English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

I t is a great pleasure to have this joust with Igor who is not only a brilliant thinker about the mind, but also a great intellectual sparring partner. Igor has expressed his dissent from the view that I’m going to advance in support of the motion which is that ‘computer models of the mind are invalid’, in his book: The World in My Mind, My Mind In The World: Key Mechanisms of Consciousness in Humans, Animals and Machines. He actually devotes five pages of the final chapter to what he calls ‘Tallis’s complaint’ which I am now going to make public. The point of issue is whether computer models of the mind are valid. I am going to argue that they are not because the computational theory of mind is invalid. Igor may go on to argue that even if the computational theory of mind is invalid it may still be useful in the sense of being fruitful. I will leave him to prove that and simply suggest to you that while barking up the wrong tree may give a sense of progress it is illusory. So what is the computational theory of mind? It is that the mind is to the brain as the software is to the hardware, or to put it slightly differently the mind is a set of computer programs implemented in the wetware of the brain. It is difficult to recall that this notion was once revolutionary and counterintuitive, because it now seems to be hardwired into the thinking of most people who have views on the nature of consciousness. And so in arguing against the conventional unwisdom one has not only to demonstrate that minds are not computers, but also to explain why such a daft idea should seem halfway plausible. Arguments against the computer theory of mind therefore, should include an explanation of the aetiology and persistence of the delusion. My case has four strands which are addressed in Igor’s book, but I will restate them naively, as it were, to bring everyone up to speed on our dispute. The first is the unremarkable claim that computers are not conscious. The second is the equally unremarkable claim that consciousness is not computational. The third will assert that minds are conscious and if consciousness is not computational then neither are minds. And finally, if there is time, I will examine the language by which those who believe in the computational theory of mind are selfdeceived into thinking that the gap between body and mind, or between neural activity and consciousness, is crossed. Let me begin with a something rather uncontentious – computers are not conscious. Very few people, even proponents of the computational theory of mind believe that present-day computers are conscious, in the sense of being aware of themselves and of the world around them, or of being capable of happiness or despair. Super Crays with gigabytes of RAM are as zomboid as pocket calculators. Most computational theorists are hesitant in accusing current computers of being conscious. The commoner claim is that while we do not at present have conscious computers, we soon shall have, or eventually, develop them. We should be suspicious of such claims. I am old enough to remember the predictions of people like Marvin Minsky, who claimed that by 1990, computers would be so sophisticated that they would be able to manage without us and would treat us rather in the way that we treat household pets. We should be sceptical for another reason. Those who claim that one day we shall have conscious computers are not able to specify what additional features conscious computers of the future must have. There is much hand-waving: conscious computers will be more ‘complex’ or have a different kind of architecture – for example, a parallel architecture, based on so-called neural networks. I have yet to see a definition of complexity that will make consciousness seem inevitable, or even more likely, in the artefact possessing it. And there is no evidence that currently available, massively parallel, computers are more self-aware, are more prone to suffering or joy, or able to experience the sound of music or the smell of grass than their serial counterparts. The intuitions guiding those who imagine the conscious computers of the future are often a little naı̈ve. They include the notion that feedback loops or ‘re-entry’ will awaken circuits to self-consciousness. The smartest artefacts, with the most subtle feedback mechanisms and self-monitoring, are unaware of their smartness, or of being a self or of their monitoring. The most striking testimony to the fact that very few people really believe that computers will one day be conscious is the way that the goalposts recede as computing power advances. Computers a million times more powerful than those available to Alan Turing when he put forward his famous test for machine intelligence are still not credited with the ability to truly think. Turing argued that the question of whether a machine could think should be replaced by the question, ‘are there imaginable digital computers which would do well in the imitation game?’ In the imitation game, the computer Journal of Information Technology (2008) 23, 55–62 & 2008 JIT Palgrave Macmillan Ltd. All rights reserved 0268-3962/08 $30.00

‘Computer Models of the Mind are Invalid’