Chapter Four
Chapter Four
Grid and group
It is illuminating to consider ritual as a restricted code. But more problems arise in applying this insight than I am ready to handle. Bernstein argues that the restricted code has many forms; any structured group that is a group to the extent that its members know one another very well, for example in cricket, science or local government, will develop its special form of restricted code which shortens the process of communication by condensing units into pre-arranged coded forms. The code enables a given pattern of values to be enforced and allows members to internalize the structure of the group and its norms in the very process of interaction. Much of the writings and conference proceedings of anthropologists, or of every other body of scholars, would have to be classed as ritualistic or restricted code in so far as the citing of fieldwork, the reference to (often impossible) procedures, the footnotes etc., are given as pre-coded items of social interaction. Allegiances, patronage, clientship, challenge of hierarchy, assertion of hierarchy and so on, these are being obliquely and silently expressed along the explicit verbal channels. If this is so, then Bernstein, by working within the broad framework of a dichotomy of restricted and elaborated codes, is at the stage of Durkheim when he distinguished mechanical and organic solidarity, or of Maine, distinguishing societies governed by contract or by status. As he himself says, the distinction between restricted and elaborated codes must be relative within a given culture or within the speech forms of a given group. Thus the question of whether there are primitive cultures in which all speech is in the restricted code is meaningless, since it ascribes absolute value to the definition. Bernstein would suppose that in any social group there are some areas of social life more responsible for policy decisions and more exposed to the need to communicate with outsiders. Therefore in any tribal system he would expect to find some people who had been forced to develop a more elaborated code in which universal principles can be made explicit and meanings detached from a purely local context. I am not convinced of this myself. If the situations requiring policy decisions were only part of a repetitive cycle it would be possible to discuss them fully in terms of preorganized units of speech. Only the need for innovation in policy would call forth the effort to use an elaborated speech code. This question poses intriguing problems of method for the ethnolinguists. But it is not central to my theme. More pertinent is how to use the idea of the restricted code to interpret different degrees of ritualization.
If ritual is taken to be a form of restricted code, and if the condition for a restricted code to emerge is that the members of a group should know one another so well that they share a common backcloth of assumptions which never need to be made explicit, then tribes may well vary on this basis. One can well suppose that the pygmies might never get to know each other very well. Their social intercourse might be likened in intensity and structure to that of the provisioners of a French seaside resort who move down from Paris in June to open their shops and hotels for the tourist season. They know each other quite well, there is a field of common assumptions to be sure, but it by no means exhausts their interests. They could be expected to develop a restricted code with reference to their local concerns. So we can also suppose that the pygmies and the Persian nomads who join their respective hunting or pastoral camps for a season and may not necessarily be together for next year, use one restricted code for those of their common concerns to which an enduring social structure corresponds, and variant forms of restricted code for communication within their own families. This analogy from speech codes suggests good reason for the poverty of ritual forms in the two cases. It fits the Durkheimian premise that society and God can be equated: to the extent that society is confused in its structure of relations, to that extent is the idea of God poor and unstable in content.
The restricted code is used economically to convey information and to sustain a particular social form. It is a system of control as well as a system of communication. Similarly ritual creates solidarity and religious ideas have their punitive implications. We would expect this function to be less and less important the less is effective social coherence valued. We cannot therefore be surprised that the pygmies have not developed the punitive aspects of religion. They are content with a minimal level of organization. Here again a range of comparison is suggested which would predict something about the presence and absence of ritualism in human societies. We need some way of comparing the value set on organization and social control. It is all very well to illustrate my theme by references to exotic tribes. At some point the problem of comparison must be brought under control. Not only is it dubious practice to compare preachers with pygmies. It is just as dubious to compare hunters with pastoralists, or hunters in Africa with hunters in Australia. I will try to control this problem of cultural variation by staying as much as possible within a given culture. But first, the task is to adapt Bernstein's diagram of systems of family control. It was designed to reflect the increasing influence of the division of labour in industrial society on two variables, speech and techniques of control. Our first step, then, is to eliminate the effect of the division of labour by choosing slightly different variables. Since Bernstein's work relates to the structure of London families it is concerned with personal face-to-face relations. Consequently it needs very little adaptation for tribal society. His two lines measure different aspects of what he calls positional behaviour in families. Where the division of labour has least effect, the speech code and the control system support a differentiated structure of relations in the family. If we want to follow his work closely we must first do violence to the subtlety of his thought. In Bernstein's diagram, speech codes respond to the pressure exerted from the decision-making areas at the centre-top of industrial society for more and more verbal articulateness. Family control systems respond to the same set of pressures demanding children capable of mastering intellectual abstractions concerning human behaviour. His quadrant was designed to show how the two responses are not produced in the same combination in all sectors of industrial society. The area of maximum structuring of social relations in the family is on the left: the area of maximal openness and freedom from structuring is on the right. In the bottom right the individual emerges as free as possible from a system of socially structured controls. His diagram illustrates some effects of a single pressure to move from the positional to the personal control system. The vertical line expresses changes in the use of speech. It shows the possibility of speech being used as an intensifier of positional control, with this possibility diminishing as the central pressure to be intellectually, verbally and symbolically free of the local positional structure develops. The people who have been freed most completely from structured personal relations are among those most involved in the complexity of modern industrial structure. Inevitably this model has to be dismantled to be adapted to tribal society. In what follows we are working with only a very crude and limping parody of his idea.
The task can be simplified if we recall what it is essentially that he is doing. He is deriving cosmology from control systems, or rather showing how cosmology is a part of the social bond, according to the following principles. First, any control system, since it has to be made reasonable (be justified, validated or legitimated as Weber put it), must appeal to ultimate principles about the nature of man and of the cosmos. This applies even at the family level. Second, that the control system interacts with the media of control (speech, ritual). Third, that certain consistencies hold between the coding of the medium and the character of the control system. That they should match is a long-run prediction. In a short run the transition process might obscure the match. Our task starts therefore by identifying the control aspects of the cosmology.
Somewhere far away from the level of the English family and home, some machinery is grinding out a set of social pressures. Naked power is decently clothed and made legitimate. Its demand to be made legitimate reaches into the most intimate recesses, even into the dealings the English mother has with her own child. She learns to assert her control in certain ways and to justify her authority by reference to general principles. The child is thus indoctrinated into the assumptions of his society. His curiosity is checked or roused, his expectations for himself are set in the most hidden way-not by the overt doctrines handed out, but by what is left implicit. Bernstein exposes two implicit world views carried in our styles of speech. He finds them generated in two distinguishable systems of control. To match his exercise, we should look to systems of control and the hidden assumptions which validate them. We are not ready to deal with how the media vary. This is for the next chapter. So, leaving aside speech codes, at this stage I would need to produce a comparison of control systems which will contrast an entirely personal form of relationships, unstructured by fixed principles, with a system equivalent to his positional family. We can concentrate, it seems, upon the interaction of individuals within two social dimensions. One is order, classification, the symbolic system. The other is pressure, the experience of having no option but to consent to the overwhelming demands of other people. Consider order first. Social relations demand that categories be clarified and orientations given. Order is the basic requirement for communication. It could conceivably be possible to compare symbolic systems according to the clarity of definition given to the categories used. There is a hint of such a programme in the first pages of Primitive Classification.
For us, in fact, to classify things is to arrange them in groups which are distinct from each other and are separated by clearly determined lines of demarcation .... At the bottom of our conception of class there is the idea of a circumscription with fixed and definite outlines. Now one could almost say that this conception of classification does not go back before Aristotle. ... Not only has our present notion of classification a history, but this history itself implies a considerable pre-history. It would be impossible to exaggerate, in fact, the state of indistinction from which the human mind developed. Even today a considerable part of our popular literature, our myths, and our religions is based on a fundamental confusion of all images and ideas. They are not separated from each other, as it were, with any clarity. ... If we descend to the least evolved societies known, those which the Germans call by the rather vague term Naturvölker, we shall find an even more general mental confusion.
The authors go on to compare this weakness of definition to the growth of consciousness in the individual from childhood to adulthood: distinctions when they first appear are fragmentary and unstable; only gradually does a steady circumscription of elements of experience lead to classification. However this is not the basis for a comparison of classification systems which I propose to use here. I shall take it as axiomatic that the clarity of bounding of different categories within the total system does not vary, or that, if it does become fuzzy here and rigid there, this is not a difference which I wish to take into account. I shall instead try to compare the overall articulation of the categories which constitute a world view. A classification system can be coherently organized for a small part of experience, and for the rest it can leave the discrete items jangling in disorder. Or it can be highly coherent in the ordering it offers for the whole of experience, but the individuals for whom it is available may enjoy access to another competing and different system, equally coherent in itself, from which they feel free to select segments here and there eclectically, not worrying about the overall lack of coherence. Then there will be conflicts, contradictions and uncoordinated areas of classification for those people. In effect, loss of coherence results in a narrowing of the total scope of the classification system. We can therefore take the scope and coherent articulation of a system of classification as one social dimension in which any individual must find himself. I shall call it grid.
As Durkheim himself has powerfully argued, any given classification system is itself a product of social relations. Bernstein's example of the positional family above shows people putting pressures on one another in terms of classifications. When the pressures are strong and when they uphold a set of classifications, then a process of mutual reinforcement is at work. Such a social system is likely to remain stable, unless counter-pressures develop from outside or unless new knowledge weakens the credibility of the classifications. In either case, the social change will be wrought in the other dimension, that of action or pressure. To draw the dimension of grid vertically up from zero towards more and more comprehensive articulations allows us to consider what absence of classification would mean. The zero would represent a blank, total confusion with no meaning whatever. Rulelessness could be anomy, the suicide's doubt. It could be the mystic's moment of dissociation when all classifications are in abeyance. It could also represent, as the quotation from Primitive Classification suggested, the child's first undifferentiated awareness. To distinguish among these possibilities a little, let us separate the publicly accepted classification system from the private one. An increasingly coherent but entirely private system of classification would point away from communication with others, eventually to madness. This world of private thought we draw downwards from zero.
On the horizontal axis, draw pressure, increasing from zero to the right. At zero, no demands are made on the individual. He is free of pressure. This means he is alone. But another case would have to be located on the vertical line. When pressures and counter-pressures completely balance out, the point of indecision would be recorded here. It is the moment before conversion and commitment. Towards the right he is increasingly under the bond of other people. For reasons which will be clear later, I call the tendency towards the maximum personal control the line of group. A child's life starts far along that line (since he is completely controlled by others) and low on the line of grid: as he grows he may be progressively freed from personal pressures and progressively indoctrinated in the prevailing classification system. If he is clever at internalizing the categories and their implications, he can turn them to his own defence against personal tyranny. He can even use them to tyrannize. To allow for this, we can extend the horizontal line from zero to the left. On this side the individual has escaped pressure from other people. He is exerting pressure on them.
Although the public grid of classifications is used by other people to control the individual, he can evade it if ever the insulation breaks down. The mutual reinforcing of grid and group keep the system stable only if it is perfectly insulated. But perfect insulation is rare and there is some scope for change. We have now in hand a device which could consider social change as a dynamic process. We can see the individual under strong pressure to accept a system of classification which degrades him and commits him to a life of servitude. We could assess the other options open to him, and the relative weight of competing pressures. But this is not the exercise I am attempting. Our problem is to find some relation between cosmological ideas and characteristics of social relations. I shall argue that several systems which spread in different patterns across the diagram are liable to develop recognizable trends in the way that the universe is constituted. The first task is to investigate more closely the properties of the chart.
Above the horizontal line is the area of public classification. The social system will always be centred here. Close to the line and below it lie the fringe elements, the marginal sectors of society: the more to the right they are found, the weaker their option not to be exploited by others to the left operating the public system of classifications; towards the left and zero are the voluntary outcastes, tramps, gipsies, rich eccentrics, or others who retain their freedom, at a cost. This line across the page separates the area of conformity from innovation. Given the way we have defined the vertical dimension we are not suggesting that anyone is dreaming up new conceptual systems from scratch. What is private and innovatory is the way the common cultural categories are articulated. Progress further down that line to greater coherence of the private philosophy depends on an accompanying isolation from social pressures. Beyond a certain point of originality the thinker can give up any reasonable expectation of his ideas being received. This follows from the relation between grid and group above the horizontal line of zero. The framework of institutional life and the distribution of power is the result of a long-term adaptation between social pressures and classification. The big push that changes classification must be big enough to redistribute power as well. To the far right the fringe area of private thought is socially null. It is under more pressure than it can exert. Far to the left it is in high public esteem: hordes of people to the far right would be applauding each new impulse emanating from the far lower left. It is worth pausing to consider how a person can be located in that quadrant. A musician can innovate, a painter, inventors and writers too. If his idea be ignored, he is still on the right. For most of his long working life the Flemish painter James Ensor endured that fate and revenged himself on the public that denied him honour by cruel caricatures. If successful, though, the innovator may see the public system of classification change in his own lifetime. If he wants to stay original, he will have to keep thinking of something new to surprise them with, or devise a technique for maximizing the unexpected, as John Cage has done for his music. To remain free of the public system of classification, the person needs above all not to covet its rewards. Every glance he cocks towards the prize-giving juries makes him vulnerable to their criticisms and liable to be sucked into the general grid. Thus, though it is difficult to stay there, it is possible for people to pass in their lifetimes through different points in the bottom left quadrant.
The bottom right quadrant can be filled for infancy. Here is the personal family in which the child is controlled by being made sensitive to an inventory of his parent's aches and pains: no public system of classification is used to explain the universe and his place in it, but (theoretically at least) he is taught to develop a classificatory system of his own. However, publicly known categories implicitly underline their behaviour and he is quick to deduce them as he grows older. It is surely impossible for an adult to accept heavy social pressure and yet to develop a privately articulated philosophy. If he wishes to have intellectual privacy he must inevitably achieve a solitary state, and so the tendency would be for such a person to move across from right towards the vertical line of no control.
This leads us to considering further the relations between the distribution of power and the coherence of public classifications. It is axiomatic that a steady pattern of control is needed for a coherent system of classification. The more distinguishable places in the control system and the more these are co-ordinated into a lasting hierarchy of responsibilities, the more the public classification system differentiates its categories. So a society spread across the diagram from the highest point mid-way between public grid and group on the right and across to some high point mid-way at the top left is a complex social system. Time depth and corporate institutions are implied in that pattern. Conversely, one political shake-up makes many classifications irrelevant and drags down the coherence of the symbolic system; the expectation of continuing change sets the level lower still. A society that is spread across the diagram at a low level of classification is likely to become one that is continually subject to political upheaval and a changing profile for the distribution of authority. This will be important for our theme.
We should now examine the different ways these dimensions organize our material. Some tribal systems will be spread mostly through the top right-hand side of the diagram without showing on the left. A classic instance of high classification which anthropologists would recognize are the Tallensi of the Volta Region of Ghana as described by Meyer Fortes in the colonial period. Here the public system of rights and duties equips each man with a full identity, prescribing for him what and when he eats, how he grooms his hair, how he is buried or born. Most Tallensi, probably all, are under pressure from the others. The chiefs and priests are no exception. The person whose soul is in revolt is regarded as abnormal and needing special ritual curing. In this society piety is the order of the day, piety towards senior kinsmen and piety to the dead, even though the ancestors are seen as aggressive punishers. The only enemy is the rank outsider, bound by no ties of clanship. A few miserable old women outlawed as witches are either hounded from village to village or merely tolerated. Who knows their thoughts? If they are totally mystified by the public grid which rejects them we could locate them below the horizontal line, though far to the right where options are weakest. I will argue later that a social system characterized by high classification would display the same cosmological bias. Strong grid and strong group will tend to a routinized piety towards authority and its symbols; beliefs in a punishing, moral universe, and a category of rejects.
Any bureaucratic system which is sufficiently secure and insulated from criticism will tend to think the same way. This is the monastic life, or the military society. Most clearly it is the stable tribal system discovered by anthropologists in Africa in the colonial era just before and after the Second World War. It is no accident that a functional analysis produced an equilibrium model of primitive society at that time. For the colonial regime itself provided the insulation and protection from the effects of war and famine. It tended to freeze the native social systems into patterns of reinforcement and stability.
However, the effect was not the same in Central Africa as it was in Ghana. Here the long nineteenth-century wars with Arabs and other slave-raiders had already broken up the local social structures before the colonial freeze came down. The tribes in the region around Lake Nyasa are very differently characterized in the writings of the nineteen fifties when labour migration, cash crops, and taxation accelerated the process of change. Here we also find small communities. But as to grid they come much lower on the line of coherent classification than the Tallensi. Their culture promises them contradictory rewards and holds out impossible goals. They believe that it is good to be loyal and obedient and never to split a village into factions. They also believe that the proper ambition of every man is to become head of his own village-impossible without disloyalty and friction. They put immense pressure on one another and strive incessantly to define and close the circle of their friends. Accusations of witchcraft are the political idiom of outcasting and re-definition of social boundaries. The broad, normative concept of a human being for whom moral obligations are binding is contrasted with that of the man-eating witch. To convict a rival of witchcraft is to finish him politically. This is the second of the main types of social environment I shall refer to throughout the book. For convenience I shall call it small group. It is a social system which clusters low on the right side of the diagram. Its members know one another and can count their ranks and prospects of promotion. They are not conscious of remote control by leaders located far to the left. Hemmed in and face to face, their destiny is in their own hands and they meet it with intrigue and jealousy. The contrast of small group with the previous case of high classification ranges over many aspects. For example, high classification requires a well-defined category of rejects and anomalous persons. But small group broadens the category of potential rejects to include the whole range of acquaintance, male, female, kinsman and unrelated.
The third type to distinguish from these two is the society which spreads widely across our diagram instead of being tidily clustered on the right. The leaders in the small-group case are down in among their community, struggling against their peers. In this third case which I shall call strong grid, the leaders are remote powerful beings, rarely seen face to face. We shall need to deal separately with the social environment of the leaders and of those who are subject to them.
During the colonial period, for reasons we have suggested, anthropology was much concerned with the properties of corporate groups and with rights and duties transmitted down enduring channels of control. Colonialism itself checked internal evolution and limited tribal political systems to the mere replacement of personnel in a fixed pattern of office. But research in newly independent countries, and above all in newly discovered New Guinea, has focused attention on what is called the network of links a man has to a circle radiating out from himself. In a complex society, networks are the minimum level at which social relations can be investigated. They are the sustaining base line of social ties from which corporate institutions arise. But if corporate organization is so weak that each man has to muster support ad hoc for every venture, a system of networks and temporary action-sets may describe the way the whole society functions. Philip Gulliver has summarized ably the problems of description and analysis which the anthropologist faces in such societies. I wish to concentrate on one of two possible variations of the network. In the case which Philip Gulliver himself describes, of the Ndendeule in Tanzania, no person stays in any position of eminence over others, there are no chiefs and there are no effective boundaries to the spread of the open network in all directions. For each man the meanings of society are centred upon himself but the meanings are the same for him as for others. By contrast, in other variants, it is possible for leaders to become effective and to entrench their power in their lifetime at least. Such a leader will gather his own network of allegiances powerfully round himself and create a centre of reference for the rest of society. The Big Man system, as it is called in New Guinea, is found all over the world, in Indonesia, among northern Californian Indians, in the Philippines. I take it for my fourth social type, to contrast with the other patterns. Its interest is the widespread across the diagram of grid and group at a low level of classification. Success breeds. There are few overriding community interests to check the leader's impetus. The greater his influence, the more support he attracts. A positive feedback propels him further out to the left; it increases the subjection of his followers, so they move to the right. If his success in wealth and war encourages him, he may end by eroding their existing system of obligations and become a law unto himself. Then the inevitable trend would be to lower the level of classification for everyone in his orbit. He has made their lineages and ancestral shrines less meaningful for them than his own favour. The big categories however, solvency, worth, equity, remain as containers into which a changing synthesis of meaning is poured from year to year. However, the leader has to reckon with rivals creeping near to him in eminence. The world of his peers is a sparse and fluctuating scene of coalitions. Each is bent on success. If they are realistic, their followers recognize that right goes with might and line up accordingly. This type of social system in its various stages has now been frequently and well described. There are many more examples from New Guinea. The interesting difference between them is the range within which the competition of Big Men must use existing corporate institutions or can override them and in doing so attaches large parts of the public system of classification to the whims of the Big Men themselves.
We have now distinguished three types of social environment: high classification, small group, strong grid which includes the heroic society of competing Big Men and that of their followers. The latter come low on the vertical line of classification because coherence is achieved only at some very general level of abstraction which is compatible with the syncretizing rivalry of distant giants. But the spread across the diagram expresses the strong control which these people experience. Recruited and harnessed to a competition which seems to hold glittering rewards for all, they find themselves trying to work a complex system of rules. In the name of the rules the Big Men justify their demands. Whether it be rules of monetary exchange, debt and credit, or rules of etiquette and hospitality, the system constitutes an oppressive grid. Londoners too know what this can mean. As a system of control industrial society is impersonal. Some more than others feel their lives controlled, not by persons, but by things. They wander through a forest of regulations, imponderable forces are represented by forms to complete in triplicate, parking meters, inexorable laws. Their cosmos is dominated by objects of which they and fellow humans are victims. The essential difference between a cosmos dominated by persons and one dominated by objects is the impossibility of bringing moral pressures to bear upon the controllers: there is no person-to-person communication with them. Hence the paradox that some of the people whose metaphysics are most fuzzy and who respond only to very diffuse symbols-in short, who in their cosmology are most like pygmies and Arizona peyotists-are those who are much involved in certain sectors in industrial society. To this paradox I shall return.
For the leaders who have spiralled down far to the left the same impersonal rules of exchange are made like rungs on a ladder of promotion. The Big Men live in a world of noble pacts, hard bargains, dastardly betrayals and revenges. Apart from the exotic cases given, there are examples nearer home. Our ancient Anglo-Saxon vengeance and inheritance laws defined a set of responsible kin radiating from each particular individual. The Norse sagas expressed a corresponding world view.
With these four social types distinguished, it will be possible to show that they generate distinctive cosmologies. The system of control is validated by a typical bias in the system of belief. These tendencies are the subject of this book, for they make their own typical demands on the media of expression and thus produce natural systems of symbolic behaviour. A brief summary of the types of belief would go as follows. With high classification, piety and sacralized institutions, strong boundaries between purity and impurity; this is the prototype, the original Durkheimian system in which God is Society and Society is God, where all moral failings are at once sins against religion and the community. With small group there is less confidence in the power of God to protect the faithful, a dualist cosmology reckons with the power of demons and their allies; justice is not seen to prevail. Strong grid tends to a pragmatic world view, sin is less understood than shame for loss of personal honour, face, or solvency. In the first type a profit and loss calculus applies to the spiritual economy of the whole community; strong grid focuses on the honour of the individual, the number of supporters he can summon up, the control he has over his women folk. Strong grid divides between the heroic society of the Big Men and the recurrent millennial tendencies of their subjects. Finally, the positions near and around zero should be specially noticed. When public classification and pressures are withdrawn or cast aside, the individual left alone with himself develops a distinctive cosmology, benign and unritualistic.
As Durkheim suggested, this experience is the beginning of consciousness, with all the emotional force that that implies. The sense of escape from others and of self-discovery is possible with any shift towards the left of the diagram. Out here, especially below the vertical line, where the individual is articulating his own classificatory system, the thinker does not see his fellow human beings as the principal determinants of social life. Fellow humans do not put their imprint on the world as models of controlling influence. In consequence, the cosmos is not anthropomorphic. There is less call for articulate forms of social intercourse and no need for a set of symbols with which to send and receive specific communications. Thus, we have already identified one area of the diagram in which there will be less regard for ritual. Furthermore, it suggests another dimension which is not on the diagram, that lying between density and sparsity. When populations are sparse and social relations infrequent, interrupted, and irregular, a person does not have the impression of inhabiting a man-dominated world. What preoccupations about his fate he may entertain concern drought, pasture, livestock, movements of game, pests, or growth of crops. He is controlled by objects, not persons. Objects do not respond to personal modes of approach. Fellow humans are fellow sufferers.
It is tempting to try to assimilate whole cultures to the general outlook of individuals dropping to near zero. But sparsity conceals too many variables; better to stick to those on the diagram. There is ample material there for explaining the similarity between the world view of pygmies in the Ituri forest and that of certain Londoners deeply implicated in industrial society. First, we should turn to the media of social relations. If the pattern of social relations put their stamp upon speech forms, as Bernstein's work shows, they no doubt put a pattern upon non-verbal forms of communication as well. If the speech forms thus produced themselves control the kind of social responses possible in a given social environment, we should expect the usage of the body for communication to exert a parallel constraint.