Shortly before his last illness began, I wrote to him, reminding him of his advice to me and
proposing
that we might publish a joint letter, offering the same advice to others.
Richard-Dawkins-The-Devil-s-Chaplain
It simply, as Diamond explains, becomes medicine.
Conversely, if a technique devised by the President of the Royal College of Physicians consistently fails in double-blind trials, it will cease to be a part of 'orthodox' medicine.
Whether it will then become 'alternative' will depend upon whether it is adopted by a sufficiently ambitious quack (there are always sufficiently gullible patients).
180
? But isn't it still an arrogance to demand that our method of testing should be the scientific method? By all means use scientific tests for scientific medicine, it may be said. But isn't it only fair that 'alternative' medicine should be tested by 'alternative' tests? No. There is no such thing as an alternative test. Here Diamond takes his stand, and he is right to do so.
Either it is true that a medicine works or it isn't. It cannot be false in the ordinary sense but true in some 'alternative' sense. If a therapy or treatment is anything more than a placebo, properly conducted double- blind trials, statistically analysed, will eventually bring it through with flying colours. Many candidates for recognition as 'orthodox' medicines fail the test and are summarily dropped. The 'alternative' label should not (though, alas, it does) provide immunity from the same fate.
Prince Charles has recently called for ten million pounds of govern- ment money to be spent researching the claims of 'alternative' or 'complementary' medicine. An admirable suggestion, although it is not immediately clear why government, which has to juggle competing priorities, is the appropriate source of money, given that the leading 'alternative' techniques have already been tested - and have failed - again and again and again. John Diamond tells us that the alternative medicine business in Britain has a turnover measured in billions of pounds. Perhaps some small fraction of the profits generated by these medicines could be diverted into testing whether they actually work. This, after all, is what 'orthodox' pharmaceutical companies are expected to do. Could it be that purveyors of alternative medicine know all too well what the upshot of properly conducted trials would be? If so, their reluctance to fund their own nemesis is all too understandable. Never- theless, I hope this research money will come from somewhere, perhaps from Prince Charles's own charitable resources, and I would be happy to serve on an advisory committee to disburse it, if invited to do so. Actually, I suspect that ten million pounds' worth of research is more than would be necessary to see off most of the more popular and lucrative 'alternative' practices.
How might the money be spent? Let's take homeopathy as an example, and let us suppose that we have a large enough fraction of the grant to plan the experiment on a moderately large scale. Having given their consent, 1000 patients will be separated into 500 experimentals (who will receive the homeopathic dose) and 500 controls (who will not). Bending over backwards to respect the 'holistic' principle that every individual must be treated as an individual, we shall not insist on giving all experimental subjects the same dose. Nothing so crude.
SNAKE OIL
181
? THEY TOLD ME, HERACLITUS
Instead, every patient in the trial shall be examined by a certified homeopath, and an individually tailored therapy prescribed. The different patients need not even receive the same homeopathic substance.
But now comes the all-important double-blind randomization. After every patient's prescription has been written, half of the patients, at random, will be designated controls. The controls will not in fact receive their prescribed dose. Instead, they will be given a dose which is identical in all respects to the prescribed dose but with one crucial difference. The supposed active ingredient is omitted from its prepara- tion. The randomizing will be done by computer, in such a way that nobody will know which patients are experimentals and which controls. The patients themselves won't know; the therapists won't know; the pharmacists preparing the doses won't know, and the doctors judging the results won't know. The bottles of medicine will be identified only by impenetrable code numbers. This is vitally important because nobody denies placebo effects: patients who think they are getting an effective cure feel better than patients who think the opposite.
Each patient will be examined by a team of doctors and homeopaths, both before and after the treatment. The team will write down their judgement for each patient: has this patient got better, stayed the same, or got worse? Only when these verdicts have all been written down and sealed will the randomizing codes in the computer be broken. Only then will we know which patients had received the homeopathic dose and which the control placebo. The results will be analysed statistically to see whether the homeopathic doses had any effect one way or the other. I know which result my shirt is on, but - this is the beauty of good science - I cannot bias the outcome. Nor can the homeopaths who are betting on the opposite. The double-blind experimental design disempowers all such biases. The experiment can be performed by advocates or sceptics, or both working together, and it won't change the result.
There are all sorts of details by which this experimental design could be made more sensitive. The patients could be sorted into 'matched pairs', matched for age, weight, sex, diagnosis, prognosis and preferred homeopathic prescription. The only consistent difference is that one member of each pair is randomly and secretly designated a control, and given a placebo. The statistics then specifically compare each experimental individual with his matched control.
The ultimate matched-pairs design is to use each patient as his own control, receiving the experimental and the control dose successively, and never knowing when the change occurs. The order of administering
182
? the two treatments to a given patient would be determined at random, a different random schedule for different patients.
'Matched pairs' and 'own control' experimental designs have the advantage of increasing the sensitivity of the test. Increasing, in other words, the chance of yielding a statistically significant success for homeopathy. Notice that a statistically significant success is not a very demanding criterion. It is not necessary that every patient should feel better on the homeopathic dose than on the control. All we are looking for is a slight advantage to homeopathy over the blind control, an advantage which, however slight, is too great to be attributable to luck, according to the standard methods of statistics. This is what is routinely demanded of orthodox medicines before they are allowed to be advertised and sold as curative. It is rather less than is demanded by a prudent pharmaceutical company before it will invest a lot of money in mass production.
Now we come to an awkward fact about homeopathy in particular, dealt with by John Diamond, but worth stressing here. It is a funda- mental tenet of homeopathic theory that the active ingredient - arnica, bee venom, or whatever it is - must be successively diluted some large number of times, until - all calculations agree - there is not a single molecule of that ingredient remaining. Indeed, homeopaths make the daringly paradoxical claim that the more dilute the solution the more potent its action. The investigative conjuror James Randi has calculated that, after a typical sequence of homeopathic 'succussive' dilutions, there would be one molecule of active ingredient in a vat the size of the solar system! (Actually, in practice, there will be more stray molecules knocking around even in water of the highest attainable purity. )
Now, think what this does. The whole rationale of the experiment is to compare experimental doses (which include the 'active' ingredient) with control doses (which include all the same ingredients except the active one). The two doses must look the same, taste the same, feel the same in the mouth. The only respect in which they differ must be the presence or absence of the putatively curative ingredient. But in the case of homeopathic medicine, the dilution is such that there is no difference between the experimental dose and the control! Both contain the same number of molecules of the active ingredient - zero, or whatever is the minimum attainable in practice. This seems to suggest that a double- blind trial of homeopathy cannot, in principle, succeed. You could even say that a successful result would be diagnostic of a failure to dilute sufficiently!
There is a conceivable loophole, much slithered through by 183
SNAKE OIL
? THEY TOLD ME, HERACLITUS
homeopaths ever since this embarrassing difficulty was brought to their attention. The mode of action of their remedies, they say, is not chemical but physical. They agree that not a single molecule of the active ingredient remains in the bottle that you buy, but this only matters if you insist on thinking chemically. They believe that, by some physical mechanism unknown to physicists, a kind of 'trace' or 'memory' of the active molecules is imprinted on the water molecules used to dilute them. It is the physically imprinted template on the water that cures the patient, not the chemical nature of the original ingredient.
This is a scientific hypothesis in the sense that it is testable. Easy to test, indeed, and although I wouldn't bother to test it myself, this is only because I think our finite supply of time and money would be better spent testing something more plausible. But any homeopath who really believes his theory should be beavering away from dawn to dusk. After all, if the double-blind trials of patient treatments came out reliably and repeatably positive, he would win a Nobel Prize not only in Medicine but in Physics as well. He would have discovered a brand-new principle of physics, perhaps a new fundamental force in the universe. With such a prospect in view, homeopaths must surely be falling over each other in their eagerness to be first into the lab, racing like alternative Watsons and Cricks to claim this glittering scientific crown. Er, actually no they aren't. Can it be that they don't really believe their theory after all?
At this point we scrape the barrel of excuses. 'Some things are true on a human level, but they don't lend themselves to scientific testing. The sceptical atmosphere of the science lab is not conducive to the sensitive forces involved. ' Such excuses are commonly trotted out by practitioners of alternative therapies, including those that don't have homeopathy's peculiar difficulties of principle but which nevertheless consistently fail to pass double-blind tests in practice. John Diamond is a pungently witty writer, and one of the funniest passages of this book is his description of an experimental test of 'kinesiology' by Ray Hyman, my colleague on CSICOP (the Committee for the Scientific Investigation of Claims of the Paranormal).
As it happens, I have personal experience of kinesiology. It was used by the one quack practitioner I have - to my shame - consulted. I had ricked my neck. A therapist specializing in manipulation had been strongly recommended. Manipulation can undoubtedly be very effective, and this woman was available at the weekend, when I didn't like to trouble my normal doctor. Pain and an open mind drove me to give her a try. Before she began the manipulation itself, her diagnostic
184
? technique was kinesiology. I had to lie down and stretch out my arm, and she pushed against it, testing my strength. The key to the diagnosis was the effect of vitamin C on my arm-wrestling performance. But I wasn't asked to imbibe the vitamin. Instead (I am not exaggerating, this is the literal truth), a sealed bottle containing vitamin C was placed on my chest. This appeared to cause an immediate and dramatic increase in the strength of my arm, pushing against hers. When I expressed my natural scepticism, she said happily, 'Yes, C is a marvellous vitamin, isn't it! ' Human politeness stopped me walking out there and then, and I even (to avoid hassle) ended up paying her lousy fee.
What was needed (I doubt if that woman would even have under- stood the point) was a series of double-blind trials, in which neither she nor I was allowed to know whether the bottle contained the alleged active ingredient or something else. This was what Professor Hyman, in John Diamond's hilarious description of a similar case, undertook. When, predictably, the 'alternative' technique ignominiously flunked the double-blind test, its practitioner delivered himself of the following immortal response: 'You see? That is why we never do double-blind testing any more. It never works! '
A large part of the history of science, especially medical science, has been a progressive weaning away from the superficial seductiveness of individual stories that seem - but only seem - to show a pattern. The human mind is a wanton storyteller and, even more, a profligate seeker after pattern. We see faces in clouds and tortillas, fortunes in tea leaves and planetary movements. It is quite difficult to prove a real pattern as distinct from a superficial illusion. The human mind has to learn to mistrust its native tendency to run away with itself and see pattern where there is only randomness. That is what statistics are for, and that is why no drug or therapeutic technique should be adopted until it has been proved by a statistically analysed experiment, in which the fallible pattern-seeking proclivities of the human mind have been systematically taken out of the picture. Personal stories are never good evidence for any general trend.
In spite of this, doctors have been heard to begin a judgement with something like, 'The trials all say otherwise, but in my clinical experience . . . ' This might constitute better grounds for changing your doctor than a suable malpractice! That, at least, would seem to follow from all that I have been saying. But it is an exaggeration. Certainly, before a medicine is certified for wide use, it must be properly tested and given the imprimatur of statistical significance. But a mature doctor's clinical experience is at least an excellent guide to which hypotheses
SNAKE OIL
185
? THEY TOLD ME, HERACLITUS
might be worth going to the trouble and expense of testing. And there's more that can be said. Rightly or wrongly (often rightly) we actually do take the personal judgement of a respected human individual seriously. This is so with aesthetic judgements, which is why a famous critic can make or break a play on Broadway or Shaftesbury Avenue. Whether we like it or not, people are swayed by anecdote, by the particular, by the personal.
And this, almost paradoxically, helps to make John Diamond such a powerful advocate. He is a man whom we like and admire for his personal story, and whose opinions we want to read because he expresses them so well. People who might not listen to a set of nameless statistics, intoned by a faceless scientist or doctor, will listen to John Diamond, not just because he writes engagingly, but because he was dying while he wrote and he knew it: dying in spite of the best efforts of the very medical practices he was defending against opponents whose only weapon is anecdote. But there is really no paradox. He may gain our ear because of his singular qualities and his human story. But what we hear when we listen to him is not anecdotal. It stands up to rigorous examination. It would be sensible and compelling in its own right even if its author had not previously earned our admiration and our affection.
John Diamond was never going to go gentle into that good night. When he did go it was with guns blazing, for the splendidly polemical chapters of Snake Oil occupied him right up to the end, working against . . . not so much the clock as time's winged chariot itself. He does not rage against the dying of the light, nor against his wicked cancer, nor against cruel fate. What would be the point, for what would they care? His targets are capable of wincing when hit. They are targets that deserve to be hit hard, targets whose neutralization would leave the world a better place: cynical charlatans (or honestly foolish dreamers) who prey on gullible unfortunates. And the best part is that although this gallant man is dead, his guns are not silenced. He left a strong emplacement. This posthumous book launches his broadside. Open fire, and don't stop.
186
? 5
EVEN THE RANKS OF TUSCANY
Stephen Jay Gould and I did not tire the sun with talking and send him down the sky. We were cordial enough when we met, but it would be disingenuous to suggest that we were close. Our academic differences have even been spun out to book length, by the philosopher Kim Sterelny in Dawkins vs Gould: Survival of the Fittest,^ while Andrew Brown, in The Darwin Wars: How Stupid Genes Became Selfish Gods,TM goes so far as to divide modern Darwinians into 'Gouldians' and 'Dawkinsians'. Yet, despite our differences, it is not just the respect due to the dead that leads me to include in this book a section on Stephen Gould with a largely positive tone.
'And even the ranks of Tuscany' (Steve would have completed the
quotation from his formidable literary memory) 'Could scarce forebear to
107
cheer'. Macaulay celebrated the admiration that can unite enemies in
death. Enemies is too strong a word for a purely academic dispute, but
admiration is not, and we were shoulder to shoulder on so much. In his
review of my own Climbing Mount Improbable, Steve invoked a collegiality 08
between us, which I reciprocated, in the face of a shared enemy:'
In this important uphill battle for informing a hesitant (if not outrightly hostile) public about the claims of Darwinian evolution, and for explaining both the beauty and power of this revolutionary view of life, I feel collegially entwined with Richard Dawkins in a common enterprise.
He was never ashamed of his immodesty, and I hope I may be forgiven for
sharing with my readers the one occasion when he was good enough to
include me in it: 'Richard and I are the two people who write about
109
evolution best . . . ' There was a 'but' of course, but I must press on.
The book reviews that follow, spaced many years apart, show what I hope will be read as an equal collegiality, even where they are critical. Ever Since Darwin was the first collection of Gould's celebrated essays from Natural History. It set the tone for all ten of them, and the 'barbed rave' tone of Rejoicing in Multifarious Nature (5. 1) could also serve for any.
187
? EVEN THE RANKS OF TUSCANY
The Art of the Developable (5. 2), though written in 1983, has not pre- viously been published. It is a joint review of Peter Medawar's Pluto's Republic and the third of Gould's collections of essays from Natural History. It was commissioned by the New York Review of Books but eventually, for reasons that I no longer recall, the publication fell through. Years later I sent the review to Steve, and he expressed warm disappointment that it had never been published. Medawar was one of my intellectual heroes, and Gould's too: it was another thing we had in common. My title, 'The Art of the Developable', unites Medawar's Art of the SolubleTM with Gould's long interest in the evolution of development.
Wonderful Life is, in my view, a beautiful and a misguided book. It is also misguiding: its enthusiastic rhetoric leading other authors to absurd conclusions far beyond Dr Gould's intentions. I developed this aspect fully in 'Huge Cloudy Symbols of a High Romance', one of the chapters of my Unweaving the Rainbow. Reprinted here as Hallucigenia, Wiwaxia and Friends (5. 3), the title given it by the Sunday Telegraph, is my review of Wonderful Life itself.
Human Chauvinism and Evolutionary Progress (5. 4) is my review of Full House, a book that was renamed by the British publishers Life's Grandeur. The review was published as a matched pair with Steve's review of Climbing Mount Improbable. The Editor of Evolution thought it would be amusing to invite each of us to review the other's book simultaneously, knowing the existence, but not the content of the other's review. Gould's review had the characteristic title 'Self-help for a hedgehog stuck on a molehill'. Full House is all about the idea of progress in evolution. I agree with Gould's objections to progress as he saw it. But in this review I develop two alternative meanings of progress which I think are important and are not vulnerable to his objections. My intention was not just to review a book but to make a contribution to evolutionary thinking.
Stephen Gould was my exact contemporary but I always thought of him as senior, probably because his prodigious learning seemed to belong to a more cultivated era. His lifelong colleague Niles Eldredge, who was kind enough to send me the text of his moving eulogy, said that he had lost an elder brother. Years ago it seemed natural to me to ask Steve's advice when I was travelling in America and was invited to have a televised 'debate' with a creationist. He said that he always refused such invitations, not because he was afraid of 'losing' the debate (the idea is laughable) but for a subtler reason which I accepted and never forgot.
Shortly before his last illness began, I wrote to him, reminding him of his advice to me and proposing that we might publish a joint letter, offering the same advice to others. He enthusiastically agreed, and suggested that I should prepare a
188
? draft on which we could later work together. I did so but, sadly, 'later' nev? r came. When I heard of his sudden death, I wrote to Niles Eldredge, asking if he thought Steve would have wished me to publish the letter anyway. Niles encouraged me to do so, and, as Unfinished Correspondence with a Darwinian Heavyweight (5. 5), it closes this section.
For good or ill, Steve Gould had a huge influence on American scientific culture, and on balance the good came out on top. It is pleasing that, just before his death, he managed to complete both his magnum opus on evolution and his ten-volume cycle of essays from Natural History. Although we disagreed about much, we shared much too, including a spellbound delight in the wonders of the natural world, and a passionate conviction that such wonders deserve nothing less than a purely naturalistic explanation.
EVEN THE RANKS OF TUSCANy
189
? Rejoicing in Multifarious Nature111 Review of Ever Since Darwin by S. J. Gould
'The author shows us what is revealed when we remove the blinkers which Darwin stripped from biology a century ago. ' Some overkill there, or an excitingly paradoxical striptease technique? The first essay in the book discusses Darwin's own coyness in not revealing his theory until 20 years after he thought of it, and I shall return to this. The quotation from the jacket blurb gives a false impression, for Stephen Gould's writing is elegant, erudite, witty, coherent and forceful. He is also, in my opinion, largely right. If there are elements of paradox and overkill in Dr Gould's intellectual position, they are not to be found within these covers. Ever Since Darwin is a collection of essays which first appeared as a regular monthly column in Natural History. Skilfully edited to flow in eight main sections, the 33 essays, of which I can mention only a sample, reinforce my feeling that scientific journalism is too important to be left to journalists, and encourage my hope that true scientists may be better at it than journalists anyway.
Gould's collection begins to bear comparison with P. B. Medawar's immortal The Art of the Soluble. And if his style does not quite make the reader chortle with delight and rush out to show somebody - anybody - the way Medawar's does, Gould is to be thanked for some memorable lines. No doubt puritan killjoys of Science for 'The People' will denounce the vivid and helpful anthropomorphism in 'Reproduce like hell while you have the ephemeral resource, for it will not last long and some of your progeny must survive to find the next one. ' But on second thoughts they may be too busy plotting the abolition of slavery in ants, or brooding over the deviationism of:
Natural selection dictates that organisms act in their own self-interest . . . They 'struggle' continuously to increase the representation of their genes at the expense of their fellows. And that, for all its baldness, is all there is to it; we have discovered no higher principle in nature.
190
? Ever since Darwin we have known why we exist and we have known at least how to set about explaining human nature. I agree that natural selection is 'the most revolutionary notion in the history of biology' and I would toy with substituting 'science' for 'biology'. Childishly simple as it is, nobody thought of it until centuries after far more complicated ideas had become common currency, and it is still the subject of misunderstanding and even apathy among educated people. A microcosm of this historical enigma is the subject of Gould's first essay. Just as humankind waited centuries longer than our hindsight deems necessary before discovering natural selection, so Darwin delayed his own publication 20 years after he first thought of the theory in 1838. Gould's explanation is that Darwin was afraid of the psychological implications of his idea. He saw what Wallace would never admit, that the human mind itself must be a material product of natural selection. Darwin, in fact, was a scientific materialist.
In another essay Gould is encouraged by the genetic closeness of humans and chimpanzees to speculate that 'inter-breeding may well be possible'. I doubt it, but it is a pleasing thought and Gould surely exaggerates when he rates it 'the most . . . ethically unacceptable scientific experiment I can imagine'. For my ethics, far less acceptable experiments are conceivable, and actually done in animal physiology laboratories every day, and a chimp/human hybridization would provide exactly the come-uppance that 'human dignity' needs. Gould is, in general, rather good at puncturing human speciesist vanity, and in particular he will have nothing to do with the myth that evolution represents progress towards man. This scepticism informs his valuable account of 'Bushes and ladders in human evolution', and fires his scorn for attempts to rank human races as primitive or advanced.
He returns to the attack on progress in the very different guise of the theory of orthogenesis, the idea that evolutionary trends have their own internal momentum which eventually drives lineages extinct. His telling of the classic Irish Elk story gains freshness from his intimacy with the fossils of the Dublin Museum and gives the lie to the myth that palaeontology is dry and dull. His conclusion that the proverbially topheavy antlers were important in social life is surely right, but he may underestimate the role of within-species competition in driving species extinct. Large antlers could directly have caused the extinction of the Irish Elk while at the same time, right up to the moment of extinction, individuals with relatively large antlers were out-reproducing indivi- duals with relatively small antlers. I would like to see Gould come to terms with the 'orthoselective' impact of 'arms races' both between and
REJOICING IN MULTIFARIOUS NATURE
191
? EVEN THE RANKS OF TUSCANY
within species. He seems to approach this in his essays on the 'Cambrian explosion'.
Natural history can be sold for its intrinsic fascination, but it is much better used to make a point. Gould tells us about a fly that eats its mother from the inside, about 17-year cicadas and 120-year bamboos, and about uncanny fish-decoying mussels. He employs the useful trick of first opening the reader's mind by boggling it, then filling it with the important biological principle. One principle I would have liked to have heard more on is that of the limitation of evolutionary perfection: 'Orchids are Rube Goldberg machines; a perfect engineer would certainly have come up with something better' (Rube Goldberg is the American Heath Robinson). My own favourite example, inherited from an undergraduate tutor, is the recurrent laryngeal nerve. It starts in the head, goes down into the chest, loops round the aorta, then goes straight back into the head again. In a giraffe this detour must be wasteful indeed. The human engineer who first designed the jet engine simply threw the old propeller engine out and started afresh. Imagine the contraption he would have produced if he had been constrained to 'evolve' his jet engine by changing a propeller engine one bit at a time, nut by nut and bolt by bolt!
While on the problem of perfection, I think Gould exaggerates the relevance of 'neutral mutations'. Molecular geneticists are understandably interested in DNA changes as molecular events, and any that have no effect on protein function may reasonably be called neutral mutations. But to a student of whole organism biology they are less than neutral; they are not, in any interesting sense, mutations at all! If the molecular neutralists are right, their kind of neutral mutation will forever be hidden from the field biologist and from natural selection. And if a field biologist actually sees variation in phenotypes, the question of whether that variation could be selectively neutral cannot be settled in the biochemistry laboratory.
Several essays touch on aspects of the relationship of Darwinism to human society and politics. There is much humane good sense here and I agree with most of it. Although 'sociobiology' is inspiring excellent research, Gould is right that it has also led to some second-rate bandwagoneering. 'But was there ever dog that praised his fleas? ' asked Yeats. Perhaps a dog may be held responsible for the fleas he sheds, but only to a small extent. At the AAAS meeting in Washington in 1997, Gould and I witnessed an organized attack on his most distinguished Harvard colleague. * Gould well deserved his ovation for the apt Lenin quote with which he disowned the rabble. But as he watched those
*A glassful of water was thrown sideways at Professor E. O. Wilson (subsequently exaggerated in various accounts to a 'pitcher of iced water, poured over him').
192
? pathetic fleas ineffectually hopping around the stage chanting, of all things, 'genocide', did he wonder with a little itch of conscience on which dog they had been sucking?
The Epilogue is forward-looking and whets our appetite for Volume 2, which I earnestly hope will be forthcoming. * One theme which I know Gould has already carried further in his Natural History column is his dislike for 'the ultimate atomism' of regarding organisms as 'temporary receptacles . . . no more than instruments that genes use to make more
112
genes like themselves'. In describing this as 'metaphorical nonsense'
Gould underestimates the sophistication of the idea, first cogently
113
expressed in its modern form by George C. Williams. The dispute is
largely semantic. Inclusive fitness is defined in such a way that to say 'the individual works so as to maximize its inclusive fitness' is equivalent to saying 'the genes work so as to maximize their survival'. The two forms are each valuable for different purposes. Both contain an element of personification; it is dangerously easier to personify organisms than to personify genes. The gene selection idea is not naively atomistic, as it recognizes that genes are selected for their capacity to interact productively with the other genes with which they are most likely to share 'receptacles'; this means the other genes of the gene pool; and the gene pool may therefore come to resemble a 'homeostatically buffered system' tending to return to (one of) its evolutionarily stable state(s). Irrevocable determination by genes is no part of the idea nor is anything remotely approaching a 'one gene, one trait' mapping from genotype to phenotype. In any case it has nothing to do with 'supreme confidence in universal adaptation', which is as likely to be found among devotees of 'individual selection' or 'species selection'.
'I will rejoice in the multifariousness of nature and leave the chimera of certainty to politicians and preachers': a resounding conclusion to a stimulating book - the work of a free and imaginative scientific mind. The final, sad paradox is this. How can a mind capable of such rejoicing, open enough to contemplate the shifting splendour of three thousand million years, moved by the ancient poetry written in the rocks, how can such a mind not be bored by the drivelling ephemera of juvenile pamphleteers and the cold preaching of spiteful old hardliners? No doubt they are right that science is not politically neutral. But if, to them, that is the most important thing about science, just think what they are missing! Stephen Gould is well qualified, and strategically placed, to strip away even those dark blinkers and dazzle those poor unpractised eyes.
*In fact, ten volumes were eventually published, the last one, / have Landed, at the time of his death.
REJOICING IN MULTIFARIOUS NATURE
193
? '%t0 M mem
The Art of the Developable
The acknowledged master of biological belles lettres has long been Sir Peter Medawar. If there is a younger biologist or an American biologist that bears comparison, it is probably in both cases Stephen Jay Gould. It was therefore with anticipation that I received these two collections of essays, reflections by leading and highly literate biologists on their subject and its history and philosophy.
Pluto's Republic is one of those titles that cannot be mentioned without an immediate explanation, and Sir Peter begins thus:
A good many years ago a neighbour whose sex chivalry forbids me to disclose [it takes a Medawar to get away with this kind of thing nowadays] exclaimed upon learning of my interest in philosophy: 'Don't you just adore Pluto's Republic? ' Pluto's Republic has remained in my mind ever since as a super- latively apt description of that intellectual underworld which so many of the essays in this volume explore. We each populate Pluto's Republic according to our own prejudices . . .
Here I nursed a mischievous half-hope that Stephen Gould might be found among the denizens of Medawar's private underworld - his more sanctimonious cosignatories of a notorious letter to the New York Review of Books about 'sociobiology' (13 November 1975) are prominent in mine. But Gould is several cuts above those former associates of his and he is not among Medawar's targets. Indeed, they share many targets, IQ- metricians for instance.
Most of the essays in Pluto's Republic have appeared twice before, first as book reviews or transcripts of lectures, then in previous anthologies
116
such as The Art of the Soluble and The Hope of Progress, which were
presumably reviewed at the time. Although I shall therefore give Pluto's Republic less than half my space in this joint review, I vigorously repudiate any mutterings about such second order anthologizing being too much of a good thing. The earlier books have long been out of
194
14 Hen's Teeth and Horse's Toes by Stephen Jay Gould
Review of Pluto's Republic by Peter Medawar'
and 11
? print, and I have been scouring the second-hand bookshops ever since my own Art of the Soluble was stolen. I discovered when I reread them here that I had many favourite passages word-perfect in memory. Who indeed could forget the opening sentence of the 1968 Romanes Lecture, 'Science and Literature'? T hope I shall not be thought ungracious if I say at the outset that nothing on earth would have induced me to attend the kind of lecture you may think I am about to give. ' At the time this prompted the apt rejoinder from John Holloway: 'This lecturer can never have been thought ungracious in his life. '
Or listen to Medawar on another great biologist, Sir D'Arcy Thompson:
. . . He was a famous conversationalist and lecturer (the two are often thought to go together, but seldom do), and the author of a work which, considered as literature, is the equal of anything of Pater's or Logan Pearsall Smith's in its complete mastery of the bel canto style. Add to all this that he was over six feet tall, with the build and carriage of a Viking and with the pride of bearing that comes from good looks known to be possessed.
The reader may be hazy about Logan Pearsall Smith and Pater, but he is left with the overwhelming impression (since he probably is familiar with the idiom of P. G. Wodehouse) of a style that is undoubtedly bel, and may very well be canto. And there is more of Medawar in the passage quoted than Medawar himself realized.
Medawar continually flatters his readers, implying in them an erudition beyond them, but doing it so that they almost come to believe in it themselves:
'Mill,' said John Venn in 1907, has 'dominated the thought and study of intelligent students to an extent which many will find it hard to realise at the present day'; yet he could still take a general familiarity with Mill's views for granted . . .
The reader scarcely notices that Medawar himself is still taking a general familiarity with Mill's views for granted, although in the reader's own case it may be far from justified. 'Even George Henry Lewes found him- self unable to propound his fairly sensible views on hypotheses without prevarication and pursing of the lips. ' The reader's knowing chuckle is out before he realizes that actually he is in no position to respond knowingly to that 'even'.
Medawar has become a sort of chief spokesman for 'The Scientist' in the modern world. He takes a less doleful view of the human predica- ment than is fashionable, believing that hands are for solving problems rather than for wringing. He regards the scientific method - in the right
THE ART OF THE DEVELOPABLE
195
? EVEN THE RANKS OF TUSCANY
hands - as our most powerful tool for 'finding out what is wrong with [the world] and then taking steps to put it right'. As for the scientific method itself, Medawar has a good deal to tell us, and he is well qualified to do so. Not that being a Nobel Prize winner and a close associate of Karl Popper is in itself an indication that one will talk sense: far from it when you think of others in that category. But Medawar not only is a Nobel Prize winner, he seems like a Nobel Prize winner; he is everything we think a Nobel Prize winner ought to be. If you have never understood why scientists like Popper, try Medawar's exposition of the philosophy of his 'personal guru'.
He read Zoology at Oxford, and early in his career made important contributions to classical Zoology, but was soon drawn into the highly populated and highly financed world of medical research. Inevitably, his associates have been molecular and cell biologists, but he seldom had any truck with the molecular chauvinism which plagued biology for two decades. Medawar has a good appreciation of biology at all levels.
He has also inevitably associated with doctors, and the preoccupa- tions and sympathies of a doctor pervade several of these essays, for example his sensitive reviews of books on cancer and psychosomatic heart disease. I especially enjoyed his blistering contempt for psycho- analysis: not a lofty, detached contempt for any ordinary pretentious drivel, but a committed contempt, fired by a doctor's concern. Psychoanalysts have even had their say over the puzzle of Darwin's long illness, and Medawar is at his withering best in telling us about it.
For Good, there is a wealth of evidence that unmistakably points to the idea that Darwin's illness was 'a distorted expression of the aggression, hate and resentment felt, at an unconscious level, by Darwin towards his tyrannical father'. These deep and terrible feelings found outward expression in Darwin's touching reverence towards his father and his father's memory, and in his describing his father as the kindest and wisest man he ever knew: clear evidence, if evidence were needed, of how deeply his true inner sentiments had been repressed.
Medawar, when he smells pretentious pseudoscience, is a dangerous man. His famous annihilation of Teilhard de Chardin's The Phenomenon of Man might have been thought an unfair attack on the dead, but for the extraordinary influence Teilhard exerted (and still exerts: Stephen Gould tells us that two journals established to discuss his ideas still flourish) over legions of the gullible including, I am afraid, my juvenile self. I would love to quote huge chunks of what is surely one of the great destructive book reviews of all time, but must content myself with
196
? two sentences from Medawar's typically barbed explanation of the popular appeal of Teilhard.
Just as compulsory primary education created a market catered for by cheap dailies and weeklies, so the spread of secondary and latterly of tertiary education has created a large population of people, often with well-developed literary and scholarly tastes who have been educated far beyond their capacity to undertake analytical thought . . . [The Phenomenon of Man] is written in an all but totally unintelligible style, and this is construed as prima-facie evidence of profundity.
Medawar's Herbert Spencer Lecture, and his review of Arthur Koestler's Act of Creation, are more respectful of his victims, but pretty punchy nevertheless. His review of Ronald Clark's Life of J. B. S. Haldane is enlivened by personal reminiscence, and reveals a sort of affection for the old brute which seems to have been reciprocated.
I remember Haldane's once going back on a firm promise to chair a lecture given by a distinguished American scientist on the grounds that it would be too embarrassing for the lecturer: he had once been the victim of a sexual assault by the lecturer's wife. The accusation was utterly ridiculous and Haldane did not in the least resent my saying so. He didn't want to be bothered with the chairmanship, and could not bring himself to say so in the usual way.
But if Haldane did not in the least resent Medawar's saying so, one cannot help wondering whether this was only because Medawar must have been one of the very few people Haldane ever met who could look him levelly in the eye, on equal terms intellectually. Peter Medawar is a giant among scientists and a wicked genius with English prose. Even if it annoys you, you will not regret reading Pluto's Republic.
In 1978 the Reviews Editor of a famous scientific journal, whose nature prudence forbids me to disclose, invited me to review Stephen
Jay Gould's Ever Since Darwin, remarking that I could 'get my own back' on opponents of 'genetic determinism'. I don't know which annoyed me more: the suggestion that I favoured genetic 'determinism' (it is one of those words like sin and reductionism: if you use it at all you are against it) or the suggestion that I might review a book for motives of revenge. The story warns my readers that Dr Gould and I are supposed to be on opposites sides of some fence or other. In the event, I accepted the commission and gave the book what could fairly be described as a rave review, even, I think, going so far as to praise Gould's style as a creditable second best to Peter Medawar's. *
*See 'Rejoicing in Multifarious Nature' (pp. 190-3).
THE ART OF THE DEVELOP ABLE
197
? EVEN THE RANKS OF TUSCANY
I feel inclined to do the same for Hen's Teeth and Horse's Toes. It is another collection of essays reprinted from Gould's column in Natural Histor)'. When you have to turn these pieces out once a month you must pick up some of the habits of the professional working to a deadline - this is not a criticism, Mozart did the same. Gould's writing has some- thing of the predictability that we enjoy in Mozart, or in a good meal. His volumes of collected essays, of which this is the third, are put together to a recipe: one part biological history, one part biological politics (less if we are lucky), and one part (more if we are lucky) vignettes of biological wonder, the modern equivalent of a mediaeval bestiary but with interesting scientific morals instead of boring pious ones. The essays themselves, too, often seem to follow a formula or menu. As appetizer there is the quotation from light opera or the classics, or sometimes its place is taken by a piece of reassuring nostalgia; a reminiscence from a normal, happy, very American childhood world of baseball stars and Hershey bars and Bar Mitzvahs - Gould, we learn, is not just one of your pointy-headed intellectuals but a regular guy. This homely informality softens the conspicuous erudition of the main course - the fluency in several languages, the almost Medawarian familiarity with literature and humanities - and even gives it a certain (un-Medawarian) charm (compare Gould himself on Louis Agassiz: '. . . the erudition that has so charmed American rustics . . . ').
Gould's own respect for Medawar is evident. The idea of science as 'the art of the soluble' provides the punchline for at least four of the essays: 'We may wallow forever in the thinkable; science traffics in the doable', '. . . science deals in the workable and soluble'; and two essays end with explicit quotations of the phrase. His view of Teilhard de Chardin's style is similar to Medawar's: '.
180
? But isn't it still an arrogance to demand that our method of testing should be the scientific method? By all means use scientific tests for scientific medicine, it may be said. But isn't it only fair that 'alternative' medicine should be tested by 'alternative' tests? No. There is no such thing as an alternative test. Here Diamond takes his stand, and he is right to do so.
Either it is true that a medicine works or it isn't. It cannot be false in the ordinary sense but true in some 'alternative' sense. If a therapy or treatment is anything more than a placebo, properly conducted double- blind trials, statistically analysed, will eventually bring it through with flying colours. Many candidates for recognition as 'orthodox' medicines fail the test and are summarily dropped. The 'alternative' label should not (though, alas, it does) provide immunity from the same fate.
Prince Charles has recently called for ten million pounds of govern- ment money to be spent researching the claims of 'alternative' or 'complementary' medicine. An admirable suggestion, although it is not immediately clear why government, which has to juggle competing priorities, is the appropriate source of money, given that the leading 'alternative' techniques have already been tested - and have failed - again and again and again. John Diamond tells us that the alternative medicine business in Britain has a turnover measured in billions of pounds. Perhaps some small fraction of the profits generated by these medicines could be diverted into testing whether they actually work. This, after all, is what 'orthodox' pharmaceutical companies are expected to do. Could it be that purveyors of alternative medicine know all too well what the upshot of properly conducted trials would be? If so, their reluctance to fund their own nemesis is all too understandable. Never- theless, I hope this research money will come from somewhere, perhaps from Prince Charles's own charitable resources, and I would be happy to serve on an advisory committee to disburse it, if invited to do so. Actually, I suspect that ten million pounds' worth of research is more than would be necessary to see off most of the more popular and lucrative 'alternative' practices.
How might the money be spent? Let's take homeopathy as an example, and let us suppose that we have a large enough fraction of the grant to plan the experiment on a moderately large scale. Having given their consent, 1000 patients will be separated into 500 experimentals (who will receive the homeopathic dose) and 500 controls (who will not). Bending over backwards to respect the 'holistic' principle that every individual must be treated as an individual, we shall not insist on giving all experimental subjects the same dose. Nothing so crude.
SNAKE OIL
181
? THEY TOLD ME, HERACLITUS
Instead, every patient in the trial shall be examined by a certified homeopath, and an individually tailored therapy prescribed. The different patients need not even receive the same homeopathic substance.
But now comes the all-important double-blind randomization. After every patient's prescription has been written, half of the patients, at random, will be designated controls. The controls will not in fact receive their prescribed dose. Instead, they will be given a dose which is identical in all respects to the prescribed dose but with one crucial difference. The supposed active ingredient is omitted from its prepara- tion. The randomizing will be done by computer, in such a way that nobody will know which patients are experimentals and which controls. The patients themselves won't know; the therapists won't know; the pharmacists preparing the doses won't know, and the doctors judging the results won't know. The bottles of medicine will be identified only by impenetrable code numbers. This is vitally important because nobody denies placebo effects: patients who think they are getting an effective cure feel better than patients who think the opposite.
Each patient will be examined by a team of doctors and homeopaths, both before and after the treatment. The team will write down their judgement for each patient: has this patient got better, stayed the same, or got worse? Only when these verdicts have all been written down and sealed will the randomizing codes in the computer be broken. Only then will we know which patients had received the homeopathic dose and which the control placebo. The results will be analysed statistically to see whether the homeopathic doses had any effect one way or the other. I know which result my shirt is on, but - this is the beauty of good science - I cannot bias the outcome. Nor can the homeopaths who are betting on the opposite. The double-blind experimental design disempowers all such biases. The experiment can be performed by advocates or sceptics, or both working together, and it won't change the result.
There are all sorts of details by which this experimental design could be made more sensitive. The patients could be sorted into 'matched pairs', matched for age, weight, sex, diagnosis, prognosis and preferred homeopathic prescription. The only consistent difference is that one member of each pair is randomly and secretly designated a control, and given a placebo. The statistics then specifically compare each experimental individual with his matched control.
The ultimate matched-pairs design is to use each patient as his own control, receiving the experimental and the control dose successively, and never knowing when the change occurs. The order of administering
182
? the two treatments to a given patient would be determined at random, a different random schedule for different patients.
'Matched pairs' and 'own control' experimental designs have the advantage of increasing the sensitivity of the test. Increasing, in other words, the chance of yielding a statistically significant success for homeopathy. Notice that a statistically significant success is not a very demanding criterion. It is not necessary that every patient should feel better on the homeopathic dose than on the control. All we are looking for is a slight advantage to homeopathy over the blind control, an advantage which, however slight, is too great to be attributable to luck, according to the standard methods of statistics. This is what is routinely demanded of orthodox medicines before they are allowed to be advertised and sold as curative. It is rather less than is demanded by a prudent pharmaceutical company before it will invest a lot of money in mass production.
Now we come to an awkward fact about homeopathy in particular, dealt with by John Diamond, but worth stressing here. It is a funda- mental tenet of homeopathic theory that the active ingredient - arnica, bee venom, or whatever it is - must be successively diluted some large number of times, until - all calculations agree - there is not a single molecule of that ingredient remaining. Indeed, homeopaths make the daringly paradoxical claim that the more dilute the solution the more potent its action. The investigative conjuror James Randi has calculated that, after a typical sequence of homeopathic 'succussive' dilutions, there would be one molecule of active ingredient in a vat the size of the solar system! (Actually, in practice, there will be more stray molecules knocking around even in water of the highest attainable purity. )
Now, think what this does. The whole rationale of the experiment is to compare experimental doses (which include the 'active' ingredient) with control doses (which include all the same ingredients except the active one). The two doses must look the same, taste the same, feel the same in the mouth. The only respect in which they differ must be the presence or absence of the putatively curative ingredient. But in the case of homeopathic medicine, the dilution is such that there is no difference between the experimental dose and the control! Both contain the same number of molecules of the active ingredient - zero, or whatever is the minimum attainable in practice. This seems to suggest that a double- blind trial of homeopathy cannot, in principle, succeed. You could even say that a successful result would be diagnostic of a failure to dilute sufficiently!
There is a conceivable loophole, much slithered through by 183
SNAKE OIL
? THEY TOLD ME, HERACLITUS
homeopaths ever since this embarrassing difficulty was brought to their attention. The mode of action of their remedies, they say, is not chemical but physical. They agree that not a single molecule of the active ingredient remains in the bottle that you buy, but this only matters if you insist on thinking chemically. They believe that, by some physical mechanism unknown to physicists, a kind of 'trace' or 'memory' of the active molecules is imprinted on the water molecules used to dilute them. It is the physically imprinted template on the water that cures the patient, not the chemical nature of the original ingredient.
This is a scientific hypothesis in the sense that it is testable. Easy to test, indeed, and although I wouldn't bother to test it myself, this is only because I think our finite supply of time and money would be better spent testing something more plausible. But any homeopath who really believes his theory should be beavering away from dawn to dusk. After all, if the double-blind trials of patient treatments came out reliably and repeatably positive, he would win a Nobel Prize not only in Medicine but in Physics as well. He would have discovered a brand-new principle of physics, perhaps a new fundamental force in the universe. With such a prospect in view, homeopaths must surely be falling over each other in their eagerness to be first into the lab, racing like alternative Watsons and Cricks to claim this glittering scientific crown. Er, actually no they aren't. Can it be that they don't really believe their theory after all?
At this point we scrape the barrel of excuses. 'Some things are true on a human level, but they don't lend themselves to scientific testing. The sceptical atmosphere of the science lab is not conducive to the sensitive forces involved. ' Such excuses are commonly trotted out by practitioners of alternative therapies, including those that don't have homeopathy's peculiar difficulties of principle but which nevertheless consistently fail to pass double-blind tests in practice. John Diamond is a pungently witty writer, and one of the funniest passages of this book is his description of an experimental test of 'kinesiology' by Ray Hyman, my colleague on CSICOP (the Committee for the Scientific Investigation of Claims of the Paranormal).
As it happens, I have personal experience of kinesiology. It was used by the one quack practitioner I have - to my shame - consulted. I had ricked my neck. A therapist specializing in manipulation had been strongly recommended. Manipulation can undoubtedly be very effective, and this woman was available at the weekend, when I didn't like to trouble my normal doctor. Pain and an open mind drove me to give her a try. Before she began the manipulation itself, her diagnostic
184
? technique was kinesiology. I had to lie down and stretch out my arm, and she pushed against it, testing my strength. The key to the diagnosis was the effect of vitamin C on my arm-wrestling performance. But I wasn't asked to imbibe the vitamin. Instead (I am not exaggerating, this is the literal truth), a sealed bottle containing vitamin C was placed on my chest. This appeared to cause an immediate and dramatic increase in the strength of my arm, pushing against hers. When I expressed my natural scepticism, she said happily, 'Yes, C is a marvellous vitamin, isn't it! ' Human politeness stopped me walking out there and then, and I even (to avoid hassle) ended up paying her lousy fee.
What was needed (I doubt if that woman would even have under- stood the point) was a series of double-blind trials, in which neither she nor I was allowed to know whether the bottle contained the alleged active ingredient or something else. This was what Professor Hyman, in John Diamond's hilarious description of a similar case, undertook. When, predictably, the 'alternative' technique ignominiously flunked the double-blind test, its practitioner delivered himself of the following immortal response: 'You see? That is why we never do double-blind testing any more. It never works! '
A large part of the history of science, especially medical science, has been a progressive weaning away from the superficial seductiveness of individual stories that seem - but only seem - to show a pattern. The human mind is a wanton storyteller and, even more, a profligate seeker after pattern. We see faces in clouds and tortillas, fortunes in tea leaves and planetary movements. It is quite difficult to prove a real pattern as distinct from a superficial illusion. The human mind has to learn to mistrust its native tendency to run away with itself and see pattern where there is only randomness. That is what statistics are for, and that is why no drug or therapeutic technique should be adopted until it has been proved by a statistically analysed experiment, in which the fallible pattern-seeking proclivities of the human mind have been systematically taken out of the picture. Personal stories are never good evidence for any general trend.
In spite of this, doctors have been heard to begin a judgement with something like, 'The trials all say otherwise, but in my clinical experience . . . ' This might constitute better grounds for changing your doctor than a suable malpractice! That, at least, would seem to follow from all that I have been saying. But it is an exaggeration. Certainly, before a medicine is certified for wide use, it must be properly tested and given the imprimatur of statistical significance. But a mature doctor's clinical experience is at least an excellent guide to which hypotheses
SNAKE OIL
185
? THEY TOLD ME, HERACLITUS
might be worth going to the trouble and expense of testing. And there's more that can be said. Rightly or wrongly (often rightly) we actually do take the personal judgement of a respected human individual seriously. This is so with aesthetic judgements, which is why a famous critic can make or break a play on Broadway or Shaftesbury Avenue. Whether we like it or not, people are swayed by anecdote, by the particular, by the personal.
And this, almost paradoxically, helps to make John Diamond such a powerful advocate. He is a man whom we like and admire for his personal story, and whose opinions we want to read because he expresses them so well. People who might not listen to a set of nameless statistics, intoned by a faceless scientist or doctor, will listen to John Diamond, not just because he writes engagingly, but because he was dying while he wrote and he knew it: dying in spite of the best efforts of the very medical practices he was defending against opponents whose only weapon is anecdote. But there is really no paradox. He may gain our ear because of his singular qualities and his human story. But what we hear when we listen to him is not anecdotal. It stands up to rigorous examination. It would be sensible and compelling in its own right even if its author had not previously earned our admiration and our affection.
John Diamond was never going to go gentle into that good night. When he did go it was with guns blazing, for the splendidly polemical chapters of Snake Oil occupied him right up to the end, working against . . . not so much the clock as time's winged chariot itself. He does not rage against the dying of the light, nor against his wicked cancer, nor against cruel fate. What would be the point, for what would they care? His targets are capable of wincing when hit. They are targets that deserve to be hit hard, targets whose neutralization would leave the world a better place: cynical charlatans (or honestly foolish dreamers) who prey on gullible unfortunates. And the best part is that although this gallant man is dead, his guns are not silenced. He left a strong emplacement. This posthumous book launches his broadside. Open fire, and don't stop.
186
? 5
EVEN THE RANKS OF TUSCANY
Stephen Jay Gould and I did not tire the sun with talking and send him down the sky. We were cordial enough when we met, but it would be disingenuous to suggest that we were close. Our academic differences have even been spun out to book length, by the philosopher Kim Sterelny in Dawkins vs Gould: Survival of the Fittest,^ while Andrew Brown, in The Darwin Wars: How Stupid Genes Became Selfish Gods,TM goes so far as to divide modern Darwinians into 'Gouldians' and 'Dawkinsians'. Yet, despite our differences, it is not just the respect due to the dead that leads me to include in this book a section on Stephen Gould with a largely positive tone.
'And even the ranks of Tuscany' (Steve would have completed the
quotation from his formidable literary memory) 'Could scarce forebear to
107
cheer'. Macaulay celebrated the admiration that can unite enemies in
death. Enemies is too strong a word for a purely academic dispute, but
admiration is not, and we were shoulder to shoulder on so much. In his
review of my own Climbing Mount Improbable, Steve invoked a collegiality 08
between us, which I reciprocated, in the face of a shared enemy:'
In this important uphill battle for informing a hesitant (if not outrightly hostile) public about the claims of Darwinian evolution, and for explaining both the beauty and power of this revolutionary view of life, I feel collegially entwined with Richard Dawkins in a common enterprise.
He was never ashamed of his immodesty, and I hope I may be forgiven for
sharing with my readers the one occasion when he was good enough to
include me in it: 'Richard and I are the two people who write about
109
evolution best . . . ' There was a 'but' of course, but I must press on.
The book reviews that follow, spaced many years apart, show what I hope will be read as an equal collegiality, even where they are critical. Ever Since Darwin was the first collection of Gould's celebrated essays from Natural History. It set the tone for all ten of them, and the 'barbed rave' tone of Rejoicing in Multifarious Nature (5. 1) could also serve for any.
187
? EVEN THE RANKS OF TUSCANY
The Art of the Developable (5. 2), though written in 1983, has not pre- viously been published. It is a joint review of Peter Medawar's Pluto's Republic and the third of Gould's collections of essays from Natural History. It was commissioned by the New York Review of Books but eventually, for reasons that I no longer recall, the publication fell through. Years later I sent the review to Steve, and he expressed warm disappointment that it had never been published. Medawar was one of my intellectual heroes, and Gould's too: it was another thing we had in common. My title, 'The Art of the Developable', unites Medawar's Art of the SolubleTM with Gould's long interest in the evolution of development.
Wonderful Life is, in my view, a beautiful and a misguided book. It is also misguiding: its enthusiastic rhetoric leading other authors to absurd conclusions far beyond Dr Gould's intentions. I developed this aspect fully in 'Huge Cloudy Symbols of a High Romance', one of the chapters of my Unweaving the Rainbow. Reprinted here as Hallucigenia, Wiwaxia and Friends (5. 3), the title given it by the Sunday Telegraph, is my review of Wonderful Life itself.
Human Chauvinism and Evolutionary Progress (5. 4) is my review of Full House, a book that was renamed by the British publishers Life's Grandeur. The review was published as a matched pair with Steve's review of Climbing Mount Improbable. The Editor of Evolution thought it would be amusing to invite each of us to review the other's book simultaneously, knowing the existence, but not the content of the other's review. Gould's review had the characteristic title 'Self-help for a hedgehog stuck on a molehill'. Full House is all about the idea of progress in evolution. I agree with Gould's objections to progress as he saw it. But in this review I develop two alternative meanings of progress which I think are important and are not vulnerable to his objections. My intention was not just to review a book but to make a contribution to evolutionary thinking.
Stephen Gould was my exact contemporary but I always thought of him as senior, probably because his prodigious learning seemed to belong to a more cultivated era. His lifelong colleague Niles Eldredge, who was kind enough to send me the text of his moving eulogy, said that he had lost an elder brother. Years ago it seemed natural to me to ask Steve's advice when I was travelling in America and was invited to have a televised 'debate' with a creationist. He said that he always refused such invitations, not because he was afraid of 'losing' the debate (the idea is laughable) but for a subtler reason which I accepted and never forgot.
Shortly before his last illness began, I wrote to him, reminding him of his advice to me and proposing that we might publish a joint letter, offering the same advice to others. He enthusiastically agreed, and suggested that I should prepare a
188
? draft on which we could later work together. I did so but, sadly, 'later' nev? r came. When I heard of his sudden death, I wrote to Niles Eldredge, asking if he thought Steve would have wished me to publish the letter anyway. Niles encouraged me to do so, and, as Unfinished Correspondence with a Darwinian Heavyweight (5. 5), it closes this section.
For good or ill, Steve Gould had a huge influence on American scientific culture, and on balance the good came out on top. It is pleasing that, just before his death, he managed to complete both his magnum opus on evolution and his ten-volume cycle of essays from Natural History. Although we disagreed about much, we shared much too, including a spellbound delight in the wonders of the natural world, and a passionate conviction that such wonders deserve nothing less than a purely naturalistic explanation.
EVEN THE RANKS OF TUSCANy
189
? Rejoicing in Multifarious Nature111 Review of Ever Since Darwin by S. J. Gould
'The author shows us what is revealed when we remove the blinkers which Darwin stripped from biology a century ago. ' Some overkill there, or an excitingly paradoxical striptease technique? The first essay in the book discusses Darwin's own coyness in not revealing his theory until 20 years after he thought of it, and I shall return to this. The quotation from the jacket blurb gives a false impression, for Stephen Gould's writing is elegant, erudite, witty, coherent and forceful. He is also, in my opinion, largely right. If there are elements of paradox and overkill in Dr Gould's intellectual position, they are not to be found within these covers. Ever Since Darwin is a collection of essays which first appeared as a regular monthly column in Natural History. Skilfully edited to flow in eight main sections, the 33 essays, of which I can mention only a sample, reinforce my feeling that scientific journalism is too important to be left to journalists, and encourage my hope that true scientists may be better at it than journalists anyway.
Gould's collection begins to bear comparison with P. B. Medawar's immortal The Art of the Soluble. And if his style does not quite make the reader chortle with delight and rush out to show somebody - anybody - the way Medawar's does, Gould is to be thanked for some memorable lines. No doubt puritan killjoys of Science for 'The People' will denounce the vivid and helpful anthropomorphism in 'Reproduce like hell while you have the ephemeral resource, for it will not last long and some of your progeny must survive to find the next one. ' But on second thoughts they may be too busy plotting the abolition of slavery in ants, or brooding over the deviationism of:
Natural selection dictates that organisms act in their own self-interest . . . They 'struggle' continuously to increase the representation of their genes at the expense of their fellows. And that, for all its baldness, is all there is to it; we have discovered no higher principle in nature.
190
? Ever since Darwin we have known why we exist and we have known at least how to set about explaining human nature. I agree that natural selection is 'the most revolutionary notion in the history of biology' and I would toy with substituting 'science' for 'biology'. Childishly simple as it is, nobody thought of it until centuries after far more complicated ideas had become common currency, and it is still the subject of misunderstanding and even apathy among educated people. A microcosm of this historical enigma is the subject of Gould's first essay. Just as humankind waited centuries longer than our hindsight deems necessary before discovering natural selection, so Darwin delayed his own publication 20 years after he first thought of the theory in 1838. Gould's explanation is that Darwin was afraid of the psychological implications of his idea. He saw what Wallace would never admit, that the human mind itself must be a material product of natural selection. Darwin, in fact, was a scientific materialist.
In another essay Gould is encouraged by the genetic closeness of humans and chimpanzees to speculate that 'inter-breeding may well be possible'. I doubt it, but it is a pleasing thought and Gould surely exaggerates when he rates it 'the most . . . ethically unacceptable scientific experiment I can imagine'. For my ethics, far less acceptable experiments are conceivable, and actually done in animal physiology laboratories every day, and a chimp/human hybridization would provide exactly the come-uppance that 'human dignity' needs. Gould is, in general, rather good at puncturing human speciesist vanity, and in particular he will have nothing to do with the myth that evolution represents progress towards man. This scepticism informs his valuable account of 'Bushes and ladders in human evolution', and fires his scorn for attempts to rank human races as primitive or advanced.
He returns to the attack on progress in the very different guise of the theory of orthogenesis, the idea that evolutionary trends have their own internal momentum which eventually drives lineages extinct. His telling of the classic Irish Elk story gains freshness from his intimacy with the fossils of the Dublin Museum and gives the lie to the myth that palaeontology is dry and dull. His conclusion that the proverbially topheavy antlers were important in social life is surely right, but he may underestimate the role of within-species competition in driving species extinct. Large antlers could directly have caused the extinction of the Irish Elk while at the same time, right up to the moment of extinction, individuals with relatively large antlers were out-reproducing indivi- duals with relatively small antlers. I would like to see Gould come to terms with the 'orthoselective' impact of 'arms races' both between and
REJOICING IN MULTIFARIOUS NATURE
191
? EVEN THE RANKS OF TUSCANY
within species. He seems to approach this in his essays on the 'Cambrian explosion'.
Natural history can be sold for its intrinsic fascination, but it is much better used to make a point. Gould tells us about a fly that eats its mother from the inside, about 17-year cicadas and 120-year bamboos, and about uncanny fish-decoying mussels. He employs the useful trick of first opening the reader's mind by boggling it, then filling it with the important biological principle. One principle I would have liked to have heard more on is that of the limitation of evolutionary perfection: 'Orchids are Rube Goldberg machines; a perfect engineer would certainly have come up with something better' (Rube Goldberg is the American Heath Robinson). My own favourite example, inherited from an undergraduate tutor, is the recurrent laryngeal nerve. It starts in the head, goes down into the chest, loops round the aorta, then goes straight back into the head again. In a giraffe this detour must be wasteful indeed. The human engineer who first designed the jet engine simply threw the old propeller engine out and started afresh. Imagine the contraption he would have produced if he had been constrained to 'evolve' his jet engine by changing a propeller engine one bit at a time, nut by nut and bolt by bolt!
While on the problem of perfection, I think Gould exaggerates the relevance of 'neutral mutations'. Molecular geneticists are understandably interested in DNA changes as molecular events, and any that have no effect on protein function may reasonably be called neutral mutations. But to a student of whole organism biology they are less than neutral; they are not, in any interesting sense, mutations at all! If the molecular neutralists are right, their kind of neutral mutation will forever be hidden from the field biologist and from natural selection. And if a field biologist actually sees variation in phenotypes, the question of whether that variation could be selectively neutral cannot be settled in the biochemistry laboratory.
Several essays touch on aspects of the relationship of Darwinism to human society and politics. There is much humane good sense here and I agree with most of it. Although 'sociobiology' is inspiring excellent research, Gould is right that it has also led to some second-rate bandwagoneering. 'But was there ever dog that praised his fleas? ' asked Yeats. Perhaps a dog may be held responsible for the fleas he sheds, but only to a small extent. At the AAAS meeting in Washington in 1997, Gould and I witnessed an organized attack on his most distinguished Harvard colleague. * Gould well deserved his ovation for the apt Lenin quote with which he disowned the rabble. But as he watched those
*A glassful of water was thrown sideways at Professor E. O. Wilson (subsequently exaggerated in various accounts to a 'pitcher of iced water, poured over him').
192
? pathetic fleas ineffectually hopping around the stage chanting, of all things, 'genocide', did he wonder with a little itch of conscience on which dog they had been sucking?
The Epilogue is forward-looking and whets our appetite for Volume 2, which I earnestly hope will be forthcoming. * One theme which I know Gould has already carried further in his Natural History column is his dislike for 'the ultimate atomism' of regarding organisms as 'temporary receptacles . . . no more than instruments that genes use to make more
112
genes like themselves'. In describing this as 'metaphorical nonsense'
Gould underestimates the sophistication of the idea, first cogently
113
expressed in its modern form by George C. Williams. The dispute is
largely semantic. Inclusive fitness is defined in such a way that to say 'the individual works so as to maximize its inclusive fitness' is equivalent to saying 'the genes work so as to maximize their survival'. The two forms are each valuable for different purposes. Both contain an element of personification; it is dangerously easier to personify organisms than to personify genes. The gene selection idea is not naively atomistic, as it recognizes that genes are selected for their capacity to interact productively with the other genes with which they are most likely to share 'receptacles'; this means the other genes of the gene pool; and the gene pool may therefore come to resemble a 'homeostatically buffered system' tending to return to (one of) its evolutionarily stable state(s). Irrevocable determination by genes is no part of the idea nor is anything remotely approaching a 'one gene, one trait' mapping from genotype to phenotype. In any case it has nothing to do with 'supreme confidence in universal adaptation', which is as likely to be found among devotees of 'individual selection' or 'species selection'.
'I will rejoice in the multifariousness of nature and leave the chimera of certainty to politicians and preachers': a resounding conclusion to a stimulating book - the work of a free and imaginative scientific mind. The final, sad paradox is this. How can a mind capable of such rejoicing, open enough to contemplate the shifting splendour of three thousand million years, moved by the ancient poetry written in the rocks, how can such a mind not be bored by the drivelling ephemera of juvenile pamphleteers and the cold preaching of spiteful old hardliners? No doubt they are right that science is not politically neutral. But if, to them, that is the most important thing about science, just think what they are missing! Stephen Gould is well qualified, and strategically placed, to strip away even those dark blinkers and dazzle those poor unpractised eyes.
*In fact, ten volumes were eventually published, the last one, / have Landed, at the time of his death.
REJOICING IN MULTIFARIOUS NATURE
193
? '%t0 M mem
The Art of the Developable
The acknowledged master of biological belles lettres has long been Sir Peter Medawar. If there is a younger biologist or an American biologist that bears comparison, it is probably in both cases Stephen Jay Gould. It was therefore with anticipation that I received these two collections of essays, reflections by leading and highly literate biologists on their subject and its history and philosophy.
Pluto's Republic is one of those titles that cannot be mentioned without an immediate explanation, and Sir Peter begins thus:
A good many years ago a neighbour whose sex chivalry forbids me to disclose [it takes a Medawar to get away with this kind of thing nowadays] exclaimed upon learning of my interest in philosophy: 'Don't you just adore Pluto's Republic? ' Pluto's Republic has remained in my mind ever since as a super- latively apt description of that intellectual underworld which so many of the essays in this volume explore. We each populate Pluto's Republic according to our own prejudices . . .
Here I nursed a mischievous half-hope that Stephen Gould might be found among the denizens of Medawar's private underworld - his more sanctimonious cosignatories of a notorious letter to the New York Review of Books about 'sociobiology' (13 November 1975) are prominent in mine. But Gould is several cuts above those former associates of his and he is not among Medawar's targets. Indeed, they share many targets, IQ- metricians for instance.
Most of the essays in Pluto's Republic have appeared twice before, first as book reviews or transcripts of lectures, then in previous anthologies
116
such as The Art of the Soluble and The Hope of Progress, which were
presumably reviewed at the time. Although I shall therefore give Pluto's Republic less than half my space in this joint review, I vigorously repudiate any mutterings about such second order anthologizing being too much of a good thing. The earlier books have long been out of
194
14 Hen's Teeth and Horse's Toes by Stephen Jay Gould
Review of Pluto's Republic by Peter Medawar'
and 11
? print, and I have been scouring the second-hand bookshops ever since my own Art of the Soluble was stolen. I discovered when I reread them here that I had many favourite passages word-perfect in memory. Who indeed could forget the opening sentence of the 1968 Romanes Lecture, 'Science and Literature'? T hope I shall not be thought ungracious if I say at the outset that nothing on earth would have induced me to attend the kind of lecture you may think I am about to give. ' At the time this prompted the apt rejoinder from John Holloway: 'This lecturer can never have been thought ungracious in his life. '
Or listen to Medawar on another great biologist, Sir D'Arcy Thompson:
. . . He was a famous conversationalist and lecturer (the two are often thought to go together, but seldom do), and the author of a work which, considered as literature, is the equal of anything of Pater's or Logan Pearsall Smith's in its complete mastery of the bel canto style. Add to all this that he was over six feet tall, with the build and carriage of a Viking and with the pride of bearing that comes from good looks known to be possessed.
The reader may be hazy about Logan Pearsall Smith and Pater, but he is left with the overwhelming impression (since he probably is familiar with the idiom of P. G. Wodehouse) of a style that is undoubtedly bel, and may very well be canto. And there is more of Medawar in the passage quoted than Medawar himself realized.
Medawar continually flatters his readers, implying in them an erudition beyond them, but doing it so that they almost come to believe in it themselves:
'Mill,' said John Venn in 1907, has 'dominated the thought and study of intelligent students to an extent which many will find it hard to realise at the present day'; yet he could still take a general familiarity with Mill's views for granted . . .
The reader scarcely notices that Medawar himself is still taking a general familiarity with Mill's views for granted, although in the reader's own case it may be far from justified. 'Even George Henry Lewes found him- self unable to propound his fairly sensible views on hypotheses without prevarication and pursing of the lips. ' The reader's knowing chuckle is out before he realizes that actually he is in no position to respond knowingly to that 'even'.
Medawar has become a sort of chief spokesman for 'The Scientist' in the modern world. He takes a less doleful view of the human predica- ment than is fashionable, believing that hands are for solving problems rather than for wringing. He regards the scientific method - in the right
THE ART OF THE DEVELOPABLE
195
? EVEN THE RANKS OF TUSCANY
hands - as our most powerful tool for 'finding out what is wrong with [the world] and then taking steps to put it right'. As for the scientific method itself, Medawar has a good deal to tell us, and he is well qualified to do so. Not that being a Nobel Prize winner and a close associate of Karl Popper is in itself an indication that one will talk sense: far from it when you think of others in that category. But Medawar not only is a Nobel Prize winner, he seems like a Nobel Prize winner; he is everything we think a Nobel Prize winner ought to be. If you have never understood why scientists like Popper, try Medawar's exposition of the philosophy of his 'personal guru'.
He read Zoology at Oxford, and early in his career made important contributions to classical Zoology, but was soon drawn into the highly populated and highly financed world of medical research. Inevitably, his associates have been molecular and cell biologists, but he seldom had any truck with the molecular chauvinism which plagued biology for two decades. Medawar has a good appreciation of biology at all levels.
He has also inevitably associated with doctors, and the preoccupa- tions and sympathies of a doctor pervade several of these essays, for example his sensitive reviews of books on cancer and psychosomatic heart disease. I especially enjoyed his blistering contempt for psycho- analysis: not a lofty, detached contempt for any ordinary pretentious drivel, but a committed contempt, fired by a doctor's concern. Psychoanalysts have even had their say over the puzzle of Darwin's long illness, and Medawar is at his withering best in telling us about it.
For Good, there is a wealth of evidence that unmistakably points to the idea that Darwin's illness was 'a distorted expression of the aggression, hate and resentment felt, at an unconscious level, by Darwin towards his tyrannical father'. These deep and terrible feelings found outward expression in Darwin's touching reverence towards his father and his father's memory, and in his describing his father as the kindest and wisest man he ever knew: clear evidence, if evidence were needed, of how deeply his true inner sentiments had been repressed.
Medawar, when he smells pretentious pseudoscience, is a dangerous man. His famous annihilation of Teilhard de Chardin's The Phenomenon of Man might have been thought an unfair attack on the dead, but for the extraordinary influence Teilhard exerted (and still exerts: Stephen Gould tells us that two journals established to discuss his ideas still flourish) over legions of the gullible including, I am afraid, my juvenile self. I would love to quote huge chunks of what is surely one of the great destructive book reviews of all time, but must content myself with
196
? two sentences from Medawar's typically barbed explanation of the popular appeal of Teilhard.
Just as compulsory primary education created a market catered for by cheap dailies and weeklies, so the spread of secondary and latterly of tertiary education has created a large population of people, often with well-developed literary and scholarly tastes who have been educated far beyond their capacity to undertake analytical thought . . . [The Phenomenon of Man] is written in an all but totally unintelligible style, and this is construed as prima-facie evidence of profundity.
Medawar's Herbert Spencer Lecture, and his review of Arthur Koestler's Act of Creation, are more respectful of his victims, but pretty punchy nevertheless. His review of Ronald Clark's Life of J. B. S. Haldane is enlivened by personal reminiscence, and reveals a sort of affection for the old brute which seems to have been reciprocated.
I remember Haldane's once going back on a firm promise to chair a lecture given by a distinguished American scientist on the grounds that it would be too embarrassing for the lecturer: he had once been the victim of a sexual assault by the lecturer's wife. The accusation was utterly ridiculous and Haldane did not in the least resent my saying so. He didn't want to be bothered with the chairmanship, and could not bring himself to say so in the usual way.
But if Haldane did not in the least resent Medawar's saying so, one cannot help wondering whether this was only because Medawar must have been one of the very few people Haldane ever met who could look him levelly in the eye, on equal terms intellectually. Peter Medawar is a giant among scientists and a wicked genius with English prose. Even if it annoys you, you will not regret reading Pluto's Republic.
In 1978 the Reviews Editor of a famous scientific journal, whose nature prudence forbids me to disclose, invited me to review Stephen
Jay Gould's Ever Since Darwin, remarking that I could 'get my own back' on opponents of 'genetic determinism'. I don't know which annoyed me more: the suggestion that I favoured genetic 'determinism' (it is one of those words like sin and reductionism: if you use it at all you are against it) or the suggestion that I might review a book for motives of revenge. The story warns my readers that Dr Gould and I are supposed to be on opposites sides of some fence or other. In the event, I accepted the commission and gave the book what could fairly be described as a rave review, even, I think, going so far as to praise Gould's style as a creditable second best to Peter Medawar's. *
*See 'Rejoicing in Multifarious Nature' (pp. 190-3).
THE ART OF THE DEVELOP ABLE
197
? EVEN THE RANKS OF TUSCANY
I feel inclined to do the same for Hen's Teeth and Horse's Toes. It is another collection of essays reprinted from Gould's column in Natural Histor)'. When you have to turn these pieces out once a month you must pick up some of the habits of the professional working to a deadline - this is not a criticism, Mozart did the same. Gould's writing has some- thing of the predictability that we enjoy in Mozart, or in a good meal. His volumes of collected essays, of which this is the third, are put together to a recipe: one part biological history, one part biological politics (less if we are lucky), and one part (more if we are lucky) vignettes of biological wonder, the modern equivalent of a mediaeval bestiary but with interesting scientific morals instead of boring pious ones. The essays themselves, too, often seem to follow a formula or menu. As appetizer there is the quotation from light opera or the classics, or sometimes its place is taken by a piece of reassuring nostalgia; a reminiscence from a normal, happy, very American childhood world of baseball stars and Hershey bars and Bar Mitzvahs - Gould, we learn, is not just one of your pointy-headed intellectuals but a regular guy. This homely informality softens the conspicuous erudition of the main course - the fluency in several languages, the almost Medawarian familiarity with literature and humanities - and even gives it a certain (un-Medawarian) charm (compare Gould himself on Louis Agassiz: '. . . the erudition that has so charmed American rustics . . . ').
Gould's own respect for Medawar is evident. The idea of science as 'the art of the soluble' provides the punchline for at least four of the essays: 'We may wallow forever in the thinkable; science traffics in the doable', '. . . science deals in the workable and soluble'; and two essays end with explicit quotations of the phrase. His view of Teilhard de Chardin's style is similar to Medawar's: '.
