| 书目名称 | Predicates and Their Subjects | | 编辑 | Susan Rothstein | | 视频video | | | 丛书名称 | Studies in Linguistics and Philosophy | | 图书封面 |  | | 描述 | .Predicates and their Subjects. is an in-depth study of the syntax-semantics interface focusing on the structure of the subject-predicate relation. Starting from where the author‘s 1983 dissertation left off, the book argues that there is syntactic constraint that clauses (small and tensed) are constructed out of a one-place unsaturated expression, the predicate, which must be applied to a syntactic argument, its subject. The author shows that this predication relation cannot be reduced to a thematic relation or a projection of argument structure, but must be a purely syntactic constraint. Chapters in the book show how the syntactic predication relation is semantically interpreted, and how the predication relation explains constraints on DP-raising and on the distribution of pleonastics in English. The second half of the book extends the theory of predication to cover copular constructions; it includes an account of the structure of small clauses in Hebrew, of the use of `be‘ in predicative and identity sentences in English, and concludes with a study of the meaning of the verb `be‘. | | 出版日期 | Book 2004 | | 关键词 | Syntax; predicative; semantic; semantics; subject; syntactic | | 版次 | 1 | | doi | https://doi.org/10.1007/978-94-010-0690-3 | | isbn_softcover | 978-1-4020-2058-2 | | isbn_ebook | 978-94-010-0690-3Series ISSN 0924-4662 Series E-ISSN 2215-034X | | issn_series | 0924-4662 | | copyright | Kluwer Academic Publishers 2004 |
| 1 |
Why ‘Subject’ is a Grammatical Concept |
Susan Rothstein |
|
Abstract
The traditional approach to subjects, predicates and predication is essentially pragmatic, and defines the subject in terms of a semantic/pragmatic predication relation explained in terms of ‘aboutness’. As for the syntax, it proposes that the syntactic subject is the syntactic expression of the entity that the sentence is about, and a sentence is said to ‘say something about its subject’. This subject-predicate relation has often been taken for granted as being in some sense ‘given’, and the subject and predicate are frequently assumed to be the basic semantic elements which a proposition is composed out of. Chapter 1 of this book reviews some of the issues that these assumptions raise, and aims to clear the ground for the discussion of grammatically or structurally grounded theories of predication which constitutes the bulk of this book. I am not, at this stage, interested in the question of which structural definitions of predicate, subject, and predication are to be preferred; rather I want to identify some of the difficulties which arise when we try to define these concepts pragmatically, and to show why we are led to search for purely structural definitions.
|
| 2 |
The Grammatical Theory of Predication |
Susan Rothstein |
|
Abstract
In the previous Chapter, we examined the intuition that a proposition is an instantiation of a predication relation, and established the difficulties which arise in trying to define ‘subject’ and ‘predication’ in terms of semantic/ pragmatic notions of ‘aboutness’, ‘topic’, ‘pivotal argument’ and so on. We saw also that there is good grammatical evidence that a sentence has a binary structure, and that it is divided into two constituents of different kinds which are traditionally known as argument and predicate. The burden of the evidence thus indicates that the distinguished nature of the subject position is a grammatical fact, which follows not from the inherent structure of assertions, but from the compositional structure of sentences.
|
| 3 |
The Syntactic Properties of Subjects |
Susan Rothstein |
|
Abstract
Pleonastics are syntactic arguments that do not enter into thematic relations of any kind. It has frequently been proposed in the government-binding framework that there is a generalisation that pleonastics occur only in subject position and not in object position. Postal and Pullum (1987), in a careful presentation of potential counterexamples, suggest that this generalisation is not true. However, the theory of grammatical predication that I am presenting in this Chapter predicts that pleonastics should appear only in subject positions, and explains why they cannot appear anywhere else. The potential counterexamples presented by Postal and Pullum divide into two groups; some of them are genuine pleonastics which appear in what turn out to be subjects of small clause complements of ECM verbs (small clauses like those discussed in Chapter 2:2.3.) The others are examples of . in true object position which, despite appearances, turn out not to be a pleonastic . but a theta-marked object of the main verb. The first set, the true pleonastics, are the topic of this section; the second set, which includes sentences like those in (1), are the topic of the second half of Chapter 7 (see also Rothstein 1995a).
|
| 4 |
Predication as a Thematic Relation |
Susan Rothstein |
|
Abstract
In the previous Chapters I discussed several kinds of structures which are evidence for a non-lexical/thematic analysis of subjecthood, since they are cases in which the subject of a primary predicate is not assigned a thematic argument by the head of the predicate. These included constructions where the subject is a pleonastic, constructions where the DP in subject position is moved there from a lower position, and clauses where the main predicate simply doesn’t assign an external theta-role. The predicates in these constructions are predicates by virtue of their syntactic structure and not by virtue of their lexical structure. A ‘distinguished argument’ theory which defines predicate and predication in terms of thematic arguments has to explain how XPs which do not assign external theta-roles enter into predication relations with their subjects. The way in which lexical, non-movement theories of subjects deal with these problems is very different from the way in which a movement-based theory like Williams’ can deal with it (whether the movement is actual, in a ‘strong’ derivational theory of syntax, or ‘virtual’ in a representational type theory in which DP chains are built at S-structure in conformance with certain constraints, as in Rizzi (1986)). I will discuss non-movement accounts of subject hood in Chapter 6 and 7; here I will consider in depth the theta-based theory developed by Williams. My argument will be that it is not possible to combine a syntactic theory which includes movement and NP (or rather, DP) chains with a distinguished argument account of subjects, predicates and predication.
|
| 5 |
The Syntactic Forms of Predication |
Susan Rothstein |
|
Abstract
In Chapter 2 I argued for a syntactic predication relation which cannot be reduced to a thematic relation. I argued that this syntactic relation is a saturation relation between a predicate and an argument, its subject, and I claimed that there was a locality relation between subject and predicate, namely that subject and predicate have to c-command each other. The discussion in Chapter 2 concentrated on primary, or clausal, predication relations, and in particular on the predication relation within small clause complements of . and other ECM verbs, where the predication relation is realised in its simplest form. By this I mean that small clause predication has two basic characteristics, and the predication relation is more complicated if either of these properties is changed. Small clause predication is . predication, by which I mean that the subject and predicate c-command each other, and it is . predication, which means that the subject and predicate form a constituent. If either of these properties does not hold of a predication relation, then we need to say something more than what was said in Chapter 2.
|
| 6 |
Interpretation |
Susan Rothstein |
|
Abstract
In this Chapter, we discuss how the structures that I have been arguing for up to now are interpreted. The crucial syntactic point that I have been making is that certain syntactic constituents, namely maximal projections of lexical heads, are syntactically one-place predicates, no matter where they occur in the sentence, and no matter what their thematic structure is. I make some basic assumptions about the relation between the syntactic structure and the semantic interpretation: that syntactic predicates translate as semantic predicates, and that if a is the semantic translation of XP, then α needs to be applied to the denotation of the syntactic argument of which XP is syntactically predicated. Put together, this means that in the semantic representation, the denotations of syntactic predicates must all be of the form λx.α, independently of the structural configuration in which they appear, and independently of the thematic adicity of α. I shall propose that this is effected by means of a rule of predicate formation which operates automatically at the predicate XP node in the tree, and which prefixes the semantic interpretation of a predicate XP by a λx operator. (Note that the theory uses x as a distinguished variable by means of which the grammar keeps track which argument ends up in subject position, and I’ll talk more of this in the course of the Chapter.) I’ll assume that a verbal or adjectival expression which has assigned all its internal arguments denotes a set of events, and that the effect of predicate formation is to take a verbal/adjectival expression of type into >, whether it is a main predicate or an adjunct predicate, and independent of whether it has an external argument to assign. In this Chapter, I’ll be concerned with the interpretation of clausal and adjunct predicates which do assign an external argument, and the structures that we will be giving an interpretation for are those that we have argued for in the previous Chapters. I’ll concentrate on VP and AP predicates, and I’ll leave predicative NPs and AP modifiers to Chapter 9. I’ll turn to an extensive discussion of pleonastic subjects in the next Chapter.
|
| 7 |
The Semantics of Pleonastics |
Susan Rothstein |
|
Abstract
As has been clear from Chapter 1, pleonastics are a major issue for any theory of subjects, predication or clausal structure. The obligatory presence of nonthematic DPs in subject position with no apparent role in the interpretation of the clause is a challenge for any theory which wants to try and explain what the nature of clausal structure is. And for a compositional theory of interpretation, which assumes that syntactic operations are correlated with semantic operations, the question is what semantic operation could be paired with inserting an obligatory, but meaningless DP in subject position. Depending on the kind of theory, the problem takes a slightly different shape, and suggests a different kind of solution. Recent theories which seriously discuss the issue, in particular Sag (1982), Dowty (1985), Gazdar, Klein, Pullum and Sag (1985) and Chierchia (1989) assume that pleonastic . denotes a dummy or null element ⊥, which Dowty (1985), following Lauri Kartunnen, calls the ‘ugly object’; the semantic correlate of inserting a pleonastic in the subject position is functional application to ⊥. The question is how to motivate such an operation. I am only going to discuss the . pleonastic here: as McCloskey (1991) shows, . has very different properties from ., and needs to be discussed separately.
|
| 8 |
Predication Structures In Modern Hebrew Identity Constructions |
Susan Rothstein |
|
Abstract
This Chapter presents an analysis of simple Hebrew identity sentences which I offer as a case study in the syntax of predication. Briefly, the facts are that Modern Hebrew (henceforth just ‘Hebrew’) allows assertions to be made with matrix small clauses; consisting of just a subject and a non-verbal predicate. It also allows ‘inflected small clauses’, where a pronominal copula, either personal (PronH) or impersonal (PronZ), realises Infl, and takes a non-verbal maximal projection as a complement. As far as the personal pronominal copula is concerned, it looks as if it is optional, since both inflected and non-inflected forms of the small clauses occur. However, in identity sentences, only the inflected form is possible. The point of this Chapter is to argue that the absence of uninflected identity small clauses follows from the requirement that every clause is an instance of a syntactic predication structure. I shall argue that where the post-copula XP is inherently a predicate, a predicate structure is formed whether or not Inflection is present, but that when the post-copula element is an argument and referential, Inflection is necessary to create a syntactic predication structure; furthermore, Infl is necessary to trigger the type shift which allows the identity sentence to be interpreted as a semantic predication structure.
|
| 9 |
Copular Constructions in English |
Susan Rothstein |
|
Abstract
This Chapter looks at the structure of copular sentences in English. There are two basic issues which concern us. The first, which follows directly on from the discussion of Hebrew copular constructions in the previous Chapter, is the fact that in English too, small uninflected clauses are possible, but not if the small clause is an expression of identity. In the previous Chapter we saw that in Hebrew a semantically null Inflectional element is necessary in identity Statements to create a syntactic predication structure, but not in small clauses, where the predication structure is created by the relation between the inherently predicative XP and the subject. In English, the same phenomenon occurs, but there are two major differences. First, we don’t get the contrast in matrix sentences, since English doesn’t allow uninflected matrix clauses, except in the ‘echoic’ questions and exclamations discussed in Akmajian (1984), and noted in Chapter 2. Second, English requires Inflection to take a VP complement, and thus the contrast is not between small clauses and inflected small clauses, but between small clauses and IPs with infinitivally marked VPs, as illustrated in(l):
|
| 10 |
The Meaning of ‘be’ |
Susan Rothstein |
|
Abstract
In the analysis of copula constructions that I have presented in the previous Chapter, copula . adds nothing to the meaning of the sentence in which it occurs. It denotes the identity function on predicates, λ E.E, and its semantic ‘contribution’ to the VP it heads is limited to the fact that if its complement is an argument DP, it triggers a type-shifting operation which raises the DP from type d to type . Instead its contribution is structural. The copula is a V which heads a VP syntactic predicate, which is a constituent at which predicate formation occurs. The complement of . is interpreted as an expression with a free x variable, denoting a set of events. As we saw in the derivations in examples (15) and (27) in the previous Chapter, predicate formation at the VP level introduces a λ,-operator which binds the x variable (in both predicative and identity sentences), thus allowing the constituent to be predicated of the subject. The Hebrew inflectional copula, discussed in Chapter 8, also need make no independent semantic contribution, and similarly denotes the identity function on expressions, but at type >. It also triggers a type raising Operation from d → < d, < e,t > > in its complement in identity (equative) sentences.
|
|
|
发表于 2025-3-21 18:09:10
