1
The social determination of ritual
Social science originates in the radical idea thai society and culture are natural, not God-given, phenomena and, therefore, are governed by general laws of an earthly character. Many thinkers and writers directly and indirectly related to Ihe mainstream of social science have refused this starling point, but the attempt has always failed to change this basic premise, because the denial of the natural basis of culture and society negates the essential prerequisite for the idea of social science. Social science is therefore a materialist study in that it assumes that social phenomena are, in the end, to be explained as natural phenomena and it studies the factors creating society and culture. This is not to say that non-human material phenomena such as the properties of plants, animals and the soil and of man as a non-social being are what determine the ideas of human being. Such a view of materialism is a simple misunderstanding, making it patently false, since the influence of ideas on history is clear all around us. Rather, materialism sees that ideas also are. in the end, products of a complex yet natural process taking place in history.
AH social science, therefore, attempts to explain the way the mechanisms of production and reproduction continuously produce social and cultural phenomena. Many social scientists would shy away from such a bold and bald statement of their position» partly because they may not want tobe reminded of the impious basis of their endeavour, but also for two better reasons. The first is theoretical: They want to avoid the impression that they have already found what they are looking for. The second reason has two sides, one factual (he other ethical. Explaining other peoples1 beliefs and feelings can look like a refusal to accept the personal value of the emotions of joy and sorrow, even of illumination, of the participants in the practices analysed, and this is especially so in the field of religion. This is a particular danger in a study such as this, which detects in religious practices the expression of domination and militarist violence. Yet not to recognise the power and value of much of what is done for the participants is simply to introduce factual inaccuracy. When such an omission is used to distance ourselves from the people whose lives we have briefly shared in field-work and, as a result, a way in which wc deny the continuity between them and
I
us» il becomes morally objectionable. Only a few anthropologists, such as V. Turner, have had (he courage lo make clear the continuity between (heir personal beliefs and those of the people they were studying (Turner 1962). But the recognition of this continuity, which I have tried to make clear in this book by using language (hat recalls our own culture, need not, in the end, make us feel, as Turner sometimes implies, that the phenomena are beyond explanation or criticism. Quite the contrary, it is precisely because of this continuity that we cannot see these phenomena as though they concerned another species, as the easy condemnations of cthnoccntrism often invite us so to do. This attitude is not only repulsive but also counter-productive, as it makes us forget where we are heading and so lose any clear guiding purpose.
This book is an attempt to understand the factors determining a social phenomenon, a religious ritual, in a given historical period. It docs not pretend, for reasons to be explained in the last chapter, to account for the phenomenon totally. It is, therefore, both a theoretical book, in that it proposes general conclusions, and also a book about specific events in specific places at specific times. This hybrid nature has been characteristic of anthropology since the time of Mal-inowski and Radcliffe-Brown and has. I believe, been one of its strengths, enabling it to avoid the empty platitudes 'pure1 theory often means and the pointless particularity of some recent studies.
The main body of this book consists of an analysis of a particularly rich ritual practised by the Merina of Madagascar that surrounding the circumcision of young boys. I saw this ritual performed eight times during a period of field-work between 1964 and 1966 and once during field-work in 1971. I also have indirect information about many other performances. The first substantive part of this book (Chapters 4-6) examines how symbolic analysis can explain the meaning of these performances in their social and cultural context and what connections can be established, causal or otherwise, with phenomena other than the ritual itself, using the various approaches available.
Chapter 7 deals with the history of the ritual, which, with difficulty, can be traced back almost two hundred years. These two centuries were particularly eventful for Madagascar in general and the Merina in particular, as they span a period when a large state developed, when (he population was converted to Christianity, when colonisation by Europeans came and went, and when one of the most bloody anticoloniai revolts in Africa south of the Sahara took place. Finally, Chapter 8 puts the historical and symbolic approaches together in order to explain what happens to such a ritual over time.
Generally, this study gives an example of how religion is involved in the rise of states, colonial rule and 'disestablishment', and, more generally still, how it is related to power, authority and violence. What changes and does not change in the ritual in some circumstances is one of the main concerns here, and this will be looked at in as much detail as possible, and perhaps beyond the endurance of the non-specialist reader, who might be advised to skip Chapters 4 and 7.
Such detail is. however, necessary if we want (o move towards an understanding of how this kind of ritual is historically determined, since, in the end. causal propositions in the social sciences can be tested only historically.
As a preliminary, it is worthwhile reviewing briefly how the problem of the social determination of ritual has been approached or avoided in anthropology, to show why this study has taken the approach it has. The topics raised here in a preliminary way will be discussed again in greater depth in Chapter 8.
Perhaps the first writer to stress ritual in his analysis of society and to attribute a key role to it was the unorthodox theologian W. R. Robertson-Smith. Not only were his theories illuminating in themselves, but his influence on subsequent writers can hardly be overestimated His theory of ritual offers us perhaps one of the clearest statements of the determination of ritual by other factors - in this case, clan organisation, which, it is implied, is the organising principle of the rather hazy society he is writing about.
He begins by insisting on the primacy of ritual over belief as the core of religion. His arguments for this are varied. First, he argues that in a preliterate society there is simply no fonim where beliefs can be expressed or discussed outside the context of ritual. Second, he maintains that what matters to the participants, in themselves and others, is acting and participating rather than spec* ulating. Consequently, one is not concerned in a primitive society with what other people think about the supernatural; one is only concerned about their proper participation. Third, and perhaps most interestingly in the light of what this study reveals, Robertson-Smith suggests that rituals are much more stable historically than beliefs (Robertson-Smith 1899, pp. 23-31).
Ttien Robertson-Smith turns lo a consideration of the relationship bciwccn social groupings and ritual. In particular, he looks at sacrifice. Sacrifices, for him. are partly devices for restoring errant members of the community to the group, but principally they serve to reaffirm the solidarity of (he group in a communal meal. The content of sacrifice is seen by Robertson-Smith as direcüy determined by the requirement of incorporation and reincorporation. It is partly because according to Robertson-Smith commensality has for all human beings (he psychological meaning and effect of solidarity that sacrifices take the form of a meal, but also because, for himt this communal meal originally consisted in the eating of the totem of the clan. This totem was believed to share the same mystical substance as the clan members, and its communal ingestion entailed transformation of the participants into the same substance. Robertson-Smith's work retains much interest for us today, and the themes he dealt with will recur throughout this study, His discussion of the primacy and historical stability of ritual fits well (he particular case discussed in this book, although I should be wary of accepting this view for all cases, especially for societies of a much less hierarchic character than the Merina that seem to manifest a real taste for meta* physical speculation (Lévi-Strauss 1964, 1966, 1968, 1971; Endicott 1979).
What is particularly striking in Robertson-Smith is the central proposition that
ritual is determined by social organisation. His book is perhaps (he simplest statement of the determination or ritual by non-ritual elements in the literature until we come to the writings of certain members of the American cultural-ecology school (Rappaport 1979). This makes Robertson-Smith's theory a useful starling point for examining the problems that arise from such a position.
There is no doubt that Robertson-Smith was mistaken about the nature of clans and sacrifices, both among the ancient Semites and in other parts of the world, as E. E- Evans- Pri Ichard (1965, pp. 51-3) pointed out with emphatic contempt, but this kind of theory has been re-echoed again and again, often with much better evidence (for example, in Radcliffe-Brown 1952).
The first point to note about this type of argument is that it hinges on a set of correspondences that are taken to be causal relationships. In Robertson-Smith's theory, the elements seen as corresponding are clans, ritual congregation and the emotions of commcnsalily The clans cause the ritual for the purposes of solidarity, and this maintains the clans. The argument is therefore ideological in that the effect of the commensal ritual, solidarity, is made to be its cause, implying a conscious or unconscious motivation on (he part of the originators of the ritual - a motivation that would have to have been based on the same kind of knowledge of determination as (hat of Robertson-Smith himself. That is lo say, the purpose of the initiators would have to have been based on knowledge of the effects of commcnsality, and if those who continue the ritual do not slate (his any more, it is because they have forgotten the wise intent of their ancestors, because they cany on the institution merely from habit, or because they conceal the real purpose of the ritual from others and perhaps from themselves as well by some strange process of the unconscious. This type of explanation, which has often been labelled functionalist, is both undemonstrated and totally unlikely. It has been incisively criticised again and again (Necdham 1962: Jarvie 1965; Spiro 1965).
Even if we accepted (hat rituals such as sacrifices do maintain the unity of clans, or other social groups, we should still be a long way from understanding how they are formed. The question would arise whether the unity-maintaining function of the ritual accounted for all aspects of the ritual; otherwise the connection might be simply partial or even trivial. For example, there is a correspondence between the smaller unit of political administration in Britain and the congregation of Anglican churches at the parish level; but, clearly, to explain the Anglican religion in terms of the needs of this administrative unit would be preposterous, Robertson-Smith goes farther than many of his successors in accounting for details of the rituals he is dealing with in terms of the need for solidarity; for example, he explains sacrifice, the communal meal, the choice of sacrificial animal, but even he fails to account in sociological terms for the pre* supposition that underlies all this: the notion of totemism, which, unlike his predecessors, he just takes as given. By doing this, he leaves us with the problem that we are not even given the basic requirement for a theory of causation, a clear
siaiemeni that one type of phenomenon accounts in part or in whole for another. Wc are left simply with congruence; causation is only imputed.
The problem arises again when a somewhat different type of determination is proposed for a ritual in the work of :< Rappaporl in his adventurous book Pigs for the Ancestors (1967). He argues that ritual should be seen as an adaptive mechanism of the human species in a given ecological balance. Rappapon seems to be arguing in his book, or at least in a subsequent article (Rappapon 1977), that a complex ritual of a New Guinea people is explained by its effect on the ecological balance, in particular by reducing the pig population when it reaches a critical threshold. The ritual then sets off a negative feedback, restoring the ecological balance. But, having given an explanation of this type, Rappaporl finds that it raises a whole range of further questions. In particular, why, if the deleterious effects of the rising pig population are clear to the people - and Rappaport implies thai they are (1967. pp. 158-9) - should the message have to be given in ritual form at all? His answer is that ritual makes the message imperative and beyond discussion because of its sacred character and that it is the only way in which such important actions can be enforced in uncentraliscd societies-
Rappapon, like a number of other functionalists, does not imply an original founder for the ritual, rather he replaces this mythical figure with natural selection that explains the continuance of this particular rituaľ The form of the argument is, therefore, that culture, like genetic mutations, produces a large number of random forms and that the useful ones persist while the harmful ones disappear. In slightly less courageous form this argument is quite common in anthropology when it dissolves into merely noting the simultaneous occurrence of two phenomena - the weak functionalism of Jarvie (1965), Congruence, however, either is trivial or implies in a disguised form, as it usually docs, the kind of argument clearly stated by Rappaporl.
The problems in this son of interpretation are many and have been discussed often (Friedman 1974, Rappaport 1977), but overall they are the same ones raised by Robertson-Smith's study. Stressing the effect of the ritual on the pig population and other ecological balances may be just a trivial observation. Since this ritual, like so many others, involves killing animals (in this case pigs), it inevitably affects the pig population. Passing from this observation of effect to the specification of cause is quite impossible. It would require Rappaport to show that somehow populations with the type of ritual he analyses outlive populations without it- It would lead to the implication that the ritual is like an instinct, something beyond conscious motivation. (Otherwise the notion of natural selection could not apply.) It would require an explanation of why the Tsembaga were unable to dispose of pigs in ways other than through this particular ritual. AH this cannot be done.1 In other words, the appearance of the demonstration of determination of the ritual is achieved in this explanation, as in that of Robertson-Smith, by confounding effect and cause. The same criticism could be directed at many other writers who have tried to explain rituals as the products
of a simple determining process - Radcliffe-Brown, Freud. Gluckman, to cite only a few.
Another type of explanation of the determination of ritual by social phenomena is to be found in the work of a number of Marxist writers. Generally they draw on [he notion of ideology. Ideology for them is a device whereby a ruling class imposes its ideas, values and image of reality on those whom it dominates. Few Marxist writers have have discussed ritual, but those who do, such as M Godclicr, see ritual as pari of ideology and as determined by this purpose (Go-dclier 1973, Chapter 5), or more generally for maintaining order. Here again, however, we find the same problems as those discussed in relation to the writings of Robertson-Smith and Rappapon A bizarre scenario seems to be implied for the origins of ideology, or of rituals that create and perpetuate ideology. Ii would involve ideology's being created as a plot by cynical rulers who deliberately in* vent subtle and totally convincing mystifying devices for the domination of others, or invoking, as Rappapon does, a theory of natural selection that explains persistence. Above all it makes the significance and power of religion for participants quite beyond comprehension. Although these writers tend to ignore the rather absurd notions implicit in their theory of ideology, the implications are nevertheless there. Otherwise one would have to make the totally un-Marxisl supposition that ideology was created by other than human agency. Also present is the old problem thai the demonstration of the alleged facl that ideological apparatuses such as rituals perform the functions of ideology does not mean (hat these functions are the cause of their being as they are.
Robertson-Smith, Rappaport and Godelier therefore all run into exactly the same difficulties. Their theories either imply a form of intentional action thai is quite unlike anything we know and thai they do not document, or. if they do not, their observations become nothing more than the noting of congruences that have nothing necessarily causal in character. Perhaps the weakness of such an approach can best be shown by pointing out that the opposite conclusions can be reached from similar data with equal plausibility. This is precisely the position of two other writers on ritual, N. Fustel de Coulange and Emile Durkhcim, who were able to use Roberixm-Smith s conclusions for their totally different theories. For them, in a variety of ways, it was the ritual, the emotions and the concepts embodied in it, that creates the social units with which they are congruent.
Fustel de Coulange's argument is that the concept of ancestor worship was to shape all Roman history and create the notion of the ancient city and its laws on the model of the original patriarchal family or minima) lineage (Fustel de Coulange. 1868). His work is, however, rather ambiguous as a general theory of determination, because he also shows how this concept of the city was undermined, and later defeated, by other types of groupings, not religious or ritual in origin, but corresponding to status groups or classes. A much more complex picture therefore emerges than the declared idealist posture of the author leads
us lo expect. None the less, the fact remains thai precisely the opposite point of view lo that of Robertson Smith can be deduced from similar data with as much* or as little, plausibility.
Many of the views of Fustcl de Coulange were taken over in Durkhcim's work. particularly in his book The Elementary Forms of the Religious Life (Durkheim, 1912). Unlike Fustcl de Coulange. Durkheim sees religion and ritual as ultimately determined by material conditions, especially demography. In so far as he sees a relationship between rituals and such social categories as clans, however, it is the former that determine the latter. Durkheim sees ritual as the device by which the categories of understanding organising our perception of nature and of society are created and given their categorical, hence inevitable, and compulsive nature, ln other words, the theories of Durkheim and Fustcl dc Coulange are mirror opposites of those of such writers as Robertson-Smith. Rap* paport and Godelier, because a connection between ritual and social organisation docs not itself constitute a demonstration of causation. It appears to be a demonstration of causation because all the writers hypothesise an unknown historical period when the one clement directly and in a flash created the other. Of course, if such historical events had taken place the matter would be settled. However, none of the people concerned give any evidence of this historical moment, and as soon as wc try lo imagine itt it becomes preposterous. Determination cannot be demonstrated by hypothetical history, only by real history.
In spite of the apparent differences among these writers they all have one thing in common: Their treatment is highly reductionist. This is another familiar criticism of functionalist theories but it remains none the less valid. The problem is that any theory that explains rituals in terms of their socially regulative functions pays attention lo only a few aspects of very rich and complex phenomena. There is no way that such functions can account for the complexity of what we lind, as will be shown, yet again, with the example that forms the subject matter of this book.
The reason for this reduclionism is actually no different from the reason for the failure to demonstrate causation in functionalist theories. It is the false notion of history that such theories imply. In these theories the social need for solidarity, or domination, or ecological balance, created the ritual. The moment when this occurred in envisaged as a very simple one: Before, there was nothing significant; afterwards, there was the ritual A ritual created so simply can therefore only be accounted for by Ihe simple function it is required to perform. Of course rituals arc much more complex, but since the theory implies such simple punc* tual historical creation, the theoretician must just ignore what is in front of him or her and only note what fits. As will be argued more fully throughout this book, il is only when one looks at real hisiory in all its complexity, a hisiory that has no starling point, that we can understand ihe possibility of an intricate process of determination producing a phenomenon as complex as the ritual we shall examine.
The realisation of Ihc rcductionism inhcrenl m functionalist analysis led in anthropology to a different approach lo ritual. This reaction is exemplilied in the writings of theorists such as Horton, but is already implied in the work of Radin and the later works of Evans-Přilehaní and even earlier writers (Evans-Prilchard 1956). Their approaches vary significantly, as will be discussed in Chapter 8, but for the sake of brevity here ihey can be grouped together under ihc label of intcllcctuattsts and symbolists* Basically such authors sec religion us an explanation of the world and man in fairly traditional theological terms. They sec religion as a speculation on nature and an intellectual accommodation of the beyond. They are very critical of functionalism. as they recognise it as a device for ignoring the content of ritual. They refer to Robertson-Smith's view on the unimportance of belief as proof of ihis.
Certainly the reaction ihey represent has shown that there is much more to be seen in rituals than has previously been supposed. This is also true of other writers who take a somewhat similar attitude, with different stylistic modes of presentation» using various diagrammatical or logical devices for presenting their data. These would include Biede I man. Maybury-Lewis and Willis, The close analysis of the content of ritual thai this antifunctionalist approach produced has clearly been a major advance and I have attempted lo follow the examples ihey offer in Chapters 4 and 5 The whole approach has. however, raised problems as fundamental as those of functionalists Perhaps the first and most fundamental of these difficulties was Robertson-Smith's starling question. Why, if rituals arc lo be seen as intellectual speculation, do they lake the form of ritual? Why is participation so important^ Why do rituals have power? Finally, how and why do they come about? As soon as we begin lo attempt an answer we seem lo be thrown back lo ihc funciionalisl questions and at the same lime to their methods.
In ortler to avoid reductionism, the intellectually have made of rituals something lhat they clearly arc not; a discourse on the nature of man in the universe, indeed, as has been shown by G. Lewis (1980) this misrepresentation leads to totally unwarranted additions in that (he intcllcctualists complement what they can observe with other deductions in order to produce the kind of scheme that in ihe anthropological literature is referred to as cosmology. The antifunctionalists therefore fall into another form of reductionism that may well be even more dangerous.
Thus the problem of inlcllcctualist and symbolic approaches is very similar to that of functionalist ones: The former theories assume lhat rituals arc as they arc in order to fulfil a single function, lhat of explanation. This view also implies a pseudo-history, a just-so story of an absolute beginning, when a thinker, unbound by society or the necessities of life or for that matter any previous intellectual preconceptions, worked out the whole thing. This is totally ludicrous and contradicts everything we know of real human beings in history, with the possible exception of certain philosophers who have never been attributed with initiating rituals of the type discussed here. Of course most of the writers who adopt
such a position would deny any such strange scenario, bui again* as for the functionalists, unless they believe that the schemes they construct fell ready-made from heaven, they cannot escape the implication.
Not only docs the substitution of false history for real history lead to the adoption of untenable iheories of causation, it also leads as it did for the functionalists to reductionism. With such a simple imputed cause nearly all the aspects that make the ritual a ritual and not a Platonic dialogue have to be ignored. Here, indeed, these authors rejoin some of the Marxist writers who see in ideology an 'alternative theory' of the world to the true one. What they both miss is thai the image so created is of a different kind to non-ritual knowledge because of the way it is created phenomenologically and historically. Again, we find that it is only if wc return to the reality of the historical process that the complexity and the many facets of rituals can begin to be accounted for. The functionalist, the intellectuals! and the symbolist positions that have dominated anthropology will be shown, therefore, to be equally, and for very similar reasons, unsatisfactory.
It may appear that wc arc. therefore, doomed in the study of ritual to be endlessly bouncing between two walls - a functionalist wall, which implies a process of formation that is clearly wrong and that leads us to ignore most of what can be observed, and an intellcctualist or symbolist wall, which fails in any way to explain or place rituals in their social context, and as a result leads us lo misrepresent our data by making us pretend that it is a different type of activity than it patently is. that is* by making it look as (hough it were a theological treatise.
However, the picture in anthropology is not quite so bleak. There have been some writers who have attempted to combine the sociological with the symbolic approaches. Foremost amongst these has been V. Turner. It is impossible to do justice to Turner's very subtle theory here, and in any case perhaps its greatest strength lies in the fruitful method of analysis of particular cases it implies. However, what he proposes is that we should note the symbolic side of ritual, together with its emotional and its sociological aspects What he fails to do is to bring these together convincingly, except by suggesting that the ambiguity and complexity of symbolism make it suitable for social manipulation. Although this point is undoubtedly valid, it does not explain how the social brings about the symbolic, and Turner denies strongly that this will ever be possible.
The reason is that the social considered by Turner is the very shon term, for example a segmentation dispute, a struggle between two different potential leaders. On a short time-scale it is quite clear that the social does not bring about the symbolic, and to suggest that it docs, as is done in functionalist theories, is, as Turner clearly sees, wrong. This does not mean, however, that the establishment of a causative link between the history of the social formation and the ritual cannot be attempted. It is simply that the connection cannot be understood either synchronically, or in the very short term. Indeed, this book will show that for the Mcrina circumcision ritual, two hundred years are quite insufficient fordoing
more Chan beginning lo understand (he nature Of determination. The reason is nol far to seek. People act in terms of what they know and what they know is the product of their historically constructed culture* They may transform and change this culture but they do not do it from a zero starling hase Because of this the study of determination must be not a study of initial creation, but of the principles of transformation.
It is this problem that forms the approach used in this book. The circumcision ceremony will be considered not only as a symbolic system but also as a symbolic system being created in history, as far as it is possible to do so» because this proCttS of creation OCCUR on a much longer lime-scale than anthropologists have usually been willing to consider. This point is implied in most of the work of Lévi-Slrauss but never demonstrated by him, simply because he has never done a genuinely historical study. It is attempted here, although, as must inevitably be the case, the historical determination of the ritual through time can he only partially examined. The reason is simple: The processes of formation the ritual must have imply a much longer time-scale lhan the two hundred years examined here. Because only two hundred years will have been examined, only a pan of the ritual can be accounted for. This should nol surprise us. After all. we know that there are much larger continuities in culture lhan arc covered by two centuries. Indeed, the idea of culture areas, which is accepted implicitly or explicitly by most anthropologists, inevitably implies this.
Of course these consideration* apply to any cultural phenomenon hut the general problem becomes somewhat modified when we turn to ritual. The reason for this lies in the very character of ritual as a special kind of activity. One of the criticisms of the inlcllectu3lis< position is» as we saw. that it presents rituals as though they were an intellectual treatise: as cosmologies. A number of writers have pointed out how misleading this is. Foremost amongst ihc anthropologists are Bateson (1958) and Turner (1967). who have emphasised in different ways the emotive and sensuous aspects of ritual More recently Sperber (1974) has shown how totally wrong it is to approach ritual and symbolism as though the task in hand was a matter of decoding and translation More practically a number of anthropologists (Bloch 1974, Rappaport 1979. Tambiah 1979. Lewis 1980. Ahem 1981) have tried to isolate the particular nature of the communication used in ritual to understand what relation the symbolic content of [he ritual has to other types of communication. In spite of a number of disagreements and varying emphasis all these writers agree that what characterises ritual is that it lies somewhere between an action and a statement. Because of this any representation of the 'argument" of a ritual as such is misleading. But because ntual does have elements that are like statements, because it does retain some propositional force, the opposite altitude, which would deny any validity to an analysis of content of ntual would also be misleading* We are. therefore, faced with a genuine difficulty, which originates in part from the literary and narrative techniques anthropologists must use if they are to be readable. When representing the nature of
ritual, they can neither analyse rituals as (hough they were propositions nor totally ignore their propositional character. The solution adopted here follows the lead of several of the writers already mentioned: It first analyses the ritual symbolically as though it were a proposition and then heavily qualities this presentation, in pan by examining its historical destiny.
This is merely (o recognise the special and difficult nature of rituals and explains why the study of ritual has offered such a challenge and has figured ever more prominently in anthropology. On the whole, the problem has been faced by ever refining the formal analysis of ritual with help from communication theory, psychology, philosophy» even literature and literary criticism- I have no doubt of the value of these approaches and I feel we have genuinely advanced in our understanding of ritual. However, I also have a feeling that these studies, including my own, represent attempts lo grasp what, in the end, it is impossible to grasp: what rituals mean to the participants and the onlookers. This type of search for meaning» although not pointless, has no end.
There is, however, another way of doing the same thing that for a moment at least bypasses the difficulty, and that is to look at what happens to rituals in the course of history. Quite simply, if rituals are a special kind of phenomena, it follows that they will be manifested in history in a special way. If this is so, it is also reasonable to assume that the special way rituals arc manifested in history will reveal what kind of phenomena they are. The advantage of this approach is that, unlike other attempts at getting at the nature of ritual through seeing what it means for the participants, it uses as its basis much more easily graspable phenomena: the way rituals are affected by events.
There is. however. ye( a further advantage in (his approach: It enables us to bring together the functional and the intellectual sides of the study of ritual in that it enables us to follow the interplay between the two aspects without reducing one to the other. The aim of this book is therefore to show that the meaning and nature of a ritual can be understood in the process of its historical formation and that by this means the recurring problems of the study of ritual can be overcome. But some attempt will be made beyond this to reconsider the question of ideology in general and to see how some of the conclusions reached concerning ritual can illuminate the more fundamental questions raised by this controversial concept. It seems to mc that it is only by combining the knowledge of what the experience of ideology isf something that anthropologists with the experience of field-work behind them arc able to do. and the knowledge of history that we can advance our understanding in this area.