4
Enactments of meaning
Roy A. Rappaport, Ritual and Religion in the Making of Humanity (Cambridge: Cambridge University Press, 1999), pp. 107-138.
The complex relationship between the self-referential and the canonical streams of ritual's messages is best approached through further exploration of the relationship between saying and doing. After a preliminary discussion of general principles of efficacy we will consider a crucial indexical message, intrinsic to ritual's very form, alluded to at the end of the second chapter, in the absence of which the canon would be inconsequential. We approach here what I understand to be ritual's fundamental office and will discuss in this light the establishment of convention in ritual and the social contract and morality that inheres in it. These observations provide grounds for taking ritual to be humanity's basic social act.
At the end of the last chapter I argued that participation in rituals might not only indicate aspects of performers' contemporary states but impose transforming decisions on those states. A clumsy bit of sleight-of-hand may seem to be poorly hidden in that argument. To claim on the one hand that supercision, for example, indicates the achievement of a certain stage in a boy's maturation and, on the other, that it imposes a dichotomous decision on that process, may seem either ingenuously confused or disingenuously confusing. To indicate a condition would be one thing, to transform it another. The two, however, are not being confused. They are being conflated. Some, if not most, of the self-referential messages occurring in ritual do not merely "say something" about Trfc state of the performer. They "do something" about it. That a Tahitian boy indicates something about his ontogenetic stage by having himself supercised is indubitable. It is no less indubitable that he has done something about it. Similarly, by dancing at a kaiko, a Maring man signals iris pledge to help his hosts in warfare.1 Dancing cannot help but
107
108      Ritual and religion
Enactments of meaning
109
signal that pledge because, among the Maring, dancing brings that pledge into effect. This is further to say that that signal is indexical (and not merely symbolic) because the acts of dancing and pledging are one and the same. Such a pledge can, of course, be violated but, on grounds discussed in chapter 2, it may be asserted that there is no way to lie about having made the pledge.
It was earlier argued that many, if not all, of ritual's self-referential messages are indexical in character. I would now emphasize a point noted in passing in chapter 1: that the indexical nature of acts signalling conventional states (such as pledging) is a consequence of their accomplishment of whatever it is that they indicate. We earlier noted that the relationship between sign and signified in indexical self-referential transmissions is the inverse of what it is in more familiar symbolic transmissions. In the casual usage of everyday, we usually, and rather carelessly, take signs, indexical or otherwise, to report, describe, represent, denote, designate, reflect or otherwise signify states of affairs existing independent of, and usually previous to, our references to them. In the case of the ritual acts and utterances with which we are concerned, the sign brings the state of affairs into being and - here is the sleight-of-ritual -having brought it into being cannot help but indicate it. We shall return, a little later in this chapter, to ritual's one indispensable indexical message. We are further concerned in this chapter with the ways in which conventional states are transformed and, at a deeper level, how conventions themselves are established in ritual. Because, however, the significance of what I have been proclaiming to be ritual's most profound indexical message is best elucidated in such a context, we will first consider principles of efficacy.
1. The physical and the meaningful
Both the occurrence of ritual and ritual's contents form and transform that upon which they are imposed, but, as we noted in chapter 2, not primarily by force of energy or expenditure of matter. What is often called their "power"2 rests upon other means or principles. We may discern in nature two general classes of efficacy - the physical and the meaningful. The efficacy of what Leach calls "technique," achieving as it does its results through the deployment of matter and energy in accordance with the laws of physics and chemistry is, largely if not entirely, physical, but prayer is not, nor is ritual, nor are words. Their efficacy is grounded in principles of communication.
Bateson has noted some general differences between the two classes:
when you enter the world of communication, of organization, etc., you leave behind the whole world in which effects are brought about by forces and impacts and energy exchange. You enter a world in which "effects" - and I am not sure one should still use the same word - are brought about by differences [bits of information] ...
The whole energy relation is different. In the world of mind nothing - that which is not - can be a cause. In the hard sciences, we ask for causes and we expect them to exist and be "real." But remember that zero is different from one, and because zero is different from one, zero can be a cause in ... the world of communication. The letter which you do not write can get an angry reply; the income tax form that you do not fill in can trigger the Internal Revenue boys into energetic action, because they, too, have had their breakfast, lunch, tea and dinner and can react with energy which they derive from their metabolism. The letter which never existed is no source of energy ... what we mean by information, the elementary unit of information, is a difference which makes a difference, and it is able to make a difference because the neural pathways along which it travels and is continually transformed are themselves provided with energy...
The pathways are ready to be triggered. We may even say that the question is already implicit in them.
There is, however, an important contrast between most of the pathways of information inside the body and most... outside ... The [external] differences ... are first transformed into differences ... of light or sound and travel in this form to my sensory ... organs. The first part of their journey is energized in the ordinary hard science way, from "behind." But when the differences enter my body by triggering an end organ, this type of travel is replaced by travel which is energized at every step by the metabolic energy latent in the protoplasm which receives the difference, recreates or transforms it and passes it on.
(1972d: 452-453; emphasis in original)
Because, in possible disagreement with Bateson, I take information to be no more than one form of meaning and because I do not think that all forms of meaning can be reduced to information in the strict sense, I designate the class into which ritual falls as that of "meaningful (rather than informational) acts."
To distinguish the meaningful and the physical as two distinct classes of efficacy, is not to propose that they are separate or separable in nature. That ritual achieves its effects through the communication of meanings does not imply that it does not both consume and mobilize energy and material, nor that technique proceeds in some mindless way without the guidance of meaning. Matter-energy devoid of information is necessarily devoid of life: information and other forms of meaning separated from matter-energy could be conceived as mathematical or formal abstraction, pure spirit or Platonic ideal, but the act of conceiving it as such, being a biological process like all other acts of conception,
110      Ritual and religion
Enactments of meaning       111
could not occur in the absence of matter and energy. Although there seems to be no direct relationship between the meaningfulness of messages and the amount of energy required to transmit them, the communication of meaning, both informational and of higher order, always requires energy and sometimes matter as well. Even speech is propelled by energy, and information is inscribed on paper, engraved on stone, combined in DNA and encoded on magnetized tape. Conversely, any change detected by an organism in the energy flux to which it is subjected, or any change it detects in its material environment conveys information to it, for information is, in one of its aspects, detected, or at least detectable, difference. Moreover, as Bateson observes, matter-energy and informational processes frequently cannot be separated from each other because the receipt of information leads organisms to expend their own energy to bring about effects that may include both material and informational elements. Matter-energy processes and informational processes, although they can be distinguished, are inseparable in nature, and it is of interest that the myths of many peoples, including both Australian aborigines and ancient Hebrews, describe creation, either in whole or in part, as an act or set of acts, imposing form upon an already existent but inchoate primordial matter (Bateson 1972b: xxiii ff.). Creation, this is to say, is conceived as the informing of substance and the substantiation of form. Higher-order meaning comes later.
Information and matter-energy processes may be inseparable in nature, but they, the objects they affect, and the ways in which they achieve their effects may be distinguished. Matter and energy, it seems clear, operate most effectively, that is with most predictable results, upon inert materials. The efficacy of information and other forms of meaning, on the other hand, rests not only upon the ability of senders to encode and transmit information, but upon the ability of others to receive those messages, that is to recognize, comprehend and take account of them.
Not all entities are equally capable of being informed. Mythic acts of creation aside, insensible and inert objects cannot receive messages at all. They can be formed and transformed and arranged so that they can convey information to others but they can't be informed. Organisms, social systems, and perhaps some machines constructed by humans can be informed, that is, receive messages and modify their behavior or understandings in light of such receptions. Obviously there are great differences among systems in their receptive capacities. Dogs probably have greater absolute capacity for being informed than do snails, and humans greater capacities than dogs.3 While there are not known to be
differences in the average capacities of the memberships of different human populations for receiving and processing information, there surely are differences between individuals per se. Moreover, common sense, if not empirical studies, suggest that there are differences among human groups as groups in their capacities to receive information and to be transformed by the information they receive, owing to differences in their size, literacy, technology, and special institutions for receiving, storing, interpreting and disseminating information.4
That there are differences in what members of different human groups take to be meaningful is, of course, the raison d'etre of comparative anthropology. Be this as it may, it is obvious that few Americans would be informed of much by a speech in Mongolian, and the same may be said of some of the non-verbal as well as verbal messages transmitted in rituals. Most American youths might be deeply impressed by being subincised, but none, it is probably safe to say, would be hastened on the road to sociological manhood by such an operation, as would young Walbiri or Arunta. The ritual form may be universal, but all human rituals include signs specific to the society, church, or congregation in which they are performed, and the arrangement of sign elements into liturgical orders is also in some degree socially and culturally specific. If the reception of the messages encoded in liturgical orders requires that they be in some sense "understood" (which is not to say that they may not at the same time be mysterious), then it is necessary that they be in some way learned, because the understanding of signs only conventionally related to their referents could not possibly be specified genetically. This is to say that participants must be trained, indoctrinated or otherwise prepared to receive the messages rituals transmit. The work of Campbell (1959), Erikson (1966), Turner (1969) Wallace (1966) and Goodenough (1990) suggests that the ability to be informed by ritual is itself established in the individual, in part in the course of a series of rituals starting in early infancy and proceeding to maturity. Erikson (1966) has referred to this process of preparation as "ritualization," and we shall return to it in a later chapter. Here it is necessary to note only that the informative capacity of ritual, its ability to form and transform, rests not only upon its special mode of transmission but also upon its reception by specially prepared receivers.
The effects that can be achieved by the deployment of motion and energy on the one hand and meaning on the other are also obviously different. Physical effects - weights lifted, ditches dug, billiard balls knocked into pockets, acids neutralized, metals smelted, plants cultivated
112      Ritual and religion
Enactments of meaning      113
- must rely upon physical processes for their achievement. Conventional effects, on the other hand - princes transformed into kings and words into promises, the profane made sacred and truces declared - can only be achieved by meaningful acts. Whatever energy may have been required by a young man to fulfill the responsibilities of knighthood, or even required to complete his transformation to knighthood, did not flow into his shoulder from the sword with which he was dubbed, nor from the voice of whomever it was who dubbed him. It came, to recall Bateson, from his breakfast and dinner. Dubbing transformed him into a knight not by the force of its blow but by informing him of his knighthood, or better, informing him with knighthood. (In fact, according to the Oxford Unabridged Dictionary, a man is dubbed to knighthood, the transformative aspect of being informed being emphasized.) Although certain physical acts were performed with certain objects in dubbing, whatever transformation took place in the passage to knighthood was neither a physical alteration of the young man, nor was it effected by a process that can be comprehended as physical, chemical or biological. The transformation was, and could only be, effected by communicating meaning to whomever was the locus of the transformation, and to other concerned persons, in accordance with the conventions of the society in which it was occurring. The same may be said of the transformations in group membership and belligerency status effected by the planting of rumbim among the Maring, and even of those rituals in which the initiate is subjected to severe pain and by which he or she may be left physically marked for life: scarified, subincised, circumcized, canines removed, septum pierced, fingers lopped off. The significant transformations produced by such operations are obviously not physical, nor are they outcomes of the physical changes per se wrought by such operations but of the meanings those changes carry.
Ritual acts, such as body mutilation, planting rumbim and dubbing are, to use Van Neumann's term, "markers," that is "observable bundles, units or changes of matter-energy whose patterning bears or conveys the informational symbols from the ensemble" (James Miller 1965: 164). It is, of course, a matter of great interest that humans, who have a range of codes and markers to choose from, employ precise, subtle, energetically and materially inexpensive speech for the transmission of some messages and comparatively crude, expensive and sometimes painful physical acts for the transmission of others. We touched upon this question peripherally in the last chapter, and will discuss it in greater detail in the next.
To distinguish between the domains in which the physical and meaningful prevail is not to declare that the boundary between them is sharp or clear. It is unlikely that any sea has ever been parted by prayer or turned back by command, and we may be equally confident that no prince has ever been transformed into a king, no man and woman into husband and wife, by matter and energy alone. But prayer as well as drugs may have an effect upon the physical well-being of those praying and even upon the health of those for whose sakes prayers are offered. So may sorcery. Accounts of "voodoo death" are well-documented (for recent discussions of such matters see Lex 1979, d'Aquili and Laughlin 1979) and provide us with reason to believe that the efficacy of some rituals rests upon the ability of human organic processes to translate information conventionally encoded in such utterances and acts as cursing, bone pointing and shamanic projectile removal into chemical and neural signals. These, in turn, may have further physical consequences, either beneficial or harmful, for the organism receiving the message. It is significant that the ritual acts initiating attempts to achieve organic effects are often simple and easy to observe, but the subsequent neural and hormonal processes directly producing the effects are not. They are extremely complex and they are hidden from direct view. The precise nature of the causal principles relating the act to its ultimate effect is, thus, obscure and even mysterious. The location, within human organic processes, of the boundary between the domains of the physical and the meaningful is not well known but cannot be a sharp one. It is plausible to believe that the very obscurity of this region is one basis of notions concerning the occult efficacy of ritual words and acts.
2. Speech acts
There are important differences among the occurrences that have been rather casually included here in the class of meaningful acts. Events, such as dubbings, rumbim plantings and supercision must be distinguished from messages which simply inform receivers of conditions in their social or physical environments.
First, such rituals are more likely to inform the participants of changes in themselves or perhaps it would be better to say with changes in themselves, than they are of changes in their surroundings.
Secondly, whereas simple messages concerning environmental conditions leave the responses of receivers to their own devices, rituals specify their responses, often very precisely. When, in medieval Europe, a youth
114
Ritual and religion
Enactments of meaning
115
was dubbed he was not stripped of his status to become anything desire or imagination suggested to him. He became a knight and nothing but a knight. When an alien grasps the rumbim of a Maring local group he becomes a de jure member of that local group, and nothing else. When they uproot rumbim. Maring husbandmen become potential warriors. When they are supercised it is t'aure'are'a, and nothing else, that Tahitian boys become.5
Thirdly, while a message concerning the state of the social or physical environment may lead us to undertake an action which will transform ourselves or the conditions surrounding us, at least some rituals themselves complete the transformations with which they are concerned. Dubbing, we have observed, does not tell a youth to be a knight, nor does it tell him how to be a knight. It makes him a knight.
Ritual is full of conventional utterances and acts which achieve conventional effects. "I dub thee to knighthood," "I name this ship the Queen Elizabeth," "I swear to tell the truth." "I promise to support you in warfare," "We find the defendant guilty." The importance of such utterances in the conduct of human affairs is so patent as to obviate the need to argue it, but philosophers, in the last few decades especially, have given considerable attention to their peculiar characteristics. J. F. Austin (1962) has called them "performative utterances" and "illocutionary acts," J. R. Searle (1969) includes them among what he calls "speech acts," F. O'Doherty (1973) refers to an important sub-class as "factitive" acts or utterances, J. Skorupski (1976) uses the term "operative acts" for a class resembling them closely.
It is important to make clear that the force of what I shall call "performatives," following Austin's earlier and simpler terminology, does not depend in any simple and direct way upon the effect of these acts and utterances upon the minds and hearts of those exposed to them. Whether or not he has reservations, planting rumbim joins a man to the group with whom he plants it. Regardless of what they may think or feel about it those who are excommunicated, outlawed, found guilty or demoted in rituals properly conducted by authorized persons are thereby excommunicated, outlawed, made felonious or degraded. If authorized persons declare peace in a proper manner, peace is declared whether or not the antagonists are persuaded to act accordingly. This is not to say that acts and utterances which are performative may not be persuasive, threatening, inspiring or otherwise affect the receiver in ways inducing him to act in particular ways. In the language of speech act theory they may have "perlocutionary" as well as "illocutionary" force.6 It is to say
that an action having a conventional effect is completed in the gesture or utterance itself.
Performatives differ in the scope of the action they complete. If an authorized person, following a proper procedure, names a ship the Queen Elizabeth, the ship is so named. Others may, if they like, call it "Hortense," but its name happens to be Queen Elizabeth, and that's really all there is to it. On the other hand, if a man has danced at another group's kaiko, thereby promising to help his hosts in warfare, that is not all there is to it. for it remains for him to fulfill his pledge and he may fail to do so. The naming, which not only constitutes an action but actually brings into being the state of affairs with which it is concerned, is of the class of performatives that we may call "factive."7 Whereas many actions completed in ritual - dubbings, declarations of peace, marriages, purification - are factive. it is obvious that all are not. Some - among which are those that Austin called "commissives" (1962: 150ff.) - do not bring into being the states of affairs with which they are concerned, but merely bring into being the commitment of those performing them to do so sometime in the future.8
3. The special relationship between rituals and performativeness
While many liturgies are performative, where some sort of performative act is the main point of the performance, transforming war into peace, restoring purity to that which has been polluted, joining men and women in wedlock, performativeness is not confined to ritual. There is no advantage to be gained, for instance, in taking the publican's utterance "The bar is closed" to be ritual, but when he says "the bar is closed" it is thereby closed, and you are not likely to get another drink. Performatives are not confined to ritual, but there is a special relationship between ritual and performativeness.
First, the formal characteristics of ritual enhance the chances of success of the performatives they include. Like any other acts performatives can fail. If, for instance, I were to dub one of my junior colleagues Knight of the Garter he would not thereby become a Knight of the Garter, even if the conduct of the ritual were letter perfect. Conversely, if Queen Elizabeth dubbed Princess Anne's horse to knighthood it probably wouldn't make him a knight, even granted the well-known English tolerance of eccentricity. And if a befuddled cleric recited the funeral liturgy rather than the marriage service I doubt if the couple standing before him would thereby become objects of mourning (Austin 1962: passim). All of these instances of faulty performatives are of ritual
116
Ritual and religion
Enactments of meaning      117
performatives and ritual performatives can misfire. The ludicrous nature of these instances suggests, however, that they are less likely to do so than are other non-ritualized performatives because the formality of liturgical orders helps to insure that whatever performatives they incorporate are performed by authorized people with respect to eligible persons or entities under proper circumstances in accordance with proper procedures. Moreover, the formality of ritual makes very clear and explicit what it is that is being done. For instance, if one Maring casually said to another whom he happened to be visiting, "I'll help you when next you go to war" it would not be clear whether this was to be taken as a vague statement of intent, as a prediction of what he would be likely to do, or as a promise, nor would it necessarily be clear what might be meant by "help." To dance this message in a ritual, however, makes it clear to all concerned that a pledge to help is undertaken, and it is conventionally understood that that help entails fighting. Ritual, this is to say, not only ensures the correctness of the performative enactment, but also makes the performatives it carries explicit. It generally makes them weighty as well. If a message concerning the current states of participants is communicated by participation in ritual it will not be vague, and the formality, solemnity and decorum of ritual infuses whatever performatives the ritual incorporates with a gravity that they otherwise might not possess. In sum, simply by making their performatives explicit rituals make clear to their participants just what it is that they are doing and thus they specify, at least tacitly, what would constitute an abuse or violation of whatever obligations those performatives entail. Clear definition itself may reduce the likelihood of abuses and violations simply by leading people to "think twice" before acting. To put this a little differently, clear definition, which is intrinsic to the formality of ritual, itself possesses perlocutionary force, and so do the gravity, solemnity and decorum characteristic of many rituals. Reflex-ively, the perlocutionary force inhering in the formality of a ritual supports whatever performatives are enacted in that ritual.
There are two other closely related reasons for considering the perfor-mativeness of ritual. First, the association of the sacred and occult with performatives in magical and religious rituals may mystify their conventional nature, and this may enhance their chances of success. To take the state of affairs established by a king's enthronement to derive from the sacramental virtue of crown and chrism may be more effective with respect to the maintenance of the social order over which the king reigns than would be the recognition of enthronement as a naked performative.
a mere conventional act, the effects of which could, for instance, in response to the short-run displeasure of his subjects, easily be reversed.
Second, as Ruth Finnegan (1969: 50) has suggested, albeit rather unspecifically, the "truth lying behind" assumptions concerning what is often called "the magical power of words" may be related to their illocutionary force or performativeness. It may be proposed, rather more specifically, that the magical power of some of the words and acts forming parts of liturgies derives from the relationships between them and the conventional states of affairs with which they are concerned. As we have already observed, the relationship of performatives to the states of affairs with which they are concerned is the inverse of that of statements. The facts, events or situations to which a statement refers presumably exist independent of and previous to the statement referring to them, and a statement is assessed true if it accords in some sufficient degree to those previously existing and independent states of affairs. Since performatives bring about the facts, events, situations, etc., with which they are concerned, these facts are subsequent to and contingent upon them. Performatives, and most unambiguous factives, are self-fulfilling: they make themselves true in the sense of standing in a relationship of conformity to the states of affairs with which they are concerned. In light of this it may be suggested that the performativeness, and more especially the factiveness of ritual acts and utterances provide a basis for occult efficacy in general, including the magical power of words in particular. Ritual's words do, after all, bring conventional states of affairs, or "institutional facts" into being, and having been brought into being they are as real as "brute facts" (Searle 1969: chapter 2, passim). It may also be that magical power is attributed to other words by extension of the principle of factiveness beyond the domain of the meaningful, in which it is clearly effective, into the physical, in which it is not, but we must be very careful about stipulating the limits of the effects ritual can accomplish. We have already argued in somewhat different terms that their illocutionary force may be augmented by perlocutionary force. The effectiveness of persuasion, threat, cajolery, inspiration and ecstasy may well thrust beyond the purely conventional, and beyond discursive consciousness into the organic, as in the cures of healing rituals and in the injuries of ensorcellment.
4. Ritual's first fundamental office
Performatives are not confined to ritual, and there seems to be more to some or even all liturgies than the performatives that they incorporate.
118
Ritual and religion
Enactments of meaning      119
Indeed, some liturgies may not seem to include performatives in anv simple sense at all. Many religious rituals do not seem to be directed toward achieving simple conventional effects through conventional procedures. If, however, simple performativeness is not criterial of ritual, something like it, but of higher order, may be. Although not all rituals are obviously and simply performative, performativeness itself may be made possible by ritual. We approach here the conjunction of formality and performance noted but not discussed in chapter 2. We come, this is to say, to what is intrinsic to the act of performing a liturgical order, and thus to the heart of the relationship between the self-referential and canonical.
The characteristic of liturgical orders salient here is the simple fact of their performance; that they must be performed. Without performance, there is no ritual, no liturgical order. Records or descriptions of liturgies performed in Ur and Thebes survive but they are merely about liturgies not themselves liturgies. They are remains of the dead, for the liturgical orders they recall are no longer given life and voice by the bodies and breath of men. Performance is not merely one way to present or express liturgical orders but is itself a crucial aspect or component of the messages those orders carry. The following may seem involuted. The involution is intrinsic to the phenomenon, and not to my account of it. A liturgical order is a sequence of formal acts and utterances, and as such it is realized - made real, made into a res - only when those acts are performed and those utterances voiced. This relationship of the act of performance to that which is being performed - that it brings it into being - cannot help but specify as well the relationship of the performer to that which he is performing. He is not merely transmitting messages he finds encoded in the liturgy. He is participating in - that is, becoming part of - the order to which his own body and breath give life.
To perform a liturgical order, which is by definition a more or less invariant sequence of formal acts and utterances encoded by someone other than the performer himself, is necessarily to conform to it. Authority or directives, therefore, seem intrinsic to liturgical order (see Maurice Bloch 1973). The account just offered suggests, however, something more intimate and perhaps more binding than whatever is connoted by terms like "authority" and "conformity." The general notion of communication minimally implies transmitters, receivers, messages, and channels through which messages are carried from transmitters to receivers. Sometimes, furthermore, as in the case of canonical messages, which ritual s performers find already inscribed in prayer books or prescribed by tradition, transmitters should be distinguished from encoders, whose
identities may be lost in time and whose dicta are, in part for that very reason, timeless. We earlier noted a peculiarity of ritual communication, namely that in ritual the transmitter and receiver are often one and the same. At least the transmitter is always among the most important receivers. Now we note another of ritual's peculiarities. To say that performers participate in or become parts of the orders they are realizing is to say that transmitter-receivers become fused with the messages they are transmitting and receiving. In conforming to the orders that their performances bring into being, and that come alive in their performance, performers become indistinguishable from those orders, parts of them, for the time being. Since this is the case, for performers to reject liturgical orders being realized by their own participation in them as they are participating in them is self-contradictory, and thus impossible. Therefore, by performing a liturgical order the participants accept, and indicate to themselves and to others that they accept whatever is encoded in the canon of that order.
This act of acceptance is the first of ritual's fundamental offices. The self-referential and the canonical are united in the acceptance of the canon. Acceptance is the self-referential message intrinsic to all liturgical performances, the indexical message without which liturgical orders and the canonical messages they encode would be without consequence, nonexistent, or vacuous. It is not a trivial message because humans are not bound to acceptance of particular conventional orders by their genotypes. They are often free not to participate in rituals if they do not care to, and refusal to participate is always a possibility, at least logically conceivable, by potential actors. Participation, and thus acceptance, always rests in logic and in some degree in fact, upon choice. Such choices may sometimes be extremely costly, but are always possible.
5. Acceptance, belief, and conformity
The assertion that acceptance is intrinsic to liturgical performance may still seem to be either dubious or indubitable. It therefore requires some elaboration and clarification in this section and the next.
First, acceptance is not belief. The concept of belief is difficult to define and the occurrence of belief difficult to establish (see R. Needham 1972). Let us say that the term "belief" at least suggests a mental state concerning, or arising out of, the relationship between the cognitive processes of individuals and representations presented to them as possible candidates for the status of true. As such, "belief" is a second-order process, that is, one concerned with the relationship between a first order
120      Ritual and religion
Enactments of meaning 121
process and external reality. By this account, belief is an inward state, knowable subjectively if at all, and it would be entirely unwarranted either for us or for participants or witnesses to assume that participation in a ritual would necessarily indicate such a state.
Acceptance, in contrast, is not a private state, but a public act, visible both to witnesses and to performers themselves. People may accept because they believe, but acceptance not only is not itself belief; it doesn't even imply belief. Ritual performance often possesses perlocutionary force, and the private processes of individuals may often be persuaded by their ritual participation to come into conformity with their public acts, but this is not always the case. Belief is a cogent reason, but far from the only reason, for acceptance. Conversely, belief can provide grounds for refusals to accept. Reformers and heretics, for the very reason that they believe deeply in certain postulates concerning the divine, may refuse to participate in the rituals of religious institutions they take to have fallen into error or corruption.
This account suggests that although participation in liturgical performance may be highly visible it is not very profound, for it neither indicates nor does it necessarily produce an inward state conforming to it. Such a view is widely held by critics of religion who are inclined to take ritual participation to be nothing more than empty or even hypocritical formalism, a view reflected in one of the term's common modern meanings, formal behaviour devoid of substance or consequence. But. paradoxically, it may be, and it surely has been implied by religion's defenders, that the acceptance indicated by liturgical performance, being independent of belief can be more profound than conviction or sense of certainty, for it makes it possible for the performer to transcend his or her own doubt by accepting in defiance of it. Even the most devout, indeed especially the most devout, sometimes harbour doubts or even voice scepticism concerning propositions expressed in liturgies to which they scrupulously conform, and acceptance in this deep sense has much in common with certain Christian notions of faith. Fehean O'Doherty, a Catholic priest, writes "faith is neither subjective conviction nor experienced certitude, but may be at its best where doubt exists" (1973: 9). and Paul Tillich has said that faith necessarily includes an element ot uncertainty or doubt (1957: 16 ff). It is also of interest in this regard that Judaism does not require the devout to believe, for belief is not subject to command. It does, however, demand of them that they accept the law, and this acceptance is signalled by, and is intrinsic to, conformity to the ritual observances that pervade all of life.
Be this as it may, there may well be, and often are, disparities between the act of acceptance and the inward state associated with it. One can accept publicly not only that which one doubts but that which one privately despises or secretly denies. But if acceptance is intrinsic to performance it is not vitiated by secret denial. To recognize that secret denial may hide beneath the acceptance inhering in the act of performance is to recognize that the grounds of acceptance may vary widely, that acceptance is not necessarily founded upon belief, and that it does not even necessarily imply the subjective state termed "approval."
Acceptance, then, can be unconvinced and "insincere," but insincerity does not nullify acceptance. In what appears to be a flaw of sufficient seriousness to vitiate its meaningfulness lies the very virtue of acceptance through liturgical performance. Its social efficacy lies in its very lack of profundity, in the possibility of disparity between the outward act and the inward state. The distinction between belief and acceptance corresponds to the distinction made in the third chapter between the public and private. Participation in ritual demarcates a boundary, so to speak, between private and public processes. Liturgical orders, even those performed in solitude, are public orders and participation in them constitutes an acceptance of a public order regardless of the private state of belief of the performer. We may cite here Austin's views on a great range of performatives - promising, swearing, repudiating, commending, assessing - among which accepting is to be included:
we must not suppose ... that what is needed in addition to the saying of the words in such cases is the performance of some internal spiritual act, of which the words are then the report. It's very easy to slip into this view at least in difficult portentous cases ... In the case of promising - for example, "I promise to be there tomorrow" - it's very easy to think that the utterance is simply the outward and visible ... sign of the performance of some inward spiritual act of promising, and this view has been expressed in many classic places. There is the case of Euripedes' Hippolytus who said "My tongue swore to, but my heart did not" -perhaps it should be "mind" or "spirit" rather than "heart." but at any rate some kind of backstage artiste. (1970: 236)
It is gratifying to observe in this very example [that of Hippolytus] how excess of profundity, or rather solemnity, at once paves the way for immorality. For one who says "promising is not merely a matter of uttering words! It is an inward and spiritual act!" is apt to appear as a solid moralist standing out against a generation of theorizers: we see him as he sees himself, surveying the invisible depths of ethical space, with all the distinction of a specialist in the sui generis. Yet he provides Hippo-
122      Ritual and religion
Enactments of meaning      123
lytus with a let-out, the bigamist with an excuse for his "I do" and the welsher with a defence for his "I bet." Accuracy and morality alike are on the side of the plain saying that our word is our bond (Austin 1962: 10).
Acceptance in, or through, liturgical performance may reflect an inward state of conviction; it may also encourage "the mind," "the heart" and "the spirit" into agreement with itself. It does not necessarily do either, however, and therefore it does not eliminate all of the shenanigans of which the mind, the heart, the spirit, and other "backstage artistes" may be capable, but my argument, based on Austin's, proposes that although liturgical performance does not eliminate insincerity, it renders it publicly impotent. It is the visible, explicit, public act of acceptance, and not the invisible, ambiguous, private sentiment, which is socially and morally binding.
Because public and private processes are (and must be) related, but only loosely related, a range of what Austin (1962: 95ff., passim) called "infelicities" - insincerities and the like - are possible. But if, somehow, public orders could be required to depend upon the continuing belief, sincerity, goodwill, conviction or enthusiasm of those subject to them, the possibility of insincerity or deceit would surely be replaced by the high probability of non-order or disorder because of the near impossibility of meeting such a standard. This is not to say that the private processes may not be important in the dynamics of ritual. In a later chapter we shall take up belief and religious experience. It is simply to recognize that the private states of others are in their nature unknowable and even one's own attitudes may not always be easy to ascertain, for we are inclined to be ambivalent about matters of importance, like the conventions to which we are subordinate, and private states are likely to be volatile. "Common belief" cannot in itself provide a sufficiently firm ground upon which to establish public orders, even in very simple societies. We cannot know if a belief is common, for one thing, and whereas belief is vexed by ambivalence and clouded by ambiguity acceptance is not. Liturgical orders are public, and participation in them constitutes a public acceptance of a public order, regardless of the private state of belief. Acceptance is not only public but clear. One either participates in a liturgy or one does not; the choice is binary and as such it is formally free of ambiguity.9 While ritual participation may not transform the private state of the performer from one of "disbelief" to "belief," our argument is that in it the ambiguity, ambivalence and volatility of the private processes are subordinated to a simple and unambiguous public act, sensible both to the performers themselves and
t0 witnesses as well. Liturgical performance is, thus, a fundamental social act, for the acceptance intrinsic to it forms a basis for public orders which unknowable and volatile belief or conviction cannot.
That a liturgical order is accepted in its performance does not, furthermore, guarantee that the performer will abide by whatever rules or norms that order encodes. We all know that a man may participate in a liturgy in which commandments against adultery and thievery are pronounced, then pilfer from the poor box on his way out of church, or depart from communion to tryst with his neighbor's wife. To recognize such sordid realities is not to agree that liturgical acceptance is hypocritical, trivial or meaningless (Douglas 1973: 30), nor is it to dismiss claims for the social efficacy of acceptance through liturgical performance. It is, in fact, to affirm them, for such violations do not nullify acceptance, nor render it trivial. It is in such instances that the importance of ritual acceptance is most dramatically demonstrated. The primary function or metafunction of liturgical performances is not to control behavior directly, but rather to establish conventional understandings, rules and norms in accordance with which everyday behaviour is supposed to proceed. Participation in a ritual in which a prohibition against adultery is enunciated by, among others, himself may not prevent a man from committing adultery, but it does establish for him the prohibition of adultery as a rule that he himself has both enlivened and accepted. Whether or not he abides by that rule, he has obligated himself to do so. If he does not, he has violated an obligation that he himself has avowed. The assertion here is similar to those of Austin (see above) and of the philosopher John Searle, who has argued that
when one enters an institutional activity by invoking the rules of that institution one necessarily commits oneself in such and such ways, regardless of whether one approves or disapproves of the institution. In the case of linguistic institutions like promising [and accepting] the serious utterance of words commits one in ways which are determined by the meaning of the words. In certain first person utterances the utterance is the undertaking of an obligation. (1969:189)
Searle later notes that the notion of obligation is closely related to those of accepting, recognizing, acknowledging. This suggests that there is no obligation without acceptance, and perhaps that morality begins with acceptance.10 We may also note that while the acceptance of conventional undertakings, rules and procedures is possible outside of ntual, the formal and public nature of liturgical performance makes it very clear that an act of acceptance is taking place, that the acceptance is
124      Ritual and religion
Enactments of meaning 125
serious, and what it is that is being accepted. In Austin's terms (1962: passim) it is "explicitly performative." In sum, it is not ritual's office to ensure compliance but to establish obligation.
6. Performativeness, metaperformativeness, and the establishment of convention
We may now return to the assertion that although all ritual may not include simple performatives - conventional procedures for achieving conventional effects - something formally similar to simple performativeness, but of higher order, is intrinsic to ritual's form, and that this characteristic of ritual makes performatives possible.
Austin (1962: 26ff.) listed six conditions that must be fulfilled if performatives are to be successful (see also Searle 1969: passim). These include a number of obvious stipulations already noted at least tacitly -that they be performed by properly authorized persons under proper circumstances, and that they be executed correctly and completely. We have observed that the formality of ritual goes a long way to assure that those conditions are met, but ritual's contribution to performativeness is not limited to its service as a protocol, conformity to which assures full and correct performance. Its significance is much more fundamental.
Austin states the first and most basic condition for performative success, that which he labels A.l, as follows: "There must exist an accepted conventional procedure having a certain conventional effect, the procedure to include the uttering of certain words [or the performance of certain symbolic acts] by certain persons in certain circumstances" (1962: 14) (emphasis mine). Conventional effects cannot be achieved without conventions for achieving them. If young men are to be transformed into knights there must be a procedure for doing so, and this procedure must be acceptable to the relevant public. We may also note, although Austin does not, that the acceptance of a procedure for dubbing knights tacitly but obviously entails an acceptance of the convention of knighthood itself. Yet further, if the young man is "armed as a knight for the service of Christ" by priests or bishops, an acceptance of Christ's divinity is also entailed (Marc Bloch 1961: ch. 33).
Austin's basic condition seems obvious, but it is not trivial because its violation is possible. A performative attempt could misfire because, tor instance, no one but the performer recognized the procedure which it employed or the state of affairs it sought to achieve. An attempt by an American citizen living in the United States to win a divorce by repeating to his wife "I divorce you" three times would not rid him of matrimony s
burdens, nor would any divorce procedure whatsoever succeed in a society not recognizing divorce at all (Austin 1962: 27). Conventional procedures and conventional states, or even entire conventional codes, may be accepted by some and not by others and there are surely changes in the conventions of any society through time, with the scope of their applicability expanding or contracting, some disappearing altogether while others appear. Conventions can cease to exist because they are no longer accepted (Austin 1962: 30), as, for instance, in the case of the code of honor of which duelling was a part. It is unlikely, to say the least, that in contemporary United States or Great Britain a slap of a glove across a cheek would lead to a duel. The conventions of which this ritual act was an element are no longer accepted. They are, as we say, "history."
Austin stipulated as requisite to the effectiveness of performatives that relevant conventions exist and be accepted, but he gave only scant attention to the ways in which this prerequisite might be fulfilled. The argument being presented here is that ritual may fulfill it. To establish a convention - a general public understanding, a regular procedure, an institution - is both to ascribe existence to it and to accept it. The two are hardly distinct, as Austin (1962: 26) understood, for the existence of a convention, given the meaning of the word, is a function of its acceptance (see Bateson 1951: 212ff.).n To perform a liturgy is at one and the same time to conform to its order and to realize it or make it substantial. Liturgical performance not only recognizes the authority of the conventions it represents, it gives them their very existence. In the absence of performance liturgical orders are dead letters inscribed in curious volumes, or insubstantial forms evaporating into the forgotten. A ritual performance is an instance of the conventional order to which it conforms. Conversely, a ritual performance realizes the order of which it is an instance. Participants enliven the order that they are performing with the energy of their own bodies, and their own voices make it articulate. They thereby establish the existence of that order in this world of matter and energy; they substantiate the order as it informs them.
If performatives are to be understood as conventional procedures for achieving conventional effects, rituals are, by this account, more than simple performatives. We have already noted that a dubbing did more than transform a particular young man into a knight; it also repeatedly established (accepted the existance of) a conventional procedure for transforming young men into knights. It further established and reestablished the conventions of knighthood itself, and of the divinity of the god in whose name and service knights were dubbed. The mass, in
126      Ritual and religion
Enactments of meaning
127
contrast, establishes a more general conventional understanding of the relationship of humans to the divine. The act of acceptance intrinsic to ritual performance is not simply performative, as are specific conventional acts occurring within rituals - crowning, marrying, dubbing, purifying - but meta-performative. Rituals do more than achieve conventional effects through conventional procedures. They establish the conventions in terms of which those effects are achieved.
The establishment of convention is the second of ritual's fundamental offices. It is fundamental because all of ritual's simple performative functions are founded upon, or presume it. and so may conventional procedures outside of ritual itself. It is fundamental, this is to say. because the establishment of convention is what might be called a "we/afunction" making possible the fulfillment of particular functions by the particular conventions established.
It is fundamental in a second, formal sense, because the establishment as convention of whatever is encoded in canon is intrinsic to the form of ritual, that is, to the performance of more or less invariant sequences of formal acts and utterances not encoded by the performer. We observe here the profound importance of invariance and formality. These are the features that maintain constant that which is accepted. In the absence of such constancy that which is accepted would not be conventional. Indeed, acceptance would be inconsequential, meaningless, or even logically impossible if the canon were made up afresh by each participant for each performance.
We note in passing that as one of "the realities" lying behind notions of the magical power of words may be simple performativeness or factiveness mystified, so may widespread notions concerning the creative power of The Word rest upon meta-performativeness or meta-factiveness mystified, upon the realization of conventions through participation in invariant liturgical orders.
7. Ritual and daily practice in the establishment of convention
That the obligations clearly and explicitly accepted in liturgical performance are nullified by neither disbelief nor violation has a significance transcending the problems of insincerity and deceitfulness. We approach here a matter of profound importance, the relationship of convention to behavior and, more particularly, difficulties in establishing convention through ordinary usage. I have argued that one of ritual's fundamental offices is the establishment of convention, but no claim was made that convention is established only in ritual. It therefore may seem in this
regard that ritual is no different from usage or practice in general. Convention, be it noted, may also be established by decree. There are important differences, however, between liturgical performance and other means for establishing convention.
First, we may contrast liturgical performance with quotidian practice. As Bateson (1951: 214) long ago remarked, "every statement in a given codification is an affirmation of that codification and is therefore in some degree metacommunicative (when I say T see the cat' I am implicitly affirming the proposition that the word 'cat' stands for what I see)". The core meaning of the term "code" is linguistic, in some usages it denotes vocabularies and the rules for combining their elements into larger meaningful units without reference to or restriction upon what can or may he said. While linguistic conventions may be taken to be paradigmatic of those established in ordinary practice, it should be kept in mind that the concept of code has been extended beyond language by some cognitive anthropologists, who tend to see cultures generally as complex codes made up of "shared finite cognitive set[s] of rules for the socially appropriate construction and interpretation of messages and behaviour" (Kernan 1972: 333, cf. Frake 1964). But a liturgy is not a code in this wide and semantically unspecifying sense. It is a more or less fixed sequence of stereotyped actions and utterances and as such what can be expressed in it is narrowly circumscribed. Permissible variations in some aspects of performance do allow or require the participants to encode indexical messages, but the content of the canon, the invariant aspect of the liturgy in respect to which the indexical messages may vary, is fixed, and therefore the range of indexical messages that may be transmitted in any liturgy is restricted. Moreover, as we have noted, the participants do not encode but only transmit the invariant messages the canon embodies. Since this is the case, the term "liturgical order"' seems more appropriate than "liturgical code." It follows that the acceptance of an order, because it is in its nature highly restrictive, is therefore more socially consequential and significant than the affirmation of a more or less unrestrictive code. High valuation of the qualities of the consequential and the restrictive, as previous discussion suggests, invites the application of liturgical order, for liturgy tends to make explicit precisely what is being stipulated, it is in its very form constricting, and further, liturgical form and decorum tends to make its substance seem grave.
It would, of course, be mistaken to impose the simple dichotomous distinction of "codes" versus "orders" upon the conventions organizing social life. Linguistic codes and the conventional dogmatic understand-
128      Ritual and religion
Enactments of meaning 129
ings embodied in some liturgies stand at opposite ends of a continuum of constraint. Between them are stretched the conventions which organize not merely what people say to or about each other or the world, but what they do to and with each other and the world around them. Little is known about the order of this continuum, whether certain domains of culture are more likely to be subject to narrower or more rigorous constraints than others, or whether variation from one society to the next is wide. We shall return to this general question in a later chapter when we discuss sanctification. For now it may be noted that comparatively high degrees of constraint seem entailed by the invariant nature of liturgy, and it may further be suggested that the more highly motivated people are to violate a convention or the more consequential its violation is deemed to be, the more likely it is to be established in liturgy than in daily practice, or the more closely and strongly will it be associated with conventional understandings that are so represented (see the discussion of sanctification in chapter 10 below). Be this as it may, certain conventions, for instance, those of speech, emerge out of ordinary usage and are maintained by ordinary usage in sufficient stability to allow meaningful and orderly social interaction. In such cases, "the norm is identical with the statistical average" (Leach 1972: 320). It may be suggested that variation with respect to such conventions can be comfortably tolerated and day-to-day usage may be allowed to establish, maintain or change them. But ordinary practice or usage is not in itself sufficient to establish all conventions, nor are statistical averages arising out of behavior always coextensive with conventions.
First, it is impossible for ordinary usage to establish conventions to which no ordinary usage corresponds. Such conventions include most importantly, and perhaps exclusively, the understandings upon which religions are founded, dogmas and mysteries concerning gods and the like which, being typically without material referents and always being taken to be extraordinary, cannot grow out of ordinary usage. The fundamental importance of these conventions will be discussed in later chapters.
Secondly, statistical averages arising out of usage represent no more than common practice, summations of behavior, and the utility ot summations of behavior for guiding or assessing the behavior of which they are summations is limited, at best. They are particularly inadequate, first, in the case of conventions with moral import, for they tend to reduce the notion of immorality to deviation from a statistically average behavioral range - to that which "is not done," and the moral to that
which "is done." Such a notion of morality and immorality is, on the one hand, descriptively erroneous - no society operates with such a conception of morality - and, on the other hand, operatively inadequate. No society could so operate, at least for long, because it would be without any means for assessing common practice itself, and a common practice may, even in terms of the moral code of the society in which it occurs, be vicious, anti-social or self-destructive. Common practice, statistical average, ordinary usage will have difficulty establishing conventions concerning aspects of social life that are obviously restrictive, obviously arbitrary, highly charged emotionally, especially dangerous, or require obedience conventions, this is to say, that demand of individuals that they subordinate their self-interests to the common good. Behavioral variation may be less tolerable with respect to these matters than with respect to linguistic usage, and uncertainty as to the precise nature of the conventions themselves, a different matter, may be even less tolerable than variations in the practices which they presumably direct. Ordinary usage always varies, and in ordinary usage rules and conventions are frequently violated. Leach was generally pointing in the right direction but did not go far enough when he suggested that "if anarchy is to be avoided, the individuals who make up a society must from time to time be reminded of the underlying order that is supposed to guide their social activities. Ritual performances have this function for the group as a whole. They momentarily make explicit what is otherwise fiction" (1954: 16). Although usage may not be faithful to it, that which is represented in a liturgical order is not a fiction, (except in the sense of being "made up," rather than given by non-human nature, see Geertz 1973: 15), and the performance does more than remind individuals of an underlying order.
It is well to make explicit an assumption tacit in our general argument. The orders of societies, like the order of the universe in general, tend to degenerate into disorder. Their material elements disintegrate or decay into non-functioning fragments if they are not maintained, and their meaningful elements, including conventional understandings and rules, dissolve into error, nonsense, ambiguity, vagueness, hypocrisy and mean-inglessness unless continually clarified, corrected and re-established. Far from clarifying and reasserting conventions, the vagaries of practice may tend to erode them. It is therefore necessary to establish at least some conventions in a manner which protects them from dissolution in the variations of day-to-day behavior and the violations in which history abounds. Liturgy does not simply remind people of the orders which usage - behavior and history - violates and dissolves. It establishes and
130      Ritual and religion
Enactments of meaning 131
ever again reestablishes those orders. Liturgy preserves the conventions it encodes inviolate in defiance of the vagaries of ordinary practice, thereby-providing them with existence independent of, and insulated against, the statistical averages which characterize behavior. That "everyone does it" exonerates no one. For people of the Book adultery would remain a sin even if every married person indulged in it.
It should be noted, however, that the violation of convention is not always simply a matter of entropy, chaos or anarchy asserting itself against an ideal but not fully realized order. The constitution of some societies is such that the violation of some conventions is not only frequent but systematic, and yet the convention has a vital part to play in the life of the society. Gluckman long ago (1954) considered certain African rituals in this light, and we may note that among the Maring and other Highland New Guinea peoples, a strong patrilocal patrilineal ideology prevails. It is putatively patrilineal clans that hold territories and putatively patrilineal sub-clans that claim smaller tracts (Rappaport 1968: ch. 2). Rituals, addressed largely to patrilineal ancestors, are conducted by these groups at special places on the land which they and their deceased ancestors are said to occupy together. But the exigencies of life and death are such that the demographic fortunes of these small groups (sub-clans among the Tsembaga ran, in 1963, from almost none to about thirty-five persons, clans from sixteen to seventy persons), fluctuate widely, and fluctuations may lead to the violation of the patrilineal patrilocal ideal. Groups must maintain their strength vis-a-vis their neighbors, and when their numbers are low the members of a group will attempt to attract outsiders to settle among them. Their kinship terminology, which is Iroquois on ego's generation, but generational on all descending generations, as well as on the second ascending generation and above, seems well suited to the assimilation of strangers, obliterating as it does distinctions between agnates, other cognates and affines in two generations, and rumbim planting can be seen as the beginning of a process by which cognates are transformed into agnates. This ritual transformation of non-agnates into agnates is able to preserve the conventions of patrilocality and patrilineality, if not patrifiliation, inviolate in the face of continual violation in usage (see LiPuma 1990). This is of considerable adaptive importance. Densities sometimes become high in the New Guinea highlands and therefore it sometimes becomes necessary or desirable to exclude people who would like to immigrate (Meggitt 1965b). Agnatic rules, necessarily violated by practice but preserved by ritual, provide a basis for such exclusion when there is need.
The formality definitive of ritual and distinguishing it from ordinary behavior is clearly of importance in preserving the conventions it encodes from the errors and trespasses of daily practice. Because preservation is virtually entailed by ritual's formality, and because the acceptance as well as the precise stipulation of convention is intrinsic to ritual's form, ritual may well be without functional, or metafunctional, equivalents.
Ritual, to be sure, is not altogether unique in establishing conventions at the same time that it insulates them from the variations and violations of behavior. Conventions may also be promulgated by decree and maintained by force. But the acceptance of those subject to a decree is not intrinsic to the promulgation of that decree. In contrast, it is one and the same ritual act that both realizes and accepts a liturgical order. Furthermore, the act of acceptance establishes an obligation with respect to the convention accepted, an obligation that is not specifically undertaken and may not be felt by those subject to decrees.
It is of interest in this regard that even in those instances in which conventions are self-consciously promulgated by kings or parliaments the act of promulgation and those participating in it are surrounded by ritual. Kings are crowned, public officers sworn into office, meetings of parliaments are ceremonially opened and closed, and their deliberations set within a more or less invariant procedure. Moreover, their decrees may be accepted, albeit indirectly and non-specifically, by those subject to them in such ritual acts as pledges of allegiance to the entities, or symbols thereof, from which the promulgators derive their authority. To accept an order is to ascribe legitimacy to its terms. To ascribe legitimacy to its terms is to oblige oneself to abide by them, or to put it a little differently, to agree to their application as a set of standards against which the acceptor's own actions are properly judged. Insofar as participation in a liturgical order is an acceptance of that order, it legitimizes that order.
It may also be suggested, although there is no way of knowing, that ritual, in the very structure of which both authority and acquiescence are implicit, may well be the primordial means by which humans have established conventions. The conditions that make it possible for some men to promulgate conventions by directives to which other men must conform seem to have developed relatively recently, probably not antedating by much, if at all, the appearance of plant and animal cultivation 10,000 years or so ago.12 Ritual, on the other hand, does not require superordinate human authorities to establish conventions and must have antedated procedures that do.
132      Ritual and religion
Enactments of meaning       133
8. The morality intrinsic to ritual's structure
The performance of a liturgy not only brings conventions into being but invests them with morality. Moral dicta are not explicit in all liturgies, but morality, like social contract, is implicit in ritual's structure.
We have, following Searle, noted that obligation is entailed by the acceptance intrinsic to participation in ritual. Breach of obligation, it could be argued, is one of the few acts, if not, indeed, the only act that is always and everywhere held to be immoral. Homicide, for instance, is not. There are conditions, so common as to require no illustration, under which killing humans is laudable or even mandatory. What is immoral is, of course, killing someone whom there is an obligation, at least tacit, not to kill. A similar point can be made about most or possibly even all other specific acts generally taken to be immoral. Breach of obligation is of a higher order of generality than any such specific breach as murder, rape or robbery and it may be suggested that it is breach of obligation that transforms otherwise morally positive, neutral or empty acts into crimes such as murder or robbery. Breach of obligation may, then, be the fundamental immoral act, the element in the absence of which an act cannot be construed to be immoral, in the presence of which it is ipso facto immoral. The topic is a difficult one, and surely cannot be settled here. I will only emphasize that failure to abide by the terms of an obligation is universally stigmatized as immoral. To the extent, then, that obligation is entailed by the acceptance intrinsic to the performance of a liturgical order, ritual establishes morality as it establishes convention. The establishment of a convention and the establishment of its morality-are inextricable, if they are not, in fact, one and the same.
We may refer again in this regard to the relationship of performatives to the states of affairs with which they are concerned. Austin initially tried to say that performatives differ from statements in that performatives are neither true nor false (1970: 233ff.) whereas statements are either true or false. Later he found this view to be questionable because certain performatives, notably verdictives, are supposed to stand in a relationship to states of affairs similar to that of true statements to states or affairs. Later he found this not always to be the case. Performatives do, however, differ from statements in a related way which he did not note but which does have to do with truth, and with the foundations of morality.
In discussing their indexical nature, we have observed that the relationship of performatives to the states of affairs with which they are concerned is the inverse of that of statements or descriptions. Statements
report autonomously existing states of affairs. Performative acts realize states of affairs. The inverse nature of these relationships has obvious implications for assessment. The adequacy of a descriptive statement is assessed by the degree to which it conforms to the state of affairs that it purports to describe. If it is in sufficient conformity we say that it is true, accurate or correct. If it is not we say that it is false, erroneous, inaccurate or lying. The state of affairs is the criterion by which the truth, accuracy or adequacy of a statement is assessed. In the case of performatives there is an inversion. If, for instance, a man is properly dubbed to knighthood and then proceeds to violate all of the canons of chivalry, or if peace is declared in a properly conducted ritual but soon after one of the parties to the declaration attacks the other, we do not say that the dubbing or the peace declaration were faulty, but that the subsequent states of affairs are faulty. We judge the state of affairs by the degree to which it conforms to the stipulations of the performative act. Liturgical orders provide criteria in terms of which events - behavior and history -may be judged. As such, liturgical orders are intrinsically correct or moral. Morality is inherent in the structure of liturgical performance prior to whatever its canons explicitly assert about morality in general or whatever in particular may be taken to be moral. Morality derives ultimately not from statements about what may be right and wrong but from what liturgy establishes as right or wrong. To put it a little differently, to establish a convention independent of usage is to establish an "ought" against which the "is" of behavior may be judged.13
The establishment^ morality is clearest in the case of simple factive-ness and the commissive implications thereof. It is patently immoral to act incompatibly with the terms of a conventional state of affairs that one has ritually participated in bringing into being. My argument implies, however, that morality is also intrinsic to the meta-factiveness of jritual, that is, to the establishment of particular conventions and conventional orders. One who violates not merely the terms of a conventional state of affairs, but of the conventional order defining such states of affairs is not guilty of a simple immoral act, but of apostasy. It is of interest here that in Zoroastrian Persia and Vedic India states of affairs that departed from the proper liturgically established order were designated by terms that also seem to have meant "lie," druj in Persia and anrta in India (Duchesne-Guillemin 1966: 26ff., N. Brown 1972: 252ff., Orlin 1976). What may be called "Vedic lies" or "Zoroastrian lies" (Rappaport 1979b), states of affairs that their perpetrators are aware do not conform to prevailing liturgically established orders, are the inverse
134      Ritual and religion
Enactments of meaning 135
of "vulgar lies," statements that their transmitters believe misrepresent the states of affairs which they purport to report.
9. Ritual and myth, and drama
-Acceptance entails neither belief nor obedience. To say that in performing a canon the participant accepts whatever conventional understandings, principles, rules or procedures it encodes is simply to say that he has obligated himself to abide by its terms regardless of his private opinions and feelings about them. In slightly different terms, an act of acceptance invests the objects accepted with the qualities of correctness, propriety, legitimacy and morality, and thus establishes them as criteria in terms of which common practice, behavior, events, history and especially the acceptor's own conduct, may be judged.
It is obvious but nevertheless worth making explicit, if only to take issue with a loose anthropological truism, that the ritual relationship of performers to what they are performing distinguishes ritual from myth on substantive as well as formal grounds. Ritual actions cannot be seen as simply "exemplifying in another medium the cultural values that find verbal expression in statements about the world, society, man - statements which we call beliefs and which are elaborated in narratives or myths" (La Fontaine 1972: xvii). And while it may be, as in Leach's phrase of four decades ago (1954: 12), that "myth is the counterpart of ritual, myth implies ritual, ritual implies myth," they are not, to complete the famous dictum, "one and the same." They are never one and the same, even when they are about the same things, or even when the ritual is simply an enactment of the myth. Myth as such carries no self-referential information, nor does its telling either presuppose or establish any particular relationship between the myth and he or she who recounts it. The narrator may tell it as priest to novice, entertainer to audience, sorcerer to apprentice, father to child, French structuralist to students, literary critic to ladies' club, folklore collector to those who read his anthology. This is to say that the telling of a myth, even in the case of the priest, does not necessarily imply that the teller accepts the myth as part of an order in which she herself participates, or to which she subordinates herself. The relationship of narrators to the myths they narrate is often unknown to auditors or readers, and is, at any rate, irrelevant to them. What may be relevant to them is the substance of the myths themselves, something which can as well be read as seen, heard, voiced or acted. As narrators do not necessarily accept the myths which they narrate as anything more than stories, neither do their audiences, and such accept-
ance as they might accord to such stories is responsive to their perlocu-tionary force, the ability of the stories to move them, and not through conformity to their form. Myths, like rituals, can "die" (Eliade 1963), but they do not, as do rituals, become dead letters if they are preserved only in writing.
In contrast to myths, rituals even when they seem to be no more than detailed reenactments of myths always stipulate a relationship between performers and that which they perform. Such rituals communicate more than their myths. They communicate the indexical message of the participants' acceptance of those myths as well.
It may be suggested that myth and drama are closer to being "one and the same" than are myth and ritual. A drama based upon a myth is no more nor less than an enactment of a version of that myth. As such, its performance, unlike ritual performance, does not indicate acceptance of the narrative being played. But ritual and drama have sometimes been taken to be closely related in one way or another and Jane Harrison long ago (1913), observed that the origins of Western drama lay in ritual. She observed that the term "drama" comes from the Greek dromenon, literally, "thing done," but early denoting religious ritual. It may be well, in view of their putative relationship or similarity, to note in more detail than was appropriate in chapter 2 some further differences between ritual and drama.
Perhaps most important is that which distinguishes an audience from a congregation. The congregation participates in the ritual, with all that participation entails. The audience at a performance of a Western drama merely watches and listens. It is present for the performance, but is not part of it. A congregation is generally required to do things in the course of a ritual: sing, dance, read responsively, kneel, eat, drink. In contrast, the members of a Western audience are not required to do anything and may even be required to do nothing. Whereas a congregation joins the celebrant in performing the acts that comprise the ritual, an audience does not join the actors in the performance of a drama. The actors act on one side of the proscenium arch, the audience refrains from action on the other.
Secondly, the acts of those who celebrate rituals express or enliven the orders to which the congregation acquiesces. But in theatre, actors play parts which, when they come together in the totality of the drama, comprise fictions which none of those present need or is expected to accept as anything other than fiction. Even when audiences accept
136      Ritual and religion
Enactments of meaning
137
dramas as great fiction, true as some of Shakespeare's plays are true, they do not accept them as literal reality, in some degree concrete, of which they are parts, but as representations of some sort. As in the case of myth audience, acceptance rests upon the ability of drama to persuade or move, and not upon its demand to act in conformity to its form. In Austin's terms, drama, like myth without ritual, has at best perlocu-tionary but not illocutionary force.
It is interesting in this regard to compare the "acts" of those who participate in rituals and those who "act" in drama. For one who performs a ritual "to act" is to take an action that affirms or even brings into being a significant order and also states his acceptance of it. It may even transform that order or himself. The ritual act, this is to say, "does something," it is an action that is meant to affect the world and it is likely to do so. To act in a drama, in contrast, is not to take an action affecting the world, but only to imitate doing so. That acting in a drama is not acting in the non-dramatic sense is, moreover, clearly signalled to those present by a whole set of context markers setting the dramatic action apart from"real life": the seats, the curtain, the program providing the worldly names of those who, for an hour or two, will act out a temporary identity in the playwright's words and the director's gestures. To act in the dramatic sense is precisely not to act in the non-dramatic sense. This contrast is strongly suggested by the alternative term for drama in English. It is "play," a term which, of course, also denotes lack of earnestness. Whereas a worshipper takes part in a ritual, thus participating in the enduring order that his own performance helps bring into being, an actor plays a part in a play, a part which evaporates when the curtain falls and when his own identity is supposed to return to guide his actions once again.
We note here what distinguishes ritual from drama, but of course, particular events are not always purely one or the other. Some performances include elements of both, or better, stand somewhere on a continuum lying between the polar forms. We may think here of miracle and passion plays and concert performances of religious music in churches. The effectiveness of some performances may arise out of the ambiguities of this continuum, and so may the failures of others. "Living theatre," for instance, is likely to be unsuccessful because an audience is asked to take upon itself congregation-like duties while, lacking the well-rehearsed certainties of liturgy, not knowing in what they are being asked to participate or with whom they are participating. But surely, through time, the character of some performances changes, the relationship ot
those present being transformed from that of congregation to audience or audience to congregation (see Kapferer 1983: esp. ch. 8). Harrison argues (1913: 35-38) that the transformation from dromenon to drama in ancient Greece may be traced archaeologically through changes in the use of space occurring during the fifth and sixth centuries BC. At the beginning of the period there was only the orchestra, a round area in which everyone present joined together in the dances of the dithyramb, the spring festival. Later, however, a theatre, that is to say rows of seats, came to tier the hillside above the orchestra, and spectators came to be separated from dancers and actors. According to Harrison this separation from the action in space was concurrent with and inseparable from a trend toward the detachment of those present from the action taking place, and she distinguished ritual from drama on the grounds of the distinction between participation and contemplation, that is to say, congregation and audience. Conversely, in our own day there seems to be a transformation of audiences into, or at least in the direction of, congregations in certain performances, notably rock concerts. In the absence of canon in these events it is not surprising that their stars are virtually apotheosized.
10. Ritual as the basic social act
To summarize, the existence of a conventional order is contingent upon its acceptance; in fact a rule or understanding cannot be said to be a convention unless it is accepted. In ritual, however, acceptance and existence entail each other, for a liturgical order is perforce accepted in its realization, in, that is to say, the performance which gives it substance. Since obligation is entailed by acceptance, and the breaking of obligation is per se immoral, the existence, acceptance and morality of conventions are joined together indissoluably in rituals; they are, in fact, virtually one and the same. The same cannot be said of principles, rules, procedures or understandings established by proclamation, or legislation on the one hand or by daily practice on the other. This is to say that there is a logically necessary relationship between the form which is ritual, the performance of more or less invariant sequences of formal acts and utterances not encoded by the performers, and the messages rituals contain concerning both what is performed and the relationship of the performer to what he performs. Ritual is not merely another way to "say things" or "do things" that can be said or done as well or better in other ways. The form which is ritual is surely without communicational equivalents and thus, possibly, without functional or metafunctional
138      Ritual and religion
equivalents. That ritual's abilities are intrinsic to its form and in indissoluble association only with its form, goes far to account for its ubiquity.
In attending to ritual's form we must not lose sight of the fundamental nature of what it is that ritual does as a logically necessary outcome of its form. In enunciating, accepting and making conventions moral, ritual contains within itself not simply a symbolic representation of social contract, but tacit social contract itself. As such, ritual, which also establishes, guards, and bridges boundaries between public systems and private processes, is the basic social act.