Epistemology is the attempt to legitimate a philosophical domain, which Rorty believes we can no longer maintain. With the advent of the new science of the sixteenth and seventeenth centuries, philosophy’s role, in terms of access to theoretical knowledge about the world, was displaced. In addition to providing knowledge based on empirical experience and observed fact, new science also carried apparently uncontroversial norms of progress. It thus represented a serious legitimation challenge to the formerly uncontested domain of philosophical reflection. Rorty argues that Cartesian epistemology was tailor-made to meet this challenge, arguing the line that doubts can be raised about any empirical claim whatsoever as well as claiming that doubt cannot be alleviated by experience. Thus, the philosophical domain of epistemology was preserved. In order to get beyond this conception, with its concomitant consequences for knowledge of the world and other people, Rorty believes that we have to break the picture of the mind as a mirror of reality once and for all.
The core of the argument behind representational epistemology is that vocabulary is optional and mutable. Rorty sets out to show that this si not the case. His primary challenge and arguments are against mirroring through an extension of arguments from Sellars, Quine, Kuhn, Wittgenstein and Davidson towards a general critique of the concept of the mind inherited from seventeenth and eighteenth century philosophy. Kant is a specific target for Rorty, for it is through the Kantian picture of concepts and intuitions ‘getting together to produce knowledge’ that the idea of epistemology is confirmed as a specifically philosophical endeavour. If we do not have a distinction between what is contingent and what is necessary then, Rorty claims, ‘we will not know what would count as a “rational reconstruction” of our knowledge. We will not know what epistemology’s goal or method could be’.
The Kantian concept of the mind works on the picture of a mind’s structure producing thoughts or representations through working on empirical content. These representations are judged according to how accurately they mirror reality.
Rorty combines Sellars and Quine in order to challenge the notion that epistemology is at the core of philosophy. He argues that neither philosopher took their own arguments to the logical limit, and so both end up attacking the same distinction: Quine from the position of anti-linguisticism (mental entities are replaced by notions of meaning or structure) and Sellars from the attack on the myth of the given. As Rorty notes, ‘Sellars and Quine invoke the same argument, one which bears equally against the given-versus-nongiven and the necessary-versus-contingent distinctions. The crucial premise of this argumet is that we understand knowledge when we understand the social justification of belief, and thus have no need to view it as accuracy of representation’.
The upshot of pursuing Sellars and Quine’s arguments is that we see knowledge as a ‘conversation…[a] social practice, rather than an attempt to mirror nature’. Rorty terms this ‘epistemological behaviourism’. In epistemological behaviourism, legitimation of our practices (and claims) is no longer achieved through reference to a set of context-transcendent standards, but through conversation. Relinquishing the limits of knowledge to what is purely conversational marks a point of departure between Rorty and many of his ‘friendly’ critics such as Putnam, McDowell and Dennett, who, would baulk at going so far though approve of Rorty’s historical scepticism towards the context-transcending ambitions of philosophy.
From his claim about the conversational standard of knowledge, Rorty has also drawn the charges of relativism and subjectivism. He defends his position in Truth and Progress, stating that his ‘strategy for escaping the self-referential difficulties into which “the Relativist” keeps getting himself is to move everything over from epistemology and metaphysics into cultural politics, from claims to knowledge and appeals to self-evidence to suggestions about what we should try’.
The difference between epistemological behaviourism and relativism or subjectivism is further demonstrated in light of Rorty’s criticism of the notion of representation. Both relativism and subjectivism are products of the representationalist paradigm. Rorty makes use of Davidson’s criticism of the scheme-content distinction and of the correspondence theory of truth in order to back up his rejection of any philosophical project that upholds distinctions between what is made and what is found, the subjective and the objective, appearance and reality. He does not deny that these distinctions do not have an application, but he maintains that the application is always bound by context and interests; as such, there is nothing useful or interesting to be said about truth in general.
Epistemological behaviourism is also distinct from a strand of idealism that asserts the primacy of language or thought over an unmediated world. This follows from his appropriation of Davdison’s theory of meaning. Conversationalism does not give priority to the subjective or the objective; it is rather the other side of his anti-representationalism, which denies that we are related to the world in anything other than causal terms. There is nothing useful that we can say with respect to the view that the world limits our ways of coping with it.
By attacking the notion that the world constrains rational agents’ thoughts and behaviours, Rorty has raised criticism from those who take the natural sciences as their primary reference point. The first claims that, by denying the chief process of science i.e. the effort to learn the truth of things by allowing ourselves to be constrained in our beliefs about the world, Rorty is denying the very idea of Science. The second, internal, criticism, tries to show that scientists would not be motivated to continue in their work if Rorty’s view of science were to prevail because it would cease to be the useful thing that Rorty thinks that it is.
Rorty’s relationship with natural science is more complex than it sometimes appears. He says, for example, that he tends to view natural science ‘as in the business of controlling and predicting things, and as largely useless for philosophical purposes’ and yet he spends a great deal of time elaborating on an alternative view of intellectual virtue that draws on the virtues embodied in good science. Good science here is not linked to better and better representations, but rather success is predicated on the model of rationality that scientific practice espouses, which leads to democratic exchange of view. In this sense, then, we can see how/why Rorty eschews science as philosophically significant.
Rorty is not denying that there are any uses at all of notions like truth, knowledge or objectivity. His point is that these notions always demonstrate particular features of their varying contexts of application. When we abstract from these contexts, we are left with hypostatisations, which are incapable of providing us with any guide to action at all. Thus, we do not have a concept of objective reality that can be invoked to explain the success of some set of norms of warrant, or to justify some set of standards over another.
The linking of truth with justification is perhaps Rorty’s clearest statement about a theory of truth. As late as 1982, he was still attempting to articulate a view of truth derived from James, namely that true is what is good or useful for us to believe. After this, following Davidson, he rejects all attempts to explicate truth in terms of other concepts. These days, Rorty’s position has evolved again; truth has various important uses, but it does not itself name a goal towards which we can strive, over and above warrant and justification. That is not to say that truth is reducible to warrant, but rather to say that there is nothing deep or substantial that we can say about the concept. We have only semantic explanations for why it is the case that a sentence is true when its truth conditions are satisfied. We have no measure for truth apart from increasing warrant, which for Rorty is a key element of why the concept is so useful. Like goodness, sentences can only ever by analytically certified as true by virtue of its possession of some other property.