From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

1. Introduction

Timothy Williamson has recently proclaimed that the linguistic turn in philosophy has come to the end – that “fewer and fewer of those who accepted the label ‘analytic philosophy’ for their work would also claim to take the linguistic turn” (Williamson 2004: 107). Peter Hacker has answered by claiming that what Williamson understands by the phrase ‘linguistic turn’ is misguided, and as far as analytic philosophy has passed the turn, it would be to best to get “back again”, because it is what philosophy is good for (Hacker 2007: 125). Brian Leiter has identified the current American philosophy to be divided into two camps that, we can see, reflects the meta-philosophical tension between Williamson and Hacker. “Quinean naturalists”, Leiter says, are constructive philosophers who believe that the problems of philosophy are “real”, while “Wittgensteinian quietists” dissolve philosophical problems, rather than solve them (Leiter 2004: 2-3). Richard Rorty has commented the issue – in a paper contributed to this very occasion three years ago – by marking, and clearly echoing Leiter’s division, that “people who call themselves “naturalists” typically see little value in Wittgenstein’s work” and “doubt that what Gustav Bergmann dubbed “the linguistic turn” was a good idea” (Rorty 2007: 160-1).

In what follows, I will discuss the ‘Williamson/Hacker-controversy’ over the linguistic turn from the light of Rortian pragmatism. Rorty, as we know, has his own contribution in the history of the linguistic turn – if not more, but editing and naming the book (Rorty 1967) from which the very notion of the ‘linguistic turn’ derived its widespread use. Even it might look that Rorty’s position on this debate is easy to locate, the matter actually is more complicated if a more closer look is taken to it.¹ I will focus on one particular feature in this controversy: the idea of representation; more precisely: (1) how the linguistic turn is related to the issue of representation over-all, and (b) what is the relation between mental and linguistic representation.

In the first part of my paper I will critically discuss Williamson’s views, and there one could see how a Rortian is teamed up with Hacker in this issue. In the second part I will discuss Rorty’s views relation to Hacker’s ideas, and here we, perhaps surprisingly, will find agreements with Williamson. However, it needs to be noted that a Rortian position I here reconstruct is motivated by Rorty’s writings – these are views that I think Rorty should maintain, but I am not sure if he really did (in the case I got Rorty wrong). Rorty’s own stance towards the linguistic turn, and of its merits, is not easy to grasp, even though it is fair to say that he seemed to hold quite stabile view since the mid-seventies.

2. The linguistic turn and the quest for representationalism

The central idea that Williamson promotes in his paper “Past Linguistic Turn?” is the idea that if there was any merit in the linguistic turn that was, by taking the theme of language as the prima concern of philosophy, helping to recognize the representational nature of human thinking and language. More concretively, it would mean that the topic of intentionality and aboutness is to be stressed. However, the gain of this approach was not profoundly seen until the topic of mental representation was arisen within the current philosophy of mind. So Williamson concludes:

One might therefore classify both thought and language together under the more general category of representation, and argue that the linguistic turn was just the first phase of the representational turn, on which the goal of philosophy is the analysis (in a generous sense) of ‘representation’. (Williamson 2004: 108)

However, the model of the linguistic turn Williamson has in his mind is quite recent one, famously articulated by Michael Dummett. The central idea of it is “the proper method of analysing of thought consists in the analysis of language” (Dummett 1978: 458). Even though Dummett attributes it to Frege, and thereby reads him as the founder of analytic philosophy, it is highly controversial if the case is more of projection of Dummett’s own position than Frege’s own.

To any person who has read Rorty’s Philosophy and the Mirror of Nature (1979) the point Williamson here addresses, by treating the topic of ‘mental representation’ almost as a novelty in recent philosophy of mind, hits quite hard; it cannot be more in opposition the very moral of Rorty’s book². The whole point of the book is to see the topic of mental representation as the core problem of the modern philosophy, and the linguistic turn is just a fairly recent 20^th centurion move within the scheme of this bigger agenda. In Rorty’s metaphorical terminology, a language is a small ‘Mirror of Nature’ whereas the mind is a big one. Even though Rorty did not thought anymore by the time of writing the book (contra to the optimism of Rorty 1967) that the linguistic turn was a meta-philosophical progress – analysing the language or using linguistic methods would not get the problems solved any better – it was still a turn worth making. Namely, by focusing into language – by seeing the old problems as problems concerning language – they could be more easily to over-come. According to Rorty, the “pragmatization” or “naturalization” of the very notion of language within the course of analytic philosophy – by late-Wittgenstein, Quine, Sellars, Davidson, and, finally, Brandom – had the result that the tradition was working for “self-destruction”. To make a long story short, one could say that the very profit of the linguistic turn, as Rorty sees the case, was to get rid of the whole idea of representationalism.

It should be noted that there is a sense in the notion of representationalism that seems to separate Williamson’s discussion from the one Rorty does in Philosophy and the Mirror of Nature. Williamson is not worried, or even motivated, by the existence of epistemological sceptic, but sees representationalism, and the problematic concerning it, in home with “the revival of metaphysical theorizing, realist in spirit” (Williamson 2004: 111). Starting from post-Kripkean philosophy he seems to be believe that “throwback to pre-Kantian metaphysics” seems to not just passing the linguistic turn but – to use Rorty’s expressive term – whole ‘Kantian-epistemologically-centred philosophy’. Williamson remarks that the phase is over and appeals to Kant’s authority “ring hollow, for they are unbacked by any argument that has withstood the test of recent time” (ibid: 111). Hacker, on the other hand, accuses Williamson of historical blindness, and offers a ‘historical turn’ for analytic philosophers like Williamson to take – for example, Kant would have much worthwhile to say about how thinking alone can yield knowledge of the reality – the presupposition Williamson takes for granted, by seeing logic and mathematics offering evidence that it does (see Williamson 2004: 127). However, the latter is debatable as it is based on the account we have of the nature of logical and mathematical knowledge (Hacker 2007: 135). Hacker, like Williamson, does link the linguistic turn with Kantian legacy, but makes a contrary conclusion: it would be a drawback to ignore the legacy of Kantian critical philosophy.

3. Mental vs. linguistic representation

As far as the topic of representation goes, Hacker is keen on showing that the very idea of mental representation is an incoherent notion. Here we can hear a pure Wittgensteinian speaking: a thought or a belief does not represent anything because by definition representation is a medium that needs non-representational properties in virtue of which it can represent what it does (such as written word has visual properties or a spoken word aural properties). The mental contexts are “all message, and no medium” (ibid: 139). Only language can be representational, and thinking only insofar as it is to be seen as language-like – and even the latter is a mistaken idea.

So it is noteworthy that even though Hacker raises doubts about the representational feature of thinking, this is not at all the case with language. Whereas Rorty sees the gain of the linguistic turn to lead – via the step of transforming the issues concerning thinking into those of language and then changing our account of the latter – into antirepresentationalism, Hacker sees it as straightening out the misguided idea of thoughts as representational into its proper place – the view, as we can see, he shares with Williamson (who, in fact, does not even recognize the theme of representation prior the linguistic turn). What is for Rorty a co-incidental shift from Big Mirror into a smaller one, is for Hacker a problem finally properly and thoroughly constructed. As far as thinking was viewed to be representational prior Wittgenstein, that was, one could conclude, a mistake. The point of Hacker’s argument against mental representationalism is to stress the relevance of priority of language as far as representation goes, while in the case of Williamson the linguistic feature seems to be a sub-issue in a larger issue of representation, and in the preferable order of the studies, secondary into that of mental representation.

Hacker’s meta-philosophical account of the issue of representation is different. For him, it is not a goal of philosophy – the problem of philosophy an sich to be solved – as Williamson sees it; the goal is, according to Hacker, that “both the concept of language and the concepts of thought and thinking are sources of confusion and hence fit for subjects of philosophical reflection”. So he concludes, by provoking the lasting or re-recognizable value of the linguistic turn:

The aim of philosophy is the clarification of the forms of sense, in one way or another, are conceptually puzzling […] the charge of philosophy[…] is the extirpation of nonsense. […] The prize is […] a proper understanding of the structure and articulations of our conceptual scheme, and disentangling the conceptual confusions. (Hacker 2007: 140).

From a Rortian point of view, what Hacker is trying here to do is to find a difference where there is not one. For Rorty, Williamson is consistent and in a right track by recognizing the similar function of the same phenomenon; the same philosophical problematic is to be drawn in respect to both mental and linguistic contexts. It is just the views Williamson and Rorty hold of the matter are in total opposition to each other – it is one’s representationalism against other’s antirepresentationalism. The problem with Hacker’s view, and seemingly with each of “quietist Wittgensteinians”, who Rorty (2007: 161) prefers to call “Wittgensteinian therapists” ³, contra to “pragmatic Wittgensteinians” like him, is the appeal to the category of “nonsense”. The argument to back up this is based on the pragmatic idea that “there is nothing interesting to be said about the distinction between sense and nonsense”, because “if we adopt the social-practice view of language, there seems no way to reconstruct the relevant idea of ‘confusion’” (Rorty 2007: 170). However, a Hackerian could claim that the problem with a Rortian position is that, even though Rorty personally gives such a light weight into it, it leaves one unarmed to defend oneself against the ‘non-critical’ movements – there is no such criteria by which we could illegalize the status of Post-Kripkean metaphysical school, applauded by Williamson. Rorty recognizes Kantian origin in the talk of nonsense: it was Kant’s originality “to erect general theory about proper and improper use of concepts”, and thereby, he laid a “general theory of representation” (ibid: 170). The effect of the linguistic turn (being a theory of not of representation in general, but of linguistic representation) is that “nonsense” became “a term of philosophical art” (ibid: 171). As we saw, Hacker is explicit of this continuity between Kantian critical philosophy and the linguistic turn. It is just he and Rorty have opposite views of its outcome.

4. Closing Remarks

Over-all, I think what is noteworthy that, no matter which side we take on the issue, all of the contributions of Williamson, Hacker and Rorty do address in their own particular way the meta-philosophical significance of the linguistic turn (even though one of the participants of the discussion seemingly seem approach it quite lighten-heartedly, absorbing the anti-historical, arrogant style, sometimes typical to best analytic philosophy). The issue in hand is very much the Kantian legacy of critical philosophy. One way is to pretend like the criticism never happened; the whole Kantian phase was a side walk from the traditional ‘secure path of science’, that is, philosophy-as-metaphysics, but luckily we are now back in a right track. This is very much presupposed in Post-Quinean naturalism with its Post-Kripkean metaphysical spirit. The other way is to keep the Kantian spirit alive and retain the Kantian critical self-image of philosophy. This is to keep the flag of the linguistic turn on. This kind of spirit is to be found among not just Oxfordian ‘Wittgensteinians’ but Pittsbourghian ‘Neo-Hegelians’ as well. The third is to shift radically the self-image of philosophy, to wholeheartedly forget the pretension of the ‘secure path of science’. This, as we know, has been Rorty’s explicit theme and which, as we also know, be it good or not, did not gather much applauds from his colleagues, at least within the analytic philosophy.