Why be conscious?

19 August 2005

A good question. A paper (pdf) by Lee Pierson and Monroe Trout asks, what is consciousness for ? If consciousness has evolved in human beings, it must have had some survival value, but what exactly was it? This reminds me of the way Wallace, the co-discoverer of evolution, eventually came round to the view that the theory he and Darwin had discovered could not account for human intelligence. Consider a simple hunter-gatherer, he said; nothing in his way of life requires the ability to do calculus or write symphonies - yet we have developed these supererogatory abilities.

Wallace was surely mistaken, but the question remains. Pierson and Trout's answer is, in a word volition. They believe consciousness gives humans (and other creatures - no species prejudice here) the ability to rise above the hard-wired responses of simpler creatures and manage their behaviour in a more strategically effective way. This is possible because consciousness confers the ability to direct one's attention, keeping it on an object for sustained periods, or for that matter diverting it to a new object on the fringes of awareness.

This is a pretty good answer, I think, though it doesn't strike me as outstandingly novel. We've heard before of how lower animals (and according to some even primates) are liable to have their attention dominated by what they can actually see here and now; the unfortunate chimp, for example, who finds it impossible to choose the smaller reward even though he has seen, and even seems to understand, that he is always given the reward he doesn't reach for.

Pierson and Trout range much more widely than that, however, and in doing so they sometimes become less convincing. Reading the paper is a bit like flying a plane above some low-lying cloud: in places you can see the ground unrolling quite nicely, but in others alarming peaks rear up unexpectedly without much warning. It may be that the theory suffers from being expounded in a short paper, and that a longer treatment would clear up some of these cloudy areas.

I'm not at all sure, to take a fairly fundamental example, what metaphysical status they are assigning to consciousness. They say very clearly that they regard it as an emergent property of neural activity, but they also seem to lean towards fields and other more novel explanations. The great advantage of emergence, surely, is that it gives you something unexpected without requiring new fields or other apparatus. There's also an awkwardness about saying that an emergent phenomenon has strong causal relations with the substrate it arises from - or at least, some elucidation is needed.

Another feature of the theory which seems to need more explanation is its anti-determinism. Pierson and Trout are very clear that the operation of volition is not deterministic. They repeat one dreadful argument against determinism. This goes as follows: according to determinists our beliefs are determined by our circumstances, but that means that the determinists' own views were simply dictated by fate, and they have therefore no grounds for saying that determinism is actually true: determinism undermines itself. A pretty weak argument: the fact that my views were caused by something hardly means that they are false: indeed, if they were caused by the right kind of thing - relevant evidence and valid arguments - it might be a reason for thinking them true.

I suspect that Pierson and Trout just take it for granted that the sort of higher-level control they are proposing requires something non-computational; but that's far from clear - at the very least it requires some clearer supporting argument. The real problem is that it isn't at all clear how consciousness helps, either. How do subjective experiences allow us to direct our attention? For that matter, how does consciousness allow us to make decisions about which object to direct our attention towards?

I suspect that there might be interesting answers to some of these questions, and perhaps Pierson and Trout could give them - but they don't seem to have done so to me in the current paper.