On theories that don't matter

Many years ago I thought of writing an essay entitled ‘On theories that don’t matter’. The basic idea was that there are three kinds of scientific theory. Some are true (as far as we can tell), and so matter. Others are demonstrably false, and so don’t matter scientifically, although they may be fruitful in other ways or historically interesting. And then there is a third kind of theory, which claims that it is true that there is – and can be - no specific category of knowledge we can call ‘truth’ (or indeed ‘falsehood’).

If such a theory is false, then it goes the same way as any other false theory. But if it is true, and it succeeds in proving that there is no such as thing as truth or falsehood, then there is still no reason to think that they matter. After all, they deny the reality of the one criterion by which something can be thought to matter from a scientific point of view – its truth. If they are true then there is no such thing as truth, so they are not true.

But unfortunately, if a theory that undermines the very notion of truth and falsehood is found to be true (whatever that might mean) then there is no such thing as truth in general, so no possibility of ‘scientific’ knowledge at all.

So it is a serious problem for scientists everywhere. And yet many of the most important (or at least most prominent) theories of the last two centuries have been of this third, epistemologically nihilistic kind. Behaviourism, evolutionary psychology, meme theory – all of them render the very idea of truth meaningless. If they are false, they are taking up a lot of intellectual bandwidth; and if they are true then we are up the creek.

Nor is this just about science. Philosophically speaking too, many – I suspect most – of the major schools of western thought have implicitly taken a similarly self-contradictory perspective. Empiricism is obviously incapable of delivering anything like truth, since there is no reason to believe that any accumulation of particular experiences, no matter how assembled, would tell us the truth. On the contrary, it simply implies that experience is little more than an accumulation of prejudices. But rationalism equally self-contradictory. Although it argues that we don’t enter into experience naked, it never seems to explain why we should think that the structures we bring to experience represent anything better than systematic experience.

Philosophy has differed from most of science in that it generally admits to these problems. Kant and Hume, for example, were entirely explicit about the impossibility at arriving at the truth. Science has seldom been quite so consistent. Indeed, some of its main practitioners have adopted a positively Olympian position, and propagating the truth of theories that, if applied to themselves, would cause them to disappear into their own unintelligibility.

For example, had behaviourism been true, it would also be true that the only reason I believed behaviourism to be true was that it was ‘reinforcing’ to believe it. Or rather, I did not even ‘believe’ it; I simply said that I believed it. What is more, I said that I believed it for the same reason (or rather, the same cause) that I like strawberries or that I am afraid of giant spiders – because it is reinforcing to do so. At no point does any actual belief take place, and in no sense can any of my statements be thought of in terms of truth or falsehood. Likewise for B.F. Skinner and his supporters – whenever they made a statement in favour of behaviourism or offered evidence in favour of their hypotheses, they made them – and we said we believed them – solely because it was reinforcing to do so. Truth and falsehood never came into it. Or rather, I say that truth and falsehood never come into it, exactly as though that were a meaningful proposition, but in fact I’m just saying this particular sting of words because my personal reinforcement history makes me do it. Under different personal circumstances I might have said that truth and falsehood were absolutely essential. For all I know, that is just what people did say, when behaviourism was still in vogue. Or rather, I say ‘for al I know’, but…

And so on, down the dark tunnels of infinite regression. In the end, if behaviourism is true then behaviourism – and literally everything else you or I or any other human being has ever said - is utterly empty.

Likewise for evolutionary psychology. According to this account, we ‘know’ about the world as we do because evolution has found it adaptive to look at it one way rather than another – the way, say, a crocodile looks at the world. Not only does this offer us no particular epistemic access to the world but, according to evolutionary psychology, it is utterly improbable that we could ever have evolved any way of looking at the world that was any closer than the demands of reproductive fitness required. And even if our evolved worldview were absolutely spot-on, there is no way we could know that it was, and no way we could kept it that way if evolution decides to head off in some other direction.

It is surprising just how many supposedly scientific theories of human knowledge are denying any strictly epistemic grounds (including evidence, argument, logic, mathematics, scientific method, and so on) for believing things. Of course, if they are true then it is not a self-contradiction, because nothing is a self-contradiction, because there are no criteria of logical coherence that make any sense any more. Even so, I find it quite worrying. Well, I say that I find it worrying, but…

All these theories – and many more - are undermined by the same problem, namely that they deny either the reality of or the relationship between the knowing structure we usually call the subject and the world of objects we may or may not be able to know. For a theory to matter doesn’t demand that we have any particular view about what subject and object are or exactly how they are linked together, but it does demand that you have some idea that knowledge is a relationship of this general kind. Eliminate either side and knowledge dissolves into dust.

Which is of course just what all the theories I referred to above do. They either deny that one side (usually the subjective side) exists or undermine its independent reality. Like a viral infection (a metaphor widely employed by meme theorists), a pattern of reinforcement or an advantageous adaptation invades your brain, takes command of its resources to replicate itself, and there’s nothing you can do about it. A meme or an adaptation or a reinforcement can only be displaced from our apparently quite brainless brains through another equally brainless process of selection or reinforcement.

There’s nothing you can do about it. But then, there’s not much ‘you’ left over to do anything. ‘You’ are little (and in some versions nothing) more than the sum of the adaptations or memes or reinforcements in your head – which is to say, the sum of your current prejudices. Can you simply criticise and reject a prejudice on the grounds that it violates your goals or values or that it simply doesn’t make sense? Richard Dawkins seems to think so. But then both your values and the criteria and mechanisms by which you make sense of things are themselves just more prejudices you had already been infected by.

So even if you could evict them, what difference would it make? Such prejudices play no special role in the workings of your mind, such as providing the higher order integrative function that is usually assigned to logic or mathematics or regulating activity and experience through values. There is no possibility of either verification or falsification. On the contrary, all of these theories are inherently trivialising. Nothing rises above the level of a catchy tune, a good feeling or a neat reproductive trick.

Rather oddly, something else all these theories share is a comprehensive disregard for what is already known about how our intelligence operates, which is anything but in the manner of reinforcement, meme and selection’s shared method of trial and error. Intelligence proceeds by a process of abstracting higher level principles, values and goals, from action in the world, which in turn provide it with a capacity for insight and criticism. Through these, intelligence studies the empirical surfaces and functional utility of things for both the underlying structure and the existential qualities that can only be reached through a process of systematic, active construction. Since reinforcement and selection and all the rest are completely anathema to any such process, plainly they cannot offer us any explanation of intelligent activity and experience.

Hence all these supposedly scientific theories’ fundamental self-contradiction. Once one lapses into the view that an idea or value or method colonises my mind through some process of exogenous variation and selection over which I exercise no subjective control, as opposed to the idea that I believe in certain values, symbols, methods and purposes because I have good reasons to do so, then it is hard to see why one should not apply the same logic not only to every aspect of science itself, such as the notion of objectivity or scientific method, but also to the particular theory that is making this claim. But of course, once the idea of objectivity is abandoned as possessing no more intrinsic value than a catchy tune, then that is the end of science. As with any reductionism, all such theories make science unthinkable, for they argue that what counts as ‘science’ depends solely on what notion of science is currently especially infectious. and what counts as ‘thinking’ is whatever happens to be in my head.

Or perhaps one should privilege certain players in this particular game (e.g., scientists). But that still leaves the problems of exactly why one should privilege any particular disciplines, concepts or values in this way, and where the line should be drawn between the privileged and the non-privileged. There seems to be no compelling answer to the former, and in the absence of such an answer, there is no hope of an answer to the latter. And even if there were compelling reasons for privileging certain critical components of human experience, that does nothing to explain how such privileged areas could have come into existence.

If, on the other hand, ‘subjective control’ and ‘good reasons’ are taken to include at least the objectively testable coherence, consistency, completeness and correctness of ideas, values, and so on, then plainly the dual reduction of intelligence to biology and culture that underlies so much contemporary thinking cannot be completed. There are no doubt very many undigested aspects of our biology and culture that prejudice and compromise science, but it is doubtful whether any such failing of reason could not be overcome by its further development.