The Unnatural Nature of Science Page 13
From that moment Epstein dropped everything else and started working on the tumour. ‘Slogging’ would be a better description. Material from tumours was flown in from Africa, and he and his group used all the standard procedures for isolation of viruses. All of them failed: the results were, without exception, negative. But he didn’t enter Holton’s graveyard. He and his team continued the search and tried to grow the tumour cells in culture. Again, failure was complete. Failure continued for over two years; but, although all the laboratory evidence was against the idea that the tumour was caused by a virus, they persisted. ‘But it had to be right. It just had the feel of being right. And that’s why one carried on.’ Then one wintry Friday afternoon a sample arrived from Africa which was cloudy and looked contaminated with bacteria. But Epstein examined it under the microscope and saw that it was cloudy because the tumour had broken up into huge numbers of single cells. Immediately he was reminded of an American group who grew tumours not as lumps, as he had been trying, but by breaking the tumour up into single cells. So he tried to grow the tumour as single cells, and this worked. This was the breakthrough, and soon after he identified a virus growing in the cells – the Epstein–Barr virus had been discovered.
Some final cases come from Einstein’s work. Popper has quoted Einstein’s statement that ‘The general theory of relativity will be untenable if the prediction it made about the gravitational shift of spectral lines were not observed.’ But Einstein stuck to his theory even though the prediction was not confirmed during his lifetime. The other example is a very famous case of prediction: Einstein’s prediction, again from the general theory of relativity, of the gravitational bending of light. This was confirmed by an English expedition led by Arthur Eddington to observe the eclipse in 1919, and the results from the eclipse created enormous publicity both for relativity and for Einstein. But for Einstein the results seem to have been of much less importance. According to his student Ilsa Rosenthal-Schneider, who was with him when Eddington’s cable arrived announcing that measurements had confirmed the theory, Einstein remarked, ‘But I knew that the theory is correct.’ What, she asked, if the prediction had not been confirmed. ‘Then I would have been sorry for the dear Lord – the theory is correct.’ This confident view was again expressed later. ‘I do not by any means find the chief significance of the general theory of relativity in the fact that it has predicted a few minute observable facts, but rather in the simplicity of its foundation and in its logical consistency.’ And Eddington himself even stated that one should not ‘put overmuch confidence in the observational results that are put forward until they have been confirmed by theory’.
From examining the history of the field following the confirmation, the historian of science Stephen Brush has concluded that the main value of a successful forecast is favourable publicity: the results from the eclipse put relativity theory much higher on the scientific agenda and provoked other scientists to try to give plausible alternative explanations. But light bending could be considered as reliable evidence for Einstein’s theory only when those alternatives failed, and then its contribution was independent of its discovery.
Publicity may seem a strange virtue to ascribe to a scientific experiment but, as we now recognize, that is precisely what scientists need for the survival of their ideas. Science is partly about consensus, and if one’s ideas are not widely known they may be neglected. As will be seen, it is social issues of this type that have led sociologists of science to question whether science is anything more than a social construct.
It is unfashionable among historians of science to take what Herbert Butterfield called a Whig view of history – to interpret the past in terms of progress, as opposed to seeing it as a series of events that have no particular direction. But it is precisely in this respect that science, once again, is special: for the history of science is one of progress, of increased understanding. Of course there have been errors, innumerable social influences, but, given a reasonable time scale, depending on the subject, progress has been a characteristic of science over the last few centuries. And in the last fifty years the progress in, for example, understanding biology at the molecular level has been astonishing. Science is progressive in that the truth is being approached, closer and closer, but perhaps never attained with certainty. But very close approximation can be a great achievement and is infinitely better than error or ignorance. Philosophers are much involved with such problems.
6
Philosophical Doubts, or Relativism Rampant
If science is an unnatural process, quite different from ordinary thinking, it might be thought possible to state clearly what the nature of science is and to define scientific method. If only this were so! In fact, defining the nature of science and scientific method with rigour and consistency turns out to be extremely difficult. It is even doubtful that there is a scientific method except in very broad and general terms. Perhaps scientists themselves have helped to create the illusion that method in science is highly ordered, for they write almost all their papers as if there were a scientific method. There is a format of ‘Introduction’ followed by ‘Methods’ then ‘Results’ and finally the ‘Discussion’. But, as Peter Medawar pointed out, the scientific paper is a kind of fraud, for its neat format bears no relation to the way in which scientists actually work: imagination, confusion, determination, passion – all the features associated with scientific creativity have been purged from it.
For scientists, defining the nature of science is of only marginal interest, for it has no impact on their day-to-day activities. For philosophers of science, and for some sociologists, by contrast, the nature of science and the validity of scientific knowledge are central problems. These observers have found the nature of science puzzling, and some have even come to doubt whether science is, after all, a special and privileged form of knowledge – ‘privileged’ in that it provides the most reliable means of understanding how the world works. While providing no real threat to science they have become an increasingly vocal group, with an unfortunate influence on the study of science and its history.
It is the very progress of science that presents the basic problem. If science provides the best understanding of the world, how should one regard, for example, the ideas about phlogiston that were held before the discovery of oxygen and the understanding of its role in combustion? If those who believed in phlogiston could be so wrong, how can we be sure that the same upheaval will not occur in current areas of science? The whole history of science is filled with new discoveries and the overthrow or modification of ideas which were held to be true. So in what sense, then, is scientific knowledge a true description of the world, and what right have we to call it ‘privileged’?
The vast majority of scientists would not be interested in such problems. They would probably just argue that the older theories were the best available at the time, and almost always some, perhaps many, features of an old theory will be incorporated into its successor. Scientists have to accept the possibility that their most strongly held view may turn out to be wrong, but some concepts have been so widely tested that it is extremely unlikely that they will suffer this fate. Even those who are dubious about the privileged nature of science do not direct their criticisms at the results of science itself – that the earth goes round the sun, that water is made of two hydrogen atoms and one of oxygen, or that DNA is the genetic material. The attention of the philosophers, rather, is focused on the nature of scientific knowledge and how it is acquired.
The philosopher Willard Quine, for example, argues – contrary to the experience of scientists – that scientific theories are never logically determined by data, so there are always, in principle, alternative theories that will fit the data more or less adequately. He also argues that any theory can be saved from being falsified by modifying the criteria that are used to decide what counts as a good theory. On this view, widely held by philosophers, any set of empirical observations can always be explained by an infinite
number of hypotheses. This view is true only if the hypotheses differ in some very minor manner, like the difference between two numbers at the 100th decimal point. In practice scientists are not concerned with such minute differences except in cases where they will have a real impact on their theories and predictions. Scientists are concerned not with absolute truth but with theories that provide understanding of the phenomena involved. The criteria for a good theory have already been mentioned (Chapter 1), but it seems that it is up to those who really believe that an infinite number of theories are possible to demonstrate this by providing satisfactory alternatives to classical Newtonian mechanics or to genetic theory. As yet none are forthcoming, and anyone who has tried to construct even a simple quantitative theory to account for some observations will know just how difficult it can be even to get one model to work.
Kuhn’s views on incommensurability, with his emphasis on social processes determining the acceptance of a theory (Chapter 5), can lead one to a relativistic view of science. For if there really is no rational way of choosing between rival theories, for choosing between one paradigm or theory and another, then it seems that science may be a mere social construct and that a choice of scientific theories becomes like fashion, a matter of taste. If this were really true then scientific ideas would be merely a reflection of a particular set of social and cultural conditions, and science could not merit the so-called privileged position assigned to it. But such a conclusion is not valid. Although social processes play a role in science, scientists change theories because the new ones provide a better correspondence with reality; because, like Darwin’s theory of evolution, they provide a better explanation of the world. While the initial stages of acceptance of one or other of competing theories may have a strong social aspect that involves fashion, power groupings and so on, the main criterion will eventually be how well the theory explains the phenomena.
The emergence of molecular biology is a clear example of a scientific revolution, but not in the way that Kuhn would have us believe. The members of the biological scientific community were not confronted with rival and incommensurable theories between which they found it difficult to choose: rather, scientific advances gave rise to a new set of ideas which completely changed the mode of thought or, in Kuhn’s term, the paradigm. Instead of thinking about cells in terms of energy and metabolism, the paradigm changed to information, so that, for example, the questions that were now asked about proteins – key constituents of cell function (Chapter 1) – were not about the source of the energy to make them but about the information for the ordering of the amino acids. Of course there was some resistance to the new ideas and the molecular biologists were evangelical in trying to persuade others. They undoubtedly also used rhetoric. But the evidence from the structure of DNA and other key discoveries was so persuasive that almost everyone – certainly the young – got caught up in the excitement of what is clearly a new age for biology and one which has brought spectacular advances. As the American evolutionary biologist Ernst Mayr has emphasized, it is probably true that philosophers of science have ignored advances in biological science, to their cost. By almost always drawing their examples from physics, they have missed out on revealing examples of scientific progress in other fields, particularly molecular biology.
One of the widely quoted criteria for characterizing science has been Karl Popper’s emphasis on falsification rather than verifiability. However, the importance of falsification was also made clear by others like the French biologist Claude Bernard in his book on experimental medicine in 1865. In real life, scientists often do not conform to this formula for doing science, as we have seen in Chapter 5, but there are also some philosophical problems in this approach. It is claimed that verifying a theory is a rather weak way of establishing its validity, and so it becomes difficult to define the conditions under which a scientific theory can be said to be true. Take the trivial hypothesis that all swans are white, or that sodium burns with a yellow flame – ‘trivial’ because, although they are often used as models for thinking about the ‘truth’ of scientific ideas, there are not really hypotheses or theories but are just simple correlations from observations and are totally lacking in the richness and explanatory powers of real theories. Popper has argued that the truth or otherwise of these statements cannot be guaranteed on the grounds that they are supported by numerous observations, and so has led the attack on the so-called ‘inductive’ basis for verification.
If scientists have made thousands of observations that confirm that all swans are white or that sodium burns with a yellow flame, this is, Popper says, no reason to believe that the statement is true. As demonstrated long ago by Hume, induction – inferring relationships from repeated instances – is logically untenable. By contrast, only negative instances – falsifications – provide evidence that can be trusted. If one swan is found that is black, then the hypotheses that all swans are white is falsified definitively. ‘… there is no more rational procedure than the method of … conjecture and refutation; of boldly proposing theories; of trying our best to show that these are erroneous; and of accepting them tentatively if our critical efforts are unsuccessful,’ says Popper. But would one really give up one’s lifelong experience on seeing just one black swan? As described earlier, many scientists would not – and would be unwise to do so – for how could one be sure it was really a black swan? Would one not want several examples? If so, one is back with induction. This approach thus avoids the whole question of how scientists actually decide whether or not a theory is refuted or verified. But at least its emphasis on bold conjecture points to a feature of science on which all scientists would agree: science is not just the growth of organized factual knowledge but is a creative endeavour which aims at understanding (Chapter 5). On the negative side, Popper’s argument only partly helps define what science is, for, although scientific ideas must be falsifiable, just because ideas are falsifiable does not mean that they are part of science. Absurd ideas are falsifiable but are not part of science, as will be discussed in Chapter 7.
Scientists have an unstated set of criteria for choosing one theory rather than another – and these, moreover, encapsulate some of the main aims of science. In addition to dealing satisfactorily with the phenomena it tries to explain, the theory should have as broad a scope as possible and so encompass a wide range of phenomena. It should be able to predict new relationships and offer scope for further development. It should also be as simple as possible, with a minimum number of hypotheses.
Many of the problems associated with the philosophy of science have their roots in philosophy in general and are not peculiar to science. They are problems relating, for example, to the nature of reality and truth. The existence and nature of ordinary objects such as tables and chairs are held by some philosophers to be problematical. Some philosophers would accept their existence as real, some would deny their real existence, and others would claim that they reflect only external influences on our senses. Thus philosophers are divided among schools of thought whose descriptions – materialism, metaphysical realism, objectivism and so on – hint at their preferred position. But these are the problems of philosophers, and we should not become confused through their inability to deal satisfactorily with the nature of reality and whether or not there is a real world. It seems that Ludwig Wittgenstein may have been saying just this: ‘What we find out in philosophy is trivial: it does not teach us new facts, only science does that. But the proper synopsis of the trivialities is enormously difficult and has immense importance. Philosophy is in fact the synopsis of trivialities.’
More generally, if philosophers are correct about the essentially unknowable nature of the world, then this is a problem relevant not just to science but to all knowledge. It must presumably apply to statements like ‘The sun always rises in the east’ and ‘Pigs cannot fly’. For those philosophers who live in a world where they really have doubts about reality, their world is even more unnatural than the world of scientific idea
s, but in a quite different way. I have no doubt about the difficulties that philosophers face or the ingenuity they have shown in dealing with such problems. I do, however, strongly deny the relevance of these problems to science. It is essential not to mix up the philosophers’ problems in dealing with truth, rationality and reality with the success or otherwise of science. My own position, philosophically, is that of a common-sense realist: I believe there is an external world which I share with others and which can be studied. I know that philosophically my position may be indefensible, but – and this is crucial – holding my position will have made not one iota of difference to the nature of scientific investigation or scientific theories. It is irrelevant.
It is not my intention to argue that science has a claim to absolute validity – indeed, one of the main features of science is that its adherents must be prepared, in principle, to change their minds in the face of evidence. I also must accept that scientists operate within a framework of usually unstated assumptions that the physicist and historian Gerald Holton has called themata. The themata underlie – even underpin – the scientific endeavour and are independent of its subject-matter, experiments and analyses. Copernicus, for example, believed that nature is God’s temple and that humans can discern its design and its constant laws – an idea that resonates through Galileo and Newton. Two themata that pre-exist in modern science are the ideas of simplicity and beauty. To these, in physics at least, is coupled the conviction that, as the physicist Steven Weinberg has said, ‘we will find the ultimate laws of nature, the few simple general principles which determine why all of nature is the way it is …’ This echoes Newton, who, after showing how his theory of gravity enabled him to deduce, in detail, the motion of the planets, wrote, ‘I wish we could derive the rest of the phenomena of nature by the same kind of reasoning.’ And Einstein taught that the noblest aim of science was to grasp the totality of physical facts, leaving out not a single datum of experience. How unnatural, in a way, these themata are: for what in the myriad and varied events of our daily life gives a hint that such a unity – beautiful and simple – might exist?