The only direct manifestations of language are utterances and inscriptions, and it is we who imbue them with significance. So language is at best an abstraction, and cannot be a medium through which we take in the world nor an intermediary between us and reality. It is like a sense organ, an organisational feature of people which allows them to perceive things as objects with a location in a public space or time, or as events with causes and effects. (Davidson).
The fact of representation, be the vehicle thoughts, utterances or inscriptions, is taken to be central to our ability to posit true or accurate statements. That which is an accurate representation is said to be true, to correspond to the facts, to mirror nature. Davidson finds such talk unfortunate. For him, philosophy would be transformed if we were to set aside the traditional view of mental or linguistic representation, as many problems, such as scepticism, realism and antirealism, the subjective-objective distinction, would simply fall away.
Davidson attacks the idea of facts as facts in the world to which our thoughts, utterances etc correspond. Thus, if one is to give substance to the notion of representation, one must also be prepared to defend the claim that there are facts. As Davidson argues, ‘the correct objection to correspondence theories [of truth] is…that such theories fail to provide entries to which truth vehicles…can be said to correspond’. If there are no facts, then truth makers or objects of knowledge cannot function. Further, if there are no representations there is no sense to be made of cultural relativism, for example. Indeed, Davidson writes, ‘it is good to be rid of representations…for it is thinking that there are representations that engenders thoughts of relativism’. Relativism presupposes that a viable distinction can be made between that which is represented and the representation itself. For Davidson, this dualism between a ‘conceptual scheme’ and ‘empirical content’ is untenable, precisely because he believes that there are no distinct facts to which true utterances correspond.
For Rorty, Davidson has made it impossible to continue talk of the ‘problems of philosophy’ by undermining the scheme-content distinction: ‘if one gives up thinking that there are representations, then one will have little interest in the relation between mind and the world or language and the world’. In giving up the presupposition that there are bits of the world that make our sentences true, we dispense with the contemporary quarrel between realists and antirealists, because we need not worry about trying to ‘distinguish between those true sentences which correspond to ‘facts-of-the-matter’ and those which do not’. Rorty sees Davidson as continuing the tradition of anti-representationalism (Quine, Sellars, Wittgenstein etc), which is the ‘attempt to eschew discussion of realism by denying that the notion of ‘representation’, or that of the ‘fact of the matter’ has any useful role in philosophy’. In short, Rorty uses Davidson’s argument against facts as a reason to claim him as a champion for his anti-representational crusade. Let’s return to Davidson’s argument to see if this reading of him bears out.
Davidson claims that both conceptual relativism and the distinction between ‘scheme’ and ‘content’ are nonsensical. His argument against the former intertwines with the latter, which the former is meant to presuppose. He deploys two primary arguments against the scheme-content distinction: 1) the anti-scheme argument, which involves an appeal to the conditions something must satisfy if it is to qualify as a conceptual scheme; and 2) the anti-content argument, which relies upon the successful rejection of facts. The second argument is most relevant for our discussion of anti-representationalism.
There are four distinct ways of characterising the relationship between scheme and content. Both reality and uninterpreted experience are contenders for the role of content, whilst we find that conceptual schemes either organise (in the sense of systematising) or fit (in the sense of describing). The four characterisations are thus: schemes organise reality, organise experience, fit reality or fit experience. Davidson argues that none of these characterisations are viable. Most pertinent is his argument against schemes fitting reality as this connects with talk of an ontology of facts. Talk of fitting is talk of whole sentences as opposed to the ‘referential apparatus of language’; it is a sentence that ‘successfully faces the tribunal of experience…provided that it is borne out by the evidence’, that is to say the totality of sensory experience. Similarly, for a set of sentences to fit the totality of experience is for each of the sentences in the set to be true. Davidson allows that such entities can be called ‘posits’, for they can be contrasted with what they are not: sensory experience, or in other words, uninterpreted content.
Davidson then makes his move against schemes fitting either experience or, reality, for, he argues, ‘the trouble is that the notion of fitting the totality of experience, like the notion of fitting the facts, or of being true to the facts, adds nothing intelligible to the simple concept of being true’. To speak of sensory experience is just to express a view about the nature of evidence, and it does not add a new entity to the universe against which we can test conceptual schemes: ‘all the evidence there is is just what it takes to make our sentences or theories true. Nothing, however, no thing, makes sentences of theories true’. On Davidson’s conception of a scheme, there must be something extralinguistic for a true sentence or belief to match up to in order for the schemes-fitting-reality narrative to succeed. The something extralinguistic can either be the world itself or an individual fact: Davidson is claiming that ‘neither will work because each trades on the idea that the entity in question ‘makes the sentences true’.
Davidson makes use of a slingshot argument to show why he thinks that individual facts are not the something extralinguistic; ‘a slingshot argument imposes very definite constraints on what theories of facts…must look like, constraints that many proposed theories are incapable of satisfying’ (Neale: 1999). But then Davidson seems to go further than this and also rejects the idea of the world existing independently in order to defeat the view that individual sentences are made true by the world. Certainly in the course of rejecting the idea that schemes organise the world he points out that this position presupposes that there are entities out there in the world that require organisation. But at other points he does seem to suggest that the world is one of two things that makes a sentence true, for example Tarski’s Convention T reveals that ‘the truth of an utterance depends on just two things: what the words as spoken mean, and how the world is arranged…Two interpreters, as unlike in culture, language, and point of view as you please, can disagree over whetehr an utterance is true, but only if they differ on how things are in the world they share, or what the utterance means’. However, this is of no consolation to the correspondence theorist, for ‘it is no more illuminating to be told that a sentence is true if and only if it corresponds to the world than it is to be told that a sentence is true if and only if it is true, states a truth, or fits the facts’ (Neale: 1999). These are all just an idiomatic meaning of the phrase ‘is true’.
The crucial point here (one that perhaps Rorty misses) is that although Davidson has argued that if there are no individual facts then we cannot say a true sentence corresponds to the facts, he does not explicitly claim that there can be no representations of objects or events. Whilst he does not accept instances of ‘a represents b’ where b refers to a fact, he nevertheless accepts instances where a is an object and b is a person, place or event. Rorty, however, wants to draw more from Davidson’s conclusions in order to fit him into the tradition of anti-representationalism, namely, the ‘claim that no linguistic items represent any non-linguistic items’. But Davidson is only saying that there are no sentence of the form a represents b where a is anything whatsoever and b is a fact. He does so on the basis that there are no entities to serve as the value b, for ‘nothing can be said to represent a fact since there are no facts to be represented’ (Neale: 1999).