In contrast to the scorn which Rorty’s name now provokes in some quarters, it’s arresting to see the esteem in which he was held by Roy Bhaskar in the late 80s, albeit in the context of a trenchant philosophical critique. He commends Rorty’s “eloquent critique of the epistemological problematic” but intends to argue that Rorty remains captive to this problem field in ways he himself fails to recognise (Bhaskar 1989: 146). In doing so, he advocates a philosophical post-narcissism which is capable of elaborating “non-anthropocentric pictures of being” through taking Rorty’s project of ‘de-divinisation’ and pursuing it much further than Rorty was either willing or able to do (Bhaskar 1989: 147).
His initial target is Rorty’s account of science, particularly his easy imputation of chronic success in “the prediction and control of nature”. In this claim Rorty reveals himself to have accepted Hempelian assumptions about natural science, in effect committing himself to a basically positivist account. Much of Bhaskar’s critique proceeds from systematically exploring the ambiguities which are entailed by Rorty’s failure to distinguish between the intransitive (ontological) and transitive (epistemological) dimensions of science. Once we begin to draw this distinction, Rorty’s constant invocations of ‘redescription’ come to seem much more modest in their conclusions, though Rorty himself fails to recognise this:
Thus redescribing(td) the past in a revolution way can cause(id) radical new changes, including a new identity, self-definition or auto-biography: but it cannot retrospectively cause(id) old changes, alter the past (as distinct from its interpretation). It is not surprising that Rorty should slip from transitive to intransitive uses of terms like ’cause’ – it is endemic to empirical realism, the epistemological definition of being in terms of (a particular empiricist concept of) experience. (Bhaskar 1989: 152).
Bhaskar’s point is not to impute anti-realism to Rorty, though the latter surely does come to this in his later work. For Bhaskar “the crucial questions in philosophy are not whether to be a realist or an anti-realist, but what sort of realist to be (an empirical, conceptual transcendental or whatever realist); whether one explicitly theorises or merely implicitly secretes one’s realism and whether and how one decides, arrives at or absorbs one’s realism” (Bhaskar 1989: 153). Bhaskar is in agreement with Rorty’s repudiation of the ‘Archimedean point’ outside human history and the notion of ‘correspondence’ as standing between world and language. However he finds it problematic, as well as internally inconsistent, for Rorty’s realism to adopt such a whiggish approach to actually existing science – imputing a continual extension of our capacity to ‘control and intervene’ with one hand while bracketing the philosophy of science with another. He shares Rorty’s anti-foundationalism and applauds his “vigorous assault on its attendant ocular metaphors, mirror imagery and overseer conception of philosophy” (Bhaskar 1989: 157).
So what’s the problem? Rorty’s peculiarly positivistic stance finds expression in his assumption that an individual represents a closed system. Bhaskar addresses this point in a dense critique which I won’t attempt to summarise but is an astonishingly accomplished analysis which is worth studying in detail (Bhaskar 1989: 161-162). His attention is to better understand “A Tale of Two Rortys”: a tension which runs through his work and precludes him from offering either an adequate understanding of scientific activity or a sustainable account of human freedom. In essence he finds himself reproducing a linguistified version of the Kantian distinction between people as empirical selves and as moral agents. Rorty is attempting to combine a physicalism which sees individuals as closed causal systems, in which it is possible (in principle) to predict every movement of a person’s body by reference to microphysical states, with an affirmation of the discursive freedom of human beings.
However it is this freedom to ‘re-describe’ which is the cause of all the problems. He fails to distinguish between objects changing and requiring a new description and an unchanged object being redescribed. In this sense ‘redescription’ comes to be detached from the characteristics of the objects being redescribed. Yet this is central to Rorty’s account of human freedom:
Man is the describing, redescribing being. Among the entities man can describe in a new, and abnormal, way, is himself. By making a new, incommensurable description of herself ‘stick’, she makes it true; and thus ‘gives birth to’ (to use Harold Bloom’s term) or ‘creates’ herself – which is to say ‘overcomes’ her previous or past self. Moreover, only by describing herself in a totally novel way can she capture or express her idiosyncrasy, uniqueness – or rather achieve it, achieve her individuation – for anything else would reduce her to a (more or less complex set of formula(e), a token of a type (or set of types). Such radical self-redescription (which could be nicknamed ‘me-‘ or ‘we-‘ description) is the highest form of description. For not only does the redescription redescribe the describer; but in the process of redescription – of wining it, of making it stick, of achieving recognition for it – it makes the (re)description true; so achieving the identity of subject and object, by creating it. (Bhaskar 1989: 171)
On this picture we are left with a notion of freedom as “caprice, discourse, capricious discourse and creative discourse” (Bhaskar 1989: 173). Even this highest form of freedom within Rorty’s account, the possibility of ‘creative discourse’, falls short because it operationalises freedom in abstraction from the material dimension of social life. Rorty’s account makes it difficult to see how we could ever come to identify or transform structures which engender a diminution of human freedom. It also fails to recognise the constraining effects they may have on freedom even in his own narrow understanding of it. As Bhaskar observes, “it is now easy to see how the notion that ‘man is always free to choose new descriptions’ can encourage the voluntaristic position that man is always free to choose any description” (Bhaskar 1989: 176). Rorty’s discursive freedom should not be repudiated in and of itself but should rather be contextualised in terms of a much deeper sense of freedom and, crucially, a notion of emancipation which “depends upon the transformation of structures rather than just the amelioration of states of affairs” (Bhaskar 1989: 178).