Ritual, History and Power

Selected Papers in Anthropology

Maurice Bloch

• Publisher: Routledge
• Year: 2020
• Print ISBN: 9781845200213
• eBook ISBN: 9781000324549
• Edition: 1
• Topic: Social Sciences
• Subtopic: Philosophy History & Theory
• Index: Social Sciences

This volume provides a collection of some of Maurice Bloch's most important work, including influential essays on power, hierarchy, death and fertility.

1 The past and the present in the present¹

This paper starts by considering the old problem of how to account for social change theoretically and criticizes some of the models used because either they see the social process in terms used by the actors and so are unable to explain how it is that actors can change those terms, or they see the mechanisms of change as occurring in terms totally alien to the actors and so are unable to explain how these mechanisms can be transformed into meaningful action. The source of this problem is traced to Durkheim's notion that cognition is socially determined. By contrast it is argued that those concepts which are moulded to social structure are not typical of knowledge but only found in ritual discourse, while the concepts using non-ritual discourse are constrained by such factors as the requirements of human action on nature. This means that there are terms available to actors by which the social order can be criticized since not all terms are moulded by it. Finally it is suggested that such notions as social structure only refer to ritualized folk statements about society, statements expressed in ritual discourse precisely with those concepts which are given as demonstrations of the theory of the cultural relativity of cognition. The Durkheimian correlation between society and cognition is merely a correlation of only certain ethical statements and certain aspects of cognition. This type of discourse is present in different types of society in varying amounts according to the degree of instituted hierarchy that these societies manifest. Anthropological theories about the conceptualization of time are given as an example of the general argument.

In this paper I want to follow Malinowski in two ways: first in his style of argument which, as I see it, is based on the belief that one might as well hang for a sheep as for a lamb, and second by using Malinowski's highly realistic view of the anthropologist's subject matter as a tool for criticizing other theories. For Malinowski what was to be studied was a long conversation² taking place among the people with whom we live during fieldwork and in which we inevitably join: a long conversation where not only words are exchanged but from time to time also things, animals, people, gestures and blows, but where nonetheless language plays a most prominent part. For him everything was to be found there, in that conversation. His view of the theoretical importance of the past followed naturally from this. On the one hand, since the past cannot be seen in this ongoing conversation, it has no explanatory value, and on the other hand, when it does appear in discourse as a subject matter it has to be explained in terms of the present.

The first of these conclusions — that the past has no explanatory value — is clearly wrong. The long conversation which the anthropologist observes has begun long before he came and indeed it has begun long before any of the people the anthropologist meets have been born. As in all discourse these observed statements and acts must be related both semantically and in terms of shared communicative conventions such as words, syntax, etc. to those that have preceded them and so, if only for this, the present cannot be understood apart from the past in that it answers it. However, a semantic relation to past utterances is not the only requirement of meaning, and Malinowski was right in stressing something other anthropologists often forget, that what is said must also be adapted to the tasks in hand. So even if we reject Malinowski's general view about the past we can nonetheless ask with him what explains the appearance of the past as a subject matter in the present?

First, however, I want to look once again at Radcliffe-Brown's theory of social structure. A denunciation of this has become an essential part of all public lectures in anthropology but what concerns me here is that the very frequency of these denunciations itself raises interesting problems. Why, if Radcliffe-Brown's theories were so wrong, were they so evidently fruitful? Or, to put the matter more specifically, if what he was talking about was not, as I think is now clear, a science of society and culture, what was it and why cannot he be dealt with once and for all? I want to answer this last question first, leaving the earlier and more fundamental one until later.

Radcliffe-Brown inherited two key propositions from Durkheim and these are what concern us here: 1) that society is a homogeneous, organized and self-reproducing entity; 2) that the categories of understanding and systems of classification are social in origin: that is, that constructive influence on them comes from such things as the form of social groups and the linkage which exists between them, and not from constraints from the extra-social world. Now although this second proposition concerning the social determination of cognition is the more adventurous it has hardly been theoretically challenged by anthropologists. The first, on the other hand, that society is an organic, harmonious, rule-governed static system, has come in for continual criticism. Actually most of the critics of the organic view of society have accepted the Durkheimian point about cognition, but I want to show that the organic view of society is implied by the notion that society determines cognition and that it is because of the acceptance of this latter point that the criticisms of the static nature of social structure fail.

The reasons why the theory of the social origin of cognitive systems has gone unchallenged are not at all clear, but it is in part because this theory is linked to a belief adhered to by most anthropologists: that different cultures or societies have fundamentally different systems of thought. Durkheim himself used, as part of evidence against the view that cognitive systems were primarily constrained by nature, data which showed that different peoples had different ideas of such things as time, space, animal species, causation and so on (Durkheim, 1912). Since we had different systems of thought, he argued, but all lived in the same world, the differences must come from society. Similarly American cultural anthropology has inherited, via Boas, the theories of German romantics like Herder: that every people have their own proper view of the world (Lowie, 1937). Equally, from the left, the Durkheimian theory of the origin of cognitive system has gone largely unchallenged. This is because many versions of Marxism, forgetting Marx's own distinction between ideology and knowledge,³ also rely on the Hegelian notion of the relativity of cognition. This is especially true of some of the recently influential Althusserian versions. Given such broad agreement it is not surprising that the view that cognitive systems are socially determined has again been recently powerfully put by such varied but influential writers as Lévi-Strauss (1962a), Douglas (1966; 1970; 1975), Geertz (1973) and Godelier (1973), to mention only anthropologists.

If the view of the social determination of cognition and classification is everywhere, so have for quite a time been criticisms of the notion of social structure. These all focus on the point that although social structure is claimed to be a theory of society, it only concentrates on very limited aspects of the natural phenomena, and that with its emphasis on the reproduction of the system it fails to account for change and conflict.

Thus criticisms of Radcliffe-Brown's position have focused on two points: the first is that it is not true that societies stay the same, and that therefore some room has to be made in the theory for change, and the second is that rules of behaviour, since they are not necessarily followed, are not 'all the story'. In most cases these two points are linked, and the solution offered to these failings in Radcliffe-Brown's notion of social structure is the construction of a two-level model of society which incorporates in a variety of ways the Marxist-inspired distinction between superstructure and infrastructure. The clearest of these theories is perhaps Firth's distinction between the level of social organization and the level of social structure (Firth, 1964). For Firth social structure is much what Radcliffe-Brown meant by the phrase, while social organization is the pattern produced by people following or not following the rules of social structure. Systematic disobedience of these rules leads to social change at the level of organization and Firth suggests that somehow things can reach such a point that changes become necessary in the social structure. Such a formulation seems at first sight to do what was intended, that is to modify the original theory of social structure so that it can account for change, but that is an illusion. The reason lies in the fact that the level of organization, the presumed source of change in the social structure, is contained within the level: social structure. The level of organization can only be apprehended in terms of the social theory of the actors; their system of social classification and rules, which is what is referred to by the phrase 'social structure'. Organization is a matter of following or not following rules; rules which apply to roles recognized by the people studied. This means that within such a theoretical framework, although deviance is accounted for, it is not possible to understand how the rules and the social categories which give deviance meaning can, themselves, be changed, since they are given in the very language within which social organization is discussed. This kind of difficulty also exists in the many similar theoretical formulas associated with such writers as Gluckman and his many followers. This is the problem which Leach's formulation in Political Systems of Highland Burma tries to overcome, though in the end he too comes up against the same problem. In order that the rules will not contain the range of possible actions, he suggests that we should have three levels not two; 1) a level of shared meanings common throughout the area he studies; 2) a level of rules which are not necessarily consistent one with another and which are chosen ad hoc by the actors in terms of a third level; 3) a level of enlightened self-interest very similar to Firth's 'social organisation'. In this way Leach is able to account for changes in rules. While, for Firth, choices are decisions about whether to obey or disobey rules, for Leach choices are concerned with which rule to obey. Leach, however, has to face the problem that the actors must be able to communicate among themselves. Clearly they cannot choose any system because then their actions would stop being meaningful one for another, and so Leach shows that all these varying rules embody the same meaningful categories, understanding of which is shared in the whole geographical area. This does avoid the difficulty concerning communication but it brings us back in a different way to the earlier problem which we saw in Firth's theory. Instead of actions being bounded by the meanings given to them by rules they are bounded by the meaning given to them by the shared concepts. Thus, for the same reason that Firth's theory cannot account for the creation of new rules, Leach's theory cannot account for the creation of new concepts.

This seems at first sight a strange problem because it is difficult to see why some of the actors at a certain point in the social process cannot say: this social system is no good at all, let us take a fresh look at the situation and build up a new system. The reason why they cannot, within the theoretical framework discussed, lies in the unanalyzed notion of the social determination of thought. Simply if all concepts and categories are determined by the social system a fresh look is impossible since all cognition is already moulded to fit what is to be criticized.

Strangely the problem also exists in a theory which seems to do precisely what is needed since it offers to the actors a source of knowledge which can be used to challenge the social order. This is a theory expressed by a varied group of writers, mainly French, who have drawn their inspiration from Marx (Meiliassoux, 1972: Terray, 1969: Godelier, 1966; 1973). This position would criticize Leach and Firth in the way outlined above, and as a solution propose that the infrastructure be constructed in a way that is totally external to either rules or concepts, in terms of the rationale of the processes of production and reproduction. History could then be seen as the interaction of two levels of different nature, neither one being reduced to the other, so that a continually progressive dialectic could exist between them. This kind of theory would therefore truly achieve the elusive goal of a dynamic system which takes into account the shared system of meanings of people without its movement being caught within it. The problem there, however, comes from the fact that the infrastructure is seen as external to the concepts of the actors. Now for it to be a source of criticism of the social order it means that people must apprehend it in terms available to them and which are different from and incompatible with those of the dominant social theory. This means terms not determined by it. Otherwise the infrastructure, however contradictory to the dominant social theory, is never transformed into action and just carries on in its own sweet way, totally irrelevant to the processes of history.

One can put the problem generally by saying that, if we believe in the social determination of concepts, as all the writers mentioned so far tacitly or explicitly do, this leaves the actors with no language to talk about their society and so change it, since they can only talk within it. This problem explains why anthropologists are continually producing pictures of society similar to those of the historians who so well explained the logic of the feudal system that they also explained why peasants' revolts could not occur. It also explains why anthropologists' work leaves us totally unprepared for the dramatic and revolutionary changes which are occurring in the very areas they have studied.

So what evidence is there for the crucial and apparently testable proposition of the social determination of knowledge from recent studies of cognition? Well, anthropological studies of cognitive systems at present seem strangely contradictory. On the one hand we have the work of such writers as Lévi-Strauss, Geertz, Douglas and Willis, which stress the variation in systems of classification of such things as animals, plants, colours, and which, in various ways, link these systems of cognition to social structure. On the other hand we also have completely different studies associated principally with the work of Kay (1975), Berlin (1972), Berlin and Kay (1969), Berlin et al. (1973) and Bulmer (1967; 1968; 1970) which come up with totally different findings: basically, that colour, plant, animal and even human classifications are based on identical criteria and produce identical classes and sub-classes varying only in degree of elaboration. I believe the contradiction between these two types of studies can be explained, and it is to this that I now turn.

First let us look at what is probably the most fundamental claim, repeatedly made by cultural relativists - that concepts of time are closely bound to social organization and therefore vary from society to society. This is not a topic that, as far as I know, has actually been examined by Berlin and Kay or their associates. But social scientists as varied as Durkheim (1912), Boas (1966), Lévi-Strauss (1962a, 1966), and Evans-Pritchard⁴ have loved to tell us that the notion of time, which we feel is self-evident, can be experienced in other cultures in totally different ways, not as linear but perhaps as static or as cyclic. This is a really popular claim to make among anthropologists, if only for the reason that if it, and all it implied, were true, all academic subjects, especially the better funded ones such as physics, should really become simply a sub-trade of anthropology. This indeed was almost the position of Whorf (1956) but it is also rather carelessly implied by many others.

In one sense at least what they say is true: that is, if the claim about the relativity of concepts of time is upheld, it is so fundamental that it inevitably justifies the conclusion that all aspects of culture are relative. However, even before we look at this proposition critically, something ought to make us suspicious. In its full baroque form the relativists' argument will have us believe that we can produce a whole range of different concepts of time for different cultures. However, an examination of this 'range' reveals that it by and large boils down to only two notions of time. On the one hand we have concepts rather like our own folk everyday concept of linear durational time, and on the other hand a concept of a static notion of time often referred to as cyclic, the two words referring to the same sort of evidence. Before proceeding, however, one point must be made. In reducing the evidence to two types I am, of course, talking about claims concerning the perception of duration not the ways in which time is divided up, or metaphorically represented. These are, of course, legion but are not relevant to our argument. The Malagasy used to, and still sometimes now, divide the day in terms of the parts of the house reached by the rays of the sun. This works because of the strict orientation of their houses, but it only tells us that they use different types of clocks from us. This is not what I am talking about. Let us return to the more fundamental claim that different people perceive time differently. First of all there are a priori arguments against this type of claim which have been formulated in varying ways by, amongst others, Gellner (1968) in answer to Winch, Max Black (1959) in answer to Whorf, and in philosophy by Ayer (1973) in answer to the New Hegelians. The most recurrent such point is contained in Wittgenstein's famous remark that 'if lions could speak we could not understand them':⁵ In other words, that communication with creatures with a fundamentally different system of ideas and life is not possible, and surely people with a different concept of time would in this respect be like lions, since everyone agrees about the particularly fundamental nature of this proposition. On the other hand, the existence of anthropology itself bears witness to the fact that it is possible, if with certain difficulty, to communicate with all other human beings, however different their culture. Wittgenstein's zoology brings to mind another remark by a supporter of cognitive relativity and shows its naivety: that is, the criticism made by Evans-Pritchard of anthropologists who tried to understand other cultures in terms which made sense to them. He ridiculed this type of reasoning by describing it as 'if I were a horse' arguments; the implication being that for the anthropologist to pretend to reconstruct the thought processes of other people is as ridiculous as trying to reconstruct the thought processes of horses.⁶ But surely, there is no reason to believe that if horses could speak, we would understand them any better than lions, while Evans-Pritchard's whole work is a demonstration that, with help from the anthropologists, we can indeed understand the Azande or the Nuer. This is possibly because of a fact that Evans-Pritchard seems to have overlooked: that neither he, nor other anthropologists, study horses. In other words, if other people really had different concepts of time we could not do what we patently do, that is communicate with them. Evidence for such a conclusion also comes from a completely different source, and that is the mass of recent studies of syntax and semantics of different languages that have been carried out by American linguists. Disagreements and polemics in this field are many, but at least consensus seems to be emerging on one point, and that is that the fundamental logic employed in the syntax of all languages is, Whorf notwithstanding, the same. The implications of this for notions of time are clear. The logic of languages implies a notion of temporality and sequence and so if all syntax is based on the same logic, all speakers must at a fundamental level apprehend time in the same way...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Contents
• Acknowledgements
• Foreword
• 1 The past and the present in the present
• 2 Symbols, song, dance and features of articulation: Is religion an extreme form of traditional authority?
• 3 The disconnection between power and rank as a process: an outline of the development of kingdoms in central Madagascar
• 4 Marriage amongst equals: an analysis of the marriage ceremony of the Merina of Madagascar
• 5 From cognition to ideology
• 6 Hierarchy and equality in Merina kinship
• 7 Descent and sources of contradiction in representations of women and kinship
• 8 Almost eating the ancestors
• 9 The ritual of the royal bath in Madagascar: the dissolution of death, birth and fertility into authority
• Notes
• Bibliography
• Index