Durkheimian anthropology and religion

This work is licensed under the Creative Commons | © Maurice Bloch. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau5.3.019


Durkheimian anthropology and religion

Going in and out of each other’s bodies

Maurice BLOCH, London School of Economics and Political Science


In memory of Skip Rappaport

Emile Durkheim’s work has always been criticized for reifying the social and situating it in an indeterminate zone between actors’ consciousness and positive facts.1 In this chapter, however, I am not concerned with exploring whether this criticism of the founder of French sociology’s work is justified. My purpose instead is to show that it is possible to retain some aspects of Durkheim’s conclusions about the nature of religion and of the social with types of argument quite different from those he employed. My framework here is that of modern evolutionary natural science and recent understandings of the specificities of the human mind/brain.

Such an evolutionist perspective tends to make social/cultural anthropologists uncomfortable. I hope that as they read on, they discover that an evolutionist perspective does not necessarily lead to the dangers they envision; and that it can even be reconciled with some of their most cherished ideas that will emerge all the stronger as a result.[286]

But because one might as well hang for a sheep as for a lamb, I begin my argument much further back than is usual in evolutionary anthropology with a consideration of the very earliest stages of life on earth, when unicellular organisms associated together to form multicellular units in the Cambrian era.

During this crucial transition, and for millions of years, it was far from clear whether those early multicellular organisms were one or many because they were in an in-between stage. This biological conundrum still exists, in varying ways and to varying degrees, for many subsequent and more complex forms of life. An extreme example is coral, about which one can argue equally plausibly either that the minute units of which it consists are separate organisms or that whole coral branches (or even whole reefs) are one single animal.

The difficulty of isolating the “individual” does not only apply for such exceptional life forms. The issue of identifying the specific unit on which natural selection acts arises in respect of all living things and has become particularly acute in modern biology. Does natural selection occur at the gene level or on combinations of associated genes? Or is it at the level of the individual? Or on a larger group that shares genes to differing degrees (Stotz and Griffiths 2004)?

This sort of question is particularly problematic when we are dealing with social species. Is it the bee or the hive that is the animal? After all, the bees in a hive are as genetically identical as are the different bits of the human body, and a hive possesses only one set of working reproductive organs.

The biological problems do not end there. When does an embryo become separate from its mother? Is a live spermatozoid a unit? More generally, how far are parents one with their children, and are descendants of individuals their continuation or new units? Are descent groups one body? Do members of one caste have unique distinctive types of blood? Are nations one people? Are we all the children of God in the brotherhood of Christ? Is society, as Durkheim claimed, more than the sum of the constituent individuals?

Here, those readers who have already given me up as some sort of biological reductionist, indifferent to the higher purpose of cultural anthropology, might summon a ?icker of interest with these more familiar disciplinary questions. They may even begin to hope that I might have something to say about religion and ritual, which, after all, is what this book is about. I shall get there . . . eventually. And indeed, my prime purpose in this chapter is to consider the theoretical implications of the way I have just managed to slither from a discussion of the structure of coral to hoary classical subjects in anthropology and even to central tenets of some interpretations of the Christian religion.

If the reader is totally unsympathetic to the approach, however, I propose they will already have revelled in identifying a familiar sleight of hand: representing facts about the world as if they were just that, without having ?rst recited the anthropologists’ exorcism prayer.

I humbly acknowledge that everything I say is nothing but an epiphenomenon of my present cultural position and time and that this inevitably leads me to essentialize a particular cultural position and then mercilessly impose it on defenseless people.[287]

In other words, I have been guilty of suggesting that my scientific knowledge, a mere elitist manifestation of my own culture, is somehow the basis of the propositions made by those people around the world who say things like this: “The members of our group, which has existed since the beginning of time, share a distinctive type of bone”; or “Our lineage consists of one body”; or “Initiation reunites us with our ancestors”; or “Ask not for whom the bell tolls; it tolls for thee.”

I would have thus committed all the category mistakes in the book. Especially in having forgotten the fact that the cultural creates an impenetrable screen between what is and our cultural representations. Familiar arguments of this kind might be partly justified as ?rst steps when we teach an introduction to anthropology,2 but in this chapter I argue that when left in categorical form, they are as misleading as the ethnocentricism that anthropologists love to denounce.

We can start with a classic and familiar polemic as a way of introducing the theoretical position I shall adopt here.

In the bad-old days, so the story goes, anthropologists used to think that kinship was based on the fact that people go in and out of each other’s bodies. Indeed, they might have stressed that the physical separation of a child from its mother takes quite a while, with intermediate phases such as breast feeding and child care. Some of these earlier vulgar anthropologists went as far as to suggest that the care given by fathers to infants was somehow the consequence of having gone into the mother during sexual intercourse. They argued that these “natural” foundations were the common base of all different kinship systems (Yanagisako and Collier 1987: 30–35).

Such naïveté, however, was soon to be severely disciplined by developments in our subject. First, anthropologists stressed the old platonic point that humans do not live in the world as God or the scientists see it, but via their own understanding of it (I don’t see why this does not apply to other animals, too). From this they argued that the foundation (i.e., going in and out of each other’s bodies) cannot be the direct foundation of social knowledge. This correction was, however, soon deemed not to have been severe enough. It was not simply that people saw the world “through a glass darkly”; it was that they did not see it at all. There was no such fact as that people went in and out of each other’s bodies; they were just accidental cultural representations of which my particular formulation is only one among many. Thus, to talk of different, culturally constructed kinship systems as if they were cultural interpretations of a single reality was a fallacy. In a wonderful metaphor, David Schneider explained that if you went out into the world armed with a kinship-shaped cutting tool, you inevitably got kinship-shaped pieces. By this he implied that if the tool had had any other shape than the western-shaped kinship tool, which would be the case with the tools used by the “others,” you would have gotten a quite different shape (Schneider 1984, 198).

I have always liked this metaphor of Schneider’s because, as a child, I used to spend much time watching my grandmother making biscuits. She would roll out a large flat pancake of dough on the marble of the kitchen table, and with a few ancient tin tools she would cut out various shapes. This is exactly what Schneider [288]has in mind. But the other reason why I like his metaphor is that what is wrong with it is also obvious. The world in which people go in and out of each other, the denounced foundation, is not (as Schneider’s analogy suggests) inert, undifferentiated, and flat like biscuit dough. It has a shape, and this shape, while it does not determine the way the world will be represented, severely restricts the parameters of what is likely.

Plato also used a culinary metaphor to talk about the world. For him, however, the world was more like a roast chicken than pastry, and unless you really wanted to make things difficult for yourself, you would “carve it at the joints,” wherever they occurred on the animal you were serving up.

Indeed, it is the dialectic between the facts of sex and birth and the cultural representations of these phenomena that most promises to advance our understanding of the nature of human beings, which, of course also involves the cultural (and hence historical) aspect. But examination of this dialectic is what the Schneiderian rhetoric makes impossible by refusing to allow us to ask what the representations “are about” and what the world is like. A trivial objection to the effect that not all languages have a word for what anthropologists call “kinship” puts a stop to any consideration of the really important questions about our species.

And there is yet something else that is obscured by Schneider’s figure of speech. The cutting tools, which represent concepts in the metaphor, also have to be explained. There is no doubt that these tools are the products of specific histories but they nevertheless have had to be usable by the minds of the human beings who employ them. Here again, the world interacts in a challenging way with the representations that cultural anthropologists study. It is banal to stress merely that the world we live in is culturally constructed; what is of interest is the indirect relation of the construction to what is constructed and how the construction is used.

This chapter, however, will not pursue the implications of the link between the fact that we go in and out of each other’s bodies in birth and sex and the cultural representations of this fact in kinship systems. Many (I do not include myself among them) might feel that this topic has grown tiresome. I merely evoke the controversy to stress that because all cultures interpret, and have to interpret, the fact that we go in and out of each other in sex and birth, they also have to interpret the consequent fact that for us (as with coral) there is indeterminacy concerning the physical boundaries of individuals. For instance, the so-called “descent theorists” of my anthropological youth were fascinated with groups of people who declare themselves to be “one body”; in other words, corporate groups. These statements are interesting not because they are flights of fancy proving yet again that the world we live in is culturally constructed but because they are in part motivated by the very real fact of the indeterminacy and arbitrariness of the boundaries of biological units.

My focus in this chapter concerns another real fact about human beings that, although it concerns a matter different from kinship, is not altogether unrelated to it. Indeterminacy and arbitrariness of boundaries are not simply the result of the sexual character of our species and the way it reproduces itself. They are also due to another feature of Homo sapiens. Individuals go in and out of each other because of certain characteristics of the human nervous system. This form of interpenetration is as material as sex and birth; but unlike sex and birth it is more or less unique to our species (Povinelli et al. 2000; Decety and Somerville 2003).[289]

I have already mentioned that, although the boundaries of individual units are arbitrary among all living forms, this ambiguity takes on a special, perhaps more extreme form in social animals because the social—of itself and by definition— continually reconnects the individuals whom time and genealogical distance are separating. Such a process occurs in a variety of ways in different life forms because of the mechanisms that make the social differ according to the species concerned. So it is not surprising that the specific basis of human sociability is a product of those capacities of our species that make it distinctive (Humphrey 2002).

One thing that normal human babies do at about 1 year old, but our nearest relatives, the chimpanzees, never do, is point at things, not because they want what they designate—they do this, but so do chimps—but because they want the people around them to pay attention to the same things. In other words, they want the people they are with to adjust their minds in harmony with theirs—in short, to share intentionality (Gopnik 1993; Tomasello and Rakoczy 2003; Tomasello 1999). This demonstrative pointing is one of the first stages of the development of that unique and probably most important of human capacities: the ability to “read” the mind of others, a capacity that is somewhat oddly referred to as “theory of mind” (TOM for short). This ability continues to develop from the age of 12 months on until the child reaches the age when it can be shown that the child “knows” that other people act in terms of the beliefs or concepts they hold, rather than in terms of how the world is (Wimmer and Perner 1983). By “know,” I simply mean that the child and, of course, the adult, acts in terms of their reading of the beliefs of alter and continually adjusts her behavior accordingly. I do not mean that the person who does this is necessarily conscious of the process (a point to which I shall return in a moment). The whole process is far too complex and too rapid for that to be possible. Nonetheless, the importance of TOM can hardly be overestimated. Those familiar with Gricean theories of linguistic pragmatics will realize that it can be argued, convincingly in my opinion, that this continual mind reading is what makes linguistic communication, and indeed all complex human communication, possible (Sperber and Wilson 1986).

It is legitimate to think that to talk of the mutual mind reading on which our social life is based is, at best, simply a metaphor; at worst, a mystification. However, I want to stress that the metaphor refers to an empirical phenomenon of interpenetration, even though admittedly we don’t stick our finger into each other’s brains in some kind of mental intercourse.

Just how material the process of mind reading may be has become clearer in the light of recent neurological findings. For instance, many researchers now argue that the unique human ability to read the mind of those with whom we interact is ultimately based on a much more general feature of the brain that is not confined to humans: the so-called “mirror neurones” (Gallese and Goldman 1998).

Perhaps the term is misleading. What is being referred to is an observation that has been made possible by modern neural imagery. The term mirror neurones means that exactly the same neurones are activated in our brains when, for example, we see someone raising their arm to point at the ceiling as when we perform the action ourselves. In other words, the action of alter requires from us a part of the same physiological process: the neural part as the action of ego. Indeed, a moment’s reflection makes us realize that, even without the arcane and somewhat [290]contested biology of mirror neurones, the very nature of human communication must involve something like this (Decety and Somerville 2003).3

Let us consider a simple act of linguistic communication. Here I follow Sperber and Wilson’s theory of relevance fairly closely (Sperber and Wilson 1986). For my message to come across when I say, for example: “Today we honor the memory of Roy Rappaport” a mechanism must occur that enables you to penetrate my brain and align yours so that its neuronal organization resembles mine. In order to do this, we both had to use a tool, sound waves in this case, but it cannot possibly be the sound waves, as such, that carried my meaning to you. Sound waves, poor things, are just sound waves. The reality is that sound waves enable me to modify your brain, or mind, so that its neuronal organization in part resembles mine, admittedly in a very limited way. And, of course, the ability to communicate in this way—to connect our neurones—is what makes culture possible because culture must ultimately be based on the exchange of information. This can then be combined with other information and then transformed or reproduced through time and across space in a uniquely human way.

The parallel neuronal modification implied by communication has further important implications. Let us assume, for the sake of argument, that it is possible for an individual to create ex nihilo a representation. Such a representation could then be said to be under that individual’s control because the process that produced it would be hers alone. However, when the representation comes from someone else’s brain (i.e., when it comes via the process of communication, which is in fact always the case, though to varying extents), the representation of one brain colonizes another. This process, whether it is conscious or subconscious, is the basis of all communication. In such a case, the created neuronal activity of one brain is the material existing in another. By this means, the brains of different individuals interpenetrate materially so that the boundaries that we believe to be obvious become problematic. What I am saying is very similar to what some writers, especially Ed Hutchins, call “distributed cognition” (Hutchins 1995). However, I would distance my argument from them on one minor point. Hutchins, in talking about this phenomenon, likes to refer to minds “not bounded by the skin” as if some sort of extra-biological process existed. I am too literal-minded to feel comfortable with such phraseology, which makes the process in question appear surreal. The process of interpenetration I am discussing is straightforward and biological.

My other difference from the distributed cognition folks is not a disagreement; I simply would like to push their insights further. Hutchins is famous for his demonstration of the way the knowledge necessary to navigate a big ship is not held in the head of any one crew member; it is distributed in a group. In an action such as coping with an emergency, each individual does his job as best he can in the light of his own knowledge, but in doing so he relies on other individuals who have other bits of knowledge necessary to navigate the ship that he does not and does not need to have. This is what Hutchins calls distributed cognition. For this type of reliance [291]on the knowledge of others to be possible, the different individuals need to trust that the others know what they are doing and are well intentioned. This means that people can then act on what they know is incomplete knowledge, but which they trust is completed by the knowledge others have, to the extent of acting on that which they do not need fully to understand. It is not that they rely wholly on others; they rely on others at the very moment they rely on their own knowledge.

By using this particular formulation, I deliberately align what I am saying with the point made by a group of philosophers who, following Hilary Putnam and the “deference” theorists, stress that social life is based on trust of others; basically on the default assumption that these others with whom we are in contact are normally competent and cooperative. In other words, because of our theory of mind adaptation, we continually interpenetrate as we communicate and hold as true information that makes sense only because it is also contained or continuous with that in other minds (Putnam 1975; Burge 1993; Orrigi 2000). This is the nature of human cognition, which is essentially social. Such a state of affairs makes it possible that the content of knowledge stored in an individual is not to be understood nor consciously sought to be understood, but this individual is likely to be aware of the solidarity on which the whole system of social cognition is based, and this may be greatly valued. This is a point to which I shall return.

* * *

I started this chapter by arguing that for all living things, humans included, the distinctness of the units of life is far from clear. Furthermore, I argued that for people this fact is commonly represented in the kinship systems that are about this reality in culturally varied and specific ways. For social animals the problem of the blurring of individual boundaries is compounded by the very nature of their sociality. Individuals in social species are, to varying degrees, materially continuous with each other. Because humans are social animals, this problem applies to them. In their case, this state of affairs is brought about by the tool that makes human sociability possible: the hard-wired human capacity referred to as theory of mind. Such an assertion, however, raises the same question that I touched on in the discussion of sex and birth: What are the cultural implications, if any, of this fact? The necessity to ask this difficult question is precisely what is missing from much of the work of evolutionists such as Tooby and Cosmides (Cosmides and Tooby 1992) and even Rappaport.

The parallel with kinship may help to advance the argument, but at the same time it highlights an obvious difficulty. When anthropologists study kinship systems, they are studying representations of phenomena having to do with obvious empirical processes, of which no one can be unaware: going in and out of each other’s bodies. When we examine the interpenetrations of minds, however, we are dealing with phenomena not so easily consciously perceived. The continual mutual reading of minds on which communication depends is like grammar: it is and has to be subconscious, if only to operate at the necessary speed. But, if that is so, how is it possible that an awareness of this process could occur, a necessary step for it to take explicit form in cultural representations? To approach this question, I ask the reader to accompany me on a detour, away from purely theoretical considerations and toward a brief description of an empirical case.[292]

About a year ago, I decided to do a new (for me) type of field research in the remote Malagasy forest village in which I have been working on and off for nearly 40 years. I carried out what is probably the most typical experiment used to demonstrate the development of children’s understanding of TOM in front of any villagers who were available at the time; I then asked the adults watching to make sense of what they had just seen. By inviting them to give me their interpretation of what was going on, I placed my informants in the same situation as that in which professional psychologists normally find themselves in the lab. The experiment in question is usually called the “false belief task.” In the version I used, I showed a child two hats, and I placed sweets under one of them in the front of the child and everyone else present. I then asked a member of the audience to leave the house and, showing the child what I was doing, I switched the treasure to the other hat. I then asked the child—this is the key question—under which hat the person who had just gone out of the house would look for the sweets when they returned. The results in the Malagasy village were, as expected, much the same as those reported from all over the world. Younger children say that the person who left the house will look under the hat where the sweets actually are, while older children say that the person will look under the hat where he or she saw them put, but where, of course, they no longer are. This difference is usually interpreted by psychologists to mean that the younger child has not yet subconsciously understood that other people do not necessarily know what they know. To put it more theoretically and somewhat differently, the younger child has not yet subconsciously understood that people act in terms of their possibly false beliefs, not in terms of what the world is actually like.

The adult Malagasy villagers’ interpretation of the experiment was not all that different from that of professional cognitive psychologists. After a bit of prodding and reflection, the commonest explanation was that younger children have not yet learned to lie, so they do not understand that other people can also lie. For reasons that I cannot go into here, I take this to mean that the younger children are represented by them as naive empiricists, while they believe the older children and adults know that people can deceive and therefore look for the communicative intention of the speaker because they do not simply trust appearances that could be manipulated by people.

I then used the discussion of the results of this experiment, which had been conducted in front of villagers, as a springboard for a more general discussion about the nature of thought. During these continuing discussions, the villagers explained that thought was an activity through which one matched one’s action to one’s purpose. Thought, they reasoned, is thus a feature of all animals. Fleas, for example, also think because they hide in the seams of garments in order not to get caught. Humans, however, are superior to other animals in that they have an extra tool— language— that enables them to achieve the purpose of their thought more efficiently especially through indirectness and deceit.

When I consider the very detailed information on mind, thought, and cognitive development that I obtained through this work from the largely unschooled Malagasy in this remote village, I am, above all, struck by the familiarity of the ideas they expressed and their similarity with our own folk view. I am also impressed by the correspondence between their views and those of the psychologists. And, [293]indeed, when I look at the few other ethnographic studies of folk theories of mind and thought we possess, I find this general family likeness again and again (Gubser 1965; Rosaldo 1980).4

These similarities inevitably raise the question of what causes such recurrences. The obvious answer is that they are triggered by an awareness of the same actual universal human cognitive process. This explanation, however, runs into the difficulty discussed previously, that mental processes such as the workings of the mind operate below the level of consciousness while what I was told in the discussions that followed the experiments was clearly explicit and conscious.

But is this difficulty as serious as it seems? Or, to put it another way, following the arguments of a number of cognitive scientists (Jackendoff 1987; Block 1993; Humphrey 2002), is the barrier between the conscious and the subconscious as impenetrable as the objection assumes? The comparison with grammar, alluded to previously, suggests otherwise. When we speak to or comprehend others, we do not consciously obey grammatical rules; nevertheless, we can become aware of the existence of such rules when, for example, somebody makes a “grammatical” mistake. Indeed, it is probably as a result of such “mistakes” that folk grammarians the world over can build their theories. Although these folk grammatical theories vary probably because of a great variety of historical and cultural factors, it would surely be perverse not to accept that their obvious similarities are caused by the way grammar actually works and that this can be accessed to a degree.

The situation with theory of mind is probably similar, perhaps also based on reflection prompted by instances of faulty or difficult communication. For example, much of the general speculation about the nature of mind and thought in the data I collected In the Malagasy village was linked to explicit reflection on the abilities and limitations of a co-resident deaf and dumb man. It seems that this sort of more familiar and recurrent event causes the same kind of continual attempt to understand the psychology of thought and communication as was artificially stimulated by my experiment. This is probably why villagers were so willing, enthusiastic even, to engage in the discussion of the experiment I had conducted once their initial resistance was overcome. The intellectual challenge it presented was not as unusual or bizarre as it might at first seem from the outside. Of course, this more ordinary speculation was not done in the jargon of modern psychology, but with the cultural tools available. But even these unsophisticated tools and vocabulary must have been developed in relation to psychological processes that actually occur and are known to occur. It is not surprising, therefore, that similar ideas and representations should crop up, again and again, in different cultural and historical contexts. In making this claim, I am not arguing for any direct determinism between the actual working of the mind and people’s theories about it. Many other factors are clearly involved in each case. The working of the mind is difficult for the Malagasy to understand and represent, as indeed it is for any psychologist. It involves peeping past barriers of many kinds by means of thought or practical experiments, but both parties do this and for neither party is this completely impossible.[294]

To illustrate such complexity and to begin to approach the subject of religion and ritual, I return to my case study.

When the Malagasy villagers so emphatically insisted that thought always, directly or indirectly, was a matter of matching ends and means, I was naturally led to ask them about dreams. Were these not a case of thought without a practical end in view? The commonest answer I was given to such a question was negative. Dreams, I was assured, occurred when other people entered you and thought through you in order to achieve their ends. In this way, local cognitive theory was made coherent with a theory of interpenetration with which I had become familiar when I studied Malagasy ancestor worship. This is because it is through dreams that ancestors manifest themselves most typically, and make their desires known.

This local theory of dreams is radically different from what is found in many other cultures, including that of professional psychologists. This, however, does not mean that as soon as we touch on phenomena that are usually labelled religious, we inevitably move away from concerns cognate with those of professional cognitive science. The idea that dreams are really other people, especially ancestors, thinking through you for their own ends is part of that much more general idea that previous generations, dead forebears, living elders, or absent members of the family are speaking through you as you consciously or subconsciously “quote” them. Not only are you expected to utter the words of other wise people because you trust and rely on them. But a person’s forebears are thought to be continually acting through him or her. Indeed, to allow that to happen willingly is to show respect and to act morally. Morality is thus experienced, less as a matter of individual choice and more as a matter of submission and recognition of the presence of others who penetrate you. As soon as we rephrase the Malagasy concept of ancestors in this (to my mind) ethnographically more accurate way, we find that we have returned to the familiar territory of scientific theories of distributed cognition and deference mentioned previously. In the very area in which my Malagasy co-villagers could be represented as most exotic—notably their beliefs in the power of ancestors—we find them very close to Hutchins and Putnam. Even their belief in the penetration of the young by elders and ancestors turns out to be built on an implicit recognition of the effect of interpenetration made possible by TOM—on the real fact that knowledge is distributed.

The point I want to stress is that the operation of theory of mind and the nature of the distribution of knowledge in society are neither unknown nor fully known by the Malagasy villagers I studied. Furthermore, they are aware of the unsatisfactory partial nature of their knowledge, often commenting on this during the discussions that followed the experiments. And, as a result of their realization of the incompleteness of their knowledge, when the chance arises, as when I showed them the false belief task or when they observed the deaf and dumb man, they eagerly seize the opportunity to find out more about their own and others’ mental processes. In that inquisitiveness they are no different from professional scientists. Like them, their knowledge is incomplete, but also like them they strain to know more about a reality that, in the case of psychological processes, is common to all human beings and partly accessible. Of course, as in the case of the scientists, but probably to a greater extent, there are many other factors that interact with the villagers’ theoretical speculation and representations, and this multiplicity of factors produces systems that are only partly scientifically motivated. However, it is the [295]commonality of the enterprises and the reality of the world they engage with that explains the continuity between scientific discussion of such things as theory of mind and the cultural representations of largely unschooled Malagasy villagers and Western scientists.

The bodily interpenetration of TOM is thus, to a certain degree, known by Malagasy villagers, and this knowledge combines in varying ways and in varying contexts with other types of knowledge. This leads to partial continuities between scientific and folk understandings of the interpenetration of individuals and of the consequent provisionality of levels of individuation. It is to these that I now turn.

A central implication of TOM is that all social relation implies interpenetration, so the arbitrariness of boundaries within the social fabric applies not just to people who are related but also between all human beings who are in contact. Awareness of this ensures that ideologies of individualism are always, to varying degrees, negated by ideologies based on the realisation of interconnection, as Mauss stressed in his seminal essay on the gift (Mauss 1923-1924).

Knowledge of interpenetration and of the lack of clear boundaries, as well as the emotions that are an integral element of the way these phenomena are experienced, is what is meant by that most Durkheimian of words: solidarity. The presence of this sentiment, at its most general, is one that is difficult to put one’s finger on, because it seems rarely made explicit or the subject of reflexive discourse. However, from my reading of ethnography and from my own experience, it would seem that a default assumption in most cultures is that there is a potential moral obligation to any stranger with whom one might come into contact or, to put it in a different way, that the very fact of entering into a relationship implies being consubstantial and therefore morally obligated. Perhaps the most familiar manifestation of this phenomenon is the obligation of hospitality toward strangers, a moral imperative that recurs, admittedly in different forms, in so many unrelated cultures but that, as far as I know, has been little theorized at a comparative level by modern anthropologists. This general unspecific morality is probably an epiphenomenon of the very nature of human communication.

There are, however, many instances of much more specific and elaborate awareness of the lack of boundary between individuals. Many of them seem to fall in the domain that is usually labelled as religion, though some are of a less amiable and more threatening form. I have already mentioned the Malagasy interpretation of dreams and its link with ancestor worship. Ancestor worship is found all over the world in a variety of forms and is often linked to the lack of bodily differentiation within descent groups. Other examples are witchcraft-like ideas that often take the form of a belief in the secret and evil penetration of one’s body by a consuming other made possible by the existence of communication. More obvious still are beliefs in spirit possession that seem to crop up all over the world. These beliefs are an extreme representation of the colonizing nature of social relations because they involve the total invasion and replacement of one individual’s intentional mind by that of another.

In a somewhat different way, the realization of the interpenetration of individuals and of the context dependence of boundaries also seems present in many political movements and religions. The idea of a corporal unity beyond the individual is well documented for certain forms of Christianity, Islam, and devotional Hinduism. They [296]emphasize an alternative “brotherhood” to that based on interpenetration of sex and birth, thereby highlighting the comparability of the two types of interpenetration at the same time as using the one to challenge the other. These ideas become most explicit in the mystical forms of these religions, for example in Sufism or devotional Hinduism, in which the theme of the interpenetration of the bodies of the devotees and the lack of boundaries of their bodies takes an extreme and dramatic form.

Perhaps, however, it is in ritual that the conscious and culturally encoded awareness of lack of boundedness is clearest. This, of course, was one of Durkheim’s key points, but what he stressed was the effervescence of highly dramatic rituals.

There is no doubt that feelings of transcendence of individuality and even of dissolution of self into a greater whole occur in many of the manifestations that we would label as ritual. Furthermore, these may well be part of the realization of the empirical lack of boundary of human individuals. However, many rituals are simply not like that. One universal feature of ritual, however, is deference, if only because it is at the very core of the meaning of the English term. Deference is, as noted previously, the acceptance of the content of other minds without necessarily knowing the whys and wherefores of the propositions and actions one performs. As argued in different ways by Putnam, Burge, and Hutchins, this is characteristic of knowledge in society and implies cognitive interpenetration. Ritual is an extreme case of this. In ritual one accepts that the motivation for meaning is to be found in others one trusts (Bloch 2004). In other words, it is not only that one surrenders one’s intentionality to others but also that one is aware of this happening. Recourse to ritual is therefore to be understood not only as awareness of neural interpenetration, a submission to other minds, but also as a celebration of such awareness.

Of course, religious and ritual representations are not simply realizations of the fact that we interpenetrate each other as we interact and that the boundaries separating individuals are provisional and alterable. In each and every case, much more is involved that might indeed be more important in the particular case. I am simply saying that the social, sexual, and reproductive characteristics of the human species means that we go in and out of each other’s bodies in at least three different ways, and that this implies an indeterminacy of the level of relevant differentiation. In the case of birth and sex, the interpenetration is inevitably, though variously, cognized. In the case of TOM, the matter is more complicated, however. The working of TOM is normally below consciousness as is also the interpenetration it involves. However, because the boundary between the conscious and the subconscious is not sharp and because we have tools to traverse it (such as experiments or the existence of deaf and dumb relatives), we can use our hazy awareness of the process to interpret and speculate about such phenomena as dreams, the relation with ancestors, and many other central aspects of human life. This knowledge—the raw material of interpenetration—becomes a resource and an idiom that can become central in many representations that we would label as moral, religious, or ritual. It is this line of causation from the fact of interpenetration to its conscious representations by different people in different ways that makes it possible to slither from the biological to the cultural, including the religious.

In so far as this causal chain involves a direct connection between the social, the moral, the religious, and ritual, such an argument is inevitably reminiscent of Durkheim’s theories. After all, the central argument of The Elementary Forms of [297]the Religious Life is that religion, by means of ritual, is a projection of the intuition of the dependence of the individual on society, and of the individual’s incompleteness—an intuition that gives rise to the impression of the presence of a superior transcendental element: the religious (Durkheim 1912).

My admiration for this great anthropologist cannot but be heightened by the similarity of our arguments. Much of what I have said is what he said long ago, though from a totally different epistemological base. For this reason, it is also essential to stress the profound difference between his argument and my own, if only to clarify the status of what I have been arguing.

Unlike Durkheim, I am not proposing a general theory of “religion.” Like most modern anthropologists, I do not believe that the term religion has any general analytical value. To seek the essence of religion would, therefore, inevitably run into the circularity for which The Elementary Forms of the Religious Life has been criticized. In any case, awareness of the provisional nature of individual boundaries occurs in many kinds of cultural representations that could never reasonably be termed religious. For the same reason, I am not arguing that the interpenetrations of kinship and TOM are the origin of the religious; any such claim would be meaningless because for me what anthropologists call religion is merely a ragbag of loosely connected elements without a common core.

Most importantly, however, I differ from Durkheim in his understanding of causation. For Durkheim the social, which comes from we know not where, mysteriously causes the cultural, which then gives us the tools to invent what is, irrespective of what the world is like. This idealist fantasy would be worth elaborating only as a quaint example of an archaic conceit if it did not in my opinion still resemble much contemporary anthropological theorizing.

What I am proposing is more straightforward, more modest, more materialist, and anchored in evolutionary theory. The source of the social is to be found in the cognitive capacities of humans, though, of course, the evolutionary line of causation between the social and the cognitive is not unidirectional but rather, as argued by Humphrey and Tomasello, a single process. This socio/cognitive means that, even more than is the case for nonsocial animals and differently from the case for other social animals, the boundaries between human individuals are partial at best. This fact and our consequent bodily connectedness, which supplements and sometimes competes with the connectedness of kinship, are fuzzily available to our consciousness. It is this awareness that becomes a recurrent element in a great variety of representations in different cultures, representations that we must not forget are different kinds of phenomena from the simply psychological. It is these kinds of awareness that Durkheim examined under the label “solidarity.” And, furthermore, the types of solidarity he identified are often, though not always (as he also stressed) manifest in what we call religion and ritual.


Bloch, M. 2004. “Ritual and deference.” In Ritual and memory: Toward a comparative anthropology of religion (eds) H. Whitehouse, and J. Laldlaw, 65–78. Walnut Creek, CA: Altamira Press.[298]

Block, N. 1993. “The computer model of the mind.” In Readings in philosophy and cognitive science (ed) A. Goldman, 819–832. Cambridge, MA: MIT Press.

Burge, T. 1993. “Content preservation.” The Philosophical Review 102, 457–488.

Cosmides, L., and J. Tooby. 1992. “Cognitive adaptation for social exchange.” In The adapted mind: Evolutionary psychology and the generation of culture (eds) J. H. Barkow, and L. Cosmides, and J. Tooby, 163–228. Oxford: Oxford University Press.

Decety, J., and J. A. Somerville. 2003. “Shared representations between self and other: A social cognitive neuroscience view.” Trends in Cognitive Sciences 7, 527–533.

de Waal, F. B. M.. 1996. Good natured: The origins of right and wrong in humans and other animals. Cambridge, MA: Harvard University Press.

Durkheim, E. 1912. Les formes elementaires de la vie religieuse. Paris: Alcan. (Translated 1915. The elementary forms of the religious life. London: Allen Unwin).

Gallese, V., and A. Goldman. 1998. “Mirror neurones and the simulation theory of mind.” Trends in Cognitive Science 12, 493–501.

Gopnik, A. 1993. “How we know our minds: The illusion of first person knowledge of intentionality.” Behavioural and Brain Science 1, 90–101.

Gubser, N. 1965. The Nunamiut Eskimos: Hunters of caribou. New Haven, CT: Yale University Press.

Humphrey, N. 2002. The mind made flesh: Essays from the frontiers of psychology and evolution. Oxford: Oxford University Press.

Hutchins, E. 1995. Cognition in the wild. Cambridge, MA: MIT Press.

Jackendoff, R. 1987. Consciousness and the computational mind. Cambridge, MA: MIT Press.

Mauss, M.. 1923–1924. “Essai sur le don: Forme et raison de l’echange dans les societes archaiques.” L’Annee Sociologique, seconde serie, 30–186.

Origgi, G. 2000. “Croire sans comprendre.” Cahiers de philosophie de VUniversite de Caen, 34, 191–201.

Povinelli, D. J.,, J. M. Bering, and S. Giambrone. 2000. “Towards a science of other minds: Escaping the arguments by analogy.” Cognitive Science 24, 509–554.

Putnam, H. 1975. “The meaning of ’meaning.’.” Minnesota Studies in the Philosophy of Science 7: 131–193.

Rosaldo, M. 1980. Knowledge and passion: Ilongot notions of self and social life. Cambridge: Cambridge University Press.

Schneider, D. 1984. A critique of the study of kinship. Ann Arbor: University of Michigan Press.

Sperber, D., and D. Wilson. 1986. Relevance: Communication and cognition. Oxford: Blackwell.

Stotz, K., and P. Griffiths. 2004. “Genes: Philosophical analyses put to the test.” History and Philosophy of the Life Sciences 26, 5–28.

Tomasello, M. 1999. The cultural origins of human cognition. Cambridge, MA: Harvard University Press.[299]

Tomasello, M., and H. Rakoczy. 2003. “What makes human cognition unique? From individual to shared to collective intentionality.” Mind and Language 18 (2), 121–147.

Wimmer, H., and J. Perner. 1983. “Beliefs about beliefs: Representation and constraining function of wrong beliefs in young children’s understanding of deception.” Cognition 13, 103–128.

Yanagisako, S. J., and J. F. Collier. 1987. “Toward a unified analysis of gender and kinship.” In Gender and kinship: Essays toward a unified analysis (eds) J. F. Collier and S. J. Yanagisako, 14–50. Stanford, CA: Stanford University Press.


Maurice BLOCH is Emeritus Professor of Anthropology at the London School of Economics and Political Science and the author of hundreds of articles and many books. He has carried out fieldwork among irrigated rice cultivators and shifting agriculturalists in Madagascar, and in other parts of the world including Japan. Partly because of his French background he has combined British and French approaches and was instrumental in introducing the revival in French Marxist theory to British anthropologists. His interests have focused on the notion of ideology and he has written on ritual and language. He is now working on how to relate the findings of cognitive psychology with anthropology.


Editors’ Note: This is a reprint of Bloch, Maurice. 2007. “Durkheimian anthropology and religion: Going in and out of each other’s bodies.” In Religion, anthropology, and cognitive science, edited by Harvey Whitehouse and James Laidlaw, 63–88. Ritual studies monograph series. Durham, NC: Carolina Academic Press. We thank the author for granting permission to reprint this essay and we remind the reader that we retain the style of the original.

1. An earlier version of this was given to the American Association for the Anthropology of Religion as a Rappaport lecture. I would like to thank R. Astuti, E. Keller, G. Orrigi, A. Yengoyan, and D. Sperber for comments on an earlier version.

2. They are what would be used to dismiss as irrelevant studies such as those of Cosmides and Tooby (1992) about cheater detection.

3. It is also important to remember the importance of sharing of emotions, which is highly relevant to the argument of this chapter and goes in the same direction as the evidence on TOM. It is not considered here, but I hope to do so in another publication. See de Waal 1996.

4. Rosaldo’s book, in fact, emphasizes the exotic character of Ilongot psychology, but I am struck that, in matters of cognition at least, Ilongot conceptualization is very familiar.