LING 2120 Syntax

what is acquired first, grammatical or lexical items

grammatical morphemes

Children acquiring a language with a rich inflectional morphology (like Hungarian) acquire grammatical morphemes earlier than kids acquisition a language with poor inflectional morphology (like English

Most early words and sentences don’t have:

- plural marking

- determiners

- tense and agreement

- auxiliaries/copula verbs and modal verbs

Syntax relates to…

how words in a sentence are organized into phrases and sentences that convey meaning

• Part of the study of syntax concerns mapping out the rules that govern how verbs combine with noun phrases and other constituents to yield coherent meanings

Generativist beleive _______ while Constructivist beleive _____

syntactic categories are innate

knowledge of syntactic categories emerge from the

input

Need to use syntax to learn about the …

semantics (syntactic bootstrapping) but

the child also needs the semantics to learn the syntax (semantic bootstrapping)

• If there are parts that the child already knows, the child can bootstrap themselves into the rest

bootstrapping

refers to the process of using the resources you have at hand to solve a problem by yourself, without explicit help from any external source

Phonological bootstrapping

The use of phonological properties of the speech stream (e.g., acoustic and distributional cues) to identify linguistic units (words) and the boundaries between them (forms only, not meanings

Syntactic bootstrapping

the use of syntactic properties (argument structure) to narrow down the meanings of words

Semantic bootstrapping

learners first learn some basic semantic properties of words, and then they can use this semantic (meaning) information to bootstrap into syntax

pinker: Semantics helps children identify how objects and events they experience in the real world are mapped to syntax

Children first learn some basic semantic properties of words, then they use this semantic information to bootstrap into syntax.

Linguists define lexical categories according to their

grammatical function (e.g., nouns are words that can be preceded by a determiner; verbs are words that can be marked for tense), rather than meaning

• But there are consistent and reliable correspondences between lexical categories (nouns; verbs) and semantic features (objects; events):

- nouns refer to people, animals and physical objects

- verbs refer to actions and changes of stat

Semantic bootstrapping assumes that…

children are able to pick out what is going on (event) and who is doing what, through experience with the real world; i.e., that children map semantic categories to syntactic categories relatively easily

If/Then.

f children know/can learn the basics of sentence structure...Noun vs. verb: If, through experience with the real world, young children can distinguish objects from events

• Subject vs. object: If they know that subjects are typically animate and agents of actions (volitionally causing an event or change of state) and if they know that objects are themes/patients (acted-upon entities)

...And if they know...

• The meanings for some nouns (like dog, cat, daddy, banana)

...Then:

• When they observe a scene where subject, verb and object are consistent with the

correspondences identified above (agent, event, theme), they can determine from sentences like:

(a) The dog is chasing the cat

(b) Daddy is eating a banana

that English is head-initial (VO) and that subjects precede predicates (SVO

nouns are proceeded by…

determiners

verbs are..

marked for tense

1. Children use phonological information to identify phrase and clause

boundaries.

2. Children use semantic/conceptual information to assign syntactic

categories to to sequences of sounds.

3. Children use argument structure information to learn verb meaning

1. phono bootstrapping

2. semantic bootstrapping

3. syntactic boot

The holophrastic stage

• 1 word stage

• At the onset of production, children typically produce single-word utterances

• Adults often interpret these single-word utterances as meaning more than the word itself

• comp is greater than production

in the holophrastic stage, Children show sensitivity to the position of

verb heads and complements at this stage

head

determines category of phrase

complement

phrase that is needed to complete the meaning of the head

english is head…

initial

preferential looking task

To test knowledge of syntax, even when the child is producing only single words (or no words at all)

“The cat chases the dog”

Results: Children spend significantly longer looking at the appropriate screen.

---

Children looked longer at the correct scene, regardless of whether the

language was head-initial or head-final.

• Children’s behavior must be driven by abstract knowledge about sentence

structure, given that novel verbs were used. In other words, their knowledge

of word order is not tied to specific lexical items.

how do you know a single word stage is ending?

repetition of the same word

chained word utterances, more than one word with a pause

do we count morphemes or words?

morphemes MLU

telegraphic stage

The two-word combinations produced by children largely reflect the word order

of the target language:

• Examples:

baby cry = The baby is crying

1 vs 2 word stage

• Children at the one-word stage demonstrate that they know whether their language is head-iniOal or head-final.

• The productions of children at the two-word stage largely reflect the word order constraints of the target language

• speech contains mostly content words

unaccusative verbs

have theme subjects instead of subjects, which moves to sub position

have NO AGENTS. either a patient or theme that originates in the object position and moves to subject position

lexical vs functional

lexical - nouns, verbs ( carry info about meaning)

functional- TP, DP, CP

types of argument structure (3)

1. intransitive ( one argument) Ex: emma walks

2. transitive ex: emma pushes ben

3. ditransitive ex: emma puts a sandwich in her mouth

components of a tree (3)

1. spec

2. head

3. comp

is inflection found within the VP?

no.

because inflectional information such as tense is not part of the argument structure of the verb

• The functional structure must encode the abstract functional syntactic organization of a sentence via functional phrases, like IP (Inflectional Phrase)

note:

- auxiliaries (is, have) and modal verbs (will, should, can etc) are located in I as well.

SA

OA

IND

subject agree

object agree

indicative mood

Do children show evidence of abstract syntactic structure within the functional domain (i.e., above the VP) when they are in the telegraphic stage

yes

finite versus nonfinite

finite is inflected and marked for tense, nonfinite is not and is ungrammatical in isolation

Children’s early multiword utterances contain

nonfinite sentences

is “this writes” finite

yes. it has 3rd person agreement, marked in present tense in English

is “i like toys” finite

cant tell bc there is not agreement for present tense

is “mama will come back” finite

yes. “will”

T/F: child grammar is of a similar nature to adult grammar

true

truncation

underlying syntactic representations are optionally truncated

small clause hypothesis

underlying syntactic representations consist of VP structure and nothing higher

• early grammar is lexical-thematic, functional categories are programmed to emerge/mature later (about age 3

Unique Checking Constraint (UCC)

Tense or agreement are optionally included in the derivation when the UCC is operational

• UCC falls away with maturation

problem with small clause hyp

1. At the same point in development that children produce nonfinite sentences,

they also produce finite (i.e., adult-like) sentences.

2. The same child can alternate between producing nonfinite and finite sentences (optional infinitives)

3. Children’s finite forms must contain the functional structure necessary to express tense (i.e., IP) contra the Small Clause Hypothesis

For every adult sentence, the top node of the syntactic tree is

complementizer phrase (CP)

For children, the structure above XP is not projected: the tree will be

truncated if XP is lower than CP (e.g., IP, VP, NP)

• If a structure projects only up to VP: the result will be nonfinite (no tense marking, i.e. no IP)

• If a structure projects up to IP: the result will be finite, i.e. tense marking is produced

• If a structure is projected all the way up to CP: then all nodes below CP (including IP!) will be projected; evidence for this structure in children’s productions would be e.g. a question with tense marking

can a kid in the same day produce a IP,VP,CP

yes

The canonical position for subjects is the

Specifier of IP. Although it is thought that subjects are probably generated within VP and then raise into the Specifier of IP position, this is where they occur in the adult language

• If IP is missing because the child has projected VP as the highest node in a given sentence, then there is no real place for the subject

T/F:If the top node of the tree is VP, then the grammatical subject position (specifier of IP) will not be available

• Prediction: Subjects will be omitted from truncated structures more than from non-truncated structures (supported: Sano & Hyams 1995)

• When subjects are produced, they will appear in Specifier of VP

true

Crismas effect

OI never occur with WH questions because if a structure is projected all the way up to CP: then all

nodes below CP (including IP!) will be projected; evidence

for this structure in children’s productions would be, e.g.,

a question with tense marking

what can truncation account for? (3)

1. order of verb form with respect to adverb/negator

2. subject omission with root infinitives

3. lack of root infinitives in wh-questions (though English is an exception)

Syntax

- how words in a sentence are organized into phrases and sentences that convey meaning

How verbs combine with np

Consented with the expression of functional aspects of language such as tense, negation, etc

Argument structure

Argument structure- how verbs combine with np and other constituents to yield meanings

LPLA

kids come to know things they have no evidence for

DPLA

- developmental path that kids take in the acq of a particular phenomenon

Why they develop along the path they do

Bootstrapping

the process of using the resources you have at hand to solve a prob by yourself, without help from a source

How do they bootstrap?

1. Through prosody, rhythm, and phonotactics ( called phonotactic bootstrapping)

2. Syntactic bootstrapping- kids exploit the semantic properties of argument structure to narrow down the meanings of verbs

Example:

“She gorped the ball”

→ 2 participants → likely a caused action

“She gorped”

→ 1 participant → likely a self-action

👉 Structure helps them guess the meaning of “gorp”

1. Semantic bootstrapping- learners first learn some basic semantic prop of words, they they use the meaning to bootstrap into syntax

Example:

If a child hears:

“The dog runs”

They think:

“dog” = a thing → noun

“runs” = an action → verb

👉 Meaning helps them build grammar

Pinker

concluded kids should be able to determine that there is a category of words that generally label objects and another category of words that generally label events

Kids come to the lang learning task with assumptions about sentence structure

The dog is chasing the rabbit

dog=agent

Rabbit =patient

The learner can infer that this lang has SVO structure

Pinker extended this to agents that were not people or concrete objects and verbs to not label tangible events

The decision evoked a harsh response

The child can know that nouns can be preceded by determiners (the) so they know decision is a noun

-ed on evoked tells them it is a verb

Subject always comes before the word verb

Structure dependent distributional learning

The procedure of using the rules of the mini grammar to determine the sentence structure of a non transparent sentence. ( like:The decision evoked a harsh response)

What is it called when a child figures out the structures of a non transparent sentence?

Structure dependent distributional learning

When is the one word stage?

16-19 months

T/F:Kids learn the word order of their language quite early. Showed that children who produce only one word at a time, which is the one word stage at 16 to 19 months, correctly interpret sentences like the Big Bird is tickling Cookie Monster. In that the big bird is acting upon the cookie monster not the other way around

true

Do kids use correct order in their own productions?

yes

At what stage do correct adultlike word orders arise?

2/3 word stage

Functional structure

provides info about things like time of event, polarity of the sentence (pos or neg), modality ( event may/could occur)

• Marked by inflectional morphology

Is functional structure variable? Is argument structure?

Yes

NOOOOOOO, the verb must ALWAYS have a subject

What is the glue that words of a sentence are bound together

Functional categories

What is an indication of language disorder above age 4?

Omission of functional categories

telegraphic speech

functional categories are missing

Is functional categories a part of the verbal or nominal domain?

both

Perfective aspect

expresses notion that the event was completed already

Where do all functional things occur in a tree?

ABOVE the verb, they are not a part of the argument structure

SA3a= 3rd person subject agreement

Finiteness- abstract version of tense

• Something has finiteness if it can be located in time

Nonfinite- cannot be located in time

Infinitive- base form of the verb ( to infront)

• Ex: to get

Bare verb- get

Optional infinitives (OI)

nonfinite verbs seem to be optional, as kids alternate between finite and non

Why do kids do this?

Answered by the truncation hypothesis.

Truncation hypothesis

Characterized by the omission of functional categories. omission is not absolute.

There is not a stage in which each and every functional element in child language is missing.Children sporadically produce functional elements.

Everything above a given node is truncated

Rizzi

Put forward a theory that accounts for the optionality of inflection in child language and has a mechanism to deal with observed cross linguistic variation. 

• Argued in every adult sentence, the top node of the syntactic tree is a complementizer phrase (CP)

In the sentence: I think that john is happy, what is the cp?

That

Also wh words

t=

traces ( where things move from)

T’s need..

Transcripts

Rizzi says the cp is always there even if it is…

Empty

Called axiom

Do kids have the axiom of a cp yet?

No. for them the axiom is an xp ( any node in the tree).

2 features of the truncation hypothesis:

1. Any node may be specified as the top of the tree (may go all the way up to CP, or just a VP)

2. Once a node is specified, any node under that must be present in the structure

Ex: cp is present which means everything must be

Top node may be cp/ip/vp/np

Why is omission of functional elements so sporadic in kid speech?

They have the option to project their tree to any node

Where does this evidence come from?

Contingencies in the data. 

T/F:In adult French, finite verbs always go before negation, while infinitive verbs always go after negation.

true

What claim is that the child's tree consists of the VP structure and nothing above that?

small clause hypothesis

In french, negation occurs in front of a ____ verb and after a ____ verb

Nonfinte

Finite

Are optional infinitives an error of Commission in the sense that the child lacks knowledge of the inflectional system?

no

The fact that children obey the distributional properties of finite and non finite verbs shows that…

They have knowledge of the functional structure in their language and they are using this functional structure to determine the position of the verbs.

If the child projects anything below a TP..

An OI will occur

If the top node is TP

It will not occur. 

This is why they are optional, the root is specified variably, meaning sometimes OI occur and sometimes they dont. explained by the truncation hypothesis

If the top node is anything below TP, there is nowhere for a french verb to arise , so it will remain in the..

VP. If it his TP, the verb can rise which gets the verb to the left of negation

Does truncation neatly explain OI in kid language?

yes

Null subject contingencies

found more with OI than finite verbs

T/F: subjects are gen in the VP but rise to the TP spec position.

True

What if the TP is missing with kids? There is no real place for the subject, creating a null subject ( in OI)

Reasons for null subjects:

Telegraphic speech

They think their language is like Italian and allows null subjects. They would need to figure out from their input that is not allowed

Crismas effect

OI NEVER occur with Wh questions

OI occur only when the top node of a tree is below TP, with WH questions, the top node MUST be a cp, therefore, absence of OI

What can the truncation hypothesis not explain?

OI do not occur in many languages

English is unusual with OI (ex: elle sit floor is a nonfinite verb but COULD be an OI. it is hard to tell because bare verbs have non visible inflectional morphology

In english we cant use verb positon to tell us

The emission of an inflectional and functional ailments does not mean that those syntactic categories are absent from the child's language. It just means that what is missing is the axiom that the top node must always be a CP.