Robert Stainton

Our exposition is framed around two questions: What interpretive effects can linguistic utterances have? What causes those effects? Lepore and Stone make an empirical case that some effects are contributions to the public record of a conversation determined by linguistic conventions—following Lewis—while non-contributions (our term) produced by imagination offer no determinate content—following Davidson. They thereby replace the old semantics–pragmatics divide by eliminating conversational implicature altogether. We critique Lepore and Stone’s position on empirical grounds, presenting cases in which contributions are made non-conventionally. We also critique their view methodologically, presenting a dilemma by which they either cannot handle many cases using their framework or they do so in an ad hoc fashion. We conclude by suggesting Relevance Theory as an alternative that follows Lepore and Stone’s purported methodology and handles many of their empirical cases.

This chapter focuses on the speech act of assertion. In the first section, it considers whether what makes something an assertion is a matter of semantics — where, in this context, ‘semantic’ is used in the sense of being a matter of linguistic convention. In the second section, it considers the determinants of asserted content, and whether they are semantic (in a different, specialized sense, to be explained). Specifically, it asks, given that an action is an assertion, whether its specific content can be fixed solely on the basis of its form/content together with certain antecedently settled referents for context sensitive items. The answer is that whether we have an assertion at all, and if so what its content is, has rather more to do with pragmatics than has often been assumed.

This chapter lays out five different ways of reading the context principle: methodological, metasemantic, pragmatic, semantic, and psychological. It notes several rationales for embracing the principle. It then objects to the principle, on several of its readings, from non-sentence use. The suggested result, in the face of this objection, was three parts consistency and two parts inconsistency: (a) the first reading of the principle would be largely untouched; (b) the second would be left unsupported; but (c) the other readings would be outright falsified.

This chapter focuses on language-thought relations. It argues that grasping a proposition does not require deploying internally a sentence of natural language that expresses it. This alone threatens the view that thought is inner speech. It then considers a specific subcase, viz. where the proposition so grasped is part of an argument. It points out that, if premises/conclusions can be conveyed sub-sententially, then there are things that are not items of natural language but which have logical form.

This chapter discusses what is going on in apparent cases of sub-sentential speech acts. It begins with a relatively broad-strokes account of how agents can communicate using ordinary words and phrases; this provides the ‘general view’. The general view is taken to be the core of the positive position. A preferred way of spelling out the general view is presented — the so-called ‘specific view’ — which introduces internal mental representations of various sorts. The result is a view that is not only pragmatic, but representational.

There are two obvious kinds of manoeuvres for rejecting the existence of full-fledged non-sentential speech acts, thereby avoiding the implications canvassed briefly in Chapter 1. The first is to deny that there are full-fledged speech acts performed at all. The second obvious kind of manoeuvre is to deny that the examples are truly non-sentential, alleging instead that every apparently sub-sentential speech act is actually an utterance of some kind of sentence. Jason Stanley points out that one must consider a ‘divide-and-conquer’ strategy, explaining away some cases as underlyingly sentential and other cases as not full-fledged speech acts. This chapter aims to rebut this ‘divide-and-conquer’ strategy.

This chapter discusses the plausibility of syntactic ellipsis in abstraction, surveying data-based arguments both for and against an ellipsis account (from case assignment, binding phenomena, licensing conditions, etc.). It is shown that the evidence against syntactic ellipsis heavily outweighs the evidence for it. It is argued that sub-sentential speech acts cannot be explained away as really involving syntactically elliptical sentences. Hence,premise one of the two-premise argument still stands: namely that speakers genuinely can utter ordinary words and phrases in isolation, and thereby perform full-fledged speech acts.

This chapter looks at semantic ellipsis: expressions that are not syntactically sentential, but nevertheless have characters that yield propositional contents given a context. Relevant examples include ‘Attention!’ and ‘No shirt, no shoes, no service’. The idea is to use such cases to explain away apparently sub-sentential speech: an attempt is made to assimilate the cases under discussion (like ‘Nice dress’ and ‘On the stoop’) to commands such as ‘Attention!’. This attempted assimilation is rejected on two fronts. First, it would require that human languages contain masses of semantically elliptical sentences. Second, postulating these masses of elliptical sentences would introduce lots of new ambiguities — which would be otiose, since a language that lacked such ambiguities would be used in just the ways we actually observe.

This chapter examines the idea that an ordinary sentence is spoken — either actually produced, or just intended/recovered — in seemingly sub-sentential speech. It discusses the idea that ‘shorthand’ in some sense is at work. What emerges is that it is very implausible to maintain that an ordinary sentence is produced in these cases, and that the only senses in which ‘shorthand’ might truly be at work are ones that restate, rather than reject, premise one outlined in chapter 1: namely, that speakers genuinely can utter ordinary words and phrases in isolation, and thereby perform full-fledged speech acts.

This chapter explains three background issues additional to those presented in Chapter 1. First, it contrasts three senses of ‘sentence’ — syntactic, semantic, and pragmatic — and draws attention to three corresponding senses of ‘elliptical’. Second, it defends a commitment to an open evidence-base for the philosophical study of language, in the face of arguments that take off either from the nature of mind, or from the ontology of language. Finally, it briefly notes a commitment to an empirical thesis about the mind, viz. that it is structured into faculties, including in particular, one specific to language.

This chapter introduces and explains the two key premises around which the book is built. Premise 1 says that speakers genuinely can utter ordinary words and phrases in isolation, and thereby perform full-fledged speech acts. Premise 2 says that if speakers genuinely can utter ordinary words and phrases in isolation, and thereby perform full-fledged speech acts, then such-and-such implications obtain. What emerges repeatedly from the discussion of the two premises, and in several different ways, are two issues about words and thoughts: whether mere words (as opposed to sentences) can be used to state complete thoughts, and what the answer to this question entails about the general issue of how language (i.e., ‘words’) relates to thinking (i.e., ‘thoughts’). It is these two issues about words and thoughts that give rise to the title of the book.

Philosophical reflection on death dates back to ancient times, but death remains a most profound and puzzling topic. Samantha Brennan and Robert Stainton have assembled a compelling selection of core readings from the philosophical literature on death. The views of ancient writers such as Plato, Epicurus, and Lucretius are set alongside the work of contemporary figures such as Thomas Nagel, John Perry, and Judith Jarvis Thomson. Brennan and Stainton divide the anthology into three parts. Part I considers questions about the nature of death and our knowledge of it. What does it mean to be dead? Is it possible to survive death? Is the end of life a mystery? Part II asks how we should view death. What (if anything) is so bad about dying? If death is nothingness, should it be feared or regretted? Part III examines ethical questions related to killing, particularly abortion, euthanasia and suicide. Is killing ever permissible? Under what conditions or circumstances?

This chapter introduces a pragmatics-oriented approach to non-sentential speech, and defends it against two recent attacks. Among other things, it rehearses and elaborates a defence against the idea that much, or even all, of such speech is actually syntactically elliptical — and hence should be treated semantically, rather than pragmatically. The chapter is structured as follows. Section 1 introduces the phenomenon, contrasts semantic versus pragmatic approaches to it, and explains some of what hinges on which approach is taken. Section 2 presents Jason Stanley's objections to the pragmatics-oriented approach, and his counterproposal that all truth-conditional effects of context on what is asserted can be traced to elements of underlying structure. Section 3 canvasses numerous varieties of ellipsis. It focuses on what kind of ‘ellipsis’ is required if the pragmatics-oriented approach is actually to be rejected rather than being recast in other terms. Section 4 responds to Stanley, and shows that the two known varieties of ellipsis which would turn the trick for the semantics-oriented approach are not empirically plausible when applied to the cases in question. In further defence of this conclusion, Section 5 presents a family of syntactic arguments — from Peter Ludlow and others — in favour of ellipsis (and a concomitant semantics-based approach), and replies to these latter arguments. Section 6 returns explicitly to issues of the semantics-pragmatics boundary and draws some larger lessons about what sub-sentential speech does and does not entail, including what it does not entail _vis-à-vis_ the distinctiveness of linguistic communication, and the viability of truth-conditional semantics.

In his recent book, The Elm and the Expert, Fodor attempts to reconcile the computational model of human cognition with information‐theoretic semantics, the view that semantic, and mental, content consists of nothing more than causal or nomic relationships, between words and the world, or (roughly) brain states and the world. In this paper, we do not challenge the project. Nor do we show that Fodor has failed to carry it out. instead, we urge that his analysis, when made explicit, turns out rather differently than he thinks. In particular, in some places where he sees problems, he sometimes shows that there is no problem. And while he says two conceptions of information come to much the same thing, his analysis shows that they are very different. This is rather strange.