“This book contains the first attempt to trace development all the way from genes through the mind to culture.” The authors’ illusion that this is so is owing at least in part to their neglect or ignorance of the thought of many others who have attempted to arrive at a conception of human heredity that brings in minds and culture as well as genes.

Although Herbert Spencer groped his way toward the notion, I believe that Thomas Hunt Morgan,1 the most influential geneticist since Mendel, was the first to realize that in addition to an ordinary genetic evolution in the Darwinian mode, human beings also enjoy an “exogenetic” evolution in the Lamarckian mode: the latter is mediated not through chromosomes but through indoctrination and learning, example and imitation (“aping”)—a form of evolutionary change written about in addition by J.S. Huxley, C.H. Waddington, and myself.

T.H. Morgan’s formulation is so unfamiliar that it is worth quoting in full, especially since it is a model of clear thinking and writing.

While biologists have come to reject the theory of the inheritance of acquired characters by means of the germ-cells, nevertheless they recognize the fact that the human race has succeeded in another way in transmitting certain traits acquired in one generation to the next. There are, then, in man two processes of inheritance: one through the physical continuity of the germ-cells; and the other through the transmission of the experiences of one generation to the next by means of example and by spoken and written language. It is his ability to communicate with his fellows and train his offspring that has probably been the chief agency in the rapid social evolution of man. In the animal kingdom we find many cases in which the young are protected and cared for by their parents. Such beginnings furnish the background out of which has evolved the more complex relation of parents and offspring in the human race, where a prolonged period of childhood furnishes exceptional opportunities for the transmission of tradition and experience.2

“Cultural evolution” is not a very good description because it could be taken to connote evolution of culture instead of an evolution mediated through culture, so perhaps Julian Huxley’s “psychosocial evolution” is better, though I myself prefer the wholly neutral “exogenetic” or (A. J. Lotka’s term) “exosomatic” evolution. Because of the crucial importance of the cultural nexus between one generation and the next, S. A. Barnett has thought to replace Homo sapiens or Homo faber by Homo docens (teaching man) as the specific designation of man. The trouble with Homo sapiens is the temptation it offers to the cynic to contrast man’s titular wisdom with his actual lack of it.

The most important differences between the ordinary and exogenetic evolution are that the latter is reversible and that the knowledge, skills, and manners acquired in one generation can be transmitted to the next (for this reason it is said to be an evolution “in the Lamarckian mode”).

The authors make some gestures to the politesse of scholarship, which usually amounts to not much more than conceding that others have tried to grope their way toward the concepts we profess to be the first to expound. They mention as having an affinity with their own concepts Richard Dawkins’s “meme,”3 a sort of hereditary package which he sees as “the unit of cultural transmission” or as a “unit of imitation.” But they fail to mention the mneme of Ewald Hering,4 which Dawkins developed. For Hering the formation and transmission of mnemes in every situation in which hereditary transmission may be said to occur is an important general characteristic of living organisms.

Genes, Mind, and Culture is written in the prose style and with the literary manners which have earned sociology the amused contempt of scholars throughout the world and which are the despair of the many serious and highly literate sociologists who object to their subject’s being thought of as so much codswallop.5 Here is how the authors start one of several early summaries of their argument:

Within the framework of the [authors’] newly proposed comparative social theory, mankind is classified as a eucultural species, in which mental activity is based to a large extent on reification and symbolization and the young are socialized through purposive programs of teaching.

During socialization an array of behaviors and artifacts, which we have termed culturgens, are processed through a sequence of epigenetic rules. These rules are the genetically determined peripheral sensory filters, interneuron coding processes, and more centrally located cognitive procedures of perception, learning, and decision making. They affect the probability of transmitting one culturgen as opposed to another. The probability distributions themselves constitute bias curves, which will be used in later chapters to link human cognition to patterns of social behavior.

Indeed, the book is written so badly and with so little evidence of any attempt to be intelligible or to persuade that I feel some special explanation is called for. I am very sorry to say that in my opinion the explanation is that the authors’ purpose has been above all else to impress, to show what a profound subject sociobiology is and what deep fellows the authors are. Such an intention is understandable if not excusable in the practitioners of a would-be science such as sociobiology still is. The same charge could at one time have been brought against biologists, who sought to impress their fellows by the adoption of a comically pretentious and grandiloquent terminology. Thus “pseudopodia”—presumably so called to distinguish them from real feet—are the bulges and protrusions of the body by means of which Amoeba probes its way along a substratum; this is a silly nomenclature to be sure but my favorite idiocy is the description as a pseudonavicella—an as-it-were-little-boat—of a reproductive element in the life history of a parasite of the earthworm.


The stylistic enormities of bad sociologists are not of this kind but take the form of a constitutional obscurity through the use of unexplained abstractions in a convoluted prose style which disgusts more often than it impresses. The present authors are not in these respects typical sociologists for they have a style of obfuscation wholly their own. We are plunged into deep water right away. We learn that “evolutionary biologists have hesitated to extend the concept of biological causation and natural selection to the study of culture.” Such biologists are said to have been inhibited by the prevalence of what may be called the “Promethean-gene hypothesis” and by being in thrall to the notion of the “psychic unity of mankind.” But what the hell is the “Promethean-gene hypothesis” and what is meant by the “psychic unity of mankind?” Perhaps we need not bother with the meanings because the authors have little use for either notion.

In a passage of characteristic felicity of thought and expression, the late Professor A. S. Eddington wrote, “We need scarcely add that the contemplation in natural science of a wider domain than the actual leads to a far better understanding of the actual.”6 But perhaps such thought was in the authors’ minds when they wrote:

The trajectory of human history can be plotted when the arrays of genetic-cultural properties are deduced across many imaginable species and beyond the arbitrary limits of human variation and are used to define the mathematical elements needed to describe the evolutionary process.

After reading this passage I turned curiously to the bibliography to see if it betrayed any awareness of Professor Karl Popper, who, in three famous papers published in 1944 in the journal Economica,7 did an accomplished job of discrediting “historicism,” the aberration of thought to which the quoted passage belongs. The bibliography revealed no such awareness. But historicism is one of the two principal cultural diseases by which sociobiology is afflicted. (The second is “geneticism,” to which I turn below.)

Historicism is the belief that there may be formulated predictive laws of social transformation and of the historical process generally—laws, as it were, of human destiny. That the authors seem unaware of the decisive arguments against it is discouraging; but they proceed to chide some sociobiologists—presumably the amateur sociobiologists to whom I refer below—for holding views so manifestly foolish that we hardly need the assurance that they are “radically incorrect.” “Behavior,” the present authors gravely inform us, “is not explicit in the genes, and mind cannot be treated as a mere replica of behavioral traits.”

Such passages gave me a valuable clue about how to interpret this book—I mean, how to account for its having been written at all. It appears to me that sociobiology is of two kinds, amateur and professional. That which has reached the scholarly level we find in some Sunday newspapers is amateur sociobiology—a quasi-scientific interpretation of human behavior and the social structures deriving from it, sustained mainly by the belief that evolutionary changes will come about if it is desirable that they should do so “for the benefit of the species,” and sustained also by the belief that the behavior of animals, particularly of primates, casts a bright light upon the behavior of human beings, especially upon aggression and defense of territory. These amateurish views have been castigated so often and so expertly that no further comment is called for here. Richard Dawkins, citing a number of offending documents, remarked, “The trouble with these books is that their authors got it totally and utterly wrong.” 8

These failings give us the clue to the understanding of the present book. Wilson and Lumsden are determined to be taken as professional sociobiologists who do not subscribe to the simple view that “genes prescribe behavior,” and as if to prove that they are, they have produced this pretentious and subtly grandiloquent book.


One of the symptoms of the authors’ ambition to show how very scientific they are is the repeated disfigurement of the text by all manner of impressive-looking mathematical formulations, often making use of functions—e.g., the “Fokker-Planck” function—which, being wholly unfamiliar to most biologists, will surely be unintelligible to them.

I myself was once rebuked by implication for writing in this style. I had written a paper on human growth full of impressive-looking equations representing the shape of a human being as a function of his age, using many inadequately explained functions of perhaps needless complexity. I did not write the paper to show what a clever fellow I was, but was not above hoping that such a construction might be put on it. Later, out of curiosity, I asked a distinguished anatomist friend, Professor S. Zuckerman, what he did when in the course of reading a paper he came across mathematical formulas he did not fully understand. “I hum them,” he said, and I never played the trick again.

Pros though they may be, Lumsden and Wilson do not escape the charge of being guilty of a second great aberration of thought that runs through the whole of sociobiology. The first, as I have already said, is historicism. The second comparable aberration, so grave as to amount to a cultural disease of modern biology, is that which I have called “geneticism.”9 Geneticism is the belief that human social structures, the rise and fall of nations, and all the nuances of human behavior and of linguistic usages (such as the ability or inability to pronounce th) are genetically determined. Geneticism is genetic determinism in an extreme form. One of the most offensive of its assumptions, of which psycholinguistics is flagrantly guilty, declares that any common characteristics human beings possess—including, perhaps, the ability to use certain basic sentence structures such as the subject-predicate form—must somehow be genetically encoded and part, therefore, of our inheritance.

This sounds reasonable, and it might be true. But if we try to demonstrate it we are stuck: the reason is that Mendelism is a theory about the differences between individuals; there is a genetic theory about the ability to taste the compound phenylthiocarbamide (very bitter, I am told) because some people can’t taste it. If we could all taste it, the idea of its being genetically encoded would be an unsupported surmise. The same reasoning applies to the bizarre circumstance that the urine of some people smells disgusting after eating asparagus while the urine of others does not. Here too the Mendelian inheritance of this character difference gives us the assurance that it is some specific entry in DNA.

I know one biologist—a distinguished botanist—who cultivates a geneticism so extreme that he has dismissed contemptuously the whole notion that there can be a heredity mediated through cultural transmission. He reviled T. H. Morgan for proposing that such a system of heredity might exist, saying that such a notion would open the door to Lamarckism and Lysenkoism. He implied that all the enormities of Russian persecution of geneticists could be debited to Morgan, an accusation as silly as blaming Gregor Mendel for the gas chambers.

Lumsden and Wilson are not of this kidney; indeed, their specific purpose in writing this book—as opposed to the more general purpose outlined above—was to bring mind and culture into the genetic scheme of things. They clearly feel that it is pusillanimous to declare that the connection between genetic and cultural heredity is that “genetic evolution” provides merely for the capacity to evolve by culture. “In this book,” they say, “we propose a very different view in which the genes prescribe a set of biological processes, which we call epigenetic rules, that direct the assembly of the mind.” Human beings are said to be the only species to enjoy “eucultural” status with faculties of learning, imitation, teaching, and reification.

This last is a good example of the kind of word that people of literary judgment tend to shun; others on the blacklist are “chiliasm” and “chiliastic.” Other nonwords such as “acultural” or “protocultural” designate inferior cultural status.

“Reification,” for Lumsden and Wilson, distinguishes human beings from all other species. Its distinctive operations are “(1) the production of concepts, and (2) the continuously shifting reclassification of the world.” For according to the authors, small-brained animals filter out most sensory signals at the periphery and respond only to a small repertoire of “sign stimuli” among those that remain. The human mind, by contrast, absorbs a multitude of chaotic sensory stimuli most of which are of no immediate relevance. From these it constructs an “internal reality.”

These arguments form the basis of a number of glib asseverations in support of which various authorities are cited. Among these are that human beings tend to adopt a binary classification to break up continuously varying configurations (for example, in-group/out-group, child/adult, sacred/profane), the demarcation lines of which are reinforced by ritual and taboo. Moreover, wholly new phenomena described by Huxley as “mentifacts” encompass processes that are emotionally potent but only weakly comprehended by the rational portions of the mind. “Thus gods, spirits, and totems can be interpreted as the outward representations of sanctifying, group-binding activities of the mind…. Metaphors are created to link more directly perceived physical phenomena with those less easily grasped.” We go on to learn without surprise that “abstraction and symbolization appear to have been the primary achievements in the evolution of euculture.”

The agents of transmission of culture from generation to generation—the mnemes or memes of alternative terminologies—are, as we have seen, referred to as culturgens—“(from L. cultur(a), culture),” we are told, a little parade of learning that would not have been made more impressive by pointing out that gen comes from a Greek root. When the notion is exemplified, any mystery surrounding it disappears; thus culturgens might comprise “an assortment of food items, an array of carpenter’s tools, a variety of alternative marriage customs to be adopted or discarded, or any comparable array of choices.” In my view,10 that which is transmitted from generation to generation in culturally mediated evolution is the objective world of “things of the mind” or their material embodiments—the world of theories, dissertations, arguments and instructions, and, in addition, material embodiments of thoughts or acts of mind such as a scheme or organization of a joint stock company or a nation.

These together form what Popper11 classified as “World Three.” The objectivity of this world, which Popper has particularly insisted upon, is given away by the language in which we speak of it—as when we “grasp” an argument or “size up” an idea. One of the authors’ principal preoccupations is how culturgens interact with the more familiar genes of Mendelism in the process of cultural transmission—a subject to which we shall recur. Authors who have no gift for making their meaning clear can sometimes express themselves more easily in diagrams and charts. Lumsden and Wilson cannot: their diagrams make what is already obscure still more difficult to follow.

The main problem the authors grapple with is, as I say, the interaction of culturgens and genes. This problem is of the same general kind as that which confronted embryologists in the dark days before DNA. At one time there was thought to be an antithesis between preformation and epigenesis. According to the doctrine of preformation, development was essentially an unfolding of that which was already there—a notion belonging to the days when the role of the female in reproduction was not understood, the female being regarded (as in many Eastern countries she still is) merely as soil for the germination and growth of the male seed. Imperfect microscopy combined with a perverse willingness to regard such a notion as true made it possible to see a homunculus neatly packaged in a spermatozoon.

The alternative to this view was that in development the organs and other differentiated parts of the embryo arose out of an ostensibly homogeneous or structureless germ. This process was referred to as “epigenesis,” but no convincing or even marginally plausible scheme of how epigenetic development could come about and yet provide for heredity was ever formulated. Needless to say, the antithesis is now of merely antiquarian interest: we simply think about these problems in a different way. Something is preformed without doubt and this is the chemically encoded genetic message carried in the chromosomes; and epigenesis is usually understood as that which brings it about that different genes or sets of genes are awakened and act in some cells and remain silent in others.

This led to a reorientation of attitude toward the developmental process. Older natural historians had been inclined to ask themselves by what process the fertilized egg of a frog could “turn into” a frog. It is not merely a verbal, but a genuinely conceptual distinction to realize that the fertilized egg of a frog does not turn into a frog: it is a frog—just a very young one, that’s all. The epigenetic influences are those that turn something which is genetically and potentially a frog into that which is not only genetically a frog but an actual frog in the most ordinary sense. Nevertheless, according to Lumsden and Wilson, “In spite of the relative sophistication of the field of developmental psychology, the epigenetic rules have never been systematically described, and the data concerning them have remained scattered and unconnected to evolutionary theory.”

This is very true, but although there has been no systematic description of epigenetic influences in the development of individual organisms, the developmental physiology of the Thirties was full of attempts to grapple with the problem which seemed to me then and seem to me now as inefficacious as those currently under consideration. I remember especially the proliferation of abstract concepts such as “morphogenetic fields,” “axial gradients,” “individuation fields,” and the supposed influences of such agents as “evocators” and “eidogens.”

I have not the least doubt that Wilson, an educated biologist, is thoroughly well aware of the embarrassing mess developmental biology got itself into in the Thirties and Forties. I think though that it is a conceptual weakness of the authors’ treatment that they do not, so far as I remember, call attention forth-rightly to a very important difference of principle between development and heredity in individual organisms and in societies. With individual organisms there is genuinely an end and a starting anew—that is, a gap between generations.

This is not to say that members of a parental generation die before the filial generation starts again, though sometimes there is a close approximation to it. “Like an operatic hero,” J.B.S. Haldane once remarked, “the Pacific salmon dies after mating.” Societies differ from individual organisms in a way that makes cultural heredity easier to understand, for with societies there is not normally an end and a starting anew: a society consists at any time of older people who teach or provide models and younger people imitating and learning—all mixed up in such a way that people can learn from each other as cells cannot—and can play with each other too, an activity that may I suppose conduce to learning.

I do think it possible to make unduly heavy weather over cultural inheritance. If we take an objective view of the “things of the mind,” the inheritance by one generation from its predecessor of the know-how to make shoes is not more mysterious in principle than the inheritance of the shoes themselves: the instructions about how to make shoes are (as the saying is) “passed on”—and so are the shoes.

A further cause of complaint arises out of what seems to me to be the authors’ gratuitous geneticizing. They place great emphasis on the existence of culturgens in sets of alternative forms to be adopted or discarded. Perhaps they had in mind the model of Mendelian genes that may also exist in alternative forms or “alleles” of which only one is transmitted in a gamete. They go on to say that if the development of the members of a society “is genetically constrained in such a way that the same culturgen is selected each time, the transmission is said to be pure genetic transmission.” What this formulation amounts to is that the authors treat conventional Mendelian genes as the agencies which provide the epigenetic climate that favors the bringing out of one culturgen rather than another.

I think that this is a perverse way of looking at things and that the authors have got it the wrong way round. I believe it would be much more in keeping with the way that biologists look at things to regard the culturgens as providing the epigenetic climate that affects the mode of expression of Mendelian genes. A concrete example will make my point clear. In human populations there is a maleficent gene which, if inherited from both parents, deprives its possessor of the power to metabolize the ubiquitous protein constituent phenylalanine, leading to gross defects in the development of the nervous system and, if uncorrected, to low-grade mental deficiency. (About 1 percent of mental defectives are victims of this inborn error of metabolism—phenylketonuria.)

If, however, the society inherits the information necessary for the recognition of the disease at or shortly after birth, and for preparing diets especially low in phenylalanine—if, in short, the potential victim can be brought up in a nearly phenylalanine-free microcosm, then the offending gene has no opportunity to make itself manifest. Here then the culturgen determines the expression of the gene. Of course, the relationship between Mendelian gene and culturgen could be read the other way around: no one would think to prescribe a diet low in phenylalanine for anyone except a potential victim of phenylketonuria, so the Mendelian gene could be said to affect the expression of culturgens. I believe, though, that the way I put it comes more naturally to biologists.

The authors’ emphasis upon the existence of arrays of alternative culturgens seems to me to distract attention from the fact that not all transmission and acquisition of culture—and not even the more important part of it—turn on a choice between alternatives. In the development of, for example, a frog it may be that only a smaller part of the DNA has to do with the Mendelian process that directs the development of one kind of frog rather than another. Surely the greater part has to do with the inheritance of frogginess as such—with the characteristics common to all frogs rather than with those that differentiate one frog from another—the subject of Mendelism. May it not be much the same in cultural heredity—that the greater part of it has to do with the inheritance of such characteristics as the ability to teach and learn, or imitate, and to communicate, irrespective of what is taught or learned or communicated? Contrary to the authors, it is for an understanding of these general capacities that we look first to biology.

The glossary accompanying this book is in keeping with the main text. It is obscure, pretentious, full of showing off and in places downright perverse. A professional biologist for forty-six years, I have never before heard the words “epigene” and “epigenotype”: their invention and use reminds me of those bad old days of embryology to which I referred above, when embryologists such as C.H. Waddington were, as it seemed to me, laying down a smoke-screen behind which they could conceal their conceptual nakedness. I cite without comment one entry in the glossary—

Culturgen (pronounced “kul’ tur jen”) The basic unit of culture. A relatively homogeneous set of artifacts, behaviors, or mentifacts (mental constructs having little or no direct correspondence to reality) that either shares without exception one or more attribute states selected for their functional importance or at least shares a consistently recurrent range of such attribute states within a given polythetic set. Culturgens can be mapped into nodelink structures in long term memory and in many instances can be treated as identical with them.

In reading this book I grieved at how very far it fell short of Wilson’s ambitious claim in his principal work on the subject—that sociobiology had become a branch of biology of the same stature as, for example, molecular biology. Indeed it has become nothing of the kind. The present work reminds me of a bad political manifesto—one that loses more votes than it wins. Certainly it has lost mine. It makes me sorry, too, for my sociologist friends.

The study of animal behavior in the style that has come to be known as ethology is genuinely helpful to psychologists of human behavior, if only to show that human behaviors or proclivities such as aggression (real of simulated), sexual rivalry, showing off, play, and the use of tools, so far from being characteristically human, have deep evolutionary roots and in that light may be easier to understand. I do not believe any such claim can be made for biology vis-à-vis sociology, and have yet to be convinced that biology can throw any new light upon the structure and behavior of human societies or upon the problems to which they give rise.

I find myself very much in sympathy with the criticisms of sociobiology explicitly expressed by Kenneth Bock. As one reviewer summarized Bock’s conclusion: “The universals of biology and genetic theory cannot account for recent history or the differences between cultures.”12

This Issue

July 16, 1981