John Collins

The paper argues for a model of polysemy based on the blueprint offered by Paul Pietroski whereby the meaning of a lexical item is an instruction to fetch a concept from an address. We show that the bare idea of fetching admits of a deep construal, where a concept is fetched, and a shallow construal, where the instruction merely links a lexical item to an address without automatically retrieving anything from the address; retrieval only occurs when the item is embedded within a syntactic structure. We offer considerations in favour of the shallow construal, which is consistent with a root conception of lexical items.

The concern in this chapter will be with the different roles of laws and luck in linguistics, and specifically in what ways various phenomena of spoken language depend upon accidental or “lucky” facts that are not sufficiently stable to feature in laws, and so should not serve as the _foci_ of linguistic theory. This “nomological” conception is what drives the Chomskyan study of an “I-language” as an internal computational system underlying human linguistic “competence,” as opposed to the more widespread, commonsensical view, advocated by, e.g., David Lewis and Michael Devitt, of language as fundamentally concerned with conventional “performance.” The chapter briefly reviews some of the crucial data that support this Chomskyan conception in the cases of syntax, phonology, and at least an internalist semantics. With regard to this last case, numerous examples are discussed that suggest that an externalist, referentialist semantics is indeed too subject to “lucky” contextual effects to be nomologically explanatory. The chapter concludes by pointing how such a denial of an externalist semantics does not have the dire consequences for metaphysical realism that both friend and foe of Chomskyan linguistics sometimes seem to assume.

Sentence meanings and propositions are intimate, but don’t exactly get on. Two familiar divergences are (i) the diversity of linguistic forms (within and between languages) that intuitively ‘say’ the same thing, such as active/passive pairs, and (ii) the contextual determination of propositional content that the type linguistic form leaves undetermined. The paper doesn’t raise doubts about these phenomena, but offers a general framework under which sentence meaning is best theorised as mere constraints upon what can be said with a given sentence, where propositions themselves, as what can be said, are artefacts that do not require any theorising beyond an understanding of the states, linguistic or otherwise, they are employed to track for our descriptive and explanatory purposes. Two principal considerations will be presented in favour this general picture: semantic underdetermination and, more especially, polysemy/copredication. The former is not the familiar contextual underdetermination, but the more significant condition that linguistic material alone essentially leaves propositions underdetermined, whatever the context. The latter is where a lexical item can express, selectively or simultaneously, distinct extensions relative to distinct predicates.

Speech production models typically assume that such production begins with the message that the speaker wants to convey. Such a message must be formatted in such a way that the other stages in speech production go swiftly, which means that it must be formatted according to the expressive demands and powers of the language that the speakers employ. Levelt and followers think that there can be a re-formatting stage that translates thoughts, couched in some other non-linguistic format, into messages. We argue that such a translation is particularly challenging, and propose an account of thinking that dispenses with the translation process. According to our proposal, linguistic creatures think in lexical concepts, that is, concepts that correspond to morphemes of the language they speak. Such lexical concepts are representations of sounds or of external signs that learners acquire in the process of word learning and are able to combine into phrases.

Recently, several philosophers have discussed conjunction reduction and other tests for polysemy/homonymy with reference to the potential ambiguity of certain expressions central to various philosophical debates. However, it has been argued that the conjunction reduction tests are not decisive in settling such matters, because some polysemous terms are acceptable under conjunction reduction. For example, conjunctions involving polysemous book and lunch can be reduced: The book is heavy and interesting and Lunch was delicious but took forever. Against this contention, the paper argues that polysemy does not undermine ambiguity tests. Terms that are acceptable in copredicational contexts in some or even in many cases also generate zeugma in other cases. We also argue, however, that conjunction reduction and similar tests have holes: monosemous terms can also give rise to zeugmatic readings. We analyze two classes of verb as potential examples.

According to Chomsky and followers, natural language is a computational system that generates syntactic structures that are counterfunctional with respect to communication. Consequently, language is more appropriately considered as being “designed” for thought rather than communication. In this paper, we argue that, while natural language, understood as an internal computational system along standard generative lines, is recruited for distinctive human thinking, such recruitment also requires, and is strongly influenced by, a process we dub lexical externalisation. We first show that there is good reason to believe that the atomic items that go to form words essentially include phonological information of externalised words. After that, we explore what externalisation implies with respect to the relation between lexical items and concepts. We suggest that externalisation has profound effects on what concepts we think with, as well as how concepts relate to lexical items (i.e., how thinking processes ultimately relate to linguistic computations).

The dissertation gives an account of the principles guiding the rational revision of belief. I develop a non-probabilistic account of belief revision. My central thesis is the claim that there are two quite different kinds of rational belief change; two methods suited to two different sorts of situation. I call these methods updating and supposing. This claim, presented in Chapter Two, is argued on the basis of results proved in Chapter One. Chapters Three and Four are applications of the distinction between updating and supposing to two areas of current philosophical debate: the semantics of conditionals; and the theory of deliberation dynamics. The discussion of conditionals appeals to the temporal asymmetry in knowledge. Why is it that we know more about the past than about the future? Chapter Five offers a new answer to that old question, based on an account of recording devices as physical systems stable under perturbation. ;I hope to have shown that many of these philosophical questions--questions about belief revision, decision theory, and conditionals, that have traditionally been framed in probabilistic terms--can be most fruitfully discussed without ever mentioning probability.

This chapter describes some basic, often puzzling features of intentionality, with an eye to its role not so much in ordinary folk ascriptions but in serious psychological explanations, especially in many of Noam Chomsky's own presentations of his theory. It then considers Chomsky's censure of the notion, leading him to deny what would seem to be the explicit intentionalisms on which he seems to rely. Implicit in Chomsky's treatment of grammar is the idea that the positing of the language faculty offers an “abstract” perspective on a physical system (the human brain) that implements the relevant features of the abstracta in ways we currently don’t understand. Two initial problems might be raised for this picture, one regarding inference and the other, generalization. The proposed model is effectively saying that theorists track states and generalizations of the linguistic system in an explicitly representational format, but what the theory is true of need not have such representational properties.

What is Truth? is a collection of original philosophical articles by many of the central figures in the field. Most of the contributions are focused on deflationism, for and against, although other approaches have a fair airing, and some novel accounts are presented. The intrinsic worth of many of the papers apart, the interest of the collection arises, I think, from its bringing into relief a number of problematic lacunae within the extant deflationisms, which, I predict, will be the main area of controversy in the years to come. The currently entertained options about truth may be usefully coordinated in the space opened by Frege's work. Roughly, Frege distinguished two types of construction in which the adjective true predicatively occurs. In the first type, the predicate takes a full finite complement and has an expletive subject; e.g., It is true that violets are blue (these types may be read as extraposed versions of the type (e.g.) [the thought] that violets are blue is true.) For these constructions, Frege suggested that the contents they express are flat with the contents of their complements, i.e., to entertain the truth of a thought is not to picture the thought or its constituents in relation or correspondence with any external elements, be they facts or objects; rather, the contents of truth predications may be expressed by the contents to which truth is predicated. Hence, we have the origin of the idea behind the familiar schema (ST) TRUE(P) iff P (Let the left flank be proxy for any type of truth predication to an explicitly presented full sentential clause. Let ST to be a generalisation over the various schemata that have been forwarded.) The second type of construction which occupied Frege is where the adjective is predicated of a quantifier noun phrase with no sentential complement; e.g., Everything Bob said is true..

The paper articulates and defends the view that paired structures of mentally 'represented' phonological and semantic features should, for all theoretical purposes, replace the notions of proposition and sentence. Following Chomsky, I refer to such pairs as expressions (EXP). In the first part, I elaborate the notion of an EXP and contrast it with that of sentence/proposition. The paper's second part questions a range of considerations which putatively show that propositions are fundamental to our understanding of meaning and cognitive attitudes. I argue that while these considerations are effective against the theoretical worth of sentences, as usually understood, they may be accommodated by the EXP-conception or otherwise eschewed.

A plausible thought about vagueness is that it involves semantic incompleteness. To say that a predicate is vague is to say (at the very least) that its extension is incompletely specified. Where there is incomplete specification of extension there is indeterminacy, an indeterminacy between various ways in which the specification of the predicate might be completed or sharpened. In this paper we show that this idea is bound to founder by presenting an argument to the effect that there are vague predicates which cannot be sharpened in such a way as to meet certain basic constraints (of penumbral connection and public accessibility) that must be imposed on the very notion of a sharpening.

The theory that structured propositions are complex act-types has been independently articulated by Peter Hanks and Scott Soames. The present paper argues that the role of the act in such theories is supererogatory, for the individuation conditions of the act-based propositions remain wholly at the level of concepts and their formal combination, features which the traditional structured proposition theorist endorses. Thus, it is shown that the traditional problems for structured propositions are only ameliorable on the act conception by appeal to the very resources of the traditional conception. It is also shown that the act theories have no act-based account of quantification and other operator-involving relations. An elementary account of propositions is sketched, after Fodor, which retains the virtues of the traditional structural conception without falling into ’Platonism’.

The paper considers our ordinary mentalistic discourse in relation to what we should expect from any genuine science of the mind. A meta-scientific eliminativism is commended and distinguished from the more familiar eliminativism of Skinner and the Churchlands. Meta-scientific eliminativism views folk psychology qua folksy as unsuited to offer insight into the structure of cognition, although it might otherwise be indispensable for our social commerce and self-understanding. This position flows from a general thesis that scientific advance is marked by an eschewal of folk understanding. The latter half of the paper argues that, contrary to the received view, Chomsky's review of Skinner offers not just an argument against Skinner's eliminativism, but, more centrally, one in favour of the second eliminativism.

Generics are exception-admitting generalisations, which find expression in apparently diverse linguistic forms. A standard claim is that there is a hidden linguistic unity to genericity in the form of a covert operator, Gen. This article surveys and rejects a range of considerations that purport to show Gen to be syntactically essential to the explanation of a range of linguistic phenomena connected to genericity. The conclusion reached is that genericity is not a specifically linguistic property insofar as it does not supervene upon any especial linguistic device; rather, genericity supervenes on broader cognitive competencies within linguistic constraints.

Much of the best contemporary work in the philosophy of language and content makes appeal to the theories developed in generative syntax. In particular, there is a presumption that—at some level and in some way—the structures provided by syntactic theory mesh with or support our conception of content/linguistic meaning as grounded in our first-person understanding of our communicative speech acts. This paper will suggest that there is no such tight fit. Its claim will be that, if recent generative theories are on the right lines, syntactic structure provides both too much and too little to serve as the structural partner for content, at least as that notion is generally understood in philosophy. The paper will substantiate these claims by an assessment of the recent work of King, Stanley, and others.