Wittgenstein, Non-Factualism, and Deflationism

AbstractAmongst those views sometimes attributed to the later Wittgenstein are included both a deflationary theory of truth, as well as a non-factualism about certain regions of discourse. Evidence in favor of the former attribution, it is thought, can be found in Wittgenstein's apparent affirmation of the basic definitional equivalence of 'p' is true and p in §136 of his Philosophical Investigations. Evidence in favor of the latter attribution, it might then be presumed, can be found in the context of the so-called 'private language argument', wherein Wittgenstein provides an expressivist treatment of first-person present tense sensation utterances. In this paper, by contrast, I will argue that Wittgenstein's later philosophy is best understood as endorsing neither a non-factualism about sensation utterances, nor a deflationism about truth. Wittgenstein should instead be understood as offering a 'mixed' view of sensation utterances according to which some while not others are apt for expressivist treatment. Moreover, he should be thought of as identifying truth-conditions with semantic 'correctness-conditions', and thus truth with semantic 'assertibility'.Keywords: Wittgensteinnon-factualismdeflationarytruthassertibilityprivate language Notes1 See, e.g., Bolton, Citation1979: pp. 159–60; Dummett, Citation1978: pp. xxxiv, 317; Kripke, Citation1982: p. 86; McGinn, 1984, p. 71: p. 71; Hacker,Citation1996: p. 191; and Blackburn, Citation1998: p. 166. Of these, Hacker and McGinn each explicitly use the term 'deflationary,' while Dummett and Kripke use the term 'redundancy theory'. Blackburn uses 'deflationary' and 'redundancy' interchangeably. Bolton along with Hacker, further, employs the term 'Ramsiean', without making any distinction between this term and the term 'deflationary'; and McGinn uses the term 'redundancy' while similarly not making any such distinction. With a few exceptions (e.g., Vision: Citation2004, Citation2005), commentators are unlikely to make any distinction between the attribution of a 'redundancy' versus a 'deflationary' theory of truth, and nor is it typical to distinguish between what I call the 'equivalence thesis' and one or more of what I label 'deflationary theses'. For this reason, at this point I simply use the term 'deflationism' as a blanket term to cover each of the aforementioned attributions, although as the discussion progresses it will become clear that that term is in fact ambiguous between (at least) two different positions, one of which presents a potential inconsistency with non-factualism, and one of which does not.2 The charge of 'non-factualism' is most famously leveled against Wittgenstein by Kripke (Citation1982), but specifically in regards to the former's views about content ascriptions. Though I believe that many of the insights developed in this paper can be brought to bear against Kripke's reading, a robust critical treatment of that reading is unfortunately beyond the scope of this paper. In regards to Wittgenstein's views on sensation utterances, it is more common to characterize Wittgenstein as an 'expressivist,' e.g., Hacker (Citation1996: p. 132); Sluga (Citation1996: p. 340); Blackburn (1998: pp. 161–4); Fogelin (Citation1999: p. 170). Nevertheless, since expressivist treatments of other domains, such as ethics, are often thought to commit one to a non-factualist analysis of the relevant statements, the concern remains that Wittgenstein, in defending an expressivist treatment about first-person present tense sensation reports, will on those grounds be justly characterized as adhering to a non-factualism about substantial segments of psychological discourse. This could be viewed as problematical in light of Boghossian's (1990) arguments.3 Boghossian, 1990.4 James, Citation1907.5 Strawson, Citation1964.6 Quine, Citation1986.7 It is not completely obvious how, according to Boghossian's (1990) somewhat over simplified division of deflationary theories of truth into the 'performative' and 'disquotational' versions, one would categorize 'minimalist' versions such as that developed by Horwich (Citation1998), for instance, or 'prosentential' versions such as that defended by Brandom (Citation1994), amongst others. While Horwich's minimalism obviously bears some important resemblance to disquotational theories, it nevertheless differs in subtle but important ways, such as in taking truth to be not only a property (characterizable in terms of an equivalence schema and various of its derivable consequences), but also a property of propositions, where these are thought to be importantly distinct from sentences. Brandom's emphasis on the anaphoric role played by the truth-predicate, by contrast, makes it less obvious in which camp, if either, he belongs. However, while this may pose a complication for Boghossian, we need not pause to consider it in this context, since insofar as Wittgenstein is a deflationist (I shall argue that he is not), he is certainly neither a 'minimalist' (in Horwich's sense), nor a prosententialist. In the first case, this can be seen in amongst other things the prominence given by Horwich to 'propositions' as abstract objects importantly distinct from sentences (a view which Wittgenstein denies). In the second, it can be seen in the fact that, although there is no reason to suppose that the anaphoric role played by the truth-predicate could not reasonably be incorporated within a Wittgensteinian 'description' of its use and so meaning, he would not, it seems to me, reduce the focus of such a description to this particular role, and in any case puts no specific emphasis, within his later writings, upon this aspect of its usage.8 Boghossian, 1990: p. 164.9 Boghossian's charge of straightforward incompatibility between deflationism and non-factualism, it should be noted, is a controversial one, having been criticized by several able commentators (e.g., Soames (1999: p. 251–5, Bave (2009: p. 309–10)). Other authors, moreover, such as Divers and Miller (Citation1994), Wright (Citation1992), and Horwich (Citation1993), have developed arguments similar to Boghossian's, which exploit the idea that any 'minimalist' theory of truth implies the truth aptness of ethical statements. These results have however been disputed by Jackson, Oppy, and Smith (Citation1994), and Smith (Citation1994). Since it is not my present aim to assess the soundness of Boghossian's argument nor, a fortiori, to assess the seriousness of the purported inconsistency between ethical non-cognitivism and minimalism, any detailed discussion of this literature is unfortunately beyond the scope of this paper. As mentioned above, my goal is instead to explore the prima facie problem posed for Wittgenstein by the apparent incompatibility Boghossian alludes to, in an effort to shed interpretive light on his remarks on both sensation utterances, and on truth. If it turns out that there exists no straightforward incompatibility of the sort alluded to by Boghossian, this will not in any case undermine any of the central claims made in this paper, but will in fact simply diffuse the potential charge of inconsistency stemming from Boghossian's argument, by way of a shorter but far less illuminating route.10 Winch, Citation1983.11 See, e.g., Churchland, Citation1984, and Dennett, Citation1978.12 See, e.g., Mackie, Citation1977, and Ayer, Citation1952: p. 107.13 Notably, non-factualism as characterized by (1) differs from an error theoretic approach to a region of discourse, since according to the error theorist but not the non-factualist, predicates in the given region DO denote properties, thought those properties fail to be instantiated by anything (see Boghossian, 1990: p. 159).14 Boghossian offers a gloss on these conditions to the effect that, in order to qualify as truth –conditional on a deflationary theory, a sentence 'must admit of coherent embedding within negation, the conditional, and other connectives, and within contexts of propositional attitude' (1990: p. 163). However, there are good reasons to ignore this gloss on the conditions, and to focus instead on the conditions themselves. Specifically, the further requirements specified in the gloss are not in fact simply implicit in the conditions, but are supplemental to them. This can be seen in the fact that what the two conditions taken on their own require is that, in order to qualify as apt for embedding in the truth-functions, (that is be truth-conditional) a sentence must only be significant and declarative. The gloss on the conditions, by contrast, specifies that over and above being significant and declarative, a sentence must already admit of coherent embedding in the connectives in order to qualify as truth-apt. It cannot, for instance, be part of the position which Boghossian chides Ayer for maintaining to be consistent with deflationism (1990: p. 165), that ethical utterances, e.g., are coherently embeddable within truth functors, since qua expressivist, this is precisely what Ayer denies. Consequently, attempting to incorporate the aforementioned gloss on the two conditions within Boghossian's official characterization of deflationism, then, results in a complicating ambiguity between two distinct positions. Since the position which Boghossian in fact goes on to argue is inconsistent with non-factualism, however, is that according to which satisfaction of the two conditions, taken strictly on their own, are jointly sufficient for truth-conditionality, its best to treat his gloss on the two conditions as an unfortunate slip.15 The formulation is Richard Rorty's (Citation1986: p. 334).16 Boghossian, 1990: p. 165.17 Wittgenstein, 1961 6.53.18 Analogously, Read (Citation2007: Part III) has suggested that 'realism' versus 'anti-realism' (manifest in the context of Dummett's [Citation2000] reflections on time, for instance) amounts to a spurious debate based in a spurious conceptual distinction; one which can flourish, moreover, only in the context of a failure to adequately appreciate and apply Wittgenstein's novel therapeutic (as opposed to theoretical) approach to classical philosophical problems.19 Wittgenstein, 1958: §§290–91.20 .Wittgenstein, 1958: §36.21 Wittgenstein, 1958.22 This point is also emphasized by Baker (Citation1998: p. 333).23 Wittgenstein, 1958.24 Ibid.25 Wittgenstein, 1953: §§134–6.26 Vision (Citation2004: p. 131) develops a similar distinction between what he calls the 'semantic thesis' of deflationism, and its 'metaphysical thesis'. According to the semantic thesis: 'the left-hand and right hand sides of the bi-conditionals mean the same thing (viz., have the same cognitive content)', while according to the metaphysical thesis: '"Is true" does not express a property, or truth has no (substantial) nature, or "is true" is not a predicate, or …(the bi-conditionals) express all there is to TRUTH.' While I am in broad sympathy with Vision's drawing of these important distinctions between elements of deflationism, strictly our distinctions do not overlap, since, in light of important cases in which semantic content and cognitive content can come apart (e.g., proper names) I would not be prepared to formulate the 'semantic thesis' in precisely the way Vision formulates it.27 In other words, nothing Wittgenstein says here indicates how he would understand the use of the truth predicate in what Scott Soames calls 'environments of type 2' (1999: p. 230), that is 'cases in which we predicate truth of some proposition or set of propositions that we do not explicitly assert, display, or produce' (ibid.).28 A very similar point is made by Vision, Citation2004: p. 132.29 Winch, Citation1983: p. 401.30 E.g., Putnam, Citation1981.31 For a more robust conception of the sort of logical space in which we might locate Wittgenstein's views on truth, without succumbing either to a deflationary conception of truth, or assimilating these views with any competing alternative 'theory' of truth, see Diamond, Citation2003.32 Brandom, Citation2000: p. 57.33 Rodych, Citation2011.34 See Putnam, Citation1994, and Diamond, Citation2003, for more in-depth explanations, of how and why cases like these pose a problem for epistemic readings of 'assertibility'.35 Except perhaps in the uncontroversial sense that we in some cases make the facts what they are by e.g., building bridges in certain locations, undertaking certain career paths, getting married, etc. Wittgenstein (1981).36 Rodych, Citation2011.37 Wittgenstein (1996, §4).38 This is suggested by Jacobsen (Citation1997), for instance.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.