On The Autonomy of Linguistic Meaning Frege and many following him, such as Dummett, Geach, Stenius and Hare, have envisaged a role for illocutionary force indicators in a logically perspicuous notation. Davidson has denied that such expressions are even possible on the ground that any putative force indicator would be used by actors and jokers to heighten the drama of their performances. Davidson infers from this objection a Thesis of the Autonomy of Linguistic Meaning: symbolic representation necessarily breaks any close tie with extra-linguistic purpose. A modified version of Frege's ideal is here propounded according to which an expression is a force indicator just in case it indicates force in any speech act in which it occurs. It is shown that attainment of such an ideal would not have deprived Frege of what he desired of a perspicuous notation. In elaborating this ideal we also espouse an illocutionary conception of validity and argue that the ideal is one to which English conforms: parenthetical speech act verbs in the first person present indicative active are force indicators in the modified sense. But the mere possibility of force indicators in the modified sense is enough to show to be at best a half-truth Davidson's Autonomy Thesis.

Direct Reference and Implicature On some formulations of Direct Reference the semantic value, relative to a context of utterance, of a rigid singular term is just its referent. In response to the apparent possibility of a difference in truth value of two sentences just alike save for containing distinct but coreferential rigid singular terms, some proponents of Direct Reference have held that any two such sentences differ only pragmatically. Some have also held, more specifically, that two such sentences differ by conveying distinct conversational implicata, and that a conflation of implicatum with semantic content leads speakers to judge such sentences capable of differing in truth value. It is argued here that this latter defense of Direct Reference employs false explanans, on the ground that speakers conflate semantic content with implicatum only in quite special cases, and we have independent grounds for thinking that sentences reporting speech acts and attitudes are not cases of this sort.

Quantity, Volubility, and Some Varieties of Discourse Abstract: Grice's Quantity maxims have been widely misinterpreted as enjoining a speaker to make the strongest claim that she can, while respecting the other conversational maxims. Although many writers on the topic of conversational implicature interpret the Quantity maxims as enjoining such volubility, so construed the Quantity maxims are unreasonable norms for conversation. Appreciating this calls for attending more closely to the notion of what a conversation requires. When we do so, we see that eschewing an injunction to maximal informativeness need not deprive us of any ability to predict or explain genuine cases of implicature. Crucial to this explanation is an appreciation of how what a conversation, or a given stage of a conversation, requires, depends upon what kind of conversation is taking place. I close with an outline of this dependence relation that distinguishes among three importantly distinct types of conversation.

Illocutions, Implicata, and What a Conversation Requires An approach is offered to the prediction and explanation of quantity implicata (implicata whose calculation depends upon adversion to Grice's maxim of Quantity) that, unlike the majority of approaches available, does not construe Quantity as requiring speakers to make the strongest claim that their evidence permits. Central to the treatment offered is an elaboration of the notion of what a conversation requires as appealed to in the Cooperative Principle and in the Quantity maxim. What conversations require is construed as depending, at any given point, upon (i) the aim(s) of the conversation taking place, (ii) the conversational record, which includes such features as common ground and salience relations among objects, and (iii) any proffered illocution calling for a reply. In accounting for this third dimension a partial characterization is provided of the speech acts of assertion and interrogation in terms of their role in constraining the progress of the conversation in which they occur.

Indeterminism and the Thin Red Line (with Nuel Belnap, University of Pittsburgh) Abstract: Central to the idea of indeterminism is this: At a given moment in the history of the world there are a variety of ways in which affairs might carry on. In considering indeterminism we concentrate on a generalization of McTaggart's "B-series," which describes the before/after relations that moments bear to one another, with no reference to whether any of those moments are past, present or future. One common way of conjoining the aspect of indeterminism formulated above with an understanding of our world deriving from a construal of the B-order as a series, is to hold that at a given moment in the history of our world from which there are a variety of ways in which affairs might carry on, one of those ways is asymmetrically privileged over against all others as being what is actually going to happen. It used to be said of the British Empire that it was maintained by a thin red line of soldiers in service to the Queen. We express the view just sketched by saying that from among the lines along which history might go, subsequent to an indeterministic moment, one of those lines is the course along which history will go, and it is both thin and red. We argue that in spite of its stalwartness the Thin Red Line cannot be maintained.

Symmetry Arguments for Cooperation in the Prisoners' Dilemma (with Cristina Bicchieri, Carnegie-Mellon University) Abstract: A variety of philosophers, decision theorists, and game theorists have advanced arguments according to which rational agents who play a one-shot Prisoners' Dilemma (PD) should choose to cooperate if they know either that they are in identical circumstances, or are in some sense identical twins. The reasoning in favor of the cooperative solution raises the question whether the PD with the appropriate "Identicality" assumption is inconsistent, since there is a separate argument employing dominance reasoning that favors non-cooperation. We argue here that the question can only be answered relative to a clarification of the Identicality assumption. There turns out to be only one interpretation of the Identicality assumption that justifies cooperation, but this interpretation may be controversial. On another interpretation of this assumption the description of the PD game is indeed inconsistent, while on the remaining interpretations of that assumption the argument for the cooperative solution is fallacious. Seeing the inconsistency teaches us a lesson concerning the kinds of postulates that can be added to the description of a game. Seeing the fallacy will motivate a more careful treatment of the relation of modal concepts to the notion of rational choice, and will help us to illuminate the relevance to game theory of technical tools developed by logicians.

Reflections on Reflection: van Fraassen on Belief (with Christopher Hitchcock, Rice University) Abstract: In 'Belief and the Will' van Frasssen employed a diachronic Dutch Book argument to support a counterintuitive principle called Reflection. There and subsequently van Fraassen has put forth Reflection as a linchpin for his views in epistemology and philosophy of science, and for the Voluntarism (first person reports of subjective probability are undertakings of commitments) that he espouses as an alternative to Descriptivism (first person reports of subjective probability are merely self-descriptions). Christensen and others have attacked Reflection, taking it to have unpalatable consequences. We prescind from the question of the cogency of diachronic Dutch Book arguments, and focus on Reflection's proper interpretation. We argue that Reflection is not as counterintuitive as it appears--that once interpreted properly the status of the counterexamples given by Christensen and others is left open. We show also that Descriptivism can make sense of Reflection, while Voluntarism is not especially well suited to do so.

Moore's Many Paradoxes Over the last two decades J.N. Williams has developed an account of the absurdity of such utterances as "It's raining but I don't believe it" that is both intuitively plausible and applicable to a wide variety of forms that this so-called Moorean absurdity can take. His approach is also noteworthy for making only minimal appeal to principles of epistemic or doxastic logic in its account of such absurdity. We first show that Williams places undue emphasis upon assertion and belief: It is similarly absurd for a person to accept a proposition P as a supposition for the sake of argument while denying that her state of mind is one of supposing P, yet Williams has no account of this. Williams' approach is then modified to account for such a case. That modification employs a principle of doxastic logic that is at least plausible as the one on which Williams relies, while being unlike his principle in applying to cases other than belief.

Illocutionary Force and Semantic Content Illocutionary force and semantic content are widely held to occupy utterly different categories in at least two ways: (1) Any expression serving as an indicator of illocutionary force must be without semantic content, and (2) no such expression can embed. A refined account of the force/content distinction is offered here that (a) does the explanatory work that the standard distinction does, while, in accounting for the behavior of a range of parenthetical expressions, (b) shows neither (1) nor (2) to be compulsory. The refined account also motivates a development of the "scorekeeping model" of conversation, helps to isolate a distinction between illocutionary force and illocutionary commitment, and reveals one precise respect in which meaning is only explicable in terms of use.

Attitude Ascription's Affinity to Measurement The relation between two systems of attitude ascription that capture all the empirically significant aspects of an agent's thought and speech may be analogous to that between two systems of magnitude ascription that are equivalent relative to a transformation of scale. If so, just as an object's weighing eight pounds doesn't relate that object to the number eight (for a different but equally good scale would use a different number), similarly an agent's believing that P needn't relate her to P (for a different but equally adequate interpretive scheme could use a different proposition). In either case the only reality picked out by any system of ascription is what is common to all equivalent rivals. By emphasizing some contrasts between decision theory and belief-desire psychology, it is argued that if attitude ascription is appropriately analogous to measurement then not only is being related to a proposition an artifact of the system of representation chosen, so are belief and desire.