This paper strives to characterize the relation between accent placement and discourse in terms of independent constraints operating at the interface between syntax and interpretation. The Givenness Constraint requires un-F-marked constituents to be given. Key here is our definition of givenness, which synthesizes insights from the literature on the semantics of focus with older views on information structure. AvoidF requires speakers to economize on F-marking. A third constraint requires a subset of F-markers to dominate accents.The characteristic prominence patterns of "novelty (...) focus" and "contrastive focus" both arise from a combination of the Givenness Constraint and AvoidF. Patterns of prominence in questions as well as in answers to questions are explained in terms of the constraints, thanks in part to the way in which the Givenness relation is defined. Head/argument asymmetries noted in the literature on Focus Projection are placed in the phonology-syntax interface, independent of discourse conditions. Deaccenting follows when AvoidF is ranked higher than constraint(s) governing head/argument asymmetries. (shrink)
The sentence Irving was closer to me than he was to most of the others contains a quantifier, most of the other, in the scope a comparative. The first part of this paper explains the challenges presented by such cases to existing approaches to the semantics of the comparative. The second part presents a new analysis of comparatives based on intervals rather than points on a scale. This innovation is analogized to the move from moments to intervals in tense semantics. (...) The remainder of the paper is concerned with an interval-based semantics of degree in relation to issues other than the comparative proper. The paper begins with a discussion of the role negative polarity has played in studies on the semantics of comparatives. (shrink)
(1) is an example of an adjectival comparative. In it, the adjective important is flanked by more and a comparative clause headed by than. This article is a survey of recent ideas about the interpretation of comparatives, including (i) the underlying semantics based on the idea of a threshold; (ii) the interpretation of comparative clauses that include quantifiers (brighter than on many other days); (iii) remarks on differentials such as much in (1) above: what they do in the comparative and (...) what they do elsewhere in the language; (iv) the relationship between comparatives and other Degree constructions (e.g. as important, too impor- tant); and (v) the types of phrases in which comparatives are found (adjective: tighter versus noun: more water). Given the nature and purpose of this essay, I have tried not to presuppose background in formal semantics and I have departed from standard practice in journal articles by, as much as possible, not interrupting the flow with footnotes and references. There are two appendices. The first provides more analytical detail and there I do rely on formal techniques of natural language semantics. The second covers the sources for the ideas surveyed here. (shrink)
In the formation of extended noun phrases, expressions are used that describe some dimension. Weight is described by each of the prenominal expressions in heavy rock, too much ballast, 2 lb rock, 2 lbs of rocks. The central claim of this paper is that the position of these types of expressions within the noun phrase limits the kinds of dimensions they may describe. The limitations have to do with whether or not the dimension tracks relevant part-whole relations. An analogy is (...) made between these constraints and the well-known constraints on thematic relations that are incurred by the position of a noun phrase in a clause. A proposal is made about the meanings of expressions like too much and 2 pounds which explains their common cross-categorial distribution and this informs the analysis of their use in noun phrases. A position is taken on the meaning of the count mass distinction which, in conjunction with the hypothesis about dimensions, explains asymmetries in the distribution of prenominal adjectives with count and mass nouns. (shrink)
This paper begins with a discussion ofcumulativity (e.g., ‘P(a) & P(b) implies P(a+b)’), formalized using a verb phrase operator. Next, the meanings of distributivity markers such aseach and non-distributivity indicators such astogether are considered. An existing analysis ofeach in terms of quantification over parts of a plurality is adopted. However,together is problematic, for it involves a cancellation or negation of the quantification associated witheach. (The four boys together owned exactly three cars could not be true if each of the boys (...) owned three cars, though the same sentence withouttogether could be.) A refinement of this idea of negated quantification over parts is proposed as an analysis for non-distributivity operators. It is then worked out in the system described in Cooper (1983), in which positive and negative extensions are assigned. Presuppositions connected with plural reference are considered at this point as well. Finally, the cumulativity operator is argued to be quantificational and therefore sensitive to contextual domain selection. This context sensitivity is claimed to be the source of distributive readings that appear in the absence of modifiers likeeach and non-distributive or collective readings that arise withouttogether. (shrink)
An introductory text in linguistic semantics, uniquely balancing empirical coverage and formalism with development of intuition and methodology. -/- This introductory textbook in linguistic semantics for undergraduates features a unique balance between empirical coverage and formalism on the one hand and development of intuition and methodology on the other. It will equip students to form intuitions about a set of data, explain how well an analysis of the data accords with their intuitions, and extend the analysis or seek an alternative. (...) No prior knowledge of linguistics is required. After mastering the material, students will be able to tackle some of the most difficult questions in the field even if they have never taken a linguistics course before. -/- After introducing such concepts as truth conditions and compositionality, the book presents a basic symbolic logic with negation, conjunction, and generalized quantifiers, to serve as the basis for translation throughout the book. It then develops a detailed compositional semantics, covering quantification (scope and binding), adverbial modification, relative clauses, event semantics, tense and aspect, as well as pragmatic phenomena, notably deictic pronouns and narrative progression. -/- A Course in Semantics offers a large and diverse set of exercises, interspersed throughout the text; those labeled “Important practice and looking ahead” prepare students for material to come; those labeled “Thinking about ” invite students to think beyond the content of the book. (shrink)
This dissertation is based on the compositional model theoretic approach to natural language semantics that was initiated by Montague (1970) and developed by subsequent work. In this general approach, coordination and negation are treated following Keenan & Faltz (1978, 1985) using boolean algebras. As in Barwise & Cooper (1981) noun phrases uniformly denote objects in the boolean domain of generalized quanti®ers. These foundational assumptions, although elegant and minimalistic, are challenged by various phenomena of coordination, plurality and scope. The dissertation solves (...) these problems by developing a ¯exible process of meaning composition, as ®rst proposed by Partee & Rooth (1983). Flexible interpretation involves semantic operations without any phonological counterpart, which participate in the interpretation process and change meanings of overt expressions. The dissertation introduces a novel ¯exible system where a small number of operations describe the behaviour of complex phenomena such as `non-boolean' and, the scope of inde®nites and the semantics of collectivity with quanti®cational NPs. The proposed theory is based on a distinction between two features of meanings in natural language. (shrink)
There are predicates that I call “stubbornly distributive” based on what happens when they are combined with plural count noun phrases. I will use these stubbornly distributive predicates to identify and analyze a certain subset of mass nouns which I call “multi-participant nouns”. Traffic and rubble are multi-participant nouns but furniture and luggage turn out not to be. Importantly, ‘typical’ mass nouns like water are multiparticipant nouns.
The association of only with focus is explained in terms of (a) a semantics for only which makes no mention of focus and (b) discourse appropriateness conditions on the use of focus and principles of quantifier domain selection. This account differs from previous ones in giving sufficient conditions for association with focus but without stipulating it in the meaning of lexical items. Detractors have contended that foci have different pragmatic import depending on whether or not they are associated with a (...) higher operator. I give evidence against this claim. Others argue that there is no deterministic connection between intonational focus and association. One argument for this is the fact that association readings are possible even when nothing in the scope of the operator is focussed. The present account predicts the absence of intonational focus in these cases and explains how the readings come about. The wide variety of associating operators provide incentive for pursuing accounts like the present one based on independent principles of grammar. (shrink)
This paper grew out of a reaction to Elisabeth Selkirk's contribution to the Handbook of Phonology (Goldsmith 1996). Section 1.2 of that article is concerned with syntactic and semantic aspects of the placement of pitch accents in English. As will be seen in the data to be presented below, the constellation of pitch accents in an utterance is determined in part by properties of the preceding discourse, including the distinction between new and old information. This means for example, that a (...) phrase containing accent is appropriate in a different set of contexts than that same phrase unaccented. Since accents are assigned within words but the pragmatics can make reference to phrases, the following question arises. Given a word that is accented, which phrases count as containing accent as far as the pragmatics is concerned and which not? This question is known as the projection problem, though I have characterized it in slightly different terms than is normally done, for reasons to be made clear below. As Selkirk shows, projection is sensitive to argument structure and other syntactic notions (see Gussenhoven(1984), Rochement(1986), Schmerling(1976) and Selkirk 1984)). In Selkirk's system, a sentence can have one or more foci and these foci are marked with an F in the syntactic tree. Focussing a phrase leads to a certain kind of.. (shrink)
I. Intervals & the Unique Witness Property (1) John is taller than Mary. (2) \!d \!dS John is d-tall ‚ (d B dS) ‚ Jill is d S-tall. (3) \!d \!dS ϕ(d) ‚ (d B dS) ‚ ψ(dS). (p-p) (4) SNEAKERS. Grant expresses interest in a pair of sneakers. I offer to buy them for him, having the impression that they cost somewhere in the $20-$30 range. We arrive at the store and to my horror I discover that the sneakers (...) cost $150. (shrink)
It is natural to think of comparisons in terms of points on a scale. Jack is taller than Jill if the point associated with Jack on the height scale is higher than Jill’s point. Jack is much taller than Jill is if Jack’s point is separated from Jill’s by a sizable amount. It is also natural to think of temporal discourse in terms of points on a time line. The analogy between the two is worth taking seriously.
This paper strives to characterize the relation between accent placement and discourse in terms of independent constraints operating at the interface between syntax and interpretation. The GIVENness Constraint requires un-F-marked constituents to be GIVEN. Key here is our definition of GIVENness which synthesizes insights from the literature on the semantics of focus with older views on information structure. AvoidF requires speakers to economize on F-marking. A third constraint requires a subset of F-markers to dominate accents.