Codex HTML

Does Ethology Throw Any Light on Human Behavior?
Sir Peter Medawar

Although biologists take to it very kindly, the idea that the behavior of animals can throw any light at all upon the behavior of men is so far from self-evident that it would have been regarded with the utmost derision by the more tough-minded philosophers and philosophic thinkers of the seventeenth and eighteenth centuries --- I mean, by men of the stature of Thomas Hobbes and Samuel Johnson.

Dr Johnson, in particular, would have been deeply outraged: `Sir,' he would have said --- he certainly would have said that, but what else he would have said can only be conjectural, though I think it might have run: `is not the possession and exercise of moral judgement precisely the distinction between mankind and the brute creation? Show me an earthworm or marmoset that can tell the difference between right and wrong.' Thomas Hobbes would have pointed out --- and in chapter 12 of Leviathan did, in effect, point out --- that it is only by virtue of characteristics that differentiate civilised mankind from lesser beings that the life of man is anything but `solitary, poor, nasty, brutish and short'.

Scientists of the twentieth century cannot be expected to quake in their shoes at the thought of the opinion that might have been held of them by eighteenth-century philosophers, however skillful and tough-minded they may have been, for they were simply not in possession of all the information that would have made it possible for them to form a definitive opinion. In particular, they were unaware of the evolutionary descent of man; but even if they had been appraised of it their reaction would probably have been `What of it?' and Dr Johnson, wielding, as ever, the butt-end of his pistol, would have demanded to know what great and illuminating new truth about mankind followed from our realisation of his having evolved.

No one would answer this question today with anything like the blandly asseverative self-confidence that would have been characteristic of social Darwinists such as Francis Galton and Ernst Haeckel. To them it would have seemed obvious that mankind was stratified into superior and inferior human beings, and that all were engaged in a struggle for ascendancy -- it would have seemed obvious, indeed, that `man is a fighting animal', It is, therefore, particularly satisfactory to put on record that it was, above all, ethologists who discredited the simple-minded and socially destructive beliefs characteristic of social Darwinism. Some human beings are aggressive, to be sure, but it is by studying ourselves and not by studying animals that we recognize this trait in mankind; indeed, it is perhaps not unfair to say that those who know most about aggression in animals are most cautious in imputing any such thing as an aggressive `instinct' to mankind.

This still leaves us, however, with the rather hectoring question of just what great new truth has been learnt about man as a result of the recognition of his evolutionary descent.

I now believe the question is wrongly put and that it embodies a false conception of the nature of scientific progress. All scientists despise the ideology of `breakthroughs' --- I mean the belief that science proceeds from one revelation to another, each one opening up a new world of understanding and advancing still farther a sharp line of demarcation between what is true and what is false. Everyone actually engaged in scientific research knows that this way of looking at things is altogether misleading, and that the frontier between understanding and bewilderment is rather like the plasma membrane of a cell as it creeps over its substratum, a pushing forward here, a retraction there --- an exploratory probing that will eventually move forward the whole body of the cell.

It is, indeed, not a grand ethological revelation that the scientist should seek from his awareness of the evolutionary process, but rather an enlargement of the understanding made possible by a new or wider angle of vision, a clue here and an apt analogy there, and a general sense of evolutionary depth in contexts in which it might otherwise be lacking. In such an exercise as this, as Robert Hinde points out, we often have as much to learn from the differences as from the similarities between human beings and animal models. In short, I think ethology is one of the many areas of thought in which a philosophic understanding of the nature of the scientific process is salutary: in real life, science does not prance from one mountain top to the next.

A great deal obviously depends upon whether or not it is possible to establish genuine homologies between the behavior of human beings and that of the collateral descendants of their remote ancestors. To my mind, there can be no question that we can do so: consider, for example, the highly complex and teleonomically related behavioral and physiological activities that go to make up sexually appetitive behavior, mating, conception, gestation, parturition, lactation, suckling and the care of the young. It is hardly conceivable that this entire complex scenario should have sprung into being for the first time with the evolution of Homo sapiens; and there is no reason nowadays why we should put ourselves to the exertion of imagining that it could have done so. The same applies to appetitive behavior as it relates to the conventional everyday context of seeking food.

Nevertheless, even if the existence of these homologies is conceded, we must not expect too much to follow from them; in particular, we should not expect a great variety of psychologically illuminating insights, for ethology, often stops short at just the level at which psychology begins, that si, stops short of explaining the nature and origin of the differences between individuals. As I pointed out in an early essay on this very subject, what is interesting about our human propensity for loving or hating is why one person loves a second and hates a third, just as what is interesting about our human dietetic habits is not the physiological need for food or the behavior that goes with it, but why any one human being will eat this and not that, here and not there, and now and not then. Psychology has had to do with the differences between human beings, ethology --- at least in its early days --- with the characteristics they have in common. I went on to say: `It is no great new truth that human beings are ambitious; what is interesting about ambition is why in one person it should take the form of wanting to become a great musician and in another of wanting to raise a large family, and in a third (for this, too, is an ambition) of wanting to do nothing at all.' These questions seem to me to belong to the domain of psychology and not to ethology at all.

I do not think there is any general answer to the question of how exact are the homologies between, say, human and primate behavior. Scientifically, it would be completely sterile to start with the presumption that homologies were illusory and the correspondences invariably inexact. Methodologically, the sensible thing is to start with the hypothesis that the homologies are fairly close; for such a hypothesis can be made the basis of action and can put us on the right wavelength for making observations of the kind which will either falsify it or strengthen our belief that there may be something in it. For example, if it were to be shown that maternal deprivation had a psychologically damaging effect on rhesus monkey babies, then there would be a case prima facie for supposing that the same might be true of human beings. If the idea is wrong it will be shown to be so, for it si the great and distinctive strength of science that we need not persist long in error if there is a genuine determination to expose our ideas to tough critical analysis. In general, it is illuminating to recognise that such human --- sometimes all too human --- activities as play, showing off and sexual rivalry are not psychic innovations of mankind, but have deep evolutionary roots.

I mention it purely as an aside that the ancient origins and deeply programmatic character of human reproductive behavior makes it extremely unlikely that a prudential system of sexual morality can be reinstated or the population explosion contained merely by exhortation or appeals to reason, however cogently worded. It also makes it extremely unlikely that there is any easy psychological solution of the problem of family limitation. On the subject of behavioral homologies as they apply to man, I cannot do better than to quote the fine closing paragraph of Darwin's The Descent of Man:

We are not here concerned with hopes or fears , only with the truth as far as our reason permits us to discover it, and I have given evidence to the best of my ability. We must, however, acknowledge as it seems to me, that man with all his noble qualities, with sympathy which he feels for the most debased, with benevolence that extends not only to other men but to the humblest living creature, with his godlike intellect which has penetrated into the movements and constitution of the solar system --- with all these exalted powers --- man still bears in his bodily frame the indelible stamp of his lowly origin.

For our own purposes this quotation must, of course, be modified so as to read `behavioral repertoire' instead of `bodily frame'. With this modification, the only word we could reasonably cavil at is the word `indelible', for we are by no means at the mercy of our bodily inheritance. For example: human evolution has taken such a course that haemolytic disease of the newborn is almost inevitable in certain situations unless we take steps to circumvent it, but we can and do take steps to do so. The same applies to human behavior: our social conventions and institutions do inhibit and to some extent prevent the more exuberant forms of the behavior so often described as `bestial' though in reality it is more characteristic of man than of beast.

I should like to quote what has always seemed to me to be a very wise passage by Alfred North Whitehead in his Introduction to Mathematics: `It is a profoundly erroneous truism, repeated by all copybooks and by eminent people when they are making speeches, that we should cultivate the habit of thinking of what we are doing. The precise opposite is the case. Civilisation advances by extending the number of operations we can perform without thinking about them.' I find this statement compellingly true of much human behavior that can be described as skilled, for if somebody asks us what seven sixes are (or anything else from our time-tables) the last thing on earth we want to do is go back to Peano and Frege or to Russell and Whitehead to puzzle out what the answer to such a searching question can be: we want to snap it out without hesitation; and likewise the process of learning to drive a car, obeying traffic signals and avoiding obstacles, etc., is in effect learning not to think about actions which did, at one time, require anxious deliberation (`Now let me see, which of these many pedals will arrest the motion of the car?'). We are learning to give learned behavior the polish and fitness for purpose which we describe colloquially as instinctive, so that we can drive a car safely even if we are tired, preoccupied and in a hurry. In the same essay I wrote:

Paradoxically enough, learning is learning not to think about operations that once needed to be thought about; we do, in a sense, strive to make learning `instinctive', i.e. to give learned behavior the readiness and aptness and accomplishment that are characteristic of instinctive behavior. [But that is only half the story.] The other half truth is that civilisation also advances by a process that si the very converse of that which Whitehead described: by learning to think about, adjust, subdue and redirect activities which are thoughtless to begin with because they are instinctive. Civilisation also advances by bringing instinctive activities within the domain of rational thought, by making them reasonable, proper and co-operative. Learning, therefore, is a twofold process: we learn to make the processes of deliberate thought `instinctive' and automatic, and we learn to make automatic and instinctive processes the subject of discriminating thought.

Anybody who professes to discern a moralising flavor in what I have been writing is perfectly right: it is exactly what I intend. I think we shall have to get used to the idea that moral judgements should intrude into the execution and application of science at every level --- and in no context more exigently than the interpretation of science for the benefit of laymen.

Anyhow, these reflections on Whitehead lead me naturally to see one or two points I should like to make on the special importance of language and the learning process and their relevance to the conception and measurement of `intelligence'.

Having dismissed the idea that we must look to ethology for great revelations about the nature of human behavior, and also the idea that scientific understanding lurches forward from one revelation to another, I should like to put on record my belief that the great contribution of ethology to the understanding of human behavior is methodological: indeed, there are already signs that some of the more eclectic psychiatrists, anxious as ever to be in the swim and in the forefront of opinion, are starting to speak about ethology as if they had invented it themselves. This will be all to the good if it has the effect of persuading them to direct upon human beings the intent and candid gaze to which ethologists owe their success in the study of animal behavior.

The very last thing I have in mind is anything like the at one time popular sociological exercise known as `mass observation' --- a kind of inductivism gone mad, as I remember it --- which was dedicated to the proposition that if only one could collect enough information about what people actually do and actually say at home, in pubs and in buses etc., then some great new truth about human behavior would, of necessity, emerge. So far as I am aware, no truth of any kind emerged, great or small, except perhaps that it takes all kinds to make a world. No: I was allowing myself to cherish the hope that ethological methods might one day make it possible to build up a biologically well-founded psychology or even a psychopathology to take the place of the weird farrago of beliefs which forms the basis of modern psychoanalytic psychotherapy, a system of beliefs which persists because it has never been found wanting, and has never been found wanting because it has never been exposed to any evaluation.

I have defended on several occasions the view that what is characteristic of human beings considered as animals is not the making and use of tools but the communication from one person to another, specially in the next generation, of the knowledge and know-how required to do so. It is by virtue of this faculty that human beings come to enjoy a kind of cultural evolution which has converted us into animals that are simultaneously aerial, terrestrial and submarine, possessing X-ray eyes and sense organs sensitive enough to feel the heat of a candle at distance of a mile. In briefest summary, because there is no point in going over old ground again, the main characteristics of this distinctively human form of evolution are:

  1. It is Lamarckian in style, that is, unlike ordinary or `endosomatic' evolution it embodies a learning process; for what is learned is passed on and becomes part of the evolutionary heritage.
  2. Cultural heredity is mediated through non-genetic channels --- hence the use of the terms `exogenetic' or `exosomatic' evolution to describe the process and distinguish it from ordinary genetic or Darwinian evolution.
  3. It is reversible. Not only in theory --- as it is possible in theory for a kettle to boil when it is put on ice --- but in good an earnest. We could return to the Stone Age in one generation.

This system of evolution is the characteristic to which we owe our clear-cut biological supremacy over all other organisms, because it has conferred almost unlimited capabilities upon us, including even that of leaving the earth and going to live elsewhere in the solar system.

This characteristically human system of heredity calls for and depends upon the existence of language and other forms of conceptual communication (one is reminded that Dr. George Steiner refers to man as a `language animal' rather than as a `tool-making animal'). The existence of and our dependence on exogenetic evolution must place a specially high selective premium upon such capabilities as teachability and imitativeness (a word for which, in this context, the vernacular term `aping' seems uncannily apt) because these form the causal nexus of cultural heredity --- points made very well by Barnett in a recent lecture entitled `Homo docens'.

With this system of heredity in mind the distinction between programmed and learned behavior loses some of its force, for an episode of behavior can be `programmed' in the sense that the physical plant and the functional capabilities necessary for its execution must be ready-made even though the behavior itself has to be learned. Barnett himself attaches considerable importance to the observations of Kuo on the killing of rats and mice by cats. It is clear that although a cat's aptitude and general capability for killing mice is `laid on' developmentally, cats are much more likely to kill mice if they have observed their mothers doing so.

The authors surmise that this kind of behavior may be characteristic of Felidae generally, noting, however, that no experiments on the subject have yet been done on tigers or lions; but if we reflect on the importance attached to teaching and the entire apparatus of pedagogy in human beings it seems difficult to resist the hypothesis that the same principle is true of human beings.

In the light of these considerations I wonder increasingly at the naïveté of those psychologists who believe that prowess in certain so-called `intelligence tests' provides a measure of so-called `innate intelligence' which is virtually unaffected by the subject's age and which cannot be taught or influenced by experience. If this were true of any human performance, my first reaction would be to dismiss that performance as relatively unimportant and certainly not one that could be made the basis of a measure of intelligence, for I believe that the endowments that have made human beings what they are, are above all imitativeness and teachability.

For these and other reasons I believe that the IQ concept and some of its practitioner should now be relegated to that dusty, cavernous and ill-lit building called the Museum of Social Darwinism, in which other principal exhibits are Francis Galton, Ernst Haeckel and Alfred Rosenberg, and many others who have misunderstood the bearing of biology on human affairs or who have propagated mischievous views in the name of science.


When the conference at which I read the paper was planned, the title I had submitted for a contribution that I thought would come well from a general biologist was: `Does the study of the behavior of animals throw any light on the behavior of Man?', and this is the topic on which I wrote my paper. The title was such a mouthful, however, that I changed it to its present form.

Put colloquially, my answer to the original question can be seen from the text to be: `Not very much, really --- it is by studying human beings themselves that we learn about their behavior.' If, however, we put the question in the form in which it actually appears at the head of my contribution, then I think the proceedings of the conference make it possible to say with some confidence: `Yes, indeed.' This puts me under an obligation to try to describe what differentiates ethology from psychology in a conventional sense.

I think ethology has two distinctions: in trying to make teleonomic sense of behavioral performances that might seem to inexperienced observers to be a stream of incoherent and functionless activities, ethologists are not yet importuned by an insistent and urgent need to find a causal explanation of every phenomenon they observe. Closely related to to this is the welcome truth that ethology, unlike some psychological systems, is not yet crabbed and confined by the doctrinal tyranny of any pre-existing explanatory system. These two characteristics give ethology the freshness and spontaneity which other biologists find so enviable, and which are sadly so lacking from many of the older and more conventional branches of zoology.

I turn now to the consideration of a number of comments and criticisms that have been made about the first version of this paper.

Apropos of exogenetic evolution, Professor Niko Tinbergen, who was unfortunately unable to be present at the meeting, expressed some doubt about the total reversibility I claimed for it, and in particular about my contention that human beings could revert to the Stone Age in one generation if the cultural nexus between one generation and the next were to be wholly severed. The matter can only be resolved by a purely notional experiment in which we are to imagine the development and likely fate of a community consisting of a number of human babes reared as if by magic on a desert island --- kept alive, indeed, but without the benefit of any of that inheritance which is passed on from one generation to the next by precept, books or word of mouth. Without that inheritance --- without any formal schooling or indoctrination in the useful arts --- would they not regenerate among themselves into something like a paleolithic culture?

This experiment becomes more and more vague and unsatisfactory the more one thinks about it, so there is no point in going on about it. I admit, though, that the `Stone Age' was a rhetorical over-simplification; the point I really want to make, which I think most people concede, is that exogenetic evolution is reversible.

Another point made by Tinbergen was that we should seek analogies no less often than homologies --- meaning by analogies `similarities convergently evolved in less closely related species in adaptation to similar niches. I accept this criticism completely, and smiled at the justice of the rebuke given to someone trained in a school of zoology for which the notion of `homology' was the central --- even the energising --- concept of the whole of biology.

I am not very well up in the literature of ethology and was not therefore surprised to learn several example unfamiliar to me of exogenetic evolution or cultural transmission among animals. Tinbergen, in particular, referred me to `acculturation' in Japanese macaques and referred me to a passage by Hinde; in addition, Mr Gady Katzir called my attention to a description of a cognate phenomenon by Marais.

Amid the general discussion that followed my paper I reminded the members of the symposium that molecular genetics began with the complete solution, by intent and single-minded research, of the phenomenon of pneumococcal transformation, and I went on to ask if there were not some comparable phenomenon in ethology, the complete interpretation of which would have an effect analogous to that of an aircraft's breaking cloud; I referred to the transformation of male into female sexual behavior and vice versa by the injection of hormones characteristic of the opposite sex --- phenomena of which a number of specially apt examples are given by Lehrman. A complete interpretation of any one of these phenomena, I argued, might have an effect on ethology comparable to the effect of O. T. Avery's work on the subsequent growth of molecular genetics. In the discussion which followed it became clear that the members of the conference were not at all sure that any such parallel between ethology and the growth of molecular biology could be drawn.

It may, in any case, turn out that the evolution I have in mind will be conceptual rather than one that turns upon the analysis of behavioral phenomena. Commenting on my paper both Hinde and Bateson professed the rather strong feeling that the Lorenzian distinction between instinctual and learned behavior is no longer useful: `all behavior is both genetically and environmentally influenced; all learned behavior is genetically influenced'.

Methodologically speaking, I think this is a rather weak declaration and part of it, at least, is tautologous. Obviously no behavioral performance could take place without the physical and physiological means to execute it --- there must be nerves and muscles and other apparatus of the kind we normally think of as being `laid on' by developmental processes. It could be said of every character trait whatsoever that its determination is partly natural and partly nutural, yet we do know of character differences that are wholly genetic in determination in the usual sense --- for example, a human being carrying the blood-group gene associated with group A will be of group A in almost any environment that is capable of sustaining life. It is an old, vexed question, and it will not be solved in the context of behavior until Aubrey Manning's ambition for the foundation of the genetics of behavior is realised.

Typed 26--27 June 1996