92 THE FOREST OF SYMBOLS
Classes and the Geometrical Theory of Points," Proceedings of the London Mathematical Society, XXI.
Laming, Annette. 1959. Lascaux. Harmondsworth: Penguin Books.
Leakey, L. S. B. 1931. The Stone Age Cultures of Kenya Colony, London: Oxford University Press.
Leib, Arthur. 1946. Folklore, LVIL 128-133.
Mountford, Charles P. 1962. Oceania and Australia (Art of the World
Series, VIII). London: Methuen. Needham, Rodney, i960. "The Left Hand of the Mugwe," Africa, XXX,
No. 1.
Nikhilinananda, Swami, trans. 1963. Upanishads. New York: Harper, Row.
Richards, A. I. 1956. Ckisungu. London: Faber & Faber.
Skeat, W. W., and Blagden, C. O. 1906. PagaK Races of the Malay Peninsula. London: Macmillan.
Smith, Robertson W. 1912. "A Journey in the Hedjaz." In J. S. Black and G. Crystal, eds., Lectures and Essays. London: A. & C. Black.
Spence, Lewis. 1911. Article on "Cherokees," Hastings Encyclopaedia of Religion and Ethics, III, 506-507.
Summers, Roger. 1958. Inyanga. London: Cambridge University Press.
Turner, Victor W. 1961. Ndembu Divination: Its Symbolism and Techniques (Rhodes-Livingstone Paper 31). Manchester University Press.
Vergiat, A. M. 1936. Les Rites Secrets des Primitifs de VOuhangui. Paris: Payot.
Wescott, Joan. 1962. "The Sculpture and Myths of Eshu-Elegba, The Yoruba Trickster," Africa, XXXII, No. 4.
White, C. M. N. 1961. Elements in Luvale Beliefs and Rituals (Rhodes-Livingstone Institute Paper 32). Manchester University Press.
Zaehner, R. C. 1962. Hinduism. London: Oxford University Press.
CHAPTER IV
Betwixt and Between: The Liminal Period in Rites de Passage*
In this paper, I wish to consider some of the sociocultural properties of the "liminal period" in that class of rituals which Arnold van Gennep has definitively characterized as "rites de fassage." If our basic model of society is that of a "structure of positions," we must regard the period of margin or "liminality" as an interstructural situation. I shall consider, notably in the case of initiation rites, some of the main features of instruction among the simpler societies. I shall also take note of certain symbolic themes that concretely express indigenous concepts about the nature of "interstructural" human beings.
Rites de passage are found in all societies but tend to reach their maximal expression in small-scale, relatively stable and cyclical societies, where change is bound up with biological and meteorological rhythms and recurrences rather than with technological innovations. Such rites indicate and constitute transitions between states. By "state" I mean here "a relatively fixed or stable condition" and would include in its meaning such social constancies as legal status, profession, office or calling, rank or degree. I hold it to designate also the condition of a person as determined by his culturally recognized degree of maturation as when one speaks of "the married or single state" or the "state of infancy." The term "state" may also be applied
* Read at the Annual Meeting of the American Ethnological Society in Pittsburgh, March 1964. First published in The Proceedings of the American Ethnological Society (1964).
93
94
THE FOREST OF SYMBOLS to ecological conditions, or to the physical, mental or emotional con tion in which a person or group may be found at a particular time \ man may thus be in a state of good or bad health; a society in a state of war or peace or a state of famine or of plenty. State, in short, is a more inclusive concept than status or office and refers to any type of stable or recurrent condition that is culturally recognized. One may, I S1]p, pose, also talk about "a state of transition," since J. S. Mill has, after all, written of "a state of progressive movement," but I prefer to reg; transition as a process, a becoming, and in the case of rites de passt even a transformation—here an apt analogy would be water in process' of being heated to boiling point, or a pupa changing from grub to moth. In any case, a transition has different cultural properties from those of a state, as I hope to show presently.
Van Gennep himself defined "rites de passage" as "rites which accompany every change of place, state, social position and age." To point up the contrast between "state" and "transition," I employ "state" to include all his other terms. Van Gennep has shown that all rites of transition are marked by three phases: separation, margin (or limen), and aggregation. The first phase of separation comprises symbolic behavior signifying the detachment of the individual or group either from an earlier fixed point in the social structure or a set of cultural conditions (a "state"); during the intervening liminal period, the state of the ritual subject (the "passenger") is ambiguous; he passes through a realm that has few or none of the attributes of the past or coming state; in the third phase the passage is consummated. The ritual subject, individual or corporate, is in a stable state once more and, by virtue of this, has rights and obligations of a clearly defined and "structural" type, and is expected to behave in accordance with certain customary norms and ethical standards. The most prominent type of rites de passage tends to accompany what Lloyd Warner (1959, 303) has called "the movement of a man through his lifetime, from a fixed placental placement within his mothers womb to his death and ultimate fixed point of his tombstone and final containment in his grave as a dead organism—punctuated by a number of critical moments of transition which all societies ritualize and publicly mark with suitable observances to impress the significance of the individual and the group on living members of the community. These are the important times of birth, puberty, marriage, and death." However, as van Gennep, Henri Junod, and others have shown, rites de passage are not confined to culturally defined life-crises but may
gTWlXT AND BETWEEN: LIMINAL PERIOD 95 mpany any change from one state to another, as when a whole aCjbe g°es t0 war' 01 w^en ^ atEests to tne passage from scarcity to *knty by performing a first-fruits or a harvest festival. Rites de yas-
ae too, are not restricted, sociologically speaking, to movements between ascribed statuses. They also concern entry into a new achieved status, whether this be a political office or membership of an exclusive club or secret society. They may admit persons into membership of a religious group where such a group does not include the whole society, or qualify them for the official duties of the cult, sometimes in a graded series of rites.
Since the main problem of this study is the nature and characteristics of transition in relatively stable societies, I shall focus attention on rites de passage that tend to have well-developed liminal periods. On the whole, initiation rites, whether into social maturity or cult membership, best exemplify transition, since they have well-marked and protracted marginal or liminal phases. I shall pay only brief heed here to rites of separation and aggregation, since these are more closely implicated in social structure than rites of liminality. Liminality during initiation is, therefore, the primary datum of this study, though I will draw on other aspects of passage ritual where the argument demands this. I may state here, partly as an aside, that I consider the term "ritual" to be more fittingly applied to forms of religious behavior associated with social transitions, while the term "ceremony" has a closer bearing on religious behavior associated with social states, where politico-legal institutions also have greater importance. Ritual is transformative, ceremony confirmatory.
The subject of passage ritual is, in the liminal period, structurally, if not physically, "invisible." As members of society, most of us see only what we expect to see, and what we expect to see is what we are conditioned to see when we have learned the definitions and classifications of our culture. A society's secular definitions do not allow for the existence of a not-boy-not-man, which is what a novice in a male puberty rite is (if he can be said to be anything). A set of essentially religious definitions co-exist with these which do set out to define the structurally indefinable "transitional-being." The transitional-being or "liminal persona" is defined by a name and by a set of symbols. The same name is very frequently employed to designate those who are being initiated into very different states of life. For example, among the Ndembu of Zambia the name mwadi may mean various things: it may stand for "a boy novice in circumcision rites," or "a chief-
96 THE FOREST OF SYMBOLS
designate undergoing his installation rites," or, yet again, "the first or ritual wife" who has important ritual duties in the domestic family Our own terms "initiate" and "neophyte" have a similar breadth of reference. It would seem from this that emphasis tends to be laid on the transition itself, rather than on the particular states between which it is taking place.
The symbolism attached to and surrounding the liminal persona is complex and bizarre. Much of it is modeled on human biological processes, which are conceived to be what Levi-Strauss might call "isomorphic" with structural and cultural processes. They give an outward and visible form to an inward and conceptual process. The structural "invisibility" of liminal fersonae has a twofold character. They are at once no longer classified and not yet classified. In so far as they are no longer classified, the symbols that represent them are, in many societies, drawn from the biology of death, decomposition, ca-tabolism, and other physical processes that have a negative tinge, such as menstruation (frequently regarded as the absence or loss of a fetus). Thus, in some boys' initiations, newly circumcised boys are explicitly likened to menstruating women. In so far as a neophyte is structurally "dead," he or she may be treated, for a long or short period, as a corpse is customarily treated in his or her society. (See Stobaeus' quotation, probably from a lost work of Plutarch, "initiation and death correspond word for word and thing for thing." [James 1961, 132]) The neophyte may be buried, forced to lie motionless in the posture and direction of customary burial, may be stained black, or may be forced to live for a while in the company of masked and monstrous mummers representing, inter alia, the dead, or worse still, the un-dead. The metaphor of dissolution is often applied to neophytes; they are allowed to go filthy and identified with the earth, the generalized matter into which every specific individual is rendered down. Particular form here becomes general matter; often their very names are taken from them and each is called solely by the generic term for "neophyte" or "initiand." (This useful neologism is employed by many modern anthropologists).
The other aspect, that they are not yet classified, is often expressed in symbols modeled on processes of gestation and parturition. The neophytes are likened to or treated as embryos, newborn infants, or sucklings by symbolic means which vary from culture to culture. I shall return to this theme presently.
The essential feature of these symbolizations is that the neophytes
bETWIXT AND BETWEEN: LIMINAL PERIOD 97
are neither living nor dead from one aspect, and both living and dead from another. Their condition is one of ambiguity and paradox, a confusion of all the customary categories. Jakob Boehme, the German mystic whose obscure writings gave Hegel his celebrated dialectical "triad," liked to say that "In Yea and Nay all things consist." Liminal-ity may perhaps be regarded as the Nay to all positive structural assertions, but as in some sense the source of them all, and, more than that, as a realm of pure possibility whence novel configurations of ideas and relations may arise. I will not pursue this point here but, after all, Plato, a speculative philosopher, if there ever was one, did acknowledge his philosophical debt to the teachings of the Eleusinian and Orphic initiations of Attica. We have no way of knowing whether primitive initiations merely conserved lore. Perhaps they also generated new thought and new custom.
Dr. Mary Douglas, of University College, London, has recently advanced (in a magnificent book Purity and Danger [1966]) the very interesting and illuminating view that the concept of pollution "is a reaction to protect cherished principles and categories from contradiction." She holds that, in effect, what is unclear and contradictory (from the perspective of social definition) tends to be regarded as (ritually) unclean. The unclear is the unclean: e.g., she examines the prohibitions on eating certain animals and crustaceans in Leviticus in the light of this hypothesis (these being creatures that cannot be unambiguously classified in terms of traditional criteria). From this standpoint, one would expect to find that transitional beings are particularly polluting, since they are neither one thing nor another; or may be both; or neither here nor there; or may even be nowhere (in terms of any recognized cultural topography), and are at the very least "betwixt and between" all the recognized fixed points in space-time of structural classification. In fact, in confirmation of Dr. Douglas's hypothesis, liminal fersonae nearly always and everywhere are regarded as polluting to those who have never been, so to speak, "inoculated" against them, through having been themselves initiated into the same state. I think that we may perhaps usefully discriminate here between the statics and dynamics of pollution situations. In other words, we may have to distinguish between pollution notions which concern states that have been ambiguously or contradictorily defined, and those which derive from ritualized transitions between states. In the first case, we are dealing with what has been defectively defined or ordered, in the second with what cannot be defined in
98 THE FOREST OF SYMBOK
static terms. We are not dealing with structural contradictions whi we discuss liminality, but with the essentially unstructured (which V is at once destructured and prestructured) and often the people them- % selves see this in terms of bringing neophytes into close connection with deity or with superhuman power, with what is, in fact, often r""; regarded as the unbounded, the infinite, the limitless. Since neophytes "I are not only structurally "invisible" (though physically visible) and -ritually polluting, they are very commonly secluded, partially or com- ^ pletely, from the realm of culturally defined and ordered states and " statuses. Often the indigenous term for the liminal period is, as among Ndembu, the locative form of a noun meaning "seclusion site" (kun- * kunka, kung'ula). The neophytes are sometimes said to "be in another place." They have physical but not social "reality," hence they have to be hidden, since it is a paradox, a scandal, to see what ought not to be there! Where they are not removed to a sacred place of concealment they are often disguised, in masks or grotesque costumes or striped with white, red, or black clay, and the like.
In societies dominantly structured by kinship institutions, sex distinctions have great structural importance. Patrilineal and matrilineal moieties and clans, rules of exogamy, and the like, rest and are built up on these distinctions. It is consistent with this to find that in liminal situations (in kinship-dominated societies) neophytes are sometimes treated or symbolically represented as being neither male nor female. Alternatively, they may be symbolically assigned charac- i teristics of both sexes, irrespective of their biological sex. (Bruno Bettelheim [1954] has collected much illustrative material on this point from initiation rites.) They are symbolically either sexless or bisexual and may be regarded as a kind of human ■prima materia—as undifferentiated raw material. It was perhaps from the rites of the Hellenic mystery religions that Plato derived his notion expressed in his Symposium that the first humans were androgynes. If the liminal period is seen as an interstructural phase in social dynamics, the symbolism both of androgyny and sexlessness immediately becomes intelligible in sociological terms without the need to import psychological (and especially depth-psychological) explanations. Since sex distinctions are important components of structural status, in a structureless realm they do not apply.
A further structurally negative characteristic of transitional beings is that they have nothing. They have no status, property, insignia, secular clothing, rank, kinship position, nothing to demarcate them
BETWIXT AND BETWEEN: LIMINAL PERIOD 99
structurally from their fellows. Their condition is indeed the very totype 0f Sacred poverty. Rights over property, goods, and services inhere in positions in the politico-jural structure. Since they do not occupy such positions, neophytes exercise no such rights. In the words of King Lear they represent "naked unaccommodated man."
I have no time to analyze other symbolic themes that express these attributes of "structural invisibility," ambiguity and neutrality. I want n0w to draw attention to certain positive aspects of liminality. Already we have noted how certain liminal processes are regarded as analogous to those of gestation, parturition, and suckling. Undoing, dissolution, decomposition are accompanied by processes of growth, transformation, and the reformulation of old elements in new patterns. It is interesting to note how, by the principle of the economy (or parsimony) of symbolic reference, logically antithetical processes of death and growth may be represented by the same tokens, for example, by huts and tunnels that are at once tombs and wombs, by lunar symbolism (for the same moon waxes and wanes), by snake symbolism (for the snake appears to die, but only to shed its old skin and appear in a new one), by bear symbolism (for the bear "dies" in autumn and is "reborn" in spring), by nakedness (which is at once the mark of a newborn infant and a corpse prepared for burial), and by innumerable other symbolic formations and actions. This coincidence of opposite processes and notions in a single representation characterizes the peculiar unity of the liminal: that which is neither this nor that, and yet is both.
I have spoken of the interstructural character of the liminal. However, between neophytes and their instructors (where these exist), and in connecting neophytes with one another, there exists a set of relations that compose a "social structure" of highly specific type. It is a structure of a very simple kind: between instructors and neophytes there is often complete authority and complete submission; among neophytes there is often complete equality. Between incumbents of positions in secular politico-jural systems there exist intricate and situa-tionally shifting networks of rights and duties proportioned to their rank, status, and corporate affiliation. There are many different kinds of privileges and obligations, many degrees of superordination and subordination. In the liminal period such distinctions and gradations tend to be eliminated. Nevertheless, it must be understood that the authority of the elders over the neophytes is not based on legal sanctions; it is in a sense the personification of the self-evident authority of
ioo THE FOREST OF SYMBOLS
tradition. The authority of the elders is absolute, because it represents the absolute, the axiomatic values of society in which are expressed the "common good" and the common interest. The essence of the complete obedience of the neophytes is to submit to the elders but only in so far as they are in charge, so to speak, of the common good and represent in their persons the total community. That the authority in question is really quintessential tradition emerges clearly in societies where initiations are not collective but individual and where there are no instructors or gums. For example, Omaha boys, like other North American Indians, go alone into the wilderness to fast and pray (Hocart 1952, 160). This solitude is liminal between boyhood and manhood. If they dream that they receive a woman's burden-strap, they feel compelled to dress and live henceforth in every way as women. Such men are known as mixuga. The authority of such a dream in such a situation is absolute. Alice Cummingham Fletcher tells of one Omaha who had been forced in this way to live as a woman, but whose natural inclinations led him to rear a family and to go on the warpath. Here the mixuga was not an invert but a man bound by the authority of tribal beliefs and values. Among many Plains Indians, boys on their lonely Vision Quest inflicted ordeals and tests on themselves that amounted to tortures. These again were not basically self-tortures inflicted by a masochistic temperament but due to obedience to the authority of tradition in the liminal situation—a type of situation in which there is no room for secular compromise, evasion, manipulation, casuistry, and maneuver in the field of custom, rule, and norm. Here again a cultural explanation seems preferable to a psychological one. A normal man acts abnormally because he is obedient to tribal tradition, not out of disobedience to it. He does not evade but fulfills his duties as a citizen.
If complete obedience characterizes the relationship of neophyte to elder, complete equality usually characterizes the relationship of neophyte to neophyte, where the rites are collective. This comradeship must be distinguished from brotherhood or sibling relationship, since in the latter there is always the inequality of older and younger, which often achieves linguistic representation and may be maintained by legal sanctions. The liminal group is a community or comity of comrades and not a structure of hierarchically arrayed positions. This comradeship transcends distinctions of rank, age, kinship position, and, in some kinds of cultic group, even of sex. Much of the behavior recorded by ethnographers in seclusion situations falls under the
BETWIXT AND BETWEEN: LIMINAL PERIOD 101
principle: "Each for all, and all for each." Among the Ndembu of Zambia, for example, all food brought for novices in circumcision seclusion by their mothers is shared out equally among them. No special favors are bestowed on the sons of chiefs or headmen. Any food acquired by novices in the bush is taken by the elders and apportioned among the group. Deep friendships between novices are encouraged, and they sleep around lodge fires in clusters of four or five particular comrades. However, all are supposed to be linked by special ties which persist after the rites are over, even into old age. This friendship, known as wubwambu (from a term meaning "breast") or wulunda, enables a man to claim privileges of hospitality of a far-reaching kind. I have no need here to dwell on the lifelong ties that are held to bind in close friendship those initiated into the same age-set in East African Nilo-Hamitic and Bantu societies, into the same fraternity or sorority on an American campus, or into the same class in a Naval or Military Academy in Western Europe.
This comradeship, with its familiarity, ease and, I would add, mutual outspokenness, is once more the product of interstructural liminality, with its scarcity of jurally sanctioned relationships and its emphasis on axiomatic values expressive of the common weal. People can "be themselves," it is frequently said, when they are not acting institutionalized roles. Roles, too, carry responsibilities and in the liminal situation the main burden of responsibility is borne by the elders, leaving the neophytes free to develop interpersonal relationships as they will. They confront one anodier, as it were, integrally and not in compartmentalized fashion as actors of roles.
The passivity of neophytes to their instructors, their malleability, which is increased by submission to ordeal, their reduction to a uniform condition, are signs of the process whereby they are ground down to be fashioned anew and endowed with additional powers to cope with their new station in life. Dr. Richards, in her superb study of Bemba girls' puberty rites, Chisungu, has told us that Bemba speak of "growing a girl" when they mean initiating her (1956, 121). This term "to grow" well expresses how many peoples think of transition rites. We are inclined, as sociologists, to reify our abstractions (it is indeed a device which helps us to understand many kinds of social interconnection) and to talk about persons "moving through structural positions in a hierarchical frame" and the like. Not so the Bemba and the Shilluk of the Sudan who see the status or condition embodied or incarnate, if you like, in the person. To "grow" a girl into a
102
THE FOREST OF SYMBOLS woman is to effect an ontological transformation; it is not merely i convey an unchanging substance from one position to another by quasi-mechanical force. Howitt saw Kuringals in Australia and I ha\ seen Ndembu in Africa drive away grown-up men before a circumc sion ceremony because they had not been initiated. Among Ndembi men were also chased ofF because they had only been circumcised ; the Mission Hospital and had not undergone the full bush seclusion according to the orthodox Ndembu rite. These biologically mature men had not been "made men" by the proper ritual procedures. It is the ritual and the esoteric teaching which grows girls and makes men. It is the ritual, too, which among Shilluk makes a prince into a king, or, among Luvale, a cultivator into a hunter. The arcane knowledge or "gnosis" obtained in the liminal period is felt to change the inmost nature of the neophyte, impressing him, as a seal impresses wax, with the characteristics of his new state. It is not a mere acquisition of knowledge, but a change in being. His apparent passivity is revealed as an absorption of powers which will become active after his social status has been redefined in the aggregation rites.
The structural simplicity of the liminal situation in many initiations in offset by its cultural complexity. I can touch on only one aspect of this vast subject matter here and raise three problems in connection with it..This aspect is the vital one of the communication of the sacra, the heart of the liminal matter.
Jane Harrison has shown that in the Greek Eleusinian and Orphic mysteries this communication of the sacra has three main components (1903, 144-160). By and large, this threefold classification holds good for initiation rites all over the world. Sacra may be communicated as: (1) exhibitions, "what is shown"; (2) actions, "what is done"; and (3) instructions, "what is said."
"Exhibitions" would include evocatory instruments or sacred articles, such as relics of deities, heroes or ancestors, aboriginal chur-ingas, sacred drums or other musical instruments, the contents of Amerindian medicine bundles, and the fan, cist and tympanum of Greek and Near Eastern mystery cults. In the Lesser Eleusinian Mysteries of Athens, sacra consisted of a bone, top, ball, tambourine, apples, mirror, fan, and woolly fleece. Other sacra include masks, images, figurines, and effigies; the potter)' emblems (nihusd) of the Bemba would belong to this class. In some kinds of initiation, as for example the initiation into the shaman-diviner's profession among the Saora of Middle India, described by Verrier Elwyn (1955), pictures
BETWIXT AND BETWEEN: LIMINAL PERIOD 103
and icons representing the journeys of the dead or the adventures of supernatural beings may be shown to the initiands. A striking feature of such sacred articles is often their formal simplicity. It is their interpretation which is complex, not their outward form.
Among the "instructions" received by neophytes may be reckoned such matters as the revelation of the real, but secularly secret, names 0f the deities or spirits believed to preside over the rites—a very frequent procedure in African cultic or secret associations (Turner 1962a, 36). They are also taught the main outlines of the theogony, cosmogony, and mythical history of their societies or cults, usually with reference to the sacra exhibited. Great importance is attached to keeping secret the nature of the sacra, the formulas chanted and instructions given about them. These constitute the crux of liminality, for while instruction is also given in ethical and social obligations, in law and in kinship rules, and in technology to fit neophytes for the duties of future office, no interdiction is placed on knowledge thus imparted since it tends to be current among uninitiated persons also.
I want to take up three problems in considering the communication of sacra. The first concerns their frequent disproportion, the second their monstrousness, and the third their mystery.
When one examines the masks, costumes, figurines, and such displayed in initiation situations, one is often struck, as I have been when observing Ndembu masks in circumcision and funerary rites, by the way in which certain natural and cultural features are represented as disproportionately large or small. A head, nose, or phallus, a hoe, bow, or meal mortar are represented as huge or tiny by comparison with other features of their context which retain their normal size. (For a good example of this, see "The Man Without Arms" in Chisungu [Richards 1956, 211], a figurine of a lazy man with an enormous penis but no arms.) Sometimes things retain their customary shapes but are portrayed in unusual colors. What is the point of this exaggeration amounting sometimes to caricature? It seems to me that to enlarge or diminish or discolor in this way is a primordial mode of abstraction. The outstandingly exaggerated feature is made into an object of reflection. Usually it is not a univocal symbol that is thus represented but a multivocal one, a semantic molecule with many components. One example is the Bemba pottery emblem Coshi wa ng'oma, "The Nursing Mother," described by Audrey Richards in Chisungu. This is a clay figurine, nine inches high, of an exaggeratedly pregnant mother shown carrying four babies at the same time,
io4 THE FOREST OF SYMBOLS
one at her breast and three at her back. To this figurine is attached a riddling song:
My mother deceived me! Coshi wa ng'omai So you have deceived me; I have become pregnant again.
Bemba women interpreted this to Richards as follows:
Coshi wa ng'oma was a midwife of legendary fame and is merely addressed in this song. The girl complains because her mother told her to wean her first child too soon so that it died; or alternatively told her that she would take the first child if her daughter had a second one. But she was tricking her and now the girl has two babies to look after. The moral stressed is the duty of refusing intercourse with the husband before the baby is weaned, i.e., at the second or third year. This is a common Bemba practice (1956, 209-210).
In the figurine the exaggerated features are the number of children carried at once by the woman and her enormously distended belly. Coupled with the song, it encourages the novice to ponder upon two relationships vital to her, those with her mother and her husband. Unless the novice observes the Bemba weaning custom, her mothers desire for grandchildren to increase her matrilineage and her husband s desire for renewed sexual intercourse will between them actually destroy and not increase her offspring. Underlying this is the deeper moral that to abide by tribal custom and not to sin against it either by excess or defect is to live satisfactorily. Even to please those one loves may be to invite calamity, if such compliance defies the immemorial wisdom of the elders embodied in the mbusa. This wisdom is vouched for by the mythical and archetypal midwife Coshi wa ng'oma.
If the exaggeration of single features is not irrational but thought-provoking, the same may also be said about the representation of monsters. Earlier writers—such as J. A. McCulloch (1913) in his article on "Monsters" in Hastings Encyclopaedia of Religion and Ethics—are inclined to regard bizarre and monstrous masks and figures, such as frequently appear in the liminal period of initiations, as the product of "hallucinations, night-terrors and dreams." McCulloch goes on to argue that "as man drew little distinction (in primitive society) between himself and animals, as he thought that transformation from one to the other was possible, so he easily ran human and
BETWIXT AND BETWEEN: LIMINAL PERIOD ,05
animal together. This in part accounts for animal-headed gods or animal-gods with human heads." My own view is the opposite one: that monsters are manufactured precisely to teach neophytes to distinguish clearly between the different factors of reality, as it is conceived in their culture. Here, I think, William James's so-called "law of dissociation" may help us to clarify the problem of monsters. It may be stated as follows: when a and b occurred together as parts of the same total object, without being discriminated, the occurrence of one of these, a, in a new combination ax, favors the discrimination of a, h, and x from one another. As James himself put it, "What is associated now with one thing and now with another, tends to become dissociated from either, and to grow into an object of abstract contemplation by the mind. One might call this the law of dissociation by varying concomitants" (1918, 506).
From this standpoint, much of the grotesqueness and monstrosity of liminal sacra may be seen to be aimed not so much at terrorizing or bemusing neophytes into submission or out of their wits as at making them vividly and rapidly aware of what may be called the "factors" of their culture. I have myself seen Ndembu and Luvale masks that combine features of both sexes, have both animal and human attributes, and unite in a single representation human characteristics with those of the natural landscape. One ikishi mask is partly human and partly represents a grassy plain. Elements are withdrawn from their usual settings and combined with one another in a totally unique configuration, the monster or dragon. Monsters startle neophytes into thinking about objects, persons, relationships, and features of their environment they have hitherto taken for granted.
In discussing the structural aspect of liminality, I mentioned how neophytes are withdrawn from their structural positions and consequently from the values, norms, sentiments, and techniques associated with those positions. They are also divested of their previous habits of thought, feeling, and action. During the liminal period, neophytes are alternately forced and encouraged to think about their society, their cosmos, and the powers that generate and sustain them. Liminality may be partly described as a stage of reflection. In it those ideas, sentiments, and facts that had been hitherto for the neophytes bound up in configurations and accepted unthinkingly are, as it were, resolved into their constituents. These constituents are isolated and made into objects of reflection for the neophytes by such processes as componental exaggeration and dissociation by varying concomitants.
jo6 THE FOREST OF SYMBOLS
The communication of sacra and other forms of esoteric instruction really involves three processes, though these should not be regarded as in series but as in parallel. The first is the reduction of culture into recognized components or factors; the second is their recombination in fantastic or monstrous patterns and shapes; and the third is their recombination in ways that make sense with regard to the new state and status that the neophytes will enter.
The second process, monster- or fantasy-making, focuses attention on the components of the masks and effigies, which are so radically ill-assorted that they stand out and can be thought about. The monstrosity of the configuration throws its elements into relief. Put a man's head on a lion's body and you think about the human head in the abstract. Perhaps it becomes for you, as a member of a given culture and with the appropriate guidance, an emblem of chieftainship; or it may be explained as representing the soul as against the body; or intellect as contrasted with brute force, or innumerable other things. There could be less encouragement to reflect on heads and headship if that same head were firmly ensconced on its familiar, its all too familiar, human body. The man-lion monster also encourages the observer to think about lions, their habits, qualities, metaphorical properties, religious significance, and so on. More important than these, the relation between man and lion, empirical and metaphorical, may be speculated upon, and new ideas developed on this topic. Liminality here breaks, as it were, the cake of custom and enfranchises speculation. That is why I earlier mentioned Plato's self-confessed debt to the Greek mysteries. Liminality is the realm of primitive hypothesis, where there is a certain freedom to juggle with the factors of existence. As in the works of Rabelais, there is a promiscuous intermingling and juxtaposing of the categories of event, experience, and knowledge, with a pedagogic intention.
But this liberty has fairly narrow limits. The neophytes return to secular society with more alert faculties perhaps and enhanced knowledge of how things work, but they have to become once more subject to custom and law. Like the Bemba girl I mentioned earlier, they are shown that ways of acting and thinking alternative to those laid down by the deities or ancestors are ultimately unworkable and may have disastrous consequences.
Moreover, in initiation, there are usually held to be certain axiomatic principles of construction, and certain basic building blocks that make up the cosmos and into whose nature no neophyte may inquire.
107
BETWIXT AND BETWEEN: LIMINAL PERIOD
Certain sacra, usually exhibited in the most arcane episodes of the liirrinal period, represent or may be interpreted in terms of these axiomatic principles and primordial constituents. Perhaps we may call these sacerrima, "most sacred things." Sometimes they are interpreted fry a myth about the world-making activities of supernatural beings "at the beginning of things." Myths may be completely absent, however, as in the case of the Ndembu "mystery of the three rivers" (which I have described, pp. 61-65). mystery (mpangit) is
exhibited at circumcision and funerary cult association rites. Three trenches are dug in a consecrated site and filled respectively with white, red, and black water. These "rivers" are said to "flow from Nzambi," the High God. The instructors tell the neophytes, partly in riddling songs and partly in direct terms, what each river signifies. Each "river" is a multivocal symbol with a fan of referents ranging from life values, ethical ideas, and social norms, to grossly physiological processes and phenomena. They seem to be regarded as powers which, in varying combination, underlie or even constitute what Ndembu conceive to be reality. In no other context is the interpretation of whiteness, redness, and blackness so full; and nowhere else is such a close analogy drawn, even identity made, between these rivers and bodily fluids and emissions: whiteness — semen, milk; redness = menstrual blood, the blood of birth, blood shed by a weapon, etc.; blackness = feces, certain products of bodily decay, etc. This use of an aspect of human physiology as a model for social, cosmic, and religious ideas and processes is a variant of a widely distributed initiation theme: that the human body is a microcosm of the universe. The body may be pictured as androgynous, as male or female, or in terms of one or other of its developmental stages, as child, mature adult, and elder. On the other hand, as in the Ndembu case, certain of its properties may be abstracted. Whatever the mode of representation, the body is regarded as a sort of symbolic template for the communication of gnosis, mystical knowledge about the nature of things and how they came to be what they are. The cosmos may in some cases be regarded as a vast human body; in other belief systems, visible parts of the body may be taken to portray invisible faculties such as reason, passion, wisdom and so on; in others again, the different parts of the social order are arrayed in terms of a human anatomical paradigm.
Whatever the precise mode of explaining reality by the body's attributes, sacra which illustrates this are always regarded as absolutely sacrosanct, as ultimate mysteries. We are here in the realm of
io8 THE FOREST OF SYMBOLS
what Warner (1959, 3-4) would call "nonrational or nonlogical symbols" which
arise out of the basic individual and cultural assumptions, more often unconscious than not, from which most social action springs. They supply the solid core of mental and emotional life of each individual and group. This does not mean that they are irrational or maladaptive, or that man cannot often think in a reasonable way about.them, but rather that they do not have their source in his rational processes. When they come into play, such factors as data, evidence, proof, and the facts and procedures of rational thought in action are apt to be secondary or unimportant.
The central cluster of nonlogical sacra is then the symbolic template of the whole system of beliefs and values in a given culture, its archetypal paradigm and ultimate measure. Neophytes shown these are often told that they are in the presence of forms established from the beginning of things. (See Cicero's comment [De Leg. II. 14] on the Eleusinian Mysteries: "They are rightly called initiations [beginnings] because we have thus learned the first principles of life.") I have used the metaphor of a seal or stamp in connection with the ontological character ascribed in many initiations to arcane knowledge. The term "archetype" denotes in Greek a master stamp or impress, and these sacra, presented with a numinous simplicity, stamp into the neophytes the basic assumptions of their culture. The neophytes are told also that they are being filled with mystical power by what they see and what they are told about it. According to the purpose of the initiation, this power confers on them capacities to undertake successfully the tasks of their new office, in this world or the next.
Thus, the communication of sacra both teaches the neophytes how to think with some degree of abstraction about their cultural milieu and gives them ultimate standards of reference. At the same time, it is believed to change their nature, transform them from one kind of human being into another. It intimately unites man and office. But for a variable while, there was an uncommitted man, an individual rather than a social -persona, in a sacred community of individuals.
It is not only in the liminal period of initiations that the nakedness and vulnerability of the ritual subject receive symbolic stress. Let me quote from Hilda Kuper's description of the seclusion of the Swazi chief during the great Incwala ceremony (1961, 197-225). The Incwala is a national First-Fruits ritual, performed in the height of summer when the early crops ripen. The regiments of the Swazi
B£TWIXT AND BETWEEN: LIMINAL PERIOD I09
nation assemble at the capital to celebrate its rites, "whereby the nation receives strength for the new year." The Incwala is at the same time "a play of kingship." The king's well-being is identified with that of the nation. Both require periodic ritual strengthening. Lunar symbolism is prominent in the rites, as we shall see, and the king, personifying the nation, during his seclusion represents the moon in transition between phases, neither waning nor waxing. Dr. Kuper, professor Gluckman (1954), and Professor Wilson (1961) have discussed the structural aspects of the Incwala which are clearly present in its rites of separation and aggregation. What we are about to examine are the interstructural aspects.
During his night and day of seclusion, the king, painted black, remains, says Dr. Kuper, "painted in blackness" and "in darkness"; he is unapproachable, dangerous to himself and others. He must cohabit that night with his first ritual wife (in a kind of "mystical marriage"— this ritual wife is, as it were, consecrated for such liminal situations).
The entire population is also temporarily in a state of taboo and seclusion. Ordinary activities and behavior are suspended; sexual intercourse is prohibited, no one may sleep late the following morning, and when they get up they are not allowed to touch each other, to wash the body, to sit on mats, to poke anything into the ground, or even to scratch their hair. The children are scolded if they play and make merry. The sound of songs that has stirred the capital for nearly a month is abruptly stilled; it is the day of hacisa (cause to hide*). The king remains secluded; ... all day he sits naked on a lion skin in the ritual hut of the harem or in the sacred enclosure in the royal cattle byre. Men of his inner circle see that he breaks none of the taboos ... on this day the identification of the people with the king is very marked. The spies (who see to it that the people respect the taboos) do not say, "You are sleeping late" ot "You are scratching," but "You cause the king to sleep," "You scratch him (the king)"; etc. (Kuper 1947, 219-220).
Other symbolic acts are performed which exemplify the "darkness" and "waxing and waning moon" themes, for example, the slaughtering of a black ox, the painting of the queen mother with a black mixture—she is compared again to a half-moon, while the king is a full moon, and both are in eclipse until the paint is washed off finally with doctored water, and the ritual subject "comes once again into lightness and normality."
In this short passage we have an embarrassment of symbolic riches. I will mention only a few themes that bear on the argument of this
no THE FOREST OF SYMBOLS
paper. Let us look at the king's position first. He is symbolically invisible, "black," a moon between phases. He is also under obedience to traditional rules, and "men of his inner circle" see that he keeps them. He is also "naked," divested of the trappings of his office. He remains apart from the scenes of his political action in a sanctuary or ritual hut. He is also, it would seem, identified with the earth which the people are forbidden to stab, lest the king be affected. He is "hidden." The king, in short, has been divested of all the outward attributes, the "accidents," of his kingship and is reduced to its substance, the "earth" and "darkness" from which the normal, structured order of the Swazi kingdom will be regenerated "in lightness."
In this betwixt-and-between period, in this fruitful darkness, king and people are closely identified. There is a mystical solidarity between them, which contrasts sharply with the hierarchical rank-dominated structure of ordinary Swazi life. It is only in darkness, silence, celibacy, in the absence of merriment and movement that the king and people can thus be one. For every normal action is involved in the rights and obligations of a structure that defines status and establishes social distance between men. Only in their Trappist sabbath of transition may the Swazi regenerate the social tissues torn by conflicts arising from distinctions of status and discrepant structural norms.
I end this study with an invitation to investigators of ritual to focus their attention on the phenomena and processes of mid-transition. It is these, I hold, that paradoxically expose the basic building blocks of culture just when we pass out of and before we re-enter the structural realm. In sacerrima and their interpretations we have categories of data that may usefully be handled by the new sophisticated techniques of cross-cultural comparison.
BIBLIOGRAPHY
Bettelheim, B. 1954. Symbolic Wounds. Glencoe: Free Press. Cicero, M. Tullius. 1959. De Legibus. Ed. by de Plinval. Paris: Les Belles Lettres.
Douglas, Mary. 1966. Purity and Danger. London: Routledge & Kegan Paul.
Elwin, Verrier. 1955. The Religion of an Indian Tribe. London: Geoffrey Cumberlege.
Gennep, A. van. i960. The Rites of Passage. London: Routledge & Kegan Paul.
BETWIXT AND BETWEEN: LIMINAL PERIOD MI
Gluckman, Max. 1954. Rituals of Rebellion in South-East Africa. Manchester University Press.
Harrison, Jane E. 1903. Prolegomena to the Study of Greek Religion. London: Cambridge University Press.
fjocart, A. M. 1952. The Life-Giving Myth. London: Methuen.
James, E. O. 1961. Comparative Religion. London: Methuen.
James, William. 1918. Principles of Psychology. Vol. 1. New York: H. Holt.
Kuper, Hilda. 1947. An African Aristocracy. London: Oxford University Press, for International African Institute.
McCulloch, J. A. 1913. "Monsters," in Hastings Encyclopaedia of Religion and Ethics. Edinburgh: T. & T. Clark.
Richards, A. I. 1956. Ghisungu. London: Faber & Faber.
Turner, V. W. 1962. Chihamha, the White Spirit (Rhodes-Livingstone Paper 33). Manchester University Press.
Warner, W. L. 1959. The Living and the Dead. New Haven: Yale University Press.
Wilson, Monica. 1959. Divine Kings and the Breath of Men. London: Cambridge University Press.