Recently there have been a number of attempts from students of mathematical and physical backgrounds to reestablish a dualistic approach to mind and behaviour. Most notable has been the work of Penrose [1], however, his basic idea has origins in the work of Popper and Eccles [2]. Hameroff [ThisIssue] provides an example of their acceptance of the division that Descartes made between mind and body. Unsatisfied (probably with good reason) with any of the present-day materialistic accounts of, "our mental experience," he makes a specific proposal that a purely speculative scientific theory (quantum gravity) provides the conduit between mind and body.

The essential premise from which an appeal to quantum gravity may make sense is Penrose's arguments [1] that the mind can perform non-computable operations. Certainly, quantum mechanics supposes the incomprehensible property that the world seems to exist in a superposition of possible states and every so often plumps for one or another. If one follows Penrose and Hameroff and assumes that the hypothetical quantum gravity will also have the same property, then such a phenomena may offer a route for non-computability in the brain. We should also reflect upon the example they provide to establish that the mind performs non-computable operations. Penrose [1] claims that although a human mind can comprehend the proof of G\"odel's theorem, such an understanding can not come from within a formal computational system. This may well be a sensible argument against some claims made in classical artificial intelligence (although Sloman [3] provides a thorough defense), however, it hardly establishes non-computability as a yard-stick with which to measure mental explanations. Indeed, just as our appreciation of the truth of G\"odel's theorem comes from taking a wider, and less exact, perspective than just the formal proof; there seems (just as Turing [4] thought) to be no a priori reason why a purely mechanistic system could not also bring knowledge from outside a formal system (nay, general knowledge) to aid its `understanding'.

However, Hameroff, is also concerned with terms like 'free will' which any mechanistic system would find hard to provide. Certainly something indeterminate, like quantum gravity, offers more hope for `free will'. Penrose and Hameroff [5] propose that the microtubules of the cell cytoskeleton could be used to bring some of the weirdness of quantum mechanics into the classical realm. Penrose and Hameroff's idea is that, "the picture I have is that for a while, these quantum computations go on and they keep themselves isolated from the rest of the material for long enough - perhaps something of the order of nearly a second - that the kinds of criteria I was talking about take over from the standard quantum procedures, the non-computational ingredients come in, and we get something essentially different from standard quantum theory."

Penrose and Hameroff's belief [5,6] is that systems of microtubules might sustain large-scale quantum-coherent activity, with, "individual OR occurrences constituting conscious events." Because the microtubules are tubes, Hameroff suggests that their insides could be in some way isolated from the random fluctuations (heat) in the environment, this is, of course, crucial to maintaining quantum coherence for extended periods of time. However, we do not find such arguments convincing since microtubules are dynamic entities existing in a balance between polymerisation and depolymerisation [7]. Notably, individual tubulin dimers are constantly being added and removed from the open ends of the microtubules and individual microtubules have a half life of about 10 minutes.

To support his argument that microtubules might somehow be involved in the mind, rather than neurons and the nervous system, Hammeroff, following Penrose, points to the complex behaviour of single celled organisms. Unicellular organisms, by their nature, lack nervous systems, yet they are capable of complex behaviours such as chemotaxis and object avoidance. Hameroff and Penrose both cite the ciliate Paramecium which has a cell cytoskeleton rich in microtubules. These microtubules act as skeletal elements anchoring the cilia, and resist the forces the cilia generate as they propel the animal through the water with powerful rowing strokes. Hameroff follows Penrose in suggesting that the microtubules of a paramecium's cytoskeleton may also play a role in performing the computations necessary for its complex behaviour by accessing the quantum mechanical realm in the same way as he proposes for the brain (presumably since he supposes paramecium compute by using his "OR" procedure noted above, Penrose must also believe paramecium have 'free will' and 'experience'). However, through careful experimental work, and detailed computational models [8,9], the computational processing systems underlying the complex behaviour of unicellular organisms are now rather well understood. Protein molecules act as the computational elements in living cells, and they perform computations by their complex chemical interactions. Even bacteria are capable of complex behaviour, and in the case of clonal bacteria, they can achieve much higher degrees of behavioural coordination than paramecium ever shows. The complex behaviour exhibited by bacteria (for example chemotaxis) can be modelled from "first principles" from the chemistry of the complex interacting molecules and the reactions which drive metabolism. As Bray [6] notes, "In unicellular organisms, protein-based circuits act in place of a nervous system to control behaviour." Microtubules are nowhere to be seen and, as such, can hardly be the necessary elements required to do whatever computing might be needed for behaviour.

It is worth noting that in Hameroff and Penrose's theory nothing is specified to do the work of the mind. That is achieved by an amorphous (immaterial?) quantum entity which interacts with 'Platonic logic embedded in space-time'. The implications of such an opinion are unclear, one possible reading would imply that since there must, by definition, be only one Platonic logic, then all agents performing under it must hold the same opinions - just as all mathematicians agree on \pi or G\"odel's theorem. Neither is it clear how appealing to such a Platonic space furnishes an explanation for 'free will' or 'experience' that is any more informative than a mechanistic attempt.

Clear though it is that consciousness is certainly, for many, a mystery needing explanation, Hameroff and Penrose's attempt to offer the microtubules within in neurons the same place in their theory as Descartes' use of the pineal gland, is yet again making the brain a, now quantum, antenna for the mind. The new conduit is identified by drawing from new and speculative ideas in physics. This can be a dangerous form of logic, for that fact that two areas of science are not understood does not imply that they are connected. Irrespective of how exciting these new ideas coming from physics are, it seems a little churlish of their proponents, as can be the want of physicists, to try to explain all phenomena from their level of explanation. We fear the necessary insights are some distance away.

References

[1] Penrose, R. (1989) The emperor's new mind. Oxford, OUP.
[2] Popper, K. and Eccles, J. (1977) The self and its brain. Berlin, Springer-Verlag.
[3] Sloman, A. (1992). The Emperor's Real Mind. Artificial Intelligence. 56. pp 355-396.
[4] Turing, A.M. (1950). Computing machinery and intelligence. Mind LIX, no. 2236, pp 433-460.
[5] Hameroff, THIS ISSUE
[6] Penrose, R. (1996) The large the small and the human mind. Cambridge University Press. Cambridge.
[7] Alberts, B., Bray, D., Lewis, J., Raff, M., and Watson, J.D., 1994. The molecular biology of the cell. Garland Publishing Inc., New York.
[8] Bray, D. (1995) Protein molecules as computational elements in living cells. Nature. 376, p307
[9]. Bray, D., & Bourret, R.B. 1995. Computer-analysis of the binding reactions leading to a transmembrane receptor-linked multiprotein complex involved in bacterial chemotaxis. Molecular biology of the cell. 6. no. 10, pp 1367-1380.