Spooky systems and magic bullets

Of all the frustrating, murky biological phenomena it is consciousness, with its quintessential subjectivity, that exercises a surprisingly strong pull on a growing group of physical scientists and mathematicians. The latest moth to the flame is Alexander Cairns-Smith, an organic chemist. The particular path that the author follows, although it is not stated explicitly until two thirds of the way through the book, is one whereby we might obtain a scientific model of what consciousness "is like". Curiously, the author dismisses the more understandable aim of discovering what consciousness actually is without explaining clearly why, and without giving any estimate of just how much, or how little, he hopes his model will achieve.

It is a model that is nonmathematical, physical and "spooky". Indeed, the theme of spookiness is with the reader from the outset. Cairns-Smith's vision of the brain is one viewable as three different systems, accounts of which give the book its overall structure.

System one is the world of the chemist-physicist, where molecular and submolecular interactions occur without reference to any grander design. Cairns-Smith is clearly at home here: he writes with authority and the ease of metaphor and image that betokens the experienced teacher. On the other hand general readers will find certain passages very tough going and might wonder why they have to take on board a morass of technical material occupying a good third of the book, yet which has no obvious link to consciousness.

Nonetheless, even a rather cursory read would reveal the take-home message in the author's view of the fundamental components of the physical world. It is a world where atoms are not autonomous, where events are not local but global, and where energy and matter blur into each other. Nothing is quite what it seems.

Having explored the concept of matter, we are shown how matter underlies life. It is a shame that the "spookiness" or otherwise of life itself is not discussed. However valuable thorough discussion of enzymes, ATP and so forth might be, a more engaging slant for the general reader would have been to demolish once and for all the vitalist stance. One feels that the author missed an obvious opportunity here to show that life can be reduced to familiar, nonmagical elements - but that the emergent properties of those elements are, for the time being at least, exclusive to biological systems. Surely it is this counterintuitive aspect of seemingly magical emergent properties that agrees rather well with Cairns-Smith's basic disposition.

Another more helpful key word that is stressed, however, is "communication". This idea dominates once we enter system two. Here we deal with signalling within the physical brain, from one neuron to another. Cairns-Smith rather misleadingly refers to this system as "neural computation", presumably to draw a sharp distinction with the ethereal, holistic consciousness that will constitute level three. On the other hand, the term might mislead the casual reader into assuming the physical brain is like a computer, a stance which Cairns-Smith does not particularly try to sell, even though he persistently describes neuronal connections (as many have before him) as "wiring", an inflexible, inelegant and to my mind inaccurate metaphor.

Sometimes the images pay off, for example, the action of cell adhesion molecules is described as "neurons waving their sugary flags", while axonal transport is crisply portrayed as cellular products efficiently "going by rail". On the other hand there are times when the analogies do not speak immediately to the nonphysical scientist: the paradoxically passive process of generation of action potentials is compared to Westinghouse brakes, rather than, say, the more banal yet familiar example of a deflating balloon squirting through the air. However, the overall impression is of a writer making an enormous effort to convey obscure and difficult ideas in an accessible way. It would be very surprising if Cairns-Smith hit the bull's eye with every metaphor every time.

A more worrisome problem is that by attempting a rather detailed review of the brain from the neuronal level, "bottom up", the reader tends to slip uncomfortably between stools. Too much space is devoted to too much detail for the general reader to be able to see the wood for the trees. And there are features of the wood that are relevant but have been omitted. For instance, it would be helpful to know that the firing of action potentials is not the only way of observing discrete populations of neurons at work. The elegant work by Grinvald and Aertsen showing how neuronal assemblies of varying size can form and reform within fractions of a second, may well be of relevance to anyone interested in seemingly "spooky" brain functions. Similarly, discussions of clue-laden phenomena such as synesthesia, phantom limb pain or prosopagnosia would be more likely to rivet the reader's interest than even the most cursory mention of ligand-gated ion channels and G proteins.

Meanwhile, golden opportunities slide away. We are given a relatively detailed neuroanatomical description of the brain, but are not treated to any discussion of the rationale and problems of localisation of function. The chemical structures of key transmitters are shown in a figure, but not their distribution in the brain. System two could have served as an invaluable bridge between the matter of system one and the mind of system three. As it stands, there are no clear clues, and readers are left with the impression that they have leapfrogged over the physical brain in order to relate molecules, atoms and other particles directly to consciousness.

Cairns-Smith as much admits this bias in the final chapter, which is couched as an engaging dialogue between himself (presumably), "Advo", and "Crit", a hypothetical adversarial chum. Advo: "How is it [consciousness] affected by system two? How does it act on system two?" Crit: "That I can answer easily. Don't know." True, none of us knows, but the phenomena of consciousness (system three) that are clearly rooted in perturbations in system two, for example phenomena involving mood-modifying drugs or, more specifically, blindsight, get very little airing.

Even though Cairns-Smith would probably agree that system three is most likely to arise from system two in some way, there is a feeling of rupture, a noncontinuity between the boring old slavish neurons plodding away algorithmically, and the will-o'-the-wisp of consciousness as it flits about the brain, delocalised and defiantly special.

Consciousness is hastily defined as "feelings", but as this profile is filled out during the narrative, we start to encounter some contradictions. On the one hand Cairns-Smith refers to consciousness when one is not paying attention as "spread out", but a little later he speaks of the relative simplicity of an actual experience at any one instant. It is hard to know how the author actually views consciousness since he spends very little time exploring any of its characteristics.

Still, a less-than-clear picture of consciousness should not deter us from the author's central theme: a quantum version of events along the lines originally proposed, we read, by Danah Zohar and Ian Marshall. Only in this final section of the book does Cairns-Smith's picture finally become less opaque. Having laboured in the early chapters through the fuzzy interconnections of time and space, energy and matter, we can now apply a comparable approach to mind and brain, where wonderfully ambiguous, delocalised quantum events collapse into the humdrum Newtonian world of the physical brain. Like Roger Penrose, Cairns-Smith has a vision of quantum coherence across banks of neurons: in his case, however, the orchestration takes the form of a Bose condensate (a Marxist-like organisation of particles whereby the individual is subsumed into the general whole so as to pull together towards a collective end).

I have two immediate scientific objections to this idea being actually implemented in the brain. First, Cairns-Smith conjures up the idea of "specialist brain cell proteins" that play a crucial, and presumably committed role as "suitable oscillators" in the generation of consciousness, but which thereby come perilously close to resembling that anathema of nonvitalist scientists and philosophers alike: magic stuff. But perhaps I am reading too much into terminology. A second anxiety is far more serious, since it concerns whether quantum events could ever unfold in the first place. The temperature in our heads is so high as to make it extremely improbable that boson condensation could play a significant role in brain function. Granted, Cairns-Smith acknowledges this problem, but to dismiss the issue by saying that "it is by no means implausible" that natural selection has circumvented the problem in an as-yet-mysterious way, just will not do.

A more general complaint is that all the favourite topics relating to the quality of the first-person experience are not elucidated at all. Whether or not Newtonian grounded action potentials are replaced by quantum coherence, the philosopher is no nearer to enlightenment on how such events might be translated into, or even correspond with, a red quale. Neuroscientists too might be frustrated, since no attempt is made to interpret phenomenological events such as Alzheimer's disease, schizophrenia (although they are mentioned) or even dreams, in terms of the model. As it stands therefore, the concept of boson condensation being "like" consciousness, currently has only limited usefulness. On the other hand no book of this sort should be dismissed for not producing the magic bullet of the physical basis of the mind. Cairns-Smith has a story to tell and he does so, eloquently and well.

Susan Greenfield is lecturer in pharmacology, University of Oxford, and Gresham professor of physic.