Stephen Barker

In this paper we present a new metaphysical theory of material objects. On our theory, objects are bundles of property instances, where those properties give the nature or essence of that object. We call the theory essential bundle theory. Property possession is not analysed as bundle-membership, as in traditional bundle theories, since accidental properties are not included in the object’s bundle. We have a different story to tell about accidental property possession. This move reaps many benefits. Essential bundle theory delivers a simple theory of the essential properties of material objects; an explanation of how object coincidence can arise; an actual-world ground for modal differences between coincident objects; a simple story about intrinsic properties; and a plausible account of certain ubiquitous cases of causal overdetermination.

The arrow of chance –understood as objective single case propensity –and the arrow of causation go together. Where an event c causes another event e, then c contributes to a chance of e that is realized with e’s occurrence. Where there is at a time t a chance of e occurring at some other time, and e occurs, then something c at t caused e. Having argued for this coalescence of the arrows of chance and cause, I then focus on what it means. I reject the idea that the coalescence is explained by an analysis of cause in terms of counterfactuals and chance – as those dedicated to the chance-raising analysis of causation propose. I also argue against the idea that we explain it by analysing chance in terms of prior notions of cause and objective probability. What the coalescence shows, I argue, is that causation is a specific pattern of events involving realized chances. I then show that we can analyse dispositions in terms of chance.

There is a wide-spread belief amongst theorists of mind and language. This is that in order to understand the relation between language, thought, and reality we need a theory of meaning and content, that is, a normative, formal science of meaning, which is an extension and theoretical deepening of folk ideas about meaning. This book argues that this is false, offering an alternative idea: The form of a theory that illuminates the relation of language, thought, and reality is a theory of language agency. In a nutshell, the theory of language agency is a theory of competence, without being a theory of understanding or grasping rules. It is a theory of cognitive structure and language production. This theory distils all there is to say about language, thought, and reality. It does not supplement a theory of truth-conditions or semantic norms. It is not the explanation of how a speaker, qua cognitive system causally embedded in a larger reality, is able to use a language with some pre-existing semantic characterization. There is no pre-existing semantic characterization. Nevertheless, there are facts of meaning, as good as any other facts. The dissolution of the theory of meaning is accompanied by another disappearance. That is the disappearance of metaphysical questions in a number of domains. Once we complete the theory of language agency, then just as theoretical questions about meaning disappear, certain theoretical questions about existence disappear. Having provided a theory of the language agency for talk of meaning, fact, property, relation, and proposition, there is no question left over about what meanings, facts, properties, relations, and propositions are. There is no theory to be given of their natures. This is not because they have primitive irreducible natures. Rather it is because, in a sense to be clarified in this work, they lack natures. I call this approach to language agency Global Expressivism. That is because it generalizes some of the insights brought to the study of value-language by expressivists. However, it removes these insights from the clouding affects of attempting to make expressivism a semantic theory. Expressivism about value fails as a semantic theory of value talk. However, global expressivism can succeed as a theory of all talk because it is not a semantic theory but a theory of language agency, wherein the theory of meaning is replaced by a theory of talk about meaning.

At the heart of semantics in the 20th century is Frege’s distinction between sense and force. This is the idea that the content of a self-standing utterance of a sentence S can be divided into two components. One part, the sense, is the proposition that S’s linguistic meaning and context associates with it as its semantic interpretation. The second component is S’s illocutionary force. Illocutionary forces correspond to the three basic kinds of sentential speech acts: assertions, orders, and questions. Forces are then kinds of acts in which propositions are deployed with certain purposes. I sketch a speech-act theoretic semantics in which that distinction does not hold. Instead of propositions and forces, the theory proposes proto-illocutionary acts and illocutionary acts. The orthodox notion of a proposition plays no role in the framework, which is a good thing, since that notion is deeply problematic. The framework also shows how expressionists, who embrace a sophisticated speech-act framework, face no Frege-Geach embedding problem, since the latter assumes the Sense/Force distinction.

This book develops an alternative approach to sentence- and word-meaning, which I dub the speech-act theoretic approach, or STA. Instead of employing the syntactic and semantic forms of modern logic–principally, quantification theory–to construct semantic theories, STA employs speech-act structures. The structures it employs are those postulated by a novel theory of speech-acts. STA develops a compositional semantics in which surface grammar is integrated with semantic interpretation in a way not allowed by standard quantification-based theories. It provides a pragmatic theory of truth, a treatment of logically complex discourse as expressive cognitive states, and a background metaphysics in which the world is a totality of logically simple states of affairs. The book also puts forward an account of how intentional states provide the simple, representational foundation for a superstructure of speech-act structures–a system of thoughts–that far outruns the expressive power of the intentional foundation. In short, it provides an account of cognitive foundations of a language and a naturalistic reduction of semantics through an expressive theory of semantic norms.

Emotivist, or non-descriptivist metaethical theories hold that value-statements do not function by describing special value-facts, but are the mere expressions of naturalistically describable motivational states of (valuing) agents. Non-descriptivism has typically been combined with the claim that value-statements are non-cognitive: they are not the manifestations of genuine belief states. However, all the linguistic, logical and phenomenological evidence indicates that value-statements are cognitive. Non-descriptivism then has a problem. Horgan and Timmons propose to solve it by boldly combining a non-descriptivist thesis about value with the claim that value-judgements are after all cognitive. Although possessing many attractive features, I argue that their framework fails to deliver the promised results; it suffers from a certain internal incoherence about the concept of content and mis-characterizes the descriptive/non-descriptive content distinction required by nondescriptivism.

A comprehensive theory ofeven if needs to account for consequent ‘entailing’even ifs and in particular those of theif-focused variety. This is where the theory ofeven if ceases to be neutral between conditional theories. I have argued thatif-focusedeven ifs,especially if andonly if can only be accounted for through the suppositional theory ofif. Furthermore, a particular interpretation of this theory — the conditional assertion theory — is needed to account foronly if and a type of metalinguistic negation ofQ if P. We therefore have evidence that the currently accepted approaches to conditionals are basically wrong about the semantic forms they attribute toif P, Q.11.

The standard view about counterfactuals is that a counterfactual (A > C) is true if and only if the A-worlds most similar to the actual world @ are C-worlds. I argue that the worlds conception of counterfactuals is wrong. I assume that counterfactuals have non-trivial truth-values under physical determinism. I show that the possible-worlds approach cannot explain many embeddings of the form (P > (Q > R)), which intuitively are perfectly assertable, and which must be true if the contingent falsity of (Q > R) is to be explained. If (P > (Q > R)) has a backtracking reading then the contingent facts that (Q > R) needs to be true in the closest P-worlds are absent. If (P > (Q > R)) has a forwardtracking reading, then the laws required by (Q > R) to be true in the closest P-worlds will be absent, because they are violated in those worlds. Solutions like lossy laws or denial of embedding won't work. The only approach to counterfactuals that explains the embedding is a pragmatic metalinguistic approach in which the whole idea that counterfactuals are about a modal reality, be it abstract or concrete, is given up.

Must counterfactual analyses of causation appeal to chances and chance-raising in order to tame indeterministic causation? It is generally thought so. 1 Against the grain, I contend that appeal to chance-raising is not required to analyse chancy causation. In Section 1 below I argue that the standard cases motivating the chance-raising analysis – cases such as bombardments of radioactive atoms causing the decay of those atoms – should be treated as instances of preemption. Such cases, I urge, are open to exactly the same kind of analysis as cases of preemption in a deterministic setting. With that thought in mind, I set out to provide a unified counterfactual analysis of causation for both deterministic and indeterministic cases, without appeal to chance-raising. In Section 2, I outline an initial sketch for the deterministic case of the theory. The account is a descendant of Lewis's quasi-dependence analysis but expressed in terms of counterfactual embedding. I show how this theory copes with preemption, trumping and over-determination. In Section 3, I broach the vexed issue of the non-transitivity of cause. I accept that cause is not transitive, and modify the theory given in Section 2 to derive the final theory. The overall picture is roughly this: c caused e if and only if c and e occur and had c not occurred a disposition to a causal path issuing in e would not have been manifested in the circumstances. I show in Section 4 how this theory applies to chancy causation.

I argue that a new solution to the semantic paradoxes is possible based on truth-making. I show that with an appropriate understanding of what the ultimate truth and falsity makers of sentences are, it can be demonstrated that sentences like the liar are alethically undecidable. That means it cannot be said in principle whether such sentences are true, not true, false, not-false, neither true nor false, both true and false, and so on. I argue that this leads to a solution to the semantic paradoxes that appears to be free of revenge problems, allows us to maintain classical logic and the validity of the T-schema.

It seems to be generally accepted that (a) counterfactual conditionals are to be analysed in terms of possible worlds and inter-world relations of similarity and (b) causation is conceptually prior to counterfactuals. I argue here that both (a) and (b) are false. The argument against (a) is not a general metaphysical or epistemological one but simply that, structurally speaking, possible worlds theories are wrong: this is revealed when we try to extend them to cover the case of probabilistic counterfactuals. Indeed a type of counterfactual probability exists which cannot be expressed in possible worlds terms at all. The argument against (b) emerges when we look at the form of an adequate account of both probabilistic and non-probabilistic counterfactuals. I do this by sketching and defending an approach to counterfactuals that, first, invoke a generalized notion of cause as primitive and, secondly, is algorithmic in form: counterfactuals are evaluated algorithmically in terms of other counterfactuals, without vicious circularity. Structures like possible worlds do not play a role either in general truth-conditions or in evaluation. They are simply the wrong sorts of structures.

Can hybridism about moral claims be made to work? I argue it can if we accept the conventional implicature approach developed in Barker (Analysis 2000). However, this kind of hybrid expressivism is only acceptable if we can make sense of conventional implicature, the kind of meaning carried by operators like ‘even’, ‘but’, etc. Conventional implictures are a form of pragmatic presupposition, which involves an unsaid mode of delivery of content. I argue that we can make sense of conventional implicatures, but doing so requires we embrace a form of pure, non-hybrid expressivism. This is a cognitivist expressivism I have developed elsewhere. We need cognitivist expressivism to make sense of how we evaluate—judge as correct or incorrect—implicature-bearing sentences. Once we embraced the possibility of this pure expressivism, we might as well be pure expressivists about normative discourse too. I show how we can do that. The motivations for a specifically hybrid theory are dialectically undercut.

My goal is to illuminate truth-making by way of illuminating the relation of making. My strategy is not to ask what making is, in the hope of a metaphysical theory about is nature. It's rather to look first to the language of making. The metaphor behind making refers to agency. It would be absurd to suggest that claims about making are claims about agency. It is not absurd, however, to propose that the concept of making somehow emerges from some feature to do with agency. That's the contention to be explore in this paper. The way to do this is through expressivism,. Truth-making claims, and making-claims generfally, are claims in which we express mental states linked to our maipulation of concepts, like truth. In particular, they express disposition to undertake derivations using inference rules, in which introduction rules have a specific role. I then show how this theory explains our intuitions about truth's asymmetric dependence on being.

If we seek to analyse causation in terms of counterfactual conditionals then we must assume that there is a class of counterfactuals whose members (i) are all and only those we need to support our judgements of causation, (ii) have truth-conditions specifiable without any irreducible appeal to causation. I argue that (i) and (ii) are unlikely to be met by any counterfactual analysis of causation. I demonstrate this by isolating a class of counterfactuals called non-projective counterfactuals, or NP-counterfactuals, and indicate how counterfactual analyses of causation must appeal to them to account for the correct causal judgements we make. I show that the truth-conditions of NP-counterfactuals are specifiable only by irreducible appeal to causation. A dilemma then holds: if counterfactual analyses of causation eschew appeal to NP-counterfactuals they are empirically inadequate, but if they appeal to NP-counterfactuals they are circular and thus conceptually inadequate.

I argue that indefinite descriptions are referring terms. This is not the ambiguity thesis: that sometimes they are referring terms and sometimes something else, such as quantifiers. No. On my view they are always referring terms; and never quantifiers. I defend this thesis by modifying the standard conception of what a referring term is: a modification that needs to be made anyway, irrespective of the treatment of indefinites. I derive this approach from my speech-act theoretic semantics. The basic thought is that referring terms have as their meanings speech-acts of a certain kind called proto-referring acts. These are acts in which speakers advertise or present intentions to denote, where denotation is a word-world relation, and advertising an intention is acting as if one has intentions, where it is open whether one has them or not, or whether the referring term used denotes or not. I show how this works for proper names. The meaning of a proper name is the speech-act proto-referring act type defined by a certain referential tree. This gives us the basis for an account of proper name meaning irrespective of denotation: a uniform treatment of full and empty names. Applied to indefinites, we can capture cases where speakers perform proto-acts—in which they advertise an intention to denote something—where they intend to denote, but others where they do not, but they still perform the proto-act: advertising an intention to denote. Two cases are ‘Fred saw a hippo’ and ‘Jane did not see a hippo’. In both cases U performs the same proto-act, represent it as Rpro, but possesses the advertised intention in the first case, but not in the second. In the first case, Rpro gains referential content—its descriptive content is expanded to include seen by Fred, whereas in its second token use it has no content augmentation or sentential determination of reference. In the second case ‘a hippo’ is an empty referring term, just as ‘Pegasus’ is empty in ‘Fred did not see Pegasus; he does not exist’. But in both sentences ‘a hippo’ functions in the same basic way: in both cases the basic proto-referring act Rpro is performed. Using this approach, I show how definite descriptions can be construed as indefinites with added meaning.

I use the principle of truth-maker maximalism to provide a new solution to the semantic paradoxes. According to the solution, AUS, its undecidable whether paradoxical sentences are grounded or ungrounded. From this it follows that their alethic status is undecidable. We cannot assert, in principle, whether paradoxical sentences are true, false, either true or false, neither true nor false, both true and false, and so on. AUS involves no ad hoc modification of logic, denial of the T-schema's validity, or obvious revenge.

I argue that figurative speech, and irony in particular, presents a deep challenge to the orthodox view about sentence content. The standard view is that sentence contents are, at their core, propositional contents: truth-conditional contents. Moreover, the only component of a sentence’s content that embeds in compound sentences, like belief reports or conditionals, is the propositional content. I argue that a careful analysis of irony shows this view cannot be maintained. Irony is a purely pragmatic form of content that embeds in compound sentences. The standard view cannot account for this fact. I sketch out a speech-act theoretic framework that can. But in accepting this alternative framework we are giving up on the whole idea of an autonomous semantics.