Monday, 28 February 2011

Social scientists against the Truth!

However, in an alternative postmodern worldview, the concept of an objective reality against which knowledge can be validated is discarded and with it the ‘correspondence theory’ of truth as the basis for understanding validity. [Paterson & Duxbury 2007: 536]

Why do social scientists not like the correspondence theory of truth? Both the paper I read last week and a conversation I had with Harry Collins’ colleague Rob Evans a couple of years ago are instances of this hostility. But I wonder whether it is a correctly directed.

I assume that the correspondence theory is something on the following lines.
  • ‘S’ is true if and only if ‘S’ corresponds to a fact (the fact that S).
It contrasts with other substantial theories such as the coherence theory (where truth consist in membership of a maximally coherent set of beliefs) and insubstantial theories such as the redundancy or deflationary theory. It addresses an intuition that when a belief or sentence/utterance is true it is so in virtue of something, something to which the belief or sentence/utterance answers. Something makes it true.

Why should one be suspicious of the correspondence theory? I think that it is helpful to distinguish between two things: correspondence slogans and correspondence theories. The former are innocent. Following the convention of using a ‘snowbound triviality’ then if one thinks that ‘S’ is true then one might assert:
  • S. (Snow is white)
  • ‘S’ is true. (‘Snow is white’ is true.)
  • It is true that S. (It is true that snow is white.)
  • It is a fact that S. (It is a fact that snow is white.)
  • ‘S’ states a fact. (‘Snow is white’ states a fact.)
and hence perhaps
  • ‘S’ corresponds to a fact. (‘Snow is white’ corresponds to a fact.)
That last case may sound stilted but it does not seem to introduce anything beyond what the others involve. Perhaps it could be introduced first via a notion of fitting a fact. (Or one can imagine it being used in a language class. Sentences in the language being taught might be left in a jumble and the task would be to match them to corresponding facts, either stated in English or depicted in pictures.) We can learn what facts are, and how to speak about them, in just this way. That is, assuming that Little Ludwig has mastered asserting that snow is white, I can teach him these other complexities on that basis. If he is prepared to claim that snow is white then he should also be prepared to accept that that sentence (or the utterance made using it) is true and that it is a fact that...

So far, so innocent. What is not innocent is attempting to use the correspondence slogan to explain truth and hence use it as the basis for a correspondence theory, a theory of the nature of truth. That is, one uses the right hand side of:

‘S’ is true if and only if ‘S’ corresponds to a fact (the fact that S)

to shed light on the left. But this faces two obvious and familiar problems. It requires both that one can explicate the correspondence relation and the relevant fact which makes the sentence true in terms initially independent of the left hand side invocation of ‘S’ being true. Sadly, it is far from clear that this can be done. Without using the sentence (or one like it), how can one articulate the sentence-shaped chunk of reality to which it answers? And how can a sentence correspond to that fact, independently picked out, in any other way than the truistic grammatical connection that depends on already knowing how and when to use the sentence?

The problems do not stem from merely talking of ‘correspondence’ and ‘facts’. That can be quite innocent. They stem from attempting to explain truth in these terms, when truth seems the more basic notion.

But if one does not distinguish between these two things (the slogan and the theory) then the rejection of an explanatory connection can escalate into rejection of the very idea of aiming to get the facts right. Then, given the close connection between it is a fact that S, it is true that S, ‘S’ is true and, just, S then the rejection of one can have some wild consequences when one is obliged to reject the others.

The temptation is often then to think that in the absence of full-blown world-involving truth-is-out-there truth, one can still have, and make do with, something like justification. Justification seems homely and modest. It is within our ken (it depends on no favour from the world). Invoking truth, facts and the nature of the world seems hubristic and something outside our ability to ensure (it depends on a favour from the world). But the problem is that once those have gone, reconstructing a form of justification worth its salt looks to be impossible since justification’s purpose is to increase the likelihood of truth. Without that, what makes a putative justificatory move justificatory? What does it mean to say that it is justificatory if one cannot explain justification by invoking truth? One cannot just make do with justification and give up on truth.

Why are social scientists dismissive of correspondence slogans and hence of truth itself? I suspect that it follows from a misreading of what even an explanatory correspondence theorist (let alone the mere use of a correspondence slogan) aims to do. It aims to say in what truth consists, or what it comprises. It does not aim to provide practical guidance to getting to the truth. My suspicion is that social scientists are very well aware of the complex social processes by which knowledge of truth is arrived at, for example in the sciences and science-guided social practices (such as medical care). But they assume that a correspondence theory naively assumes that the facts, to which true beliefs or sentences/utterances answer, just sit around patiently waiting to be vacuumed up. Since they eschew such naive inductivism (or a particularly stupid form of positivism) they reject the appeal to facts so understood in characterising truth (both when used as a theory but also as a mere slogan). But in fact, what they reject is a myth. There is no reason to assume that the facts are like that or that they must be thought of like that when saying that truths correspond to them. The reasons for dismissing a correspondence theory are quite other.

My hunch is thus that the target of their scepticism is wrong. It shouldn’t be truth and the innocent aim of trying to get the facts right. Truth is a friend to academics. It may be dull but it serves as a short-hand for normative standards. What the social scientists should target (what they should be sceptical of) are their particular (object) facts. Social constructionists of science should target the particular (usually physical) scientific facts they wish to explain in other terms. Social constructionists of mental illness should target the putative facts about illness they wish to question or explain away. Leave truth in general out of it.