Tim Ingold
Department of Social Anthropology
University of Manchester
Oxford Road
Manchester M13 9PL


January 1999
Final, revised text, April 1999


Human beings are restless creatures. They are always moving about. But far from being haphazard these movements are, for the most part, highly controlled. In many cases this control is achieved through long and frequent practice; we may then speak of it as a skill. Now it is evident that people raised in different environments, and following different ways of life, also possess a range of different skills. As an anthropologist I am particularly concerned to understand the nature of these differences. It has long been conventional to attribute them to something called 'culture'. Whether culture is genuinely unique to human beings, or present - albeit in rudimentary forms - in non-human species, has been much debated. There is general agreement, however, on two points. First, humans rely on culturally acquired skills to an extent unparalleled elsewhere in the animal kingdom. Secondly, whatever biological differences may exist among human beings are irrelevant so far as their acquisition of culture is concerned. Or to put it another way, every creature born of man and woman should, in principle, be capable of acquiring the skills appropriate to any form of cultural life. 'One of the most significant facts about us', wrote the anthropologist Clifford Geertz, 'may finally be that we all begin with the natural equipment to live a thousand kinds of life but end in the end having lived only one'.1

Consider, for example, that most widespread of human movement skills: the ability to walk on two feet. Every human new-born, barring accident and handicap, has the potential to develop full bipedality. In that sense we are inclined to suppose that walking is an innate capacity, one for which - as Geertz would say - humans are naturally equipped. It is part of our biological make-up, given from the start rather than culturally acquired. Yet we also know that people in different societies are brought up to walk in very different ways. One of the first to recognise the significance of this fact, as an index of cultural variation, was the French ethnologist Marcel Mauss. In an essay of 1934 entitled 'Techniques of the Body', Mauss attempted a preliminary and very sketchy classification of the extraordinarily diverse postures and gestures adopted by people around the world in their most ordinary activities, whether at rest (sleeping, squatting, sitting, standing) or in movement (walking, running, jumping, climbing, swimming). Here are some of his notes on the subject of walking:

The habitus of the body being upright while walking, breathing, rhythm of the walk, swinging the fists, the elbows, progression with the trunk in advance of the body or by advancing either side of the body alternately (we have got accustomed to moving all the body forward at once). Feet turned in or out. Extension of the leg. We laugh at the 'goose-step'. It is the way the German army can obtain the maximum extension of the leg, given in particular that all Northerners, high on their legs, like to take as long steps as possible. In the absence of these exercises, we Frenchmen remain more or less knock-kneed...


For the adult in any society, Mauss concluded, walking is an acquired technique. There is no 'natural way' of going about it.2

How are we to reconcile these two views of walking: as innate capacity and acquired skill? One commonly proposed solution is to argue that while humans are naturally endowed with the anatomy that makes bipedal locomotion a practical possibility, and the behavioural propensity or 'instinct' to put it to effect, precise directions about how to walk are passed on from generation to generation as part of a cultural tradition. This tradition includes rules and representations laying down standards of propriety, perhaps specific to age, sex and gender, that walkers are enjoined to follow, and in terms of which their performance is evaluated and interpreted. Thus while the capacity to walk is a biological universal, particular ways of walking are expressive of social values. Would it not suffice, then, to combine the biology of human nature with the sociology of cultural difference to produce a complete 'biosocial' account of the ways people walk?

Mauss's answer was that it would not. For the link between human nature and culture can only be established by way of a third term, namely what is called the 'human mind'. The discipline that exists to study the human mind is, of course, psychology. Any account of the relation between biological and sociological dimensions of human existence must leave room, said Mauss, for the 'psychological mediator'.3 Though Mauss failed to elaborate on the reasons for this, and probably lacked the conceptual tools to do so, these have been clearly spelled out in more recent contributions to the field of culture theory which have drawn much of their inspiration from developments in cognitive science. In a nutshell, the argument goes that if rules and representations for the generation of culturally appropriate behaviour are to be transmitted from one mind to another, across the generations, then certain devices must already be in place that enable the novice to 'decode' the input of sensory data drawn from observations of the behaviour of experienced practitioners, and thereby to reconstruct these rules and representations inside his or her own head. Or as Roy D'Andrade, one of the pioneers of cognitive anthropology, has put it, the transmission of specific cultural content, in the form of programmes, depends upon the functioning of universal cognitive capacities, or processors.4

We are perhaps most familiar with this idea in the case of language learning, where it is supposed that the child's acquisition of his or her mother tongue depends on the pre-existence, in the mind, of an innate language acquisition device (LAD) that is able to process the input of speech sounds so as to establish a system of grammatical and syntactical rules for the production of well-formed and comprehensible utterances. By the same token, there ought to exist a 'walking acquisition device' - some kind of cognitive module dedicated to the construction of a culturally specific programme for bipedal locomotion from observations of other people's movements. Such a device should, in principle, be as universal to the human mind as is the predisposition to walk on two feet to the body. It seems, therefore, that to complete our picture of the walking, talking human being we have to put together three things: (i) the human body with its built-in anatomical structures and capacities of movement (limbs for walking, vocal tract for speech); (ii) the human mind with its hard-wired, computational architecture of processing mechanisms, and (iii) the assemblage of culturally specific representations or programmes whose transmission across the generations these mechanisms make possible.

I refer to this idea of the human being as the sum of three complementary parts, namely body, mind and culture, as the complementarity thesis. It is backed, as I shall show, by a formidable intellectual alliance between the theoretical paradigms of neo-Darwinism in biology, cognitive science in psychology, and culture theory in anthropology. Far from advocating this alliance, I intend to argue that it is dangerously misconceived. Before doing so, however, I should explain how its constituent parts fit together. I start with the biology.

The complementarity thesis

1. Evolutionary biology

The central claim of Darwinian biology is that human beings, along with creatures of every other kind, have evolved, through a process of variation under natural selection. This claim, however, rests on a critical assumption, namely that the growth and maturation of the individual organism - its ontogenetic development - is a separate matter from the evolution of the species to which it belongs. To be sure, what an organism does during its life is both a consequence of, and has consequences for, the evolution of its kind. Its life-history is not, however, a part of that evolution. In its Darwinian conception, evolution is not a life process. If we ask what evolves, it is not the living organism itself, nor its manifest capabilities of action, but rather a formal design specification for the organism technically known as the genotype. By definition, the genotype is given independently of any particular environmental context of development. Its evolution takes place over numerous generations, through gradual changes brought about by natural selection in the frequency of its information-bearing elements, the genes. Ontogenetic development is then understood as the process whereby the genotypic specification is translated, within a certain environmental context, into the manifest form of the phenotype.

Most contemporary biologists regard the phenotype as the outcome of an interaction, over the course of a life-cycle, between genotype and environment. Indeed I have heard this called the 'first law of biology'. But the formula is misleading on two counts. First, since 'environment' apparently includes everything relevant to the development of an organism barring the genes themselves, genes cannot interact with an environment but only in an environment with other entities that are, of course, also interacting with one another. So why should the environment always be defined in relation to the genes rather than to any other of the myriad interactants in the cell? The answer brings me to the second point, which is that the equivalence accorded to genes and environment in the interactionist formula is an illusion. In reality the distinction between genes and environment is mapped onto a far more ancient and deep-seated one in the Western tradition of thought, between form and substance. Thus the genotype is privileged as the locus of organic form, while the environment is supposed merely to provide the material conditions for its substantive realisation. Of course, an organism may develop different features in changed environments, but these differences are regarded as no more than alternative phenotypic 'expressions' of the same basic design. Only when the design itself changes is evolution said to occur.

Turning to the particular case of humankind, it follows from this account that it must be possible to specify what a human being is, independently of the manifold conditions of development under which humans grow up and live out their lives. This possibility is, in fact, entailed in the very assumption that human beings together make up a species - that is, a class of entities that may be grouped together on the grounds of their possession of certain design features transmitted along lines of descent from a common ancestral source. The sum of these features amounts to what is commonly known as 'human nature'. This notion, as we know, is far older than Darwin: what Darwinian theory added was the claim that human nature is the product of an evolutionary process. So if walking, for example, is part of human nature, then we have to suppose that it has its basis in a design specification - a kind of programme for the assembly of a functioning bipedal apparatus - that has evolved alongside all the other elements of the complete genotypic endowment that each one of us receives at the point of conception. It is in this sense that human beings are said to be universally equipped with an innate capacity to walk on two feet, regardless of how they walk in practice, or of whether they walk at all - or go everywhere by car! Specific ways of walking have not themselves evolved, they are just alternative phenotypic realisations of a pre-established, genotypic trait.

2. Cognitive science

Now just as neo-Darwinian biology presumes that there exists a context-independent specification for the design of the body, so in the field of psychology, cognitive science posits a similarly independent specification for the architecture of the mind. This architecture includes the various cognitive mechanisms or processing devices which, as I have already shown, would have to be in place before any kind of transmission of cultural representations could take place at all. As to the problem of where these mechanisms come from, cognitive scientists generally assume that this has already been solved by evolutionary biology. Since the information specifying the mechanisms cannot be transmitted culturally, we are left with only one possibility: it must be transmitted genetically - that is, as one component of the human genotype. Indeed by and large, in the literature of cognitive science, the postulation of innate mental structures is taken to require no more justification than vague references to genetics and natural selection.

In the same way that evolution has provided human beings with a body that can walk and a vocal apparatus that allows them to produce speech, so, we are told, it has also provided a mind furnished with the acquisition devices that enable them to take on representations for walking in culturally particular ways, and for speaking particular languages. However this union of evolutionary biology and cognitive science is not without its contradictions, which are proving to be a particular source of difficulty for the new discipline of evolutionary psychology to which it has given birth. The trouble lies with the distinction between innate and acquired structures. This distinction lies at the very heart of cognitive science's account of how the mind works. A mind without innate mechanisms - that is, one conceived as a 'blank slate' or tabula rasa - apparently could not learn, since it would have no way of making sense of the data of experience. And without learning there could be no transmission of representations across generations, and hence no culture.

Yet the majority of evolutionary biologists profess to have long since discarded the innate/acquired dichotomy. The architecture of the organism, they say, is neither innate nor acquired, but the outcome of a lifelong interaction between endogenous, genetic factors and exogenous, environmental ones. Now it is one thing to claim, as biologists do, that every organism starts out with a design specification (the genotype) encoded in the materials of heredity, quite another to claim - as does cognitive science - that every human being comes into the world with a preformed mental architecture. For such an architecture, in order to function, would have to exist not merely in the virtual guise of a design, but already 'hardwired' in the brain. Somehow or other, in order to kick-start the process of cultural transmission, strands of DNA have magically to transform themselves into information processing modules. This is rather like supposing that merely by replicating the design of an aircraft, on the drawing board or computer screen, one is all prepared for take-off. I shall return below to the attempts of evolutionary psychology to resolve this dilemma.

3. Culture theory

The final component of the trilogy is a certain notion of culture, conceived as a corpus of knowledge or information that can be transmitted across the generations independently of the situations of its practical application in the world. This notion goes back to a celebrated definition by anthropologist Ward Goodenough, who in 1957 pronounced that 'A society's culture consists of whatever it is one has to know or believe in order to operate in a manner acceptable to its members'.5 Notice how this definition effectively separates the process by which cultural knowledge is acquired from the way it is expressed in outwardly observable behaviour. One obtains the knowledge in order to be able to operate or function in the world. The underlying logic of this separation is identical to that which separates genotypic endowment from phenotypic expression in evolutionary biology. Just as the genotype contains a context-independent specification for the design of the organism, so the transmitted cultural information contains a context-independent specification for its behaviour, consisting of what have variously been described as plans, programmes, schemata, representations, recipes, rules and instructions. And where the genotype is said to be 'realised' in the context-specific form of a certain phenotype, through a process of development within an environment, so is culture said to be 'expressed' in the life-history of the individual by way of his or her environmentally situated behaviour.

This analogy, in turn, underwrites theories of so-called gene-culture coevolution, which start from the premise that in human populations, two mechanisms of inheritance or information transmission operate in parallel: one genetic, the other cultural.6 Each of us receives from our predecessors one set of genes, and another set of cultural instructions or 'memes', and together these pull the strings in the development of behaviour. Though in recent years the concept of the meme has been vigorously promoted in the popular writings of biologist Richard Dawkins, the idea that culture consists of particles of heritable information analogous to genes is scarcely novel. As long ago as 1956 the anthropologist Clyde Kluckhohn, participating in an interdisciplinary group that had teamed up to explore the analogies between genetic, linguistic and cultural evolution, had coined the expression 'cultural genotype' to refer to the pattern of rules and representations underlying manifest behaviour. Numerous similar suggestions have been advanced since then, one of the more recent coming from sociobiologist E. O. Wilson and his collaborator Charles J. Lumsden, who christen the analogue of the gene the 'culturgen', even going so far as to recommend how their new-fangled term should be pronounced!7

We are now in a position to sum up the relation between processes of biogenetic and cultural transmission, as generally understood within the framework of the complementarity thesis. Consider a sequence of individuals linked along a line of genealogical descent. In each generation the anatomical equipment of the body, together with the information-processing capacities of the mind, are built to genetic specifications established through natural selection. And in each generation the mind's capacities are filled with information from which are built the programmes that put the bodily equipment to use in culturally specific ways. Thus, to return to our examples of walking and talking, the genes provide the necessary instructions for building a workable bipedal apparatus and vocal tract, as well as for the assembly of the walking acquisition device and its linguistic counterpart, the LAD. With the aid of these devices, individuals of each generation are able to take on board the rules that enable them to walk, to paraphrase Goodenough, in a manner acceptable to the members of their society, and to speak correctly in the particular language of their community.

I hope, from this summary, to have made two things clear. The first is how evolutionary biology, cognitive science and culture theory conspire with one another to produce a synthetic account of the living, acting human being as a creature of three components, of genotype, mind and culture. The second is that all three approaches - in biology, psychology and anthropology - share one fundamental premise in common: namely that the bodily forms, intellectual capacities and behavioural dispositions of human beings are specified independently and in advance of their involvement in practical contexts of environmental activity. Yet in each of the three disciplines the dominant paradigms have come under attack, and for similar reasons. Neo-Darwninism has been criticised for its inability to offer an adequate account of ontogenetic development, cognitive science for its removal of the mind from human beings' bodily engagement in the world, and culture theory for its separation of knowledge from situations of practical application.

My contention is that by combining these three lines of criticism, coming respectively from developmental biology, ecological psychology and the anthropological theory of practice, it should be possible to produce a counter-synthesis infinitely more powerful than the prevailing biopsychocultural orthodoxy. In this synthesis the conventional divisions between body, mind and culture would be dissolved, not - as in the more extreme versions of sociobiology or cultural constructionism - by reducing everything to one or other of these terms, but through a unitary focus on the whole organism-person, undergoing a process of growth and development within an environment and contributing through its presence and activity to the development of others. I shall now go on to spell out the elements of this alternative view.

Developmental biology

To begin, let me return to the analysis of walking. The conventional account, as we have seen, has it that walking is a capacity universal to human beings and that, as such, it is 'programmed' into the genotype. To this, culture is supposed to add further instructions that specify the particular ways of walking appropriate to people in society. But what, precisely, does it mean to say that I, along with all my fellow humans, possess a capacity to walk? Do we all, likewise, have a capacity to swim, to relax for long periods in a squatting position, or to carry things on our heads? I can indeed swim, though plenty of people, even in my own society, cannot. Yet like everyone else who has been brought up to sit on chairs, I find that having to squat for any length of time is acutely uncomfortable - though I am told that, with sufficient training, this can be overcome. Along with the great majority of inhabitants of the Western world, however, I am quite incapable of carrying things on my head, at least without manual support.

Are we to conclude, then, that unlike walking, carrying things on the head is not innate to humans but culturally acquired? What of the capacity to read and write? Any catalogue of alleged human universals tends to project the image that people of affluent, Western societies - probably including the majority of readers of this volume - have of themselves. Thus where we, as privileged members of such societies, can do things that they - people of 'other cultures' - cannot, this is typically attributed to the greater development, in ourselves, of universal human capacities. But where they can do things that we cannot, this is put down to the particularity of their cultural tradition. To reason along these lines is to apply just the kind of double standards that have long served to reinforce the modern West's sense of its own superiority over 'the rest', and its sense of history as the progressive fulfilment of its own, markedly ethnocentric vision of human potentials. Once we level the playing field of comparison, however, only one alternative remains: that all human beings must have been genotypically endowed, at the dawn of history, with the 'capacity' to do everything that they ever have done in the past, and ever will do in the future - not only walk, talk, swim and squat but also to read and write, do the pole-vault, ride on horseback, drive cars or fly aeroplanes.

The fact of the matter, however, is that human babies are no more born walking and talking than they are born swimming or squatting: these are bodily skills whose development presupposes an environment that includes already competent caregivers, a range of supporting objects and surfaces, and a certain medium or terrain. Given the requisite environmental conditions, these skills are more or less bound to develop; yet in every case their development depends on a process of learning that is embedded in contexts of interaction with other persons and things. A moment's reflection will show that the same must be equally true of every other bodily skill that humans have ever practised, regardless of its degree of cultural particularity, or the level of social or artefactual scaffolding entailed in its acquisition. And this points inexorably to one conclusion: that the notion of capacity is quite vacuous unless it refers back to the overall set of conditions that must be in place, not only in the individual's genetic constitution but also in the surrounding environment, to make the subsequent development of the characteristic or capability in question a realistic possibility.

The point I am making could perhaps be clarified by way of a negative example. Human beings may nowadays be able to fly planes, but despite numerous and determined attempts, they cannot fly unaided. Does this not, then, establish some kind of 'bottom line'? Whatever environmental conditions may obtain, there are certain things that human beings potentially can do, and certain things that they definitely cannot. As a thought experiment, one could ask how much genetic change would be needed to turn a human being into something like a bat. The answer, no doubt, would be quite a lot - enough to rule out the possibility of such a transformation for our immediate descendants! But while the genetic difference may provide some of the explanation for why humans cannot fly and bats cannot walk, it would be a serious mistake to infer from this that the particular genetic constitution of the bat encodes, within itself, a design for constructing the mechanism of flight, or conversely, that human genes encode a design for building the apparatus of bipedalism. The source of the error lies in the identification of genetic differences with formal traits, a trick which, as Paul Weiss noted long ago, automatically vests genes with exclusive responsibility for organisation and order.8 It is one thing to claim that without certain genetic modifications having taken place in the lines of descent leading, respectively, to bats and humans, bats could not fly and humans could not walk. But it is quite another to speak of the establishment, in these lineages, of 'genes for flying' or 'genes for walking'.

I do not for one moment mean to deny that every organism starts life with its complement of DNA in the genome. However for the genome to encode any kind of design specification, it would be necessary to suppose that some means exists of 'reading off' this specification from the sequence of DNA base pairs, that is independent of any developmental process. No such means has ever been demonstrated. There is, in truth, only one reading of the genome, and that is the process of ontogenetic development itself. It follows that there can be no design for the organism other than its actual phenotypic form, as it emerges within a particular developmental context. Hence the genotype, conceived as a context-independent design specification, does not exist. It is, quite simply, a figment of the biological imagination. What this means, in general terms, is that the forms and capacities of human and other organisms are attributable, in the final analysis, not to genetic inheritance but to the generative potentials of the developmental system, that is, the entire system of relations constituted by the presence of the organism, including its genes, in a particular environment. As the philosopher of biology Susan Oyama has pointed out, only within the context of such a system can we possibly say what any gene, or cluster of genes, is 'for'.9 And so too, in the particular case of human beings, there can be no determination of what a human being is, no human nature, apart from the manifold ways in which humans become, as they live out their lives in diverse communities and native environments.

Thus in learning to walk, the infant learns to walk in the approved manner of his or her society: it is not as though the latter is somehow added on to a generalised bipedality that has miraculously appeared of its own accord, in advance of the infant's entry into the world. Hence there is no such thing as 'bipedal locomotion', as distinct from the various ways in which people actually walk; no pre-programmed 'essence' of the activity that is isolable from the real-time performance of the activity itself.10 And since ways of walking are properties of neither genes nor 'culture' (conceived as a package of transmissible information), but rather of developmental systems, to account for their evolution we have to understand how such systems are constituted and reconstituted over time. The key to this understanding lies in the recognition that humans, like all other creatures, set up through their own actions the environmental conditions both for their own future development and for that of others to which they relate. Thus they figure not as passive 'sites' of evolutionary change but as creative agents, producers as well as products of their own evolution. We therefore seek in vain for the evolutionary origins of a skill like walking. For far from having been fixed genetically, at some time in the ancestral past, such skills continue to evolve in the very course of our everyday lives. They have not originated yet, and they never will.

Clearly, neither orthodox evolutionary biology nor its complement in the field of culture theory is able to offer a coherent account of human development. According to the complementarity thesis, every human being is in part ready-made genetically, in part moulded through the superimposition, upon this preformed substrate, of pre-existing norms and values. This moulding process is conventionally regarded as one of enculturation or socialisation. Thus in childhood, as anthropologist Walter Goldschmidt puts it, the infant is allegedly transformed 'from a purely biological being into a culture-bearing one'.11 Real humans, however, are not like that. Rather, they grow in an environment furnished by the presence and activities of others. To be sure, as people go through life, they grow out of certain ways of doing things, and grow into others. But no-one has ever grown out of biology, or grown into society or culture. We do not progress, in the course of our lives, from a stage of biological incompletion as 'mere' organisms, to one of social completion as fully fledged persons. For every one of us is fully and indissolubly organism and person from beginning to end. And there is nothing in the least 'mere' - that is, partial or residual - about being an organism!

If walking, then, is simultaneously biological and social, this is not because it is the sum of separately endowed components, the one belonging to our common genetic heritage as members of the human species, the other added on through our education in the standards of a specific tradition. Walking is certainly biological, in that it is part of the modus operandi of the human organism. But it is only thanks to the person's involvement in a social world that he or she can undergo normal development as an organic being. Human infants do not learn to walk in isolation, and even adults rarely walk alone. In everyday practice a person's movements, his or her step, gait and pace, are continually responsive to the movements of others in the immediate environment. Indeed it is largely in this responsiveness that the skill of walking lies. And it is in this respect, too, rather than in its expression of values that somehow reside in an extra-somatic domain of collective representations, that walking is pre-eminently social.

Ecological psychology

Let me turn, now, from biology to psychology. I have already shown that it is impossible to derive a design specification for the organism from its genetic constitution alone, independently of the conditions of its development in an environment. For cognitive science this problem is further compounded, for if the theory of learning as the transmission of cultural information is to work, the requisite processing devices must already exist, not merely - as it were - 'on the drawing board', but in the concrete hardwiring of human brains. Attempts in the literature to resolve this problem, insofar as it is even recognised, are confused and contradictory. To cut a rather long and tangled story short, they boil down to two distinct claims. One is that the concrete mechanisms making up what evolutionary psychologists call the mind's 'evolved architecture' are reliably constructed, or 'wired up', under all possible circumstances. The other is that these universal mechanisms proceed to work on variable inputs from the environment to produce the diversity of manifest capabilities that we actually observe.

Consider the specific example of language acquisition. Here, the alleged universal mechanism is the aforementioned 'language acquisition device' (LAD). All human infants, regardless of the language or languages they may end up speaking in later life, are supposed to come pre-equipped with such a device. It is a fact, say evolutionary psychologists John Tooby and Leda Cosmides (without citing any evidence to support their claim), that even individuals who survive a childhood of aberrant social isolation, though they may never acquire a language and remain incapable of speech, will nevertheless possess 'the same species-typical language acquisition device as everyone else'.12 During a well-defined stage of development, this device is said to be activated, operating on the input of speech sounds from the environment so as to install, in the infant's mind, the grammar and lexicon of the particular language spoken in his or her community. It would thus appear that language acquisition is a two-stage process: in the first, the LAD is constructed; in the second, it is put into service in order to furnish the capacity so established with specific syntactic and semantic content.




This model of cognitive development is summarised in figure 1. Notice how the model depends on factoring out those features of the environment that are constant, or reliably present, in every conceivable developmental context, from those that represent a source of 'variable input' from one context to another. Only the former are relevant in the first stage (the construction of innate mechanisms); only the latter are relevant in the second (the acquisition of culturally specific capabilities). Once again, the parallel with walking is instructive. One of the conditions for learning to walk, obviously, is that there be a ground surface to walk on. It is reasonable to assume, without going into the wilder fantasies of science fiction, that this condition is universally fulfilled. Yet how could the infant, taking his or her first steps, encounter 'ground', as a concrete condition of development, not only as distinct from, but also prior to, such diverse 'walk-on-able' surfaces as sand, asphalt, meadow and heath, all of which call for different modalities of gait, balance and footwork? And how, again, could such a ground be free from all contours? Bizarre as it may seem, this is precisely how the ground with which we first come into contact through the feet would have to be experienced, if we are to hold on to the notion that specific techniques of walking are superimposed upon an innate biophysical substrate in the human body. And just the same kind of partitioning in the child's experience of the environment is entailed by the notion that competence in his or her mother tongue is acquired upon the base of a preformed 'language instinct'.

Of course for comparative analytic purposes it is sometimes helpful, even essential, to sift the general from the particular, or to establish a kind of 'lowest common denominator' of development. But real environments are not partitioned in this way. Let me continue for a moment with the example of language learning. Recent research has shown that from well before birth, infants are sensitive to the surrounding ambience of sound, and above all to the mother's voice. Thus the human baby comes into the world already attuned to certain environmentally specific sound patterns. From birth onwards, it is surrounded by an entourage of variously accomplished speakers who provide support in the form both of contextually grounded interpretations of the infant's vocalisations and of demonstrations, or attention-directing gestures, to accompany their own. This environment, then, is not a source of variable input for a preconstructed 'device', but rather furnishes the variable conditions for the growth or self-assembly, in the course of early development, of the neurophysiological structures underwriting the child's capacity to speak. As the conditions vary, so these structures will take manifold forms, each differentially 'tuned' both to specific sound-patterns and to other features of local contexts of utterance. These variably attuned structures, and the competencies they establish, correspond of course to what appear to observers as the diverse languages of the world. In short, language - in the sense of the child's capacity to speak in the manner of his or her community - is not acquired. Rather, it is continually being generated and regenerated in the developmental contexts of children's involvement in worlds of speech. And if language is not acquired, there can be no such thing as an innate language learning device.13

What applies specifically in the case of language and speech also applies, more generally, to other aspects of cultural competence. Learning to walk in a particular way, or to play a certain musical instrument, or to practise a sport like cricket or tennis, is a matter not of acquiring from an environment representations that satisfy the input conditions of preconstituted cognitive devices, but of the formation within an environment, of the necessary neurological connections, along with attendant features of musculature and anatomy, that underwrite the various skills involved. In short, the systems that actually generate skilled activity are not hard-wired but 'softly assembled'.14 This conclusion, however, puts paid to one of the key ideas behind the thesis of the complementarity of body, mind and culture, namely that cultural learning is like filling a universal, genetically specified container with culturally specific content. The notion that culture is transmissible from one generation to the next as a corpus of knowledge, independently of its application in the world, is untenable for the simple reason that it rests on the impossible precondition of a ready-made cognitive architecture. The condition is impossible because no matter at what point in the life-cycle one might choose to identify a particular structure or mechanism - even before birth - a history of development in a certain environment already lies behind it.

In truth, nothing is really transmitted at all. For the growth of practical knowledge in the life history of a person is a result not of information transmission but of guided rediscovery. By this I mean that in each successive generation, novices learn through being placed in situations in which, faced with certain tasks, they are shown what to do and what to watch out for, under the tutelage of more experienced hands. To show something to someone is to cause it to be made present for that person, so that he or she can apprehend it directly, whether by looking, listening or feeling. Here the responsibility of the tutor is to set up the conditions in which the novice is afforded the possibility of such unmediated experience. Placed in a situation of this kind, the novice is instructed to attend to this or that aspect of what can be seen, touched or heard, so as to get the 'feel' of it for him- or herself. This is not a matter of replicating memes, culturgens or any other such particles of cultural information. For what each generation contributes to the next are not rules and representations for the production of appropriate behaviour, but rather the specific circumstances under which successors, growing up in a social world, can develop their own embodied skills and dispositions, and their powers of awareness and response. Learning in this sense is tantamount to what James Gibson, who pioneered the approach known as ecological psychology, called an 'education of attention'.15

Ecological psychologists reject the view, which lies at the crux of cognitive science, that individuals acquire the knowledge needed to operate in the external world through a processing, in the mind, of sensory inputs delivered to it from the receptor organs of the body. This view, they argue, artificially separates the activity of the mind in the body from the reactivity of the body in the world, and in so doing merely perpetuates a mind-body split that has bedevilled our thinking since the days of Descartes. An ecological approach, to the contrary, takes as its point of departure the condition of the whole organism-person, indivisibly body and mind, actively engaged with salient components of the environment in the practical tasks of life. Human beings, like other animals, get to know the world directly, by moving about in the environment and discovering what it affords, rather than by representing it in the mind. Thus meaning is not the form that the mind contributes, by way of its innate or acquired schemata, to the flux of 'raw' sensory data, but is rather continually being generated within the relational contexts of people's practical engagement with the world around them.

In line with his ecological principles, Gibson maintained that we learn to perceive by a fine-tuning or sensitisation of the entire perceptual system, comprising the brain and peripheral receptor organs along with their neural and muscular linkages, to particular features of our surroundings. Through this process, the human being emerges not as a creature whose evolved capacities are filled up with structures that represent the world, but rather as a centre of awareness and agency whose processes resonate with those of the environment. Knowledge, then, far from lying in the relations between structures in the world and structures in the mind, mediated by the person of the knower, is immanent in the life and experience of the knower as it unfolds within the field of practice set up through his or her presence as a being-in-the-world. With this conclusion we have reached the point where we can cross the final barrier, from the psychology of perception to the anthropology of cultural difference. It is to the latter that I now turn.


The anthropological theory of practice

I have shown that in the terms of the complementarity thesis, culture consists of packages of rules and representations that are available for transmission across the generations, independently of their practical application in contexts of perception and action. These rules and representations add up to what cognitive anthropologists have taken to calling 'cultural models'. Culture, suggests Bradd Shore in a recent volume, is best seen as a very large and heterogeneous collection of such models. Corresponding closely to what many psychologists call schemas, these models are said to furnish people with 'what they must know in order to act as they do, make the things they make, and interpret their experience in the distinctive way that they do'.16 But just as the formal design that orthodox evolutionary biology ascribes to an organism's genotype is allegedly unaffected by the vagaries of its life history, so the knowledge held in the cultural models is supposed to remain aloof from the 'hands on' business of doing, making and experiencing. Acquired from predecessors and stored in memory, whence it will be passed along to successors, this knowledge is supposed to be expressed in practice, but not to be generated in it.

However this view of culture - as what Roy D'Andrade has called '"pass it along" type information'17 - has not gone unchallenged. Indeed there is a powerful movement within contemporary anthropology that would reject it altogether. One of the most influential figures in this movement has been Pierre Bourdieu. In a series of works dedicated to the elaboration of a comprehensive theory of practice, Bourdieu has attempted to show how knowledge, rather than being imported by the mind into contexts of experience, is itself generated within these contexts in the course of people's involvement with others in the ordinary business of life. What he has in mind is the kind of practical knowhow that we associate with skill - a knowhow that we carry in our bodies and that is notoriously refractory to codification in terms of any system of rules and representations. Think of the technique involved in tying shoelaces, or breaking an egg, or doing the ironing. Skills such as these are developed not through formal instruction but through the repeated and often wordless performance of tasks involving specific postures and gestures, or what Bourdieu calls a particular body hexis. 'A way of walking, a tilt of the head, facial expressions, ways of sitting and of using implements' - all these, and more, comprise what it takes to be an accomplished practitioner, and together they furnish a person with his or her orientations in the world.18

One way of putting this would be to say that these skills and orientations are 'embodied'. This is not to suggest, as do many writers in social and cultural theory, that the human body should be understood as a kind of vessel into which social and cultural content can be poured, or a surface upon which it can be inscribed. Such a view would be mistaken, on two counts. First, it would render the body passive and inert; secondly, its movements would be reduced to mere signs, directing attention elsewhere in the search for what they stand for - namely, to a realm of attitudes, beliefs or mental states that floats like a mirage above the road we tread in real life.19 The point is rather that in treading this road, in our movement along a particular way of life, the body undergoes processes of growth and decay, and that concomitantly, particular skills, habits, capacities and strengths, as well as debilities and weaknesses, are enfolded into its very constitution - in its neurology, musculature, even its anatomy. To adopt a distinction suggested by the social historian Paul Connerton, this is a matter of incorporation rather than inscription.20 Walking, for example, could be said to be embodied in the sense of being developmentally incorporated through practice and training in an environment. The same, in principle, should go for every other practical skill.

Having arrived at this point, however, I can see no further justification for upholding a distinction between the body and the organism. Surely the body - with its powers of autonomous movement, active and alive to the world - is the organism. But so, for that matter, is the mind. Indeed one could just as well speak of 'enmindment' as of 'embodiment', for to develop certain patterns of movement in the world is, at one and the same time, to develop certain modalities of attending to it. If mind, as Gregory Bateson so passionately argued, 'is not limited by the skin', but rather extends outwards into the environment along the multiple pathways of sensory involvement,21 so likewise the body is not a static, self-contained entity but given in movement, undergoing continual growth and development along the lines of its manifold environmental relationships. Body and mind, therefore, are not two separate things but two ways of describing the same thing - or better, the same process, namely the activity of the organism-person in his or her environment. What, after all, is movement, as Thelen asks, 'but a form of perception, a way of knowing the world as well as acting on it?' Walking, for example, could be described as way of getting about, but equally as a way of getting to know the environment, primarily by way of contact through the feet, but also thanks to the sights and sounds that the movement affords.

These observations offer a new resolution to one of the oldest of anthropological conundrums. Take two people from different backgrounds, and place them in the same situation: they will differ in what they make of it. Why should this be so? Cognitive anthropologists would respond that it is because they are handling the same input of sensory data in terms of dissimilar cultural models or representational schemas. The theory of practice, however, suggests an alternative answer. Our two characters perceive their surroundings differently because they have been trained, through previous experience of carrying out diverse practical tasks involving particular bodily movements and sensitivities, to orient themselves in relation to the environment and to attend to its features in different ways. The difference, in other words, lies not in the ways people represent the environment inside their heads, but in the ways they discover what it affords for their activities. Crucially, this implies that how people perceive will depend upon how they move, including how they walk. I have noted already that a large part of the skill of walking lies in the responsiveness of one's movement to the movements of others in the vicinity. But it responds, too, to ever-changing conditions in the non-human environment. That it does so is immediately apparent if we pause to imagine what walking would be like if it did not.

As an illustration, let me return to Marcel Mauss's observations on the subject of walking, which I cited at the outset. We laugh, he said, at the goose-stepping German soldier. Why? Because his movements are so oddly mechanical. No-one naturally walks like that: indeed if they did, they would forever be tripping over things. The goose-step is only possible on the artificially monotonous surface of the parade-ground. Under ordinary conditions the human gait, though rhythmic, is never metronomic, nor do the feet or knees follow exactly the same trajectory from step to step. Over fifty years ago one of the pioneers in the study of human movement, the Russian scientist Nicholai Bernstein, observed a similar phenomenon in a study of blacksmiths at work.23 He found that while the trajectory of the tip of the skilled smith's hammer, in repeated blows on the anvil, was remarkably consistent, the trajectory of his limbs was never precisely the same from one blow to the next. How can it be, Bernstein asked, that the motion of the hammer rather than that of the limbs is reliably replicated, when it is only by way of the limbs that the hammer is made to move? The smith's movements, he reasoned, cannot be the output of a fixed motor programme. Rather, the secret of dexterity must lie in the continual adjustment or 'tuning' of movement in response to an ongoing perceptual monitoring of the emergent task. In short, the smith watches and feels as he works.

In just the same way, the skilled walker adjusts his or her footwork to a monitoring - which may be auditory and visual as well as tactile - of ever variable ground conditions. From this, there follows a conclusion of capital importance. Neither walking nor any other kind of skilled practice can be reduced to the mechanical application of a fixed motor programme or formula. Hence also, it cannot be through the replication of such formulae that skills are passed from generation to generation. Traditional models of social learning, as we have seen, separate the intergenerational transmission of information specifying particular techniques from the application of this information in practice. First, a model or generative schema is established in the novice's mind from observations of the movements of already accomplished practitioners; secondly, the novice imitates these movements by running off exemplars of the technique in question from the schema. Now I do not deny that the learning of skills involves both observation and imitation. But the former is no more a matter of forming internal, mental representations of observed behaviour than is the latter a matter of converting these representations into manifest practice. For the novice's observation of accomplished practitioners is not detached from, but grounded in, his own active, perceptual engagement with his surroundings. And the key to imitation lies in the intimate co-ordination of the movement of the novice's attention to others with his own bodily movement in the world.

This is what is meant, colloquially, by getting the 'feel' of things. And it is this, too, that marks the progression from clumsiness to dexterity. The clumsy practitioner is precisely one who implements mechanically a sequence of received instructions, while remaining insensitive to the evolving conditions of the task as it unfolds. Conversely, to have a feel for what one is doing means moving in a way that is continually and subtly responsive to the nuances of one's relations with relevant aspects of the environment. To achieve such fluency of performance it is not enough to observe; one has also to undertake repeated practical trials. But in these trials it is the task, rather than the precise trajectory of movement, that is repeated. The novice or apprentice, engaged in such trials, is not 'acquiring culture', as though it could be simply downloaded into his or her head from a superior source in society, but is rather embarked upon the process that anthropologist Jean Lave has called 'understanding in practice'.24 And our conclusion, that such understanding calls for a fine-tuning of skills of action and perception through repeated trials within an environment, is fully consistent not only with the ecological approach in psychology, reviewed above, but also with a biological focus on the generative dynamics of developmental systems.



At this point we can return to the claim of Clifford Geertz, cited at the outset, that while all humans come into the world with the 'natural equipment' to live any kind of life, in the end they live only one. Human life, in this view, is conceived as a movement from the universal to the particular, or from biology to culture, entailing a gradual filling up of capacities and a closing down of possibilities. I can only conclude that such a view is fundamentally mistaken. The fact is that our bodily equipment, if we can call it that, is not ready-made but undergoes continual formation in the course of our lives. Even the skeleton, for example, grows in body that is actively doing things, and its precise form is liable to bear the mark of these activities.25 And the growth of the body is an aspect of the very same developmental process by which we gain proficiency in the skills appropriate to the particular kind of life we lead. So what do we begin with? What is already in place at the moment of inauguration of a new human life-cycle? The answer, as we have seen, is not an open-ended design specification in the guise of the genotype, but rather the total system of relations comprised by the presence of the fertilised egg with its complement of DNA, in a womb, in the body of the mother-to-be, who in turn is alive and active within a particular environment. In short, what each of us begins with is a developmental system.

Human beings, then, are not born biologically or psychologically identical, prior to their differentiation by culture. There has to be something wrong with any explanatory scheme that needs to base itself on the manifestly ludicrous claim - in the words of John Tooby and Leda Cosmides - that 'infants are everywhere the same'.26 Even parents of identical twins know this to be untrue! The source of the difficulty lies in the notion that culture is an extra ingredient that has to be 'added in' so as to complete the human being. We have found, to the contrary, that all those specific abilities that have classically been attributed to culture - to walk in a certain way, to speak a certain language, to sit or squat, and so on - are in reality incorporated, through processes of development, as properties of human organisms. In that sense, they are fully biological. Culture, then, is not superorganic or supra-biological. It is not something added to organisms but a measure of the differences between them. And these differences, as I have shown, arise from the ways in which they are positioned vis--vis one another, and non-human components of the environment, in wider fields of relationship.

Now if, by evolution, we mean differentiation and change over time in the forms and capacities of organisms, then we must surely allow that skills of the kind we have been concerned with here - being biological properties of organisms - must have evolved. We cannot, however, attribute this evolution to changing gene frequencies. No-one would seriously suggest that people from different backgrounds walk in different ways, or speak different languages, because of differences in their genetic make-up. But nor does it make sense, as we have seen, to suppose that these differences are due to something else, namely culture, that overwrites a generalised biological substrate. Walking and talking are no more the operations of a mind impregnated by culture than they are of a body designed by natural selection. They are rather developmentally enhanced achievements of the whole organism-person, at once body and mind, positioned within an environment. And to account for these achievements, what we need is nothing less than a new approach to evolution, one that sets out to explore not the variation and selection of intergenerationally transmitted attributes, but the self-organising dynamics and form-generating potentials of relational fields.

I do not deny that cumulative changes may take place, over successive generations within a population, in the frequencies with which particular genes are represented. Nor do I deny that these changes can be explained, at least in part, by the logic of natural selection. What I do deny, however, is the existence of any link between changes on the one hand, in gene frequencies, and on the other, in the forms and capacities of organisms, that is independent of the dynamics of development. In orthodox evolutionary biology, this link is established by way of the concept of the genotype. Remove this concept, and you take away the keystone without which the entire edifice of neo-Darwinian theory collapses. Natural selection, in short, may occur within evolution, but does not explain it. Only by going beyond the theory of evolution through variation under natural selection, and by considering the properties of dynamic self-organisation of developmental systems, can we hope to discover the possible consequences of those changes that can be explained by natural selection for the evolutionary process itself.

The root source of the explanatory poverty of neo-Darwinian theory is not hard to find. It lies in what one of its principal architects, Ernst Mayr, calls 'population thinking'.27 Modern biology, Mayr insists, requires us to think of evolutionary change as aggregated over populations of numerous discrete individuals, each of which is uniquely specified in its essential constitution independently of, and prior to, its life in the world. This way of thinking, however, systematically disrupts any attempt to understand the generative dynamics of developmental systems. How, after all, can one hope to grasp the continuity of the life process through a mode of thought that can only countenance the organic world already shattered into a myriad of fragments? All it can do is to count up the pieces. What we need, instead, is a quite different way of thinking about organisms and their environments. I call this 'relational thinking'. It means treating the organism not as a discrete, prespecified entity but as a particular locus of growth and development within a continuous field of relationships. It is a field that unfolds in the life activities of organisms and that is enfolded (through processes of embodiment or enmindment) in their specific morphologies, powers of movement and capacities of awareness and response. Our conception of evolution, then, is more topological than statistical. But only with such a conception, I contend, can we understand the evolutionary process from within, recognising that we ourselves are no more capable of watching from the sidelines than are creatures of any other kind, and that like them, we participate with the whole of our being in the continuum of organic life.


Figure 1

The two stages of cognitive development according to the complementarity thesis. In the first stage the human genotype interacts with the constant component of the environment to produce the universal mechanisms of the mind's evolved architecture. In the second, this architecture operates on variable environmental inputs to produce culturally specific capabilities. (Source: Tim Ingold, "From Complementarity to Obviation: On Dissolving the Boundaries Between Social and Biological Anthropology, Archaeology and Psychology'. Zeitschrift fr Ethnologie 123 [1988]: 21-52, p.43.)



1. Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1973), 45.

2. Marcel Mauss, Sociology and Psychology: Essays, tr. B. Brewster (London: Routledge and Kegan Paul, 1979), 102, 114-15.

3. Ibid., 101.

4. Roy G. D'Andrade, "The Cultural Part of Cognition". Cognitive Science 5 (1981): 179-195.

5. Cited in R. G. D'Andrade, "Cultural Meaning Systems", in Richard A. Shweder and Robert A. LeVine, eds, Culture Theory: Essays on Mind, Self and Emotion (Cambridge: Cambridge University Press, 1984), 89.

6. William H. Durham, Coevolution: Genes, Culture and Human Diversity (Stanford, CA: Stanford University Press, 1991).

7. R. W. Gerard, Clyde Kluckhohn and Anatol Rapoport, "Biological and Cultural Evolution: Some Analogies and Explorations". Behavioral Science 1 (1956): 6-34; Charles J. Lumsden and Edward O. Wilson, Genes, Mind and Culture (Cambridge, Mass.: Harvard University Press, 1981), 7.

8. Paul Weiss, "The Living System: Determinism Stratified", in Arthur Koestler and J. R. Smythies, eds, Beyond Reductionism: New Perspectives in the Life Sciences (London: Hutchinson, 1969), 35.

9. Susan Oyama, The Ontogeny of Information: Developmental Systems and Their Evolution (Cambridge: Cambridge University Press, 1985).

10. This point has been powerfully argued, with specific reference to walking, by Esther Thelen. See Esther Thelen, "Motor Development: A New Synthesis". American Psychologist 50 (1995): 79-95.

11. Walter Goldschmidt, "On the Relationship Between Biology and Anthropology", Man (N.S.) 28 (1993): 341-359.

12. John Tooby and Leda Cosmides, "The Psychological Foundations of Culture", in Jerome H. Barkow, Leda Cosmides and John Tooby, eds, The Adapted Mind: Evolutionary Psychology and the Generation of Culture (New York: Oxford University Press, 1992), 45.

13. A. DeCasper and M. Spence, "Prenatal Maternal Speech Influences Newborns' Perception of Speech Sounds", Infant Behavior and Development 9 (1986): 13-50; Patricia Zukow-Goldring, "A Social Ecological Realist Approach to the Emergence of the Lexicon: Educating Attention to Amodal Invariants in Gesture and Speech", in Cathy Dent-Read and Patricia Zukow-Goldring, eds, Evolving Explanations of Development: Ecological Approaches to Organism-Environment Systems (Washington, DC: American Psychological Association, 1997); C. H. Dent, "An Ecological Approach to Language Development: An Alternative Functionalism", Developmental Psychobiology 23 (1990): 679-703.

14. On this idea, see Thelen, op. cit., and Andy Clark, Being There: Putting Brain, Body and the World Together Again (Cambridge, Mass.: MIT Press, 1997), 42-5.

15. James J. Gibson, The Ecological Approach to Visual Perception (Boston: Houghton Mifflin, 1979) 254.

16. These lines are quoted from Naomi Quinn and Dorothy Holland, "Culture and Cognition", in Dorothy Holland and Naomi Quinn, eds, Cultural Models in Language and Thought (Cambridge: Cambridge University Press, 1987), 4. See Bradd Shore, Culture in Mind: Cognition, Culture and the Problem of Meaning (New York: Oxford University Press, 1996), 44.

17. D'Andrade, op. cit., 179.

18. Pierre Bourdieu, Outline of a Theory of Practice (Cambridge: Cambridge University Press, 1977), 87.

19. On these points, see Michael Jackson, Paths Toward a Clearing: Radical Empiricism and Ethnographic Inquiry (Bloomington: Indiana University Press, 1989), 122-3.

20. Paul Connerton, How Societies Remember (Cambridge: Cambridge University Press, 1989), 72-3.

21. Gregory Bateson, Steps to an Ecology of Mind (London: Granada, 1973), 429.

22. Thelen, op. cit., 89.

23. Nicholai A. Bernstein, "On Dexterity and its Development", in Mark L. Latash and Michael T. Turvey, eds, Dexterity and its Development (Mahwah, NJ: Lawrence Erlbaum Associates, 1996).

24. Jean Lave, "The Culture of Acquisition and the Practice of Understanding", in James W. Stigler, Richard A. Shweder and Gilbert Herdt, eds, Cultural Psychology: Essays on Comparative Human Development (Cambridge: Cambridge University Press, 1990), 309-27.

25. See, for example, Theya Molleson, "The Eloquent Bones of Abu Hureyra", Scientific American 271 (1994): 60-65.

26. Tooby and Cosmides, op. cit., 33.

27. Ernst Mayr, The Growth of Biological Thought (Cambridge, Mass.: Harvard University Press, 1982), 45-7.