Unmaking mysteries

Sir Peter Medawar is the most distinguished biologist in Britain today. His work on immunology during the 1950s is the inspiration of all modern transplantation surgery, and was judged worthy of a Nobel Prize. In addition to his technical work, he also writes and lectures on science to a broader audience, and with more grace and wit than any one else who has tried his skill in that field. He has held many of the high offices open to a scientist in this country; he has delivered with uniform excellence the most prestigious lectures sponsored by the universities and learned societies; and, as the readers of this journal and its New York counterpart know, he is outstandingly the best reviewer of scientific books in the English-speaking world.

But despite all that, Medawar’s essays are less well-known in literary circles than they deserve. They are widely known among scientists, wherein his wittier put-downs are recalled with chortling delight, stumblingly misremembered in the fashion of Monty Python as retold by undergraduates. His audience need not be so confined. One might fear that they would be hard work – ‘No man,’ as Dr Johnson remarked, ‘reads a book of science from pure inclination’ – but they are always easy going, and require no esoteric knowledge. One might fear that the praise lavished upon his style is only relative to the prevailing standards of literacy among scientists: but they are sufficiently well-written, intelligible, interesting and amusing to stand comparison with the very best in any field. Now, with the publication of Pluto’s Republic, the last excuse for ignorance has been extinguished. Medawar’s essays have previously appeared in three anthologies, The Uniqueness of the Individual (1957), The Art of the Soluble (1967), which was the best of the three, and The Hope of Progress (1972), all of which are now out of print, and have been available only to those sharp-eyed collectors who were prepared to hunt them through the random classifications and dusty shelves of second-hand bookshops. That skill is no longer required. The new volume contains all of The Art of the Soluble, and most of The Hope of Progress. To these are added the contents of a slim volume on Induction and Intuition in Scientific Thought, a few more recent, uncollected essays, and one, on ‘Expectation and Prediction’, that is, in its present form, new. The Uniqueness of the Individual, I may add, was reprinted, with a new introduction, by Dover Books in 1981.

As for their themes, we can let the author speak for himself. ‘In so far as the essays collected have a central or recurrent theme, it is in the attempt to answer the following questions: what is science, what kind of person is a scientist, and what kind of act of reasoning leads to scientific discovery and the enlargement of the understanding? I seldom put these questions directly, and answer them only incompletely and bit by bit.’ More often than not, he approaches the questions backhandedly, by means of the negative instance. Such, in fact, is the meaning of Pluto’s republic, which Medawar calls ‘an intellectual underworld’ that ‘we each populate according to our own prejudices’. It is, in other words, Medawar’s Inferno, with the difference that he does not torment his victims.

What a crowd Pluto’s citizens are! Obscurantists and rhapsodic intellecters, futurologists and economic forecasters; inductivists; mystical theologians like Teilhard de Chardin; mystical humanists like Arthur Koestler, who, although ‘a very clever and knowledgable man’, ‘has no real grasp of how scientists go about their work’; there are advocates of such doctrines as historicism, scientism and poetism; there is (after a balanced judgment) Herbert Spencer, whose ‘system of general evolution does not really work’; and then IQ psychologists, who ‘give the impression of being incapable of learning anything from anybody’; and everyone else who doubts the hope of progress, such as Francis Galton, for the ‘air of almost exultant scorn in his description of the uselessness of a man’s trying to better himself beyond the degree of his innate capacities’; there are other kinds of psychologists too – psychoanalysts, with their system that combines ‘conceptual barrenness with enormous facility of explanation’, and existential psychiatrists, who, ‘because some families may create an environment conducive to mental disorder’, would abolish all family life, and would treat madness by constructing a microcosm of ‘understanding’ around it such that madness no longer appears mad: and all of them can meet the editor of Nature, John Maddox, who does not take the prophets of ecological disaster seriously enough, and Norman St John-Stevas, who had the misfortune to let off one of the sillier arguments against abortion in Medawar’s hearing, and is briskly propelled across the Styx on page 24.

Such are the citizens of Pluto’s republic: a hotch-potch of Medawar’s victims from two decades, with certain features in common. Many of them have preyed upon the name of science, in order to gain respectability for their own ideas. In this, however, they have been unsuccessful, since they do not understand the true nature of science: they seize on its superficial features, and imitate its manners. They are seduced by measurement and jargon. IQ psychologists misapply measurement; and Teilhard and Koestler invent their own ridiculous terms, while playing fast and loose with the existing technical vocabulary of science. ‘Teilhard,’ says Medawar, ‘habitually and systematically cheats with words.’ ‘For the most part consciousness is treated as a manifestation of energy, though this does not help us much because the word “energy” is itself debauched; but elsewhere we learn that consciousness is a dimension, or is something with mass, or is something corpuscular and particulate which can exist in various degrees of concentration, being sometimes infinitely diffuse.’ Similarly, ‘as to style, Koestler overdoes it ... We aren’t quite sure when he intends to be taken literally: for example, what about “A concept has as many dimensions in semantic space as there are matrices of which it is a member”?’ People who treat technical terms in this fashion can demonstrate, all too easily, abstract resemblances that do not, in any enlightening sense, exist.

Although Medawar exposes, with relish, the folly of Pluto’s citizens, he has positive purposes too. He includes a minority of essays on scientists who, we must conclude, belong among the warbling birds and fresh pastures of the fields of Elysium rather than in Pluto’s republic. Such are Jim Watson, despite his childish understanding of science; J.B.S. Haldane, despite his political ineptitude; and D’Arcy Thompson. But more importantly, Medawar uses unscientific ideas to explore his own philosophy of science. Practising scientists rarely bother to develop their philosophy; some are called upon from time to time to deliver their thoughts upon the subject, whereupon, according to character, they either repeat orthodoxies or try to be original, only to achieve silliness. None of this matters: it is all, as Medawar would say, part of the comedy of science, and scientists themselves take no notice of what is said. Medawar, however, is an exception. He has directed his critical powers at general questions of the nature of science, and he has things to say that are worth our attention. As he explains the errors of eugenics, of induction, of psycho-analysis, and the rest, his successful veteran ideas accumulate into a body of positive philosophy. His philosophy is practical, realistic, broad-ranging, subtle. It is equipped to destroy errors, but it is not purely reactionary. Although Medawar nowhere states his philosophy as a whole, it does exist, and can be gathered together and expressed.

Medawar accepts what he (and others) call the hypothetico-deductive philosophy of science. His main source for it is Sir Karl Popper. Medawar is well-known as Popper’s leading disciple; his superlatives have even been emblazoned on the cover of Popper’s The Logic of Scientific Discovery to lubricate the sales of that work. But Popper himself is not the subject of an essay. Indeed, Medawar devotes more space to a more distant source, the Victorian Master of Trinity, William Whewell. Popper’s presence here is between the lines, but he is present in all of Medawar’s philosophical reflections.

Since Medawar’s main purpose is to distinguish science from non-science, and as Popper is known to have supplied, in his criterion of falsifiability, a means to exactly that distinction, we might expect it to be Medawar’s main theme. It is not. The effects of that disastrous criterion in my own subject, evolutionary biology, have made me sceptical of its value, so I was interested to see what Medawar would make of it. Sometimes he is approving, and states that, while disproof is possible, proof is not. On another occasion, in a foot-note, he goes back on this. Most often he implicitly ignores it: when he writes of scientific tests, he avoids all mention of falsification, to write of the ‘critical evaluation’ or the ‘critical scrutiny’ of the ‘verifiable deductive consequences’ of hypotheses. If we were to take a single word to stand for the scientific method that Medawar recommends, it would be, not falsification or verification, but criticism.

Criticism, too, is a Popperian theme. But Medawar makes more use of Popper’s distinction between the origin and the test of an idea, between ‘having an idea and trying it out’. The imaginative origin of hypotheses is an important part of Medawar’s philosophy. Hypotheses, he believes, cannot be generated by any mechanical procedure or formulated method: their origin is unanalysable. The scientific method, according to Medawar, proceeds through two stages, and he may strike his victims for a misunderstanding, or violation, in relation to either stage. He does not just parrot ‘unfalsifiable’ at them in the manner of vulgar Popperians: in fact, as we have seen, he silently passes by the whole issue of falsifiability, to provide a more realistic analysis of science. Let us see how he uses his philosophy to deal with four different victims.

The fault of psychoanalysis is an imbalance between imaginative hypothesis and critical scrutiny. It is imaginative all right, but it is insufficiently critical: in place of proper criticism, ‘a lava flow of ad hoc explanation pours over and around all difficulties, leaving only a few smoothly rounded prominences where they might have lain.’ What is wrong is not that their ideas are untestable, but that they will not test them.

Induction is the main competing philosophy of science that Medawar concerns himself with. It is, he believes, the philosophy of science which is accepted by the majority of non-scientists. It is also wrong. It thus needs to be disproved, in public, and Medawar takes on the task. Induction is presented as the theory that scientific knowledge is obtained by the mechanical accumulation of all the facts on a subject, in order to discover, by classification, the ‘laws of nature’. The imagination is thus completely ignored. It implies that there could be what Medawar calls a ‘calculus of discovery’, a sure method that would predictably lead to knowledge. He believes the origin of our hypotheses is unpredictable.

He also blames induction for the separation of ‘reason’ and ‘imagination’ by the poets of Romanticism, from which spring the twin errors of ‘scientism’ and ‘poetism’. The Romantics, he tells us, split reason and imagination apart, and treated them as if they were in conflict: but they are put together again in the hypothetico-deductive system in which imagination and reason are indispensably allied as successive stages in the enlargement of understanding.

Historicism is also based on a mistaken, inductivist idea of the origin of scientific discoveries. He defines it as the belief that ‘a predictive science of history can exist.’ He opposes it with an argument that he attributes to Popper. The future will be influenced by scientific discoveries: but scientific discovery is inherently unpredictable: therefore the future is unpredictable. Medawar illustrates the argument by the case of a discovery, electrical power, that was not foreseen by Marx, and which falsified one of Marx’s predictions.

Medawar also scatters around some more original philosophy. I will confine myself to a single illustration. It concerns the kind of problems that should be worked upon. We may accept that knowledge is obtained by testing hypotheses, but there are hypothses and hypotheses, tests and tests. Which should be chosen? The following passage, which is taken from his review of Jim Watson’s Double Helix, irritated a few literary intellectuals when it was first published. Its point is that it is better to work on problems that matter. Watson, with Francis Crick, discovered the structure of DNA. How much did Watson owe to luck? Medawar replies: not much, ‘except in the trite sense’. ‘In England a schoolboy of Watson’s precocity and style of genius would probably have been steered towards literary studies. It just so happens that during the 1950s, the first great age of molecular biology, the English schools of Oxford and particularly of Cambridge produced more than a score of graduates of quite outstanding ability – much more brilliant, inventive, articulate and dialectically skilful than most young scientists; right up in the Watson class. But Watson had one towering advantage over all of them: in addition to being extremely clever he had something important to be clever about.’

What is the historical context of Medawar’s science and philosophy? He went up to Oxford, in 1932, and read zoology as an undergraduate. Zoology was then presided over (if not totally ruled) by the frail, shy, aging and selfish Professor Goodrich. Goodrich was a perfect fossil from Victorian zoology. He believed that all zoology could be done with a microtome and a microscope, and kept experimental science, as much as possible, out of his course, and department. His zoology was as useless, and (to those few with eyes to see) as beautiful, as a work of art. If Medawar still appreciates the classical beauty of that subject, most obviously in his appreciation of D’Arcy Thompson, it was not enough to satisfy him. After graduation, he walked to the other end of South Parks Road, to the William Dunn School of Pathology, and requested space, to work on tissue culture, from Howard Florey. He was allowed in, and assisted, in a minor way, in the main project of that laboratory, penicillin.

He had set out on the path to research that matters. In so doing, he stands out among the zoologists who graduated from Goodrich’s school. Those that stayed in biological research were more likely to increase our knowledge of the microanatomy or evolutionary relations of the main groups of animals. If I choose J.Z. Young as an example, it is only because he will be known to the readers of this journal: I do not intend any disrespect. J.Z. Young’s career runs parallel to Medawar’s. They were both turned into biology at Marlborough, both went up to Magdalen College, Oxford to read zoology (Young was Medawar’s tutor), both later became professors at University College, London, and Young is, like Medawar, one of the grand figures of zoology. The significance of his great work The Anatomy of the Nervous System of Octopus Vulgaris would be recognisable to Goodrich: that of Medawar’s later work would not.

Florey was not a man to fiddle around with merely elegant work. He was impatient for discovery, important discovery. If Florey’s school was one influence directing Medawar to important work, the other was the Second World War, with all its distresses and suffering. A man with bad skin burns at that time simply died. There was no question of transplantation. If skin, or any other organ, is grafted from one individual to another, the immune system of the recipient kills the graft, and rejects it. It was thought that immunity was so deeply built into the organisation of the body as to be unchangeable. Transplantation was thus impossible. Such was the belief that Medawar challenged, and by 1956 had triumphantly overthrown. He went down from Oxford to become Professor of Zoology first at Birmingham (1947) and then at University College, London (1951); he later directed the National Institute of Medical Research at Mill Hill.

To consider his work, we need to use the concept of ‘immunological tolerance’. The body’s immune system only destroys foreign cells: it can recognise its own cells, which it leaves unharmed. The immune system is said to tolerate the cells of its own body. Now for Medawar’s work. Consider two strains of mice: call them A and B. If cells from a B-strain mouse are injected into an A-strain mouse, they are killed by its immune system. Now let some B cells be injected, not into an adult A-strain mouse, but into an embryo of about seventeen days’ age. These B cells are not rejected. Because the immune system takes some time to ‘learn’ which cells are its own, if you catch it early enough, it will later learn to treat the foreign, injected B cells as ‘self’. Now graft the skin of a B-strain mouse onto the grown-up adult of an A mouse that has been injected with B cells as an embryo. A normal A-strain mouse would reject the graft. But the injected mice do not. They have acquired a tolerance for B-strain cells. The rejection of grafts is not, after all, inevitable.

Another check on this interpretation is possible. The lymphocytes, the main effector cells of the immune system, of a normal A mouse should reject a B skin graft. A large injection of lymphocytes from a normal A-strain mouse into the experimental A mice that had acquired tolerance to B cells should therefore cause a graft of foreign B skin to be rejected. Medawar and his colleagues performed this experiment too, and it worked.

These results were at that time our most important knowledge of immunological tolerance. But they had a broader influence than that. They are the inspiration of modern transplantation surgery. They did not supply transplantation surgery with a new method, but with the inspiration to develop new methods. They brought within the realm of the possible something that had been thought impossible. I suspect that, if we are to understand Medawar’s opposition to such figures as Francis Galton, and to IQ psychology, we must see it against the background of his work on transplantation.

That great Baconian motto, ‘the effecting of all things possible’, gives both a title and a theme to the final essay of Pluto’s Republic, which, Medawar tells us, he liked ‘best both to prepare and to deliver’. As he is aware, it is not fashionable to preach the possibility of progress. ‘Anybody nowadays who dared to suggest that the plight of man might not be wholly desperate would get a sharp rap over the knuckles in any literary weekly.’ I would not have it thought that fortnightlies are soft on such matters, but I find his confidence more refreshing than wicked. He is certainly impatient with moping mysticism and ‘drooping despondency’. ‘If complacency is to be deplored,’ he said, ‘so too is humility. Humility is not a state of mind conducive to the advancement of learning.’ These are not, as was said of another unfortunate doctor, the imbecilities of a senile optimism: if Medawar is in pursuit of scientific success, and if he is unashamedly confident on the way, he has made a philosophy of criticism too.

The themes of these essays are always interesting; their substance is always sensible and, at best, both important and highly original. But the real delight lies not in the subjects or in the ideas but in the style. It is continually graceful and delightful, and even the less brilliant essays, such as the discussion of ethology and human behaviour, are of seasoned elegance, spiced with wit. Style is, it appears, one of Medawar’s interests. The essays abound with comments on the subject, from D’Arcy Thompson, whose On Growth and Form, Medawar thinks, ‘is beyond comparison the finest work of literature in all the annals of science that have been recorded in the English tongue’, through Teilhard, who writes in ‘that tipsy, euphoristic prose-poetry which is one of the more tiresome manifestations of the French spirit’, to Herbert Spencer, with his ‘hideous, powerful prose – the writing of a man who, lacking and perhaps contemputous of the stylistic graces, is absolutely determined to be understood’.

Medawar’s own prose aims, in exposition, at simplicity, but it amuses by its combination of naughtiness and grandeur. He builds up long, stately sentences, which unfold to say exactly what is wrong with an idea. ‘The thoughts I have been criticising are thus not really thoughts at all, but thought-substitutes, declarations of the kind public people make on public occasions when they are desperately hard up for things to say.’ He pulls no punches. ‘I have not chosen these examples to poke fun at them, ridiculous though I believe them to be, but simply to illustrate the Olympian glibness of psychoanalytic thought.’ He delivers judgments that are sometimes teasing, sometimes frightening. ‘The Phenomenon of Man cannot be read without a feeling of suffocating, a grasping and flailing around for sense.’ There are his arch, often outrageous asides. On the discovery of the structure of DNA, and the molecular biology revolution: ‘It is simply not worth arguing with anyone so obtuse’ – all physicists? – ‘as not to realise that this complex of discoveries is the greatest achievement of science in the 20th century.’ Then there is the lightly worn authority with which he refers to that ‘aberration of science’ which ‘has come to be called “scientism” ’ – a term, I believe, of Medawar’s own invention. Medawar has – what he questionably attributes to that Freemason D’Arcy Thompson – ‘a most resolute determination to unmake mysteries’, and he has what he calls ‘intelligence as the humanist understands the term’ – ‘a combination of strong understanding and commonsensical judgments’.