Zoltan Szabo

This paper is a defense of substantive explanations in semantics. It begins by offering a diagnosis of why the view that semantic theories are merely descriptive has been widely accepted in philosophy and suggests that these grounds are not compelling. Then it argues that semantic explanations don’t have a uniform direction—upwards or downwards the syntactic tree. There is an explanatory division within the lexicon: the meanings of content words (nouns, verbs, adjectives, and adverbs) are semantically fundamental, while the meanings of function words (auxiliaries, connectives, copulas, derivational morphemes, determiners, expletives, prepositions, etc.) are derivative.

According to the traditional doctrine of logical form, sentences have an underlying structure which lays bare their inferential profiles. Logical form is supposed to be inherent in the sentence (not ascribed to it as a result of formalization), and it is supposed to capture its logical (not merely syntactic or semantic) features. This chapter argues against the existence of logical form in this sense by casting doubt on the coherence of the idea that certain inferences are valid solely in virtue of their form. And while the idea that some inferences are valid in virtue of their form and facts of logic is perfectly coherent, these inferences are arguably unavailable in natural languages. Moreover, appeal to semantic competence is no help, because logical competence is distinct from our knowledge of language. The upshot of the argument is that we should abandon the Davidsonian hope that an adequate compositional semantics could explain why logical validities are valid, or how we know that they are.

According to a traditional view, semantics is primarily concerned with what is said and pragmatics with what is implicated. This introductory chapter discusses the Gricean underpinning of this view and argues that it can be made more precise if we make two Austinian distinctions: between a locutionary and an illocutionary sense of ‘what is said’, and between an illocutionary and a perlocutionary sense of ‘what is meant’. Once these distinctions are drawn the pretense of terminological innocence disappears from the traditional view. The chapter ends with six sensible lines each of which could _in principle_ be called the semantics/pragmatics divide. They jointly define the border area whose mapping is the concern of the chapters in this volume.

In this paper, we provide a comprehensive survey of the space of possible analyses of the phenomenon of quantifier domain restriction, together with a set of considerations which militate against all but our own proposal. Among the many accounts we consider and reject are the ‘explicit’ approach to quantifier domain restric‐tion discussed, for example, by Stephen Neale, and the pragmatic approach to quantifier domain restriction proposed by Kent Bach. Our hope is that the exhaustive discussion of this special case of context‐dependence will provide guidelines for how to decide, for an arbitrary case of context‐dependent discourse, whether it should be treated syntactically, semantically, or pragmatically.

0. Abstract In this paper, I argue that although the behavior of adjectives in context poses a serious challenge to the principle of compositionality of content, in the end such considerations do not defeat the principle. The first two sections are devoted to the precise statement of the challenge; the rest of the paper presents a semantic analysis of a large class of adjectives that provides a satisfactory answer to it. In section 1, I formulate the context thesis, according to which the content of a complex expression depends on the context of its utterance only insofar as the contents of its constituents do. If the context thesis is false, the content of some complex expression is not compositionally determined. In section 2, using an example due to Charles Travis, I construct an objection to the context thesis based on the behavior of the adjective ‘green’. In section 3 and 4, I look at some of the difficulties surrounding the semantics of ‘good’, which provide the motivation for the thesis that most adjectives are contextually incomplete one-place predicates. In section 5, I discuss how ‘green’ and other color adjectives can be treated within such a semantic theory. Since this theory is compatible with the context thesis, the objection against the compositionality of content looses its force.

Felicitous underspecification—apparently flawless use of context-sensitive words in contexts where they cannot be assigned unique semantic values—is rather common in ordinary speech. King presents a hypothesis about the mechanism conversational participants employ handling felicitous underspecification, one that fits the rich data he surveys well. I will begin by illustrating how King’s account could be put to use in making sense of what happens in a real life conversation. Then I will point out certain shortcomings of the explanation and offer suggestions about how they might be overcome.

In a recent book, Logical Form: between Logic and Natural Language, Andrea Iacona argues that semantic form and logical form are distinct. The semantic form of a sentence is something that (together with the meanings of its parts) determines what it means; the logical from of a sentence is something that (all by itself) determines whether it is a logical truth. Semantic form does not depend on context but logical form does: for example, whether ‘This is this’ is a logical truth depends on whether the two occurrences of ‘this’ are used to demonstrate the same individual. I respond by claiming that logical form is indifferent to reference and is sensitive only to obligatory co-reference. When the speaker intends both occurrences of ‘this’ to be interpreted the same way the logical from of ‘This is this’ is a=a, while in a context where the speaker has no such intention it is a=b. This proposal allows a much more conservative revision of the traditional picture than the one suggested by Iacona. Instead of identifying the logical form of a natural language sentence by seeking a formalization in an artificial language, we obtain it through abstraction from its syntactic analysis: replacing the non-logical expressions by schematic letters, making sure that we use identical ones if and only if the speaker intended co-reference.

Dickie (2020) presents an argument against the traditional, broadly Gricean view of conversation. She argues that speakers must sometimes be more specific than required for sharing knowledge on a topic of common concern. Her proposed solution is to claim that the goal of conversation is not just sharing knowledge but also sharing cognitive focus. In response, I argue that her proposal faces both conceptual and empirical difficulties, and that the traditional view can handle the problem of non-specificity by acknowledging that in order to sustain mutual trust, conversational participants should be less than optimally efficient.

This article presents three more-or-less-traditional considerations for compositionality. The first is that the usual statement of the compositionality principle is massively ambiguous. One of the eight available readings rules out all sources of multiplicity in meaning in complex expressions besides the lexicon and the syntax. Others are more permissive—how much more is not always clear. The second claim is that traditional considerations in favour of compositionality are less powerful than is often assumed. Compositionality is best construed as an empirical hypothesis on meanings expressed in natural languages. Finally, the third claim is that, even if compositionality is true, most of the debates in philosophy, linguistics, and psychology surrounding compositionality will remain open. These debates tend to be about significantly stronger theses.