A recent issue of the Times Literary Supplement has excited considerable attention because of a long article by C.P. Snow, a rebuttal to the critics of Two Cultures. The article is dignified, reasonable, and entirely predictable, essentially a pièce justificative, adding nothing new to the controversy or to the original essay apart from the substitution of DNA for the Second Law of Thermodynamics as a test of scientific literacy. The rest of the issue, however, is more noteworthy. Entitled “The Art of Science,” it may be taken as a subtle rebuke to Sir Charles and a corrective to the popular image of the Two Cultures—a scientific culture that is essentially rational, logical, and evidential, and a literary culture that is essentially intuitive, emotive, and imaginative. What the TLS is now suggesting is that the scientific enterprise itself is, at its best, a kind of art, and that the two cultures have more in common than is generally thought.

The theme is best stated in the article “Imagination and Hypothesis” by Dr. P.B. Medawar, the distinguished English biologist, Nobel Prize winner, and director of the National Institute for Medical Research. Dr. Medawar refutes five of the most commonly accepted myths about science: the myth of a “scientific mind,” of “the scientific method,” of observation and deduction à la Bacon, of induction à la Mill, and of a “natural law” arrived at by rational, objective processes. Dr. Medawar concludes:

Scientists and laymen should become aware of the inspirational character of scientific discovery and of the hypothetico-deductive foundation upon which scientific understanding rests…. At every level of endeavor scientific research is a passionate undertaking, and the Promotion of Natural Knowledge depends above all else upon a sortie into what can be imagined but is not yet known.

The following article, “Portrait of the Scientist as Artist,” makes much the same point. The Professor of Chemistry at Cambridge, H.C. Longuet-Higgins, agrees that “the scientist is by nature an artist,” that “his ways of thought are primarily intuitive and imaginative,” and that the fiction of a “scientific method” was invented to “account for the unaccountable: the demonstrable capacity of the human mind to create ideas out of the void and to discriminate between the significant and the trivial.”

There should be no need to remind us, as Dr. Medawar does, that “hypothesis” and “imagination” are not pejorative words. We do, however, have to be reminded that they introduce a problematic quality into science. Science does not merely accumulate certainties; it also accumulates uncertainties. The more science knows, the more it knows how little it knows. Yet the desire for certainty is so pervasive that even scientists succumb to the popular superstitions of science.

As if to dramatize this point, the same issue of the TLS contains a perfect specimen of the pseudo-scientific mentality, in which hypothesis and imagination of a not particularly distinguished order are given far more authority than they merit. The example is the more egregious because it occurs in an area of science that is, or should be, at every moment acutely aware of the fragile, fragmentary, and tentative nature of its hypotheses. The review is of a book by Richard Carrington on The Million Years of Man which, according to the reviewer, starts with the stellar universe and the history of the earth, goes through the “probable evolution of life,” describes the development of man, mind, and civilization, and concludes with a chapter on the future of “human destiny and the individual.” Whether the author makes more of the “probable” character of these wide-ranging deductions I do not know. But the reviewer takes issue with him on only two points—the interpretation of the Paleolithic cave paintings and the future of humanity. For the rest the reviewer has only praise, singling out for special commendation the “charming reconstructions of prehistoric scenes, especially one of a Mesolithic riverside settlement where the girls are wearing clothes that look remarkably like modern gym tunics.”

American anthropologists and museum curators tend to clothe their prehistoric women in sarongs rather than gym tunics, but they produce equally charming and confident reconstructions. Evolution, a recent addition to the “Life Nature Library,” has illustrations and text in the same glossy finish and sharp focus, the brilliance of the photographs conspiring with the lucidity of the prose to induce in the reader a state of hypnotic acquiescence. There are, for example, sketches of the faces of the shark, lizard, opossum, lemur, monkey, gorilla, and man, accompanied by the explanation that in these “successive stages” of progress, “the eyes work their way to the front of the head for true binocular vision” and “the head becomes increasingly spherical—the most efficient shape for maximum brain in minimum skull.” The pictures and captions fit so neatly together that it is a shame to be reminded of Von Helmholtz’s remark that the human eye had “every possible defect that can be found in an optical instrument, and even some which are peculiar to itself”; or of Dr. Medawar’s complaint, in an earlier article, that the entire human structure is sadly deficient in many respects. The “successive stages” of progress that Life attributes to organic life may be neither so successive nor so progressive as this or any other tidy scheme would have it.


It also seems presumptuous to point out to Ruth Moore and the editors of Life (who are jointly listed as the authors of this book, in consultation with various professors and curators) that the process of reconstructing a head from fragments of bone is considerably less certain and precise than is here made out. Miss Moore, indeed, has good reason to appreciate just how uncertain and imprecise (not in details but in crucial respects) such reconstructions may be. It is generally regarded as bad form to make too much of the Piltdown fraud, and the Life book casually dismisses it as a “find which many years later embarrassingly turned out to be a hoax.” Yet Miss Moore herself was unfortunate enough, on the very eve of its exposure as a fraud, to have published a book in which Piltdown man was given pride of place in the human lineage. And then, as now, Miss Moore was only reflecting the best scientific opinion. For forty years, until less than ten years ago, some of the most eminent paleontologists in the world not only gave credence to what Miss Moore euphemistically describes as a “hoax,” but soberly argued the merits of various reconstructions of a skull combining a modern human brain with an ancient ape jaw, performed anatomical experiments demonstrating the exact muscular changes that must have taken place in the evolutionary process, and proved mathematically that the required changes could well have taken place in the available period of time. Neither Miss Moore nor even the paleontologists most deeply committed to particular interpretations of Piltdown man seem to have learnt much from the experience, and the present volume is full of the same kind of confident assertions based on the same kind of evidence—evidence that may or may not be true, that may or may not be relevant.

If it is suggested that a Life book or a museum exhibit is too easy a mark, a recently translated work by one of the world’s leading paleontologists, G. H.R. von Koeningswald’s Evolution of Man, can be taken in illustration. What is conspicuous here are not the glib certitudes of the popularizer, but the practical dilemmas of the conscientious working scientist. Almost every page provides evidence of the infinitely complicated and problematic state of paleontology:

As we can see, paleontologists base their classifications chiefly on the canine-premolar section of the jaw, which is far too frequently missing. Added to this is the difficulty that no one is quite certain of the form of the original human canine. Hence types with large canines are far too readily classified as pongids [simians], on the unwarranted assumption that the teeth of modern apes can be used as a general criterion.

A great difficulty arises from the claim that, since australo-pithecines made fire and tools, they must be considered full “men.” We saw that Dart called his Makapan specimen Australo-pithecus prometheus, the fire-raiser, because of the charcoal near him. However, the mere discovery of pieces of charcoal does not entitle us to say that they were produced by men, for spontaneous bushfires are a far more likely explanation.

More equivocal still is the tool problem. No matter what the real explanation of the strange indentations on baboon skulls may be, they were certainly not caused in the way Dart has suggested. According to him, a number of evenly damaged humeri must have been used as percussion instruments, but we agree with Zapfe that the damage was more probably caused by byenas.

Unfortunately we know nothing about the Pleistocene ancestry of either the gorilla or the chimpanzee, and little about their possible relationship with Tertiary European types. The only species whose prehistory is known to any extent is the orangutan.

The evolution of the rest of the skeleton [apart from the jaw] can be reviewed very briefly—very little is known about it.

Yet the book is as prolifie in unwarranted assumptions and contradictions as in qualifications and cautions. The first quotation cited above, for example, on our ignorance of the “form of the original human canine,” is followed immediately by the statement: “Sivapithecus was said by Pilgrim to be related to man, but its large canines make this classification doubtful”!

Of all scientists it is the scientist turned historian who is most readily victimized by the myths described by Dr. Medawar. An intellectual regression seems to set in as soon as the scientist regresses into history. That the scientist is not the most reliable guide to his own history—nor, perhaps, the best arbiter between the Two Cultures—is suggested by the avalanche of Darwiniana released by the centenary of the Origin of Species four years ago, when some of the most estimable scientists were responsible for some of the most mindless pieties.


On the other hand, recent weeks have seen the appearance of three volumes of selections from the works of leading protagonists in the original Darwinian controversy, each volume well chosen, well edited, and provocative, both in their introductions and texts. One was edited by a historian, the other two by literary critics whose more normal medium is the Hudson Review or Kenyon Review. The three volumes, apart from the suggestive circumstances of their editing, provide a fascinating exercise in dialectic, and their simultaneous publication is providential for the reviewer. Asa Gray was Darwin’s chief supporter in America; Louis Agassiz was Darwin’s—and Gray’s—chief critic in America; and Darwin himself was ambivalent towards Gray’s main thesis while being curiously dependent upon one of Agassiz’ main theories. The complications and anomalies revealed in these volumes go far to undermine both the conventional view of Scientific Method and the conventional reading of scientific history.

Although the Origin of Species is supposed to have burst like a bombshell upon the unsuspecting scientific world, to say nothing of the lay world, the public debate between Gray and Agassiz on the subject of evolution started several months before the appearance of the Origin. Indeed, the issue was joined two years earlier, with the publication of Agassiz’ “Essay on Classification,” which is an almost point-by-point refutation of the main arguments of the Origin. Refutation, and at the same time anticipation of the Origin. For Agassiz was as much a progenitor of Darwin as an antagonist. The burden of Agassiz’ five-volume work on fossil fishes was, as he put it, the “correspondence between the succession of fishes in geological times and the different stages of their growth in the egg.” This parallelism between paleontology and embryology Darwin claimed to be “by far the strongest single class of facts” in favor of his theory. When Huxley, several years before the Origin, disputed one of Agassiz’ favorite examples of such parallelism, Darwin confessed that “though I saw how excessively weak the evidence [of Agassiz] was, I was led to hope in its truth.” Five years later, describing Agassiz’ theory in the first edition of the Origin, he combined skepticism and hope in equal proportions:

I must follow Pictet and Huxley in thinking that the truth of this doctrine is very far from proved. Yet I fully expect to see it hereafter confirmed, at least in regard to subordinate groups. For this doctrine of Agassiz accords well with the theory of natural selection.

Although subsequent years brought no further confirmation of Agassiz’ theory and no reason to discount the evidence of Pictet and Huxley, the final edition of the Origin, published in 1872, saw the triumph of hope over doubt. The theory was then ascribed to “Agassiz and several other highly competent judges”; the sentence about Pictet and Huxley was omitted; and the passage ended confidently: “This view accords admirably well with out theory.”

The affinity between Darwin and Agassiz went far beyond this one theory. As Gray ‘himself said: “There is a significant correspondence between the rival theories [Darwin’s Natural Selection and Agassiz’ anti-evolutionism] as to the main facts employed. Apparently every capital fact in the one view is a capital fact in the other. The difference is in the interpretation.” The point should be noted by biographers of Darwin, who almost invariably present the Voyage of the Beagle as a triumphant voyage of discovery. Every mention of a fossil, every allusion to a resemblance between a fossil and a living species, every reference to the geographical distribution of species, the relations of species in the same area or in different areas, the range of variations within species—all these are pounced on as premonitions of the Origin. And the assumption is made that Darwin not only discovered the theory of natural selection as a logical or necessary deduction from these facts, but that he discovered the facts themselves. Yet, as Gray admitted and Agassiz demonstrated, these facts about fossils, embryology, homology, geographical distribution, etc., were not only the common property of all naturalists; they could be used as effectively against the theory of natural selection, indeed against any theory of evolution, as for it.

Where Agassiz differed from Darwin was in denying the utilitarian and functional character of these facts. The correspondence between fossil forms and living embryos, or between species widely disparate in time and place, suggested to him a “unity of plan”—but a unity of plan revealing no purpose or use except to testify to the will of God. For what utilitarian function, he protested, was served by the fact that the embryo of a living egg re-capitulated the ancient succession of fossils? Why, in the name of utility, should an arctic species differ only in the smallest details from a tropical one, while in important respects exhibiting the same unity of plan? And why should species exist only as “categories of thought”—for Agassiz agreed with Darwin that species had no material reality—if it were not to demonstrate the “intellectual connection” imposed by God upon nature?

It is curious that Agassiz, who generally figures in the history of Darwinism as the champion of religion against science, should have been remembered by Henry Adams as the only professor who contributed to his education during four years at Harvard, and by William James as the scientist who taught him to love the concrete and to distrust the abstract. It is also curious that the very qualities admired by Adams and James, qualities that made him one of the most respected scientists of his generation, should have been so largely responsible for his opposition to theories of evolution. His “Essay on Classification” was as much an attack on the metaphysics of Natural Theology as on the science of the evolutionists. It was the teleological character of both, the ascription of purpose and design to nature, that he condemned as a violation of the true facts of nature and of the true dignity of God.

There is little doubt that the theory of Natural Selection, as presented in the Origin and as understood both by Darwin and by most of his disciples, was essentially teleological. Thus Asa Gray, in his enthusiastic review of the book, praised it for showing how the system of nature had “received at its first formation the impress of the will of its Author, foreseeing the varied yet necessary laws of its action throughout the whole of its existence, ordaining when and how each particular part of the stupendous plan should be realized in effect.” Darwin was so pleased with Gray’s other articles on the Origin that he undertook to publish them as a pamphlet, and it was he who suggested the sub-title: “Natural Selection not inconsistent with Natural Theology.” He also added to the second edition of the Origin a motto from Bishop Butler to the same effect. (Paley already appeared in the text in support of the view that “no organ will be formed…for the purpose of causing pain or for doing an injury to its possessor.”) Later, to be sure, particularly under the prompting of Huxley, (who, however, himself professed to believe in a “wider teleology” of natural law), Darwin began to suspect that Gray was assigning too much power to God and too little to Natural Selection, and that the universe was not entirely beneficent in its design. But his suspicions of Natural Theology did not extend to teleology, and when, as late as 1874, Gray congratulated him for having restored teleology to science, Darwin thanked him warmly: “What you say about Teleology pleases me especially, and I do not think any one else has ever noticed the point.”

If the recent selection of Darwin’s writings by Stanley Edgar Hyman is superior to most others, it is because it was compiled by a literary critic whose method is peculiarly well suited to eliciting motifs, such as the teleological one, that were paramount in the original conception of Darwinism and that have since been generally ignored. In an essay published last year in The Tangled Bank. Hyman undertook to consider Darwin’s writings as “imaginative organizations, as though they were poems.” The results of that undertaking obviously inform the present selection, and it is unfortunate that he could not reprint the whole of that essay as a commentary on this selection. For the essay is a notable contribution to the “Art of Science” and a salutary corrective to the ideas of most scientific historians.

Mr.Hyman, coming at Darwin from a new direction, by-passes the usual myths and conventions. He explains why it was the Origin. rather than its many predecessors and anticipators, that produced the revolutionary effect it did: “This enormous effect was achieved by the power of scientific argument, that is by the book’s rhetorical organization.” He quotes the opening words of the final chapter of the Origin: “As this whole volume is one long argument…” He observes that “the evidence to establish the idea of evolution by natural selection inductively was not really available in 1859,” that many of Darwin’s processes turn out on closer examination to be plausible hypotheses, and his causes tautologies,” and that the vehemence of the controversy hardly suggested “the characteristic effect of a scientific demonstration.” “Perhaps the most surprising thing in the Origin,” he comments, “is that it repeatedly calls not for an act of understanding but for an act of imagination.”

And he gets bolder as he goes on. If the characteristic effect of the Origin was not that of “scientific demonstration,” it was that of a “dramatic poem,” a “tragic drama.” Even more, it was a poem of a special kind, “something like a sacred writing, a scripture.” Darwin’s metaphors and imagery were consistently teleological, “In reality, of course, Darwin’s teleology is as sacred and supernatural as Paley’s but with all seeing Mother Nature substituted for God the Father.” “The appeal of the Origin is not only for imagination, wonder, marvel, rather than ratiocination, but ultimately for belief and faith.” And its rapid acceptance confirmed its religious character, for it followed the typical sudden and dramatic pattern of religious conversion, rather than the slow, laborious process generally associated with scientific demonstration and persuasion.

All this may sound wild and bizarre, the playful fancies of a littérateur. Yet to one who has studied Darwin closely, they are obvious, literal truths. If Mr. Hyman has any failing, it is in not being daring enough. It is unfortunate that, having come so far, he has not gone further, and that he has contented himself with minor literary insights and refrained from significant judgments of a scientific or philosophical order. Thus he does not take seriously enough his own remarks about the rhetorical nature and effect of the Origin. When he says that the book called for an act of “imagination” rather than “understanding,” he is too easily satisfied with such familiar varieties of imagination as the hypostatization of nature or the personification of plants and insects. Yet he has presented evidence enough, and there is far more of the same kind, to suggest that the act of imagination was a response not to this or that passage of poetic writing but to the entire rhetorical structure of the Origin. He quotes Asa Gray: “The interest for the general reader heightens as the author advances on his perilous way and grapples manfully with the most formidable difficulties.” But he is too intent upon his own quest for metaphors and images to examine the perilous way taken by Darwin in his “mythic quest.” (It is significant that Mr. Hyman’s selection includes only the briefest excerpt from the Origin, offering instead an early, barer sketch of the theory.)

One of Darwin’s contemporaries analyzed the nature of the “one long argument” of the Origin: first the reader was asked to concede the “mere possibility of imagining” a given proposition; having made this concession he then found that the possibility of imagining the proposition was itself put forward as a reason for supposing the probable truth of the proposition; the probable truth of this proposition, in turn, was taken as the basis for the possibility of imagining another proposition, and so on, ad seriatim. By this Alice-in-Wonderland logic, the compounding of possibilities led not to a diminution of probability, as in conventional logic, but to an augmentation of probability. And the reader became more and more committed to the theory as he allowed his imagination to be more and more committed to conjecture. All the difficulties Darwin so manfully grappled with were in fact overcome only in imagination. When he was confronted with what he himself took to be the “gravest and most obvious” objection to his theory, his only reply was to propose an “imaginary illustration” and a “hypothesis.” And when even such imaginary solutions failed him, he made the failure itself an accomplice:

Does not the difficulty rest much on our silently assuming that we know more than we do? I have literally found nothing so difficult as to try and always remember our ignorance. I am never weary, when walking in any new adjoining district or country, of reflecting how absolutely ignorant we are why certain old plants are not there present, and other new ones are, and others in different proportions…Certainly a priori [i.e., according to his theory] we might have anticipated that all the plants anciently introduced into Australia would have undergone some modification; but the fact that they have not been modified does not seem to me a difficulty of weight enough to shake a belief grounded on other arguments.

This is surely not the “scrupulous regard for negative evidence” that Mr. Hyman refers to and that idolators of science celebrate. It is, however, one path taken by the scientific imagination—and not, perhaps, the most devious one.

Another instance: Mr. Hyman points out that Darwin was conducting experiments with garden peas at exactly the same time as Gregor Mendel, but that he failed to deduce from these experiments the laws of genetics because he did not attend to the same facts as did Mendel. Variations of color provided Mendel with his chief clue; but Darwin was too engrossed with those variations suggesting conflict, struggle, and competition to pay attention to so passive a characteristic as color. Elsewhere Mr. Hyman contrasts the dramatic accounts of the “war of nature” and “struggle for existence” in the Origin with the frigidly neutral formulations of modern genetic-oriented Darwinians. In both cases, however, he fails to draw any significant conclusions from these contrasts. If he seriously followed through on his own principle that metaphor, imagery, and rhetoric are not a “superficial matter of ‘style’,” but rather “the work of the moral imagination, imposing order and form on disorderly and anarchic experience,” he would have had to conclude that Darwin and Mendel ordered nature in quite different ways, and that Natural Selection and genetics are not as easily reconciled as the neo-Darwinians suppose.

William James once said: “Different minds have a scent for different kinds of truth.” Philosophers of science expounding upon the “art of science,” historians of science reminding us of the confusion and ambiguities involved in actual scientific discoveries, literary critics examining scientific texts as works of “moral imagination”—all these may help us appreciate the cogency of James’s remark.

This Issue

December 12, 1963