The world in my mind, my mind in the world by Igor Aleksander, Exeter, Imprint Academic, 196 pp, £17.95 hbk ISBN 1 84540 021 6, published 17 May 2005, and
Consciousness: creeping up on the hard problem by Jeffrey Gray, Oxford, Oxford University Press, 341 pp, hbk ISBN 0 19 852090 5 £29.95 published 17 June 2004
(This is the version submitted. It may have been edited before publication.)
These two books – both on the hard problem of consciousness – could hardly be more different. Igor Aleksander, an engineer whose ideas are inspired by building robots, wonders what is required to make them conscious. Jeffrey Gray, a well-known psychologist who died last year, amassed a vast amount of evidence from neuroscience, philosophy and psychology to construct his own solution to what some people believe is the last remaining mystery for science.
Both begin by outlining the “hard problem”, a phrase coined by philosopher Dave Chalmers. The problem is to understand how subjective experiences, or “what it’s like to be”, can arise from the activity of physical brain cells; or, to put it a different way, to explain how the redness of red or the ineffable smell of lavender (those qualities that philosophers call “qualia”) can be created by the brain. This really does seem to be a mystery because brains and private feelings appear to be utterly different kinds of thing – so how can one explain the other?
Amidst all the confusion there are two major responses to the mystery. On one side the “nothing extra” camp believes that once all the structures and functions of the brain have been explained there will be no mystery left over – as happened with “phlogiston” or the once mysterious “life force”. In other words, there is no such thing as “consciousness itself” and the task is to reduce consciousness to something else. Some versions equate consciousness with the functions the brain carries out. Some claim that consciousness simply is brain activity. The trouble is that so far no one has explained how this can be so.
Closer to everyday intuitions is the “something extra” side, who believe that consciousness is, as it seems to be, something additional to other brain functions, with its own powers, properties and functions. Here the task is to understand how something so peculiar and indefinable can possibly be produced by the brain or correlated with brain activity. In this view consciousness really is mysterious, and the hard problem is really hard.
Aleksander puts himself firmly in the reductionist “nothing extra” camp by saying that mechanism and sensation are as inseparable as H2O and water – they are identical. So he is not looking for the neural correlates of consciousness but for the “Neural IDENTITY of Being Conscious”. Conscious sensations are neural activity, he says, because believing anything else would be tantamount to believing in ghosts. He proposes five axioms that derive from his own first-person, conscious experience, including the feeling of experiencing an outside world, having memories, selective attention, planning, and emotions. He then sets about explaining how these can be applied both to artificial systems such as robots and to living ones such as ourselves.
The critical function, according to Aleksander, is the process he calls “depiction”, a kind of very rich, multimodal representation of the world whose contents become what we are conscious of. However, Aleksander never explains what is meant by this identity, nor why the contents of such depictions end up as subjective experiences when the contents of other kinds of representation do not.
Some people, myself included, have argued that consciousness is an illusion. Aleksander rejects this idea but gets into a fearful muddle in the process. He seems to confuse the idea of visual illusions, when we make mistakes about what is out there in the world, with what is known as “Grand Illusion” theory when we make mistakes about the nature of our own experience, for example, wrongly believing that we experience a stream of rich and detailed representations of the world when in fact we have only fleeting scraps of perception and an illusion of richness and continuity.
Aleksander’s most important contribution is in considering how a machine could be conscious and what would make it so, but in the end I was not convinced by his claim that the process of depiction holds the key. Although he claims that mechanism and conscious experiences are the same thing, he still lapses into saying that the brain “generates” or “supports” consciousness as though consciousness were, after all, a mysterious added extra.
Jeffrey Grey, by contrast, does not shrink from tackling the mystery head on. Indeed he fully embraces the idea that the brain generates qualia, and tries to explain how. For example, there is the difficult question of what consciousness is for and why it evolved. If you think that brain activity and subjective experiences are the same thing (or that functions such as perception and memory are) then you have no problem with evolution because natural selection favours those functions and abilities, and you do not have to explain consciousness as well. But if, like Gray, you think that consciousness is an added extra then you have to explain what it is for.
According to Gray, qualia are indeed added extras. He even suggests, I think uniquely among consciousness theorists, that the pairings between meanings and qualia are potentially flexible – that evolution might have attached different qualia to different functions, or none at all. So there must be a reason why they are linked up the way they are. He claims that qualia, once created, can be put to the service of a great variety of cognitive processes. So they have very important functions that are quite separate from the functions of the brain processes that create them.
The function of qualia cannot be the fast on-line control of behaviours, even though it may feel that way. Gray explains why not using both everyday examples – such as the tennis player who hits the ball back long before she can have consciously seen it coming – and detailed descriptions of many fascinating experiments. It is precisely such experimental findings that have led many theorists to treat consciousness as an illusion, or as nothing but the functions of the brain, or even as an epiphenomenon, but not Gray. Instead he suggests that consciousness acts as a late error detection mechanism. The human cognitive system relies on a comparator system that, in discrete moments of about a tenth of a second, makes plans, generates predictions about what will happen next, and compares them with sensory perceptions. All this is done unconsciously but then the outputs of the comparator, the mismatches, finally “enter consciousness”. Once “in consciousness” the slow time-scale there ensures that we experience a smooth and stable constructed world, and errors can be detected for improving future actions.
Gray’s is an unusual and daring theory but I don’t think it works because it leaves the central mystery intact. What does it mean to say that something “enters consciousness”? Is consciousness some kind of place in the brain or mind? Or, if not, what happens to neural activity to make it turn into, or create, or generate, qualia? I don’t believe the theory provides answers. Nevertheless this is a superb book. Gray’s descriptions of the experiments are clear and accurate, his explanations of the competing theories are lucid and helpful, and his own theory is an interesting challenge to all of us still tearing our hair out over the mysteries of consciousness. Everyone working in consciousness studies should read it.