2

TO INNER EXPERIENCE

Those on the New Left who are in revolt against empty rituals do not readily see themselves walking in the footprints of Wycliffe and of ardent Protestant reformers. Yet if we can make the leap from small exotic cultures to our European religious tradition, we can make the easier transition between anti-ritualism in a secular and in a religious context. We are then able to see that alienation from the current social values usually takes a set form: a denunciation not only of irrelevant rituals, but of ritualism as such; exaltation of the inner experience and denigration of its standardized expressions; preference for intuitive and instant forms of knowledge; rejection of mediating institutions, rejection of any tendency to allow habit to provide the basis of a new symbolic system. In its extreme forms anti-ritualism is an attempt to abolish communication by means of complex symbolic systems. We will see, as this argument develops, that it is a viable attitude only in the early, unorganized stages of a new movement. After the protest stage, once the need for organization is recognized, the negative attitude to rituals is seen to conflict with the need for a coherent system of expression. Then ritualism re-asserts itself around the new context of social relations. Fundamentalists, who are not magical in their attitude to the Eucharist, become magical in their attitude to the Bible. Revolutionaries who strike for freedom of speech adopt repressive sanctions to prevent return to the Tower of Babel. But each time this movement of revolt and anti-ritualism gives way to a new recognition of the need to ritualize, something has been lost from the original cosmic ordering of symbols. We arise from the purging of old rituals, simpler and poorer, as was intended, ritually beggared, but with other losses. There is a loss of articulation in the depth of past time. The new sect goes back as far as the primitive church, as far as the first Pentecost, or as far as the Flood, but the historical continuity is traced by a thin line. Only a narrow range of historical experience is recognized as antecedent to the present state. Along with celebrating the Last Supper with the breaking of bread, or the simplicity of fishermen-apostles, there is a squeamish selection of ancestors: just as revolutionaries may evict kings and queens from the pages of history, the anti-ritualists have rejected the list of saints and popes and tried to start again without any load of history.

But swings of the pendulum do not take us far enough in the interpretation of anti-ritualism. There is still the long secular trend to be accounted for which has resulted in a lack of sensitivity to condensed symbols, and at the same time a general preoccupation with lack of meaning. The move away from ritual is accompanied by a strong movement towards greater ethical sensitivity. Thus we find Christian denominations in the United States less and less distinguishable from one another and from the Jewish community, less and less willing to refer to doctrinal differences, and all equally committed to programmes of social betterment. This trend has been well described by Herberg in his Protestant, Catholic, Jew (1960) and documented by intensive research by Neal (1965). To understand it, however, I find myself drawn to the work of another sociologist whose research is specially relevant.

Ritual is pre-eminently a form of communication. Socio-linguistics provide us with an angle of approach. Basil Bernstein is a sociologist whose thought descends through Durkheim to Sapir (Bernstein, 1965: 148). His special concern is to discover how speech systems transform the experience of speakers. By a line of inquiry as subtle in its perception as it is powerful in scope he seeks to apply Sapir's insight about the controlling influence of language on culture.

It is quite an illusion to imagine that one adjusts to reality essentially without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the real world is to a large extent unconsciously built up on the language habits of the group.... We see and hear and otherwise experience very largely as we do because the language habits of our community predispose certain choices of interpretation.

(Sapir, 1933: 155–69)

This present book is an essay in applying Bernstein's approach to the analysis of ritual. It will help us to understand religious behaviour if we can treat ritual forms, like speech forms, as transmitters of culture, which are generated in social relations and which, by their selections and emphases, exercise a constraining effect on social behaviour. Even when we have summarized a little of what Bernstein has said, and then applied it to ritual as a medium of communication, we are still a long way from using his insights to understand anti-ritual. To this I hope to return in the last chapter.

Bernstein very cogently distinguishes his argument from that of Whorf (1941) and others who have treated language as an autonomous cultural agent and failed to relate its formal patterns to the structure of social relations. Indeed, before Bernstein it was difficult to see how such a relationship could be established. For certainly, in large areas of its internal development, language follows its own autonomously given rules. It is not surprising, as he has alleged (1965), that contemporary sociologists often seem to ignore the fact that humans speak, unless the sociologists are specially concerned with speech, in which case they emphasize its integrating or divisive functions. Speech tends to be treated as a datum, something taken for granted. If it is true that the analysis of speech as a social institution (one as basic as family and religion) has scarcely been broached, anthropologists are in no way to feel smug about the analysis of ritual. They do not make the mistake of neglecting this field; nor do they suppose that ritual is merely divisive or integrative in social relations. The data are piled up in great stacks of analysis of particular tribal symbolic systems which express the social order. But why some tribes should be pious and others irreverent or mercenary, why some are witch-ridden and others not, are questions which have only been entertained in sporadic fashion. As for the deeper question of whether symbolic forms are purely expressive, merely ‘the means of solving specific problems of communication or reflection’ as Sapir put it, or whether they interact on the social situations in which they arise, and whether their effect is constraining and reactionary – these questions are not systematically approached. Still less have anthropologists developed a frame of discourse in which their tribal studies can be related to ourselves. This is the point at which a revolutionary insight into language as a social process can help us.

Bernstein starts with the idea that there are two basic categories of speech, distinguishable both linguistically and sociologically. The first arises in a small-scale, very local social situation in which the speakers all have access to the same fundamental assumptions; in this category every utterance is pressed into service to affirm the social order. Speech in this case exercises a solidarity-maintaining function closely comparable to religion as Durkheim saw it functioning in primitive society. The second category of speech distinguished by Bernstein is employed in social situations where the speakers do not accept or necessarily know one another's fundamental assumptions. Speech has then the primary function of making explicit unique individual perceptions, and bridging different initial assumptions. The two categories of speech arise in social systems which correspond to those which Durkheim indicated as governed by mechanical and organic solidarity. So Bernstein would deserve the attention of anthropologists if only because he is sympathetic to a Durkheimian sociology of knowledge, one which was originally worked out by comparing ritual as a medium of communication in tribal and in industrial society. He says that:

different speech systems or codes create for their speakers different orders of relevance and relation. The experience of the speakers may then be transformed by what is made significant or relevant by different speech systems. As the child learns his speech, or, in the terms I shall use here, learns specific codes which regulate his verbal acts, he learns the requirements of his social structure. The experience of the child is transformed by the learning generated by his own, apparently voluntary acts of speech. The social structure becomes in this way the sub-stratum of the child's experience essentially through the manifold consequence of the linguistic process. From this point of view, every time the child speaks or listens, the social structure is reinforced in him and his social identity shaped.

(Bernstein, 1970: 124)

He distinguishes two different types of linguistic code. One he calls the elaborated code, in which, as he says, the speaker selects from a wide range of syntactic alternatives, which are flexibly organized; this speech requires complex planning. In the other, which he calls the restricted code, the speaker draws from a much narrower range of syntactic alternatives, and these alternatives are more rigidly organized. The elaborated code is adapted to enable a speaker to make his own intentions explicit, to elucidate general principles. Each type of speech code is generated in its own type of social matrix. As I understand it, the differences between the two coding systems depend entirely on the relation of each to the social context. The restricted code is deeply enmeshed in the immediate social structure, utterances have a double purpose: they convey information, yes, but they also express the social structure, embellish and reinforce it. The second function is the dominant one, whereas the elaborated code emerges as a form of speech which is progressively more and more free of the second function. Its primary function is to organize thought processes, distinguish and combine ideas. In its more extreme, elaborate form it is so much disengaged from the normal social structure that it may even come to dominate the latter and require the social group to be structured around the speech, as in the case of a university lecture.

It is essential to realize that the elaborated code is a product of the division of labour. The more highly differentiated the social system, the more specialized the decision-making roles – then the more the pressure for explicit channels of communication concerning a wide range of policies and their consequences. The demands of the industrial system are pressing hard now upon education to produce more and more verbally articulate people who will be promoted to entrepreneurial roles. By inference, the restricted code will be found where these pressures are weakest. Professor Bernstein's research in London schools and families finds that the codes are instilled into children from their earliest infancy by their mothers. Each speech system is developed in its corresponding system of family control. He asks mothers of working-class and middle-class families how they control their children under five; what happens if the child won't go to bed? won't eat? breaks the crockery? From their detailed responses he constructs a distinctive pattern of values, a distinctive concept of the person and of morality.

Let me describe the two kinds of family role system. Restricted codes are generated in what he calls the positional family. The child in this family is controlled by the continual building-up of a sense of social pattern: of ascribed role categories. If he asks ‘Why must I do this?’ the answer is in terms of relative position. Because I said so (hierarchy). Because you're a boy (sex role). Because children always do (age status). Because you're the oldest (seniority). As he grows, his experience flows into a grid of role categories; right and wrong are learnt in terms of the given structure; he himself is seen only in relation to that structure. The child's curiosity in working-class or some aristocratic families is harnessed to the task of sustaining his social environment. Let me quote briefly from Bernstein himself. Differences in speech are taken to be

indices of a particular form of communication; they are not in any sense accidental but are contingent on a form of social structure. These differences I shall argue indicate the use of a linguistic code. It is a (restricted) code which does not facilitate the verbal elaboration of meaning; it is a code which sensitizes the user to a particular form of social relationship which is unambiguous, where the authority is clear-cut and serves as a guide to action. It is a code which helps to sustain solidarity with the group at the cost of verbal signalling of the unique difference of its members. It is a code which facilitates the ready transformation of feeling into action. It is a code where changes in meaning are more likely to be signalled non-verbally than through changes in verbal selections.... How does this way of translating experience come about? What in the culture is responsible for the speech system? . . . Different social structures will generate different speech systems. These speech systems or codes entail for the individual specific principles of choice which regulate the selections he makes from language at both the syntactic and lexical level. What the individual actually says, from a developmental perspective, transforms him in the act of saying.

As the child learns his speech or in our terms learns specific codes which regulate his verbal acts he learns the requirements of his social structure. From this point of view every time the child speaks the social structure of which he is a part is reinforced in him, and his social identity develops and is constrained. The social structure becomes for the developing child his psychological reality by the shaping of his acts of speech. If this is the case, then the processes which orient the child to his world and the kind of relationships he imposes are triggered off initially and systematically reinforced by the implications of the speech system. Underlying the general pattern of the child's speech are initial sets of choices, in-built preferences for some alternatives rather than others, planning processes which develop and are stabilized through time – coding principles through which orientation is given to social, intellectual and emotional referents.

(Bernstein, 1964: 56–7)

When a child learns a restricted code he learns to perceive language in a particular way. Language is not perceived as a set of theoretical possibilities which can be transferred into a facility for the communication of unique experience. Speech is not primarily a means for a voyage from one self to the other. In as much as this is so then areas of the self are not likely to be differentiated by speech and so become the object of special perceptual activity. It is also likely that the motivations of others will not serve as starting points for inquiry and verbal elaboration. Of some importance, the identity of the individual will be refracted to him by the concrete symbols of his group rather than creating a problem to be solved by his own unique investigations.... A critical aspect of the family is the means of expression of authority, particularly the type of verbal interaction authority relationships create. I shall argue that associated with parents limited to a restricted code is a specific form of authority relation. Authority can be expressed so as to limit the chances of verbal interaction with the relationship, or authority can be expressed so as to increase verbal interaction. The area of discretion available to the child may be reduced to an uncompromising acceptance, withdrawal or rebellion within the authority relationship, or the social context of control may permit a number of responses on the part of the child.... If the appeals are status-oriented then the behaviour of the child is referred to some general or local rule which constrains conduct, ‘shouldn't you clean your teeth’, ‘you don't behave yourself like that on a bus’, ‘children in grammar schools are expected to behave rather differently’. Status appeals may also relate the child's behaviour to the rules which regulate his conduct with reference to age, sex or age relationships, e.g. ‘Little boys don't play with dolls’, ‘you should be able to stop doing that by now’, ‘you don't talk to your father, teacher, social worker, etc., like that’. These are important implications of status appeals. If they are not obeyed the relationship can quickly change to reveal naked power and may become punitive. Status appeals are impersonal. They rely for their effectiveness upon the status of the regulator. The effect of these appeals is to transmit the culture or local culture in such a way as to increase the similarity of the regulated with others of his group. If the child rebels he is challenging very quickly the culture of which he is a part, and it is this which tends to force the regulator into taking punitive action.

(ibid.: 59–60)

By contrast, in the family system which Professor Bernstein calls personal a fixed pattern of roles is not celebrated, but rather the autonomy and unique value of the individual. When the child asks a question the mother feels bound to answer it by as full an explanation as she knows. The curiosity of the child is used to increase his verbal control, to elucidate causal relations, to teach him to assess the consequences of his acts. Above all his behaviour is controlled by being made sensitive to the personal feelings of others, by inspecting his own feelings. Why can't I do it? Because your father's feeling worried; because I've got a headache. How would you like it if you were a fly? or a dog? The child tends to be controlled by person-oriented appeals:

In these appeals the conduct of the child is related to the feelings of the regulator (parent) or the significance of the act, its meaning is related explicitly to the regulated, to the child, e.g. ‘Daddy will be pleased, hurt, disappointed, angry, ecstatic if you go on doing this’. ‘If you go on doing this you will be miserable when the cat has a nasty pain’.... Control is effected through either the verbal manipulation of feelings and through the establishment of reasons which link the child to his acts. In this way the child has access to the regulator as a person and he has access to the significance of his own acts as they relate to him as consequences.... The status-oriented appeals rely for their effectiveness upon differences in status whereas the person-oriented appeals rely more upon the manipulation of thought and feeling.

(ibid.: 60)

In this way the child is freed from a system of rigid positions, but made a prisoner of a system of feelings and abstract principles. The personal system of family control is well adapted to develop verbal skills: the child will do better in school examinations as a result of his control of the elaborated code. He may shoot forward to the top of the wider society, become Prime Minister, Head of UNO; the sky's the limit. Underlying this family system is anxiety about the child's development and educational success. It is probably not inspired by ambition. More likely it is inspired by the knowledge that in a changing world the only ticket anyone can hold for staying in a privileged niche is education. The child is being educated for a changing social environment. As his parents move from one town or country to another in response to the need for professional mobility, the child grows in a family system which is relatively unstructured, a collection of unique feelings and needs. Right and wrong are learnt in terms of his response to those feelings. Instead of internalizing any particular social structure, his inside is continually stirred into a ferment of ethical sensibilities. We can immediately and from our own experience recognize this as the basis for the move from ritual to ethics. There is no need to indicate the clichés from the literary and philosophical output of the last 100 years which validate the system.

To sum up Basil Bernstein's approach in diagrams: in diagram 1 the horizontal arrow expresses the way that patterns of family control are progressively detached from the immediate social structure of the family and local community and progressively co-ordinated with the demands of the wider industrial social structure (Bernstein, 1970). Diagram 2 studies the effect of the same industrial pressures upon speech. Verbal communication is progressively detached from its service to the immediate social context and elaborated for its use in the widest social structures of all.

Diagram 1 Family control

In the process, note that as speech sheds its social harness, it becomes a very specialized, independent tool of thought. Basil Bernstein has plausibly suggested that the emancipation of speech from social control underlies some variations in religious worship. Diagram 3 is the result of our discussions together. It is very impressionistic and designed more than anything to help follow in imagination the kind of transitions that can be studied in this framework.

Admittedly, there are several difficulties about this diagram. To understand it we should look first at squares A and B. A represents most primitive cultures in which speech forms are firmly embedded in a stable social structure. The primary use of language is to affirm and embellish the social structure which rests upon unchallengeable metaphysical assumptions. In such a system we would expect to find that the admired virtues are those which unquestioningly uphold the social structure, and the hated sins are transgressions against it. Since individual motivation is irrelevant to the demand for performance, we would expect to find little reflection on the notion of the self; the individual is hardly concerned as a complex agent. On the contrary, the self is seen as a passive arena in which external forces play out their conflicts. This would be the social structure to give rise to totemic thought systems and to art forms which celebrate social dichotomies and confrontations. In these the relation of the individual to society is hardly considered. This general class emerges as appropriate wherever literacy is low and the social structure stable.

Diagram 2 Speech codes

Diagram 3 General cosmological ideas

In square B, speech and thought have been elaborated as specialized tools for decision-making, but the social structure still retains a strong grip on its members, even to the extent that its underlying assumptions are not challenged. Elaborated speech in this case is still in the service of the social structure, but uses the philosophical reflections at which it has become adept for examining and justifying those assumptions. This would be the square for Aristotle. The result of this reflection of speech and thought on the social structure would be an awareness of the demands of the latter upon the individual and of the possibility of the individual not responding adequately. Truth and duty would be the primary virtues. They express the confidence that the social structure rests upon a rational foundation which justifies its claim to allegiance. As a result of the capacity for reflection and as an expression of the new independence of thought, we would expect the self to be accorded a much more active role. The danger of the individual rejecting the claims of society would here be recognized, though condemned. Does it help to suggest that classical drama, Oedipus and El Cid, depicts these attitudes?

In squares C and D the social structure has lost its grip. Square C, according to Bernstein, is unstable, a transitional phase. For example a mother belonging to the professional classes by her own education and aspirations, married into a working-class environment, might bring up her children by the techniques of personal control, but through the rest of their social relations they would be obliged to use speech of the restricted code. Here the individual is valued above the social structure; hence the literature of revolt, Rimbaud, or D. H. Lawrence.

In so far as there is a literature for this square, we have to assume that individuals reared in it have made in their lifetime the move from C to D, have become verbally articulate in elaborate codes.

We can understand square D most easily, for it includes ourselves. I cannot go further without trying to be more specific about who, in our contemporary society, fits into squares B and D. What is the distribution of people using elaborated speech codes between positional and personal family control systems? Start with square B. The positional family develops on the assumption that roles should be defined clearly and the elaboration of speech, in so far as it is used to sustain role patterns, reduces ambiguity. Here we would expect to find aristocracies whose aspirations are relatively fixed and whose role structure is clearly ascribed. Also certain sections of the middle class will be here. The military profession, for example, demands unambiguous allocation of roles; the legal profession lives by reducing role ambiguity. There are other highly educated sectors of modern society whose profession encourages them to favour positional control systems. The work of engineers, concerned primarily with abstract relations between material objects, does not lead them to use the elaborated code to reflect critically on the nature of social relations. That they should tend towards positional family systems becomes clearer when we see how the other square, D, is filled. Here are the people who live by using elaborated speech to review and revise existing categories of thought. To challenge received ideas is their very bread and butter. They (or should I say we) practise a professional detachment towards any given pattern of experience. The more boldly and comprehensively they apply their minds to rethinking, the better their chances of professional success. Thus the value of their radical habit of thought is socially confirmed, and reinforced. For with the rise to professional eminence comes the geographical and social mobility that detaches them from their original community. With such validation, they are likely to raise their children in the habit of intellectual challenge and not to impose a positional control pattern. How much more likely are they to prefer personal forms of control if the area of their professional thinking deals with human relations: psychologists, anthropologists, novelists, philosophers, political scientists. The professions which deal with the expression of personal feelings rather than with abstract principles are also found here. This is the square in which ideas about morality and the self get detached from the social structure. This would be the niche in which to consider Existentialism and the deep pre-occupation of our day with the technical process of artistic creation.

The positional child, who knows the pattern in which he belongs, cannot understand the anguish which Sartre has described so poignantly in Words, the biography of his first ten years of life (1967). In the home of his domineering and histrionic grandfather he and his widowed mother were appendages, serving mainly to provide the grandparents with emotional satisfaction. From infancy he was tormented with the consciousness that his existence had no justification. There was no society organized in general categories of age, sex and hierarchy in which his developing role could be seen as necessary to some overall pattern. Only the promise of personal success, the renown of future genius, could justify his life in the patternless adult world which made such unconvincing pretence of valuing him for his loveable personality. This dominant anxiety of childhood is clearly related to his later philosophical position. Bernstein suggests that problems of self-justification arise in the personal family – and was not the Reformation about self-justification?

Some may argue that Bernstein's contrast of positional with personal family is no different from the old distinction between achieved and ascribed status. This is an error. The positional/personal contrast rides at a higher level of abstraction. In the personal family, true enough, all roles have to be achieved, but the converse does not hold good. In some positional families, within the ascribed framework, important roles have to be achieved. For example in military families there is a strong emphasis on achievement. The contrast of achieved/acquired status needs to take account of the different realms of achievement.

For reasons which may now be clear, it is easier for us (writers and readers) to recognize and sympathize with the aspirations of one raised in a personal family than with the person raised in the positional family. Certainly I believe that the Bog Irish have suffered from such a blank in the imaginative sympathy of their pastors, whose personal outlook conforms to their special niche in the professional ladder. The latter might take a more generous view of tenacious ritualism if they saw the impoverished power of response to condensed symbols of all kinds which lies in the direction they are leading and if they could value ritualistic forms of commitment as such. It must be difficult for the child reared on abstract principles in the personal family to draw moral lines, to be bound by promises – for unquestioned boundaries have never been part of his upbringing. The child in the positional family grows into a set of unchallenged categories, which are expressed by non-verbal symbols as well as by words.

Bernstein's work on the social structuring of speech in London families challenges the anthropologist in many, very difficult, fields. At first sight all ritual would seem to be a form of restricted code. It is a form of verbal utterance whose meanings are largely implicit; many of them are carried along standardized non-verbal channels. Indeed, since Malinowski no one has thought to interpret the language of magic apart from the symbolic actions and apart from the whole social context. Ritual is generally highly coded. Its units are organized to standard types in advance of use. Lexically its meanings are local and particular. Syntactically it is available to all members of the community. The syntax is rigid, it offers a small range of alternative forms. Indeed so limited does the syntactic range tend to be that many anthropologists find that a simple binary analysis is sufficient to elucidate the meanings of myth or ritual symbols. Bernstein himself has suggested that his definitions should be applicable to other symbolic forms – he has suggested music (1965: 166). He has also recognized that in any one case elaboration and restriction will be relative. Obviously there are technical difficulties in applying this comparison of speech and ritual forms. However, at first glance we seem to have a ready-made solution to our opening question. The causes of anti-ritualism today in middle-class European and American communities would appear to be a predictable result of a process of socialization in which the child never internalizes a pattern of social statuses and never experiences authoritative control which exalts the self-evident property of a social system to command obedience. Symbols of solidarity and hierarchy have not been part of his education. Consequently a form of aesthetic experience is closed to him. As Bernstein emphasizes:

It is important to realize that a restricted code carries its own aesthetic. It will tend to develop a metaphoric range of considerable power, a simplicity and directness, a vitality and rhythm; it should not be disvalued. Psychologically, it unites the speaker to his kin and to his local community.

(ibid.: 165)

It is tempting to equate the restricted code with ritualism and to leave the matter there. In many anthropologists’ accounts of a pastoral or hunting economy there appears the outline of an ox or wild pig or antelope quartered and subdivided, with a legend indicating the category of kin to which each segment is allocated. Such a chart showing the correct distribution of game or of sacrificial meat summarizes the main social categories. Similarly the first fruits celebrations of agriculturalists. Each feast reaffirms the categories visibly and publicly. Primitive rules of purity also support the social categories and give them external, physical reality. Clearly the words which accompany these distributions carry a small part only of the significance of the occasion. The comparable situations in family life would be the spatial layout of chairs in the living room which convey the hierarchy of rank and sex, the celebration of Sunday dinner, and for some families, presumably those in which a restricted code is used, every meal and every rising, bathing and bed-time is structured to express and support the social order. Bernstein's fully personal family, then, would be one in which no meals were taken in common and no hierarchy recognized, but in which the mother would attempt to meet the unique needs of each child by creating an entirely individual environment of time-table and services around each one of her brood; early supper for this one to go to choir practice, late supper for that one coming back from an excursion, hospitality for that other one's friends and so on; food selections too would be on an individual basis. How could such a child ever learn to respond to a communally exerted authority? His ears would not be attuned to catch the unspoken messages of a restricted code. Hence some of the deafness and antipathy to ritualism in our day.

This would be fine and an end of the argument if, as was commonly held, all primitive peoples were ritualist and if the movement away from magicality were indeed able to be plotted along a graph showing more and more the effects of the division of labour on family behaviour. But I have mentioned already the unritualistic pygmies; then there are the Basseri who have so little that can be called religion, the Anuak who are much more interested in counteracting witchcraft than in worshipping God or indulging in metaphysical speculations, and probably a host more of so-called primitive tribes who share with the most industrially advanced nations a lack of interest in ritual. We shall need to look closely at the social structures of these tribes, to find a set of variables which will be consistent both with the Bernstein effect among ourselves and with what is known about primitive social structure and cosmology. This exercise will take us a long way from Bernstein's analysis, but I intend to return to it at the end. No doubt readers are more interested in themselves than in exotic tribesmen and I regret that I am not able to develop more fully the parallel between the positional home and the primitive ritualist. Let me at least include some suggestions thrown out by Bernstein (in a personal communication) on the types of religious behaviour he would expect to be associated with types of family control. The primitive ritualist, in his ascribed social system, expresses cosmic orientations and moral directives in condensed symbols. The home which is organized around positional values has comparable methods of explanation and control. In a community composed of such homes God would be also known through the restricted code. Theological concepts about him would not be fully elucidated, he would be known by his attributes as manifested in the social structure. Knowing God would be subject to the same restrictions as knowing the mother: the code of speech would not provide means for reflecting upon or inspecting the relationship verbally. The religious cult would be expected to correspond in style to the family rituals and therefore to be fixed and ritualistic. Similarly the definition of sin would be more concerned with specific external actions than with internal motivation.

In diagram 3 the idea of the self was progressively detached from the social structure. So, as ritualism declines, the Idea of God becomes more intimate. But as God comes nearer he is diminished in glory and power. This hypothesis can be recognized as thoroughly Durkheimian. For the cosmology, based on its particular hierarchy of values and upholding a particular pattern of behaviour, is derived from society. As the grip of his immediate society on the individual tightens or slackens something happens to his religious attitudes.

There is an awkward paradox in this presentation. For as a Londoner gets drawn more and more into the vortex of industrial society his religious ideas seem to approximate more and more to those of the pygmy. He believes in spontaneity, friendship, freedom, and goodness of heart: he rejects formality, magic, doctrinal logic-chopping and condemnation of his fellow humans for their wrong-doings. This paradox is due to a distortion in the comparison caused by the effects of the division of labour. Pygmies cannot be equated with preachers, journalists and dons. The argument will have to go a long way before we can pick up this paradox and resolve it. In the meanwhile, note what the Bernstein effect amounts to. As a result of definable pressures on home and school there is an increasing tendency to rear children by personal, elaborated speech code methods. This produces a child acutely sensitive to the feelings of others, and interested in his own internal states. It follows that such an education will predispose a person to ethical preoccupations, for while it opens up his vocabulary of feeling it also denies him any sense of pattern in his social life. He must therefore look for some justification of his existence outside the performance of set rules. He can only find it in good works on behalf of humanity in general or in personal success, or both. Hence the drive towards a purely ethical religion.