Naturalism without a subject: Huw Price’s pragmatism

ABSTRACT

Huw Price has developed versions of naturalism and anti-representationalism to create a distinctive brand of pragmatism. ‘Subject naturalism’ focuses on what science says about human beings and the function of our linguistic practices, as opposed to orthodox contemporary naturalism’s privileging of the ontology of the natural sciences. Price’s anti-representationalism rejects the view that what makes utterances contentful is their representing reality. Together, they are to help us avoid metaphysical ‘placement problems’: how e.g. mind, meaning, and morality fit into the natural world. By combining subject naturalism and his own ‘global’ version of expressivism with Robert Brandom’s inferentialism about content, Price proposes a pragmatist ‘anthropology’ as a replacement for substantively metaphysical approaches to placement problems. In this paper I argue that Price’s project cannot succeed, and that this shows something important about what form pragmatism ought to take. Price’s view doesn’t work because no subject naturalist vocabulary is sufficient to describe any assertional practice; there is no way to connect his expressive-functionalist explanations to the practices and concepts which are their subject – nor, even, to the human subjects who are the focus of a philosophical anthropology. I close by suggesting how we might improve on these shortcomings of Price’s pragmatism.

KEYWORDS:

• Pragmatism
• content
• Huw Price
• naturalism
• Robert Brandom

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Many dissenters from contemporary naturalism are not anti-naturalists, but reformers hoping to improve (as they see it) naturalism. See e.g. (De Caro and Macarthur 2004) and (2010).

2 Of varying kinds, from eschewing ‘representation’ entirely (Rorty), reconstructing representation in inferentialist terms (Brandom), or constructing a pragmatic account of representation itself (Peirce’s Semiotic). The commonality is a rejection of, roughly, a Cartesian notion of the relation of mind to world as between ‘internal’ representations of an ‘external’ world.

3 Pragmatism’s relation to metaphysics is contested. Should we reconstruct metaphysics, or reject it entirely? I won’t address this here, but I think pragmatism rejects metaphysics as traditionally conceived. A pragmatist argument for the necessity of metaphysics via a critique of Price is (Legg and Giladi 2018).

4 Or is it neo-pragmatism? I won’t hash out terminology here or judge who is or isn’t a pragmatist. ‘Pragmatism’ is both a historical philosophical movement and a set of ideas related by family resemblance with or direct influence from that movement. For me ‘pragmatism’ includes everyone from Peirce and Dewey to Price and Cheryl Misak.

5 Cf. (Price and Macarthur 2007, 230f.).

6 An idiom of Brandom’s (2008, 12). See §2.5 below.

7 Contrast other pragmatist critiques of Price: Heney’s (2015) Peircean criticism of "Truth as convenient friction" (Price 2011, 163-83), Brandom’s critique of his global anti-representationalism (Brandom 2013), and Legg and Giladi’s (2018) critique of his rejection of metaphysics. A recent non-pragmatist critique of Price’s view is (Knowles 2017).

8 Similarly, Jackson (1998) calls them ‘location problems’.

9 A concept from Sellars's "Philosophy and the Scientific Image of Man" (Sellars 1963, 1-40). The scientific image is the world as construed by the natural sciences, of physical (chemical, biological) objects behaving according to natural laws. This image is very different from our ‘Manifest Image’ of ourselves and our world as rich with things like furniture, persons, rights and wrongs, economies, etc.

10 If there are no such truth-makers or realizers, this calls for expressivism, fictionalism, or eliminativism about that concept. See Jackson (1998).

11 Often written with an upper-case ‘R’ to denote its status a substantive theory, I won’t follow this convention, taking it that the word’s reference to a substantive theoretical view is captured by its ‘-ism’ suffix.

12 I won’t here justify the pragmatist stance, but take it on board to see if Price’s version can do what he hopes.

13 One of the first steps Price takes is arguing that subject naturalism is theoretically prior to object naturalism, and that the latter needs validation from the former perspective, which it can’t get (2013, 6–15). As before, I will just take this on board as part of trying to see if the overall project can work.

14 The Pricean and Brandomian pragmatist adopts the view common to Dewey, Wittgenstein, and Sellars that in general, to understand a concept is to have mastered the use of a word in the appropriate language-game(s).

15 Normative language is licit in functional explanations, of how something ought to normally work, as defined by its function. E.g., a heart ought to pump blood, not because it is subject to a rule it ought to follow, but because when a heart doesn’t pump blood it is not working properly. This functional sense of normativity is not the kind of normativity the naturalist excludes, since we have already gotten a naturalistic grip on it via selection effects, such as evolution. The excluded kind of normativity is the categorical, rule-governed kind, which is notoriously naturalistically recalcitrant (efforts to explain the latter in terms of the former are ongoing, if as yet unsuccessful). Daniel Dennett helpfully calls this distinction that between instrumental normativity and social normativity (20177).

16 By ‘intentional’ here I mean relating to intentionality—also notoriously naturalistically recalcitrant.

17 Price does consider the possibility that ‘the human component [of explanations of why we use certain vocabularies] might be entirely pruned away … and [we would] be left with a bare description of nature’ (2011, 30). Although he thinks ‘[in] practice … that this limit is out of reach—that the contribution on our side never goes to zero’ (ibid.), this is not because of naturalism’s limitations, but rather a limitation imposed on our judgements by our contingent dispositions to ‘go on in the same way’ in the same particular way. So, it is dispositional contingency (they might have been different had we evolved differently, say) that prevents a bare description of nature, not the fact that a bare description of nature might leave out something distinctively human beyond the narrowly natural. This suggests to me that the restrictions on subject naturalist vocabularies I describe and critique in the main text are inherent in how Price conceives naturalism. See §3.3, below, and cf. (Beasley 2015).

18 I use ‘narrow’ as a proleptic suggestion that perhaps a ‘broader’ but still naturalistic vocabulary is possible.

19 For the argument for taking expressivism global, see (Price and Macarthur 2007) and (Price 2019, 143–148).

20 See (Brandom 1994). A sort of Reader’s Digest version is (Brandom 2000), and a brisk article-length summary is (Brandom 2010).

21 A paradigm example of the ‘sparer’ view is Dretske's causal-informational semantics (Dretske 1981). I believe the most promising version of this sort of view belongs to Ruth Millikan; however, though it is more 'robust', it is still ultimately causal-functional in nature (1984); (2017).

22 See (Price 2011, 261 n.5), (2013, 176), and (2019, 134). He has called the view ‘non-cognitivism’ (Price 2011, 112) or ‘non-factualism’ (Price 1988), but now sticks with ‘expressivism’.

23 There’s still a sense in which what we do and say is connected to our environment. Price distinguishes between two types of ‘representation’ (in a theoretically ‘light’ sense): ‘i-representations’, the conceptual or propositional content of our linguistic utterances, and ‘e-representations’, which are tracking, indicating, and feedback relations that causally link us and (some of) our utterances to the environment (2013, 36). For Price, distinguishing these is seeing how content and correspondence come apart without giving up the idea that a lot, but not all, of what we talk about is responsive to what is ‘out there’ in our environment.

24 Though as a naturalist perhaps he wouldn’t recognize a distinction between ‘philosophical’ anthropology and anthropology per se, except maybe in terms of methods and the problems each focuses on. I take it he’d think that here we have a continuum, not a sharp divide. Cf. (Price 2011, 11; 30) and (2013, 181ff.); see also §2.2 above.

25 See (Price 2013, 48ff.), (2011, 188; 199; 208).

26 Following Ryle (1950) and Sellars (1957). See also (Brandom 2015), and Michael Williams (2013, 128–144).

27 Cf. the genealogies of ‘knowledge’ by Edward Craig (1990) and of ‘truth’ by Bernard Williams (2002).

28 In Brandom’s technical terms, such a metavocabulary V_X is a vocabulary ‘VP-sufficient’ (that is, a Vocabulary that describes a Practice) to specify a practice P_Y, where P_Y is ‘PV-sufficient’ (that is, a Practice in which a Vocabulary is used) to deploy the vocabulary V_Y (2008, 9ff.). For ease of exposition, I won’t use these terms here, using wordier but slightly less technical phrases to say the same thing.

29 Emphasis in original. In response to Brandom (2008, 12), Price agrees that the two formulations are equivalent (2019, 146 n.27).

30 By the performances being subject to evaluation in the light of the rule, or by the rule being the thing participants in the practice are entitled to cite for justification.

31 This breathless summary is indebted to (Brandom 2010) and (Peregrin 2014).

32 Price commits to this when he says that such stances determine appropriateness conditions for assertions while also adopting Brandom’s account of assertion, which gives appropriateness (normativity) centre-stage in the determination of content. Can Price just reject this aspect of Brandom’s view? No: he needs a non-representationalist account of content in order to avoid the problems with his early, ‘flat’ version of global expressivism. So he is committed to whatever makes inferentialism work as an account of the uniformity of assertion which overlays our practices’ functional pluralism.

33 The texts on rule-following most helpful to me (not to say I agree with them all) are: Baker and Hacker (1984, 2009); Boghossian (1989); Brandom (1994); (Hattiangadi 2007); (Hymers 2009); (Kripke 1982), and (McDowell 1998).

34 A reviewer asked: Wouldn’t this impugn any naturalist argument whatsoever, and isn’t this too great a claim? Whether or not it would do so would need to be demonstrated; at the moment, I claim only that this argument pertains to Price’s version of subject naturalism pragmatism—which is all my argument in this paper requires, since I’m focused on whether Price’s view is a workable version of pragmatism.

35 This also undermines Price’s criticism of Brandom on this score (Price 2011, 315-21). However, Brandom’s account may itself have a weak spot here in its use of only normative, but not intentional, vocabulary; see (Dennett 2010; Hattiangadi 2003; Rödl 2010). If so, however, it is not beyond shoring-up—something I hope to address in other work.

36 (Williams 2013) is admirably clear on this distinction.

37 See e.g. (Dewey [1938] 1986, 374) and (Brandom 2011, 26).

38 First suggested by McDowell (2004) and developed by others; see De Caro and Macarthur (2004, 2010) and Macarthur (2004).

39 The category ‘natural non-scientific’ is from (Macarthur 2004). A caveat: it is not clear that the orthodox naturalist’s restriction of what counts as ‘science’ to, basically, the natural sciences (or even just physics) is a legitimate restriction that the liberal naturalist will ultimately want to accept, thus calling into question the invidious distinction between ‘scientific’ and ‘non-scientific’ implied by ‘natural non-scientific’. For the present, this need not concern us; however, I have taken up this issue in other work (Beasley 2019).

40 One avenue to explore is the naturalism of the early pragmatists, specifically Dewey (although some disagree that Dewey’s naturalism plays well with McDowell’s; see [Godfrey-Smith 2010] and [Welchman 2008]). Price himself argues that McDowell’s liberal naturalism faces a dilemma: either accept that a non-metaphysical ‘external view’ on linguistic practices is possible (the subject naturalist view) or be saddled with an extreme form of quietism even McDowell would reject (2015). If my argument in this paper is correct, then Price’s argument against McDowell fails, since that argument relies on the idea that such an ‘external view’ is even possible, which, if I’m right, it isn’t.

41 Though in a nuanced way; see (Price 2013, 191–194), and (Price 2011, 319ff.).

42 Here again the issue of the relation between function and content arises—although it makes sense that the need for coordination among hominids could explain the emergence of assertion-practices, that’s different from saying what assertions are insofar as they are contentful utterances. Thanks to Carl Sachs for helping me clarify this.

43 Brandom sketches out how this might go (2008, 179), but whether it avoids the problems I impute to Rorty and Price is an open question I take up elsewhere. Regardless, he is clear about rejecting global anti-representationalism, preferring to pragmatically reconstruct the representational notions of ordinary descriptive discourse (Brandom 2013, 85–111).

44 I’d like to thank Bob Brandom, Diana Heney, Henry Jackman, Oliver Lean, Mark Migotti, Huw Price, and Carl Sachs for conversations about the ideas presented here; and especially Bob, Oliver, Mark, Carl, and anonymous referees for their very helpful comments on previous versions. Thanks also to audiences at the 2017 Canadian Philosophical Association Congress and the University of Calgary Department of Philosophy. This research was supported by the Social Sciences and Humanities Research Council of Canada.

Additional information

Funding

This work was supported by Social Sciences and Humanities Research Council of Canada.