Syntax - Relative Clauses

Relative clauses Syntax I place where we see wh-movement, • Another besides in explicit questions (either in the main clause or embedded) is in relative clauses. •The book which I read •The woman who(m) I met consist of a head noun (book, woman) • These and then what appears to be a wh-question that further specifies the referent of the head noun. Relative clauses • Relative clauses serve to modify the head noun. • Kind of like adjectives, or PP modifiers. • The unhappy students. • The students from Vancouver. • The students who solved the problem. • So where would you put them? Differences between questions and relative clauses • The “question” inside a relative clause has a couple of odd properties, not shared with regular main clause or embedded questions. • • • • • Relative clauses structure of a • The relative clause: CP [clause-type:Rel, uwh*] is • Aadjoined to the NP, like an adjective, or a PP modifier. meaning is essentially “the man with the property of • The being the answer to ‘Who did I meet?’ ” see in a moment that C [Rel] can be pronounced as • We’ll either Ø or as that. • • *The problem what I solved. Which/that/Ø In addition to being able to say • We can also say • • The problem which I solved. The book which Mary read The book that Mary read • and The book Mary read The problem which I will solve. The problem I solved. The problem that I solved. • And they all mean the same thing. So we expect that they would all have basically the same structure (they all have a question adjoined in the nP)—so where is the wh-word in the last two? Op Op • The secret to these last two kinds of relative clauses is Op, the silent wh-word. • • It is also possible to pronounce that with Op, giving us: That is, the book which Mary read and the book Mary read are really exactly the same except that in one case you pronounce the wh-word, and in the other, you don’t. • the book [CP Opi that [TP Mary read ti ]] • • Why can’t we pronounce that with which? • • the book [CP whichi Mary read ti ] the book [CP Opi Mary read ti ] Doubly-Filled COMP filter *the book [CP whichi that [TP Mary read ti ]] Op • • The Doubly-Filled COMP filter is the traditional “explanation”: of Op? Is there really wh-movement of • Skeptical Op, a silent wh-phrase? Doubly-Filled COMP filter: *[CP wh-word if/that/for…] • I read the book [ • You can’t pronounce both a wh-word and C at the same time. Thus: • *I read the book [ • • • • ! the book [CP Opi [TP Mary read ti ]] ! CP whichi [TP Mary said [CP that [TP Bill bought ti ]]]]. CP whichi [IP Mary wonders [CP who [TP bought ti ]]]]. • I read the book [ the book [CP Opi that [TP Mary read ti ]] ! the book [CP whichi [TP Mary read ti ]] CP Opi (that) [TP Mary said [CP that [TP Bill bought ti ]]]]. • *I read the book [ ! *the book [CP whichi that [TP Mary read ti ]] CP Opi (that) [TP Mary wonders [CP who [TP bought ti ]]]]. Op we have a silent wh-word, why can’t we ask • Ifquestions with it? Where did Mary buy this book t ? • The place [Op Mary bought this book t ] • When did Mary buy this book t ? • The time [Op Mary bought this book t ] • Why did Mary buy this book t ? • The reason [Op Mary bought this book t ] • How did Mary buy this book t ? • The way [Op Mary bought this book t ] • *Op did Mary buy this book t ? • • See why? i i i i i i i i i i i i i i i i i i Op • Recoverability condition:The content of a null category must be recoverable. • the place [Op (that) Mary bought that book t ] • the day [Op (that) Mary bought that book t ] • the reason [Op (that) Mary bought that book t ] • the way [Op (that) Mary bought that book t ] i i i i i i i i • In each case, we can tell what the wh-phrase is by looking at the head noun. Weak Crossover QR read every book. • Sue For every book x, Sue read x. • [every book]i [TP Sue read ti ]. finishing up with one more argument • We’re for Quantifier Raising. The idea being that, in order to be interpreted, quantifier phrases must move to a position outside the clause (adjoining to TP). This is almost never audible movement, but there is evidence that it happens anyway, abstractly (“covertly”). • *Who does his roommate like t ? • Who t likes his roommate? i i i i i first argument was based on weak crossover. • The Audible (“overt”) movement of wh-phrases seems to be constrained by WCO. If this is simply a property of movement, then finding WCO effects with quantifiers constitutes an argument that quantifiers also move. Crossover (WCO): A coindexed pronoun • Weak cannot intervene between an operator and its variable. WCO • • [Every girl]i [TP ti likes heri roommate]. • • [Every girl]i [TP heri roommate likes ti ]. • WCO is about moving over a bound variable—so if WCO rules out this meaning, there must have been movement. There must have been QR. For every girl x, x likes x’s roommate. i ACD • Here’s another reason to believe in QR, antecedent contained deletion. This one’s kind of complicated, so hang on tight. • First, we need to talk about VP ellipsis. • • Mary bought a record, and Bill did too. For every girl x, x’s roommate likes x. [TP Mary -ed [vP buy a record]] and [TP Bill -ed [vP buy a record]] too. VP ellipsis VP ellipsis • Mary bought a record and Bill bought a tape. will consider the process of VP ellipsis to be • We one of deletion under identity. ellipsis is allowed when a preceding VP is • VP identical. • Underlyingly: -ed [ Mary sleep] and -ed [ ! Mary bought a record and Bill did too. interpret this, you need to use the content • To of the preceding VP. bought a record and Bill did (buy a • Mary record) too. vP vP Bill sleep] too. deletion: • Before Mary -ed [ t sleep] and Bill -ed [ t sleep] too • Pronunciation: Mary -ed [ t sleep] and Bill -ed [ t sleep] too vP vP Mary slept and Bill did too vP vP VP ellipsis VP ellipsis as long as two VPs in sequence look • So, identical (where traces of movement look identical to one another—they sound the same), we are allowed to pronounce the second one very quietly. an extreme case of • Like Mary bought a record and ! ! Bill bought a record too. • • • Note that identity is actually fairly abstract. • Before deletion: John -ed [vP t sleep] and Mary will [vP t sleep] too John slept and Mary will too. John slept and Mary will • i • Bill i likes [the record that Mary bought]. likes the record that Mary bought and Sue • Bill does too. likes the record that Mary bought and Sue • Bill does (like the record that Mary bought) too. QR and ACD too. The inflectional features of v don’t matter for identity; the verb doesn’t inherently have a tense suffix. VP ellipsis with relative clauses • Now, consider a DP with a relative clause: • the record [Op that Mary bought t ]. sleep ACD • • • • • Bill likes every book Mary does. • The deleted VP is contained in the antecedent VP (antecedent-contained deletion) Bill [vP likes every book Opi Mary [vP likes ti ]]. vP: likes [every book Op Mary likes t ] vP: likes t Those aren’t the same. VP ellipsis shouldn’t work, but yet it does. Where do quantifiers go? • But now let’s consider what QR would do. • Every book that Mary likes is a quantifier. • Quantifiers have to move up past the subject by LF. • Bill likes every book Mary does. (before covert movement): • Pronunciation Bill [ likes [every book Op Mary [ likes t ]]]. • Every student left. • [Every student] [ t left ] • We need a variable in subject position, so • LF: [every book Op Mary [ • Believe me that it is also moving j vP j vP vP j likes tj ]]i Bill [vP likes ti ]. • But now the VPs are identical. So QR allows us to explain ACD in a natural way. i TP i QR must be moving the quantifier out of TP, to somewhere higher then TP. somewhere lower than CP. Adjunction to TP Adjunction to TP • In order to accommodate this, we need to formulate a new position to which quantifiers move. One difference between QR (adjunction to TP) and movement to SpecTP is in the motivations. • Moving to SpecTP or moving to SpecCP is motivated by some need of T (EPP: T needs a DP in its specifier) or C ([Q] C needs a [+WH] in its specifier). TP QP TP TP subj • This position is going to be T! T vP adjoined to TP. • QP TP subj Moving a quantifier (QR) is required because the quantifier needs to get out of the TP (for interpretation). TP itself has no need for quantifiers. T! T vP What have we done? What’s left to do? Adjunction to TP we could say that moving to Spec is • So, something that happens if the moving thing is pulled (T is pulling up a subject to satisfy its own needs, not the needs of the moving subject) or pushed (quantifiers move to satisfy their own needs, not the needs of the T). • • People have a very complex knowledge of their native language. TP QP subj • An XP that is pulled up goes into Spec. • An XP that is pushed up adjoins. • They can distinguish combinations of words TP T! T What do they know? vP on the basis of whether they are or are not part of their language. • What do they know? • How do they know it? Formalization • We tried to formulate a model of how English speakers determine what is English. • Rock the put me is not English. • Why? • Well, the comes before nouns like rock. • Subjects take a special (nominative) form. • Verbs like put require a putter, a puttee, and a place. • We know the words, the words have certain properties (some are verbs, some are ditransitive, etc.). • Not all possible properties seem to matter for everything. To pronounce put we need know that it starts with /p/, but that kind of property doesn’t seem to matter for word order. To understand integer, we need to know about negative numbers, but this also doesn’t seem to matter for syntax. Formalization • We observed what properties do seem to matter, and named them as features. • When features combine properly, the result is English. • Verbs need all of their arguments, subjects and verbs agree. • So, we formalize “combine properly” as satisfying each lexical item’s requirements. We check to be sure the features match. Observation Formalization • Because things seem to “move around” (they are in some sense needed in two places in the structure, but are pronounced in only one of them), we hypothesized that some kinds of feature checking needs to happen over very short distances, requiring the addition of another copy of the thing whose feature is being checked up by the matching feature. We called these features “strong” (because they have stronger requirements). How do we know this? • From there, we explored some of the kinds of • We also confronted the fact that some of • Words like himself (anaphors) seem to need to • Principle A? How does a child know that an knowledge we have about what makes for a good English sentence. be below, but close to, the thing they refer to. • A class of functional elements (modals, auxiliary have and be) seem to have to show up somewhere to the left of (above) not and adverbs. Language variation • Taking as given that all human languages must be acquirable, and noting that children seem to quickly achieve a level of language knowledge that seems to go beyond what can be simply determined by the input, we look across languages to see what their task needs to entail? • Properties all languages have can be taken to follow somehow from the structure of the human brain (“principles”). Places where languages differ in constrained ways can be taken to be options “known” to the child and selected among (“parameters”). these things seem to be simply unlearnable, or at least not learnable in a way that matches what children actually seem to do. antecedent outside the binding domain doesn’t qualify? • Particularly since other languages (like Japanese) allow anaphors to take long-distance antecedent? What’s left? • Lots. There’s lots left. Lots and lots. • We’ve looked at the basic structure of clauses, how they are put together, and how they are embedded within one another. • We’ve got a foundational sampling of the kinds of things we need to build clauses, and how languages can differ in this. • But if you pick up a newspaper and read for just a short time, you’ll quickly find sentences we have not yet worked out an analysis for. What’s next? Syntax II • For any of these things (e.g., Pat and Chris left, *Who did • One somewhat general trend is that the closer we • We look at what these hypotheses would predict • Like the NP which ultimately needed to be made of Pat and leave?, I want Chris out of my sight, I made Pat cry, Chris seems angry, Pat scrubbed the floor clean/naked), there are various hypotheses we could advance. elsewhere, look for the simplest hypotheses, and look across languages for similar phenomena to see how languages can and can’t differ with respect to what’s possible, and whether those differences correlate with other differences in ways that might allow both things to follow from a single parameter. Take-home points • Human knowledge of language is extraordinarily complex. • We know a lot, but we do not yet know everything. We have models of this knowledge that are very successful at predicting what is possible, but there are still gaps. look at the behavior of areas of the tree, the more fine-grained distinctions we seem to need in order to account for the full range of things people know about the syntax of their language. DP-nP-NP, or the VP which became vP-VP, there’s some reason to think that TP and CP also need be “split” into some number of more specific phrases. • And smart people disagree about what the best analysis is, research continues.