The Unnatural Nature of Science

by Lewis Wolpert

  The physicist John Barrow has listed a further set of assumptions:

  • There is an external world separable from our perception.

  • The world is rational: ‘A’ is not equal to ‘not A’.

  • The world can be analysed locally – that is, one can examine a process without having to take into account all the events occurring elsewhere.

  • There are regularities in nature.

  • The world can be described by mathematics.

  • These presuppositions are universal.

  These assumptions may not be philosophically acceptable, but they are experimentally testable and they are consistent with the ability of science to describe and explain a very large number of phenomena.

  Has philosophy in fact influenced science? Many of the leading physicists at the beginning of the century were well schooled in philosophy, and the German physicist Ernst Mach had strong views on the nature of science. However, an interest in philosophy was just part of the ‘normal’ intellectual cultural environment in Germany at that time. Today, things are quite different, and the ‘stars’ of modern science are more likely to have been brought up on science fiction. They view the philosophy of science as, in Holton’s phrase, a ‘debilitating befuddlement’, and it has been remarked that the physicist who is a quantum mechanic has no more knowledge of philosophy than the average car mechanic. Not only are most scientists ignorant of philosophical issues, but science has been totally immune to philosophical doubts. In this century at least, science has generally been wholly unaffected by the philosophers of science, though some Nobel laureates, like the neurophysiologist John Eccles, claim that their work has been greatly influenced by Popper. Another possible exception is psychology, where there is a link because psychology is closely related to problems that have origins in philosophy, such as the nature of knowledge and how the brain thinks.

  Even distinguished philosophers of science like Hilary Putnam recognize the failure of philosophy to help understand the nature of science. They have not discovered a scientific method that provides a formula or prescriptions for how to make discoveries. But many famous scientists have given advice: try many things; do what makes your heart leap; think big; dare to explore where there is no light; challenge expectation; cherchez le paradox; be sloppy so that something unexpected happens, but not so sloppy that you can’t tell what happened; turn it on its head; never try to solve a problem until you can guess the answer; precision encourages the imagination; seek simplicity; seek beauty … One could do no better than to try them all. No one method, no paradigm, will capture the process of science. There is no such thing as the scientific method.

  Just because it may be difficult to define exactly what is meant by ‘life’, for example, and thus whether or not certain machines or computer programs are or are not alive, that in no way means that there is no distinction between living and inanimate systems. Science is a complex social process, and no simple-minded description in terms of Kuhn’s paradigms or Popper’s falsification will provide an adequate description. The demarcation problem is real only in the sense that science is rich, varied, heterogeneous and complex. Its edges may be blurred, but the core is solid.

  For reasons that may be connected with the peculiar nature of science, we have a situation in which the ideas of Karl Popper and Thomas Kuhn seem to be far better known among non-scientists, especially those working in the humanities, than the ideas of almost any contemporary scientist. Another widely quoted philosopher of science is Paul Feyerabend, who, in his book Against Method, urges his readers to ‘free society from the strangling hold of an ideologically petrified science just as our ancestors freed us from the stranglehold of the One True Religon’. All these ideas would not matter if they remained in the philosophical domain, but unfortunately they are sometimes used to undermine the scientific enterprise itself, on the grounds that if science’s attitudes towards truth and the role of evidence are philosophically untenable then the whole of science is also suspect.

  A less philosophical and more pragmatic approach to understanding the nature of science is to examine how scientists do their work. It would be of great interest to know more about the social interactions between scientists and to see how these interactions, and also interactions with other parts of society, influence how scientists work. Scientists do not work in a cultural or social vacuum. In Chapter 5 I raised briefly the issue of the ‘sociobiology’ of science, and for a formal analysis, one might look to those sociologists who study science.

  The more traditional sociologists of science, as represented by the work of Robert K. Merton, sought to understand social processes in science and tried to define, for example, the procedures that most scientists accept and adopt in their work. Older sociologists like Emile Durkheim even excluded science from the sociology of knowledge, on the grounds that it was a special case. I am a great admirer of the sociologist Max Weber, and it is reassuring to know his attitude to science and what he thought it meant: ‘It means the knowledge or belief that if one but wished one could learn it at any time. Hence, it means that principally there are no mysterious incalculable forces that come into play, but rather that one can, in principle, master all things by calculation.’ Weber recognized science’s explanatory power: one no longer needs ‘recourse to magical means’. He recognized both the power of the rational experiment and that science does make presuppositions, like accepting the rules of logic.

  More recently, however, some sociologists have identified themselves with what is called the Strong Programme of the Sociology of Science. This approach takes the view that the very nature of belief and rationality in science requires explanation and the same sort of analysis as non-rationality. No distinction appears to be drawn between good and successful science and what most scientists would regard as second-or third-rate work. Those who hold to the Strong Programme believe that all knowledge is essentially a social construct and so all science merits the same attention. All knowledge is regarded as relative to the social environment in which it is constructed. This new-style sociology, which claims the relativity of science, is called the sociology of scientific knowledge and is known by the acronym SSK.

  The SSK approach to science is as follows. It feels bound to ask whether a belief is part of the routine cognitive and technical competence handed down from generation to generation and is supported by the authorities of the society. Is it transmitted by established institutions or supported by accepted agencies involved in social control? Is it bound up with patterns of vested interest? Does it have a role in furthering shared goals, whether political or technical or both? What are the practical and immediate consequences of particular judgements that are made with respect to the belief? The most striking feature of this approach is that it says nothing about the belief’s contribution to understanding, its correspondence with reality or its internal logical consistency.

  The SSK programme on one central point is explicit: ‘we should abandon the idea of science as a privileged or even a separate domain of activity and enquiry.’ For such sociologists of science, like Steve Woolgar, the certainties about science – that is, the old beliefs in its cultural uniqueness – have gone. Relativism is strongly defended by proponents such as Barry Barnes and David Bloor, who claim that the real threat to a scientific understanding of knowledge and cognition is posed by those who oppose relativism and who grant certain forms of knowledge, such as science, a privileged status. They hold strongly to what they call their equivalence postulate, which is that all beliefs are on a par with one another with respect to the causes of their credibility. For them the incidence of all beliefs, without exception, calls for empirical investigation, and beliefs must be accounted for by finding the specific local causes for their credibility. Such strong statements make one wonder whether they accept the reality of everyday items like cups of tea.

  Even statements that 2 + 2 = 4 are treated as legitimate targets for sociological questioning, and so too are logic
and rationality. It is claimed that, ‘By looking at reason and logic, we find that reason, logic and rules are post-hoc rationalizations of scientific and mathematical practices, not their determining force.’ In other words, SSK is an extremely ambitious programme with very large, not to say extravagant, claims. It is thus necessary to examine some of the evidence on which the claims are based and what new insights this approach has given us. One can say at once that there have been very few SSK studies of mathematics and logic, and it is not unreasonable to ignore SSK claims to success in those disciplines. But in biology and physics there have at least been a number of studies, and I will describe some of these.

  When they discover a new law, a new phenomenon, a new object, scientists believe that the discovery relates to an existing external world. To count as a discovery, their findings should be novel and preferably important. A very different view is taken by those who adhere to the SSK programme, who wish to emphasize that it is the social content that determines whether or not something is called a discovery. Instead of asking about the characteristics of scientific discovery, the newer perspective tends to ask: given that scientists’ actions and beliefs can be organized in various ways, by what interpretive practices is science made to exemplify a certain kind of rationality?

  The discovery by Mendel of the fundamental laws of genetics has been examined by Augustine Brannigan within this sociological framework. Contrary to the widely held view that Mendel’s paper of 1866 was ignored until it was rediscovered in 1900, Brannigan argues that it was less the content of Mendel’s paper than the context within which it appeared that led to it being hailed as a discovery in 1900. That context, Brannigan wishes to emphasize, was related both to a priority dispute between the geneticists Carl Correns and Hugo de Vries and to disputes about its relevance to evolution. There is, in fact, evidence that Mendel’s paper was not completely neglected when it was first published but was quoted a number of times, though no one gave it any prominence or suggested that it was significant. There is some uncertainty as to where and when de Vries, who arrived at laws similar to Mendel’s, first came across Mendel’s work. Whatever the case, when Correns, who had also discovered similar laws, received a reprint of de Vries’s article on his discoveries, on 21 April 1900, he at once sent off a paper in which he announced results similar to those of de Vries but gave priority to Mendel. It is not unreasonable to suggest that he did this in order to resolve a priority dispute. It is also reasonable to see Bateson’s vigorous support for Mendel reflecting in part his belief that Mendel’s results supported his views that discontinuous variation was the key feature in evolution. Thus, in Brannigan’s view, Mendel’s fame is due less to his science than to how this was used by others to promote their own positions.

  To be identified with a discovery is prestigious in science and is a major reward for a scientist. We should not be surprised that Correns may have used Mendel’s work in order to prevent de Vries being given priority. Priority disputes are indeed a common feature of science, and the assigning of credit certainly reveals complex social interactions. Merton has described what he calls the Matthew Effect – scientists who are already eminent get a disproportionate amount of credit, at the expense of those less well known. This may be due to a scientist’s reputation making a discovery more visible and giving it respectability. The sociologists are correct in claiming that the success or failure of a scientific idea may, initially at least, be due to much more than ‘pure’ scientific criteria.

  Mendel’s discovery, later confirmed by de Vries and Correns, showed that inheritance of characters could be understood in terms of the transmission of discrete characters that maintained their identity from generation to generation. These ‘discrete characters’ were only later to be identified as genes. A major feature of Mendel’s work was to allow the study of inheritance to be expressed mathematically and to make it possible to state laws as to how a character would be inherited in subsequent generations. From a scientist’s viewpoint, Mendel’s approach was new and fundamental. As the molecular biologist François Jacob points out, Mendel’s achievement was similar in its power to the introduction of statistical mechanics into physics: he concentrated on a small number of characters with sufficiently striking differences for discontinuity to be introduced. With Mendel, biological phenomena acquired the rigour of mathematics. This was not by accident, for in the introduction to his paper Mendel says ‘so far, no generally applicable law governing the formation and development of hybrids has been successfully formulated’ and he points to the difficulty of the task. He then speaks of his study in terms of the large number of experiments required as well as the necessity to arrange the different forms ‘with certainty according to their separate generations’. He very carefully chose his experimental material with this in mind. The distinguished geneticist R. A. Fisher some time ago remarked that people find in Mendel’s paper whatever they are looking for. Yet Brannigan argues that Mendel was not ahead of his time and his reputation was modest because his identity with his contemporaries was so complete.

  What the sociologists do not illuminate is why no one had done Mendel’s classic experiment before. It is a similar problem to that posed by the enormous gap between Aristotle and Galileo with respect to thinking about motion. There is no doubt that social factors must play a role, but, unless they believe that Mendel and Galileo made significant discoveries, it is unlikely that sociologists will analyse them with respect to this question. If they treat all science in so dispassionate and detached a manner as not to single out great achievements, they may be missing the core of the problem. Scientific discovery cannot be judged only in social terms but must also take into account the new understanding or knowledge it provides.

  Another example of historical analysis relates to phrenology – the so-called science of interpreting brain function and capability from the size and shape of the head. Phrenology began with a Viennese, Dr Franz Gall, who, with his assistant Johann Spurzheim, in the late eighteenth century put forward these three main principles: the brain is the organ of mind; it is made up of a number of separate organs, each related to a distinct mental faculty; and the size of each organ is a measure of the power of its associated faculty. Thirty-six faculties were listed, including love of children, blandness, prudence and dignity. In Edinburgh at the beginning of the nineteenth century, an enthusiastic disciple of Spurzheim was George Combe, an eclectic scholar with no formal scientific training. Opposition to phrenology in Edinburgh came both from the anatomists at the university and from those who taught the philosophy of the human mind, particularly Professor Sir William Hamilton. The Scottish moral philosophers held the view that mind is an immaterial entity which is both single and indivisible, and this contrasted sharply with the view of the phrenologists. The philosophers were critical of the thirty-six faculties, and asked why there was not among them, for example, a special faculty for love of horses. (However, that was perhaps not quite fair, for there was a region associated with the love of animals.) The phrenologists and those in the university were in vigorous dispute between 1803 and 1828, when the case for the phrenologists collapsed because of evidence relating to observations on the brain. A crucial issue was the frontal sinuses, since, according to Hamilton, they vary greatly and mask the development of about one-third of the so-called phrenological organs. Another point of criticism was that neither the phrenological organs nor the associated faculties were clearly defined – almost any observation could be confirmed. Hamilton also found that the size of the cerebellum – the supposed organ of sexual activity – was larger in females, completely contrary to the phrenologists’ expectation.

  However, this quite conventional view that scientific evidence resolved the issue has been criticized because of its neglect of the social dimension of the controversy. It has been suggested that the viewpoints of the opposing factions were incommensurable, and that the war between the phrenologists and the moral philosophers should, in large measure, be treated
as a conflict between university professors and those exposed to university teaching, on the one hand, and those not associated with the university, on the other. There was, for example, considerable support for phrenology from the lower, middle and working classes. The phrenologists’ emphasis upon empirical methods in mental science reflected their socially based anti-elitism, and their great claim was that phrenology was founded on observations which anyone could make and so enabled the ordinary man to discover the truth. They thought, mistakenly, that science is a common-sense activity.

  The sociological viewpoint does not regard the difference in the intellectual bases of the two sides as important, even though untrained amateurs have rarely made a significant contribution to any important scientific controversy. Moreover, the conflict was not really within science but between those in science and those outside science. It seems that the SSK analysis is based on the belief that anyone can do science and that the difference in training would not be relevant. Of course, in the end, it is only the fit with reality that matters.

  It is curious what topics the SSK have chosen, or not chosen, in order to study relativism in its historical context. Hardly any of the major achievements of, for example, modern biology – the gene and DNA, electrophysiology, biochemistry and so on – has been the subject of an analysis which supports the relativists’ case; and among the examples of creativity given earlier it is not easy to see how the discovery of messenger RNA (Chapter 4) or the structure and role of DNA (Chapter 1) could merely be social constructs. They could only appear to be so to someone ignorant of the complex science involved.