In the concluding essay of an adventurous collection, Stephen Jay Gould observes that most 'classic stories' in science are wrong. There are good reasons why he is right. In their reconstruction of the past, practising scientists have been apt to celebrate the insight of those who anticipated their own ideas, tacitly dismissing those who were blind to where the future would lie. The result has often been sterile histories, distorted by a preoccupation with confirming the present. The apocalyptic aspects of science, with the next breakthrough just around the corner, may add to the distortion by a more general undervaluation of the past. And the distortion is often sealed by an appeal to history for corroboration of fashionable stereotypes of scientific method, the classic discoveries having been made by 'prepared minds' whose interrogation of nature was conducted according to the canons of inductivism, hypothetico-deductivism or some transcendent hybrid. Consequently, science carries along a false history which, like a recessive gene, can pass undetected from one generation to the next. Not one of the least justifications for serious scholarship in the history of science is that it can rectify the distorted vision which the textbook traditions enshrine. It is a justification which Gould happily accepts. Whilst the majority of the 30 essays which compose Hen's Teeth and Horse's Toes divulge the latest news in natural history, at least a third have the additional merit of bringing a critical history of science to a larger audience. Gould's iconoclastic remark occurs in a discussion of a 19th-century 'fact', the history of which raises interesting questions indeed. This 'fact' was the ability of sires to influence subsequent progeny not fathered by them. One example was familiar and acceptable to Charles Darwin: the successive offspring of Lord Morton's mare. Crossed with a quagga (a now extinct zebra with stripes confined to neck and forequarters), the Arab mare delivered a hybrid with stripes in evidence. Subsequently mated with a black Arab stallion, the mare again produced an offspring resembling the quagga. This curious form of action at a distance was given a name (telegony) and even inspired a major programme of experimental breeding. As a 'fact' it was comfortably embedded in most of the genetic theories of the time (including Darwin's own) and was only rejected when August Weismann made it impossible in theory - the theory he erected on the continuity of the germ plasm and its protection from extraneous influence. Now for the moral. By contrast with the adage that an unquestioned theory may be overthrown by one novel fact, we have the illuminating case of an unquestioned fact being overthrown by one novel theory.

LRB 1 December 1983 | PDF Download