- Title How to Make a Monster
- Author Butler, Marilyn Arbib, Michael
- Publication Title The Daily Telegraph
- Collection The Daily Telegraph
- Date Wednesday, Nov. 2, 1994
- Issue Number 43347
- Page Number 16
- Place of Publication London, England
- Language English
- Document Type Article
- Publication Section News
- Source Library Telegraph Media Group
- Copyright Statement © Telegraph Media Group Limited.
SCIENCE HOW TO MAKE A MONSTER As a new film about Frankenstein opens this week, Prof Marilyn Butler examines how Mary Shelley's tale owed more to the scientific thought of her age than modern popularisations admit IN 1992, I met a researcher hired to find out what Californians think of scientists. She had seen a particularly neat response: a sketch of the Monster from the 1931 film of Frankenstein, as acted by Boris Karloff. Even Mary Shelley's original novel of 1818 can be told in a sentence, as the story of an inventor who assembled a human form from body parts, brought it to life, and ran away. In an age of genetic engineering, Frankenstein remains the most damaging sketch of the modern scientist, too clever by half and deeply irresponsible. True, commentary on the novel has ranged far beyond its scientific content. Frankenstein invites discussion about topics such as personal ethics, nurture, family break-up and mental breakdown, Yet none of these applications of the plot do away with a newly emerged fact, or at least an overwhelming probability, that the tale Mary Shelley first told at Geneva in June 1816 referred to a scientific debateoftheday. In London between 1814 and 1817, two surgeons, John Abernethy (1766-1831) and William Lawrence (1783-1867), disputed the so called "vitalist issue" (how life is to be defined and where it originates) in public lectures at the Royal College of Surgeons. It has recently come to light that Mary and her husband Percy Shelley knew both is easier to document Mary Shelley's debts to Lawrence's physioiogy. zoology and ethnography in the two volumes she had to add to make a novel. In volume II of her original Frankenstein, the Creature tells the story of his survival, self-education and yearning to socialise. In volume HI, Shelley compares that creaturely instinct to be loving and familial with Frankenstein's loneliness, egotism, and seeming aversion to his own marriage. In 1819, Lawrence brought out Lectures on Physiology. Zoology and the Natural History of Man. In the section he calls "A Natural History of Man", he discusses the scientific implications of narratives of children found in the wild, on which Shelley plainly draws for the Creature's early experience. Later in that work, he debates the degeneracy and in-breeding of Europe's refined classes. In effect, Mary Shelley has followed the Abernethian "bad science" of the experiment with a scientific alternative, Lawrence's state-of-the-art overview of humanity's progress. Before Branagh, at least, the Lawrence passages have not been much utilised. Shelley's anonymous book was suspiciously reviewed in the influential Quarterly Review, an opponent of irreverence and materialism, and next year Lawrence's Lectures were savaged. The Quarterly called on th Rl Cllgf y debaters, and that Lawrence was a close friend. In 1814, Abernethy addressed a pious public, as well as fellow scientists, when he said that the material sciences could not account for the nature and origins of life; something "superadded" was still wanted. William Lawrence, the Royal College of Surgeons to make Lawrence withdraw the most offensive passages, and undertake not to write again in the same vein. The Royal College suspended him until he withdrew the book. The fate of Lawrence's book is a significant event in the history of science, and a key to the post-publication ki Ahi Probability: Prof Marilyn Butler Abernethy's former pupil, emerged as the champion of the materialists, who held that life was immanent within the body and a property of matter. The first of the iectures he gave in March 1816 gave an impressive overview of new insights into physiology from the Continent. The second, Life, rebutted Abernethy's case as a mish-mash of out-of-date science and irrelevant theology. Already a cause celebre in the spring of 1816, when the Shelleys, Lord Byron and the doctor Polidori were in London, Lawrence's lectures came out in book form in June. By then, all four writers had become acquainted as neighbours on the lakeside near Geneva. On 15 June, they discussed ''the principle of life, and whether there was any probability of its ever being discovered". Let us suppose they were talking about the vitalist debate, and Lawrence's sceptical contribution to it. The following night, four of them undertook to produce a horror story each. What could be more likely than that the experiment Mary Shelley chose as her topic is essentially Abernethy's allegedly farfetched solution to the life issue put into practice as an untidy, half-baked attempt to vivify a very dead body with something superadded, analogous to electricity? My suggestion that the inventor Frankenstein partly caricatures Abernethy remains guesswork. It implications of the experiment. As Percy Shelley's widow, Mary Shel pp evolution of Frankenstein. At this point, Frankenstein was notorious enough to suffer piracy, in the form of two unauthorised London stage adaptations of 1823 which drew attention to the daring materialist ley was now tarred with her husband's well known atheism as well as Lawrence's materialism, until she performed surgery on her text. For her novel's third edition in 1831, Mary Shelley added remorseful passages, which made Frankenstein a more sympathetic and religious character. She made his scientific education seem dangerous, an intellectual false start. She removed details about the family's marriages that touched on genetic concerns. What remains in the novel of Lawrence — the sympathetic side of the Creature, the "good" empirical natural science — is hardly noticed. The 1831 edition is the version modern readers have generally read, the version that, from the 1820s, the public and the media forced on the author. They took Frankenstein to their hearts as the ultimate realisation of a threatening story: the mad scientist and his frightful progeny. D Marilyn Butler is Rector of Exeter College, Oxford, and the editor of Mary Shelley's 'Frankenstein': the 1818 edition (Oxford: World's Classics) Branagh is the inventor and de Niro the monster in Frankenstein, which opens on Friday Boris Karloff as the Creature in the 1931 film inspired by Mary Shelley's third, altered, edition Shelley's Gothic theme survives in the quest for artificial intelligence in computers. A book by Sir Roger Penrose suggests it is doomed to fail. Neuroscientist Prof Michael Arbib is not so sure FIVE years ago, mathematician Roger Penrose published The Emperor's New Mind, a scientific epic that enraged artificial intelligence experts by insisting that their efforts to solve the mystery of consciousness would never succeed, His response to their criticisms, Shadows of the Mind, finds him even more entrenched in that view, The book may charm and convince readers by alternating scientific popularisations with impressively detailed mathematical argument, which Penrose says m3y be skipped. None the less, he fails to prove his assertions. Part I of Shadows tries to show that consciousness is inherently impossible to simulate on a computer; Part II asserts that yet unknown advances in physics, a fusion of quantum mechanics and general relativity in a new theory called quantum gravity, will yield a scientific (but non-computable) account of consciousness. Shadows is essentially a replay of The Emperor's New Mind but with the first part augmented to respond to many criticisms of the earlier work, and the second augmented with a discussion of cellular structures called microtubules as a bridge from quantum gravity to neurons, brain cells. For the former he builds a strawman, a theorem-proving robot irrelevant to a theory of consciousness while in the lat ter he clutches at straws (microtubules, to be exact). The first part rests on the fate of the search for an axiom system whose theorems yielded all mathematical truths. In 1931 Kurt Godel dashed hopes for such a system, proving that any consistent, arithmetical logic must be incomplete: if L denotes th incom that logic, he showed the statement "L is consis* tent" could be coded in numerical form, was true by assumption, yet was not a theorem of L. This led many people to claim that Godel's theorem showed that machines could not be as intelligent as humans, arguing that any machine with a claim to intelligence would be equivalent to a computer whose sole job was to derive theorems of some consistent logic L. We humans know something (namely, L is consistent) that the machine cannot know, and therefore we are smarter. Penrose says we are aware of something the machine is not aware of (namely, L is consistent) and therefore consciousness is non-computable. In discussing Godel's theorem, my 1987 book Brains. Machines, and Mathematics draws on the result that a computer can be designed such that, whatever formal system L it embodies, it can ^produce itself in a form that embodies the old L with the consistency claim for L appended as a new axiom. It needs no consciousness to add this axiom. Therefore Penrose rests his claim that human consciousness is non-computable on the ability to detect a "missing truth" that can be computed by an unconscious machine! He insists on the assertion "any machine with a claim to intelligence would be equivalent to THE REALITY a computer whose sole job is to derive theorems of some consistent logic." However, the ability to verify proofs of theorems is a very small portion of the mathematician's mental apparatus. As Mary Hesse and I argued in The Construction of Reality, human knowledge can be fallible and metaphorical. Little of it approaches the "unassailable truth" required for the straw-man But fallibility does not imply non-computability. Indeed we offered a computational account of a fallible knower. It is not based on a fixed logical system, but changes as the knower makes mistakes or encounters new aspects of the world. We even outlined a cornputational theory of consciousness in this framework, where Godel's limitation does not apply. The very fact that Penrose and I reach opposite conclusions demonstrates that mathematicians are not well described as "generators of unassailable truth" and so there is no reason for a psychologist to build unassailability into a model of the mind, whether computational or otherwise. The second part of Shadows starts with a lively exposition of quantum theory and of the key paradox that concerns what the process of measurement means. Penrose suggests this may be resolved by a future theory of quantum gravity. Disturbingly, he ri th avity. never tries to characterise consciousness to set goals for this theory, save for some informal remarks. He concedes that consciousness is a property of brains, but cites almost none of the vast literature of brain research. Most neuroscientists agree that microtubules provide a "skeleton" for the neuron with two Penrose: builder ofthe straw-man functions: to control the neuron's shape, and to transport molecules to and fro between the cell body and the synapses where one neusends "messages" to another. Penrose suggests that the network of microtubules might be the site of quantum activity in the brain that might correspond to his solution of the above paradox, and this might yield the non-computability he believes he has shown necessary for consciousness. This "mighty" chain does not bring a "science of consciousness" any closer. The nature of consciousness remains an open question, but there is no convincing eyidence that it rests on noncomputability, quantum gravity, or microtubules, which can be found in the brains of ants. My bet is that the key lies in studies of brain damage, addressing such data as those on one patient who has no consciousness of an object's size — she cannot convey it verbally or in pantomime • yet can correctly preshape her hand when moving to grasp it! Penrose offers no clues as to how such phenomena might be explained. □ Michael A. Arbib is professor of computer science and neurobiology at the University of Southern California. Shadows of the Mind: On Consciousness, Computation and a New Physics ofthe Mind, by Sir Roger Penrose, is out tomorrow (Oxford University Press, £16-99).