Arthur Koestler
Arthur Koestler; drawing by David Levine

When this extraordinary book—part treatise and part biology and psychology copy book, part independent scientific speculation and part romantic Naturphilosophie—was published in England earlier this year, it caused something of a sensation. No one quite knew how to take it. Some reviewers (and it was reviewed profusely and often at great length) professed to think that, however improbably, The Act of Creation is itself a true act of scientific creation, conceivably the greatest and certainly the most ambitious work in the life sciences since Darwin’s epochal Origin of Species. On the other side, perhaps misled by Koestler’s reputation as a novelist (it is not without significance that Koestler has been mainly a political novelist for whom the work of imaginative literature may be at the same time a moral and political act), by his inappropriately lively and witty style, by displays of subjective reactions presumably irrelevant to questions of objective scientific understanding and truth, and by his bland refusal to be daunted by any intellectual problem, no matter how technical or complex, many were unable to take him seriously in the role of scientist In their view, Koestler does not fully realize what is involved in scientific inquiry, and his book, for all its learning, must be judged as the misplaced product of an incurably poetic imagination. Yet on both sides of the controversy, the critics seemed uncertain of their own reactions, and before they were through they had usually managed to hedge their bets by radical, if piecemeal, concessions to the opposing point of view.

I scarcely blame them. For one thing, The Act of Creation contains not one but two (no doubt related) books of very different ranges, perspectives, and styles. The first is beautifully readable and, for the most part, intelligible to any informed twentieth-century reader; the other is highly technical, occasionally unintelligible except to specialists and sometimes (I suspect) not even to them, but at the same time immensely ranging, not to say visionary, in its purview. For another thing, both books, for all their differences, have similar defects. Here I am not speaking of factual errors, most of which could no doubt be removed without affecting the main drift of Koestler’s arguments. The underlying difficulty concerns the intention of the book as a whole. What is Koestler really up to? Would it be an egregious mistake to take the book at its face value? But, then, what precisely is its face value? Practicing scientists will be, have been, disposed to dismiss Koestler as a possibly gifted but presumptuous and uncritical “writer” who has wandered unaccountably into a field for which, by training and aptitude, he is intrinsically unfitted. Literary men, especially those who have erected a wall between “imaginative literature” and other forms of writing, whether scientific, historical, criticial, or philosophical will “submit” that Koestler is at once making clear at the theoretical level what was always apparent from his practice, namely, that he is not an artist but an artificer, a maker and joiner who has been filling in time while his story-teller’s imagination has lain fallow. Either way, it seems impossible to accept his book as a “normal” contribution to biological and psychological science. And the point is merely sharpened by Professor Sir Cyril Burt’s statement, in an admiring if cautious Foreward, that Koestler had a scientific education, has “visited” in many places of learning where serious psychological research is conducted, has a formidable knowledge of the “literature” of the subject, and enjoys the intimate friendship of some of the most original investigators in contemporary science, “from nuclear physics to experimental neurology.”

To complete the picture, Koestler himself has created further uncertainty by a rather testy and cryptic reply to one reviewer who had made the important point, which others have also noted, that Koestler not only does not adequately distinguish, but apparently fails to appreciate, the salient difference between the scientist’s original moment of insight or discovery and the objective validation of his hypothesis, and between his (the scientist’s) “moment of truth” and the achievement of public scientific knowledge. Even if, as Koestler contends, there is in all human thought a continuous gradient from “objective” to “subjective” reactions, this does not prove what Koestler evidently thinks it proves. For scientific objectivity concerns, not a characteristic response on the part of scientific discovers, whether “subjective” or “objective,” but conformity to a system of impersonal practices for testing hypotheses.

If Koestler were willing to concede this, his two main arguments could still be true: that the scientist’s original insight bears a close analogy to the artist’s creative act; and that certain psychological (and ultimately physiological) conditions are common to them both. Why is he unwilling to make this saving concession which would seem to leave his theory of creation virtually intact? This question, I believe, is but one side of a more general question about the nature of his achievement. What no one seems to have asked is this: why does Koestler always stretch the range of his ideas to, and beyond, the breaking point? And what is the source of his passion for melding attitudes or temperaments? These questions are not answered by the charge that he lacks scientific discipline or that he suffers from the artist’s supposed incapacity for handling general ideas. When he chooses, Koestler can be as restrained and as analytical as the most proper academic scientists. Nor is he lacking in self-criticism. He makes clear at the outset that he has “no illusions about the prospects of the theory…it will suffer the inevitable fate of being proven wrong in many, or most, details…What I am hoping for is that it will be found to contain a shadowy pattern of truth, and that it may stimulate those who search for unity in the diverse manifestations of human thought and emotion.” Moreover, as it seems to me, Koestler is quite right in contending that there is no such thing as the artistic or the scientific temperament. The ideas both of art and of science concern forms, not of temperament, but of achievement. And to either form of achievement there is no one temperamental throughway. In my judgment, the real source of Koestler’s “defects” as a scientist is, rather, an overriding concern, not fully appreciated either by himself or by his critics, which can only be called metaphysical. And it is as metaphysics that The Act of Creation as a whole must be judged. However, before I can even begin to suggest why, and how, this is so, we must first take a closer look at his theory itself.


The Act of Creation is, I believe, a more truly creative work than any of Koestler’s novels and a more adequate revelation both of his powers as a writer and of his enduring interests as a man. In his own view, however, Koestler the quondam novelist and Koestler the scientific psychologist are merely chips off the same block. According to him, the creative faculty in whatever form is owing to a circumstance which he calls “bisociation.” And we recognize this intuitively whenever we laugh at a joke, are dazzled by a fine metaphor, are astonished and excited by a unification of styles, or “see,” for the first time, the possibility of a significant theoretical breakthrough in a scientific inquiry. In short, one touch of genius—or bisociation—makes the whole world kin. Or so Koestler believes. And why should this not be so? That one man is tone deaf and another good at numbers, or that one has an eye for facts and another is impervious to them proves nothing to the contrary. Koestler is talking about—if he will forgive the word—the fundamental mechanism of genius, not the specific “skills,” native or acquired, that may be necessary to particular acts of creation. On this score, he merely underlines a point we all must have known before, namely, that it is not the skill or facility that makes the creative artist or scientist, but what he does with it. In short, Koestler is interested in that transfiguring movement of the mind which turns cleverness into the creative act and talent into the stroke of genius. He believes that he can define the essential conditions of that movement. He also thinks, perhaps incontinently, that its more primitive analogues are in evidence everywhere in the organic world.

The theory of bisociation is explained and profusely illustrated at the human level in the lengthy but continuously fascinating first book of The Act of Creation. Here Koestler is at his ease. He entertains while he enlightens and persuades, and even though one may not really be shown in the end just why a bisociative act occurs in one situation but not in another, one is nonetheless enabled virtually to “see” Gutenberg in the act of inventing the printing press, Kepler (always a special hero of Koestler’s) in the process of discovering his three great laws of planetary motion, and Darwin moving up to and then lighting upon, as Darwin himself put it, “the long-sought-for law of nature that solved the problem of the origin of the species.” Technically, “bisociation” may best be understood by distinguishing it from unitary, habitual associations on a single “plane” of experience. Connections of the latter sort, involving merely the articulation of an established routine, are manifest in such standard examples as, say, the association of clouds with rain, of fox with bagels, of the word “Fido” with Fido. Bisociation, on the contrary, occurs only when a new connection is made between two (or more) independent contexts of association which Koestler variously calls “frames of reference,” “universes of discourse,” or “types of logic.”1


Koestler shrewdly introduces his study of bisociation through an account of humor which, although (in my judgment) too intellectualistic, is one of the most convincing parts of his book. Humorous bisociation involves the perception of a situation in two habitually incompatible associative contexts. This “causes” an abrupt transfer of the train of thought from one “frame of reference” to another governed, as he puts it, by “a different logic or ‘rule of the game’.” In humor, however, certain emotions, “owing to their greater inertia,” cannot follow such rapid intellectual moves and so, “discarded by reason, they are worked off along channels of least resistance in laughter.” The emotions involved are, in particular, those of the self-assertive, aggressive-defensive type, which are based on the sympathetic-adrenal system and tend to beget bodily activity. “Laughter is a luxury reflect which could arise only in a creature whose reason has gained a degree of autonomy from the urges of emotion, and enables him to perceive his own emotions as redundant—to realize that he has been fooled.” There is no space to cite a fair sample of the many jokes and witticisms (both good and bad) which Koestler so perceptively and convincingly analyzes. Consider the following “memorable statement” which appeared (ye gods!) in Vogue:

Belsen and Buchenwald have put a stop to the too-thin woman age, to the cult of undernourishment.

This horror, as Koestler says, “makes one shudder, yet it is funny in a ghastly way, foreshadowing the ‘sick jokes’ of a later decade. The idea of starvation is bisociated with one tragic, and another, utterly trivial, context.” The result, in spite of ourselves, is an uncontrollable titter. Let me add only that here, above all, it is not just ideas that are bisociated, but whole ideo-motor-affective Gestalten of which the “idea” provides merely the nucleus.

Scientific discovery also involves the same bisocation of “matrices.” (This is the technical term that Koestler uses in referring more generally to the standardized routines, frames, abilities, skills, etc. that are governed by sets of rules or, as he calls them, “codes.”) The difference lies in the fact that in science we find “a blend of passions in which both the self-asserting and self-transcending tendencies participate—symbolized by the Mad Professor and the Benevolent Magician of folk lore.” Here, however, both tendencies are sublimated and each is offset by the other. This development, Koestler argues, is foreshadowed in the exploratory behavior of “clever animals” like the chimpanzee which, in Köhler’s classic experiment, discovered, after many unsuccessful efforts, that he could rake a banana into his cage by fitting two short hollow sticks together. The chimp’s original motivation was to get the banana, but his new discovery “pleased him so immensely” that he kept repeating the trick and forgot to eat it. “Eureka!” he would no doubt have said had he possessed that verbal skill, “I have it!” For Koestler in fact Archimedes’ transfixed cry expresses the archetypal jubilation of all intellectual discoverers.

Koestler’s account of artistic creativity is no less intriguing than his discussions of creativity in humor and in science. He begins with an explanation of what he calls “the logic of the moist eye” in which he contrasts with great skill the underlying responses involved in comedy and tragedy. But this is merely preliminary to a complex and subtle treatment of forms of verbal and visual creation (here one wonders whether it was merely the pressure of time that is responsible for the salient omission of an extended discussion of music, which is at once the formal and the expressive art, par excellence. It must suffice to say that, in Koestler’s view, whereas laughter is sparked by the collision of matrices, and scientific discovery by their integration, aesthetic activity arises from a “juxtaposition” of matrices in which there occurs an emotional catharsis, that is, “the rise, expansion, and ebbing away of the self-transcending emotions.” Here, also, however, there is a characteristic insistence upon “intellectual illumination—seeing something familiar in a new light….” The emotional catharsis simply follows the bisociative intellectual illumination. To my mind, the arch-foe of mechanistic theories in psychology and biology has himself fallen into a rather mechanistic view of the role of emotion in art. I would suggest both that emotion is present in the bisociative confrontation and that pre-established forms of emotional reactions by themselves be at least central aspects of the artist’s creative act.

There is no space for adequate resumé of Book Two (entitled Habit and Originality) which sets out the general bio-psychological theory of which the theory of human creativity developed in Book One is, in principle, merely a special, if also to most of us the most interesting and important case. As Koestler puts it, whereas Book One represents an “upward approach” that moves from part to whole and examines the creative act in its most complex human forms, Book Two represents a “downward approach” that moves from the complex to the elementary and from the whole to its parts. Thus, if Koestler is right, The Act of Creation represents the meshing of a novel theory of man’s highest mental processes into a universal thesis which applies to, and in principle helps to “explain,” certain patterns of change to be found in all forms of life “from embryonic development to symbolic thinking.” All such processes of development are “governed” by certain “rules of the game” which give them appearance of purpose. Whether phylogenetically or ontogenetically acquired, these rules, or “codes” as Koestler calls them, operate on all levels of life. And on all levels, there occur creative “fusions” or “marriages” (bisociations) of such coded routines, skills, or matrices.

As one perhaps might have anticipated, the basic model of the creative act is sexual reproduction. Here, I am bound to say, it is not easy to know how far fact leads figure, or figure fact. Koestler not only uses his models; he so pulls and hauls them, partly by metaphorical and analogical extension, partly by contextual redefinition, that one does not always know in the end what illuminates and what is being illuminated. In any case, Koestler is not suggesting, I take it, that sex is actually the source of all creative advances in the organic world. And while there are echoes of, as well as many references to, Freud (and to a lesser extent Jung) in the book, it would be a radical mistake to suppose either that Koestler is a Freud gone mad or that he is offering a kind of quasi- or pseudo-analytic account of creation of the sort one might expect from a freely bisociating psychological novelist. On the contrary, the whole bent of Koestler’s approach is in direct conflict with the mechanistic side of a theory of development which would explain all mature forms of human behavior in terms of certain infantile or pre-natal fixations.

The old fashioned notion of organic hierarchies is fundamental to Koestler’s “general theory.” Roughly, his contention is that any complex organism is an hierarchial system of units, subunits, sub-sub-units, and so on. However, an organic hierarchy is not simply an “order of rank” of the sort exhibited in the “pecking hierarchy” of the farm yard. As in the case of a military organization with its squads, platoons, companies, etc., each unit in an organic hierarchy possesses a certain wholeness and autonomy, although each lower order unit is also subordinate to that just above it in the hierarchy to which it belongs. In individual organisms, Koestler believes, such hierarchical organization is exhibited not only in their main organs, but also in their cells and cell parts. Moreover, on every level, there are at work, according to him, “homologous” principles, so that any activity or process at one level has its homologue at every other level. In fact, all higher-order mental processes have their lower-order bodily equivalents and vice versa. Koestler thus claims that the “dichotomy” of “self-assertion and participation” holds on all levels of bodily action and reaction, and, conversely, that to every form of “bodily” self-maintenance and self-repair there is a corresponding higher “mental” homologue. Somewhat perplexingly, however, homologous principles are evidently thought to apply not only within hierarchies internal to individual organisms but also across species to the great hierarchy of natural organic kinds. Thus, for example, in the “general alarm reactions” of injured animals there occurs something very like the “stress” undergone by artists and scientists in the act of creation.

Koestler goes very far with such comparisons. Indeed, such verbal phenomena as rhythm and rhyme, assonance and pun are even said to be “vestigial echoes” of the “primitive pulsations of living matter.” These remarks, apart from any question about their sense, plainly show Koestler’s inability to limit the application of his notion of organic hierarchies. The result is that we find ourselves in increasing doubt as to what a genuine hierarchical organization or connection really is. I can understand, for instance, how in principle a cell-part can be viewed as standing in a homologous hierarchical relation to the cell to which it belongs, but not, so help me, if I am asked to think of such a relation as holding between verbal rhythms and “primitive pulsations of living matter”—whatever they are.

Here, I believe, is a type of difficulty that is both recurrent in any symptomatic of Koestler’s thought: Having formulated what looks to be a significant concept or thesis he then stretches it so far and applies it so loosely that a question is eventually raised both about its own rules of meaning and about the sort of game Koestler is playing with it. And, in a word, he has a way of driving an idea so far into the ground that it seems to lose its point; or else, to follow his own analogy, he shoots it so far into the air that it simply passes beyond the range of our intellectual vision. This is a great pity, not so much because it raises doubts about Koestler’s discipline as a scientific thinker (similar doubts can be raised about many if not most seminal minds in science), but also, and more importantly, it makes it much harder than necessary to do justice to the main drift of his thinking. One has constantly to remind oneself that the unfruitfulness of a particular analogy does not entail the failure of analogical thinking as such, and that the misapplication of a theory is indeed a misapplication only, and not necessarily an evidence against the theory. On the other side, it is essential to realize that questions of success and failure in such matters are entirely relative to the degree of precision with which concepts are defined and theories formulated. What would be an absurd misapplication of a theory in mathematical physics may be merely suggestive in biology; what seems anomolous in biochemistry may be splendidly analogical in political “science” or theology. Concepts applied with a certain precision to the sphere of human action may have no meaning or else have to be radically redefined, when applied for scientific purposes within the domains of physics and chemistry.

The key phrase in the preceding sentence is of course “for scientific purposes.” Where our task is not, as in physical science, to describe and to predict phenomena, but, as in metaphysics, to give a unifying sense of the lay of the land, or, as in theology, to prepare a unifying basis for worship or a direction for the devotional intent, such lack of precision, or such misapplication, may be harmless or even exciting. Why not say that one touch of nature makes the whole world kin if the point in saying it is to bring Achilles out of his tent? Why not call the principle of creation a Father, a bisociative secretion, or a swerve in the matrix of life, if that moves, or comforts, or helps the human spirit itself to “bisociate,” to analogize, to think and feel in ways that are truly life-enhancing.

As William James would say, I am in some sense on Koestler’s side against those who suppose that the concepts of the reflex, the routine, the skill, the method, and the institution provide adequate bases, or models, for the understanding of human life or of nature. Man betrays himself when he “reduces” himself to a system of routines, to a creature who simply “adapts” himself to “situations” by skills already available to him. Koestler is on the side of freedom and life, or at any rate the good life, and freedom are deeply intertwined. Still, I am not sure that Koestler will be pleased by my favor. For even if his theories on certain reaches turn out to have explanatory value, they are not, as they stand, scientific theories. Or so it seems to me. What, then, are they? Fundamentally, the role of the theories of bisociation and of organic heirarchies, not to say, of their union, is metaphysical. Now of course old-line positivists will regard this as equivalent to a charge that Koestler’s theories are essentially meaningless. But this charge holds only for scientific reference and verification. I have said, partly in criticism of Koestler himself, that he is too much beholden to the model of the frame of reference. He really does seem to think that the jester is simply one who laughs off his bisociative “discovery” and that the artist is one who drains his “discovery” (shades of Aristotle, whom Koestler abominates) into an emotional catharsis. This is a mistake. The point, however, is that there are forms of thought-feeling-action, and forms of expression that subtend them, which do not have a “discovery,” at least in the scientific sense, as the consummation to be wished. Koestler’s theses are truly creative, but they are creative in the metaphysical rather than in the scientific dimension: that is to say, they are fresh, exciting principles of orientation that suggest approaches to a Weltanschauung which I find exhilarating to contemplate. Their “verification” is not so much perceptual as emotional-volitional. Here, so to say, is not a thesis, but a vision, not a theory, but a unifying picture which, in an age of specialization, of bifucations,—and walls—is a refreshment of the soul. One touch of genius may indeed make the whole world kin. Let us see.

This Issue

December 17, 1964