sign in sign up
<<
Home , Syntax, Theta criterion

Previously, in LX522… CAS LX 522 n We were talking about q-roles, the “ I slots” that predicates (e.g., ) have. These are the “roles” that the participant play in the event. n As part of their lexical entry, verbs have a list of q-roles that they assign, a list of required Week 6a. Case and checking (with participants. a little more q-Theory) n Kick: Agent, Theme n Jog: Agent n Introduce: Agent, Theme, Goal

Common thematic relations The q-criterion

n Agent: initiator or doer in the event n The q-criterion: n Theme: affected by the event, or undergoes the n every q-role in the q-grid is assigned to exactly one action argument. n every argument is assigned exactly one -role. n Bill kicked the ball. q n Experiencer: feel or perceive the event n The second half protects us against superfluous arguments. But it’s hard to evaluate this if we n Bill likes pizza. don’t know what an argument is. n Goal: n It’s hard to say, actually. There are some further n Bill gave the book to Mary. (Recipient) concepts that we should have before we can even n : a statement, can be true/false. start to state this accurately. For now, let’s just n Bill said that he likes pizza. suppose that DPs and CPs are necessarily arguments, and PPs usually aren’t.

Theta Grids Theta Grids n We can formalize the about q-roles in n Johni gave [the book]j [to Mary]k. the lexical entry for a by using a theta grid, like so: give Source/Agent Theme Goal give Source/Agent Theme Goal i j k i j k

n The columns each represent a q-role, the indices in The first q-role is the lower row will serve as our connection to the assigned to the . The other q-role are actual arguments; e.g. It is the external q-role. internal q-roles. It is often designated by n Johni gave [the book]j [to Mary]k. underlining it.

1 Theta Grids The in action n The q-roles in the theta grid are obligatory. n An example: push. (Optional things like on the hill are not in the q-grid). push Agent Theme i j give Source/Agent Theme Goal i j k n Billi pushed the shopping cartj. n Fine, push assigns two q-roles, one (the external q-role) is n Adjuncts are related to the verb via thematic assigned to Bill, the other (the internal q-role) is assigned relations (e.g., instrument, location, etc.), but an to the shopping cart. There are two arguments here, each does not get a q-role. They are optional. gets a q-role. n *Billi pushed. (j?) n *Billi pushed the shopping cartj the corner?.

The Theta Criterion in action The EPP n An example: cough. n With the Theta Criterion in our toolbox, let’s take a look at a special kind of (which will cough Agent turn out to tell us something important about i syntax). n It rained. n i Bill coughed. n It snowed. n Fine, cough assigns one q-role (the external q-role), to Bill. n How many q-roles does rain assign? There are one arguments here, and it gets a q-role. n If we think about it, it doesn’t really mean n *Billi coughed the shopping cart?. anything at all. It is not a participant in the event; it really can’t be getting a q-role. (cf. also Spanish).

The EPP The EPP

n So, the theta grid for rain really looks like n Given the q-Criterion and the fact that rain this: doesn’t have any q-roles to assign, what’s it doing there? And why doesn’t it violate the q-Criterion? n As to the first , the conclusion that syntacticians have come to is that the it is there due to a separate constraint, which goes by the EPP.

2 The EPP Features and Case n The EPP IP must have a . n Recall that in English have several Case forms, indicating their n More informally, all have subjects. grammatical function (subject, non- subject): n Because rain has no arguments (no q-roles), a special, n Nominative (subject): contentless (it) has to be inserted to in He, she, they, … order to have a grammatical sentence. This kind of n Accusative/objective (non-subject): “empty it” is called an expletive or a pleonastic Him, her, them, … pronoun. It is not an argument (in this use). n But what’s wrong with *Him left? n We stipulate that it is not subject to the q-criterion.

Features and Case Features and Case

n What stops us from picking him, a [+Past] n The intuition is that subjects (things in the I, and leave, and Merging in order to specifier of IP, at least for I like [+Past]) produce him left? must have nominative Case. I needs [+Nom] in its specifier. IP * IP IP * IP DP I¢ DP I¢ DP I¢ DP I¢ He Him He Him I VP I VP [+Nom] I VP [+Acc] I VP -ed leave -ed leave -ed leave -ed leave

Specifier features Features and Case

n To encode this requirement, we posit a n When we the DP with [-ed leave] to form second type of on the lexical items SpecIP, the features of the specifier are “checked” of category I: the specifier features. against the (specifier) features of the . If they match, they are removed from the to-do list, n Specifier-features are requirements; they they are “checked off”. are features that must be found in the specifier. IP * IP n If I has a [+Nom] specifier feature, SpecIP DP I¢ DP I¢ (the specifier of IP) must have a [+Nom] He Him [+Nom] I VP [+Acc] I VP feature. -ed leave -ed leave [+Past, …] [+Past, …] spec: [+Nom] spec: [+Nom]

3 Features and Case Subject

n If we finish with unchecked features of this kind, n The same kind of thing rules out *I has left. the derivation crashes, the sentence is no good. n Here, the problem is that the subject is [1sg] n These features are uninterpretable and so if they (a.k.a. [+1,-2,-Pl]), has is [3sg] (a.k.a. [-1,-2,-Pl]) are still there when we try to compute the meaning of the structure, we can’t. IP * IP IP * IP DP I¢ DP I¢ DP I¢ DP I¢ He Him He I [+Nom] I VP [+Acc] I VP [3sg] I VP [1sg] I VP -ed leave -ed leave has gone has gone [+Past, …] [+Past, …] [-Past, …] [-Past, …] spec: [+Nom] spec: [+Nom] spec: [3sg] spec: [3sg]

Complement features Subject agreement () n The [1sg], [3sg] features are fundamental to the meaning n Heads also can impose similar requirements on the kind of the pronouns, so they are not uninterpretable. But of that they have as a . For example, specifier features on has are still uninterpretable and has requires the perfective (-en) form of the must be checked off. verb. (I’ll use [+n] as a shorthand for [+Participle, n Only uninterpretable features are checked off (deleted +Perfect]) from the to-do list) when satisfied. n We can encode these as complement features. IP * IP IP * IP DP I¢ DP I¢ DP I¢ DP I¢ He I He He [3sg] I VP [1sg] I VP [3sg] I VP [3sg] I VP has gone has gone has gone has go [-Past, …] [-Past, …] comp: [+n] [+n] comp: [+n] [-n] spec: [3sg] spec: [3sg] spec: [3sg] spec: [3sg]

Features Head features

n Lexical items have three kinds of features. n Interpretable: Fundamental to the meaning, n Head features: Primary features… crucial to interpreting the meaning of the structure n Specifier features: Uninterpretable features n [3sg] on pronouns, [D] on that must be checked against the features of the specifier (at last projecting Merge). n Uninterpretable: Not part of the meaning, but nevertheless part of the . Must be n Complement features: Uninterpretable eliminated (checked off) by the end of the features that must be checked against the derivation. features of the complement (at first projecting n [+Nom] on determiners, [+Nom] on I, [3sg] on Merge). auxiliaries

4 Features and Case Case and the DP n Fantastic. Now why is this grammatical n How about objective (accusative) Case? (assuming that will and meet are the same as in I How would we encode the fact that You will meet you)? will meet her is fine, but *You will meet she is ungrammatical? IP IP DP I DP I¢ ¢ You I VP D NP I VP will Ø Pat will V DP V DP meet meet her D NP the students

Back to q-theory? Details… n We started off looking at argument n Trying to encode the Agent q-role as a requirements that verbs impose, in terms of specifier feature brings up a complication: their q-grids, the q-roles that they need to The Agent is in SpecIP, not SpecVP, so assign. how would a specifier feature of V be n Can we think of these in terms of checked? complement and specifier features? n (We don’t want Agent to be a specifier feature of I, because not all subjects are agents—it n Pat kicked the ball. depends on the specific verb). n Pat slept.

Percolation? Clarification/preview n Radford introduces a concept called percolation n Although they seem to be doing the same thing to handle problems like this. If V has a specifier so far, let me stress: feature that is not checked off before its last projection (VP), the requirement is “passed up the tree” to the next head (I), and becomes a n requirement of I. Case and q-roles are not n Using this, we could say that if V has an Agent the same thing. specifier feature, it can be passed up to I and satisfied by having an Agent in SpecIP. n Case marks structural position in the tree. n For now (these two weeks), we will assume this is what happens. After the midterm, we’ll see n q-roles are assigned to participants in an event. that we don’t really need percolation.

5 Case ≠ q-role T n After the midterm, we will look at how to T T handle these, but notice: n Active: I pushed him. T n Passive: He was pushed. n Masc3sg is the Theme (q-role) in both cases, but the T T (Objective/Accusative Case) in the active form, and the subject (Nominative Case) in the passive T T form. n Also: T n The door opened. (Door=Nom, Theme) T n I opened the door. (Door=Acc, Theme)

6