GRS LX 700 Observations Language Acquisition and ...€¦ · forms a signiÞcant part of the PLD...

1

Week 1. Introduction

GRS LX 700Language Acquisition

andLinguistic Theory

Observations

! Adults have a grammar of their native language.

! Some things are English. Some are not.! Dog threes cat a chased.

! Anchovies I like, but pepperoni I can’t stand.

! Eight hungry capybaras danced frantically.

! You cannot stop a philosopher with a thesis.

! Mary saw her in the mirror.

! Mary saw her duck in the mirror.

Deduce the rules:

! 1, 2, 3, __ , __ , __? (5, 7, 11)

! Sue said Mary borrowed my book about salsa.

! Sue said that Mary borrowed my book about salsa.

! What did Sue say Mary borrowed?

! What did Sue say that Mary borrowed?

! Who did Sue say borrowed my book about salsa?

! Who did Sue say that borrowed my book about salsa?

! What did Sue say Mary borrowed my book about?

Yet people know this stuff…

! Adult native speakers uniformly andoverwhelmingly agree.

! To know English is to have knowledgeof (how to determine) which sentencesare possible and which are impossiblein English.

! How one comes to have this knowledgeis going to be our primary focus.

Grammar

! People eventually end up with a system withwhich they can produce (and rate) sentences: agrammar.

! Even if a native speaker of English has neverheard either of these sentences before, s/heknows which one is possible in English andwhich one isn’t:

15) Eight very adept sea lions played trombones.16) Eight sea lions very adept trombones played.

How do people know this?

! Every native speaker of English knows thesethings.

! Nobody who speaks English as a first languagewas explicitly taught (growing up) “You can’tquestion a subject in a complement embeddedwith that (if you want dessert, young lady).” or“You can’t use a proper name if it’s c-commanded by something coindexed with it.”

! Trying to use any simple kind of general learningprinciple based on (analogy to) the sentences youget seems almost sure to lead you astray.

2

That’s the setup

! Language involves a complex grammar.

! Adults end up with knowledge of thisgrammar, quite uniformly.

! Children seem to go through advancingstages of language sophistication; they arelearning, the end result being the adultlanguage system.

! The question: What is the nature of thechildren’s learning?

Linguists, great and small

! As linguists trying to figure out the grammaticalsystem of a language, we…! Look at which sentences are grammatical

! Look at which sentences are ungrammatical

! Compare them to describe generalizations about whatthe crucial factors are differentiating the grammaticalfrom the ungrammatical.

! Check the predictions of the hypothesizedgeneralization by looking at more complex sentences.

! Are kids just little linguists?

Kids are not just littlelinguists.

! *What did you see the book about on the table?! *Who did Mary say that bought coffee?

! Eight very adept sea lions played trombones.

! Linguists’ theories: built by considering bothgrammatical and ungrammatical sentences (oftenof a fairly complex type).

! Kids: Don’t hear ungrammatical sentences, noreven all of the grammatical sentences (often of asimpler type).

Why negative evidence?

! A kid hears: Jane said Mary lost her keys.

! Maybe a kid hears this in all three contexts.! So, her can refer to a previous female referent, inside or

outside the sentence.

! A kid hears: Jane said Mary voted for her.! This time, her is Jane.

! So far, the kid never heard such a sentence whereher would be Mary. But it could be the very nextsentence, in principle.

! Maybe after waiting long enough, the kid assumesit’s ungrammatical?! But what’s long enough? And how much can the kid

really be keeping track of?

So how do they do it?

! They imitate what they hear?! Kids get a lot of questions and imperatives.

! Kids produce a lot of declaratives.

! Me playing.

! What do you think what the puppet has eaten?


! By analogy? (Associationism/connectionism?)! That is optional.

! Questions are formed by putting the question word inthe front.

! Who do you wanna invite?

! Who do you wanna take you to the airport?

! Kick-kicked, sing-sang, mail-membled.

3


! One hypothesis holds suggests that parentsactually help kids along (though notconsciously).

! It’s well known that people seem to instinctivelytalk to little kids in kind of a weird way;exaggerated intonation, simpler words, morerepetition. “Baby talk” or as it is sometimesknown, “Motherese”.

! Many have entertained the idea that thissimpler, more carefully articulated, speechmight guide kids along the path of languageacquisition.

Some properties of“Motherese”

! Slower speech, longer pauses

! Higher pitch, greater pitch range

! Exaggerated intonation and stress

! More varied loudness

! Fewer disfluencies

! More restricted vocabulary

! More rephrasings

! More repetitions

! Shorter, less complex utterances

! More imperatives and questions

! Fewer complex (multiclause) sentences

Does “Motherese” driveacquisition?

! Initially tempting, perhaps, but no.

! If “Motherese” were crucial for acquisition, itmust be available to all language acquirers,universally.

! Several documented cultures don’t even speakto the kids until they reach linguisticsophistication. (Of course, they’re exposed tolanguage in the environment, but not directed atthem in “Motherese”)

Does “Motherese” driveacquisition?

! If you give a 4-month old the choice ofwhether to listen to “Motherese” or tonormal adult-directed speech, the kid willchoose to listen to “Motherese”…

! …so it is quite likely that “Motherese”forms a significant part of the PLD for thekid, but it can’t be necessary for successfullanguage acquisition.

Simpler isn’t really better

! Linguists look to complex sentences todifferentiate between predictions of differenthypotheses about how the grammar works.

! Generally, prior to considering complexsentences, the data underdetermines thegrammar; there are (at least) two systemscompatible with the data observed so far.

! If linguists need to look to complex sentences tofigure out the intricacies of the rules (which alladult native speakers seem to end up with), kidsshould need this information too.

Positive and negative evidence

! Kids need to know the grammaticalsystem by the time they are adults.

! Kids hear grammatical sentences(positive evidence)

! Kids are not told which sentences areungrammatical(no negative evidence)

! Let’s consider no negative evidencefurther…

4

Negative evidence

! Negative evidence (information that a givensentence is ungrammatical) could come invarious conceivable forms.! “The sentence Bill a cookie ate is not a sentence in

English, Timmy. No sentence with SOV wordorder is.”

! Upon hearing Bill a cookie ate, an adult might! Offer negative reinforcement

! Not understand

! Look pained

! Rephrase the ungrammatical sentence grammatically

Kids resist instruction…

McNeill (1966)! Nobody don’t like me.

! No, say ‘nobody likes me.’

! Nobody don’t like me.

[repeats eight times]

! No, now listen carefully; say ‘nobody likes me.’

! Oh! Nobody don’t likes me.


Braine (1971)! 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?


Cazden (1972) (observation attributed to Jean Berko Gleason)

! My teacher holded the baby rabbits and we patted them.

! Did you say your teacher held the baby rabbits?

! Yes.

! What did you say she did?

! She holded the baby rabbits and we patted them.

! Did you say she held them tightly?

! No, she holded them loosely.

Negative evidence viafeedback?

! Do kids get “implicit” negative evidence?

! Do adults understand grammatical sentencesand not understand ungrammatical ones?

! Do adults respond positively to grammaticalsentences and negatively to ungrammaticalones?

Approval or comprehension?

Brown & Hanlon (1970):! Adults understood 42% of the grammatical

sentences.

! Adults understood 47% of the ungrammatical ones.

! Adults expressed approval after 45% of thegrammatical sentences.

! Adults expressed approval after 45% of theungrammatical sentences.

This doesn’t bode well for comprehension or approval asa source of negative evidence for kids.

5

Kids’ experience differs

! Parents respond differently! Eve & Sarah’s parents ask clarification

questions after ill-formed wh-questions.

! Adam’s parents ask clarification after well-formed wh-questions…and after past tenseerrors.

! How can kids figure out what correlateswith grammaticality in their situation?

Feedback disappears

! Adam and Sarah showed almost no replycontingencies after age 4…

! But they still made errors after age 4

! And they still stopped making thoseerrors as adults (learning didn’t cease).

Three possible types offeedback

! Complete: consistent response, indicatesunambiguously “grammatical” or“ungrammatical.”

! Partial: if there is a response, it indicates“grammatical” or “ungrammatical”

! Noisy: response given to both grammatical andungrammatical sentences, but withdifferent/detectible frequency.

Statistics (from Marcus 1993)

Suppose response R occurs 20% of the time forungrammatical sentences, 12% of the time forgrammatical sentences.

Kid gets response R to utterance U, there’s a63% chance (20/32) that U is ungrammatical.Guess: ungrammatical, but 38% chance ofbeing wrong.

Kid doesn’t get response R, 52% chance (88/168)it’s grammatical. Guess: grammatical, but48% chance of being wrong.

Statistics (from Marcus 1993)

Suppose response R occurs 20% of the time forungrammatical sentences, 12% of the time forgrammatical sentences.

Suppose kid got response R to U, and is 63%confident that U is ungrammatical—ok, butnowhere near good enough to build a grammar.

This is a serious task, a kid’s going to want to besure. Suppose kid is aiming for 99% confidence(adults make at most 1% speech errors of therelevant kind—pretend this reflects 99%confidence).

Lacking confidence

! Based on R (20%-12% differential), they’d haveto repeat U 446 times (and compile feedbackresults) to reach a 99% confidence level.

! Based on various studies on noisy feedback, arealistic range might be from 85 times (for a35%-14% differential) to 679 times (for a 11.3%-6.3% differential).

! This sounds rather unlike what actuallyhappens.

6

In a way, it’s moot anyway…

! One of the striking things about child languageis how few errors they actually make.

! For negative feedback to work, the kids have tomake the errors (so that it can get the negativeresponse).

! But they don’t make enough relevant kinds oferrors to determine the complex grammar.

Yes-no questions

17) The man is here.18) Is the man here?

Hypothesis 1: Move the first is (or modal, auxiliary)to the front.

Hypothesis 2: Move the first is after the subjectnoun phrase to the front.

19) The man who is here is eating dinner.

Yes-no questions

19) The man who is here is eating dinner.20) *Is the man who here is eating dinner? (*H1)

21) Is the man who is here eating dinner? (! H2)

No kid’s ever said (20) to mean (21), which wouldhave been necessary to distinguish hypotheses 1and 2… Why not?

It seems that kids don’t even entertain Hypothesis 1.And that’s fine, because it seems like Hypothesis 1 is

a kind of rule not found in any adult language.

Abstract principles

! Principle C: Nothing coreferential can c-command a proper name.

*Hei believes Johni’s teacher.

Hisi teacher believes Johni.

! Study of adult grammar reveals that c-commandis the appropriate abstract notion, defined onsyntactic structures. But how do kids learn aboutc-command? You can’t hear c-command.

! What’s more, study of adult grammar reveals thatPrinciple C holds in every language!

! Kids don’t make as many mistakes as would beneeded for hypothesis testing.

! Kids seem to receive no relevant negativeevidence while learning language anyway.

! Kids learn fast.! Kids become adults with all the grammatical

knowledge pertaining thereto (uniform, highlycomplex)

! Kids come to know abstract principles (likePrinciple C) without access to evidencedetermining them. In many cases, these principlesare observed in all human languages. “Poverty ofthe stimulus”

So, we’ve got… Having language = being human

! A linguistic capacity is part of being human.

! Like having two arms, ten fingers, a visionsystem, humans have a language faculty.

! Specification of having arms instead of wings,etc., is somehow encoded genetically.

! Structure of the language faculty ispredetermined, like the structure of the visionsystem is.

! The language faculty (tightly) constrainswhat kinds of languages a child can learn.

! =“Universal Grammar” (UG).

7

Universal Grammar

! UG tightly constrains the learning process.

! Study of syntax, phonology, etc., is generallytrying to uncover properties of Language, tospecify what kind of languages a child can learn,to see what kinds of restrictions UG places onlanguage.

! But kids don’t just enter the world speaking likeadults—there’s development.

! And, adults don’t all end up speaking the samelanguage—there is learning.

Game Plan

! The goal of the first part of the course is todiscover what we can about UG, about thislanguage acquisition device, through looking atfirst language development.

! What do kids know and when? How could theycome to know these things? What have beensome of the major discoveries concerningdevelopment of syntax?

Coming up—L1A

Formal learnability

Principles and parameters

How to set a parameter

Triggers, subset principle

Syntax at age 2

Root clause nonfinite verbs

Null subjects

Case errors

Theories of development

Weak and strong continuity

Experimental evidence

Optional Infinitives/ATOM

Binding theoryHow early is Principle B?

PassivesHow early are passives?

MaturationDo some innate grammaticalprinciples mature?

Wh-questionsProperties of development,implications for syntactictheory

Coming up—L2A

L1A ! L2A? How so?

Knowledge of language

Course of development

Parameter settings?

Critical period hypothesis

Windows of opportunity

“Access” and “Transfer”

What are the effects of the L1

on the L2’ers knowledge?

What role does UG play in

L2A?

Experimental resultsCourse of developmentSensitivity to universalconstraintsEffects of instruction

Other factorsMajor models

Vainikka & Young-Scholten,Krashen, White, Flynn,Schwartz & Sprouse, …

Learnability

! The Principles & Parameters model isdesigned to address the learnabilityproblem children face:! Languages are very complex.! Languages differ (something has to be learned).

! Children get insufficient and variable evidenceto deduce the uniform rules of grammar theyend up with.

! Children have adult-like grammars relativelyquickly.

! The proposed solution to the apparentparadox is to suppose that to a large extent allhuman languages are the same. The grammaticalsystems obey the same principles in allhuman languages.

UG Japanese

English

Principles and Parameters

8

! Languages differ, but only in highly limited ways.! In the order between the verb and the object.

! In whether the verb raises to tense

! …

UG Japanese

English


! This reduces the task for the child immensely—allthat the kid needs to do is to determine from theinput which setting each of the parameters needsto have for the language in his/her environment.

UG Japanese

English


The standard picture

! The way this is usually drawn schematically islike this. The Primary Linguistic Data (PLD)serves as input to a Language AcquisitionDevice (LAD), which makes use of thisinformation to produce a grammar of thelanguage being learned.

LADPLD grammar

The standard picture

! This isolates the innately specified languagefaculty into a single component in the picture.The LAD contains (a specification for) all ofthe principles and the parameters, and has aprocedure for going from PLD to parametersettings.

LADPLD grammar

! We may be able to avoid confusion later,though, if we differentiate the innatelyprovided system into its conceptualcomponents.

! This is my rendition of a way to think aboutUG, parameters, and LAD.

LAD

PLD

UG

SubjacencyBinding Theory

Modeling human languagecapacity

! UG provides the parameters and contains thegrammatical system (including the principles,like Subjacency, Binding Theory, etc.) thatmakes use of them.

! LAD sets the parameters based on the PLD.Responsible for getting language to kids.

LAD

PLD

UG



9

! The idea behind this diagram is that UG issomething like the shape of language knowledge.! Knowledge of language can only take a certain,

innately pre-specified “shape”.

! A system with this “shape” has certain properties,among them Binding Theory, Subjacency, … thePrinciples.

LAD

PLD

UG



! The Parameters are different ways in whichstored knowledge can conform to the “shape”of UG.

! The LAD is a system which analyzes the PLDand sets the parameters.

LAD

PLD

UG



! So two languages which differ with respect toone parameter setting might be representedkind of like this.! This is of course a cartoon view of things, but

perhaps it might be useful later.

Language

A Language

B

Principles and Parameters Principles and Parameters

! So what are the Principles and Parameters?

! Good question! —and that’s what theoreticallinguistics is all about.

! Since 1981, many principles and parameters havebeen proposed. As our understanding of languagegrows, new evidence comes to light, and previousproposals are discarded in favor of bettermotivated ones. It’s hard to keep a current tally of“the principles we know of” because of the activenature of the field.


! Some of the (proposed) Parameters that havereceived a fair amount of press are:! Bounding nodes for Subjacency

! Binding domain for anaphors and pronouns

! Verb-object order

! Overt verb movement (V moves to tense)

! Allowability of null subject (pro) in tensed clauses

! We’ll look at each of them in due course…

Verb-object order

The parameter for verb-object order (moregenerally, the “head parameter” setting out theorder between X!-theoretic head and complement)comes out as:

! Japanese: Head-final (X follows complement)! English: Head-initial (X precedes complement).

Figuring out which type the target language is isoften fairly straightforward. Kids can hearevidence for this quite easily. (Not trivial,though—consider German SOV-V2)

10

Principle A

22) Sam believes [that Harry overestimates himself]

23) Sam-wa [Harry-ga zibun-o tunet-ta to] it-ta]Sam-top Harry-nom self-acc pinch-past-that say-past‘Sam said that Harry pinched him(self).’

Principle A

! Principle A. A reflexive pronoun musthave a higher antecedent in its bindingdomain.

! Parameter: Binding Domain! Option (a): domain = smallest clause

containing the reflexive pronoun! Option (b): domain = utterance containing the

reflexive pronoun

But how can you set thisparameter?

! Every sentence a kid learning English hears isconsistent with both values of the parameter!

! If a kid learning English decided to opt for the“utterance” version of the domain parameter,nothing would ever tell the kid s/he had made amistake.

! S/he would end up with non-English intuitions.


! A kid learning Japanese can tell rightaway that their domain is the sentence,since they’ll hear sentences where zibunrefers to an antecedent outside the clause.


! The set of sentences allowed in English is asubset of the set of sentences allowed inJapanese. If you started assuming the Englishvalue, you could learn the Japanese value, butnot vice-versa.

Sentences allowed in Japanese (domain = utterance)

Sentences allowed in English (domain = clause)

Subset principle/defaults

! Leads to: The acquisition device selects themost restrictive parametric valueconsistent with experience. (Subsetprinciple)

! That is, for the Principle A domainparameter, you (a LAD) start assumingyou’re learning English and switch toJapanese only if presented with evidence.

11

What it takes to set a parameter

! Binding domain parameter! Option (a): Binding domain is clause.

! Option (b): Binding domain is utterance.

! English = option a, Japanese = option b.

E

J


! Binding domain parameter! Kids should start under the

assumption that theparameter has the Englishsetting.

! If they hear only Englishsentences, they will stick withthat setting.

! If they hear Japanesesentences, they will haveevidence to move to theJapanese setting.

E

J


! Null subject parameter! Option (a): Null subjects are permitted.

! Option (b): Null subjects are not permitted.

! Italian = option a, English = option b.

E

IVery sensible. Now, let’s

consider another parameter of

variation across languages.


! The Subset principle says thatkids should start with the Englishsetting and learn Italian if theevidence appears.

! But even English kids are well-known to drop subjects early onin acquisition. As if had the Italiansetting for this parameter.

E

I

Moreover…

! English kids hear looks good and seems ok and stopthat right now. Why don’t they end up speakingItalian? If they mis-set the parameter, how couldthey ever recover?

! Italian kids hear subjectless sentences—whydon’t they interpret them as imperatives orfragments (so as not to have to change theparameter from the default)?

Triggers

! It seems like actual occurrence of nullsubjects isn’t a very good clue as towhether a subject is a null subjectlanguage or not.

! Are there better clues? If a strappingyoung LAD were trying to set the nullsubject parameter, what should it lookfor?

12

Triggers

! Turns out: Only true subject-drop languagesallow null subjects in tensed embedded clauses.

24) *John knows that [— must go]. (English)

25) Juan sabe que [— debe ir]. (Spanish)‘Juan knows that [he] must go.’

! Perhaps the LAD “knows” this and looks forexactly this evidence. Null subjects in embeddedtensed clauses would be a trigger for the(positive setting of the) null subject parameter.

Triggers

! A potential problem with the proposedsubject-drop trigger is that it requires complexsentences—you need to look at an embeddedsentence to check for the trigger.

! Such sentences might be too complicated forkids to process.

! Degree-1 learnability: Triggers need look nolower than 1 level of embedding.

! Degree-0 learnability: Triggers need lookonly at main clauses.

Triggers

! Many who work on learnability haveadopted the hypothesis that triggersneed to be degree-0 learnable.

! Subjacency. *[wh … [" … [# … t … ] ]where " and # are bounding nodes.

Bounding node parameter for IP:! Option (a): IP is a bounding node (English).

! Option (b): IP is not a bounding node (French, Italian).

IP and TP

are often

used inter-

changeably

Triggers

! Thus, a kid learning French couldn’t chooseoption (b) by hearing this…

28) Violà un liste de gens… ‘there is a list of people…’

[à qui on n’a pas encore trouvé [quoi envoyer t t ]]to whom one has not yet found [what to send]]

! …since that’s a degree-2 trigger. But…

Triggers

29) Combien as- [IP tu vu [DP t de personnes]]?

How-many have you seen of people

‘How many people did you see?’

! If IP were a bounding node, this should be

ungrammatical in French, so this can serve

as (degree-0) evidence for option (b).

Triggers

! Principles are part of UG

! Parameters are defined by UG

! Triggers for parameter settings are definedas part of the LAD.

13

Navigating grammar spaces

! Regardless of the technical details, theidea is that in the space of possiblegrammars, there is a restricted set thatcorrespond to possible human grammars.

! Kids must in some sense navigate thatspace until they reach the grammar thatthey’re hearing in the input data.

Learnability

! So how do they do it?

! Where do they start?

! What kind of evidence do they need?

! How much evidence do they need?

! Research on learnability in languageacquisition has concentrated on theseissues.

Are we there yet?

! There are a lot of grammars to choose from, evenif UG limits them to some finite number.

! Kids have to try out many different grammars tosee how well they fit what they’re hearing.

! We don’t want to require that kids remembereverything they’ve ever heard, and sit there andtest their current grammar against the wholecorpus of utterances—that’ a lot to remember.

Are we there yet?

! We also want the kid, when they get to theright grammar, to stay there.

! Error-driven learning! Most theories of learnability rely on a kind of

error-detection.! The kid hears something, it’s not generable by

their grammar, so they have to switch theirhypothesis, to move to a new grammar.

Plasticity

! Yet, particularly as the navigation progresses,we want them to be zeroing in on the rightgrammar.

! Finding an error doesn’t mean that you (as akid) should jump to some random othergrammar in the space.

! Generally, you want to move to a nearbygrammar that improves your ability to generatethe utterance you heard—move in baby steps.

Triggers

! Gibson & Wexler (1994) looked at learningword order in terms of three parameters(head, spec, V2).

! Their triggering learning algorithm says ifyou hear something you can’t produce, tryswitching one parameter and see if ithelps. If so, that’s your new grammar.Otherwise, stick with the old grammarand hope you’ll get a better example.

14

Local maxima

! A problem they encountered is that there arecertain places in the grammar space where youend up more than one switch away from agrammar that will produce what you hear.

! This is locally as good as it gets—nothing next toit in the grammar space is better—yet if youconsider the whole grammar space, there is abetter fit somewhere else, you just can’t get therewith baby steps.

Local maxima

! This is a point where any move you make isworse, so a conservative algorithm will neverget you to the best place. Something a workinglearning algorithm needs to avoid. (And kids,after all, make it).

Backing up a few steps

! Of course, there’s a long way to gobetween being plunked down in the worldand evaluating whether you’re hearingnull subjects in embedded tensed clauses(or, conversely and degree-0 learnably,expletives like in it’s raining).! What are the words?

! Which ones are verbs?

l! k\t"ætˆts"\sˆtgosain!

ænd\mˆnivæn! si" \brˆd#?

Learning language is hard.

! Kids have to find the words. And the referents.

! Extraction: identify grammatical units.

! Segmentation: analyze identified units intocomponent parts.

! We will disregard these important points inorder to proceed analyzing the development ofsyntax.

Do kids have syntacticcategories?

! Once they’ve got the words, have thekids categorized them correctly?

! Do kids categorize the linguistic worldin terms of the same kinds of categoriesadults do? (e.g., noun, verb, …)

! Evidence is hard to come by.

15

So, do kids have syntacticcategories?

! There’s not really any clear way to know atthe earliest (one word) stages.

! One view is that the null hypothesis (whichwe adopt, lacking evidence to the contrary)should be that kids do have adult-likesyntactic categories.

! Continuity. Kids end up being adults withadult syntactic categories; if they initiallycategorize words differently, we need toexplain how they change their categorizationto the adult type.

MLU

! Kids’ linguistic development is oftenmeasured in terms of Mean Length ofUtterance (MLU).! Can be measured in various ways (words,

morphemes)

! Gives an idea of kids’ normal utterance length

! Seems to correlate reasonably well with otherqualitative changes in kid productions

2-year olds

! Around 2 years old

! Around MLU 1.75

! Around 400 words in the vocabulary

! 1-3 word utterances

! Word order generally right

! Grammatical words (the, is) generallymissing

2 1/2 year olds

! About 2 1/2 to 3 years

! About MLU 2.25

! About 900 words in the vocabulary

! Some grammatical devices (past tense -ed, verbal-ing).

! Over-regularization errors (He goed in the house),indicating they’ve grasped the rule of past tenseformation.

! Single clause sentences

3 and 4 year olds

! About 3 to 3 1/2 years, MLU about 2.75,about 1200 words, beginning to usesyntactic transformations (Is Daddy mad?Where is he going?)

! About 3 1/2 to 4 years, MLU about 3.5,about 1500 words, multi-clause sentences,still some over-regularization

4 and 5 year olds

! 4-5 years, MLU around 4, about 1900words, using more conjunctions andtemporal terms (before, after), gain somemetalinguistic awareness.

! After 5, MLU stays about the same (nolonger predictive), sentences get morecomplex, vocabulary increases (moreslowly), over-regularization decreases…

16

Do kids at the one-word stagehave/know syntactic structure?

! Early attempt to answer the question.

! Based on comprehension—kids clearlyunderstand more than they can produce.

! de Villiers & de Villiers (1973), kidsaround MLU (mean length of utterance) 1to 1.5 asked to act out the truck pushes thecar, and got it right only about a third ofthe time.

Do kids at the one-word stagehave/know syntactic structure?! Hirsh-Pasek & Golinkoff (1991),

preferential looking task. Lessburdensome task. Significant preferencefor correct screen (word order & role).

Hey,Cookie

Monster is

tickling Big

Bird." #

How do we describe multi-word utterances?

! Syntactically, in the same terms as theadult grammar? (continuity)

! Or discontinuously? (For some reason,people seem to think this is simpler…)! Thematic (agent+action, action+theme, …)

! Pivot (P1 + O, O + P2, O + O, O)

! “Limited scope formulas” (here+X, want+X)

Syntactic approach

! Continuity: "

! VP VP

V PP V PPsit sit

P NP P NPon chair chair

Why 2 words?

! Maybe they omit words they don’tknow?! Well, but they do omit words they know.

! A kid who’s used hurt before, documented assaying baby cheek to mean ‘baby hurt cheek.’

! Pinker (1984): Processing bottleneck! A 2-word utterance “filter”

! Kids “grow out” of this constraint.

! Still, kind of mysterious. What’s easier?

Evidence for structure

! Recall also the Hirsh-Pasek & Golinkoff(1991), preferential looking task.

! Structure plays a crucial role in figuringout which screen to look at.

Hey,she’s

kissing the

keys." #

17

The second green ball

! Challenge to assumption that kids have structure?Matthei (1982) 3;9-6;3 ‘get the second green ball.’

! When faced with this:

Do they pick the second and green ball or thesecond green ball?

! Kids did terribly—about half the time wrong.

…beware of the task

! However, why chance? Why not always “secondand green”?

! This tends to suggest kids didn’t really “get” thetask. In fact, they made the same mistake withthis array and “pick the second ball”.

! So the problem is probably with ordinalnumbers and manipulating subsets…

…beware of the task

! Additionally, the kids could see the arraythe whole time, so kids may well havedecided on which object to pick by thetime they heard “pick the second…”

! Hamburger & Crain (1984) re-did theexperiment, hiding the array until therequest was complete—kids’ error ratedropped to 14%.

Intermediate moral

! It’s not easy to run a successfulexperiment—you have to be sure thatwhat you’re testing for isn’t beingobscured by other cognitivelimitations.! Act out The truck pushes the car.

! Pick the second green ball.

One-substitution

! Anecdotal evidence:! nice [yellow pen], nice one (1;11)

! Hamburger & Crain (1984): ‘Point to the firstgreen ball. Ok. Now, point to the second one.’! Note: “Failure” wouldn’t tell us anything here,

since one could also legitimately mean ball—but ifkids take one to mean green ball, that’s evidencethat kids do have the syntactic sophistication toreplace N! with one.

! Nevertheless, 42 / 50 kids interpreted it asgreen ball.

Some properties of kidspeak

! Kids’ language differs from adultlanguage in somewhat predictable ways.These can serve as clues to kids’grammatical knowledge. Up to around 3or so…! Case errors for nouns

! Some word order errors! Omitted subjects

! Verbs not (always) fully inflected

18

Word order errors?

! Languages vary with respect to word order! SVO English, French, Mandarin, …

! VSO Tagalog, Irish, …

! SOV Japanese, Korean, Turkish, …

! SOV+V2 German, …

! Clahsen (1986) reports that German kids don’tmanage to put the verb in second position untilthe finite/nonfinite distinction is “mastered.”

! But at that point the change was immediate:Sentence-syntactic properties are storedseparately from word’s category properties.

Word order errors?

! Surprisingly few—95% correct in English, DP-internal order (*black the dog) may be at 100%.

! Yet there are a number of things like: Doggy sew.! It appears that in these cases, it is theme+V

without an expressed agent. When agent isexpressed, themes are in their place.

! Sounds like an unaccusative or a passive—perhapsthey are treating the verb in these cases asunaccusatives? An attractive idea—but for thefact that young kids are bad at passives andunaccusatives.

Word order errors

! Occasionally, postverbal subjects occur—butthese seem to occur with likely unaccusatives withpostverbal subjects on occasion: going it, come car,fall pants. (cf. adult Mandarin , or Italian, whichwould allow that).

! Alternative approach to Doggy sew might betopicalization: Doggy, you sew—if kids actuallycan’t do passives and unaccusatives, then thismight be the only explanation (short of pureperformance error).

The Bennish optative

! Anecdote about Ben, from Sadock (1982)! SVO normally, but in optative (wish) constructions,

he uses a weird word order.

! Intransitives (subject follows verb)! Fall down Daddy. ‘Daddy should fall down’

! Eat Benny now. ‘Let Benny eat now.’

! Sit down Maggie, Mommy.‘Maggie should sit down, Mommy.’

! Transitives (subject marked with for)! Pick up Benny for Daddy.

‘Daddy should pick Ben up.’

! Read a story for Mommy.‘Mommy should read a story.’

The Bennish optative! He’s marking transitive subjects with for, but

leaving intransitive subjects and objectsunmarked.

! In the optative, Ben treats transitive subjectsdifferently, and objects and intransitive subjectsthe same way.

! This pattern is reflected in a type of adultlanguage as well. Ergative languages marksubjects of transitives differently from bothobjects and intransitive subjects.

! Accusative languages (like English) mark objectsdifferently (I left, I bought cheese, Bill saw me).


! Perhaps Ben’s language is ergative in theoptative mood. (An option for adultlanguages, though clearly not in hisparents’ language)

! Further evidence:! Ergative case marker is often homophonous

with marker for possessive (cf. Inuktitut -upused for both), and Ben uses for (his ERGmarker) in possessive constructions as well.

! That’s a nose for Maggie ‘That’s Maggie’s nose.’

19


! Further evidence:! Ergative languages are almost invariably split often

along semantic lines. Sadock takes the optativerestriction to be of this type (cf. Georgian,nominative-accusative most of the time, except in thesubjunctive and aorist, where it is ergative-absolutive)

! Ben’s not really making word order errors,exactly—he just thinks he’s speaking Georgian.His errors come from among the options.

Pre-subject negation

! Kids will say things like:! No I see truck

! Not Fraser read it

! No lamb have a chair either.

! Anaphoric no? ‘No, I see the truck.’

! Often distinguishable from context, andthey are not all anaphoric.

Pre-subject negation

! Déprez & Pierce 1993 looked at these, andproposed that not Fraser read it comes froma failure to raise the subject out of SpecVPto SpecIP. That is, here, Fraser is still in itsVP-internal subject position.

! Some believe this, some don’t, but it’s awell-known analysis.

Case errors

! English pronouns exhibit Case! Nom: I, he, she, they

! Acc: me, him, her, them

! Gen: my, his, her, their

! Kids seem to make errors until at least 2.! me got bean

! her do that

! me eye

! In general, it is often overgeneralization of Acc.

Overuse of accusative

! Default case: Acc in adult English (Schütze 1997)! Me too.

! What, me cheat?! Never!

! Me, I like pizza.

! It’s me.

! —Who did this? —Me.

! So, “overuse of accusative” may well be justusing a default form for nouns which don’t havecase.

Default Case

! Russian (Babyonyshev 1993): Default caseappears to be Nom.

! Russian kids make basically no errors insubject case.

! …but they overuse Nom in other positions(e.g., Nom instead of Acc on an object).

20

Default Case

! German (Schütze 1995): Default case alsoappears to be Nom:! Was? Ich dich betrügen? Nie!

‘What? I cheat on you? Never!’

! Der, den habe ich gesehen.‘He, him I saw.’

! Object case errors are more common thansubject case errors, and usually involveovergeneralization of Nom.

Determiners

! Kids will also often leave out determiners.! Hayley draw boat.

! Turn page.

! Reading book.

! Want duck.

! Wayne in garden

! Daddy want golf ball.

Subject drop

! Even in languages which don’t allow nullsubjects, kids will often leave subjects out.! No turn.

! Ate meat.

! Touch milk.

! Dropping the subject is quitecommon—dropping other things (e.g.,object) is quite rare.

Subject vs. object drop

1578Object

436157Subject

SEA

Root infinitives

! Another, fairlyrecently-noticedaspect of kid speech isthat they will useinfinitive verbssometimes whenadults would usefinite verbs. In lots oflanguages.

! French:! Pas manger la poupée

not eat[inf] the doll

! Michel dormirMichel sleep[inf]

! German:! Zahne putzen

teeth brush[inf]

! Thorstn das habenThorsten that have[inf].

! Dutch:! Ik ook lezen

I also read[inf.]

Root infinitives

! English kids do this too, it turns out, butthis wasn’t noticed for a long time.! It only write on the pad (Eve 2;0)

! He bite me (Sarah 2;9)! Horse go (Adam 2;3)

! It looks like what’s happening is kids areleaving off the -s.

! Taking the crosslinguistic facts intoaccount, we now think those are nonfiniteforms (i.e. to write, to bite, to go).

21

Root infinitives

! However, children learning somelanguages seem to show very few rootinfinitives or none at all.! Italian, for example.

! Often these languages with very few rootinfinitives! Allow null subjects

! Have fairly complex agreement morphology

Pulling it all together

! Kids sometimes use nonfinite verbs.

! Kids sometimes leave out the subject.

! Kids sometimes use the wrong Case onthe subject (looks like a default Case).

! Kids sometimes get the word order wrong(specifically, with respect to negation andfor V2).

! Kids generally leave out determiners.

Kid grammars

! A major research industry arose trying toexplain how these properties of childspeech come about (and how they relate toeach other) in terms of the grammaticaland/or performance abilities of children.

$

$ $

$

$ $

$ $

$

$

GRS LX 700 Observations Language Acquisition and ...€¦ · forms a signiÞcant part of the PLD...

Documents

Transcript of GRS LX 700 Observations Language Acquisition and ...€¦ · forms a signiÞcant part of the PLD...