Week 8. Control and PRO CAS LX 522 Syntax I. Back to business… Mary is likely to leave. Mary is likely to leave. Mary starts in SpecVP, gets a -role

Week 8. Control and PROWeek 8. Control and PRO

CAS LX 522CAS LX 522Syntax ISyntax I

Back to Back to business…business…

Mary is likely to leave.Mary is likely to leave. Mary Mary starts in SpecVP,starts in SpecVP,

gets a gets a -role from -role from leaveleave..

Adjlikely

AdjAdjP

DPiMary

VVPT

[pres]

T

TP DS

VPTto

T

TP

leave

VV

C

C

[–Q]

CP

Vbe

Recall…Recall…


gets a gets a -role from -role from leaveleave.. MaryMary moves up to the moves up to the

embedded SpecTP toembedded SpecTP tosatisfy the EPP.satisfy the EPP.

MaryMary still doesn’t have still doesn’t have Case.Case.

Adjlikely

AdjAdjP

DPiMary

tj

VVPVj+T

be+[pres]

T

TP

VPTto

T

TP

leave

VV

ti

C

C

[–Q]

CP

Recall…Recall…


gets a gets a -role from -role from leaveleave.. MaryMary moves up to the moves up to the

embedded SpecTP toembedded SpecTP tosatisfy the EPP.satisfy the EPP.

MaryMary still doesn’t have still doesn’t have Case.Case.

MaryMary moves up to main moves up to main clause SpecTP, satisfying clause SpecTP, satisfying the EPP and getting the EPP and getting Case.Case.

Adjlikely

AdjAdjP

DPiMary

tj

VVPVj+T

be+[pres]

T

TP SS

VPTto

T

TP

leave

ti

VV

ti

C

C

[–Q]

CP

(Note how wewrite multipletraces)

Recall…Recall…

This happens This happens because because likelylikely assigns only one assigns only one --role, an internal role, an internal --role.role.

LikelyLikely does not does not assign Case, and so assign Case, and so Mary Mary must keep must keep moving, both to moving, both to satisfy the EPP and satisfy the EPP and to get Case.to get Case.

Adjlikely

AdjAdjP

DPiMary

tj

VVPVj+T

be+[pres]

T

TP SS

VPTto

T

TP

leave

ti

VV

ti

C

C

[–Q]

CP

Reluctance to leaveReluctance to leave

Now, consider:Now, consider: Mary is reluctant to leave.Mary is reluctant to leave.

This looks very similar to This looks very similar to Mary is likely to Mary is likely to leaveleave..

Can we draw the same kind of tree for it?Can we draw the same kind of tree for it?

How many How many -roles does -roles does reluctant reluctant assign?assign?


Reluctant Reluctant has has twotwo -roles to assign.-roles to assign. One to the one feeling the reluctance One to the one feeling the reluctance

((ExperiencerExperiencer)) One to the proposition about which the reluctance One to the proposition about which the reluctance

holds (holds (PropositionProposition))

Leave Leave has one has one -role to assign.-role to assign. To the one doing the leaving (To the one doing the leaving (AgentAgent).).

In In Mary is reluctant to leaveMary is reluctant to leave, what , what -role does -role does Mary Mary get?get?


In In Mary is reluctant to leaveMary is reluctant to leave,, MaryMary is doing the leaving, gets is doing the leaving, gets

AgentAgent from from leaveleave.. MaryMary is showing the reluctance, is showing the reluctance,

gets gets ExperiencerExperiencer from from reluctantreluctant..

And we have a problem:And we have a problem: Mary Mary appears to be getting two appears to be getting two --

roles, in violation of the roles, in violation of the --criterioncriterion..

ReluctanceReluctance……

Mary is reluctant to Mary is reluctant to leave.leave.

ReluctantReluctant assigns its assigns its --roles within AdjP as roles within AdjP as required, required, MaryMary moves moves up to SpecTP in the up to SpecTP in the main clause by SS.main clause by SS.

But what gets the But what gets the --role from role from leaveleave, and , and what satisfies the EPP what satisfies the EPP for the embedded for the embedded clause?clause?

Adjreluctant

Adj

AdjP

DPiMary

tj

VVPVj+T

is

T

TP

SS

VPTto

T

TP

leave

ti

?

VV

?



There must be There must be somethingsomething there, there, getting the getting the -role and -role and satisfying the EPP.satisfying the EPP.

But we can’t see it.But we can’t see it.

It’s a phonologically It’s a phonologically empty (Ø) DP. We will empty (Ø) DP. We will call it PRO.call it PRO.

Adjreluctant

Adj

AdjP

DPiMary

tj

VVPVj+T

is

T

TP

SS

VPTto

T

TP

leave

ti

?

VV

?



There must be There must be somethingsomething there, there, getting the getting the -role and -role and satisfying the EPP.satisfying the EPP.

But we can’t see it.But we can’t see it.

It’s a phonologically It’s a phonologically empty (Ø) DP. We will empty (Ø) DP. We will call it PRO.call it PRO.

Adjreluctant

Adj

AdjP

DPiMary

tj

VVPVj+T

is

T

TP

SS

VPTto

T

TP

leave

ti

VV

tk

DPkPRO

ReluctanceReluctance…… Mary is reluctantMary is reluctant

[PRO to leave].[PRO to leave].

PRO does not get Case.PRO does not get Case. *Mary is reluctant Bill to *Mary is reluctant Bill to

leave.leave. In fact, PRO In fact, PRO cannotcannot get get

Case.Case. *Mary is reluctant for to *Mary is reluctant for to

leaveleave Mary is reluctant for Bill to Mary is reluctant for Bill to

leaveleave PRO refers (like a pronoun PRO refers (like a pronoun

or an anaphor) to or an anaphor) to MaryMary..

Adjreluctant

Adj

AdjP

DPiMary

tj

VVPVj+T

is

T

TP

SS

VPTto

T

TP

leave

ti

VV

tk

DPkPRO

If there’s a PRO,If there’s a PRO,how do we know?how do we know?

Mary is reluctant [PROMary is reluctant [PROmm to leave] to leave]

MaryMaryii is likely [ is likely [ ttii to leave]. to leave].

These two sentences look These two sentences look very muchvery much alike—when faced with a sentence alike—when faced with a sentence that looks like this, how do we know that looks like this, how do we know which kind it is?which kind it is?

If there’s a PRO,If there’s a PRO,how do we know?how do we know?

Best method for finding PRO:Best method for finding PRO: Count Count the the -roles. If there appear to be fewer -roles. If there appear to be fewer arguments than arguments than -roles (in a -roles (in a grammatical sentence), there must be grammatical sentence), there must be a PRO.a PRO.

Another way is to try with Another way is to try with idiomsidioms like like The cat is out of the bagThe cat is out of the bag or or The cat’s The cat’s got your tonguegot your tongue or or The jig is upThe jig is up..

IdiomsIdioms

For something to have an idiomatic For something to have an idiomatic interpretation (an interpretation not interpretation (an interpretation not literally derivable from its literally derivable from its component words), the pieces need component words), the pieces need to be very close together at DS.to be very close together at DS. It is likely that the jig is up.It is likely that the jig is up. It is likely that the cat is out of the bag.It is likely that the cat is out of the bag. It is likely that the cat has your tongue.It is likely that the cat has your tongue.

IdiomsIdioms

It is ok if the pieces of the idiom move It is ok if the pieces of the idiom move away after DS, we can still get the away after DS, we can still get the idiomatic interpretation:idiomatic interpretation: [The cat][The cat]ii is likely is likely ttii to have your tongue. to have your tongue. [The cat][The cat]ii is likely is likely ttii to be out of the bag. to be out of the bag. [The jig][The jig]ii is likely is likely ttii to be up. to be up.

The important thing is that they are The important thing is that they are together at DS (the together at DS (the -role needs to be -role needs to be assigned by the predicate to the noun)assigned by the predicate to the noun)

IdiomsIdioms If we break up the pieces, then we lose If we break up the pieces, then we lose

the idiomatic interpretation and can only the idiomatic interpretation and can only get the literal meaning.get the literal meaning. The cat thinks that it is out of the bag.The cat thinks that it is out of the bag. The cat thinks that it has your tongue. The cat thinks that it has your tongue.

With PRO sentences (“control With PRO sentences (“control sentences”), we also lose the idiomatic sentences”), we also lose the idiomatic reading.reading. #The cat is reluctant to be out of the bag.#The cat is reluctant to be out of the bag. #The cat attempted to have your tongue.#The cat attempted to have your tongue. #The jig tried to be up.#The jig tried to be up.

IdiomsIdioms

The reason for this is that the idiomatic The reason for this is that the idiomatic subject and the idiomatic predicate were subject and the idiomatic predicate were never together…never together… The cat is reluctant [PRO to be out of the bag]The cat is reluctant [PRO to be out of the bag] The cat attempted [PRO to have your tongue]The cat attempted [PRO to have your tongue] The jig tried [PRO to be up]The jig tried [PRO to be up]

Unlike with raising verbs:Unlike with raising verbs: [The jig][The jig]ii is likely [ is likely [ ttii to be up] to be up]

ControlControl

PRO is similar to a silent pronoun; it PRO is similar to a silent pronoun; it gets its referent from somewhere gets its referent from somewhere outside its sentence. In many outside its sentence. In many situations, however, PRO is situations, however, PRO is forcedforced to to co-refer to a preceding DP, unlike a co-refer to a preceding DP, unlike a pronoun.pronoun. BillBillii thinks that he thinks that hei/ji/j is a genius. is a genius. BillBillii is reluctant PRO is reluctant PROi/*ji/*j to leave. to leave.

We say that PRO is We say that PRO is controlledcontrolled (here by the matrix subject).(here by the matrix subject).

Subject and object Subject and object controlcontrol

There are actually two different kinds There are actually two different kinds of “control verbs”, those whose of “control verbs”, those whose subject controls an embedded PRO subject controls an embedded PRO and those whose object does.and those whose object does.

BillBillii is reluctant [PRO is reluctant [PROii to leave] to leave] reluctantreluctant is a is a subject control predicatesubject control predicate

JohnJohnii persuaded Bill persuaded Billjj [PRO [PROjj to leave] to leave] persuadepersuade is an is an object control predicateobject control predicate

PROPROarbarb

Finally, there is a third use of PRO, in Finally, there is a third use of PRO, in which it gets which it gets arbitrary referencearbitrary reference and and means something like “someone/anyone”.means something like “someone/anyone”. [PRO[PROarbarb to leave] would be a mistake. to leave] would be a mistake.

The conditions on which interpretation The conditions on which interpretation PRO can/must get are referred to as PRO can/must get are referred to as Control TheoryControl Theory, although to this day the , although to this day the underlying explanation for Control underlying explanation for Control remains elusive.remains elusive.

““Control theory”Control theory”

For now, what control theory consists of is For now, what control theory consists of is just marking the theta grids of specific just marking the theta grids of specific predicates (predicates (persuadepersuade, , reluctantreluctant) with an ) with an extra notation that indicates when an extra notation that indicates when an argument is a controller.argument is a controller.reluctant Experiencer

controllerProposition

i j

persuade Agent Themecontroller

Proposition

i j k

““Control theory”Control theory”

Predicates that have a controller marked Predicates that have a controller marked are are control predicatescontrol predicates. When the controller . When the controller is the external argument, it is a is the external argument, it is a subject subject controlcontrol predicate predicate, otherwise it is an , otherwise it is an object object controlcontrol predicate predicate..reluctant Experiencer

controllerProposition

i j

persuade Agent Themecontroller

Proposition

i j k

The PRO conundrumThe PRO conundrum Back when we talked about Binding Theory, Back when we talked about Binding Theory,

we said that DPs come in one of three types, we said that DPs come in one of three types, pronounspronouns, , anaphorsanaphors, and , and R-expressionsR-expressions..

PRO is a DP, so which kind is it?PRO is a DP, so which kind is it? It gets its reference from elsewhere, so it can’t be It gets its reference from elsewhere, so it can’t be

an R-expression.an R-expression. It is sometimes It is sometimes forcedforced to get its referent from an to get its referent from an

antecedent, like an anaphor and unlike a antecedent, like an anaphor and unlike a pronoun.pronoun.

But that referent is outside its clause, meaning it But that referent is outside its clause, meaning it can’t be an anaphor (the antecedent would be too can’t be an anaphor (the antecedent would be too far away for Principle A). Plus, it’s not far away for Principle A). Plus, it’s not alwaysalways forced (PROforced (PROarbarb), like a pronoun.), like a pronoun.

The PRO conundrumThe PRO conundrum Back when we talked about Binding Theory, Back when we talked about Binding Theory,

we said that DPs come in one of three types, we said that DPs come in one of three types, pronounspronouns, , anaphorsanaphors, and , and R-expressionsR-expressions..

PRO is a DP, so which kind is it?PRO is a DP, so which kind is it?

Conclusion:Conclusion: It doesn’t seem to be any one of It doesn’t seem to be any one of the three. the three. It doesn’t seem to fall neatly It doesn’t seem to fall neatly under Binding Theoryunder Binding Theory

……hence, we need “Control Theory” to deal hence, we need “Control Theory” to deal with the distribution and interpretation of with the distribution and interpretation of PRO.PRO.

The PRO conundrumThe PRO conundrum

These weird properties of PRO are These weird properties of PRO are sometimes taken to be the cause of another sometimes taken to be the cause of another generalization about PRO generalization about PRO (the “PRO (the “PRO theorem”)theorem”)

PRO PRO cannotcannot get Case. get Case.

That is, PRO is forbidden from any position That is, PRO is forbidden from any position where Case would be assigned to it (hence, where Case would be assigned to it (hence, it cannot appear in SpecTP of a finite it cannot appear in SpecTP of a finite clause—only a nonfinite clause)clause—only a nonfinite clause)

Control TheoryControl Theory

Despite the fact that PRO does not submit Despite the fact that PRO does not submit to Binding Theory, there are some binding-to Binding Theory, there are some binding-theory-like requirements on control of PRO.theory-like requirements on control of PRO.

PRO is only obligatorily controlled by a c-PRO is only obligatorily controlled by a c-commanding controller.commanding controller.

[Bill[Billjj’s mother]’s mother]ii is reluctant [PRO is reluctant [PROi/*ji/*j to leave] to leave]

PRO: One possiblePRO: One possiblepiece of supportpiece of support

Let’s think back to Binding Theory.Let’s think back to Binding Theory. Principle A says that Principle A says that anaphors must be anaphors must be

bound within their binding domainbound within their binding domain, and , and we take binding domain to be the we take binding domain to be the clause.clause. *Bill wants [Mary to meet himself]*Bill wants [Mary to meet himself]

However, now consider:However, now consider: Bill is reluctant to buy himself a gift.Bill is reluctant to buy himself a gift. Bill promised Mary to buy himself a gift.Bill promised Mary to buy himself a gift.

Why are these allowed?Why are these allowed?

PRO: One possiblePRO: One possiblepiece of supportpiece of support

BillBillii is reluctant [PRO is reluctant [PROii to buy himself to buy himselfii a gift] a gift]

BillBillii promised Mary [PRO promised Mary [PROii to buy himself to buy himselfii a gift] a gift]

*Bill*Billii promised Mary promised Maryjj [PRO [PROii to buy herself to buy herselfjj a gift] a gift]

*Bill*Billii promised Mary promised Maryjj [PRO [PROii to buy him to buy himii a gift] a gift]

BillBillii promised Mary promised Maryjj [PRO [PROii to buy her to buy herjj a gift] a gift]

*Bill*Billii is reluctant [PRO is reluctant [PROii to buy him to buy himii a gift] a gift]

While it’s true that While it’s true that BillBill is outside of the is outside of the binding domain of binding domain of himselfhimself, and hence , and hence BillBill cannotcannot be the antecedent for be the antecedent for himselfhimself, PRO , PRO isis in the binding domain and its reference is in the binding domain and its reference is controlled.controlled.

PRO: recapPRO: recap

Although we can’t see that PRO is Although we can’t see that PRO is there, all of our theoretical there, all of our theoretical mechanisms point to its being there.mechanisms point to its being there. EPP says that clauses need a subject.EPP says that clauses need a subject. The The -criterion says that there must be -criterion says that there must be

exactly as many arguments as exactly as many arguments as -roles.-roles. Binding Theory indicates something is Binding Theory indicates something is

present inside embedded clauses.present inside embedded clauses. If the rest of our theory is right, it If the rest of our theory is right, it

seems that PRO seems that PRO mustmust be there. be there.

Back to raisingBack to raising

So far, we’ve only talked about So far, we’ve only talked about is likelyis likely, but , but there are a couple of other raising verbs as there are a couple of other raising verbs as well.well. [The cat][The cat]ii seemsseems [ [TPTP ttii to be out of the bag]. to be out of the bag].

[The cat][The cat]ii appearsappears [ [TPTP ttii to have his tongue]. to have his tongue].

[The jig][The jig]ii provedproved [ [TPTP ttii to be up]. to be up].

[The cat][The cat]ii beganbegan [ [TPTP ttii to get his tongue]. to get his tongue].

What these verbs have in common is that What these verbs have in common is that they have no external they have no external -role and an internal -role and an internal Proposition Proposition -role.-role.


In fact, nothing keeps us from piling In fact, nothing keeps us from piling raising verbs one atop the other:raising verbs one atop the other: [The cat][The cat]ii seems [ seems [ ttii likely [ likely [ ttii to get his to get his

tongue]].tongue]]. [The jig][The jig]ii began [ began [ ttii to seem [ to seem [ ttii likely [ likely [ ttii to be to be

up]]]up]]]

In these cases, the subject moves from In these cases, the subject moves from SpecTP to SpecTP, only receiving Case at SpecTP to SpecTP, only receiving Case at the last stop, satisfying the EPP at each TP.the last stop, satisfying the EPP at each TP.


Raising verbs will cause anything in a Raising verbs will cause anything in a complement TP that isn’t getting Case to complement TP that isn’t getting Case to move up to their SpecTP.move up to their SpecTP.

Passive arguments:Passive arguments: [The sandwich][The sandwich]ii seems [ seems [ ttii to have been [ eaten to have been [ eaten ttii]]]]

Even expletive Even expletive itit:: ItItii began [ began [ ttii to rain] to rain] ItItii began [ began [ ttii to seem [ to seem [ ttii likely [ likely [ ttii to rain]]] to rain]]]

Here, Here, itit was inserted to satisfy the EPP in the was inserted to satisfy the EPP in the most embedded TP, but then raised from most embedded TP, but then raised from SpecTP to SpecTP to satisfy the rest of their SpecTP to SpecTP to satisfy the rest of their EPP conditions.EPP conditions.

Side note: ChainsSide note: Chains

[The jig][The jig]ii began [ began [ ttii to seem [ to seem [ ttii likely [ likely [ ttii to be to be up]]]up]]]

Some time ago we saw the term Some time ago we saw the term chainchain applied to the concept of applied to the concept of positions occupied positions occupied by a (moving) constituent in a structureby a (moving) constituent in a structure..

Here, the Here, the chain chain for for The jigThe jig is: is: ( [The jig]( [The jig]ii, , ttii , , ttii , , ttii ) )

……referring to all the places its been in the referring to all the places its been in the tree. tree.

Side note: ChainsSide note: Chains

[The jig][The jig]ii began [ began [ ttii to seem [ to seem [ ttii likely [ likely [ ttii to be to be up]]]up]]]

Chain: ( [The jig]Chain: ( [The jig]ii, , ttii , , ttii , , ttii ) )

If we consider the If we consider the chainchain as a coherent entity, as a coherent entity, we can state conditions in a slightly nicer we can state conditions in a slightly nicer way:way: Every (argument) chain gets exactly one Every (argument) chain gets exactly one -role.-role. Every (argument) chain receives Case. Every (argument) chain receives Case. (except PRO’s)(except PRO’s)

Doing this allows us to avoid saying every Doing this allows us to avoid saying every argument gets case argument gets case at some pointat some point, and a , and a --role role at some different pointat some different point..

Italian subjectsItalian subjects

Many languages have the property Many languages have the property that when the subject is understood that when the subject is understood (often in the cases where in English we (often in the cases where in English we would use a pronoun subject), it can be would use a pronoun subject), it can be just left out entirely. For example, just left out entirely. For example, Italian:Italian:

Parlo.Parlo. Parli.Parli.speak-1sspeak-1s speak-2sspeak-2s‘I speak’‘I speak’ ‘You speak’‘You speak’

Italian subjectsItalian subjects

So what about the EPP and the So what about the EPP and the --criterion?criterion? Clearly Clearly ‘speak’‘speak’ assigns a assigns a --role, and presumably the Italian role, and presumably the Italian SpecTP needs to be filled as well.SpecTP needs to be filled as well.

This sounds like a familiar question… This sounds like a familiar question… should we hypothesize that the should we hypothesize that the subject in these sentences is PRO?subject in these sentences is PRO?

Little Little propro

There is one important difference There is one important difference between the Italian null subject and between the Italian null subject and PRO, namely the null subject in Italian PRO, namely the null subject in Italian appears in a position that gets Case.appears in a position that gets Case. Io parlo.Io parlo.

I speak-1s ‘I speak’I speak-1s ‘I speak’ Since PRO Since PRO cannotcannot appear in a Case- appear in a Case-

marked position, we have to take this marked position, we have to take this to be something similar but different: to be something similar but different: Little Little propro..

Little Little propro

Little Little propro is really just a regular pronoun, only is really just a regular pronoun, only null. It doesn’t have the fancy control null. It doesn’t have the fancy control properties exhibited by PRO, it appears in properties exhibited by PRO, it appears in Case-marked positions.Case-marked positions.

Languages seem to be divided into those Languages seem to be divided into those which have little which have little propro and those which don’t, and those which don’t, often correlating with the amount of often correlating with the amount of agreement on the verb (rich agreement makes agreement on the verb (rich agreement makes it more likely that a language will have it more likely that a language will have propro). ). Languages with Languages with propro are often called “ are often called “propro-drop-drop languageslanguages” or “null subject languages”.” or “null subject languages”.

Features and checkingFeatures and checking

An elaboration…An elaboration…

We assume that we have a lexicon We assume that we have a lexicon full of items (“words”) that get full of items (“words”) that get inserted into terminal nodes of the inserted into terminal nodes of the tree. These items can be considered tree. These items can be considered to be little collections of properties, to be little collections of properties, or “features.”or “features.”


What do we know about What do we know about sheshe in English? in English? It’s a DIt’s a D It’s pronounced “she”It’s pronounced “she” It has nominative CaseIt has nominative Case It is 3rd personIt is 3rd person It is singularIt is singular It is feminineIt is feminine

These things we know are all properties, These things we know are all properties, or or featuresfeatures, of the lexical item , of the lexical item sheshe. (These . (These are the are the grammatically relevant grammatically relevant properties properties anyway…)anyway…)

Features of TFeatures of T

Now, let’s think about T.Now, let’s think about T. English T has features like [past] or English T has features like [past] or

[pres], and sometimes we’ve written [pres], and sometimes we’ve written [past] as [past] as -ed -ed to indicate its to indicate its pronunciation.pronunciation.

But what determines the (regular) But what determines the (regular) pronunciation of the affix in T?pronunciation of the affix in T? I walk. You walk. He walks. They walk.I walk. You walk. He walks. They walk. I walked. You walked. He walked. They I walked. You walked. He walked. They

walked.walked.

Features of TFeatures of T It seems that It seems that bothboth the tense feature and the tense feature and

the person specification of the subject the person specification of the subject affects how T is pronounced.affects how T is pronounced.

Why?Why? The modern approach to this phenomenon The modern approach to this phenomenon

(which often goes by the name of (which often goes by the name of Spec-Spec-Head AgreementHead Agreement) is to suppose that there ) is to suppose that there are features both on T and on the subject are features both on T and on the subject (for person, number) and that when they (for person, number) and that when they are in a Spec-Head relationship, the are in a Spec-Head relationship, the features are features are close to each otherclose to each other..

Spec-head agreementSpec-head agreement

The reason it is important for the features The reason it is important for the features to be close to each other is that the to be close to each other is that the syntax needs to be able to syntax needs to be able to checkcheck to make to make sure the features match. Spec-head sure the features match. Spec-head counts as “close”.counts as “close”. *I walks. *He walk*I walks. *He walk..

If the subject has different person If the subject has different person features from the tense/agreement suffix features from the tense/agreement suffix in T, then the sentence is ungrammatical.in T, then the sentence is ungrammatical.

SpecTPSpecTP Another thing SpecTP is famous for its ability to Another thing SpecTP is famous for its ability to

host nominative case-marked subjects.host nominative case-marked subjects. This is implemented in the same way, by This is implemented in the same way, by

analogy to agreement.analogy to agreement. To say that finite T is a nominative case assigner To say that finite T is a nominative case assigner

is to say that it has a feature [(Assign) Nom], is to say that it has a feature [(Assign) Nom], and DPs like and DPs like II and and hehe have a feature [Nom]. have a feature [Nom].

A subject “getting Case” in SpecTP is then not A subject “getting Case” in SpecTP is then not exactly exactly gettinggetting Case so much as it is Case so much as it is checkingchecking to to be sure that the Case it has is the right one.be sure that the Case it has is the right one.

Case Case hashas to be checked to be checked (guilty until proven (guilty until proven innocent).innocent).

SpecTPSpecTP

This is really just another way to state the This is really just another way to state the Case Filter (“DPs need (to check their) Case Filter (“DPs need (to check their) Case”) but it’s now in terms of a more Case”) but it’s now in terms of a more specific understanding of what it means to specific understanding of what it means to “assign Case”.“assign Case”.

This also means that the “government This also means that the “government radius” is a way to characterize the radius” is a way to characterize the positions which are positions which are closeclose enough for enough for feature checking to occur.feature checking to occur.


There is a distinction between features There is a distinction between features that need to be checked and features that need to be checked and features that do not.that do not. Case features like [nominative] need to be Case features like [nominative] need to be

checked.checked. These are the kinds of features These are the kinds of features which often motivate movement.which often motivate movement.

Category features like [D] on a determiner Category features like [D] on a determiner are fine as they areare fine as they are, they don’t need to be , they don’t need to be checked against anything else.checked against anything else.

Features and checkingFeatures and checking Another point worth observing about Another point worth observing about

checking features like [Nom] on a DP is checking features like [Nom] on a DP is that that it only happens onceit only happens once. Once you’ve . Once you’ve checked to be sure that the Case is right, checked to be sure that the Case is right, you’re fine—in fact, you you’re fine—in fact, you can’tcan’t check it a check it a second time.second time.

For this reason, sometimes people think For this reason, sometimes people think of the features as being removed when of the features as being removed when checked (like on a checklist). Either way, checked (like on a checklist). Either way, you only check them once.you only check them once.

Case checking seems Case checking seems symmetricalsymmetrical

Recall that we said T has a feature [Assign Recall that we said T has a feature [Assign Nom], and this is checked against the [Nom] Nom], and this is checked against the [Nom] feature of a subject like feature of a subject like wewe in order to in order to validate the Case on the subject.validate the Case on the subject.

There is actually reason to think that There is actually reason to think that bothboth the [Assign Nom] feature on T and the the [Assign Nom] feature on T and the [Nom] feature on the DP need to be checked[Nom] feature on the DP need to be checked—and that each can happen only once.—and that each can happen only once. Finite T Finite T needsneeds to check Nom on a DP. to check Nom on a DP. DPs DPs needneed to check Case. to check Case.

Moving Moving toto Case positions Case positions

Consider:Consider: It is likely that we will leave.It is likely that we will leave. *We*Weii are likely that are likely that ttii will leave. will leave.

What’s the problem with the second What’s the problem with the second one?one?


*We*Weii are likely that are likely that ttii will leave. will leave.

WeWe moved up to the finite SpecTP, and moved up to the finite SpecTP, and checked off its [Nom] feature with [Assign checked off its [Nom] feature with [Assign Nom] feature of T. Both are now Nom] feature of T. Both are now inactivated.inactivated.

But then But then wewe is moved up to the matrix is moved up to the matrix SpecTP. Yet SpecTP. Yet wewe no longer has an active no longer has an active [Nom] feature (it’s been checked already), [Nom] feature (it’s been checked already), so the matrix T can’t get rid of its [Assign so the matrix T can’t get rid of its [Assign Nom] feature.Nom] feature.


It It isis possible to move solely for the EPP if possible to move solely for the EPP if there is no Case to check (i.e. in a nonfinite there is no Case to check (i.e. in a nonfinite TP).TP). [The sandwich][The sandwich]ii is likely is likely ttii to have been eaten to have been eaten ttii..

So, we So, we couldcould have moved have moved wewe to the matrix to the matrix SpecTP—something SpecTP—something elseelse went wrong. went wrong. *We*Weii are likely that are likely that ttii will leave. will leave.

And what went wrong is that this leaves the And what went wrong is that this leaves the matrix SpecTP without a DP to check its matrix SpecTP without a DP to check its [Assign Nom] feature against.[Assign Nom] feature against.

So where are we?So where are we?

The generalizations here are:The generalizations here are: The Case Filter:The Case Filter:

DPs are inserted into the structure with a Case DPs are inserted into the structure with a Case feature which must be checked.feature which must be checked.

Case assigners are inserted into the structure with Case assigners are inserted into the structure with a Case-assignment feature which must be checked.a Case-assignment feature which must be checked.

A DP cannot move from a Case-checking A DP cannot move from a Case-checking position to another Case-checking position.position to another Case-checking position. Case checking only happens once (de-activating Case checking only happens once (de-activating

the Case feature on both the Case-assigner and the the Case feature on both the Case-assigner and the DP)DP)

WhWh-questions-questions

This was kind of complicated, but it This was kind of complicated, but it was worth going through in order to was worth going through in order to set up set up whwh-questions for next time.-questions for next time.

To get us started:To get us started: WhWh-questions-questions are information-seeking are information-seeking

(not yes/no) questions, in English (not yes/no) questions, in English involving one or more of the “involving one or more of the “whwh--wordswords” (” (whowho, , wherewhere, , whatwhat, , whenwhen, , whywhy, , howhow, …), …)


In a In a whwh-question, we find that we do the -question, we find that we do the same inversion that happens with yes-no same inversion that happens with yes-no questions… (moving T to C).questions… (moving T to C). WillWillii Bill Bill ttii eat lunch? eat lunch?

……plusplus, we move the , we move the whwh-word into SpecCP:-word into SpecCP: WhatWhatjj will willii Bill Bill ttii eat eat ttjj ? ?

This movement of This movement of whwh-words is similar, but -words is similar, but different, from the DP movement we’ve seen different, from the DP movement we’ve seen so far with passives and raising verbs.so far with passives and raising verbs.


With yes-no questions, we posited a With yes-no questions, we posited a [+Q] [+Q] C at the head of CP, which caused C at the head of CP, which caused the movement of T to C.the movement of T to C.

For For whwh-questions, we can think of a -questions, we can think of a different kind of C, a different kind of C, a [+Q, +WH][+Q, +WH] C, C, which prompts which prompts bothboth the movement of T the movement of T to C to C andand the movement of the the movement of the whwh-word -word into SpecCP.into SpecCP. (So, yes-no questions would have a (So, yes-no questions would have a [+Q, –[+Q, –

WH]WH] C) C)


What causes the movement of the What causes the movement of the whwh--word to SpecCP is considered to also be word to SpecCP is considered to also be a case of feature checking.a case of feature checking.

In this case, the In this case, the C has a [+WH] feature C has a [+WH] feature to checkto check, and the , and the whwh-words have [+WH] -words have [+WH] features that can be checked against itfeatures that can be checked against it..

So, the So, the whwh-word is brought up into -word is brought up into SpecCP to bring the features close SpecCP to bring the features close enough for checking, and then presto! enough for checking, and then presto! everybody wins.everybody wins.


Interestingly, looking at English, Interestingly, looking at English, [+WH] feature checking appears not [+WH] feature checking appears not as symmetrical as Case checking. In as symmetrical as Case checking. In particular, moving just particular, moving just oneone whwh-word -word to SpecCP seems to be sufficient.to SpecCP seems to be sufficient. Who gave what to whom?Who gave what to whom?

That is, all of the other That is, all of the other whwh-words -words can remain, seemingly “unchecked”.can remain, seemingly “unchecked”.


[+WH] C [+WH] C mustmust check its [+WH] feature. check its [+WH] feature. WhWh-words -words maymay check their [+WH] feature. check their [+WH] feature.

In a sense, In a sense, English English whwh-movement provides -movement provides a pretty good motivation for a “feature” a pretty good motivation for a “feature” view of these phenomenaview of these phenomena. It appears that . It appears that [+WH] C has a “need” which a [+WH] C has a “need” which a whwh-feature -feature can satisfy, and once satisfied (even with can satisfy, and once satisfied (even with other other whwh-words around), everything is fine.-words around), everything is fine.

Documents

Week 8. Control and PRO CAS LX 522 Syntax I. Back to business… Mary is likely to leave. Mary is likely to leave. Mary starts in SpecVP, gets a -role