The Unnatural Nature of Science

by Lewis Wolpert

  When it was suggested to Pasteur that many of his great achievements depended on luck, he replied – I’m sure with more than a little irritation – ‘In the field of observation in science, fortune only favours the prepared mind.’ It is not by chance that it is always the great scientists who have the luck.

  In 1908 Alexander Fleming passed his final medical examination and was awarded the Gold Medal of the University of London. He wrote a thesis on ‘Acute Bacterial Infection’ for the competition for the Cheadle Medal at his medical school, which he won. In this essay he described what were then thought to be all the defences against bacteria that the medical profession could offer. These were, in addition to the patient’s own resistance, the use of antiseptics, some antibacterial agents for specific bacteria (such as mercury for syphillis) and vaccines. Vaccines were the passion of his chief, Almoth Wright, and the treating of bacterial infections was his major preoccupation. Just a year later Fleming had the opportunity to administer Salvarsan, the chemical discovered by Paul Ehrlich which killed the organism that caused syphillis. The astonishing success of this treatment must have made Fleming realize that bacteria could be killed by specific chemical treatment. But, how, he must have wondered, was one to find such chemicals? During the 1914–18 war Fleming was in France, where he observed that antiseptics were powerless to prevent infection of wounds: the results were, unfortunately, often better if no antiseptics at all were used.

  If all this is not enough to persuade the reader of Fleming’s prepared mind, his discovery of lysozyme must remove all doubts. In 1922 he added a little of the mucous from his nose, at a time when he had a cold, to a bacterial culture. Around the drop, all the bacteria were killed. With great care and patience, he showed that the active component of the mucous which caused this was a natural constituent of tears, which he called lysozyme. This, he thought, was the body’s natural protection against bacteria. So, when that fateful mould of Penicillium floated on to his bacterial plate in 1928 and he saw that all round the mould the bacteria had been killed, no mind could have been better prepared. His biographer André Maurois reports Fleming as saying:

  I had often seen such contamination before. But what I had never seen before was staphylococci undergoing lysis [breakdown] around the contaminating colony. Obviously something extraordinary was happening. With the background that I had, this was much more interesting than my staphylococcal research, so I switched promptly. I am now glad that for years my interest had been directed to antiseptics and that some years before I had found in a somewhat similar manner another naturally occurring antiseptic, lysozyme. But for the previous experience it is likely that I should have thrown the plate away, as many bacteriologists had done before me.

  Of course it was fortunate that the mould landed on Fleming’s plate, but was that chance any less or more likely than that he had been born at the particular time, had become a doctor and had then chosen microbiology? Thousands, if not millions, of small events shape all our lives. Why focus on just one? To do so is quite misleading. To designate some scientific discoveries as serendipitous can be equally so. If Bobby Jones had discovered penicillin while playing golf, that, perhaps, would have been an example of serendipity.

  In 1896 Henri Becquerel was experimenting with uranium salts which emitted light in the dark after being exposed to the sun. He had concluded that the sun had caused the uranium to emit some sort of radiation, because it could blacken a photographic plate. Because, apparently, the sun failed to shine and so delayed further experiments, he developed the photographic plate nevertheless, found that the plate was blackened even though it had not been exposed to the sun and so discovered radioactivity. But this had nothing to do with luck: to develop the plates without exposing the uranium to sunlight was an obvious control that any scientist would be expected to do.

  The discovery of the vulcanization of rubber relates more to technology than to science, and indeed many, if not most, of the examples that are supposed to illustrate serendipity are concerned more with the discovery of a process or substance that has useful applications rather than being related to pure science. Charles Goodyear devoted an enormous amount of effort to making rubber impervious to temperature changes – in fact it became an obsession. He treated rubber with a variety of substances, including sulphur, with no success. In 1844, by accident, he allowed a mixture of rubber and sulphur to touch a hot stove and to his surprise he found that it was only slightly charred but, most dramatically, was flexible and tough, over a wide range of temperatures. Vulcanization had been discovered and it seems to me the perfect example of how technology advanced both before and after the Greek discovery of science. No science was required, only curiosity, and the common sense to try a variety of methods and to select the ones that work.

  In modern science I am always impressed by the fact that it is always the best scientists who seem to be the luckiest. Of course, advance in science, being a journey into the unknown, must inevitably confront scientists with the unexpected. This is not luck or chance: it is of the very nature of science, for as one explores phenomena or ideas at the frontiers of scientific knowledge it is the unexpected that provides the clues to guide further work. In recognition of this, Bruce Alberts, a distinguished molecular biologist, has cogently argued against giving too many grants to any research scientist, otherwise the scientist no longer has contact with the phenomena and merely receives filtered reports from more junior scientists. Under these circumstances, Alberts rightly observes, important and unexpected observations will escape notice by the leading scientist who has the skill and experience to recognize their significance.

  There are, indeed, numerous examples where scientists have, with hindsight, missed the importance of a particular event. In a sense, Aristotle failed miserably in this respect: he should have recognized the contradiction in his ideas about falling bodies, and he certainly could have discovered laws relating to the swing of the pendulum. It shows the absurdity of the idea of serendipity when one realizes that it was nearly 2,000 years after Aristotle before Galileo took notice of the pendulum-like swing of an altar lamp. A more recent example is provided by Peter Medawar. In Pluto’s Republic he points out how he and his colleagues missed the significance of an observation which, if correctly interpreted, would have led them to recognize a new and very important phenomenon in immunology. This was the reaction of a graft against the host to which it had been transplanted; at the time, they were focusing their attention on the reaction of a host to a graft. We are surrounded all our lives by innumerable ‘facts’ and ‘accidents’. The scientist’s skill is to know which are important and how to interpret them.

  5

  Competition, Cooperation and Commitment

  Among the many misconceptions of science are that scientists either pursue truth in a dispassionate manner, their only reward and aim being the better understanding of the world, or that they are entirely competitive and selfish. While both have some elements of truth, these are misleading images. Scientists are emotionally involved in their work, and, in addition to the joys of discovery, the social interactions between scientists play a fundamental role in setting scientists’ goals. Scientific knowledge is cumulative, and scientists have a special relationship to other scientists both because they are in competition with them and because they want their esteem, so they cooperate with them. Scientists want other scientists to accept their ideas, but the acceptance of new ideas is more complex than just judgements about verification or falsification. Scientists do not like to give up their ideas or accept those of others without good reasons.

  Compared to the creative arts, science is ultimately an anonymous enterprise. Scientists add to the body of scientific knowledge, and it is in essence irrelevant that some are made temporarily famous through a discovery, for in the long run their ideas are incorporated into a common body of public knowledge. For example the invention of the calculus, in the seventeenth century, revolutionized mathematics and is the basis of all m
odern applied mathematics and engineering. But, other than historians, no one is interested that it was discovered independently by Leibniz and by Newton, who fought bitterly about priority, and no one would now read their almost impenetrable papers. As ideas become incorporated into the body of knowledge, the discoverers, the creators (of whom there may be many), simply disappear. Likewise, no one reads Watson and Crick’s original paper if they want to know about DNA, or Darwin if they wish to understand evolution (though it must be admitted that to read The Origin of Species can still be very rewarding). Thousands of scientists have contributed to our understanding of DNA and evolution, and their knowledge has been distilled into general and specialized textbooks. Scientific papers have a short life – even really important ones are no longer referred to after a few years. Scientists cannot work in isolation, because the enterprise is essentially cumulative.

  Compare all this with the arts: for painters, novelists and poets, the original creation is all-important. The artist does not contribute to a common enterprise; the artist’s work is not assimilated into a larger body and its essence is its individuality. For the scientists, by contrast, the aim is to get others to accept their ideas, to obtain consensus. As the mathematician David Hubert once expressed it, the importance of a scientific work can be measured by the number of previous publications it makes it superfluous to read.

  A peculiar feature of science which has important implications for the social behaviour of scientists is that discoveries can be made only once. Once a particular discovery has been made, others cannot make it – though it will, of course, open up new possibilities. The general theory of relativity or evolution by natural selection or the structure of DNA cannot be discovered again. Shakespeare’s Hamlet was not a discovery: it didn’t stop others from writing plays even on related subjects. But Watson and Crick’s discovery of the structure of DNA was quite different: once they had discovered the structure, no one else could do the same and a major problem had been solved. Writing Hamlet solved no problems in this sense. Knowing the structure of DNA opened up an enormous field of research and there were other discoveries to be made – indeed, several more Nobel Prizes have since been won for work on DNA. Watson and Crick were themselves building on the accumulated knowledge provided by many other workers. And there is another important feature: if Watson and Crick had not discovered the structure of DNA, one can be virtually certain that other scientists would eventually have determined it. With art – whether painting, music or literature – it is quite different. If Shakespeare had not written Hamlet, no other playwright would have done so.

  For all these reasons, the strategy that scientists adopt in relation to their work and their colleagues is very different from that of artists. Artists are not subject to the criteria of validation and falsification that are central to the social activity of scientists. Artists may plagiarize, but they cannot falsify in the same sense as scientists can: they cannot cheat.

  We are thus confronted with the ‘sociobiology’ of science. Sociobiology is defined as ‘the systematic study of the biological basis of all social behaviour’, and sociobiologists ask questions as to why animals engage in the forms of behaviour that are observed. What strategy should scientists adopt to maximize the success of their ideas, which are, in a sense, the scientists’ children? Also, how should scientists behave with respect to their subject and their colleagues so as to be most successful? These are questions that sociobiologists ask about animals. An often discussed question, for example, is the basis of altruism. For animals, the answer is framed in terms of the advantage it gives to the survival of the animal’s genes. (As the geneticist J. B. S. Haldane perceptively remarked, he would lay down his life if it saved the lives of eight cousins, since that would ensure a better chance for the survival of his genes.) Other questions relate to the investment that animals make in producing and rearing their offspring, which clearly has resonance with scientists’ devotion to their ideas. Yet other ideas deal with competitiveness between animals and the extent to which, within a species, aggressiveness pays off. This gave rise to the important concept of an evolutionary-stable strategy: the strategy adopted by animals of the same species, with, say, metaphorical hawklike and dovelike characters competing for the same resource, such that it could not be displaced by a better strategy.

  Scientists cannot be treated as idealized animals and it is not legitimate to apply a sociobiological analysis to them. However, it does not seem unreasonable to assume that scientists wish to maximize the success of their ideas. Success can be thought of in terms of selection of their ideas by the community in the field in which they work. This is associated with personal success, which involves advancement in relation to jobs, promotion, praise by one’s peers, money for supporting research, some personal financial rewards and, on occasion, prizes. The value to the individual scientist of each of these rewards will vary, but they are closely interlinked and can be lumped together under the rubric of esteem by other scientists.

  In order to promote the success of their ideas, and hence themselves, scientists must thus adopt a strategy of both competition and collaboration, of altruism and selfishness. Each must balance his or her behaviour, in relation for example to sharing information, in these terms. Artists are confronted with such choices to a much lesser extent. Another special feature that characterizes modern science is the enormous number of collaborative research projects. Single-author papers are now a rarity in the scientific literature. Many papers have four or five authors, and in some cases in subatomic physics the number of names attached to the paper may be fifty or even more.

  It may not be unreasonable to think that the strategy scientists adopt is one that is entirely competitive and self-seeking, since there are, in a sense, only a limited number of golden discoveries to be made in any one field at any one time. Once that ‘gold’ has been claimed, the other ‘prospectors’ are left penniless. But this view ignores the intensely cooperative nature of the scientific enterprise. Scientific success is not only about making discoveries about nature but about persuading other scientists of the validity of your ideas. In the process, one has to be part of a community which, with time, has developed quite a rigorous set of unstated norms for acceptable behaviour. Included in these norms are the ideas that science is public knowledge, freely available to all; that there are no privileged sources of scientific knowledge – ideas in science must be judged on their intrinsic merits; and that scientists should take nothing on trust, in the sense that scientific knowledge should be constantly scrutinized. In addition, there have arisen a set of rules for the sharing of materials. In molecular biology, for example, once a paper is published which contains information on specific genes or proteins, then the authors are duty-bound to provide materials from their laboratory which enable other workers to pursue work on those genes or proteins. They may, of course, require that future research be collaborative, but it is not acceptable for them to keep all the materials for themselves.

  There is an almost prurient fascination in the media with both competition and fraud in science. It is as if these contaminate the purity of science, and they are viewed almost in the same way as someone of note in the religious world being discovered to be wholly immoral. Competition between scientists is regarded as, at the very least, indecent – quite alien to the image of the ivory-tower scientists pursuing knowledge for its own sake. But this is to fail to understand the special nature of the scientific enterprise and how scientists interact with one another. Scientists have to adopt a special strategy in order to be successful. They have both to compete and cooperate. Carl Djerassi, the chemist who first synthesized the birth-control pill, is one of the very few distinguished scientists who have written a novel about science; it is not surprising that he made fraud and the Nobel Prize its central themes. J. B. S. Haldane is reported to have said that his great pleasure was to see his ideas widely used even though he was not credited with their discovery. That may have been fine for someo
ne as famous and perhaps noble as Haldane, but for most scientists recognition is the reward in science.

  There are cases where scientists have plagiarized the work of others and where results have been manufactured to support a particular hypothesis. It is inevitable that among the many thousands involved in scientific research there should be a small number who behave dishonestly and quite against the accepted norms. In several cases even distinguished scientists have been involved, by putting their name on a paper containing fraudulent results obtained by a junior colleague. They may, in some instances, not have had the time to examine the primary data in detail and so also have been deceived, since it is one of the dangers of ever-increasing collaborative work that scientists must have complete trust in the colleagues with whom they collaborate. For the functioning and the image of science, fraud is inexcusable; but for the advancement of science in the long run it really does not matter much, because it is so rare. Moreover, many respectable papers will themselves turn out to be wrong or irrelevant. A fraudulent result in an important area will soon be discovered when others fail to replicate the work, and this is exactly what has happened in several cases. More subtle is the scientist’s desire to ‘massage’ the results so as to support a viewpoint. Distinguished scientists have been accused of doing just this. Mendel’s results that established his ideas on inheritance were, it is claimed, just too good to be believable. The desire to present one’s results in the best light can be difficult to resist. In the example to be given below, Millikan will be seen to have been highly selective about which results he published when measuring the charge on the electron.