week 3. further notes on non-finite root forms grs lx 700 language acquisition and linguistic theory

Week 3. Further notes onWeek 3. Further notes onnon-finite root formsnon-finite root forms

GRS LX 700GRS LX 700Language Language

Acquisition andAcquisition andLinguistic Linguistic

TheoryTheory

Where we areWhere we are

There is evidence that children There is evidence that children around the age of 2:around the age of 2: Allow nonfinite verbs in main clauses—Allow nonfinite verbs in main clauses—in non-NS languages.in non-NS languages.

Differentiate between the syntax of Differentiate between the syntax of finite and nonfinite verb forms.finite and nonfinite verb forms.

Show evidence from word order of Show evidence from word order of functional structure above the VP.functional structure above the VP.

Subject case and Subject case and dodo-support suggest -support suggest that TP/AgrP can be missing from the that TP/AgrP can be missing from the child representations.child representations.

Proposals concerning Proposals concerning TP/AgrPTP/AgrP

Wexler (1998): Unique Checking Constraint.Wexler (1998): Unique Checking Constraint. For adults, subjects need to check a feature For adults, subjects need to check a feature both on T and on Agr. For kids, only one is both on T and on Agr. For kids, only one is possible (so either T or Agr must be left out, possible (so either T or Agr must be left out, or UCC must be violated).or UCC must be violated).

PredictsPredicts: NS/OI. Other reflexes, perhaps : NS/OI. Other reflexes, perhaps object shift in Korean.object shift in Korean.

Rizzi (1993/4): Truncation.Rizzi (1993/4): Truncation. Adults know that CP=root. Kids don’t, so they Adults know that CP=root. Kids don’t, so they will sometimes stop early.will sometimes stop early.

PredictsPredicts: If TP is missing, so is CP.: If TP is missing, so is CP. Legendre et al. (2000): Like UCC, within Legendre et al. (2000): Like UCC, within OT.OT.

Rizzi and truncated Rizzi and truncated treestrees

Rizzi (1993/4): Kids lack the CP=root Rizzi (1993/4): Kids lack the CP=root axiom.axiom.

The result (of not having The result (of not having CP=rootCP=root) is ) is that kids are allowed to have that kids are allowed to have truncated truncated structuresstructures—trees that look like adult —trees that look like adult trees with the tops chopped off.trees with the tops chopped off.

ImportantlyImportantly: The kids don’t just leave : The kids don’t just leave stuff out—they just stop the tree stuff out—they just stop the tree “early.” So, if the kid leaves out a “early.” So, if the kid leaves out a functional projection, s/he leaves out functional projection, s/he leaves out all all higherhigher XPs as well. XPs as well.

Truncation: < TP < CPTruncation: < TP < CP

If kid selects anything lower If kid selects anything lower than TP as the root, the result than TP as the root, the result is a is a root infinitiveroot infinitive—which can be —which can be as big as any kind of XP below TP as big as any kind of XP below TP in the structure.in the structure.

Note in particular, though, it Note in particular, though, it can’tcan’t be a CP. be a CP.

So: we expect that evidence of CP So: we expect that evidence of CP will correlate with finite verbs.will correlate with finite verbs.

Truncation: TP < AgrSPTruncation: TP < AgrSP

Pierce (1989) looking at French Pierce (1989) looking at French observed that there are almost observed that there are almost no root infinitives with no root infinitives with subject clitics—this is subject clitics—this is predicted if these clitics are predicted if these clitics are instances of subject agreement instances of subject agreement in AgrS; if there is no TP, in AgrS; if there is no TP, there can be no AgrSP.there can be no AgrSP.

Truncation: TP <> NegP?Truncation: TP <> NegP? There is some dispute in the syntax There is some dispute in the syntax literature as to whether the position of literature as to whether the position of NegP (the projection responsible for the NegP (the projection responsible for the negative morpheme) is higher or lower than negative morpheme) is higher or lower than TP in the tree.TP in the tree.

If NegP is higher than TP, we would expect If NegP is higher than TP, we would expect not to find negative root infinitives.not to find negative root infinitives.

But we But we do do find negative RIs—(Pierce 1989): in the find negative RIs—(Pierce 1989): in the acquisition of French, negation follows finite verbs and acquisition of French, negation follows finite verbs and precedes nonfinite verbs (that is—French kids know the precedes nonfinite verbs (that is—French kids know the movement properties of finiteness, and thus they have movement properties of finiteness, and thus they have the concept of finiteness).the concept of finiteness).

So, is TP higher than NegP?So, is TP higher than NegP? Hard to say conclusively from the existing French data Hard to say conclusively from the existing French data because there are because there are not manynot many negative root infinitives—but negative root infinitives—but further study further study couldcould lead to a theoretical result of this lead to a theoretical result of this sort about the adult languages.sort about the adult languages.

S O VS O Vfinfin??

Usually (Poeppel & Wexler 1993) German Usually (Poeppel & Wexler 1993) German kids put finite verbs in second kids put finite verbs in second position, and leave nonfinite verbs at position, and leave nonfinite verbs at the end.the end.

OccasionallyOccasionally one finds a one finds a finitefinite verb at verb at the end.the end.

Rizzi suggests we could look at this as Rizzi suggests we could look at this as an instance of a kid choosing AgrSP as an instance of a kid choosing AgrSP as root, where CP is necessary to trigger root, where CP is necessary to trigger V2.V2.

*Truncation?: Where *Truncation?: Where train go?train go?

Truncation predictsTruncation predicts: If TP is missing, : If TP is missing, then CP should be missing.then CP should be missing.

But Bromberg & Wexler (1995) observe that But Bromberg & Wexler (1995) observe that bare verbs do appear in bare verbs do appear in whwh-questions in -questions in child English. child English. WhWh-questions implicate CP, -questions implicate CP, bare verbs implicate bare verbs implicate somethingsomething missing (TP missing (TP or AgrP). So, truncation can’t be right.or AgrP). So, truncation can’t be right.

Guasti notes that although the logic here Guasti notes that although the logic here works, English is weird in this respect: works, English is weird in this respect: pretty much all other languages pretty much all other languages dodo accord accord with the prediction.with the prediction.

““ATOM”ATOM” Schütze & Wexler Schütze & Wexler propose a propose a mmodel odel of this in which of this in which the case errors the case errors are a result of are a result of being able to being able to either either oomit mit AAgrSP or grSP or TTense.ense.

For a subject to For a subject to be in nominative be in nominative case, AgrSP must case, AgrSP must be therebe there (TP’s (TP’s presence is presence is irrelevant).irrelevant).

For a finite verb, For a finite verb, both both TP TP and AgrSP must be there.and AgrSP must be there. English inflection (3sg English inflection (3sg present –present –ss) relies on ) relies on both. both. If one or the other If one or the other is missing, we’ll see an is missing, we’ll see an infinitiveinfinitive (i.e. bare (i.e. bare stem).stem).

Thus, predicted:Thus, predicted: finite finite (AgrSP+TP) verbs show Nom (AgrSP+TP) verbs show Nom (AgrSP), but only half of (AgrSP), but only half of the nonfinite verbs (not the nonfinite verbs (not both AgrSP and TP) show both AgrSP and TP) show Nom (AgrSP). We should Nom (AgrSP). We should notnot see finite+Acc.see finite+Acc.

Implementing the UCCImplementing the UCC

The basic idea: The basic idea: In adult clauses, In adult clauses, the subject needs to move the subject needs to move bothboth to to SpecTP SpecTP and (then)and (then) to SpecAgrP. to SpecAgrP.

This needs to happen because T This needs to happen because T “needs” something in its “needs” something in its specifier (≈EPP) and so does Agr.specifier (≈EPP) and so does Agr.

The subject DP can “solve the The subject DP can “solve the problem” for both T and for Agr—problem” for both T and for Agr—for an adultfor an adult..

Implementing ATOMImplementing ATOM Implementation:Implementation: For adults: For adults:

T needs a D feature.T needs a D feature. Agr needs a D feature.Agr needs a D feature. The subject, happily, The subject, happily, hashas a D feature. a D feature. The subject moves to SpecTP, takes care The subject moves to SpecTP, takes care of T’s need for a D feature (the of T’s need for a D feature (the subject “checks” the D feature on T). subject “checks” the D feature on T). The T feature loses its need for a D The T feature loses its need for a D feature, but the subject still has its feature, but the subject still has its D feature (the subject is still a DP).D feature (the subject is still a DP).

The subject moves on, to take care of The subject moves on, to take care of Agr.Agr.

Implementing ATOMImplementing ATOM

Implementation:Implementation: For kids: For kids: Everything is the same except that Everything is the same except that the subject can only solve the subject can only solve one one problem before quittingproblem before quitting. It . It “loses” its D feature after “loses” its D feature after helping out either T or Agr.helping out either T or Agr.

Kids are constrained by the Kids are constrained by the Unique Unique Checking Constraint Checking Constraint that says that says subjects (or their D features) can subjects (or their D features) can only “check” another feature once.only “check” another feature once.

So the kids are in a bind.So the kids are in a bind.

Implementing ATOMImplementing ATOM Kids in a pickle:Kids in a pickle: The only options open to The only options open to the kids are:the kids are: Leave out TP Leave out TP (keep AgrP, the subject can solve (keep AgrP, the subject can solve Agr’s problem alone). Agr’s problem alone). Result: nonfinite verb, Result: nonfinite verb, nom case.nom case.

Leave out AgrP Leave out AgrP (keep TP, the subject can solve (keep TP, the subject can solve T’s problem alone). T’s problem alone). Result: nonfinite verb, Result: nonfinite verb, default case.default case.

Violate the UCC Violate the UCC (let the subject do both things (let the subject do both things anyway). anyway). Result: finite verb, nom case.Result: finite verb, nom case.

No matter which way you slice it, the kids No matter which way you slice it, the kids have to do something “wrong”. At that have to do something “wrong”. At that point, they choose randomly (but cf. point, they choose randomly (but cf. Legendre et al.)Legendre et al.)

Minimalist terminologyMinimalist terminology Features come in two relevant kinds: Features come in two relevant kinds: interpretable interpretable and and uninterpretableuninterpretable..

Either kind of feature can be involved in a Either kind of feature can be involved in a “checking”—only interpretable features “checking”—only interpretable features survive.survive.

The game is to have no uninterpretable The game is to have no uninterpretable features left at the end.features left at the end.

““T needs a DT needs a D” means “” means “T has an T has an uninterpretable [D] featureuninterpretable [D] feature” and the subject ” and the subject (with its normally interpretable [D] (with its normally interpretable [D] feature) comes along and the two features feature) comes along and the two features “check”, the interpretable one survives. “check”, the interpretable one survives. UCC=D uninterpretable on subjects?UCC=D uninterpretable on subjects?

Are kids really UG-Are kids really UG-constrained?constrained?

So, aren’t TP and AgrSP required So, aren’t TP and AgrSP required by UG? Doesn’t this mean kids by UG? Doesn’t this mean kids don’tdon’t have UG-compliant trees? have UG-compliant trees?

Actually, perhaps no. UG requires Actually, perhaps no. UG requires that all features be checked, but that all features be checked, but it isn’t clear that there is a UG it isn’t clear that there is a UG principle that requires a TP and principle that requires a TP and an AgrP in every clause.an AgrP in every clause.

Are kids really UG-Are kids really UG-constrained?constrained?

Perhaps what requires TP and AgrP are Perhaps what requires TP and AgrP are principles of (pragmatic) principles of (pragmatic) interpretationinterpretation……

You need TP so that your sentence is You need TP so that your sentence is “anchored” in the “anchored” in the discoursediscourse.. You need AgrSP … You need AgrSP … whywhy? Well, perhaps ? Well, perhaps something parallel…? Wexler doesn’t really something parallel…? Wexler doesn’t really say…say…

Regardless, kids can check all the Regardless, kids can check all the uninterpretable features even without uninterpretable features even without TP or AgrSP; hence, TP or AgrSP; hence, they can still be they can still be considered to be UG-constrainedconsidered to be UG-constrained..

NS/OI via UCCNS/OI via UCC

An old idea about NS languages is that An old idea about NS languages is that they arise in languages where Infl is they arise in languages where Infl is “rich” enough to “rich” enough to identifyidentify the subject. the subject.

Maybe in NS languages, AgrS does not Maybe in NS languages, AgrS does not needneed a D a D (it may in some sense be (it may in some sense be nouny enough to say that it nouny enough to say that it isis, or , or already already hashas, D)., D).

If AgrS does not need a D, the subject If AgrS does not need a D, the subject is free to check off T’s D-feature and is free to check off T’s D-feature and be done.be done.

Is there any way to see Is there any way to see the effects of UCC even in the effects of UCC even in

NSNS languages? languages? Italian: Italian: Mary has laughedMary has laughed.. Suppose that auxiliaries (like Suppose that auxiliaries (like havehave) also ) also have a D-feature to be checked as the have a D-feature to be checked as the subject (in the adult language) passes subject (in the adult language) passes through.through. Not crazy:Not crazy: (All) the students (all) have (All) the students (all) have (all) left.(all) left.

UCC-constrained kids will have to drop UCC-constrained kids will have to drop something (the auxiliary or T), even in something (the auxiliary or T), even in Italian.Italian.

Lyons (1997) reports that a “substantial Lyons (1997) reports that a “substantial proportion of auxiliaries are omitted in proportion of auxiliaries are omitted in OI-age Italian.”OI-age Italian.” Ok, Ok, maybemaybe. Consistent, anyway.. Consistent, anyway.

One open question…One open question…

The UCC says you can only use a D-The UCC says you can only use a D-feature on a DP to check against a feature on a DP to check against a functional category functional category onceonce..

This explains why sometimes TP is This explains why sometimes TP is omitted (keeping AgrSP) and omitted (keeping AgrSP) and sometimes AgrSP is omitted (keeping sometimes AgrSP is omitted (keeping TP).TP).

but if but if GEN infin. GEN infin. comes from comes from omitting omitting bothboth TP and AgrSP, what TP and AgrSP, what could ever cause that (particularly could ever cause that (particularly given given Minimize ViolationsMinimize Violations)?)?

Theories of missing Theories of missing structurestructure

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

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

Theories of missing Theories of missing structurestructure

““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.

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

Legendre et al. (2000)Legendre et al. (2000) Wexler: During OI stage, kids sometimes Wexler: During OI stage, kids sometimes omit T, and sometimes omit Agr. Based on a omit T, and sometimes omit Agr. Based on a choice of which to violate, the choice of which to violate, the requirement to have T, to have Agr, to requirement to have T, to have Agr, to have only one.have only one. (cf. “Kids in a pickle” slide)(cf. “Kids in a pickle” slide)

Legendre et al.: Looking at development Legendre et al.: Looking at development (of French), it appears that the choice of (of French), it appears that the choice of whatwhat to omit is systematic; we propose a to omit is systematic; we propose a system to account for (predict) the system to account for (predict) the proportion of the time kids omit T, Agr, proportion of the time kids omit T, Agr, both, neither, in progressive stages of both, neither, in progressive stages of development.development.

Optimality TheoryOptimality Theory

Legendre et al. (2000) is set Legendre et al. (2000) is set in the in the Optimality TheoryOptimality Theory framework (often seen in framework (often seen in phonology, less often seen phonology, less often seen applied to syntax).applied to syntax).

““Grammar is a system of ranked Grammar is a system of ranked and violable constraints”and violable constraints”


In our analysis, one constraint is In our analysis, one constraint is Parse-T, which says that tense must Parse-T, which says that tense must be realized in a clause. A be realized in a clause. A structure without tense (where TP structure without tense (where TP has been omitted, say) will violate has been omitted, say) will violate this constraint.this constraint.

Another constraint is *F (“Don’t Another constraint is *F (“Don’t have a functional category”). A have a functional category”). A structure structure withwith TP will violate this TP will violate this constraint.constraint.


Parse-T and *F are in Parse-T and *F are in conflictconflict—it —it is impossible to satisfy both at is impossible to satisfy both at the same time.the same time.

When constraints conflict, the When constraints conflict, the choice made (on a language-choice made (on a language-particular basis) of which particular basis) of which constraint is considered to be constraint is considered to be “more important” (“more important” (more highly more highly rankedranked) determines which ) determines which constraint is satisfied and which constraint is satisfied and which must be violated.must be violated.


So if *F >> Parse-T, TP will be So if *F >> Parse-T, TP will be omitted.omitted.

and if Parse-T >> *F, TP will and if Parse-T >> *F, TP will be included.be included.


Grammar involves Grammar involves constraintsconstraints on on the representations (e.g., SS, the representations (e.g., SS, LF, PF, or perhaps a combined LF, PF, or perhaps a combined representation).representation).

The constraints exist in The constraints exist in all all languageslanguages..

Where languages differ is in how Where languages differ is in how important each constraint is with important each constraint is with respect to each other constraint.respect to each other constraint.

Optimality Theory: big Optimality Theory: big picturepicture

Universal Grammar Universal Grammar isis the constraints the constraints that languages must obey.that languages must obey.

Languages differ only in how those Languages differ only in how those constraints are ranked relative to constraints are ranked relative to one another. (So, “parameter” = one another. (So, “parameter” = “ranking”)“ranking”)

The kid’s job is to re-rank The kid’s job is to re-rank constraints until they match the constraints until they match the order which generated the input that order which generated the input that s/he hears.s/he hears.

Legendre et al. (2000)Legendre et al. (2000)

Proposes a system to predict Proposes a system to predict the proportions of the time the proportions of the time kids choose the different kids choose the different options among:options among: Omit TPOmit TP Omit AgrSPOmit AgrSP Omit Omit bothboth TP and AgrSP TP and AgrSP Include Include bothboth TP and AgrSP TP and AgrSP (violating UCC)(violating UCC)

French v. EnglishFrench v. English

English: T+Agr is pronounced likeEnglish: T+Agr is pronounced like //ss/ if we have features [3, sg, / if we have features [3, sg, present]present]

//eded/ if we have the feature [past]/ if we have the feature [past] //ØØ/ otherwise/ otherwise

French: T+Agr is pronounced like:French: T+Agr is pronounced like: danserdanser NRFNRF a danséa dansé (3sg) past(3sg) past je danseje danse 1sg (present)1sg (present) j’ai danséj’ai dansé 1sg past1sg past

The ideaThe idea

Kids are subject to conflicting Kids are subject to conflicting constraints:constraints: Parse-TParse-T Include a projection for tenseInclude a projection for tense Parse-AgrParse-Agr Include a project for Include a project for agreementagreement

*F*F Don’t complicate your tree withDon’t complicate your tree withfunctional projectionsfunctional projections

*F*F22 Don’t complicate your tree soDon’t complicate your tree somuch as to have much as to have twotwo

functionalfunctionalprojections.projections.

The ideaThe idea

Sometimes Parse-T beats out *F, Sometimes Parse-T beats out *F, and then there’s a TP. Or and then there’s a TP. Or Parse-Agr beats out *F, and Parse-Agr beats out *F, and then there’s an AgrP. Or both then there’s an AgrP. Or both Parse-T and Parse-Agr beat out Parse-T and Parse-Agr beat out *F*F22, and so there’s both a TP , and so there’s both a TP and an AgrP.and an AgrP.

But what does But what does sometimessometimes mean? mean?

Floating constraintsFloating constraints

The innovation in Legendre et al. The innovation in Legendre et al. (2000) that gets us off the ground (2000) that gets us off the ground is the idea that as kids re-rank is the idea that as kids re-rank constraints, the constraints, the positionposition of the of the constraint in the hierarchy can constraint in the hierarchy can get somewhat fuzzy, such that two get somewhat fuzzy, such that two positions can positions can overlapoverlap..

*F*F

Parse-TParse-T


*F*F

Parse-TParse-T

When the kid evaluates a form When the kid evaluates a form in the constraint system, the in the constraint system, the position of Parse-T is fixed position of Parse-T is fixed somewhere in the range—and somewhere in the range—and winds up sometimes outranking, winds up sometimes outranking, and sometimes outranked by, *F.and sometimes outranked by, *F.


*F*F

Parse-TParse-T

(Under certain assumptions) (Under certain assumptions) this predicts that we would this predicts that we would see TP in the structure 50% see TP in the structure 50% of the time, and see of the time, and see structures without TP the structures without TP the other 50% of the time.other 50% of the time.

French kid dataFrench kid data

Looked at 3 French kids from CHILDESLooked at 3 French kids from CHILDES Broke development into stages based on Broke development into stages based on a modified MLU-type measure based on a modified MLU-type measure based on how long most of their utterances were how long most of their utterances were (2 words, more than 2 words) and how (2 words, more than 2 words) and how many of the utterances contain verbs.many of the utterances contain verbs.

Looked at tense and agreement in each Looked at tense and agreement in each of the three stages represented in the of the three stages represented in the data.data.


Kids start out using 3sg Kids start out using 3sg agreement and present tense for agreement and present tense for practically everything (correct practically everything (correct or not).or not).

We took this to be a “default”We took this to be a “default” (No agreement? Pronounce it as (No agreement? Pronounce it as 3sg. No tense? pronounce it as 3sg. No tense? pronounce it as present. Neither? Pronounce it as present. Neither? Pronounce it as an infinitive.).an infinitive.).


This means if a kid uses 3sg or This means if a kid uses 3sg or present tense, we can’t tell if present tense, we can’t tell if they are they are reallyreally using 3sg (they using 3sg (they mightmight be) or if they are not using be) or if they are not using agreement at all and just agreement at all and just pronouncing the default.pronouncing the default.

So, we looked at non-present tense So, we looked at non-present tense forms and non-3sg forms only to forms and non-3sg forms only to avoid the question of the avoid the question of the defaults.defaults.

French kids dataFrench kids data

We found that tense and agreement We found that tense and agreement develop differently—specifically, develop differently—specifically, in the first stage we looked at, in the first stage we looked at, kids were using tense fine, but kids were using tense fine, but then in the next stage, they got then in the next stage, they got worseworse as the agreement improved. as the agreement improved.

Middle stage: looks likeMiddle stage: looks likecompetitioncompetition between T between Tand Agr for a single node.and Agr for a single node.

A detail about countingA detail about counting We counted non-3sg and non-present verbs.We counted non-3sg and non-present verbs. In order to see how close kids’ utterances In order to see how close kids’ utterances were to adult’s utterances, we need to know were to adult’s utterances, we need to know how often how often adultsadults use non-3sg and non-present, use non-3sg and non-present, and then see how close the kids are to and then see how close the kids are to matching that level.matching that level.

So, adults use non-present tense around 31% of So, adults use non-present tense around 31% of the time—so when a kid uses 31% non-present the time—so when a kid uses 31% non-present tense, we take that to be “100% success”tense, we take that to be “100% success”

In the last stage we looked at, kids were In the last stage we looked at, kids were basically right at the “100% success” level basically right at the “100% success” level for both tense and agreement.for both tense and agreement.

Proportion of non-Proportion of non-present and non-3sg present and non-3sg

verbsverbs

0%

5%

10%

15%

20%

25%

30%

35%

40%

3b 4b 4c

non-presentnon-3sgadult non-presadult non-3sg

Proportion of non-Proportion of non-finite root formsfinite root forms

0%

5%

10%

15%

20%

25%

30%

35%

3b 4b 4c

NRFs

A model to predict the A model to predict the percentagespercentages

Stage 3b (first stage)Stage 3b (first stage) no agreementno agreement about 1/3 NRFs, 2/3 tensed formsabout 1/3 NRFs, 2/3 tensed forms

*F*F22 *F*FParseTParseT

ParseAParseA


Stage 4b (second stage)Stage 4b (second stage) non-3sg agreement and non-non-3sg agreement and non-present tense each about 15% present tense each about 15% (=about 40% agreeing, 50% (=about 40% agreeing, 50% tensed)tensed)

about 20% NRFsabout 20% NRFs

*F*F22 *F*FParseTParseTParseAParseA


Stage 4c (third stage)Stage 4c (third stage) everything appears to have everything appears to have tense and agreement (adult-like tense and agreement (adult-like levels)levels)

*F*F22 *F*FParseTParseTParseAParseA

Predicted vs. observed—Predicted vs. observed—tensetense

0%

5%

10%

15%

20%

25%

30%

35%

40%

3b 4b 4c

non-presentpredicted non-pres

Predicted vs. observed—Predicted vs. observed—agr’tagr’t

0%

5%

10%

15%

20%

25%

30%

35%

40%

3b 4b 4c

non-3sgpredicted non-3sg

Predicted vs. observed—Predicted vs. observed—NRFsNRFs

0%

5%

10%

15%

20%

25%

30%

35%

3b 4b 4c

NRFspredicted NRFs

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 leaving intransitive subjects , but leaving intransitive subjects and objects unmarked.and objects unmarked.

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

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

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

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

Further evidence:Further evidence: Ergative case marker is often Ergative case marker is often homophonous with marker for possessive homophonous with marker for possessive (cf. Inuktitut (cf. Inuktitut -up-up used for both), and used for both), and Ben uses Ben uses forfor (his ERG marker) in (his ERG marker) in possessive constructions as well.possessive constructions as well.

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

The Bennish optativeThe Bennish optative Further evidence:Further evidence:

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

Korean negation?Korean negation?

The UCC is about checking D, and that The UCC is about checking D, and that happens not only for subjects but for happens not only for subjects but for objects.objects.

In English objects don’t have to double-In English objects don’t have to double-check, but are there effects in other check, but are there effects in other languages?languages?

Hagstrom (2000) looks at errors with Hagstrom (2000) looks at errors with negation made by children learning negation made by children learning Korean at about the same age that, in Korean at about the same age that, in other languages, kids are producing root other languages, kids are producing root infinitives.infinitives. Fairly technical and minimalist, but if you Fairly technical and minimalist, but if you survive Wexler 1998, you’re most of the way survive Wexler 1998, you’re most of the way there.there.

Korean negation?Korean negation? Short Form Negation in Korean:Short Form Negation in Korean:

Chelswu-ka Chelswu-ka pappap-ul -ul anan-mek-ess--mek-ess-ta.ta.Chelswu-nom Chelswu-nom ricerice-acc -acc negneg-eat-past-decl-eat-past-decl‘Chelswu didn’t eat rice.’‘Chelswu didn’t eat rice.’

Common OI-age kid error:Common OI-age kid error:

na na anan pappap mek-e mek-eI I negneg ricerice eat-decl eat-decl‘I don’t eat rice.’‘I don’t eat rice.’

Negation errors in Negation errors in child Koreanchild Korean

Generalization about child errors with Generalization about child errors with SFN:SFN:VP-internal material is privileged in VP-internal material is privileged in its ability to occur between its ability to occur between anan and the and the verb in child errorsverb in child errors..

Subjects (Subjects (except subjects of except subjects of unaccusativesunaccusatives) never appear between ) never appear between anan and the verband the verb

Objects often doObjects often do Adverbs often doAdverbs often do

Negation errors in Negation errors in child Koreanchild Korean

Can this error be made to follow from the Can this error be made to follow from the UCC (you can’t check a D-feature twice)?UCC (you can’t check a D-feature twice)?

Kid errors seem to involve a structure Kid errors seem to involve a structure like:like:

negneg […VP material… ] […VP material… ] verbverb

suggesting that adult negation has a suggesting that adult negation has a movement that kids are failing to do:movement that kids are failing to do:

[…VP material…][…VP material…]ii neg tneg tii verbverb

One movement down…One movement down… For the UCC to apply, there need to be two For the UCC to apply, there need to be two movements. Do adults move the object movements. Do adults move the object twice?twice?

Adults also seem to perform a Adults also seem to perform a secondsecond movement of the object; the adverb movement of the object; the adverb calcal ‘well’ must immediately precede the verb ‘well’ must immediately precede the verb (unlike other adverbs)—but presumably the (unlike other adverbs)—but presumably the object object originallyoriginally (at D-structure) falls (at D-structure) falls between between calcal and the verb. Hence: and the verb. Hence:

objectobjectii … … calcal … … ttii verbverb

That’s two movementsThat’s two movements

So, the object (and some of the So, the object (and some of the VP-internal material) seems to VP-internal material) seems to have to move have to move twice twice in negative in negative sentences, once to get around sentences, once to get around calcal (in any kind of sentence), and (in any kind of sentence), and again to get around again to get around anan (neg). (neg).

That’s what we need to get off That’s what we need to get off the ground if we want to the ground if we want to attribute this error to the UCC.attribute this error to the UCC.

The proposalThe proposal

In Korean, the object moves to In Korean, the object moves to SpecAgrOP (step one) and checks a D-SpecAgrOP (step one) and checks a D-feature:feature:

AgrOPAgrOP

DPDPii AgrOAgrO

AgrOAgrO[D][D] VPVP

calcal VP VP

VV ttii


Then, AgrOP moves to an Then, AgrOP moves to an AgrNegP above negation, to AgrNegP above negation, to check a D-feature:check a D-feature:

AgrNegPAgrNegP

AgrOPAgrOPii AgrNegAgrNeg

AgrNegAgrNeg[D][D] NegPNegP

anan NegNeg

NegNeg ……ttii……


The kid can only do The kid can only do oneone of of those movements if it obeys the those movements if it obeys the UCC, since each one requires UCC, since each one requires the same D-feature (contributed the same D-feature (contributed by the object).by the object).

So, the kid must eitherSo, the kid must either ignore the UCC, orignore the UCC, or omit AgrOP, oromit AgrOP, or omit AgrNegPomit AgrNegP

PredictionsPredictions 1: Omit AgrNegP (retaining AgrOP):1: Omit AgrNegP (retaining AgrOP):

Object moves (over Object moves (over calcal) to SpecAgrOP. AgrOP (with ) to SpecAgrOP. AgrOP (with calcal and object) remains below NegP. and object) remains below NegP.

an object cal verb an object cal verb (*, non-adult-like)(*, non-adult-like) 2: Omit AgrOP (retaining AgrNegP)2: Omit AgrOP (retaining AgrNegP)

Object (nearest thing with a D-feature) moves Object (nearest thing with a D-feature) moves directly to SpecAgrNegP, over directly to SpecAgrNegP, over anan and and calcal..

object an cal verb object an cal verb (*, but needs (*, but needs calcal to be to be present)present)

3: Violate UCC (keep AgrOP & AgrNegP)3: Violate UCC (keep AgrOP & AgrNegP) object cal an verbobject cal an verb (adult-like) (adult-like)

4: Omit both AgrOP & AgrNegP?4: Omit both AgrOP & AgrNegP? an cal object verb an cal object verb (*, without (*, without cal cal looks like looks like 1)1)

Met?Met?

Sadly, the experiments haven’t Sadly, the experiments haven’t been done and the examples been done and the examples haven’t been reported in the haven’t been reported in the literature.literature. We need We need errorserrors with with transitivetransitive verbs involving verbs involving short-form short-form negationnegation and the adverb and the adverb calcal…… Possibly fairly easy elicitation Possibly fairly easy elicitation experiment that can be done…experiment that can be done…

Predictions for Predictions for unaccusativesunaccusatives

Unaccusative “subjects” start out in Unaccusative “subjects” start out in object position, and must presumably object position, and must presumably move through many more projections move through many more projections (AgrOP, AgrNegP, TP, AgrSP)(AgrOP, AgrNegP, TP, AgrSP)

UCC kid can still just do UCC kid can still just do oneone.. Only one (of five) will yield a non-Only one (of five) will yield a non-adult order: keep AgrOP and you get: adult order: keep AgrOP and you get: an an subject cal verbsubject cal verb..

Turns out: kids make only about 10% Turns out: kids make only about 10% (detectible) errors with unaccusatives (detectible) errors with unaccusatives (vs. about 30% with transitives). A (vs. about 30% with transitives). A successful prediction?successful prediction?

So…So… The UCC seems to be pretty successful The UCC seems to be pretty successful in explaining why either TP or AgrSP in explaining why either TP or AgrSP are often omitted for kids in languages are often omitted for kids in languages like French, German.like French, German.

The connection to the NS/OI The connection to the NS/OI generalization is reasonable to explain generalization is reasonable to explain why we don’t seem to see OIs in why we don’t seem to see OIs in Italian.Italian.

The more general prediction that the The more general prediction that the UCC makes about double-movements to UCC makes about double-movements to check D-features may well be borne out check D-features may well be borne out by the facts of Korean negation. by the facts of Korean negation.

week 3. further notes on non-finite root forms grs lx 700 language acquisition and linguistic theory

Documents