Will this erring be read?

In 1989 Roger Penrose put forward a revolutionary thesis. He argued that consciousness was associated not with macroscopic events in the nervous system as classically described, but with phenomena at the quantum level with all the uncertainty that that entails. In brief, his arguments were as follows. There are theorems in mathematics that cannot be proven by algorithms, but that mathematicians can see to be true. In so far as the nervous system operates algorithmically like computers (as usually accepted), it could not see the truth of such theorems. Some unspecified noncomputational mechanism must therefore be at work in the brain. Since individual particles could not affect the nervous system as we know it, coherent aggregates of particles (the Bose-Einstein condensate - particles that have all their attributes in common, as in a laser beam) must be formed somewhere in order to affect the conventional nervous system. Penrose believes such aggregates occur in the micro-tubules, which are part of the cell's cytoskeleton: he gives a series of cunning arguments to render this idea plausible. For example, unicellular organisms behave in a complex and purposive way: they have no nerve cells but they do have a cytoskeleton. Again, substances that render people unconscious, anaesthetics, are thought to affect the microtubules.

Penrose is meticulous in his use of evidence, but his highly speculative arguments are not completely persuasive. Although there are true propositions that are incapable of being proved within a given calculus, they might be proven within a higher level one (a metalanguage). Furthermore, one of the implications of Penrose's arguments appears to be that there are no true propositions that people cannot prove, but this is almost certainly untrue. For example, I cannot prove that I will perform a particular action (even if there is no change in the external world) since attempting to decide may change my state and I cannot take such changes into account without an infinite regress. Nor can I be certain that the decision process will halt before a solution is found, as opposed to being abandoned without a result being reached. Second, Penrose gives no examples from everyday life in which noncomputational methods must be used. Third, he is vague about how such methods operate in order to solve problems. He is reduced to saying that we do not at present understand the collapse of the wave function (the interconnection between the world of fundamental particles and the macroscopic world) and that once this is understood, which would require a revolution in physics, we shall understand noncomputational methods and also the nature of consciousness. This is a wish and a prayer, not a promise. Finally, if all consciousness depends on nonconscious quantum phenomena, the attempt of cognitive science to correlate consciousness with computational mechanisms would be a waste of effort, despite the successes it has already produced.

It was only a matter of time before someone would seize on Penrose's ideas and use them in ways he never intended. Ostensibly, Chris Nunn attempts to contrast Penrose's way of looking at consciousness and the brain with the classical view. In fact, the book defends Penrose's ideas by extending them to explore a number of bizarre phenomena - some real, most imaginary.

Nunn claims that telepathy can be explained by the Einstein-Podolsky-Rosen phenomenon - the fact that if two particles become "entangled" the attributes of one can be fixed by an observation on the other, however far away they are, even though before the observation is made, these attributes are not determined. Similarly, according to Nunn, psychokinesis is achieved by particles in the brain being somehow in tune with particles in the object affected by psychokinesis. No account of how quantum waves operate to achieve these effects is supplied. Nunn uses similar arguments to explain Jung's archetypes, and even the tendency for members of a group to think in the same way as one another. Talking to one another is apparently not sufficient. He also believes that the indeterminacy of the collapse of the wave function underlies freedom of the will: this opinion is inconsistent with the predictability of most human actions and if decisions are influenced by random events, the concepts of credit and blame would have to be abandoned.

Awareness contains many errors: for example, NMDA is not a neurotransmitter as alleged; we see the world in three dimensions not two; according to "identity theory" consciousness is not "the result" of conventional brain activity, it is such activity (or aspects of it) under a different description. Some of Nunn's arguments are, to put it mildly, odd. For example, he infers that consciousness must have evolutionary value because it uses up 30 per cent of the body's energy: hence, it must be useful or it would not have evolved in the face of its disadvantageous energy consumption. But surely it is the brain not consciousness that uses energy: there is no need to believe that awareness requires any extra energy whatever.

It is unclear whether Nunn understands what he writes, but it is certain that few if any of his readers will. Awareness may well make Penrose feel queasy.

Stuart Sutherland is emeritus professor of experimental psychology, University of Sussex.