by Simon Mawer
The invaders brought with them famine and cholera. The hospital just up the hill from the monastery filled to overflowing. The bells of the Augustinian church tolled almost continually for the dead (until the authorities forbade the practice because it was damaging to the citizens’ morale). And all that torrid, pestilential summer, Mendel wrote and revised and rewrote.
The paper on hybridization in the garden pea was published in the Proceedings of the Brünn Society for the Study of Natural Science. It was sent to 120 other societies and organizations around Europe. Copies went to the Universities of Vienna and Berlin, to the Royal Society and the Linnaean Society of London, to the Royal Horticultural Society at Kew, to Uppsala and Paris and Rome and St. Petersburg. No one read it. This was one of the three most significant and famous papers in the whole history of biology,5 and no one took any notice. He also had forty offprints made of the paper, but we know of the fate of only five of them. To whom the others were sent, we just don’t know. One imagines Darwin, one imagines Haeckel, one imagines Huxley, one imagines Purkyn?. But we don’t know.
By the time of publication, the Prussians had gone and the city of Brünn appeared peaceful again. More than that, it appeared unchanged. Once again the Lord Lieutenant had taken up his position in the city. Once more the Estates were meeting in the Landhaus. The Empire, that shambolic collection of German, Magyar, Slav, Italian, and Jew, had been left untouched. Its borders were entire. Once more the military bands played in the Augarten—Strauss they played, Strauss, Strauss, Strauss—for all the world as though the Imperial Army had not just been defeated in war.
It is one of the dangers of the historical perspective to mistake the momentous for the mundane. Nothing much had changed except that the balance of Middle Europe had been reset. Nothing much had changed except that the German people had stumbled incoherently—they could hardly be accused of efficiency in the matter—a further step toward the apocalypse. Nothing much had changed except that an unknown friar, shortly to be elected abbot, had discovered the mechanics of inheritance and had, all unbeknownst to himself, created a new science that was to be taken up by the Gesellschaft für Rassenhygiene (the Society for Racial Hygiene) in 1905 and the Nazi Party two decades later. It was a science that would ultimately lead to the ovens of Auschwitz.
1. Handyside et al. Lancet i: 347–49 (1989); Coutelle et al. British Medical Journal 299: 22–24 (1989).
2. Restriction enzyme SfcI.
3. Versuche über Pflanzen-Hybriden, 1866. What would now be called polygenic inheritance: a brilliant further insight into genetic theory. This idea would quite escape the so-called rediscoverers of his work at the start of the twentieth century.
4. Letters to Nägeli, July 1870 and September 1870.
5. There is no competition. The other two are the Darwin-Wallace paper on evolution by means of natural selection, delivered to the Linnaean Society (1858); and the Crick-Watson letter to Nature on a suggested structure of DNA (Nature, 1953). There are no papers greater than these; on these hang all the law and the prophets.
Mendel cheated. Oh yes, that’s the story. Useful, isn’t it? The Stalinists, in their desperation to demonstrate that Mendelism was a fraud, nothing more than a capitalist-fascist plot, used this calumny to support their point of view: Mendel cheated, genetics is a lie, Lysenko is right, the environment is everything, man can be molded by his circumstances, the revolution will create a true socialist environment, and man will fit perfectly into the earthly paradise like a hand into a glove. And thus the great experiment of Communism finds justification for its view and millions have to die before everyone (well, the majority at least) tumbles to the fact that the evidence is now stacked against the hypothesis of an earthly paradise and the great experiment can be brought to a close.
Mendel cheated.
But it was not some Soviet toady laboring away in a genetics laboratory somewhere in Omsk or Tomsk who caught the great man out; it was Sir Ronald Fisher. I particularly like the use of the title—it makes the accusation so much more authoritative. Sir Ronald Aylmer Fisher (1890–1962), graduate in mathematics at Cambridge University, sometime professor of eugenics at London University.
Eugenics? Does your mind stall? Do you feel shivers down your spine? Does the flesh on the back of your neck crawl? Oh yes—there was a professor of eugenics at London University. Fisher occupied the Galton Chair of Eugenics, founded by Charles Darwin’s cousin and first held by a brilliant racist called Karl Pearson (he of the chi-square test and the Pearson correlation coefficient, biologists and statisticians please note). There was a chair of eugenics at Cambridge as well, but the university had the decency to change the name of their department to plain genetics in 1943 when Fisher moved there; by that time, presumably, the stench from continental Europe was becoming unbearable. In London the senses cannot have been so acute: the title of the Galton Chair was not changed to plain “Genetics” until 1961.
So, Mendel cheated. The problem is, you see, his results were too good.
EXAMPLE:
F1 generation total: 1,064 pea plants; of which 787 tall and 277 dwarf.
Theoretical ratio 3:1. Actual ratio 2.84:1.
It is rather close, isn’t it? But that is not quite the point. You could toss a coin one hundred times and find that it came up heads forty-eight times and tails fifty-two times, and no one would be too surprised. But if you claimed that every time you repeated the experiment it came up similarly close to 50:50, people might start getting suspicious. The tall:dwarf values I have just quoted are almost the worst that Father Gregor found. His other experimental ratios are all as close or closer to the ideal 3:1.
2.96:1 3.01:1 2.95:1 3.15.1 2.82:1 3.14:1
There they are, the actual values. The problem with Mendel’s work is that time after time, repetition after repetition, his results were simply too close to the expected ratios. Expected by whom? By Mendel, of course. It was Sir Ronald Fisher who showed that the probability that Mendel could come consistently so close to his expected ratios by pure chance was so small as to be negligible. Considering Mendel’s 3:1 ratios alone, the probability of his having got greater deviations from the expected ratios than he actually found is .95. In laymen’s terms, Mendel had a ninety-five-percent chance of getting worse results than he did. Put backwards, he had only a five-percent chance of getting as perfect a set of results as he did. Ergo, Mendel cheated. Putting all his known results together, the probability of his having got greater deviations from the expected ratios than he actually found is .99993. That means that he had a 99.993-percent chance of getting worse results than he did. Put backwards, he had only a .007-percent chance of getting as perfect a set of results as he did, which is no chance at all.
So, he cheated. Mendel spent a decade of his life on his breeding experiments on the garden pea, further tested the validity of his theories on Antirrhinum, Matthiola, Fuchsia, Campanula, and a further eighteen species, and, thanks to the idiot Nägeli, wasted God knows how much time on trying to repeat the work on Hieracium—and all the time he cheated.
The trouble is, he was right.
The ninth edition of the Encyclopaedia Britannica was published with an article on “Hybridisation” by G. J. Romanes. Romanes was one of the most devoted disciples of Charles Darwin, and had consulted the great man at length during the preparation of the article. Darwin recommended that he read W. O. Focke’s book Die Pflanzenmischlinge, and, moreover, he actually lent Romanes his own copy. This book, published in 1881, outlined Mendel’s work on Pisum, Phaseolus, and Hieracium, and also mentioned him in the historical section, which Darwin particularly recommended Romanes to read. Romanes duly researched and wrote the article, and the name G. Mendel duly appeared in the bibliography Darwin’s copy of Die Pflanzenmischlinge was duly returned, the pages for the work on the Papilionaceae still uncut, as they remain today; which is doubly curious as one of those pages (110) also mentions Darwin’s own work with garden peas. Indeed, the references to Mendel and to Darwin ar
e immediately adjacent to each other, the two names separated by “(loc. cit.)” and a period.
Who was it who said, “You will find it a very good practice always to verify your references”? Darwin didn’t check his references, and neither did Romanes. They cheated. And Darwin needed Mendel. Oh yes, indeed, Darwin needed Mendel. As far as he had any clear ideas about a matter that was of prime importance to his theory of evolution by natural selection, Darwin believed in the blending theory of inheritance. That is, he held that offspring tend to be a blend of their parents’ characteristics.
The trouble is, he was wrong.
By logical extension of this blending theory, a species should show less and less variation over a number of generations. Any artist knows that if you take the whole spectrum of colored paints and solemnly mix them together in pairs, finally and inevitably you will end up with muddy brown. And any naturalist knows that this is not what happens with plants and animals. Like anyone else with eyes, Darwin looked around him and, rather than muddy brown, found a bewildering, dazzling range of variation within each species. He saw, to use the technical term, polymorphisms. So to account for this undeniable variety he further postulated a high degree of spontaneous variation—what we would now call mutation.
The trouble is, he was wrong again. A high mutation rate implies an instability in the genetic material, which in turn would mean that you couldn’t guarantee what you inherited from your parents. In such a case, natural selection simply wouldn’t occur because the genes selected in one generation wouldn’t necessarily be passed on to the next. By the time they got there, they would probably have mutated.
No, what Darwin needed was Mendel. And he recommended a book that referred to Mendel’s work (a total of fourteen separate citations), and Romanes even quoted Mendel in his bibliography, and neither of them verified the references.
In fact Mendel had already seen this difficulty of the blending theory in the Origin of Species and come to Darwin’s rescue—in his Pisum paper he points out that if you cross parents differing in seven pairs of characters and then you allow the hybrid offspring to self-fertilize, in the second generation you will have 2,187 different genetic constitutions. He even generalized the rule, in one of his most brilliant insights: if n designates the number of characteristic differences in two parental plants, then 3′ is the number of genetically different individuals produced in the second generation after self-fertilization. Assuming that all the character pairs show complete dominance, then 2′ is the number of different combinations of phenotypes that would occur. Thus, in the case of his seven pairs of characters, you would obtain 128 different phenotypic combinations. That’s where the variety that Darwin needed so desperately comes from, from a reshuffling and recombination of Father Gregor’s factors; and the full explanation is there in his original paper. And no one noticed.
Later, in response to criticism of this very weakness, Darwin moved toward a belief in the inheritance of acquired characteristics. In The Variation of Animals and Plants Under Domestication (1868—note the date; it was just two years after Mendel’s paper was published) he envisaged body cells shedding hereditary particles, called gemmules or pangenes, into the blood. These entirely fictitious things, these fabrications, were visualized by him as models of the cells from which they come. They subsequently assemble to form the sex cells and thereby get passed on to the next generation. But because they originate from body cells they may therefore be affected by whatever has happened to the parent cells. Thus the effects of the environment on the body cells will end up being inherited.
That’s Darwin.
The trouble is, he was wrong again. He also seemed unaware that this particular theory completely contradicted the blending theory. Of course, it is never difficult for human beings to hold two contradictory beliefs at the same time. Look how many believe in a merciful and loving God, despite all the evidence to the contrary. Oh no, contradictory beliefs are by no means impossible; but they’re not very scientific.
At exactly the same time as Mendel was working so brilliantly, so doggedly, with such piercing insight into the matter of inheritance, August Weismann of the University of Freiburg im Breisgau performed an experiment of mind-boggling stupidity to disprove the theory of the inheritance of acquired characters. This experiment involved chopping off the tails of mice. Weismann bred mice through five generations, more than nine hundred of the wretched animals, laboriously chopping off all their tails.
And by the fifth generation? Mice with tails.
I wonder whether his colleagues tried to hush the whole thing up. Or maybe he himself tried to keep it quiet, working late into the evening when no one else was around and keeping the cages and cages of tailless mice behind locked doors. Chop, chop, chop. Disposal would have been a problem. What did the cleaning lady imagine the good professor was up to? Or did he wrap the tails in newspaper and slip them into some rubbish bin on the way home? Chop, chop, chop. Did Weismann imagine he was contributing to the sacred body of man’s knowledge? Chop, chop, chop. Five generations. At least the iniquity of the fathers is only visited upon the children unto the third and fourth generation. There is something quintessentially Teutonic about Weismann’s insistence on going one further than God. He showed this talent in other directions, becoming the first honorary chairman of the Society for Racial Hygiene.
Mendel kept mice. I’ve told you that. I’ll bet he didn’t do anything so idiotic as cut their tails off.
I keep mice. We have thousands of them in the animal room at the laboratories, tiny, mewing creatures with pink noses and twitching whiskers. Some of them are monstrously deformed.
Let us listen to E. B. Ford, sometime Emeritus Professor of Ecological Genetics at Oxford University and friend of that enthusiastic eugenicist Leonard Darwin:
The total number of plant and animal species now described lies between 1,100,000 and 1,200,000. It would have been far less in Mendel’s day, but still very large. Yet he based his views upon a single species: the edible pea, Pisum sativum. It is true that he corroborated them to a slight extent by work on, unfortunately, a related plant: the bean, Phaseolus. He also published the results of his experimental crosses with the hawk-weeds, Hieracium … Thus Mendel’s conclusions, though probably developed from a consideration of living organisms in general, were really only established from his monumental study on peas. Are the principles apparently derived from experiments upon a single species really applicable to over a million others, exhibiting all the diversity of animal as well as of plant life? It seems questionable indeed. Oddly enough, it would not have done so had Mendel merely used one other, chosen with discrimination …
Antirrhinum, Aquilegia, Calceolaria, Campanula, Carex, Cheiranthus, Cirsium, Dianthus, Ficaria, Fuchsia, Geum, Hieracium, Ipomoea, Linaria, Lychnis, Malus, Matthiola, Mirabilis, Phaseolus, Pirus, Potentilla, Prunus, Tropaeolum, Verbascum, Veronica, Viola, Zea.
Oddly enough, Mendel tried as best he could. It was just that everyone else was too stupid to understand what he had done. One wonders how E.B. himself would have measured up …
Cyril Burt cheated, of course. We all know that now (or almost all of us, but there are already some revisionists around). The curious thing is, we should have known it all along. We should have looked into his figures, poked around in them, looked up his references (tricky, that one, because a good number of them simply didn’t exist), generally picked at the fabric of his work to see whether it would come apart at the seams. By we I mean they, of course—the people who took it all at face value and actually encouraged its use in education: in the eleven-plus examination. Pigeonholing at eleven years old. The children of the middle class go to the grammar schools; the children of the working class go to the secondary moderns.
What Cyril Burt set out to show was that intelligence is inherited, or such a large portion of it as makes no difference. He did it by testing people’s intelligence. He tested people at random, he tested members of families unto the third and f
ourth generation, he tested identical twins. And he came to the conclusion that intelligence is about as inherited as, say, shortsightedness.
Burt used an array of tests and then labored long and hard at the mathematical analysis of his results. He also used straightforward personal observation, a curious method that appears to go something like this: you chat with someone; you are an expert in such matters; off the top of your head you decide what his IQ score is; you are right. He did all this and, following in the footsteps of Spearman, who had come to the conclusion that intelligence is a unitary thing, which he called g, and whatever you’ve got you’ve got, and we can’t do much about it, Burt decided that if you can identify a person’s intelligence at an appropriate age (eleven was the earliest age possible for reliable identification), then you can decide what education such a person needs and deserves. There’s nothing crueler than raising false hopes in a child, is there? No more attempt to train a dwarf as a basketball player than give an average man an academic education.
Now, I don’t want to kick a man when he’s down (actually I do—given my particular disadvantage, it’s the only way I get the opportunity), nor do I wish to speak ill of the dead, although if the libel laws don’t allow you to speak ill of them when they are alive, then I don’t see that you’re left with much alternative. But the blunt fact is, Cyril Burt was a fraud. He invented data to fit his prejudices and he even invented coworkers to fit his data. He was a lifelong scandal, and all in the name of genetics. But I did all right by him. I passed the eleven-plus. So, probably, did you, if your parents couldn’t afford a private school and you’re anything near my age and are reading this book.
Consider this little gem: shortly after the death of Alfred Binet in 1911, one of his admirers, Henry Goddard, administered the Binet test (adapted for English-language use) on behalf of the U.S. Public Health Service to immigrants at Ellis Island. He used two women to administer the tests, because women are gentler and more sympathetic. In 1913, working with Hungarians, Italians, Russians, and Jews, these two ladies discovered that 80 percent of these immigrants were feebleminded, the percentage differing little from group to group. Eighty percent.
