Sensitive Semantics Nothing has had a greater impact on 20th-century analytic philosophy than Freges Sense/Reference (Sinn/Bedeuting) distinction.

What has received less attention1, no doubt at least in part due to Freges own disdain for natural language (Sprache), is his distinction between that of sinn (the context-neutral element of a sentence which orbits its referential axis) and kraft (that which determines the contextual significance of that sentences being uttered; the assertoric, imperative, interrogative and optative force which captures the sentences mood). Whilst not wishing to sanction such a divide in principle I dont believe it works I do wish to sanction it in practice. First, as it serves as a useful tool for dividing two approaches to the subject (there are those that endorse it; those that dont); second, because it raises interesting questions about the subjects nature and future, for if Michael Dummett is right, without it, philosophy2 has no future, at least not in any systematic sense3. In essence, this dissertation can be seen as an attempt to rebut Dummetts claim. That a) analytic philosophy has a future without the sense/force distinction and b) the distinction on its own is no guarantee of future progress, entails that the division, so conceived, provides neither necessary, nor sufficient grounds for the execution of the kind of project Dummett has in mind and so should be jettisoned. The case will be argued on contextualist grounds. I believe it to be both novel and important. Contextualism, it will be argued, at least in its contemporary semantic form, has been most clearly and radically articulated by Charles Travis. Nothing, however, as far as the following polemic goes turns on this; it is enough that he is held to be the villain of the piece by the relevant parties4. That the current constellation of positions that endorse a context-neutral semantics of one form or another5 renders his view unpopular, however, affords no grounds for its being discarded for we are not, as Austin would put it, playing old Harry with the underdog; the devil is in the detail alone. It will be argued, therefore, that whilst numerous objections both can and should be raised for the case against, an abandonment of contextualist principles would constitute, at best, a pyrrhic victory (or loss, as the case may well be). As criticisms of this position are developed, we will put more flesh on the bones of the view, or better, views of Travis. For the time being, let us start with Austin: It seems to be fairly generally realized (sic.) nowadays that if you just take a bunch of sentences...impeccably formulated in some language or other, there can be no question of sorting them out into those that are true and those that are false; for (leaving out of account so called analytic sentences) the question of truth and falsehood does not turn only on what a sentence is, nor yet on what it means, but on,

1 2

Dummett asides! Need to point out that this might refer to analytic philosophy, or philosophy tout court 3 Reference Dummett 4 Include protagonists here Charles and Lepore, Jerry Fodor and Tim Williamson 5 Reference C&L, Davidson, minimalist & deflationist accounts Horwich etc.

speaking very broadly, the circumstances in which it is uttered. Sentences are not as such either true or false. (Austin 1962b, pp.110-111 quoted in Travis 2008 p.26) Travis grants Austin proved badly mistaken about the first point. Contemporary analytic philosophy, he thinks, has taken a recent turn for the worse. We are told, nevertheless, that he was right about the rest, and that rest constitutes the core idea of occasion-sensitivity (Ibid): viz that the illocutionary force of an utterance is determined by context alone, and that this holds the key to determining that which is said by whom it is said of at the time of their saying it (simpliciter, truth conditions). Before continuing, let us first make clear both what is and is not at issue here: to wit, those indexical sophistries the correct treatment of which both flexible and inflexible semanticists would not contest, at least not in any significant way. That I am here is meaningful if and only if the corresponding demands of modal saturation (most notably time, place and identity) are satisfied is trivial, handled both sensitively and insensitively in a convergent manner789. Once token reflexivity is thrown into the medley, however, thought is tied less loosely to the linguistic forms within which it is expressed: that the same thought manifests itself within different sentential contexts10 combined with the fact that isomorphic sentential expressions can express a multiplicity of thoughts11 is substantial and allows for an unshadowing of mentalese1213 which its antagonists cannot account for. If adaptability is held to be a virtue, evolutionists notwithstanding, then clearly contextualism is not without its charm. What is it, outside of context, for an object to be blue14? What is it, for that matter, for an object to be an object15? Earlier on, reference was made to contextualist semantics. In fact the position we find ourselves in, should we take seriously the truths of which Travis so speaks, extends beyond this. Occasion sensitivity, taking the former point as given, places extensive demands upon, not just analytic philosophy, but also metaphysics, epistemology and cognitive science16, the likes of which cannot be met by much of the current work being undertaken in those fields. That the world is not ontologically self-sufficient (it requires context to do some work on its behalf); that expanding and contracting contextual modalities force a reconsideration of our basic conception of knowledge (post-Gettier); that the characteristic lack of systematicity
6 7

Need to reference Reference Charles and Lepore leads to problems with their account (ultimately a moderate contextualism rather than a semantic minimalism 8 Insert Travis quote (pp2-3 OS) 9 Davidson quote (there was an event (performed by Jones (at midnight) that was the buttering of the toast)) [symbolise] is this what we really mean?! 10 Insert example 11 Insert quote from Travis pigs grunting 12 Of course in ruling out both rigid and flaccid identity designation and so too systematicity 13 Need to add that antagonists might accept one side of the dichotomy, but only occasion sensitivity endorses both hence unsystematic 14 Rerference Travis? 15 Reference Austin S&S 16 One could of course go on here

vital to semantic flexibility both violates and invalidates the principle of compositionality required to run both computational and connectionist models of mind (on at least two counts) and more generally, that the inaccessibility of a context-neutral semantics vitiates against the possibility of devising a coherent, deflationist or minimalist account of truth17 (parochialism, at the risk of equivocation, is anything but parochial) all pose a threat to much, if not most, of what is currently at stake in contemporary philosophy. But all of this, of course, requires a level of demonstration that would be beyond the modest scope of this proposal. Such points, nevertheless, will be touched upon as the work progresses. There is a second, yet adjacent issue at play here concerning the possibilia of progression in philosophy; more precisely, how one moves forward in the subject given the absence of a system which proceeds according to generally agreed methods of enquiry [of a special sort]18 (TOE p. 455). I have hinted at and will develop one way in which such an obligation may be met; one that both acknowledges Dummets general attitude that the philosophy of language is the foundation of the rest of philosophy19 since it is only by the analysis of language that we can analyse thought (TOE p. 442) but resists his inference that it is difficult to see how a systematic theory of meaning for a language is possible without acknowledging the distinction between sense and force20. (TOE p. 450). Such a conception might comprise a third way of handling progress21 which evades the constraints a commitment to Dummetts general principle obliges whilst still preserving a progression of sorts. Whilst not explicit in Travis there is, nevertheless, an implicit global commitment to the view that occasion sensitivity pervades all areas of the discourse22. That a) penumbral indeterminacy and a rejection of sharp cut-offs present in Aristotles Condition (et al), insofar as it precipitates a reconsideration of what we might call true; b) the extent to which the expansion and contraction of modal possibilia 23 impinges upon the epistemic conditions which legislate over factive meaning (bleats meaning sheep and the possibility of ringer situations - A Sense of Occasion et al); c) the unshadowing of thought militates against the compositional account of cognition required to construct a representational model of mind (and the subsequent productivity and systematicity of thought which regiment our conceptual repertoire and so manufacture the modal infinities of mentalese); but perhaps most importantly, and simply, that d) meaning, in the traditional sense of the word, is far more sensitive to occasion than we might otherwise imagine, indicates that the causal influence exerted by radical contextualism is pervasive. As to the question of where next?, as we shall see, far from being thwarted, the contextualist practitioner is actually spoilt for choice.

17 18

Insert Crispin Wright quote to show that the two are not synonymous Which will be de 19 See note 2 of Grahamsit article WSUP 20 For without systemacity, there can be, as we shall see, for Dummett, no progress 21 In relation to Dummets two conceptions of progress and systematicity (CAP) 22 st Reference Travis article on Philosophy in the 21 Century 23 See for example Millar p. 342

The second sense concerns Wittgensteins conception of philosophy; most notably the view that philosophy is therapeutic through and through24. For if the role of philosopher consists in nothing more than shewing the fly the way out of the fly-bottle25 what form of progress, if any, can be envisioned beyond the clarification of characteristic kinds of intellectual confusion the only remedy for which is extended and patient treatment (TOE p. 439)? Paradoxically, as I hope to show, Dummett argues that Wittgenstein offers a more robust account than Austin, whose philosophy is not a therapy, but an empirical study [of particular words and their particular uses] incapable of systematisation (TOE p. 440). It will be argued that this is perhaps not so, but that Wittgensteins contextualism does offer us a future of sorts and that progress in philosophy more of a kind of work on oneself (PO p. 161), whilst slow, can nevertheless be significant. I see no reason to view the above two positions as divergent; proceeding in different directions, perhaps, but ultimately applying the same method of criteria. Much of the above exposition, however, turns on earlier arguments presented by Michael Dummett, and it is to these we now turn.

24 25

reference reference