The Singularity is Near (2005)
These books would not be interesting in respect of consciousness theory, if it were not that Susan Blackmore, who is prominent in consciousness studies, turned to Kurzweil to help refute Penrose’s quantum consciousness theory. This seems an odd choice, as Kurzweil was firstly an entrepreneur in the computer industry and later diversified into writing popular ‘futurist’ books. He was not a specialist in either physics or neuroscience. In respect of consciousness, Kurzweil’s main argument is that computer technology will progress fast enough for himself and others to download themselves into computers before the middle of the 21st century. In ‘Are We Spiritual Machines’ Kurzweil is therefore mainly concerned that Penrose’s version of quantum consciousness should not get in the way of people being downloaded into computers, rather than with entering into any deeper discussion of the nature of consciousness.
His first argument is that the brain has more than enough capacity to do what it does without using quantum computing. In fact, some researchers argue that known brain capacity is not enough to resolve the perception of ambiguous objects (1.- 6.). Kurzweil does not appear to have done his homework here, confidently stating that ‘no one has suggested human capabilities that would require a capacity for quantum computing’.
He goes on to give a garbled account of how the Penrose/Hameroff theory arose. He refers to it being ‘pointed out to Penrose that neurons were too big for quantum computing’, whereas this was obvious to anyone familiar with physics. Penrose in fact offered only the vaguest suggestions for possible mechanisms in his first book. Kurzweil then refers to Penrose ‘coming up with the tubule theory’ whereas this is a product of pre-existing work by Stuart Hameroff, who became Penrose’s partner in developing the Orchestrated Objective Reduction (Orch OR) theory of consciousness.
However, the real point is that Penrose is talking about ‘understanding’ and by extension consciousness, as distinct from information processing. In the Penrose/Hameroff hypothesis, the quantum feature is primarily required to generate consciousness. In contrast, Kurzweil is really only discussing information processing, and gives us no clue as to what it is that makes conscious processing of information in the brain different from non-conscious processing either in or out of the brain. He does not seem to be aware of this as a problem.
Kurzweil’s second book ‘The Singularity is Near’ (2005) attempts a slightly more serious discussion of Penrose, but it was first book, written in 1999, that Blackmore refers to. Kurzweil starts by quoting Koch who wrote ‘Quantum mechanics is mysterious, and consciousness is mysterious. Q.E.D. Quantum mechanics and consciousness must be related.’ The quote is an intentionally silly suggestion. Needless to say, Penrose never said or implied anything of the sort. It does little credit to the standards of discussion in consciousness studies for this sort of thing to be highlighted. Good science is not done in this way.
In his own discussion, Kurzweil starts by conflating two unrelated arguments. First of all, he goes back to the argument proposed in ‘Are We Spiritual Machines’ that human capabilities do not require the brain to have the extra capacity provided by quantum computing. He says that no one has suggested such a thing, which is not in fact true, as other investigators besides Penrose have argued that there may be a capacity problem (1. – 6.). He then introduces a quote from Seth Lloyd to support his claim. As a physicist, Lloyd outlines the main real objection to quantum consciousness, which is the speed of quantum decoherence in the normal conditions of the brain. This is an important objection, but it is absolutely nothing to do with the Kurtzweil argument it is called on to support.
Kurtzweil also trundles the old mantra, peppering much of consciousness studies, that there is no scientific evidence or that there is little scientific evidence for such and such. The casual reader can take this to mean that a particular proposition has been shown to be false, whereas what it usually means is that little or no research has been undertaken in the particular area, often because funding would never be forthcoming. A further disconcerting feature of both these books is that important and controversial claims often go unreferenced, in fact the first book has no references at all. Further discussion in the second book again misses the point that Penrose/Hameroff are discussing consciousness, rather than just the modelling of information processing. Again, Kurzweil far from being someone to refer to on consciousness seems to largely ignore it as a distinct feature. His main concern, as in the first book, seems to be to prevent Penrose getting in the way of his own pet project of downloading minds into computers.
Finally, the author discuss Gödel’s theorem, the related work of Turing and the Church-Turing thesis. Gödel argued that propositions that could not be proved by a particular system of axioms could still be obviously true. Kurtzweil tries to turn this round by means of a strong interpretation of the Church-Turing thesis, to the effect that because the human brain comprises matter and energy, and that these are governed by the laws of physics, then any human understanding can be expressed in terms of an algorithm, which is by definition something that can be handled by a computer. This argument essentially works by assuming the thing it sets out to prove, because the Penrose proposition is that some physics, notably objective reduction of the wave function is non-computable.
Tags: Kurzweil Posted by