep in mind about the process.
Falsifiable hypotheses
When we formulate a hypothesis, it is important that we make a [pb_glossary id="495"]falsifiable hypothesis[/pb_glossary], which is a hypothesis that is logically possible to be proven wrong. It is important that your hypothesis is falsifiable so that you can do the next step in the scientific method, test your hypothesis! If it is impossible to fail your test, then perfoming the test doesn't help us figure anything out. Let's take an example: our hypothesis is that ghosts exist but are invisible, have no smell or taste, make no sounds, and cannot interact with the physical world whatsoever. How could we test this hypothesis? We cannot see, smell, touch, hear, or taste a ghost. We cannot trap a ghost as evidence. This hypothesis is unfalsifiable. There is no test we could do that would show that it is wrong. Therefore, it is not a good hypothesis. [caption id="attachment_1079" align="aligncenter" width="200"] Figure 1: A decorative ghost. Photo by Juan Vargas. Used under Pexels license.[/caption]
Confirmation bias
Another thing to watch out for when doing scientific research, or in fact any kind of critical thinking, is [pb_glossary id="1063"]confirmation bias[/pb_glossary]. Confirmation bias is a human tendency to look for evidence and information that is consistent with what they already believe. Here is an example. Let's pretend you work at a bakery and you believe that the macarons are the best and most popular item sold at your bakery. Every time someone buys a macaron, you may think, "See! Macarons are the best! Everyone likes them!" But if someone buys something else, you may not notice it or dismiss it as an exception. This would be a simple example of confirmation bias at work. In order to counter against this confirmation bias, you could count how many times people buy each kind of treat. Sometimes, when you do this, you may be surprised at the numbers! [caption id="attachment_1064" align="aligncenter" width="300"] Figure 2: Four macarons. Photo by Arminas Raudys. Used under Pexels license.[/caption]
Negative evidence
Another thing you could do to counter confirmation bias is to consciously train yourself to watch for [pb_glossary id="1022"]negative evidence[/pb_glossary]. Negative evidence is evidence that something is not possible, did not occur, or is absent. If you wanted to look for negative evidence that macarons are the most popular treat, you might count every time someone does not buy a macaron. Negative evidence is very important in linguistic analysis. In linguistics, negative evidence often takes the form of ungrammaticality. When we test a hypothesis about the structure of language, we should consider both what patterns we predict to be possible and what patterns we predict to be impossible according to our hypothesis, and we should check for both. In the last section, we compared the use of we and us by looking at both positive and negative evidence. The positive evidence were the examples of the sentences we could say, such as those in (1).

(1) a. We wrote a story.

b. You gave us a gift.

c. A story was written by us.

d. We were given a gift.

The negative evidence were the examples of the sentences that are awkward or unusual, such as those in (2).

(2) a. *Us wrote a story.

b. *You gave we a gift.

c. *A story was written by we.

d. *Us were given a gift.

If we only had the examples in (1), we wouldn't know for sure whether we and us were interchangeable or not. It could be just a coincidence that we was always used in subject position and us was always used in object position. It is the examples in (2) that show us that it matters what position they go in by showing us where they can't go.
Necessary and sufficient evidence
When you are making a hypothesis, you should think about what kind of evidence would allow you to draw conclusions about the accuracy of your hypothesis. There are two kinds of evidence to consider: [pb_glossary id="1072"]necessary evidence[/pb_glossary] and [pb_glossary id="1074"]sufficient evidence[/pb_glossary]. Necessary evidence is evidence that must be the case in order for your hypothesis to be true. For example, let's say that you think your sister ate the last cookie from the cookie jar. In order for this hypothesis to be true, the cookie jar must be empty. If there are any cookies left in the cookie jar, then it is impossible that your sister ate the last cookie. An empty cookie jar is necessary evidence that your sister ate the last cookie from the cookie jar. [caption id="attachment_1073" align="aligncenter" width="200"] Figure 3: A cookie jar full of cookies. Photo by Alina Matveycheva. Used under Pexels license.[/caption] However, even if the cookie jar were empty, this would not be sufficient evidence that your sister ate the last cookie. Maybe someone else ate it! In order to prove that it was your sister who ate the last cookie, you'd have to catch her in the act, or at least with crumbs around her face. [caption id="attachment_1075" align="aligncenter" width="200"] Figure 4: A girl eating a cookie. Photo by Charles Parker. Used under Pexels license.[/caption] But even so, in order to prove she ate the last cookie (and not the second-last one), you need to catch her taking the last cookie from the cookie jar and eating it. This would be sufficient evidence that your sister ate the last cookie from the cookie jar. Let's apply this to a linguistic example. In the last section, our initial hypothesis was that we and us are perfect [pb_glossary id="517"]synonyms[/pb_glossary]. In this case, the fact that we and us have the same meaning is necessary evidence. It could not be true that they are perfect synonyms if they do not share the same meaning. It is not possible to get sufficient evidence, though. In order to prove that they are perfect synonyms, you would have to show that they are always interchangeable in every possible context. Since there are infinite possible contexts, it is not possible to test them all. When we make our hypothesis, thinking about what constitutes necessary and sufficient evidence can help us decide what data to look for to test our hypothesis. It can also help us decide whether the hypothesis is falsifiable. If a hypothesis is falsifiable, it should be possible to identify the kind of evidence that would prove its opposite.

Key takeaways

Hypotheses should be falsifiable. That is, it should be possible to prove them wrong.

Confirmation bias is when you only notice the evidence that supports what you already believe. Everyone has confirmation bias to some extent. To counter confirmation bias, we should intentionally look for evidence that contradicts what we believe.

Negative evidence is evidence that something is not possible or does not exist. In linguistics, negative evidence often takes the form of ungrammatical utterances.

Necessary evidence is evidence that must be true in order for your hypothesis to be true.

Sufficient evidence is evidence that, on its own, is enough to prove your hypothesis.

Check yourself!
[h5p id="15"]
References and further resources

For a general audience

🧠 Veritasium. 2014. The most common cognitive bias. YouTube. https://www.youtube.com/watch?v=vKA4w2O61Xo

Academic sources

Davis, Henry, Carrie Gillon, and Lisa Matthewson. 2014. How to investigate linguistic diversity: Lessons from the Pacific Northwest. Language 90(4): e180–e226. https://muse.jhu.edu/article/563094/summary

4.2. Methods of data collection

juliedoner — Wed, 22 Nov 2023 19:58:13 +0000
There are three main ways we collect data for use in morphology and syntax: [pb_glossary id="1442"]corpus[/pb_glossary] studies, [pb_glossary id="1489"]linguistic elicitation[/pb_glossary], and experimentation.
Types of data collection

Corpus studies
A [pb_glossary id="1442"]corpus[/pb_glossary] study is based on a collection of found real-world data. A corpus (plural: corpora) can be compiled from written text, such as a collection of social media posts, newspaper articles, or books, or it can be compiled from video or audio recordings, such as a set of talk show interviews, political speeches, or podcasts. Sometimes researchers compile their own corpora, but there are also several corpora that have already been compiled and are available for use, such as the Corpus of Contemporary American English (COCA). If a word or construction has been found in a corpus, we say it is [pb_glossary id="1815"]attested[/pb_glossary]. If it has not been found, we say it is [pb_glossary id="1027"]unattested[/pb_glossary].
Linguistic elicitation
In linguistic elicitation, linguists work with a user of a language to collect linguistic data from that language. Elicitation is often bilingual, which means that data collection is mediated by another language. The linguist may ask the language user to translate a word, phrase, or sentence into the target language, or the linguist may construct examples in the target language and ask the language user if the constructed examples are acceptable. Linguistic elicitation can also be performed even if the linguist and the language user have no language in common, although it is more difficult. Daniel Everett gave a demonstration of how to perform a monolingual elicitation at the 2013 Linguistic Institute, which was recorded and posted here. To perform a monolingual elicitation, you can use props or images to elicit words and act out situations or use videos to elicit sentences. Although bilingual elicitation is a lot easier, it also increases the likelihood that the language user will be influenced by the mediating language. For example, if two word orders are permitted in the target language, a language user might be more likely to use the word order that matches with the word order of the mediating language, even if it is the less common one in the target language. Another method of data collection that is closely related to elicitation is introspection. If you are a speaker of the language that you are studying and you make up your own data based on your own personal judgments, that is called introspection.
Experimentation
An experiment is a highly controlled procedure, usually done in a research lab. The researcher will recruit multiple research participants, and ask them to perform the same task, in as close to the same conditions as possible, and observe their behaviour. Experiments in morphosyntax can be simple, such as a survey which collects grammaticality judgments from a large number of people, or more complex, such as an eye-tracking study that pays attention to where research participants are looking when reading a sentence. A pause or slower reaction time can be a clue that the structure is more complex.
Classification of data collection methods
We can classify these methods of data collection by whether they collect observational data or targeted data. Targeted data is data that is specifically sought out by the researcher to test a hypothesis. Both elicitation and experimentation are targeted. In elicitation, the researcher asks the language user for specific constructions that provide evidence for or against their hypothesis. In experiments, the researcher designs the methods of their experiment to provide evidence supporting or contradicting their hypothesis. Observational data, on the other hand, is naturally occurring data that has been observed. Corpus studies use observational data, as the data was created for independent purposes and later analyzed. We can also classify these methods of data by whether they are categorical or quantitative. Categorical data is sorted into categories, such as grammatical or ungrammatical. Elicitation results in categorical data, based on the judgments of the language user. Quantitative data is data that has been counted and statistically analyzed. Corpus studies result in quantitative data, such as the frequency of a particular word or construction. Experiments also result in quantitative data, such as the rate at which research participants exhibit a particular behaviour.

Table 1. Classification of data collection methods

Categorical data Quantitative data

Observational data Corpus study

Targeted data Elicitation Experiment

Which methods are best?
Each of the methods have different strengths and weaknesses. Which method is most appropriate for a given study will vary based on the research question and several other factors. It's best, if possible, to use multiple methods--hopefully the different methods will all converge on the same results! In this next section, we will look at two factors that should be considered when choosing a method: negative evidence and the resources of the language community.
Negative evidence
In Section 2.2, we learned that [pb_glossary id="1022"]negative evidence[/pb_glossary], which is evidence that something is not possible, did not occur, or is absent, is important for linguistic analysis. Which methods might produce negative evidence? Let's do a little thought experiment. Let's say we looked through the entire publishing history of The Toronto Star to look for the two sentences in (1) and didn't find them.

(1) a. The prime minister of Canada ate a salami sandwich.

b. *The ate of salami a Canada ministerial prime sandwich.

Does this mean that both sentences (1a) and (1b) are not possible? According to my intuition, sentence (1a) is a possible sentence, but it just hasn't been used by The Toronto Star. On the other hand, sentence (1b) is not a possible sentence of English. We can't tell the difference between a sentence that is possible but [pb_glossary id="1027"]unattested[/pb_glossary] and a sentence that is not possible at all just by looking at the sentences that have been produced in the past. Since language is creative and [pb_glossary id="164"]productive[/pb_glossary], not all possible sentences have been produced in the past. Arguably, most possible sentences haven't been produced yet. Because of this, no conclusions can be drawn if the pattern you are looking for doesn't show up in your corpus. As Carl Sagan famously said, "Absence of evidence is not evidence of absence." Corpus studies, since they result in observational data and not targeted data, cannot provide negative evidence. Elicitation and experimentation, on the other hand, can provide negative evidence, depending on their design.
Resources of the language community
Another important factor to consider when choosing a data collection method is the resources of the language community. If you are doing a study on English, it is easy to recruit a large number of English speakers to participate in a research study, and there are multiple electronic corpora that can be searched with just a few clicks. But English is one of the best-resourced languages in the world, and the situation is not the same for many minority and endangered languages. However, minority and endangered languages are crucial for helping us understand the breadth of the diversity of human language!
Minority and endangered languages are exactly the languages for which it may be impossible to gather large numbers of participants, and where the absence of literacy, the age of the speakers, and other factors make certain types of experiments unfeasible. (Davis et al. 2014: e187)
Even within well-resourced languages like English, corpora have biases. There are numerous dialects of English, which may all get mixed together in a single corpus, obscuring the differences between the dialects and making it unclear which groups of people use which language patterns. Minority dialects might not show up at all, or show up infrequently enough that they get excluded by the statistical analysis.
Check yourself!
[h5p id="25"]
References and further resources

For linguistics students

🔍 Everett, Daniel. 2013. Monolingual demonstration. Linguistic Institute. https://www.youtube.com/watch?v=sYpWp7g7XWU

Academic sources

📑 Davies, Mark. (2008–) The Corpus of Contemporary American English (COCA). https://www.english-corpora.org/coca

Davis, Henry, Carrie Gillon, and Lisa Matthewson. 2014. How to investigate linguistic diversity: Lessons from the Pacific Northwest. Language 90(4), e180–e226

3.1. What is a grammar?

juliedoner — Wed, 22 Nov 2023 20:37:28 +0000
If you look up the word grammar in a dictionary, you will find that this word has many different meanings and usages, three of which are particularly relevant to the study of morphology and syntax.
Grammar as a description of a particular language
One definition of grammar is a collection of rules that describe a particular language. Sometimes when people refer to the grammar of a language, they mean just the syntax rules, but other times, they mean all of the rules of the language, including phonological rules. However, remember that when we are talking about the rules of the language, we mean [pb_glossary id="498"]descriptive[/pb_glossary] rules, not [pb_glossary id="497"]prescriptive[/pb_glossary] rules. Descriptive rules are not rules like a speed limit, where some authority tells you how fast you should drive and you can choose to ignore or follow the rule—and if you ignore it, you might have to pay a fine or even go to jail. [caption id="attachment_1200" align="aligncenter" width="300"] Figure 1. A speed limit sign. Photo by Shahnoor Habib Munmun. Used under CC BY 3.0 license.[/caption] Rather, these are rules like the law of gravity, which exists because scientists studied the properties of the world and wrote a law that describes what's happening. You don't choose whether or not to follow the law of gravity. If you do break it, and start floating away, you won't get in trouble! Instead, scientists will be very curious about what's going on! Some examples of descriptive rules of English are listed in example (1).

(1) a. Nasalize vowels that precede a nasal consonant. phonological rule

b. Mark plurals with the suffix -s. morphological rule

c. Include a subject before the verb in a sentence. syntactic rule

[caption id="attachment_1289" align="aligncenter" width="1024"] Figure 2. A bookshelf at a university library in the Indigenous languages section. This section contains reference and pedagogical grammars, as well as dictionaries, texts in Indigenous languages, and academic books on specific topics in Indigenous linguistics. What kinds of books can you spot in this picture? Photo by Julie Doner.[/caption] Pedagogical grammars are designed to help people learn a language in order to speak, write, or sign it. They often contain example dialogues, vocabulary, and practice exercises. They will also often explain grammatical patterns using non-technical vocabulary, ideally starting with the most useful constructions and vocabulary. [caption id="attachment_5428" align="aligncenter" width="600"] Figure 3: A page from a pedagogical grammar of Yoruba (Mosadomi 2011). Yoruba is a member of the Niger-Congo language family and is spoken primarily in Nigeria with over 43 million native speakers (Eberhard et al. 2023).[/caption] A reference grammar is designed for linguists or other scholars to look up the properties of a particular language. Reference grammars may be purely descriptive, simply describing the properties of the language in question, or they may have a theoretical approach, using the properties of a particular language to make claims about human language in general. Some reference grammars will only cover morphology and syntax, but others may cover phonology as well. [caption id="attachment_1291" align="aligncenter" width="1024"] Figure 4: A picture of page 35 of Bloomfield (1957), a classic reference grammar of Nishnaabemwin (also known as Eastern Ojibwa). Nishnaabemwin belongs to the Algonquian language family and is spoken by about 220 people in Canada, according to the 2021 census. It is spoken in and around Lake Huron, especially Manitoulin Island (Eberhard et al. 2023). Photo by Julie Doner.[/caption]
Grammar as a model of Language
In the study of syntax and morphology, the most important meaning of the word grammar is a scientific model of the entire system and structure of the human capacity for language. In other words, any theory of morphosyntax which accounts for how all of the components of language structure work together to result in the possible languages of the world could be considered a grammar. A scientific [pb_glossary id="1409"]model[/pb_glossary] is a representation of a system, an organism, a process, or a phenomenon. In this case, what we are modelling is the human capacity for Language. This does not mean we are studying a particular language, like French, Anishinaabemowin, or American Sign Language, but the cognitive ability of humans to use language. This model of Language has to account for the ways that diverse languages are similar, but also for the ways that languages are different. It needs to account for how children acquire language, both the process and the timing. It should also be compatible with all of the individual rules of particular human languages, since all of those languages are the result of the human capacity for Language. A model like this consists of a collection of hypotheses. Each of these hypotheses needs to be tested and revised to help us develop the most accurate model of Language. As we test and revise our models, we look to see what kind of data it can explain. Chomsky (1965) proposed three levels of adequacy:

An [pb_glossary id="1206"]observationally adequate[/pb_glossary] grammar accounts for all of the real-world data that we observe.

A [pb_glossary id="1207"]descriptively adequate[/pb_glossary] grammar not only accounts for the real-world data that we observe, but also speaker judgments about what is or is not ungrammatical, therefore including [pb_glossary id="1022"]negative evidence[/pb_glossary].

An [pb_glossary id="1208"]explanatorily adequate[/pb_glossary] grammar accounts for all of the above, while also explaining how language is acquired.

We aim for an explanatorily adequate grammar, as that kind of model enables us to explain the most kinds of data.
Check yourself!
[h5p id="18"]
References and further resources

Reference materials

Bloomfield, Leonard. 1957. Eastern Ojibwa: Grammatical sketch, texts and word list. Ann Arbor: The University of Michigan Press.
Mosadomi, Fehintola. 2011. Yorùbá Yé Mi: A Beginning Yorùbá Textbook. Austin: University of Texas at Austin. https://coerll.utexas.edu/yemi
Academic sources

Chomsky, Noam. 1965. Aspects of the theory of syntax. Cambridge, MA: MIT Press.

📑 Eberhard, David M., Gary F. Simons, and Charles D. Fennig, eds. 2023. Ethnologue: Languages of the world. Twenty-sixth edition. Dallas, TX: SIL International. http://www.ethnologue.com.uml.idm.oclc.org

2.6. Becoming a linguist: Scientific writing in morphosyntax

juliedoner — Wed, 22 Nov 2023 20:39:14 +0000
As we mentioned earlier in this chapter, morphosyntax is a science because it follows the scientific method. When we write about morphology and syntax, we use a genre of scientific writing. However, not all scientific fields follow the same conventions. In this section, we will learn about the genre of scientific writing used in morphology and syntax.
The structure of a morphosyntax paper
A lot of scientific fields use a formulaic structure known as the IMRaD structure, which stands for "Introduction, Methods, Results, and Discussion." Although all of those components show up in a morphosyntax paper, we rarely use that structure in morphosyntax. The IMRaD structure is used for papers that discuss formal experiments and analyze their results using statistics. Although morphologists and syntacticians sometimes publish papers about formal experiments in the IMRaD format, the vast majority of morphology and syntax papers follow a less formulaic structure. Papers should still begin with an introduction and end with a conclusion, but the body of the paper typically weaves together description of data, analysis of data, and discussion of the implications of the data throughout.
The introduction and conclusion
The introduction of a paper should typically be less than a fifth of the total length of the paper. In it, you should clearly state the main claim of your paper and summarize your main arguments. You should also contextualize your main claim, explaining why it is important and what theories or frameworks you are working with. Because the structure of each morphology and syntax paper is unique, most papers end the introduction with a description of the structure of the paper. This is often called the overview, outline, or roadmap. Here's an example from Alhailawani (2023):
This article is organized as follows. In section 2, I provide an overview of languages with P-drop [Preposition-drop]. I also discuss the two main approaches to P-drop, with representative examples. In section 3, I set out the JA [Jordanian Arabic] P-drop facts and note the similarities and differences between JA and other P-drop languages. In section 4, I discuss and refute an analysis of JA P-drop in terms of pseudo noun incorporation. Section 5 shows that a silent P analysis of P-drop fares better at capturing the JA P-drop data. In section 6, I argue for an analysis of P-drop in JA where PP and DP are present, showing that the analysis neatly captures the JA P-drop facts. Section 7 is a brief conclusion.
Your conclusion should repeat most of the same information as the introduction, except perhaps in a briefer format. You should re-state your main claim and re-summarize your main arguments. You should also return to the contextualization of your paper, restating why it is important and how your paper contributes to the field. Sometimes, a conclusion can also suggest areas for further study. Here's the conclusion from the same paper about Jordanian Arabic (Alhailawani 2023):
In this article, I provided a detailed description of P-drop in JA. I showed that the main properties of P-drop found in other P-drop languages hold in JA. I refuted an analysis of P-drop in terms of pseudo noun incorporation (Gehrke and Lekakou 2013, Hall 2019), showing that the prepositionless goal noun in JA does not behave like a pseudo-incorporated nominal. I showed that the noun in JA P-drop exhibits the typical properties of a regular referential DP. I applied some diagnostics of PP-hood to show that there is a silent PP structure involved in JA P-drop. Thus, I argued for an analysis of P-drop in which a PP structure is present but unpronounced (Myler 2013, Biggs 2014, Bailey 2018, among others). The silent P is licensed via incorporation of P into v (den Dikken 2010), which in turn explains the absence of P-drop with manner of motion verbs. Also, the goal DP is assigned case in situ via Agree with a higher p head in a split p/P structure (Svenonius 2007). P-drop is a relatively understudied phenomenon due to its restrictive nature. The arguments and analysis presented in this article contribute to the ongoing debate on the underlying mechanisms involved in P-drop.

Let's practice analyzing this writing so we can learn from it!

Compare the conclusion from the outline from the introduction. Match the parts of the outline to the restatement in the conclusion. How is the wording different? Which one provides more detail? What other differences do you notice between the outline from the introduction and the conclusion?

Where does the author contextualize their paper, showing how it contributes to the field?

Remember that you're not writing a mystery novel! It can be tempting to make a big reveal at the end with your final analysis, or describe all the false paths you went down when trying to complete your analysis, but doing so will typically be frustrating and confusing to your reader. Instead, present it simply: say what you're going to do in your paper, do it, and then say what you've done. It may feel repetitive to write like that, but remember that you've (hopefully) thought about the subject of your paper for a while and are very familiar with it, but your reader is probably learning about it for the first time and does not know your thinking process. Good, clear writing often feels repetitive to the author. For example, notice how the example outline from the introduction and the conclusion from Alhailawani (2023) repeat many of the same points. If you are reading someone else's paper, it is often a good idea to read the abstract, the introduction, and the conclusion of the paper first. This will help you decide if this is the right paper for you to read right now, and if so, which sections will be the most important or relevant for you. You should also usually be able to get an idea of the main point of the paper, which can help you to understand the paper as a whole.
The body
Before you start writing the body of a morphology or syntax paper, I would recommend making a list of the topics you are going to discuss in your paper and then deciding what order they should go in. Think about what your reader needs to already know in order to understand each part of your paper. Sometimes one argument builds on another argument, which determines which order they should go in. The body of your paper should be divided into several sections with headers, especially if your paper is longer than about four pages. Each section should also be structured kind of like a whole paper: every section should have an introduction and conclusion that explains the main idea of the section and connects the content of the section back to the topic of the paper. This is recursive, so for even longer papers, your sections should be divided into subsections, and so on. The headers for each section should be numbered and should have titles that clearly indicate what the section is about. Some papers number the introduction as section 1, but others number it as section 0 or give it no number at all.
Style in morphology and syntax papers
Sometimes when students begin to write university-level papers, they try to use complex vocabulary and sentence structure to look more sophisticated. But the best papers take complex ideas and explain them in simple language. You want to be as clear and concise as you are able. Do not use a more complex vocabulary when simpler terms will do.
There seems to exist a belief that professional scientific writing needs to have a certain tone – impersonal, dense, abstract. I often see it in students who try to imitate that drab style [...] A lot of academics do write that way; nobody thinks that academics ought to write that way. In fact, editors and readers usually very much prefer if you don’t write that way.
-Rohrer 2024

One of the biggest obstacles to writing clearly and concisely in academic writing is that we are trying to explain intricate ideas. Academic writing often turns out complex because the ideas that the author is trying to convey are complex. On the other hand, it is very impressive if you manage to explain a complex idea in simple language, without simplifying the concept. Another big obstacle to writing clearly and concisely in academic writing is that we are trying to be very precise. We use technical terms as appropriate because technical terms have precise definitions in our academic community. Our sentences may also become complex when we try to accurately convey exactly what we mean. When we aren't sure about our conclusions, or if our conclusions only apply in some cases, we try to be very clear about how sure we are and how our understanding is limited. For example, don't say always unless you're sure it actually happens always. Instead, write usually or often or maybe even just sometimes. Keep in mind that many academics are actually not very good writers. If you find a paper that is hard to read, try to figure out if it's because the writing is not clear or because the ideas are difficult. There are some papers that you should not be imitating. Instead, what you should learn from those papers is what not to do. Academics also tend to use abbreviations, as you no doubt observed in the two examples above from Alhailawani (2023). Try not to use too many abbreviations; it can become very overwhelming for the reader. But it is alright to use a few. However, when you introduce an abbreviation for the first time, make sure you specify what the abbreviation means. This is usually done by writing out the term in full the first time, with the abbreviation afterwards in parentheses, as shown below (Alhailawani 2023):
In Jordanian Arabic (JA), directional prepositions like ʕa and li ‘to’ can be dropped with certain motion verbs like yiruuħ ‘go’ and yewsal ‘arrive' ...
The reverse can also be done, with the portion that does not appear in the abbreviation in parentheses, as with the following example from Alhailawani (2023):

In Greek and some varieties of British English, the complement of some motion verbs can optionally appear without an overt preposition or a determiner. [...] This phenomenon is known as P(reposition)-drop

(Ioannidou and den Dikken 2009)

Word choice in scientific and linguistics writing
In scientific writing, we need to be very precise with our language. In this section, I will describe some of the words that we need to be especially careful with.
Proof, evidence, and fact
You should very rarely claim that you have proved something or that some piece of evidence is proof of something. The scientific definition of proof means you can show that it holds in all cases. There is no possible scenario where it could be wrong, not even a very unlikely scenario. Most often, proof is used for logical proofs, which is a formal way of demonstrating that certain base assumptions logically lead to particular conclusions. When students use the words proof or prove, usually evidence or provides evidence for is more appropriate wording. When scientists use the word fact, they mean something that has been observed and confirmed repeatedly, to the point that we accept it as true; however, even when it has been observed hundreds or even thousands of times, it has still not been considered proven!
Theory, hypothesis, and model
As we went through the scientific method demonstration in Section 2.1, you may have noticed that we tried to stay skeptical about our conclusions. Because of this, and because we want to be very precise about how sure we are, some words are used differently in everyday speech than they are used in scientific discourse. For example, in everyday usage, theory might mean a hunch, idea, or speculation, but in scientific writing, theory means a set of hypotheses about the cause of some event or phenomenon that have been scientifically tested. In some cases, a scientific theory might still be under debate, but in other cases, it might be generally accepted by the scientific community, although ultimately unprovable. In everyday speech, a hypothesis might mean any guess, but in scientific writing, it means an educated guess or an informed prediction, based on observation of data. When scientists are trying to describe how a complex system works, they might make a series of hypotheses known as a [pb_glossary id="1409"]model[/pb_glossary]. For example, Figure 1 shows a model of a carbon atom. This is not what a carbon atom actually looks like if you look at it through a very powerful microscope. Instead, it is a schematic representation of the parts of a carbon atom and their relationships to each other. [caption id="attachment_1412" align="aligncenter" width="256"] Figure 1: A model of a carbon atom. Photo by SrKellyOP. Used under CC0 1.0 license.[/caption] In this textbook, we will create and test models of language known as grammars. They will also be schematic representations of the structure of language, rather than a "picture" of what is happening in our brains.
Significant
In everyday speech, significant just means important. However, in scientific writing, significant is short for statistically significant, which means that there is at least a 95 percent chance that, if you repeat the experiment, you will get results that are the same or more extreme results. In other words, there is a five percent chance or less that your results were a fluke due to random chance. This value is known as the p-value, and the 95% threshold is often described in decimals as p-value ≤ 0.05 'a p-value less than or equal to 0.05'. To avoid confusion, it is best to reserve the word significant for its statistical meaning in scientific writing. Instead, you can use words like crucial, important, or critical.
Clearly, obvious, and without doubt
Sometimes writers will write that something is clearly the case, or that an argument is obvious, or that we can know something without doubt. Even experienced writers do this! Unfortunately, if something is obvious or clear to you, it does not mean it is obvious or clear to everyone. It can be very frustrating as a reader when an author says something is obvious instead of explaining it. If it is so clear and obvious, then it should be easy for you to take a few lines and explain it!

Writing anxiety

You do not need to feel guilty or ashamed if you are anxious about writing. Feeling anxious about writing is normal. Even famous authors get writer's block! The important thing is to not let your anxiety stop you from writing. I know that is easier said than done, but here are some tips to help you:

Start early and break up the project into small pieces. Perhaps your goal for one day is just to find a research paper, and you will read it another day. Write an outline of your paper on a different day than you start writing.

Make a more manageable goal. If it feels overwhelming to write 200 words, make the goal to write just 100 words that day. Some days, just writing one word might be a good goal. Often, once you get started, you might get going and write more than your goal, which feels really good! But if not, even a small amount of writing every day can give you lots of momentum.

Don't worry about being perfect. Just try to start writing. You can come back and edit it later, but it is easier to edit something than nothing.

Find a format or a genre that is less scary to you, like an email to a loved one or a list. Use that writing to start, and then edit it into formal writing later.

If you're stuck, write about why you're stuck.

Record yourself describing your paper out loud, then use speech-to-text to get a first draft.

Meet up with friends or classmates to work independently alongside each other. Turn off your Wi-Fi and focus on just writing for an hour. Having someone else who is expecting you to show up at a specific time and write can help you avoid procrastination.

Key takeaways

Experimental linguistics papers follow a formulaic Introduction, Methods, Results, and Discussion structure, but most papers in morphosyntax present the data and the analysis interwoven together throughout the body of the paper. Since the structure is not formulaic, most authors give an overview of the structure of the paper at the end of the introduction.

In the introduction of your paper, you should clearly state your main claim, explain why your paper matters, and outline the evidence you will use to support your main claim. There should be no surprise reveals in the conclusion!

Your conclusion should restate most of the information you gave in the introduction.

The best papers explain complex ideas in the simplest language possible, without simplifying the ideas.

Be careful about your word choice in scientific writing. For example, the word proof should be reserved for logical proofs, and the word significant should be reserved for statistical significance.

Check yourself!
[h5p id="22"]
References and further resources

For students
Batchelor, LaRoyce. 2024. Growing a writing practice: Non-extractive writing. Pressbooks. https://pressbooks.openedmb.ca/nonextractivewriting/ 🔍📑 Macaulay, Monica. 2011. Surviving linguistics: A guide for graduate students, 2nd edition. Somerville, MA: Cascadilla Press. Rohrer, Julia. 2024. Writing about technical topics in an accessible manner. The 100% CI. https://www.the100.ci/2024/12/01/writing-about-technical-topics-in-an-accessible-manner/
Sources for examples

Alhailawani, Mohammad. 2023 The syntax of silent directional prepositions in Jordanian Arabic. Canadian Journal of Linguistics. 68 (2): 229–249.

6.5. Signed language morphology

juliedoner — Fri, 28 Jun 2024 16:40:16 +0000

8.10. Becoming a linguist: Finding academic sources

juliedoner — Wed, 31 Jul 2024 21:48:40 +0000

Where to find academic sources
Academic sources are a bit trickier to find than newspaper articles, popular books, or blogposts. How google scholar references of other sources university library university A-Z guide Ask a librarian
One common mistake students make when they are doing research is to just use the first few sources they find as their only sources. Instead, you should read one or two sources and use what you learned from your first sources to refine your research questions, identify gaps in your sources, and do targeted searches for the specific information that you are missing.

Should I pay for access to academic articles?

Have you ever looked for a resource for a homework assignment, only to find one that looks perfect, but it's behind a paywall? Did you know that you likely can get access to that article for free? Here are the places you should check:

Search your university library database. Most university libraries subscribe to large bundles of academic journals. It is very likely your library has an electronic copy of most academic articles published in the last few decades. The library also has many ebooks and physical books. Search for it in your university library database!

Check the author's website. Many academics post pre-publication versions of their papers on their website or in institutional repositories. Unlike general publishing, academic authors aren't usually paid royalties for articles (although they may receive royalties for books), and they want people to read their work!

Ask a librarian. If you cannot find it in the library database, ask a librarian! Many university libraries will have an "ask us" or "chat with us" pop up in the corner of their website where you can have a live conversation with a librarian (not a bot!) during library hours. Sometimes the library does have a copy of the article, but there was a mistake with the cataloguing. The librarian may also suggest an inter-library loan.

Ask for an inter-library loan. There is probably a form you can fill out on your university library webpage to request an inter-library loan. This is when the library borrows a source from another library. In the old days, they used to mail books to each other! Now, it is more likely you will get a scanned PDF emailed to you. This may take a few days to a few weeks.

Email the author. Authors are usually happy to share their work with students and others who are interested. They may not be allowed to post their article on their website based on the copyright agreement with the publisher, but they may be able to share it by email or point you to another source which is not paywalled.

Ask your instructor. Your instructor likely has a large personal electronic and physical library, and probably has copies of many sources related to the course material. They also have more experience navigating the university library system and the culture of academia. If you are having trouble tracking down a source, ask for help!

Request that the library purchase the article. If you think your source will be helpful to other students, as well, you might request that the library purchase the source (or suggest to your instructor that they do so). There is usually a form to suggest a purchase on the library website. This will likely take too long to be useful to you, but it is one way you can help the students who will come after you and to signal your interest in a particular source to your library.

If you cannot find the source in a timely manner, chances are that you can do without it, for most undergrad assignments. To find a related article on the same topic, you can look in the references list (which is often not paywalled) or do a search for articles that cite the paper of interest. Some academics, researchers, and activists have created large repositories of PDFs of books and academic articles online. Usually these databases are hosted and compiled in countries where copyright laws are less strict. The URL for these sites changes frequently, because of the actions of the publishing companies, so they can be tricky to find. Many students and academics use these databases frequently. Be aware that there may be legal repercussions for using such sites, and that doing so could also affect future employment opportunities, as some employers have policies about downloading and storing copies of pirated content.

Key takeaways

Check yourself!
[h5p id="9"]
References and further resources

For linguistics students

🔍 Conrad, Kirby. 2023. Citation sprinkles. Medium. https://kconrod.medium.com/citation-sprinkles-3ffeca947aa1

8.8. Features and subcategories

juliedoner — Sat, 10 Aug 2024 22:06:03 +0000

8.9. Do all languages have the same syntactic categories?

juliedoner — Sat, 10 Aug 2024 22:06:27 +0000

7.6. Clauses without tense

juliedoner — Tue, 20 Aug 2024 15:30:52 +0000

Embedded clauses with special structure

To-infinitives

Gerund clauses

Small clauses

Case Study: Colonial practices in the early linguistic research of Franz Boas and George Hunt

juliedoner — Wed, 29 Jan 2025 18:51:58 +0000
Early modern linguistics and anthropology in North America evolved from an entangled relationship in the nineteenth century into two distinct disciplines today. Originally, the overarching trend in these fields was to arrange cultures on a hierarchy of ‘progress.’ Early scholars essentially studied particular non-Western societies through the lens of classical evolutionism theory, meaning they were compared to Western cultural values. In regards to languages, these were typically studied for their similarity or differences to commonly spoken European languages. Due to these colonial origins, the most prominent and noted names in these fields are male European names (Eriksen 2015). You may have heard of Franz Boas, who is credited as a ‘founding father’ of American anthropology and an influential figure in modern American linguistics. Franz Boas is one of many well-known scholars attributed in the histories of linguistics and anthropology. There were, of course, numerous individuals conducting research during this early era of linguistics who did not receive as much credit. What kind of person do you think was more likely to receive credit? Indigenous linguists and ethnographers, for example, have contributed enormously to the origins of modern linguistics. Frequently referred to as informants, Indigenous researchers and ethnographers were integral to the early founding of American anthropology and linguistics. They also deserve the immense recognition that their counterparts are given (Bruchac 2018: 19). Valuable earlier work by Indigenous linguists and women is marginalized to this day. Part of acknowledging the colonial history of linguistics includes learning which scholars and intellectuals are being prioritized and who is being overlooked; the next step is to change this. Have you heard of the linguist and ethnographer George Hunt? What if we told you that he collaborated with Boas for over four decades?
Who was George Hunt?
George Hunt (1854 - 1933) was a Tlingit-English linguist, ethnographer, interpreter, and collector who was born and raised in Kwakwaka'wakw territory (in so-called Fort Rupert, British Columbia, Canada). Hunt spoke Tlingit, English, and Kwak’wala (Berman 1994: 12). Hunt worked as an interpreter and guide for missionaries, colonial government officials, and research expeditions due to his skill as a translator and his cultural knowledge (Berman 1994: 24–27; Bruchac 2018: 21). [caption id="attachment_6158" align="aligncenter" width="280"] Figure 1. "George Hunt" (1898) by Harlan Ingersoll Smith is in the public domain.[/caption] Hunt is most known for his collaborations with, and contributions to, the ethnographic work of Franz Boas. Franz Uri Boas (1858 - 1942) was a German-American anthropologist who is considered to be a founding influence in modern American anthropology and its schools of thought. Boas is attributed with popularizing the theory of cultural relativism, a view considered radical in his time because it began to counter racist ideologies. Cultural relativity is the idea that a culture and its beliefs, standards, and thoughts should be understood within its cultural framework – not in the context of different cultures (Bruchac 2018: 20; Eriksen 2015: 765–766). [caption id="attachment_6164" align="aligncenter" width="280"] Figure 2. "Franz Boas" (circa 1915), attributed to the Canadian Museum of History, is in the public domain.[/caption] Cultural relativism developed as a response to classical social evolutionism and ethnocentrism. Evolutionist anthropologists formed racist ‘classifications’ and ranked cultures hierarchically in comparison with Western “‘levels of development’,” ethnocentrically judging cultures against their own Western ideals (Eriksen 2015: 766). Despite his progressive cultural relativism, Boas still acted superior to Hunt and was ethnocentric in his approach. For example, Hunt wrote about some food-related topics, but Boas deemed these papers as less important due to his own cultural context (Bruchac 2018: 34). Hunt and Boas’ working relationship evolved from Boas hiring Hunt as an expedition guide, to later hiring Hunt as an informant, consultant, collector, and author (Bruchac 2018). Throughout their decades-long correspondence, Boas exploited Hunt by encouraging Hunt to source and obtain cultural information and material items for museum exhibits. Hunt sometimes violated traditional protocol to secure material for Boas, such as secretly excavating Koskemo gravesites to collect human skulls and funeral items (Bruchac 2018: 31).
Undervalued contributors: Lucy Homikanis Hunt and Tsukwani Francine Hunt
Despite Hunt’s Tlingit background, he was mostly an informant about the Kwakwaka’wakw culture and language. He wrote extensively about Kwakwaka’wakw practices, procedures, and other aspects of daily life, including Kwak’wala “myths, folktales, genealogical histories, and other narratives … prayers, ceremonial and everyday songs, and dreams” in his manuscripts (Berman 1994: 37). Hunt accessed certain Kwakwaka’wakw ceremonies and different types of knowledge through his two marriages to his first and second wives and their high-ranking families. In 1872, George Hunt married his first wife Lucy Homikanis Hunt, the daughter of a Kwakwaka’wakw chief (Bruchac 2018). Hunt depended on Lucy for her invaluable knowledge and manuscript revision. She often reviewed Hunt’s writing, at times suggesting manuscript topics. Lucy “participated in recounting, recording, and translating traditions that were virtually unknown to her husband” and aided him when he forgot to include something (Bruchac 2018: 34–35). Her death in 1908 left Hunt in mourning. Hunt temporarily stopped communication with Boas, struggling for years to write with his grief and without her vital contributions (Bruchac 2018: 35). [caption id="attachment_6155" align="aligncenter" width="718"] Figure 3. Back row, left-to-right: George Hunt, Lucy Homikanis Hunt, Franz Boas. Seated at front: Anislaga Mary Ebbetts Hunt (George’s mother). The Hunt children stand and sit around them. "The Hunt family at Tsaxis" (1894) by O.C. Hastings is used under the American Philosophical Society's open access policy.[/caption] Nearly a decade later, Hunt married his second wife, a high-ranking Kwakwaka’wakw woman named Tsukwani Francine Hunt (Bruchac 2018: 35). Their marriage inspired Hunt to continue his ethnographic work with Boas. While sleeping, Tsukwani sometimes sang and spoke to spirits. With Hunt nearby transcribing what she said, he recorded what was otherwise inaccessible information for Boas (Bruchac 2018: 37). Hunt usually credited his female informants in his manuscripts, but Boas did not include these attributions in the texts’ final versions (Bruchac 2018). Boas diminished the important cultural role of Kwakwaka’wakw women “as carriers and inheritors of tradition” when he dismissed the vital contributions of Lucy Homikanis Hunt and Tsukwani Francine Hunt (Bruchac 2018: 24). [caption id="attachment_6177" align="aligncenter" width="280"] Figure 4. Tsukwani Francine Hunt. "A Chief's daughter - Nakoaktok" by Edward S. Curtis is in the public domain.[/caption]
Integral insights, discredited: Boas and Hunt
Boas also devalued Hunt’s writing abilities and transcriptions. Hunt often spelled Kwak’wala words closer to their phonemic transcription, meaning he represented more of the individual sounds in his writing. Boas called these practices ‘errors,’ despite some actually being linguistic insights. Boas complicated his orthography by excessively marking vowel allophones or variants and omitting secondary articulatory features on consonants. Instead of using Boas’ method, Hunt’s transcriptions would, for example, represent certain vocalic segments as non-phonemic schwas preceded by a labialized consonant. This “treatment of non-phonemic schwa” influenced Boas’ later unpublished writing about Kwak’wala root structure (Berman 1994: 46). Although Hunt was technically credited by Boas, “Boas asserted intellectual dominance by characterizing himself as Hunt’s language instructor, editor, and proofreader” in his acknowledgments (Bruchac 2018: 41). Hunt and Boas co-authored three books on Kwak’wala, with Boas publishing additional articles and texts under his own name from Hunt’s ethnographic collection (Bruchac 2018: 21). Seeing “Boas’ name alone on the cover of most of these volumes” misrepresents Hunt’s immense contributions to these ethnographic and linguistic materials (Berman 2001: 204). Boas and Hunt’s forty years of work is invaluable, but it is not without bias. Hunt and Boas only give partial insight to Kwakwaka’wakw life at the time. While “Boas’ output on the Kwakwaka’wakw was dependent upon Hunt’s vast knowledge and ceaseless labor,” Hunt himself was dependent on the essential input and generosity of his wives Lucy and Tsukwani, their families, and other Kwakwaka’wakw members (Berman 2001: 204). However, their level of input is not enough. Even still, their voices are filtered through Hunt and Boas’ writing. Lucy and Tsukwani are cultural knowledge carriers; this filtration is clearly unnecessary. Hunt and Boas’ work only tells one part of the story. This history begs the questions, whose voices are centred and what might be missing?

Connecting this history to modern linguistics

How can we learn from Boas’ mistakes?

What do we need to keep in mind when using Boas and Hunt’s research in current times?

Hunt and Boas’ research notes are still actively being used in modern linguistics research. Is it possible to use it ethically and critically? Why and how, or why not?

Who should control the archives of Boas and Hunt’s work?

Key takeaways

key takeaway insert for coding

Check yourself!
[h5p id="65"]
References and further resources

Academic sources

Berman, Judith. 1994. Raven and sunbeam, pencil and paper: George Hunt of Fort Rupert, British Columbia. Ms., University of Pennsylvania. https://www.academia.edu/33041956/Raven_and_Sunbeam_pencil_and_paper_George_Hunt_of_Fort_Rupert_British_Columbia

Berman, Judith. 2001. Unpublished materials of Franz Boas and George Hunt: A record of 45 years of collaboration. In Gateways: Exploring the legacy of the Jesup North Pacific expedition, 1897-1902, ed. Igor Krupnik and William W. Fitzhugh, 181–213. Washington, D.C.: Arctic Studies Center, National Museum of Natural History, Smithsonian Institution.

🔍 Bruchac, Margaret M. 2018. Finding our dances: George Hunt and Franz Boas. In Savage kin: Indigenous informants and American anthropologists, 20–47. Tucson: University of Arizona Press.

Eriksen, Thomas H. 2015. Anthropology and history. In International encyclopedia of the social & behavioral sciences, 2nd edition, ed. James D. Wright, 1: 765–771. Elsevier. https://doi.org/10.1016/B978-0-08-097086-8.03025-7

Chapter 2 practice exercises

juliedoner — Wed, 22 Nov 2023 20:39:43 +0000

Data analysis

Exercise 1. Prescriptive or descriptive?
Beginner [2.3] Consider the following sentences. Some of them are prescriptively incorrect and some are descriptively ill-formed. Which are which?

a. I didn't say nothing.

b. Who should I pass the present to?

c. The cat black chased the mouse.

d. My daughter seems to happily go wherever we take her.

e. Is supper ready yet?

Exercise 2. Applying the scientific method
Intermediate-advanced [2.1, 2.2] In Section 2.1, we used the scientific method to identify the difference between we and us. In this exercise, we will use the scientific method to identify the difference between our and ours.
a. Come up with some sentences that use the words our and ours.

b. Make an observation about your sentences. Do you notice any patterns?

c. Turn your observation into a hypothesis.

d. Check to see if your hypothesis is [pb_glossary id="495"]falsifiable[/pb_glossary]. What would be the [pb_glossary id="1072"]necessary[/pb_glossary] and [pb_glossary id="1074"]sufficient[/pb_glossary] evidence to support your hypothesis? What is the opposite hypothesis? What would be the necessary and sufficient evidence to support the opposite hypothesis?

e. Come up with some sentences that test your hypothesis. Make sure you use both positive and [pb_glossary id="1022"]negative evidence[/pb_glossary].

f. Do you need to revise your hypothesis? Why or why not?

Exercise 3. Mental vs. written lexicon
Beginner [2.4] Which of the following pieces of information would you expect to be encoded in the mental lexicon? Which would typically be encoded in a dictionary?
a. The plural of desk is desks.

b. The words guarantee and warranty have the same historical origin.

c. The idiom raining cats and dogs means that it's raining very hard.

d. The abbreviation tl;dr means 'too long; didn't read' and is written at the beginning of long internet posts followed by a summary of the post.

e. My two-year-old niece calls juice /d͡ʒud͡ʒu/.

f. The plural of child is children.

Communication and study skills

Exercise 4. Scientific terminology
Beginner [2.6] The following paragraph misuses scientific terminology. Rewrite the paragraph using terminology correctly. Hint: If you're stuck, re-read the subsection on word choice in scientific writing from Section 2.6.
There is a significant distinction between the pronouns we and us. The following examples prove that they are not interchangeable.

(1) a. We are happy.

b. *Us are happy.

c. Joni saw us.

d. *Joni saw we.

From looking at the examples in (1), it's obvious that we and us are used differently.

Exercise 5. Personal language reflection
Intermediate [2.3] Write a paragraph describing your language background. Some questions you may answer in your response include:
a. What language(s) do you speak?

b. What language(s) did your ancestors speak?

c. Do the language varieties you speak have prestige in your society?

d. How is your identity connected to your language(s)?

e. How did you and your family come to speak the languages and language varieties you speak?

f. How do prescriptivist attitudes affect you, your family, and your community?

Exercise 6. Citation practice
Beginner [2.6] The following definition is taken from the 4th edition of Syntax: A Generative Introduction by Andrew Carnie on page 27.
Syntax (as part of grammar): The level of linguistic organization that mediates between sounds and meaning, where words are organized into phrases and sentences.
-Carnie 2021: 27

a. How would you write and cite a direct quote of this definition? Write it out using proper format for an in-text citation using parentheses.

b. How would you cite a paraphrase of this definition? Re-write the definition in your own words, using proper format for an in-text citation using parenthesis.

c. All of the information needed to write a bibliography entry for this textbook is listed below. Write a bibliography entry in the style of the Canadian Journal of Linguistics.

Title: Syntax: A Generative Introduction Edition: 4th Author: Andrew Carnie Publisher: Wiley Blackwell Year: 2021 Publication location: Malden, MA

Exercise 7. Evaluating sources
Intermediate [1.3, 2.5, 2.7] Consider the following sources. For each source, determine (i) if it is an academic source, (ii) the primary sub-field(s) of linguistics relevant to the source, and (iii) write a citation for the source using Canadian Journal of Linguistics citation style.
a. Domains in Michif phonology by Nicole Rosen

b. Blackfoot language by Glenbow Museum

c. Modeling the noun morphology of Plains Cree by Conor Snoek, Dorothy Thunder, Kaidi Lõo, Antti Arppe, Jordan Lachler, Sjur Moshagen, and Trond Trosterund

d. The Mohawk language by Marianne Mithun

e. English-Cayuga/Cayuga-English dictionary by Frances Froman, Alfred Keye, Lottie Keye, and Carrie Dyck

Research and application

Exercise 8. Dictionary vs. lexicon
Intermediate [2.4] In this exercise, we will compare a dictionary with your mental lexicon.
a. Think of a word you know that you expect to not be in the dictionary. Why not? Then look it up in a dictionary and see if it’s there.

b. Find a word in a dictionary that is not in your mental lexicon. Why do you think is it not in your mental lexicon?

Language Journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 9. Prescriptivism
Advanced [2.3] Research the prescriptivist attitudes of your language. Here are some questions to guide you. Not all of them will be applicable to all languages.
a. What are some of the different language varieties of your language? Is one more prestigious than the others?

b. How was your language affected by colonialism? Has it been a colonial language? Is it in contact with a colonial language?

c. Is there a language academy that determines what is "proper" use of the language?

Exercise 10. Pronouns
Beginner [2.1] What pronouns are used in your language? Does your language use different pronouns in subject and object position?

Chapter 8 practice exercises

juliedoner — Fri, 24 Nov 2023 23:59:35 +0000

Data analysis

Exercise 1. Identifying parts of speech
Beginner-Intermediate [8.1-8.6] Label the parts of speech of all of the words in the following sentences taken from A Tale of Two Cities by Charles Dickens. Labels you should use: noun (N), pronoun (PN), determiner (Det), verb (V), tense (T), preposition (P), adjective (Adj), adverb (Adv), conjunction (Conj), complementizer (Comp), and negation (Neg).
a. There were a king with a large jaw and a queen with a plain face, on the throne of England. b. With drooping heads and tremulous tails, they mashed their way through the thick mud, floundering and stumbling between whiles, as if they were falling to pieces at the larger joints c. There was a steaming mist in all the hollows, and it had roamed in its forlornness up the hill, like an evil spirit, seeking rest and finding none. d. Two other passengers, besides the one, were plodding up the hill by the side of the mail. e. In those days, travellers were very shy of being confidential on a short notice, for anybody on the road might be a robber or in league with robbers. f. Our booked passenger showed in a moment that it was his name. g. The guard, the coachman, and the two other passengers eyed him distrustfully. h. Take that message back, and they will know that I received this, as well as if I wrote. i. I shall not go to bed till night; but I want a bedroom, and a barber. j. He had a healthy colour in his cheeks, and his face, though lined, bore few traces of anxiety.

Exercise 2. Identifying parts of speech
Intermediate [8.1-8.5] Consider the Spanish examples below. Based on this data, form hypotheses about criteria that can be used to identify parts of speech in Spanish. If it's helpful, use the chart included below to organize your answers. Examples (1b)-(1d) are taken from Como agua para chocolate by Laura Esquival. Note: There may not be enough data to fill out all of the cells in the chart, but see what you can find!
Glossing abbreviations: CONT = continuous (ongoing action, DEM = demonstrative (i.e., this or that), DET = determiner, DIM = diminutive (a small thing), F = feminine, M = masculine, PL = plural, PRES = present, PROG = progressive, REFL = reflexive, SG = singular.

(1) a. La viej-it-a está espera-ndo el autobús.

DET.F old-DIM-F be.PRES.3SG wait.PROG DET.M bus

'The old lady is waiting for the bus.'

b. Éste dorm-ía apacible-mente junto a su herman-a.

DEM.M sleep-PAST.3sg peaceful-ly together to his sibling-F

'This one slept peacefully next to his sister.'

c. Tita la escuchaba con gran interés.

Tita 3SG.F listen.PAST.CONT.3SG with big interest

'Tita was listening to her with great interest.'

d. Giró la cabeza y sus ojos se encontraron con los de Pedro.

turn.PAST.3SG DET.F head and her.PL eye.PL REFL met.PAST.3PL with DET.PL of Pedro

'She turned her head and her eyes met Pedro's.'

Part of speech Syntactic distribution Inflectional morphology Derivational morphology

Noun

Verb

Adjective

Adverb

Preposition

Determiner

Exercise 3.
One way that we can classify verbs into subcategories is through a property called lexical aspect.
Exercise 4. Language evolution
Intermediate [8.1, 8.5] Consider the following sentences. What is the part of speech of because in the sentences in (2)? (2) a. I ate my lunch early because I was hungry. b. Because of the changes in communication style since the advent of the Internet, new types of slang are being used. Now consider the sentences in (3), which represent newer, more playful use of the word because. What is the part of speech of because in the sentences in (3). (3) a. I have secrets because reasons. b. Language is changing because Internet. See also the book Because Internet by Gretchen McCulloch on the ways that Internet has been changing language!
Communication and study skills

Exercise 5. Predicate-argument flexibility
Intermediate [8.9] The following examples, from a language called Gitksan, illustrate a property of several languages called predicate-argument flexibility. Explain how these examples demonstrate why we should not use semantic criteria to identify the part of speech of a word. Gitksan is spoken in British Columbia and belongs to the Tsimshianic language family. Abbreviations AFF = affirmative particle, CN = common noun connective, SX = subject extraction marker.

(3) a. hanaq̕=ɬ simʔo:gid(-id)=ist

woman=CN chief(-SX)=AFF

"The chief is a woman.'

b. simʔo:git=ɬ hanaq̕=ast

chief=CN woman=AFF

"The woman is a chief.'

(4) a. hanaq̕=ɬ c̕aw-ad=ast

woman=CN smart-SX=AFF

'The smart one is a woman.'

b. c̕aχʷ=ɬ hanaq̕=ast

smart=CN woman=AFF

'The woman is smart.'

(5) a. w̓itxʷ=ɬ hanaq̕=ast

arrive=CN woman=AFF

'The woman arrived.'

b. hanaq̕=ɬ w̓itxʷ-id=ist

woman=CN arrive-SX=AFF

'The one who arrived is a woman.'

[Gitksan; Davis et al. 2014: 196-197]

Exercise 6. Word creation
Intermediate [8.1, 8.5] Invent a new word in a language you speak and compose a dictionary entry for it by answering questions (a)-(f) and then reflect on the properties of your word by answering questions (g)-(h).
a. How do you spell your new word? b. What language does your new word belong to? c. How do you pronounce your word? (Use IPA if you know it) d. What is the part of speech of your new word? e. What is the definition of your new word? f. Provide an example sentence containing your new word. If your new word is not an English word, gloss your example. g. Identify morphological or syntactic characteristics that support the part of speech you identified in question (d). Explain how it supports your classification. You can make reference to your example sentence in (f). h. How hard would it be to get your word to catch on? Explain, making reference to the part of speech of your invented word.
Note: This assignment is inspired by Frindle, a book by Andrew Clements about a kid who invents a new noun, frindle, which means ‘pen,’ and convinces all his friends to use the word, too. The new word catches on and even makes it into the dictionary a few years later. Exercise 7. Part 1. Modals and auxiliaries are each subcategories of the T category. Write a short 1-2 page essay based on the data below explaining (a) why they should both be considered part of the T category and (b) but why they should be split into two subcategories. multiple auxiliaries, not multiple modals inflected auxiliaries, not inflected modals Part 2. Must and have to have almost the same meaning, but must is a modal. Is have to also a modal? Explain whether have to has the same properties as modals. Construct examples using have to to support your argument. Hint: You can use the data from Part 1 as a starting part.
Exercise 7. Slang analysis
Intermediate [8.1-8.6] Think of a slang word or a word that has only recently entered English or another language you speak. What part of speech is it? Construct examples that back up your answer. Is it an open or closed part of speech?
Language journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 8. Identifying parts of speech
Intermediate [8.1-8.6] Find examples of nouns, verbs, adjectives, and adverbs in your language. What syntactic and morphological criteria can you use to identify them?
References and further resources

Sources for examples

Davis, Henry, Carrie Gillon, and Lisa Matthewson. 2014. How to investigate linguistic diversity: Lessons from the Pacific Northwest. Language 90(4): e180–e226.

3.2. Generative Grammar

juliedoner — Thu, 30 Nov 2023 06:26:43 +0000
In this textbook, the model that we will be learning together belongs to the family of models called [pb_glossary id="1213"]Generative Grammar[/pb_glossary]. It was Noam Chomsky that came up with the idea that models of grammar should be generative. He defines generative grammar as "a system of rules that in some explicit and well-defined way assigns structural descriptions to sentences" (Chomsky 1965: 8). In other words, a generative grammar uses rules to generate or "build" the structure of sentences. Santorini and Kroch (2007) define it as "an [pb_glossary id="1344"]algorithm[/pb_glossary] for specifying, or generating, all and only the grammatical sentences in a language." What's an algorithm? It's simply any finite, explicit procedure for accomplishing some task, beginning in some initial state and terminating in a defined end state. Computer programs are algorithms, as well as recipes, knitting patterns, the instructions for assembling an Ikea bookcase, or a list of steps for balancing your checkbook. An important point to keep in mind is that it is often difficult to construct an algorithm for even trivial tasks. A quick way to gain an appreciation for this is to describe how to tie a bow. Like speaking a language, tying a bow is a skill that most of us master around school age and that we perform more or less unconsciously thereafter. But describing (not demonstrating!) how to do it is not that easy, especially if we're not familiar with the technical terminology of knot-tying. In an analogous way, constructing a generative grammar of English is a completely different task than speaking the language, and much more difficult (or at least difficult in a different way)! Just like a cooking recipe, a generative grammar needs to specify the ingredients and procedures that are necessary for generating grammatical sentences. Not all models of grammar use a generative framework. In other kinds of grammar models, language is produced by repeating memorized fragments or by probabilistic modeling, which is more similar to how Large Language Models produce language.

Noam Chomsky

[caption id="attachment_1210" align="alignright" width="225"] Figure 1: Noam Chomsky in 2004. Photo by Duncan Rawlinson. Used under CC BY-NC 2.0 license.[/caption] Noam Chomsky (1928- ) is perhaps the most well-known linguist in the world. Beginning with his 1955 dissertation Transformational Analysis and his 1957 book Syntactic Structures, Chomsky revolutionized the ways we think about language and linguistics and founded the modern field of linguistics. As we already learned in this section, Chomsky was the first to explicitly model language as a rule-governed system, launching the study of Generative Grammar. Another foundational proposal by Chomsky was the idea of [pb_glossary id="180"]Universal Grammar[/pb_glossary], which is the idea that humans are genetically endowed with the capacity for language. We will learn more about Universal Grammar in the remainder of this chapter. Grammatical theory has changed a lot since 1955, but Chomsky has been a key player throughout this time. Most of the grammatical models that have been developed since then were either developed within a Chomskyan framework, sometimes by Chomsky himself, or in direct opposition to it. Chomsky is also well-known for his political writings and activism. At the time of this writing, Chomsky, although retired, is still publishing papers and giving talks. He is professor emeritus at the Massachusetts Institute of Technology and a laureate professor at the University of Arizona.

Some useful distinctions
When we are talking about our model of grammar, there are some useful distinctions we should make.
Competence vs. performance
Sometimes when we produce language, words don't come out exactly the way we intend. Because of this, we need a distinction between competence and performance. If you have linguistic [pb_glossary id="169"]competence[/pb_glossary] in a language, then you have acquired the grammatical rules necessary to produce the language in question. If you have linguistic [pb_glossary id="170"]performance[/pb_glossary], the language you produced conforms to the rules of the grammar you are using. Most of the time, we have both competence and performance. But it is possible to have competence without performance. For example, if you are drunk or sleepy, you are more likely to misspeak. You still have the rules of grammar in your head, so you have competence. But you may have trouble accessing or implementing the grammar rules, so the end product of your language use does not conform to the rules of grammar in your head, so you do not have performance. You can also have performance without competence. For example, say you memorize a sentence from a language you don't speak. You can repeat it, perhaps even flawlessly, so you have linguistic performance. However, you do not have the grammatical rules in your mind necessary to construct that sentence from scratch -- you are only repeating what you have memorized -- so you don't have competence in that language.
I-language vs. E-language
The next distinction we should make is between [pb_glossary id="1257"]I-language[/pb_glossary] and [pb_glossary id="1258"]E-language[/pb_glossary]. I-language stands for internal language and refers to the system of grammatical rules that an individual language user has in their mind. Everyone has a slightly different I-language. E-language, on the other hand, stands for external language, and refers to how language is externalized, including how it is used in a community. Since everyone has slightly different I-languages and because of the effects of linguistic performance, the E-language might not be consistent. In Generative Grammar, what we are trying to model is the properties of human I-languages. However, we cannot access I-language directly, since it is a cognitive object. Instead, we infer the properties of I-language from the properties of particular E-languages. I-languages is also sometimes called Language (with a capital L) while e-language can be called language (with a small l).
Synchronic vs. diachronic
It is sometimes useful to look at how language changes over time, which is called the [pb_glossary id="1264"]diachronic[/pb_glossary] study of language. Although historical linguistics can be very interesting indeed, our model of grammar needs to be a model of language at a particular time, which is called the [pb_glossary id="1266"]synchronic [/pb_glossary] study of language. Often, this means studying modern languages, but it can also mean studying a historical language at a particular period. The history of a language is not encoded in its grammar. Most speakers, unless they have specifically studied it, do not know the history of the languages they speak (and if they have studied it, they very likely did so after their language was acquired as young children). Because the history of the language is not part of what most speakers know, we cannot use a historical explanation in our model of grammar. The historical explanation can be useful for explaining why the grammar has one set of rules and not another, but the rules themselves need to work as a system independent of where they came from. Let's use riding a bicycle as an analogy. I could know the history of the bicycle, where each piece of metal was mined and smelted, and where the bike was assembled. All of those processes had to happen in order for the bike to exist for you to ride, but that knowledge isn't necessary to be able to ride the bike. What is necessary is that your bike is properly assembled, with the peddles linked to one of the wheels with a chain. When you peddle your bike, the peddles move the chain, which in turn rotates the wheel and moves the bike forward. This chain reaction between the parts of your bike is kind of like a grammar. The pedals, the chain, and the wheel are the different parts of a system that work together to make your bike work. It doesn't matter whether the chain on your bike is the original one or has been replaced, it just matters that it is working now. In the same way, it is not necessary to know the history of your language in order to use the language. When we are trying to explain how our model of Language works, it doesn't matter if a particular rule was original to the language or borrowed from a different language -- it is part of the system now, and we need to explain how it works now. [caption id="attachment_1267" align="aligncenter" width="300"] Figure 2: A bicycle. Photo by LUM3N. Used under Pixabay Content License.[/caption]
Check yourself!
[h5p id="19"]
References and further resources

Attribution
Portions of this section are adapted from the following CC BY NC source:
↪️ Santorini, Beatrice, and Anthony Kroch. 2007. The syntax of natural language: An online introduction. https://www.ling.upenn.edu/~beatrice/syntax-textbook

For a general audience

Enos, Casey. No date. Noam Chomsky. Internet Encyclopedia of Philosophy. https://iep.utm.edu/chomsky-philosophy

Academic sources

Chomsky, Noam. 1955. Transformational Analysis. Doctoral dissertation, University of Pennsylvania.

Chomsky, Noam. 1957. Syntactic Structures. The Hague/Paris: Mouton.

Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.

3.3. The acquisition of morphosyntax

juliedoner — Thu, 30 Nov 2023 07:01:47 +0000
As mentioned in Section 3.1, the best model of grammar should be [pb_glossary id="1208"]explanatorily adequate[/pb_glossary], which means that it should be able to account for all the data we observe, all of the grammaticality judgments we collect, and how the rules of grammar are acquired. Because of this, although Language Acquisition is an entire field of study on its own, morphologists and syntacticians also care about the language acquisition process.
Learning vs. acquisition
You may have noticed that we have said a lot about language "[pb_glossary id="1313"]acquisition[/pb_glossary]" but not about language "learning." This is because, in linguistics, we distinguish between the two. When we say "acquisition," we mean the development of something (in this case, language) subconsciously, by instinct. When we say "learning," on the other hand, we mean that it developed consciously, through study and deliberate practice. One piece of evidence that children acquire language, rather than learn it, is the course and timing of language development. Children typically acquire the majority of their grammar by the age of 4, before they go to school. This happens regardless of the language(s) they are exposed to. It also happens in different cultural contexts, in communities with vastly different traditions on how to interact with and speak to young children (see Figueroa in press and references therein). Children even begin learning the [pb_glossary id="482"]prosody[/pb_glossary] of the language in their environment before they're even born (Gervain 2013)! Children will usually not even repeat utterances that have grammatical structures that they have not yet acquired—to the point that asking children to repeat an experimenter's utterance can be used in acquisition studies (for example, Kidd et al. 2006). Compare that to the study of language in a classroom as a teenager or adult. In language classrooms, we often memorize and recite vocabulary and grammatical patterns. It takes effort and often makes your brain hurt! If you've ever studied a language as an adult or a teenager, think about how the process differed for you from acquiring a language as a child. If you're like me, there were times that you could recite the verb conjugations for a test, but when it came time to use them in a real conversation, you couldn't find the right form in your mind quick enough. This is what it feels like when you've learned the grammar consciously, but you haven't acquired it subconsciously. One of the puzzles of language acquisition is the fact that children do not seem to receive negative evidence. The majority of the utterances produced by the adults in their environment are grammatical. If the adults misspeak and produce an ungrammatical utterance, there is often nothing to indicate that they misspoke. It is difficult to identify even indirect ways children might get negative evidence. Some studies show that parents comprehend children's ungrammatical utterances as frequently as their grammatical utterances, and that parents rarely correct children's ungrammatical utterances, and instead respond to the content of their child's speech (Brown and Hanlon 1970, as cited in Marcus 1993). When parents do correct their children's grammar, it doesn't usually work very well. Some famous examples include the following:
For experimental purposes, I have occasionally made an extensive effort to change the syntax of my two children through correction. One case was use by my two-and-a-half-year-old daughter of other one as a noun modifier. Over a period of a few weeks I repeatedly but fruitlessly tried to persuade her to substitute other + N for other one + N. With different nouns on different occasions, the interchanges went somewhat as follows:
"Want other one spoon, Daddy." "You mean, you want THE OTHER SPOON." "Yes, I want other one spoon, please, Daddy." "Can you say ‘the other spoon’?” “Other one spoon.” “Say ‘other’.” “Other.” “Spoon.” “Spoon.” “Other spoon.” “Other. spoon. Now give me other one spoon?”
Further tuition is ruled out by her protest, vigorously supported by my wife. Examples indicating a similar difficulty in using negative information will probably be available to any reader who has tried to correct the grammar of a two- or three-year-old child.
(Braine 1971: 160-161, as cited in Marcus 1993)

Child: Nobody don’t like me. Mother: No. say “nobody likes me.” Child: Nobody don’t like me. [Eight repetitions of this dialogue follow.] Mother: No, now listen carefully, say “NOBODY LIKES ME.” Child: Oh! Nobody don’t likes me.
(McNeill 1966: 69, as cited in Marcus 1993)

If we, as linguists, require negative evidence to figure out how grammar works, how can children manage without it?
Rule-based language learning
Let's consider a toy version of English that contains three-word sentences consisting of a noun, a verb, and another noun. The toy version contains sentences like (1) that are sensible given the real world as well as sentences like (2) that aren't, but that might be useful in fairy tale or science fiction contexts.

(1) a. Cats detest lemons.

b. Children eat tomatoes.

c. Cheetahs chase gazelles.

(2) a. Lemons detest cats. (“Secret life of citrus fruits”)

b. Tomatoes eat children. (“Attack of the genetically modified tomatoes”)

c. Gazelles chase cheetahs. (“Avenger gazelle”)

Again for the sake of argument, let's assume a (small) vocabulary of 1,000 nouns and 100 verbs. This gives us a list of 1,000 x 100 x 1,000 (= 100 million) three-word sentences of the type in (1) and (2). Numbers of this magnitude are difficult to put in human perspective, so let's estimate how long it would take a child to learn all the sentences on the list. Again, for the sake of argument, let's assume that children can memorize sentences quickly, at a rate of one sentence a second. The entire list of three-word sentences could then be memorized in 100 million seconds, which comes to 3.17 years. So far, so good. However, the minute we start adding complexity to Toy English, the number of sentences and the time it would take to memorize them quickly mushrooms. For instance, adding only 10 adjectives to the child's vocabulary would cause the number of five-word sentences of the form in (3) to grow to 10 billion (100 million x 10 x 10).

(3) a. Black cats detest green peas.

b. Happy children eat ripe tomatoes.

c. Hungry cheetahs chase speedy gazelles.

Even at the quick rate of one sentence per second that we're assuming, the list of all such five-word sentences would take a bit over 317 years to learn. Clearly, this is an absurd consequence. For instance, how could our child ever come to know, as every English speaker plainly does, that the sentence in (4) is ungrammatical? If grammatical knowledge were based purely on rote memorization, the only way to determine this would be to compare (4) to all of the 10 billion five-word sentences and to find that it matches none of them.

(4) *Cats black detest peas green.

And even after performing the comparison, our fictitious language learner still wouldn't have the faintest clue as to why (4) is ungrammatical! In addition to this thought experiment with its comically absurd consequences, there is another reason to think that language acquisition isn't entirely based on rote memorization - namely, that children use what they hear of language as raw material to construct linguistic rules. How do we know this? We know because children sometimes produce rule-based forms that they have never heard before. One of the earliest demonstrations that children acquire linguistic rules, rather than simply imitating the forms of adult language, was the well-known [pb_glossary id="3130"]wug test[/pb_glossary] (Berko 1958). In it, the psycholinguist Jean Berko used invented words to examine (among other things) how children between the ages of 4 and 7 form plurals in English. She showed the children cards with simple line drawings of objects and animals and elicited plurals from them by reading them accompanying texts like (5).

(5) This is a wug. Now there is another one. There are two of them. There are two ___.

More than 75% of the children pluralized the invented words cra, lun, tor, and wug in exactly the same way that adults did in a control group: they added the sound -z to the word (Berko 1958: 159-162). Since none of the children had encountered the invented words before the experiment, their response clearly indicates that they had acquired a plural rule and were using it to produce the novel forms. Children are also observed to produce novel rule-based forms instead of existing irregular adult forms (for instance, comed or goed instead of came or went). This process, which is known as [pb_glossary id="1337"]overregularization[/pb_glossary], is further illustrated in (6) (Marcus et al. 1992: 148-149, based on Brown 1973).

(6) a. beated, blowed, catched, cutted, doed, drawed, drived, falled, feeled, growed, holded, maked, sleeped, standed, sticked, taked, teached, throwed, waked, winned (Adam, between the ages of 2 and 5)

b. drinked, seed, weared (Eve, between the ages of 1½ and 2)

Overregularized forms don't amount to a large fraction of the forms that children produce overall (less than 5% in the case of past tense forms, according to Marcus et al. 1992: 35), but they are important because they clearly show that even the acquisition of words can't be completely reduced to rote memorization.

Skinner vs. Chomsky and the debate on how children acquire language

B.F. Skinner was a psychologist who worked within a theory called behaviourism. In the theory of behaviourism, you can train people and animals to behave in certain ways by using positive and negative reinforcement (giving or taking away a desired stimulus) and positive and negative punishment (giving or taking away an undesired stimulus). In 1957, Skinner published a book called Verbal behaviour in which he argued that children's language development could be explained by behaviourism. Basically, a babbling baby would one day, by coincidence, babble dada and the parents would be excited and encourage the baby to repeat those sounds. Over time, the baby would learn to associate dada with their father, and then to learn increasingly complex language structures. Two years later, in 1959, Chomsky published a review of Skinner's book that eventually became even more famous than Skinner's original book! When Chomsky's review first came out, Chomsky was still relatively unknown, and Skinner mostly ignored it. He never formally refuted it, but continued to claim that Chomsky misunderstood his argument.

Check yourself!
[h5p id="20"]
References and further resources

Attribution
Portions of this section are adapted from the following CC BY NC source:
↪️ Santorini, Beatrice, and Anthony Kroch. 2007. The syntax of natural language: An online introduction. https://www.ling.upenn.edu/~beatrice/syntax-textbook

Comedy and satirical linguistics

🎉 Dresher, B. Elan. 2010. The Geese Rethink Innateness. Toronto Working Papers in Linguistics 32. https://twpl.library.utoronto.ca/index.php/twpl/article/view/13954

For a general audience

🧠 CrashCourse. 2020. Language Acquisition: Crash Course Linguistics #12. https://www.youtube.com/watch?v=Ccsf0yX7ECg

🧠 NOVA PBS Official. 2010. The secret life of scientists and engineers: Jean Berko Gleason and wugs. https://www.youtube.com/watch?v=ElabA5YICsA

Academic sources

Berko, Jean. 1958. The child's learning of English morphology. Word 14: 150–177.

Brown, Roger. 1973. A first language: The early stages. Cambridge, MA: Harvard University Press.

Braine, Martin D. 1971. On two types of models of the internalization of grammars. In Dan Slobin (ed.), The ontogenesis of grammar. New York: Academic Press. 153–186.

Brown, Roger & Camille Hanlon. 1970. Derivational complexity and order of acquisition in child speech. In John R. Hayes (ed.), Cognition and the development of language. New York: Wiley. 11–53.

Chomsky, Noam. 1959. Review of Verbal behaviour by B.F. Skinner. Language 35(1): 26–58.

Figueroa, Megan. (to appear). Language Development, Linguistic Input, and Linguistic Racism. WIREs Cognitive Science.

Gervain, Judit. 2018. The role of prenatal experience in language development. Current Opinion in Behavioral Sciences 21: 62–67.

Kidd, Evan, Elena Lieven, & Michael Tomasello. 2006. Examining the role of lexical frequency in the acquisition and processing of sentential complements. Cognitive Development 21(2): 93–107.

Marcus, Gary, Steven Pinker, Michael Ullman, Michelle Hollander, T. John Rosen, and Fei Xu. 1992. Overregularization in language acquisition. Monographs of the Society for Research in Child Development 57(4). Chicago: University of Chicago Press.

Marcus, Gary F. 1993. Negative evidence in language acquisition. Cognition 46(1): 53–85.

McNeil, David. 1966. Developmental Psycholinguistics. In Frank Smith and George A. Miller (eds.), The Genesis of Language: A Psycholinguistic Approach. Cambridge: MIT Press. 15–84.

Skinner, B.F. 1957. Verbal behaviour. New York-London: Appleton-Century.

3.7. Becoming a linguist: Empirical and theoretical arguments

juliedoner — Thu, 30 Nov 2023 07:02:04 +0000
There are two main kinds of arguments, empirical arguments and theoretical arguments. Empirical arguments are based primarily on observation of data, while theoretical arguments are based primarily on ideas. In this class, we will be using both. Often, more complex arguments combine both theoretical and empirical arguments.
Empirical arguments
Empirical arguments are based on data. They have three parts: make relevant observations, make a theoretical claim, and explain how your observations are related to your claim. The parts don’t have to be in this order, but they should be in an order that makes sense. In this chapter, the fact that children's knowledge of language is rule-based (as discussed in Section 3.3) is an example of an empirical argument. We observed examples of children's behaviour, such as their response to corrections from their parents, their pluralization of [pb_glossary id="247"]nonce words[/pb_glossary], and their use of [pb_glossary id="1337"]overregularization[/pb_glossary]. Based on our observations, we concluded that children's language is rule-based rather than memorized.
Making observations
When you make an empirical argument, you need to start by describing the surface properties of your data. You should be answering the question, “What are the non-controversial facts?” These are mostly observations that a non-linguist could make! Some examples of observations that you might start which are relevant to morphology and syntax include:

What does the sentence mean?

Which nouns and pronouns in the sentence refer to the same entity?

Is the sentence ambiguous? In other words, does it have more than one meaning?

What is the word order? Which words precede or follow which other words?

Are other word orders possible?

Is the sentence grammatical?

What similarities are there between this word and other words?

There are lots of observations that you could make about any sentence, but you need to stick to the ones that are relevant to your theoretical claims.
Making a theoretical claim
Although an empirical argument is based on some observation, it leads you to make some theoretical claim about the structure of your data. In this part, you should be answering the question, "What conclusions can you draw based on your observations?" This is where you talk about things you can’t see or hear, but that you infer based on what you do see or hear. Some examples of theoretical claims you might make in this course include:

Whether a word is [pb_glossary id="197"]simplex[/pb_glossary] or [pb_glossary id="198"]complex[/pb_glossary].

The classification of a word or sentence based on its structure.

The part of speech of a word (e.g., whether a word is a noun or a verb), as we will learn about in Chapter 5.

Whether a string of words behaves as a unit, which we will learn about in Chapter 16.

Backing up your claim
To make a complete argument, you also need to explain how your your observations provide evidence for your claim. You cannot assume that your readers will be able to draw the same conclusions as you! In some cases, you should also think about other claims people might think to make about the same data, and why your analysis is better.
Theoretical arguments
Theoretical arguments are based on ideas, logic, and reasoning, instead of data. Because of this, we can never trust them as much as empirical arguments. However, we wouldn’t be able to get as far in science without theoretical arguments, too! In this chapter, the Poverty of the Stimulus argument in Section 3.4 is an example of a theoretical argument. A theoretical argument has two parts: premises and a conclusion.
Laying out your premises
A theoretical argument is based on premises, which are the ideas that you use as a foundation to draw your conclusion. They are your claims about how things work. A premise can be a conclusion from a previous (theoretical or empirical) argument, or it can be an easily observable fact (e.g., language is infinite; humans acquire language; not all languages are the same). For your argument to work, you need to be relatively certain that your audience will accept your premises. The Poverty of the Stimulus argument has two premises: that language is infinite and that infinite systems are unlearnable. One of the ways to argue against a theoretical argument is to claim that one (or more) of the premises is false.
Drawing a conclusion
Your premises should logically lead to a conclusion. It's important that you actually state what that conclusion is, though, instead of making your readers figure it out. Another way to argue against a theoretical argument is to claim that the conclusion does not actually follow from the assumptions.
Consider your assumptions
Every time you make an argument, you make theoretical assumptions. These can be foundational, like “I am assuming generative grammar is correct,” or structural, like “I’m assuming that two things can’t occupy the same spot,” or methodological, like “I’m assuming that grammaticality judgments are reliable.” A lot of times, we are making assumptions without realizing it, so it’s always good to spend some time thinking about what assumptions we are making. Decide whether you should be making your assumptions, or whether you should dig deeper and investigate some of them. Sometimes it’s useful to make assumptions, though, to see where a theory will lead us, or because if we had to start from scratch with no assumption every time, we wouldn’t make it very far very fast. Decide whether you need to state your assumptions. You don’t need to state any assumptions that you are pretty sure your audience shares. This happens a lot in a class setting—we share the assumptions of the class materials. You also don’t need to state any assumptions that are pretty obvious from your argument. But beware! Sometimes what is obvious to you will not be obvious to your reader. When in doubt, state your assumptions.
Presenting your argument
When you write an argument, you should clearly state all of the components of your argument and how they are connected. Sometimes you will have to make a chain of arguments. For example, if we were to argue that a language has prepositions, we first have to identify an adposition and which noun it goes with, and then show that the adposition goes before that noun. Sometimes a chain of argumentation will include both empirical arguments and theoretical arguments. For example, our argument that children use rule-based language actually has both empirical and theoretical components. As you lay out all of the arguments in your line of reasoning and each of their components, think about what order it is best to lay out your claims. Do any of your points depend on another point? Always make sure you use technical terminology where appropriate. Technical terminology is useful because it is normally concisely defined to avoid ambiguity. For example, saying un- is a morpheme is more precise than saying un- is part of a word.
Check yourself!
[h5p id="23"]

2.4. The lexicon and dictionaries

juliedoner — Tue, 12 Dec 2023 04:51:34 +0000
When we learn a language, there are two types of things we know. First, we know the descriptive grammar rules of the language, which we will learn more about in Chapter 3, and we know vocabulary from the language. We store our vocabulary in our [pb_glossary id="1379"]mental lexicon[/pb_glossary]. The lexicon is the part of our memory where we store our knowledge of all the words we know. This includes how to pronounce the word, what the word means, and anything else about the word that is not predictable. That is, anything about the word that we have to memorize rather than that we can predict based on the grammar rules of the language. For example, for a noun with an irregular plural like child ~ children or a verb with an irregular past tense like go ~ went, we would have to store the irregular form in our lexicon. But for nouns with regular plurals like cat ~ cats and verbs with regular past tenses like walk ~ walked, we would not store it, instead we would build it from the rule. Likewise, some complex words have unpredictable or [pb_glossary id="1694"]non-compositional meanings[/pb_glossary]. For example, the word edible doesn't simply mean "can be eaten"; it means "can be eaten without ill effect to your health." The idiosyncratic component of meaning must also be stored in the lexicon. Psycholinguistic studies show that productive complex words that are frequent may also have their own entries in the lexicon and may be accessed as unanalyzed wholes (Sims-Williams 2021). This means that in future language change, the stem and the complex word created from it might take separate paths -- one can pick up idiosyncratic meanings or take on changes in pronunciation that the other does not. Some examples of this include brother, which formerly had an irregular plural brethren. The irregular plural picked up an idiosyncratic meaning (religious co-affiliation) and a new, regular, plural form, brothers was coined. Likewise, the plural of cloth used to be clothes. The irregular plural picked up the idiosyncratic meaning of garment and a new plural, cloths, was coined. Deciding exactly which parts of language are productive, and can be built from our grammar rules, versus which parts of language are memorized in our mental lexicon is an important question in the study of morphosyntax. Another related question is when we access the lexicon as we build our sentences and words. There are three main options:

We access the lexicon only at the beginning. We pick out which words or morphemes we are going to use at the beginning of the process, and then we move them to a workspace where we can combine them to build a structure.

We access the lexicon only at the end. We build a structure using abstract pieces, and then we plug in the appropriate words at the end.

We access the lexicon multiple times throughout the process, inter-weaving the selection of words and the building of structure.

Dictionaries
A dictionary is kind of similar to a mental lexicon, because it also records the vocabulary of a language. But there are some important differences between a mental lexicon and a dictionary, as summarized in Table 1.

Table 1. The differences between mental lexicons and dictionaries

Our mental lexicons Dictionaries

Created instinctively and subconsciously based on daily use. Created for specific purposes by [pb_glossary id="1389"]lexicographers[/pb_glossary].

Includes slang and informal language. Less likely to include slang and informal language.

Instantly updates with exposure to new language innovations. Lags a few years behind language innovations.

Only includes terms known by the language user. May include technical and archaic terms.

Includes [pb_glossary id="651"]idioms[/pb_glossary] and longer pieces of non-compositional meaning. Rarely includes idioms.

Highly individual. Tries to describe standard or common usage.

Linked to detailed information about the cultural background and real-world knowledge. Usually minimal information about the cultural background and real-world knowledge included.

Includes all meanings and uses of a word. Includes only the most common meanings and uses of a word.

May not include spelling and does not include [pb_glossary id="2479"]etymology[/pb_glossary]. Usually includes spelling and sometimes includes etymology.

Some of the things we will learn together are things that you might be able to look up in a dictionary, like the part of speech (noun, verb, etc.) of a word. But in this course (and most linguistics courses), we want to learn how to figure it out ourselves, not how to look it up. There are a few reasons for this. First of all, dictionaries aren't always accurate. As mentioned in Table 1, dictionaries often lag behind language change and also only include the most common or standard patterns of language. In our linguistic analyses, we want to consider all the possible structures, not just the most common ones. Second, as a linguistics student, you are training to become the kind of person who will one day be qualified to write dictionaries. You can look things up in a dictionary to confirm your analysis or get a second opinion, but if you depend only on the dictionary, then you are skipping part of the learning process! [caption id="attachment_2476" align="aligncenter" width="1024"] Figure 1: The Dictionary of Proto-Algonquian by David Pentland. This dictionary includes a lot of etymological information about numerous Algonquian languages, which would not be included in the mental lexicons of speakers. Photo by Julie Doner. Reproduced with permission.[/caption]

Key takeaways

We store our vocabulary and all of the parts of language that are not predictable from rules in our mental lexicon.

Whereas dictionaries are curated intentionally for specific purposes, mental lexicons are created subconsciously and includes all of the forms known by the language user.

You should avoid looking up the answers for the exercises in this course in the dictionary.

Check yourself!
[h5p id="21"]
References and further resources

For a general audience

🧠🔍 Curzan, Anne. 2014. What makes a word "real"? TedxUofM. https://www.ted.com/talks/anne_curzan_what_makes_a_word_real

For linguistics students

🧠 Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Section 7.4: The mental lexicon. Essentials of Linguistics, 2nd edition. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/chapter/7-4-the-mental-lexicon

🧠 Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Section 7.10: Why not the dictionary? Essentials of Linguistics, 2nd edition. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/chapter/7-10-why-not-the-dictionary

Sources for examples

Pentland, David. 2023. Proto-Algonquian Dictionary: A Historical and Comparative Dictionary of the Algonquian Languages. Winnipeg: Algonquian and Iroquoian Linguistics.

Academic sources

Sims-Williams, Helen. 2021. Token frequency as a determinant of morphological change. Journal of Linguistics 58 (3): 571–607.

Chapter 3 practice exercises

juliedoner — Fri, 15 Dec 2023 18:01:50 +0000

Data analysis

Exercise 1. Competence vs. performance
Beginner [3.2] Consider each of the following situations. Does the language user in each case have competence only, performance only, both, or neither? Explain your answer.
a. Sandy has just been to the dentist and now her mouth is numb. When she talks, it is difficult to understand her, but the numbness will wear off soon and Sandy will be able to talk clearly again.

b. Kai is 18 months old. He knows several words but doesn't combine them into sentences.

c. Cherry the dog knows the word walk and gets very excited every time she hears the word walk. When Cherry's owner twists his ankle, he tells his mom on the phone, "I can't walk." Cherry hears this, barks excitedly, and fetches her leash.

d. Nissa and Remi are chatting at a party. Nissa is telling Remi about her shopping trip, when Remi spills his drink. Nissa stops midsentence to help Remi clean up and forgets what she had been talking about.

e. A company has just released a new AI program called ASLBot and everyone is talking about it. The program is able to produce videos of a cartoon character signing in American Sign Language based on a Large Language Model. You can even turn on your webcam and have a relatively natural two-way conversation with ASLBot that seems fairly human-like. You research how ASLBot is programmed, and you discover it is programmed based on a probabilistic model of language, rather than a generative one.

Communication and study skills

Exercise 2. Theoretical and empirical arguments
Advanced [3.7] Pick an argument from the textbook or from the lecture to analyze.
a. Is it a theoretical argument or an empirical argument? Or does it have bits of both?

b. For the empirical arguments: What is the observable evidence? What is the claim? How was the claim backed up? Can you think of another way to analyze the same data?

c. For the theoretical arguments: What are the assumptions, the premises, and the conclusions? Do you agree with the premises and the assumptions?

Research and application

Exercise 3. Trick an LLM
Advanced [3.6] Explore a freely-available LLM program and see if you can find some of its weaknesses. Explain what you find.
a. Can you find a tricky grammatical construction that it misunderstands?

b. Did it give you any false information?

c. Did its responses reveal any biases or stereotypes?

Language journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 4. Find a grammar
Beginner [3.1] Are there any published grammars about your language available in your university library? Are they pedagogical, descriptive, or theoretical grammars?

3.6. Computer use of language

juliedoner — Fri, 05 Jan 2024 18:05:08 +0000
Large language models (LLMs), one type of a broad category often called "artificial intelligence," collect a large amount of data and use probabilistic and statistical methods to process language that they haven't seen before and respond to it. In the last few years, LLMs have received a lot of attention because they seem to interpret and use language in a human-like way, at first glance. However, if you look closely, these programs have some significant differences compared to the human use of language.
How LLMs learn language
Humans learn language through live interaction with other humans, in context. When infants as young as 10 months old are learning language, they pay attention to where the adults who are speaking or signing are looking, and use that information to help them figure out what words mean (Brooks and Meltzoff 2005). In contrast, LLMs typically learn language through the statistical analysis of large collections of text, called [pb_glossary id="1442"]corpora[/pb_glossary]. Text corpora have several important differences compared to live interactive speech and sign. For example, text typically indicates where words and sentences begin and end with spaces and periods, unlike speech. Many recent LLM programs are trained on very large corpora, such as large portions of the Internet. The Internet is several orders of magnitude larger than the total amount of language a human hears in their lifetime, let alone how much language input a child receives in the first four years of life, when they acquire the majority of their grammar. We can tell that LLMs do not learn language in the same way as humans because of the differences in the size and type of input. But perhaps the biggest difference in the ways LLMs use language is that they do not have the shared context and real-world knowledge that we have. The language they produce can be very convincing, but there is evidence that LLMs do not understand the language they are producing, if you know where to look. Consider exchange 1, in which ChatGPT tells you that someone wearing a blindfold is not easy to see. Even a young child can tell you that this is false. The blindfold obstructs Mary from seeing, not from being seen.
Exchange 1 Prompt: Mary and Sue are talking and Mary is blindfolded. Is Mary easy to see? Response: Mary being blindfolded implies that she cannot see, regardless of how easy or difficult she might be to see physically. The blindfold prevents her from seeing anything. So, from a literal perspective, Mary is not easy to see because she is blindfolded.
ChatGPT 3.5, May 24, 2024 Prompt by Hadas Kotek

When humans communicate, we try to recreate our interlocutor's perspective and anticipate their train of thought. We try to determine why they are communicating in the way they chose, and respond sometimes to the purpose of the communication rather than the literal meaning. An LLM does not have access to any of the real-world context or empathy needed for this kind of communication. When we, as users, access information through an LLM, we lose the relationship aspect of information seeking. The information we get from an LLM becomes separated from its source, which makes it more difficult to evaluate its reliability. This problem is amplified to the level of absurdity in situations where LLMs reproduce satirical sources as fact. For example, Google AI overview has suggested eating a rock a day based on an article by satirical news source The Onion and that adding non-toxic glue to pizza can keep the cheese from sliding off based on a Reddit comment which was presumably meant as a joke (Bender 2024).
How LLMs process and produce language
As we already saw in Section 3.3, there is evidence that children produce language using a rule-based approach. When humans use language, they usually understand it and think about its meaning. LLM programs, on the other hand, focus on key words and use statistical analysis. The way they produce language can be called a stochastic parrot, that is, "a system for haphazardly stitching together sequences of linguistic forms it has observed in its vast training data, according to probabilistic information about how they combine, but without any reference to meaning" (Bender et al. 2021: 617). In other words, it considers what words are in the input it receives and then calculates what words are likely to appear as an output. The probabilistic model of language used by LLMs results in some key differences compared to human language. First of all, LLMs are not good at extending patterns to novel forms. For example, in exchange 2, ChatGPT fails a [pb_glossary id="3130"]wug test[/pb_glossary] assessing how to pronounce the plural of the novel word mee. Since mee ends in a vowel, the plural form should be pronounced with a /z/ sound, rhyming with freeze not lease. Recall that even young children are able to correctly predict the plural form of unfamiliar words.
Exchange 2 Prompt: If ”mee” is the name of a new type of toy, and I have two of them, I have two mees. Does the plural word ”mees” rhyme with ”lease” or ”freeze”? Response: The plural word "mees" would typically rhyme with "lease." It follows the standard English pluralization rule where adding an "s" to the end of a noun typically results in a pronunciation that rhymes with "lease." So, "mees" would rhyme with "lees" rather than "freeze."
ChatGPT 3.5, May 24, 2024 Prompt by Nathan Sanders

Additionally, LLMs are not able to distinguish between rare and ungrammatical constructions, unlike human children (Nathan Sanders, p.c.). In Section 3.4, we learned that, in human language learning, we are able to use a relatively small amount of data to learn language, even though language is infinite, which is called the [pb_glossary id="179"]poverty of the stimulus[/pb_glossary]. What this means is that we can determine whether a sentence is ungrammatical even if we've never encountered the sentence before. LLMs, on the other hand, are not good at identifying ungrammaticality. The trouble LLMs have in identifying ungrammatical utterances is illustrated through exchange 3. The prompt asks ChatGPT to evaluate two wh-questions, shown in (1) and (2) below.

(1) What will you sleep after the cat eats some?

(2) You will fall asleep after the cat eats some what?

Wh-questions are normally formed by moving a wh-word such as what to the beginning of a sentence. However, there are constructions in which wh-movement is blocked, called wh-islands, including the [pb_glossary id="166"]embedded clause[/pb_glossary] after the cat eats some tuna. This means that the wh-word has to stay in place, as in sentence (2). It is ungrammatical to move it, as in sentence (1). However, ChatGPT fails to recognize that sentence (1) is ungrammatical.
Exchange 3 Prompt: I said to Alice: ”I will sleep after the cat eats some tuna”. However, Alice didn’t hear the last word I said. Which of the following would be valid ways for Alice to respond?

(1) What will you sleep after the cat eats some?

(2) You will fall asleep after the cat eats some what?

Response: Both responses are valid ways for Alice to seek clarification: (1) "What will you sleep after the cat eats some?" - Alice is asking for clarification on what you'll sleep after the cat eats, as she didn't hear the last word. (2) "You will sleep after the cat eats some what?" - Alice is asking for clarification on what specifically the cat will eat before you sleep. In both cases, Alice is attempting to fill in the missing information from what she heard you say.
ChatGPT 3.5, May 24, 2024 Prompt by Nathan Sanders

Note the interpretation that ChatGPT gives for sentence (1): "Alice is asking for clarification on what you'll sleep after the cat eats." This interpretation would arise if the wh-word was moved from the object position of the main clause, rather than from the embedded clause. This movement is not blocked by being a wh-island, however, it is still ungrammatical because the main clause verb sleeps is [pb_glossary id="617"]intransitive[/pb_glossary] and does not take an object. I followed up with ChatGPT to see if I could get it to also provide the interpretation where the wh-word comes from the embedded clause, describing what the cat has eaten, which would violate the wh-island. I did this by asking if sentence (1) was ambiguous. ChatGPT provided two interpretations. The first interpretation provided is that the speaker is asking what activity or event you will engage in after the cat eats some food. This is similar to the interpretation provided for sentence (1) in exchange 3, although it removes the reference to sleeping. The second interpretation provided in exchange 4, on the other hand, is that the speaker is asking what the cat will eat. This second interpretation has the wh-word moving from object position of eat inside the embedded clause. In other words, interpretation 2 violates the wh-island.
Exchange 4 Prompt: Is the sentence "What will you sleep after the cat eats some?" ambiguous? Response: Yes, the sentence "What will you sleep after the cat eats some?" is ambiguous because it could be interpreted in two different ways:

(1) "What will you sleep after the cat eats some?" - This interpretation suggests that the speaker is asking what activity or event you will engage in after the cat eats some food. (2) "What will you sleep after the cat eats some?" - This interpretation suggests that the speaker is asking what specific thing the cat will eat before you sleep. The ambiguity arises from the lack of clarity about whether "some" refers to an action (e.g., eating) or a noun (e.g., food).
ChatGPT 3.5, May 24, 2024

Remember, LLMs are trained on [pb_glossary id="1442"]corpus[/pb_glossary] data, which means they do not encounter [pb_glossary id="1022"]negative evidence[/pb_glossary]. This means that LLM training data would not include sentences like (1) and so it is not surprising that LLMs don't know how to handle them. However, children also do not have access to negative evidence, and children are able to distinguish between rare and ungrammatical utterances. When problems like these become known, the LLM developers often work on adjusting the design of the LLM, so it is possible that these issues may no longer arise in future iterations of ChatGPT. Even so, LLMs need specialized programming to address such issues, whereas children figure it out on their own.

Some dangers of LLMs: Hallucinations and bias

LLMs are programmed to create plausible-sounding text, but not necessarily to be accurate, truthful, or even helpful. It is not uncommon for LLMs to confidently assert false information or to make things up. When an LLM makes up false information, it is called a hallucination. Sometimes LLM hallucinations are partially true, which makes them harder to spot. For example, an LLM might create a reference list using real authors' and journals' names, but the articles themselves are not real. However, because LLMs create output that sounds like a human, it is easy sometimes to forget that it was not created by a human and should not be trusted like a human. Furthermore, the LLM will not indicate its sources, which makes it even trickier to fact-check. We don't always forget that LLMs are not human. Sometimes we remember that we are interacting with a computer, and so we treat their output as objective and universal. But this is also not true. LLMs are each trained on a particular corpus, and if the corpus has any biases, the LLM will reproduce that bias. In fact, because LLMs are statistical, it may even amplify the bias. If you play with an LLM long enough, it is easy to find these biases. For example, they may assume people's gender based on their occupation and text-to-image programs will often default to white people. At the time of this writing, nearly all LLMs are trained primarily on English data. This is another major sources of bias. LLMs are very interesting tools with lots of potential applications—but if we are to use them responsibly, we need to keep their hallucinations and their biases in mind.

Check yourself!
[h5p id="24"]
References and further resources

For a general audience

🔍 Bender, Emily. 28 May 2024. Information is relational. Mystery AI Hype Theater 3000: The Newsletter. https://buttondown.email/maiht3k/archive/information-is-relational/.

🔍 Bender, Emily. 14 June 2022. Human-like programs abuse our empathy -- even Google engineers aren't immnue. The Guardian. https://www.theguardian.com/commentisfree/2022/jun/14/human-like-programs-abuse-our-empathy-even-google-engineers-arent-immune

Bender, Emily and Chirag Shah. 13 Dec 2022. All-knowing machines are a fantasy. Institute of Art and Ideas News. https://iai.tv/articles/all-knowing-machines-are-a-fantasy-auid-2334

🔍 Biever, Celeste. 25 July 2023. ChatGPT broke the Turing test -- the race is on for new ways to assess AI. Nature 619: 686-689. https://www.nature.com/articles/d41586-023-02361-7

Chomsky, Noam, Ian Roberts, and Jeffrey Watumull. 8 Mar 2023. Noam Chomsky: The false promise of ChatGPT. The New York Times. https://www.nytimes.com/2023/03/08/opinion/noam-chomsky-chatgpt-ai.html

🔍 Dede, Chris. 6 Aug 2023. What is Academic Integrity in the era of generative Artificial Intelligence? Silver Lining for Learning. https://silverliningforlearning.org/what-is-academic-integrity-in-the-era-of-generative-artificial-intelligence

🔍 Epstein, Robert. 18 May 2016. The empty brain. Aeon. https://aeon.co/essays/your-brain-does-not-process-information-and-it-is-not-a-computer

Kotek, Hadas. 6 Nov 2023. Text-to-image models are shallow in more ways than one. Personal blog. https://hkotek.com/blog/text-to-image-models-shallow-parsing

🔍📚 Millière, Raphaël and Charles Rathkopf. 23 Nov 2023. Why it's important to remember that AI isn't human. Vox. https://www.vox.com/future-perfect/23971093/artificial-intelligence-chatgpt-language-mind-understanding

🔍 Nkonde, Mutale. 22 Feb 2023. ChatGPT: New technology, same old misogynour. Ms. Magazine. https://msmagazine.com/2023/02/22/chatgpt-technology-black-women-history-fact-check

🔍 O'Brien, Matt. 1 Aug 2023. Chatbots sometimes make things up. Is AI's hallucination problem fixable? Associated Press. https://apnews.com/article/artificial-intelligence-hallucination-chatbots-chatgpt-falsehoods-ac4672c5b06e6f91050aa46ee731bcf4

O'Neil, Cathy. 2016. Weapons of math destruction: How big data increases inequality and threatens democracy. Crown Books.

Academic sources

Bender, Emily, Timnit Gebru, Angelina McMillan-Major, and Shmargaret Smitchell. 2021. On the dangers of stochastic parrots: Can language models be too big? Proceedings of the 2021 ACM conference on fairness, accountability, and transparency. 610–623.

Bender, Emily, and Alexander Koller. 2020. Climbing towards NLU: On meaning, form, and understanding in the age of data. Proceedings of the 58th annual meeting of the association for computational linguistics. 5185–5198.

Brooks, Rechele and Andrew Meltzoff. 2005. The development of gaze following and its relation to language. Developmental Science 8 (6): 535–543.

Dingemanse, Mark. 2024. Generative AI and research integrity. Manuscript, Radboud University Nihmegen. https://osf.io/preprints/osf/2c48n

Hicks, Michael Townsen, James Humphries, and Joe Slater. 2024. ChatGPT is bullshit. Ethics and Information Technology 26 : 38. https://link.springer.com/article/10.1007/s10676-024-09775-5

Kotek, Hadas, Rikker Dockum, and David Sun. 2023. Gender bias and stereotypes in Large Language Models. Proceedings of The ACM Collective Intelligence Conference. 12–24.

4.1. Language modality

juliedoner — Mon, 08 Jan 2024 21:06:04 +0000
In linguistics, language [pb_glossary id="1502"]modality[/pb_glossary] is the medium or format through which language is conveyed. Human language shows up in at least four different modalities:

Spoken language, also known as vocal-auditory language, is produced by making sounds with the vocal tract and is perceived through hearing.

Signed language, also known as manual-visual language, is produced by making movements with the hands, face, and body, and is perceived through seeing.

Tactile language, also known as manual-tactile, is produced by making movements with the hands, face, and body, and is perceived through touch.

Graphemic language is produced by making markings such as symbols on some material and is perceived through seeing.

Most often, in natural settings, language is multi-modal, meaning we use multiple modalities simultaneously to communicate (Henner and Robinson 2023). For example, someone communicating through spoken language will also frequently use body posture and hand gestures at the same time. Our model of language needs to take into account all of the attested modalities of human language. Studying different modalities can give us different clues about which aspects of language are a result of the physical constraints of a particular modality, and which are a result of our cognitive dispositions. Historically, the field of linguistics has overlooked signed and tactile languages and focused primarily on spoken language.

A terminological note on the words signed languages and deaf

Signed languages are sometimes called sign languages. Both terms are generally acceptable, so you may encounter either one in linguistics writing. Sign languages has long been the more common term, but signed languages has recently been gaining popularity among deaf scholars. Another piece of relevant terminology that is in flux is the long-standing distinction in capitalization between uppercase Deaf (a sociocultural identity) and lowercase deaf (a physiological status). However, this distinction has been argued to contribute to elitist gatekeeping within deaf communities, so many deaf people have pushed to eliminate this distinction (Kusters et al. 2017, Pudans-Smith et al. 2019). In this textbook, we follow these prevailing modern trends by using signed languages and by not using the Deaf/deaf distinction. However, the alternatives are still widespread in linguistics writing, so you may still encounter them. For these issues, it is important to proceed with caution and follow the lead of anyone more knowledgeable than you, especially if they are deaf. If you are uncertain what usage is appropriate in a given situation with a given deaf person, ask what they prefer.

Why writing is different
Spoken, signed, and tactile languages are all primary modalities, but writing, a form of graphemic language, is a bit different. Unlike the other three modalities, writing is secondary to another modality, because it is a written representation of a language that occurs primarily in another modality, usually spoken language. For example, English writing is a representation of spoken English. Individual letters in written English represent different sounds in spoken language. Here are some ways that writing is different:

Other modalities of language are transient. That is, they disappear as soon as they are produced. Writing is the only language modality that can last independently of the language user.[footnote] Since the invention of audio and video recording devices, we can also keep records of spoken, signed, and tactile language. However, these are all recordings of a language user producing the language—they do not separate the language from the language user in the same way as writing. [/footnote] Because of this, writing is often not as interactive as other modalities of language, which, in turn, affects discourse structure and other properties.

Other modalities of language are acquired subconsciously by instinct, while writing must be consciously learned.

Unlike spoken and signed languages, writing has only been independently invented a handful of times throughout human history. Many languages do not have a written form at all. Most of the languages that do have a written form borrowed at least the idea from another language community. In contrast, spoken languages occur in every known human community with hearing populations, and deaf children who are not exposed to signed languages will develop a homesign language with their families.

Writing has different genre conventions than spoken language. Writing tends to be more formal and language change in writing lags behind change in spoken language.

Because of these differences, linguists do not treat written language the same. Because it is not subconsciously acquired, there is no reason to think it would be part of [pb_glossary id="180"]Universal Grammar[/pb_glossary]. For these reasons, it is often intentionally excluded from our model of language. However, we still sometimes use written language as a clue to patterns in spoken language, especially for older varieties of language for which only a written record remains. We have to be careful, however, that we do not dismiss all graphemic forms of language as secondary to spoken language. There are many forms of graphemic communication that are not directly linked to a spoken language, at least some of which display some complex properties (see Henner and Robinson 2023).
Dr. Melanie McKay-Cody, a deaf Cherokee professor, demonstrates that petroglyphs—stone carvings—from the Northern Ute in Utah and the Kiowa in Texas incorporate signs from North American Indian Sign Language (NAISL). She argues that these petroglyphs are a form of written signed language. Although a few different transcription methods for signed language have been proposed in modern times, none have really caught on. But perhaps there has been signed language writing for hundreds, if not thousands, of years!

Ethics and modality
As previously mentioned, the field of linguistics has primarily focused on spoken languages, which Henner and Robinson (2023) call modality chauvinism. This focus on spoken modality has its roots in ableism, since other modalities are often used by people with disabilities. The solution seems simple—include more signed languages in our studies! But even well-intentioned linguists will face some difficulties incorporating signed languages in their research. For example, the lack of a writing system for signed languages makes them more difficult to write about, contributing to their marginalization in research (Henner and Robinson 2023). Secondly, signed languages are best portrayed by video, and so signed language data cannot be anonymized and require different copyright permissions. Most importantly, however, signed and tactile modalities are used by communities which are marginalized both within linguistics and in wider society, and therefore require extra care from researchers. We will discuss more about how to conduct research with marginalized communities in Chapter 5. These extra difficulties do not excuse us from including signed and tactile languages in our research! But they must be taken into account when planning a research study.

Other modalities for other species?

Some animal species use different modalities for their communication systems. For example, ants are known to use scents to communicate. However, as far as we can tell, these communication systems do not seem to have all of the properties of language. In some science fiction settings, though, authors have also invented full language in other modalities. For example, in Becky Chambers' The Wayfarers series, a species called the Aeluon communicate through patches on their cheeks that change colours. In the movie Arrival, the alien species have a form of written language that does not seem to be secondary to spoken language. Many, many science fiction stories also have some sort of telepathic communication in their fictional worlds.

Check yourself!
[h5p id="26"]
References and further resources

Attribution
The call-out box 'A terminological note' is adapted from the following CC BY NC source:
↪️ Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Section 3.1: Modality. Essentials of Linguistics, 2nd edition. Pressbooks. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/chapter/3-1-modality/

Linguistics in the media

🎉 Chambers, Becky. 2014–2021. The Wayfarers series. Hodder & Stoughton. https://www.otherscribbles.com

Book 1: The long way to a small, angry planet

Book 2: A closed and common orbit

Book 3: Record of a spaceborn few

Book 4: The galaxy, and the ground within

🎉 Villeneuve, Denis (director). 2016. Arrival. FilmNation Entertainment, Lava Bear Films, and 21 Laps Entertainment.

For a general audience

🔍 Canadian Language Museum. n.d. Sign Languages of Canada 3D. Virtual museum exhibit. https://languagemuseum.ca/online-exhibits/sign-languages-of-canada-3d

🔍 📚 Heumann, Judy and Melanie McKay-Cody. 2023. Indigenous sign languages with Dr. Melanie McKay-Cody. The Heumann Perspective. https://judithheumann.com/melaniemckaycody/

Edwards, Terra. 2024. Going tactile: Life at the limits of language. Oxford: Oxford University Press.

🔍 Leland, Andrew. 12 May 2022. DeafBlind communities may be creating a new language of touch. The New Yorker. https://www.newyorker.com/culture/annals-of-inquiry/deafblind-communities-may-be-creating-a-new-language-of-touch

🔍 Quartz. 2016. Pro-tactile ASL: A new language for the DeafBlind. YouTube video. https://www.youtube.com/watch?v=9GrK3P15TYU

For students

🧠 Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Section 3.1: Modality. Essentials of Linguistics, 2nd edition. Pressbooks. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/chapter/3-1-modality

Academic sources

🔍📚 Henner, Jon, and Octavian Robinson. 2023. Unsettling languages, unruly bodyminds: A Crip Linguistics manifesto. Journal of Critical Study of Communication and Disability 1(1): 7–37. https://criticalstudycommunicationdisability.org/index.php/jcscd/article/view/4

Kusters, Annelies, Maartje De Meulder, and Dai O’Brien. 2017. Innovations in Deaf Studies: Critically mapping the field. In Innovations in Deaf Studies: The role of deaf scholars, ed. Annelies Kusters, Maartje De Meulder, and Dai O’Brien, Perspectives on Deafness, 1–56. Oxford: Oxford University Press.

McKay-Cody, Melanie. 2019. Memory comes before knowledge—North American Indigenous Deaf: Socio-cultural study of rock/picture writing, community, sign languages, and kinship. PhD dissertation, University of Oklahoma. https://shareok.org/handle/11244/319767

Pudans-Smith, Kimberly K., Katrina R. Cue, Ju-Lee A Wolsley, and M. Diane Clark. 2019. To Deaf or not to deaf: That is the question. Psychology 10(15): 2091–2114.

4.3. Language, dialect, and grammar

juliedoner — Mon, 08 Jan 2024 21:06:35 +0000

One language, many grammars
When we are collecting linguistic data to study, we are usually collecting data from a particular language. But it turns out that the notion of language is very messy. The notion of language seems straightforward because we are used to thinking and speaking of "the English language," "the French language," "the Swahili language," and so forth. But these terms are actually much vaguer than they seem at first glance because they cover a large number of varieties, including ones that differ enough to be mutually unintelligible. For instance, Ethnologue distinguishes 32 dialects of English in the United Kingdom alone. In addition, distinct dialects of English are spoken in former British colonies, including Australia, Canada, India, New Zealand, South Africa, the United States, and many other African, Asian, and Caribbean nations, and many of these dialects have subdialects of their own. Similarly, Ethnologue distinguishes 11 dialects of French in France and 10 dialects of Swahili in Kenya, and there are further dialects in other countries in which these languages are spoken. Moreover, we use terms like "the English language" to refer to historical varieties, which may differ profoundly from modern-day varieties. For example, present-day English differs from Old English about as much as modern English differs from German. Although the most salient differences between dialects are often phonological (that is, speakers of different dialects often have different accents), dialects of a so-called single language can differ syntactically as well. For instance, in standard French, as in the Romance languages more generally, adjectives ordinarily follow the noun that they modify. But that order is reversed in Walloon, a variety of French spoken in Belgium. The two parametric options are illustrated in (1).

(1) a. Standard French

un chapeau noir

a hat black

'a black hat'

(1) b. Walloon

on neûr tchapê

a black hat

'a black hat'

(Bernstein 1993: 25-26)
In this example, we have two different dialects of "the same language", French, differing with respect to a language parameter. The opposite is also possible: two "different languages" that are basically the same. For example, the same linguistic variety spoken on the Dutch-German border may count as a dialect of Dutch or German depending on which side of the political border it is spoken, and the same is true of many other border dialects as well. This is called a dialect continuum, where each adjacent language variety is quite similar, but the extreme ends of the continuum are quite different from each other. According to Max Weinreich, "a language is a dialect with an army and a navy." A striking (and sad) confirmation of this aphorism concerns the recent terminological history of Serbo-Croatian. As long as Yugoslavia was a federal state, Serbo-Croatian was considered a single language with a number of regional dialects. The 14th edition of Ethnologue, published in 2000, still has a single entry for Serbo-Croatian. In the 15th edition, published in 2005, the single entry is replaced by three new entries for Bosnian, Croatian, and Serbian. As the previous discussion has shown, the notion of language is based more on sociopolitical considerations than on strictly linguistic ones. By contrast, the term 'grammar' refers to a particular set of parametric options that a speaker acquires. The same language label can be associated with more than one grammar (the label "French" is associated with grammars that have adjective-noun order and ones with noun-adjective order), and a single grammar can be associated with more than one language label (as in the case of border dialects). In other words, one particular [pb_glossary id="1258"]E-language[/pb_glossary] might be a mishmash of many different [pb_glossary id="1257"]I-languages[/pb_glossary], as discussed in Section 3.2.
Mutual intelligibility
It is important to distinguish the concept of shared grammar from [pb_glossary id="1511"]mutual intelligibility[/pb_glossary]. To a large extent, standard English and many of its nonstandard varieties are mutually intelligible even where their grammars differ with respect to one parameter or another. On the other hand, it is perfectly possible for two or more varieties that are mutually unintelligible to share a single grammar. For instance, in the Indian village of Kupwar (Gumperz and Wilson 1971), the three languages Marathi, Urdu, and Kannada, each spoken by a different ethnic group, have been in contact for about 400 years, and most of the men in the village are bi- or trilingual. Like the standard varieties of these languages, their Kupwar varieties have distinct vocabularies, thus rendering them mutually unintelligible to monolingual speakers, but in Kupwar, the considerable grammatical differences that exists among the languages as spoken in other parts of India have been virtually eliminated. The difference between standard French and Walloon with respect to prenominal adjectives is another instance of this same convergence phenomenon. Here, too, the adjective-noun order in Walloon is due to language contact and bilingualism, in this case between French and Flemish, the other language spoken in Belgium; in Flemish, as in the Germanic languages more generally, adjectives ordinarily precede the nouns that they modify.
One speaker, many grammars
Finally, we should point out that it is perfectly possible for a single speaker to acquire more than one grammar. This is most strikingly evident in balanced bilinguals. Speakers can also acquire more than one grammar in situations of syntactic change. For instance, in the course of its history, English changed from an object-verb order language to a verb-object order language, and individual speakers during the transition period (which began in late Old English and continued into Middle English) acquired and used both parametric options. Speakers can also acquire more than one grammar in situations of diglossia or stable syntactic variation. For instance, English speakers who speak a non-prestigious dialect of English at home might acquire the prestigious parametric variants at school, and switch between them depending on context.
Check yourself!
[h5p id="27"]
References and further resources

Attribution
This section is adapted from the following CC BY NC source:
↪️ Santorini, Beatrice, and Anthony Kroch. 2007. The syntax of natural language: An online introduction. https://www.ling.upenn.edu/~beatrice/syntax-textbook

Academic sources

Bernstein, Judy. 1993. Topics in the syntax of nominal structure across Romance. Ph.D. dissertation, City University of New York.

📑 Eberhard, David M., Gary F. Simons, and Charles D. Fennig (eds.). 2023. Ethnologue: Languages of the World. Twenty-sixth edition. Dallas, Texas: SIL International. http://www.ethnologue.com.uml.idm.oclc.org

Gumperz, John J., and Robert Wilson. 1971. Convergence and creolization: A case from the Indo-Aryan/Dravidian border in India. In Dell Hymes, ed., Pidginization and creolization of languages. Cambridge: Cambridge University Press. 151–167.

5.1. Language and community

juliedoner — Mon, 08 Jan 2024 21:07:40 +0000
As social scientists, we have a responsibility to think critically about where our data comes from—not only so that we can ensure it is reliable and that the collection methods fit our research objectives, but also to ensure that our data has been collected in an ethical manner. We are studying human language, which means that we are collecting our data from people and from communities, and we need to consider the impact of our research on these people and communities. In this chapter, we will look at some of the ethical issues involving using and collecting data, especially data from Indigenous languages, within the context of colonialism. In the past, linguistics as a field has tended to separate the data we collect from the communities that use them, to treat the data as abstract ideas that belong to no one, or perhaps to the linguist who records them. However, such an approach can cause harm; the languages we study often have emotional, cultural, historical, and spiritual significance to the communities using them. As Christine Mallinson writes, "language and society are inseparable, language is a cultural artifact that belongs to those people and communities who use it, and addressing power dynamics and inequalities surrounding language is central to social justice, equity, and inclusion" (Charity Hudley, Mallinson & Bucholtz 2024b: xiv). Likewise, we, as researchers, bring our own history, culture, and context to bear as we do research. As Clemons and Lawrence (2020: e259) state, "research subjectivities affect the kinds of questions that get asked as well as the interpretation of findings."
Emotionless, passionless, abstract, intellectual, academic research is a goddamn lie, it does not exist. It is a lie to ourselves and a lie to other people. Humans—feeling, living, breathing, thinking humans—do research. When we try to cut ourselves off at the neck and pretend an objectivity that does not exist in the human world, we become dangerous, to ourselves first, and then to the people around us.
Eber Hampton (1995: 52, as cited in Wilson 2008: 100-101)

As such, I will begin this chapter by introducing myself and my position with regards to linguistic research and Indigenous communities.

Positionality statement
I, the author of this textbook, am a white settler person of mixed Germanic and Celtic descent. Most of my ancestors settled in Canada between 5 and 8 generations ago in the 1700s and 1800s. Most bought, were given, or rented land to farm. Some were the first white settlers in their area, such as David and Mary Thompson on my paternal grandmother's side, who came from Scotland and were the first to settle in Scarborough, Ontario. My ancestors settled on land covered by many different treaties in Ontario, which were signed on their behalf, including the Williams treaties, the Toronto Purchase, the Haldimand Tract and Simcoe Patent, the Huron Tract, and the Robinson-Huron Treaty. There are many ethical issues and disputes regarding these treaties. As I write this section, I cannot separate myself from my family's history. What I can do, however, is move forward in a posture of learning, one that acknowledges this history and my place in it and that seeks to move forward in a better way. This learning journey will continue for the remainder of my life. This section represents some of the things I have learned so far, from my perspective as a settler linguist who would like to honour the people who speak the languages I study as best as I can. As I wrote this section, I tried to base it on the words and shared experiences of Indigenous scholars and community members. However, the very act of collecting and summarizing these views colours them with my perspective. If at all possible, you should learn about the experiences of Indigenous people directly from them. Check out the sources marked with a ➡️ in the references sections throughout this chapter for some suggestions.

What is an Indigenous language?
In brief, an Indigenous language is a language used by or associated with an Indigenous people group. Often, Indigenous languages have historical, cultural, and/or spiritural significance for the Indigenous group(s) that use them. In its simplest meaning, indigenous is simply an adjective describing originating in a particular place. Under this definition, all languages are indigenous to somewhere, since all languages arose in a particular community in a particular place. More often, though, the term indigenous is used in a more specific manner. However, as there are more than 5,000 distinct Indigenous groups in the world (according to the United Nations), it is difficult to provide a precise definition. Instead of providing a definition of indigeneity, the United Nations Declaration on the Rights of Indigenous Peoples states that Indigenous peoples have a right to self-identify as Indigenous. This can lead to tricky situations, though, when groups with dubious claims to Indigeneity demand political Indigenous status. However, some commonalities between Indigenous groups can be identified: continued occupation of a land since before colonization, distinct cultures and political systems, and the desire to maintain institutions distinct from the majority population.
Indigenous peoples have in common a historical continuity with a given region prior to colonization and a strong link to their lands. They maintain, at least in part, distinct social, economic and political systems. They have distinct languages, cultures, beliefs and knowledge systems. They are determined to maintain and develop their identity and distinct institutions and they form a non-dominant sector of society.
-United Nations

Indigenous peoples (and their languages) are found in every region of the world. The majority of the world's languages are used by Indigenous peoples.
Indigenous peoples challenge! How many Indigenous people groups are you aware of? Try to name as many Indigenous people groups as you can from memory, such as the following:

The Indigenous people groups of where you currently live

The Indigenous people groups of the places you grew up

The Indigenous people groups of where your ancestors are from

An Indigenous people group from every continent of the world

An Indigenous people group for every letter of the alphabet

An Indigenous people group from every country of the world

Consider reading about the Indigenous people groups from regions that are especially relevant to you! Many of you will be surprised by what you find.

Key takeaways

Language is inseparable from the communities that use them.

As linguists, we have an ethical obligation to ensure that the way we collect, store, and use language data does no harm to the communities who use the languages our data is from.

It is impossible to stay completely objective; everyone has conscious and subconscious biases.

Indigenous people groups typically have the following characteristics: they have occupied their land since before colonization, are a minority population, and desire to maintain distinct cultural and political institutions.

Check yourself!
[h5p id="41"]
References and further resources
The references in this section that have at least one Indigenous author are indicated with an ➡️.
For a general audience

🔍 Oxford, Will. 2019. Indigenous Languages in Canada. Canadian Language Museum. https://languagemuseum.ca/wp-content/uploads/2022/12/indigenous_languages_in_canada_0.pdf

United Nations. No date. Indigenous peoples. Accessed May 16, 2024. https://www.un.org/en/fight-racism/vulnerable-groups/indigenous-peoples

United Nations. 2007. United Nations declaration on the rights of Indigenous peoples. https://social.desa.un.org/issues/indigenous-peoples/united-nations-declaration-on-the-rights-of-indigenous-peoples

Academic sources

Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz, eds. 2024a. Decolonizing Linguistics. Oxford: Oxford UP. https://global.oup.com/academic/product/decolonizing-linguistics-9780197755259?cc=ca&lang=en&#

Charity Hudley, Anne, Christine Mallinson, and Mary Bucholtz, eds. 2024b. Inclusion in Linguistics. Oxford: Oxford UP. https://global.oup.com/academic/product/inclusion-in-linguistics-9780197755303?cc=ca&lang=en&

Clemons, Aris, and Anna Lawrence. 2020. Beyond position statements on race: Fostering an ethos of antiracist scholarship in linguistic research (Response to Charity Hudley et al.). Language 96 (4): e254-e267.

➡️ 🔍 Wilson, Shawn. 2008. Research is ceremony: Indigenous research methods. Halifax: Fernwood Publishing.
➡️ 📑 Younging, Gregory. 2018. Elements of Indigenous Style. Brush Education Inc.
For instructors

Singerman, Adam. 2025. Talking about language endangerment and Indigenous languages in the classroom: Some dos and don'ts I have learned through fieldwork in the Brazilian Amazon. In Lauren Clemens, Vera Gribanova, and Gregory Scontras (ed). Syntax in uncharted territories: Essays in honor of Maria Polinsky. Open Access Publications from the University of California. 525-545. https://escholarship.org/uc/item/3081x700#main

4.6. Becoming a linguist: Discussing data

juliedoner — Mon, 08 Jan 2024 21:07:56 +0000
In this section, we are going to discuss how to write about linguistic data. In morphology, our data is often words or lists of words. In syntax, our data is often phrases and sentences. Two technical terms will be useful in this section. The [pb_glossary id="1569"]object language[/pb_glossary] is the language under discussion, or the language being analyzed. The [pb_glossary id="1570"] metalanguage[/pb_glossary] is the language that we are using in order to discuss the object language. In this section, I am using examples from Gonzalez (2023), which is a paper written in English about questions in Finnish and Turkish. The object languages of this paper are Finnish and Turkish, while the metalanguage is English.
Numbering data
When we include linguistic data in our papers, we usually separate it out from the main text or prose of our argument. Examples that are separated out this way are each numbered sequentially throughout the text. An example is shown below from the opening paragraph of Gonzalez (2023).
The strategy used for forming polar questions varies across languages. For instance, in languages like English, polar questions are formed using a raising intonation (marked as ↑) and subject-auxiliary inversion, as illustrated in (1). In contrast, in other languages, polar questions are formed using an interrogative particle. Japanese is an example of such a language. As shown in example (2), polar questions can be formed using the interrogative particle ka.

(1) Are you leaving ↑ ?

(2) Japanese (Uegaki 2018)

Hanako-ga hashitta-ka?

Hanako-NOM ran-KA

'Did Hanako ran?'

(Gonzalaz 2023: 2)

Related examples may share the same number, with sub-examples labelled with letters.

(4) a. Finnish

Lähti-kö Mari?

left-PolQP Mari

'Did Mary leave?'

(4) b. Turkish

Oya Dilara’yı mı öp-tü?

Oya Dilara-ACC PolQP kiss-PST

'Did Oya kiss Dilara?'

(Gonzalez 2023: 3)

Definitions, formulas, and syntax trees are also sometimes numbered alongside the examples. If you repeat an example several pages after you first discuss it, it is good to repeat the example so your reader does not need to flip back and forth. Typically, authors will give the repeated example a new number, but also tell you the original number, as shown below.
Consider again example (5), repeated below as (23).

Embedded examples
We also sometimes include examples embedded directly into a sentence, instead of setting them apart and numbering them. This is generally only done with really short examples, such as one-word examples. This format can also be used when discussing portions of a longer example. This should really only be done when the example does not need to be [pb_glossary id="1576"]glossed[/pb_glossary], either because the example is simple enough to not need it, or because the example is glossed elsewhere close by in the text. When we include examples embedded into a sentence, the word(s) in the object language are typically in italics. In fact, any time you mention a word (rather than use the word), it should be in italics. If the object language is different than the metalanguage, you should also include a translation of the word(s) in single quotes. In the following example, pu:ch ‘ask’ and ja:n ‘know’, both Hindi-Urdu words, are formatted in this way. On the other hand, kya: is not translated, because this sentence is part of a larger passage about the details of the meaning and use of kya:.
For instance, example (15) shows that kya: can be embedded under the rogative predicate pu:ch ‘ask’, but cannot be embedded under the responsive predicate ja:n ‘know’.

(15) a. Hindi-Urdu (Bhatt & Dayal 2020)

Ṭi:car=ne Anu=se pu:ch-a: [ki kya: vo ca:i piyegi:].

teacher=ERG Anu=from ask-PFV that KYA s/he tea drink.FUT.3FSG

'The teacher asked Anu whether she would drink tea.'

(5) b.

*Anu ja:n-ti: hai [ki kya: tum ca:i piyoge].

Anu.F know.HAB.F be.PRS.SG that KYA you tea drink.FUT.2MPL

(Int.) 'Any knows whether you will drink tea.'

(Gonzalez 2023: 8)

How to discuss data
Every time you include a linguistic example, you should introduce your example, present your example, and then discuss your example. Not all examples necessarily need discussion before and after. If it's a very simple example, it may be enough simply to explain it all before you present it. Use your judgment and think about what your reader needs to know to understand your example and how it fits into your argument, as well as when they need to know which information.
Introduce your example
Before you give an example, you should at least tell your reader that you're about to show them an example, referring to the example by number. In can be very disorienting if you're reading along and all of a sudden there's an example with no context. Even better, though, is if you tell your reader why you're about to show them an example and tell them what to look for. This is illustrated above with Gonzalez's introduction to her examples (1) and (2).
Present your example
After you've introduced the example, present the example. The example should be numbered, as discussed above, and [pb_glossary id="1576"]glossed[/pb_glossary], as we will discuss in Section 4.7. If the example is at all complex, guide your reader to which parts of the example they should be looking at. You can bold or underline the relevant parts. For example, Gonzalez bolded the question markers in her example (4), shown above. Ideally, you should explicitly state what you've bolded and underlined when you introduce the example. Instead of bolding or underlining, you can also describe in the text where your readers should be looking, before or after you present the example. Always clearly indicate the language that the data is from. If the entire paper is about the same language, you can do this just once, in the introduction. But if you're discussing multiple languages, every single example should be labelled with the language it comes from. Some authors do this in the prose before the example, but it's even clearer if it's incorporated into the offset example itself. Two common ways of doing so are shown below in examples (1)-(2). In (1), the name of the language is right-aligned next to the example. In (2), the name of the language appears above the example. Whichever format you choose, you should be consistent throughout your paper.

(1) This is an example. [English]

(2) English

This is another example.

Discuss your example
After you have presented your example, you should discuss it. Do not leave it to your reader to analyze the data -- that's your job as the author. Describe the pattern you see in the data and explain why the pattern matters for your argument. And be specific! Mention specific words or morphemes from your example and why they matter.
Attributing your data
Whenever you use data, you must attribute your source.

If you collected the data via introspection, this is the only time that you don't need to attribute your sources.

If you collected the data via elicitation, you should thank your speakers in your acknowledgments section. It is also a good idea to discuss your methodology at some point within your paper.

If your data comes from a corpus, the data collection methods should be explained at some point in your paper. If you use multiple corpora, each example should be marked with which corpus it came from.

If your data comes from another published source, you should provide an in-text citation next to or under the example that indicates its source. Since an example is a kind of direct quote, you should include the page number of the original source. If the source from which you got the example is not itself the original source, you should cite both the original source and the source you got it from, in the format "[original source], as cited in [source you got it from]".

Check yourself!
[h5p id="30"]
References and further resources

Sources for examples

Gonzalez, Aurore. 2023. Interrogative particles in polar questions: The view from Finnish and Turkish. Glossa 8(1): 1–47.

Chapter 4 practice exercises

juliedoner — Mon, 08 Jan 2024 21:08:11 +0000

Data analysis

Exercise 1. Glossing
Beginner [4.7] Imagine that you are conducting elicitations in Nigeria and you have just collected the following Yoruba data in (1).[footnote] The data in this exercise was provided by Tolani Akinlade. [/footnote] You have collected the pronoun paradigm and several sentences. You've also looked up all the words in the dictionary, which are listed in the table below. Gloss the following three sentences according to linguistic convention to prepare them for inclusion in a paper. Note 1: Bolu, Wale, and Shade are proper names. Note 2: Yoruba is a tonal language, but the tones are omitted in these examples.

(1) a. Yoruba

a ti jo

'We danced'

(1) b. a ti jo

'We danced'

(1) c. a ti jo

'We danced'

Table 1. Mini Yoruba dictionary

Yoruba data English translation

adiyẹ 'chicken'

ati 'and'

gbọ 'hear'

je 'eat'

jinan 'ready' (for food)

jo 'dance'

ko 'not'

n action in progress

ti past tense marker

Table 2. Yoruba pronoun paradigm

Subject pronouns Object pronouns

1sg mo mi

2sg iwọ e

3sg o ∅

1pl a wa

2pl eyin yin

3pl wọn wọn

Communication and study skills

Exercise 2. Writing about data
Intermediate [4.6] In Chapter 2, Exercise 2, you developed some data to demonstrate the behaviour of our and ours. Use the introduce, present, discuss format described in Section 4.6 to describe your data.
Research and application

Exercise 3. Source analysis
Intermediate [4.7, 4.8] Look up an academic article, a book chapter in an academic book, a dissertation, and/or a single-authored academic book in morphology or syntax. Look at how the examples are formatted and answer the following questions.
a. Find the abbreviations list. Where is it?

b. What abbreviations are common/uncommon?

c. Does the source use three or four line glossing?

d. What similarities or differences do you notice compared to the Leipzig conventions?

e. Do they restart the example numbering in every chapter/section?

f. Can you tell which examples they collected themselves vs. got from other sources? How can you tell?

g. How do they cite examples from other sources?

h. If there are examples from multiple languages, how do they indicate which examples are in which language? Is it effective or confusing?

i. Are syntax trees numbered like examples or treated like figures?

Language Journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 4. Language modality
Beginner [4.1] What is the modality of your language?
Exercise 5. Dialects
Intermediate [4.3] What are the different dialects of your language? Is your language mutually intelligible with another language?
Exercise 6. Glossing
Intermediate [4.7, 4.8] Find three sentences from your language and create properly formatted glosses for them.

4.4. What is grammaticality?

juliedoner — Mon, 08 Jan 2024 21:09:05 +0000
The aim of a generative grammar is to generate all and only the grammatical sentences of a language. Since the notion of grammaticality is basic to syntactic theory, it is important to distinguish it from notions with which it is easily confused. If you are collecting data on a language, and a speaker tells you that a sentence seems odd, they don't usually know why. It is up to the linguist to figure out why!
Grammaticality vs. making sense
First and foremost, 'is grammatical' is not the same thing as 'makes sense.' The sentences in (1) all 'make sense' in the sense that it is easy to interpret them. Nevertheless, as indicated by the asterisks, they are not grammatical.

(1) a. *Is our children learning?

b. *Me wants fabric.

c. *To where are we be taking thou, sir?

d. *The introduction explained that “the Genoese people, besides of hard worker, are good eater too, and even ‘gourmand,’ of that honest gourmandise which will not drive a man to hell but which is, after all, one of the few pleasures that mankind can enjoy in this often sorrowful world.”

Conversely, sentences can be grammatical, but not 'make sense.' Two examples are given in (2). Since the sentences are grammatical, they aren't preceded by an asterisk. Their semantic anomaly can be indicated, if desired, by a prefixed pound sign (hash mark).

(2) a. #Colorless green ideas sleep furiously. (Chomsky 1965: 149) – cf. Revolutionary new ideas appear infrequently.

b. #I plan to travel there last year. – cf. I plan to travel there next year.

Grammaticality vs. difficulty of processing
Second, 'grammatical' must be distinguished from 'easily processable by human beings.' This is because it turns out that certain well-motivated simple grammatical operations can be applied in ways that result in sentences that are virtually impossible for human beings to process. For instance, it is possible in English to modify a noun with a relative clause, and sentences containing nouns that are modified in this way, like those in (3), are ordinarily perfectly acceptable and easily understood. (Here and in the following examples, the relative clauses are bracketed and the modified noun is underlined.)

(3) a. The mouse [that the cat chased] escaped.

b. The cat [that the dog scared] jumped out the window.

But now notice what happens when we modify the noun within the relative clause in (3a) with a relative clause of its own.

(4) The mouse [that the cat [that the dog scared] chased] escaped.

Even though (4) differs from (3a) by only four additional words and a single additional level of embedding, the result is virtually uninterpretable without pencil and paper. The reason is not that relative clause modification can't apply more than once, since the variant of (3a) in (5), which contains exactly the same words and is exactly as long, is perfectly fine (or at any rate much more acceptable than (4)).

(5) The mouse escaped [that the cat chased] [that the dog scared].

The reason that (4) is virtually uninterpretable is also not that it contains recursive structure (the relative clause that modifies mouse contains the relative clause that modifies cat). After all, the structures in (6) are recursive, with up to 5 levels of embedding, yet they don't throw us for a loop the way that (4) does.

(6) a. She won.

b. The Times reported that [she won].

c. John told me that [the Times reported that [she won]].

d. I remember distinctly that [John told me that [the Times reported that [she won]]].

e. They don’t believe that [I remember distinctly that [John told me that [the Times reported that [she won]]]].

f. I suspect that [they don’t believe that [I remember distinctly that [John told me that [the Times reported that [she won]]]]].

Example (4) is unacceptable not because it is ungrammatical, but because of certain limitations on human short-term memory (Chomsky and Miller 1963: 286, Miller and Chomsky 1963: 471). Specifically, notice that in the (relatively) acceptable (5), the subject of the main clause the mouse doesn't have to "wait" (that is, be kept active in short-term memory) for its verb escaped since the verb is immediately adjacent to the subject. The same is true for the subjects and verbs of each of the relative clauses (the cat and chased, and the dog and scared). In (4), on the other hand, the mouse must be kept active in memory, waiting for its verb escaped, for the length of the entire sentence. What is even worse, however, is that the period during which the mouse is waiting for its verb escaped overlaps the period during which the cat must be kept active, waiting for its verb chased. What makes (4) so difficult, then, is not the mere fact of recursion, but that two relations of exactly the same sort (the subject-verb relation) must be kept active in memory at the same time. In none of the other relative clause sentences is such double activation necessary. For instance, in (3a), the mouse must be kept active for the length of the relative clause, but the subject of the relative clause (the cat) needn't be kept active since it immediately precedes its verb chased.
Sentences like (3) and (4) are often referred to as centre-embedding structures, and the dependencies between the subjects and their verbs are said to be nested.

The mouse that the cat chased escaped. | |_____| | |____________________________| The mouse that the cat that the dog scared chased escaped. | | |_____| | | | |_________________________| | |________________________________________________|

By contrast, the corresponding dependencies in (5) are not nested.

The mouse escaped that the cat chased. |________| |______| The mouse escaped that the cat chased that the dog scared. |________| |______| |_____|

Grammaticality across contexts
A final important point to bear in mind is that any sentence is an expression that is paired with a particular interpretation. Grammaticality is always determined with respect to a pairing of form and meaning. This means that a particular string can be grammatical under one interpretation, but not under another. For instance, (7) is ungrammatical under an subject-object-verb (SOV) interpretation (that is, when the sentence is interpreted as Sue hired Tom).

(7) Sue Tom hired.

Example (7) is grammatical, however, under an object-subject-verb (OSV) interpretation (that is, when it is interpreted as Tom hired Sue). On this interpretation, Sue receives a special intonation marking contrast, which would ordinarily be indicated in writing by setting off Sue from the rest of the sentence by a comma. In other words, the grammaticality of (7) depends on whether its interpretation is analogous to (8a) or (8b).

(8) a. *She him hired.

b. ✓Her, he hired. (The other job candidates, he didn’t even call back.)

Related to this idea is that context matters. For example, sentence (7), even though it has a grammatical interpretation, is a fairly odd way to word that sentence. However, in some contexts, it sounds more natural than in other contexts. Sometimes it is fairly easy to think of a context where a sentence is grammatical, but other times it can be quite tricky.
Check yourself!
[h5p id="28"]
References and further resources

Attribution
This section is adapted from the following CC BY NC source:
↪️ Santorini, Beatrice, and Anthony Kroch. 2007. The syntax of natural language: An online introduction. https://www.ling.upenn.edu/~beatrice/syntax-textbook

Comedy and satirical linguistics

🎉 Bakery, Yreka. 2006. New speech disorder linguists contracted discovered! Speculative Grammarian CLI(2). https://specgram.com/CLI.2/03.bakery.disorder.html

Academic sources

Chomsky, Noam. 1965. Aspects of the theory of syntax. Cambridge, MA: MIT Press.

Chomsky, Noam, and George A. Miller. 1963. Introduction to the formal analysis of natural languages. In R. Duncan Luce, Robert R. Bush, and Eugene Galanter, eds., Handbook of mathematical psychology, vol. 2, 269–321. New York: Wiley.

Miller, George A., and Noam Chomsky. 1963. Finitary models of language users. In R. Duncan Luce, Robert R. Bush, and Eugene Galanter, eds., Handbook of mathematical psychology, vol. 2, 419–491. New York: Wiley.

4.5. The validity of speaker judgments

juliedoner — Wed, 10 Jan 2024 03:19:24 +0000

Who should we ask for judgments?
Common wisdom among linguists is that we should only be asking [pb_glossary id="1534"]native speakers[/pb_glossary] for linguistic judgments. We are trying to get at the subconscious knowledge about language that is acquired by instinct during early childhood. If a speaker learned the language later in life, then their language may have subtle differences compared to a native speaker, because their judgments may be influenced by their first language. But is this common wisdom all there is to say about it? First of all, it's important to keep in mind that everyone’s judgments can give us information about language. A multilingual language user can give us information about how multilingual brains work and how different grammars interact with each other. A neurodivergent language user or a language user with a language disorder can give us information about variation in language processing. However, if we are trying to develop a grammar of a particular language, neurotypical monolinguals can give us data that has the fewest confounding factors. But it is not always as straightforward as it sounds to find a native speaker with no confounding factors. If you are studying a language using bilingual elicitation, you will need to find someone who has advanced proficiency in both languages, and there will be influence between the two languages. Even monolinguals almost always have command of more than one dialect or variety, and so there is always a chance of influence between multiple grammars. In contrast, some native speakers do not have sufficient command of their native language to act as a language informant. For example, someone who immigrates in childhood may become dominant in their second language and may even lose their native language completely if they stop having opportunities to use it. There are also some situations where the idea of a native speaker itself is problematic. For example, approximately 90% of deaf children are born to hearing parents and many of these do not learn to sign until they go to school, and thus arguably do not have a native language at all. When studying a signed language, is it better to find one of the rare "native" speakers, or is it better to work with a typical signer, one who learned the signed language a little bit later in life? Racism also plays a role in who is considered a "native speaker." It is not uncommon for “native speaker status” to be denied to those who speak less prestigious varieties of the language, especially varieties associated with people of colour, even if it is their first language. For example, consider the following excerpt from Nuzhat Amin's (1997) short reflection on how her native dialect of Pakistani English is perceived in Canada.
Another source of disempowerment for South women is the preference given to native speakers of English, and the assumption that only white people are native speakers, and therefore, that only white people know "good" English. I consider myself to be a native speaker of English on the grounds that English is the language I know best, but my colleagues -- teachers of English and English as a second language (ESL), linguists, and applied linguists -- often position me as a non-native speaker, I would say, because I am non-white and I have a Pakistani accent. When I say that I am a native speaker, there is a look of bewilderment and disbelief on their faces.
(Amin 1997: 141)

Judgments can be messy
When you read a finished, typed-up linguistics paper, the data is usually all organized, with no sign of controversy. Each piece of data is marked clearly either as grammatical or ungrammatical. However, if you ever try to do collect your own data, it soon becomes obvious that it does not start out quite so tidy. When you are working with real people, a lot of things can confuse the data. You might get two different people giving you two different judgments for the same sentence. Sometimes even the same person will give you a different judgment from one day to the next. Here are some of the reasons judgments are messy:

The language has more than one variety or dialect. Two speakers may have slight differences in their grammar. Even a single speaker might alternate between their multiple grammars from one situation to the next.

Sometimes a sentence is only grammatical in certain contexts. It's possible that the context your language consultant is imagining has changed your results. One way to reduce this problem is to set up a context for them.

If you ask several judgments in a row, especially if they're similar to each other or tricky constructions, a speaker may start to have judgment fatigue. Judgment fatigue can either cause everything to sound wrong, or cause everything to sound right.

Some constructions are marginally grammatical, which means they're not great but they're also not terrible. Marginal sentences are often marked with a question mark (?) instead of an asterisk (*).

Statistically testing judgments
The practices of using elicitation and introspection as the basis for linguistic research is sometimes criticized as not being sufficiently rigorous. Sometimes linguists who use introspection as their primary method of study get called "armchair linguists" disparagingly. However, we don't need to just assume that a particular method is or is not valid. A study by Sprouse, Schütze, and Almeida (2013) responded to these criticisms by collecting nearly 300 different examples from published papers in one of the top academic journals in linguistics and testing them using multiple different methodologies. They found that these different methodologies, including informal methods like elicitation and introspection, gave the same result 95% of the time, showing that these different methodologies meet a reasonably reliable benchmark. However, this study only tested English data published in an English journal. When a research article in linguistics goes through the peer-review process, the reviewers typically also consider the data in the paper, as well as the argument. So, when a paper is reviewed by reviewers who know the language that the paper is about, the data goes through an additional process of verification, compared to a paper that is reviewed by those who do not know the language. Therefore, errors in data from lesser-studied and lesser-known languages are more likely to slip through the review process than errors in well-known languages like English. Furthermore, making mistakes about lesser-studied languages have more severe consequences. With fewer resources available on the language, it is more likely that the mistake will be perpetuated instead of corrected. Thus, even though elicitation and introspection can provide us with reliable data, we do need to make sure we use them carefully. We should try to replicate our results—double-checking our data with the same speaker in follow-up sessions, as well as checking with multiple speakers. And if a grammaticality judgment is marginal or otherwise tricky, or if there seems to be a lot of dialectal variation, we should consider using alternative methods alongside elicitation and introspection.
Check yourself!
[h5p id="29"]
References and further resources

Academic sources

🔍📚 Amin, Nuzhat. 1997. South Englishes, North Englishes. Canadian Woman studies 17(2). https://cws.journals.yorku.ca/index.php/cws/article/view/8930

Birkeland, Anne, Adeli Block, Justin T. Craft, Yourdanis Sedarous, Sky Wang, Alexis Wu, and Savithry Namboodiripad. 2024. Rejecting nativeness to produce a more accurate and just Linguistics. Language 100 (3): e156-e194.

Sprouse, Jon, Carson T. Schütze, and Diogo Almeida. 2013. A comparison of informal and formal acceptability judgments using a random sample from Linguistic Inquiry 2001-2010. Lingua 134: 219–248.

Truan, Naomi. 2024. Whose language counts? Native speakerism and monolingual bias in language ideological research: Challenges and directions for further research. European Journal of Applied Linguistics 12 (1): 34-53.

4.7. Becoming a linguist: Glossing spoken language data

juliedoner — Fri, 12 Jan 2024 05:34:10 +0000
If the object language you are discussing differs from your metalanguage, you should gloss your examples. This means that you provide a morpheme-by-morpheme translation of your data. This is what enables us, as linguists, to analyze the structure of languages we ourselves do not speak. Normally glosses are provided in a three- or four-line format. In three-line glosses, the first line provides the object language data. The second line provides a morpheme-by-morpheme translation of the object language into the metalanguage. The third line provides a natural, idiomatic translation into the metalanguage. Gonzalez's example (4a) from the previous section is repeated below to illustrate.

(1) Finnish

Lähti-kö Mari?

left-PolQP Mari

'Did Mary leave?'

(Gonzalez 2023: 3)
A 4-line gloss is most often used when the phonological processes in the object language make it difficult to identify the morpheme boundaries. In these cases, the first line would provide the object language without modification and the second line would provide the [pb_glossary id="1577"]underlying form[/pb_glossary] of each morpheme. The last two lines are the same as in a three-line gloss.

(2) Blackfoot

Nitsiikaakaahsi'taki aotoyaakihtsiniki

nit- iik- aak- yaahssi -i'taki a- oto- yáakihts -iniki

1- intend- FUT- be.good.VTI -feel.emotion DUR- go.to.do- go.to.bed -SUBJ

'It will make me happy when you go to bed.'

(Déchaine and Wiltschko 2014: 74)

The first line: The object language data
If the object language of your paper is not written with the Latin alphabet but your metalanguage is, then it is convention to transliterate the object language into the Latin alphabet or transcribe it with the IPA. Many languages have a standardized transliteration system which you can use. For example, Mandarin is often transliterated in the system known as pinyin. If you wish to include the original orthography for any reason, you can use a 4-line gloss with the orthography as the first line and a transliteration or transcription in the second line. Secondly, you must always mark the sentence for its acceptability. If the sentence has no acceptability mark, it means that it is fully acceptable. The most common acceptability marks are shown in Table 1.

Table 1. Acceptability markings

Name Symbol Meaning Example

Asterisk * ungrammatical *Cat the apple ate.

Question mark ? weakly ungrammatical ?I have dived into the pool.

Percentage sign % variation in acceptability %I eat meat anymore.

Number sign # semantically or pragmatically ill-formed #My toothbrush ate an apple.

The most common acceptability mark is the asterisk (*), which means that it is ungrammatical, or, in other words, morphologically or syntactically ill-formed. It is also sometimes used when it is unclear why a sentence is unacceptable. The question mark (?) is used when a sentence feels weakly ungrammatical. In other words, it feels kind of off, but not fully ungrammatical. The percentage sign (%) is used when there is variation. The example in Table 1 uses positive anymore, which is a grammatical feature of some dialects of US American English and Irish English. Finally, the number sign (#), also sometimes called the hash or pound sign, is used for sentences that are semantically or pragmatically ill-formed. The example in Table 1 is semantically ill-formed because toothbrushes cannot eat. Sometimes linguists will use multiple asterisks or question marks, or combinations of asterisks and question marks, to indicate degrees of unacceptability. Sometimes linguists use parentheses to abbreviate multiple examples into one example. If a word is included in the example in parentheses, as in (3a), it means that the word in parentheses is optional. Thus, (3a) means that the sentence is acceptable either without the word quickly, as in (3b), or with the word quickly, as in (3c).

(3) a. Jenna ate the pickles (quickly).

b. Jenna ate the pickles.

c. Jenna ate the pickles quickly.

If a word is included in parentheses with an asterisk also inside the parentheses, as in (4a), that means that the word in parentheses may not be included in that position in the sentence but that the sentence is otherwise grammatical. Thus, the notation in (4a) means that the sentence in (4b), without quickly, is grammatical, but the sentence in (4c), with quickly, is ungrammatical.

(4) a. Jenna ate the (*quickly) pickles.

b. Jenna ate the pickles.

c. *Jenna ate the quickly pickles.

Finally, if a word is in parentheses with an asterisk outside of the parentheses, as in (5a), it means that the word in the parentheses is obligatory. In other words, the notation in (5a) means that a sentence with the word ate, as in (5b), is grammatical, but the same sentence, with ate omitted, as in (5c), is ungrammatical.

(5) a. Jenna *(ate) the pickles.

b. Jenna ate the pickles.

c. *Jenna the pickles.

The second line: Morpheme-by-morpheme glossing
In the second line of a three-line gloss, we include a morpheme-by-morpheme translation. The format for the second-line gloss is highly conventionalized. The most common conventions are summarized by the Leipzig Glossing Rules. If you continue on in linguistics, you should refer to the glossing rules when you read and write papers. This explanation is consistent with the Leipzig glossing rules. Each word in the first line should be aligned with their translation in the second line. When they are not lined up, the examples are much harder to read, as shown by the difference between (6a) and (6b) below. You can line them up manually using spaces and tabs or you can use an invisible table, with each word in a different column.

(6) a. Hindi-Urdu

Ṭi:car=ne Anu=se pu:ch-a: [ki kya: vo ca:i piyegi:].

teacher=ERG Anu=from ask-PFV that KYA s/he tea drink.FUT.3FSG

'The teacher asked Anu whether she would drink tea.'

b. Ṭi:car=ne Anu=se pu:ch-a: [ki kya: vo ca:i piyegi:].

teacher=ERG Anu=from ask-PFV that KYA s/he tea drink.FUT.3FSG

'The teacher asked Anu whether she would drink tea.'

(Gonzalez 2023: 8)
In the second line, we put content words in lowercase letters, and we put grammatical markers in small caps. If small caps cannot be used for some reason, uppercase letters can be used instead. Morphemes are separated out by hyphens (-) in both the first and second lines. Other markings may be optionally used for special kinds of morphemes (=, ~). If a morpheme cannot be translated by a single word, use periods (.) to separate the pieces of the gloss that all translate parts of the same morpheme. Authors often use abbreviations for the grammatical markings in the gloss. The Leipzig glossing rules include a list of common abbreviations. If you use any abbreviations that differ from Lepizig's, you need to include a key or glossary of your abbreviations somewhere in your paper. Some common places authors put their abbreviation list include a footnote at the beginning of the paper, a footnote on the first example in the paper, a note at the end of the paper, or an appendix in a book. All of the abbreviations used in this textbook are included in a back matter section.
The third line: A natural translation
The third line of a gloss should provide a natural translation into the metalanguage. The third line tells you what the example means. Sometimes an author will also include a literal translation with the abbreviation "lit.". If an example is ungrammatical, it might not really have a meaning. In this case, an author might indicate that the meaning is "intended". Finally, if you have multiple examples in a row that all mean the same thing, sometimes the author will only include the translation once, after either the first or last example.
Check yourself!
[h5p id="31"]
References and further resources

Reference materials

📑 Max Planck Institute for Evolutionary Anthropology Department of Linguistics. 2015. Leipzig Glossing Rules. https://www.eva.mpg.de/lingua/resources/glossing-rules.php

Sources for examples
Déchaine, Rose-Marie and Martina Wiltschko. 2014. Micro-variation in agreement, clause-typing and finiteness: Comparative evidence from Blackfoot and Plains Cree. In J. Randolph Valentine and Monica Macaulay, eds. Papers of the 42nd Algonquian Conference. SUNY Press. 69101.
Gonzalez, Aurore. 2023. Interrogative particles in polar questions: The view from Finnish and Turkish. Glossa 8(1): 1–47.

9.2. Different meanings of word

juliedoner — Tue, 16 Jan 2024 02:36:13 +0000
At the beginning of this textbook, we defined morphology as the structure of words. Words are also important units manipulated by the syntax. But what is a word? It turns out that the definition of word is trickier than it may at first seem, especially when we consider the different kinds of words we see in different languages. One reason wordhood is difficult to define is that, even within the same language, we use word to mean lots of different things. Consider the following situations. What is meant by word in each of them?

(1) a. You are playing a word game with friends or family in which you have to rearrange different letter tiles or cards to create a word. You rearrange them to create the word furnitures. You argue about whether it is a word.

b. You are assigned a paper for a class that is supposed to be 500 words long.

c. In a grammar class, you learn that the passive voice must be formed with the word BE; for example, was seen, is seeing, has been seen.

d. It is estimated that the English language includes approximately 1 million words (Merriam-Webster).

To sort out the differences between these different uses of word, linguists use the technical terms lexeme, word token, and word form. A [pb_glossary id="1667"]lexeme[/pb_glossary] is a word in the abstract sense, including all of the different inflectional forms that a word can take. In example (1c), what the teacher means is that the passive voice is constructed using the lexeme BE, but it doesn't have to be be itself, it can be one of its other forms. In (1d), they are estimating that English includes approximately 1 million lexemes. Lexemes are usually rendered in all caps. A [pb_glossary id="1671"]word form[/pb_glossary] is a a particular form of a lexeme. In (1a), you are arguing about whether the lexeme FURNITURE can appear with plural inflection, or, in other words, whether FURNITURE has a plural word form. In (1c), your teacher has taught you that the passive voice must have one of the word forms of the lexeme BE, including was, is, and been. Technically, a lexeme is a set of all the inflected word forms associated with the root word. However, it is cumbersome to list all of the words in the set each time we refer to it. The word form we use to refer to a lexeme is typically called the [pb_glossary id="1678"]citation form[/pb_glossary]. It is the citation form that will typically be listed in the dictionary. Languages differ as to which word form is used as the citation form, depending on convention and what makes sense given the morphological properties of the language. For example, in English, we usually use the non-finite form of verbs and the singular form of nouns as the citation form. The collection of all the word forms in a lexeme is called the [pb_glossary id="1684"]paradigm[/pb_glossary] for that lexeme. In other words, a paradigm is the set of all the inflected forms of a word. On the other hand, words that are related by derivation each belong to a different lexeme. A set of words related by derivation is called a [pb_glossary id="1685"]word family[/pb_glossary]. A [pb_glossary id="1668"]word token[/pb_glossary] is a particular occurrence of a word. In example (1b), the teacher wants you to write a paper that has 500 word tokens. Your paper will likely use the word the several times, but you will count each instance as a separate word (unlike if you were counting lexemes). Let's look at an example together. Below, you will see the entry for buy from the Canadian Oxford Dictionary (Barber 2005).

buy

▶ verb (buys, buy*ing past and past participle bought)

[transitive] purchase; obtain in exchange for money, etc.

serve to obtain: money can't buy happiness.

[transitive] procure (the loyalty, etc.) of a person by bribery, promises, etc.

win over (a person) in this way.

[transitive] get by sacrifice, great effort, etc.: dearly bought ⃒| bought with our sweat.

[transitive] informal accept, believe in, approve of: the police bought our story.

[intransitive] be a buyer for a store, etc.: buys for the furniture chain.

▶ noun informal a purchase: that car is a good buy.

The entry for buy includes at least two [pb_glossary id="1683"]homonymous[/pb_glossary] lexemes, the verb BUY and the noun BUY. These two words are related and belong to the same word family, but must be considered different lexemes because their relationship is derivational rather than inflectional, as the noun is formed by [pb_glossary id="1731"]conversion[/pb_glossary] from the verb. We might also wonder if the five different definitions of the verb buy should also correspond to different homophonous lexemes, or whether they should all be considered different uses of the same lexeme. I will save the finer points of this debate for your semantics class, but we can tell that this [pb_glossary id="1389"]lexicographer[/pb_glossary], at least, decided they were one lexeme (or else they would have listed buy₁ verb and buy₂verb, etc.), likely because the senses listed here all seem to be related to the same core sense by means of extension and metaphor. The citation form for both lexemes is buy. The entry for the verb BUY has three word forms listed in addition to its citation form at the beginning of the entry: buys, buying, and bought. The entry for the noun BUY does not have any additional word forms listed, likely because the only other word form, the plural buys, is completely regular. Overall, this dictionary entry includes 10 word tokens of different forms of buy. There is the headword, which begins the dictionary entry. The second, third, and fourth word tokens are the examples of the different word forms, buys, buying, and bought. The remaining six tokens are found in the examples, such as in money can't buy happiness.

Key takeaways

The word word has lots of different meanings. There are technical terms used to distinguish these meanings.

A lexeme is the abstract sense of a word, including all of its different inflectional forms.

A word form is a particular form of a lexeme.

A word token is a particular occurrence of a word.

Check yourself!
[h5p id="32"]
References and further resources

For a general audience

🧠 Merriam-Webster. How many words are there in English? https://www.merriam-webster.com/help/faq-how-many-english-words.

Sources for examples

Barber, Katherine. 2005. The Canadian Oxford Dictionary. 2nd ed. Oxford: Oxford University Press.

9.3. Packaging words and morphemes

juliedoner — Tue, 16 Jan 2024 02:37:52 +0000
Languages differ based on how they package meaning into a single morpheme, and how they package morphemes into a word. Traditionally, languages have been classified into four morphological types based on the structure of the word: [pb_glossary id="1789"]isolating[/pb_glossary], [pb_glossary id="1792"]agglutinative[/pb_glossary], [pb_glossary id="1793"]fusional[/pb_glossary], and [pb_glossary id="1794"]polysynthetic[/pb_glossary].
Traditional morphological types

Isolating

Isolating languages, also known as analytic languages, have an almost one-to-one morpheme-to-word ratio. Words in isolating languages are often unaffixed bare roots. Bound morphemes (morphemes that cannot stand alone) are very infrequent.

Yoruba, as shown below in (1), is an example of an isolating language. Each word in these examples contains only a single morpheme. Tense markers, such as the past tense marker ti in (1a) and the future marker ma in (1c), are independent words.

(1) a. ọkunrin ti jo. [Yoruba][footnote]The Yoruba data in these examples was provided by Tolani Akinlade.[/footnote]

boy PAST dance

'The boy danced.'

b. ọkunrin ∅ jo.

boy PRES dance

'The boy is dancing.'

c. ọkunrin ma jo.

boy FUTURE dance

'The boy will dance.'

Agglutinative
In agglutinative languages, there is a much higher morpheme-to-word ratio. In other words, there are many morphemes in a single word. Usually, there is little interaction between each morpheme and each morpheme contains a single piece of grammatical information; in other words, there is more-or-less one-to-one matching between morphemes and meaning. However, there is still usually only one root morpheme in most words (with the exception of compound words). Turkish, as shown in Table 1, is an example of an agglutinative language.
Table 1. The paradigm for ev 'house' in Turkish (adapted from Lieber 2016: 148-149)

singular plural

nominative ev ev-ler

definite-accusative ev-i ev-ler-i

genitive ev-in ev-ler-in

dative ev-e ev-ler-e

locative ev-de ev-ler-de

ablative ev-den ev-ler-den

In Turkish, nouns inflect for number (singular or plural) and [pb_glossary id="700"]case[/pb_glossary] (nominative, definite-accusative, etc.). Case marks for the role that the noun plays in the sentence (e.g., subject, object, object of preposition, etc.). Each morpheme in Table 1 has exactly one meaning. For example, -ler means plural, -i means definite-accusative, and so on. To form a word in Turkish, multiple morphemes are strung together.
Fusional
Fusional languages, on the other hand, have morphemes that express multiple pieces of grammatical information simultaneously. There is almost no one-to-one match between pieces of meaning and number of morphemes. There also tend to be multiple morphemes in a single word. Latin is an example of a fusional language, as shown in Table 2.
Table 2. The paradigm for puella 'girl' in Latin (adapted from Lieber 2016: 149)

singular plural

nominative puell-a puell-ae

gentive puell-ae puell-ārum

dative puell-ae puell-īs

accusative puell-am puell-ās

ablative puell-ā puell-īs

Latin nouns, like Turkish, also inflect for case and number. However, unlike Turkish, we cannot separate the parts of the word that indicate number and the parts that indicate case. The morpheme -am, for example, simultaneously encodes both accusative and singular, while the morpheme -ārum simultaneously encodes both genitive and plural. We cannot separate either -am nor -ārum into two parts; they are each one morpheme with two meanings. There is no morpheme that repeats down either the singular or plural columns, nor any morphemes that repeat across each case row.
Polysynthetic
Finally, polysynthetic languages often have a single word representing meanings that are associated with whole sentences in other languages. These words contain long strings of roots and affixes. Often the object, and sometimes the subject, are inflected onto the verb. Inuktitut is an example of a polysynthetic language, as shown in (2).

(2) a. ayagciqsugnarqnillruuq

ayag- ciq- yugnarqe- ni- llru- u- q

go- FUT- probably- claim- PAST- INDIC.INTR- 3SG

'He said he would probably go.'

b. uqa- limaar- vi- liu(ng)- inna- nngit- -tunga

speak- all.of- NOM- make- always- NEG- DEC.1SG

'I was not always making libraries.'

(Inuktitut; Compton and Pittman 2010: 2168)
Both of the Inuktitut words in (2) contain multiple morphemes. The example in (2a) contains root words that are associated with more than one part of speech, including the verbs ayag- 'go' and ni- 'claim' and the adverbial yugnarqe- 'probably'. The example in (2b) contains inflectional morphemes that are associated with nouns, like the nominative case marker vi- as well as inflectional morphemes that are associated with verbs, like the declarative first-person singular marker -tunga. Each of these words are translated into a whole English sentence. Example (2a) even has two clauses in the translation! In speech, we don't usually pause in between words, so why do we think that the examples in (2) are each single words? There are a few reasons. First, speakers have intuitions that they are single words. Second, the morphemes are mostly bound morphemes that cannot stand on their own. Third, many phonological processes cross morpheme boundaries but do not cross word boundaries, giving us a big clue about where word boundaries are. Polysynthetic languages are sometimes confused with agglutinative languages. Unlike agglutinative languages, a polysynthetic language will often have more than one root. Compound words, you may recall, also have more than one root. However, unlike compound words, words in polysynthetic languages contain inflectional morphemes from more than one part of speech in a single word. Another potential difference between polysynthesis and compounding is that polysynthesis is productive, while compounding varies in productivity across languages.
Synthesis, fusion, and exponence
The four morphological types above really conflate three properties: [pb_glossary id="1821"]synthesis[/pb_glossary], [pb_glossary id="1822"]fusion[/pb_glossary], and [pb_glossary id="1823"]exponence[/pb_glossary]. Not all languages fit perfectly into one of the four traditional categories, so it is sometimes useful to consider each of these properties independently. Some languages may even differ depending on the kind of morpheme. For example, Chapter 21 of the World Atlas of Language Structures (WALS) surveys how different languages package both noun and verb inflection. If you cross-reference the two maps associated with Chapter 21, you'll find that some languages, including Chukchi, Finnish, German, West Greelandic, Nenets, Russian, and Yaqui, combine case with number in a single morpheme, similar to the Latin examples in Table 2, but do not combine their tense markers with another category. Conversely, you'll find that other languages, including Aymara, Brahui, Georgian, Hindi, Spanish, and Tibetan combine their tense markers with agreement in a single morpheme but do not combine case with another category (Bickel and Nickels 2013b). It may be enough to classify both of these sets of languages as fusional in some situations, but in other situations it may be useful to make more nuanced distinctions. The four traditional categories are discrete categories, meaning a language either fits into the category or does not. The three categories of synthesis, fusion, and exponence, on the other hand, are scales, which means that a language might fall on the high end of the scale, on the low end of the scale, or somewhere in the middle.
Synthesis
Synthesis is the measure of how many morphemes may combine into a single word. A language with a high degree of synthesis will have many morphemes contained into a single word, while a language with a low degree of synthesis will have few morphemes combined into a single word. Languages with a low degree of synthesis are often called [pb_glossary id="1824"]analytic[/pb_glossary] languages.
Fusion
Fusion is the measure of how phonologically separable morphemes are from their hosts. In a language with a low degree of fusion, it is easy to identify the boundaries between different morphemes. Isolating languages can be classified as low fusion languages because most of the morphemes are separated by a word boundary, which is a very clear phonological boundary. Agglutinative languages also have a low degree of fusion. On the other hand, a language with a high degree of fusion will have morpheme boundaries that are less clear. English exhibits a higher degree of fusion in some irregular verb inflections, such as catch ~ caught, fall ~ fell, and bend ~ bent. In these forms, there is no part of the word that we can clearly identify as either a present tense or past tense morpheme. Semitic languages such as Hebrew and Arabic are well-known for using vowel changes as one of their main morphological processes (instead of just in irregular verbs like English), and so would be classified as having an even higher degree of fusion than English. Note that this meaning of fusion is very different from the category fusional discussed above, which is more closely related to the concept of exponence.
Exponence
Exponence is the measure of how many meanings can be encoded into a single morpheme. Wari', a Chapacura-Wanham language spoken in Brazil, has high exponence on the word na, which encodes the person and number of the subject, as well as several categories of verb inflection (realis, non-future, and active), for a total of five grammatical categories expressed in a single morpheme. When multiple meanings are encoded in a single morpheme in this way, it is known as [pb_glossary id="4194"]cumulative exponence[/pb_glossary].

(3) Toc na com.

drink.SG 3SG.REAL.NONFUT.ACTIVE water

'He is drinking water.'

(Wari'; Everett and Kern 1997: 339, as cited in Bickel and Nickels 2013b)
Low exponence, on the other hand, will only have one meaning encoded in each morpheme.
Redefining the traditional categories
Let us now revisit the four traditional morphological categories and redefine them in terms of their synthesis, fusion, and exponence. Isolating languages have one or close to one morpheme per word, which means they have low synthesis. The morphemes, belonging as they do to independent words, are phonologically separable, and therefore isolating languages have low fusion as well. Finally, most morphemes in isolating languages encode only one meaning, and so they also have low exponence. Agglutinative languages have many morphemes per word, and therefore have high synthesis. However, the morphemes are still phonologically separable and encode one meaning per morpheme; in this way, agglutinative languages have low fusion and low exponence. Fusional languages are characterized by having many meanings encoded into a single morpheme, which means they have high exponence. The definition of fusional languages doesn't specify their degree of synthesis or fusion, but fusional languages will tend to have fewer morphemes per word than agglutinative languages since each morpheme is carrying more information. Finally, polysynthetic languages are highly synthetic. The definition of polysynthesis does not specify the degree of fusion or exponence, but polysynthetic languages tend to have a small degree of fusion, with phonological processes crossing morpheme boundaries, and medium amounts of exponence.
Table 3. Defining the four traditional morphological categories in terms of synthesis, fusion, and exponence

Synthesis Fusion Exponence

Isolating languages low low low

Agglutinative languages high low low

Fusional languages medium unspecified high

Polysynthetic languages very high unspecified unspecified

Using The World Atlas of Language Structures (WALS)

You can explore how these morphological properties show up in the different languages of the world on The World Atlas of Language Structures (WALS). In WALS, Chapter 20 describes fusion, Chapter 21 describes exponence, and Chapter 22 describes synthesis. Each article is associated with one or more maps showing how the different properties are distributed across a sample of the world's languages. You can even combine the data from two or more maps to see how the different properties interact. WALS is put together from a descriptive functionalist framework. As such, the terminology in some chapters might differ from what we learn in class, but you can often figure it out if you read the explanations in the chapters. However, as you explore WALS, you should keep in mind that it has a couple of weaknesses. First, the authors of each chapter on WALS are experts in the topic of their chapters, but not on all of the languages in the database (which would be impossible!). Often, then, the authors have classified languages based on a surface understanding of the language from referencing grammars and other published works. Second, mostly only spoken languages are represented in the database, with the exception of Chapters 139 and 140, which are dedicated to signed languages.

Key takeaways

Language differ in how many meanings are put in a single morpheme and how many morphemes are put in a single word.

Isolating languages have a near one-to-one morpheme-to-word ratio, with each morpheme containing one meaning.

Agglutinative languages have many morphemes in each word, but each morpheme still only contains one meaning.

Fusional languages put multiple meanings into some morphemes and multiple morphemes into each word.

Polysynthetic languages have many morphemes in a single word, often the equivalent of a sentence in other languages. There may be multiple roots in a single word, as well as inflectional affixes from more than one part of speech.

These four morphological types conflate three properties. Synthesis is the measure of how many morphemes may combine in a single word. Fusion is the property of how phonologically separable morphemes are from their hosts. Exponence is the property of how many meanings can be encoded into a single morpheme.

Check yourself!
[h5p id="34"]
References and further resources

Comedy and satirical linguistics

🎉 Phlogiston, Phineas Q. 2007. Cartoon theories of linguistics part 3: Morphological typology. Speculative Grammarian CLII (3). https://specgram.com/CLII.3/09.phlogiston.cartoon.3.html.

Sources for examples

Compton, Richard, and Christine Pittman. 2010. Word-formation by phase in Inuit. Lingua 120: 2167–2192.

Lieber, Rochelle. 2016. Introducing Morphology. 2nd edition. Cambridge: Cambridge University Press.

Academic sources

🔍 Bickel, Balthasar, and Johanna Nichols. 2013a. Fusion of Selected Inflectional Formatives. In WALS Online (v2020.3) [Data set], ed. Matthew S. Dryer and Martin Haspelmath. Zenodo. https://wals.info/chapter/20.

🔍 Bickel, Balthasar, and Johanna Nichols. 2013b. Exponence of Selected Inflectional Formatives. In WALS Online (v2020.3) [Data set], ed. Matthew S. Dryer and Martin Haspelmath. Zenodo. https://wals.info/chapter/21.

🔍 Bickel, Balthasar, and Johanna Nichols. 2013c. Inflectional Synthesis of the Verb. In WALS Online (v2020.3) [Data set], ed. Matthew S. Dryer and Martin Haspelmath. Zenodo. https://wals.info/chapter/22.
Mattissen, Johanna. 2004. A structural typology of polysynthesis. Word 55 (2).

9.4. Wordhood: An unsolved problem

juliedoner — Tue, 16 Jan 2024 02:40:08 +0000
In Section 6.2, we learned about how the word word can be used in many different ways. In addition, we defined many technical terms we can use to distinguish these different uses. But we still haven't defined what a word is. There is a reason for that! The definition of word is an unsolved problem in linguistics. There is no definition of word that works for all known languages.
Some attempts at defining word
In this section, we will consider many attempts at definitions of word and discuss how they fall short. Before you read this section, though, take a moment and attempt to write a definition of word. As you read through these possible definitions, compare them to the one you wrote. Do any of the problems discussed here also apply to your definition?
Based on spelling
We could attempt to define words based on how they appear in writing. One possible definition, then could be as in (1). (1) Definition attempt 1: A string of letters written with a space on each end. The first problem with defining word based on [pb_glossary id="1843"]orthography[/pb_glossary] is that not all languages are written down. Definition (1), then, would mean that languages without a writing system would have no words at all, which does not match speaker intuition. Even languages with a writing system don't all indicate word boundaries. Classical Chinese, Ancient Greek, and Classical Latin were all written without spaces between words. In languages that have a writing system that uses spaces, the use of spaces doesn't always correspond exactly to word units. For example, English compound words can be written as one word, with a hyphen, or with a space, as illustrated in (2). (2) a. Compound words with spaces: hot dog, high school, common sense, dining room, first aid, peanut butter, post office, prime minister, search engine, remote control, washing machine, role model b. Hyphenated compound words: sister-in-law, check-in, far-fetched, free-for-all, know-how, merry-go-round, one-sided, well-being, up-to-date, self-esteem, freeze-dried, cage-free c. Compound words without spaces: greenhouse, airport, bathroom, basketball, daylight, desktop, firefly, grandmother, grapefruit, grasshopper, lipstick, mailbox Whether a compound word has a space, a hyphen, or is written as one word is just convention, and it can change over time. For example, in older texts, you may see ice-cream written with a hyphen, whereas in modern texts it tends to be written with a space instead, as ice cream. However, there is evidence, as we will discuss in Chapter 10 on compounding, that compound words behave as a single word, regardless of how they are written.
Based on uninterruptibility
We could attempt to define words based on their interruptibility. Perhaps a word is a unit that cannot have anything inserted into its middle. (3) Definition attempt 2: A unit that cannot have anything added to it, except at the edges. One of the big problems with this definition is that it doesn't account for [pb_glossary id="1846"]infixes[/pb_glossary]. For example, the Lakhota first singular subject marker -wa- is an infix.[footnote]The Lakhota first person singular subject marker alternates between being a prefix and an infix, depending on the stem word.[/footnote] This is shown in (4), where the roots máni, aphé, and hoxpé have -wa- inserted in between the first and second syllable to form the first-person singular.

(4) a. máni 'he walks' ma-wá-ni 'I walk'

b. aphé 'he hits' a-wá-phe 'I hit'

c. hoxpé 'he coughs' ho-wá-xpe 'I cough'

(Lakhota; Albright 2000: 2)
If we maintain the the definition in (3), that a word is a unit that can only be added to at its edge, data like (4) would force us to conclude that the roots máni, aphé, and hoxpé each consist of two words. This is problematic, though, because the roots only contain one morpheme—that is, one piece of meaning! How could one morpheme be spread across two words? Another challenge to the idea that a word is not interruptible are phrasal verbs. Phrasal verbs, such as turn on, blow up, break down, and call off are a subcategory of compound words that usually consist of a verb and a preposition. Phrasal verbs can have past tense markers inserted in between their two components, as in called off. Phrasal verbs are also well-known for allowing the object to occur in between the verb and the preposition, as shown in (5) to (8).

(5) a. turn on the lights

b. turn the lights on

(6) a. blow up the building

b. blow the building up

(7) a. break down the box

b. break the box down

(8) a. call off your dog

b. call your dog off

English phrasal verbs, on their own, are perhaps not the strongest piece of evidence, since we could perhaps argue that they are two words, not one. German has a similar pattern that is even more puzzling.[footnote]German data provided by Katharina Pabst.[/footnote] In German, verbs with separable prefixes are written with one word when there is an auxiliary, for example aufstehen in (9b) and zurückgeben in (10b). But in (9a) and (10a), when there is an auxiliary, the word is split into two pieces which aren't even next to each other! The verb portions of the particle verbs, stehst in (9a) and gebe in (10a), are the second word in each sentence, while the preposition portions of the particle verbs, auf in (9a) and zurück in (10a), are at the end of each sentence.

(9) German

a. Wann stehst du morgen auf?

when stand.pres.2sg you tomorrow up

'When will you get up tomorrow?'

(11) b. Wann willst du morgen aufstehen?

when want.pres.2sg you tomorrow stand.up

'When do you want to get up tomorrow?'

(10) German

a. Ich gebe meiner Tante ihr Buch zurück.

I give.pres.1sg 1sg.dat.f.poss aunt 3sg.f.dat book give.back

'I give my aunt her book back.'

(10) b. Ich muss meiner Tante ihr Buch zurückgeben.

I must.pres.1sg 1sg.dat.f.poss aunt 3sg.f.dat book give.back

'I have to give my aunt her book back.'

Based on independence
Perhaps we should define words, not based on whether we can interrupt them, but based on whether they can stand on their own. (11) Definition attempt 3: A unit that can be pronounced in isolation. The problem with this definition is, first of all, that some subcomponents of words can stand on their own. For example, clippings like math, flu, and fridge can stand on their own. Does that mean that mathematics, influenza, and refrigerator are not words, but phrases? Our intuitions (or mine at least) says that can't be right, because that would mean that the parts of mathematics, influenza, and refrigerator that don't show up in math, flu, and fridge are also words. Secondly, some words, especially functional words, can't really stand on their own, such as the and of.
Based on the interface between semantics and phonology
Perhaps, then, we could define words based on their function as a form-meaning pair. In that case, a word would be a string of sounds or signs that, together, are associated with a meaning. (12) Definition attempt 4: A form-meaning pair. However, this definition does not quite work either. There are, of course, form-meaning pairs that are bigger than a word as well as form-meaning pairs that are smaller than a word. It is easy to find form-meaning pairs that are smaller than a word. Any morpheme is a form-meaning pair, thus any multimorphemic word consists of form-meaning pairs that are smaller than a word. For example, teapot consists of two form-meaning pairs, tea and pot. There are also form-meaning pairs that are bigger than a word, which are called [pb_glossary id="1853"]idioms[/pb_glossary]. Idioms are phrases with non-compositional meaning. In other words, it is the entire phrase that is paired with a meaning. Some examples of English idioms are listed in (13).

(13) a. break a leg to wish someone luck before a performance

b. beat around the bush explain or request something indirectly

c. hit the sack go to bed

d. on the ball prepared, ready

e. raining cats and dogs raining very hard

f. spill the beans tell a secret

g. under the weather sick

Based on phonological domains
Perhaps we can define words based on the limits of phonological processes, such as stress assignment. (14) Definition attempt 5: The domain in which phonological processes such as stress occur. The first problem with this definition is that functional words such as the and of often do not receive stress at all, in which case they would not count as words. The second problem is that the domain of phonological processes vary, both within the same language and across different languages. For example, the two roots in a compound word sometimes behave like they're in the same phonological domain, and sometimes they do not. Let's consider vowel harmony in Finnish, which causes all vowels in the same word to match in backness. In (13), the adessive marker appears as -llä if the vowels in the stem are front vowels, as in (13a), and as -lla if the vowels in the stem are back vowels, as in (13b).

(15) a. pöydä-llä

table-ADESS

'on the table'

b. kadu-lla

street-ADESS

'in the street'

(Finnish; Julien 2002: 24)
The same pattern does not hold in compound words, as shown in (16). The first root, pää 'head', has front vowels, while the second root, kaupunki 'city', has back vowels. Vowel harmony does not cross the boundary between the two compound words.

(16) pää-kaupunki

head-city

'capital'

(Finnish; Julien 2002: 24)
In contrast, compound words in Greek do behave as a single domain for stress assignment, as shown in (17). In this compound word, there is only one stress, marked by the accent on the final vowel.

(17) ksilo-θimonyá

wood-stack

'wood stack'

(Greek; Julien 2002: 17)
If this compound word behaved like two separate words, we would expect there to be a stress on each root. Instead, there is only stress on the head of the compound word.
Based on speaker intuition
Let's attempt one last definition of word. Perhaps we can consider words to be a unit according to the intuition of the language users. (18) Definition attempt 6: A unit that stands on its own according to the intuition of language users. The difficulty with this definition is that literate language users don't normally have intuitions about wordhood that are separate from their writing system. Illiterate language users are becoming more difficult to find, due to global advances in education. The rules from the writing system might reflect underlying grammatical principles, but they also might not.
What do we do now?
We do not have a consistent way of defining words across all contexts and languages. This could mean lots of different things. It could mean that we just haven't identified the right definition yet. It could mean that there is no universal definition of word, but that it is defined language-by-language. It could also mean that the word word refers to more than one thing, and using the same word in all of these different contexts has confused us! Finally, it could mean that there really is no such thing as words in the grammar at all. Let's look at these last two ideas more closely.
Different kinds of word
One possible solution is to claim that there are different domains that are sometimes called words: phonological words, grammatical words, and orthographic words. Sometimes these different domains coincide, and refer to the same string, but sometimes they don’t. The different wordhood domains could be defined as follows:

Phonological words are a unit in the prosodic structure of a sentence. They are the domains in which word-level phonological processes occur, including stress assignment. Phonological words are the smallest unit that can stand on their own.

Grammatical words are a unit in morphosyntactic structure. They correspond to the terminal nodes in syntax trees. In other words, they are the units that syntax can manipulate. This definition is probably the least satisfying of these three, since we sometimes put morphemes that cannot stand on their own in terminal nodes of syntactic trees.

Orthographic words are a unit we use in writing, based on spelling convention.

In other cases, linguists divide the notion of word into lexeme, word token, or word form, as discussed in Section 8.2.
Maybe words aren't real
Another possible solution to this problem is to assume that words aren't actually a real component in our model of human language. Julien (2002) is one linguist who hypothesizes this:
...my working hypothesis in the following will be that 'word' in the nonphonological sense is a distributional concept. That is, if a given string of morphemes is regarded as a word, it simply means that the morphemes in question regularly appear adjacent to each other and in a certain order. The reason the morphemes show such behavior is to be found in their syntax. But notably, the structural relation between the morphemes is not directly relevant for the word status of the string; it only matters insofar as some structural arrangements of morphemes may result in independent distribution and internal cohesion, whereas others may not. Crucially, if wordhood cannot be associated with any particular structural morpheme configuration, it follows that grammar cannot have at its disposal any specific word-forming devices. If a word is just the accidental outcome of the manipulation of morphemes that takes place in syntax, it must be the case that words come into being in our perception; that is, words are perceived rather than formed.
(Julien 2002: 36)

Basically, Julien is arguing that words are just strings of morphemes that appear together frequently enough that we treat them in a special way, rather than from some special configuration in syntactic structure.

Key takeaways

There is no way to define word that works for all languages because orthographic, phonological, morphological, syntactic, and semantic boundaries don't all align in the same way across words and across languages.

One possible explanation for this is that phonological words, grammatical words, and orthographic words are three different kinds of units that only sometimes coincide.

Another possible explanation is that words are strings of morphemes that appear together frequently but that don't correspond to any particular structure.

Check yourself!
[h5p id="35"]
References and further resources

Academic sources

Albright, Adam. 2000. The productivity of infixation in Lakhota. UCLA Working Papers in Linguistics 0: 1–19.

Julien, Marit. 2002. Syntactic heads and word formation. Oxford: Oxford University Press.

Chapter 9 practice exercises

juliedoner — Tue, 16 Jan 2024 02:40:35 +0000

Data analysis

Exercise 1. Dictionary search
Beginner [9.2] Look at the dictionary entry for the word meaning 'canoe' in this Ojibwe dictionary.
a. What is the citation form for this word? b. Which word forms does it list? c. Identify all of the word tokens of the word for canoe on the website. List two examples here. d. How many lexemes do you think are represented by this one dictionary entry? Explain your reasoning.

Exercise 2. Linguistic analysis
Intermediate [9.3] Consider the following data from four different languages. Based on the data provided here, classify each language either as isolating, agglutinative, fusional, or polysynthetic. Explain the reasoning behind your classification choice. Abbreviations used in these examples include 1, 2, 3 = 1st, 2nd, 3rd person, ACC = accusative (object pronoun), CPL = completive, DAT = dative (indirect object pronoun), DET = determiner, FUT = future, NOM = nominative (subject pronoun), PL = plural, PROG = progressive (action in progress), SG = singular

(1) q'ai-hwagait-tsagam-yu'k-det

still-CPL-toward_shore-move-3PL

'they moved still far away toward the shore'

(Tsimshian; Boas 1911: 325, as cited in Mattissen 2004: 191)

(2) Le are k-u-k'yek le t'oyt'ik.

DET 3SG PROG-3SG-kick DET ball

'S/he is/was kicking the ball.'

(K'iche'; data from Elsa Sacalshot)

(3) a. bɨ vuch-a-th tsɨ

1SG.NOM see-1SG-2SG.ACC 2SG.ACC

'I will see you.'

(Kashmiri; Wali & Koul 1997: 228, as cited in Manetta 2014: 246)

b. su vuch-iy (tse)

3SG.NOM see-2SG.DAT 2SG.DAT

'He will see you.'

c. su vuch-i tǝmis

3SG.NOM see-FUT 3SG.DAT

'He will see him.'

(Kashmiri; Manetta 2014: 246)

(4) Kua kitia e au a Moka.

PERF see ERG 1SG ABS Moka

'I saw Moka.'

(Niuean; Massam 2020: 4)

Communication and study skills

Exercise 3. Wordhood
Advanced [9.4] Polysynthetic languages cause particular challenges for the concept of wordhood. Choose one of the attempted definitions of word from Section 9.4 and discuss whether that definition could handle polysynthetic languages.
Exercise 4. Dictionary entry production
Intermediate [9.2] Think of a slang word or a word that has only recently entered English or another language you speak. Construct a dictionary entry for the word. Include the citation form, inflectional forms (if any), part of speech, pronunication, a definition, and examples of the word in use. Hint: Use a published (paper or online) dictionary as a model for how to format the dictionary entry.
Exercise 5. Word games
Advanced [9.2, 9.4] Look up the rules of a word game and answer the following questions. You can use the rules from the game Codenames copied below or choose another word game.
a. Think about the rules of the game. Which uses of word are relevant to the game mechanics (for example, lexeme, word form, word token, etc...)? Does the game need to limit productive formation of new words, or would creative formation of new words make the game more fun? b. If the game defines what is or is not a word for the purposes of the game, does the game's definition of a word have any problems? Can you think of anything that you would consider a word that is excluded by the game's definition, or anything that you would not consider a word that is included? If you were involved in the game design, would you recommend any changes to how word is defined in the instructions? c. If the game does not define what is a word, how would you define a word for the purposes of this game? Remember to use non-technical terminology, so that players will be able to understand the instructions. Why did you choose the definition you chose?

In the game Codenames, one player gives clues to the rest of their team to help them pick out their team's secret words out of a set of 25 words. The catch is that the clue can only be one word long! Because of this, the game instructions spend some time defining what counts as a single word. Here is an excerpt.
Sometimes you have to make judgment calls about what is valid and what is not. Different groups may prefer to play the game differently.
Compound words
English has three ways to write a compound word. Greenhouse is one word. Pack rat is two words. Mother-in-law is hyphenated. Technically, only greenhouse can be a one-word clue. You can decide to allow any compound words. However, in no case should a player be allowed to invent compound words. Lunar squid is not a valid clue for MOON and OCTOPUS.
Proper names
Proper names are always valid clues if they follow the other rules. George is a valid clue, but you might want to specify whether you mean George Washington or George W. Bush. Your group can agree to count proper names as one word. This would also allow titles such as The Three Musketeers. Even if you don't allow multi-word proper names, you might want to make an exception for place names like New York. Spymasters should not be allowed to make up names, not even names that turn out to be real. Sue Mee is not a valid clue for CHINA and LAWYER.
Acronyms and abbreviations
Technically, CIA is not one word. But it is a great clue. You can decide to allow common abbreviations like UK, lol, and PhD. And words like laser, radar, and sonar are always allowed, even thought they originated as acronyms.
Homonyms
Some people prefer to allow a more liberal use of homonyms. You can allow knight to be a clue for night-related things if that makes the game more fun for you.
(Chvátil 2015: 7)

Research and application

Exercise 6. Etymology
Beginner-Intermediate [9.1] Look up the following words in an etymological dictionary. What is their word formation process? How can you tell? Be sure to use the technical terms for different formation processes discussed in Section 9.1. Note: Sometimes the dictionary entry will tell you directly. Other times, you will have to read the etymology and decide for yourselves based on the description of the word's history.
a. Scuba (n) b. Brainwash (n) c. Camcorder (n) d. Traceable (adj) e. Bandersnatch (n) f. Party (v) g. DIY (adj/n) h. Gym (n) i. Boomerang (n) j. Scavenge (v) k. Exam (n) l. Okay (adj/adv) m. Wabe (n) n. Keyboard (n) o. Labradoodle (n) p. Yuppie (n)

Exercise 7. WALS
Advanced [9.3] Choose one language you don't know much about and look it up in WALS chapters 20, 21, and 22.
a. Provide a brief introductory statement about your language. (e.g., What is its name, what language family is it from, and where is it spoken?) b. Describe your language in terms of fusion, exponence, and synthesis. c. Based on what you found, does your language fall neatly into one of the traditional four categories? Why or why not?

Language journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 8
Beginner [9.2] Find a dictionary for your language. Pick a dictionary entry, and identify the lexeme, citation form, word forms, and word tokens in that entry.
Exercise 9
Intermediate [9.1] Is there an etymological dictionary available for your language? If so, find and describe the etymology of three different words in your language with three different word formation processes.
Exercise 10
Advanced [9.3] Describe the synthesis, fusion, and exponence of your language. Does your language fit well into one of the categories of isolating, agglutinative, fusional, or polysynthetic? Explain your reasoning.
References and further resources

Sources for examples
Chvátil, Vlaada. 2015. Codenames. Czech Games Edition.
Manetta, Emily. 2014. Markedness and syncretism in Kashmiri differential argument encoding. In Pritha Chandra and Richa Srishti (eds.), The Lexicon-syntax interface: Perspectives from South Asian languages. Amsterdam: John Benjamins. 245–270.
Massam, Diane. 2020. Niuean. Oxford: Oxford University Press.
Mattissen, Johanna. 2004. A structural typology of polysynthesis. Word 55 (2).

9.1. Word formation processes

juliedoner — Tue, 16 Jan 2024 02:41:03 +0000
Sometimes, new words are added to a language, either due to [pb_glossary id="1742"]neologism[/pb_glossary] [niˈɑləd͡ʒɪzm̩ ] (the creation of a new word within the language itself) or due to [pb_glossary id="1732"]borrowing[/pb_glossary] (when one language, the recipient language, adapts words or other grammatical features from another language, the donor language). [pb_glossary id="1743"]Taboo avoidance[/pb_glossary] is a special kind of neologism in which a word is replaced (either by a modified pronunciation or by a newly invented word) to avoid an obscene or taboo meaning, such as gosh for God. Taboo meanings are often centred on sexuality, bodily waste, and religion. In this section, we will be using < and > symbols from historical linguistics to show the historical direction of change. For example, X > Y means that the older form X was replaced by the newer form Y, while Y < X means the opposite, that the newer form Y originated from the older form X.
Deriving
One of the most common ways to form new words is by adding new [pb_glossary id="161"]morphemes[/pb_glossary]. There are two main kinds of morphemes, [pb_glossary id="1723"]inflectional[/pb_glossary] morphemes and [pb_glossary id="1724"]derivational[/pb_glossary] morphemes. We will learn more about the difference between them in Chapter 9, but for now, you can think about the two terms this way: inflectional morphemes add grammatical information to a word, such as plurality or a tense marker, while derivational morphemes create a new word. Some examples of English derivational processes are shown in Table 1.
Table 1. Examples of derivation in English

Stem Derivational morpheme Derived word

a. new -ness newness

b. play -ful playful

c. playful -ness playfulness

d. friend -ly friendly

e. friendly un- unfriendly

f. friend -ship friendship

As shown by Table 1 examples (c) and (e), you can add more than one derivational morpheme to the same word. When a grammatical pattern can be extended in predictable manners, it is called [pb_glossary id="164"]productivity[/pb_glossary]. Some derivational morphemes are productive (like -ish) and others are not (like -ness), as shown in Table 2. While -ish can be used with all six stems, -ness can only be used with examples (a) to (c). We will discuss productivity more in Chapter 13.

Table 2. The productivity of -ish and -ness

Adjective stem with -ish with -ness

a. new newish newness

b. blue bluish blueness

c. slow slowish slowness

d. long longish *longness

e. strong strongish *strongness

f. popular popularish *popularness

[pb_glossary id="1731"]Conversion[/pb_glossary] is a special kind of derivation, in which you change the part of speech of a word without changing the form of the word, for example changing the verb walk to the noun walk.

Borrowing

Another way that a language might acquire new words is through contact with another language. In borrowing, the borrowed words are called [pb_glossary id="1737"]loanwords[/pb_glossary] and are often [pb_glossary id="1733"]nativized[/pb_glossary], which means that the pronunciation, and sometimes the morphology, is changed to match the regular patterns of the recipient language. For example, the Innu word for pancake is tekanep, taken from French des crêpes /dekʀεp/. The determiner des is incorporated into the word, the /d/ is devoiced, /ʀ/ becomes /n/, and a vowel is added to break up the consonant cluster. Borrowing can happen for a variety of reasons, including prestige (e.g., numerous French and Latin loanwords in academic English) and need (e.g., animal words such as moose, caribou, chipmunk, skunk, opossum, and raccoon have been borrowed into English from Algonquian languages). The direction of borrowing can often be inferred based on knowledge of phonology, sound change, morphology, cognates, real world geography, ecology, cultural differences, history, and so forth. Borrowing often occurs within particular semantic domains (food, religion, politics, and so forth), especially when very different cultures come into contact. The same source word may take different paths and be borrowed multiple times into the same language. This may be because two languages are in contact over long periods of time and borrow the same word at multiple points in its history, with different pronunciations as the source language undergoes change. It also happens if the same word is borrowed from different dialects or varieties. For example, captain, chef, and chief all originate from caput, the Latin word for ‘head’, via French. Captain and chief were both borrowed in about the 14th century from the Old French words capitaine and chief. Capitaine came from the Latin capitaneus, a derived form of the word caput to mean ‘leader’ rather than ‘head’. Over the next few centuries, the French word chief became chef. It was then borrowed again into English in the 18th century as part of the phrase chef de cuisine ‘head of the kitchen'. A [pb_glossary id="1734"]calque[/pb_glossary] or [pb_glossary id="1751"]loan translation[/pb_glossary] is a specific type of borrowing that includes a direct translation of an idiomatic loanword, rather than a phonological nativization, for example English black market < German Schwarzmarkt, literally ‘black market’.
Root creation
Sometimes a word is invented without basis on any previously existing words, which is called [pb_glossary id="1735"]root creation[/pb_glossary]. This happens most often in product naming and fiction. Some examples include google ‘a search engine brand’ or lightsaber 'a laser sword from the series ‘Star Wars’. Sometimes, a product name expands to include all similar objects, regardless of brand. This is called [pb_glossary id="1736"]genericization[/pb_glossary]. Some examples include kleenex, thermos, or popsicle, which now refer to any facial tissue, double-walled food storage container, or frozen treat on a stick, regardless of brand. However, all three started as brand names originally. The word google is also well on its way in genericization. Do you call it googling even if you’re using a different search engine?
Shortening
Shortening an existing word is called [pb_glossary id="1738"]clipping[/pb_glossary] or [pb_glossary id="1746"]truncation[/pb_glossary], like with math < mathematics, phone < telephone, or flu < influenza. Sometimes people reanalyze a monomorphemic word to be polymorphemic, and then remove the extra morphemes. This is called [pb_glossary id="1739"]backformation[/pb_glossary]. For example, the noun editor existed before the verb edit. However, people misinterpreted the /-əɹ/ ending of editor to be the same as the derivational morpheme -er 'one who does V', and assumed that there must be a verb to edit to go with the noun editor. It is impossible to tell whether a word is a backformation without looking up which form came first in an etymological dictionary. Other examples of backformation include haze < hazy, diagnose < diagnosis, or Old English pise ‘pea’ (plural pisan) > Modern English pea (plural peas), with the singular formed by stripping of the supposed plural -s in the pise, even though pise was originally singular and had no plural morpheme.
Combining
Sometimes new words are coined by combining existing words. If you combine two roots and keep the whole of both roots, you’ve made a [pb_glossary id="189"]compound word[/pb_glossary], such as rainbow, blackboard, or greenhouse. In contrast, if one or both words is not fully preserved in the resulting form, this is called a [pb_glossary id="1741"]blend[/pb_glossary] or [pb_glossary id="1752"]portmanteau[/pb_glossary]. Examples of blends include motor hotel > motel, smoke + fog > smog, or breakfast + lunch > brunch. If you only keep the first letter (or two) of each word in a phrase, it’s an [pb_glossary id="1747"]acronym[/pb_glossary] if you pronounce it like a word, as with laser < light amplification by stimulated emission of radiation or radar < radio detecting and ranging. On the other hand, if each letter in the abbreviation is pronounced as a letter, it is known as an [pb_glossary id="1748"]initialism[/pb_glossary] or [pb_glossary id="1749"]alphabetism[/pb_glossary]. Some examples include FBI > Federal Bureau of Investigation or UK > United Kingdom.

Did you know? The Internet slang initialism LOL is claimed to have originated in Canada! Wayne Pearson, an IT consultant from Calgary, claims to have coined it as a teenager on a chatroom for Calgarians in the 1980s.
When a word that already exists is claimed to be an acronym, this is known as a [pb_glossary id="1750"]backronym[/pb_glossary]. There has been a recent trend of backronyms going viral on social media, masquerading as the true origin of the word. For example, a false idea now circulating says that news stands for notable events weather and sports, when the term really comes from the word new from the early 15th century. It is best to check an etymological dictionary before clicking "share" on such posts.

Etymological dictionaries
People often guess the history of a word incorrectly, and [pb_glossary id="1740"]folk etymologies[/pb_glossary] which are not historically accurate often circulate in communities. Sometimes folk etymologies can even drive language change! For example, penthouse comes from the Old French word apentis meaning 'attached building.' Due to similarity in meaning and pronounciation with the word house, its spelling changed over time to become the modern penthouse. If you want to know the history of a word, you can look it up in an etymological dictionary. The best etymological dictionary for English is the Oxford English Dictionary (OED). The OED is paywalled, but you can likely access it online through your university’s library. Another good online etymological dictionary to explore is the Online Etymology Dictionary by Douglas Harper. Although it is not technically an academic source, this is a free and reliable etymological dictionary of English. There are many more etymological dictionaries and other resources out there that cover a wide variety of languages! Check your university library to see what you can find!

Key takeaways

New words are added to language in a variety of ways. Neologisms can come from other languages, which are called loanwords, they can be formed within the language itself, or they can be coined in a process called root creation.

Word formation processes that involve shortening an existing word include clipping and backformation.

Word formation processes that involve combining existing words including compounding, blending, initialisms, and acronyms.

The symbols < and > are used to show the direction of change in the study of historical linguistics. These symbols point at the newer form.

You can look up the origins of a word in an etymological dictionary, such as the Oxford English Dictionary (OED).

Check your understanding!
[h5p id="33"]
References and further resources

Attribution
↪️ This section was adapted from lecture notes prepared by Nathan Sanders. Used with permission.
For a general audience

🧠 CrashCourse. 2020. Language Change and Historical Linguistics: Crash Course Linguistics #13. https://www.youtube.com/watch?v=dNkMC92kFLA.

🎉 Hudes, Sammy. 2015, October 7. What's it like to coin the term LOL? Calgary Herald. https://calgaryherald.com/news/local-news/whats-it-like-to-coin-the-term-lol.

Reference materials

📑 Harper, Douglas. 2024. Online Etymology Dictionary. https://www.etymonline.com.

📑 Oxford University Press. 2023. Oxford English Dictionary. https://www.oed.com.

11.1 Hierarchical structure in compound words

juliedoner — Wed, 14 Feb 2024 05:53:56 +0000

As we saw in the previous chapter, some languages, including English, have a productive process through which new words can be formed that involves putting two words together. The resulting words are called [pb_glossary id="189"]compounds[/pb_glossary].

Table 1. Compounding in English

rice pot → rice-pot

swan boat → swan-boat

phone case → phone-case

This process is, theoretically, infinitely [pb_glossary id="187"]recursive[/pb_glossary], meaning that we can continue to make new words from existing words.

Table 2. More compounding in English

rice-pot rack → rice-pot-rack

swan-boat jacket → swan-boat-jacket

phone-case store → phone-case-store

Rice-pot has two morphemes, rice and pot. When we put them together, we make something whose meaning is the combination of both. This idea is represented in terms of [pb_glossary id="624"]bracketing structure[/pb_glossary].
[latex] \ex. \a. \textit{rice-pot} : [ [ rice ] [ pot ] ] \b.\textit{swan-boat} : [ [ swan ] [ boat ] ] [/latex]
Inside each pair of brackets is one meaning "unit." So rice-pot involves three pairs of brackets: one around rice, one around pot and one around the compound rice-pot. Same with swan-boat.

This way of representation allows us to iteratively represent more complex compounds, as in (2).
[latex] \setcounter{ExNo}{1} \ex. \a. \textit{rice-pot-rack} : [ [ [ rice ] [ pot ] ] [ rack ] ] \b. \textit{swan-boat-jacket} : [ [ [ swan ] [ boat ] ] [ jacket ] ] [/latex] When we build larger compounds, we are creating a meaningful unit, and then adding to it. So first I create the word rice-pot, and it has a meaning like, "a pot for rice." Then I create the word rice-pot-rack, which has the meaning "a rack for pots for rice." In other words, the meaning of the entire word depends on the meaning of its parts.
Of course, the more complex our compounds become, the more difficult it is to read the bracketing structure. So another way to express the exact same information is by using a tree, as in (3) and (4).
[latex] \setcounter{ExNo}{2} \ex. \begin{tikzpicture}[mytree] \node(top) {rice-pot} child {node {rice}} child {node {pot}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {rice-pot-rack} child {node {rice-pot} child {node {rice}} child {node {pot}} } child {node {rack}}; \end{tikzpicture} \ex. \begin{tikzpicture}[mytree] \node(top) {swan-boat} child {node {swan}} child {node {boat}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {swan-boat-jacket} child {node {swan-boat} child {node {swan}} child {node {boat}} } child {node {jacket}}; \end{tikzpicture} [/latex] Each [pb_glossary id="625"]node[/pb_glossary] in the tree corresponds to one pair of brackets. Thus, trees and brackets provide the exact same amount of information, it's just that trees do it in a visually more appealing way. But it's always possible to state a bracketing structure as a tree, and vice versa. I can actually make my representations slightly more informative by adding in category information. The following trees and bracketing structures give us all the preceding information, and they additionally tell us what the category of each morpheme is. [latex] \setcounter{ExNo}{4} \ex. \a. \begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ rice}} child {node {N \\ pot}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {N} child {node {N} child {node {N \\ rice}} child {node {N \\ pot}} } child {node {N \\ rack}}; \end{tikzpicture} \b. [$_N$ [$_N$ rice ] [$_N$ pot ] ] \hfill [$_N$ [$_N$ [$_N$ rice ] [$_N$ pot ] ] [$_N$ rack ] ] \ex. \a.\begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ swan}} child {node {N \\ boat}}; \end{tikzpicture} \hspace{100pt} \begin{tikzpicture}[mytree] \node(top2) {N} child {node {N} child {node {N \\ swan}} child {node {N \\ boat}} } child {node {jacket}}; \end{tikzpicture} \b. [$_N$ [$_N$ swan ] [$_N$ boat ] ] \hfill [$_N$ [$_N$ [$_N$ swan ] [$_N$ boat ] ] [$_N$ jacket ] ] [/latex] Note that it's not necessary to explicitly state the compound words that are formed at each junction, because I can just look lower in the tree to figure out what that word is. The reason it's good to list the category, though, is that you can combine more than one category. We therefore want to know which category "projects," that is, provides the part of speech to the resulting word. [latex] \setcounter{ExNo}{6} \ex. \textit{black-bird} : \begin{tikzpicture}[mytree] \node(top) {N} child {node {A \\ black}} child {node {N \\ bird}}; \end{tikzpicture} \hspace{50pt}[$_N$ [$_A$ black ] [$_N$ bird ] ] \ex. \textit{sweet-talk} : \begin{tikzpicture}[mytree] \node(top) {V} child {node {A \\ sweet}} child {node {V \\ talk}}; \end{tikzpicture} \hspace{50pt}[$_V$ [$_A$ sweet ] [$_V$ talk ] ] [/latex]
Hierarchical structure
These tree and bracketed diagrams illustrate one property of language: words and sentences have [pb_glossary id="2142"]hierarchical structure[/pb_glossary]. A hierarchical structure is a way of organizing elements. You might encounter hierachical structure in a workplace, where, for example, a worker might report to their team leader, the team leader might report to a department head, and the department head might report to the CEO. In morphology trees, though, instead of reporting to people higher up the chain of command, hierarchical structure represents containment relationships. The nodes higher up in the structure contain the nodes that branch off below them.
Structural ambiguity

What does the following compound mean?
[latex] \setcounter{ExNo}{8} \ex. baby-tree-planter [/latex]
Many of you will say that it means, "someone who plants baby trees." And many of you will say that it means "a tree-planter who is a baby." Both of these are right; it's possible to get both readings of the compound.
If we wanted to represent each meaning, we would choose different structures. The "baby-trees" meaning would have the following representations. [latex] \setcounter{ExNo}{9} \ex. ``A planter of baby trees'' : \begin{tikzpicture}[mytree] \node(top) {N} child {node {N} child {node {N \\ baby}} child {node {N \\ tree}} } child {node {N \\ planter}}; \end{tikzpicture} [/latex] The reason is that we want "baby tree" to be a unit of meaning, i.e., a [pb_glossary id="2190"]constituent[/pb_glossary], because that meaning describes the kind of planter it is. On the other hand, the "tree-planting-baby" meaning would have the following representation [latex] \setcounter{ExNo}{10} \ex. ``A tree-planter who is a baby'' : \begin{tikzpicture}[mytree] \node(top) {N} child {node {N \\ baby}} child {node {N} child {node {N \\ tree}} child {node {N \\ planter}}}; \end{tikzpicture} [/latex] The reason is that we want "tree-planter" to be a unit of meaning first, because ultimately we're describing a kind of tree-planter. This kind of ambiguity is called [pb_glossary id="627"]structural ambiguity[/pb_glossary], because we can represent the ambiguity structurally.
There is another kind of ambiguity called [pb_glossary id="2193"]lexical ambiguity[/pb_glossary]. The word bank is lexically ambiguous, because it means two different things (a financial institution and the side of a river). But it's not structurally ambiguous because the two meanings do not correspond to different structures.
The take-away point here is that different meanings correspond to different structures. More abstractly, the different meanings are the result of packaging the information in different ways.
Check yourself!
[h5p id="36"]
References and further resources

Attribution

This section is adapted from the following CC BY NC SA source:

↪️ Gluckman, John. n.d. Chapter 3: Brackets and trees. The science of syntax. https://pressbooks.pub/syntax/chapter/brackets-and-trees

1.5. Becoming a linguist: Myths and misconceptions about academic integrity

markc — Wed, 28 Feb 2024 21:50:45 +0000
While students might have an idea of what falls under academic integrity, some things are not always intuitive and so they may find themselves breaking rules by accident. Some common myths are…
Myths: Using resources

“I only need to cite people that I’m quoting.”
You need to cite any source that has information that you are using. This could be statistics, theoretical information, examples, or examples. When in doubt, cite it. If you are unsure how to cite something properly, ask your instructor or a librarian. By citing, not only are we attributing information to people who have contributed to the field, but we are also keeping track of how information has been developed.
“Artificial Intelligence is allowed because it’s like a spell checker.”
As of now, there is no consensus on the use of AI for assignments. Check the syllabus and ask your instructor before using it. For example, while some instructors may allow you to use AI in specific contexts for specific purposes, you should assume it is not allowed for graded work by default. Writing about a course topic in your own words helps you to recognize what you do or don’t understand, remember the material, and begin to think critically about it. Using AI to help you write runs counter to these goals.
"I can re-use a paper I wrote for another course in this course."
Submitting the same work to two different courses is considered self-plagiarism in many cases. It is considered an academic offence because you would receive credit twice for doing the work once. It is good to see points of connection between your different courses, but if you re-submit a paper you wrote for another course, your paper will not demonstrate the new material you learned in this course. You also lose the opportunity to go deeper into the topic and to continue to improve your writing skills. In some cases, instructors may allow or even encourage you to build on and develop a project you started for another course or assignment; however, you should also check with your instructor first. You need to be transparent about which work is re-used and which is new.
Myths: Working with classmates

“I can work on assignments with classmates.”
For the most part, assignments are meant to be done alone. Even collaborating and changing up your answers a bit counts as inappropriate collaboration. While you might encounter courses where collaboration on assignments is permitted, this is not the norm. Although working together can help you improve your learning in some cases, instructors also want to make sure you understand and are learning all of the material, not just some of it. Check your course syllabus for your individual course’s policies.
“If my groupmates are not doing the work, I have to carry the group project.”
On the contrary, writing the name of someone who has not contributed sufficiently on a piece of work could be a form of personation and thus a violation of academic integrity. Instead of trying to “carry” a group project, contact the instructor. The purpose of group projects is to learn how to collaborate with others on something larger and to learn from each other, which is facilitated through everyone putting in the work. In addition, it is not fair for your groupmate to get credit for an assignment they did not contribute to.
“I am only responsible for my own part of a group project.”
If you divide up a group project, you are still responsible for all of the parts. You should read over your groupmates' parts and make sure you understand them and can explain them. If a groupmate plagiarizes, the entire group may be held responsible.
“I can have someone else sign the attendance sheet or write the test for me.”
No. Claiming to be someone else is not permitted, especially when it involves forging a signature. If you have to miss a graded component of the course, contact the instructor as soon as possible to see what can be done. The instructor is likely required to make reasonable accommodations for many kinds of personal or medical emergencies, according to the policies of your university.
The final word
As a rule of thumb, when you are unsure of what is allowed and what is not, refer to the course syllabus, your institution’s academic integrity guidelines, and even your instructor if needed. Instructors would rather help you learn how to cite properly than try to bust you for failing to cite accurately by accident.
Things to do
While we often focus on what not to do, here are some things that you should do:
Read the syllabus
Read the syllabus and your institution’s academic integrity guidelines. There is a lot of information, but learning about them and understanding the general principles of academic integrity will help you. This is especially important if you are a student from another country because academic integrity rules vary around the world.
Pick appropriate groupmates
Form appropriate groups for projects. Put otherwise, recall that students have differing goals in a course. While some students just want to pass, others want an A+. In the event that A+ students and D students are in the same group, there is a greater chance of students accusing others of not pulling their weight or asking their groupmates to do too much work.
Set internal deadlines
Set internal deadlines for group projects. Similarly, by coming up with your own internal deadlines for a group project, you can be assured that everyone will have contributed sufficiently before the deadline is closing in, reducing the chance that people will perceive others as not helping.
Organize your notes
Keep your research notes organized. Clearly label your notes with their sources. Make sure your notes also clearly distinguish which parts are copied word-for-word and which are in your own words, for example, by using quotation marks. In some cases, an annotated bibliography might also be helpful. This applies to writing assignments the most, but keeping track of where you got your information and taking notes on it before you have started writing (or during the writing process) makes it easier to accurately cite sources and understand how theories and other ideas have developed.
Consider a citation manager
Consider using a citation manager. These tools can store all of your citations, allowing you to generate references from them. They can also be used as a database for sources that you have read in the past, allowing you to more quickly find the source of particular information if you need it again in the future.

Key takeaways

There are a variety of reasons to cite things, including your own work.

You must work alone unless you are given permission to work in a group. You should not put anyone's name on a group assignment if they did not contribute.

If you're unsure of what's allowed and what's not, read the syllabus and assignment instructions. If necessary, ask your instructor as well.

Check yourself!
[h5p id="37"]
References and further resources

🦬 University of Manitoba. Academic integrity. https://umanitoba.ca/student-supports/academic-supports/academic-integrity

🦬 University of Manitoba Faculty of Arts. Academic integrity. https://umanitoba.ca/faculties/arts/student_resources/student_responsibilities_integrity.html

🦬 University of Manitoba Libraries. Citation management software: Intro to citation managers. https://libguides.lib.umanitoba.ca/citationmanagers

6.2. Affixes

juliedoner — Sat, 11 May 2024 13:03:34 +0000
[pb_glossary id="2865"]Affixes[/pb_glossary] are bound morphemes that are attached to a stem, usually linearly. There are 5 kinds, categorized based on where they are attached: prefixes, suffixes, infixes, circumfixes, and suprafixes.
Prefixes
[pb_glossary id="688"]Prefixes[/pb_glossary] are attached to the beginning of a stem, as shown in (1). In (1a), the prefix re- is attached to the stem work to mean 'work again.' In (1b), the prefix un- is attached to the stem lock to mean 'reverse the action of locking.'

(1) a. re-work

b. un-lock

There can be more than one prefix in the same word. For example, in (2), the prefix un- is attached to the stem lock to mean 'reverse the action of locking,' and then the prefix re- is attached to the stem unlock to mean 'reverse the action of locking again.' These are both prefixes, since they are attached to the beginning of the stem, even if it is no longer at the beginning of the stem after other affixes have been added.

(2) re-un-lock

Prefixes are marked with a hyphen at the end, as with re- and un-, to show that they are bound and that they must be attached to a stem on their right edge.
Suffixes
[pb_glossary id="684"]Suffixes[/pb_glossary] are attached to the end of a stem, as shown in (3). In (3a), the suffix -ity is attached to the stem equal to mean 'the state of being equal'. In (3b), the suffix -s is attached to the stem cat to mean 'multiple cats.'

(3) a. equal-ity

b. cat-s

Just as with prefixes, there can be more than one suffix, as shown in (4). In (4), the suffix -ity is attached to the stem equal to form equality meaning 'the state of being equal,' and then the suffix -s is attached to the stem equality to mean 'multiple states of being equal.' These are both suffixes, since they are attached to the end of the stem, even if -ity is no longer at the end of the stem in the final form of the word. Note that -ity changes spelling to -itie when the plural morpheme -s is added, but it is the same morpheme (it is pronounced the same and it has the same meaning).

(4) equal-itie-s

Suffixes are marked with a hyphen at the beginning, as with -ity and -s, to show that they are a bound morpheme that attaches to the stem on its left side.
Infixes
[pb_glossary id="1846"]Infixes[/pb_glossary] are attached inside a stem, as shown in the Hoava data in (5). Hoava is an Austronesian language spoken in the Solomon Islands. In Hoava, the infix -in- is inserted before the first vowel of the stem, turning verbs and adjectives into nouns. In (5a)-(5d), this results in an infix. However, in vowel-initial words like (5e)-(5f), this looks like a prefix, because the position before the first vowel of the word happens to also be the edge of the word. We still call it an infix, though, because the overall pattern that can explain the data as a whole is an infixation pattern.

(5) Hoava

a. to 'alive' to 'life'

b. hiva 'want' hiva 'wishes'

c. va-bobe 'cause to be full' va-bobe 'filled object'

d. ta-poni 'be given' ta-poni 'gift'

e. asa 'grate' asa 'pudding of grated cassava'

f. edo 'happy' edo 'happiness'

(original source, Davis 2003, adapted from Kalin 2022: 644)
Note also that the infixation pattern applies to the stem, not the root. This means that the infix might sometimes appear inserted inside the root, as in (5a)-(5b), or might sometimes apper in an affix that has already been attached to the root, as in (5c)-(5d), where it appears inside the causative prefix and the passive prefix, respectively. Many students confuse infixes with prefixes or suffixes that are not at the word edge, like un- in (2) or -ity in (4). Infixes are fairly rare. If you think you have found an infix in your analysis of a language, you should look closely to see if you really can't analyze the infix as a a prefix or a suffix that is not at the word edge. If you're confused, think of a stack of books as a metaphor. Figure 1 shows a stack of five books. We can think of the middle book as the root and the books on top of it and under it as the prefixes and suffixes. The second book from the top is not at the top or the bottom of the pile, but it is also not inside of another book. It is in between two books. Likewise, prefixes and suffixes might be in between two affixes, instead of at a word edge, but they will not be inside of another affix. [caption id="attachment_2881" align="aligncenter" width="300"] Figure 1. A stack of five books. Photo by Julie Doner.[/caption] In Figure 2, on the other hand, I've added a sixth book, in the middle of the second book from the top. This book is inside another book. It is like an infix, which appears inside of another morpheme. Likewise, in (5a), the nominalizing morpheme -in- appears inside of the morpheme to 'alive' in the word tino 'life.' Neither the sequence /t/ nor /o/ on their own are morphemes. Instead, the morpheme to is interrupted. [caption id="attachment_2880" align="aligncenter" width="300"] Figure 2. A stack of books with one book inside of another book. Photo by Julie Doner.[/caption] In English, infixation only occurs with expletives, as in (6), or in language play, as with Homer Simpson style -ma- infixation in (7) or Hip Hop style -iz- infixation in (8). All three of these infixes occur before the primary stressed syllable of the word.

(6) a. important important

b. fantastic fantastic

c. Kalamazoo Kalamazoo

(McCarthy 1982, as cited in Yu 2007: 1)

(7) a. saxophone saxophone

b. violin violin

c. Michelangelo Michalangelo

(Yu 2004, as cited in Yu 2007: 1)

(8) a. house house

b. soldiers soldiers

c. ahead ahead

(Viau 2002, as cited in Yu 2007: 1)
When describing infixes, it is important to identify and describe the pattern of where they get inserted, as well as their form and meaning. To identify the pattern, you may need to refer to the edge of the word (either the beginning or end of the word) or to the syllable or prosodic structure of the word. Some examples of where an infix might be inserted include:

Before/after the first/last consonant of the stem

Before/after the first/last vowel of the stem

Before/after the first/last syllable of the stem

Before/after the primary stress of the stem

When mentioning an infix in isolation, it should be written with a hyphen on each side, as with -in-, -ma-, and -iz-, to show that it is a bound morpheme that is inserted inside of the stem. When a word containing an infix is segmented, as in examples (5)-(8), the infix should be marked with angle brackets <> on either side. This shows that it is an infix and not a string of suffixes or prefixes. Infixation is a little bit controversial. Some linguists, such as Kalin (2022), argue that infixes are really prefixes or suffixes that have been displaced.
Circumfixes
[pb_glossary id="2918"]Circumfixes[/pb_glossary] are affixes that appear in two pieces, on either side of the stem. As with infixes, circumfixes are rare. If you propose a circumfix, you need to have a compelling reason why it should be analyzed as one circumfix instead of two separate morphemes. The data in (9) illustrates a morphological pattern that can be analyzed as a circumfix. In Tagalog, an Austronesian language of the Philippines, the circumfix ka- -an can be affixed to a noun to mean 'a group of X', as shown in (9). Neither the first nor second part of the circumfix can appear without the other. In other words, there is no word kaintsik or intsikan. Instead, both morphemes must always appear together.

(9) Tagalog

a. intsik 'Chinese person' ka-intsik-an 'the Chinese'

b. pulo 'island' ka-pulu-an 'archipelago'

c. tagalog 'Tagalog person' ka-tagalog-an 'the Tagalogs'

(Lieber 2016: 88)
Circumfixes are written in the form X- -Y, where X is the part of the circumfix that goes at the beginning of the stem and Y is the part of the circumfix that goes at the end of the stem. This shows that it is a bound morpheme and that it is attached around the stem. Circumfixes can be considered a special instance of [pb_glossary id="2942"]parasynthesis[/pb_glossary], where one meaning is signaled simultaneously by more than one morphological process, since it is encoded simultaneously by both a prefix and a suffix.
Suprafixes
Finally, [pb_glossary id="2921"]suprafixes[/pb_glossary] are morphemes that are encoded in the [pb_glossary id="482"]prosodic[/pb_glossary] structure of a word. Sometimes morphological distinctions in prosody are analyzed as suprafixes and included as a kind of affix, or sometimes they are considered to be a form of [pb_glossary id="2928"]non-concatenative morphology[/pb_glossary], which will we discuss shortly in Section 6.4. Suprafixes can create contrast in meaning through pitch, length, or stress. An example of a tone suprafix is shown in (10) for the language Tsuut'ina, a Dene language of Alberta. In Tsuut'ina, a high tone is used to mark an incomplete action while a low tone is used to marked a complete action.

(10) incomplete action complete action gloss

a. -k'ós -k'òs 'to throw'

b. -tšúʔ -tšùʔ 'to seize'

(Cook 1971: 169)
An example of a length suprafix can be found in Estonian, a Uralic language spoken in Estonia, in Northern Europe, as shown in (11). The [pb_glossary id="2959"]genitive[/pb_glossary] [pb_glossary id="700"]case[/pb_glossary] is used to mark possession on nouns, while the [pb_glossary id="2960"]partitive[/pb_glossary] case is used to mark partialness, similar to the English word some. For some Estonian nouns, the only distinction between genitive case is the length of a syllable. In (11a), the genitive form of vodka viina, has a long vowel. In the partitive case, the syllable is lengthened, creating an overlong vowel, viiːna. In (11b), the long consonant in the genitive linna 'city' becomes overlong in the partitive linnːa. Finally, in (11c), the diphthong in the genitive laulu 'song' becomes long in the partitive lauːlu.

(11) genitive case partitive case gloss

a. viina viiːna 'vodka'

b. linna linːna 'city'

c. laulu lauːlu 'song'

(Spahr 2016: 152)
Some English words show a stress shift between noun-verb pairs, where the noun has initial stress and the verb has final stress, as shown in (12). This could be considered a kind of suprafix affecting stress.

(12) verb form verb meaning noun form noun meaning

a. re-CORD 'to convert sound or text into a permanent form' RE-cord 'a log of past happenings'

b. per-MIT 'to allow' PER-mit 'a document recording authorization'

c. re-BEL 'to engage in an act of defiance' RE-bel 'someone who regularly engages in acts of defiance'

d. pro-GRESS 'to move forward or onward' PRO-gress 'movement forwards or onwards'

e. pre-SENT 'to give a gift or award publicly' PRE-sent 'a gift'

Tonal and length suprafixes can be written with a hyphen preceding a placeholder V (for vowel) or σ (Greek letter sigma, for syllable) and tone or length markers, as shown in (13). There are many different markers used to describe tone, depending on preference and the number of tones in use in the language in question; any of these notations can be used, but be sure to explain your notation if you are not using a well-known system. (For an overview of notation systems for tone, see Section 3.12 of Essentials of Linguistics, 2nd edition). Suprafixes are usually marked with a preceding hyphen, similar to suffixes, to show that it is a bound morpheme and because suffixes are more common cross-linguistically than prefixes. Similar to infixes, this notation should be supplemented with a description of which vowel or syllable it applies to. There is not really a good notation to use for suprafixes affecting stress; these patterns should be described in words.

(13) notation meaning

a. -V˦ a vowel with a mid-high tone

b. -σ́ a syllable with a high tone

c. -Vː a long vowel

Key takeaways

Prefixes attach to the left of the stem, suffixes attach to the right of the stem, infixes attach internally to the stem, circumfixes attach to either side of the stem, and suprafixes are encoded in the prosody of the stem.

Infixes and circumfixes are rare. If you think you have found an infix or a circumfix, double check that a simpler analysis, using prefixes and suffixes instead, is not possible.

Affixes are marked with hyphens to show which side of the stem they attach to.

Check yourself!
[h5p id="39"] Consult the following data in answering the check yourself questions:

(14) Juliesh (conlang)

a. solabi 'I walk'

b. sobok 'I swim'

c. filabi 'we walk'

d. fobok 'we swim'

e. sobilabi 'I walked'

f. fibibok 'we swam'

References and further resources

For students

📑 Anderson, Catherine, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi. 2022. Section 3.12: Tone and intonation. Essentials of Linguistics, 2nd edition. Pressbooks. https://ecampusontario.pressbooks.pub/essentialsoflinguistics2/chapter/3-12-tone-and-intonation/
Lieber, Rochelle. 2016. Introducing morphology. 2nd edition. Cambridge: Cambridge UP.
Academic sources

Cook, Eung-Do. 1971. Vowels and tones in Sarcee. Language 47 (1): 164-179.

🔍 Kalin, Laura. 2022. Infixes really are (underlyingly) prefixes/suffixes: Evidence from allomorphy on the fine timing of infixation. Language 98 (4): 641-682.

Spahr, Christopher. 2016. Contrastive representations in non-segmental phonology. Doctoral dissertation, University of Toronto.

Yu, Alan. 2007. A natural history of infixation. Oxford: Oxford UP.

6.1. The building blocks of morphology

juliedoner — Sat, 11 May 2024 13:03:54 +0000
A word can be broken up into several [pb_glossary id="161"]morphemes[/pb_glossary], which, as we learned in Section 1.1, are the smallest meaningful units of language. There is more than one kind of morpheme. In this section, we will learn about roots and affixes and about free and bound morphemes.
Roots, stems, and affixes
A [pb_glossary id="3076"]root[/pb_glossary] is the core morpheme of a word, while [pb_glossary id="3077"]affixes[/pb_glossary] are the morphemes that are attached to it. Some examples are shown in (1).

(1) Word Root Affixes

a. librarian library -an

b. unthinkable think un-, -able

c. workers work -er, -s

d. dismissiveness miss dis-, -ive, -ness

e. declassifiable class de-, -ify, -able

Words are built by starting with the root and adding one morpheme at a time. The piece of the word that hosts an affix at each stage is called the [pb_glossary id="3078"]stem[/pb_glossary] or the [pb_glossary id="3079"]base[/pb_glossary]. Sometimes students mix up roots and stems. A root is always a single morpheme, but a stem may be complex, consisting of multiple morphemes. The root of a word does not change as you add more affixes, but the stem is different for each affix. Let's consider the word declassifiable as an example, which contains four morphemes: de-, class, -ify, and -able, with the hierarchical structure shown in (2).

(2) a. [latexpage] \begin{tikzpicture} [+preamble] \usepackage{forest} [/preamble] \begin{forest} for tree={parent anchor=south,child anchor=north} [Adj [V [de-] [V [N [class] ] [-ify] ]] [-able] ] \end{forest} \end{tikzpicture}

(2) b. [Adj [V [de-] [V [N[class] ] [-ify] ] ] [-able] ]

The root of the word declassifiable is class. The first affix that is attached to the root is -ify. The stem of -ify is the root class. Next, de- is attached; the stem of de- is the complex form classify. Finally, -able is affixed; the stem of -able is the complex form declassify. As you can see, complex words like declassifiable are built from the inside out, starting with the root. Affixes are added to either edge of the root, one at a time. We will discuss more about how to determine which order affixes are attached in Chapter 11.
Free and bound morphemes
Another way we can classify morphemes is based on whether they are [pb_glossary id="3080"]free[/pb_glossary] or [pb_glossary id="3081"]bound[/pb_glossary]. A free morpheme can stand on its own, while a bound morpheme must always be attached to another morpheme. Some examples of free morphemes in English include library, think, blue, class, and couch. Some examples of bound morphemes in English include -s, -ify, -ed, in-, dis- and -ish. We use hyphens to show whether a morpheme is free or bound. Free morphemes are not marked with hyphens while bound morphemes are marked with hyphens. The hyphens are always placed on the side of the affix that attaches to the stem. Affixes should always have a hyphen. In some languages, like English, most roots are free and do not need a hyphen. In other languages, most roots are bound, and do need a hyphen. For example, in Spanish, most nouns are marked with a gender suffix. Since the roots cannot occur on their own without the gender suffix, these would be considered bound roots. Some examples, shown in (3), are the feminine noun casa 'house' which consists of the bound root cas- and the feminine suffix -a or the masculine noun cerezo 'cherry' which consists of the bound root cerez- and the masculine suffix -o.

(3) a. Spanish

cas -a

house -F

'house'

(3) b. cerez -o

cherry -M

'cherry'

Key takeaways

Roots are the core morpheme of a word. Affixes are morphemes that are attached to the root.

The piece of a word that hosts an affix is called a stem or base.

Morphemes that can stand on their own are called free morphemes and morphemes that cannot stand on their own are called bound morphemes.

Check yourself!
[h5p id="40"]

6.4. Reduplication and non-concatenative morphology

juliedoner — Sat, 11 May 2024 13:04:16 +0000

Reduplication
[pb_glossary id="3757"]Reduplication[/pb_glossary] is a special morphological process that involves full or partial copying of the stem. Some examples of full reduplication in Javanese are shown in (4). In all of these examples, reduplication indicates plurality.

(4) bare form bare form gloss reduplicated form reduplicated form gloss

a. baita ship baita~baita various ships

b. səsupe ring səsupe~səsupe various rings

c. omaha house omaha~omaha various houses

(Uhlenbeck 1978: 90, as cited by Booij 2005: 35)
Javanese also has partial reduplication. Whereas the reduplicated forms in (4) repeat the entire stem, the reduplicated forms in (5) only reduplicate the initial consonant. Note that the vowel that has been added in the reduplicated forms is not copied from the stem; in all cases, it is a /ə/, regardless of the vowels of the stem. In all of the examples in (5), reduplication indicates a change from a noun into a verb.

(5) bare form bare form gloss reduplicated form reduplicated form gloss

a. gəni fire g~ə-gəni to warm oneself by the fire

b. jawah rain j~ə-jawah to play in the rain

c. tamu guest t~ə-tamu to visit

(Uhlenbeck 1978: 90, as cited by Booij 2005: 35)
Javanese uses many different reduplicative patterns with a variety of meanings. For a thorough description of many of the ways reduplication is used in Javanese, see Miyake (2011). According to the Leipzig glossing rules, reduplication should be marked with a tilde (~) connecting the reduplicant to the stem.
Non-concatenative morphology

Affixes, for the most part, join the root on its edges, resulting in a linear string of morphemes, but not all morphological processes result in a linear string of morphemes. Morphological processes that produce affixes in a linear string are called [pb_glossary id="3677"]concatenative[/pb_glossary], from the Latin word catena 'chain', while morphological processes that do not result in a linear string are called [pb_glossary id="3680"]non-concatenative[/pb_glossary]. We have already seen some examples of non-concatenative morphology; [pb_glossary id="2921"]suprafixes[/pb_glossary], which modify the [pb_glossary id="482"]prosody[/pb_glossary] of a word, are one type of non-concatenative morphology.
Root-and-pattern morphology
A well-known example of non-concatenative morphology is [pb_glossary id="3706"]root-and-pattern morphology[/pb_glossary], which is common in the Semitic language family. In Semitic languages, many roots consist only of three consonants. These are known as [pb_glossary id="3707"]triconsonantal roots[/pb_glossary]. Different sets of vowels are interwoven between the consonants in specific patterns, creating different but related word forms. These vowel patterns are sometimes called [pb_glossary id="3708"]transfixes[/pb_glossary], because they are inserted across the root. An example from Hebrew is shown below in (1).

(1) Transfix pattern Pattern meaning Hebrew verb Gloss

a. CaCaC intransitive verb gadal He grew.

b. CiCeC to cause X gidel He raised.

c. CuCaC to be caused to be X gudal He was raised.

d. hiCCiC to make X higdil He enlarged.

e. huCCaC to be made to be X hugdal He was enlarged.

(adapted from Ussishkin 2000: 659)

Example (1) shows several Hebrew words formed from the triconsonantal root gdl. In the first column, we see several different transfixes that can be applied to the root. Each transfix pattern includes three Cs which represent the placement of the root consonants. The transfix pattern also indicates which vowels are interwoven into the word and where, as well as if there are any additional consonants, such as the initial /h/ in examples (d) and (e). Triconsonantal roots are bound roots, because they cannot appear on their own, without the vowels of a transfix.

Ablaut

[pb_glossary id="3499"]Ablaut[/pb_glossary], also called [pb_glossary id="3730"]internal change[/pb_glossary], is when a morphological process is marked by changing the vowel of the stem. Ablaut is used in many Germanic languages, including English. Ablaut is similar to root-and-pattern morphology, except it only affects one vowel.

Some examples of ablaut in the formation of the English plural are shown in (2). As you can see, there are many different vowels that may be affected by ablaut in English. For example, /oʊ/ changes to /aɪ/ in the plural of mouse in (2a), and /u/ changes to /i/ in the plural of goose in (2b) or tooth in (2c), and so forth.

(2) singular plural

a. mouse mice

b. goose geese

c. tooth teeth

d. woman women

Ablaut also shows up in English verbs. As shown in (3), there are two forms in English that may have ablaut compared to the [pb_glossary id="3738"]bare form[/pb_glossary]: the past tense and the past participle. The [pb_glossary id="3731"]past participle[/pb_glossary] is used in the English perfect (have sung, have come, etc.) or the passive (was sung, was found). Some verbs, like sing in (3a), have a different vowel for each of the three forms. Others use the same vowel for the bare form and the past participle but a different vowel for the past tense, such as come in (3b). Others have the same vowel for the past tense and past participle but a different vowel for the bare form, such as find in (3c) or get in (3d). Finally, drive in (3e) has a different vowel in all three forms—the bare form and past participle may be spelled the same, but they're pronounced with different vowels—but the past participle additionally has the regular past participle suffix.

(3) bare form past tense past participle

a. sing sang sung

b. come came come

c. find found found

d. get got got

e. drive /dɹɑiv/ drove /dɹoʊv/ drive-n /dɹɪvən/

Key takeaways

Reduplication is a morphological process whereby the stem is copied. The entire stem can be copied, which is called full reduplication, or just part of the stem, which is called partial reduplication.

Non-concatenative morphology are morphological processes that do not result in a linear string. Suprafixes, root-and-pattern morphology, and ablaut are all examples of non-concatenative morphology.

Root-and-pattern morphology is common in Semitic languages and consists of triconsonantal roots that are combined with a transfix. Transfixes indicate which vowels should be used and the arrangement of the consonants.

Ablaut is when a morphological category is marked by a vowel change.

Check yourself!
[h5p id="45"] Consult the following data in answering the check yourself questions:

(4) Javanese

a. abang 'red'

b. abang-abang 'some what red, reddish'

c. biru 'blue'

d. biru-biruan 'bluish'

(Miyake 2011: 50)

(5) Morrocan Arabic

a. ktəb 'write'

b. ktab 'book'

c. ktuba 'books'

d. katib 'writer'

(Noamane 2018)

(6) a. sing-song

b. ding-dong

c. chat-chat

d. dilly-dally

References and further resources

For a general audience

🎉 McLendon, Lisa. November 9 2020. Why ablaut reduplication is top-top. ACES: The society for editing. https://aceseditors.org/news/2020/why-ablaut-reduplication-is-tip-top

For students

Booij, Geert. 2005. The grammar of words: An introduction to linguistic morphology. Oxford: Oxford University Press.

Academic sources

🔍 Miyake, Yoshimi. 2011. Reduplication in Javanese. Asian and African Languages and Linguistics 6: 45-59.

Noamane, Ayoub. 2018. Morphological causatives in Moroccan Arabic: Word-based or root-based? Asinag 13: 217-240.

Uhlenbeck, E. M. 1978. Studies in Javanese Morphology. The Hague: Martinus Nijhoff.

Ussishkin, Adam. 2000. Root-and-pattern morphology without roots or patterns. Proceedings of the North East Linguistics Society 30(2).

6.7. Clitics

juliedoner — Sat, 11 May 2024 13:04:59 +0000
A clitic is a morpheme which has some of the properties of an independent word and some properties of an affix. The word clitic comes from the Latin word for leaning and is related to words like incline. You can think of a clitic as a morpheme that leans on another word, which is called the host. Clitics depend on a word that they are adjacent to. An enclitic leans on the word that comes before it, and so looks kind of like a suffix, while a proclitic leans on the word that comes after it, so looks kind of like a prefix. According to the Leipzig glossing rules, you may indicate that a morpheme is a clitic by using an equal sign (=) instead of a hyphen (-), but this is an optional rule. There are lots of different kinds of clitics. What unites them all is that they don't behave quite like a word, and they don't behave quite like an affix either. In this section, we will look at some examples of some of these different kinds of clitics.
Bound morphemes which modify an entire phrase
One group of morphemes that can be categorized as clitics are bound morphemes which modify an entire phrase. One example of this kind of clitic is [pb_glossary id="2959"]genitive[/pb_glossary] 's in English. As shown in example (1), genitive 's indicates possession. For example, in (1a), 's attaches to Sharon and indicates that the house belongs to Sharon.

(1) a. Sharon's house

b. [My little sister]'s house

c. [My little sister with the green shirt]'s house

d. [My little sister with the green shirt and the mischievous grin]'s house

e. [My little sister that ran a marathon]'s house

Like an affix, 's cannot stand on its own—it is a bound morpheme. However, unlike an affix, it doesn't modify just the word it attaches to; it modifies an entire phrase. This is shown in examples (1c)-(1e). In these three examples, 's attaches to the words shirt, grin, and marathon, respectively. However, these phrases do not mean that the house belongs to the shirt, the grin, or the marathon. In all three sentences, the house belongs to the sister, since sister is the head of the phrase. In other words, 's modifies the entire phrase it attaches to, not just the one word it attaches to.
Bound morphemes which attach to a word they do not modify
Another category of clitics are bound morphemes that attach to a word they do not modify. This is the behaviour of determiners in the language Nisg̱a'a. In Nisg̱a'a, determiners cliticize to the word that appears before them, even if they don't belong to the same phrase. This is shown below in (2). The noun Mary takes the [pb_glossary id="3949"]proper noun[/pb_glossary] determiner =t, which encliticizes to the verb giba before Mary. Likewise, the noun hlgiikw 'sister' takes the [pb_glossary id="3950"]common noun[/pb_glossary] determiner =hl, which encliticizes to the word Mary before it.

(2) Nisg̱a’a

Yukwt gibas Maryhl hlgiikwt.

yukw =t giba -t [=t Mary] [=hl hlgiikw -t]

IPFV =3 wait -3 [=PN Mary] [=CB sister -3]

'Mary is waiting for her sister.'

(Tarpent 1987: 211-212, as cited in Forbes 2024)

Morphemes that occur in a different location than expected
Some morphemes are classified as clitics simply because they appear in a different location than expected. Object pronouns in Romance languages are often called clitics for this reason. Consider the French example below. The typical word order of French is subject-verb-object, as shown in (3a). The object, which is the entity that the event happens to, in this case la fille 'the girl,' appears after the verb when it is a full noun phrase. However, when the object is a pronoun, as in (3b), it appears before the verb instead.

(3) a. French

(subject verb object)

Pierre voit la fille.

Pierre see.3SG the girl

'Pierre sees the girl.'

(3) b. (subject object verb)

Pierre la voit.

Pierre her see.3SG

'Pierre sees her.'

Clitic pronouns in Romance languages have some very unique and well-known properties. For example, some Romance languages have a construction called clitic doubling where a noun phrase appears twice—both as a full noun phrase and as a clitic pronoun, as shown in (4). In this sentence, the indirect object is marked twice: once right before the verb as the pronoun le and once at the end of the sentence as the full noun phrase a Juan.

(4) Spanish

Carolina le dio un libro a Juan.

Carolina DAT3SG gave a book DAT Juan

'Carolina gave a book to Juan.'

(Belloro 2007: 6)
Another unique property of clitic pronouns in some Romance languages is that they can undergo a process of clitic climbing. Clitic climbing is when a clitic moves from an embedded clause to the main clause, as shown in example (5). In the Catalan example in (5), the 2nd person singular clitic pronoun t' semantically belongs in the embedded clause with the verb veure 'see'. However, it shows up in the main clause attached to he vingut 'have come,' even though it has no semantic association with the event of coming.

(5) Catalan

T'-he vingut a veure.

you- have.1SG come to see.INF

'I have come to see you.'

(Solà 2002: 228)
Another special kind of clitic that appears in an unexpected position are called second-position clitics. These clitics always appear after the first word in their phrase, no matter what the word is. Some examples from Serbo-Croatian are shown in (6). In these examples, there is a string of three second-position clitics, smo 'are', mu 'him', and je 'her.' When there is a cluster of clitics like this, it is known as a clitic group. In sentence (6a), the clitic group appears after the subject pronoun mi 'we'. In (6b), the subject is dropped, so the clitic group moves to appear after the verb predstavili 'introduced'. Sentence (6c) is a question, so the clitic group appears after the question word zašto 'why'. Regardless of what the first word of the sentence is, the clitic group always appears immediately after it, in second position.

(6) a. Serbo-Croatian

Mi smo mu je predstavili juče.

we are him.dat her.acc introduced yesterday

'We introduced her to him yesterday.'

(6) b. Predstavili smo mu je juče.

introduced are him.dat her.acc yesterday

'We introduced her to him yesterday.'

(6) c. Zašto smo mu je predstavili juče?

why are him.dat her.acc introduced yesterday

'Why did we introduce her to him yesterday?'

(Bošković 2016: 28)

Words that cannot bear stress
Another reason a word might be considered a clitic is if it is prosodically deficient. Generally, this means that it does not usually bear stress. Many functional words, including English words such as the, a, and of might be considered clitics for these reasons.

Key takeaways

Clitics are bound morphemes that have some properties of an independent word and some properties of an affix.

Clitics that attach to the beginning of their host are called proclitics and clitics that attach to the end of their host are called enclitics.

Some unusual properties of clitics include that they might appear in unexpected positions, they might attach to a word that they do not modify, or they might be a bound morpheme that modifies an entire phrase.

Check yourself!
[h5p id="48"] Consult the following data in answering the check yourself questions:

(7) a. Comanche

tɨasi-se nɨ tɨhka

again-DM I eat

'Again I ate.'

(7) b. *nɨ tɨhka

I eat

'I eat.'

(7) c. tɨhka nɨ

eat I

'I eat.'

(McDaniels 2008, as cited in Bošković 2016: 35)

References and further resources

Academic sources

Belloro, Valeria. 2007. Spanish clitic doubling: A study of the syntax-pragmatics interface. PhD thesis, State University of New York at Buffalo.

Bonet, Eulalia. 2019. Clitics and clitic clusters in morphology. Oxford Research Encyclopedia of Linguistics. https://oxfordre.com/linguistics/display/10.1093/acrefore/9780199384655.001.0001/acrefore-9780199384655-e-519

Bošković, Željko. 2016. On second position clitics crosslinguistically. In Formal studies in Slovenian syntax: In honor of Janez Orešnik, ed. Franc Lanko Marušič and Rok Žaucer. Amsterdam: John Bemjamins. 23-53.

Forbes, Clarissa. 2024. Tsimshianic. In The Languages and linguistics of Indigenous North America: A comprehensive guide, Volume 2, ed. Carmen Dagostino, Marianne Mithun, and Keren Rice. Walter de Gruyter. 985-1012.

Solà, Jaume. 2002. Clitic climbing and null subject languages. Catalan Journal of Linguistics 1: 225-255.

6.3. Inflection and derivation

juliedoner — Sat, 11 May 2024 13:16:34 +0000
Another way we can classify affixes is whether they are [pb_glossary id="1723"]inflectional[/pb_glossary] or [pb_glossary id="1724"]derivational[/pb_glossary]. In this section, we will spend just a little bit of time on the difference between inflection and derivation to get the basic idea. We will go much deeper into their differences in Chapter 9.
Inflection
Inflectional morphemes encode the grammatical properties of a word. Some common examples of inflectional morphemes include [pb_glossary id="660"]plural[/pb_glossary] markers on nouns, as shown in (1); [pb_glossary id="3550"]tense[/pb_glossary] and [pb_glossary id="3551"]aspect[/pb_glossary] markers on verbs, as shown in (2); or [pb_glossary id="3489"]comparative[/pb_glossary] and [pb_glossary id="3496"]superlative[/pb_glossary] markers on adjectives, as shown in (3).

(1) singular noun plural noun

a. cat cat-s

b. chair chair-s

c. bush bush-es

(2) bare verb past tense past participle present participle

a. walk walk-ed walk-ed walk-ing

b. sew sew-ed sew-n sew-ing

c. sleep slep-t slep-t sleep-ing

(3) bare adjective comparative superlative

a. red redd-er redd-est

b. long long-er long-est

c. happy happi-er happi-est

Generally speaking, we don't consider inflectional forms of the same stem to be different words, but to be different forms of the same word. The collection of all of the inflectional forms of a root are called the [pb_glossary id="1684"]paradigm[/pb_glossary] for that word. For example, the row (1a) is the paradigm for the noun cat, row (2a) is the paradigm for the verb walk, and row (3a) is the paradigm for the adjective red. A word without any inflectional morphemes is often called the [pb_glossary id="3738"]bare form[/pb_glossary]. In examples (1)-(3), the bare form of each paradigm is in the first column. The bare form may be a root or it may be a complex form containing both a root and derivational morphemes. We can also talk about an inflected word form as having [pb_glossary id="634"]morphosyntactic features[/pb_glossary]. For example, cat would have the feature [+singular] and cats would have the feature [+plural]. Morphosyntactic features classify words into groups based on similar behaviour. For example, English words with the feature [+plural] share the behaviour of being marked with a plural marker and having the meaning of 'more than one'.
Derivation

In contrast, derivational morphemes create words of new [pb_glossary id="616"]parts of speech[/pb_glossary] or words with new meanings. For example, the verb scare can be changed into a new word, the adjective scary, by adding the derivational morpheme -y. Although scare and scary share the same root scare, they are different words. The verb scare refers to an event and belongs to a verbal paradigm and can take verbal inflectional morphology, such as scared and scaring, while the adjective scary refers to a quality and belongs to an adjectival paradigm and can take adjectival morphology, such as scarier and scariest. Derivational morphemes don't always change the part of speech of the word, though. Morphemes that change the meaning of the word enough that we want to call it a new word, even if it keeps the same part of speech, are still called derivational morphemes. For example, the prefix re-, attaches to verbs to form new verbs that mean to perform the action again, as in (4). Both scared and rescared are verbs.

(4) a. My sister scared me.

b. My sister scared me again. She rescared me.

Sometimes derivational morphemes do both; they change the meaning substantially and change the part of speech. For example, adding the suffix -er to a verb creates a noun that identifies the person who performed the action, known as an agentive noun, or adding -able to a verb creates an adjective meaning that the action is possible to perform. Derivational word forms based on the same root belong to the same [pb_glossary id="1685"]word family[/pb_glossary], but each has their own, separate, inflectional paradigm. For example, the word family of scare includes the verb scare (5a), the adjective scary (5b), the noun scare (5c), and the adverb scarily (5d). It also includes more complex forms such as the repetitive verb rescare (5e), the agentive noun scarer (5f), and the adjective scareable which expresses possibility (5g). The word family even includes words containing multiple derivational morphemes, such as the word scariness in (5h), which consists of three morphemes: the verb root scare, the adjective-forming suffix -y (spelled with an when it is not at the end of the word), and the noun-forming suffix -ness.

(5) a. My sister scared me. verb

b. the scary costume adjective

c. a big scare noun

d. The ghost moved scarily. adverb

e. After I calmed down, my sister rescared me. repetitive verb

f. the scarers at the haunted house agentive noun

g. My husband is easily scareable. adjective of possibility

h. the scariness of this costume noun derived from the adjective

While it is often possible to list the complete paradigm for a word, it is not possible to list the complete word family of a root, since we can build infinitely complex words out of one root.

Key takeaways

Inflectional morphemes encode the grammatical properties of a word.

The list of the different inflectional forms of a word is called a paradigm.

We can formally indicate the inflectional properties of a word using morphosyntactic features.

Derivational morphemes create new words by changing the part of speech of a word, substantially changing its meaning, or both.

The list of different words derived from the same root is called a word family.

Check yourself!
[h5p id="44"]

6.6. Allomorphy

juliedoner — Sat, 11 May 2024 13:24:53 +0000
Sometimes a morpheme changes form in different contexts. The different forms of a morpheme are called [pb_glossary id="3178"]allomorphs[/pb_glossary]. For example, the plural allomorph of English takes many different forms, as shown in (1). These include predictable variation, such as variation between [pb_glossary id="3179"]voiceless[/pb_glossary] /s/ after voiceless consonants as in cats in (1a), [pb_glossary id="3180"]voiced[/pb_glossary] /z/ after voiced consonants as in dogs in (1b), and epenthetic /əz/ after [pb_glossary id="3181"]sibilants[/pb_glossary] as in bushes in (1c). The English plural form also has unpredictable, or irregular, variation, including vowel changes as in foot versus feet in (1d), irregular suffixes such as ox versus oxen in (1e), and plural forms borrowed from other languages as with cactus versus cacti in (1f). Sometimes multiple strategies combine as with child versus children in (1g), which exhibits both an irregular suffix -ren and a vowel change from /aɪ/ to /ɪ/. Finally, sometimes the plural form is the same as the singular form, as with sheep. These are all examples of plural allomorphs in English.

(1) singular singular pronunciation plural plural pronunciation

a. cat /kæt/ cat-s /kæts/

b. dog /dɑg/ dog-s /dɑgz/

c. bush /bʊʃ/ bush-es /bʊʃəz/

d. foot /fʊt/ feet /fit/

e. ox /ɑks/ ox-en /ɑksən/

f. cactus /kæktɪs/ cacti /kæktaɪ/

g. child /t͡ʃaɪld/ child-ren /t͡ʃɪldɹən/

h. sheep /ʃip/ sheep /ʃip/

We can classify allomorphs based on two properties. First, we consider the relationship between the different forms of the morpheme. If the different allomorphs are related to each other through a regular phonological pattern, they are called [pb_glossary id="3481"]phonological allomorphs[/pb_glossary]. If they are not, they are called [pb_glossary id="704"]suppletive[/pb_glossary] allomorphs. Second, we consider the contexts that trigger allomorphy. Allomorphy can be triggered by [pb_glossary id="3521"]phonological[/pb_glossary], [pb_glossary id="3522"]morphological[/pb_glossary], or [pb_glossary id="3523"]lexical conditioning[/pb_glossary].
Phonological allomorphs
When the form of an allomorph is predictable based on phonological context, it is called a [pb_glossary id="3481"]phonological allomorph[/pb_glossary]. In other words, phonological allomorphs are in complementary distribution based on a predictable phonological context. It is possible to write a morphophonological rule to describe where each allomorph is used. For example, the plural allomorphs in (1a)-(1c) are phonological allomorphs. The plural morpheme surfaces as /əz/ after [pb_glossary id="3181"]sibilants[/pb_glossary], as /s/ after other voiceless consonants, and as /z/ elsewhere. We can write two morphophonological rules which, taken together, describe this pattern, as in (2).

(2) a. plural -z → -əz / sibilant _

b. plural -z →-s / voiceless C _

We can also write a distribution diagram, as in (3).

(3) -əz / sibilant _

plural -s / voiceless C _

-z / elsewhere

Suppletion
On the other hand, [pb_glossary id="3486"]suppletive[/pb_glossary] allomorphs are not predictable from phonological context. [pb_glossary id="3487"]Weak suppletion[/pb_glossary] has some phonological similarity to the other allomorphs. Some example of weak suppletion can be identified in English past tense forms, shown in (4).

(4) present present pronunciation past past pronunciation

a. buy /baɪ/ bought /bɑt/

b. catch /kɛt͡ʃ/ cought /kɑt/

c. teach /tit͡ʃ/ taught /tɑt/

The present tense forms buy, catch, and teach have nothing in common with each other that would explain why they do not take the regular -ed past tense marker, which is why these past tense forms are considered suppletive and not phonological. However, each present tense and past tense pair share the same initial consonant—/b/ for buy and bought, /k/ for catch and caught, and /t/ for teach and taught—which is why these are considered weak suppletion. In contrast, in [pb_glossary id="3488"]strong suppletion[/pb_glossary], the allomorphs do not share any phonological similarity. For example, as shown in (5a), the root for the verb go and its past tense went have no phonological similarity. Likewise, shown in (5b), the root good has no phonological similarity with the [pb_glossary id="3489"]comparative[/pb_glossary] form better or the [pb_glossary id="3496"]superlative[/pb_glossary] form best. Therefore, good exhibits strong suppletion. On the other hand, the relationship between better and best could be considered weak suppletion, because they share the initial /bɛ/.

(5) a. bare root go /goʊ/

a. past tense went /wɪnt/

b. root good /gʊd/

b. comparative better /bɛɾəɹ/

b. superlative best /bɛst/

Where does suppletion come from?

Historically, suppletion has many different possible sources. One common source of suppletion is when a [pb_glossary id="164"]productive[/pb_glossary], historical process has died out, leaving only a few remnants in the modern grammar. This is the source of the plural form feet. Old English nouns followed many different conjugation patterns, much like Latin. Some, including foot, had a regular [pb_glossary id="3499"]ablaut[/pb_glossary] pattern, where plurality was marked by a vowel change in the stem, between the singular fot and the plural fet (OED). Eventually, though, the plurals in English went through a process of [pb_glossary id="3500"]regularization[/pb_glossary], until today, where the majority of words take an -s plural marker. However, some of the more frequent words, including foot, kept the ablaut plural form. Another common source of suppletion is through [pb_glossary id="3501"]analogy[/pb_glossary], when a word adopts a similar pattern to another word. For example the word catch is borrowed from French cacher, meaning to hunt (OED), so at first it may be surprising that it has a suppletive past tense that looks like it may have been left behind by an older ablaut pattern of Old English. [pb_glossary id="3502"]Etymologists[/pb_glossary] think the past tense caught came from analogy from other words ending in -t͡ʃ which also have irregular pasts, such as latch and teach (OED). Another source of suppletion is when two [pb_glossary id="1684"]paradigms[/pb_glossary] merge. For example, the past tense of GO, went, comes from a different verb, wend, meaning to turn or change direction. It has been used as the past tense of GO since the 15th century, replacing the old past tense forms of GO, which were also suppletive (OED). In general, suppletion is much more common in frequent words, since children need to be exposed to the irregular pattern frequently enough to learn it. You can look up the historical development of English words in the Oxford English Dictionary, which can likely be accessed online for free through your university library. Search for the word you're interested in, and then click the Etymology tab to learn about its history.

Conditioning
The choice of which allomorph is used where can depend on different factors: it can be phonologically, morphologically, or lexically conditioned. When allomorphy is phonologically conditioned, it means the choice of allomorph depends on phonological context. For example, the regular forms of the plural allomorph are phonologically conditioned. The plural morpheme –s appears as [əz] after a sibilant, [s] after a voiceless consonant, and [z] elsewhere. Morphological conditioning, on the other hand, depends on the morphological context, usually the grammatical properties of the morpheme, such as an inflectional category. For example, in Spanish, the verb meaning ‘to go’ uses the stem ir in the infinitive and future, va- in the present and imperfective past, and fu- in the perfective past. This is morphological conditioning because it depends on morphological features such as present, infinitive, or future. Finally, lexical conditioning depends on the properties of the stem. For example, the Persian plural marker appears as –an for human nouns and –ha for non-human nouns, which is an example of lexical conditioning based on semantic properties. Lexical conditioning also includes any form that is not predictable (i.e., must be memorized). The irregular plural -ren in children is an example of lexical conditioning that is not predictable. There is nothing about the word child that predicts that it would have an irregular plural.

Spelling changes vs. allomorphy

One mistake that students sometimes make is misidentifying a spelling change as allomorphy. Allomorphy describes only changes in pronunciation, not changes in spelling. Sometimes a change in spelling and allomorphy co-occur, as in (6). In (6a), the morpheme invade in its bare form ends with a /d/ sound and is spelled with a /d/. In (6b), when -ion has been suffixed, the pronunciation of the final segment of the stem changes from /d/ to /ʒ/.

(6) a. invade /ɪnveɪd/

b. invas-ion /ɪnveɪʒən/

Sometimes, there is allomorphy without a change in spelling, as in (7). In (7a), the morpheme electric in its bare form ends with a /k/ sound. However, after -ity has been suffixed, as shown in (7b), the /k/ sound changes to an /s/ sound. On the other hand, if -ian is suffixed, the /k/ sound changes to an /ʃ/ sound. All three sounds /k s ʃ/ are spelled with a .

(7) a. electric /əlɛktɹɪk/

b. electric-ity /əlɛktrɪsəti/

c. electric-ian /əlɛktriʃən/

Finally, sometimes there is a change of spelling without allomorphy, as in (8). In (8a), the morpheme happy in its bare form ends with an /i/ sound, spelled with a . When -ness is suffixed, the spelling changes from to , but the pronunciation stays the same.

(8) a. happy /hæpi/

b. happi-ness /hæpinəs/

Remember, as discussed in Section 4.1, that writing as a modality is secondary to spoken language. Written language is conscious, learned by convention, and, as shown in these examples, does not always accurately reflect the subconscious grammatical patterns of language.

Key takeaways

A morpheme can take different forms in different contexts, which is called allomorphy.

When the form of an allomorph is predictable based on phonological context, it is called a phonological allomorph.

If the form of an allomorph is not predictable based on phonological context, it is called a suppletive allomorph. Weak suppletion is when the allomorphs bear some resemblance, while strong suppletion is when the allomorphs bear no resemblance.

Allomorphy can also be classified based on what triggers the allomorphy. Phonologically conditioned allomorphs arise due to phonological context, morphologically conditioned allomorphs arise due to the inflectional properties of the stem, and lexically conditioned allomorphs arise due to the lexical properties of the stem, such as its semantic properties or idiosyncratic properties of the stem.

Check yourself!
[h5p id="43"]

2.5. Becoming a linguist: Identifying academic sources

juliedoner — Wed, 05 Jun 2024 06:35:09 +0000
In many of your university-level assignments, you will be asked to engage with academic or scholarly sources. In fact, unless you are told otherwise, you should probably assume that you will be expected to use academic sources by default. In this section, we will learn how to identify academic sources and what the review process is like for academic sources.
What is an academic source?
One of the main ways that an academic source is different from other kinds of publications is that an academic source is usually focused on [pb_glossary id="3798"]original research[/pb_glossary]; that is, the author of the source is not only reporting information that is already known about the topic, but is coming up with something new. For example, they could have gathered new data, developed a new theory, or extended an old theory to a new context. This means that academic sources are often very narrow in topic and quite specialized. Another way academic sources are quite different from other kinds of resources is the [pb_glossary id="3645"]peer review[/pb_glossary] process. When an article or another piece of scholarship is peer reviewed, the editor will contact other experts on the topic and ask them to read and critique the manuscript. There will usually be two or three reviewers, who will provide detailed comments on the manuscript and give a recommendation of whether it is suitable for publication. Often, an article will go through multiple rounds of review before it gets published. Different journals and publishers have slightly different processes, but most often the process will be either blind, which means the author doesn't know who the reviewers are, or double-blind, which means the author doesn't know who the reviewers are and the reviewers don't know who the author is. This process is most typical of an academic journal, but there are other kinds of sources worth mentioning, as well. Most books at a typical bookstore do not describe original research or go through a peer review process, but academic books do. Dissertations include original research but don't technically go through a peer review process; however, they are defended by the author and approved by a panel of experts, and so they have roughly equivalent status. On the other hand, they are often not as polished as published research. In contrast, other kinds of publications do not present original research and are not peer-reviewed. Most books in a typical bookstore and the magazines at the check-out counter are what would be called "popular publications." They are written by professional writers—not necessarily an expert in the topic—for a general audience. Although editors fact-check the content, they are not peer reviewed by experts and have few citations. Trade publications fall in between scholarly and popular publications. They are written for a specific group of people, and so are more specialized, but they still do not report original research, nor are they peer reviewed. An example might be a magazine for dentists summarizing the most recent dentistry tools and techniques.

Table 1. Kinds of publications

Academic source Popular publication Trade publication

Content Original research News, opinions, and trends News and opinions about a specific profession or area

Audience Researchers General People belonging to a particular profession

Author Researchers Journalist or professional writer Member of the field and/or professional writer

Citations Extensive Few Few

Review Peer-reviewed Fact-checked Fact-checked

There are also a lot of sources that are not peer reviewed but are sometimes used in academic writing. For example, researchers might post their work in progress on their website or on repositories before it is peer reviewed. Researchers also present their work in progress at conferences, and sometimes write up their presentation in a conference proceedings paper. Depending on the conference, proceedings papers may or may not be peer reviewed. Occasionally, you may even see someone cite a blog post, a social media post, an email, or even a chat in the hallway (usually cited as personal communication, abbreviated p.c.). This is because, no matter how prestigious or not your source is, you need to give appropriate credit.
A common place for linguists, in particular, to post and share their work in progress is on LingBuzz. Although many of the articles are legitimate work-in-progress by academic linguists, most of this work is not peer-reviewed (yet) and anyone can post an article.
When you are doing research, part of your job as a researcher is to evaluate the reliability of your sources. Any source potentially has errors, but a source that has been peer reviewed likely has fewer errors than one that has not been peer reviewed, and an informal source prepared by an expert is more likely to have fewer errors and more nuance than one prepared by a non-expert. Part of how you evaluate the source is by checking whether it is peer reviewed and whether it is authored by an expert. Another big part of evaluating sources is using your critical thinking skills to evaluate the content of the source itself. However, as a student, it is best to use peer-reviewed sources as much as possible. First of all, as a student, you are less likely to identify errors and inconsistencies than someone more experienced. Secondly, if you are doing research for an assignment, likely your instructor wants you to practice identifying and using academic sources.

The peer review process on Wikipedia

Most of us have looked something up on Wikipedia—probably even this week! We rely on it frequently in our daily lives, but professors will often say not to use it for your assignments. How come Wikipedia is good enough for our day-to-day, but not for your homework? Wikipedia uses a crowd-source volunteer model. That means that anyone with Internet access can create a new Wikipedia page or edit an existing one. You do not even need to login! However, anyone can also flag your edits as incorrect. With millions of users accessing Wikipedia every day, many errors get flagged and corrected very quickly. However, there are two very big differences between the review process of Wikipedia and the peer review process. First of all, review on Wikipedia is not systematic. It is possible that a minor edit on a little-used page might get missed for quite some time. Second, Wikipedia is not necessarily reviewed by experts. There are some experts on Wikipedia, but there are also lots of amateurs. This means that it is possible that a commonly-believed falsehood might survive on Wikipedia if the majority of volunteer editors believe it to be true. The kind of false information most likely to survive on Wikipedia is a plausible-sounding fact that most people don't know. For example, in How I accidentally started a Wikipedia hoax about Amelia Bedelia, journalist EJ Dickson describes a prank edit she and a friend, Evan, posted in 2009. Their prank edit claimed that the author of the Amelia Bedelia children's series, Peggy Parish, spent some of her childhood in Cameroon and based her series on a Cameroonian maid. EJ and Evan expected it to be flagged and removed right away, but not only did it survive Wikipedia's fact-checkers for 5 and a half years, it got repeated by several reputable sources, including Peggy Parish's own nephew! It is still true that Wikipedia can be a very useful source in many ways. However, for anything that matters, you should double-check the information. Edits in Wikipedia are supposed to have sources listed (although they don't always). Follow the citation to the original source and double-check that the source is reputable and that it actually says what Wikipedia says it says, and then cite the original source.

How to identify an academic source
If you read enough academic papers, you will probably start to notice some patterns about who publishes a lot on certain topics and where they publish, and start to form an understanding about which journals are the most prestigious. However, for most students, this doesn't really start happening unless they pursue graduate studies. How do you decide who is an expert and which journals are reputable when you're just starting out?
The author
One of the first steps is to look at the author. Read their bio and google them. Some things to check:

Do they have a university affiliation?

Is their research specialty related to the topic of the source?

Graduate students and professors writing in their research area are likely reliable authors.
The web host
Where did you find the article? If you found it through your university library, that is likely a good sign. If it is hosted on an academic publisher's website, that is also a good sign. However, watch out for websites where people create profiles and share their own research, like ResearchGate and Academia.edu. Many legitimate researchers do post their published and unpublished research on such websites, and Google Scholar searches will find results from those websites. But anyone can create a profile and post an article. Double-check the author's credentials before citing work posted by the author, especially if it is unpublished.
The publisher
If the paper is formally published, who published it? If it's published by a university press, it is likely a scholarly source. Additionally, some for-profit publishers specialize in academic publishing, while others specialize in trade or popular publications, so who the publisher is can be a big clue about what kind of publication it is. You can google a publisher to discover their specialty. Some academic publishers who publish a lot of work in linguistics written in English include:

Cambridge University Press

Elsevier

John Benjamins

Kluwer

MIT Press

Mouton de Gruyter

Oxford University Press

Springer

Wiley Blackwell

The open access revolution in linguistics

Have you ever been frustrated by the paywalls on academic articles? You're not alone! Academic articles are researched, written, and edited by academics. These academics do not usually get paid by the journals for their work; instead, it is considered necessary service to their profession. Publishers may take care of copy-editing, formatting, printing, and web-hosting of the final product, but over time, more and more of that work has been automated or has been passed on to the unpaid academic editors. And yet, publishers continue to increase the subscription fees that university libraries have to pay in order to get access to the journal articles. In 2015, the editors of the journal Lingua submitted a proposal to Elsevier, the for-profit publisher that operated the journal, to move Lingua to an open access model. They proposed that everyone should be able to read Lingua for free and that authors would retain copyright on their articles. Elsevier refused, so the entire editorial board of Lingua quit and started a new, open access journal called Glossa. Elsevier recruited a new editorial board and is still publishing Lingua, but it dropped from one of the top journals in linguistics to a marginal one. Meanwhile, Glossa has become very prestigious and popular, and is fully open access! More recently, the editors of another journal, Syntax, are making a similar move. In 2024, Wiley Blackwell, the publisher of Syntax, eliminated the journal's paid editorial staff and assigned all the tasks to a generalized team who aren't trained in formatting linguistic data. As a result, the founding editor and several others have resigned and are in the process of founding a new open access journal named Syntactic Theory and Research.

The editorial board
Academic publishers will have editorial boards for their journals and for book series. It is the editorial board who arranges for peer review and ensures that the published work meets the standards of the scholarly field (and not just general publication standards). Editorial boards of academic publications should be made up of established academics with expertise in relevant fields, and will typically be professors at universities. Just like you can google the authors of papers, you can google the editorial boards to determine whether they are reputable scholars.

"Publish or perish" and predatory journals

Academia is very competitive, and there is a lot of pressure for early career academics to publish papers in order to get a job and then get promoted to a tenured position—especially since there are far more qualified candidates than positions available. As a result, a number of "predatory journals" have sprung up to exploit graduate students and other inexperienced academics. These predatory journals will often send recruitment emails inviting people to submit articles, with many of the same properties as other phishing emails. Some signs of a predatory journal include:

A very broad topic area, or publication of research articles outside of their stated topic area

A very short review process (e.g., less than three months)

A cost to the author to publish

No editorial board or an editorial board that does not consist of established scholars

Few in-text references (e.g., less than 2-3 per paragraph)

Short bibliographies that cite mostly general resources

Low quality content (e.g., presenting opinion as fact, lack of engagement with opposing views, methods are not described in enough detail to be replicated)

Key takeaways

An academic source typically is written by an expert, presents original research, is peer reviewed, and includes extensive citation.

The peer review process is when other experts on the topic read and critique a manuscript and give a recommendation to the editor for whether the manuscript should be published.

To identify whether a source is academic, look for the following clues:

The author has a university affiliation and is writing within their research specialty.

The source is posted on a publisher's website, rather than a website where authors share material directly.

The publisher is an academic publisher or university press.

The editorial board consists of established academics with relevant expertise.

Check yourself!
[h5p id="49"]
References and further resources

For a general audience

Dickson, EJ. 29 July 2014. How I accidentally started a Wikipedia hoax about Amelia Bedelia. Daily Dot. https://www.dailydot.com/unclick/amelia-bedelia-wikipedia-hoax/

Walkden, George. 9 March 2024. Editors of Syntax resign, found new journal. Open Access Linguistics. https://oaling.wordpress.com/2024/03/09/editors-of-syntax-resign-found-new-journal/

Wexler, Ellen. 5 November 2015. What a mass exodus at a linguistics journal means for scholarly publishing. The Chronicle of Higher Education. https://www.chronicle.com/article/what-a-mass-exodus-at-a-linguistics-journal-means-for-scholarly-publishing/

6.9. Becoming a linguist: How to read an academic article

juliedoner — Wed, 05 Jun 2024 06:41:56 +0000

Academic articles vs. textbooks
You are probably used to reading textbooks, but reading academic articles may be a new skill. Here are some important differences between academic articles and textbooks to keep in mind. The most obvious difference between a textbook and an academic article is the length. While a textbook is a whole book, an academic article is short. In linguistics, they are often between 15 and 50 pages long. A textbook is usually used for learning or studying, while an academic article is usually used for research. This changes how you read! When you are reading a textbook, every detail might be important, and your professor will probably expect you to understand all or most of the reading. On the other hand, your professor will probably not expect you to understand every part of an academic article. When you are reading an academic article, you shouldn't pay equal attention to all parts of the paper. Think about why you are reading that paper, and focus on those parts. For example, if you are studying the same language as the paper, you will spend more time studying the data. If you are trying to do a similar study on your own, you will spend more time studying the methods. If you care most about the theory the paper is arguing for, then you will spend more time studying the argumentation and conclusions. A textbook will assume that you know nothing about the topic. The purpose of the textbook is to teach you the topic! An academic article, in contrast, assumes you have background knowledge. This is the bit that makes reading academic articles the most difficult. When you are just starting out, of course you won't have all of the background knowledge! Morphosyntax as a field is big enough that even experienced researchers won't have all of the background knowledge necessary to understand the article. Academic articles, remember, typically present original research, which means that they are presenting new information. The article will contain brand new ideas, so even the most experienced professor will be learning something new from the article. Here are some strategies that can help you navigate your gaps in knowledge:

Most of the time, you just need to understand the main claim of the article, not all of the details. Work on figuring out that main claim, and don't stress over all of the details that you don't understand.

Try to figure out which pieces of background knowledge are really necessary to understanding the main goal of the paper, and which are not that important. Don't look up every word you don't know.

Remember your goal for reading the paper! Do you care more about the data, the methods, or the theory? Spend most of your time and energy trying to understand the parts that are relevant to you.

Pay attention to the year the paper was written. If you know a little bit about the history of linguistics, the year is a big clue about what kinds of assumptions and frameworks the author was working with.

Sometimes the paper uses a technical term that is new to you, but maybe you do know it, just by a different name. If it looks familiar but they call it something else, don't doubt yourself! It is good to double check that they mean the same thing, though.

Once you have identified a gap in your knowledge that you really need to fill in order to meet your research goals, how do you go about learning what you need to know? Of course, you can search online, but you may find that online searches might turn up simplified or irrelevant explanations, depending on the term. Here are some ways that you can get more specialized answers, ordered from easiest to most difficult:

Ask your instructor for advice or ideas about where to go. Your instructor may be able to give you specific advice about individual topics.

Look up terms in a linguistics glossary or dictionary, such as the one at the end of this textbook or Utrecht's Lexicon of Linguistics.

Find a textbook that introduces and explains the topic.

Find a cover article that explains the topic. A cover article is a special kind of academic article that, instead of presenting new original research, gives an overview of the current ideas and research on a given research area.

Find a dissertation or book that addresses the topic, and read the introduction or literature review section of the dissertation or book.

Finally, in morphology and syntax, we are often working with a model of grammar. A textbook will present a complete model. It may be a simplified model, and the textbook author has thought about how all the pieces integrate into the whole. On the other hand, an academic article may present just one piece and go deep into that one piece. The author of the academic article might not have an opinion about how this piece interacts with all of the other pieces in the model.
Reading techniques: The three pass method
Believe it or not, most academics do not read an article from beginning to end straight-through. Instead, they bounce around strategically, making decisions about where they want to focus and how much time and energy they need to spend on understanding the article. In this section, I will describe the three pass method to reading an academic article which can help you begin to read more strategically. This description is adapted from Keshav (2002) with elements from Tatman (2016).
The first pass: A strategic skim
The first pass through the article should be quick, maybe even only 10-15 minutes. You're only going to read parts of the article, and your main goal of the first pass is to decide if this is the right article for you to spend your time on.
Step one: Read the title, abstract, and keywords carefully
As you're reading, try to figure out what the main goal of the paper is. Decide if you want to spend more time on this article, and if so, which parts are most relevant. Ask yourself why you are reading this paper. Which questions are you trying to answer? Does this article address those questions?
Step two: Scan the introduction and conclusion
Skim through the introduction and conclusion. Try to determine what the main conclusions of the paper are and what kind of evidence was used. At this point, you may also wish to note any terms you don't know that are used more than once in both the introduction and the conclusion. Consider looking them up now, so that you are better equipped to understand the meat of the paper in the second pass. Remember, you don't need to look up every term, just the ones that seem like they will be important.
Step three: Flip through the paper and read the headings and skim the data
In this step, you are figuring out how the paper is structured. You are trying to determine which sections will be most relevant to the questions you're trying to answer, and how the different sections relate to each other. You can use this information to pace yourself in the second pass, especially if the article is long enough that you won't be able to read it in one sitting. You might also choose to only read portions of the paper in the second pass. As you skim the paper, pay attention to what kind of data is used. For example, the data may consist of full sentences, individual words, morphological paradigms, or statistics. Which language(s) does the data come from? Is it adult or child language? Also pay attention to what kinds of formalisms and diagrams are used. For example, the paper may include semantic formula, syntax trees, tables, charts, graphs, or formalisms that are unfamiliar to you. You can use the data and the formalisms used to get a bigger picture of what kind of paper it is: the theoretical framework, the methodology, and the subfields that are relevant.
Step four: Stop and evaluate
At the end of the first pass, take some time to answer the following questions (adapted from Keshav 2007).

Category: What kind of paper is it? Is it summarizing previous work or presenting a new analysis? What subfield(s) does it fall under? What methodologies are used? Is it descriptive, describing the properties of some language, or theoretical, arguing for a theoretical position, or both?

Context: What theoretical framework does it adopt? Is it one you are familiar with? What other papers is it related to?

Corrections: Did you notice any errors or faulty assumptions?

Contributions: How does this paper advance the field?

Clarity: Is it well-written?

At this point, decide whether to keep reading. You may choose to stop after the first pass if you don't think the paper will be useful to you after all, either because it wasn't quite the right topic or methodology, because you think it makes faulty assumptions, or because you think you already got the information you need from the first pass.
The second pass: A quick read-through
In the second pass, you will either read the paper through from beginning to end or read just the sections that you picked out as most relevant to you. In the second pass, you will read more carefully but you still aren't trying to understand every detail. Instead, you should focus mainly on trying to understand the main idea of the paper. After the second pass, you should be able to explain the main argument of the paper as well as the main supporting evidence. If you don't understand a paragraph after reading it a couple times, move on. The second pass shouldn't take you much longer than 2-3 hours, depending on the length of the paper. As you read, formulate questions based on what you don't understand and based on further research goals:

What don't you understand? Pick out terms that seem important that you don't know. Are there parts of their argument that leave you confused?

What is something you'd like to know more about? If you had to write a research paper about something, what ideas does this paper give you?

There are many reasons why you might still not understand the main ideas of the paper after the second pass. The paper may be in a topic area, framework, or methodology that you aren't familiar with, with lots of unfamiliar terminology and/or formalisms. In other words, you don't have sufficient background. The paper may also be poorly written. Finally, maybe you're just tired. If you don't understand the paper after the second pass, there are several possible responses. You could rest and try again. This may help simply because the second time trying to understand something is easier than the first time. You could do more background reading and research to fill in your gaps of understanding, and then try again. Or you can set it aside and hope that it's not going to be important to your course or your career. The second pass is enough depth for most papers you read. It is enough for most course assigned readings. However, if you are leading a discussion about the paper, writing a critique or review of the paper, or using it as the main resource for your own research, you should probably move on to the third pass.
The third pass: A deep reading
If you choose to do a third pass, this is when you try to understand every single detail in the paper. Consider every step of the argumentation and every piece of data. Read the footnotes. Look for flaws. Ask yourself, if you were writing this paper, would you have made the same choices and drawn the same conclusions?

Key takeaways

Reading an academic article requires different strategies than other kinds of reading.

When reading an academic article, you should keep in mind your purpose for reading it and strategize which parts of the article to spend more time on.

Do not expect to understand all of the details in an academic article, especially in an area that is new to you.

Most academics usually skim the title, headings, introduction, and conclusion before they read an article as a whole, to help them decide if they want to read it more carefully and which parts to focus on.

Check yourself!
[h5p id="51"]
References and further resources

For students

Fruehwald, Josef. August 22, 2022. Reading a technical paper. Lin517: Natural Language Processing. https://jofrhwld.github.io/teaching/courses/2022_lin517/resources/reading/

Kerstens, Johan, Eddy Ruys, and Joost Zwarts. 2001. Lexicon of Linguistics. Utrecht University. https://lexicon.hum.uu.nl/

Keshav, Srinivasan. 2007. How to read a paper. ACM SIGCOMM Computer Communication Review 37 (3): 83-84.

Tatman, Rachael. January 29 2016. How to read a linguistics article in 8 easy steps. Making Noise & Hearing Things. https://makingnoiseandhearingthings.com/2016/01/29/how-to-read-a-linguistics-article-in-8-easy-steps/

Chapter 5 practice exercises

juliedoner — Tue, 25 Jun 2024 02:46:37 +0000

Data analysis

Exercise 1
Classify each of the following scenarios according to its model of community engagement from Czaykowska-Higgins (2009) The options are: (a) linguist-focused model, (b) advocacy research, (c) empowering research, or (d) community-based language research.
a. Joshil Patel has been writing his PhD dissertation on the Dene language. He has lived in one particular Dene community for months at a time as he has collected his data, during which time the local school has been implementing a Dene language immersion program, but they are short on language resources. Joshil has helped the community apply to government grants for the language immersion program. As part of his research, Joshil recorded several stories told by elders in Dene. Joshil types up the stories, prints and binds them, and gifts them to the school as a thank you for their contribution to his dissertation research.

b. Sam Klein just got hired as a professor at the University of Manitoba. Once they got settled in, Sam researches which local Indigenous organizations are involved in language revitalization. Sam reaches out to several of those organizations and offers their help and expertise. One of the organizations asks for help with their ongoing dictionary project. Sam begins by attending several meetings as a spectator, and eventually volunteers to join the committee working on the morphological analysis of the complex verbs of the language. Sam is using their status and experience as a university professor to get the committee’s discoveries published in an academic journal, with several community members as co-authors. One of the volunteers on the project is a bright high school student. Sam helps the high school student apply to the linguistics program at UM.

c. Jane Clark is very interested in wordhood and in developing a cross-linguistic definition of words. She wants to test her hypothesis on a polysynthetic language. She reads up on Inuktitut in the university library. She applies to the university ethics board for conducting an experiment on speakers of Inuktitut. She then travels to an Inuit community and recruits Inuktitut speakers to participate in her experiment. She pays participants $30 for completing her 30-minute experiment.

d. Yuxi Liu and Zihao Wu are writing a grammar of the Taiwanese language Amis. Yuxi and Zihao have visited the Amis communities frequently as they work on their project. They had several community members give comments on the outline and each draft of their grammar. Based on feedback from the community, Yuzi and Zihao have included a glossary of technical linguistic terms so that the grammar can be used by community members and not just by other linguists. Yuzi and Zihao will thank the community members in the acknowledgements section of their book.

Communication and study skills

Exercise 2
Research a language and write an introduction to that language that emphasizes the relationship between the language and its speakers and humanizes them.
Exercise 3
Read the Truth and Reconciliation Commission’s Calls to Action and/or the Missing and Murdered Indigenous Women and Girls Calls for Justice. Choose either one call to action or one call to justice and discuss how it is relevant to linguistic research. How can you create a reciprocal and respectful relationship with the communities whose languages we are studying?
Research and application

Exercise 4
Choose a place that is significant to you, such as where you live, where you grew up, or where your ancestors are from. What languages are/were spoken there? Spend extra time looking for languages and varieties that are not well-known. Are there Indigenous or endangered languages that you’ve never heard of before? How has colonialism and history affected the language landscape of this place?
Language journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 5
Is your language an Indigenous language? How do you know?
Exercise 6
Research the history of linguistic research on your language. In what ways were colonial practices in use?
Exercise 7
Write a profile about one of the communities that use your language.

5.2. Colonialism and decolonization

juliedoner — Tue, 25 Jun 2024 02:50:23 +0000
The Canadian Encyclopedia defines colonialism as "the process through which a foreign people establish control over a territory and, if applicable, its Indigenous peoples" (Hale 2023). Colonial control can be exercised in a variety of ways, and often includes control of a people's language. Shoemaker (2015) describes 12 different forms of colonialism, two of which are relevant to us: settler colonialism and extractive colonialism.
Settler colonialism

Settler Colonialism. Large numbers of settlers claim land and become the majority. Employing a “logic of elimination,” as Patrick Wolfe put it in the American Historical Review, they attempt to engineer the disappearance of the original inhabitants everywhere except in nostalgia.
-Shoemaker 2015

Settler colonialism was the major form of colonialism deployed in Canada and many other nations. Throughout much of Canadian history, "desirable" settlers were actively recruited, while "undesirable" settlers were discouraged and Indigenous peoples were killed, displaced, or assimilated by force in order to clear the land for settlers. One major tool of forced assimilation used in Canada and elsewhere is the Indian Residential School System. Children were forcibly removed from their families and taken to boarding schools where they were horribly mistreated, as well as forbidden from speaking their languages and practicing their cultures. The stated aim of these residential schools was assimilation, as stated openly by Prime Minister John A. Macdonald in parliament.
When the school is on the Reserve the child lives with its parents, who are savages; he is surrounded by savages, and though he may learn to read and write his habits and training and mode of thought are Indian. He is simply a savage who can read and write … Indian children should be withdrawn as much as possible from the parental inﬂuence, and the only way to do that would be to put them in central training industrial schools where they will acquire the habits, modes and thought of white men.
-John A. Macdonald, House of Commons, 9 May 1883 as cited by Starblanket 2018: 179

The Truth and Reconciliation Commission classifies these practices as a "cultural genocide" (2015). The act of forcibly transferring children from one group to another group is also classified as an act of genocide according to the United Nations (1948). The last residential school didn't close until 1996 and communities are still healing and rebuilding. Many community elders and language users are residential school survivors. Other forms of forced assimilation are still in force, as well, such as the placement of Indigenous children in foster care with non-Indigenous families. When we do linguistic research with communities that have experienced forced assimilation, we need to keep this in mind. There may be only a few language users left, whose time and energy belongs to the community first, not to outsider linguists. Because language use was punished in the residential schools, survivors may have trauma associated with the language. Others may have purposefully not taught the language to their children in order to protect them. Community members may be rightfully upset if settlers and other outsiders seem to have more access to their own language than they do.
Extractive colonialism

Extractive Colonialism. All the colonizers want is a raw material found in a particular locale: beaver fur, buffalo hides, gold, guano, sandalwood. The desire for natural history specimens and ethnographic artifacts could also be considered extractive colonialism. A slash-and-burn operation, extractive colonialism does not necessarily entail permanent occupation, but it often seems to follow. Extractive colonizers might destroy or push away indigenous inhabitants to access resources but more typically depend upon native diplomatic mediation, environmental knowledge, and labor.
-Shoemaker 2015

Although aspects of settler colonialism are still at play in Canada, we have in many cases transitioned to an extractive colonial society.

Extractive colonialism in Manitoba

Extractive colonialism is alive in well in Canada. It is part of our everyday lives, even though it may be invisible to many of us. Let's look more closely at two instances of extractive colonialism that is ongoing in Manitoba: Winnipeg's water supply and Manitoba Hydro.

Winnipeg aqueduct *coming soon*. Read Aqueduct by Adele Perry or watch Freedom Road to learn more.

Starting in the 60s, Manitoba Hydro has diverted a major river and built several dams in Northern Manitoba. It is continuing to build expensive dams, even though we are already generating more power than we need. The initial construction of these dams has a huge ecological impact, involving clear cutting forests, moving entire communities, and building quarries and roads, in addition to the dam itself. The ongoing impact of Manitoba Hydro's management of water levels also has a huge impact. Large fluctuations in water level have swept away entire beaches, downed millions of trees, killed wildlife, and have destroyed the local fishing population and fishing-based economy. The rapid erosion also introduces mercury into the water. The ongoing erosion makes it dangerous to travel along the shorelines, while the trees in the water and flooded islands make it dangerous to boat. The town of South Indian Lake used to have the second largest whitefish fishery in North America, but now 85% of the community relies on social assistance (Wilt 2016). Meanwhile, local hydro employees receive discount heating while Indigenous populations pay a "rural" premium—which, together with the destruction of the fishing industry, creates a wealth gap largely along racial lines. Meanwhile, Manitoba Hydro continues to request and receive augmented flow licenses every year, allowing them to manipulate water levels more than the initial agreements. Read "How green energy has hurt First Nations in the North" by James Wilt or watch Green Green Water to learn more.

Extractive colonialism doesn't only apply to physical resources like fur and oil, though; it also applies to knowledge. When we do research on a language, we need to be careful to protect the well-being of the community the data comes from, keeping in mind that we are guests and that the language belongs to the community, not to the researchers. We also need to remember that our research depends on community members' knowledge, labour, and expertise, and that we would not be able to do our research without them.
Decolonization
Decolonization, on the other hand, refers to the removal of colonial elements. This means that we have to first recognize the ways that our systems are colonial in nature and then restructure them to remove those colonial structures. In the next few sections, we will first learn about the colonial history of linguistics and then we will look at models for how to move forward in a good way.

Key takeaways

Settler colonialism is when settlers attempt to become the majority population group of a land, displacing or assimilating the original population.

The Indian Residential School System in Canada was a form of genocide used to force the assimilation of Indigenous peoples.

In Indian Residential Schools, children were typically forbidden from speaking their languages, which still has far-reaching impact on Indigenous language use today.

Extractive colonialism is when settlers collect resources from the settled land, often using Indigenous labour or expertise, prioritizing their desire for resources over the well-being of Indigenous communities.

Research conducted on Indigenous peoples can be a form of extractive colonialism if it is not done in a good way.

Decolonization is the process of identifying colonial systems and removing them from our society.

Check yourself!
[h5p id="42"] References and further resources The references in this section that have at least one Indigenous author are indicated with an ➡️.
For a general audience

Hale, Karl. 2023. Colonialism in Canada. The Canadian Encyclopedia. https://www.thecanadianencyclopedia.ca/en/article/colonialism-in-canada

➡️ McLeod, Angelina. 2019. Freedom Road. National Film Board of Canada. https://www.nfb.ca/series/freedom-road-series/

Mikkelson, Dawn, and Jamie A. Lee (Directors). 2005. Green green water. Culture unplugged. https://www.cultureunplugged.com/play/6661/Green-Green-Water

Perry, Adele. 2016. Aqueduct: Colonialism, resources, and the histories we remember. Winnipeg: ARP Books.

➡️ Shoemaker, Nancy. 2015. A Typology of Colonialism. Perspectives on History.

Truth and Reconciliation Commission of Canada. 2015. Honouring the truth, reconciling for the future: Summary of the final report of the Truth and Reconciliation Commission of Canada. https://ehprnh2mwo3.exactdn.com/wp-content/uploads/2021/01/Executive_Summary_English_Web.pdf

United Nations. 1948. Convention on the prevention and punishment of the crime of genocide. https://www.un.org/en/genocideprevention/genocide-convention.shtml

➡️ Wa Ni Ska Tan Alliance. 2024. Augmented flow: License to destroy documentary. https://www.youtube.com/watch?v=ZdrEwEt-V7o

Wilt, James. 19 October 2016. How green energy has hurt First Nations in the North. Vice. https://www.vice.com/en/article/4w58mq/how-green-energy-has-hurt-first-nations-in-the-north

Academic sources

Carleton, Sean. 2021. 'I don't need any more education': Senator Lynn Beyak, residential school denialism, and attacks on truth and reconciliation in Canada. Settler Colonial Studies 11 (4): 466-486. https://www.tandfonline.com/doi/full/10.1080/2201473X.2021.1935574

➡️ Fontaine, Lorena. 2017. Redress for linguicide: Residential schools and assimilation in Canada. British Journal of Canadian Studies 30(2): 183–204.

MacDonald, David & Graham Hudson. 2012. The Genocide question and Indian residential schools in Canada. Canadian Journal of Political Science 45: 427–449.

➡️ Starblanket, Tamara. 2017. 'Kill the Indian in the child': Genocide in international law. InIndigenous peoples as subjects of international law, ed. Irene Watson. Abingdon, Oxon: Routledge. 171-200.

8.7. Case study: French possessive adjectives are not adjectives

juliedoner — Wed, 03 Jul 2024 20:58:57 +0000
One common difference between descriptive grammars written by linguists and pedagogical grammars is how they use technical terminology. For instance, in French textbooks, the term adjectifs possessifs ‘possessive adjectives’ is commonly used. This might suggest to students that these words are actually adjectives. In reality, they are determiners. Hence, we will be calling them possessive determiners to be consistent with our final conclusion. We can analyze how adjectives, determiners, and possessive determiners are used in sentences to demonstrate this. Before showing why possessive determiners are not adjectives, it might be useful to explain why someone might think that they are adjectives in the first place. First, adjectives in French have to agree in number (singular vs. plural in French) with the noun that they modify. When they fail to agree, the result is an ungrammatical phrase. As shown in (1), since the noun personnage 'character' is singular, adjective principal 'main' in its singular form can modify it. However, if we try using the plural form, principaux 'main', the result is an ungrammatical phrase.

(1) a. French

le personnage principal

the.SG character.SG main.SG

'the main character'

(1) b. *le personnage principaux

the.SG character.SG main.PL

The same is true of possessive determiners. They must also agree in number with the noun that they modify. In (2), if we use singular possessive determiner mon 'my' with the singular noun stylo 'pen', the result is grammatical. Again, however, if we try using plural possessive determiner mes 'my' with the singular noun, we get an ungrammatical phrase.

(2) a. French

mon stylo

my.SG pen.SG

'my pen'

(2) b. *mes stylo

my.PL pen.SG

Additionally, both adjectives and possessive determiners need to agree in noun class (masculine vs. feminine in French) as well. As shown in (3)[footnote]These examples were based on a passage from Manitoban author Gabrielle Roy's La route d'Altamont.[/footnote], the feminine noun catin 'doll' must go with feminine possessive determiner ma 'my' and feminine adjective blonde 'blonde'. If either is replaced with its masculine variant, mon 'my' and blond 'blond', then the phrase is ungrammatical.

(3) a. French

ma catin blonde

my.F doll.F blonde.F

'my blonde doll'

(3) b. *mon catin blonde

my.M doll.F/td> blonde.F

(3) c. *ma catin blond

my.F doll.F blond.M

(3) d. *ma catin blond

my.M doll.F blond.M

In terms of agreement, then, adjectives and possessive determiners are quite similar and so it is not unreasonable for someone to group the two together in some way. However, other determiners, such as articles, must also agree in number and noun class with their corresponding noun. For number agreement, in (4), singular noun cheval 'horse' must go with singular determiner le 'the' while plural noun chevaux 'horses' must go with plural determiner les 'the'.

(4) a. French

le cheval

the.SG horse.SG

'the horse'

(4) b. *les cheval

the.PL horse.SG

(4) c. *le chevaux

the.SG horse.PL

(4) d. les chevaux

the.PL horse.PL

'the horses'

For noun class agreement, in (5), masculine noun livre 'book' can co-occur with masculine determiner le 'the' but not feminine determiner la 'the'.

(5) a. French

le livre

the.M book.M

'my book'

(5) b. *la livre

the.F book.M

At this stage, adjectives, determiners, and possessive determiners are all behaving the same way. Intuitively, however, speakers of French know that determiners and adjectives do not belong to the same category. As such, possessive determiners can only logically be adjectives, determiners, or some third category. One way to test the category of possessive determiners is to look at ways in which adjectives and determiners behave differently and then see if possessive determiners act more like adjectives, more like determiners, or in a way different from each of the two. The first test concerns the relative word order of adjectives, determiners, and possessive determiners against cardinal numbers. As a reminder, cardinal numbers are used for counting things (one, two, three, etc.). This can be contrasted with ordinal numbers for example, which order things (first, second, third, etc.). In French, determiners such as les and possessive determiners such as mes go before cardinal numbers, as shown in (6) and (7) respectively.

(6) French

les deux femmes

the two women

'the two women'

(7) French

mes deux frères

my two brothers

'my two brothers'

Conversely, adjectives come after cardinal numbers. As shown in (8), the adjectives jaunes 'yellow' and jeunes 'young' come after the cardinal number trois 'three'. This is true whether the adjective comes before or after the noun that it modifies.

(8) a. French

les trois livres jaunes

the three books yellow

'the three yellow books'

(8) b. les trois jeunes garçons

the three young boys

'the three young boys'

The second test that can be used is the deletion test. In a sentence, removing an attributive adjective such as petit 'little' does not make a sentence ungrammatical, as shown in (9).

(9) a. French

Elle aime le petit chien.

she likes the little dog

'She likes the little dog.'

(9) b. Elle aime le chien.

she likes the dog

'She likes the dog.'

However, sentences do become ungrammatical if we remove a determiner (10) or a possessive determiner (11).

(10) a. French

Il veut un chat.

he wants a cat

'He wants a cat.'

(10) b. *Il veut chat.

he wants cat

(11) a. French

Je vois mes amis.

I see my friends

'I see my friends.'

(11) b. *Je vois amis.

I see friends

Finally, we can use the substitution test. Typically, you can replace words of one syntactic category with a word of the same category but not of another category. In French, if we try to substitute an adjective like bon 'good' for another adjective like grand 'big', the result is grammatical (12a-b). Conversely, if we try to replace an adjective with a determiner (12c) or a possessive determiner (12b), the result is ungrammatical.

(12) a. French

On a un bon livre.

we have a good book

'We have a good book.'

(12) b. On a un grand livre.

we have a big book

'We have a big book.'

(12) c. *On a un le livre.

we have a the book

(12) d. *On a un mon livre.

we have a my book

Additionally, if we have a sentence with a determiner in it, it can be freely substituted by another determiner (13a-b) or a possessive determiner (13c) but not an adjective (13d).

(13) a. French

Elle mange la poutine.

she eats the poutine

'She's eating the poutine.'

(13) b. Elle mange une poutine.

she eats a poutine

'She's eating a poutine.'

(13) c. Elle mange ma poutine.

she eats my poutine

'She's eating my poutine.'

(12) d. *Elle mange grande poutine.

she eats big poutine

Note that the examples directly above also show that possessive determiners can be freely substituted by determiners but not adjectives. Based on the three tests above—the relative word order compared to cardinal numbers, the deletion test, and the substitution test—it can be seen that determiners and possessive determiners behave in one way whereas adjectives behave in a different way. This suggests that possessive determiners are in fact determiners and not adjectives. This section has shown only three tests that provide such evidence, but there are in fact more. Hopefully this section has shown you that pedagogical grammars often have analyses that do not fit well into modern morphosyntactic analysis and that even provide more difficult explanations in some cases. In terms of possessive determiners, if they were adjectives, there is no natural explanation for why they always go before nouns, come after cardinal numbers, cannot co-occur with other determiners, can be substituted by determiners and not adjectives, and cannot be deleted when “modifying” a noun. Under the analysis that possessive determiners are determiners, learners know that they can simply put the possessive determiners in place of where they would put a determiner. The only reason why these problems seem to seldom come up among anglophone learners is that these constructions are quite similar to what we have in English.

Key takeaways

So-called "possessive adjectives" in French actually pattern with determiners.

Evidence that they are possessive determiners include agreement patterns, word order, obligatoriness, and substitution.

Pedagogical grammars and traditional language-specific terminology doesn't always match up with linguistic analysis.

Check yourself!
[h5p id="47"] Consult the following data in answering the check yourself questions:

(14) a. French

Nous buvons nos boissons.

we drink our drinks

'We're drinking our drinks.'

(14) b. *Nous buvons boissons.

we drink drinks

(14) c. Nous buvons les boissons.

we drink the drinks

'We're drinking the drinks.'

References and further resources

Sources for examples

Roy, Gabrielle. 1966. La route d'Altamont. Montréal: Editions HMH.

6.8. Becoming a linguist: Anatomy of an academic article

juliedoner — Fri, 02 Aug 2024 05:06:37 +0000

The first page
The first page of an academic article has lots of useful information, as shown in the annotated picture of the first page of Bjorkman (2022) in Figure 1. I've marked five areas where you should look when first navigating an academic article, which I will explain below. This article is published in the Canadian Journal of Linguistics. Other journals will have different layouts, but they will generally be similar and have many of the same elements. [caption id="attachment_4072" align="aligncenter" width="1152"] Figure 1. The first page of an academic journal article, Bjorkman (2022), published by the Canadian Journal of Linguistics. Used under CC BY 4.0 license.[/caption]
Box 1 (red): The publication information
At the top of the page, as part of the header, the publication information of the article is listed. In this case, it includes the journal title, the volume and issue numbers, the page numbers, the year of publication, and the DOI, which is a unique number assigned to academic articles to help you find them. You will need this information in order to write a citation for the article. In some journals, this could be listed on the bottom of the page or on a cover page before the article begins, instead.
Box 2 (purple): The title
This is the name of the article, and one of the first things you will look at when deciding whether you should read an article. The title of this article is Some structural disanalogies between pronouns and tense. What can you tell about this article based on this title?

The paper is probably written in English, since the title is in English.

It is about the structure of [pb_glossary id="670"]pronouns[/pb_glossary] and [pb_glossary id="3550"]tenses[/pb_glossary], and so is probably related to the fields of morphology, syntax, and/or semantics.

The word disanalogies suggests that it is about how pronouns and tense are different from each other.

Box 3 (orange): The author's information
The author of the article is usually listed underneath the title, alongside other information about the author, such as their affiliation. This article is written by Bronwyn Bjorkman who is affiliated with Queen's University in Kingston, Ontario. This article also provides the author's email. If you do wish to contact an author, I recommend googling them, in case their contact information has changed since the publication of the paper.
Box 4 (green): The abstract
Underneath the author's information is a paragraph labeled abstract. When abstract is used as an adjective, it usually means the opposite of concrete. But when abstract is used as a noun, especially in academic settings, it usually means a summary that you can use to decide whether you want to engage with a bigger work. You can think of the abstract as a movie trailer or the blurb on the back of a book. Not all journal formats include a header labeling the abstract. Sometimes, the abstract will just be found at the beginning of the paper in a box or a different font. More rarely, a paper may not have an abstract at all. Here is the text of this article's abstract:
A pronominal analysis of tense goes back to Partee (1973), motivated by a series of proposed parallels between the interpretation of tenses and that of pronouns. This article revisits Partee's interpretive parallels, as well as two more identified in Kratzer (1998), in light of subsequent developments in work on both temporal relations and on pronouns. The goal of this article is not to argue for or against a pronominal analysis of tense, but instead to make clearer the syntactic and semantic space within which such an analysis is situated, especially given that pronouns have been given increasingly complex syntactic representations even as tense has remained syntactically simplex. -Bjorkman 2022: 143
Here's what you can tell from this abstract:

Partee (1973) and Kratzer (1998) argue that there are some parallels between tense and pronouns, and they use those parallels to provide an analysis of tense. Note: Barbara Partee and Angelika Kratzer are both very well-known and influential semanticists. Their status combined with the age of their articles suggests that their analyses of tense have been very important, perhaps foundational.

These analyses are about the interpretation of tense and pronouns, which suggests that this article may include some semantic analysis.

More work has been done on tense and on pronouns since Partee's and Kratzer's articles were published, which is why their analyses are worth revisiting.

The article does not necessarily contradict Partee and Kratzer.

One difference between pronouns and tense is that pronouns tend to be analyzed as having complex syntactic structures but tense as having simpler structure.

Because the CJL is a bilingual publication, it provides abstracts for all of its papers in both French and English (even though the papers themselves are not bilingual). In this case, the French abstract is below the English one, labeled Résumé.
Different kinds of academic materials have different kinds of abstracts, with some differences. For example, the abstract of an academic article, like this one, is only about a paragraph long. It won't usually include references or headings (although in some other fields, it may). It is usually provided right before the article it is about. The abstract for a conference presentation, in contrast, is often submitted anonymously months before the conference begins, and is used to determine which presentations are included in the conference. The selected abstracts will then be de-anonymized and published on the conference website, so that people can use them to decide which presentations to attend. Conference abstracts are usually one or two pages long, with strict page limits set by the conference organizers. Headings are usually used to conserve space and direct the readers' attention, and an abbreviated reference list may be included at the end.

Box 5 (blue): Keywords
Underneath the abstract, you will often find keywords, although not all journals include keywords for their articles. These keywords give you more clues about what the article is about. The keywords will often indicate which languages are being analyzed, which subfields of linguistics the article can be classified under, and which grammatical properties are being analyzed. The keywords for this paper are tense, pronouns, temporal semantics, and syntax–semantics interface. Here's what you can tell from the keywords:

We can confirm that the main topics of the paper are tense and pronouns (which we already figured out from the title and abstract).

The main subfield of this paper is probably semantics, but it is also relevant to syntax.

This paper also includes French keywords underneath the French abstract, under the label Mots-clés.
The body of the paper
The body of the paper will probably be divided into numbered sections and subsections. As mentioned in Section 2.6, morphosyntax papers don't usually have a formulaic structure. Instead, the structure will typically be described in an outline at the end of the introduction. Throughout the paper, you may find tables, graphs, linguistic data, syntax trees, and other formalisms. These will typically be numbered and discussed in the text. It is not considered acceptable to include data and formalisms without providing a description, relevant context, and explanation. If you are not sure how to interpret the data or formalisms, the discussion in the text may help you to understand them better.
The fine print

Footnotes and endnotes
Academic papers tend to have a lot of additional notes. If they are found at the bottom of each page throughout the paper, they are called footnotes. If they are found at the end of the paper, they are called endnotes. Papers will either use one or the other, not both. When you are writing a paper, don't go out of your way to include footnotes or endnotes. If it is important, it should be included in the main text, not in a note. In published papers, many of the footnotes or endnotes get added through the review process, as responses to the reviewers. Footnotes and endnotes may contain points of clarification or additional details. A lot of researchers hide their problems and limitations in their notes, so if you are writing a critique of a paper or looking for ideas for further research, reading the notes can be helpful. In some citation styles, references are included in the notes. As described in Section 2.7, we typically use in-text citations in linguistics instead.
Acknowledgements
There will typically be an acknowledgements section included in the article. This may be found as a footnote on the title or in a special section at the end of the article. If the researchers collected data using elicitation, their language consultants will often be thanked in the acknowledgements section, giving you more information about how the data was collected.
Abbreviations list
If there is glossed linguistic data, an abbreviations list should be included in the article. This may be found in a footnote on the first example or at the end of the article in a special section.
References
At the end of the article, there will be a references list. The references list should include the full bibliographic information of every source cited in the paper. The format of the references list used in the Canadian Journal of Linguistics is described in detail in Section 2.7.

Key takeaways

On the front page of an academic article, you should be able to find the title, the author's information, the bibliographic information of the article, a summary of the article called an abstract, and sometimes article keywords.

The main body of the paper will usually be divided into numbered sections and subsections. The organization of the paper will often be described at the end of the introduction.

Data and formalisms included throughout the paper will be numbered and should be described in the main text.

Footnotes and endnotes can be a good place to find problems with the paper or areas of further research.

Papers will often have acknowledgements and abbreviations list that can provide information to help you interpret the data in the paper.

Academic articles should always end with a references section that provide the full bibliographic information of all sources cited in the article.

Check yourself!
[h5p id="50"]
References and further resources

Sources for examples

Bjorkman, Bronwyn. 2022. Some structural disanalogies between pronouns and tenses. Canadian Journal of Linguistics 67 (Special Issue 3: Pronouns): 143-165.

6.10. Guided reading: Is n't a clitic or an affix?

juliedoner — Fri, 02 Aug 2024 05:09:28 +0000
In this section, we are going to read through an academic article together, Cliticization vs. inflection: English N'T by Arnold Zwicky and Geoffrey Pullum. You should be able to find the paper in your university library. At the end of this page, there is a glossary of technical terms used in this paper that haven't been covered yet in this textbook.
The first pass
As discussed in Section 6.9, the first thing you should do when reading an academic article is to read the abstract and keywords carefully. The abstract of this paper isn't labeled, but is found in a smaller font right under the authors' affiliations. This paper doesn't have keywords, so we can skip that step.
Questions to ask while you're reading the abstract:

Why do you think I picked this article to read? How is it related to the course content in general and to this chapter, specifically?

Which ideas and terms from this chapter show up in the abstract?

What year was this paper written?

Next, skim the introduction and conclusion. For this step in this paper, I would skim the unlabeled section at the beginning, Section 1: Background, and Section 5: Further Observations. When you're scrolling or flipping through the pages between sections 1 and 5, pay attention to the titles of the headers and the kinds of data. Since this paper has uninformative section headers, you may wish to read the first sentence or two of each section as well as the header. When you get to Section 5, you will probably notice that it is not a conclusion, but rather in-depth analysis and contextualization. At this point, I would start skimming to find the conclusion—it turns out the true conclusion is just the final paragraph of the paper.
Questions to ask while you're skimming the introduction, conclusion, headers, and data:

What is the main conclusion of this paper?

Which subfield(s) would you classify this paper into?

Is this paper more descriptive or theoretical? That is, does it describe a property of a particular language or more generally argue for a certain kind of model of Language?

How is this paper structured?

What kinds of data are found in this paper?

What technical terms terms show up a lot? Are there any that you don't know or that you need a refresher on?

How is this paper different than the papers that came before? What new ideas did it introduce?

Can you tell yet why this paper matters? What is its contribution to the field of linguistics?

The structure of Zwicky and Pullum (1983)

Here, I provide a reverse outline of this paper. A reverse outline is when you take a completed paper and you make an outline of it. This can be useful as a reading and study tool, but it can also be a useful to step for evaluating and revising the structure of paper drafts that you have written. 0. unlabeled (page 502)

Context: Sometimes you can be surprised even by well-studied languages.

Main claim: This paper discusses one surprise: that n't is an inflectional affix, not a clitic.

1. Background (pages 502-504)

Describes and motivates 6 criteria for distinguishing between clitics and inflectional affixes.

2. Criteria A-D (pages 504-505)

Demonstrates how the first four criteria apply to non-controversial examples of clitics and inflectional affixes.

These first four criteria are all related by being about how the clitics and affixes combine with their hosts and stems.

3. Criteria E-F (pages 505-506)

Demonstrates how the last two criteria apply to non-controversial examples of clitics and inflectional affixes.

These last two criteria are related by being explained if cliticization happens later than other syntactic and morphological operations.

4. The contracted negator n't (pages 506-510)

Applies the six criteria to n't.

Demonstrates that n't behaves like an affix and not a clitic for all 6 criteria.

5. Further observations (pages 510-512)

Uses cross-linguistic evidence to demonstrate that it is not surprising for there to be an inflectional affix marking negation.

Argues against the alternative explanation that n't is a special clitic.

The second pass
Now go back to the beginning of the paper and read it through from beginning to end. In this guided reading section, we will do a deep dive into parts of Section 4, which is both the most important part of the paper and also where most of the more complex ideas are found. I will leave the rest for you to read on your own.
Beginning of section 4, page 506
At this point of the paper, the authors have presented 6 criteria for distinguishing inflectional affixes from clitics and have illustrated how those criteria work for some non-controversial examples of both inflection and clitics, to establish that these criteria are valid. Now, they are using these six criteria to test a more controversial morpheme: negative n't, and will show that it behaves like an inflectional affix. Note that one of the authors of this paper, Zwicky, had previously adopted the opposite analysis—and has changed his mind when he considered the evidence more fully. This section doesn't go through the criteria in the same order as the rest of the paper. Instead, they start with Criterion E: "Syntactic rules can affect affixed words, but cannot affect clitic groups." Back on page 504, the reason for this criterion is explained. They are working under the assumption that the model of grammar is structured in such a way that all syntactic operations occur before any phonological operations, and that cliticization also occurs after all syntactic operations. On the other hand, word formation processes, including inflection, happen before syntax.

(1) Order of operations assumed by Zwicky and Pullum (1983)

Module of grammar Morphology → Syntax → Phonology

Operations inflection SAI cliticization

To test this criterion, they look at Subject-auxiliary inversion (SAI), which is the process involved in yes-no question formation. SAI is unquestionably a syntactic operation, and so it should occur after inflection but before cliticization, as shown in (1). This predicts that inflectional affixes should move with the auxiliary in yes-no questions, as shown in (2). Here, different inflectional forms of DO can undergo SAI: do, does, and did.

(2) a. Do you want a cookie?

b. Does he want a cookie?

c. Did you want a cookie?

It also predicts that clitics should not move with the auxiliary, which is also what happens, as shown in (3) (Zwicky and Pullums's (8)).

(3) a. You could've been there.

b. *Could've you been there?

What does n't do? It can move, so it behaves like an affix according to criterion E. This is shown in (4) (Zwicky and Pullums's (4)), where it undergoes SAI.

(3) a. You haven't been there.

b. Haven't you been there?

Assumptions about a model of grammar

As shown in (1), this paper assumes a certain order of operations in the grammar. First, words are built using morphological rules in the lexicon. Then, the words are combined into phrases and sentences in the syntax component. Finally, the sentences are interpreted by phonology, where it is decided how the sentence should be pronounced. These assumptions show up in subtle ways throughout the paper, for example in their use of the word lexicalization. They don't explicitly state their assumptions, because this was the model of grammar that most generative linguists in the 1980s had adopted. But this is not longer the case! In the early 1990s, two linguists, Morris Halle and Alex Marantz, proposed a new model of grammar called [pb_glossary id="4368"]Distributed Morphology[/pb_glossary] with different assumptions about the relationship between morphology and syntax. They noticed that there are many similarities between the rules that form words and the rules that form phrases, and so they argued that having one set of rules that takes care of both morphology and syntax is a simpler, and therefore better, model of grammar than one that has two separate components. Distributed Morphology is now a very popular model of the architecture of grammar, but it has still not been adopted universally. A few syntacticians still assume a separate morphological component. Meanwhile, more models hypothesizing different relationships between morphology and syntax continued to be proposed. If you would like to know more, the basics of Distributed Morphology are explained in a relatively straightforward way on this FAQ by Rolf Noyer.

Table 1, page 508
Zwicky and Pullum's Table 1 shows the complete list of auxiliaries that can host n't in the authors' grammars, providing evidence for criteria A, B, and C.. The table includes the bare form of the auxiliary, the pronunciation of the bare form, the form of the auxiliary with n't, and the pronuniciation of the form of the auxiliary with n't. Note that the transcriptions here aren't in standard IPA, which is common in older papers like this when typing unusual characters was more difficult.
Table 1: Non-standard transciption used in Zwicky and Pullum (1983)

Zwicky and Pullum's symbol IPA symbol Meaning

u
ʊ high back lax rounded vowel

š ʃ voiceless post-alveolar fricative

n̥ n̩ syllabic alveolar nasal

Criteria A states that "clitics can exhibit a low degree of selection with respect to their hosts, while affixes exhibit a high degree of selection with respect to their stems." In other words, clitics can usually attach to anything, but affixes are more picky. The fact that this table is even possible provides evidence that n't behaves like an affix—n't can only attach to auxiliaries and the full list of possible stems is short enough that it can fit on one page! Criteria B states that "arbitrary gaps in the set of combinations are more characteristic of affixed words than of clitic groups." This table shows that n't behaves like an affix because it cannot attach to auxiliaries like may and am. There is no phonological reason why may should not be able to host n't—mayn't is a possible word in English—making this an arbitrary gap. The auxiliary am is a bit more difficult to pronounce with n't, but some dialects do allow it, suggesting that, again, this is an arbitrary gap. Criteria C states that "morphophonological idiosyncrasies are more characteristic of affixed words than of clitic groups." In other words, there is more lexically conditioned [pb_glossary id="480"]allomorphy[/pb_glossary] for affixes than for clitics. This is shown throughout the table, in the many differences in pronunication between the bare form and the affixed form. For example, the vowel in do /du/ changes in don't /dont/ and the /l/ in shall /ʃæl/ is dropped in shan't /ʃænt/. There is also not a consistent pattern that can predict whether the negative form is one syllable or two—for can, both options are possible!
Criterion D, page 509
Criterion D states that "semantic idiosyncracies are more characteristic of affixed words than of clitic groups," which means that words with affixes are more likely to have special non-predictable meanings than combinations of clitics and hosts. In this section, Zwicky and Pullum show that not and n't don't have the exact same meanings in all contexts, but the difference is subtle. This difference in meaning shows that n't behaves more like an affix than a clitic. This section is one of the trickiest to understand, especially if you haven't studied much semantics before. The argument in this section depends on the notion of [pb_glossary id="4248"]scope[/pb_glossary], which is a word used to describe which parts of a sentence are modified by a semantic operator. In this case, the semantic operator is negation. Compare sentences (4a) and (4b). In (4a), the word not only modifies rotten. That is, the scope of negation is only the word rotten. In (4a), Hentie did eat a fruit, but the fruit was fresh. In (4b), on the other hand, the word not modifies the entire sentence; in other words, not takes scope over the entire sentence. In (4b), we do not know whether Hentie ate a fresh fruit. All we know is that the action of eating a rotten fruit did not occur.

(4) a. The fruit Hentie ate was not rotten.

b. It is not the case that Hentie ate a rotten fruit.

When there are two semantic operators in the same clause, they interact. Which order you compose them changes the meaning of the sentence. For example, in (5), we see the difference between must(not(P)) and not(must(P)). In semantics, P is used as a placeholder for a [pb_glossary id="4243"]proposition[/pb_glossary], which is a semantics term meaning a statement that can be true or false.

(5) semantic notation meaning English example

a. must(not(P)) It must be that case that not P. You must not come early. You are required to stay away until the appropriate time, because we will be fumigating the room, and it will not be safe to enter.

b. not(must(P)) It is not the case that P must be. It is not necessary for you to come early. I don't need your help to set up, but you can come early if you want to.

The difference in scope between not and n't is illustrated in (6) and (7) (from in-text examples from Zwicky and Pullum 1983: 509). The modal must not only has one possible scope meaning. The sentence you must not go home can only mean that you are not allowed to go to home. The sentence with n't instead of not has the same meaning.

(6) must(not(P)) not(must(P))

a. You must not go home. ✔︎ ✘

b. You mustn't go home ✔︎ ✘

The pattern is different with can, though. In these sentences, an accent (é) is used to mark stress. With the full-word form in (7a), the sentence is ambiguous. It can mean that you have the option of not going home, but it can also mean that you are not allowed to go home. The sentences with n't in (7b) and (7c), on the other hand, only have one possible meaning; they can only mean that you are not allowed to go home, which is the opposite scope order that was allowed with mustn't.

(7) can(not(P)) not(can(P))

a. You can nót go home. ✔︎ ✔︎

b. You cánnot go home ✘ ✔︎

c. You can't go home. ✘ ✔︎

The relationship between must not and mustn't is predictable—they share the same pattern as each other. In contrast, the relationship between can not on the one hand and can't and cannot on the other is not predictable. We then have a semantic idiosyncracy in the meaning of a word using n't, indicating that n't behaves more like an affix than a clitic.
Don't mix up preposition with an E and proposition with an O. Both are technical terms in linguistics, but are very different things! A preposition is a part of speech that typically indicates location, such as in, on, at, or under. A proposition, on the other hand, is a statement that can be true or false, such as I enjoy eating cake.

Reading with a writer's eye

As you read this paper, did you think of any ways that you would write it differently that would make it clearer and easier to read? Doing so is a good way to become a better writer. Here are some changes that I think would make this paper a bit easier to read:

I'd give the headers more transparent headings. For example, instead of labeling Section 2 "Criteria A-D," I'd label it something like "Relationship between a bound morpheme and its host."

Throughout Sections 2 and 3, I'd make sure there was at least one example of an inflectional affix and of a clitic for each criteria, as much as possible.

I'd repeat the relevant criteria in Section 4 so the reader doesn't need to scroll back to find them.

At the beginning of Section 4, I'd tell the reader I was discussing criterion E, instead of leaving it for the reader to infer.

What about you? Do you have any suggestions for how to make this paper clearer?

Glossary of terms used in this paper
The following is a list of technical terms used in this paper that have not yet been introduced in this textbook, as well as some advanced words used in scientific discourse that aren't quite technical terms. Agreement: When the features on one word gets transferred to or shared with another word. Auxiliary verbs: Verbs like have, be, and do that appear before a main verb. Conjunctive: Having the meaning of 'and'. Constituent: A group of words behaving like a syntactic unit. Copula: A verb used to introduce non-verbal [pb_glossary id="4142"]predicates[/pb_glossary], such as am in I am a linguist. Feed: When one rule or process must occur before another one, the first rule is said to feed the second. This is a term borrowed from [pb_glossary id="194"]phonology[/pb_glossary]. Finite verbs: Verbs which are inflected for [pb_glossary id="3550"]tense[/pb_glossary] (as opposed to infinitive verbs, which are not). Formative: A word you can use in morphology to describe a linguistic element when you don't want to be specific about whether it is a word, a morpheme, or something else. Government: A particular structural relationship in syntax, largely no longer in use. Emphatic stress: When you pronounce a word more forcefully for emphasis. When a word has a clitic variant, the clitic usually can't be used if there is emphatic stress on it. Idiosyncracy: A pattern or feature that is unique to an individual. In this paper, the individual is the morphological [pb_glossary id="3078"]stem[/pb_glossary]. Attachment of an [pb_glossary id="3077"]affix [/pb_glossary]to a stem is more likely to have exceptions or unusual patterns than attachment of a clitic to a host. Inchoative/causative verb: An inchoative is a verb expressing that a certain state has begun. A causative is a verb indicating that the subject caused a resulting state. For example, verbs such as moisten and soften mean 'cause to become moist/soft.' Lexical items: Items formed in the [pb_glossary id="1379"]lexicon[/pb_glossary]. That is, they are formed by morphological rules, not syntactic rules. Lexicalization: The process by which words are added into the [pb_glossary id="1379"]lexicon[/pb_glossary]. The paper seems to be referring, here, to how auxiliary + n't constructions were previously combined syntactically but were so frequent in combination that they got added to the lexicon. Once they had their own lexical entries, the combinations were able to pick up idiosyncratic meanings and forms. Monosyllables: Words which contain only one syllable. Neologism: Newly-formed word. Node: The point at which several branches come together in a syntax tree. Parochial: Limited in use and scope. Readjustment rule: A rule that adjusts the output of syntactic rules to prepare it for the phonology. Selection: When a word or morpheme is "picky" about what things can attach to it to form a new word or a phrase. Sentential negation: When the entire sentence is negated, not just a word or a phrase inside the sentence. Simple clitic: A clitic which has an equivalent full-word form that can appear in the same positions. Special clitic: A clitic which either has no equivalent full-word form or it has an equivalent full-word form that appears in different positions. Stipulate: To add an assumption to our model without any motivation other than that it makes the data work out. Subject-auxiliary inversion (SAI): The word order change involved in the formation of yes-no questions in English, in which the subject and the auxiliary swap positions. For example, the statement "You are awake" becomes the yes-no question "Are you awake?" when the subject you swaps position with the auxiliary are. Superlative: An inflectional affix on adjectives to mean 'most'. Syllabic: Behaving as the nucleus of a syllable (i.e., instead of a vowel). Syncategoremic: Something that doesn't quite have meaning on its own, but only has meaning when combined with something else. Transformation: A rule of grammar that rearranges the order of words, used in some grammatical models of the 1960s-1980s. They have generally been replaced by movement rules in many modern models of grammar.
Check yourself!
[h5p id="52"]
References and further resources

Academic sources

Noyer, Rolf. n.d. Distributed Morphology: Frequently Asked Questions list. https://www.ling.upenn.edu/~rnoyer/dm/home.html

Zwicky, Arnold, and Geoffrey Pullum. 1983. Cliticization vs. inflection: English N'T. Language 59 (3): 502-513.

8.1. Determining part of speech

juliedoner — Sat, 10 Aug 2024 22:02:39 +0000

The [pb_glossary id="616"]part of speech[/pb_glossary] of a word, also called its syntactic or lexical category, is a classification of its behaviour. Some examples of parts of speech include nouns, verbs, adjectives, adverbs, prepositions, and so on. You likely learned a little bit about these categories in school, with the definitions described in Table 1. These traditional definitions are also sometimes called semantic definitions, since they classify part of speech based on the meaning of words. But, as you will see shortly, the traditional semantic definitions have some problems. In linguistics, we don't use the traditional definitions to determine part of speech. Instead, we use distributional criteria, as will be discussed below.

Table 1: Traditional semantic definitions of parts of speech

Part of speech Traditional definition Examples

Noun A person, place, or thing child, school, toy, Canada

Verb Actions or states eat, sleep, grow, know

Adjective Describes quality, quantity, or extent rich, far, green, numerous

Adverbs Describes manner, quality, or degree quickly, very, carefully, partially

Prepositions Location or origin in, from, on, at

Problems with the traditional approach

The same word can belong to multiple different categories
One big problem with the traditional definitions of Table 1 is that the same word can belong to multiple different categories. For example, in (1a), the word work is a verb; in (1b), it is a noun; and in (1c), it is an adjective.

(1) a. We work at a joke factory.

b. Have you finished your work yet?

c. My work clothes are filthy.

All three instances of the word work in (1) have related meaning—they are all associated with some kind of labour. However, they have very different functions and behaviours in each sentence. Therefore, when we are determining the part of speech of a word, it is these functions and behaviours that we depend on to classify words.
Abstract nouns and stative verbs
Not all words are easy to classify according to the traditional semantic definitions, either. What part of speech is idea? You might know it is a noun, but how do you know that? It's definitely not a person or a place. It is debatable whether it's a thing. Or what about the noun destruction? That’s an action, so we might think it's a verb! The destruction of the city by Godzilla describes an event—it’s clearly not a person, a place, or a thing. However, the behaviour and function of destruction is that of a noun. Likewise, not all verbs are actions, and some verbs stretch the notion of state. The sentenceTasha owns a car contains the verbs owns.  Is owning an action or a state? Only under a very loose definition. What about the verb seems in Abigail seems tired. Is seem an action or a state?
Words can change category
Another piece of evidence that part of speech should not be determined by semantic definitions is that a word can change syntactic category over time without changing its meaning. One example of this is the word fun. For most speakers, fun is a noun. Compare the sentences in (2) and (3) and notice how fun has the same behaviour as other nouns, such as water. It does not share the same behaviour as adjectives, such as hard.

(2) a. That was a lot of fun.

b. That was a lot of water.

c. *That was a lot of hard.

(3) a. I had more fun than her.

b. I had more water than her.

c. *I had more hard than her.

For younger speakers and in informal speech, though, fun can also behave like an adjective. Like other adjectives, fun can take comparative and superlative inflection for these speakers, forming the words funner and funnest. For these younger speakers, fun can behave the same way as adjectives like hard, as shown in (4). Nouns, on the other hand, cannot appear in this kind of structure, as shown for water in (4b).

(4) a. Swimming is funner than running.

b. *Swimming is waterer than running.

c. Swimming is harder than running.

Different categories in different languages
Another piece of evidence that we cannot use meaning to define syntactic category comes when we compare languages. For instance, if you’ve studied a Romance language, you’ve learned that certain concepts which are expressed in English using adjectives are expressed in Romance using verbs and nouns. For example, if you want to express hunger in English, you would say I am hungry, using the adjective hungry. In contrast, in many Romance languages and in Swahili, you would say the equivalent of I have hunger, using a noun, as shown in (5). If we (as English speakers) were to use meaning to define word category, we might be tempted to call words like hambre, faim, and njaa adjectives meaning 'hungry.' But this would be wrong, because they actually behave as nouns, as the literal translations in (5) suggest. Meaning doesn’t help us to define categories within a language, and it doesn’t help us define across categories across languages.
People can identify the category of made-up words
Perhaps the most fascinating piece of evidence that we cannot use semantic criteria to classify part of speech comes from the fact that people can reliably identify the category of made-up words, even when they don't know their meanings. Consider below an excerpt from the poem Jabberwocky from Through the Looking-Glass by Lewis Carroll. These lines contain five made-up words, which are bolded below. (6) And as in uffish thought he stood, The Jabberwock, with eyes of flame, Came whiffling through the tulgey wood, And burbled as it came! Even though we don't know what the words mean, we can tell that uffish and tulgey are likely adjectives, a Jabberwock is likely a noun, and whiffling and burbled are likely nouns. We cannot be using semantic criteria, since we don't even know what these words mean, so we must be using some other kind of evidence to determine the part of speech of these words.
Distributional criteria
If we cannot use semantic criteria, then what do we use? We use three criteria for determining part of speech:

Syntactic distribution, which is the words that it can appear adjacent to.

[pb_glossary id="1723"]Inflectional[/pb_glossary] morphology, which is how the word can be inflected.

[pb_glossary id="1724"]Derivational[/pb_glossary] morphology, which are the derivational morphemes that it contains.

Let's look again at the stanza from the Jabberwock to see these criteria in action. (7) And as in uffish thought he stood, The Jabberwock, with eyes of flame, Came whiffling through the tulgey wood, And burbled as it came! We can tell that Jabberwock is a noun because it appears after the word the, which is an example of syntactic distribution. For an example of inflectional morphology, consider the word burbled. It takes the past tense marker -ed, which is a property of verbs. Finally, uffish uses the suffix -ish, which is a derivational morpheme that appears on adjectives. At this point you may be wondering if this is circular reasoning. We have decided to call things nouns because they behave like nouns. And we have decided how nouns behave by looking at nouns. The observation here is that there are groups of words that have similar behaviour. The names of the categories are essentially arbitrary: we could call them blickershnitzels, and the observation would still be true and useful. Table, dog, teacher, brick, honesty, and so on, all behave in a similar fashion in some respects, and we give this group of words the name noun. Lick, sit, own, jump, describe, believe, and so on, also all behave in a similar fashion in some respects, and we give this group of words the name verb. When we say that something is a noun and has category N, all we’re saying is that that thing acts like the other noun-words. Likewise, when we say that something is a verb, we’re simply saying that that thing acts like the other verb-words. In the next few sections, we will go through the common syntactic, inflectional, and derivational properties of the different parts of speech. You can use these different behaviours to classify words by part of speech. However, as someone who is able to use words in different languages, you already know these properties, at least subconsciously. We just need to learn to be consciously aware of these patterns. We can also look at data from a language we don't speak to find patterns of behaviour of different words and identify parts of speech in these other languages. Of course, different languages have different word order, as well as different morphological patterns, and so the criteria won't be exactly the same across languages. Some linguists question whether all languages have the same categories at all! But there does seem to be some common patterns across languages. Many of the criteria we use in one language are parallel to similar criteria in other languages.
Categorizing phrases
Individual words are not the only things that have part of speech. Groups of words that behave as units—called phrases—also have distributional patterns that we can use to categorize them. Plus, there’s a pattern. Phrases have a head—the main word of the phrase—and the part of the speech of the head of the phrase will predict the distributional patterns of the phrase. For example, a phrase headed by a noun is called an NP (short for noun phrase). Any NP can be the subject of a sentence, as shown in (8), even if we add more words to the phrase, or if we switch the head noun for a a different head noun. (8) a. [the apple] is red b. [the sweet apple] is red c. [the apple on the ground] is red d. [the apple that fell off the tree] is red e. [some paint] is red f. [my bicycle] is red Likewise, any phrase headed by a verb is a VP, any phrase headed by a preposition is a PP, and so forth.

Key takeaways

We classify words based on their behaviour and function. These classifications are called part of speech or syntactic category.

Traditional definitions of parts of speech depend on semantic criteria, but there are several difficulties with determining part of speech based on meaning.

Linguists use syntactic distribution, inflection, and derivation to classify words into part of speech.

Phrases can also be classified by the part of speech of their head and share the distribution of other phrases with the same kind of head.

Check yourself!
[h5p id="53"]
References and further resources

Attribution

Portions of this section are adapted from the following CC BY NC SA source:

↪️ Gluckman, John. n.d. Chapter 1: Syntactic categories. The science of syntax. https://pressbooks.pub/syntax/chapter/chapter-2-parts-of-speech/

8.2. Nouns

juliedoner — Sat, 10 Aug 2024 22:03:00 +0000
We can identify nouns based on their inflectional morphology, derivational morphology, and syntactic distribution. In terms of inflectional morphology, nouns may inflect for person, number, gender, and/or case. In English, nouns inflect for number, with singular and plural forms. English pronouns also have person and gender distinctions, in addition to number. There are many different kinds of derivational morphemes that modify nouns. Some examples of derivational morphemes in English are shown below in Table 1.
Table 1. English derivational morphemes

suffix examples

a. -ment employment, basement

b. -ness friendliness, emptiness

c. -ity sincerity

d. -ty certainty

e. -(t)ion devotion

f. -ation expectation

g. -ist specialist, linguist

h. -ant attendant,

i -ery shrubbery, mockery

j. -ee employee, invitee

k. -ship hardship, friendship

l. -aire billionaire,

m. -acy advocacy

n. -let piglet

o. -ling underling

p. -hood neighbourhood,

q. -ism socialism

r. -ing fencing

(adapted from Carnie 2021: 47)
In terms of syntactic distribution, noun phrases typically show up in argument positions, both the arguments of verbs and the arguments of prepositions, as shown in (1). (1) a. The apple is red. subject b. I ate the apple. direct object c. I gave the apple a shine. indirect object d. There was a worm in the apple. object of preposition Nouns can be modified by determiners and adjectives, and so their syntactic position can often be described with reference to determiners and adjectives. In English, nouns can appear after either determiners or adjectives, or appear on their own, as shown in (2). (2) a. I ate the apple. after a determiner b. I ate the red apple. after an adjective c. I ate apples. on its own
Inflectional values on nouns
The most common kinds of inflection found on nouns include case, person, number, and gender. The latter three of these (person, number, and gender) tend to have close interactions and the group is often referred to as φ-features, named after the Greek letter phi.
Case
Case is used to mark the role a noun phrase plays in a sentence. Some languages do not mark case at all, some mark case on on only some nouns, some mark case on all nouns, and some mark case on determiners. In English, we have case distinctions only on pronouns. We use the nominative case form in the subject position of finite clauses, the genitive case for possessives, and the accusative case form on the objects of active voice clauses. In English, the accusative case also acts as the default and shows up in additional contexts, as well. (3) a. I saw them. 1SG.NOM (no gloss for saw) 3PL.ACC b. They saw me. 3PL.NOM (no gloss for saw) 1SG.ACC The pronoun paradigm for English is shown in Table 1. In addition to case, English pronouns also have person, number, and gender distinctions, which we will discuss later in this section.
Table 2. English personal pronouns

Features Morphological forms

Person Number Gender nominative case accusative case genitive case

1st singular -- I me mine

2nd singular -- you you yours

3rd singular masculine he him his

3rd singular feminine she her hers

3rd singular inanimate it it its

3rd singular non-binary or unspecified they them theirs

1st plural -- we us ours

2nd plural -- you you yours

3rd plural -- they them theirs

In English, we use the accusative case for both direct and indirect objects, but in many other languages, they use the accusative case only for direct objects, and have another case, the dative case, for indirect objects. German, as shown in (4), is one example of a language that uses dative case. The example in (4) also illustrates how in some languages, including German, case is marked on determiners rather than on the noun itself. (4) Sie hat dem Mann das Buch geschenkt. she.NOM has the.M.DAT man the.N.ACC book given 'She has given the man the book.'' German (Beermann 2001, as cited in Alexiadou et al. 2010: 2). Some languages have extensive case marking systems that they use in addition to or instead of prepositions. Hungarian is an extreme example of this, with 21 cases, as shown in (5).

(5) Hungarian

hajó nominative

hajó-t accusative

hajó-ban inessive

hajó-ból elative

hajó-ba illative

hajó-n superessive

hajó-ról delative

hajó-ra sublative

hajó-nál adessive

hajó-tól ablative

hajó-hoz allative

hajó-ig terminative

hajó-nak dative

hajó-val instrumental-comitative

hajó-képp formal

hajó-ul essive

hajó-ként essive-formal(-similitive)

hajó-vá translative-factitive

hajó-ért causal-final

hajó-nként distributive

hajó-stul sociative

(Tompa 1968: 206-209, as cited in Iggesen 2013)
Languages such as English, German, and Hungarian have what is called nominative-absolutive case alignment, which means that we use the same case marker for the subjects of transitive and intransitive verbs, and a different case marker for the objects of transitive verbs. This is shown in Figure 1.

(6) a. She saw her.

b. She slept.

[caption id="attachment_5076" align="aligncenter" width="830"] Figure 1: A schematic representation of nominative-accusative case alignment.[/caption] On the other hand, other languages mark the sole argument of intransitive verbs with the same case as the object of transitive verbs, which is called absolutive case. The case that is used to mark only subjects of transitive verbs is called ergative case. Ngarinyman, a Pama-Nyungan language of Australia, is an example of an ergative-absolutive language. For example, sentence (7a) is intransitive, and so the sole argument garu 'child' appears in absolutive case. In (7b), when the same verb is used transitively, the subject, yawardarlu 'horse' appears in ergative case. The object in (7b) appears only as agreement on the verb.

(7) Ngarinyman

a. Bard wan -inya garu.

fall.down change.location -pst child.abs

'The kid fell down.'

(7) b. Yawarda -rlu bardba -yi yuwa -ni.

horse -erg fall.down -1sg.obj put -pst

'The horse threw me off.'

(Denniss 2023: 142)
We can see another transitive-intransitive pair in (8). Sentence (8a) is intransitive, and the sole argument warlagu 'dog' is again marked in absolutive case. In the transitive version, the subject warlagu 'dog' is marked with ergative case while the object, janga 'woman' is marked in absolutive.

(8) Ngarinyman

a. Rungab garr -any warlagu.

bark be.at -pres dog.abs

'The dog is barking.'

(8) b. Warlagu -rlu rungab nya -nga jan.ga.

dog -erg bark perceive -pres woman.abs

'The dog barked at the woman.'

(Denniss 2023: 145-146)
[caption id="attachment_5077" align="aligncenter" width="862"] Figure 2: A schematic representation of ergative-absolutive case alignment.[/caption] Many languages use a combination of nominative-accusative and ergative-absolutive case. This is called split ergativity. The contexts in which each case system is used can be determined by tense, person, or other factors. The names and meanings of some common cases are listed in Table 3.
Table 3. The meanings of some common cases.

Case Meaning

nominative subject of a finite verb

accusative object of an active verb

ergative subject of a transitive verb

absolutive argument of an intransitive verb or object of a transitive verb

dative indirect object of a verb

genitive possessor of another noun

ablative movement away from

vocative addressee

locative location

instrumental a tool or means used to perform the action

Person
Nouns in most languages are categorized as first, second, or third person. First person nouns refer to the person speaking or signing. Second person nouns refer to the person being addressed. Third person nouns are everyone and everything else. First person plural pronouns typically mean the person speaking or signing and people associated with them, rather than meaning that multiple people are speaking or signing (although, in rare cases, it can mean that, too). On the other hand, second person plural pronouns can either mean that you are addressing a group or that you are referring to the person you are addressing and the people who are associated with them. Some languages make a distinction between inclusive and exclusive first person plural pronouns. Inclusive first person pronouns include the addressee in their reference (meaning 'us and you'), while exclusive ones do not (meaning 'us and not you'). Niuean, an Austronesian language spoken on the island of Niue, is an example of a language with inclusive and exclusive pronouns. Niuean's pronoun paradigm is shown in Table 4.
Table 4. Niuean personal pronouns (Seiter 1980: 49)

singular dual plural

1st exclusive au maua mautolu

1st inclusive taua tautolu

2nd koe mua mutolu

3rd ia laua lautolu

Some languages also have an obviative, which is sometimes referred to as the fourth person. In languages that have an obviative, only one third person noun (usually the most topical one) will be marked as proximate while the remaining 3rd person nouns will be marked with the obviative, as shown below in (9). In the Oji-Cree example in (9a), the subject is in the first person, so the third person object awaasihs 'the child (proximate)' is marked with the proximate, not the obviative. But in (9b), there is a third person subject. Therefore, the third person object awaasihsan 'the child (obviative)' is marked in the obviative with the suffix -an. Oji-Cree is an Algonquian language spoken in Manitoba and Northern Ontario. (9) Oji-Cree (Oxford 2017: 2) a. Niwaapamaa awaasihs. ni-waapan-aa awaasihs-∅ 1-see-DIR- child-PX.SG 'I see the child.' b. Owaapamaan awaasihsan O-waapam-aa-n awaasihs-an. 3-see-DIR-OBV child-OBV 'S/he sees the child.'
Number
Number indicates how many instances of the noun there are. In English, we have two number categories: singular and plural, although singular is unmarked. In languages like English which have both a singular and plural, the singular means that there is exactly one while the plural means that there is more than one. The plural marker in English is-s, although there are also a number of nouns which take irregular plurals, such as children, feet, and cacti. (10) a. cat singular b. cat-s plural Some languages also have a dual marker, which indicates that there is exactly two instances of the noun. In languages with a dual marker, singular means exactly one, dual means exactly two, and plural means three or more instances of the noun, as shown in (11). (11) a. taalib student.M.SG 'student' b. taalib-een student.M.DU 'two students' c. taalib-iin student.M.PL 'students' Gulf Arabic (Holes 1990: 148, as cited in Aikhenvald 2018: 2) A few languages have a trial, which marks exactly three, and in which case the plural would be used for four or more instances of the noun, as shown in (12). (12) a. ngayi 'I (singular)' b. ngarrgu 'we two' (dual) c. ngarrgunime 'we three' (trial) d. ngagurr 'we (many)' (plural) Ngan'gityemerri (Reid 1990: 118-118, as cited in Aikhenvald 2018: 2) Even fewer languages go higher than three, but languages with a quadruple or a quintuple form do exist. One is American Sign Language. Some languages also have a paucal, as shown in (13), which indicates that there are only a few. (13) a. áine ŋára woman that:3SG 'that woman' (singular) b. áine ŋara-díaru woman that-dual 'those two women' (dual) c. áine ŋara-díato woman that-paucal 'those few women' (paucal) d. áine ŋára-di woman that-PL 'those women' (plural)
Manam (Lichtenberk 1983: 67, as cited in Aikhenvald 2018: 3)
Some nouns arbitrarily take a plural marker no matter how many of the item there are, such as pants, scissors, and (eye)glasses. Often, these are items that typically come in two roughly equal or symmetrical parts, but not all items which consist of a pair require the plural marking, showing that it is an arbitrary grammatical rule that must be encoded in the lexicon. For example, bicycle, eyes, chopsticks, or socks all also typically come in pairs or two parts but have both singular and plural forms.
Gender
In grammar, gender refers to noun classes. In languages with extensive gender systems, all nouns belong to a noun class which is part of the lexical entry of that noun. Other words in the phrase will agree with the gender of the noun. For example, Russian nouns trigger agreement on the verb, as shown in (14). With a masculine noun such as žurnal 'magazine' in (14a), the verb appears in the masculine bare form. With a feminine noun such as kniga 'book' in (14b), the verb appears with the feminine suffix -a. Finally, with a neuter noun such as pis'mo 'letter' in (14b), the verb appears with the neuter suffix -o. (14) a. Žurnal ležal na stole. magazine lay.M on table. 'The magazine laid on the table.' b. Kniga ležal-a na stole. book lay-F on table 'The book lay on the table.' c. Pis'mo ležal-o na stole. letter lay-N on table. 'The letter lay on the table' Russian (Corbett 2013) In some languages, the noun class is marked explicitly with a morpheme on the noun, but in other languages, it only shows up on other words that agree with the noun, as with the Russian example in (14). In SiSwati, a Bantu language spoken in Eswatini and South Africa, the noun class markers are indicated both on the noun and the words that agree with it, such as adjectives. This is illustrated below in (15) for class 2 marker ba- in (15a), class 3 marker li- in (15b) and class 7 marker si- in (15c). Note that, unlike in many Indo-European languages, the noun classes in Bantu languages like SiSwati are not based on gender. Instead all of the human nouns belong to class 1 (when singular) and class 2 (when plural). (15) a. ba-fati la-ba-k^hulu Cl2.PL-woman ADJ-Cl2.PL-big 'big women' b. li-ɬombe le-li-k^hulu Cl3.SG-shoulder ADJ-CL3.SG-big 'big shoulder' c. si-lwane le-si-k^hulu Cl7.SG-animal ADJ-Cl7.SG-big 'big animal' In some languages, nouns belonging to different classes have distinct inflectional paradigms. This occurs in Russian, for example, as shown in Table 5. Notice how each gender has different suffixes for each case category.
Table 5. Inflectional paradigms of different gender singular nouns in Russian

Inflectional class Gender Nominative Accusative Genitive Dative Instrumental Locative Gloss

I masculine žurnal žurnal žurnala žurnalu žurnalom žurnale 'magazine'

II feminine gazeta gazetu gazety gazete gazetoj gazete 'newspaper'

III feminine kost' kost' kosti kosti kost'ju kosti 'bone'

IV neuter pis'mo pis'mo pis'ma pis'mu pis'mom pis'me 'letter'

The different meanings of sex and gender

Many people confuse the terms sex and gender. To make matters even more confusing, linguists use gender in more than one way. Let's define these terms precisely so that we can be clear about what we mean.
Sex
Definition: A set of biological attributes in humans and animals, including hormone level and expression, reproductive anatomy, chromosomes, and gene expression (CIHR 2023). Categories: Sex is usually divided into male and female, but there is variation in how these different biological attributes are expressed, including some who have attributes of both sexes or neither (Fausto-Sterling 2000). Property of: People and animals.
(Personal) gender
Definition: The socially constructed roles, behaviours, and expressions of a person based on how people perceive themselves and interact with others (CIHR 2023). Categories: Some common gender categories include woman, man, girl, boy, or non-binary person (CIHR 2023). Property of: People and maybe animals.
Semantic gender
Definition: Nouns that refer to people or animals sometimes carry semantic gender, which means that the personal gender of the person it refers to (or the sex of an animal) is encoded in the meaning of the word. This includes words like sister and brother, for which the semantic gender is part of the root word, but also words like duke and duchess, which include semantic gender in a separate morpheme. An increasing number of non-binary forms are being coined, such as nibling for the child of a sibling. Categories: Depending on the context, these may be referred to as male, female, and unspecified, or as masculine, feminine, and epicene. Property of: Words, mostly nouns. Morphemes encoding primarily semantic gender are usually derivational morphemes.
Grammatical gender
Definition: Grammatical gender refers to noun classes in grammar. Unlike semantic gender, these noun classes may trigger agreement in other words or determine which inflectional paradigm a word uses. Categories: In some languages, including many Indo-European languages, grammatical gender has distinctions between masculine and feminine or between masculine, feminine, and neuter. Other languages, including many Algonquian languages, have distinctions between animate and inanimate nouns. Languages in the Bantu family have approximately 16 noun classes (half singular and half plural), and all humans belong to the same noun class, as we saw for SiSwati above in example (15). Property of: Words, mostly nouns. Morphemes encoding grammatical gender are usually inflectional morphemes.

In languages which use grammatical gender, many nouns may have a grammatical gender which corresponds to the semantic gender of the entity to which it refers, but there are also nouns which do not have semantic gender and are assigned grammatical gender arbitrarily. For example, milk does not have semantic gender, and it is arbitrarily assigned masculine gender in French and feminine gender in Spanish, as shown in (16). (16) a. le lait DEF.SG.M milk 'the milk' French b. la leche DEF.SG.F milk 'the milk' Spanish There are are even nouns that have a semantic gender that doesn't match its grammatical gender. One rather well-known example of this is the German word mädchen 'girl' which is semantically female but grammatically neuter. (17) das mädchen NOM.DEF.SG.N girl 'girl' German The distinction between semantic and grammatical properties can also be made for other inflectional categories, including number (as we saw earlier for pants and scissors) and animacy. For example, in Meskwaki, an Algonquian language from the Iowa region, most plants use inanimate gender in everyday use, but may be marked with animate gender when they are being spoken to or in religious contexts. This is illustrated below for strawberry in (18). (18) a. ahteːhimin-i strawberry-SG.INAN 'strawberry (inanimate)' b. ahteːhimin-a strawberry-SG.AN 'strawberry (animate), when addressed' Meskwaki (Dahlstrom 1995: 58) On the other hand, the noun raspberry is always animate—even in everyday contexts—and cannot appear with inanimate gender. (19) a. wiːtawiːh-a raspberry-SG.AN 'raspberry' b. *wiːtawiːh-i raspberry-SG.INAN Meskwaki (Dahlstrom 1995: 58)

Singular you and they

In English, the pronoun you is used for both the singular and plural, but this was not always the case. You started out as a plural pronoun, and the pronouns thou (nominative), thee (accusative), and thy (genitive) were the singular pronouns. Here are some examples of the archaic singular pronoun from the Shakespeare play Hamlet. (11) a. Thou seest the heavens (Act II Scene IV) b. I'll give thee a wind. (Act I Scene III) c. ...and everyone did bear thy praises in his kingdom's great defence (Act I Scene III) In Early Modern English, at the time of Shakespeare, you could be used as a plural pronoun or it could be used as a singular pronoun in more formal contexts and to show respect to the addressee. Over time, the polite you form became more and more common. Some prescriptivists didn't like this change and complained about it, as illustrated by the quote below.
There has never been a greater perversion of language than the using the pronoun you, in the place of thou. [...] Using you for thou causes us to depart from some of the fundamental rules of grammar. For example: we are taught that "a verb should agree with its nominative number and person."
-Thomas 1857: 114-115

Now, in modern English, you is firmly established as a singular pronoun, replacing thou/thee/thy entirely. Some dialects have even coined new plural pronouns such as y'all or youse to re-establish the distinction between second person singular and plural. Using the plural pronoun to show respect to single addressees is also used in other languages. For example the plural second person pronoun vous is used as a polite form to address single addressees in French, instead of the singular second person pronoun tu, which, depending on the dialect, is used only for close friends and family. The pronoun they also originated as a plural pronoun but has been used as a singular pronoun in certain contexts since at least the 1400s (Balhorn 2004). Since the other third person singular pronouns specify gender or animacy, singular they is often used when the gender is unknown or irrelevant, as in (12). (12) a. (Seeing an unidentified distant figure ) They're waving at us. b. (After answering a phone call) They had the wrong number.
(Bjorkman 2017: 1)

Singular they was also used when it referred back to a noun modified by a [pb_glossary id="5728"]quantifer[/pb_glossary], even if the gender was specified, as shown through the historical examples in (13). (13) a. Shakespeare (A Comedy of Errors, 1623) There's not a man I meet but doth salute me As if I were their well-acquainted friend b. Austen (Pride and Prejudice, 1813) Both sisters were uncomfortable enough. Each felt for the other, and of course for themselves.
(Bjorkman 2017: 3)
More recently, they has begun to be used as a singular pronoun for specific people whose gender is known. Some modern English speakers accept neither the sentences in (14) or (15). Some accept the sentences in (14) but not (15), and some accept all of the sentences in both (14) and (15). Sentences that are grammatical for some people but not all are marked with a percentage sign (%). Subscript letters, called co-indexes, are included beside nouns to show that they refer to the same entity. (14) a. %The professor_i said they_i cancelled the exam. b. %Our eldest child_i broke their_i leg. c. %I'll let my cousin_i introduce themselves_i.
(Bjorkman 2017: 2)
(15) a. %Janet_i said they_i cancelled the exam. b. %Thomas_i broke their_i leg. c. %I'll let my sister_i introduce themselves_i.
(Bjorkman 2017: 2)

Several different researchers have suggested that the different judgment patterns for sentences like those in (14)-(15) are the result of a few small differences in people's grammar. Bjorkman (2017) argues that some people have a grammar that requires referential nouns to be specified for gender, even if it is not morphologically marked, while others have a grammar that allows nouns to be unspecified for gender. Konnelly and Cowper (2020) argue that there are three distinct grammars for they in competition, representing different stages of a change in progress, as described in (15). (15) a. Stage 1 singular they: (quantified antecedent, or antecedent of unknown/irrelevant gender), as shown in (12)-(13) b. Stage 2 singular they: antecedent of known gender, but ungendered description//name, as in (14) c. Stage 3 singular they: antecedent of any gender, no restriction on description/name, as in (15) Camilliere et al. (2021) did a sociolinguistic study that shows that people's linguistic behaviour does cluster into the three distinct patterns proposed by Konnelly and Cowper. How people rate singular they in different contexts is correlated with social attitudes toward transgender people and prescriptivist language, as well as age and social distance between the speaker and the referent of the pronoun. Some people feel like gendered pronouns like she or he do not fit them and may request that others refer to them by using they. We can show respect for them by using the pronouns that they request. However, sometimes even people who want to use singular they to refer to specific people may struggle to because their grammar is more restrictive, and they may find the change in usage difficult to acquire at first. The good news, though, is that learning and internalizing a new grammar is entirely possible with practice! Kirby Conrod, a linguist who uses they pronouns, gives some tips on how to do that in their blog.

Classifiers
Classifiers are similar to gender, in that nouns are lexically assigned to different groups, often based on some semantic criteria. However, unlike gender, classifiers do not trigger agreement on other words and there may be hundreds of classifiers in use in a given language. Mandarin is one example of a language that uses noun classifiers. In Mandarin, nouns must always be marked with a classifier when they are combined with a numeral, as shown in (20). The classifiers can indicate either individual objects, as in (20a), or containers, as in (20b) and (20c). (20) a. liang ben shu two CL book 'two boxes' b. liang xiang shu two CL:box book 'two boxes of books' c. liang ping jiu two CL:bottle wine 'two bottles of wine' Mandarin (Zhang 2007: 45) Classifiers in Mandarin often encode the shape of the object. For example, the classifier tiao indicates long and rope-like objects. Words such as she 'snake', tui 'leg', kuzi 'pair of pants', he 'river', and bandeng 'bench' all appear with the classifier tiao (Zhang 2007: 44). Classifier ke is used for small, solid, and kernel-like objects, such as huangdou 'soybean', ya 'tooth', dingzi 'nail', and zidan 'bullet' (Zhang 2007: 46). Classifiers can also appear on the verb in some languages, to indicate the semantic properties of one of its arguments. This is how handshape classifiers are often used in American Sign Language. It also occurs in spoken language. For example, in the Papuan language Waris, the classifier put- 'round objects' indicates that the object argument, a coconut, is round. (21) sa ka-m put-ra-ho-o coconut 1SG-to VCL:ROUND-get-BENEFACT-IMPERATIVE 'Give me a coconut' (lit. coconut to-me round.one-give) Waris (Aikhenvald 2000: 3)
Derived nouns
There are some derivational morphemes that are common on nouns cross-linguistically, which are summarized in Table 6. But there are also many, many possible derivational meanings that might only show up in a single language. Some derivational morphemes change the part of speech of their stem (as in Table 7 and Table 8, which list derivational morphemes that change verbs and adjectives into nouns), whereas others keep the same part of speech (as in Table 9).
Table 7. Deverbal nouns (V → N) (adapted from Haspelmath and Sims 2010: 87)

agent noun refers to the subject of the verb English drink → drink-er

Arabic ħamala 'carry' → ħammaal 'carrier'

patient noun refers to the object of the verb English invite → invit-ee

instrument noun refers to an instrument used to perform the verb English blend → blend-er

Spanish picar 'mince' → pica-dora 'meat grinder'

action noun refers to the action of the verb English destroy → destruction

Arabic otkry-t' 'discover' → otkry-tie 'discovery'

Table 8. Deadjectival nouns (A → N) (adapted from Haspelmath and Sims 2010: 87)

quality noun refers to the quality of having the property of the adjective English bold → bold-ness

Japanese atarasi-i 'new' → atarasi-sa 'newness'

person noun refers to an entity with the property of the adjective English short → short-y

Russian umn-yj 'smart, clever' → umn-ik 'clever guy'

Table 9. Denominal nouns (N → N) (adapted from Haspelmath and Sims 2010: 87)

diminutive noun refers to a smaller, younger, cuter, or more intimate version of the noun English basin → basin-ette

Spanish gat-o 'cat' → gat-it-o 'little cat'

augmentative noun refers to a larger version of the noun English church → megachurch

Russian borod-a 'beard' → borod-išča 'huge beard'

status noun refers to the having the status or rank of the noun English child → child-hood

→

inhabitant noun refers to one who inhabits the noun English Canada → Canad-ian

Arabic Miṣr 'Egypt' → miṣr-iyyu 'Egyptian'

female noun refers to a noun having feminine semantic gender English duke → duchess

German König 'king' → König-in 'queen'

Key takeaways

Cross-linguistically, nouns may inflect for person, number, gender, and or case.

Person marks whether the noun is a conversation participant.

Number marks how many of the noun there are.

Gender marks the inflectional noun class of the word.

Case marks the role of the noun in the sentence.

Nouns typically show up in argument position of a verb or the object of a preposition.

Nouns may be modified by determiners or adjectives.

There are many derivational nouns that occur on nouns, including some that turn adjectives and verbs into nouns.

Check yourself!
[h5p id="60"]
References and further resources

For a general audience

Canadian Institutes of Health Research. 2023. What is gender? What is sex? https://cihr-irsc.gc.ca/e/48642.html

Conrod, Kirby. 2020. Pronouns 101: Introduction to your loved one's new pronouns. Medium. https://kconrod.medium.com/pronouns-101-introduction-to-your-loved-ones-new-pronouns-3fef080266d0
Fausto-Sterling, Anne. 2000. Sexing the Body. New York: Basic Books.
McCulloch, Gretchen, and Kirby Conrod. 2020. The grammar of singular they: Interview with Kirby Conrod. Lingthusiasm episode 43. https://lingthusiasm.com/post/615600862742609920/lingthusiasm-episode-43-the-grammar-of-singular

Thomas, P.B. 1857. Thou and you. The R.I. Schoolmaster 3 (4): 114-115.

For students

Haspelmath, Martin and Andrea Sims. 2010. Understanding Morphology. London: Hodder Education.

Academic sources

Aikhenvald, Alexandra. 2000. Classifiers: A typology of noun categorization devices. Oxford: Oxford University Press.

Aikhenvald, Alexandra. 2018. Number systems in grammar - position paper. Langugae and Culture Research Centre 2018 Workshop. https://www.jcu.edu.au/__data/assets/pdf_file/0004/789700/Number-systems-in-grammar-Position-paper.pdf

Balhorn, Mark. 2004. The rise of epicene they. Journal of English Linguistics 32 (2): 79-166.

Bjorkman, Bronwyn. 2017. Singular they and the syntactic representation of gender in English. Glossa 2 (1): 80.

Camilliere, Sadie, Amanda Izes, Olivia Leventhal, and Daniel J. Grodner. 2021. They is changing: Pragmatic and grammatical factors that license singular they. Proceedings of the Annual Meeting of the Cognitive Science Society 43: 1542-1548. https://escholarship.org/uc/item/3tc9s9b0

Corbett, Greville. 2013. Number of genders. In WALS, ed. Matthew Dryer and Martin Haspelmath. Zenodo. https://wals.info/chapter/30

Dahlstrom, Amy. 1995. Motivation vs. predictability in Algonquian gender. In Papers of the 26th Algonquian Conference, ed. David Pentland. Winnipeg: University of Manitoba. 52-66.

Denniss, Jessica. 2023. Non-conflating roots: The syntax of light verbs and complex predicates in Ngarinyman. PhD thesis, University of Toronto.

Iggesen, Oliver. 2013. Number of cases. In WALS, ed. Matthew Dryer and Martin Haspelmath. Zenodo. https://wals.info/chapter/49
Konnelly, Lex, & Elizabeth Cowper. 2020. Gender diversity and morphosyntax: An account of singular they. Glossa 5 (1): 40.
Oxford, Will. 2017. Proximate DP, obviative KP: Balancing the morphosyntax and pragmatics of obviation. Proceedings of the 2017 annual conference of the Canadian Linguistic Association.

Seiter, William. 1980. Issues in Niuean syntax. New York: Garland Press.
Thomas, P.B. 1857. Thou and you. The R. I. Schoolmaster 3 (4): 114-115. https://www.jstor.org/stable/44786667
Zhang, Hong. 2007. Numeral classifiers in Mandarin Chinese. Journal of East Asian Linguistics 16 (1): 43-59.

8.4. Adjectives and adverbs

juliedoner — Sat, 10 Aug 2024 22:03:40 +0000
As with the other parts of speech, adjectives and adverbs can be identified based on their morphological and syntactic distribution.
Adjectives

Inflection on adjectives
Many adjectives inflect into comparative and superlative forms. The comparative means to a greater degree, while the superlative means to the utmost degree. Some examples of adjective paradigms are shown in Table 1. In English, shorter words tend to take the comparative suffix -er and the superlative suffix -est, while longer words use more and most periphrastically.
Table 1. The inflectional paradigms of some English adjectives

base form comparative superlative

happy happi-er happiest

black black-er black-est

good better best

quiet quiet-er quiet-est

intelligent more intelligent most intelligent

beautiful more beautiful most beautiful

Derivation of adjectives
Some common derivational morphemes that apply to adjectives are listed in (1)

(1) a. -ing the dancing cat

b. -ive indicative

c. -able readable

d. -al traditional

e. -ate intimate

f. -ish childish

g. -some tiresome

h. -(i)an reptilian

i. -ful wishful

j. -less selfless

k. -ly friendly

l. un- unhappy

(adapted from Carnie 2021: 48) Cross-linguistically, derivational morphemes that form adjectives commonly come from verbs, nouns, or other adjectives. Two common types of deverbal adjectives (V --> A), shown with examples in Table 2, are adjectives that describe the verbal action as possible, called the facilitative, or that attribute the verbal action as typical of a person, called the agentive.
Table 2. Deverbal adjectives (V → A) (adapted from Haspelmath and Sims 2010: 89)

facilitative adjective English break → break-able

Basque jan 'eat' → jan-garri 'edible'

agentive adjective English →

Spanish habla-r 'talk' → habla-dor 'talkative'

Some common kinds of denominal adjectives (N --> A) are shown in Table 3. Relational adjectives are adjectives that describe relatedness to the noun. Proprietive adjectives describe having the property of the noun. Privative adjectives describe the lack of the noun. Finally, material adjectives describe the material something is made of.
Table 3. Denominal adjectives (N → A) (adapted from Haspelmath and Sims 2010: 89)

relational (= 'related to N') adjective English →

Russian korol' 'king' → korol-evskij 'royal'

proprietive (= 'having N') adjective English wonder → wonder-ful

Ponapean pihl 'water' → pil-en 'watery'

privative (= 'lacking N') adjective English peanut → peanut-free

Russian vod-a 'water' → bez-vod-nyj 'waterless'

material adjective English wood → wood-en

German Kupfer 'copper' → kupfer-n 'made of copper'

Some common kinds of deadjectival adjectives (A --> A), shown in Table 4, include attenuative adjectives, which create a weaker form of the adjective; intensive adjectives, which create a stronger form of the adjective, and negative adjectives, which create a negative form of the adjective.
Table 4. Deadjectival adjectives (A → A) (adapted from Haspelmath and Sims 2010: 89)

attenuative adjective English →

Tzutujil kaq 'red' → kaq-koj 'reddish'

intensive adjective English →

Turkish yeni 'new' → yep-yeni 'brand new'

negative adjective English →

German schön 'beautiful' → un-schön 'ugly'

Syntactic distribution of adjectives
Adjectives typically modify nouns, and so their distribution can often be described with respect to nouns and other things that modify nouns. In English, adjectives come before nouns, but after determiners. (2) a. the happy child b. a blue book c. that long drive Adjectives can also appear in copular constructions, following the verb BE. Watch out, though nouns, verbs, and prepositions can also appear following BE! (3) a. The child will be happy. b. The book is blue. c. That drive was long. Adjectives can be modified by adverbs, so their distribution can also be described with respect to adverbs. Adjectives in English typically follow the adverbs that describe them. (4) a. very quick b. so happy c. more readable
Adverbs

Inflection on adverbs
Adverbs don't take much inflection, although they can occasionally be used comparatively with the word more, as in (5). (5) She ran the race more quickly than I did.
Derivation of adverbs
Many adverbs in English use the suffix -ly.
Syntactic distribution of adverbs
Adverbs can occur in many different locations, including at the beginning or end of the clause, as in (6). (6) a. Unfortunately, I dropped my ice cream at the park. b. I dropped my ice cream at the park, unfortunately. Adverbs can also modify adjectives, as well as other adverbs. For example, in (7a), the adverb very modifies the adjective happy. In (7b), the adverb so modifies the adverb very. (7) a. very happy b. so very happy Adverbs can also modify verbs, usually appearing at the beginning or end of the verb phrase. (8) a. I quickly went to the store. b. I went to the store quickly. In fact, adverbs can appear in so many positions that it may be easier to describe where they can't appear. They cannot appear in between determiners and nouns, as in (9a), and they cannot appear after copula BE, as in (9b). (9) a. *the quickly dog b. *This dog is quickly. You may have noticed something about this distribution—it is the opposite of the distribution of adjectives. In other words, the distribution of adjectives and adverbs are in [pb_glossary id="5218"]complementary distribution[/pb_glossary].

One category or two?

You may be familiar with the idea of complementary distribution from phonology, where a complementary distribution pattern can be used to argue that two segments are allophones of the same phoneme. When two kinds of things are in complementary distribution, this can be used as evidence that they are actually variants of a single category. Adjectives and adverbs also have many things in common:

They both describe as their basic functions, although adjectives typically describe nouns while adverbs describe clauses, verbs, and other adjectives.

They both can take the derivational morpheme -ly (compare the adjective friend-ly to the adverb quick-ly).

They both can appear in the comparative form.

Because of these reasons, it has been proposed that they are two subcategories of one bigger category, A. However, most morphosyntacticians continue to distinguish them, and so we will, as well.

Key takeaways

Adjectives may inflect into comparative and superlative forms, and adverbs may have a comparative form.

There are many derivational morphemes that form adjectives from verbs, nouns, and other adjectives.

Adverbs can be formed out of adjectives with the derivational morpheme -ly.

In English, adjectives appear in between determiners and nouns or after the copula BE. Adverbs can appear anywhere except where adjectives can appear.

Check yourself!

8.6. Common pitfalls

juliedoner — Sat, 10 Aug 2024 22:05:26 +0000
In this section, we will consider some common mistakes students make when classifying words according to part of speech.
Overlapping distribution

Some syntactic categories can appear in the same place as another. For example, both adjectives and nouns can appear after a determiner, as shown in (1).

(1) a. the happy woman

D Adj N

b. the woman

D N

This test is not actually ambiguous; you just have to remember that adjectives appear in between a determiner and a noun, while nouns will not typically be followed by another noun.
Ambiguous affixes
Some affixes are ambiguous. When an affix is ambiguous, we have to use additionally criteria to conclusively determine the part of speech of a word. For example, -ly can form both adjectives and adverbs. In (2a), -ly attaches to the noun friend to form the adjective friendly. In (2b), on the other hand, -ly attaches to the adjective quick to form the adverb quickly.

(2) a. The friendly cat purred at me.

b. The cat ran quickly.

There are also two prefixes un- that can get mixed up. One un- is called the negative un- and it takes adjectives and turns them into negative adjectives, such as unhappy, unable, or unease. The other un- is called the reversive un- which means to undo an action, such as in unlock, unbend, and unclog. It's important not to get these two homophonous morphemes mixed up because negative un- turns adjectives into adjectives and reversive un- turns verbs into verbs. The [pb_glossary id="4483"]present participle[/pb_glossary], which is formed by attaching -ing to a verb stem, can be used as a [pb_glossary id="691"]progressive[/pb_glossary] verb, as a noun, or as an adjective. In some contexts, it is even ambiguous. We can tell that annoying in (3a) is a verb because it takes an object, me. We can tell that annoying in (3b) is an adjective because it appears in between a determiner and a noun. However, (3c) is ambiguous. The word annoying could be an adjective in a [pb_glossary id="686"]copular[/pb_glossary] construction or a progressive verb.

(3) a. This rash has been annoying me. progressive verb

b. This annoying rash won’t go away. adjective

c. This rash is annoying. ambiguous

When present participles are used as nouns, as in (4a), the subject of the verb appears in genitive case and the object of the verb requires a preposition of. When a present participle is used as a non-finite verb, as in (4b), the subject will be in accusative case and the object does not need a preposition. If there is no subject or object, though, as in (4c), it is ambiguous between a verb and a noun.

(4) a. His winning of the race surprised me. noun

b. Him winning the race surprised me. non-finite verb

c. Winning surprised me. ambiguous

The [pb_glossary id="3731"]past participle[/pb_glossary], which is typically formed by attached -en to a verb stem, is also ambiguous. It can also be used as an adjective, as well as in forming [pb_glossary id="523"]passive voice[/pb_glossary] and [pb_glossary id="4488"]perfect aspect[/pb_glossary]. In (5a), we can tell that stolen is a verb in the passive voice because of the by-phrase by the thief. In (5b), we can tell that stolen is a verb in the perfect aspect because of the auxiliary verb had. In (5c), we can tell that stolen is an adjective because it appears in between a determiner and a noun. In (5d), stolen is ambiguous; it could be an adjective in a [pb_glossary id="686"]copular[/pb_glossary] construction or a passive verb.

(5) a. The diamond was stolen by the thief. passive voice verb

b. The thief had stolen the diamond. perfect aspect verb

c. The stolen diamond was recovered. adjective

d. The diamond was stolen. ambiguous

Homonyms

Some words have homonyms of a different category. For example, the word bear meaning the kind of animal is a noun in most contexts while the word bear meaning to carry is a verb in most contexts. For lexical words like bear and bear, it is fairly easy to keep them straight, but it is more difficult when one or more of the two words belongs to a functional category. For example, there are two words to with different functions, and they belong to different parts of speech: it can be a preposition or it can be a [pb_glossary id="4511"]non-finite[/pb_glossary] tense marker. In (5a), we can tell that to is a preposition, because it is followed by a noun phrase. In (5b), we can tell that to is a non-finite marker because it is followed by a verb.

(5) a. I went to the store. noun

b. I want to go home. verb

Nouns that look like prepositions
There are some nouns that can indicate location or origin, which some students mistake for prepositions. These nouns may even have equivalent prepositions with the same meaning. For example, back in (1a) and top in (2a) are both nouns. We can tell that they are nouns because they cannot introduce the following noun phrase directly; instead, they need the preposition of.

(1) a. to the back of the door noun

b. behind the door preposition

(2) a. on top of the image noun

b. above the image preposition

Key takeaways

Some criteria for determining part of speech can be ambiguous. Be aware of these ambiguities and use two or more criteria to be sure.

The present and past participle forms can be used either as verbs, as nouns, or as adjectives. Some instances are ambiguous.

Some words have homonyms in different categories. Be aware of these, especially the difference between preposition and non-finite to.

Check yourself!
[h5p id="54"]

Chapter 1

juliedoner — Wed, 11 Oct 2023 18:02:56 +0000
This is the first chapter in the main body of the text. You can change the text, rename the chapter, add new chapters, and add new parts.

blah

juliedoner — Wed, 11 Oct 2023 18:29:27 +0000

Part 3: Syntax

markc — Thu, 19 Oct 2023 17:29:20 +0000

Part 3: Syntax

markc — Thu, 19 Oct 2023 17:30:10 +0000

Blah.

markc — Thu, 19 Oct 2023 17:32:37 +0000
Blah.

What are morphology and syntax?

markc — Thu, 19 Oct 2023 17:34:41 +0000
Blah.

The scientific method in morphosyntax

markc — Thu, 19 Oct 2023 17:36:32 +0000
Blah.

Linguistic data in morphosyntax

markc — Thu, 19 Oct 2023 17:36:55 +0000
Blah.

Parts of speech

markc — Thu, 19 Oct 2023 17:37:16 +0000
Blah.

What is a word?: A cross-linguistic perspective

markc — Thu, 19 Oct 2023 17:37:46 +0000
Blah.

Theories of grammar and language acquisition

markc — Thu, 19 Oct 2023 17:38:01 +0000
Blah.

What are morphemes?

juliedoner — Fri, 20 Oct 2023 02:38:11 +0000

Theories of grammar and language acquistion

juliedoner — Tue, 31 Oct 2023 17:28:04 +0000

8.5. Functional parts of speech

juliedoner — Sat, 10 Aug 2024 22:09:48 +0000
***THIS IS A DRAFT***
Functional and lexical parts of speech
So far we've only looked at lexical parts of speech. [pb_glossary id="4437"]Lexical[/pb_glossary] words are generally easier to define and encode the meaning of the sentence. There are also [pb_glossary id="4438"]functional[/pb_glossary] parts of speech. Functional parts of speech are much harder to define and have grammatical functions. Functional parts of speech include determiners, auxiliaries, modals, conjunctions, and complementizers. Adpositions are also a kind of functional part of speech, but they are more on the edge. For example, some prepositions, like of, are very functional and difficult to define, while other prepositions, such as under, are much easier to define.
Open and closed parts of speech
We can also distinguish between [pb_glossary id="4439"]open[/pb_glossary] and [pb_glossary id="4440"]closed[/pb_glossary] parts of speech. Open parts of speech can easily have new words added to them, while it is generally harder to add words to a closed parts of speech. Most lexical parts of speech are open and most functional parts of speech are also closed.

Frindle

Frindle is a children's book by Andrew Clements about a fifth-grade boy who decides to start calling pens by a new name, frindle, and gets his friends and classmates to start using the new word, too. Based on what you know about open and closed parts of speech, do you think it's realistic that frindle would catch on? Why or why not?

Identifying functional parts of speech

Adpositions
If you recall from Section 1.2, some languages use [pb_glossary id="249"]prepositions[/pb_glossary], which appear before nouns, and some languages use [pb_glossary id="250"]postpositions[/pb_glossary], which appear after nouns. Both use the symbol P. The word [pb_glossary id="253"]adposition[/pb_glossary] can be used if you don't want to distinguish between prepositions and postpositions. The function of adpositions is to introduce a noun phrase, indicating its relation to the rest of the sentence. Some examples of English prepositions include before, after, to, from, of, under, over, through, near, off, on, for. Prepositions in English often appear before nouns and determiners, as shown in (1).

(1) a. down the stairs

b. by Tuesday

c. under the sea

d. on a boat

e. near Winnipeg

f. from my bag

English also has [pb_glossary id="617"]intransitive[/pb_glossary] prepositions, which do not have objects, as shown in (2). Intransitive verbs usually appear either immediately after the verb or immediately after the object.

(2) a. I picked the bag up.

b. I picked up the bag.

c. The machine broke down.

d. The dean called the meeting off.

e. The dean called off the meeting.

f. She eased up on the brakes.

Determiners
The syntactic position of determiners can typically be described with reference to nouns. In English, determiners appear before nouns. There are several subclasses to the category of determiners, as listed in Table 1.
Table 1. Subclasses of the determiner category

Articles the, an, a

Demonstratives this, that, these, those

Quantifiers every, some, many, most, few, all, each, any...

Numerals one, two, three...

Possessive determiners my, your, his, her, its, our, their

Some wh-words which, whose

Determiners may agree with any of the inflectional features of the noun. Nominal inflectional features may also appear on the determiner instead of the noun.

Generally, the semantics of determiners is to pick out which specific entity or entities are being referred to. Whereas cat on its own could be referring to any cats in the world, or even the idea of cats in general, phrases like the cat indicates that there is a single contextually-relevant cat that is being referred to, a cat indicates that there is an individual relevant cat but doesn't specifically tell you which one, and so on.

Cross-linguistically, then, determiners also encode values such as definiteness and specificity. A definite determiner, like English the, indicates that the reference of the noun is identifiable in the context, while an indefinite determiner, like English a(n), indicates that the reference of the noun is not identifiable. Specificity indicates whether the referent is a unique entity.

(3) a. I want to eat that cookie. definite, specific

b. I want to eat the biggest cookie you can make. definite, non-specific

c. My sister gave a cookie to her friend. indefinite, specific

d. Let's go to the bakery and see if there's a cookie that looks good. indefinite, non-specific

In some languages, definiteness and specificity might inflect on the noun instead of on the determiner.

Pronouns
Pronouns are an interesting part of speech. On the one hand, they have the same distribution as noun phrases, so they look like they might be a subcategory of nouns. On the other hand, unlike nouns, pronouns are a functional part of speech: they usually can be defined using only inflectional features, such as person, number, and gender. Like other functional parts of speech, pronouns are also a closed class. It is not easy to coin new pronouns. Because of this, some people argue that maybe pronouns are a subclass of determiners rather than a subclass of nouns. Some evidence for this approach comes from (4). Although pronouns usually stand on their own, in some contexts a pronoun can co-occur with a noun, as in (4a) us students. However, there are not contexts where a pronoun can co-occur with a determiner, as in (4b). (4) a. Us students have gathered together to create a formal petition for less homework. b. *The us/we would like less homework.

Neo-pronouns

psychological studies showing generic he does not get interpreted as gender-neutral cited in Bjorkman 2017 (Martyna 1978; MacKay & Fulkerson 1979; Gastil 1990; Foertsch & Gernsbacher 1997; Miller & James 2009). Long history of trying to coin a replacement (citations) Newly coined pronouns are called neopronouns. However, because pronouns belong to a closed class, neopronouns don't often catch on easily. It is more common for existing pronouns to be used generically, even when it results in awkward or clunky phrases, such as, generic you, coordinated she or he, or the very formal one. Similarly, some people do not feel like gendered pronouns such as she or he describe them well. Some of these people choose to use pronouns like they or it, but others coin new pronouns that they feel fit them better. For example, in this blog, neopronoun Vagrant Gautam user explains why and how xe chose which pronouns feel the best to xem. Many different English gender-neutral neopronouns have been coined and are in use. Some are shown below in Table 2.
Table 2. Examples of some gender-neutral neopronouns in English

nominative accusative genitive reflexive

xe xem xyrs xemself

hi hir hirs hirself

ze zir zirs zirself

ey em eirs emself

fae faer faers faerself

One reason functional categories are hard to coin is closed class words are more likely to have irregular inflection, so people don't know how to predict their inflection and the kinds of agreement they might trigger. For example, if you look at only one column in Table 2, do you feel like you could predict the forms in the other columns? Neopronouns are being coined in other languages, too. For example, iel in French is a blend of the masculine pronoun il and the feminine pronoun elle. To read more about neopronouns in other languages, check out this article. examples of neo-pronouns in other languages: iel in French We might think that it would be better to just switch to using gender-neutral pronouns for everyone all the time. (I saw a story that Finnish speakers were borrowing English pronoun pins to advertise their identity - fact check this) Gender-neutral grammar is not correlated with gender ideology in culture (examples, citations)

Question words
Question words are a kind of pro-form, which means that they replace phrases of another part of speech. We can figure out what part of speech a question word is by looking at what it replaces. We can do that by answering the question and determining the part of speech of the answer. Some question words are ambiguous, and can be answered by phrases from more than one part of speech.
Table 3. Some possible parts of speech of question words in English

Question word Part of speech Example

a. who N Who ate the cookie? The cookie monster.

b. what N What did the cookie monster eat? A cookie.

D What TV show did you watch? This one.

c. where P Where did you watch it? On the couch.

N Where did you watch it? Here.

d. when N When did you watch it? Last week.

Adv When will you watch it? Soon.

e. why C Why did you watch it? Because I was bored.

N Why did you watch it? No reason.

f. how P How did you watch it? On the TV.

g. which D Which show did you watch? This one.

h. whose D Whose cookie was eaten? Elmo's.

Tense

Tense markers in English include auxiliaries, modals, and the non-finite marker to, as listed in Table 4.

Table 4. Subclasses of the tense category

auxiliaries HAVE, BE, and DO

modals will, would, should, shall, can, could, may, might, must, ought

non-finite to

In English, elements in the tense category appear in between the subject and the verb.

Each of the three tense subclasses have different properties.

You can have more than one auxiliary, but only one modal and one to (and you can't have a modal and a to together, either)

Except for some dialects, like Appalachian English, where they do allow double modals

auxiliaries and modals invert in question formation, but not to

auxiliaries inflect for person, number, and tense like a verb, but modals and to do not

Auxiliaries, especially, seem to have many properties in common with verbs, so some people categorize auxiliaries as subclasses of verbs. But they also have some properties of T! If you continue studying syntax, you will discover why!

***Distinguish between auxiliary and main verb uses of HAVE BE and DO***

Conjunctions
Conjunctions (Conj) connect two units of the same type. English only has four conjunctions: and, or, nor, neither...nor, and either...or.
Complementizers
Complementizers (C or Comp) put one clause inside another. Some of the complementizers in English are that, for, if, and whether.
Negation
Negation (Neg): not

Particles

Free morphemes which encode only inflectional information are often called particles. Calling something a particle is a way to avoid classifying the word into a part of speech, often because it is difficult to identify its part of speech. In some cases, there may be only one or two particles that exhibit the same behaviour, and so there is really little basis for assigning it its own part of speech category.

Key takeaways

Check yourself!

References and further resources

For a general audience
Abraham, Amelia. 2020. Gender neutral pronouns are important: Not everyone identifies with 'he' or 'she'. Vogue India. https://www.vogue.in/culture-and-living/content/gender-neutral-pronouns-are-important-not-everyone-identifies-with-he-or-she Conrod, Kirby and Vagrant Gautam. 2021. Guest Lecture in pronouns: Vagrant. Medium. https://kconrod.medium.com/guest-lecture-in-pronouns-vasundhara-1c61d3023e0c

7.1. Predicates

juliedoner — Sat, 17 Aug 2024 03:18:55 +0000
A clause or simple sentence typically consists of a [pb_glossary id="4639"]subject[/pb_glossary] and a [pb_glossary id="4142"]predicate[/pb_glossary]. The subject is who or what the sentence is about and the predicate is what the sentence is asserting about the subject. We will learn more about subjects in Section 7.2. In the sentences in (1), the subject is underlined and the predicate is in bold.

(1) a. My friend rides the bus to campus.

b. Cats like milk.

c. Some professors are strict.

d. The library is down the street.

e. Taylor Swift is a singer-songwriter.

The word predicate can be used in two main ways. Sometimes it can refer to just the [pb_glossary id="629"]head[/pb_glossary] of the predicate (usually a verb), or it can refer to the whole phrase. In (1), the whole phrases are bolded. Although most predicates are verbs, such as ride in (1a) and like in (1b), there are also non-verbal predicates. Non-verbal predicates include adjectival predicates, as in (1c), prepositional predicates, as in (1d), or nominal predicates, as in (1e). In English, non-verbal predicates always appear with the verb BE, which is known as the [pb_glossary id="686"]copula[/pb_glossary]. The copula does the grammatical job of bearing tense and agreement marking, since adjectives, prepositions, and nouns can't. Although there are multiple kinds of copular constructions with slightly different meanings, you can think of the copula roughly as an equal sign, indicating that the subject has the property denoted by the non-verbal predicate. Although in English non-verbal predicates always have a copula, some other languages have what is known as a zero copula. In these languages, there may be no verb at all when there is a non-verbal predicate. Some examples of nominal predicates with zero copula from Sinhalese (Indo-European language family, spoken in Sri Lanka), Tubu (Saharan language family, spoken in Niger and Chad), and Pitjantjatjara (Pama-Nyungan family, spoken in Australia) are shown below. (2) a. Sinhalese

unnaehee

hungak

presidde

kene-k

he

very

famous

person-nom

‘He is/ was a very famous person.’

(Gair 1970: 145, as cited in Stassen 2013) b. Tubu

sígən

līfi

3sg.emph

orphan

‘He is/was an orphan.’

(Lukas 1953: 170, as cited in Stassen 2013) c. Pitjantjatjara

wait

nglayayala

man

doctor

‘The man is/ was a doctor.’

(Doughlas 1959: 55, as cited in Stassen 2013) Some other languages have zero copula sometimes but an overt copula in other contexts. For example, Maltese, an Afro-Asiatic language spoken in Malta, has a zero copula in the present tense and an overt copula in the past tense. (3) Maltese

a.

Albert

tabib

Albert

doctor

‘Albert is a doctor.’

b.

Albert

kien

tabib

Albert

be.3sg.m.pst

doctor

‘Albert was a doctor.’

(Stassen 2013) According to WALS, just over half of the world's languages always use an overt copula (Stassen 2013).

Key takeaways

Sentences consist of a subject and a predicate.

The predicate is the part of the sentence that asserts something about the subject.

Most predicates are verbs, but there are also adjectival, nominal, and prepositional predicates.

Non-verbal predicates are introduced by a copula in many languages.

Some languages use a zero copula, either all of the time or limited to some contexts.

Check yourself!
[h5p id="55"]
References and further resources

Academic sources

Stassen, Leon. 2013. Zero copula for predicate nominals. In World Atlas of Language Structures Online, ed. Matthew Dryer and Martin Haspelmath. https://wals.info/chapter/120

7.2. Arguments

juliedoner — Sat, 17 Aug 2024 03:19:13 +0000
Most verbs can't stand on their own. Some verbs describe the properties of one entity, while other verbs describe the relationship between several entities. Because of this, verbs often require certain phrases be included with them in a sentence. The requirement is called [pb_glossary id="656"]selection[/pb_glossary], and the phrases that are required are called [pb_glossary id="703"]arguments[/pb_glossary].
Transitivity
We classify verbs based on how many arguments they require, which is called transitivity. An [pb_glossary id="617"]intransitive[/pb_glossary] verb has one argument, typically the subject. A [pb_glossary id="631"]transitive[/pb_glossary] verb has two arguments, typically a subject and direct object. A [pb_glossary id="4660"]ditransitive[/pb_glossary] verb has three arguments, typically a subject, direct object, and indirect object.
Table 1. Kinds of transitivity

Kind of verb Number of arguments Typical arguments Example

Intransitive 1 Subject The cat slept.

Transitive 2 Subject and direct object The cat chased a mouse.

Ditransitive 3 Subject, direct object, and indirect object The cat left me a dead mouse.

Subjects
The [pb_glossary id="4639"]subject[/pb_glossary] is typically (but not always) the one performing the action. Subjects in English can be identified because they go before the verb and because they trigger agreement on the verb in the present tense. If the subject is singular, the verb will be singular, as shown in (1a) where the cat is singular, so we get the singular form of the verb is. If the subject is plural, the verb will be plural. This is shown in (1b), where the subject the cats is plural and so we get the plural form of the verb are.

(1) a. The cat is purring.

b. The cats are purring.

If you're not sure what the subject is in English, you can switch a sentence to present tense and look for subject-verb agreement.

The many facets of subjecthood

In a prototypical sentence, a subject has many properties:

The subject appears in a particular position. In English, the subject comes before the verb.

The subject agrees with the verb.

The subject is what the sentence is about.

The subject is the one performing the action.

The subject is morphologically marked as being the subject.

However, a subject doesn't have all of these properties in all sentences. For example, in (2), the sentence is about lunch, but the subject is we. (2) As for lunch, we can eat some sandwiches. In [pb_glossary id="523"]passive[/pb_glossary] constructions, the entity that performs the action is deleted or included in a by-phrase. The subject of a passive is the entity that undergoes the action. This is shown in (3). An [pb_glossary id="4759"]active [/pb_glossary]sentence is in (3a), with the one performing the action, we, in subject position. In the passive sentence in (3b), however, the one performing the action is included in an optional by-phrase. The entity that undergoes the action, some sandwiches, is in subject position.

(3) a. We ate some sandwiches.

b. Some sandwiches were eaten (by us).

Finally, sometimes the subject has a special form. This happens for some pronouns in English, as we already saw in Section 2.1. The first person plural pronoun appears as we in subject position but us in other positions, as shown in (4).

(4) a. We gave you a gift.

b. You gave us a gift.

However, the subject of a [pb_glossary id="4511"]non-finite[/pb_glossary] clause does not appear in the special form for subjects. If we turn the clause in (4a) into a non-finite clause, as in (5), we get us not we.

(5) a. [For us to give you a gift] would be kind.

b. *[For we to give you a gift] would be kind.

c. They want [us to give you a gift].

d. *They want [we to give you a gift].

When we discuss the subject, we are usually referring to the structural position—in English, this is the position before the verb. If we want to refer to a phrase that has some of these other properties, we have other words for that. For example, what the sentence is about is called the topic, the one performing the action is called the agent theta role, and the morphological form associated with subjects is called [pb_glossary id="4742"]nominative[/pb_glossary] [pb_glossary id="700"]case.[/pb_glossary] We will learn more about theta roles and case later in this textbook.

Objects
In English, objects usually appear after the verb. There are two kinds of object: [pb_glossary id="4669"]direct objects[/pb_glossary] and [pb_glossary id="4670"]indirect objects[/pb_glossary]. If there is only one object, it is a direct object, as shown in (6). Direct objects typically undergo the action described by the verb. (6) The cat drank some milk. In (6), the verb is drank. The subject is the cat and the direct object is some milk. Ditransitive verbs have two objects. Ditransitive verbs usually describe a change of possession. The direct object undergoes the action. That is, it is the entity that switches possession. The indirect object is usually the entity that has possession of the object at the end of the action. The two objects in ditransitive verbs can appear in either order, but the two orders have some structural differences. The sentences in (7), which have the same meaning, illustrate this difference.

(7) a. My sister gave me her cat. double object construction

b. My sister gave her cat to me. prepositional ditransitive

The sentence in (7a) is called the [pb_glossary id="664"]double object construction[/pb_glossary]. In (7a), both objects are noun phrases. The indirect object, who receives the entity being transferred, is me and appears first. The direct object, which is the entity being transferred, is her cat and appears second. The sentence in (7b) is called the [pb_glossary id="4677"]prepositional ditransitive[/pb_glossary]. In prepositional ditransitives, the direct object her cat appears first. It is a noun phrase. The indirect object to me comes second. It is a prepositional phrase. If you're not sure which object is the direct object and which is the indirect object, rephrase the sentence so that it uses a preposition. The one that comes with a preposition is the indirect object. The differences between the two ditransitive constructions are summarized in Table 2.
Table 2. Ditransitive constructions

Construction First object Second object

Double object construction NP indirect object NP direct object

Prepositional ditransitive NP direct object PP indirect object

Adjuncts
Many sentences also include many phrases that aren't selected for by the verb or by any other word in the sentence. These are optional and add extra information. They are called [pb_glossary id="4682"]adjuncts[/pb_glossary]. Some examples of adjuncts are shown in (8). In these examples, the adjuncts are in parentheses. This shows that they are optional, since the sentence is still grammatical if we leave them out.

(8) a. The cat slept (on the porch).

b. The (black) cat slept.

c. The cat slept (peacefully).

d. The cat slept (because it was tired).

Key takeaways

A verb has requirements for phrases it must appear with, which is called selection. The phrases it selects are called arguments.

We classify verbs based on how many arguments they have, which is called transitivity. Intransitive verbs take one argument, transitive verbs take two arguments, and ditransitive verbs take three arguments.

Subjects appear before the verb and trigger subject-verb agreement.

Objects appear after the verb. Direct objects undergo the event and indirect objects are the endpoint of the event.

Ditransitives allow the objects to appear in either order. If the indirect object comes first and both objects are NPs, it is called a double object construction. If the indirect object comes second and is a PP, then it is a prepositional ditransitive.

Phrases that aren't selected for by the verb are called adjuncts.

Check yourself!
[h5p id="56"]

7.3. Clauses

juliedoner — Sat, 17 Aug 2024 03:19:29 +0000
Consider the following sentences.

(1) a. I think [that spaghetti is delicious].

b. We should have spaghetti for supper [because it is delicious].

c. [Spaghetti is delicious] and [garlic bread is delicious, too].

d. [That spaghetti is delicious] is obvious.

e. We should have spaghetti [which is delicious] for supper.

f. [Because spaghetti is delicious], we should have it for supper.

g. I want [to have spaghetti for supper].

h. [Having spaghetti for supper] is a good idea.

i. We will eat some spaghetti [when it is ready].

All of the sentences in (1) have two clauses. Look at the parts of the sentences that are in square brackets. Each of them has its own verb and arguments and can stand on its own as a sentence (if you take out the that and because), even though it is also part of a larger sentence. These are all clauses. A clause consists of a predicate and its arguments. All of the sentences in (1) have two clauses, but a sentence can have more than two. The sentence in (2) has six clauses! Sentence (2) rambles a bit and is repetitive, but it is grammatical.

(2) [I think [that we should have spaghetti, [which is delicious], for supper]] and [we should have garlic bread, too,] [because I really want [to have spaghetti and garlic bread]].

The easiest way to find all of the clauses in a sentence is to find the predicates. There will normally be one clause for each predicate. In (2), I have bolded the predicates of each clause. Then, find the arguments and adjuncts that belong to each predicate. Sometimes a clause has a special word to introduce the clause, like that and because. These are called complementizers and belong to the clause, too. There may also be a wh-word that belongs with the clause.

Serial verb constructions

Some languages allow a construction called a [pb_glossary id="4733"]serial verb[/pb_glossary] construction, in which multiple verbs form a single predicate. We can tell that they all belong to the same predicate because they share the same arguments—the same subject and the same object. Below are some examples of serial verb constructions from Tabas, an Austronesian language spoken in Indonesia; Dâw, a Makú language spoken in Northwest Amazonia; and Igbo, a Niger-Congo language spoken in Nigeria. (3) a. Tabas (Bowden 2001: 297, as cited in Aikhenvald and Dixon 2006: 2) n=babas welik n-mot do 3SG=bite pig 3SG=die REAL 'It bit the pig dead.' b. Dâw (Aikhenvald and Dixon 2006: 2) yõ:h bə:-hãm-yɔw medicine spill-go-happen.straight.away 'The medicine spilt straight away.' c. Igbo (Lord 1975: 27, as cited in Aikhenvald and Dixon 2006: 2) ó tì-wà-rà étéré à he hit-split.open-TENSE plate the 'He shattered the plate.' Because there are multiple verbs in a serial verb construction, we might think that there are multiple clauses, too. However, since all of the verbs in a serial verb construction still belong to the same predicate, we can still say that there is one clause for each predicate. Serial verb constructions have different properties in different languages. For example, the verbs in a serial verb construction might all be together in one word, as in (3b) and (3c), or might be separated into multiple words, as in (3a). They might all be next to each other, as in (3b) and (3c), or there might be words in between the different verbs, as in (3a). Inflection might appear on all of the verbs in the construction, as in (3a) or just on one of the verbs, as in (3c).

There is more than one kind of clause. Clauses are generally classified based on their relationship to other parts of the sentence. Some clauses also have special structure and can be classified based on their internal structure. Over the next few sections, we will learn how to classify clauses based on these two properties.
Main and embedded clauses
If you look at the examples of clauses in (1), you may notice that some clauses are inside of other clauses. The ones that are inside are called embedded clauses. You may also see them referred to as subordinate or dependent clauses. The clauses that they are inside of are called main clauses. They can also be called independent, matrix, or root clauses. Let's look again at example (1a), repeated below in (3). In this sentence, the embedded clause is that spaghetti is delicious. The main clause is I think that spaghetti is delicious. The embedded clause is the object of the main clause verb think, which makes the entire embedded clause part of the main clause.

(3) I think [that spaghetti is delicious].

Argument clauses
Some clauses are arguments of a verb. As we already saw, the embedded clause in (3) is the object of the main clause verb. An embedded clause can also be in subject position. This is the case in (1d) and (1h), repeated below in (4).

(4) a. [That spaghetti is delicious] is obvious.

b. [Having spaghetti for supper] is a good idea.

In (4a), the embedded clause that spaghetti is delicious is the subject of the predicate is obvious. In (4b), the embedded clause having spaghetti for dinner is the subject of the predicate is a good idea.
Adjunct clauses
Clauses can also be adjuncts. Remember that adjuncts are optional modifiers. Adjunct clauses can modify the entire main clause, a verb, or a noun. Some examples of adjunct clauses are found in (1b), (1e), (1f), and (1i), and are repeated below in (5). In (5a) and (5b), the embedded clause because it is delicious modifies the entire main clause. In (5c), the embedded clause which is delicious modifies the noun spaghetti. In (5c), the embedded clause when it is ready modifies the verb eat. All of these embedded clauses are adjuncts; they are not required to form a complete sentence.

(5) a. We should have spaghetti for supper [because it is delicious].

b. [Because spaghetti is delicious], we should have it for supper.

c. We should have spaghetti [which is delicious] for supper.

d. We will eat some spaghetti [when it is ready].

Kinds of sentences
We can also classify sentences based on how many clauses it has and how the clauses are combined. If a sentence only has one clause, it is called a simple sentence. The single clause in a simple sentence will always be a main clause. Some examples of simple sentences are shown in (6).

(6) a. Spaghetti is delicious.

b. We should have spaghetti for supper.

c. A spaghetti dinner is obvious.

If a sentence has two or more conjoined main clauses, it is called a compound sentence. Example (1c), repeated below as (7), is an example of a compound sentence.

(7) [Spaghetti is delicious] and [garlic bread is delicious, too].

Finally, if a sentence has one or more embedded clauses, it is called a complex sentence. A complex sentence will also always have at least one main clause, as well. All of the sentences in (1) except the compound sentence in (1c) are complex sentences. The complex sentences from (1) are repeated below in (8).

(8) a. I think [that spaghetti is delicious].

b. We should have spaghetti for supper [because it is delicious].

c. [That spaghetti is delicious] is obvious.

d. We should have spaghetti [which is delicious] for supper.

e. [Because spaghetti is delicious], we should have it for supper.

f. I want [to have spaghetti for supper].

g. [Having spaghetti for supper] is a good idea.

h. We will eat some spaghetti [when it is ready].

Key takeaways

A clause consists of a predicate, all of the arguments and adjuncts that modify the predicate, and sometimes a complementizer.

A main clause can stand on its own, while an embedded clause is inside of a main clause.

An embedded clause can be in subject, object, or adjunct position.

A simple sentence has one main clause, a compound sentence has two or more main clauses, and a complex sentence has one or more embedded clauses.

Check yourself!
[h5p id="57"]
References and further resources

Academic sources

Aikhenvald, Alexandra, and R. M. W. Dixon. 2006. Serial verb constructions: A cross-linguistic typology. Oxford: Oxford UP.

7.5. Relative clauses

juliedoner — Tue, 20 Aug 2024 15:30:25 +0000
Relative clauses are a kind of clause that modifies nouns. Some examples are shown in (1).

(1) a. The witch trapped Rapunzel, [who __ had very long hair], in a tower.

b. The princess [ __ trapped in the tower] had very long hair.

c. The princess [that the witch had kidnapped __ ] was very lonely.

d. The tower [where the princess was trapped __ ] was isolated in the woods.

Relative clauses are always adjuncts to a noun phrase. In (1), the nouns that are modified by the relative clauses are bolded. Any noun phrase can be modified by a relative clause. Relative clauses also always have a gap. In (1), the gaps are marked with underscores. The gap can be in subject position, as in (1a) and (1b), in object position, as in (1c), or in adjunct position, as in (1d), but the gap will always have the same meaning as the noun that the relative clause modifies. We can illustrate this by rephrasing the complex sentences in (1) as two simple sentences, as shown in (2).

(2) a. The witch trapped Rapunzel in a tower. Rapunzel had very long hair.

b. The princess had very long hair. The princess was trapped in the tower.

c. The princess was very lonely. The witch had kidnapped the princess.

d. The tower was isolated in the woods. The princess was trapped in the tower.

Relative clauses may use complementizer that, as in (1c), a question word such as who in (1a) or where in (1d), or neither, as in (1b). If it has neither a complementizer or a question word, it is called a reduced relative clause. Interestingly, it is not possible to use both that and a question word. This is shown in (3). The relative clause can be introduced by who as in (3a) or that as in (3b), but cannot be introduced by both at the same time, in either order, as shown in (3c)-(3d).

(3) a. The witch trapped the princess, [who __ had very long hair], in a tower.

b. The witch trapped the princess [that __ had very long hair] in a tower.

c. *The witch trapped the princess [who that __ had very long hair] in a tower.

d. *The witch trapped the princess [that who __ had very long hair] in a tower.

Relative clauses have very similar structure as embedded content questions. In fact, relative clauses that use question words look identical to embedded content questions—the only difference is their position in the main clause. Relative clauses modify nouns, but embedded content questions are in the argument position of a verb.

(4) a. The witch trapped the princess, [who __ had very long hair], in a tower.

b. I wondered [who __ had very long hair].

(5) a. The tower [where the princess was trapped __ ] was isolated in the woods.

b. I asked [where the princess was trapped __ ].

Factive clauses
Not all clauses that modify noun phrases are relative clauses. Another kind of clause, called a factive clause, also modifies noun phrases. Unlike relative clauses, factive clauses do not have a gap. Some examples of factive clauses are shown in (6).

(6) a. The fact [that Rapunzel was trapped] makes me sad.

b. I wonder when Rapunzel got the idea [that she could escape].

Factive clauses have the same structure as embedded declarative clauses. The only difference is that they modify a noun, instead of being in the argument position of a verb.

Key takeaways

Relative clauses are adjuncts to a noun phrase and have a gap.

Relative clauses have similar structure to embedded content questions.

Factive clauses also modify noun phrases but do not have a gap.

Check yourself!
[h5p id="59"]

7.4. Statements, questions, and commands

juliedoner — Tue, 20 Aug 2024 15:34:37 +0000
Besides classifying clauses based on their structural position, we can also classify them based on their function. We will discuss three kinds of function here: declaratives, interrogatives, and imperatives. Note that the main clause and the embedded clause of the same sentence might have different functions.
Declaratives
Declarative clauses make a statement. Most clauses are declarative.

(1) a. I like chocolate.

b. My father was going to the store yesterday.

c. The plane will land in Frankfurt tomorrow.

In English, embedded declarative clauses can be introduced with the complementizer that or no complementizer at all.

(2) a. It's not unusual [that I like chocolate].

b. Jennie told you [my father was going to the store yesterday].

c. I expect [that the plane will land in Frankfurt tomorrow].

Interrogatives
Interrogatives clauses make a question. There are two main kinds: polar questions and content questions.
Polar questions
Polar questions, also called yes-no questions, are questions that can be answered with yes, no, or maybe. In English, main clause polar questions are formed by subject-auxiliary inversion. That is, the auxiliary verb and the subject switch places compared to the equivalent declarative clause.

(3) a. Do I like chocolate?

b. Was my father going to the store yesterday?

c. Will the plane land in Frankfurt tomorrow?

English embedded polar questions do not have subject-auxiliary inversion and are introduced by the complementizers if or whether.

(4) a. You wonder [whether I like chocolate].

b. Jennie asked [if father was going to the store yesterday].

c. I'm not sure [whether the plane will land in Frankfurt tomorrow].

According to WALS, only about 1% of languages form polar questions by changing the order of words, like English does. The most common way of forming polar questions is to use a question particle, which is the strategy used by 61% of the languages in the sample on WALS. An example of a question particle in Maybrat, a language isolate of Indonesia, is shown in (5). In Maybrat, the question particle a appears at the end of the clause. (5) Maybrat (Dol 1999: 200, as cited in Dryer 2013) ana m-amo Kumurkek a 3PL 3-go Kumurkek Q 'Are they going to Kumurkek?' Other common strategies for marking polar questions in the world's languages are to use intonation (18%) or verbal morphology (17%) (Dryer 2013). An example of a language that uses verbal morphology is Hunzib, a Nakh-Daghestanian language spoken in the North Caucasus region of Russia, near the border with Georgia, as shown in (6).

(6) eƛ’e-čó-y

go-PRES.1/2-Q

‘Are you going?’

Content questions
Content questions, also called wh-questions, are questions that cannot be answered with yes, no, or maybe. In English, main clause content questions are formed by subject-auxiliary inversion and by moving a question word to the beginning of the clause. The question words in (7) are in bold. Question words are also often called wh-words.

(7) a. Who likes chocolate?

b. What do I like?

c. Where was my father going yesterday?

d. Why was my father going to the store yesterday?

e. Whose father was going to the store yesterday?

f. When will the plane land in Frankfurt?

g. How will the plane land in Frankfurt tomorrow?

h. Which airport will the plane land at tomorrow?

Embedded content questions in English do not have subject-auxiliary inversion, but the question word still moves to the beginning of the clause.

(8) a. I asked [who likes chocolate].

b. They wondered [what I like].

c. We wanted to know [where my father was going yesterday].

d. I questioned [why my father was going to the store yesterday].

e. They asked [whose father was going to the store yesterday].

f. We were wondering [when the plane will land in Frankfurt].

g. My kid is asking [how the plane will land in Frankfurt tomorrow].

h. They need to know [which airport the plane will land at tomorrow].

Many other languages also move the question word to the beginning of the clause to form content questions, but in other languages, the question word stays in its argument position. This is called wh-in situ. Mandarin is one example of a wh-in situ language, as shown in (9). The declarative sentence in (9a) has the same word order as the interrogative sentence in (9b), unlike the English equivalents.

(9) a. Yuehan xihuan Mali.

John like Mary

'John likes Mary.'

b. Yuehan xihuan shei?

John like who

'Who does John like?'

(Dong 2009: 11)
Imperatives
Imperatives make a command. In English, imperatives have a silent subject and the verb appears in its bare form.

(10) a. Eat chocolate!

b. Go to the store!

c. Take a plane to Frankfurt tomorrow!

Mixing functions
A complex sentence can have main and embedded clauses with different functions, as shown in Table 1.
Table 1. Combinations of main and embedded clause functions

Main declarative Main polar question Main content question Main imperative

Embedded declarative Ben said [that Ava is hungry]. Did Ben say [that Ava was hungry]? When did Ben say [that Ava was hungry]? Tell them [that Ava is hungry]!

Embedded polar question Ben wondered [if Ava was hungry]. Did Ben ask [if Ava was hungry]? When did Ben wonder [if Ava was hungry]? Ask [if Ava is hungry]!

Embedded content question Ben asked [what Ava wants for lunch]. Did Ben wonder [what Ava wants for lunch]? Why did Ben ask [what Ava wants for lunch]? Ask [what Ava wants for lunch]!

Key takeaways

Declarative clauses make a statement, interrogative clauses ask a question, and imperative clauses give a command.

Interrogative clauses come in two types: polar and content questions.

A complex sentence can have main and embedded clauses with different functions.

Check yourself!
[h5p id="58"]
References and further resources

Academic sources

Dryer, Matthew. Polar questions. In World Atlas of Language Structures Online, ed. Matthew Dryer and Martin Haspelmath. https://wals.info/chapter/116

Hongyuan, Dong. 2009. Issues in the semantics of Mandarin questions. PhD thesis, Cornell University.

Chapter 7 practice exercises

juliedoner — Wed, 21 Aug 2024 21:41:23 +0000

Data analysis

Exercise 1. Predicates
Intermediate [7.1, 7.2, 7.3] Determine how many clauses are in each of the following sentences. Then identify the predicates in each clause. For each predicate, determine if it is verbal, nominal, adjectival, or prepositional. For verbal predicates, determine its transitivity.

a. The members of the other team were the winners.

b. The hockey player shot the puck.

c. Every player skated.

d. Most players were on the ice.

e. The losing team was sad.

Exercise 2. Verb arguments and adjuncts
Beginner-Intermediate [7.2] Identify the subjects, direct objects, indirect objects, and adjuncts to the verbs in the following sentences.

a. The player passed the puck to his teammate at the last second.

b. The players skated quickly towards the net.

c. The skates were sharp.

d. The coach celebrated because her team won the game.

Exercise 3. Simple, complex, and compound sentences
Beginner [7.3] Determine whether the following sentences are simple, complex, compound, or both complex and compound.

a. It was very cold in the hockey arena.

b. The goalie blocked the puck and his team cheered.

c. You can become a good skater if you practice a lot.

d. The winning team won because they had better teamwork.

e. The people in the audience enjoy hockey but they dislike how cold the arena is.

f. The team that lost was sad.

Exercise 4. Embedded clauses
Intermediate [7.3] Put square brackets around the embedded clauses in the following sentences. Classify each embedded clause as a subject, object, or adjunct clause.

a. I knew that my favourite team won.

b. If the home team scored more than three goals, a local restaurant would give out free pizza.

c. That the teams went to a shootout didn't surprise me.

d. Some people play hockey because it's fun and other people play hockey because it's good exercise.

Exercise 5. Declaratives, questions, and imperatives
Beginner [7.4] Determine whether each of the clauses in the following sentences are declaratives, polar questions, content questions, or imperatives.

a. Pass me the puck!

b. You should pass me the puck.

c. Why didn't you pass me the puck?

d. Did you pass the puck to someone else?

e. Do you know why you passed the puck to someone else?

f. Why did you ask if I passed the puck to someone else?

g. The referee wondered if the puck went into the net.

h. How did the referee decide that the puck went in?

Exercise 6. Relative clauses
Advanced [7.5] Identify the relative clauses in the following sentences, if any. For each relative clause, circle the noun phrase that it modifies and put an underscore marking the gap.

a. We had never heard of the team that lost the game.

b. The team that had better teamwork won.

c. The player whom the team trained scored a goal.

d. The goalie whose team had the puck for most of the game didn't move a lot.

e. She was the coach that the team was looking for.

f. The idea that they might win the tournament motivated the team to practice.

g. The player with whom the opposing goalie had a rivalry was always happy to score.

h. The team with better players does not always win.

Exercise 7. Clauses without tense
Advanced [7.6] Identify the to-infinitives, gerund clauses, and small clauses in the following sentences. Note that sentences may contain clauses of more than one type.

a. We called him Puck McDuck.

b. The players needed to practice more.

c. Playing ball hockey is a way for people to play hockey in warmer weather.

d. We consider the coach the best in the city.

e. She encourages newcomers to improve by skating around the rink.

Communication and study skills

Exercise 8. Polar questions
Advanced [7.4] Read WALS chapters 116 and and 92 about polar questions and write a paragraph describing the ways that polar questions can vary across languages.
Research and application

Exercise 9. Relative clause positions
Advanced [7.4] Some languages put relative clauses before the noun they modify and other languages put relative clauses after the noun they modify, as shown in map 90A on WALS. Look at what other word order patterns are described by maps on WALS by going to the Chapters section then filtering for "Word Order" in the Area column. Make a hypothesis about whether another word order pattern might be correlated with the word order between nouns and relative clauses. Then cross-reference the two maps. Do you notice any patterns?
To cross-reference two maps on WALS, follow these steps:

Open one of the two maps.

Above the map, it should say, "You may combine this feature with another one. Start typing the feature name or number in the field below." Below that, there should be a box with the map you're looking at listed. Type the number of the second map you want to look at in the box, below the name of the first map.

Click submit.

Here is an example of map 83A cross-referenced with map 90A.

Language Journal

At the end of Chapter 1, we introduced the Language Journal. You should have picked a language to build a profile for as you work through this textbook. You can add to your profile with these questions.

Exercise 10. Overt and zero copulas
Intermediate [7.1] Does your language use overt or zero copulas? Give examples to back up your answer.
Exercise 11. Serial verb constructions
Advanced [7.3] Does your language use serial verb constructions? Give an example to back up your answer.
Exercise 12. Polar and content questions
Intermediate [7.4] How does your language form polar questions and content questions? Give examples to back up your answer.
Exercise 13. Relative clauses
Advanced [7.5] What do relative clauses look like in your language? Give examples to back up your answer.
Exercise 14. Non-finite clauses
Advanced [7.5] What do non-finite clauses look like in your language? Give examples to back up your answer.

Practice exercises

juliedoner — Tue, 31 Oct 2023 23:50:31 +0000
Exercise X. What do you think Artificial Intelligence (AI) programs that can create novel language tell us about Universal Grammar and the innateness of language? Note: Since the emergence of AI language models, they have been argued to prove, disprove, and be irrelevant to the theory of Universal Grammar. There are a lot of factors to consider and it will take a while for the field to figure them all out.

Do AI models get more input or less input than a child? Is the input that AI models get of the same quality or a different quality?

Is the language that AI produces the same as the language that humans produce?

Even though AI produces language that is similar to human language, does the AI understand the language it produces?

Play around with an AI program and see if you can trick it. Can you find a tricky grammatical construction that it misunderstands?

Why do we do morphology and syntax?

juliedoner — Fri, 03 Nov 2023 21:37:41 +0000

The scientific method

juliedoner — Tue, 14 Nov 2023 07:24:56 +0000

Acknowledgments

juliedoner — Wed, 15 Nov 2023 05:29:45 +0000
This book is meant to serve as a companion to LING 325/725 Syntax I at the University of Kansas. The ideas presented below are the direct result of a number of works including the syntax classes taught at the University of Pennsylvania and the University of California, Los Angeles. A great debt is also owed to Andrew Carnie's influential textbook which I have relied upon in previous versions of this class. I am deeply indebted to the following students for creating videos for this book:

Andrew Collins

Keira Dobbs

Amaya Madden

Corey McCulloch

Alesha Pisciotta

Morgan Robertson

Malena Schoeni

Emma Scott

Particular thanks is owed to Pearl Nelson-Greene for remaking the majority of videos in this book and to Aron Finholt for proofreading the first draft. Finally, thanks as well to Drake Prebyl for extensive feedback on the first draft and for creating a tree-drawing program.

Semantics

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Psycholinguistics

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Neurolinguistics

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

First language acquisition

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Second language acquisition

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Appendix

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000
This is where you can add appendices or other back matter.

Glossary

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

The Science of Syntax

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000

Acknowledgments

juliedoner — Wed, 15 Nov 2023 05:32:42 +0000
This book is meant to serve as a companion to LING 325/725 Syntax I at the University of Kansas. The ideas presented below are the direct result of a number of works including the syntax classes taught at the University of Pennsylvania and the University of California, Los Angeles. A great debt is also owed to Andrew Carnie's influential textbook which I have relied upon in previous versions of this class. I am deeply indebted to the following students for creating videos for this book:

Andrew Collins

Keira Dobbs

Amaya Madden

Corey McCulloch

Alesha Pisciotta

Morgan Robertson

Malena Schoeni

Emma Scott

Particular thanks is owed to Pearl Nelson-Greene for remaking the majority of videos in this book and to Aron Finholt for proofreading the first draft. Finally, thanks as well to Drake Prebyl for extensive feedback on the first draft and for creating a tree-drawing program.

Building a theory

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000

Manipulating structure

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000

Tying up loose ends

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000

Syntax and other linguistic fields

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000

What are we trying to model?

juliedoner — Wed, 22 Nov 2023 20:38:18 +0000

Lexicons and dictionaries

juliedoner — Wed, 22 Nov 2023 20:38:46 +0000

Language judgments

juliedoner — Mon, 08 Jan 2024 21:06:52 +0000

Introduction

juliedoner — Wed, 11 Oct 2023 18:02:56 +0000
This is where you can write your introduction.

About the author

juliedoner — Tue, 31 Oct 2023 17:17:46 +0000

Accessibility statement

juliedoner — Tue, 31 Oct 2023 17:17:59 +0000
We are committed to meeting and exceeding accessibility standards in our design and delivery of this textbook. This textbook follows the accessibility guidelines found in the Accessibility Checklist for OER Development from eCampus Ontario, which was based on Conestoga Library Services' Adapting or Creating OER accessibility checklist. One major development is that interlinear glosses were built using novel code, which is available in the backmatter of this text. Ideally, the use of such code will raise the accessibility standards for interlinear glosses in other OERs and beyond. The new code gives the following advantages.

Screen readers can read glosses as tables and thus allow more flexible navigation of them (i.e., users can view them in a grid format). It also means that users can quickly hop from gloss to gloss via navigating to the next or previous table more quickly.

The glosses are formatted consistently in terms of visuals, mostly conforming to traditional glosses, allowing sighted users to read them more easily than if they were formatted using regular tables.

Numbers are rewritten in transparent text for glosses with more than one sub-example. This means that screen readers will read the number again while sighted users will not be distracted by seeing the number repeated, even if using an external application that changes the background colour (i.e. for a dark mode interface).

Other standard accessibility practices undertaken include but are not limited to the following.

High colour contrast.

Screen reader navigability, as tested using NonVisual Desktop Access (NVDA).

Alt-text to describe images.

The ability to use the OER in multiple forms, including PDF download as well as use on both computers and mobile devices.

In addition to formatting, we also used simple, approachable language and broke up the text into smaller passages. There are also future opportunities to improve accessibility, including but again not limited to the following.

Improve screen readers' ability to accurately read non-English forms in interlinear glosses and within the text.

Anything that users requiring improved accessibility tell us. We are open to feedback. See Contact the author for information on how to provide feedback.

Statement on the inclusion of Indigenous languages

juliedoner — Tue, 31 Oct 2023 17:18:54 +0000

Acknowledgements

juliedoner — Tue, 31 Oct 2023 17:19:07 +0000
OER working group including Glenn and Janice audiences of OE Global and OE Conference Nicole Connor Heather Twitterlings Rebus Iwona Kristy from Campus Manitoba for glosses

Khashayar Hamidzadeh for feedback on accessibility for blind users

Language consultants

Language Consultant(s)

Hindi

Punjabi one anonymous consultant

Russian

Tagalog

Urdu

Yoruba Tolani Akinlade

A note to instructors

juliedoner — Tue, 31 Oct 2023 17:20:02 +0000

Contact the author

juliedoner — Tue, 31 Oct 2023 17:20:13 +0000

Report an issue
If you would like to report an issue with this textbook, please do so using this form.
Report an adoption
If you have adopted this text for use in a course, please let me know by filling out this form.
Collaboration and suggestions
If you would like to give suggestions or contribute to the textbook, please e-mail me at julie dot doner at umanitoba dot ca. I am especially open to:

Language profiles, especially from community members

Problem sets and practice exercises

Suggestions of hidden curriculum topics to include

Non-English examples, especially of understudied or minoritized languages

Applications to current events

Interesting resources to add to the References and further resources section

Suggestions about how to further promote Equity, Diversity, Inclusion, and Decolonization

Multimedia resources

I am open to student-written contributions.

Revisions history

juliedoner — Tue, 31 Oct 2023 17:21:39 +0000
As The Linguistics Analysis of Word and Sentence Structures is an online textbook, it will be continuously updated. Major revisions, such as the addition of new content, will normally be listed here. Minor revisions, such as the correction of typos and clarification of wording, will not normally be listed.

Version # Date Description of changes

1.0 January 5, 2024 Initial release of Chapters 1 and 2

1.1 January 7, 2024 Initial release of Chapter 3

1.2 January 17, 2024 Initial release of Chapter 4 (less 4.8)

1.3 January 24, 2024 Initial release of Chapter 8 (previously chapter 6)

Becoming a linguist

juliedoner — Thu, 02 Nov 2023 18:19:44 +0000
Throughout this textbook, I have included a series of sections under the name Becoming a linguist. These sections have three main goals:

To help students learn the "hidden curriculum" of the field of linguistics; that is, the knowledge and skills needed to succeed in linguistics, but aren't always explicitly taught.

To incorporate writing throughout the curriculum, as with the "Writing across the curriculum" approach, to help students develop their general writing skills, learn genre-specific linguistics writing skills, and deepen their learning through writing tasks.

To guide instructors, especially new instructors, in incorporating hidden curriculum and writing instruction and assessment into their courses.

End-of-chapter exercises also include activities that support the Becoming a linguist content, especially the Communication and study skills exercises.
Becoming a linguist content
Currently, the following Becoming a linguist sections are already included in the text, with more on the way!

Section 1.4: Study tips for linguistics classes

Section 1.5: Myths and misconceptions about academic integrity (by Connor Mark)

Section 2.5: Identifying academic sources

Section 2.6: Scientific writing in morphosyntax

Section 1.7: Citation in morphology and syntax

Section 3.7: Empirical and theoretical arguments

Section 4.6: Discussing data

Section 4.7: Glossing spoken language data

Section 6.8: Anatomy of an academic article

Section 6.9: How to read an academic article

Planned sections:

Section 4.8: Glossing signed language data

Finding academic sources

Careers in linguistics

The research process

How to take notes

Applying to grad school

Navigating uncertainty and competing theories

The names of languages

Getting the most out of your course syllabus

If you have any suggestions, please feel free to contact the author.
References and further resources

More resources for students

Charity Hudley, Anne, Cheryl Dickter, and Hannah Franz. 2017. The Indispensable guide to undergraduate research: Success in and beyond college. New York: Teachers College Press.

Macaulay, Monica. 2011. Surviving linguistics: A guide for graduate students, 2nd edition. Somerville: Cascadilla Press.

Research on the hidden curriculum in linguistics and beyond

Bergenhenegouwen, George. 1987. Hidden curriculum in the university. Higher Education 16: 535-543.

Doner, Julianne, and Connor Mark. 2024. Explicit writing instruction in an OER textbook. Oral presentation at the annual meeting of the Canadian Linguistics Association. Slides.

Doner, Julianne, Lisa Sullivan, Emilia Melara & Heather Yawney. 2024. Why aren't we teaching writing?: The advantages of early explicit writing instruction in linguistics. Canadian Journal of Linguistics special issue on Teaching and Learning in Linguistics.

Pappas, Panayiotis A., Maite Taboada & Kathryn Alexander. 2019. Teaching linguistic argumentation through a writing-intensive approach. Language 95(3): e339–e363.

Petrucci, Peter R. 2002. A writing-to-learn approach to writing in the discipline in the introductory linguistics classroom. The WAC Journal 13: 133–143.

Portelli, John, 1993. Exposing the hidden curriculum. Journal of Curriculum Studies 25(4): 343-358.

Smith, Buffy. 2013. Mentoring at-risk students through the hidden curriculum of higher education. Lexington Books.

complementizer

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
A complementizer (C) is the category that heads a clause (=sentence). By definition, all clauses are headed by a complementizer.

distribution

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
The distribution of an element refers to where it occurs, including its syntactic, morphological, and even phonological context.

Yoruba Test

markc — Thu, 15 Feb 2024 18:09:13 +0000

About the cover

juliedoner — Mon, 06 May 2024 18:58:37 +0000
In March 2024, we ran a design contest for the cover of The Linguistics Analysis of Word and Sentence Structures. After receiving many excellent contributions, we selected the winning design by Ysabelle Tankeko because of its aesthetic appeal and her incorporation of linguistic elements into the design. [caption id="attachment_2764" align="aligncenter" width="1707"] Figure 1. The cover for this textbook, designed by Ysabelle Tankeko.[/caption]
About the artist
Ysabelle Tankeko is a second-year University of Manitoba student, majoring in psychology, with a minor in linguistics and biological science. In the future, Ysabelle hopes to become a speech-language pathologist or an occupational therapist. Though she is still unsure of which career path she wants to pursue, Ysabelle has a passion for helping children. During her free time, Ysabelle enjoys crafting, playing badminton, and hanging out with her dog, Akina. Ysabelle also loves picking up new hobbies and trying new things. Ysabelle has always had a passion for art and design, and with her newfound interest and appreciation for linguistics, she decided to join the textbook design contest.
Artist's note on the cover design
Regarding the textbook design, I tried to illustrate things I remember learning from LING 1000 with Dr. Will Oxford. Both morphology and syntax use tree diagrams to show the structure of words and sentences, leading to the creation of the tree. The tree is displayed upside down to reflect the structure of the tree diagram and its branches. The puzzle represents the building of words or sentences. I always thought that figuring out the structure of a tree diagram was similar to completing a puzzle. I’ve also incorporated the abbreviations of phrases found in syntax structure. Following the X-bar template, I tried to position the letters similarly to how a complete diagram would look like.

Reduplication

juliedoner — Sat, 11 May 2024 13:04:44 +0000

Indigenous languages

juliedoner — Tue, 25 Jun 2024 02:49:26 +0000

What is an Indigenous language?
In brief, an Indigenous language is a language used by or associated with an Indigenous people group. Often, Indigenous languages have historical, cultural, and/or spiritural significance for the Indigenous group(s) that use them. In its simplest meaning, indigenous is simply an adjective describing originating in a particular place. Under this definition, all languages are indigenous to somewhere, since all languages arose in a particular community in a particular place. More often, though, the term indigenous is used in a more specific manner. However, as there are more than 5,000 distinct Indigenous groups in the world (according to the United Nations), it is difficult to provide a precise definition. Instead of providing a definition of indigeneity, the United Nations Declaration on the Rights of Indigenous Peoples states that Indigenous peoples have a right to self-identify as Indigenous. This can lead to tricky situations, though, when groups with dubious claims to Indigeneity demand political Indigenous status. However, some commonalities between Indigenous groups can be identified: continued occupation of a land since before colonization, distinct cultures and political systems, and the desire to maintain institutions distinct from the majority population.
Indigenous peoples have in common a historical continuity with a given region prior to colonization and a strong link to their lands. They maintain, at least in part, distinct social, economic and political systems. They have distinct languages, cultures, beliefs and knowledge systems. They are determined to maintain and develop their identity and distinct institutions and they form a non-dominant sector of society.
-United Nations

Indigenous Peoples (and their languages) are found in every region of the world. The majority of the world's languages are used by Indigenous Peoples.
Indigenous Peoples challenge! How many Indigenous people groups are you aware of? Try to name as many Indigenous people groups as you can from memory, such as the following:

The Indigenous people groups of where you currently live

The Indigenous people groups of the places you grew up

The Indigenous people groups of where your ancestors are from

An Indigenous people group from every continent of the world

An Indigenous people group for every letter of the alphabet

An Indigenous people group from every country of the world

Consider reading about the Indigenous people groups from regions that are especially relevant to you! Many of you will be surprised by what you find.

References and further resources
The references in this section that have at least one Indigenous author are indicated with an ➡️.
For a general audience

🔍 Oxford, Will. 2019. Indigenous Languages in Canada. Canadian Language Museum. https://languagemuseum.ca/wp-content/uploads/2022/12/indigenous_languages_in_canada_0.pdf

United Nations. No date. Indigenous peoples. Accessed May 16, 2024. https://www.un.org/en/fight-racism/vulnerable-groups/indigenous-peoples

United Nations. 2007. United Nations declaration on the rights of Indigenous peoples. https://social.desa.un.org/issues/indigenous-peoples/united-nations-declaration-on-the-rights-of-indigenous-peoples

Academic sources

➡️ 📑 Younging, Gregory. 2018. Elements of Indigenous Style. Brush Education Inc.

test

juliedoner — Fri, 09 Aug 2024 00:44:47 +0000
this is a test of the line break
in the small caps class
just a small page to test in isolation.

8.3. Verbs

juliedoner — Sat, 10 Aug 2024 22:03:15 +0000
**THIS SECTION IS INCOMPLETE** As with other parts of speech, we can identify verbs based on their inflectional morphology, derivational morphology, and syntactic distribution. In terms of inflectional morphology, nouns may inflect for tense, aspect, mood, evidentiality, transitivity, polarity, and argument agreement. In English, verbs inflect for tense and subject agreement. The English verbal system also indicates aspect, mood, voice, and polarity through periphrasis, which means the inflectional values are indicated through multiple words, rather than all on one word. There are many different kinds of derivational morphemes that modify verbs. Some examples of derivational morphemes in English are shown below in (1).

(1) affix examples

a. -ate dissipate, differentiate

b. -ize/-ise regularize, civilize

c. -ify terrify, qualify

d. re- rewrite, rewatch

e. un- undo, unlock

f. de- devalue, deselect

g. dis- disappear, discontinue

h. -en harden, blacken

i. mis- mislead, misuse

j. be- befriend, belittle

k. co- co-operate, co-exist

l. fore- foreclose, foresee

m. inter- interact, intervene

n. pre- prejudge, predict

o. sub- subvert, subdivide

p. trans- transform, transcribe

q. over- oversleep, overwork

r. out- outperform, outlast

In terms of syntactic distribution, verb phrases are typically the predicate of a clause. The distribution of verb phrases can be described with reference to the subject and tense markers. (1) Verbs can be modified by adverbs, noun phrase objects, and adpositional phrases, and so their syntactic position can also be described with reference to these. In English, verbs come before objects and prepositional phrases. Adverbs can appear on either side of the verb phrase in English. (2)
Inflectional values on verbs

Tense
Tense refers to when the event described by the verb phrase takes place. The main tense categories are past, present, and future, but some languages also have remote and/or recent tense distinctions. English arguably only has two tenses, past and non-past. Past tense is marked with -ed. (3) a. I walk. non-past b. I walk-ed. past English does have a future marker will, but in English (and in many other languages), the future acts more like a mood than a tense, as it is marked with a modal. Furthermore, the English "present tense" can be used to indicate future events, usually when they're fairly certain, in a construction known as the futurate, shown in (4). (4) a. The plane lanes tomorrow at noon. b. Class begins next Monday. c. I'm leaving in five minutes.
Aspect
Aspect refers to the internal time structure of the event. Some common aspectual distinctions include whether the event is complete (perfective) or incomplete (imperfective), whether the event is continuous (progressive), or whether the even happens repeatedly (habitual). **EXPLAIN EVENT TIME, UTTERANCE TIME, REFERENCE TIME AND ADD TIMELINES TO THE DEFINITIONS OF EACH ASPECT** English makes use of two aspectual categories: the progressive and the perfect. The progressive is marked with auxiliary BE and the present participle. It can be combined with any tense, as shown in (3). (3) a. I was sleeping. past progressive b. I am sleeping. present progressive c. I will be sleeping. future progressive The perfect (not to be confused with the perfective!) indicates an action that happened at a reference point that has relevance at the event time. It is marked with auxiliary HAVE and the past participle, as shown in (4). It can be combined with any tense. (4) a. I had eaten an apple. past perfect b. I have eaten an apple. present perfect c. I will have eaten an apple. future perfect It is also possible to combine the two aspects with each other, as in (5). Note that, when the perfect and the progressive are combined together, the first auxiliary is the perfect auxiliary HAVE, followed by the auxiliary BE in the past participle form. Thus, the word been simultaneous acts as the second part of the perfect marking (the past participle) and the progressive auxiliary. Been is followed by the main verb in the present participle form, the second part of the progressive marking. (5) a. I had been eating an apple. past perfect progressive b. I have been eating an apple. present perfect progressive c. I will have been eating an apple. future perfect progressive
Mood
Some sentences refer to events or states that have actually happened or are happening, while others indicate things like hypotheticals, alternatives to fact, predictions, and uncertain situations. We can distinguish between these different kinds of situations through mood. The two main broad categories of mood include realis mood—events and states that have actually occurred—and irrealis mood—those which have not actually occurred. **EXAMPLES OF MOOD IN DIFFERENT LANGUAGES AND DEFINITIONS OF COMMON MOOD CATEGORIES** Languages can indicate mood through verbal inflection or through periphrastic strategies such as adverbs or if-clauses. The primary way to indicate mood in English is through modals. **TYPOLOGY OF MOOD -- STRONG/WEAK MODALITY VS DEONTIC/EPISTEMIC MODALITY**
Evidentiality
Evidentiality indicates how the information encoded in the sentence is known by the speaker or signer. English does not make use of any evidential markings. **EXAMPLES OF EVIDENTIALITY IN DIFFERENT LGS AND COMMON CATEGORIES***
Transitivity
In some languages, verbs may inflect to indicate whether they are transitive or intransitive. In addition, some verbal inflection may indicate a change in the argument structure of the verb. In English, we have one inflectional category that indicates a change in argument structure: the passive voice. In the passive, the subject argument of the verb is removed and the object is promoted to subject. The English passive is marked by the auxiliary BE and the past participle, as shown in (x). The subject may also be indicated in a passive through an optional by-phrase. (x) a. The apple was eaten (by me). past passive b. The apple is eaten (by me). present passive c. The apple will be eaten (by me). future passive Note that the passive voice can be combined with the perfect, the progressive, or both. (y) a. The apple had been eaten. past perfect passive b. The apple was being eaten. past progressive passive c. The apple had been being eaten. past perfect progressive passive
Watch out! There is more than one way to delete the subject in English, and they aren't all the structural passive voice! Two other ways to delete the subject in English include the middle construction in (B) and the got passive in (C). Compare these to the active sentences in (A). (A) a. The torpedo sank the boat. b. I read the book. (B) a. The boat sank. b. This book reads easily. (C) a. The boat got sunk. b. The book got read. To identify the structural passive voice, don't only look at the meaning. Check which auxiliary and participle are being used!

Polarity
Polarity refers to whether the clause is positive or negative. In English, positive sentences are unmarked while negative sentences are marked with the word not or the inflectional morpheme n't.
Agreement
In some languages, verbs will agree in person and/or number with one or more of its arguments. **HIERARCHY OF AGREEMENT** **RICH VS. NOT RICH AGREEMENT** English has a limited amount of subject agreement. Most verbs agree with third person singular subjects in the present tense only. Agreement does not occur with other subjects or in other tenses, as shown in Table X.
Table X. The paradigm of the English verb walk

present past

1SG walk walk-ed

2SG walk walk-ed

3SG walk-s walk-ed

1PL walk walk-ed

2PL walk walk-ed

3PL walk walk-ed

The verb BE is irregular and exhibits a little more agreement, as shown in Table X, but it still does not encode enough distinctions to be considered rich agreement.
Table X. The paradigm of the English verb be

present past

1SG am was

2SG are were

3SG is was

1PL are were

2PL are were

3PL are were

Derived verbs
The most common way to derive verbs is from other verbs, as shown in Table X.
Table X. Deverbal verbs (V → V) (adapted from Haspelmath and Sims 2010: 88)

causative verb (N/A for now?) English →

Korean cwuk 'die' → cwuk-i- 'kill'

applicative verb (N/A for now?) English →

German laden 'load' → be-laden 'load into'

anticausative verb (N/A for now?) English →

Swedish öppna 'open (tr.)' → öppna-s 'open (intr.)'

desiderate verb (N/A for now?) Greenlandic sini- 'sleep' → sini-kkuma- 'want to sleep'

repetitive verb (N/A for now?) English write → re-write

reversive verb (N/A for now?) English buckle → un-buckle

Swahili chom-a 'stick in' → chom-o-a 'pull out'

There are also some derivational morphemes that create verbs out of nouns and adjectives, as shown in Tables X and X, but they are less common. English typically uses [pb_glossary id="1731"]conversion[/pb_glossary] to create denominal verbs, when a verb changes part of speech without morphological changes.
Table X. Denominal verbs (N → V) (adapted from Haspelmath and Sims 2010: 88)

'act like N' (N/A for now?) English →

Spanish pirat-a 'pirate' → pirat-ear 'pirate'

'put into N' (N/A for now?) English bottle (noun) → bottle (verb)

'cover with N' (N/A for now?) English blanket (noun) → blanket (verb)

Russian sol' 'salt' → sol-it 'salt'

Table X. Deadjectival verbs (A → V) (adapted from Haspelmath and Sims 2010: 88)

factitive (N/A for now?) English white → whit-en

Russian čern-yj 'black' → čern-it 'make black'

inchoative (N/A for now?) English ripe → rip-en

Spanish verde 'green' → verde-ar 'become green'

Key takeaways

English uses periphrasis to indicate aspect and voice through the combination of an auxiliary and participle.

auxiliary participle example

progressive aspect BE present participle I was eating an apple.

perfect aspect HAVE past participle I had eaten an apple.

passive voice BE past participle The apple was eaten.

Check yourself!

Open and closed categories

juliedoner — Sat, 10 Aug 2024 22:04:58 +0000

(TEST) Interview Stuff

markc — Wed, 25 Sep 2024 19:59:46 +0000
As shown in (1a) and (1c), when the subject is third-person singular, we get sees. Conversely, in (1b) and (1d), when the subject is third-person plural, see is used. On the other hand, the object has no effect on the conjugation of the verb.

(1) a. The monkey sees a visitor on the bench.

b. The monkeys see a visitor on the bench.

c. The monkey sees some visitors on the bench.

d. The monkeys see some visitors on the bench.

When you have a subject with a demonstrative in English, both the number of the demonstrative and the number of the noun must agree with the verb in the present tense. Otherwise, the utterance is ungrammatical. In (2a), the demonstrative and noun are both singular and the verb has third-person singular agreement, so the utterance is grammatical. Similarly, in (2h), the demonstrative and noun are plural while the verb has third-person plural agreement, so it is again grammatical. Conversely, in (2b-2g), there is some lack of agreement and the utterances are therefore ungrammatical. (2) a. This dog is cute. b. *This dog are cute. c. *This dogs is cute. d. *This dogs are cute. e. *These dog is cute. f. *These dog are cute. g. *These dogs is cute. h. These dogs are cute.

Julie's test

juliedoner — Wed, 25 Sep 2024 20:33:00 +0000
As shown in (1a) and (1c), when the subject is third-person singular, we get sees. Conversely, in (1b) and (1d), when the subject is third-person plural, see is used. On the other hand, the object has no effect on the conjugation of the verb.

(1) a. The monkey sees a visitor on the bench.

b. The monkeys see a visitor on the bench.

c. The monkey sees some visitors on the bench.

d. The monkeys see some visitors on the bench.

When you have a subject with a demonstrative in English, both the number of the demonstrative and the number of the noun must agree with the verb in the present tense. Otherwise, the utterance is ungrammatical. In (2a), the demonstrative and noun are both singular and the verb has third-person singular agreement, so the utterance is grammatical. Similarly, in (2h), the demonstrative and noun are plural while the verb has third-person plural agreement, so it is again grammatical. Conversely, in (2b-2g), there is some lack of agreement and the utterances are therefore ungrammatical.

(2) a. This dog is cute.

b. *This dog are cute.

c. *This dogs is cute.

d. *This dogs are cute.

e. *These dog is cute.

f. *These dog are cute.

g. *These dogs is cute.

h. These dogs are cute.

(2) a. This dog is cute. b. *This dog are cute. c. *This dogs is cute. d. *This dogs are cute. e. *These dog is cute. f. *These dog are cute. g. *These dogs is cute. h. These dogs are cute.

determiner

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
A determiner is syntactic category. Determiners specify properties of noun phrases.

grammar

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
A grammar of a language is a complete set of phrase structure rules.

Substitution

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
A family of constituency tests that tests constituency by replacing a string of words with another form, typically a pro-form.

pro-form

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
Various syntactic categories have different pro-forms, which can be used to substitute for those categories as a constituency test.

sentence fragment

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
The sentence fragment constituency tests asks whether the phrase that you're testing can stand on its own.

antecedent

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
An antecedent is a referring expression that forms a dependency with some lower pronoun or anaphor.

recursive

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
Something is recursive if it is a process that applies to itself. Practically, in syntax, something is recursive if it is a category that contains the same category (e.g., PP inside of PP, N inside of N, etc).

NP

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
Noun Phrase

endocentrism

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
Endocentrism is the property of being named in terms of one of the parts.

Headedness Principle

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
The principle that every head has a phrase, and every phrase has exactly one head.

transitive

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
A transitive verb has two arguments, a subject and an object.

object

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
A grammatical role. Objects are typically complements to V.

conjunction

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
See coordination.

R-expression

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
A referring expressions. R-expressions strictly refer, meaning that their reference cannot change in context.

reflexive

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A referring expression, also called an anaphor. A reflexives reference is not strict, meaning that it can change depending on context, but it is also sensitive to syntactic structure. Reflexives are subject to Condition A of binding theory.

pronoun

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A referring expression. A pronoun's reference is not strict, meaning that it can change depending on context. Note that pronouns are not nouns, they are syntactically determiners.

c-command

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A c-commands B iff A's mother dominates B.

dominate

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A dominates B if A is above B in a tree. A immediately dominates B if A dominates B and there are no other projections that dominate B which do not dominate A.

coreference

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A and B corefer if they refer to the same individual(s).

Condition C

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
The condition of binding theory that governs R-expressions. "R-expressions cannot be bound."

binding domain

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A binding domain is a minimal distance that some element makes reference to in its distribution. The binding domain for this class is the nearest dominating TP.

pro-drop

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A typological phenomenon characterized by the ability to have apparently silent pronouns. Languages like Spanish, Italian, and Swahili are pro-drop languages.

S-structure

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
Surface structure. This is the structure that built after movement and selection have been satisfied.

phrasal movement

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
Movement of phrases. There are two varieties of phrasal movement: A-movement and A'-movement. Phrasal movement contrasts with head-movement.

UTAH

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
The Uniformity of Theta Assignment Hypothesis. This hypothesis holds that each thematic role has a dedicated position in the syntax.

passivization

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A syntactic process in which an object is promoted the subject position, and a subject is demoted to an adjunct.

VPISH

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
The VP-Internal Subject Hypothesis. The hypothesis that all subjects start inside the verb phrase. This idea has been superceded by VoiceP.

expletive

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
Expletive arguments are semantically vacuous, appearing only to satisfy syntactic constraints like c-selection and/or the Subject Condition.

quantifier float

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
A process by which quantifiers appear in places that have been moved away from.

idioms

juliedoner — Wed, 15 Nov 2023 05:29:49 +0000
Idioms or "idiom chunks" are non-compositional phrases: the meanings of the constituents do not transparently reflect the meaning of the whole.

raising

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
Raising (or "raising to subject") involves movement out of one clause into the subject position of a higher clause. See also "raising verbs".

control

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
Control is the formal relationship between PRO ("big pro") and its antecedent.

elided

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
An element is considered elided if it is not pronounced, but does not display a movement relationship between the silent portion and the antecedent.

cyclic movement

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
Movement is cyclic if it occupies the same position in different clauses more than once.

Subject Condition

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
Spec-TP must be filled at S-structure. The Subject Condition guarantees that there is always a "subject" of every clause.

bound

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A is bound by B iff A and B are coreferential, and B c-commands A.

binding

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A syntactic relation. A binds B iff A and B are coreferential, and A c-commands B.

plural

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A number category, indicating non-singular.

morpheme

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A morpheme is the smallest unit of sound-meaning correspondence. All words are made up of one or more morphemes.

modifies

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A modifies B if A restricts the meaning of B. In this class, "modify" means the same as "adjoin."

coordination

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
The coordination constituency test uses conjunctions (and, but, not,...) to test for constituency.

functional

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
Functional words typically provide relationships between lexical words, or specify properties of lexical words. Their meanings are often difficult to define, and are dependent on the presence of a lexical word.

lexical

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
Lexical words have "content," and are typically easily definable, referring to things, events, actions, or abstract properties. They contrast with functional words.

evidential

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
Evidential markers indicate how the speaker learned about the information they are conveying. "Visual" evidentials indicate that the speaker saw the event."Hearsay" indicate that the speaker is reported secondhand information.

pronoun

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
A category of words. Note that pronouns are not nouns. They are determiners.

complement

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
X is a complement if its mother is a bar-level (Y') and its sister is a head (Y).

specifier

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
X is a specifier if its mother is a phrase level (YP) and it's sister a bar-level (Y').

terminal node

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
A terminal node is a node that ends in a lexical item, i.e., a word or a morpheme. Phrases and bar-levels are not terminal.

daughter

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
X is the daughter of Y if X is directly under Y (is immediately dominated by Y). This relationship can also be stated as "Y is the mother of X."

sister

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
X and Y are sisters if X and Y share the same mother---that is, if they're both directly under the same thing (immediately dominated by the same phrase).

tense-lowering

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
An instance of head-movement. Tense lowering moves T to the next lowest c-commanded head.

animate

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
An animate noun is something is alive, typically referring to something that has a conscious. Humans are animate, trees are not.

head-movement

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
Movement of heads. Head-movement is constrained by the Head-Movement Constraint. Haed-movement is distinct from Phrasal Movement.

free

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
A DP is free if it is not bound, i.e., c-commanded by a coreferential expression.

wh-movement

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
Movement of a wh-word: who, what, how, why, where, when, etc. We often use "wh-movement" as a shorthand for "A'-movement." However, while all wh-movement is A'-movement, not all A'-movement is wh-movement.

question C

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
A version of the complementizer used in questions.

polar questions

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
Yes/No questions, or questions that can be answered either "yes" or "no."

ellipsis

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
An instance of an unpronounced constituent. For example, in "Tegan called Joanna, and Sam did, too," there is an unpronounced VP [ called Joana ] in the second half of the sentence.

number

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
A morphological category referring to either singular or plural (and in some languages including dual and trial as well).

quantifiers

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
Quantifiers are sub-category of Determiners and include things like every, some, each, all, few, most, etc. Quantifiers are used to express a proportion of the associated NP.

morphology

juliedoner — Fri, 03 Nov 2023 07:49:02 +0000
The study of the structure of words; the study of the systematic co-variation of form and meaning.

morpheme

juliedoner — Fri, 03 Nov 2023 07:49:28 +0000
The smallest meaningful unit in language.

productivity

juliedoner — Fri, 03 Nov 2023 07:52:05 +0000
The property of language that allows language users to create novel utterances and have those utterances be understood by others.

embedded clause

juliedoner — Fri, 03 Nov 2023 07:54:48 +0000
A clause that is included inside of a complex sentence, as a modifier or dependent on the main clause.

possessive

juliedoner — Fri, 03 Nov 2023 07:56:03 +0000
A grammatical marking indicating possession.

prepositional phrase

juliedoner — Fri, 03 Nov 2023 07:56:51 +0000
A phrase with a preposition as its head, frequently modified by a noun phrase.

competence

juliedoner — Fri, 03 Nov 2023 07:59:25 +0000
Having acquired the rules of grammar in a particular language, regardless of the ability to produce grammatical language in a particular instance.

DP-hypothesis

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
The DP-hypothesis proposes that all NPs are encased in a DP "shell."

SVO

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
SVO languages have the word order Subject-Verb-Object in positive declaratives.

head-movement constraint

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
Informally, the head-movement constraint says that heads cannot skip over intervening heads. Formally, the head-movement constraint says that a head can only move to the nearest c-commanding or c-commanded head.

oblique

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
An oblique argument is "non-core" argument. Non-core argument are typically adjuncts, and very often appear in prepositional phrases (in English) or in semantic cases (like "instrumental") in other languages.

topicalization

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
A constituency test where some element is preposed to the front of the sentence, usually with comma intonation, e.g., "Mary, I saw."

D-structure

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
D-structure (or Deep-structure) is the level of representation before movement. It the level at which selection is satisfied must be satisfied.

Condition A

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
The condition of Binding Theory that stipulates that "An anaphor must be bound in its binding domain."

external argument

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
An external argument is any argument of the verb that is introduced outside of the VP---it is "external" to VP.

internal argument

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
An internal argument is any argument of that is introduced inside of the VP---it is "internal" to VP.

adpositional

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
An "adposition" is a functional particle or word that determines something about a following noun. Often, adpositions give information about location: on, over, under, in. English only has prepositions, because the adposition comes in front ("pre-") of the noun Many languages have "postpositions", where the adposition comes after the noun ("post-").

generative

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
A generative syntactic theory is one which proposes a set of abstract rules can can "generate" any and every human language.

suppletive

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
ADD DEFINITION

verb particle

juliedoner — Wed, 15 Nov 2023 05:29:54 +0000
A prepositional element that co-occurs with a verbal element, often affecting the meaning of the verb in subtle ways.

performance

juliedoner — Fri, 03 Nov 2023 08:00:24 +0000
The production of language that conforms to the rules of grammar of a particular language, regardless of whether the rules in question have been acquired by the language user.

categorical

juliedoner — Fri, 03 Nov 2023 08:07:16 +0000
A pattern or rule with no exceptions.

poverty of the stimulus argument

juliedoner — Fri, 03 Nov 2023 08:08:53 +0000
An argument used to support Universal Grammar based on the observation that language is infinite and the premise that infinite systems are unlearnable.

Universal Grammar (UG)

juliedoner — Fri, 03 Nov 2023 08:10:08 +0000
Chomsky's theory that there is an innate, language-specific genetic component underlying the human capacity for language.

Scientific Method

juliedoner — Fri, 03 Nov 2023 08:16:39 +0000
The method of inquiry used in science whereby researchers observe data, and then formulate, test, and revise hypotheses based on their observations.

Chomskyan

juliedoner — Fri, 03 Nov 2023 08:18:13 +0000
Any theories or models that are based on the work of Noam Chomsky.

Minimalism

juliedoner — Fri, 03 Nov 2023 08:26:05 +0000
A research program/framework in generative syntax, beginning in the 1990s and continuing through the present day, with the goal of modeling a maximally efficient computational system to account for the attested syntactic patterns of the languages of the world.

recursion

juliedoner — Fri, 03 Nov 2023 08:28:24 +0000
Iteratively applying a rule to its own output.

compound word

juliedoner — Fri, 03 Nov 2023 08:29:39 +0000
A word with two or more roots.

International Phonetic Alphabet

juliedoner — Fri, 03 Nov 2023 08:36:58 +0000
An internationally recognized set of symbols developed in order to transcribe speech sounds with a one-to-one correspondence between sound and symbol.

phonology

juliedoner — Fri, 03 Nov 2023 08:39:07 +0000
The study of the systematic rules and constraints that characterize human speech sounds and signs in and across languages.

simplex word

juliedoner — Fri, 03 Nov 2023 08:50:53 +0000
A word containing only a single morpheme.

complex word

juliedoner — Fri, 03 Nov 2023 08:51:08 +0000
A word containing two or more morphemes.

polymorphemic

juliedoner — Fri, 03 Nov 2023 08:53:00 +0000
Containing multiple morphemes.

segmented

juliedoner — Fri, 03 Nov 2023 09:33:05 +0000
(of a word) Broken up into morphemes.

syntax

juliedoner — Fri, 03 Nov 2023 18:59:33 +0000
The study of the structure of sentences and phrases; the study of how we combine words and encode the relationships between them.

linguistics

juliedoner — Fri, 03 Nov 2023 19:01:31 +0000
The scientific study of language.

phrase

juliedoner — Fri, 03 Nov 2023 19:53:52 +0000
Any grammatically well-formed string of words.

sentence

juliedoner — Fri, 03 Nov 2023 19:54:37 +0000
A grammatically well-formed string of words that communicates a complete thought.

nonce word

juliedoner — Fri, 03 Nov 2023 20:22:27 +0000
A made-up word for one-time use.

preposition

juliedoner — Fri, 03 Nov 2023 20:44:05 +0000
A word describing the relationship between a noun and another part of the phrase and that occurs before the noun. See also adposition.

postposition

juliedoner — Fri, 03 Nov 2023 20:44:32 +0000
A word describing the relationship between a noun and another part of the phrase and that occurs after the noun. See also adposition.

adposition

juliedoner — Fri, 03 Nov 2023 20:47:11 +0000
A preposition or a postposition. A word describing the relationship between a noun and another part of the phrase.

semantics

juliedoner — Fri, 03 Nov 2023 21:06:23 +0000
The study of the conventional meaning of language, including the meaning of individual words (lexical semantics) and how meaning is derived from their combination (compositional semantics).

citation

juliedoner — Wed, 08 Nov 2023 03:34:29 +0000
The attribution of quotes and information to their original sources.

replicable

juliedoner — Wed, 08 Nov 2023 07:15:46 +0000
Research that can be repeated in different contexts with similar results.

natural language

juliedoner — Thu, 09 Nov 2023 05:42:04 +0000
Language that arose out of use in a community, rather than a language that was designed or constructed.

morphosyntax

juliedoner — Thu, 09 Nov 2023 06:13:49 +0000
The study of the structure of words, phrases, and sentences. The intersection of the fields of morphology and syntax.

constructed language

juliedoner — Fri, 10 Nov 2023 00:33:47 +0000
A language designed for a particular purpose, as opposed to a natural language.

phonetics

juliedoner — Fri, 10 Nov 2023 00:40:26 +0000
The study of the physical properties of language, including how it is produced and perceived.

pragmatics

juliedoner — Fri, 10 Nov 2023 00:40:48 +0000
The study of the meaning of language in context.

discourse analysis

juliedoner — Fri, 10 Nov 2023 00:41:12 +0000
The study of how multiple sentences are combined and interact, including turn-taking in conversation.

linguistic typology

juliedoner — Fri, 10 Nov 2023 00:41:32 +0000
The study of the classification of languages according to their structural properties.

sociolinguistics

juliedoner — Fri, 10 Nov 2023 00:42:27 +0000
The study of how language is used in its social context, including variation based on broad social categories like age, gender, and region, and the intersection of language and identity.

anthropological linguistics

juliedoner — Fri, 10 Nov 2023 00:42:48 +0000
The study of the relationship between language and culture.

crip linguistics

juliedoner — Fri, 10 Nov 2023 00:45:09 +0000
A critical approach to the study of linguistics from the approach of disability studies, studying how people with differing abilities communicate, without taking a deficit approach.

historical linguistics

juliedoner — Fri, 10 Nov 2023 00:45:30 +0000
The study of how language changes over time.

psycholinguistics

juliedoner — Fri, 10 Nov 2023 00:45:49 +0000
The study of how language is processed and planned.

neurolinguistics

juliedoner — Fri, 10 Nov 2023 00:46:03 +0000
The study of how language is related to the anatomy of the brain.

language acquisition

juliedoner — Fri, 10 Nov 2023 00:46:19 +0000
The study of how children and adults acquire and learn language.

computational linguistics

juliedoner — Fri, 10 Nov 2023 00:50:01 +0000
The study of how we can model and process language with computers.

compositionality

juliedoner — Fri, 10 Nov 2023 00:54:30 +0000
The observation that the meaning of a sentence is derived by the meaning of its parts and the way that they are combined.

allomorphy

juliedoner — Fri, 10 Nov 2023 00:57:04 +0000
When the form of a morpheme varies depending on context.

prosody

juliedoner — Fri, 10 Nov 2023 01:04:01 +0000
The patterns of stress and intonation in a language.

language acquisition

juliedoner — Fri, 10 Nov 2023 01:07:58 +0000
The study of how children and adults acquire and learn language.

discourse markers

juliedoner — Fri, 10 Nov 2023 01:24:00 +0000
Words or morphemes used to manage the structure and flow of discourse.

falsifiable hypothesis

juliedoner — Fri, 10 Nov 2023 08:07:23 +0000
A hypothesis which is possible to be proven wrong by some test.

hypothesis

juliedoner — Fri, 10 Nov 2023 08:08:11 +0000
A proposed analysis or explanation made based on the observation of data, to be tested over the course of scientific investigation.

prescriptive grammar

juliedoner — Fri, 10 Nov 2023 08:08:58 +0000
A set of rules that describe someone's opinion of how language should be used.

descriptive grammar

juliedoner — Fri, 10 Nov 2023 08:09:37 +0000
A set of rules that describe how language is used based on observation.

social science

juliedoner — Tue, 14 Nov 2023 07:34:23 +0000
The study of societies.

cognitive science

juliedoner — Tue, 14 Nov 2023 07:37:22 +0000
The study of the mind.

synonym

juliedoner — Tue, 14 Nov 2023 07:55:55 +0000
A word that shares the same or similar meaning with another word in the same language.

passive voice

juliedoner — Tue, 14 Nov 2023 08:20:17 +0000
A grammatical structure in which the subject of a clause is deleted or included in an optional 'by'-phrase, and the object of the clause is promoted to subject position.

part of speech

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
synonyms: lexical category, syntactic category We classify words into part of speech based on their syntactic function. Some examples of part of speech include noun, verb, adjective, and preposition.

intransitive

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
Not having an object. Most commonly used to describe verbs that have only a single argument, the subject, although other parts of speech are also occasionally described as being transitive or intransitive.

bracketing structure

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
A representation using brackets to indicate constituents and categories in morphology or syntax.

node

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
Any point where multiple branches meet in a syntax tree, as well as the endpoint of any branch.

structural ambiguity

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
When a word or sentence can be associated with more than one hierarchical structure, each resulting in different possible meanings.

head

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
A head is the main word in the phrase which determines the properties of the phrase as a whole. In syntax, it is typically a terminal node, and is realized as a morpheme or word.

morphosyntactic feature

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
The grammatical properties of a word, morpheme, or syntactic head which are relevant to its morphology or syntax.

selection

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
The requirements of a head for what it must combine with. The head selects the type, number, and semantic role of what it combines with. There is semantic selection (s-selection) and category selection (c-selection).

double object construction

juliedoner — Wed, 15 Nov 2023 05:29:50 +0000
A ditransitive construction where both objects are NPs and the indirect object comes first, as in I gave you a book.

subject-auxiliary inversion

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
When an auxiliary verb appears in front of the subject, usually in order to form a question. Formally, this is analyzed as T-to-C movement.

suffix

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
A morpheme that attaches to the end of a stem.

copula

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
A word used to introduce non-verbal predicates, including adjectival (I am happy), nominal (I am a teacher), and prepositional (I am down the street) predicates. It has roughly the meaning of an equal sign. In English, the verb BE is used to form copular constructions.

prefix

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
A morpheme that attaches to the beginning of a stem.

progressive aspect

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
The progressive is an aspectual category that indicates an incomplete or ongoing action with respect to a particular event at a time. In English, progressive aspect is expressed using a form of the verb BE and the present participle form of the verb, e.g., 'John is eating sushi.'

case

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
A morphology property of nouns in many languages. Case can either be structural, and indicate the grammatical role (subject/object) or semantic, and indicate the thematic role of the noun.

argument

juliedoner — Wed, 15 Nov 2023 05:29:53 +0000
An argument is something that is selected for and required by the head of a phrase. Arguments include grammatical roles like "subject, object, indirect object, etc."

Old English

juliedoner — Thu, 16 Nov 2023 02:58:58 +0000
The variety of English spoken before the Norman invasion of England in 1066.

Middle English

juliedoner — Thu, 16 Nov 2023 02:59:27 +0000
The variety of English spoken after the Norman invasion of England in 1066 until approximately the year 1500.

Modern English

juliedoner — Thu, 16 Nov 2023 02:59:56 +0000
The varieties of English spoken from approximately the year 1500 until the present day.

negative evidence

juliedoner — Fri, 17 Nov 2023 21:05:41 +0000
Evidence that something is not possible, does not occur, or is absent.

unattested

juliedoner — Fri, 17 Nov 2023 21:15:37 +0000
A pattern or piece of data that doesn't show up in our data source.

confirmation bias

juliedoner — Wed, 22 Nov 2023 19:44:22 +0000
The tendency for humans to look for evidence that is consistent with their belief system.

necessary evidence

juliedoner — Wed, 22 Nov 2023 20:07:30 +0000
Evidence that must be the case if your hypothesis is true.

sufficient evidence

juliedoner — Wed, 22 Nov 2023 20:13:50 +0000
Evidence that demonstrates your hypothesis is true.

hypercorrection

juliedoner — Thu, 23 Nov 2023 00:49:06 +0000
The over-application of a rule to additional contexts.

observationally adequate

juliedoner — Wed, 29 Nov 2023 04:35:39 +0000
A model that accounts for all observed data.

descriptively adequate

juliedoner — Wed, 29 Nov 2023 04:36:08 +0000
A model that accounts for all observed data, plus also speaker judgments.

explanatorily adequate

juliedoner — Wed, 29 Nov 2023 04:36:43 +0000
A model that accounts for all observed data and speaker judgments, as well as provides an explanation for how language is acquired.

generative grammar

juliedoner — Wed, 29 Nov 2023 04:52:59 +0000
Any model of grammar that uses rules to generate or "build" a language structure. This model should be able to produce all of the grammatical sentences in the language and no others.

I-language

juliedoner — Thu, 30 Nov 2023 04:47:16 +0000
Internal language. The system of grammatical rules that an individual language user has in their mind.

E-language

juliedoner — Thu, 30 Nov 2023 05:27:52 +0000
External language. The forms of language that are produced by individuals or in a community.

diachronic

juliedoner — Thu, 30 Nov 2023 05:40:29 +0000
Describing how something (such as language) has changed over time.

synchronic

juliedoner — Thu, 30 Nov 2023 05:43:10 +0000
Describing something (such as language) at a particular point in time.

acquisition

juliedoner — Tue, 12 Dec 2023 05:21:31 +0000
Developing a skill, habit, or ability subconsciously, by instinct, instead of consciously.

overregularization

juliedoner — Tue, 12 Dec 2023 23:30:05 +0000
The process of applying a rule in contexts where it should not apply, common in child speech.

algorithm

juliedoner — Tue, 12 Dec 2023 23:51:57 +0000
A procedure, process, or system of rules used to solve a problem.

lexicon

juliedoner — Fri, 15 Dec 2023 20:25:48 +0000
The part of our memory where we store the vocabulary we have learned.

lexicographer

juliedoner — Fri, 15 Dec 2023 20:38:42 +0000
A person who compiles a dictionary.

model

juliedoner — Wed, 20 Dec 2023 00:04:53 +0000
A scientific model is a representation of a system, an organism, a process, or a phenomenon.

corpus

juliedoner — Sun, 07 Jan 2024 02:13:08 +0000
plural: corpora A body of language data collected from real-world use.

linguistic elicitation

juliedoner — Tue, 09 Jan 2024 06:29:55 +0000
Working with a user of a language to collect linguistic data from that language, through asking them to translate sentences to and from the language or by asking them to describe different situations or contexts in the language.

modality

juliedoner — Tue, 09 Jan 2024 21:23:51 +0000
The type of signal through which language is transmitted and perceived, such as spoken or signed.

negative concord

juliedoner — Tue, 09 Jan 2024 23:11:28 +0000
Agreement between multiple negative words in a sentence.

mutual intelligibility

juliedoner — Tue, 09 Jan 2024 23:21:22 +0000
When speakers from two different language varieties can understand each other.

native speaker

juliedoner — Wed, 10 Jan 2024 07:00:26 +0000
Someone who has been speaking the language in question from early childhood.

object language

juliedoner — Thu, 11 Jan 2024 09:02:18 +0000
the language under discussion or being analyzed.

metalanguage

juliedoner — Thu, 11 Jan 2024 09:02:59 +0000
the language being used to discuss the object language.

gloss

juliedoner — Thu, 11 Jan 2024 23:07:47 +0000
Provide a morpheme-by-morpheme translation of linguistic data

underlying form

juliedoner — Thu, 11 Jan 2024 23:24:41 +0000
the form of a word or morpheme before phonological processes have been applied to it

lexeme

juliedoner — Fri, 19 Jan 2024 19:00:30 +0000
a word in the abstract sense, including all of its different inflected forms

word token

juliedoner — Fri, 19 Jan 2024 19:05:36 +0000
a particular occurrence of a word

word form

juliedoner — Fri, 19 Jan 2024 19:10:20 +0000
A particular form of a lexeme, one of the possible inflected forms of a lexeme.

citation form

juliedoner — Fri, 19 Jan 2024 20:31:56 +0000
The word form used to refer to a lexeme.

homonym

juliedoner — Fri, 19 Jan 2024 21:04:26 +0000
When two or more words have the same form but different meaning.

paradigm

juliedoner — Fri, 19 Jan 2024 21:46:47 +0000
The collection of all the word forms in a lexeme is called the paradigm for that lexeme, or the set of all the inflected forms of a word.

word family

juliedoner — Fri, 19 Jan 2024 21:47:13 +0000
A set of words related by derivation.

non-compositional meaning

juliedoner — Fri, 19 Jan 2024 23:54:25 +0000
A word or phrase with meaning that is not predictable from the combination of its components.

inflection

juliedoner — Sun, 21 Jan 2024 02:06:06 +0000
A morphological process that adds grammatical information, such as number, tense, gender, or case.

derivation

juliedoner — Sun, 21 Jan 2024 02:06:52 +0000
A morphological process that creates a new lexeme.

conversion

juliedoner — Sun, 21 Jan 2024 02:30:53 +0000
Changing the part of speech of a word without changing its form.

borrowing

juliedoner — Sun, 21 Jan 2024 02:32:52 +0000
When a word from one language is adopted by another language.

nativization

juliedoner — Sun, 21 Jan 2024 02:33:45 +0000
The process of a borrowed word adopting the phonology and possibly the morphology of the language it is borrowed into.

calque

juliedoner — Sun, 21 Jan 2024 02:37:31 +0000
A borrowed phrase which is a direct translation of a phrase in another language.

root creation

juliedoner — Sun, 21 Jan 2024 02:41:03 +0000
Coining a word without basis on any previously existing words.

genericization

juliedoner — Sun, 21 Jan 2024 02:42:30 +0000
When a brand name becomes the common name for all similar items.

loanword

juliedoner — Sun, 21 Jan 2024 02:46:12 +0000
A word that has been borrowed from another language.

clipping

juliedoner — Sun, 21 Jan 2024 03:03:02 +0000
The formation of a new word by shortening an existing one.

backformation

juliedoner — Sun, 21 Jan 2024 03:04:36 +0000
A process whereby a language user reanalyzes a monomorphemic word to be polymorphemic and removes the additional morphemes.

folk etymology

juliedoner — Sun, 21 Jan 2024 03:06:42 +0000
A commonly believed etymology of a word that is not historically accurate.

blend

juliedoner — Sun, 21 Jan 2024 03:09:40 +0000
A word formed by combining portions of two or more roots. At least one of the roots must be truncated in order to be a blend.

neologism

juliedoner — Sun, 21 Jan 2024 03:16:13 +0000
The coining of a new word.

taboo avoidance

juliedoner — Sun, 21 Jan 2024 03:17:27 +0000
The coining of a word in order to avoid use of a word that is socially unacceptable.

truncation

juliedoner — Sun, 21 Jan 2024 03:24:56 +0000
The formation of a new word by shortening an existing one.

acronym

juliedoner — Sun, 21 Jan 2024 03:27:31 +0000
When the first letter or two of each word in a phrase is combined to form a new word, which is pronounced as a whole.

initialism

juliedoner — Sun, 21 Jan 2024 03:27:47 +0000
synonym: alphabetism When the first letter or two of each word in a phrase is combined to form a new word, which is pronounced as a string of letters.

alphabetism

juliedoner — Sun, 21 Jan 2024 03:28:46 +0000
synonym: initialism When the first letter or two of each word in a phrase is combined to form a new word, which is pronounced as a string of letters.

backronym

juliedoner — Sun, 21 Jan 2024 03:29:44 +0000
When a word that already exists is claimed to be an acronym.

loan translation

juliedoner — Sun, 21 Jan 2024 03:31:16 +0000
A borrowed phrase which is a direct translation of a phrase in another language.

portmanteau

juliedoner — Sun, 21 Jan 2024 03:32:35 +0000
A word formed by combining portions of two or more roots. At least one of the roots must be truncated in order to be a portmanteau.

isolating language

juliedoner — Sun, 21 Jan 2024 05:51:20 +0000
A language with more or less one morpheme per word.

agglutinative language

juliedoner — Sun, 21 Jan 2024 05:52:42 +0000
A language with one meaning per morpheme and multiple morphemes per word.

fusional language

juliedoner — Sun, 21 Jan 2024 05:53:22 +0000
A language which tends to combine multiple meanings into a single morpheme.

polysynthetic language

juliedoner — Sun, 21 Jan 2024 05:54:10 +0000
A language which combines many morphemes into a single, complex word, which often corresponds to an entire sentence.

attested

juliedoner — Sun, 21 Jan 2024 08:14:27 +0000
A word or construction that exists in a particular corpus, or, more generally, that has been observed in use in a natural setting.

synthesis

juliedoner — Tue, 23 Jan 2024 01:21:18 +0000
A scale indexing how many morphemes are contained in a word.

fusion

juliedoner — Tue, 23 Jan 2024 01:21:57 +0000
A scale indexing the degree to which morphemes are phonologically separable from their stems.

exponence

juliedoner — Tue, 23 Jan 2024 01:23:00 +0000
A scale indexing how many meanings can be combined into a single morpheme.

analytic

juliedoner — Tue, 23 Jan 2024 01:31:50 +0000
A language with a low degree of synthesis. In other words, a language with few morphemes per word.

orthography

juliedoner — Tue, 23 Jan 2024 07:15:19 +0000
The writing system of a language.

infix

juliedoner — Wed, 24 Jan 2024 00:54:05 +0000
An affix that is inserted internally to the stem.

idiom

juliedoner — Wed, 24 Jan 2024 04:44:57 +0000
A phrase with non-compositional meaning.

hierarchical structure

juliedoner — Wed, 14 Feb 2024 06:31:36 +0000
The organization of elements into ranks or levels, where each level contains or is in charge of the lower levels.

constituent

juliedoner — Fri, 16 Feb 2024 22:42:17 +0000
A group of words or morphemes that behave as a unit.

lexical ambiguity

juliedoner — Fri, 16 Feb 2024 22:43:33 +0000
When a word or sentence is ambiguous because one of the morphemes has a homonym.

metathesis

juliedoner — Wed, 27 Mar 2024 20:53:43 +0000
IPA: /məˈtæθəˌsɪs/ The reordering of phonological segments.

etymology

juliedoner — Fri, 19 Apr 2024 21:22:33 +0000
The study of the historical origins of words and morphemes, including change over time in their form and/or meaning.

affixes

juliedoner — Sat, 11 May 2024 13:07:48 +0000
bound morphemes that are attached to a stem, usually linearly.

circumfix

juliedoner — Sat, 11 May 2024 16:21:18 +0000
An affix that is inserted on both sides of the stem.

suprafix

juliedoner — Sat, 11 May 2024 16:21:53 +0000
An affix that is encoded through the prosodic structure of the stem, such as through stress, length, or pitch.

parasynthesis

juliedoner — Sat, 11 May 2024 18:39:57 +0000
A phenomenon in which one meaning or group of meanings is encoded simultaneously by more than one morphological element or process.

genitive

juliedoner — Sat, 11 May 2024 21:07:17 +0000
Case marking possession.

partitive

juliedoner — Sat, 11 May 2024 21:08:19 +0000
Case used to mark a variety of meanings including partialness and incompleteness.

root

juliedoner — Sun, 19 May 2024 06:17:29 +0000
The core morpheme of a word.

affix

juliedoner — Sun, 19 May 2024 06:18:07 +0000
A bound morpheme that attaches to a stem.

stem

juliedoner — Sun, 19 May 2024 06:18:49 +0000
The portion of a word that hosts an affix.

base

juliedoner — Sun, 19 May 2024 06:18:59 +0000
The portion of a word that hosts an affix.

free morpheme

juliedoner — Sun, 19 May 2024 06:19:36 +0000
A morpheme that can stand on its own.

bound morpheme

juliedoner — Sun, 19 May 2024 06:20:10 +0000
A morpheme that must attach to a stem.

wug test

juliedoner — Fri, 24 May 2024 21:07:48 +0000
An experimental study design whereby participants manipulate novel word forms, testing whether a word formation rule is productive.

allomorph

juliedoner — Sat, 25 May 2024 22:25:47 +0000
A variant form of a morpheme.

voiceless

juliedoner — Sat, 25 May 2024 22:26:10 +0000
Pronounced without vocal fold vibration.

voiced

juliedoner — Sat, 25 May 2024 22:26:27 +0000
Pronounced with vocal fold vibration.

sibilant

juliedoner — Sat, 25 May 2024 22:27:23 +0000
High-pitched consonant pronounced with the tongue tip, such as /s z ʃ ʒ t͡ʃ d͡ʃ/.

phonological allomorph

juliedoner — Tue, 25 Jun 2024 21:40:56 +0000
Allomorphy that is predictable based on phonological context.

suppletion

juliedoner — Tue, 25 Jun 2024 21:50:27 +0000
Allomorphy which is not predictable from phonological context.

Weak suppletion

juliedoner — Tue, 25 Jun 2024 21:55:41 +0000
Allomorphy that is not predictable from phonological context but nonetheless has some similarity with other forms in the paradigm.

strong suppletion

juliedoner — Tue, 25 Jun 2024 21:56:25 +0000
Allomorphy which is not predictable based on phonological context and has no phonological similarity to other forms in the paradigm.

comparative

juliedoner — Tue, 25 Jun 2024 21:56:56 +0000
An inflectional form of adjectives meaning "to a greater degree of X."

superlative

juliedoner — Tue, 25 Jun 2024 21:58:08 +0000
An inflectional form of adjectives meaning "to the highest degree of X."

ablaut

juliedoner — Tue, 25 Jun 2024 21:59:40 +0000
synonym: internal change A morphological process that changes the vowel in the stem.

regularization

juliedoner — Tue, 25 Jun 2024 22:00:42 +0000
A historical process of language change whereby irregular forms adopt the majority pattern.

analogy

juliedoner — Tue, 25 Jun 2024 22:01:31 +0000
A language change process whereby one word form adopts a pattern used by another word form.

etymologists

juliedoner — Tue, 25 Jun 2024 22:02:05 +0000
People who study the historical development of words.

phonological conditioning

juliedoner — Wed, 26 Jun 2024 19:54:13 +0000
When allomorphy is triggered by phonological context.

morphological conditioning

juliedoner — Wed, 26 Jun 2024 19:54:42 +0000
When allomorphy is triggered by morpological context, usually inflectional features of the stem.

lexical conditioning

juliedoner — Wed, 26 Jun 2024 19:57:19 +0000
When allomorphy is triggered by the lexical properties of the stem, such as its semantics. This also includes allomorphy which is not predictable from a pattern.

tense

juliedoner — Fri, 28 Jun 2024 19:20:53 +0000
Verbal inflection that indicates when the event took place, such as past, present, or future marking.

aspect

juliedoner — Fri, 28 Jun 2024 19:22:11 +0000
Verbal inflectional that indicates the internal time structure of an event, such as indicating that an action is complete, ongoing, repeated, etc.

peer review

juliedoner — Wed, 03 Jul 2024 21:12:01 +0000
A process where a piece of scholarship is reviewed by peers of the author. In the context of academic sources, the author and the author's peers would be expected to be experts on the topic.

concatenative morphology

juliedoner — Sat, 06 Jul 2024 20:36:09 +0000
Morphological processes that result in a linear string of morphemes, such as prefixation or suffixation.

non-concatenative

juliedoner — Sat, 06 Jul 2024 20:37:51 +0000
Morphological processes that do not result in a linear string of morphemes, such as suprafixes, transfixes, or ablaut.

root-and-pattern morphology

juliedoner — Sat, 06 Jul 2024 21:06:57 +0000
A way of forming words common in Semitic languages, in which words are formed by adding transfixes to triconsonantal roots. The transfixes determine the arrangement of the consonants and the vowels that are inserted in between the three consonants of the root, and also sometimes additional consonants.

triconsonantal root

juliedoner — Sat, 06 Jul 2024 21:07:31 +0000
A root consisting only of three consonants, as is common in root-and-pattern morphology in Semitic languages.

transfix

juliedoner — Sat, 06 Jul 2024 21:08:36 +0000
An affix that is interwoven among the segments of the root, such as in root-and-pattern morphology in Semitic languages.

internal change

juliedoner — Sun, 07 Jul 2024 01:39:36 +0000
synonym: ablaut A morphological process that changes the vowel in the stem.

past participle

juliedoner — Sun, 07 Jul 2024 01:42:05 +0000
A form of the English verb used in the perfect (e.g., have sung or have eaten) or in the passive (e.g., was sung or was eaten). The past participle is often marked with -en or shares the same form as the past tense, but there are also many verbs with an irregular past participle.

bare form

juliedoner — Sun, 07 Jul 2024 01:46:00 +0000
A word without any inflectional morphemes applied to it, or the simplest form in a paradigm.

Reduplication

juliedoner — Sun, 07 Jul 2024 03:02:19 +0000
A morphological process involving the full or partial copying of the stem.

original research

juliedoner — Sat, 13 Jul 2024 23:43:32 +0000
Research that involves the creation, discovery, or documentation of new knowledge and ideas.

proper noun

juliedoner — Thu, 18 Jul 2024 03:58:43 +0000
A noun that refers to a specific entity by its name.

common noun

juliedoner — Thu, 18 Jul 2024 03:59:27 +0000
A noun that does not refer to a specific entity by name, but rather refers to a type of entity.

predicate

juliedoner — Tue, 06 Aug 2024 03:10:37 +0000
The part of a clause that describes the event that applies to the subject. Typically the verb phrase, but non-verbal predicates also exist.

cumulative exponence

juliedoner — Wed, 07 Aug 2024 20:33:40 +0000
When multiple meanings are encoded in a single morpheme.

proposition

juliedoner — Thu, 08 Aug 2024 20:35:38 +0000
A statement that can be true or false.

scope

juliedoner — Thu, 08 Aug 2024 20:57:22 +0000
The part of the sentence that is modified by a semantic operator.

Distributed Morphology

juliedoner — Fri, 09 Aug 2024 20:10:03 +0000
A model of grammar developed in the early 1990s with three main hypotheses: (a) Late Insertion, the idea that syntax is constructed using abstract items, and vocabulary items are inserted late in the process; (b) Underspecification, the idea that vocabulary items are inserted based on being the best match to the features, and do not have to be a perfect match; and (c) Syntactic Hierarchical Structure All the Way Down, the idea that morphological rules aren't separate from syntax, but that instead, syntactic rules are used to construct both words and sentences.

lexical parts of speech

juliedoner — Sun, 11 Aug 2024 02:04:14 +0000
Parts of speech that encode the meaning of the sentence, such as nouns, verbs, adjectives, and adverbs.

functional parts of speech

juliedoner — Sun, 11 Aug 2024 02:05:06 +0000
Parts of speech that encode grammatical functions, such as determiners, complementizers, and conjunctions.

open parts of speech

juliedoner — Sun, 11 Aug 2024 02:11:49 +0000
Parts of speech that can easily have new words added to them.

closed parts of speech

juliedoner — Sun, 11 Aug 2024 02:12:09 +0000
Parts of speech that cannot easily have new words added to them.

lexical category

juliedoner — Sun, 11 Aug 2024 03:07:01 +0000
synonyms: part of speech, syntactic category We classify words into part of speech based on their syntactic function. Some examples of part of speech include noun, verb, adjective, and preposition.

syntactic category

juliedoner — Sun, 11 Aug 2024 03:07:22 +0000
synonyms: lexical category, part of speech We classify words into part of speech based on their syntactic function. Some examples of part of speech include noun, verb, adjective, and preposition.

present participle

juliedoner — Sun, 11 Aug 2024 03:39:54 +0000
An inflectional form of verbs used in the progressive aspect, formed by suffixing -ing to a verb stem.

perfect aspect

juliedoner — Sun, 11 Aug 2024 03:43:27 +0000
The perfect aspect is used to indicate an event which is completed by the indicated tense. Thus, in the "present perfect," the event of the verb has been completed by the present moment. In the "past perfect," the event of the verb has been completed at some past moment. In English, perfect is expressed using a form of the auxiliary verb HAVE and the past participle form of the following verb, e.g., "John has eaten sushi."

non-finite

juliedoner — Sun, 11 Aug 2024 04:08:50 +0000
synonym: infinitive Not marked for tense.

subject

juliedoner — Sat, 17 Aug 2024 04:07:47 +0000
What a sentence is about. Usually a noun phrase. In English, a subject will appears before the verb and triggers agreement on the verb.

ditransitive

juliedoner — Sat, 17 Aug 2024 04:43:09 +0000
Having three arguments, usually a subject, direct object, and indirect object.

direct object

juliedoner — Sat, 17 Aug 2024 04:46:43 +0000
An argument of the verb, the entity which undergoes the action.

indirect object

juliedoner — Sat, 17 Aug 2024 04:47:09 +0000
An argument of the verb, the goal or endpoint of an action.

prepositional ditransitive

juliedoner — Sat, 17 Aug 2024 04:49:46 +0000
A ditransitive construction where the direct object comes first as an NP and the indirect object comes second as a PP, as in I gave a book to you.

adjunct

juliedoner — Sat, 17 Aug 2024 04:51:21 +0000
An optional modifying phrase. The opposite of an argument.

serial verb

juliedoner — Tue, 20 Aug 2024 16:14:31 +0000
A construction where the predicate consists of a string of multiple verbs.

nominative

juliedoner — Wed, 21 Aug 2024 01:31:29 +0000
Case marking that appears on the subjects of finite clauses.

active voice

juliedoner — Wed, 21 Aug 2024 01:58:45 +0000
A clause where there is an agent in subject position. The opposite of the passive voice.

complementary distribution

juliedoner — Wed, 18 Sep 2024 22:02:06 +0000
When the distributions of two elements are opposites. Each element always appears in contexts the other element can never appear in.

quantifer

juliedoner — Wed, 20 Nov 2024 19:43:58 +0000
A kind of determiner that does not refer to specific entities but indicates some proportion of a group, for example each, every, all, many, and some.

morphology

juliedoner — Wed, 15 Nov 2023 05:29:47 +0000
Morphology is the study of morphemes. Morphemes are defined as the smallest unit of meaning.

compounds

juliedoner — Wed, 15 Nov 2023 05:29:48 +0000
Words formed from other words.

Headedness Principle

juliedoner — Wed, 15 Nov 2023 05:29:51 +0000
Every phrase has exactly one head and every head is in its phrase.

perfect

juliedoner — Wed, 15 Nov 2023 05:29:52 +0000
The perfect aspect is used to indicate an event which is completed by the indicated tense. Thus, in the "present perfect," the event of the verb has been completed by the present moment. In the "past perfect," the event of the verb has been completed at some past moment. In English, perfect is expressed using a finite form of the auxiliary verb HAVE and the participial form of the main verb, e.g., "John has eaten sushi."

complementizer

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000
A complementizer (C) is the category that heads a clause (=sentence). By definition, all clauses are headed by a complementizer.

morphology

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000
Morphology is the study of morphemes. Morphemes are defined as the smallest unit of meaning.

distribution

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000
The distribution of an element refers to where it occurs, including its syntactic, morphological, and even phonological context.

determiner

juliedoner — Wed, 15 Nov 2023 05:32:43 +0000
A determiner is syntactic category. Determiners specify properties of noun phrases.

grammar

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
A grammar of a language is a complete set of phrase structure rules.

syntactic categories

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Syntactic categories are parts of speech, referring to N (Nouns), D, (Determiners), V (Verbs), etc.

intransitive

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Intransitive verbs only every appear with a subject, and never appear with an object. In theoretical terms, an intransitive verb is a verb that c-selects for only one argument.

Substitution

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
A family of constituency tests that tests constituency by replacing a string of words with another form, typically a pro-form.

pro-form

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Various syntactic categories have different pro-forms, which can be used to substitute for those categories as a constituency test.

sentence fragment

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
The sentence fragment constituency tests asks whether the phrase your testing can stand on its own.

antecedent

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
An antecedent is a referring expression that forms a dependency with some lower pronoun or anaphor.

recursive

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Something is recursive if it is a process that applies to itself. Practically, in syntax, something is recursive if it is a category that contains the same category (e.g., PP inside of PP, N inside of N, etc).

compounds

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Words formed from other words.

bracketing structure

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
A syntactic representation using brackets to indicate constituents and categories.

node

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Any point in a tree.

NP

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
Noun Phrase

structural ambiguity

juliedoner — Wed, 15 Nov 2023 05:32:44 +0000
A structural ambiguity occurs when a sentence have more than one syntactic parse.

endocentrism

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
Endocentrism is the property of being named in terms of one of the parts.

head

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A head is a terminal node, and is realized as a morpheme or word. See also the Headedness Principle.

Headedness Principle.

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
Every head has a phrase, and every phrase has exactly one head.

transitive

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A transitive verb has two arguments, a subject and an object.

object

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A grammatical role. Objects are typically complements to V.

conjunction

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
See coordination.

feature

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
Syntactically or semantically relevant information that is associated with a head.

R-expression

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A referring expressions. R-expressions strictly refer, meaning that their reference cannot change in context.

reflexive

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A referring expression, also called an anaphor. A reflexives reference is not strict, meaning that it can change depending on context, but it is also sensitive to syntactic structure. Reflexives are subject to Condition A of binding theory.

pronoun

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A referring expression. A pronoun's reference is not strict, meaning that it can change depending on context. Note that pronouns are not nouns, they are syntactically determiners.

c-command

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A c-commands B iff A's mother dominates B.

dominate

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A dominates B if A is above B in a tree. A immediately dominates B if A dominates B and there are no other projections that dominate B which do not dominate A.

coreference

juliedoner — Wed, 15 Nov 2023 05:32:45 +0000
A and B corefer if they refer to the same individual(s).

Condition C

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
The condition of binding theory that governs R-expressions. "R-expressions cannot be bound."

binding domain

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
A binding domain is a minimal distance that some element makes reference to in its distribution. The binding domain for this class is the nearest dominating TP.

pro-drop

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
A typological phenomenon characterized by the ability to have apparently silent pronouns. Languages like Spanish, Italian, and Swahili are pro-drop languages.

S-structure

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Surface structure. This is the structure that built after movement and selection have been satisfied.

phrasal movement

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Movement of phrases. There are two varieties of phrasal movement: A-movement and A'-movement. Phrasal movement contrasts with head-movement.

UTAH

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
The Uniformity of Theta Assignment Hypothesis. This hypothesis holds that each thematic role has a dedicated position in the syntax.

passivization

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
A syntactic process in which an object is promoted the subject position, and a subject is demoted to an adjunct.

VPISH

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
The VP-Internal Subject Hypothesis, the hypothesis that all subjects start inside the verb phrase. This idea has been superceded by VoiceP.

expletive

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Expletive arguments are semantically vacuous, appearing only to satisfy syntactic constraints like c-selection and/or the Subject Condition.

quantifier float

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
A process by which quantifiers appear in places that have been moved away from.

idioms

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Idioms or "idiom chunks" are non-compositional phrases: the meanings of the constituents do not transparently reflect the meaning of the whole.

raising

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Raising (or "raising to subject") involves movement out of one clause into the subject position of a higher clause. See also "raising verbs".

control

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Control is the formal relationship between PRO ("big pro") and its antecedent.

elided

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
An element is considered elided if it is not pronounced, but does not display a movement relationship between the silent portion and the antecedent.

cyclic movement

juliedoner — Wed, 15 Nov 2023 05:32:46 +0000
Movement is cyclic if it occupies the same position in different clauses more than once.

selection

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Every head selects the type, number, and semantic role of what it combines with. There is semantic selection (s-selection) and category selection (c-selection).

Subject Condition

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Spec-TP must be filled at S-structure. The Subject Condition guarantees that there is always a "subject" of every clause.

bound

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
A is bound by B iff A and B are coreferential, and B c-commands A.

binding

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
A syntactic relation. A binds B iff A and B are coreferential, and A c-commands B.

plural

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
A number category, indicating non-singular.

morpheme

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
A morpheme is the smallest unit of sound-meaning correspondence. All words are made up of one or more morphemes.

modifies

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
A modifies B if A restricts the meaning of B. In this class, "modify" means the same as "adjoin."

coordination

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
The coordination constituency test uses conjunctions (and, but, not,...) to test for constituency.

double-object constructions

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
The double-object construction is phrase that has two "objects," like John gave Mary a book.

functional

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Functional words typically provide relationships between lexical words, or specify properties of lexical words. Their meanings are often difficult to define, and are dependent on the presence of a lexical word.

lexical

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Lexical words have "content," and are typically easily definable, referring to things, events, actions, or abstract properties. They contrast with functional words.

evidential

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Evidential markers indicate how the speaker learned about the information they are conveying. "Visual" evidentials indicate that the speaker saw the event."Hearsay" indicate that the speaker is reported secondhand information.

subject-auxiliary inversion

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
When an auxiliary verb appears in front of the subject. Formally, this is analyzed as T-to-C movement.

Headedness Principle

juliedoner — Wed, 15 Nov 2023 05:32:47 +0000
Every phrase has exactly one head and every head is in its phrase.

pronouns

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
A category of words. Note that pronouns are not nouns, they are determiners.

complement

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
X is a complement if its mother is a bar-level (Y') and its sister is a head (Y).

specifier

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
X is a specifier if its mother is a phrase level (YP) and it's sister a bar-level (Y').

terminal node

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
A terminal node is a node that ends in a lexical item, i.e., a word or a morpheme. Phrases and bar-levels are not terminal.

daughter

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
X is the daughter of Y if X is directly under Y (is immediately dominated by Y). This relationship can also be stated as "Y is the mother of X."

sister

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
X and Y are sisters if X and Y share the same mother---that is, if they're both directly under the same thing (immediately dominated by the same phrase).

tense-lowering

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
An instance of head-movement. Tense lowering moves T to the next lowest c-commanded head.

animate

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
An animate noun is something is alive, typically referring to something that has a conscious. Humans are animate, trees are not.

head-movement

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
Movement of heads. Head-movement is constrained by the Head-Movement Constraint. Haed-movement is distinct from Phrasal Movement.

free

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
A DP is free if it is not bound, i.e., c-commanded by a coreferential expression.

wh-movement

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
Movement of a wh-word: who, what, how, why, where, when, etc. We often use "wh-movement" as a shorthand for "A'-movement." However, while all wh-movement is A'-movement, not all A'-movement is wh-movement.

question C

juliedoner — Wed, 15 Nov 2023 05:32:48 +0000
A version of the complementizer used in questions.

polar questions

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
Yes/No questions, or questions that can be answered either "yes" or "no."

ellipsis

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
An instance of an unpronounced constituent. For example, in "Tegan called Joanna, and Sam did, too," there is an unpronounced VP [ called Joana ] in the second half of the sentence.

suffix

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
A morpheme that attaches to the end of a word.

number

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
A morphological category referring to either singular or plural (and in some languages including dual and trial as well).

copula

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
A "be" verb.

quantifiers

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
Quantifiers are sub-category of Determiners and include things like every, some, each, all, few, most, etc. Quantifiers are used to express a proportion of the associated NP.

prefix

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
A prefix is a morpheme that attaches at the beginning of a word.

DP-hypothesis

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
The DP-hypothesis proposes that all NPs are encased in a DP "shell."

SVO

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
SVO languages have the word order Subject-Verb-Object in positive declaratives.

progressive

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
The progressive is an aspectual category that indicates an incomplete or ongoing action with respect to a particular event at a time. In English, progressive aspect is expressed using a finite form of the verb BE and the gerundive form of the verb, e.g., "John is eating sushi."

perfect

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
The perfect aspect is used to indicate an event which is completed by the indicated tense. Thus, in the "present perfect," the event of the verb has been completed by the present moment. In the "past perfect," the event of the verb has been completed at some past moment. In English, perfect is expressed using a finite form of the auxiliary verb HAVE and the participial form of the main verb, e.g., "John has eaten sushi."

head-movement constraint

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
Informally, the head-movement constraint says that heads cannot skip over intervening heads. Formally, the head-movement constraint says that a head can only move to the nearest c-commanding or c-commanded head.

oblique

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
An oblique argument is "non-core" argument. Non-core argument are typically adjuncts, and very often appear in prepositional phrases (in English) or in semantic cases (like "instrumental") in other languages.

topicalization

juliedoner — Wed, 15 Nov 2023 05:32:49 +0000
A constituency test where some element is preposed to the front of the sentence, usually with comma intonation, e.g., "Mary, I saw."

D-structure

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
D-structure (or Deep-structure) is the level of representation before movement. It the level at which selection is satisfied must be satisfied.

Condition A

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
The condition of Binding Theory that stipulates that "An anaphor must be bound in its binding domain."

external argument

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
An external argument is any argument of the verb that is introduced outside of the VP---it is "external" to VP.

internal argument

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
An internal argument is any argument of that is introduced inside of the VP---it is "internal" to VP.

case

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
A morphology property of nouns in many languages. Case can either be structural, and indicate the grammatical role (subject/object) or semantic, and indicate the thematic role of the noun.

adpositional

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
An "adposition" is a functional particle or word that determines something about a following noun. Often, adpositions give information about location: on, over, under, in. English only has prepositions, because the adposition comes in front ("pre-") of the noun Many languages have "postpositions", where the adposition comes after the noun ("post-").

generative

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
A generative syntactic theory is one which proposes a set of abstract rules can can "generate" any and every human language.

argument

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
An argument is something that is c-selected. Arguments include grammatical roles like "subject, object, indirect object, etc."

suppletive

juliedoner — Wed, 15 Nov 2023 05:32:50 +0000
ADD DEFINITION

Verb particle

juliedoner — Wed, 15 Nov 2023 05:32:51 +0000
A verbal particle is a prepositional element that co-occurs with a verbal element, often affecting the meaning of the verb in subtle ways.

non-concatenative morphology

juliedoner — Sat, 11 May 2024 16:24:29 +0000
A morphological process that is not linear.

Authors

juliedoner — Wed, 11 Oct 2023 18:02:56 +0000

Cover

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

Table of Contents

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

About

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

Buy

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

Access Denied

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

Book Information

juliedoner — Wed, 11 Oct 2023 18:02:57 +0000

H5P listing

pressbooks — Wed, 11 Oct 2023 18:03:09 +0000

Part I. Foundations

juliedoner — Fri, 20 Oct 2023 02:35:54 +0000
In Chapters 1-6, we will learn how to:

Define the fields of morphology and syntax.

Apply the scientific method to the study of morphology and syntax.

Describe the main theoretical assumptions of generative grammar.

Describe and identify various kinds of morphemes and morphological processes..

Collect, share, present, and format morphosyntactic data.

Classify words according to part of speech.

Develop writing, argumentation, and study skills, especially as they pertain to linguistics.

1. What are morphology and syntax?

juliedoner — Fri, 20 Oct 2023 02:36:21 +0000
This textbook is an introduction to the study of the sub-fields of [pb_glossary id="160"]morphology[/pb_glossary] and [pb_glossary id="215"]syntax[/pb_glossary]. In this chapter, we begin by defining both these words and some basic concepts in both subfields. We will discuss some of the applications of morphology and syntax, and show how they interact with the larger field of [pb_glossary id="216"]linguistics[/pb_glossary]. We will also introduce some study habits that will help you succeed in the study of linguistics, as well as give an overview of some common misconceptions about academic integrity.

Learning objectives

By the end of this chapter, you should be able to:

Define the fields of morphology, syntax, and linguistics.

Describe some applications of the study of morphology and syntax and why they are relevant.

Define morphemes and distinguish between simplex and complex words.

Apply some study strategies to common tasks in linguistics courses to improve your learning, problem solving, and writing.

Understand what counts as an academic integrity violation.

2. The scientific method in morphosyntax

juliedoner — Fri, 20 Oct 2023 02:41:47 +0000
In this chapter, we will consider how we use the [pb_glossary id="184"]scientific method[/pb_glossary] in morphology and syntax. We will look more closely at how the scientific method applies in morphology and syntax and which methods we should use. We will also consider the genre of scientific writing in morphosyntax, how to identify academic sources, and how to cite in linguistics.

Learning objectives

By the end of this chapter, you should be able to:

Apply the [pb_glossary id="184"]scientific method[/pb_glossary] to simple morphosyntactic questions, including formulating a [pb_glossary id="495"]falsifiable[/pb_glossary] [pb_glossary id="496"]hypothesis[/pb_glossary].

Describe the difference between [pb_glossary id="497"]prescriptive[/pb_glossary] and [pb_glossary id="498"]descriptive[/pb_glossary] approaches to grammar and explain why we use the descriptive approach in linguistics.

Identify which information about a word would be encoded in the mental lexicon.

Identify academic sources.

Apply the conventions of scientific writing within the field of morphosyntax.

Write in-text citations and bibliography entries according to convention in linguistics where appropriate.

4. Linguistic data in morphosyntax

markc — Tue, 07 Nov 2023 19:38:38 +0000
In this chapter, we look at what types of data we consider in morphology and syntax. We will learn about how to collect it, and the strengths and weaknesses of the different kinds of data collection methods. We will also learn about how to format and present data according to linguistic convention.

Learning objectives

By the end of this chapter, you should be able to:

Describe different modalities and varieties of language using technical terms.

Describe the strengths and weaknesses of corpus studies, elicitation, and experimentation.

Discuss ethical issues in data collection and use.

Present linguistic data in a reader-friendly format, following convention.

8. Parts of speech

markc — Tue, 07 Nov 2023 19:39:50 +0000
[Placeholder.]

9. Words

markc — Tue, 07 Nov 2023 19:40:12 +0000
Words are important units in both the fields of morphology and syntax. In this chapter, we will learn all about words, including how to form new words, how to define words, and how different languages package the content of their words in different ways.

Learning objectives

By the end of this chapter, you should be able to:

Classify words by their formation process, using an etymological dictionary when necessary.

Use technical terms such as lexeme, word form, word token, citation form, word family, and paradigm appropriately in context.

Classify languages based on their degree of synthesis, fusion, and exponence.

Explain the problems in defining word.

Part II. Morphology

markc — Tue, 07 Nov 2023 19:45:40 +0000
Chapters 9-15 focus on morphology.

6. Kinds of morphemes and morphological processes

markc — Tue, 07 Nov 2023 19:47:05 +0000
[Placeholder.]

8. Segmenting words into morphemes

markc — Tue, 07 Nov 2023 19:47:30 +0000
[Placeholder.]

9. Inflection vs. derivation

markc — Tue, 07 Nov 2023 19:48:07 +0000
[Placeholder.]

10. Compounding

markc — Tue, 07 Nov 2023 19:48:33 +0000
[Placeholder.]

Pressbooks Results

Mon, 24 Feb 2025 15:52:47 +0000

11. Hierarchical structure in words

markc — Tue, 07 Nov 2023 19:48:59 +0000
[Placeholder.]

12. Morphophonology

markc — Tue, 07 Nov 2023 19:49:24 +0000
[Placeholder.]

13. Productivity and word formation

markc — Tue, 07 Nov 2023 19:49:53 +0000
[Placeholder.]

14. The lexicon

markc — Tue, 07 Nov 2023 19:50:19 +0000
[Placeholder.]

15. Clitics

markc — Tue, 07 Nov 2023 19:50:41 +0000
[Placeholder.]

Part III. Syntax

markc — Tue, 07 Nov 2023 23:02:27 +0000
Chapters 16-24 focus on syntax.

16. Constituency and hierarchy

markc — Tue, 07 Nov 2023 23:02:50 +0000
[Placeholder.]

17. Phrase structure rule grammars

markc — Tue, 07 Nov 2023 23:03:15 +0000
[Placeholder.]

18. Syntax trees

markc — Tue, 07 Nov 2023 23:03:32 +0000
[Placeholder.]

19. Binding theory

markc — Tue, 07 Nov 2023 23:03:55 +0000
[Placeholder.]

20. Predicates and arguments

markc — Tue, 07 Nov 2023 23:04:16 +0000
[Placeholder.]

21. Types of clauses

markc — Tue, 07 Nov 2023 23:04:36 +0000
[Placeholder.]

22. Intro to X-bar

markc — Tue, 07 Nov 2023 23:07:08 +0000
[Placeholder.]

23. Intro to movement

markc — Tue, 07 Nov 2023 23:07:22 +0000
[Placeholder.]

24. Different ways of doing syntax

markc — Tue, 07 Nov 2023 23:07:42 +0000
[Placeholder.]

Part IV. Morphology-syntax interface

markc — Tue, 07 Nov 2023 23:08:45 +0000
[Placeholder.]

25. Y model of grammar

markc — Tue, 07 Nov 2023 23:09:12 +0000
[Placeholder.]

26. Morphosyntactic features, selection, and agreement

markc — Tue, 07 Nov 2023 23:09:32 +0000
[Placeholder.]

27. Language universals and diversity

markc — Tue, 07 Nov 2023 23:09:53 +0000
[Placeholder.]

28. Distributed morphology

markc — Tue, 07 Nov 2023 23:10:15 +0000
[Placeholder.]

Main Body

juliedoner — Wed, 11 Oct 2023 18:02:56 +0000

What are morphology and syntax?

juliedoner — Wed, 11 Oct 2023 18:20:00 +0000

Part 1: Foundations

juliedoner — Wed, 11 Oct 2023 18:24:14 +0000

Chapter 1: What are morphology and syntax?

juliedoner — Wed, 11 Oct 2023 18:25:56 +0000

Part 2: Morphology

juliedoner — Wed, 11 Oct 2023 18:27:26 +0000

Kinds of morphemes and morphological processes

juliedoner — Wed, 11 Oct 2023 18:28:39 +0000

What are morphology and syntax?

juliedoner — Fri, 20 Oct 2023 02:14:52 +0000

Test

juliedoner — Fri, 20 Oct 2023 02:16:45 +0000

3. Theories of grammar and language acquisition

juliedoner — Tue, 31 Oct 2023 17:28:40 +0000
[Placeholder.]

Making a PSR grammar for a language you don't speak

juliedoner — Wed, 08 Nov 2023 17:32:26 +0000
One thing that linguists often do is look at languages that they don't know much about and look for the word order patterns in that language. They can then describe the patterns with phrase structure rules. The first step in figuring out the phrase structure rules for another language is to determine the constituency in your data. You can use the meaning in the English translation to determine what modifies what in the other language. Let's look at some examples together. The following examples are from Turkish. Remember that in examples from other languages, the first line is written naturally in the other language. The second line tells you what each word in the other language means. The third line tells you what the sentence means in English.
The abbreviations used in these examples are listed in the table below.

Abbreviation Meaning

1 First person (I/me)

3 Third person (it/she/he)

OBJ Object case marker

PASS Passive voice marker

POSS Possessive marker

PST Past tense

SG Singular number

Table A1.1: Turkish glossing abbreviations

Step 1: Identify the order of subject, object, and verb
The first constituents you should look for are the subject and the verb phrase. The subject will typically be the NP that performs the action, while the VP will include the verb and the object NP. The object NP is the NP at which the action is directed. The VP always includes the verb and the object (if there is an object). It never includes the subject. In sentence (1), the word order is roughly Toprak Deniz see, as we can see from the second line. But how do we know whether Toprak saw Deniz or Deniz saw Toprak? Well, we look at the English translation in the third line. The sentence means that Toprak is the one doing the seeing, not Deniz, so Toprak is the subject. Likewise, Deniz is the one being seen, so Deniz is the object.

(1) Toprak Deniz’-i gör-dü.

Toprak Deniz-OBJ see–PST

'Toprak saw Deniz.'

From this, we can hypothesize that Turkish is an SOV language, which means that sentences in Turkish follow a subject - object - verb word order by default. Turkish also has object case markers which can help us find the object (see Section 5.7 for a refresher on what case is). Not all languages use case markers, and sometimes case markers are used in unexpected ways, so we cannot always depend on case markers to help us determine which NP is the subject or the object. However, it can be a useful piece of secondary evidence! You can draw boxes around the constituents you identify so you don't get confused! In example (1) we might draw a box around Deniz’i gördü 'saw Deniz' to remind ourselves that this is a constituent, the VP

How to be a linguist: Working Hypotheses

This approach works on the assumption that the English translation in the third line has an equivalent structure as the original language, which is not always the case. Linguists will try to translate the sentence as close as possible, but sometimes the language will have a structure that does not have an equivalent in English. For example, here is what a passive sentence looks like in Turkish, although, unlike in English, it is unusual (but still possible!) to include the subject in a passive like we did here. (For a refresher on what a passive is, see Section 6.11).

(2) Deniz Toprak taraf-ın-dan gör-ül-dü.

Deniz Toprak side-3SG-POSS see-PASS-PST

'Deniz was seen by Toprak.' Or literally, 'Deniz was seen from Toprak's side.'

Because it’s passive, the one doing the seeing is not in subject position, and the one being seen is not in object position. It is important to keep in mind the difference between the structural subject and the thematic subject. The structural subject is the NP that appears in the subject position of the clause (the daughter of S). The thematic subject is the NP with the agent theta role (see Section 6.10). In most sentences, both of these pick out the same NP, the subject. However, sometimes they aren’t the same, such as in passive sentences. When you’re working on an unfamiliar language, you won’t always know if there’s something happening that is adding extra complication to your data, like if it’s a passive sentence. But we have to start somewhere! So what we do is we make a hypothesis based on the data we have. We try to make it the simplest hypothesis possible. After that, we collect more data to check our first hypothesis. If the passive sentence was the only sentence of Turkish we had, we might conclude that Turkish has OSV word order instead of SOV (depending on how seriously we took the structure of the English translation). However, once we collected more data, we’d probably notice that the passive sentence has a different word order than the others. We might also notice that the passive sentence has morphological differences. For example, it has the passive marker on the verb but it doesn’t have the object case marker on the object. We would use these clues to help us revise our initial hypothesis and conclude that sentence (2) has a change in argument structure and that Turkish really is SOV.

Step 2: Adding the modifiers to the constituents
After we identify the order between subject, object, and verb, we can start to fill out some of the other constituents, by identifying what the modifiers are modifying. For example, in sentence (3), we have the adjective siyah ‘black’ occurring between two nouns. Some languages (like English) put adjectives on the left of the noun it modifies, while others (such as French) put adjectives on the right. How do we know, then, whether Turkish is like English, and this sentence is about a black table, or whether Turkish is like French, and this sentence is about a black book?

(3) Kitap siyah masa-ya dokun-du.

book black table-OBJ touch-PST

'The book touched the black table.'

Again, we can tell by looking at the English translation in the third line. In the translation, black modifies table, so we can know that siyah modifies masaya, and therefore that siyah masaya forms a constituent. You may want to put a box around this constituent too, so we remember what we figured out.
Turkish has a phonological assimilation rule called vowel harmony which can change the pronunciation of vowels to match the backness feature of other vowels in the word. This is why the past tense marker is sometimes -dü and sometimes -du. This difference doesn’t affect the syntax, so you don’t need to worry about that difference here.

Step 3: Looking for patterns
After you've identified the constituents, you need to go through each constituent one by one and identify what goes into that constituent. At this step, the sort of questions you should ask are:

What is optional in this kind of constituent?

What is obligatory in this kind of constituent?

Which elements can be repeated, so that they need a plus sign?

What is the relative order of the elements inside a constituent?

You can identify that something is optional in a given constituent by noticing that it isn’t there each time that kind of constituent appears in your data. For example, the S rule will show the order between the subject NP and the VP. However, in some languages, including Turkish, the subject is optional in some contexts. This is shown in sentences (4) and (5). In sentence (4), there is a subject pronoun ben ‘I’. But in sentence (5), which has the exact same meaning, there is no subject pronoun! You can still tell who the subject is, but only because of the agreement marker on the verb.

(4) Ben Deniz’i gör-dü-m

I Deniz-OBJ see-PST-1SG

'I saw Deniz.'

(5) Deniz’i gör-dü-m

Deniz-OBJ see-PST-1SG

'I saw Deniz.'

Because of this, when we write the S rule for Turkish, we will put the subject NP in brackets to show that it is optional.

(6) S → (NP) VP

When we’re working with a small data set, like you will be doing on most of your homework in this class, you won’t usually be able to tell for sure whether something is obligatory. However, if it appears in all of the relevant places in your data set, you should assume it is obligatory until you find evidence to the contrary. So, for example, all of the sentences we’ve looked at in this section are transitive, and have an object. So we would not put the object NP in parentheses. (Turkish does have intransitive verbs, though, so if our data set was a bit bigger, this would be different.) Remember, we are writing our PSRs as a model that describes the data we have, not the data we expect to exist. Once we gather that further data, we can revise our hypotheses. When you are listing the members of a constituent in your PSRs, be careful that you list the constituents of just the next layer of structure. For example, let’s look at sentence (3) again.

(7) Kitap siyah masa-ya dokun-du.

book black table-OBJ touch-PST

'The book touched the black table.'

In this sentence, the VP is siyah masaya dokundu ‘touched the black table.’ Many students will look at this sentence and conclude that the VP rule for Turkish is an AdjP (for siyah), followed by an NP (for masaya), followed by a V (for dokundu), as shown in (8). But this isn’t quite right!

(8) Incorrect rules

VP → (AdjP) NP V

NP → (AdjP) N

Earlier we decided that siyah masaya ‘black table’ was a constituent. The adjective doesn’t belong directly in the VP; instead it belongs inside of the NP meaning black table, and the whole NP is inside the VP, as shown in the tree in Figure A1.4. Figure A1.4: Tree of example (7)The incorrect rules in (8) put the same AdjP in two different places—both inside the VP and inside the NP. Instead, it should only belong to the NP, as in (9). Siyah does not modify the verb. It only modifies the noun, so it belongs to the NP.

(9) Correct rules

VP → NP V

NP → (AdjP) N

Step 4: Putting it all together
In your last step, you should collect all of your PSRs in one list and then double check them. Here is the list of PSRs for the Turkish data from this section.

(10) Turkish PSRs

S → (NP) VP

VP → NP V

NP → (AdjP) N

AdjP → Adj

The first thing you should do to double check your answers is to compare all of the data in your data set to the final version of your rules. Sometimes when you revise a rule, you accidentally make it inconsistent with some data you looked at earlier in your process. I also recommend you draw a tree of one or more sentences in your data using your PSRs, to check to see if following your PSRs strictly makes the word order come out right with no pieces missing. If you only draw one sentence, choose the most complicated one. That is the one that is most likely to have a mistake! If you have to draw something that isn’t listed in your rules, then either you’ve drawn it incorrectly or there’s a mistake in your rules. Special thanks to Çağrı Bilgin for coming up with the Turkish data in this section.

Draw a syntax tree step-by-step

juliedoner — Wed, 08 Nov 2023 17:40:47 +0000
Here are some step-by-step instructions on how to analyze sentences and then draw a tree of them. Notice that before you start drawing, you should analyze the structure of the sentence! You are much more likely to draw it correctly that way.
Step One: Identify the main verb.
Don’t get distracted by auxiliary verbs! In the sentences in (1), the main verbs are bolded. Notice that in (1a), have is a main verb, while in (1d), (1e), and (1f), it is an auxiliary verb.

(1) a. I have a green spaceship.

b. I think that aliens exist.

c. That aliens exist surprised my sister.

d. You have seen an alien with a telescope.

e. Have you seen an alien?

f. What have you seen?

Step two. Find the subject of each verb.
The subject will usually be right before the verb in English (or the auxiliary verb), it will trigger subject-verb agreement, and it is often the one doing the action (but not always). You can stick a question word (who or what) before the verb, and the answer to the question will be the subject. For example, who has a green spaceship? I do. Or What surprised my sister? That aliens exist.

(2) a. I have a green spaceship. The subject of have is I.

b. I think that aliens exist. The subject of think is I, and the subject of exist is aliens.

c. That aliens exist surprised my sister. The subject of exist is aliens, and the subject of surprised is that aliens exist.

d. You have seen an alien with a telescope. The subject of (have) seen is you.

e. Have you seen an alien? The subject of (have) seen is you.

f. What have you seen? The subject of (have) seen is you.

Step three. Identify the clauses of the sentence.
There will be one clause for each main verb (but not auxiliary!). I have put brackets around each clause in (3). Notice that main clauses include the embedded clauses. Embedded clauses can be the object of the verb, as in (3b), or the subject of the verb, as in (3c).

(3) a. [I have a green spaceship].

b. [I think [[that aliens exist].

c. [[That aliens exist] surprised my sister].

d. [You have seen an alien with a telescope].

e. [Have you seen an alien]?

f. [What have you seen]?

Step four. Classify each clause.
For each clause, identify whether it is question or a statement. If it is question, identify whether it is a yes-no question or a content question. If it is a content question, figure out where the question word started before movement. You can figure out where the question word started out by saying the echo question: You have seen WHAT? or by answering the question and looking at where the answer goes: What have you seen? I have seen an alien.

(4) a. [I have a green spaceship]. This is a statement.

b. [I think [[that aliens exist]. The main clause is a statement and the embedded object clause is a statement.

c. [[That aliens exist] surprised my sister]. The main clause is a statement and the embedded subject clause is a statement

d. [You have seen an alien with a telescope]. This is a statement.

e. [Have you seen an alien]? This is a yes-no question.

f. [What have you seen]? This is a content question. The wh-phrase started in object position.

Step five. Identify the modifiers.
Identify the modifiers, such as adjectives and prepositional phrases. Figure out what they modify. Watch for ambiguity! Maybe it is possible that they can attach to more than one thing. In our set of six sentences, there are two modifiers, green in the (a) sentence and with a telescope in the (d) sentence. Green modifies spaceship. But with a telescope is ambiguous! It can modify an alien or seen. You can use constituency tests to see the two different meanings. For example, in (5), the two different meanings are illustrated with the movement test.

(5) a. It is [an alien with a telescope] that you have seen.

b. It is [an alien] that you have seen [with a telescope].

In (5a), an alien with a telescope is a constituent. This means that with a telescope modifies alien and is sister to alien (according to the Principle of Modification). This constituency structure derives the meaning that the alien has the telescope. In (5b), an alien is a constituent that does not include with a telescope. This means that with a telescope does not modify alien; instead it modifies the verb and is sister to the verb. This constituency structure derives the meanings that the telescope was used for the act of seeing.
Step six. Draw the spine of the clause.
Start at the top of the page with an S. Draw two to four branches down from the S—one for the subject, one each for the auxiliary or negation if they are there, and one for the VP. Now draw one or two branches down from the VP—one for the V head and a second one for the object if there is an object. [caption id="attachment_3751" align="alignnone" width="300"] Figure A1.22: The spine for a declarative main clause[/caption]If it is a question, also put a CP on top of the TP. [caption id="attachment_3746" align="alignnone" width="300"] Figure A1.23: The spine for a main clause question[/caption]If there is an embedded clause in your sentence, check again if it is a subject clause or an object clause. If it is a subject clause, put it under the far left branch under the S, instead of a subject NP. If it is an object clause, put it under the right branch under VP, instead of an object. Draw the spine of the embedded clause, starting at CP, with the C head on the left as either that or null. Under the CP, draw an S that contains an NP and a VP, just like you did with the main clause. [caption id="attachment_3753" align="alignnone" width="300"] Figure A1.24: The spine for a sentence with an embedded subject clause[/caption] [caption id="attachment_3752" align="alignnone" width="300"] Figure A1.25: The spine for a sentence with an embedded object clause[/caption]
Step seven. Fill in the subjects, objects, and modifiers.
Draw the subject and object NPs in the subject and object positions. If there are no modifiers, an NP will usually be two branches, one for the determiner and one for the head noun. If there are modifiers, add as many extra branches as you need. [caption id="attachment_3738" align="alignnone" width="300"] Figure A1.26: Tree diagram of "I have a green spaceship"[/caption] [caption id="attachment_3743" align="alignnone" width="257"] Figure A1.27: Tree diagram of "I think that aliens exist"[/caption][caption id="attachment_3754" align="alignnone" width="300"] Figure A1.28: Tree diagram of "That aliens exist surprised my sister."[/caption]If there are any modifiers for the verb (such as a PP or an adverb), add branches to the VP as needed. In English, adverbs can go to the left or right of the verb head, but PPs always go to the right. Remember again to watch out for ambiguity! Watch out especially for constituents at the end of the sentence. They might attach to the VP or they might attach to the object. If there are two clauses, they might also attach to either the main clause or the embedded clause! Use your constituency tests to figure it out or use the meaning to determine which part of the sentence it modifies. [caption id="attachment_3766" align="alignnone" width="398"] Figure A1.29: Tree diagram of "You have seen an alien with a telescope" where "with a telescope" modifies the verb. This structure means that you used the telescope to see the alien.[/caption][caption id="attachment_3765" align="alignnone" width="417"] Figure A1.30: Tree diagram of "You have seen an alien with a telescope" where "with a telescope" modifies "alien." This structure means that the alien has a telescope.[/caption]If one of the NPs or PPs is particularly complex, like in (6), break it down into smaller pieces, and do one piece at a time.

(6) The outlandish lady with the purple purse from Wal-Mart on her elegant arm

First, identify the head. Is this NP as a whole identifying a lady, a purse or an arm? It is identifying a lady, so lady is the head of the phrase. Which adjectives and PPs modify lady? The adjective outlandish and the PP with the purple purse from Wal-Mart on her elegant arm both modify lady. So we can draw an NP with four branches: one for the determiner the, one for the adjective outlandish, one for the head N lady, and one for the very long PP. Next you can repeat the steps with the PP. The PP with the purple purse from Wal-Mart on her elegant arm begins with the P with and contains an NP headed by purse. So under the PP, you can draw two lines, one for the P with and one for the long NP. Next, ask what modifies purse? Keep repeating these steps until you have identified all of the constituents and what each one modifies. [caption id="attachment_3695" align="alignnone" width="562"] Figure A1.31: The tree diagram for example (6).[/caption]
Step eight. If it’s a question, add the movement.
If there’s subject-auxiliary inversion, put angle brackets around the auxiliary under S, and re-write the auxiliary under the C head. Then draw an arrow from the Aux under S to the C head. [caption id="attachment_3737" align="alignnone" width="300"] [VP seen [NP an alien]]]]" width="300" height="274" /> Figure A1.32: Tree diagram of "Have you seen an alien?"[/caption]If there’s wh-movement, put the wh-phrase where it started out in the sentence, as you determined in Step 4, with angle brackets around it. Then add an extra branch to the left under the CP, rewrite the label of the wh-phrase, and draw the internal structure of the wh-phrase underneath. Then draw an arrow from where the wh-phrase started to where it ended up. [caption id="attachment_3763" align="alignnone" width="300"] [VP seen ]]]" width="300" height="287" /> Figure A1.33: Tree diagram of "What have you seen?"[/caption]
Step nine. Double check your answer!
Here are some things you should check:

Does every phrase have a head? Does every head (except Det) have a phrase?

If you read the ends of the branches from left to right, do all the words come out in the right order?

Do any branches cross? (They shouldn’t!)

Are your subject clauses in subject position and your object clauses in object position?

Are your subjects under TP, your objects under VP, and your modifiers sister to what they modify?

Are all your words labeled for part of speech?

Is movement clearly indicated with an arrow? Is it clear from your diagram where the movement started and where it ended?

Sometimes you will see trees with triangles in them. For example, there is a triangle in Figure A1.19 of Section A1.6 or in Figure A1.34 below. [caption id="attachment_3767" align="alignnone" width="340"] Figure A1.34 Example of a tree with a triangle[/caption]Triangles are a way to abbreviate the structure. Instead of drawing all the pieces of a constituent, you can write the entire constituent underneath a triangle (instead of branches). They are usually used in papers if part of the structure is not relevant. In your homework and tests, if you are asked to draw a tree and you use a triangle, you might not get full marks for the part of the tree inside the triangle--check with the grading scheme for your course to see if triangles are permitted and how they will be graded! That being said, if you are feeling very overwhelmed, you could use a triangle and you may at least get part marks. Or, even better, you can use a triangle in your rough work to help you break up the sentence into smaller, more manageable parts. Just remember to come back and finish drawing the bits you placed in triangles.
[h5p id="128"]

Syntactic categories

juliedoner — Wed, 15 Nov 2023 05:29:33 +0000

Learning Objectives

By the end of this chapter, you should be able to,

understand how languages can be described in terms of abstract descriptive rules.

use [pb_glossary id="612"]morphology[/pb_glossary] and [pb_glossary id="613"]distribution[/pb_glossary] to determine word categories

identify characteristic properties of core parts of speech (e.g., nouns, verbs, adjectives, adverbs, adpositions, etc)

To build a theory of syntax, we need to look at data. We'll start by considering the following sentences of English. [latex] \ex. \a. The mailman slept. \b. A storm rages. \c. The student cried. \d. A dog barked. \e. The senators lie. [/latex] Here's an obvious observation about these sentences: All the sentences in (1) are different. For instance, they use different words; they mean different things; some of them are in different tenses. Ask any English speaker, and they would agree: these are different sentences. Here's a not-so-obvious observation about these sentences: All the sentences in (1) are the same. Why? Well, they all can be described in the same way. They consist of a [pb_glossary id="614"]determiner[/pb_glossary] followed by a noun followed by a verb. Stated schematically, I can say that,

Sentence_English → Determiner Noun Verb

"A sentence in English consists of a determiner, followed by a noun, followed by a verb."

What have we done? We've translated English into a system of variables. For whichever values (that is, words) that we choose for each of the variables (Determiner, Noun, Verb) we will generate a sentence of English.

This means that I, as an English learner/speaker, don't have to memorize every combination of determiner, noun, and verb in English. All I have to know is this rule, and I can generate an infinite number of sentences (...as long I know the possible determiners, nouns, and verbs in English).

The goal for every generative syntactician is the set of rules that describe a human language --- and nothing else. That is, we want to find the rules that can accurately describe what is, and is not, a sentence in a language.

In general, syntactic theories (of which there are many) differ on what the rules should look like and what the background assumptions should be (that is, what is given, and what must be stated as part of the rule). But the goal is always the same: a precise and accurate description of language. This is what we are aiming for in this class.

So far we are able to capture only a limited set of data. Obviously it is possible to say many more things in English than sentences with [pb_glossary id="617"]intransitive[/pb_glossary] verbs. Let's look at some more examples.
[latex] \setcounter{ExNo}{1} \ex. \a. The happy mailman slept. \b. A terrible storm rages. \c. The new student cried. \d. A stubborn dog barked. \e. The stupid senators lie. [/latex]
Our rule for Sentence_English doesn't allow us to generate the sentences in (2), because the rule doesn't know what to do with words like happy, terrible, new, stubborn, and stupid. So let's make another rule.

Sentence_English → Determiner Adjective Noun Verb "A sentence in English consists of a Determiner, followed by an Adjective, followed by a Noun, followed by a Verb."
By itself, this rule generates all the sentences in (2), but now it over-generates, because according to the rule, every sentence in English must consist of the sequence Determiner-Adjective-Noun-Verb. So according to this new rule, the sentences in (1) are not possible sentences in English because they don't have adjectives.
So we conclude that we need two rules for English. Now our [pb_glossary id="615"]grammar[/pb_glossary] fragment of English is the following

A grammar fragment of English

Rule 1: Sentence_English → Determiner Noun Verb.

"An English sentence consists of a determiner followed by a noun followed by a verb."

Rule 2: Sentence_English → Determiner Adjective Noun Verb.

"An English sentence consists of a determiner followed by an adjective followed by a noun followed by a verb."

This is simply a logical step. Rule 1 and Rule 2 both correctly generate the sentences in (1) and (2), and they don't generate any sentences that aren't in English, like *Mailman happy slept the. Our rules predict that this sentence shouldn't be an English sentence, because our rules do not allow this sequence of elements.

Together, Rules 1 and 2 are a grammar fragment. They just describe some of what is possible in English. To continue the process and develop a complete grammar, we'd look at more data, and determine the rules that describe the data. We would do this until we developed a set of rules that could generate all the possible sentences in English, and wouldn't generate any sentences that weren't in English. We would then have a complete grammar of English.

However, there are a couple of things we seem to be missing in this grammar fragment. Foremost, these rules don't really capture the relationships between the words. For instance, intuitively, we have the sense that the happy mailman "goes together," like it's a unit, more than mailman slept. For instance, I can replace the happy mailman with he and the sentence means the same thing: He slept. I cannot replace mailman slept with anything to get an equivalent sentence: *The happy he. (This is called substitution; we'll talk about it in the next chapter.) Thus, if we want our rules to truly represent English, then we want to represent that the happy mailman goes together to the exclusion of slept.

We can actually re-write our grammar fragment to capture this grouping. We'll use the term "Determiner Phrase" (abbreviated DP) to refer to the sequence Determiner-Adjective-Noun or Determiner-Noun.

A revised grammar fragment of English Rule 1: Sentence_English → DP Verb "A sentence in English consists of a determiner phrase followed by a verb." Rule 2: DP → Determiner Noun "A determiner phrase consists of a determiner followed by a noun." Rule 3: DP → Determiner Adjective Noun "A determiner phrase consists of a determiner followed by an adjective followed by a noun."
By adding the term Determiner Phrase, we can now correctly represent that the words the happy mailman go together. Putting a determiner phrase together with a verb makes a sentence in English. I can choose either one of the DP rules, Rule 2 or Rule 3, and put it into Rule 1 to make a sentence. Importantly, our revised grammar fragment does all the things that our first attempt did, that is, it gets all the words in the right order, and does not generate sentences of English that don't exist.
This level of abstraction raises two issues.

Issue 1: What are the chunks? How do we divide up the sentence? How do we know what goes with what? For instance, how do know that there isn't a verb phrase as well? And if so, how do we determine what gets included in the determiner phrase, and what is in the verb phrase.
This topic is addressed in the chapter titled Constituency and so we'll put it aside for now.
Issue 2: What do we call the chunks? Assuming that we can divide up every sentence into chunks, what do we call each of the chunks? Stated differently, What are the categories for the variables? For example, why did I choose to call the happy mailman a Determiner Phrase and not a Noun Phrase, or an Adjective Phrase?

The second question is asking about [pb_glossary id="616"]syntactic categories[/pb_glossary]. A noun has the syntactic category of N. A determiner has syntactic category D. A verb has the syntactic category of V, an adjective has the syntactic category of A, etc. A noun phrase  is a group of words that "acts like" a noun. A determiner phrase is a group of words that "acts like" a determiner, etc.

The names are essentially arbitrary. We're simply saying that there are classes of things that behave similarly. Table, dog, teacher, brick, honesty, ..., all behave in a similar fashion (in some respects). Lick, sit, own, jump, describe, believe, ..., all behave in a similar fashion (in some respects). When we say that something is a "noun" and has category N all we're saying is that that thing "acts like" the other noun-words. Likewise, when we say that something is a verb, we're simply saying that that thing "acts like" the other verb-words.

What does it even mean to "act like noun" or "act like a verb?" In what way are "nouns" differentiated from "verbs" or "adjectives?" In defining syntactic categories, it is tempting to rely on what you were probably taught in school: nouns are "people, places, or things" and verbs are "actions," adjectives "describe," etc. That is, we can attempt to define syntactic category using meaning.

This turns out be wrong for a number of reasons. For instance, we can all agree that idea is a noun, but it's debatable whether it's a person, place, or thing. Or what about the (uncontroversial) noun destruction. That's an action! The destruction of the city by Godzilla describes an event---it's clearly not a person, a place, or a thing.
Likewise, not all verbs are actions. Tasha owns a car contains the verbs owns.  Is owning an action? Only under a very loose definition of "action." What about the verb seems in Abigail seems tired. Is seem an action? Doubtful... Perhaps most problematic for the idea that we can use meaning to define syntactic category comes when we look at languages other than English. For instance, if you've studied a Romance language, you've learned that certain concepts which are expressed in English using adjectives are expressed in Romance using verbs and nouns. This happens to be a robust strategy found across the world's languages, occurring beyond Romance languages, like, e.g., Swahili. [latex] \setcounter{ExNo}{2} \ex. \ag. Yo tengo hambre\\ 1\textsc{sg} have.1\textsc{sg} hunger\\ \trans `I'm hungry.' (Literally: `I have hunger.') \hfill Spanish \bg. J'-ai faim\\ 1\textsc{sg}-have.\textsc{1sg} hunger\\ \trans `I'm hungry' (Literally: `I have hunger.') \hfill French \cg. nina njaa\\ have.1\textsc{sg} hunger\\ \trans `I'm hungry' (Literally: `I have hunger.') \hfill Swahili [/latex] If we (as English speakers) were to use meaning to define word category, we might be tempted to call words like hambre, faim, and njaa adjectives meaning "hungry." This would clearly be wrong, because in fact they're nouns, as the literal translations suggest. So we simply cannot use meaning to determine syntactic category; it doesn't help us to define categories within a language, and it doesn't help us define across categories across languages. So how actually do I know that idea, table, hambre, faim, and njaa are nouns if I can't use meaning? Or stated differently, how can I uniquely define a class of words in any particular language? There are two kinds of reliable evidence we can use to classify syntactic categories.

Morphological properties. What are the pieces of [pb_glossary id="612"]morphology[/pb_glossary] that distinguish a particular word class?

For instance, in English only nouns can appear with plural morpheme (typically -s). Again in English, only verbs can appear with tense morphology (like past tense -ed). Morphological properties are going to be specific to a language. The morphological properties of nouns in Swahili are different from the morphological properties of nouns in English.

Distributional properties. Where do members of a particular noun appear with respect to other categories.

For instance, a determiner in English must always be the first thing in a determiner phrase. A verb in English must always be preceded by a determiner phrase. Distributional properties, too, are going to be specific to a language. In Swahili, determiners generally follow nouns, whereas in English, they strictly precede nouns.

Morphology refers to pieces of words. In any particular language, every word class may have its own specific morphological pattern. In English, only nouns can take a plural [pb_glossary id="684"]suffix[/pb_glossary]: ideas, tables, children, etc. Adjectives cannot inflect for plural in English: *reds, *sads, etc. Only verbs in English can inflect for (be marked with) tense and aspect information. walked, walking, *childed, childing. Only adjectives in English can inflect for the comparative suffix: sadder, *childer. Importantly, morphological properties are going to be language specific. For instance, in Estonian, nouns inflect for [pb_glossary id="685"]number[/pb_glossary] (singular or plural), just like in English and also for [pb_glossary id="700"]case[/pb_glossary] (which indicates the role the noun plays in the sentence).
Partial declension of the word vend "brother" in Estonian

Singular Plural

Nominative vend venna

Genitive venna venda

Partitive venda vendi

Illative vennasse vendasse

Thus, in Estonian, if a word inflects for both number and case in the same way, then we have evidence that it's a noun. Just like in English, if a word inflects for tense and aspect, then we conclude that it's a verb. If a word in English can take the comparative morpheme -er, then we conclude that it's an adjective. The second piece of evidence we can use to determine syntactic category is distributional evidence. Every word class goes in a particular position with respect to other word classes. In English, the only thing that can appear in the underlined space is a noun: The ____ is... . That is, only nouns can appear between a determiner and [pb_glossary id="686"]copula[/pb_glossary]. Or stated differently: only certain words can appear between a determiner and copula in English, and we call those words "nouns." In French, the only thing that can appear in the underlined space in a comparative construction is an adjective. [latex] \setcounter{ExNo}{3} \exg. Marie est plus {\underline{\hspace{50pt}}} que Anne\\ Marie is more {\underline{\hspace{50pt}}} than Anne\\ \trans `Marie is {\underline{\hspace{50pt}}} than Anne \hfill French [/latex] Or stated slightly differently, the only kind of word that can appear in the configuration above is a word like grande "tall," or intelligente, "smart," and we give those words the label "adjective." (Notice that adjectives in French also have a morphological distinction: they inflect for masculine and feminine gender, in addition to singular and plural.) Again, distributional properties of syntactic categories are going to be language-specific. We can't make general statements like "adjectives always precede nouns" because there are many languages in which they come after nouns. Similarly we can't say something like "verbs always come after subjects" because there are many languages in which the verb comes after the object, or even before the subject. To distinguish syntactic categories for any particular language, you have to look at the properties of that particular language.
[embed]https://www.youtube.com/embed/6I56Q1iHzJQ[/embed] Parts of speech (length: 2m 46s)

Where we're going

Now that we can identity parts of speech (using distribution and morphology), we can use that information to write our abstract rules about language, as demonstrated earlier. That is, we can translate any language into a system of variables.

We can also determine the category of groups of words. For instance, suppose I wanted to know what category the group of words happy mailman is. From distributional evidence, we would conclude that it behaves like a noun, not like an adjective (or a verb, adverb, etc). For instance, we can't use this phrase to modify another noun (5a), and it can go in the location that only nouns can go (5b).
[latex] \setcounter{ExNo}{4} \ex. \a. *The happy mailman table \b. The {\underline{happy mailman}} is {\ldots} [/latex] We're going to use this information to help us determine how to "chunk up" the sentences we find, and in turn how to write the rules that are in our grammar fragment.

Things to remember

Language description in terms of abstract rules

Using data to create a grammar fragment

How to use morphological and distributional properties of words to determine syntactic categories.

Advanced

Changing categories
One of the complications that frequently crops up when we're considering category is that words can change category. Consider the English word cut.  If I asked you what category cut was you could say, "It's a verb." Or you could say, "It's a noun." Both answers would be correct. [latex] \setcounter{ExNo}{5} \ex. \a. Sarah cut the bread. \hfil Verbal \textit{cut} \b. It was a deep cut. \hfil Nominal \textit{cut} [/latex] One question that this raises is, for any particular word, which category (if any) is the "true" form? That is, is the word cut fundamentally a verb which can be turned into a noun? Or is it a noun which can be turned into a verb? A third option is that it's fundamentally neither a noun nor a verb! It's simply some abstract semantic information which can be used either as a noun or a verb. It's also interesting to think about the limit of possibilities For cut, there is an upper bound on the category. It can be a verb, it can be a noun, and it can also be an adjective (technically a participial form): the cut bread. That's probably it. But other words can definitely have more categories. Consider the word down: [latex] \setcounter{ExNo}{6} \ex. \a. The ball is down the hole. \hfill Preposition \b. John sat down. \hfill Verbal particle \c. John downed the soda. \hfill Verb \d. The Chiefs fumbled on the first down. \hfill Noun \e. Don't talk to Mary, she's feeling down right now. \hfill Adjective [/latex] Given the abundance of forms for down, why are some "alternations" not allowed? For instance, unlike cut, the noun chop doesn't naturally have similar noun form: Mary chopped the wood. ??It was a deep chop. Why not? Why are some words restricted in "changing" categories, and others less so?
Morphological complexity in word formation
When a word is morphologically ambiguous between different categories, we call it conversion. So the difference between cut as a verb and cut as a noun is descriptively an instance of conversion. Sometimes, in fact often, there are morphological reflexes of changing category. Consider the (famous) example of the verb destroy, which has the noun form destruction. Unlike cut, the nominalization of destroy involves at least addition of the suffix -tion, plus a change in the stem of the verb from destroy to destruc- One question we can ask about such processes is how "productive" they are. By this, we mean, "How many words exhibit a similar change?" For instance, what other verbs form nouns with -tion? If we find that it's a lot of verbs, we might say that this kind of nominalization is generally productive. If it only happens with a few idiosyncratic verbs, then we might say that it isn't fully productive. English has a few fully productive category changing processes. Consider -ness affixation. Basically any adjective can be suffixed with -ness to form a noun.
Adjectival -ness nominalizations in English

Adjective Noun

happy happiness

tough toughness

mean meanness

red redness

sad sadness

Interestingly, such morphological affixation "stacks," meaning that I can add multiple affixes to a word, each one changing the category. Consider the following productive affixation processes:[footnote]Don't be fooled by the orthography! For historical (and phonological) reasons, we sometimes use a slightly different spelling for some affixes, but it's still the same affix.[/footnote]
Deverbal derivations in English

Verb Adjective (-able) Noun (-ity)

break breakable breakability

find findable findability

drink drinkable drinkability

Other languages have a much richer process of category alternations. In Dumi, a Kirati (Tibeto-Burman) language spoken in Eastern Nepal, adjectives can be made from verbs by putting -sa at the end.
Adjective formation in Dumi

Verb Gloss Adjective Gloss

rek 'sharpen' reksa 'sharp'

sur 'wash' sursa 'washed'

tuk 'keep' tuksa 'kept'

In K'iche', a Mayan language spoken in Guatemala, verbs can be suffixed with -ib'al to forms nouns meaning "location for VERBing."
Locative deverbal nouns in K'iche'

Verb Gloss Noun Gloss

atin 'bathe' atinib'al 'bath, place for bathing'

war 'sleep' warib'al 'bedroom, dormitory'

el 'leave' elib'al '(an) exit'

Overt realizations of a category changing affixes introduce a number of interesting theoretical questions. For instance, what actually is a category? Is it just the affix that attaches last? That is, since -ness always "makes" a noun, can we just say that -ness is the category noun? This is schematized below. [latex] \setcounter{ExNo}{7} \ex. sad\textsubscript{A}+ness\textsubscript{N} [/latex] But if that's true, does that mean when we have conversion (e.g., cut_Noun versus cut_Verb), do we have to say that there's a null affix, corresponding to a noun or verb? [latex] \setcounter{ExNo}{8} \ex. \a. cut\textsubscript{V}+$\emptyset$\textsubscript{N} \hfil or, \b. cut\textsubscript{N}+$\emptyset$\textsubscript{V} [/latex] Another question concerns the stacking noted earlier. Take the word breakability. This consists of three distinct morphemes, each associated with a different category. Break is a verb; -able (here spelled -abil) attaches to verbs and makes adjectives; -ity attaches to adjectives and makes nouns. It's clear that there is an order in how the elements stack. -Able has to attach to break before -ity can. What about when there is a suffix and a prefix? Is there also an ordering? Consider the word unbreakable. This also consists of three morphemes: un-break-able. At first glance, it might not seem possible to determine the order of attachment of un- and -able. But in fact, it must be the case that -able attaches first, to form the word breakable, and then un- attaches after that, to form the word unbreakable. If it happened in the opposite order, we would first make the word *unbreak---but this isn't a word! More importantly, the idea that there is a determined order of "introduction" for each piece of a word---and that order might not be the same as the linear order---will be a central observation when we start looking not at individual words, but at groupings of words, otherwise called constituents. What is remarkable is that the same processes that underlie our understanding of syntax also underlies our understanding of morphology (for the most part). These topics are discussed more in depth in LING 527/727.

Constituency

juliedoner — Wed, 15 Nov 2023 05:29:35 +0000

Learning Objectives

By the end of this chapter, you should be able to,

identify and use seven constituency tests for determining structure

understand how to interpret constituency tests to expand grammar fragments

understand the limitations of constituency tests, and what to do with false positives and negatives

The program of [pb_glossary id="702"]generative[/pb_glossary] syntax is to find the set of rules that describes a language (really that describes every language) and that doesn't describe things that aren't in a language. In chapter one, we abstracted away from the words themselves to talk about syntactic categories, which allowed us to write descriptive rules about a language. Our English grammar fragment is the following right now.

Grammar fragment for English
Sentence_English → Determiner Phrase Verb
Determiner Phrase → Determiner Adjective Noun Determiner Phrase → Determiner Noun

We are thus able to talk abstractly about what is possible in English syntax. This way of describing language raised two issues, repeated here.

Issue 1: What are the chunks? How do we divide up the sentence? How do we know what goes with what? That is, how do I know that the sequence "Determiner Adjective Noun" forms a group?

Issue 2: What do we call the chunks? Assuming that we can divide up every sentence into chunks, what do we call each of the chunks? Stated differently, What are the categories for the variables?

In response to Issue 2, we developed ways to identify syntactic categories: morphology and distribution. Thus, we can now empirically distinguish between parts of speech in any particular language because each part of speech will have specific morphological and distributional properties. I can determine when something is a noun, a verb, an adjective, etc through not only how it looks (morphology) but also where it appears with respect to other words (distribution).
Importantly, we also noticed that group of words also have syntactic categories. When I put happy and mailman together, I get something that distributes like a noun. That is, happy mailman goes in all the same places that mailman goes, like in between determiners and copulas: the happy mailman is... . (Alternatively, the group of words happy mailman doesn't act like an adjective: it doesn't do in the places that adjectives go, like in front of a noun: *the  happy mailman carpenter.) We call a string of words that forms a group a [pb_glossary id="196"]constituent[/pb_glossary]. A constituent, by definition, has a syntactic category. This is because a constituent has a syntactic distribution. As I just stated, happy mailman has the distribution of a noun. That is, the entire string of words "acts like" a noun, not an adjective, not a verb, etc. Thus, there is a sense---an intuition at this point---that happy mailman is "noun-y." The tests that are introduced below will provide empirical evidence that this is so.

What constituency means, practically

When I claim that a string of words is a constituent, I'm claiming that all of those words are grouped together under one label. Note that a single word, is, by definition, a constituent. A sentence is also, by definition, a constituent. What I'm ultimately trying to do when I figure out constituency is I'm trying figure out how information is packaged. How does any language put its pieces together to make a meaningful utterance? Precisely, in any given language, I want to know which words go with which words. This allows me to write the simplest possible rules. For instance, with our DP rule, I can capture the idea that The mailman slept and The happy mailman slept both involve a subject and verb. Identifying The mailman and The happy mailman as the same "thing" simplifies my understanding---and representation---of the English language. I can reduce sentences to grammatical roles like Subject-Verb, or Subject-Verb-Object by identifying that the the term "subject" may involve something that consists of more than one word, like the happy mailman.

It is worth noting that in English, it is difficult if not impossible to use morphology to determine constituency, so all the tests below involve distributional evidence. But morphology is actually relevant and important in other languages in determining constituency. For instance, in Swahili, adjectives must "agree" with the nouns that they modify. This typically means that the prefix of an adjective must match the prefix on a noun. [latex] \ex. \ag. {\underline{m}}toto {\underline{m}}zuri\\ child good\\ \trans `good/beautiful child' \bg. {\underline{wa}}toto {\underline{wa}}zuri\\ children good\\ \trans `good/beautiful children' [/latex] The same point can be made in Romance languages. In Spanish, adjectives modifying masculine singular nouns appear with particular distinctive morphology -o, and adjectives modifying feminine singular nouns appear with different morphology -a. It is thus possible to identify groups of words using morphology in languages which have "richer" inflectional morphology than English. However, such morphological constituency diagnostics come with their own (considerable) complications; we will not be able to address them fully in this class.
Constituency tests

1. Substitution tests
[pb_glossary id="618"]Substitution[/pb_glossary] tests are a family of tests that can be used to determine both constituency and category label. The premise behind these tests is that if you can replace the string of words in questions with something whose category you know, then the string of words must share a category with the thing you've replaced it with. For instance, if I can replace, or substitute, the tired doctor with something whose category I know, then I can conclude that the tired doctor has the same category as whatever I replaced it with.
Applying the test: The tired doctor slept. → That's right, she slept. Step-by-step:

Identify the string of words whose constituency you want to test.

Identify an appropriate substitution word, whose category you know.

Replace the string of words (and nothing else) with the chosen word.

Check the grammaticality. If it's grammatical and it means exactly the same thing, then you can conclude it's a piece of evidence in favor of treating the string as a constituent.

[embed]https://youtu.be/U6UnaJSHXcY[/embed] Substitution test demonstration (length: 1m 29s)
There are two tricky things about substitution tests. First, you need to find an "appropriate" substitution word. What is and is not appropriate changes depending on what the string of words is. For instance, it wouldn't makes sense to chose the pronoun it to replace the tired doctor, because it refers to an inanimate object (most of the time). In order to perform a substitution test, you have to make a hypothesis about what constituent you think the string of words might be. Below is a list of [pb_glossary id="209"]pro-forms[/pb_glossary] that are used to test various constituents.

Various pro-forms

Substitution of Determiner Phrases ⇒ Use a pronoun (he, she, it, they, etc). (Make sure you use the correct form of the pronoun! Don't use he when him works better, etc.)

Example: Carol saw the trees on the hill → Bill saw them, too.

Substitution of Verb Phrases ⇒ Use do so, changed to reflect the appropriate tense if necessary.

Example: Carol saw the trees on the hill → Bill did so, as well.

Substitution of Locative Prepositional Phrases ⇒ Use there. (This won't work for all prepositional phrases, only ones which refer to a location.)

Carol sat on a bench → Bill sat there, too

Substitution of clauses ⇒ Use so. (Again, this won't work for all clauses, only certain ones.)

Example: Carol thinks that Isabelle left → Sam thinks so, too

Thus, if you think it's a DP, use an appropriate pronoun. If you think it's a verb, use do so. Note that there's nothing wrong with applying more than one test. If you're not sure which substitution test to do, just do them all! Do any of them work? If so, you've learned something about the category of the constituent! But note that not all categories permit substitution. There isn't really a substitute for adjectives in English. Likewise, in many languages, there is no substitute for verb phrases. If there is no substitute for that category, then you simply cannot apply the test. (If that comes up on an assignment, just right "N/A.") The second tricky thing about substitution tests is that you often need to set up a context. For instance, the sentence Bill did so, too doesn't make sense out of context. I can't walk into a room and say this sentence out of the blue. There needs to be an [pb_glossary id="621"]antecedent[/pb_glossary] for the pro-form (do so). The same is true of any of the pro-forms above. I can't just walk into a room and say She slept, unless you know who I'm talking about. That's why when I demonstrate substitution, I typically put the sentences into a small discourse or exchange. "The tired doctor slept. That's right! She slept." The actual test is the sentence She slept, but the preceding discourse provides an antecedent for she that makes this utterance felicitous.
2. Sentence Fragments
The second constituency test we can use is called a [pb_glossary id="620"]sentence fragment[/pb_glossary] test. When you apply this test, you're asking whether the string of words in question can stand on its own. The best way to set this up is to see whether the string of words can be the response to a question.
Applying the test: Who slept? → The tired doctor. Step-by-step:

Identify the string of words whose constituency you want to test.

Write down that string of words separate from the sentence.

Try to come up with a question which could have the string as a response.

[embed]https://youtu.be/vNFLrZXgf14[/embed] Fragment test (length: 1m 12s) credit: Emma Scott [embed]https://youtu.be/rFgoTu5R_LY[/embed] Fragment test (length: 1m 39s)
Note that the fragment test alone doesn't tell you what the category of the constituent is. In contrast, the substitution test, when done correctly, does tell you about the category of the constituent. If I successfully substitute a string of words with a pronoun, then I can conclude that the string of words has the same category as the pronoun. (They're both DPs, as we'll talk about in chapter 5). If I substitute a string of words successfully with do so, then the string of words in question is a Verb Phrase, etc. This is, essentially, marshaling the fact that the classes of words share a distribution, as discussed in the previous chapter.
If you think about the sentence fragment test, in many ways, it's really just a substitution test! If I can ask the question Who slept?, and answer The tired doctor, then in a sense who has substituted for the tired doctor.

3. Movement/Displacement tests
The third constituency test we can use is also a family of tests, called movement tests (or also displacement tests). In this class we'll cover four tests. In these tests, you're taking your original sentence, and "transforming" it by moving things around. The idea behind the movement tests is that you can only move a constituent. When first learning these tests, the easiest thing to do is to memorize a "template," and then plug the sentence into the template. The general idea is that you want to label one part of the sentence as "A"---this is the string of words you're wondering about---and the other part of the sentence as "B." Then you want to put A somewhere else, leaving B alone.
a) It-clefting.
The template for it-clefting is, "It was __A__ that __B__."
Applying the test: The tired doctor slept → It was the tired doctor that slept Step-by-step:

Identify the string of words that you want to test.

Label the string of words you want to test "A," and everything else in the sentence "B."

Fill in the template, It was ___A___ that ___B___. (You may need to change the tense of was.)

Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/7T4A1gin5Io[/embed] It-clefting demonstration (length: 1m 17s)
IMPORTANT! If you find that the test works and you get a grammatical sentence, then you can conclude that A is a constituent, but you cannot conclude that B is a constituent. For example, suppose I apply the it-clefting to the following string of words.

The doctor treated a sick patient yesterday → It was a sick patient that the doctor treated yesterday.

The test indicates that a sick patient is a constituent, but it does not indicate that the doctor treated yesterday is a constituent. You can only make conclusions about the part that you've moved, not about the part that is left behind. This goes for all of the movement/displacement tests.
b) Pseudo-clefting
The template for pseudo-clefting is, "__A__ is who __B__".
Applying the test: The tired doctor slept → The tired doctor was who slept

Identify the string of words you want to test

Label the string of words you want to test "A," and the rest of the sentence "B".

Fill in the template, A was who/what B.

Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/eYG2p0CRCvQ[/embed] Pseudo-clefting demonstration (legnth: 1m 15s)

c) All-clefting
There are number of different kinds of clefts---some are language specific. The final cleft that we'll use is an all-cleft. The template for all-clefting is, "__A___ was all (that) __B__".
Applying the test: The tired doctor slept → The tired doctor was all who slept

Identify the string of words you want to test

Label the string of words you want to test "A," and the rest of the sentence "

Fill in the template, A was all who/that B.

Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/LwGpznHQmaE[/embed] All-clefting demonstration (length: 2m 14s)

d) Topicalization
The final movement/displacement test we'll use is topicalization. This is, in some ways, the easiest movement test, because all you're doing is (potentially) rearranging the sentence. You're not adding any new words in. The template for topicalization is, "__A__ , __B___". The following are all examples of topicalization. If you're a native English speaker, you'll want to read these with a slight pause at the comma. [latex] \setcounter{ExNo}{1} \ex. \a. Greta, I like, but not Talia. \b. Olives with blue cheese, I hate. \c. \textit{Anna Karenina} was okay, but \textit{War and Peace}, I loved. [/latex] Note that it's possible to topicalize a subject, but this might not result in a new word order, just a different intonational contour.
Applying the test: Maisha met the tired doctor → The tired doctor, Maisha met.

Identify the string of words you want to test

Label the string of words you want to test "A," and the rest of the sentence "

Fill in the template, __A__, __B__.

Check the grammaticality of the resulting sentence.

[embed]https://www.youtube.com/embed/6s7wCevQz8I[/embed] Topicalization (length: 1m 50s)
Like substitution tests, you want to make sure with topicalization that you've set up the context adequately. Often the best topicalization tests involves contrastive topics. You can contrast something with something else. This is what is shown in specifically in (2c), where the topicalized phrase War and Peace is being contrasted with Anna Karenina.
Keep in mind that while I've shown you four movement/displacement tests, there are many, many more. Any way you can "rearrange" a sentence, while still keeping the meaning, is potentially a constituency test.

4. Coordination
The final constituency test we'll cover is one that works in every language, but it is also the one that is easiest to give you false positives. There are a lot of ways to do the coordination test wrong.
Applying the test: The tired doctor slept → The tired doctor and Sarah slept.

Identify the string of words you want to test

Place a coordinator directly after that string of words.

After the coordinator, add something that you know is a constituent, and that you think is of the same category as the phrase you're testing. For instance, I chose "Sarah" above because I think the tired doctor might be a DP.

Check the grammaticality of the resulting sentence.

[embed]https://youtu.be/5Uc4UDw6ilc[/embed] Coordination test (length: 2m 48s)
It's very important to keep in mind that constituency tests are not infallible. We often get false positives and false negatives. That's why we have so many tests, so that we can confirm our results! We don't ever really know why a test fails. Therefore, we cannot rely on negative results, only positive results. For instance, it-clefting and pseudo-clefting give different results when testing verbs (really, verb phrases):

Nekesa grew tomatoes → *It was grow tomatoes that Nekesa did.

Nekesa grew tomatoes → Grow tomatoes is what Nekesa did.

What do we conclude when have such conflicting evidence? Well, hopefully our other tests will help us decide whether a piece is or isn't a constituent. In the end, there may be additional factors about certain tests and/or configurations that influence the grammaticality of the test. Something about it-clefting in English doesn't work with Verb Phrases. A deeper investigation into the syntax and semantics of clefts may provide an answer, but it is, unfortunately, beyond what we can cover in this class.
Where we're going
We want to know how to write the rules that describe a language. Our rules need to be informative, and they need to accurately and precisely reflect the language in question. The constituency tests inform us about how our rules should look. If I discover that happy mailman is a constituent, then this should be reflected in our rules. More importantly, if I discover that happy mailman has the distribution of a noun, then I can conclude that this entire string of words is functionally a noun. This is a major step. By claiming that there are "noun phrases" not just "nouns" we are claiming that the category of a single word can determine the category of an entire string of words. It surely cannot be accidental that happy mailman both distributes like a noun, and also has a noun in it. Similarly, the phrase very happy distributives like an adjective and has an adjective in it.

[embed]https://www.youtube.com/embed/3zFRrtsv3NM[/embed] Constituency and category (length: 1m 12s)

Looking forward a bit, our constituency tests are ultimately telling us how our trees should look. Applying constituency tests to the sentence Mary saw a dog would indicate that a dog is a constituent, saw a dog is a constituent, and there are no other constituents (except the entire sentence, Mary saw a dog). To represent these groupings based solely on constituency tests and ignoring categories, my tree would have the following structure. Each ``junction'' where two lines meet indicates that everything under it is a constituent, or one grouping of words.[latex] \begin{tikzpicture}[mytree] \node(top) {} child {node {Mary}} child {node {} child {node {saw}} child {node {} child {node {a}} child {node {dog}} } }; \end{tikzpicture}[/latex]Thus, even without labels, I can correctly represent with this tree that a dog is a constituent, and that saw a dog is a constituent. And indeed, sometimes this kind of tree is the best we can do. Our constituency tests may indicate a grouping of words for which we do not know the category. That's fine! As we noted in the last chapter, the term "category" is really just a way of saying "this is bunch of things that act alike." The actual name of the category itself is (relatively) unimportant in the long run, as long as we can correctly characterize what distinguishes (morphologically or distributionally) a group of words or word.

Things to remember

How to apply and interpret seven constituency tests

What to do when you have mixed results in your constituency tests

How constituency tests inform our syntactic representation (trees)

Advanced
It is easy to overlook how crucial---and absolutely necessary---constituency is. It is one of the most fundamental aspects of Human Language, and it is acquired by all babies at a fairly early age. The affects of constituency can be observed across a wide range of phenomena. Consider the examples (adapted from Chomsky 2021.[footnote]Chomsky, Noam. (2021). Minimalism: Where Are We Now, and Where Can We Hope To Go. Gengo Kenkyu 160: 1-41.[/footnote]). [latex] \setcounter{ExNo}{2} \ex. \a. The robbery at the pet-stores \textbf{was} criminal. \b. The robberies at the pet-store \textbf{were} criminal. [/latex] It is constituency that ultimately determines the form of the verb. We interpret the phrase The robbery at the pet stores as an entire unit, whose head is robbery. Thus, the verb indicates a singular subject: was. In contrast, when the head is plural, robberies, the verb indicates this, too: were. What is crucial here is that the form of the verb is not determined by linear order. The verb doesn't look to its immediate left to see whether that word is singular or plural. If it did, we would expect *The robbery at the pet-stores were criminal and *The robberies at the pet-store was criminal, where the choice between was and were is determined by pet-store(s). That is, we have to "see" the entire subject as a constituent in order to determine the right agreement on the verb.
Language-specific constituency tests.
We've explore above constituency tests in English. When we look cross-linguistically, we find that the same principles apply. Can you substitute the string of words with a word whose category you know? Can you displace the string of words to a different part of the sentence? Can you coordinate the string of words with something whose category you know? What is important to keep in mind is that each of these processes may ultimately look different depending on the language. For instance, in Swahili, the replacement test looks different because pronouns work different in that language. [latex] \setcounter{ExNo}{3} \ex. \ag. Juma ali-ona mtoto mrefu\\ Juma \textsc{past}-see child tall\\ \trans `Juma saw a tall child.' \cg. Juma ali-mu-ona\\ Juma \textsc{past}-see-3\textsc{sg.obj}-see\\ \trans `Juma saw him/her.' [/latex] In Swahili, object pronouns are attached directly to verb. Mu- doesn't linearly replace the string mtoto mrefu---that is, it doesn't go in the same place. But we can still use this as a constituency test, keeping in mind the independent observation that object pronouns simply have to go in a particular position. (Indeed, we could make the same point with Romance languages, in which object pronouns similarly go in front of the verb.)
Nonconfigurationality
Still, while the notion of constituency has crosslinguistic validity, the crosslinguistic picture on constituency tests is fairly complicated. We make a broad (and far from absolute) distinction between languages which are configurational, like English, and those which are nonconfigurational. In a configurational, the elements in any one sentence, have a particular order that they need to be in. For instance, in English, as a rule, adjectives comes before nouns: red coat not *coat red. A nonconfigurational language is described as having "free" word order. Consider the famous case of Warlpiri, a Pama-Nyungan language spoken in Central Australia. [footnote]Hale, Ken. 1983. Warlpiri and the Grammar of Non-configurational Languages. Natural Language and Linguistic Theory 1(1): 5-47[/footnote] [latex] \setcounter{ExNo}{4} \ex. \ag. Ngarrka-ngku ka wawirri panti-rni\\ man-\textsc{erg} \textsc{aux} kangaroo spear-\textsc{nonpast}\\ \trans `The man is spearing the kangaroo.' \bg. Wawirri ka panti-rni ngarrka-ngku\\ kangaroo \textsc{aux} spear-\textsc{nonpast} kangaroo-\textsc{erg}\\ \trans `The man is spearing the kangaroo.' \cg. Panti-rni ka ngarrka-ngku wawirri\\ spear-\textsc{nonpast} \textsc{aux} man-\textsc{erg} kangaroo\\ \trans `The man is spearing the kangaroo.' [/latex] All three orders of words are perfectly acceptable, and all three mean exactly the same thing. (The only rule is that the auxiliary element ka has to be the second thing in the sentence.) The issue that nonconfigurational languages raise is that they appear to disregard constituency. For instance, nouns and their modifiers need not be next to each other. [latex] \setcounter{ExNo}{5} \ex. \ag. \textbf{wawirri} \textbf{yalumpu} kapi-rna panti-rni\\ kangaroo that \textsc{aux} spear-\textsc{nonpast}\\ \trans `I will spear that kangaroo.' \bg. \textbf{wawirri} kapi-rna panti-rni \textbf{yalumpu}\\ kangaroo \textsc{aux} spear-\textsc{nonpast} that\\ \trans `I will spear that kangaroo [/latex] In English, we strictly cannot separate kangaroo from that, a fact which we use to determine that kangaroo and that form a constituent. So if Warlpiri allows us to do such separation, does it have constituency? The answer is yes---but it's complicated. As we'll learn later in class, the order of words we pronounce is not always the order that the words started in. That is, things move, and therefore, we sometimes have discontinuous constituents because the parts of the constituent have been separated by movement. Of course, we need to show that movement has taken place, and we have a suite a diagnostic tests for this; we'll get to them later in the class. The takeaway here is that nonconfigurationality, in particular seemingly discontinuous constituents, ultimately are not an argument against constituency in general.

Phrase-structure grammars

juliedoner — Wed, 15 Nov 2023 05:29:36 +0000

Learning Objectives

By the end of this chapter you should,

be able to build a phrase structure grammar by looking at data,

recognize and interpret constituency tests in another language,

use a phrase structure grammar to build trees,

have a basic grasp of Swahili syntax, and

understand the principles behind a data-driven approach to syntactic theory

Building a Grammar
We now to turn to the question of building an accurate phrase structure grammar. We want our phrase structure rules to be empirically correct, and so they need to correctly describe all and only the phrases that exist in a language. To emphasize a central point of the last chapter, we want our phrase structure grammar for language X to precisely reflect how language X organizes its information. In this chapter, we'll work through how we might accomplish this with Swahili.
Swahili (or Kiswahili as it is pronounced in Swahili) is a Bantu language spoken throughout East Africa by upwards of 100 million people. It is a national language of Kenya, Tanzania, and Uganda, and the Democratic Republic of Congo, and it is also spoken in Rwanda, Burundi, Malawi, Somalia, Zambia, and Mozambique. [caption id="attachment_587" align="aligncenter" width="244"] Map of region where Swahili is spoken[/caption] You can learn to speak Swahili here at KU! Courses are offered every semester and over the summer through the African and African American Studies Department.

Building a Noun phrase rule

Let's try to figure out the phrase structure rules for Swahili nouns. To do this we simply look at (a lot of) data, applying constituency tests. We start small, and then build up. For the sake of simplicity, we will not take the time to confirm all the constituency tests for each phrase that we build. But if we were being diligent, then we would do exactly that.

The following is our first set of data.[footnote]The verb changes from ulianguka to ilianguka in the last sentence because there is a plural subject.[/footnote]
[latex] \ex. \ag. mti ulianguka\\ tree fall.\textsc{past}\\ \trans `A tree fell.' \bg. mti mkubwa ulianguka\\ tree big fall.\textsc{past}\\ \trans `A big tree fell.' \cg. Ndio, huu ulianguka\\ Yes, \textsc{dem} fall.\textsc{past}\\ \trans `Yes, that fell.' \hfill [that=A big tree] \dg. mti mkubwa na mti mdogo ilianguka\\ tree big and tree small fall.\textsc{past}\\ \trans `A big tree and a small tree fell.' [/latex] These sentences provide a wealth of information. First, it looks like in Swahili (just like in English) subjects come before the verb. But based on (1a) and (1d), it looks like adjectives go after nouns in Swahili. The third and fourth sentences are constituency tests ([pb_glossary id="618"]substitution[/pb_glossary] and [pb_glossary id="663"]coordination[/pb_glossary]), showing that the sequence noun-adjective is a constituent in Swahili. With the information from (1), we can write some rules. Let's assume that the category of the noun-adjective sequence is a Noun Phrase (NP). We'll revisit that question in the next chapter.
Swahili grammar fragment (in progress)

S_S → NP V

NP → N

NP → N A

Now our grammar fragment for Swahili allows us to draw the following trees---that is, it represents Swahili's structure in the following ways. [latex] \setcounter{ExNo}{1} \ex. \ag. mti ulianguka\\ tree fall.\textsc{past}\\ \trans `A tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} \ex. \ag. mti mkubwa ulianguka\\ tree big fall.\textsc{past}\\ \trans `A big tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} child {node {A \\ mkubwa \\ big}} } child[sibling distance = 10em] {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]

There is one other phrase structure rule that we assume to be given. It's the coordination rule, shown below.

XP → XP conj XP

Here, "XP" means any phrase---"X" is a variable over categories. The important things to note about this rule is that it only allows coordination of the same category. That is, XP doesn't consist of XP conj YP. Practically, this means that a Noun Phrase can only be coordinated with another Noun Phrase. A Verb Phrase can only be coordinated with another Verb Phrase, etc. The category "conj" is short for [pb_glossary id="633"]conjunction[/pb_glossary]; there are many different conjunctions: and, but, nor, yet, ...

[embed]https://youtu.be/raks-QFct7A[/embed] Coordination (length: 1m 22s)
We have two rules for NP. There is a convention for collapsing this into one rule, which is to use parentheses. So the two NP rules above can be re-written as the single rule below.
Swahili grammar fragment (in progress)

S_S → NP V

NP → N (A)

The second rule can be read, "A Noun Phrase consists of a noun, which is optionally followed by an adjective." To be clear, the second rule contains the information of both NP rules from our first grammar fragment, it's just a conventionalized representation. Let's look at more sentences. [latex] \setcounter{ExNo}{3} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \bg. mti mkubwa mzuri mweusi ulianguka\\ tree big good black fall.\textsc{past}\\ \trans `A big beautiful black tree fell.' [/latex] We learn here that, in fact, nouns in Swahili can be followed two or three adjectives. Indeed, we could keep going! Noun Phrases in Swahili can involve an infinitive number of adjectives (up to cognitive limitations).[footnote]In fact, there actually aren't very many words whose category is adjective in Swahili (as compared to, e.g., English). Moreover, using more than one in a sentence is fairly degraded. We're putting aside some issues here for the sake of exposition.[/footnote] Again, we have a convention to express this in a single rule.
Swahili grammar fragment (in progress)

S_S → NP V

NP → N (A)+

The second rule can be read, "A Noun Phrase consists of a noun which is optionally followed by any number of adjectives." Note that we are working under the assumption that "fewer rules is better." That is, eventually, we want the phrase structure grammar that is the simplest, where "simplest" here means "uses the fewest amount of rules." [latex] \setcounter{ExNo}{4} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ mti \\ tree}} child {node {A \\ mkubwa \\ big}} child {node {A \\ mzuri \\ good}} } child[sibling distance = 14em] {node {V \\ ulianguka \\ fall.\textsc{past}}}; \end{tikzpicture} [/latex] There's a more important message that our NP rule illustrates. Notice that whenever we put two things together---whether we're making a compound or making a phrase---the result always has the category of one of the pieces we've put together. When we put together an adjective and a noun, we name the resulting constituent a Noun Phrase---not a Verb Phrase, not a Prepositional Phrase, etc. This is because the resulting constituent is still "noun-y," not "verb-y" or "preposition-y." As a rule, a phrase will always be named after one of the [pb_glossary id="362"]heads[/pb_glossary] that it contains. Thus, every phrase, by definition has a head. That is to say, a Noun Phrase contains a noun, a Verb Phrase contains a verb, an Adjective Phrase contains an adjective, etc. And vice versa! We will assume that the opposite is true as well. Every noun is in a Noun Phrase, every verb is in a Verb Phrase, and every adjective is in an Adjective Phrase. We will refer to these two properties of heads as the Headedness Principle.
The Headedness Principle Every phrase has exactly one head and every head is in its phrase.
Though we will see more motivation for this later, for now it suffices to recognize that adopting the Headedness Principle greatly reduces the number of rules we need. If we simply assume that all nouns, verbs, adjectives are in phrases, then we don't need to list two rules for each category. However, once we adopt the Headedness Principle, then we have to adjust our rules. It is no longer possible to have a rule like NP → N (A)+ because there is a head, A, that is not in its phrase. Same for V inside S_S. To fix this issue, we need a few more rules, specifically, we need at least one rule for every head.
Swahili grammar fragment (in progress)

S_S → NP VP

VP → V

NP → N (AP)+

AP → A

If you're paying close attention, you have noticed exactly one exception to the Headedness Principle: S_S does have a head! We'll return to this issue in the next chapter. For now we'll assume that the top of the tree (sometimes call the root) is "special."
With the Headedness Principle and these new rules, we can re-draw the tree from (5) as in (6). [latex] \setcounter{ExNo}{5} \ex. \ag. mti mkubwa mzuri ulianguka\\ tree big good fall.\textsc{past}\\ \trans `A big beautiful tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [sibling distance = 10em] child {node {NP} [sibling distance = 6em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} } child {node {AP} child {node {A \\ mzuri \\ good}} } } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex] One immediate benefit of adding in these extra layers is that now we have a place to put adverbs---which also need to be in a phrase! The example in (7a) demonstrates where adverbs go with respect to adjectives, and the example in (7b) demonstrates through coordination that adjectives and adverbs form a constituent [latex] \setcounter{ExNo}{6} \ex. \ag. mti mkubwa sana ulianguka\\ tree big very fall.\textsc{past}\\ \trans `A very big tree fell.' \bg. mti mkubwa sana na mweusi sana ulianguka\\ tree big very and black very fall.\textsc{past}\\ \trans `A very big and very black tree fell.' [/latex]
Swahili grammar fragment (in progress)

S_S → NP VP

VP → V

NP → N (AP)+

AP → A (AdvP)

AdvP → Adv

[latex] \setcounter{ExNo}{7} \ex. \ag. mti mkubwa sana ulianguka\\ tree big very fall.\textsc{past}\\ \trans `A very big tree fell.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [sibling distance = 10em] child {node {NP} [sibling distance = 6em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} child {node {AdvP} child {node {Adv \\ sana \\ very}} } } } child {node {VP} child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]
Let's keep working on Noun Phrases. Here is some more data.
[latex] \setcounter{ExNo}{8} \exg. mti kwenye kilima ulianguka\\ tree on hill fall.\textsc{past}\\ \trans `A tree on a hill fell.' %\bg. mti karibu-na mto ulianguka\\ %tree near river fall.\textsc{past}\\ %\trans `A tree near a river fell.' [/latex] From these sentences we learn something about [pb_glossary id="701"]adpositional[/pb_glossary] phrases. Swahili appears to have the category preposition; the adposition comes before the noun. (It is a preposition as opposed to a postposition; postpositions come after nouns.) We should add a rule like PP → P NP, but where do we put it? Notice that we have more than one thing that can come after the noun, a prepositional phrase and an adjectival phrase. Which order should they go in? In order to answer this question, we need to look at examples that includes both an adjective and a prepositional phrase. (This is the first case in which we're actually using data to clarify our model. Previously, we had just been taking data and modeling. But now we're specifically seeking our particular data to determine what our model should look like.) [latex] \setcounter{ExNo}{9} \ex. \ag. mti mkubwa kwenye kilima ulianguka\\ tree big on hill fall.\textsc{past}\\ \trans `A big tree on a hill fell.' [/latex] From this example, we conclude that adjectives precede prepositional phrases.
Swahili grammar fragment (in progress)

S_S → NP VP

VP → V

NP → N (AP)+ (PP)

AP → A (AdvP)

AdvP → Adv

PP → P NP

So now we can draw the following tree using our phrase structure grammar for Swahili. [latex] \setcounter{ExNo}{10} \ex. \begin{tikzpicture}[mytree] \node(top) {S$_S$} [level distance = 5em, sibling distance = 14em] child {node {NP} [level distance = 3em, sibling distance = 8em] child {node {N \\ mti \\ tree}} child {node {AP} child {node {A \\ mkubwa \\ big}} } child {node {PP} child {node {P \\ kwenye \\ on}} child {node {NP} child {node {N \\ kilima \\ hill}} } } } child {node {VP} [level distance = 3em, sibling distance = 6em] child {node {V \\ ulianguka \\ fall.\textsc{past}}} }; \end{tikzpicture} [/latex]
Building a Verb phrase rule

We could keep going with Noun Phrases by looking at yet more data, but let's turn to the Verb Phrase. This is often the most complicated phrase to describe, because the verb is, in a sense, the "glue" that holds a sentence together. As always, we start by looking at data.
[latex] \setcounter{ExNo}{11} \ex. \ag. Juma aliona mti\\ Juma see.\textsc{past} tree\\ \trans `Juma saw a tree.' \bg. Juma aliona mti na aliona kilima\\ Juma see.\textsc{past} tree and see.\textsc{past} hill\\ \trans `Juma saw a tree and saw a hill.' [/latex] Again, these two sentences provide an enormous amount of information. We'll concentrate on two things. First, just like in English, the [pb_glossary id="632"]object[/pb_glossary] of the verb comes after the verb. (Such languages are called verb-medial languages; we'll return to cross-linguistic variation in chapter 6.) The second point is that the verb and object form a constituent to the exclusion of the subject. This fact is demonstrated through the coordination test in (12b). We can describe both of these properties by making the minimal adjustment to our VP rule: we add in an optional NP!
Swahili grammar fragment (in progress)

S_S → NP VP

VP → V (NP)

NP → N (AP)+ (PP)

AP → A (AdvP)

AdvP → Adv

PP → P NP

Now our grammar can handle both [pb_glossary id="631"]transitive[/pb_glossary] and [pb_glossary id="617"]intransitive[/pb_glossary] verbs.
[latex] \setcounter{ExNo}{12} \begin{minipage}{.49\textwidth} \ex. \ag. Juma alicheka\\ Juma laugh.\textsc{past}\\ \trans `Juma laughed.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ Juma}} } child {node {VP} child {node {V \\ alicheka \\ laugh.\textsc{past}}} }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \ex. \ag. Juma aliona mti\\ Juma see.\textsc{past} tree\\ \trans `Juma saw a tree.' \b. \begin{tikzpicture}[mytree] \node(top) {S$_S$} child {node {NP} child {node {N \\ Juma}} } child {node {VP} child {node {V \\ aliona \\ see.\textsc{past}}} child {node {NP} child {node {N \\ mti \\ tree}} } }; \end{tikzpicture} \end{minipage} [/latex] Our grammar fragment can continue indefinitely until we have developed a full descriptive grammar of Swahili. For instance, we can develop rules for adverbs in VP (15a), for [pb_glossary id="664"]double-object constructions[/pb_glossary] (15b) (see chapter 10) and for questions (15c). [latex] \setcounter{ExNo}{14} \ex. \ag. Juma alicheka sana\\ Juma laugh.\textsc{past} very\\ \trans `Juma laughed a lot.' \bg. Juma alimnunulia Maria kitabu\\ Juma buy.\textsc{appl}.\textsc{past} Maria book\\ \trans `Juma bought Maria a book.' \cg. Je Juma alicheka?\\ \textsc{ques} Juma laugh.\textsc{past}\\ \trans `Did Juma laugh? [/latex] We won't bother going too in-depth in this endeavor, because in the next chapter we'll refine our theory slightly. But it is important at this point to re-emphasize what exactly we're trying to do when we build a phrase structure grammar. We're attempting to build an accurate description of a language. The way that we're doing this is by applying constituency tests. Note that we're also (implicitly) applying tests to determine category as well. I stipulated above that words like ulianguka and alicheka were verbs. This is not because when I translate them in English, the English word is a verb. It is because, in Swahili, the words have properties of being verbs. In particular, they inflect for person and number marking in their prefixes (in the examples above, the first vowel of the verb is always the subject agreement), and they also inflect for tense marking. (Past tense in Swahili is expressed by the prefix li-.) The words are also distributionally verbs: they occur in between subjects and objects. This basic approach, using constituency tests and identifying categories will not change as we move forward, even as we update our theoretical model. Indeed, these two things are the very heart of all syntactic research. We use empirical evidence to motivate a theoretical model. Our model will eventually be more sophisticated, but only as a response to the data that we look at.
[embed]https://youtu.be/Gjtv4zBU0bU[/embed] Tree drawing with phrase structure grammars (length: 1m 41s)

Where we're going
On the approach taken in this chapter, we assume that all languages have a distinct phrase structure grammar. That is, for any one language, the rules are idiosyncratic to that language. We just look at the data and determine what those rules are. And if language X and language Y in fact share the exact same rules, then they're the same language! (Like if x=2 and y=2 then x=y.) As we'll point out in the next chapter, though, the idea that every language is different is actually somewhat misleading. Languages turn out to be much more similar than they are different. For instance, it is robustly attested across languages that the object and verb form a constituent to the exclusion of the subject. It is also robustly attested across languages that if adjectives go after nouns in language X, then the object goes after the verb in language X as well. On the approach developed in this chapter, these similarities and patterns go unexplained. That is, given our current model, it is essentially entirely arbitrary how a language can put its pieces together, or better, how a language organizes its information. But when we truly look at a lot of cross-linguistic data, we find that this doesn't seem right. Ideally, our theoretical model isn't just adequate for, e.g., Swahili, but for all languages. This is what we're going to work towards in the next chapter. We'll look at more data, and attempt to build a theory of syntax that applies to any human language.

Things to remember

the Headedness Principle

the coordination rule

how to use constituency tests to infer structure

how to draw trees using phrase structure grammars

Advanced

Complexity in Phrase Structure Grammars
How complex can a phrase-structure grammar get? The answer is: very. Consider the following rules, taken from one of the standard syntax textbooks.[footnote]Carnie, Andrew. 2006. Syntax: A Generative Introduction, Second edition. Malden, MA: Blackwell.[/footnote]

CP → (C) TP

TP → {NP/CP} (T) VP

VP → (AdvP+) V (NP) ({NP/CP}) (AdvP+) (PP+) (AdvP+)

NP → (D) (AdjP+) N (PP+) (CP)

PP → P (NP)

AdjP → (AdvP) Adj

XP → XP conj XP

X → X conj X

One notational addition in these rules are the braces: "{.../...}". These are used to mean "either/or (but not both at once)." So the rule for TP, either an NP (what we have called DP in this textbook) or a CP can preceded T, i.e., be a subject. This accounts for sentences like The noise surprised Malika and That Jessie came home late surprised Malika. The latter sentence has a CP subject: that Jessie came home late is the subject of surprised. Notational differences aside, it is important to note in these rules that basically anything goes---as long as there is empirical motivation. So the reason that the VP is so long is because VP can have a lot of stuff in them. This is empirically necessary. But note that there really is not upper limit how how complex a rule can be. This is not a good outcome. Because it predicts that Human Languages can be random. That is, it reduces every Human Language to an unpredictable system of rules describing each phrase. This turns out to be incorrect. As we'll discuss in the next chapter, Human Languages are actually extremely systematic, meaning that, all things considered, the rules that describe a language tend to look very similar to each other. There is very little "randomness" in any given language. But the complex Phrase Structure Grammars serve a purpose. First, empirically, they show that phrases are complex creatures, often consisting of multiple distinct pieces. The Phrase Structure Grammars are complex because Human Language is complex (which is different from saying that Human Language is random/unpredictable). Second, the rules above serve a theoretical purpose.
Rule or transformation?
Sometimes, it's not clear when looking at data whether we should treat the given phrase as the result of a Phrase Structure Rule, or the result of a transformation. The term "transformation" is used to describe any changes that apply to a sentence. For instance, we typically describe topicalization as being derived via transformation. Recall that with topicalization, some constituent is moved to the beginning of the sentence. [latex] \setcounter{ExNo}{15} \ex. \a. Juma hugged the fruit seller. \b. The fruit seller, Juma hugged. [/latex] Should we make a Phrase Structure Rule to describe (16b)? The answer is usually taken to be "no." Why not? Because we can describe this kind of a sentence via a "rule of transformation," that is, a rule that describes a generalized process for turning one kind of sentence into another kind of sentence. In the case of topicalization, the rule is simple: place whatever you want to topicalize at the beginning of the sentence. Rules of transformation serve a dual purpose. On the one hand, they underscore the overall goal of reducing Human Language to a set of abstract rules that can apply to an infinite number of sentences. With a rule that describes topicalization, I can topicalize anything I want. On the other hand, rules of transformation associate particular word orders with particular meanings. This is crucial. The order of DP DP V in English is uniquely associated with the meaning of topicalization. This differentiates it then from the more basic word orders that our Phrase Structure Rules generally describe. For instance, the order V DP (a basic rule for VPs in English) doesn't "mean" a particular thing; this rule applies generally to all transitive constructions: build the house; hold a meeting; sing a tune; drop a single; ... . This is the "basic" order, and it doesn't have an inherent "meaning" associated with that order. Topicalization is different, though, because there is a meaning associated with that order. So we want a rule that applies just in that case. It is sometimes hard to know for any given word order whether it should be derived by a transformation, or whether it should be part of the basic word orderof the language. For instance, consider adjective orders in English. While adjectives are always allowed in front of a noun, sometimes adjectives can appear after a noun.[footnote]Cinque, Guglielmo. 2010. The Syntax of Adjectives: A Comparative Study. Cambridge, MA: MIT Press. [/footnote] [latex] \setcounter{ExNo}{16} \ex. \a. The only visible stars are Aldebaran and Sirius. \b. The only stars visible are Aldebaran and Sirius. [/latex] Should our DP rule include both orders of N and AP? This idea is debated, but one hint that maybe a rule of transformation can derive one of the order comes from meaning. Remember that if a particular order of words is associated with a particular meaning, then it might be better to try to derive that order via a rule of transformation. In (17), the two orders actually do have distinct meanings: the visible stars means something slightly different than the stars visible. To use the latter order (the stars visible), you have to talking about about stars that you can see right now. For instance, on cloudy night, when no one can see any stars, you can still talk about which stars are generally visible with (17a). This doesn't quite work for (17b). If no stars are visible, it sounds funny to talk about the stars visible. This difference in meaning suggests that more than a mere change in word order is going on. Rather, it suggests that there is a rule that transforms the order from AP N to N AP to map to a particular meaning. The broader takeaway is the following: As we continue moving forward in describing language, we should be paying attention to not only the patterns of word order (and category and constituency), but also whether there are patterns that associate particular words orders with particular meanings. If we find such patterns, then we can create abstractions over those patters: rules! Ultimately, we won't adopt any rules of transformation---but we will have movement, which accomplished the same thing.

X-bar syntax

juliedoner — Wed, 15 Nov 2023 05:29:36 +0000

Learning Objectives

By the end of this chapter, you should be able to,

identify intermediate levels ("bar levels") using constituency tests

understand the motivation for the X-bar schema

draw a tree using the X-bar schema,

identify the distinct positions within the X-bar schema,

diagnose the difference between adjuncts and complements

By the end of this chapter, we will have technically changed theories. We're going from Phrase Structure Grammars to X-bar Syntax. But the trick is to recognize that, even though we're going to be using the term "X-bar syntax," we aren't actually changing theories at all. The same principles, formalizations, and theoretical tools will be used in X-bar syntax that we've already been using. Our trees are just going to look more complicated. The reason that our trees will look more complicated is that we're going to discover in this chapter that there are lot of "hidden" layers. Just like when we discussed functional categories we learned to infer the presence of structure, even when we can't directly see it, we'll use the same tools to infer the presence of "bar-levels." From this, we'll make a theoretical leap. As a hypothesis about syntax, we'll adopt a uniform structural representation for all phrases. This is "X-bar syntax." It's the hypothesis that all phrases share a structure, i.e., every phrase looks identical. The hypothesis makes a number of testable predictions, which we'll explore in terms of the complement/adjunct distinction.
Deconstructing VP
Let's start by looking again at English VPs. Assume the (very simplistic) rule for VPs below.
Simplified VP rule for English (to be revised)

VP → V (DP) (PP)+

This rule allows us to minimally distinguish between [pb_glossary id="631"]transitive[/pb_glossary] and [pb_glossary id="617"]intransitive[/pb_glossary] verbs, each of which could have a modifying prepositional phrase. For instance, using this VP rule, here's the structure for The doctor will sing a song in the park. [latex] \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 6em] child {node {DP} [sibling distance = 5em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T \\ will}} child {node {VP} [sibling distance = 5em] child {node {V \\ sing}} child {node {DP} [sibling distance = 3em] child {node {D \\ a}} child {node {NP} child {node {N \\ song}} } } child {node {PP} [sibling distance = 3em] child {node {P \\ in}} child {node {DP} child {node {D \\ the}} child {node {NP} child {node {N \\ park}} } } } } }; \end{tikzpicture} [/latex] This tree correctly represents that in the park is a constituent. It also correctly represents that sing a song in the park is a constituent. However, it still is getting the constituency wrong. Look at the substitution test below. What does it show is a constituent? [latex] \setcounter{ExNo}{1} \ex. \a. The doctor will sing a song in the park. \b. and Mary will do so in the office. [/latex] What's the problem here? The test demonstrates that sing a song is a constituent, because I am able to replace that string with do so. The tree in (1) gets this wrong. The string sing a song isn't a constituent according to this tree because there is no node that contains the words sing a song and nothing else. Is our constituency test just giving us a false positive? Let's confirm. Below I'm applying the other constituency tests. [latex] \setcounter{ExNo}{2} \ex. Fragments\\ What will the doctor do in the park? {\underline{Sing a song}} \ex. \a. \textit{It}-Clefting\\ {}*It was {\underline{sing a song}} that the doctor will do in the park. \b. Pseudo-clefting\\ {\underline{Sing a song}} is what the doctor will do in the park. \c. \textit{All}-clefting\\ {\underline{Sing a song}} is all the doctor will do in the park. \d. Topicalization\\ {}*{\underline{Sing a song}}, the doctor will do in the park. \ex. Coordination\\ The doctor will [ sing a song ] and [ dance ] in the park [/latex]
So our constituency tests are telling us that sing a song is a constituent---and that it's a VP. Recall that substitution and coordination allow us to diagnose the category of the constituent. But here's the thing: the string sing a song in the park is also a VP. How do I know? It passes all the same constituency tests, and we can detect its category through substitution and coordination.
[latex] \setcounter{ExNo}{5} \ex. Substitution\\ The doctor will {\underline{sing a song in the park}}, and the nurse will {\underline{do so}}, too \ex. Fragments\\ What will the doctor do? {\underline{Sing a song in the park}} \ex. \a. \textit{It}-Clefting\\ {}*It was {\underline{sing a song in the park}} that the doctor will do. \b. Pseudo-clefting\\ {\underline{Sing a song in the park}} is what the doctor will do. \c. \textit{All}-clefting\\ {\underline{Sing a song in the park}} is all the doctor will do. \d. Topicalization\\ {}*{\underline{Sing a song in the park}}, the doctor will do. \ex. Coordination\\ The doctor will [ sing a song in the park ] and [ dance at home ] [/latex] So based on this evidence, it looks like our tree should actually have the following representation. [latex] \setcounter{ExNo}{9} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance=6em] child {node {DP} [sibling distance = 5em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T \\ will}} child {node {VP} child {node {VP} [sibling distance = 5em] child {node {V \\ sing}} child {node {DP} [sibling distance = 3em] child {node {D \\ a}} child {node {NP} child {node {N \\ song}} } } } child {node {PP} [sibling distance = 3em] child {node {P \\ in}} child {node {DP} child {node {D \\ the}} child {node {NP} child {node {N \\ park}} } } } } }; \end{tikzpicture} [/latex] This version of the tree correctly captures our empirical result: we have two distinct VP constituents. (It also correctly predicts that a song in the park is not a constituent, which you can confirm on your own.) The "trick" is that we've added a VP in-between the head and the highest VP. This projection is empirically motivated, though. It's there because our tests tell us it's there. (Note that technically, the highest VP violates the Headedness Principle.) We call this intermediate projection the "bar" level, and write it either with a line over the top (V) or an apostrophe after (V'). It's pronounced "V bar" (not "V prime"). This is just a notation for saying "I'm not the highest phrase in this phrase." To review: we've used our constituency tests to determine that there is a "hidden" level in verb phrases.
Let's look at more data.
[latex] \setcounter{ExNo}{10} \ex. \a. John will sing a song on a bike in the park and Mary will do so at the office. \b. John will sing a song on a bike and dance a jig at the park at different times. [/latex] These data are just like the data above, except that I've added in more prepositional phrases. So now sing a song on a bike can't be the highest VP in the structure. The highest VP is the entire constituent sing a song on a bike in the park. This is the topmost VP. Now in fact, we need two intermediate projections, because the constituents sing a song and sing a song on a bike and sing a song on a bike in the park are all VPs. [latex] \setcounter{ExNo}{11} \ex. [ John will [$_{VP}$ [$_{V'}$ [$_{V'}$ sing a song ] on a bike ] in the park ] ] [/latex] These data actually suggest that I can simplify my representation quite a bit, because they suggest that I can make do without the "+" symbol. All I need are three rules like the following. Note that three rules is actually simpler than one rule with a plus sign because now we can get by with less formalism. By analogy, 2 + 2 + 2 is simpler than 2 x 3 because it provides the same information with a less complex operator.
Simplified VP rule for English (revised)

VP → V'

V' → V' PP

V' → V (DP)

I've done a very tricky (but ultimately useful) thing here. I've explicitly added [pb_glossary id="622"]recursion[/pb_glossary] into my English phrase structure grammar. There's a rule that says V' (i.e., an intermediate projection between the head and the topmost VP) can consist of itself and a prepositional phrase. Why does this work? Consider the tree for John will sing a song on a bike in the park which uses the new rule above. [latex] \setcounter{ExNo}{12} \ex. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C}} child {node {TP} child {node {DP} child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ John}} } } child {node {T \\ will}} child {node {VP} child {node {V$'$} child {node {V'} child {node {V'} child {node {V \\ sing}} child {node {DP \\ a song}} } child {node {PP \\ on a bike}} } child {node {PP \\ in the park}} } } }; \end{tikzpicture} [/latex] When I build a VP now, I have an option: I can either add in an infinite number of prepositional phrases by adding in a new V' level, or I can just go directly to the head V. I've essentially created a way to make an infinitely recursive loop---with an escape hatch. Most importantly, all of these intermediate projections correlate with distinct, identifiable constituents. That is, our structure correctly represents how English packages its information.
Deconstructing TP
We can do the exact same thing with TP. Let's assume our rule for TP developed in the last chapter.
TP rule for English (to be revised)

TP → DP T VP

Now consider the following data, which suggests that there is a bar-level---an intermediate projection---in TP. [latex] \setcounter{ExNo}{13} \ex. \a. John will eat sushi and will drink sake. \b. Mary has forgiven Jed and might forgive Brad. \c. Conrad can sing and could dance. [/latex] What is this data showing? It shows that I can coordinate will eat sushi with something else. That is, there appears to be a constituent that include T and VP, but does not include the subject. This is not possible, according to our rule above. To fix this issue, we can re-write our rule, parallel to what we did with VP.
TP rule for English (revised)

TP → DP T'

T' → T VP

Adding in this extra level correctly captures what our constituency test is telling us. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 7em] child {node {DP} [sibling distance = 5em] child {node {D \\ $\emptyset$}} child {node {NP} child {node {N \\ John}} } } child {node {T$'$} [sibling distance = 4em] child {node {T \\ will}} child {node {VP} [sibling distance = 2em] child {coordinate(left1)} child {node {eat sushi}edge from parent[draw=none]} child {coordinate(right1)} } } }; \draw (left1)--(right1); \end{tikzpicture} [/latex] This tree now correctly captures the fact that T forms a constituent with VP, to the exclusion of the subject. That constituent is the bar level, T'. What if I add in more words. Here I'm adding a modifier of TP. [latex] \setcounter{ExNo}{14} \ex. \a. John definitely will eat sushi and probably will drink sake. \b. John definitely will eat sushi and drink sake. [/latex] This data now suggests that the modifier gets grouped together with T and the VP in the first example (because both constituents have a similar modifier), but in the second example shows us that T and the VP can form a constituent that is separate from this modifier (because definitely is describing both the eating and drinking events). How are we going to make this work? If Adverbs[footnote]Note that the term "adverb" is a bit misleading: this aren't "adding" to the "verb".[/footnote] like definitely, probably, etc may sometimes be distinct from the VP, and sometimes not, we need to adjust our rules about where adverbs go in TP. Just like with VP, we can fix this issue concisely by adding a recursive T' level!
TP rule for English (revised)

TP → DP T'

T' → AdvP T'

T' → T VP

This set of rules allows us to add as many modifiers of TP as we want, and we will always find that each additional modifier "creates" a subconstituent within TP. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node {DP} [sibling distance = 3em] child {node {D \\ the}} child {node {NP} child {node {N \\ doctor}} } } child {node {T$'$} [sibling distance = 6em] child {node {AdvP} child {node {Adv \\ definitely}} } child {node {T$'$} [sibling distance = 4em] child {node {T \\ will}} child {node {VP} [sibling distance = 2em] child {coordinate(left1)} child {node {eat sushi}edge from parent[draw=none]} child {coordinate(right1)} } } } }; \draw (left1)--(right1); \end{tikzpicture} [/latex]
Deconstructing NP

We'll do the exact same thing with NP. Keep in mind that we've adopted the [pb_glossary id="689"]DP-hypothesis[/pb_glossary], so all NPs are inside of DPs.

Simplified NP rule for English (to be revised)

NP → (AP)+ N

With this rule in mind, consider the patterns of one substitution shown below. [latex] \setcounter{ExNo}{15} \ex. \a. Tricia read the short book, not the long one. \b. Tricia read the short blue book, not the long one. [/latex] We observed in the last chapter that one substituted for NP, but these data suggest that that isn't quite accurate. One must be substituting for something between NP and N. Consider (16b). One is replacing the constituent blue book. This includes an adjective, so it's not just the head N, but it excludes the higher adjective short/long, so this cannot be the entire NP. We therefore find evidence in NP for an intermediate projection, and we re-write our rules accordingly. Now we can add in an infinite number of prenominal adjectives by adding in more intermediate projections.
Simplified NP rule for English (revised)

NP → N'

N' → AdjP N'

N' → N

These rules correspond to the following tree. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D \\ the}} child {node {NP} child {node {N'} child {node {AdjP} child {node {Adj \\ short}} } child {node {N$'$} child {node {AdjP} child {node {Adj \\ blue}} } child {node {N$'$} child {node {N \\ book}} } } } }; \end{tikzpicture} [/latex]
Generalizing across categories

When we look across all of our rules, we can make a few generalizations:

There's always a rule of the form XP → X', where the phrase level goes to the intermediate level. (Sometimes this rule has an element in front of X'.)

There's always a rule of the form X' → X, where the bar-level consists of the head. (Sometimes the head is followed by something on the right.)

There's always a recursive rule of the form X' →YP X' or X' → X' YP, where YP is a modifier of some sort.

If this pattern exists for every single category, it suggests that we can generalize our schema for phrase structure. We hypothesize that every phrase has the following rules. The "X, Y, Z, W" below are variables meaning "any category."

Generalized phrase structures rules (final version)

XP → (YP) X'

X' → X' ZP

X' → ZP X'

X' → X (WP)

If this is right, it means that all phrases are constructed based on these rules. In the rules above X, Y, and W are variables over categories.
What about the Headedness Principle? Have we entirely abandoned it? No! The Headedness Principle, which says that every phrase has a head and every head is in its phrase is exactly the same. But we've redefined what "phrase" means. Now a phrase minimally involves at least one X'-level. A phrase consists of all of the rules above: it is the X-bar schema. The term "phrase" is now a structural notion. The X-bar schema is a hypothesis. By looking at a lot of data, we identify that there are intermediate projections in every phrase that we look at. The subsequent phrase structure rules all end up looking very similar---suspiciously similar in fact. The X-bar schema is a way to make sense of this information. It proposes that the similarities reflect a "deep" uniformity across syntax. Our trees end up looking more complex, because now everything needs at least one intermediate projection. But the complexity of the representations comes with an overall simplification of the theory: it isn't necessary to learn a bunch of arbitrary rules for syntax. You only need the four above, because they apply to every phrasal category. [latex] \setcounter{ExNo}{16} \ex. The tired doctor cured the sick patient yesterday. Old way: \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 8em] child {node {DP} [sibling distance = 4em] child {node {D \\ the}} child {node {NP} child {node {AP} child {node {A \\ tired}} } child {node {N \\ doctor}} } } child {node {T \\ $[$past$]$}} child {node {VP} [sibling distance = 6em] child {node {V \\ cured}} child {node {DP} [sibling distance = 4em] child {node {D \\ the}} child {node {NP} child {node {AP} child {node {A \\ sick}} } child {node {N \\ patient}} } } child {node {AdvP} child {node {Adv \\ yesterday}} } } }; \end{tikzpicture} [/latex] [latex] New way: \begin{tikzpicture}[mytree, scale=.9] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance=8em] child {node {DP} [sibling distance=3em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {AP} child {node {A$'$} child {node {A \\ tired}} } } child {node {N$'$} child {node {N \\ doctor}} } } } } } child {node {T$'$} [sibling distance=6em] child {node {T \\ $[$past$]$}} child {node {VP} [sibling distance=4em] child {node {V$'$} [sibling distance = 6em] child {node {V$'$} [sibling distance = 4em] child {node {V \\ cured}} child {node {DP} [sibling distance = 3em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {AP} child {node {A$'$} child {node {A \\ sick}} } } child {node {N$'$} child {node {N \\ patient}} } } } } } } child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ yesterday}} } } } } } } }; \end{tikzpicture} [/latex]
[embed]https://youtu.be/BrTh1zqifdA[/embed] Trees in the X' schema (length: 1m 28s)

Note we also still need special rules for coordination, which look basically the same as before. Rules for coordination (final version)

XP → XP conj XP

X' → X' conj X'

X → X conj X

These rules allow you to coordinate heads (X), bar-levels (X') and phrases (XP), but you can't mix-and-match. You cannot coordinate a bar level with a head, for instance.

Positions in X'-syntax
We define four positions within the X'-schema: specifier, head, complement and adjunct. The positions are indicated in the trees below. [latex] \begin{tikzpicture}[mytree] \node(top) {XP} child {node(spec1) {YP}} child {node {X$'$} child {node {X$'$} child {node(head1) {X}} child {node(comp1) {WP}} } child {node(adj1) {ZP}} }; \node(head) at (-4,-5) {``head"}; \node(spec) at (-5,0) {``specifier"}; \node(comp) at (5,-5) {``complement"}; \node(adj) at (5,-2) {``adjunct"}; \draw[->, dashed] (head)--(head1); \draw[->, dashed] (spec)--(spec1); \draw[->, dashed] (comp)--(comp1); \draw[->, dashed] (adj)--(adj1); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {XP} child {node(spec1) {YP}} child {node {X$'$} child {node(adj1) {ZP}} child {node {X$'$} child {node(head1) {X}} child {node(comp1) {WP}} } }; \node(head) at (-4,-5) {``head"}; \node(spec) at (-5,0) {``specifier"}; \node(comp) at (5,-5) {``complement"}; \node(adj) at (-5,-2) {``adjunct"}; \draw[->, dashed] (head)--(head1); \draw[->, dashed] (spec)--(spec1); \draw[->, dashed] (comp)--(comp1); \draw[->, dashed] (adj)--(adj1); \end{tikzpicture} [/latex]

A head is the only non-phrasal [pb_glossary id="673"]terminal node[/pb_glossary] in a phrase.

A specifier is defined as, "The phrase that is [pb_glossary id="674"]daughter[/pb_glossary] of XP and [pb_glossary id="675"]sister[/pb_glossary] to X." Every phrase that meets this description is a specifier.

A complement is defined as, "The phrase that is sister to X and daughter to X." Every phrase that meets this description is a complement.[footnote]FYI Complement is spelled with an 'e'. It is not compliment.[/footnote]

An adjunct is defined as, "The phrase that is daughter to X and sister to X." Every phrase that meets this description is an adjunct.

Note that the definitions above do not make reference to linear order. It doesn't matter whether an adjunct is to the left or the right. All that matters is that its sister and mother are bar-levels. Likewise, complements can be on the left or the right of a head, as long as it is sister to the head. And a specifier can be on the left right as well, as long its mother is a phrase. Every language makes particular choices however about where specifiers and complements appear. We'll return to this momentarily. For now, we note that in English, specifiers are always on the left, and complements are always on the right.
It is conventional to use the "genealogical" terms mother, sister, and daughter when referring to the relationships between elements in a tree. The terms are meant to be transparent: A "mother" is something that has "daughters." If two things have the same mother, then they're "sisters." In the tree below, DP is the mother of D'. D' is the mother of D and NP. D and NP are sisters, and are both daughters of D'. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP}} }; \end{tikzpicture} [/latex] The analogy basically stops there. There is no such thing as an "aunt" in syntax, or a "niece."
We've seen a lot of complements. TP is a complement to CP. NP is a complement to D. DP can be a complement to V when it's an object of the verb. The only specifier we've seen is the subject. It's the specifier of TP. There is another place where a specifier is necessary: DP. Possessive structure. The other place where we observe specifiers is in possessive DPs. [latex] \setcounter{ExNo}{17} \ex. the mailman's bag \begin{tikzpicture}[mytree] \node(top) {DP} [sibling distance =7em] child {node {DP} [sibling distance =4em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ mailman}} } } } } child {node {D$'$} [sibling distance =4em] child {node {D \\ 's}} child {node {NP} child {node {N$'$} child {node {N \\ bag}} } } }; \end{tikzpicture} [/latex] We treat the possessive marker -s as a realization of D. This is because possessives are in complementary distribution with determiners: *the mailman's the bag. And because the entire DP acts as a constituent referring to a bag, not a mailman, the head of the highest DP must be the possessive D.
[embed]https://youtu.be/XQmS-ue8oko[/embed] Possessive DPs (length: 1m 51s)

Complements vs. Adjuncts

The X-bar schema illustrated above makes a prediction that some phrases that are next to a head are complements while some are adjuncts. They may look identical on the surface, but they derive from different structures.

For instance, consider the phrase the state of Kansas. In the X-bar schema, this constituent (and it definitely is a constituent) could hypothetically correspond to either tree below. The prepositional phrase of Kansas could either be a complement to state (as in the tree on the left) or an adjunct to state (as in the tree on the right).
[latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \draw (left)--(right); \hspace{200pt} \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N \\ state}} } child {node {PP} [sibling distance=2.5em] child {coordinate(left1)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right2)} } } } }; \draw (left1)--(right2); \end{tikzpicture} [/latex] The two trees actually make distinct empirical predictions. We have three tests to distinguish between complements and adjuncts. First, I just want to briefly mention that it is easy to confuse adjuncts and specifiers. When drawing your trees, remember that for every adjunct, you should have "plus one" bar-level, because adjuncts need both a bar-level which is a sister, and bar-level which is a mother. So you have to add a layer whenever you add an adjunct.
[embed]https://youtu.be/nOe_4igGwfA[/embed] Adjuncts in X'-syntax (length: 2m 15s)

Substitution

The constituency test of substitution has a different use. As we observed above, one substitution and do so substitution both target the bar-level. Since complements are under all bar-levels in a phrase, if you try to do substitution, a complement cannot be "left behind." An adjunct can, though, because there's always a bar-level that doesn't include the adjunct.
Visually, consider the two trees again. If the PP is a complement to N, then when you substitute with one everything under N' has to go too. In contrast, if the PP is an adjunct, it's possible to substitute with one and leave behind the PP. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node(nbar) {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \node(text) at (-3,-5) {\textit{one} substitutes \\ here}; \draw (left)--(right); \draw[->, dashed] (text)--(nbar); \hspace{250pt} \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node(nbar2) {N$'$} child {node(nbar1) {N$'$} child {node {N \\ state}} } child {node {PP} [sibling distance=2.5em] child {coordinate(left1)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right2)} } } } }; \node(text2) at (-3,-5) {\textit{one} substitutes \\ either place}; \draw (left1)--(right2); \draw[->, dashed] (text2)--(nbar1); \draw[->, dashed] (text2)--(nbar2); \end{tikzpicture} [/latex] So to use the substitution test to determine if something is a complement or adjunct, you simply replace the head of the phrase, and see if you can leave behind the constituent in question. Compare the sentences below. [latex] \setcounter{ExNo}{18} \ex. \a. the state of Kansas, and\\ {}*the one of Maine \b. the state near Kansas, and\\ {} the one near Maine. [/latex] As illustrated above, one-substitution crucially yields an ungrammatical result in (19a), but not in (19b). Why is this? Again, note here that one-substitution targets bar-levels; you can only replace the material that exists under a bar-level. As such, the reason for the grammaticality distinction in the examples above is that the NPs, the state of Kansas and the state near Kansas, crucially differ in the number of bar-levels they contain. That is, unlike in the NP, [ _NP state of Kansas ], where one-substitution can replace both N and PP (20a), but not N alone (20b), with the NP, [ _NP state near Kansas ], one-substitution can either replace the larger constituent containing N and PP (21a), or N alone (21b), meaning that the NP in (19) must crucially contain two bar-levels, and the NP in (20), only one. [latex] \setcounter{ExNo}{19} \ex. I visited the state of Kansas. \\ \a. I visited the \textit{one}. \hfill ( targets: [  \textsubscript{N'} [\textsubscript{N} state ][ \textsubscript{PP} of Kansas ]] ) \\\b. *I visited the \textit{one} of Kansas. \hfill ( targets: [ \textsubscript{N'} [ \textsubscript{N} state ]] )\\ \vspace{3mm} \ex. I visited the state near Kansas\\ \a. I visited the \textit{one}. \hfill ( targets: [ \textsubscript{N'} [\textsubscript{N} state ][ \textsubscript{PP} near Kansas ]] )\\\b. I visited the \textit{one} near Kansas.   \hfill ( targets: [ \textsubscript{N'} [ \textsubscript{N} state ]] )\\ [/latex] Since we have defined a complement as being the sister of the head that selects it, and the daughter of the bar-level that contains that same head (i.e. a head and its complement share a bar-level parent), while conversely defining an adjunct as being both the sister to a bar-level and the daughter to a bar-level, the fact that one-substitution can target two different constituents in (21) means that the NP, the state near Kansas, necessarily contains two bar-levels, and, therefore, must contain an adjunct. That is, the PP, near Kansas, must be an adjunct because there are two different targets (i.e. bar-levels) for one-substitution in (20); one can replace the lower bar-level to the exclusion of PP, i.e. [ _N' [ _N state ]], or it can replace the higher bar-level that includes both the lower N' and PP, i.e. [ _N' [_N' [ _N state][ _PP near Kansas ]].
Step-by-step walkthrough of the substitution test

Identify the head in question and the phrase which you want to test

Identify the appropriate substitution word for the head.

Replace the head with the substitution word, but don't replace the phrase you're testing. (Note that technically, you're substituting at the lowest bar-level, which contains the head. The lowest bar-level could, in principle also have a complement.)

Check the grammaticality.

[embed]https://youtu.be/hZvQgFIutiw[/embed] Substitution test for complements vs. adjuncts (length: 1m 6s)

[embed]https://youtu.be/4S4euaW9NDM[/embed] Substitution test for complements vs. adjuncts (length: 1m 9s) Credit Damian White Lightning

Reordering

Complements must appear as sisters to the head, and so they have a fixed position with respect to other phrases. But adjuncts can, theoretically, be reordered. This is because all that's necessary is that they have a mother that's a bar-level and a sister that's a bar-level.
For instance, consider the following trees which each have two prepositional phrases. The tree on the left treats PP₁ as an complement and PP₂ as an adjunct. If we switched their positions, then we switch their roles. [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N}} child {node {PP$_1$}} } child {node {PP$_2$}} } } }; \hspace{250pt} \node(top) {DP} child {node {D$'$} child {node {D}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N$'$} child {node {N}} } child {node {PP$_1$}} } child {node {PP$_2$}} } } }; \end{tikzpicture} [/latex] In contrast, in the tree on the right, both the PPs are adjuncts. It shouldn't matter whether we switch their positions, because they both will still be adjuncts.
What this means is that if a constituent is a complement, it cannot be reordered with another constituent. If a constituent is an adjunct, it can be.
The difference can be shown empirically in the following examples. You can't reorder of Kansas and near Maine, suggesting that of Kansas is a complement to state. In contrast, you can reorder by the ocean and near Maine, so we conclude that by the ocean is an adjunct to state. [latex] \setcounter{ExNo}{20} \ex. \a. the state of Kansas near Maine \b. *The state near Maine of Kansas \ex. \a. the state by the ocean near Maine \b. the state near Maine by the ocean [/latex]
Step-by-step walkthrough of the reordering test

Identify the head and the phrase which you want to test.

Choose an appropriate phrase that can appear with the head and phrase above.

Reorder or the two phrases with respect to the head, and check the grammaticality.

[embed]https://youtu.be/FY7XaY7VKxY[/embed] Reordering test (length: 1m 2s)

[embed]https://youtu.be/wX7EHFCTJTk[/embed] Reordering test (length: 1m 6s) Credit: Damian White Lightning

Sometimes we can't reorder phrases for semantic reasons, independent (presumably) of syntax. There happens to be a fixed order for modification that is fairly robust cross-linguistically. For instance, temporal modifiers (things that deal with "time") tend to be merged higher (or "outside") of manner modifiers (things that describe "how" the event was accomplished). Likewise, there is a fairly fixed ordering of adjectives in English (and many languages).
[latex] \setcounter{ExNo}{21} \ex. \a. Sammie cut Vaughn's hair {\underline{with scissors}} {\underline{often}} \b. ??Sammie cut Vaughn's hair {\underline{often}} {\underline{with scissors}} \ex. \a. the big brown ball \b. ??the brown big ball [/latex]
Linear ordering

As a corollary to the last test, if you can't reorder the phrases, the phrase nearest the head must be a complement. In fact, we predict that complements are always strictly adjacent to the head. Therefore, it should be impossible to insert anything between a head and complement. (Note that this test is irrelevant (N/A) if you find that you can reorder the phrase in question.)
Again, the difference is shown empirically below. [latex] \setcounter{ExNo}{22} \ex. \a. *the state that I visited of Kansas \b. the state that I visited near Maine [/latex]
Step-by-step walkthrough of linear ordering test

Identify the head and the phrase you want to know about.

Identify some other constituent that can also directly follow the head.

Put the phrase you want to know about after the new phrase, both after the head, and check the grammaticality.

[embed]https://youtu.be/hcClPu5znmQ[/embed] Linear ordering test (length: 1m 10s)
The reason that this test works is that the relationship between the head and complement is fixed: they must be sisters. But the relationship between a head and an adjunct is not fixed; there can be any number of adjuncts be between the head and the adjunct. Based on the three tests above, we find that of Kansas is a complement to state. So the following is the correct representation: [latex] \begin{tikzpicture}[mytree] \node(top) {DP} child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N \\ state}} child {node {PP} [sibling distance=2.5em] child {coordinate(left)} child {node {of Kansas} edge from parent[draw=none]} child {coordinate(right)} } } } }; \draw (left)--(right); \end{tikzpicture} [/latex]
Parameters

The order of the specifiers and complements is fixed for any one language. What does that mean? For instance, in English, the complement always comes after the head. That is, English is head-initial. With specifiers, the specifier always comes before the head. English is specifier-initial.

However, logically for each position, there are four ways things could be.

Specifier-initial : The specifier is to the left of its head (meaning to the left of the bar-level).

Specifier-final : The specifier is to the right of its head (meaning to the right of the bar-level).

Head-initial : The head is the left of its complement.

Head-final : The head is to the right of its complement.

[latex] \begin{minipage}{.49\textwidth} Specifier-initial: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {YP}} child {node {X$'$} child {node {X}} }; \end{tikzpicture} \end{minipage} \hspace{50pt} \begin{minipage}{.49\textwidth} Specifier-final: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {X}} } child {node {YP}}; \end{tikzpicture} \end{minipage} [/latex] [latex] \begin{minipage}{.49\textwidth} Head-initial: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {X}} child {node {YP}} }; \end{tikzpicture} \end{minipage} \hspace{50pt} \begin{minipage}{.49\textwidth} Head-final: \begin{tikzpicture}[mytree] \node(top) {XP} child {node {X$'$} child {node {YP}} child {node {X}} }; \end{tikzpicture} \end{minipage} [/latex]
It turns out that languages make different choices as to how they order their heads and specifiers (and adjuncts). This is called parametric variation. English is specifier-initial and head-initial. Thus, in "neutral" contexts, it will have the order of Subject-Verb-Object, or SVO.

Kashaya (a Pomoan language spoken in Northern California) has set its parameters to specifier-initial and head-final. So it's basic word order is "Subject-Object-Verb," or SOV.
[latex] \setcounter{ExNo}{23} Kashaya: \exg. Conrad Anita-to cadu\\ Conrad Anita-\textsc{acc} saw\\ \trans `Conrad saw Anita.' [/latex] Malagasy (an Austronesian language spoken on Madagascar) has set its parameters to specifier-final and head-initial. It's basic word order is "Verb-Object-Subject," or VOS. [latex] \setcounter{ExNo}{24} Malagasy: \exg. nahita ny mpianatra ny vehivavay\\ saw the student the woman\\ \trans `The woman saw the student.'' [/latex] Lastly, Hixkaryana (a Carib language spoken in Brazil) has set its parameters to specifier-final and head-final. So it's basic word order is "Object-Verb-Subject," or OVS. [latex] \setcounter{ExNo}{25} Hixkaryana: \exg. kana yan{\"i}mno b{\"i}yekomo\\ fish caught boy\\ \trans `The boy caught a fish.' [/latex]

If you're counting, you'll notice that I've said nothing about whether the two other logically possible words orders of Subject, Verb, and Object exists: VSO and OSV. These cannot be straightforwardly derived by simply stipulating specifier and head directions. Something more (i.e., movement) must be added to the theory.
Nonetheless, K'iche' (a Mayan language spoken in Guatemala) has a basic word order of "Verb-Subject-Object," or VSO. [latex] \setcounter{ExNo}{26} \exg. xril la ali Marij la a Xwaan\\ saw \textsc{det} \textsc{fem} Mary \textsc{det} \textsc{masc} John\\ \trans `Mary saw John.' [/latex] OSV is incredibly rare. This points to a very interesting distributional fact: while all possible orders exist, they don't all exist with the same frequency. SOV is the predominant word-order cross-linguistically, accounting for about 40% the world's languages. Next is SVO, accounting for about another 35%. The other categories account for the remaining 25%.
Interestingly, languages tend to be largely consistent in their head and specifier settings. That is, English is generally head-initial and specifier-initial. Kashaya is generally head-final and specifier-initial. Same with Malagasy and Hixkaryana.
It was proposed that the theory of syntax, coupled with a universal set of parameters can account for the entirety of the world's languages. That is, suppose that there is a finite number of "switches," and each language has a unique cumulative setting for each switch. This was (and still is) a popular idea. Even if it's not entirely correct, it's still a very handy way to think about language variation.

[embed]https://youtu.be/u2iza_JDWZY[/embed] Parametric Variation (length: 2m 39s) Credit: Sydney Pritchard

Recap and a look forward
This ends the first part of this book. We've built a theory of syntax: X-bar theory. We did so by looking at data, and thinking about how that data tells us. In the end, we decided that the X-bar schema was an efficient way to capture observed patterns in syntactic structures. It is worth reflecting again on what the benefit of the X-bar schema is. It is no longer necessary for us (as researchers) to try to figure out the "VP" rule, or the "NP" rule, because in fact, they're the same rule! It's just the X-bar tree. Even more, consider the task of language acquisition. Every baby is confronted with a huge challenge: acquiring the intricacies of language. That challenge is greatly simplified if the baby intrinsically knows that everything they're hearing conforms to X-bar syntax. The baby doesn't have to learn about all the different things that can go in VP, nor all the different things that can go in NP. For the most part, the baby doesn't have to figure out the constituency of the words they are hearing, because the constituency is essentially dictated by the X-bar schema (in most cases). This greatly simplifies what a baby needs to learn to acquire a language --- any language. All they need to figure out (besides the words) are the parameters. Is the language head-initial or head-final? Is the language specifier-initial or specifier-final? From here on out, we're going to stick with the X-bar schema; we won't make any more changes to the structure that we use. Going forward, we're going to start talking about manipulating structure. What happens when things start to move around?

Key Takeaways

Intermediate levels are found across all phrases.

The X-bar schema is empirically motivated by looking at constituency

The X-bar schema defines four positions: head, complement, specifier, adjunct.

Complements and adjuncts can be empirically distinguished through three tests.

Advanced
The X-bar schema only permits maximally binary branching trees. That is, according to the X-bar schema, a node can have at most two daughters. It can have fewer than two daughters, i.e., one or no daughters. But it cannot have three daughters. In light of this constraint, we should revisit two of our phrase structure rules which seem to require a node to have three daughters. Recall our rules for coordination:

XP --> XP conj XP

X' --> X' conj X'

X --> X conj X

It's not immediately clear how to map these rules into X'-syntax. Is there any way to demonstrate constituency that is consistent with binary branching structures? It turns out that there are a few tests that suggest an "asymmetric" constituent structure. Consider the following minimal, which involve (apparent) rightward movement of part of a coordinated element. [latex] \setcounter{ExNo}{26} \ex. \a. Patrick ate an apple yesterday, and also a banana. \b. *Patrick ate an apple and, yesterday, also a banana. [/latex] This data show that the conjunction and groups together with the second conjunct, not the first. Additional data comes from what is called Left-Branch Extraction. The basic observation is that sometimes you can move (something inside of) the leftmost part of a coordinate, but nothing in the rightmost part. An illustrative example comes from Japanese.[footnote]Bošković, Željko. "On the coordinate structure constraint, across-the-board-movement, phases, and labeling." Recent developments in phase theory (2020): 133-182.[/footnote] (The underlines indicate where the first word has moved from.) [latex] \setcounter{ExNo}{27} \ex. \ag. ringo-o Taro-wa [{\underline{\hspace{10pt}}} san ko] to [banana-o ni hon] tabeta\\ apple-\textsc{acc} Taro-\textsc{top} {} three \textsc{cl} and banana-\textsc{acc} two \textsc{cl} ate\\ \trans `Taro ate three apples and two bananas.' \bg. *banana-o Taro-wa [ringo-o san ko] to [{\underline{\hspace{10pt}}} ni hon] tabeta\\ apple-\textsc{acc} Taro-\textsc{top} apple-\textsc{acc} three \textsc{cl} and {} two \textsc{cl} ate\\ \trans `Taro ate three apples and two bananas.' [/latex] While an analysis of Left Branch Extraction is beyond the scope of this class, what we probably should conclude is that in coordination, the head of the coordination, i.e., and forms a constituent with the two element, to the exclusion of the first. It's possible therefore to translate our phrase structure rules for coordination above into the following X' compliant trees. [latex] \begin{tikzpicture}[mytree] \node(top){ConjP} child {node {XP}} child {node {Conj$'$} child {node {conj \\ \textit{and}}} child {node {XP}} }; \end{tikzpicture} [/latex] This tree treats the first coordinate as a specifier, and the second as a complement. And crucially, it groups the head of ConjP together with the second coordinate.

Head-movement

juliedoner — Wed, 15 Nov 2023 05:29:37 +0000

Learning Objectives

By the end of this chapter, you should be able to,

diagnose movement of T and V using adverb placement,

represent movement of heads in trees

recognize and represent do-support in English trees

Morphological Pieces in syntax
Before turning to the syntactic process of head-movement, we have to lay some groundwork. Consider a word like horses. English speakers will recognize that this word is made up of two pieces: the noun horse and the [pb_glossary id="660"]plural[/pb_glossary] suffix -s.   We call each piece of a word a [pb_glossary id="661"]morpheme[/pb_glossary]. Morphemes are the smallest units of sound-meaning correspondence. The morpheme horse cannot be divided up into any smaller unit. Hor doesn't mean anything; orse doesn't mean anything. Similarly, there's no way to further sub-divide the suffix -s into anything smaller. [pb_glossary id="612"]Morphology[/pb_glossary] is the study of word-formation. What are the (perhaps language-specific) rules for putting morphemes together to form words? How do languages "build words?" There is a long debate about "where" word-building happens. Many scholars think that there is an independent, autonomous part of the grammar that governs how words are built. Once you've built a word, you can put it into your syntactic representation. On the other hand, many scholars believe that word building is just another component of syntax. You don't build words before you put them into the structure, you build words in the tree. Why would people think this? Well, one pretty strong piece of evidence that syntax and morphology are intrinsically connected is that morphology often reflects syntactic properties. For instance, as we will discuss at length below, Tense information is very often found bundled together with the verb; in English, the past tense walked contains two morphemes, walk and -ed. We know though that English also has a Tense Phrase. In the sentence Tamisha walked to the store, the head T will bear the feature [past] but that information is realized (i.e., pronounced) as the suffix -ed on the verb. Thus, there is, at least at some level, a correspondence between the head T and the head V. For researchers that believe that word-building is done in the syntax, the feature [past] in English is pronounced (at least sometimes) as -ed. And moreover, there must be a way to get T and V together---that is, there must be a way to move either T or V---so that they form a word. This is the view that we will adopt. Unfortunately, in this class we cannot delve deeply into the very interesting (and surprisingly contentious) debate concerning how words are built. If you're interested, I recommend taking LING 527.
Tense Morphology
Let's revisit an issue we put aside from chapter 5. There, we noted that, even though there is evidence for a Tense Phrase, sometimes, the expression of Tense is as part of the verb. In Swahili, we noted that tense is a verbal prefix.

Past tense /li-/ a-li-anguka, 'she/he fell'

Present tense /na-/ a-na-anguka, 'she/he falls'

Future tense /ta-/ a-ta-anguka, 'she/he will fall'

Marking tense on the verb is robustly attested across languages. [latex] \begin{flushleft} Kannada (Dravidian) \end{flushleft} \exg. s'aNkaraswami nenne cenna:gi ha:d-id-a\\ Shankaraswamy yesterday well \textbf{sing-\textsc{past}-3\textsc{sm}}\\ \trans `Shankaraswamy sang well yesterday.' \hfill (Sridhar, 1990:220) Kurdish (Indo-Iranian) \exg. Azad-i: Perwin ma:chi: kir\\ Azad-\textsc{erg} Perwin kiss do.\textsc{past}\\ \trans `Azad kissed Perwin.' \hfill (Sarmad Majeed, \textit{pres comm}) [/latex] As we noted above, if T and V are syntactically introduced in different locations, then there must a way to move one to the other so that they're pronounced as a single word. Using an English example, suppose that the expression of past tense (the feature [past]) in English is -ed. Then it appears under the head T, and the verb walk appears under the head V. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$past$]$ \\ -ed}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] So do we move V up to T? Or do we move T down to V? It turns out that languages make different choices. We can diagnose the difference by using adverb placement. Consider first the fact that some adverbs can be placed in-between T and V: [latex] \setcounter{ExNo}{3} \ex. Sarah will often walk home. [/latex] (4) tells us that that when T and V are separate words, then an adverb can be placed between them. This must correspond to VP-adjunction on the left. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ will}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] We can use this information---the fact that some adverbs can left-adjoin to VP to diagnose how T and V get together. In English, we find that when T is a suffix, the V+T word appears to the right of the adverb. [latex] \setcounter{ExNo}{4} \ex. \a. Sarah often walked home. \b. *Sarah walked often home. [/latex] This tells us that T must move down to the V. Otherwise, how would the T information (-ed) get on the other side of the adverb? In contrast, compare what happens with adverb placement in French. Despite the fact that English and French are both canonical [pb_glossary id="690"]SVO[/pb_glossary] languages, only English allows you to put adverbs in between the subject and the verb. In French, you have to put the adverb between the verb and object. [latex] \setcounter{ExNo}{5} \begin{flushleft} French \end{flushleft} \ex. \ag. *Sarah souvent mange des fruits\\ Sarah often eat.\textsc{pres} \textsc{indef} fruit\\ \trans Intended: `Sarah often eats fruit.'\\\bg. Sarah mange souvent des fruits\\ Sarah eat.\textsc{pres} often \textsc{indef} fruit\\ \trans `Sarah often eats fruit.' [/latex] In French, the inflected verb appears to the left of the VP-adjoined adverb. It must therefore be the case that in French, the verb moves up to T. This test allows us to infer movement. Given that we can independently diagnose the location of Tense by where will occurs, we can then consider what happens when we syntactically "separate" T and V. When T and V must form a word, where does that word appear with respect to the intervening material? It appears that some languages are Tense-lowering languages, and some languages are verb-raising languages.
Tense-Lowering and Verb-Raising
Again, the placement of the adverb tells us something about how V and T get together. In particular, because adverbs appear in front of the word V+T in English, we must conclude that T moves down to V in English. How else could we get the adverb in front of Tense? [latex] \setcounter{ExNo}{5} \ex. Sarah often eats fruit. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V+T \\ walk + $[$past$]$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (T.south)--++(south:15em)-|(V.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] In contrast, because adverbs appear after the verb in French, we must conclude that V moves up to T in French. Again, how else could we get the adverb to appear after the verb----between the verb and object? [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ mange + $[$pres$]$ \\ `eats'}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ souvent \\ `often'}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {$t_{V}$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {des fruits \\ `fruits'}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (V.south)--++(south:6em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This kind of movement is called head-movement, because it involves movement of heads. In these trees, we can represent movement with arrows. The arrows in the trees have no theoretical significance. They are purely so that we can read the trees easier. It is perfectly acceptable to write the trees without the arrows. This is because the traces in the trees tell you what has moved, and where. In the English tree, I can see that the head T is not present in TP in the tree above, instead the trace t_T is there. The head T can be found in VP. Likewise, when V raises in French, the trace t_Vtakes the place of the head of VP. Traces are theoretically significant. They are absolutely crucial in our representations. If we left out traces, then there would be no head of TP or VP when T or V moved away. Keep in mind that T-lowering and V-raising are "necessity" movements. The problem with leaving T in TP in a past tense sentence in English is that the [pb_glossary id="634"]feature[/pb_glossary] [past] simply isn't a word on its own. It's a morpheme that can only be pronounced when it's attached to another word. So [past] must move somewhere, because it isn't possible to pronounce it on its own. The problem of getting T and V together goes away when T is just a word, not an affix. Thus, in English, future tense is expressed with the word will. There is no movement of T to V, and hence the adverb is pronounced in between the two, precisely where we expect it to be! [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ will}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ walk}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {home}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] In contrast, in French, future tense is still a suffix, and V still raises to T, as the adverb placement tells us. [latex] \setcounter{ExNo}{6} \ex. \ag. Sarah mang-er-a souvent des fruits\\ Sarah eat-\textsc{fut}-\textsc{3sg} often \textsc{indef} fruits\\ \trans `Sarah will often eat fruit.' \bg. *Sarah souvent mang-er-a des fruits\\ Sarah often eat-\textsc{fut}-\textsc{3sg} \textsc{indef} fruits\\ \trans [Intended: `Sarah will often eat fruit.'] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ mange + $[$future$]$ \\ `will eat'}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ souvant \\ `often'}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {$t_{V}$}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {des fruits}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (V.south)--++(south:6em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex]

Key Takeaways

To create morphologically complex words from syntactic heads, movement must take place.

In English T lowers to V; in French V raises to T

Note that the English way of doing things is much rarer cross-linguistically than the French way. As a rule, when V and T form a word together, it's because V raises to T. English is exceptional in this regard.
Tense-to-Complementizer movement
Another clear instance of head-movement occurs in questions in English. [latex] \setcounter{ExNo}{7} \ex. \a. Will Sarah eat fruit? \b. Should Becky drink milk? \c. Could the Celtics win? [/latex] Again, to get this word order, something needs to move, because normally the subject appears in front of T. In this case, T moves to the "question C," represented as C_ques. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C$_{ques}$ + T \\ $\emptyset$ + will}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } }; \draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This sort of movement is called T-to-C movement, or descriptively, "subject-auxiliary inversion." This sort of movement is also head-movement because it again seems to only affect words, not phrases. The head T displaces to a pre-sentential position. T-to-C movement can also be thought of an a "necessity" movement. We can frame this as saying that C_ques is an affix that needs to form a word with something. In English, it does so by "pulling" T to itself, like T does to V in French.
[embed]https://youtu.be/KQhBNhpS8ak[/embed] Head-movement (length: 1m 12s)

Do-support
In English, things get tricky when there are conflicting requirements. For instance, we saw that T can lower to V, and it can also raise to C. So what happens when those two movements come into conflict? What happens when there's a question, but T doesn't contain a full word, only a feature like [pres]? Let us consider the following example. [latex] \setcounter{ExNo}{8} \ex. Does Sarah eat fruit? [/latex] Something very particular happens here: in English, we insert a vacuous element do. What does do mean in this sentence? Nothing. Sarah isn't "doing eating." (For instance, we get the same do with non-action verbs: Does Sarah own a house?) Do is only there to fix a problem, namely, there is no way to pronounce the string of morphemes created when T moves to C. Crucially, C doesn't have a phonological form, and T is just a feature [pres], or at best a [pb_glossary id="684"]suffix[/pb_glossary], -s, that can only be realized on another word (as we have seen with T-to-V movement in English). [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C+T \\ $\emptyset$+[pres]}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } }; \draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] Because do doesn't mean anything, we will assume that it is merely a "trick" of phonology, or more specifically, it is a repair strategy. It repairs something that would be ineffable (meaning "can't be said"). When do-support is needed, we will therefore leave it out of the tree. There's another place where do-support applies in English: negation. [latex] \setcounter{ExNo}{9} \ex. \a. Sarah does not eat fruit. \b. *Sarah eats not fruit. \c. *Sarah not eats fruit. [/latex] Just like the adverb often, the negative element not appears between T and V: [latex] \setcounter{ExNo}{10} \ex. Sarah will not eat fruit. [/latex] But unlike adverbs, not appears to block T lowering to V. So just like when T moves to C, a feature like [pres] cannot be pronounced without some "help." We insert do to repair the situation. Again, we will not represent the word do in our tree, because we are assuming that do-support is a "post-syntactic" operation; it happens after syntax. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$pres$]$}} child {node {NegP} child {node {Neg$'$} child {node {Neg \\ not}} child {node {VP} child {node {V$'$} [sibling distance=6em] child {node {AdvP} child {node {Adv$'$} child {node {Adv \\ often}} } } child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eat}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } }; %\draw[->, dashed] (T.south)--++(south:1em)-|(C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] But why does an intervening Neg head block T to V lowering? This is explained by appealing to a general constraint on head-movement called, fittingly, the Head-Movement Constraint or HMC. Below is the "informal" version of this constraint. We'll be able to state a "formal" version once we have covered [pb_glossary id="638"]c-command[/pb_glossary] (introduced in chapter 9).
The Head-Movement Constraint (informal version) Head movement cannot "skip over" heads.

What this constraint says is that, when moving one head to another head, you can't pass an intervening head. So the thing that goes wrong in the tree above is that T can't get past Neg.

Note that the Head-Movement Constraint is formulated in such a way that intervening phrases are not a problem. So heads can skip over any and all phrases that are in-between two positions. That's why it can skip over adverbs like often. AdvP comes between T and V, but head-movement simply doesn't "care" about phrases, it only cares about heads.

Auxiliaries
English head-movement is particularly complex. We've seen above that, when necessary and possible, T lowers to V. This is an over-simplification. There are two instances where V raises to T in English: the auxiliary verbs BE and HAVE. Consider what happens when the auxiliaries are present in a question. [latex] \setcounter{ExNo}{11} \ex. \a. Is Sarah eating fruit? \b. *Does Sarah be eating fruit? \ex. \a. Has Sarah eaten fruit? \b. *Does Sarah have eaten fruit? [/latex] With auxiliary verbs, do-support doesn't apply. This makes sense since we've been treating auxiliary verbs like is and have as T-elements. So in a question, we expect that they'll move to C. However, BE and HAVE are clearly verbs---they are of category V. They bear all the morphological indicators of verbs (inflect for person, number, and tense) and they distribute like verbs (appear in places where verbs appear). We therefore conclude that sentences with auxiliaries verbs have two distinct VPs. One VP is headed by the main verb, eaten, and the other is headed by the auxiliary. We moreover conclude that auxiliary verbs raise to T. [latex] \setcounter{ExNo}{13} \ex. Sarah is eating fruit. \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eating}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } }; \draw[->, dashed] (aux.south)--++(south:1em)-|(T.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \setcounter{ExNo}{14} \ex. Is Sarah eating fruit? \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} [sibling distance=7em] child {node(C) {C+V+T \\ $\emptyset$+\textsc{be}+$[$pres$]$}} child {node {TP} [sibling distance=6em] child {node {DP} [sibling distance=1em, level distance=2em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {VP} child {node {V$'$} [sibling distance=5em] child {node(V) {V \\ eating}} child {node {DP} [sibling distance=1em] child {coordinate(left2)} child {node {fruit}edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } }; \draw[->, dashed] (aux.south)--++(south:1em)-|(T.south); \draw[->, dashed] (T.west) -| (C.south); \draw (left1)--(right1); \draw (left2)--(right2); \end{tikzpicture} [/latex] This directly predicts that auxiliaries should appear in front of adverbs like often. [latex] \setcounter{ExNo}{15} \ex. \a. Sarah is often eating fruit. \b. Sarah has often eaten fruit. [/latex]
In this class, we aren't going to deal with the various forms of the main verb. Why the [pb_glossary id="691"] progressive [/pb_glossary] aspect involves suffixing the main verb with -ing in, and why the [pb_glossary id="692"] perfect[/pb_glossary] aspect involves suffixing the main verb with -en are questions that we will not try to address in this class. We will simply assume that the morphological form of the verb that is used in any particular aspect is stipulated. Though note that we could easily arrange for eating to be the result of two distinct heads (-ing and eat) coming together through head-movement. However, in order to keep our trees as small as possible, we won't adopt this idea here.
One of the trickiest problems with English is the fact that auxiliary verbs appear to be able to ignore the [pb_glossary id="693"]head-movement constraint[/pb_glossary]. V-to-T movement is not blocked in the presence of negation. [latex] \setcounter{ExNo}{16} \ex. \a. Sarah is not eating fruit. \b. Sarah has not eaten fruit. [/latex] I know of no (good) explanation for why the HMC doesn't apply here. But that doesn't mean that the HMC isn't a valid generalization. Its effects are robustly attested across languages.
Non-auxiliary be and have
As we just saw auxiliary be and have require move to T. What about when be and have are the only verbs in the sentence? [latex] \setcounter{ExNo}{17} \ex. \a. Sarah is tired. \b. Sarah has a cold. [/latex] Does V still move to T in these sentences? Or, because they're main verbs, maybe T moves to V. This is an empirical question, meaning that we simply have to look at the relevant data. That data is adverb placement, questions, and negation. [latex] \setcounter{ExNo}{18} \ex. \a. Sarah is often tired. \b. Is Sarah tired? \c. Sarah is not tired. \ex. \a. Sarah often has a cold\\(cf, *Sarah has often a cold.) \b. Does Sarah have cold?\\(cf, *Has Sarah a cold?) \c. Sarah does not have a cold\\(cf, *Sarah has not a cold.) [/latex] Looking at the data, we see that be as a main verb patterns just like be as an auxiliary. An adverb can come between be and its complement (AdjP), be raises to C in questions, and be appears in front of not in negative sentences. This behavior makes sense if be always raises to T, whether it's an auxiliary or not. In contrast, when have is used as a main verb, it's behavior is different from have as an auxiliary. Adverbs must appear in front of have, and do-support applies in questions and negative sentences. This behavior makes sense if, with non-auxiliary have, T lowers to V. That is, main-verb have looks just like any other "real" verb, eat, walk, buy, smile, ... [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw[->, dashed] (aux.south)--++(south:.5em)-|(T.south); \draw (left1)--(right1); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} % [sibling distance=7em] child {node(C) {C \\ $\emptyset$}} child {node {TP} % [sibling distance=6em] child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(aux) {V+T \\ \textsc{have}+$[$pres$]$}} child {node {DP} child {node {D$'$} child {node {D \\ a}} child {node {NP} child {node {N$'$} child {node {N \\ cold}} } } } } } } } } }; \draw[->, dashed] (T.south) |- (aux); \draw (left1)--(right1); \end{tikzpicture} [/latex]

Summary of head-movement

In most languages in which tense and the verb are pronounced as one word, V moves to T.

In English,

For main verbs (including have), T lower to V, unless its blocked by negation or T moves to C.

For auxiliary verbs, V raises to T in all cases.

For main verb be, V raises to T in all cases.

The facts of English head-movement are particularly complicated. It is probably best if you just memorize these facts at the beginning. Keep in mind though that you can always reproduce the facts with the data above.

Where we're going: Levels in Syntax
Drawing a tree with movement requires two steps.

Draw the tree without movement first.

Re-draw the tree with movement. Do not erase the first tree.

Thus, to draw the tree for the sentences we've been looking at, it is helpful to draw to distinct trees. [latex] {\underline{\textbf{Before movement (D-structure)}}} \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {T \\ $[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {V \\ \textsc{be}}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw (right1)--(left1); \end{tikzpicture} [/latex] [latex] {\underline{\textbf{After movement (S-structure)}}} \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=1em] child {coordinate(left1)} child {node {Sarah}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {V+T \\ \textsc{be}+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(aux) {$t_{V}$}} child {node {AdjP} child {node {Adj$'$} child {node {Adj \\ tired}} } } } } } } }; \draw[->, dashed] (aux.south)--++(south:.5em)-|(T.south); \draw (right1)--(left1); \end{tikzpicture} [/latex] There are two "shortcuts" that you might try to do when first drawing trees, and they will invariably get you into trouble. The first shortcut is that you'll want to draw just one tree, putting the movement in as you go. This is a bad way to do it, because sometimes---in fact, often----there is more than one movement. For instance, questions that involve the verb be require movement of V to T, and then T to C. (See tree in above). The second way you might get into trouble is if you start erasing. Here again, you're bound to get confused once multiple movements start happening. Actually, drawing two trees serves a theoretical purpose as well. Once we've introduced the idea of movement, we can think of our syntactic "object" as existing in levels. There's one level which exists "pre-movement." We call this level the Deep-Structure or D-structure for short. Then there's the level after movement. We call this level the Surface-Structure, or S-structure for short. These levels have theoretical significance. Looking forward, we'll find that certain constraints on syntax make reference only to the D-structure, while certain other constraints are sensitive to S-structure.
Advanced

Agglutination
English is a morphologically "poor" language: it tends to use separate words, rather than affixes, to express new meanings. But many languages are more expressive, allowing morphemes to "pile up" on words. The assumption we make is that each new morpheme on a word gets there through Head Movement. Thus, when we see a morphologically complex word with a lot of affixes, this word was "built" in the syntax by iterated instances of Head Movement. Consider a morphologically complex word from the Bantu language Logoori. [latex] \exg. gu-duy-ek-an-i\\ \textsc{sm}-hit-\textsc{ac}-\textsc{EvPl}-\textsc{past}\\ \trans `It was hit repeatedly.' [/latex] The verb guduyekani consists of five different morphemes. There's the root -duy- and then the affixes gu-, -ek, -an, and -i. In the view of syntax that we're adopting in this class, each of these morphemes is added onto the root -duy as is makes its way up the tree. This is called agglutination. That is, each morpheme is the realization of a head in the tree which -duy- moves to. For instance, the suffix -an is an Event Pluralizer, and provides the meaning "repeatedly." This head, call it EvPl, gets attached to -duy- when the verb moves to that phrase. The suffix -ek is an Anticausative, and detransitivizes the verb: the Patient becomes the subject. The prefix gu- and suffix -i are both considered functional morphemes. We typically assume that "Subject Marker" (SM), is associated with the head T, because subject marking in general varies with specifications of T. (For instance, in English, Spanish, French, etc, nonfinite verbs do not agree with subjects; only finite verbs do.) And we (that is, people who study Bantu languages), typically assume that the final vowel -i is associated with the head C. The D-structure and S-structure trees are exemplified below. Note that we're ignoring where the subject is for now. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node(C) {C \\ \textit{-i}}} child {node {TP} child {node {T$'$} child {node(T) {T \\ \textit{gu-}}} child {node {EvPlP} child {node(Evpl) {EvPl$'$} child {node {EvPl \\ \textit{-an}}} child {node {ACP} child {node(AC) {AC$'$} child {node {AC \\ \textit{-ek}}} child {node {VP} child {node {V$'$} child {node(V) {V \\ \textit{-duy-} \\ hit}} } } } } } } } } }; \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} [sibling distance = 8em] child {node(C) {T+V+AC+EvPl+C \\ \textit{gu-duy-ek-an-i}}} child {node {TP} [sibling distance = 6em] child {node {T$'$} child {node(T) {$t_{T}$}} child {node {EvPlP} child {node {EvPl$'$} child {node(Evpl) {$t_{EvPl}$}} child {node {ACP} child {node {AC$'$} child {node(AC) {$t_{AC}$}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} } } } } } } } } }; \draw[->, dashed, rounded corners] (V.west)-|(AC.south); \draw[->, dashed, rounded corners] (AC.west)-|(Evpl.south); \draw[->, dashed, rounded corners] (Evpl.west)-|(T.south); \draw[->, dashed, rounded corners] ([yshift=-5pt]T.west)-|(C.south); \end{tikzpicture} [/latex] By allowing the head -duy- to move up the tree, it can "collect" the heads along the way, creating the word in the syntax. Note, though, whether something ends up being a prefix or affix is not syntactic. The fact that gu- gets pronounced as a prefix and -i as a suffix is just an unpredictable property of these affixes.
The Mirror Principle
While Head Movement doesn't determine what ends up being a prefix of suffix, it can determine the relative order of multiple prefixes and suffixes. For instance, in the trees above, I have placed the Anticausative head -ek lower in the tree than the Event Pluralizer -an. This is because in the word guduyekani, the suffix -ek is closer to root -duy-. Explicitly, the first head that -duy- moves to is -ek, so that gets attached to the root first. Then everything moves to -an, which will attach outside of -ek. Thus, the order of the morphemes mirrors the order of heads in the syntax.
The Mirror Principle[footnote]Baker, Mark (1985). "The mirror principle and morphosyntactic explanation". Linguistic Inquiry. 16: 373–415 Note that the Mirror Principle is often interpreted as a "rule" that languages have to follow. This is the wrong interpretation. The Mirror Principle is an observation about language. But there are certainly exceptions that appear to violate Mirror Principle orderings.[/footnote] Morphological order reflects syntactic order, and vice versa.
Intuitively, the Mirror Principle says that things that are lower in the tree are closer to the root, and vice versa: things that are closer to the root are lower in the tree. The Mirror Principle is both widely adopted and widely questioned. On the one hand, it is extremely successful in explaining the relative ordering of morphemes in morphological complex words. On the other hand, there are many exceptions to the Mirror Principle, some of which defy good explanation. Putting the empirical picture aside, it is worth noting that this topic is heavily debated. Adopting the Mirror Principle, as we do in this class, forces us to assume a tight relationship between the fields of syntax and morphology. This is a controversial assumption. For more on this debate, and alternative views, consider taking LING 527/727.

Science of Syntax import

juliedoner — Wed, 15 Nov 2023 05:41:41 +0000

3. Theories of grammar and language acquisition

juliedoner — Fri, 15 Dec 2023 17:55:02 +0000
In this chapter, we will introduce the basics of a [pb_glossary id="1213"]generative grammar[/pb_glossary] model and learn about some of the arguments for and against Chomsky's theory of Universal Grammar.

Learning objectives

By the end of this chapter, you should be able to:

Describe grammar as a model of the descriptive rules of language.

Distinguish between I-language and E-language, between [pb_glossary id="169"]competence[/pb_glossary] and [pb_glossary id="170"]performance[/pb_glossary], and between synchronic and diachronic approaches to explaining language properties.

Describe the evidence that children acquire language as a rule-based system.

Explain arguments for and against generative grammar and the theory of Universal Grammar.

Contrast computer use of language with human use.

Distinguish theoretical from empirical arguments and identify supporting evidence for different kinds of arguments.

5. Decolonizing linguistics

juliedoner — Tue, 25 Jun 2024 02:41:37 +0000
[placeholder]

7. Predicates, arguments, and clauses

juliedoner — Sat, 17 Aug 2024 03:13:39 +0000
placeholder

Selection

juliedoner — Wed, 15 Nov 2023 05:29:37 +0000

Learning Objectives

By the end of this chapter, you should,

be able to identify c-selectional and s-selectional properties of predicates,

be able to recognize and label thematic roles

understand the basis for UTAH and its implications

We've concerned ourselves so far with making abstract and broad rules for language. In this chapter, we're concerned with more specific things. In particular, what explains why certain words occur in certain environments. As we'll discover below, a word's syntactic environment (what it appears with) is highly idiosyncratic---within bounds. It's still the case that the category 'verb' has a particular syntactic distribution, as does noun, complementizer, etc. First a terminological note. We'll be discussing verbs heavily in this chapter, but sometimes I'll use the term predicate. Predicates are, loosely, what subjects are "about." This notion of predicate groups together verbs, adjectives, and even prepositional phrases. So in the terminology defined below, verbs can "select," adjectives can "select," prepositions can "select," etc.
C(ategory)-selection

Right now, our theory will generate an infinite number of sentences in any human language. But we've done very little to constrain our theory. For instance, nothing we've said so far will explain why (1) and (2) are not possible sentences in English.
[latex] \ex. \a. * Sarah ran Meagan \b. * Amy will kiss \ex. \a. \# The tree ran. \b. \# Chicago will kiss John. [/latex] The examples in (1) seem to be wrong for syntactic reasons (marked ungrammatical with "*"). The verb run doesn't take an object and kiss must take an object. On the other hand, (2) seem to be wrong for semantic reasons (marked infelicitous with "#"). The verbs run and kiss need an [pb_glossary id="677"]animate[/pb_glossary] subject. There's nothing in our theory so far that directly accounts for these patterns. That is, there is nothing built into our theory that makes sure that a verb like run appears without an object and with an animate subject.
Here's another example: Some verbs allow DP and CP complements, while others don't.
[latex] \setcounter{ExNo}{2} \ex. \a. John fears [$_{DP}$ the statue ] \b. John fears [$_{CP}$ that Bill left ] \ex. \a. John kicked [$_{DP}$ the statue ] \b. *John kicked [$_{CP}$ that Bill left ] [/latex]
There are some patterns that we've left implicit up until now, but if we want a coherent theory, we need to make everything as explicit as possible. What the above data teach us is that verbs make different requirements on what the rest of the structure can be. A verb like run dictates that there is only one other DP argument, and that argument is animate. A verb like kiss dictates that there are two other DP arguments, one of which is animate.

The idea that we'll flesh out this week is that on top of our X'-schema, there's a certain amount of information that is associated with each head, in particular, what it needs to combine with. We call this information the selectional properties of the head. [pb_glossary id="656"]Selection[/pb_glossary] comes in two "flavors."

Category selection (c-selection): The categorial requirements of a head.

Semantic selection (s-selection): The semantic requirements of a head.

For example, we say about the verb run that it c-selects for a DP subject and it s-selects for an animate subject. This means that it must have a DP as a subject, and that DP must refer to an animate individual. We say about the verbs kiss and kick that they c-select for a DP subject and a DP object. Both verbs require a DP subject and object, and in both cases, only the subject DP needs to be animate. Fear c-selects for a DP subject and either a DP or a CP object, etc.

As a rule, c-selection is concerned with the syntactic category of the arguments of a predicate, like whether the object must be a CP or a DP. C-selection is also concerned with the number of arguments that a predicate has. For instance, eat and consume have different c-selectional properties: it's possible for eat to only occur with a subject, while consume must occur with both a subject and an object.[footnote]Putting an asterisk outside of parenthesis means "ungrammaticality without this element." Putting asterisks inside of a parenthesis (*X) means, "ungrammatical with this element."[/footnote]
[latex] \setcounter{ExNo}{4} \ex. \a. Brenda ate (pizza). \b. Brenda consumed *(pizza). [/latex]
In short: each word (really [pb_glossary id="661"]morpheme[/pb_glossary]) dictates the syntax around it, both in terms of the category of arguments, and also the number of arguments.

Notice that c-selection very often (though not always) involves a head selecting its complement. A general rule of thumb is that all complements are selected.
Another rule of thumb is that all adjuncts are not c-selected. That is, c-selection dictates the properties of the head that must be met. Since adjuncts are typically optional, they are not c-selected. [latex] \setcounter{ExNo}{5} \ex. \a. April met Leah in the park. \b. *Met Leah in the park. \hfill (must have the subject!) \c. *April met in the park. \hfill (must have the object!) \d. April met Leah. \hfill (doesn't need the adjunct!) [/latex]

S(emantic)-selection

Every predicate (really, every head) dictates what it can, and cannot combine with. So a verb like run cannot combine with a DP object, but a verb like kiss must combine with both a DP subject and a DP object.

However, sometimes things go wrong even when the c-selectional criteria are met. For instance, the following sentences sound incorrect, even though all the c-selectional requirements are met.
[latex] \setcounter{ExNo}{6} \ex. \a. \# The tree ran. \b. \#The wall will punch John. [/latex]
These sentences sounds off for semantic reasons. Trees don't run and walls don't punch. This suggests that in addition to the categorical information, verbs also dictate the "semantic features" of what they combine with. For instance, run and punch both need [pb_glossary id="677"]animate[/pb_glossary] DP subjects. We call this requirement Semantic selection --- it's the semantic counterpart to c-selection.

Semantic selection (s-selection) : The semantic requirements of a head.

When we talk about s-selection with respect to verbs, we're typically (though not always) talking about thematic roles.

Thematic role (theta-role or Θ-role) : The semantic role assigned to each argument in a sentence.

Theta-roles are ways to classify how each argument is meaningfully related to the predicate. For instance, the subject of both run and punch is a "doer." It's performing an action, thus needs to be animate. We call this role Agent. So the subject of both run and punch is an Agent, or has the thematic role of Agent.

On the other hand, the object of punch isn't performing an action, it's affected by the action. We call this role Patient. So the object of punch has the thematic role of Patient. Every [pb_glossary id="703"]argument[/pb_glossary] of a verb has a thematic role.

Examples of thematic roles.

Agent : Someone who purposefully does something.

John broke the vase

Patient : Someone/thing that is affected, or changes

John broke the vase

Experiencer : Someone who feels or uses their mind

John loves Mary

Cause : Someone/thing that causes an action

The wave sank the ship

Goal : An endpoint, location

John went to Paris

Recipient : Someone who receives something

Mary gave John a book

Location : a place

John is sitting in the park

Instrument : Something that is used

John chopped the tree with an axe

Theme : The catch-all, if it doesn't fit into any other category.

John assisted the campaign

Tips in identifying theta-roles

Agents, and Experiencers must be animate (i.e., alive)

Causes and Agents are very similar. A Cause tends to lack a notion of "intentionality" or "purpose."

A Recipient generally involves a "transfer of possession" of some sort, which differentiates it from a Goal.

Theta-roles can vary depending on context.

When in doubt use Theme.

[embed]https://youtu.be/DKthafDLU7M[/embed] Thematic roles (length: 1m 53s) credit: Emma Scott
The point here is that, just like c-selection, different verbs have different requirements on what they s-select for. Some verbs need Agents, some need Patients, some need Experiencers, etc. Notice that, just like with c-selection, order matters. It's not just that punch needs to combine with an Agent and a Patient. The Agent must be the subject, and the Patient must be the object. We call this ordering of thematic roles the thematic structure of a verb.

Thematic structure : The "order" that the s-selectional requirements of the verb are expressed in.

So we say about the thematic structure of run that, "run has an Agent subject." And we say about the thematic structure of punch that, "punch has an Agent subject and a Patient object."

It's important to keep in mind that s-selection is a bit broader than just distribution of thematic roles. For instance, the following sentences sound off, but it's not because the thematic structure is wrong (and it's not because the c-selectional requirements haven't been met).
[latex] \setcounter{ExNo}{7} \ex. \a. \#The doctor gathered in the park \b. \#The soldier surrounded the castle. \c. \#Jesse multiplied the number two. [/latex] The reason that (8) sound wrong is that gather and surround need a plural subject, while multiply needs a plural object. This is, again, an s-selectional requirement of these verbs, not a c-selectional requirement, because the ungrammaticality stems from a semantic "mistake," not a problem with the category of the arguments.
Uniformity of Theta assignment hypothesis

If we spent a long time looking at thematic structure, we'd come up with a pretty striking correspondence between thematic structure and syntactic structure: Particular thematics roles are correlated with particular syntactic positions in a language. For instance, Agents are almost always subjects. Patients are almost always objects. Instruments are almost always in prepositional phrases or otherwise "[pb_glossary id="694"]oblique[/pb_glossary]." This old observation led to a hypothesis: the Uniformity of Theta Assignment Hypothesis or UTAH.

Uniformity of Theta Assignment Hypothesis (UTAH)

Each theta-role always appears in the same syntactic position.

We're going to assume for this class that UTAH is basically right --- although we'll revisit the question in a few chapters. In short, we're going to assume that Agents always appear in the subject position (for now assumed to be specifier of TP). Patients always appear in the object position (complement to V). Instruments always appear in a prepositional phrase, etc.

[embed]https://youtu.be/cxnVszdv4MY[/embed] Thematic structure and UTAH (length: 2m 39s)
Finally, we say that auxiliary verbs do not select for subjects. Their only requirement is that they select for a VP complement. Why? Well, the properties of a subject are never determined by the auxiliary, but rather by lower, main verb. In the examples below, the verbs start and cry determine what is an appropriate subject. [latex] \setcounter{ExNo}{8} \ex. \a. The computer/\#the child is starting. \b. \#The computer/the child is crying. [/latex] The observation that subjects are selected by the main verb, not an auxiliary, is actually extremely important, and it's something that we'll return to in chapter 10.
Where we're going

One purpose of determining selectional properties is to discover classes of words. For instance, we can talk about the class of verbs that only select for a single DP: the intransitive verbs. This are distinguished from the class of verbs that only select for two DPs: the transitive verbs.

We can likewise talk about the class of verbs that take CP complements. It surely cannot be a coincidence that in no language does the verb kick combine with a CP, but basically every language allows say to combine with a CP. In the end, we'd like to come up with an explanation for why certain classes of verbs exist. But sometimes, this is impossible. For instance, eat and consume both mean, roughly, the same thing, but only one obligatorily c-selects for a DP object.
The second reason why it's important recognize that heads dictate the syntax around them, is that we can use that information to diagnose movement. For instance, recall the [pb_glossary id="695"]topicalization[/pb_glossary] constituency test. [latex] \setcounter{ExNo}{9} \ex. Susan, Rachel hugged. [/latex] Intuitively, we think that Susan has moved to the front of the sentence. Why? Because she's who Rachel is hugging. Susan is the object of the verb. Implicit in this intuition is selection. We know (again, perhaps intuitively) that Susan is a Patient in this sentence, because hug is the kind of verb which dictates that its object is a Patient. Or stated differed, hug is the kind of verb that needs an object (due to c-selection), and so we understand that the DP Susan fills that syntactic role. This is our first taste of phrasal movement, meaning, naturally, the movement of phrases. (As opposed to head-movement, which is the movement of heads, discussed in the previous chapter.) Because we know that heads dictate what they must and must not combine with, we conclude that the DP Susan must have combined with hug to satisfy the verb's selectional requirement, and then moved to the front of the sentence.
Advanced

Binding theory

juliedoner — Wed, 15 Nov 2023 05:29:38 +0000

Learning Objectives

By the end of this chapter, you should be able to,

understand c-command

recognize the difference between R-expressions, pronouns, and anaphors

understand their distributional differences

restate the three binding conditions (A, B, and C)

understand how the distribution of referring expressions is constrained by binding theory

In this chapter we'll change gears a little. We are now going to be concerned for the first time with what are apparently non-local dependencies. By this I mean that we are going to be looking at things that interact over a syntactic distance; they aren't next to each other in the tree. We've sort of been hinting at this (e.g., with [pb_glossary id="678"]head-movement[/pb_glossary] in chapter 7) but now we'll delve into some real cases. Besides introducing Binding Theory (which is a perennially interesting topic) in this chapter, the ultimate goal here is to introduce the notion of c-command. This is, arguably, the most important thing ever discovered in syntax, and it underlies almost everything in current theory. C-command (short for "constituent command") is a structural relation. It is defined in the following way.
C-Command A c-commands B iff (if and only if),

A does not [pb_glossary id="639"]dominate[/pb_glossary] B.

B does not dominate A.

Every node that dominates A dominates B.

In prose, X c-commands its sister and everything under its sister. The following diagram illustrates c-command, as well as the notion of domination. [latex] \begin{tikzpicture} \node(top) {XP} child {node(yp) {YP}} child {node {X$'$} child {node {X}} child {node(wp) {WP}} }; \draw[->, dashed, blue] (top)..controls +(east:2) and +(east:2)..(wp) node[midway, right]{\footnotesize{domination}}; \draw[->, dashed, blue] (yp)..controls +(south west:2) and +(south:2)..(wp) node[midway, left]{\footnotesize{c-command}}; \end{tikzpicture} \begin{minipage}{.49\textwidth} \textbf{C-command relations} \begin{itemize} \item YP c-commands X$'$ \item YP c-commands X \item YP c-commands WP \item X$'$ c-commands YP \item X c-commands WP \item WP c-commands X \end{itemize} \end{minipage} \begin{minipage}{.49\textwidth} \textbf{Domination relations} \begin{itemize} \item XP dominates YP, X$'$, X, and WP \item X$'$ dominates X and WP \item[] \item[] \end{itemize} \end{minipage} [/latex] It turns out that c-command matters whenever we talk about non-local dependencies. In this context, "non-local" just means whenever two things interact which aren't in the same phrase. Binding Theory is a very clear example of the importance of c-command, but it's influence can be found in a variety of syntactic processes.
[embed]https://youtu.be/bFCy_cHKzac[/embed] C-command (length:1m 34s) Credit: Megan Spurgeon
Things that refer

Gendered language and referring expressions

In this unit, because we are talking about individuals and their properties, we are using a number of terms that are inherently gendered. It is important to note that our use of gendered terms here is purely a linguistic fact, and does not (or should not) reflect our opinions about gender. In working through the following data, please assume that the individuals in question self-identify as the gender that is indicated. That is, assume that Marta below self-identifies as female, and so we use the pronouns 'she/her' when referring to her. If Marta identified as non-binary, and chose the pronouns 'they/them/their,' the linguistic facts about the distribution of these pronouns would not change: these pronouns---that is, all pronouns---are subject to Condition B of Binding Theory.

In German, there are three linguistic expressions for referring to an individual. Each one is underlined in the following examples. [latex] \ex. \ag. {\underline{Marta}} singt\\ Marta sing.\textsc{pres}\\ \trans `Marta sings.' \bg. {\underline{Sie}} tanzt.\\ she dance.\textsc{pres}\\ \trans `She dances.' \cg. Marta liebt {\underline{sich}} sehr.\\ Marta love.\textsc{pres} \textsc{refl} very\\ \trans `Marta loves herself very much.' [/latex] The first way we can refer is to simply use an [pb_glossary id="635"]R-expression[/pb_glossary] which is short for "referential expression." R-expressions uniquely refer. Marta is an example of an R-expression because there is a unique individual named "Marta" that I'm referring to when I say that word. The reference of Marta never changes. Whenever I use the word Marta, I refer to the individual named by that word. (It is a "rigid designator.") In contrast in (2) we see a [pb_glossary id="637"]pronoun[/pb_glossary], sie. Pronouns can also refer to individuals, but crucially, their reference can change depending context. For instance, sie in (2) likely refers to Marta, but in another context, it could refer to Elise, or Helga, etc. Thus, pronouns and R-expressions share the property of being able to refer, but they differ in that R-expressions have strict reference, while pronouns do not. Finally in (3), we see a third way to refer to an individual in German: the [pb_glossary id="636"]reflexive anaphor[/pb_glossary] sich. Again, this word refers to an individual. It is used to indicate that Marta loves Marta (not Elise, not Helga, etc). Like pronouns, sich doesn't always refer to Marta. So in the sentence, Helga liebt sich sehr, then sich refers to Helga. But unlike pronouns, sich's reference isn't totally free. While sie in (2) could, in principle be anyone (who identifies as female), sich in (3) can only be Marta. It cannot be Elise, or Helga, etc.
Summary of referential terms:

R-expressions R-expressions strictly refer, meaning that they always refers to the same individual across all contexts.

Pronouns Pronouns have variable reference, meaning that the individual that they refer to can change depending on context.

Anaphors Anaphors also have variable reference, because they do not always refer to the same individual. But unlike pronouns, they are not free to refer to anyone. Their reference is fixed in any one context.

This chapter is concerned with the distribution of referential expressions. Even though all three things above can be used to refer to individuals, they all are governed by slightly different rules---and it turns out that these are fundamentally syntactic nature. That is, we will discover that the rules governing when you can and cannot say an R-expression, a pronoun, and a reflexive anaphor are defined purely in terms of syntactic structure.
Constraint on R-Expressions: Condition C
Turning back to German R-expressions (e.g, Marta), pronouns (e.g., sie), and anaphors (e.g., sich), we noted above that all three categories have pretty much the same function: they refer. That is, they pick out an individual or thing. We 'll start with the distribution of R-expressions. Consider, [latex] \setcounter{ExNo}{1} \ex. \ag. Marta sagt, da{\ss} sie singt\\ Marta say.\textsc{pres} that she sing.\textsc{pres}\\ \trans `Marta says that she sings.' \cg. Sie sagt, da{\ss} Marta singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `She says that Marta sings.' [/latex] In these sentences, I've used two different ways to refer to Marta: an R-expression and a pronoun. In principle, both sentences should be able to mean the same thing, because, as we noted above Marta always mean "Marta," and sie can mean "Marta" as well. These sentences do not mean the same thing, though. (2a) means, "Marta says that someone who identifies as female sings." That "someone" can be anyone, including Marta. So it can mean, "Marta says that Marta sings." Indeed, that is probably the preferred meaning. (2b) on the other hand means, "Someone who identifies as female says that Marta sings." What this sentence absolutely cannot mean, though, is "Marta says that Marta sings." We have a way of representing the possible meanings, using indices. I put an index as a subscript on any referring expression. The index is a "pointer." Anything with the same index points to the same thing. So in (3a), since Marta has the subscript "i", sie can have the same subscript, i.e., point to the same individual. In (3b), the two expressions cannot be co-indexed. This is what (3c) demonstrates. The two expressions bear different indices, i.e., refer to different individuals. [latex] \setcounter{ExNo}{2} \ex. \ag. Marta$_i$ sagt, da{\ss} sie$_i$ singt\\ Marta say.\textsc{pres} that she sing.\textsc{pres}\\ \trans `Marta$_i$ says that she$_i$ sings.' \bg. *Sie$_i$ sagt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `*She$_i$ says that Marta$_i$ sings.' \cg. Sie$_j$ sagt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `She$_j$ says that Marta$_i$ sings.' [/latex]
[embed]https://youtu.be/N00FWdPu0f4[/embed] Indices (length: 1m 12s)
Here's a simple hypothesis about what's going on in German.
German Reference Hypothesis #1 (to be rejected) An R-expression cannot be preceded by a co-referential pronoun.
This hypothesis says that the linear order is the problem. It predicts that no matter how long we make the sentence, we still see a similar result. And the data in (4) appear to bear out this prediction! [latex] \setcounter{ExNo}{3} \exg. *Sie$_i$ sagt, da{\ss} Karl$_j$ denkt, da{\ss} Marta$_i$ singt\\ she say.\textsc{pres} that Karl think.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `*She$_i$ says that Karl$_j$ thinks that Marta$_i$ sings. [/latex] So Hypothesis #1 looks promising! But here are some counter-examples. [latex] \setcounter{ExNo}{4} \ex. \ag. Ihre$_i$ Schwester sagt da{\ss} Marta$_i$ singt\\ her sister say.\textsc{pres} that Marta sing.\textsc{pres}\\ \trans `Her sister says that Marta sings.' \bg. Da{\ss} sie$_i$ singt, gef\"allt Marta$_i$\\ that she sing.\textsc{pres} please.\textsc{pres} Marta\\ \trans `That she$_i$ sings pleases Marta$_i$.' [/latex] So it's clear that linear order is not enough. But if linear order isn't sufficient, then there must be something structural. Let's look at the trees for a grammatical and ungrammatical sentence. Note that just like in English, possessive pronouns like ihre 'her' are analyzed as combinations of a pronoun sie 'she' and the possessive determiner 's.[footnote]These trees slightly simplify German's syntax. German has a special property called "V2." In German, the finite verb is always the second thing in the sentence in main clauses. This is derived by V-to-C movement. In embedded clauses, however, we discover that German is actually verb-final (SOV) when the verb does not move to C. [/footnote] Ungrammatical Tree (4) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance = 2ex] child {coordinate(left)} child {node {sie}edge from parent[draw=none]} child {coordinate(right)} } child {node {T$'$} child {node(T) {T+V \\ $[$pres$]$+sagt}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {CP} child {node {C$'$} child {node {C \\ da\ss}} child {node {TP} child {node {DP} [sibling distance=3ex] child {coordinate(left1)} child {node {Maria} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T2) {T+V \\ $[$pres$]$+singt}} child {node {VP} child {node {V$'$} child {node(V2) {$t_{V}$}} } } } } } } } } } } }; \draw[->, dashed] (V2.west)--++(west:1ex)-|(T2); \draw[->, dashed] (V.west)--++(west:1ex)-|(T); \draw (left)--(right); \draw (left1)--(right1); \end{tikzpicture} [/latex] Grammatical Tree (5a) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node {DP} [sibling distance = 3em] child {node {DP} [sibling distance = 2ex] child {coordinate(left)} child {node {sie}edge from parent[draw=none]} child {coordinate(right)} } child {node {D$'$} child {node {D \\ 's}} child {node {NP} [sibling distance = 3ex] child {coordinate(left3)} child {node {Schwester}edge from parent[draw=none]} child {coordinate(right3)} } } } child {node {T$'$} [sibling distance = 5em] child {node(T) {T+V \\ $[$pres$]$+sagt}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {CP} child {node {C$'$} child {node {C \\ da\ss}} child {node {TP} child {node {DP} [sibling distance=3ex] child {coordinate(left1)} child {node {Maria} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T2) {T+V \\ $[$pres$]$+singt}} child {node {VP} child {node {V$'$} child {node(V2) {$t_{V}$}} } } } } } } } } } } }; \draw[->, dashed] (V2.west)--++(west:1ex)-|(T2); \draw[->, dashed] (V.west)--++(west:1ex)-|(T); \draw (left)--(right); \draw (left1)--(right1); \draw (left3)--(right3); \end{tikzpicture} [/latex] In these trees, we observe that in the ungrammatical case, the pronoun sie  c-commands the R-expression. In the grammatical case, the pronoun doesn't. That is, when we look at what sie c-commands in each tree, we find a clear distinction between them. Specifically, we see that when sie c-commands Maria - as in (4), where sie is a sister to T', and therefore c-commands T' and everything beneath it (including the R-expression Maria) - the pronoun cannot also refer to Maria. Conversely, we see that when sie doesn't c-command Maria - as in (5a), where sie instead c-commands D' (which crucially does not include Maria) - then the pronoun can refer to Maria. This suggests a different hypothesis:
German Reference Hypothesis #2 (correct): An R-expression cannot be c-commanded by a co-referential pronoun.
Hypothesis #2 doesn't care about the linear order, i.e., the order in which we pronounce the words. Instead Hypothesis #2 cares about the hierarchical relationship between the referring expressions. It says that an R-expression cannot be c-commanded by a pronoun if they refer to the same individual. Referring to same individual is called [pb_glossary id="640"]coreference[/pb_glossary]. We have a specific term for this relationship involving c-command and coreference: binding.
Binding: X binds Y iff [if and only if],

X c-commands Y

X and Y are co-indexed (i.e., X and Y refer to the same thing).

Hypothesis #2 appears to be empirically correct for German. If we were to look at a lot of data (and control for a lot of factors), we would find that, as a rule, R-expressions cannot be c-commanded by a co-referential pronoun. Indeed, an R-expression cannot be c-commanded by anything co-refential. This restriction on R-expressions, doesn't just hold for German. It appears to be a general property of language --- basically universal. It is called [pb_glossary id="641"]Condition C[/pb_glossary] (also Principle C) of Binding Theory.
Condition C An R-expression cannot be bound, or, An R-expression must be free. ("Free" means "not bound.")
A Condition C "violation" involves two things

The R-expression being c-commanded by a co-referential expression; and

both expressions referring to the same individual.

If both of these things happen, then Condition C is violated, and the sentence is ungrammatical. For Binding Theory, we calculate c-command from the phrase level. That is, we always calculate binding by considering the relationship between referential DPs. [latex] {\ovalbox{ \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.DP$_i$ [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] \hspace{150pt} \Tree [.DP [.{D$'$} {D \\ $\emptyset$} [.NP$_i$ [.{N$'$} {N \\ John} ] ] ] ] \node(upleft) at (-1,0) {}; \node(upright) at (1,0) {}; \node(downleft) at (-1,-5.5) {}; \node(downright) at (1,-5.5) {}; \draw[thick] (upleft) -- (downright); \draw[thick] (upright) -- (downleft); \hspace{150pt} \Tree [.DP [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N$_i$ \\ John} ] ] ] ] \node(upleft) at (-1,0) {}; \node(upright) at (1,0) {}; \node(downleft) at (-1,-5.5) {}; \node(downright) at (1,-5.5) {}; \draw[thick] (upleft) -- (downright); \draw[thick] (upright) -- (downleft); \end{tikzpicture} }} [/latex] This must be the case when we consider, say, possessives. The top DP refers to something that "encompasses" the embedded DP. The only constituent that contains both phrases is the highest DP. The phrase John's dog refers to a dog, not John. [latex] \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.DP$_i$ [.DP$_j$ [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] [.{D$'$} {D \\ 's} [.NP [.{N$'$} {N \\ dog} ] ] ] ] \end{tikzpicture} [/latex]

How to talk about binding

Below are various ways students have talked about binding and binding theory. They are all wrong. [latex] \Tree [.{\ldots} XP$_i$ [.{\ldots} {\ldots} YP$_i$ ] ] [/latex]

"YP is bound to/with XP" → should be "YP is bound by XP''

"XP and YP are bound together." → should be "XP binds YP."

"YP is coindexed." → should be "YP is coindexed with XP."

"Y is binded/bounded" → should be "Y is bound."

"XP and YP c-command" → should be "XP c-commands YP."

"YP is bound by a c-commanding XP" is redundant. Being bound entails being c-commanded.

Things you can say about this binding relationship.

"XP binds YP."

"YP is bound by XP."

"XP is a binder for YP, is YP's binder."

"XP c-commands YP."

"XP and YP are coindexed," or "XP and YP have the same index."

Before turning in assignments, you should check to make sure you're not making one of these errors.

Constraint on Anaphors: Condition A

Condition C tells us about the distribution of R-expressions. We now turn to anaphors, like German sich. Here are some examples. Pay careful attention to the indices.
[latex] \setcounter{ExNo}{5} \ex. \ag. Marta$_i$ liebt sich$_i$\\ Marta love.\textsc{pres} \textsc{refl}\\ \trans `Marta$_i$ loves herself$_i$.' %\bg. *Marta$_i$ sagt da{\ss} Helga$_j$ sich$_i$ liebt \\ %Marta say.\textsc{pres} that Helga \textsc{refl} love.\textsc{pres}\\ %\trans `Marta$_i$ says that Helga loves herself$_i$.' \cg. *Martas$_i$ Bruder liebt sich$_i$\\ Marta.\textsc{gen} brother love.\textsc{pres} \textsc{refl}\\ \trans `*Marta$_i$'s brother loves herself$_i$.' [/latex]
Using the same reasoning as above, we might come to the conclusion that anaphora are governed by rules that are precisely the opposite of R-expressions.

German Anaphor Hypothesis #1: (to be revised) An anaphor must be bound.

That is, unlike an R-expression, an anaphor must be c-commanded by a co-referring expression. This looks pretty good, but it isn't quite enough. Consider the following sentence. Why does it appear to falsify the hypothesis above?
[latex] \setcounter{ExNo}{6} \exg. *Marta$_i$ sagt da{\ss} Helga$_j$ sich$_i$ liebt\\ Marta say.\textsc{pres} that Helga \textsc{refl} love.\textsc{pres}\\ \trans `Marta$_i$ says that Helga loves herself$_i$.' [/latex] The problem here is that sich is bound! Marta is coindexed with anaphor, and Marta definitely c-commands the anaphor, since Marta is a the subject of the entire sentence. It's possible to keep the idea that anaphors must be bound, but we have to add another clause in our hypothesis. The generalization is that anaphors have to be bound "close enough". That is, there is some minimal syntactic unit inside of which an anaphor must be bound. We call this minimal syntactic unit the [pb_glossary id="642"]binding domain[/pb_glossary]. For the sake of this class, we'll say that its the closest TP that dominates the anaphor.
The constraint on anaphors is Condition A of the Binding Theory.

Condition A

An anaphor must be bound in its binding domain (where the binding domain is the closest dominating TP).

For instance, consider the first sentence again,
[latex] \setcounter{ExNo}{7} \exg. Marta liebt sich sehr\\ Marta love.\textsc{pres} \textsc{refl} very\\ \trans `Marta loves herself very much.' [/latex] Here the binding domain for sich is the highest TP---which happens to be the only TP. And because Marta is also inside of this TP, sich is bound, satisfying Condition A. In (7) on the other hand, the binding domain for sich is the lowest TP. Since Marta is not in this TP, sich is free (unbound) and so violates Condition A. If there are two anaphors in a sentence, they might each have their own binding domain. [latex] \setcounter{ExNo}{8} \exg. Marta$_i$ erkl\"art sich$_i$, da{\ss} Helga$_j$ sich$_j$ liebt \\ Marta explain.\textsc{pres} \textsc{refl} that Helga \textsc{refl} love.\textsc{pres}\\ \trans `Marta$_i$ explains to herself$_i$ that Helga$_j$ loves herself$_j$.' [/latex] Again, Condition A of Binding Theory is universal---although there are some additional complications. First, there may be further constraints on an anaphor in a language. For instance, in English, anaphors must match in number and gender. Assuming that Sue identifies as female, it is not possible to say *Sue loves himself. This not because the anaphor isn't bound, but because the anaphor doesn't match the gender. An additional complication is that the things that look like anaphors in a language often have other (perhaps related) uses. For instance, reflexive pronouns in English are often used colloquially for emphasis. [latex] \setcounter{ExNo}{9} \ex. Gary likes to put his feet up after a long day of work. As for myself, I like to hit the gym. [/latex]
Constraint on pronouns: Condition B

Finally, we can look at pronouns. These are actually the most complex. In some ways they're like R-expressions. Consider the following.
[latex] \setcounter{ExNo}{10} \ex. \ag. *Marta$_i$ liebt sie$_i$\\ Marta love.\textsc{pres} her\\ \trans `*Marta$_i$ loves her$_i$.' \bg. Martas$_i$ Bruder liebt sie$_i$\\ Marta.\textsc{gen} brother love.\textsc{pres} her\\ \trans `Marta$_i$'s brother loves her$_i$.' [/latex] This is similar to Condition C because it looks like the pronoun cannot be bound by a co-referring expression. In (10a), Marta c-commands sie, and they're co-indexed, so the pronoun is bound. The result is ungrammatical. In contrast, in (10b), the pronoun is free, and the result is grammatical. Based on this data, we therefore posit the following hypothesis:

German Pronoun Hypothesis #1 (to be rejected):

Pronouns are subject to Condition C.

However, if pronouns were just like R-expressions, we would expect the sentence in (11) to be ungrammatical. (For comparison, you can reverse the pronoun and the R-expression, in which case it's a Condition C violation.)
[latex] \setcounter{ExNo}{11} \exg. Marta$_i$ sagt da{\ss} Helga sie liebt\\ Marta say.\textsc{pres} that Helga her$_i$ love.\textsc{pres}\\ \trans `Marta$_i$ says that Helga loves her$_i$.' [/latex] In fact, pronouns are also a bit like Condition A, in that they make reference to a binding domain (again assumed to be the closest TP). Here's Condition B of Binding Theory.

Condition B

A pronoun cannot be bound in its binding domain,

or,

A pronoun must be [pb_glossary id="679"]free[/pb_glossary] in its binding domain.

Summary of Binding Theory

Condition A: An anaphor must be bound in its binding domain.

Condition B: A pronoun must be free (=cannot be bound) in its binding domain.

Condition C: An R-expression must be free (=cannot be bound).

We will assume for the purpose of this class that the binding domain for all referential expressions is the closest TP.

A helpful (?) mnemonic:

Condition A covers anaphors.

Condition B covers bronouns.

Condition C is for the other thing: R-expressions.

Using Binding Theory

Because Binding Theory is assumed to be universal, that is, we can detect Conditions A, B, and C, in every language, is extremely helpful in diagnosing structure.

For instance, consider the case of [pb_glossary id="643"]pro-drop[/pb_glossary] languages. These are languages in which a pronominal subject and/or object does not need to be explicitly expressed. Swahili is pro-drop language.
[latex] \setcounter{ExNo}{12} \ex. \ag. a-li-cheka\\ 1\textsc{sm}-\textsc{past}-laugh\\ \trans `She/he laughed.' \bg. a-li-ona mti mkubwa\\ 1\textsc{sm}-\textsc{past}-see tree big\\ \trans `She/he saw a big tree.' [/latex] It's possible to think of pro-drop languages in two different ways.

Hypothesis #1: Pro-drop languages don't have pronouns. That is, there just isn't anything in the structure to represent he, she, they, etc.

Hypothesis #2: Pro-drop languages have silent pronouns. That is, there's a DP pronoun that doesn't have phonological content.

The two hypotheses are actually widely debated, with many different variations on these hypotheses. The question you can consider, What predictions does each hypothesis make with respect to Binding Theory? For instance, consider the following sentence and its English translation. (Note that Juma is typically a name for a male individual.) [latex] \setcounter{ExNo}{13} \exg. a-li-sema kwamba Juma a-li-cheka\\ 1\textsc{sm}-\textsc{past}-say that Juam 1\textsc{sm}-\textsc{past}-laugh\\ \trans `She$_j$/he$_j$ said that Juma$_i$ laughed.'\\ CANNOT MEAN: `He$_i$ said that Juma$_i$ laughed.' [/latex] The meaning that is attributed to the sentence in (13) is consistent with only one of the hypotheses above concerning pro-drop languages. Hypothesis 2 predicts that the meaning He_i said that Juma_i laughed does not exist because there is a silent pronoun binding the R-expression, thereby violating Condition C. Hypothesis 1 does not predict the unavailability of this meaning. A different explanation would have to be developed. So Binding Theory can be used to determine structure. This is true even for well-studied languages like English. Consider double object constructions. [latex] \setcounter{ExNo}{14} \ex. Mary showed Trisha a painting. [/latex] In such example, it looks like the verb show is selecting for two objects, Trisha and a painting. This suggests that double object construction should have a "flat" structure, directly contradicting our X'-schema. [latex] \begin{tikzpicture}[mytree] \node(top) {VP} child {node {V$'$} child {node {V \\ show}} child {node {DP \\ Trisha}} child {node {DP \\ a painting}} }; \end{tikzpicture} [/latex] This tree makes very clear predictions though. It says that Trisha c-commands a painting and that a painting c-commands Trisha. They mutually c-command each other. The following data suggest however that this cannot be correct. [latex] \setcounter{ExNo}{15} \ex. \a. *Mary showed her\textsubscript{i} a painting of Susan\textsubscript{i} \b. Mary showed Susan\textsubscript{i}'s friend her\textsubscript{i} (in the photograph) [/latex] This data suggests that the indirect object, the first object, c-commands the direct object, the second object, but not vice versa. We'll return to how we can make this syntactically work later. For now, the takeaway is that we can use Binding Theory to determine what the syntax looks like.

Things to remember

The difference between R-expressions, pronouns, and anaphors

The difference between Conditions A, B, and C of Binding Theory

How to use Binding Theory to determine syntactic structure

C-command

Where we're going: C-command elsewhere in syntax
We've demonstrated here the importance of c-command with respect to Binding Theory. But c-command is evident everywhere is syntax. Whenever we have a non-local dependency, we find that c-command mitigates that dependency in some way. For instance, consider again Head Movement from chapter 7. It is not clear that Head Movement also obeys c-command: you can only move to the closest c-commanding head. For instance, recall Head Movement, where a head is allowed to move up the tree ``past'' other phrases. [latex] \begin{tikzpicture} \tikzset{every tree node/.style={align=center,anchor=north, level distance = 20pt}} \Tree [.CP [.{C$'$} {C \\ $\emptyset$} [.TP [.DP [.{D$'$} {D \\ $\emptyset$} [.NP [.{N$'$} {N \\ John} ] ] ] ] [.{T$'$} \node(T){V+T \\ be+$[$pres$]$}; [.VP [.{V$'$} [.AdvP [.{Adv$'$} {Adv \\ sometimes} ] ] [.{V$'$} \node(V){V \\ $t_{be}$}; [.AdjP [.{Adj$'$} {Adj \\ hungry} ] ] ] ] ] ] ] ] ] %\draw[->, dashed] (V) ..controls +(south:4) and +(south:6)..(T); \draw[->, dashed] (V)--++(south:6em)-|(T); \end{tikzpicture} [/latex] The reason that we never see V more to, say, sometimes, is because the head Adv doesn't c-command V. That is, you can only head-move to a c-commanding head. We'll discover soon that there are other types of movement, and those two are constrained by c-command. As a general rule, you can only move to a position that c-commands your starting position. (The exception being head-lowering. ADD MORE)
Advanced
Pronouns are "special." We've discussed above the cases if pronouns that refer, that is, that point to an individuals (or individuals) in the discourse. But pronouns can sometimes have other functions. Consider the following pair of sentences.[footnote]Many people nowadays prefer the non-gendered "singular they" instead of she in sentences like this. There are prescriptive rules against this kind of singular they, it has been standard usage for hundreds of years.[/footnote] [latex] setcounter{ExNo}{16} \ex. \a. Beth thinks that she's the smartest. \b. Every student thinks that she's the smartest. [/latex] The sentences in (17) actually involve two different uses of the pronoun she. In (17a), the most natural reading is that she refers back to Beth.This is the use that we're familiar with; the pronoun is being used to refer to an individual. But what about (17b). Does she refer to a particular individual? On one reading, the answer is no. There's a reading of (17b) in she refers to different people. Consider the context in which we are talking about the students Sarah, Nekesa, Maria, and Althea (and each of these individuals identifies as female). We can understand (17b) to mean that "Sarah thinks that Sarah is the smartest" and "Nekesa thinks that Nekesa is the smartest" and "Maria thinks that Maria is the smartest" and "Althea thinks that Althea is the smartest." That is, as we run through the list of students, she refers to each of the students. (Note that she doesn't refer to every student: this sentence cannot mean "Every student thinks that every student is the smartest.") This reading doesn't treats she as a referring expressing because the pronoun doesn't pick out a particular individual in the discourse. Instead, she is a variable; she's reference varies depending on the associated quantifier. In this case, the quantifier is every. It turns out that variables have very different properties than referring pronouns: variables must be bound. For instance, the co-varying interpretation of she goes away if there is no c-command. [latex] \setcounter{ExNo}{17} \ex. *Every student$_i$'s father thinks that she$_i$'s the smartest [/latex] In (18), the variable interpretation doesn't seem to work; we can't let she mean "each of the students" in this example. (Note the use of indices here: putting an index on every student and she indicates that we should make those two DP's co-refer.) Variables can be bound "long-distance" as well: [latex] \setcounter{ExNo}{18} \ex. Every student$_i$ thinks that Bill said that she$_i$'s the smartest. [/latex] The takeaway from this is that pronouns in English (and many other languages) have two distinct functions; each function has a distinct distribution.
Referential pronouns refer to a discourse salient individual and are subject to Condition B of Binding Theory. Variable pronouns have variable reference and must be bound by a quantified expression.

Phrasal movement: A-movement

juliedoner — Wed, 15 Nov 2023 05:29:38 +0000

Learning Objectives

By the end of this chapter, you should be able to,

diagnose unaccusative and unergative verbs

draw trees representing the unaccusative/unergative distinction

understand the motivation for a low position of transitive/unergative subjects

understand the motivation for Voice

We have seen empirical evidence that the subject of the sentence (in the languages we've looked at) is in spec-TP. More specifically, we've seen evidence that spec-TP is filled at [pb_glossary id="644"]S-structure[/pb_glossary]. In this chapter, we're going to discover that the subject doesn't start there; it moves there. Again more specifically, at [pb_glossary id="696"]D-structure[/pb_glossary], spec-TP is empty. Subjects start lower in the tree and move to spec-TP. This is phrasal movement, that is, the movement of phrases (rather than heads, discussed in chapter 7). Phrasal movement is divided into two distinct domains: A-movement and A'-movement. This chapter will deal exclusively with A-movement. A'-movement is addressed in chapter 10. We'll approach phrasal movement in steps. First we'll show that some subjects can move to spec-TP. Then we'll show that all subjects move to spec-TP. We'll then deal with the question of where the subject starts from.

A quick refresher of what what was covered in chapter 8.

Verbs c-select their arguments. That is, they dictate how many, and what kind of, phrases they are required to appear with.

Verbs s-select their arguments. That is, they dictate the semantic properties, including thematic roles, of their arguments.

Every argument has one (and only one) theta-role .

Every theta-role always appears in the same "syntactic configuration" (=[pb_glossary id="646"]UTAH[/pb_glossary]).

Unaccusatives and unergatives
Consider the verbs in (1a) and (1b). [latex] \ex. \a. John sang. \b. The vase broke. [/latex] The question we will consider is whether (1a) and (1b) have the same tree. That is, if they're both intransitive verbs, do they both simply have a subject in the specifier of TP? Possibly. Call this Hypothesis #1.
Intransitive Verb Hypothesis #1 All subjects of intransitive verbs start in spec-TP.
This hypothesis is what we have been assuming: there is a dedicated subject position, and all subjects sit there. However, there's an alternative hypothesis which follows from our observation that thematic roles are associated with particular positions in the syntax (i.e., [pb_glossary id="646"]UTAH[/pb_glossary]). Notice that the parallelism between the intransitive verbs breaks down when we look at the transitive verbs. With sing, John is the subject of the transitive verb and the subject of the intransitive verb. But with break, the vase is the object of the transitive verb and subject of the intransitive verb. [latex] \setcounter{ExNo}{1} \ex. \a. \textbf{John} sang a song. \b. \textbf{John} sang. \ex. \a. John broke \textbf{the vase}. \b. \textbf{The vase} broke. [/latex] Thus, the alternative hypothesis is that, because the vase is a Patient in both the transitive and intransitive verb frames (it undergoes a change), it starts life as the object of break, and then moves to the subject position. This follows from UTAH, because UTAH holds that the Patient thematic role is always syntactically represented as the complement to V. Call this Hypothesis #2.
Intransitive Verb Hypothesis #2 Some subjects of intransitive verbs start as complements to V.
[latex] \textbf{Hypothesis \#1 D-structures} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ John}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ sing}} } } }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ the vase}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ break}} } } }; \end{tikzpicture} \end{minipage} \vspace{10pt} \textbf{Hypothesis \#2 D-structures} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {DP \\ John}} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ sing}} } } }; \end{tikzpicture} \end{minipage} \begin{minipage}{.49\textwidth} \begin{tikzpicture}[mytree] \node(top) {TP} child {node {T$'$} child {node {T \\ $[$past$]$}} child {node {VP} child {node {V$'$} child {node {V \\ break}} child {node {DP \\ the vase}} } } }; \end{tikzpicture} \end{minipage} [/latex] The idea behind hypothesis #2 is that when there is no Agent, the object of break becomes the subject by moving from comp-VP to spec-TP. [latex] \textbf{Hypothesis \#2 S-structure} \begin{tikzpicture}[mytree] \node(top) {TP} child {node(subj) {DP \\ the vase}} child {node {T$'$} child {node(T) {$t_T$}} child {node {VP} child {node {V$'$} child {node(V) {V+T \\ break + $[$past$]$}} child {node(obj) {$t_{DP}$}} } } }; \draw[->, dashed] (T.west)--++(west:.2em)|-(V.west); \draw[->] (obj) --++ (south:3em) -| (subj); \end{tikzpicture} [/latex]
The following diagnostics allow us to determine which of the two hypothesis is correct. In fact, the data suggest that hypothesis #2 is right. We distinguish two kinds of intransitive verbs.

Unergatives are intransitive verbs that lack an object, i.e., only have an [pb_glossary id="698"]external argument[/pb_glossary].

Unaccusatives are intransitives that lack a subject, i.e., only have an [pb_glossary id="699"]internal argument[/pb_glossary].

Note that when we say "lack a subject," we mean that they lack a subject at one level of representation: D-structure.
Unergatives and unaccusatives differ empirically on a number of levels. All the following diagnostics highlight the crucial difference between unergatives and unaccusatives: the former lack an internal argument, while the latter lack an external argument. That is, you can use these as tests to determine whether a verb is unaccusative or unergative.

Resultatives test.
A resultative phrase is a phrase that describes a "result state." For instance, in Keisha broke the vase into pieces, the phrase into pieces is a resultative, because it describes the result state of the vase after Keisha broke it.
Unaccusatives permit modification of the object with a "resultative" phrase, i.e., something that describes the result state of the object.
[latex] \setcounter{ExNo}{3} \ex. \a. The vase broke into pieces. \b. The water froze solid. \ex. \a. *John sang into pieces. \b. *Mary laughed solid. [/latex] The idea behind this test is that resultative phrases are in some kind of relationship with the object of the sentence. When I say, Keisha broke the vase into pieces, I'm describing the state that the vase was in after Keisha broke it. More abstractly, I'm saying something like "Keisha CAUSED the vase BE in pieces," where "CAUSE" and "BE" are abstract verbal notions. The intransitive sentence The vase broke into pieces is simply missing the CAUSE abstract verb, but the predication "the vase BE in pieces" is still present. Unergatives don't allow resultatives because the intransitive subject isn't in the same relationship with the resultative phrase: the predication "John BE in pieces" or "Mary BE solid" does not work in (5) because John and Mary were never objects of the verb (and therefore cannot be modified by the resultative).
[embed]https://youtu.be/sx29Gi_d9z8[/embed] Resultatives test (length 59s) credit: Malena Schoeni

Participial modification.

You can use unaccusatives as participial modifiers.
[latex] \setcounter{ExNo}{5} \ex. \a. The vase broke $\rightarrow$ the broken vase \b. The water froze $\rightarrow$ the frozen water \ex. \a. The man sang $\rightarrow$ *the sung man \b. the woman laughed $\rightarrow$ *the laughed woman [/latex] The idea behind the participial modification test is that the participial form of the verb still subcategorizes for an object position, but excludes a subject position.
[embed]https://youtu.be/vRhTmmUiOJU[/embed] Participial modification (length: 45s) credit: Corey McCulloch

X's way/X-self.

You can add either X's way or X-self to unergatives, but not unaccusatives.
[latex] \setcounter{ExNo}{7} \ex. \a. *The vase broke its way home \b. *The water froze itself silly. \ex. \a. John sang his way through the mall. \b. Mary laughed herself silly. [/latex] The idea behind this test is that with unergative verbs, there is no object, and so something else can be placed in that position. That is, you can "give" an unergative verb an object because comp-VP is, in a sense, "free." You can't give unaccusative verbs an object because comp-V is already occupied; there's nowhere to put an extra object.
[embed]https://youtu.be/020jEqhxDPY[/embed] X's way (length: 45s) credit: Corey McCulloch

Cognate objects.

Unergatives allow an object that expresses the same thing as the verb --- a cognate object, which is a noun that describes the same action as the verb.
[latex] \setcounter{ExNo}{9} \ex. \a. *The vase broke a big break. \b. *The water froze a big freeze. \ex. \a. John sang a song \b. Mary laughed a hearty laugh. [/latex] This test has the same purpose as the X's way/X-self test: it's diagnosing whether the object position (comp-VP) is empty. If it is, you can put a cognate object in the structure.
[embed]https://youtu.be/GWhH_Bt8YnY[/embed] Unaccusative syntax (length: 1m 31s)

Crosslinguistic evidence for the ergative/unaccusative distinction
The so-called Unaccusative Hypothesis has quite a bit of evidence in its favor, and a lot of that evidence come from crosslinguistic patterns.
Auxiliary selection.
In many languages, particularly Indo-European languages like Romance and Germanic, the auxiliary used to express past tense/perfect aspect is (at least superficially) sensitive to the distinction between unergatives and unaccusatives. In many languages, the auxiliary verb be is used for unaccusatives, and have is used for unergatives. [latex] \setcounter{ExNo}{12} German \ex. \ag. Die Vase ist gebrochen\\ the vase is broken\\ \trans `The vase broke.' \bg. Maria hat gesungen\\ Maria has sung\\ \trans `Mary sang.' [/latex]
Morphological distinctions.
In many languages, there are morphological indicators that distinguish unaccusatives and unergatives. For instance, in Romance languages, many unaccusatives obligatorily occur with the so-called reflexive clitic se/si. Unergative verbs never appear with reflexive clitic. [latex] \setcounter{ExNo}{13} \ex. \ag. La vase se casse\\ the vase \textsc{refl} break.\textsc{pres}\\ \trans `The vase breaks.' \bg. Marie (*se) chante\\ Marie \textsc{refl} sing.\textsc{pres}\\ \trans `Marie sings.' [/latex] In Logoori, a Bantu language of Kenya, many unaccusative verbs appear with the suffix -ik. Unergative verbs may never appear with this suffix. [latex] \ex. \ag. Sira a-han-i muriang'o\\ Sira \textsc{sm}-close-\textsc{fv} door\\ \trans `Sira closed the door.' \hfill Transitive \bg. muriang'o gu-han\textbf{-ik}-i\\ door \textsc{sm}-close-\textsc{ac}-\textsc{fv}\\ \trans `The door closed.' \hfill Unaccusative \ex. \ag. Sira y-iimb-i ruembo\\ Sira \textsc{sm}-sing-\textsc{fv} song\\ \trans `Sira sang a song.' \hfill Transitive \bg. Sira y-iimb-i\\ Sira \textsc{sm}-sing-\textsc{fv}\\ \trans `Sira sang.' \hfill Unergative [/latex]
To be clear, we have to be somewhat cautious about our diagnostics and evidence. While the data listed above can pick out unaccusatives, it is not guaranteed that an unaccusative verb must display all the properties listed above. For instance, the English verb arrive does not permit cognate objects, but also does not permit participial modification. [latex] \ex. \a. *Carol arrived a big arrival. \b. ??the arrived man. [/latex] Despite this divergent evidence, we generally believe that arrive is unaccusative. This is in part based on cross-linguistic evidence, and also in part based on factors we do not have time to consider here.

On passivization
We've shown above that sometimes, the subject moves to spec-TP. Indeed, there is pretty clear evidence that not all subjects start in spec-TP just from [pb_glossary id="647"]passivization[/pb_glossary]. Passivization is a syntactic process that turns a transitive verb into an intransitive verb. [latex] \setcounter{ExNo}{14} \ex. \a. John broke the vase. \hfill ``active'' \b. The vase was broken. \hfill ``passive'' [/latex]
The subject in (15b) starts as the object of break, and then gets "promoted" to the subject position. This is just like the unaccusative syntax we observed earlier: the D-structure object becomes the S-structure subject. However, passives and unaccusatives differ in a few crucial ways. Most notably, passives allow the external argument to be expressed as an [pb_glossary id="694"]oblique[/pb_glossary] argument (in a prepositional phrase). Unaccusatives do not permit this.
[latex] \setcounter{ExNo}{15} \ex. \a. The vase was broken by Tabitha \hfill passive \b. *The vase broke by Tabitha \hfill unaccusative [/latex]
This suggests that even though both passives and unaccusatives involve movement of an object to a subject, they mean slightly distinct things. The passive alternation in English is complex, and so we're going to put it aside here.

Generalized movement to spec-TP

We now know that some subjects don't start in the specifier of TP, they move there. Specifically, unaccusatives (and passives) demonstrate that not all subjects start in D-structure in the specifier of TP.

Now let's ask whether anything starts in spec-TP.That is, we used to have a uniform theory: subjects starts in spec-TP. Now we have a "split" theory: some subjects start in spec-TP, some move there. Now we're going to explore whether a uniform theory is still possible: maybe all subjects move to spec-TP.
Though we'll go over empirical evidence for this momentarily, note that purely on theoretical grounds, having subjects of transitive and unergative verbs start in spec-TP is fishy. In all other cases, selection is local, meaning that "a head dictates the syntax around it" (as discussed in chapter 8). The problem appears to be that subjects (of unergative and transitive verbs) aren't next to what is selecting them. That is, if sing is unergative and it selects for its subject, why is the subject so far away from the verb in examples like (17)? [latex] \setcounter{ExNo}{16} \ex. The man has been singing. [/latex] In the following sections, we'll again be weighing two hypotheses. The first hypothesis is the null hypothesis: subjects of transitive and unergative verbs start in spec-TP, just like we've been doing. The second hypothesis is that subjects start somewhere closer to the verb. (We'll be more precise about where shortly.) That is, nothing starts in spec-TP. Everything that we pronounce in spec-TP has moved there. The two hypotheses are sketched below. [latex] \noindent\textbf{Hypothesis \#1 ($=$the ``old'' way)}\\ \begin{tikzpicture} \Tree [.TP [.DP ] [.\textoverline{T} T [.{\ldots} \edge[roof]; {\ldots} ] ] ] \end{tikzpicture} \noindent\textbf{Hypothesis \#2 ($=$the ``new'' way)}\\ \begin{tikzpicture} \Tree [.TP [.\node(dp){DP}; ] [.\textoverline{T} T [.{\ldots} \edge[roof]; {\ldots} ] ] ] \node(t) at (.6,-3.2) {$t_{DP}$}; \draw[->, rounded corners] (t) -- ++(west:.75) -| (dp); \end{tikzpicture} [/latex]
These hypotheses make different predictions.

Hypothesis #1 predicts that a transitive/unergative subject can never be pronounced in the verb phrase, and that there is no evidence for movement to spec-TP.

Hypothesis #2 predicts that a subject can be pronounced in the verb phrase (if it hasn't moved), and that there is evidence for movement to spec-TP.

So we're looking for two kinds of data.

Is there evidence for a low position of the subject?

Is there evidence for movement to spec-TP?

There's lots of evidence for 1., and a bit of evidence for 2. Let's start with arguments that the subject can be pronounced lower than spec-TP.

Evidence for a subject low position

Small clauses.

Some clauses are "smaller" than others. The bracketed constituents are missing a tense phrase (TP).
[latex] \setcounter{ExNo}{17} \ex. \a. I saw [ Mary hug a pedestrian ] \b. I heard [ John sing ] \c. I made [ Carrie build a house ] [/latex] The subjects of hug, sing and build are not in spec-TP because there is no TP above hug, sing and build. (You can tell because hug, sing, and build do not---and cannot---appear with tense morphology: *I saw Mary will hug a pedestrian.) How could Mary, John and Carrie even get into the tree if all subjects start in spec-TP? These data suggest that Hypothesis #2 is correct, because Hypothesis #1 predicts subjects are only found in spec-TP; if there's no spec-TP, there can be no subject.
Existential constructions.
Consider the following alternation. There in (19b) is called an [pb_glossary id="649"]expletive[/pb_glossary] subject. Expletive subjects are semantically vacuous. [latex] \setcounter{ExNo}{18} \ex. \a. Three woman are climbing a mountain. \b. There are three women climbing a mountain. [/latex] Again, these example suggest that, at least sometimes, the subject of the sentence can be pronounced not in spec-TP. In the case of expletive there, something else sits in spec-TP, namely there, and so the subject three women must be somewhere else---somewhere lower.

Should we also consider the possibility that there (and expletive it) start lower in the clause? Maybe, but the rule of thumb is that it and there are not semantically selected by the verb. Expletive subjects are inserted to satisfy the requirement that every clause needs a subject, and so expletive subjects do not start in the verb phrase, they start in spec-TP.

VSO languages.

Many languages have a word order that suggests that the subject stays "low" in the clause. Consider Irish, which has canonical Verb-Subject-Object ordering for declaratives sentences.
[latex] \setcounter{ExNo}{19} \exg. Chonaic m\'e Eoghan\\ saw I Owen\\ \trans `I saw Owen.'\label{irish1} [/latex] These data make sense if V moves across the subject to T, and the subject stays low. Thus, they are most consistent with hypothesis #2. They are only consistent with hypothesis #1 with additional stipulations.
Locative inversion.

Here's some data from Zulu (from Buell 2007). In Zulu, locative phrases can be in spec-TP, triggering agreement on the verb; we see agreement between houses (noun class 10) and the subject prefix on stay in (21b).
[latex] \setcounter{ExNo}{20} \ex. \ag. Abantu abadala ba-hlala ku lezi zindlu. \\ 2people 2old 2-stay at 10these 10houses \\ \trans `Old people live in these houses.' \bg. Lezi zindlu zi-hlala abantu abadala. \\ 10these 10houses 10-stay 2people 2old \\ \trans `Old people live in these houses.' [/latex] Herero (Bantu) shows the same thing. (Data from Buell 2007 as well.) [latex] \setcounter{ExNo}{21} \ex. \ag. \`om\'uk\'az\'end\'u \'u-t\'er\'ek\`a \`ony\`am\`a p\`ong\`and\'a\\ 1woman 1-cook 9meat 16at.home \\ \trans `The woman cooks the meat at home.' \bg. p\`ong\`and\'a p\'e-t\'er\'ek\`a \`om\'uk\'az\'end\'u \`ony\`am\`a\\ 16at.home 16-cook 1woman 9meat\\ \trans `The woman cooks the meat at home.' [/latex] Notice in this pair that the English translation is the same for both sentences!
Locative inversion data are also consistent with Hypothesis #2, but are not consistent with Hypothesis #1. Under Hypothesis #2, what is happening with locative inversion is that something other than the "logical" subject gets to move to spec-TP, namely, the locative phrase. With Hypothesis #1, the word order doesn't seem possible: if the subject is always in spec-TP, how could we possible derive the word order where it appears after the verb?

Evidence for movement to spec-TP.

Quantifer float.

Some quantifiers can appear separated from the noun they combine with. This is called [pb_glossary id="650"]quantifier float[/pb_glossary].
[latex] \setcounter{ExNo}{22} \ex. \a. \textbf{All the children} will visit Santa. \hfill {\small{(the normal position)}} \b. \textbf{The children} will \textbf{all} visit Santa. \hfill {\small{(stranded! where?)}} \c. * \textbf{The children} will visit \textbf{all} Santa. \hfill {\small{(can't go here{\ldots})}} \d. * \textbf{The children} will visit Santa \textbf{all}. \hfill {\small{(can't go here{\ldots})}} [/latex]
The first two examples show us where all can go:

It can appear in front of the DP.

It can appear in front of the verb.

These data make sense if quantifiers like all can "float" in positions that the subject has occupied. Phrased differently, the data make sense if all can be "left behind" when the subject moves to spec-TP. All appears wherever the DP used to be. The ungrammatical sentences confirm this: we can't put all after the verb because the subject was never after the verb.
Voice aka "Little V"
In sum, the empirical evidence suggests that even subjects of transitive and unergative verbs start lower in the structure. The question we'll address here is where precisely the subject begins life. We said in chapter 8 that a verb like throw selects its subject. If I asked you what thematic role throw assigns to its subject, you'd probably say Agent. But consider the verb phrases in (24). All of these involve the verb throw, but the object varies. [latex] \setcounter{ExNo}{23} \ex. \a. throw a baseball \b. throw support behind someone \c. throw a match \d. throw a party \e. throw a fit \f. throw my back out \hfill (adapted from Marantz, 1984) \g. \ldots [/latex] Interestingly, we see that the thematic role of the subject depends not on the verb alone, but rather the combination of the verb and object. Throwing a baseball is not the same act as throwing a fit, or throwing my back out. In each case, the subject has a distinct thematic role, depending on both the verb and the object. The same thing can be demonstrated with the verb kill. Again, if I asked what thematic role kill assigns to the subject, you would probably say Agent. But now consider what happens when we vary the object. [latex] \setcounter{ExNo}{24} \ex. \a. kill a cockroach \b. kill a conversation \c. kill some time \d. kill an audience \e. \ldots [/latex] Again, it looks like it isn't just the verb, but the combination of verb and object that determines the semantic role of the subject. For instance, My feet are killing me, certainly cannot be attributing an Agent thematic role to my feet. Likewise, Stephen kills time by playing Switch does not attribute to Stephen the same thematic as in, Stephen killed a cockroach. The generalization is that selection of the subject depends on the entire VP, not just V. That is, the combination of verb and object together---the VP---determines the thematic properties of the subject. For our theory then, we then need a way to relate the VP to the subject. We will adopt the functional projection Voice to perform this role. [latex] \begin{tikzpicture}[mytree] \node(top) {TP} child {node {T$'$} child {node {T}} child {node {VoiceP} child {node(S2) {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} child {node {DP}} } } } } }; \node(S1) at (-3,-5) {the subject \\ at D-structure}; \draw[->, dashed] (S1) -- (S2); \end{tikzpicture} [/latex] Like other functional heads, Voice's job is to provide a relation between two phrases. Voice maps the subject to the VP---or more specifically, it relates the subject to the event described by the VP. It allows the VP to dictate the thematic properties of the subject. More generally, Voice controls the transitivity of the verb. We will assume for simplicity in this class that transitive and unergative verbs require Voice, but unaccusatives do not. This is, however, a significant simplification of the issues. [latex] \begin{minipage}{.33\textwidth} \textbf{Unaccusative schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VP} child {node {V$'$} child {node {V}} child {node {DP}} }; \end{tikzpicture} \end{minipage} \hspace{10pt} \begin{minipage}{.33\textwidth} \textbf{Unergative schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VoiceP} child {node {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} } } }; \end{tikzpicture} \end{minipage} \hspace{10pt} \begin{minipage}{.33\textwidth} \textbf{Transitive schema:}\\ \begin{tikzpicture}[mytree] \node(top) {VoiceP} child {node {DP}} child {node {Voice$'$} child {node {Voice}} child {node {VP} child {node {V$'$} child {node {V}} child {node {DP}} } } }; \end{tikzpicture} \end{minipage} [/latex]
Note that in many works, Voice is called v, pronounced "little v." We will not adopt this terminology, but be aware that Voice and v are often treated interchangeably.
Adding Voice requires a small modification to our T-lowering rule. We'll assume that instead of T lowering to V, it lowers to Voice, and that V raises to Voice. They "meet in the middle." (Nothing changes for unaccusatives; T still lowers to V.) [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Britta}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T\\ build+$\emptyset$+$[$past$]$}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {DP \\ a house}} } } } } } } }; \draw[->] (Slow.west)-|(S); \draw[->, dashed] (T)|-(Voice.west); \draw[->, dashed] (V.west)-|(Voice); \end{tikzpicture} [/latex] This is basically a stipulation. For exposition's sake, we will not bother to justify the movement of V to Voice. It is probably necessary for theory-internal reasons (in English), though if V didn't move to Voice, nothing in our theory would be significantly affected.
[embed]https://youtu.be/GzaiEoRBh_k[/embed] Voice in trees (length: 2m 27s)

Key Takeaways

Voice always introduces the external argument. That is, when it's in the tree, it serves the function of bringing in the argument that does not start inside of the VP (as the complement to V). You need Voice above the VP if,

the verb is transitive, that is, has both a subject and an object, or

the verb is unergative.

The only times you don't have Voice above the VP is if,

the verb is unaccusative, that is, has a single argument which starts as the object of the verb, or

the verb is an auxiliary verb.

In other words, unaccusatives and auxiliary verbs do not select for an external argument, and so do not need Voice.

The Unergative/unaccusative distinction beyond verbs
We have focused above on detecting unaccusative and unergative verbal predicates. It has been argued, however, that the distinction extends to other categories, notably, adjectives. For instance Hans Bennis argues based on data from Dutch that the adjective duidelijk 'clear' is unaccusative, while the adjective trouw 'loyal' is unergative.[footnote]Technically, Bennis uses the term "ergative" to mean "unaccusative." This is terminological difference, and does not matter for his analysis.[/footnote] To prove this distinction requires a separate set of diagnostics from what we've introduced above, so we will put aside the alternation for adjectives here. However, we still need a place to generate the subject in a sentence like Mary is mean. Since mean selects for a subject, and assigns it a thematic role, Mary must be selected by mean. We will assume that it is introduced in the specifier of AP, and subsequently moves to spec-TP. This conclusion is consistent with everything we've said above: spec-TP is always empty at D-structure. [latex] \begin{tikzpicture}[mytree] \node(top){CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Mary}} child {node {T$'$} child {node(T) {T+V \\ $[$pres$]$+is}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {AdjP} child {node(tS) {$t_{DP}$}} child {node {Adj$'$} child {node {Adj \\ mean}} } } } } } } }; \draw[->, dashed] (V.west)-|(T); \draw[->] (tS.west)-|(S); \end{tikzpicture} [/latex]
Morphological evidence for Voice
The reasoning for Voice so far has been largely theory internal. When we look cross-linguistically, however, we find a great deal of evidence for a Voice head. For instance, consider the morphology of the following alternations in Kunuz Nubian (data from Abdel-Hafiz 1988).
Transitive and intransitive verbs in Kunuz Nubian

Intransitive Transitive

dab 'disappear' dabir 'disappear'

bokki 'hide' bokkir 'hide'

wacci 'crack' waccir 'crack'

bassi 'leak' bassir 'leak'

What is the pattern? To make a transitive verb from an intransitive verb in Kunuz Nubian, you add the suffix -ir (with possibly vowel deletion with verb-final roots). Why is this meaningful? It shows that the difference between transitive and intransitive is not just the addition of a subject, but you actually add something to the verb.
Swahili also has a suffix -ish or -esh that does something similar.[footnote]The choice of vowel depends on the preceding vowels in the word. It is a case of "vowel harmony."[/footnote]
Transitive and intransitive verbs in Swahili

Intransitive Transitive

-rudi 'return' -rudisha 'return sth, bring sth. back'

-chelewa 'be late' -chelewesha 'make late, delay'

-fa 'die' -fisha 'destroy, cause to die'

-enda 'go, move' -endesha 'make go, drive'

-weza 'be able' -wezesha 'enable'

If there were no Voice head, then we couldn't explain why there is a systematic distinction between transitive/intransitive pairs. If we simply claim that the causative affix in Kunuz Nubian and Swahili are just how you pronounce Voice on transitive verbs, then the issue goes away. The following examples illustrate the phrasal movements and Voice. (Recall that in Swahili, V moves to T, so the head-movement goes up the tree, stopping at Voice on the way.) [latex] \setcounter{ExNo}{25} \exg. Sira a-me-chelewa\\ Sira 1\textsc{sm}-\textsc{perf}-be.late\\ \trans `Sira is late.' \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance =7em] child {node(S) {DP \\ Sira}} child {node {T$'$} child {node(T) {T+V \\ \textit{a-me-chelewa} \\ $[$perf$]$+be.late}} child {node {VP} child {node(V) {$t_{V}$}} child {node(t) {$t_{DP}$}} } } } }; \draw[->] (t)--++(south:3em)-|(S); \draw[->, dashed] (V.south)--++(south:1em)-|(T.south); \end{tikzpicture} [/latex] [latex] \setcounter{ExNo}{26} \exg. Juma a-me-chelew-esha Sira\\ Juma 1\textsc{sm}-\textsc{perf}-be.late-\textsc{caus} Sira\\ \trans `Juma made Sira late.' \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em, level distance =3em] child {node(S) {DP \\ Juma}} child {node {T$'$} child {node(T) {T+V+Voice \\ \textit{a-me-chelew-esha} \\ $[$perf$]$ + be.late + $\emptyset$}} child {node {VoiceP} child {node(t) {T$_{DP}$}} child {node {Voice$'$} child {node(Voice) {$t_{Voice}$}} child {node {VP} child {node(V) {$t_{V}$}} child {node {DP \\ Sira}} } } } } } }; \draw[->, dashed] (V)--++(south:1em)-|(Voice); \draw[->, dashed] (Voice.west)--++(west:1em)-|(T.south); \draw[->] (t)--++(south:1em)-|(S.south); \end{tikzpicture} [/latex]
Double-Object Constructions
We've been dealing exclusively with intransitive and transitive verbs, but we can also sketch a solution to transitive verbs, like give, which selects for two objects in (28), Margit and a book. [latex] \setcounter{ExNo}{27} \ex. Stacy gave Margit a book [/latex] This is called a double-object construction because the verb give appears to have two objects. We can say that the indirect object (Margit) is introduced as a specifier to VP, since this position in completely open! [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node(S) {DP \\ Stacy}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} [sibling distance = 6em] child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V+T \\ $\emptyset$+give+$[$past$]$}} child {node {VP} child {node {DP \\ Margit}} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {DP \\ a book}} } } } } } } }; \draw[->, dashed] (V)--++(south:1em)-|(Voice); \draw[->, dashed] (T)|-(Voice.west); \draw[->] (Slow.west)-|(S); \end{tikzpicture} [/latex]
Where we're going
In this chapter we introduced the idea that all subjects start lower than TP, and move there (at least in English and the languages we've looked at). This is phrasal movement, meaning the movement of phrases. Recall that the impetus for head-movement (the movement of heads) is, basically, word-formation. The suffix -ed in English just can't stand on its own; it has to be part of a word, and so either T moves to V, or V moves to T. So what is the impetus for phrasal movement? The answer to this question is much more complex. For the movements we've looked at in this chapter, we might say that the impetus for movement to spec-TP is simply that "every clause needs a subject." This idea can be syntactically encoded in the following constraint.
The Subject Condition Spec-TP must be filled at S-structure.
It's a stipulation, but it is "surface true," meaning that it appears to be a good generalization about English, and many other languages as well. There is something "special" about spec-TP in that it must be filled in the surface structure. Indeed, the idea that spec-TP "wants" to have something in it will lead us to consider other cases of movement to the subject position in the next chapter. As we'll discover in chapter 10, there are some movements that cannot be explained in the same way. It is an ongoing research question as to why things move.

Key Takeaways

identifying and representing unaccusative and unergative predicates

diagnostics for low position of (unergative/transitive) subjects

evidence for Voice

drawing trees representing movement to spec-TP

Exercises

VSO tree drawing

Here's some data from Niuean (from Woolford 1991). What do you observe about this data that bears on the question of whether the subject is low or high? (Notice the placement of the adverb!)
[latex] \setcounter{ExNo}{28} \ex. \ag. to lagomatai he ekekafo a ia\\ \textsc{fut} help \textsc{erg} doctor \textsc{abs} him\\ \trans `The doctor will help him.' \bg. Ne tutuli tumau he pusi e lapiti\\ \textsc{past} chase always \textsc{erg} cat \textsc{abs} rabbit\\ \trans `The cat always chased the rabbit.' [/latex]

Straight to spec-TP?

Consider again the data illustrating quantifier float in English, this time with the addition of the modal might. Think about the different positions that all can "float" (i.e. be left behind), and what this tells us about movement. How does the subject get to spec-TP? [latex] \setcounter{ExNo}{29} \ex. \a. \textbf{The children all} might have been visiting Santa. \b. \textbf{The children} might \textbf{all} have been visiting Santa. \c. \textbf{The children} might have \textbf{all} been visiting Santa. \d. \textbf{The children} might have been \textbf{all} visiting Santa. [/latex]

Raising versus Control

juliedoner — Wed, 15 Nov 2023 05:29:39 +0000

Learning Objectives

By the end of this chapter, you should,

be able to identify raising and control predicates using the diagnostics listed below

understand the structural difference between raising/control verbs, and be able to structural represent that difference

appreciate the arguments and evidence for PRO, and how it differs from pro.

We now know that spec-TP is empty at D-structure, and that subjects raise to spec-TP from where they are selected. This chapter addresses further issues in phrasal movement. We will mainly be concerned with the similarities and differences between the following sentences.
[latex] \ex. \a. Khufu seems to run fast. \b. Khufu wants to run fast. [/latex] The sentences of course mean different things. But they consist of the same linear sequence of words. A subject (Khufu), a finite verb (seems/wants), a nonfinite verb (to run) and an adverb (fast). The sole difference is the finite verb. In (1a), the verb is seems and in (1b) the verb is wants. Despite the surface similarities, we'll discover in this chapter that the two sentences have wildly different syntactic structures, all of which are related to the following core difference:
Want selects for a subject, seem does not.
That is, want c-selects for a subject and assigns it a thematic role. Khufu in (1b) is an argument of want and gets the thematic role Experiencer. Seem on the other does not select for a subject and does not assign it a thematic role. Instead, in (1a), Khufu is selected by the infinitive and "raises" into the main clause from the lower clause. The difference between the two verbs is sketched below. We'll provide a more precise analysis later. The important thing to keep in mind is the starting place of the subject. For verbs like seem, the subject starts inside of the lower clause. For verbs like want, the subject starts outside of the lower clause. [latex] \setcounter{ExNo}{1} \tikzstyle{every picture}+=[remember picture, inner sep=0pt, baseline, anchor=base]% \ex.{}[\textsubscript{TP} \tikz\node(high){Khufu}; [ seems [ to [ \tikz\node(t){$t$}; run fast ] ] ] ] \begin{tikzpicture}[overlay] \draw[->] (t.south) -- ++(south:1.5ex) -| (high.south); \end{tikzpicture} \tikzstyle{every picture}+=[remember picture, inner sep=0pt, baseline, anchor=base]% \ex. {}[\textsubscript{TP} \tikz\node(high2){Khufu}; [ \tikz\node(t2){$t$}; [ wants [ to [ run fast ] ] ] ] ] \begin{tikzpicture}[overlay] \draw[->] (t2.south) -- ++(south:1.5ex) -| (high2.south); \end{tikzpicture} [/latex] We call predicates like seem [pb_glossary id="652"]raising[/pb_glossary] verbs, and we call predicates like want [pb_glossary id="653"]control[/pb_glossary] verbs. All of the following diagnostics illustrate this same difference: that the starting place of the subject of a raising verb is different than the starting place of the subject of a control verb. Note that while the data is given in English, these diagnostics have cross-linguistic validity. All else equal, you can use these diagnostics to determine whether a verb embedding a nonfinite clause belongs to the class of raising or control verbs.
Raising vs. control
All of the following diagnostics allow you to determine if something is a raising or a control verb. To apply the diagnostic, you mechanically use the data given.
Expletives subjects I.
Raising verbs often permit an alternation between a referential subject and an [pb_glossary id="649"]expletive[/pb_glossary] subject. Control verbs don't. [latex] \setcounter{ExNo}{3} \ex. \a. It seems that Khufu runs fast. \b. *It wants that Khufu runs fast. [/latex] How do we know that it in (4a) is an expletive? Well, it doesn't appear to have semantic content. It doesn't refer to anything. For instance, you can't ask the question, *What seems that Khufu runs fast? These data support the idea that seem doesn't s-select for a subject, because that position can be filled by a semantically "vacuous" element. Note that many languages don't (obviously) have expletive subjects. For instance, in pro-drop languages like Spanish and Swahili, the subject position is obligatorily empty. [latex] \setcounter{ExNo}{4} \exg. (*eso) parece que Mar\'ia vio un \'arbol\\ it seem.\textsc{pres} that Mar\'ia see.\textsc{past} a tree\\ \trans `It seems that Maria saw a tree.' \exg. (*hii) i-na-onekana kwamba Beni a-na-penda magari\\ this 9\textsc{sm}-\textsc{pres}-seem that Beni 1\textsc{sm}-\textsc{pres}-like cars\\ \trans `It seems that Beni likes cars.' [/latex] Again, this supports the idea that such sentences involve a semantically empty subject, since it cannot be referred to using an overt pronoun.
How to apply the expletive subject test

Start with you original sentence → Khufu seems to run fast

Replace the subject with an expletive, and make the lower clause finite → It seems that Khufu runs fast.

Check the grammaticality. If it's grammatical, the verb is a raising verb.

[embed]https://youtu.be/yNBizEZwPoQ[/embed] Expletives I (length: 55s) credit: Keira Dobbs

It is worth asking why there are expletive subjects at all. They don't appear to serve a semantic purpose---that is, they don't mean anything. So why do expletive subjects exist? This is a complex question, and there are many ways to answer it. But one thing we can say is that, though expletive subjects do not serve a semantic function, they do serve a syntactic function, namely, the [pb_glossary id="657"]Subject Condition[/pb_glossary]. If there is nothing else that can filled spec-TP, then an expletive subject is inserted. The following are examples of other types of expletive subjects in English. [latex] \setcounter{ExNo}{6} \ex. \a. It's raining. \b. It's hot outside. \c. It was a dark and stormy night. [/latex]

Idiom chunks.

Some phrases are "non-compositional" in that the meaning isn't predictable from the parts. These are called [pb_glossary id="651"]idioms[/pb_glossary] or idiom chunks.
[latex] \setcounter{ExNo}{7} \ex. \a. The cat is out of the bag. $\longrightarrow$ ``The secret is known." \b. The shit hit the fan. $\longrightarrow$ ``There's a problem." [/latex]
We find a distinction between raising and control verbs with respect to idiom chunks.
[latex] \setcounter{ExNo}{8} \ex. \a. The cat seems to be out of the bag. \b. ??The cat wants to be out of the bag. \ex. \a. The shit seems to have hit the fan. \b. ??The shit wants to have hit the fan. [/latex] What we observe is that with raising verbs, the idiom meaning is retained. With control verbs it isn't. The a. examples sound acceptable because we can interpret the entire idiom chunk together. The control verbs lose the idiomatic meaning because the subject is actually a subject of the higher verb. Note that this diagnostic is concerned with interpretation. If we get the non-literal (idiomatic) reading after inserting the verb – as we do in (9a) and (10a) – then we conclude that the subject must have raised from the lower clause, as this is where the rest of the material from the idiomatic expression resides. This same idea explains why the idiomatic meaning is not retained with control verbs; the subject the cat in (9b) can only be interpreted as an actual animal that wants out of a bag (literal reading) because it was never in the lower clause, and therefore cannot be interpreted with the idiomatic material.
Applying the idiom chunk test

Choose your favorite idiom → The shit hit the fan.

Take the predicate you want to test (seem) and put it in the idiom → The shit seems to have hit the fan

Check whether the idiom meaning is retained. You're not looking for (un)grammaticality, rather you're looking for whether the idiom meaning survives. If the idiom meaning survives, then it's a raising verb.

[embed]https://youtu.be/VpVjkb-VTc4[/embed] Idiom test (length: 2m 49s) credit: Alesha Pisciotta [embed]https://youtu.be/6uB_ERNvhxA[/embed] Idiom test (length: 1m 1s) credit: Keira Dobbs

Expletive subjects II.

English has another expletive subject, there. It's only possible to use expletive there with a particular class of verbs, namely, verbs which assert existence, like the copula be.
[latex] \setcounter{ExNo}{10} \ex. \a. There is a book on the table. \b. There are three women in the room. [/latex] Again we see a distinction between raising and control verbs with respect to expletive there. [latex] \setcounter{ExNo}{11} \ex. \a. There seems to be a book on the table. \b. *There wants to be a book on the table. [/latex] The explanation behind this data goes as follows. Expletive there must be selected for by a particular predicate. Its appearance in (12a) follows because it has been selected for by the infinitive be in the lower clause. The reason it cannot appear in (12b) is that want cannot select for there.
Applying the expletive subject II test

Take an existential sentence → There is a book on the table.

Put the predicate you're testing (seem) into the sentence → There seems to be a book on the table.

Check the grammaticality. If it's grammatical, then it's a raising verb.

[embed]https://youtu.be/NDjYdycsRao[/embed] Expletive subjects II (length: 1m 22s) Credit: Megan Spurgeon

Scope ambiguity.

This one is a bit more complex, as it presupposes a bit of semantics. This test asks us to consider how quantified expressions are interpreted in Raising and Control structures. A quantified expression is any expression that asserts a "quantity" of a set of individuals.[footnote]In fact, quantifiers can apply to really any number of things beyond individuals. For instance, there are quantifiers over times (sometimes) and even over "worlds" (believe). For more discussion of quantifiers, I recommend taking Semantics and Syntax II. [/footnote] As example of quantified expressions, consider the sentence Every student met some professor. This sentence is ambiguous with respect to how the quantified expressions every student and some professor interact. On one reading (called the "surface scope" reading), we are talking about the list of students. As long as each student has met a professor, then the sentence is true. On this reading, the professors could be different. Timmy met Prof. Smith. Susie met Prof. Jones. Rachel met Prof. Scott. Etc.
On an alternate reading of this sentence, we are talking about a particular professor. On this reading, the sentence is true if every student met the same professor. So Timmy met Prof. Smith and Susie met Prof. Smith and Rachel met Prof. Smith. The difference between the two readings is the scope of some professor. On the first reading, some professor scope "under" the every student. On the second reading some professor scopes "above" every professor.
Now consider the following sentence with a Raising predicate, be likely.
[latex] \setcounter{ExNo}{12} \ex. Someone from Florida is likely to win the lottery. \a. $\approx$ ``There is a particular person from Florida who is likely to win the lottery (her name is Cheryl).'' \b. $\approx$ ``It's likely is that someone or other in Florida will win the lottery (but we don't know who).'' [/latex] Notice that reading in (13b) is paraphrased as if someone were in an embedded clause. This is the reading we get when we interpret the quantified expression someone lower than where it's pronounced. Crucially, this same ambiguity doesn't seem to exist with want. [latex] \setcounter{ExNo}{13} \ex. Someone from Florida wants to win the lottery. \a. $\approx$ ``There is a particular person from Florida who wants to win the lottery (her name is Cheryl).'' \b. $\approx$ ``??What is wanted is that someone or other in Florida win the lottery (but we don't know who).'' [/latex] The sentence Someone from Florida wants to win the lottery is not ambiguous in the same way. It does not have a reading in which someone is interpreted lower than the position that it is pronounced.
Applying the scope ambiguity test

Choose a sentence with an indefinite subject → Someone from Florida won the lottery

Put the predicate you're testing in the sentence → Someone from Florida seems to have won the lottery

Determine if the resulting meaning is equivalent to having an expletive subject (see expletive subject I test above) → ≈ It seems like someone from Florida won the lottery." If it is, then it's a raising verb.

Truth changing passive.

If you passivize the infinitive under a raising verb, it basically means the same as the active version. If you passivize the infinitive under a control structure, it means something very different.
[latex] \setcounter{ExNo}{14} \ex. \a. The doctor seemed to examine Mary $\approx$ Mary seemed to be examined by the doctor. \b. The doctor wanted to examine Mary $\not\approx$ Mary wanted to be examined by the doctor [/latex] Here again the explanation is that the subject of seem "inherits" whatever thematic role is assigned to the subject of the passivized verb. But the subject of want gets its own thematic role.
Applying the passivization test

Choose a sentence with a transitive verb → The doctor examined Mary.

Put the predicate you're testing (seem) in the sentence → The doctor seems to have examined Mary

Passivize your original sentence → Mary was examined by the doctor

Put the predicate you're testing (seem) in the sentence → Mary seems to have been examined by the doctor

Check whether the two sentences you've created meaning the same thing → The doctor seems to have examined Mary = Mary seems to have been examined by the doctor. If the sentences are interchangeable, then it's a raising verb.

[embed]https://youtu.be/WH82x4xtOos[/embed] Passivization (Alesha Pisciotta; length: 3m 20s)

The final two tests below pick out only a subset of control verbs. Thus, if they pass, then you can confidently conclude that you have a control predicate. If they fail, you can conclude nothing.

For-CPs

Some control predicates allow both subject positions to be explicitly filled. The subject of the infinitive appears in front of the complementizer for.
[latex] \setcounter{ExNo}{15} \ex. \a. *John seemed for Mary to read the book. \b. John wished for Mary to leave. [/latex]
Applying the for-CP test

Make a sentence with the predicate you're testing (wish) → Chi wishes to leave

Add a subject in the lower clause after for → Chi wishes for Drey to leave.

Check the grammaticality. If the sentence is grammatical, then it's a control verb.

If the sentence is not grammatical, then you can conclude nothing.

[embed]https://youtu.be/9QzYUes4TkQ[/embed] For-CPs (length: 1m 26s) credit: Morgan Robertson

Partial control.

Some control predicates allow a "mismatch" in semantic features. Sometimes the lower subject can refer to "more" than the higher subject. For instance, a verb like gather requires a plural subject.
[latex] \setcounter{ExNo}{16} \ex. \a. *The student gathered in the park. \b. The students gathered in the park. [/latex] With certain control predicates, the subject of the higher verb can be singular, even if the infinitive selects for a plural subject. This type of a "mismatch" simply isn't permitted with any raising verb. [latex] \setcounter{ExNo}{17} \ex. \a. *The student seemed to gather in the park. \b. The student wanted to gather in the park. [/latex]
Applying the partial control test

Make a sentence with a verb that requires a plural subject → The students gathered in the park.

Put the predicate you're testing (want) into the sentence, but make the subject singular → The student wants to gather in the park

Check the grammaticality

[embed]https://youtu.be/3MdDn68Sw2Q[/embed] Partial control (length: 1m 51s) credit: Morgan Robertson

Analyzing raising structures
There are a number of different predicates that share all the raising properties. Note that the term "predicate" includes verbs, adjectives, and even prepositions. The following are standardly recognized raising predicates; this is not an exhaustive list.

seem (Khufu seems to run fast.)

appear (Khufu appears to run fast.)

be likely (Khufu is likely to run fast.)

be about (Khufu is about to run fast.)

...

The analysis of raising is fairly straightforward. The subject moves from the lower clause into spec-TP of the higher clause. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ $[$pres$]$}} child {node {VP} child {node {V$'$} child {node {V \\ seem}} child {node {TP} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {Voice$'$} child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node {V \\ run}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=2em] child {coordinate(left1)} child {node(high) {Khufu} edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VP} child {node {V$'$} child {node(V) {V+T \\ seem+$[$pres$]$}} child {node {TP} child {node(mid) {$t_{DP}$}} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node(low) {$t_{DP}$}} child {node {Voice$'$} child {node(voice2) {Voice$+$V \\ $\emptyset$+run}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node(v2) {$t_V$}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } }; \draw[->, rounded corners=2pt] (low)--++(west:2)-|(mid); \draw[->, rounded corners=2pt] (mid)--++(west:2)-|(high); \draw[>-, round corners=2pt, dashed] (V)--++(west:1)-| ++(west:.5)--(T);\draw (left1)--(right1); \draw[->, round corners-2pt, dashed] (v2)--++(west:1)-|(voice2); \draw (left2)--(right2); \end{tikzpicture} [/latex] Let's break down the important things to note about this analysis.

There is [pb_glossary id="655"]cyclic movement[/pb_glossary]. The subject moves through the specifier of TP in the lower clause. The reason that this movement is necessary is that it satisfies the Subject Condition: Spec-TP must be filled at S-structure.

The TP dominating the infinitive doesn't have a CP. (This is the only time in this class we'll see a TP without a CP.) The reason that there is no CP here is that we cannot detect a CP here: we cannot use any kind of non-finite complementizer (e.g. for) with raising predicates.

There is no VoiceP in the main clause. This is because there is no external argument: seem doesn't select for a subject.

In a sense, raising verbs present a particularly interesting and complicated study in long-distance dependencies in syntax. Building on the idea in the previous chapter that phrases can move to the subject position, raising verbs illustrate that this movement can cross clauses sometimes.

Be careful! The following three things are not the same. Do not confuse them.

Raising verb: A raising verb is a verb which raises a subject out of a lower clause. It involves phrasal movement. Examples of raising verbs in English are seem and appear.

Verb raising: Verb raising is when the head V raises to T (or some other higher functional projection). It is head-movement. In English, only the auxiliaries be and have undergo raising.

Raising to subject: Raising to subject describes when a phrase moves to the subject position. This can be clause-internal---movement from the verb phrase to spec-TP, or cross-clausal: movement across a raising verb.

Analyzing control structures

It is temping to simply treat control verbs like want analogous to raising verbs, except that the subject starts in the main clause. This is shown in the tree below. (This is the D-structure.)
[latex] \begin{tikzpicture}[mytree, scale=.8] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ $[$pres$]$}} child {node {VoiceP} [sibling distance = 7em] child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {Voice$'$} [sibling distance = 6em] child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V \\ want}} child {node {TP} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node {Voice$'$} child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node {V \\ run}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \end{tikzpicture} [/latex] However, this representation is wrong. It violates two rules of the syntactic theory we've built.

The selectional requirements of run haven't been met. (Violates [pb_glossary id="656"]selection[/pb_glossary].)

Spec-TP of the lower clause isn't filled. (Violates the [pb_glossary id="657"]Subject Condition[/pb_glossary].)

Generally, if we allow trees like that above, then we need to stipulate that sometimes verbs don't select, and sometimes spec-TP doesn't have a specifier. This is a bitter pill because it really weakens the overall theory we've built.

Instead, to satisfy all of our constraints and keep the theoretical principles we've established, we are forced to posit a silent argument. We need something to be the subject of the lower clause. We call this silent argument PRO, pronounced "big pro." PRO is a pronoun, but it's a special pronoun. It only appears in particular configurations in order to satisfy grammatical constraints.
[latex] \begin{tikzpicture}[mytree, scale=.8] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ $[$pres$]$}} child {node {VoiceP} [sibling distance = 7em] child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {Voice$'$} [sibling distance = 6em] child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V \\ want}} child {node {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node {DP} child {node {D$'$} child {node {D \\ PRO}} } } child {node {Voice$'$} child {node {Voice \\ $\emptyset$}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node {V \\ run}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \end{tikzpicture} [/latex] [latex] \begin{tikzpicture}[mytree, scale=.8] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} [sibling distance=2em] child {coordinate(left)} child {node(high) {Khufu} edge from parent[draw=none]} child {coordinate(right)} } child {node {T$'$} child {node(T) {$t_T$}} child {node {VoiceP} child {node(low) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T \\ want+$\emptyset$+$[$pres$]$}} child {node {VP} child {node {V$'$} child {node(V) {$t_V$}} child {node {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} child {node {DP} child {node {D$'$} child {node(PROhigh) {D \\ PRO}} } } child {node {T$'$} child {node {T \\ to}} child {node {VoiceP} child {node(PROlow) {$t_{DP}$}} child {node {Voice$'$} child {node(voice2) {Voice$+$V \\ $\emptyset$+run}} child {node {VP} child {node {V$'$} child {node {V$'$} child {node(v2) {$t_V$}} } child {node {AdvP} [sibling distance=2em] child {coordinate(left2)} child {node {fast} edge from parent[draw=none]} child {coordinate(right2)} } } } } } } } } } } } } } } } }; \draw (left)--(right); \draw (left2)--(right2); \draw[->, rounded corners=2pt] (PROlow)--++(south:2)-|(PROhigh); \draw[->, rounded corners=2pt] (low)--++(west:2)-|(high); \draw[->, rounded corners=2pt, dashed] (v2)--++(west:1)-|(voice2); \draw[->, dashed] (T)|-(Voice); \draw[->, dashed] (V)-|(Voice); \end{tikzpicture} [/latex] PRO solves our problems because it's a DP, and it is merged in spec-VoiceP and moves to spec-TP, just like any other subject. It's "special" because it only ever appears in nonfinite clauses.
The idea behind PRO is that it acts kind of like an anaphor. Its reference must be fixed by some c-commanding expression, i.e., it has to be [pb_glossary id="659"]bound[/pb_glossary]. In all the previous examples, PRO is "controlled" by the subject.

Note though that PRO isn't actually an anaphor. PRO obeys its own rules, and has its own distribution. For instance, PRO can be "arbitrary," as in the following sentences.
[latex] \setcounter{ExNo}{18} \ex. \a. It's important to exercise. \b. To exercise is important. [/latex] The verb exercise needs a subject, so PRO must be present in these structures. But it gets an "arbitrary" reference, meaning something like "we all" or "someone generic."

Confusingly, PRO (or "big PRO") is different from pro (or "little pro").

pro is a pronoun that appears in pro-drop languages (like Swahili or Spanish). It's just a regular pronoun, but it doesn't have phonological content.

PRO only appears in one very particular position: spec-TP of non-finite clauses. It's a pronoun, but it's a special kind of pronoun.

English doesn't have pro, but it has PRO. Swahili has both pro and PRO (but they're in complementary distribution).

Evidence for PRO

You may be skeptical. Postulating the existence of something that is never pronounced and only appears in very particular circumstances should make you wary. This is good! You are acting like a scientist. PRO (or in general, the phenomenon of control of PRO) is a hypothesis. Our theory predicts that such an element should exist --- that is, that there should be something "silent" in the syntax. So we'd like to find empirical evidence for PRO.

In fact, we find a lot of empirical evidence for PRO! This is hypothesis-driven research at its finest. Based on independent principles, our theory forces us to postulate something that seems odd. But then we find evidence for it. Here's one piece of evidence from English, which utilizes what you already know about Binding Theory.

Binding theory and PRO.
The following sentences contain anaphors. Based on on Binding Theory ([pb_glossary id="697"]Condition A[/pb_glossary]), an anaphor must be bound. [latex] \setcounter{ExNo}{19} \ex. \a. It's important to trust oneself/yourself. \b. It's useless to nominate oneself/yourself for president. [/latex] If PRO didn't exist, how would the anaphor satisfy Condition A?
Where we're going
The difference between raising and control is widely debated in the current theoretical landscape. The complexities of control in particular continue to be challenge to theoretical models of syntax. (I recommend Syntax II, LING 526, for a deeper investigation into the wonders of PRO.) There is a larger issue though. Often, we come across structures in languages which we don't know how to analyze straight-away. Take for instance the sentence in (21). [latex] \setcounter{ExNo}{20} \ex. It took John an hour to get to work. [/latex] What is the relationship between John and the predicate get to work? Does John move to its surface position? Or is it generated there? This is a question of raising vs. control---which is really the question of Where do arguments start?   At this point, we've introduced the main issues with one kind of phrasal movement: A-movement. We will continue now with the other kind of phrasal movement: A'-movement. As we'll see, the two movements "look" different and also "act" different. In my opinion, A-movement is the "trickier" of the two kinds of phrasal movement. A'-movement is quite straightforward to diagnose---usually.
Exercises

Phrasal movement: A'-movement

juliedoner — Wed, 15 Nov 2023 05:29:39 +0000

Learning Objectives

By the end of this chapter, you should be able to,

recognize and represent A'-movement

determine instances of A'-movement

understand the difference between A- and A'-movement

In the last two chapters, we've been exploring the movement of phrases. This involved substantial discussion of argument structure, and in particular, the position of subjects.
In this chapter, we'll explore the second side of phrasal movement. The cases we'll look at below are in some ways more "obvious" examples of movement. We will clearly see that some constituent is being pronounced in a place where we do not expect it to be pronounced. Some examples of the movement that we'll be looking at here are given below.
[latex] \ex. \a. Who did Tamara sue? \hfill \textit{Wh-}question \b. What will Emma buy? \hfill \textit{Wh-}question \c. This book, I read. \hfill Topicalization [/latex]
As we can see in the above examples, some pre-sentential element corresponds to some "gap" later in the sentence. In all three cases, the first noun phrase corresponds to the object, which we expect to be pronounced after the verb. It is more obvious in (1) that something has moved, because not only is there something in a position that it shouldn't be, but there's also something missing after the verb where the object should be. We generally refer to this empty space as a "gap."[footnote]In this book, the terms "gap" and "trace" are functionally interchangeable. Though it should be noted that "gap" is not a formal term.[/footnote]

This kind of movement is clearly phrasal movement, as it can involve moving more than just one word (Which book will Emma buy?). Crucially however, this type of movement does not affect argument structure; who in (1a) is the object the verb, wherever we move it. Or stated differently, this type of movement can doesn't change the grammatical role of the element that has moved. The movements in the previous chapter are different: all the phrasal movements we've previously considered "created" subjects. For instance, moving the object of unaccusative V turned this object into a subject.

One of the main findings of the last few decades is that phrasal movement comes in two varieties. We looked in the last two chapters at A-movement. The examples above are called A'-movement (pronounced "A-bar movement").

A-movement : Movement to an argument position. A'-movement : Movement to a non-argument position.
The two kinds of movement have distinct, identifiable characteristics. Generally, A-movement "creates" subjects (and objects) by moving them to the "subject (and object) positions." For example, the A-movement cases we've seen have all involved movement to spec-TP, which is the "subject position" in English (and many many languages). In contrast, A'-movement does not "create" subjects and objects, it merely moves them around. In the example in (1c) above, the book is an object of read, but it's been moved to the front of the sentence. The movement to the front of the sentence didn't affect the "objecthood" of the book; it was for other reasons. So this is the core distinction between A- and A'-movement. That said, both kinds of movement also have a "signature," i.e., a core set of properties that we can use to diagnose whether something has moved via A- or A'-movement.
The landing site of A'-movement

As the above examples demonstrate, things can be moved to a pre-sentential position (meaning a position in front of the sentence). What is that position?The only phrase above TP is CP, therefore, this movement must target (i.e., land in) the specifier of CP. So [pb_glossary id="680"]wh-movement[/pb_glossary] lands in spec-CP.
[latex] \begin{tikzpicture}[mytree] \node(top) {CP} [sibling distance = 5em] child {node(wh) {DP \\ who}} child {node {C$'$} child {node {C$_{q}$+T \\ $\emptyset$+$[$past$]$}} child {node {TP} [sibling distance = 3em] child {coordinate(left)} child {node(bot) {Tarmara sue $t$} edge from parent[draw=none]} child {coordinate(right)} } }; \draw (left)--(right); \draw[->, rounded corners, red] ([xshift=2.7em]bot.south)--++(south:2ex) -| (wh.south); \end{tikzpicture} [/latex]
In general, the strategy for wh-questions in English is: "Move the wh-word to the front of the sentence, and move T to C.'' Notice that since this is a [pb_glossary id="681"]question C[/pb_glossary] (C_q), we expect to see T-to-C movement, just like with [pb_glossary id="682"]polar questions[/pb_glossary] (i.e., Did John see Mary?).
One tricky thing about wh-question formation in English involves subjects (again!). Consider the subject wh-questions below. [latex] \setcounter{ExNo}{1} \ex. \a. Who will talk to Brian? \b. What has happened? [/latex] Although it looks like there's no movement, such examples are in fact perfectly compatible with the description above. Some of the movements here are "string vacuous," meaning that they do not affect the linear order of the constituents. In the tree below, A-movement is represented in black, while A'-movement is represented in red. The idea here is that since we expect T-to-C movement to apply in question formation, the subject wh-word must move to spec-CP to yield the observed linear order. [latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node(whtop) {DP \\ who}} child {node {C$'$} child {node(C) {C$_{q}$+T \\ $\emptyset$+will}} child {node {TP} child {node(S) {$t_{DP}$}} child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Sbottom) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V \\ $\emptyset$+talk}} child {node {VP} child {node {V$'$} child {node(V) {$t_V$}} child {node {PP} [sibling distance=2em] child {coordinate(left)} child {node {to Brian} edge from parent[draw=none]} child {coordinate(right)} } } } } } } } }; \draw (left)--(right); \draw[->, rounded corners] (Sbottom)--++(south:1em)-|(S); \draw[->, red, rounded corners] (S.west)--++(west:2em)-|(whtop); \draw[->, dashed] (T.west)--++(west:2em)-|(C); \draw[->, dashed] (V.west)-|(Voice); \end{tikzpicture} [/latex] As usual, things get more complicated when do-support comes into the picture. The problem can be illustrated in the following sentence. [latex] \setcounter{ExNo}{2} \ex. Who called? [/latex] We expect to see something like *Who did call? This is because T moves to C in questions, and therefore shouldn't be able to appear on the verb. There are a couple of solutions to the puzzle --- but they all involve simply outright stipulating the solution.
For our purposes, we'll assume that T-to-C movement applies, and then a "magical" (=morphological) process allows the complex to come together with V after syntax.

Many sentences will now have at least two phrasal movements. There will almost always be something moving to spec-TP (unless there is an expletive subject), and then there might be something moving to spec-CP. There is a general and robustly attested "ordering" of movements. A-movement always precedes A'-movement.  Practically, this means that you should move to spec-TP before you move anything to spec-CP.

[embed]https://www.youtube.com/embed/7Ucj7OZ8STs[/embed] Subject wh-question (length: 3m 5s)

Pied-Piping

Often, more than just the wh-element moves. This is called pied-piping.
[latex] \setcounter{ExNo}{3} \ex. \a. Which picture did John see? \b. How big is the box? \c. Whose sister did John visit? [/latex]
In most cases, there's a very intuitive explanation for why pied-piping happens: you simply can't separate some things. For instance, can we ever separate a determiner and it's corresponding NP? Likewise, how could we get whose to move, since it's not even a constituent?
[latex] \begin{tikzpicture}[mytree] \node(top) {CP} [sibling distance = 8em] child {node(wh) {DP} [sibling distance = 5em] child {node {D$'$} child {node {D \\ which}} child {node {NP} child {node {N$'$} child {node {N \\ picture}} } } } } child {node {C$'$} [sibling distance = 5em] child {node(C) {C$_{q}$+T \\ $\emptyset$+$[$past$]$}} child {node {TP} child {node(S) {DP} [sibling distance = 1em] child {coordinate(left1)} child {node {John}edge from parent[draw=none]} child {coordinate(right1)} } child {node {T$'$} child {node(T) {$t_{T}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {Voice+V \\ $\emptyset$+see}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node(O) {$t_{DP}$}} } } } } } } }; \draw (left1)--(right1); \draw[->, red, rounded corners] (O)--++(south:2em)--++(west:27em)|-(wh.west); \draw[->, dashed, rounded corners] (V)--++(west:1em)-|(Voice.south); \draw[->, rounded corners] (Slow)-|([yshift=-3em]S.south); \draw[->, dashed, rounded corners] (T)--++(south:2em)-|(C.south); \end{tikzpicture} [/latex]
Evidence for movement
As always, we want to support our analysis with data. It's not enough to say, "Look, the wh-word is at the front of the sentence, it must have moved there." We want to show empirically that movement has occurred.
Binding.

Binding Theory and phrasal movement interact. The following sentences illustrate that movement must have taken place.
[latex] \setcounter{ExNo}{4} \ex. \a. Which picture of himself$_{i}$ did John$_{i}$ see? \b. Which picture of John$_{i}$ did he$_{i}$ see? [/latex] In (5a), we have an anaphor. Recall that anaphors are subject to [pb_glossary id="697"]Condition A[/pb_glossary] of Binding Theory. In order for that to happen, at some point himself must have been bound by John. That must have happened before movement. On the other hand, in (5b), we have an R-expression, which is subject to [pb_glossary id="641"]Condition C[/pb_glossary] of Binding Theory. Here, the movement appears to "rescue" the sentence, since if picture of John were pronounced as the object of the verb, it would be ungrammatical: *He_i saw the picture of John_i.
Languages with overt C.

Some languages actually pronounce the complementizer, like Irish:[footnote]There are a few different complementizers in Irish, which interact with A'-movement in interesting ways. The superscript "L" indicates one type of complementizer. [/footnote]
[latex] \setcounter{ExNo}{5} \exg. Cad a\textsuperscript{L} t\'a sa seomra\\ what \textsc{C} is in.the room\\ \trans `What is in the room? [/latex] After the wh-word cad, the complementizer a^Lappears, showing that the wh-word comes before (=above) C.
Wanna contraction.

There are strong intuitions about when it's possible to contract want+to into wanna in English.
[latex] \setcounter{ExNo}{6} \ex. \a. Does John wanna talk to Mary? \b. Who does John wanna talk to? \c. *Who does John wanna talk to Mary? [/latex] This pattern suggests that you can contract wanna in (7b) because the wh-word who started at the end of the sentence. In (7c) on the other hand, the gap is the subject position of the nonfinite clause: Who does John want __ to talk to Mary? We can explain this ungrammaticality if we simply say that you can't contract across traces.
Other instances of a'-movement

Embedded questions
Questions can be "embedded" under certain verbs: [latex] \setcounter{ExNo}{7} \ex. \a. Bill knows [ who saw Mary ] \b. Susan wondered [ how Mary fixed the car ] \c. Sarah decided [ which book she should buy ] [/latex]
Embedded questions are formed exactly like regular questions, except that there's no T-to-C movement in the lower clause.

Relative clauses.
Another place where we see A'-movement, in particular wh-movement, is relative clauses. [latex] \setcounter{ExNo}{8} \ex. \a. the woman [ who I saw {\underline{\hspace{10pt}}} ] is smiling \b. the man [ who {\underline{\hspace{10pt}}} laughs ] arrived. \c. I licked the fork [ which I eat with {\underline{\hspace{10pt}}} ] [/latex] Relative clauses (in English) involve the following properties:

There's a gap (a trace) inside of a CP

The CP is adjoined rightward to NP

There is A'-movement

There is no T-to-C movement inside of the relative clause.

The following is the representation of The woman who I saw was happy.
[latex] \begin{tikzpicture}[mytree] \node(top) {CP} child {node {C$'$} child {node {C \\ $\emptyset$}} child {node {TP} [sibling distance = 9em] child {node(S) {DP} [sibling distance = 5em] child {node {D$'$} child {node {D \\ the}} child {node {NP} child {node {N$'$} child {node {N$'$} child {node {N \\ woman}} } child {node {CP} child {node {DP} child {node {D$'$} child {node(ophigh) {D \\ who}} } } child {node {C$'$} child {node {C$_{RC}$ \\ $\emptyset$}} child {node {TP} child {node(S2) {DP \\ I}} child {node {T$'$} child {node(T2) {$t_{T}$}} child {node {VoiceP} child {node(S2low) {$t_{DP}$}} child {node {Voice$'$} child {node(Voice) {V+Voice+T \\ see+$\emptyset$+$[$past$]$}} child {node {VP} child {node(V2) {$t_V$}} child {node(t) {$t_{DP}$}} } } } } } } } } } } } child {node {T$'$} [sibling distance = 5em] child {node(T) {T+V \\ $[$pres$]$+\textsc{be}}} child {node {VP} child {node {V$'$} child {node(V) {$t_{V}$}} child {node {VoiceP} child {node(Slow) {$t_{DP}$}} child {node {Voice$'$} child {node(voi) {Voice+V \\ $\emptyset$+smiling}} child {node {VP} child {node {V$'$} child {node(V2a) {$t_V$}} } } } } } } } } }; \draw[->, dashed] (V2a.west)-|(voi.south); \draw[->, dashed] (V)-|(T); \draw[->, dashed] (T2)--++ (south:1em)|-(Voice.west); \draw[->] (S2low) --++(west:3em)-|(S2); \draw[->, red, rounded corners] (t)--++(south:2em)-|(ophigh); \draw[->, dashed] (V2.west)-|(Voice.south); \draw[->] (Slow.west)..controls +(west:2) and +(east:1)..(S); \end{tikzpicture} [/latex]
This tree looks much more complicated than it actually is. The trick here is realizing that there are two clauses. If I take away the constituent who I saw, I would just have the sentence, The woman is smiling, and we know how to draw that tree: the subject the woman is selected by smile and so starts in spec-VoiceP and then moves to spec-TP. The relative clause is entirely independent of this process. You build the relative clause and attach it to the NP as a modifier.

In English, there is a relatively large amount of variation in how you pronounce a relative clause. The following sentences all have identical structures.
[latex] \setcounter{ExNo}{9} \ex. \a. The woman [ who I saw {\underline{\hspace{10pt}}} ] is smiling. \b. The woman [ that I saw {\underline{\hspace{10pt}}} ] is smling. \c. The woman [ I saw {\underline{\hspace{10pt}}} ] is smiling. [/latex]
In all three cases, there has been wh-movement of who to spec-CP of the relative clause. In the first case, we pronounce who. In the second case, we don't pronounce who, but we pronounce the complementizer that. And in the third case, we pronounce neither. Since this doesn't appear to affect the meaning at all, we assume that the structure is identical in all cases.

In English, there appears to be a rule of pronunciation that says that you cannot pronounce both a wh-word and a complementizer at the same time. This has been called the Doubly-Filled Comp Filter. (Note that there is in fact dialectal variation with respect the Doubly Filled Comp filter. In some dialects, it's actually acceptable to pronounce both who and that.)

Relative clauses should not be confused with complex-NPs. The example in (11a) is a relative clause. The example in (11b) is a complex-NP. [latex] \setcounter{ExNo}{10} \ex. \a. The lie [ that Mary told {\underline{\hspace{10pt}}} ] was bad. \b. The lie [ that Mary went to Moscow ] was bad. [/latex] Relative clauses have gaps, which we analyze as movement, while complex-NPs do not.

Cross-linguistic variation in A'-movement

Not all languages display wh-movement to spec-CP. For instance, Swahili has wh-in situ.
[latex] \setcounter{ExNo}{11} \exg. Duma a-li-ona nani\\ Duma \textsc{3sg}-\textsc{past}-see who\\ \trans `Who did Duma see?' [/latex] The typological variation in (non-)movement of wh-expressions is fascinating, and it has inspired a huge amount of literature. It is a major focus of Syntax II, LING 526.
Differences between A- and A'-movement

Generally, syntax can be viewed as the investigation of silence. Many of the core syntactic problems arise in places where we expect to see something pronounced, but instead we see nothing. It is the job of the syntactician to discover why there's silence.

Option #1 : There's a silent element. For instance, pro and PRO are silent elements. They're DPs that aren't pronounced.

Option #2 : There's movement.

Option #2a : It's head-movement. We recognize this movement because it forms new words.

Option #2b : It's phrasal-movement.

Option #2b.i : It's A-movement.

Option #2b.ii : It's A'-movement.

Option #2: There's [pb_glossary id="683"]ellipsis[/pb_glossary]. In this case the silence arises through "deletion." We won't address ellipsis phenomena in this class. See LING 526.

For instance, in clefts there's clearly a gap, and if we want to analyze this structure, we have to figure out how this gap was formed. Is there a silent element? Is there movement? If there's movement, what kind of movement?
[latex] \setcounter{ExNo}{12} \ex. It was Mary that John saw {\underline{\hspace{10pt}}}. [/latex]
Clearly, this cannot be a case of head-movement; the gap does not contribute to making a new word. So this is either a case of a silent element, or phrasal movement.

In this chapter, we'll discover that different kinds of phrasal movements have a "signature." That is, it is possible to identify through diagnostic tests whether something a) has moved at all, and b) whether it has undergone A or A'-movement.

A'-movement diagnostics

Perhaps the defining characteristic of A'-movement is that it is unbounded: the moved element can be any number of clauses away from the gap.
[latex] \setcounter{ExNo}{13} \ex. \a. Who did John see {\underline{\hspace{10pt}}} \b. Who does Mary think [$_{CP}$ that John saw {\underline{\hspace{10pt}}} ] \c. Who did Bill say [$_{CP}$ that Mary thinks [$_{CP}$ that John saw {\underline{\hspace{10pt}}} ] ]? \d. Who did Carol hear [$_{CP}$ that Bill said [$_{CP}$ that Mary thinks [$_{C}$ that John saw {\underline{\hspace{10pt}}} ] ] ]? [/latex]

The term "unbounded" here has nothing to do with Binding Theory. As it applies to movement, "boundedness" refers to the ability to move across clausal domains.

Descriptively, A'-movement can cross an infinite amount (up to cognitive limitations) of CPs. This makes it different from A-movement, which cannot cross any CPs. That is, I cannot raise to subject across a CP:
[latex] \setcounter{ExNo}{14} \ex. *John thinks that {\underline{\hspace{10pt}}} saw Mary. [/latex] However, A'-movement is not an "anything goes" movement. That is, there are some movements that are impossible. For instance, it's not possible to move to a position that doesn't c-command the trace. That is, you cannot move into spec-CP of a CP in the subject position. [latex] \setcounter{ExNo}{15} \ex. \a. {} [$_{CP}$ That John came home early ] frightened Mary. \b. * [$_{CP}$ Which student (that) John came home early ] frightened $t_{DP}$ [/latex] One way to understand this is that traces are kind of like anaphors in that they are subject to a version of Condition A: they have to be c-commanded by a co-indexed element. More interestingly, even though A'-movement can cross CPs, it cannot cross all CPs. There are some clauses that you simply cannot move out of. These are called islands. It's a metaphor: the constituent is an island if nothing can get "off of" (i.e., move out of) the island. In addition to unboundedness, islands are part of the "signature" of A'-movement. That is, as a rule of thumb, A'-movement "respects" islands, meaning that all A'-movement are sensitive to islandhood diagnostics.  There are many islands; we'll cover four in this class.
Complex-NP Constraint.

Recall Complex-NPs, which are CP complements to N. (They don't have a gap). You cannot move out of CP that is complement to NP.
[latex] \setcounter{ExNo}{16} \ex. \a. Sarah heard the lie [$_{CP}$ that Susan sued Bill ] \b. *Who did Sarah heard the lie [$_{CP}$ that Susan sued {\underline{\hspace{10pt}}} ] [/latex]
Adjunct island.

You cannot do A'-movement out of a clause that is adjoined. Adjoined clauses are typically headed by complementizers like because, if, when, after, before.
[latex] \setcounter{ExNo}{17} \ex. \a. John went home [$_{CP}$ after he broke the vase ] \b. *What did John go home [$_{CP}$ after he broke {\underline{\hspace{10pt}}} ] \ex. \a. Sarah ducked [$_{CP}$ because Ben kicked the ball ] \b. *Who did Sarah duck [$_{CP}$ because {\underline{\hspace{10pt}}} kicked the ball ] [/latex]
Wh-island constraint.
You cannot do A'-movement out of something that already has a wh-movement. For instance, you cannot do wh-movement out an embedded question. [latex] \setcounter{ExNo}{19} \ex. \a. John knows [$_{CP}$ who broke the vase ] \b. *What does John know [$_{CP}$ who broke {\underline{\hspace{10pt}}} ] [/latex] Compare this with (21), which doesn't have an embedded question. [latex] \setcounter{ExNo}{20} \ex. \a. John knows [$_{CP}$ that Bill broke the vase ] \b. What does John know [$_{CP}$ that Bill broke {\underline{\hspace{10pt}}} ] [/latex]
Subject islands.

You cannot move out of a constituent that is in the subject position.
[latex] \setcounter{ExNo}{21} \ex. \a. John read [$_{DP}$ a book by Tolstoy ] \b. Who did you [$_{DP}$ read book by {\underline{\hspace{10pt}}} ]? \ex. \a. {} [$_{DP}$ A book by Tolstoy ] is on the table. \b. *Who is [$_{DP}$ a book by {\underline{\hspace{10pt}}} ] on the table? [/latex] The purpose of identifying a particular signature for A'-movement is that it helps us identify kinds of movement when we encounter new data. For instance, the following sentences all contain gaps/unpronounced elements. [latex] \setcounter{ExNo}{23} \ex. \textbf{Raising-to-Object/ECM}\\ John made Carol out {\underline{\hspace{10pt}}} to be a liar. \ex. \textbf{\textit{Tough}-construction}\\ This book was tough to read {\underline{\hspace{10pt}}} . \ex. \textbf{Pseudo-clefts}\\ Mary was who I talked to {\underline{\hspace{10pt}}} . [/latex] If we want to understand the syntax of such sentences, then we need to figure out how that gap came into existence. We can apply the tests above (including a test for unbounded dependendencies) to determine if there's movement, and what kind of movement it is.
A-movement diagnostics

We've focused here on a few island constraints, which diagnose A'-movement. There are ways to diagnose A-movement as well. The clearest way to diagnose A-movement is to use binding. A-movement creates new binding dependencies; A'-movement doesn't.
[latex] \setcounter{ExNo}{26} \ex. \a. Mary$_i$ seems to herself$_i$ {\underline{\hspace{10pt}}} to be tired. \b. *Who$_i$ does it seem to herself$_i$ that John likes {\underline{\hspace{10pt}}} ? \c. *Mary$_i$, it seems to herself$_i$ that John like likes {\underline{\hspace{10pt}}} . [/latex] When you raise to subject across seem, the resulting movement can feed Condition A: the subject can now antecede an anaphor. A'-movement does not permit this. If I question or topicalize some constituent, it cannot antecede an anaphor.
Exercises

(4)	American Sign Language (ASL)
	1INDEX 1FORCE2 2GIVE1 MONEY
	'I'll force you to give me the money.'

(5)	American Sign Language (ASL)
	a.	TWO-DAYS-AGO INDEX2 s-u-e IX3a FLOWER 2GIVE-CL:👌3a
		'Did you give Sue a flower two days ago?'

Symbol	Description	Meaning
SIGN++	two following plus symbols	Indicates that the sign is repeated in a morphological process called reduplication.
SIGN - - - - -	five following hyphens	Indicates that part of a sign is held in place while the other hand continues signing.
SIGN^SIGN	two signs joined by a caret	Indicates the formation of a compound sign.
SIGN-SIGN	two signs joined by a hyphen	Indicates that multiple words in the meta language are required to translate a single sign in the object language.
s-i-g-n	lower case letters joined by hyphens	Indicates a fingerspelled word.
"xxx"	lower case letters enclosed in quotation marks	Indicates a gesture.

(13)	French Sign Language (LSF)
		hn
	re
	PLEASE	INDEX2	BITE
	'Please, bite it.'

Abbreviation	Gesture	Meaning
aff	repeated head nods	affirmation
bl	body lean in a particular direction
hn	a single head nod	affirmation or imperative marker
neg	headshake and negative facial expression	negation
re	raised eyebrows	marks subordinate clauses
t	raised eyebrows and forward tilt of the head	topicalized constituent
wh	lowered eyebrows	content (wh-) question
y/n	raised eyebrows and head forward	yes-no question
/xxx/	various mouth gestures
)(	sucked in cheeks	'small'
()	blown out cheets	'big'

(1)	(a)	kumaʔin	ng	mangga	ang	babaʔe	[Tagalog; CREDIT?]
		eat.PAST?	NG	mango	ANG	girl
		'The girl ate a mango.'
	(b)	kumakaʔin	ng	mangga	ang	babaʔe
		eat.PRESENT?	NG	dance	ANG	girl
		'The girl is eating a mango.'
	(c)	kakaʔin	ng	mangga	ang	babaʔe
		eat.FUTURE?	NG	dance	ANG	girl
		'The girl will eat a mango.'

(2)	(a)	kʊɽiː	nət͡ʃt͡ʃiː	[Punjabi][footnote]The Punjabi data in these examples was provided by an anonymous consultant.[/footnote]
		girl.SG.FEM	dance.PAST.SG.FEM
		'The girl danced.'
	(b)	mʊndaː	nət͡ʃt͡ʃɪaː
		boy.SG.MASC	dance.PAST.SG.MASC
		'The boy danced.'
	(c)	kʊɽiː	nət͡ʃt͡ʃeːgiː
		girl.SG.FEM	dance.FUTURE.SG.FEM
		'The girl will dance.'
	(d)	mʊndaː	nət͡ʃt͡ʃeːgaː
		boy.SG.MASC	dance.FUTURE.SG.MASC
		'The boy will dance.'

(1)	noun form	verb form
a.	brood	breed
b.	doom	deem
c.	food	feed

tidy	'clean, organized'	untidy	'not clean, not organized'
happy	'in a pleasurable state of mind'	unhappy	'not in a pleasurable state of mind'
like	'similar to'	unlike	'not similar to'
fair	'equal, just'	unfair	'not equal, not just'

(1)	a.	a pumpkin pie
	b.	down the stairs
	c.	very hungry
	d.	after the truly serious and unfortunate events of last night
	e.	rode their bikes to school every day

(2)	a.	A pumpkin pie rested on the windowsill.
	b.	I rolled a ball down the stairs.
	c.	Eliud Kipchoge was very hungry after winning the marathon.
	d.	I decided to talk to my therapist after the truly serious and unfortunate events of last night.
	e.	My children rode their bikes to school every day until it got too cold.

(3)	a.	*pumpkin a pie
	b.	*the stairs down
	c.	*hungry very
	d.	*after truly the serious unfortunate and night events last of
	e.	*their rode every day to bikes school

(6)	Boumaa Fijian
	au	na	talai	Elia	i	'Orovou
	I	FUTURE	send	Elia	to	'Orovou
	'I'll send Elia to 'Orovou.'

(7)	Lezgian
	duxturrin	patariw	fena
	doctors	to	went
	'She went to doctors.'

(3)	a.	The balls kicked the toddler across the field.
	b.	The field kicked the balls across the toddler.
	c.	The kicker toddled the balls across the field.
	d.	The toddlers kicked the ball across the field.

Full phrase	Contracted form
am not	?
are not	aren't
is not	isn't

Rule	Prescriptively incorrect example	Prescriptively correct example
Do not split infinitives.	I want to quickly eat* so I’m not late.*	I want to eat quickly* so I’m not late.*
Do not end a sentence with a preposition.	That’s the person I sat next to.	That’s the person next to whom* I sat.*
Avoid the passive voice.	This book has been read by many students.	Many students have read this book.
Say and I not and me.	You walked past my brother and me.	You walked past my brother and I.
Do not use double negatives.	I didn't say nothing!	I didn't say anything!

	Older variant	Newer variant
Comparative form	more fun	funner
Superlative form	most fun	funnest

(1)	a.	A police officer shot a bystander.
	b.	A bystander was shot by a police officer.

(3)	a.	The wind bent the tree.
	b.	The tree was bent and ravaged by the constant windstorms.

(4)	a.	I ate dinner.
	b.	*Me ate dinner.
	c.	Dinner was eaten by me.
	d.	*Dinner was eaten by I.

(5)	a.	My friend and I ate dinner.
	b.	%My friend and me ate dinner.
	c.	Dinner was eaten by my friend and me.
	d.	%Dinner was eaten by my friend and I.

(3)	a.	Language is infinite.
	b.	I believe that language is infinite.
	c.	Javier says that I believe that language is infinite.
	d.	Amita thinks that Javier says that I believe that language is infinite.

(4)	a.	Zofia’s idea
	b.	Zofia’s brother’s idea
	c.	Zofia’s brother’s friend’s idea
	d.	Zofia’s brother’s friend’s mother’s idea

(5)	a.	the hole in the ground
	b.	the tree in the hole in the ground
	c.	the limb on the tree in the hole in the ground
	d.	the branch on the limb on the tree in the hole in the ground

a.	sofa
b.	beautiful
c.	bookshelf
d.	pastries
e.	recliner chair
f.	reassessment
g.	sunny
h.	spilled
i.	strawberry pie filling
j.	distressing

(1)	a.	We wrote a story.
	b.	You gave us a gift.
	c.	A story was written by us.
	d.	We were given a gift.

(2)	a.	*Us wrote a story.
	b.	*You gave we a gift.
	c.	*A story was written by we.
	d.	*Us were given a gift.

	Categorical data	Quantitative data
Observational data		Corpus study
Targeted data	Elicitation	Experiment

(1)	a.	The prime minister of Canada ate a salami sandwich.
	b.	*The ate of salami a Canada ministerial prime sandwich.