Week 2. Early Syntactic Development GRS LX 700 Language Acquisition and Linguistic Theory.

Week 2. Early Syntactic Week 2. Early Syntactic DevelopmentDevelopment

GRS LX 700GRS LX 700Language Language

Acquisition andAcquisition andLinguistic TheoryLinguistic Theory

StagesStages

““One word stage”One word stage” At 12-18 months, kids start using At 12-18 months, kids start using

identifiable words, but tend to produce identifiable words, but tend to produce only one at a time.only one at a time.

““Two word stage”Two word stage” Around 20 months more or less, kids start Around 20 months more or less, kids start

putting words together.putting words together. Do they put the words together like adults Do they put the words together like adults

do? What is the status of their linguistic do? What is the status of their linguistic (syntactic) knowledge?(syntactic) knowledge?

MLUMLU

Kids’ linguistic development is often Kids’ linguistic development is often measured in terms of measured in terms of Mean Length Mean Length of Utteranceof Utterance (MLU). (MLU). Can be measured in various ways (words, Can be measured in various ways (words,

morphemes)morphemes) Gives an idea of kids’ normal utterance Gives an idea of kids’ normal utterance

lengthlength Seems to correlate reasonably well with Seems to correlate reasonably well with

other qualitative changes in kid other qualitative changes in kid productionsproductions

2-year olds2-year olds

Around 2 years oldAround 2 years old Around MLU 1.75Around MLU 1.75 Around 400 words in the vocabularyAround 400 words in the vocabulary 1-3 word utterances1-3 word utterances Word order generally rightWord order generally right Grammatical words (Grammatical words (thethe, , isis) )

generally missinggenerally missing

2 1/2 year olds2 1/2 year olds

About 2 1/2 to 3 yearsAbout 2 1/2 to 3 years About MLU 2.25About MLU 2.25 About 900 words in the vocabularyAbout 900 words in the vocabulary Some grammatical devices (past tense -Some grammatical devices (past tense -

eded, verbal -, verbal -inging).). Over-regularization errors (Over-regularization errors (He goed in the He goed in the

househouse), indicating they’ve grasped the ), indicating they’ve grasped the rule of past tense formation.rule of past tense formation.

Single clause sentencesSingle clause sentences

3 and 4 year olds3 and 4 year olds

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

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

4 and 5 year olds4 and 5 year olds

4-5 years, MLU around 4, about 1900 4-5 years, MLU around 4, about 1900 words, using more conjunctions and words, using more conjunctions and temporal terms (temporal terms (beforebefore, , afterafter), gain ), gain some metalinguistic awareness.some metalinguistic awareness.

After 5, MLU stays about the same (no After 5, MLU stays about the same (no longer predictive), sentences get more longer predictive), sentences get more complex, vocabulary increases (more complex, vocabulary increases (more slowly), over-regularization slowly), over-regularization decreases…decreases…

Do kids at the one-word Do kids at the one-word stage have/know stage have/know

syntactic structure?syntactic structure? Early attempt to answer the question. Based on comprehension—kids clearly

understand more than they can produce.

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

Do kids at the one-word Do kids at the one-word stage have/know stage have/know

syntactic structure?syntactic structure? Hirsh-Pasek & Golinkoff (1991), Hirsh-Pasek & Golinkoff (1991),

preferential looking taskpreferential looking task. Less . Less burdensome task. Significant preference burdensome task. Significant preference for correct screen (word order & role).for correct screen (word order & role).

Hey,Cookie Monster is tickling Big Bird.

How do we describe How do we describe multi-word utterances?multi-word utterances?

Syntactically, in the same terms as the Syntactically, in the same terms as the adult grammar? adult grammar? ((continuitycontinuity))

Or Or disdiscontinuouslycontinuously? ? (For some reason, (For some reason, people seem to think this is simpler…)people seem to think this is simpler…) Thematic (agent+action, action+theme, …)Thematic (agent+action, action+theme, …) Pivot (PPivot (P11 + O, O + P + O, O + P22, O + O, O), O + O, O) ““Limited scope formulas” (Limited scope formulas” (herehere+X, +X,

wantwant+X)+X)

Syntactic approachSyntactic approach

Continuity: Continuity:

VPVP VPVP

VV PPPP VV PPPPsitsit sitsit

PP NPNP PP NPNPonon chairchair

chairchair

Why 2 words?Why 2 words? Maybe they omit words they don’t Maybe they omit words they don’t

know?know? Well, but they Well, but they dodo omit words they know. omit words they know.

A kid who’s used A kid who’s used hurthurt before, documented as before, documented as saying saying baby cheekbaby cheek to mean ‘baby hurt cheek.’ to mean ‘baby hurt cheek.’

Pinker (1984): Processing bottleneckPinker (1984): Processing bottleneck A 2-word utterance “filter”A 2-word utterance “filter” Kids “grow out” of this constraint.Kids “grow out” of this constraint. Still, kind of mysterious. What’s easier?Still, kind of mysterious. What’s easier?

Arguments for syntax…Arguments for syntax…

Conceptually:Conceptually:

Kids Kids dodo reach a point where they know reach a point where they know N and V, and they don’t seem to make N and V, and they don’t seem to make the kinds of mismatch errors you’d the kinds of mismatch errors you’d expect if they were switching from a expect if they were switching from a thematically-based categorization to a thematically-based categorization to a grammatically-based categorization.grammatically-based categorization.

Arguments for syntax…Arguments for syntax… Concretely:Concretely:

Animacy is a salient and linguistically relevant Animacy is a salient and linguistically relevant feature of nouns. But kids seem class nouns feature of nouns. But kids seem class nouns together (for the purposes of syntax and word together (for the purposes of syntax and word order) regardless of animacy.order) regardless of animacy.

Even though most subjects heard are animate, Even though most subjects heard are animate, most objects heard are inanimate, kids will most objects heard are inanimate, kids will happily use inanimate subjects or animate happily use inanimate subjects or animate objects.objects.

Kids will also happily use Kids will also happily use modifier+nounmodifier+noun combinations in both subject and object position.combinations in both subject and object position.

Kids distinguish between Kids distinguish between typestypes of nouns, of nouns, big onebig one, , big dogbig dog, but not , but not *big he*big he (though (though he bighe big).).


Semantic information probably comes Semantic information probably comes into play into play as well asas well as, but not , but not instead ofinstead of syntax.syntax.

Kaluli is an “Kaluli is an “ergativeergative language”: subjects language”: subjects of transitives get ergative case-marking, of transitives get ergative case-marking, other nouns are unmarked (“absolutive”).other nouns are unmarked (“absolutive”).

Kaluli kids use ERG first only for subjects Kaluli kids use ERG first only for subjects of transitives that are of transitives that are highly agentivehighly agentive..


Semantic information probably Semantic information probably comes into play comes into play as well asas well as, but not , but not instead ofinstead of syntax. syntax.

Russian kid reportedly used accusative Russian kid reportedly used accusative case marking case marking only for prototypical only for prototypical themesthemes (objects that changed location, (objects that changed location, for example).for example).

Structure in meaningStructure in meaning

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

Structure plays a crucial role in figuring Structure plays a crucial role in figuring out which screen to look at.out which screen to look at.

Hey,she’s kissing the keys.

The second green ballThe second green ball

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

When faced with this:When faced with this:

Do they pick the Do they pick the second and green second and green ball or ball or the the secondsecond greengreen ball? ball?

Kids did terribly—about half the time wrong.Kids did terribly—about half the time wrong.

……but the problem is the but the problem is the tasktask

However, why However, why chancechance? Why not ? Why not alwaysalways “second and green”?“second and green”?

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

So the problem is probably with ordinal So the problem is probably with ordinal numbers and manipulating subsets…numbers and manipulating subsets…

……but the problem is the but the problem is the tasktask

Additionally, the kids could see the Additionally, the kids could see the array the whole time, so kids may array the whole time, so kids may well have decided on which object to well have decided on which object to pick by the time they heard “pick by the time they heard “pick the pick the second…second…””

Hamburger & Crain (1984) re-did Hamburger & Crain (1984) re-did the experiment, hiding the array the experiment, hiding the array until the request was complete—until the request was complete—kids’ error rate dropped to 14%.kids’ error rate dropped to 14%.

Intermediate moralIntermediate moral

It’s not easy to run a successful It’s not easy to run a successful experiment—you have to be sure experiment—you have to be sure that what you’re testing for isn’t that what you’re testing for isn’t being obscured by other cognitive being obscured by other cognitive limitations.limitations. Act out Act out The truck pushes the carThe truck pushes the car.. Pick the second green ball.Pick the second green ball.

OneOne-substitution-substitution

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

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

since since oneone could also legitimately mean could also legitimately mean ballball—but —but if kids take if kids take oneone to mean to mean green ballgreen ball, that’s , that’s evidence that kids do have the syntactic evidence that kids do have the syntactic sophistication to replace Nsophistication to replace N with with oneone..

Nevertheless, 42 / 50 kids interpreted it as Nevertheless, 42 / 50 kids interpreted it as green ballgreen ball..

Some properties of Some properties of kidspeakkidspeak

Kids’ language differs from adult Kids’ language differs from adult language in somewhat predictable language in somewhat predictable ways. These can serve as clues to ways. These can serve as clues to kids’ grammatical knowledge. Up to kids’ grammatical knowledge. Up to around 3 or so…around 3 or so… Case errors for nounsCase errors for nouns Some word order errorsSome word order errors Omitted subjectsOmitted subjects Verbs not (always) fully inflectedVerbs not (always) fully inflected

Word order errors?Word order errors? Languages vary with respect to word orderLanguages vary with respect to word order

SVOSVO English, French, Mandarin, …English, French, Mandarin, … VSOVSO Tagalog, Irish, …Tagalog, Irish, … SOVSOV Japanese, Korean, Turkish, …Japanese, Korean, Turkish, … SOV+V2SOV+V2 German, …German, …

Clahsen (1986) reports that German kids Clahsen (1986) reports that German kids don’t manage to put the verb in second don’t manage to put the verb in second position until the finite/nonfinite distinction is position until the finite/nonfinite distinction is “mastered.”“mastered.”

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

Word order errors?Word order errors? Surprisingly fewSurprisingly few—95% correct in English, DP-—95% correct in English, DP-

internal order (internal order (*black the dog*black the dog) may be at 100%.) may be at 100%. Yet there are a number of things like: Yet there are a number of things like: Doggy sewDoggy sew.. It appears that in these cases, it is It appears that in these cases, it is themetheme+V +V

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

Sounds like an Sounds like an unaccusativeunaccusative or a or a passivepassive——perhaps they are treating the verb in these cases perhaps they are treating the verb in these cases as unaccusatives? as unaccusatives? Though, a red flag: Young kids Though, a red flag: Young kids are bad at passives and unaccusatives.are bad at passives and unaccusatives.

Word order errorsWord order errors Occasionally, postverbal subjects occur—Occasionally, postverbal subjects occur—

but these seem to occur with but these seem to occur with likelylikely unaccusatives with postverbal subjects on unaccusatives with postverbal subjects on occasion: occasion: going itgoing it, , come carcome car, , fall pantsfall pants. (cf. . (cf. adult Mandarin , or Italian, which would adult Mandarin , or Italian, which would allow that).allow that).

Alternative approach to Alternative approach to Doggy sewDoggy sew might might be topicalization: be topicalization: Doggy, you sewDoggy, you sew—if kids —if kids actually can’t actually can’t dodo passives and passives and unaccusatives, then this might be the only unaccusatives, then this might be the only explanation (short of pure performance explanation (short of pure performance error).error).

The Bennish optativeThe Bennish optative Anecdote about Ben, from Sadock (1982)Anecdote about Ben, from Sadock (1982)

SVO normally, but in optative (wish) SVO normally, but in optative (wish) constructions, he uses a weird word order.constructions, he uses a weird word order.

Intransitives (subject follows verb)Intransitives (subject follows verb) Fall down Daddy.Fall down Daddy. ‘Daddy should fall down’ ‘Daddy should fall down’ Eat Benny now.Eat Benny now. ‘Let Benny eat now.’ ‘Let Benny eat now.’ Sit down Maggie, Mommy.Sit down Maggie, Mommy.

‘Maggie should sit down, Mommy.’‘Maggie should sit down, Mommy.’ Transitives (subject marked with Transitives (subject marked with forfor))

Pick up Benny for Daddy.Pick up Benny for Daddy.‘Daddy should pick Ben up.’‘Daddy should pick Ben up.’

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

The Bennish optativeThe Bennish optative He’s marking transitive subjects with He’s marking transitive subjects with forfor, but , but

leaving intransitive subjects and objects leaving intransitive subjects and objects unmarked.unmarked.

In the optative, Ben treats transitive subjects In the optative, Ben treats transitive subjects differently, and objects and intransitive subjects differently, and objects and intransitive subjects the same way.the same way.

This pattern is reflected in a type of adult This pattern is reflected in a type of adult language as well. language as well. ErgativeErgative languages languages mark mark subjects of transitives differently from both subjects of transitives differently from both objects and intransitive subjects.objects and intransitive subjects.

Accusative Accusative languages languages (like English) mark objects (like English) mark objects differently (differently (I leftI left, , I bought cheeseI bought cheese, , Bill saw meBill saw me).).

The Bennish optativeThe Bennish optative

Perhaps Ben’s language is ergative in Perhaps Ben’s language is ergative in the optative mood.the optative mood. (An option for adult (An option for adult languages, though clearly not in his languages, though clearly not in his parents’ language)parents’ language)

Further evidence:Further evidence: Ergative case marker is often homophonous Ergative case marker is often homophonous

with marker for possessive (cf. Inuktitut with marker for possessive (cf. Inuktitut -up-up used for both), and Ben uses used for both), and Ben uses forfor (his ERG (his ERG marker) in possessive constructions as well.marker) in possessive constructions as well.

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

The Bennish optativeThe Bennish optative

Further evidence:Further evidence: Ergative languages are almost invariably Ergative languages are almost invariably splitsplit often often

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

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

Pre-subject negationPre-subject negation

Kids will say things like:Kids will say things like: No I see truckNo I see truck Not Fraser read itNot Fraser read it No lamb have a chair either.No lamb have a chair either.

Anaphoric Anaphoric nono? ‘No, I see the truck.’? ‘No, I see the truck.’ Often distinguishable from context, Often distinguishable from context,

and and they are not they are not all all anaphoricanaphoric..

Pre-subject negationPre-subject negation

Déprez & Pierce 1993 looked at Déprez & Pierce 1993 looked at these, and proposed that these, and proposed that not Fraser not Fraser read itread it comes from a failure to raise comes from a failure to raise the subject out of SpecVP to SpecIP. the subject out of SpecVP to SpecIP. That is, here, That is, here, FraserFraser is still in its VP- is still in its VP-internal subject position.internal subject position.

Some believe this, some don’t, but Some believe this, some don’t, but it’s a well-known analysis. it’s a well-known analysis. (See (See O’Grady for more discussion…)O’Grady for more discussion…)

Case errorsCase errors

English pronouns exhibit CaseEnglish pronouns exhibit Case Nom: Nom: II, , hehe, , sheshe, , theythey Acc: Acc: meme, , himhim, , herher, , themthem Gen: Gen: mymy, , hishis, , herher, , theirtheir

Kids seem to make errors until at least 2.Kids seem to make errors until at least 2. me got beanme got bean her do thather do that me eyeme eye

In general, it is often In general, it is often overgeneralizationovergeneralization of Accof Acc..


As kid learns Nom, it alternates with As kid learns Nom, it alternates with overgeneralized Acc in subject position.overgeneralized Acc in subject position.

Aldridge (1989): finite verbs have Nom.Aldridge (1989): finite verbs have Nom. Yet…?Yet…?

I swinging.I swinging. He hiding.He hiding.

These cases have a “silent finite These cases have a “silent finite bebe”?”? I on this one, I on this one, aren’taren’t I? I?


Bellugi (1967) observed that nominative Bellugi (1967) observed that nominative II appears for sentence-initial subjects, appears for sentence-initial subjects, but but meme marks non-sentence-initial marks non-sentence-initial subjects.subjects. I laughing.I laughing. I here.I here. When me want it?When me want it? Where me sleep?Where me sleep?

Vainikka (1993/4): no, Vainikka (1993/4): no, meme in in whwh--questionsquestions..


Another possibility: based on agentivity?Another possibility: based on agentivity? Budwig (1990): Budwig (1990): II is for low agentivity, is for low agentivity,

mymy is used for “prototypical agent” and is used for “prototypical agent” and acts to gain control.acts to gain control. I wear itI wear it (wearing microphone)(wearing microphone) My wear itMy wear it (wants to wear microphone)(wants to wear microphone)

Languages Languages dodo make case distinctions make case distinctions based on agentivity & control, so kids based on agentivity & control, so kids learning learning somesome languages will need to languages will need to attend to this.attend to this.

Overuse of accusativeOveruse of accusative

Topics? Cf., Topics? Cf., Her, I likeHer, I like… … We think not.We think not. Kids using Acc subjects don’t use “topic Kids using Acc subjects don’t use “topic

intonation”intonation” Acc subjects appear where topicalization Acc subjects appear where topicalization

should be disallowed:should be disallowed: what me play with?what me play with? there her is.there her is.

Doesn’t say anything about Doesn’t say anything about me eyeme eye, , me me daddad..

Overuse of accusativeOveruse of accusative

Default case: Acc in adult English Default case: Acc in adult English (Schütze 1997)(Schütze 1997) Me too.Me too. What, me cheat?! Never!What, me cheat?! Never! Me, I like pizza.Me, I like pizza. It’s me.It’s me. ——Who did this? —MeWho did this? —Me..

So, “overuse of accusative” may well be So, “overuse of accusative” may well be just using a default form for nouns which just using a default form for nouns which don’t don’t havehave case. case.

Default CaseDefault Case

Russian (Babyonyshev 1993): Russian (Babyonyshev 1993): Default case appears to be Nom.Default case appears to be Nom.

Russian kids make basically no Russian kids make basically no errors in subject case.errors in subject case.

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

Default CaseDefault Case

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

‘What? I cheat on you? Never!’‘What? I cheat on you? Never!’ Der, den habe ich gesehen.Der, den habe ich gesehen.

‘He, him I saw.’‘He, him I saw.’

Object case errors are more common than Object case errors are more common than subject case errors, and usually involve subject case errors, and usually involve overgeneralization of Nom.overgeneralization of Nom.

DeterminersDeterminers

Kids will also often leave out Kids will also often leave out determiners.determiners. Hayley draw boat.Hayley draw boat. Turn page.Turn page. Reading book.Reading book. Want duck.Want duck. Wayne in gardenWayne in garden Daddy want golf ball.Daddy want golf ball.

Subject dropSubject drop

Even in languages which don’t allow Even in languages which don’t allow null subjects, kids will often leave null subjects, kids will often leave subjects out.subjects out. No turn.No turn. Ate meat.Ate meat. Touch milk.Touch milk.

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

Subject vs. object dropSubject vs. object drop

0

10

20

30

40

50

60

70

Adam Eve Sarah

Percentage of Missing subjects and Objects from Obligatory Contexts

SubjectsObjects

AA EE SS

SubjecSubjectt

5757 6161 4343

ObjectObject 88 77 1515

Root infinitivesRoot infinitives

Another, fairly Another, fairly recently-noticed recently-noticed aspect of kid aspect of kid speech is that they speech is that they will use infinitive will use infinitive verbs sometimes verbs sometimes when adults would when adults would use finite verbs. In use finite verbs. In lots of languages.lots of languages.

French:French: Pas manger la poupéePas manger la poupée

not eat[inf] the dollnot eat[inf] the doll Michel dormirMichel dormir

Michel sleep[inf]Michel sleep[inf] German:German:

Zahne putzenZahne putzenteeth brush[inf]teeth brush[inf]

Thorstn das habenThorstn das habenThorsten that Thorsten that have[inf].have[inf].

Dutch:Dutch: Ik ook lezenIk ook lezen

I also read[inf.]I also read[inf.]


English kids do this too, it turns out, English kids do this too, it turns out, but this wasn’t noticed for a long time.but this wasn’t noticed for a long time. It only write on the pad It only write on the pad (Eve 2;0)(Eve 2;0) He bite me He bite me (Sarah 2;9)(Sarah 2;9) Horse go Horse go (Adam 2;3)(Adam 2;3)

It It lookslooks like what’s happening is kids like what’s happening is kids are leaving off the are leaving off the -s-s..

Taking the crosslinguistic facts into Taking the crosslinguistic facts into account, we now think those are account, we now think those are nonfinite forms (i.e. nonfinite forms (i.e. to writeto write, , to biteto bite, , to to gogo).).


However, children learning some However, children learning some languages seem to show very few languages seem to show very few root infinitives or none at all.root infinitives or none at all. Italian, for example.Italian, for example.

Often these languages with very few Often these languages with very few root infinitivesroot infinitives Allow null subjectsAllow null subjects Have fairly complex agreement Have fairly complex agreement

morphologymorphology

Pulling it all togetherPulling it all together

Kids sometimes use nonfinite verbs.Kids sometimes use nonfinite verbs. Kids sometimes leave out the subject.Kids sometimes leave out the subject. Kids sometimes use the wrong Case on Kids sometimes use the wrong Case on

the subject (looks like a default Case).the subject (looks like a default Case). Kids sometimes get the word order Kids sometimes get the word order

wrong (specifically, with respect to wrong (specifically, with respect to negation and for V2).negation and for V2).

Kids generally leave out determiners.Kids generally leave out determiners.

Kid grammarsKid grammars

A major research industry arose A major research industry arose trying to explain how these trying to explain how these properties of child speech come properties of child speech come about (and how they relate to each about (and how they relate to each other) in terms of the grammatical other) in terms of the grammatical and/or performance abilities of and/or performance abilities of children. children.

Let’s start with null Let’s start with null subjectssubjects

Until after around 2 years old, kids Until after around 2 years old, kids will often omit subjects:will often omit subjects: Drop bean.Drop bean. Fix Mommy shoe.Fix Mommy shoe. Helping Mommy.Helping Mommy. Want go get it.Want go get it.

Why?Why?

The null subject The null subject parameterparameter

Adult languages differ in whether Adult languages differ in whether they require overt subjects or not.they require overt subjects or not.

EnglishEnglish does: does: *Go to the movies tonight.*Go to the movies tonight.

ItalianItalian and and SpanishSpanish do not: do not: Vado al cinema stasera.Vado al cinema stasera. (Italian)(Italian) Voy al cine esta noche.Voy al cine esta noche. (Spanish)(Spanish)

‘(I) go to the movies tonight.’‘(I) go to the movies tonight.’

SS00 = Italian? = Italian?

Hyams (1986) proposes that kids Hyams (1986) proposes that kids learning English go through a stage learning English go through a stage during which they are speaking during which they are speaking Italian.Italian.

The “The “null subject parameternull subject parameter” has an ” has an initial setting, that of Italian.initial setting, that of Italian.

Kids use that setting until they Kids use that setting until they “reset it” to the English value.“reset it” to the English value.

Resetting the parameterResetting the parameter

Null subject languages do not have Null subject languages do not have expletives like expletives like itit or or therethere..

The English input will provide plenty The English input will provide plenty of examplesof examples No, it’s not raining.No, it’s not raining. It’s not cold outside.It’s not cold outside. There’s no more.There’s no more.

Resetting the parameterResetting the parameter

Null subject languages do not have Null subject languages do not have unstressed pronouns.unstressed pronouns.

The English input (once the kids The English input (once the kids have figured out stress/focus) will have figured out stress/focus) will still contain pronouns where they still contain pronouns where they “should” be dropped if the language “should” be dropped if the language were a null subject language.were a null subject language.

SS00 = Italian = Italian

Hyams (1986) was an early attempt Hyams (1986) was an early attempt to explain the null subjects in child to explain the null subjects in child language by making use of the language by making use of the options available among adult options available among adult languages.languages.

Suggests that kids can “mis-set” a Suggests that kids can “mis-set” a parameter (or leave a parameter set parameter (or leave a parameter set in its default state) and then recover.in its default state) and then recover.

SS00 = Italian = Italian

NobodyNobody believes this anymore. believes this anymore.

Distribution of null subjects in child English Distribution of null subjects in child English and Italian is just different.and Italian is just different.

English: no embedded null subjects (Italian English: no embedded null subjects (Italian kids do have them, Valian 1991’s 21 kids kids do have them, Valian 1991’s 21 kids uttered 123 finite subordinate clauses and uttered 123 finite subordinate clauses and none had a null subject).none had a null subject).

Other parameters seem to be set so fast that Other parameters seem to be set so fast that errors are never detectible (word order/V2/V-errors are never detectible (word order/V2/V-to-I). Parameters don’t seem to be mis-set.to-I). Parameters don’t seem to be mis-set.

Processing accounts…Processing accounts…

Kids have severely limited Kids have severely limited processing power, and so they leave processing power, and so they leave off subjects to ease the load. (Bloom off subjects to ease the load. (Bloom 1990)1990)

In favor:In favor: Length limitations even in imitationsLength limitations even in imitations Kids omit things other than subjectsKids omit things other than subjects Some kids don’t eliminate subjects, only Some kids don’t eliminate subjects, only

reduce their frequency.reduce their frequency.

Processing accounts…Processing accounts…

Contra? Hyams points out:Contra? Hyams points out: Build house…Cathy build houseBuild house…Cathy build house Go nursery…Lucy go nurseryGo nursery…Lucy go nursery Kathryn want build another house.Kathryn want build another house.

Bloom: So, no absolute limit on Bloom: So, no absolute limit on length, only a tendency to reduce length, only a tendency to reduce length.length.

Bloom (1990)Bloom (1990)

Bloom (1970) found:Bloom (1970) found: negated sentences tend to lack subjects negated sentences tend to lack subjects

more frequently then non-negated sentences.more frequently then non-negated sentences. Bloom (1990):Bloom (1990):

Hypothesis: sentences without subjects will Hypothesis: sentences without subjects will have longer VPs than sentences with have longer VPs than sentences with subjects.subjects.

Looked at past tense verbs and cognitive Looked at past tense verbs and cognitive states (states (needneed) to avoid any confusion with ) to avoid any confusion with imperatives.imperatives.


VP length (words from verb to the VP length (words from verb to the end) counted for sentences with and end) counted for sentences with and without subjects.without subjects.

ResultsResults: Mean length of VP in : Mean length of VP in sentences with subjects were sentences with subjects were (statistically) significantly shorter (statistically) significantly shorter than those without.than those without. E.g., Adam 2.333 with, 2.604 without.E.g., Adam 2.333 with, 2.604 without.

Bloom (1990)Bloom (1990) In fact, “long subjects” (lexical subjects), “short In fact, “long subjects” (lexical subjects), “short

subjects” (pronouns), and null subjects correlated subjects” (pronouns), and null subjects correlated with an increase in VP length as well.with an increase in VP length as well.

1

1.5

2

2.5

3

No subject Pronoun Lexical

AdamEveSarah


As for why subjects are dropped As for why subjects are dropped more frequently than objects by kids more frequently than objects by kids at this stage—why?at this stage—why?

Two possibilities?Two possibilities? Subjects tend to be given (old) Subjects tend to be given (old)

information (low “informativeness”, information (low “informativeness”, more expendable)more expendable)

Maybe processing “saves the heaviest Maybe processing “saves the heaviest load for last”load for last”

Hyams & Wexler (1993)Hyams & Wexler (1993)

Bloom’s (1990) approach (processing) Bloom’s (1990) approach (processing) can’t be right either.can’t be right either.

The difference between subjects and The difference between subjects and objects is big, and only rate of objects is big, and only rate of subject subject drop changes.drop changes. Adam & Eve both drop around 40-50% of Adam & Eve both drop around 40-50% of

their subjects in an early stage, and in a their subjects in an early stage, and in a later stage are down to 15-30%—later stage are down to 15-30%—meanwhile their rate of object drop stays meanwhile their rate of object drop stays around 5-10%.around 5-10%.


““Informativeness”?Informativeness”? All else being equal, the ratio of missing All else being equal, the ratio of missing

subjects to specific subjects should be subjects to specific subjects should be equal to the ratio of missing objects to equal to the ratio of missing objects to specific objects.specific objects.

Turns out that kids drop specific Turns out that kids drop specific subjects about twice as often (Adam subjects about twice as often (Adam 52%) as they drop specific objects 52%) as they drop specific objects (Adam 21%).(Adam 21%).

Hyams & Wexler (1993)Hyams & Wexler (1993) Considering Italian adults, we find exactly the Considering Italian adults, we find exactly the

same correlation Bloom reported for English same correlation Bloom reported for English kids: VP seems to be longer where there is null kids: VP seems to be longer where there is null subject, shorter with a pronoun, and shorter still subject, shorter with a pronoun, and shorter still with a lexical subject.with a lexical subject.

1

1.5

2

2.5

3


AdamEveSarah

3

3.25

3.5

3.75

4

4.25

4.5

4.75

5


Italian adult


Regardless of why the correlation holds, if Regardless of why the correlation holds, if it is a processing deficiency in kids, what it is a processing deficiency in kids, what is it for the Italian adults?is it for the Italian adults?

Seems like kids act like they’re speaking a Seems like kids act like they’re speaking a language where the null subject is a language where the null subject is a grammatical optiongrammatical option. . Note: might be Note: might be slightly different from a “null subject slightly different from a “null subject language” though. Point: dropping language” though. Point: dropping subjects is grammatical for these kids, not subjects is grammatical for these kids, not an error.an error.

Hyams & Wexler (1993)Hyams & Wexler (1993) Consider the proportion of pronouns to Consider the proportion of pronouns to

lexical subjects.lexical subjects. ““Output omission” model Output omission” model would predict would predict

that younger kids would tend to drop that younger kids would tend to drop more lexical subjects than pronouns, more lexical subjects than pronouns, compared to the ratio they wind up at.compared to the ratio they wind up at.

Grammatical omission model Grammatical omission model would would predict that younger kids would tend to predict that younger kids would tend to drop more pronouns (since some are drop more pronouns (since some are being realized as null subjects)being realized as null subjects)


We find:We find: Adam goes from about 3:1 in favor of Adam goes from about 3:1 in favor of

lexical subjects (during subject drop lexical subjects (during subject drop stage) to 1:2 (after subject drop stage).stage) to 1:2 (after subject drop stage).

When he’s dropping subjects, they are When he’s dropping subjects, they are coming out of the “pronoun” pile—the coming out of the “pronoun” pile—the number of lexical subjects is staying number of lexical subjects is staying about the same across development. about the same across development.


Ok, so maybe pronouns are more Ok, so maybe pronouns are more difficult than lexical nouns? (Doesn’t difficult than lexical nouns? (Doesn’t fit well with the length of VP result, fit well with the length of VP result, but maybe…?)but maybe…?)

Problem is: kids show a steady level Problem is: kids show a steady level of of object object pronouns throughout this pronouns throughout this time period—and output omission time period—and output omission model doesn’t have anything to say model doesn’t have anything to say about subject vs. object.about subject vs. object.

Hyams & Wexler (1993)Hyams & Wexler (1993) Basic conclusion:Basic conclusion:

Null subjects don’t seem to arise in child Null subjects don’t seem to arise in child language solely due to processing difficulty.language solely due to processing difficulty.

Rather, they seem to be Rather, they seem to be allowed allowed in the child in the child grammar.grammar.

This allows distinction between subject (high This allows distinction between subject (high rate of omission) and object (low rate of rate of omission) and object (low rate of omission)omission)

Explains the tradeoff between null subjects and Explains the tradeoff between null subjects and pronouns (and the VP length/subject correlation) pronouns (and the VP length/subject correlation) if the prinnciples governing availability of subject if the prinnciples governing availability of subject drop are similar to those at work in Italian.drop are similar to those at work in Italian.

So what allows null So what allows null subjects?subjects?

Here’s where we start to tie in to other Here’s where we start to tie in to other properties of that age.properties of that age.

Notice that in English (a non-null Notice that in English (a non-null subject language) you can have a subject language) you can have a grammatical null subject in one grammatical null subject in one context:context: I want [Ø to have a fire drill]I want [Ø to have a fire drill] [Ø to have a fire drill] would make my day.[Ø to have a fire drill] would make my day.

So what allows null So what allows null subjects?subjects?

Subjects of infinitives can be null.Subjects of infinitives can be null. Kids at the age where subjects are often Kids at the age where subjects are often

missing often use infinitive verb forms.missing often use infinitive verb forms.

Perhaps that’s the key: Perhaps that’s the key: Since kids can Since kids can use infinitives where adults can’t (main use infinitives where adults can’t (main clause main verb), this allows them to clause main verb), this allows them to use null subjects in those sentences as use null subjects in those sentences as a side effect.a side effect.

Proportion of null Proportion of null subjects in finite and subjects in finite and

non-finite clausesnon-finite clauses

0

10

20

30

40

50

60

70

80

90

100

Flem GermSGermA FrP FrN DutchH EngA

null finitenull nonfinite

Null subjects and Null subjects and infinitivesinfinitives

Perhaps we’re on to something here.Perhaps we’re on to something here. So null subjects are (for the most So null subjects are (for the most

part—not part—not completelycompletely) allowed by ) allowed by virtue of having infinitives.virtue of having infinitives.

What allows the infinitives in child What allows the infinitives in child language?language?

Several classes of Several classes of theoriestheories

No functional projections. No functional projections. (Radford) (Radford) Kids don’t have any functional Kids don’t have any functional projections (TP, CP, and so forth). This projections (TP, CP, and so forth). This comes later. No TP, no tense comes later. No TP, no tense distinction.distinction.

Structure building.Structure building. (Vainikka, (Vainikka, Guilfoyle & Noonan) Guilfoyle & Noonan) Kids start with no Kids start with no functional projections and gradually functional projections and gradually increase their functional structure.increase their functional structure.


Truncation. Truncation. (Rizzi) (Rizzi) Like structure building Like structure building but without the time course—kids have but without the time course—kids have access to all of the functional structure access to all of the functional structure but they don’t realize that sentences need but they don’t realize that sentences need to be CP’s, so they sometimes stop early.to be CP’s, so they sometimes stop early.

““ATOM” (Full competence). ATOM” (Full competence). (Wexler, …)(Wexler, …) Kids have access to all of the functional Kids have access to all of the functional structure and have a very specific structure and have a very specific problem with tense and agreement that problem with tense and agreement that sometimes causes them to leave one out.sometimes causes them to leave one out.


Figuring out the differences in Figuring out the differences in content and predictions between the content and predictions between the theories is intricate and difficult, but theories is intricate and difficult, but in doing so, we’ve learned quite a bit in doing so, we’ve learned quite a bit more about the character of child more about the character of child syntax…syntax…

Radford (1995)Radford (1995) A proposal about A proposal about Early Child EnglishEarly Child English.. Kids’ syntax differs from adults’ syntax:Kids’ syntax differs from adults’ syntax:

kids use only lexical (not functional) elementskids use only lexical (not functional) elements structural sisters in kids’ trees always have a structural sisters in kids’ trees always have a

-relation between them.-relation between them.

VPVP

NPNP V’V’manman

VV NPNPchasechase carcar

adult syntax ≠ child adult syntax ≠ child syntaxsyntax

Adults:Adults: CP—IP—VPCP—IP—VP Kids:Kids: VPVP

Evidence for absence of IP:Evidence for absence of IP: No modals (repeating, kids drop them)No modals (repeating, kids drop them) No auxiliaries (No auxiliaries (Mommy doing dinnerMommy doing dinner)) No productive use of tense & No productive use of tense &

agreement (agreement (Baby ride truckBaby ride truck, , Mommy Mommy gogo, , Daddy sleepDaddy sleep))

Absence of CPAbsence of CP

No CP system:No CP system: no complementizers (no complementizers (thatthat, , forfor, , ifif)) no preposed auxiliary (no preposed auxiliary (car go?car go?)) no no whwh-movement (imitating -movement (imitating where does where does

it go?it go? yields yields go?go?; spontaneous: ; spontaneous: mouse mouse doing?doing?))

kids bad at comprehending kids bad at comprehending whwh-object -object questions (out of canonical order). (questions (out of canonical order). (——What are you doing? —No.What are you doing? —No.))

Absence of DPAbsence of DP

No DP system:No DP system: no non-no non- elements elements

no expletives (no expletives (rainingraining, , outside coldoutside cold)) no no ofof before noun complements of nouns ( before noun complements of nouns (cup cup

teatea)) kids tend not to use determiners (kids tend not to use determiners (Hayley Hayley

draw boatdraw boat, , want duckwant duck, , reading bookreading book)) kids don’t use possessive kids don’t use possessive ’s’s, which may be , which may be

a D.a D. kids don’t use pronouns, which are kids don’t use pronouns, which are

probably D.probably D.

So why does it happen So why does it happen this way?this way?

Why VP before IP/CP?Why VP before IP/CP? you have to know what the you have to know what the -grids of -grids of

your verbs are before you can… (your verbs are before you can… (AR: AR: move the subject to SpecIPmove the subject to SpecIP)… even put )… even put the subject in the VP in the first place. the subject in the VP in the first place. Not an argument for why IP Not an argument for why IP followsfollows VP VP—simultaneous should satisfy this as —simultaneous should satisfy this as wellwell..

Kids start with VP and their system Kids start with VP and their system maturesmatures, admitting IP and CP at a later , admitting IP and CP at a later point. point. See also Vainikka (1993/4)See also Vainikka (1993/4)..

So why does it happen So why does it happen this way?this way?

Why VP before IP/CP?Why VP before IP/CP? Language parameterization lies Language parameterization lies

solely in features of functional heads solely in features of functional heads (like I and C). Kids start with the non-(like I and C). Kids start with the non-parameterized part of grammar, and parameterized part of grammar, and work their way up to the work their way up to the parameterized part. parameterized part. But—kids don’t But—kids don’t make word order errors; Japanese make word order errors; Japanese kids start talking SOV from the get-kids start talking SOV from the get-go, English kids SVO…go, English kids SVO…

The transition to IPThe transition to IP

Slightly older kids alternate between Slightly older kids alternate between Nom subjects and Acc subjects, between Nom subjects and Acc subjects, between finite verbs and nonfinite verbs.finite verbs and nonfinite verbs. One viewOne view: kids are “code-switching” between : kids are “code-switching” between

a VP grammar and an IP grammar.a VP grammar and an IP grammar. If this is the case, we expect Nom subjects to If this is the case, we expect Nom subjects to

occur in the IP grammar (with the finite occur in the IP grammar (with the finite verbs) and Acc subjects to occur in the VP verbs) and Acc subjects to occur in the VP grammar (with the nonfinite verbs).grammar (with the nonfinite verbs).

The transition to IPThe transition to IP

Radford says Radford says looklook, , they don’tthey don’t:: ““numerous” nonfinite clauses with nominative numerous” nonfinite clauses with nominative

subjects: subjects: I singingI singing, , I done itI done it.. ““frequent” finite clauses with accusative frequent” finite clauses with accusative

subjects: subjects: Me can make a henMe can make a hen, , Me didn’t paint Me didn’t paint thatthat..

But Schütze & Wexler (1996) show that the But Schütze & Wexler (1996) show that the percentagespercentages are very skewed… So, it looks are very skewed… So, it looks like predominantly, yes, Nom goes with the like predominantly, yes, Nom goes with the “IP grammar” (properly construed).“IP grammar” (properly construed).

Finiteness vs. case errorsFiniteness vs. case errors

Loeb & Leonard (1991)

subject Finite Nonfinite

he+she 436 75

him+her 4 28

% non-Nom 0.9% 27%

Finiteness vs. case errorsFiniteness vs. case errors

Schütze & Wexler (1996)

Nina

subject Finite Nonfinite

I 255 139

me+my 14 120

% non-Nom 5% 46%

The problem so farThe problem so far

Look for numbers, see if you find any. Look for numbers, see if you find any. Radford does not do a very good job Radford does not do a very good job of convincing us that what he’s telling of convincing us that what he’s telling us about is us about is representativerepresentative. He only . He only tells us that it is tells us that it is attestedattested..

We’ll have more to say about case We’ll have more to say about case errors and finiteness.errors and finiteness.

The transition to CPThe transition to CP

It has been observed that even after It has been observed that even after kids can invert yes-no questions…kids can invert yes-no questions… Did you want that one?Did you want that one?

……they fail to invert in they fail to invert in whwh-questions-questions What he can ride in?What he can ride in?

Radford suggests: C comes in two Radford suggests: C comes in two flavors, “flavors, “verbalverbal” and “” and “nonverbalnonverbal”—root ”—root clauses are clauses are verbalverbal, embedded clauses , embedded clauses are are nonverbalnonverbal, and , and I will not move to I will not move to C if C is nonverbalC if C is nonverbal..


Kids have C which isn’t specified Kids have C which isn’t specified either for either for verbalverbal or for or for nonverbalnonverbal..

The rule about moving I to C The rule about moving I to C doesn’t mention doesn’t mention unspecifiedunspecified C, so C, so I can move to unspecified C.I can move to unspecified C.

But, if a But, if a whwh-word moves into -word moves into SpecCP, then Spec-head SpecCP, then Spec-head agreement with the nonverbal agreement with the nonverbal whwh--word gives C a word gives C a nonverbalnonverbal feature, feature, prohibiting I to C movement.prohibiting I to C movement.


The problem here is that there is no The problem here is that there is no independent evidence.independent evidence.

Plus, kids are supposed to be having Plus, kids are supposed to be having trouble with subject agreement trouble with subject agreement between I and SpecIP—at the same between I and SpecIP—at the same time that they seem to be perfectly time that they seem to be perfectly able to effect agreement between C able to effect agreement between C and SpecCP…?and SpecCP…?

For next time:For next time:

Read Radford (1995)Read Radford (1995) Write up a 1-2 page summary of Write up a 1-2 page summary of

Radford 1995:Radford 1995: What are his main points?What are his main points? What is the evidence for each point?What is the evidence for each point? Is this evidence convincing? If not, why Is this evidence convincing? If not, why

not?not?

Week 2. Early Syntactic Development GRS LX 700 Language Acquisition and Linguistic Theory.

Documents

identifiable words

word stage

missing grammatical

various ways words

word utterances word

x slide

overregularization slide

right word order