A Semiotic Approach to Religion

Boris Ogibenin

Much has been said about religion, and this may be attributed to the fact that religion offers an explanation for the mental life of man. Indeed, quite frequently, such an explanation takes on as much importance as existence itself. For if we only had accurate and full descriptions of the religious systems of the many societies, ancient and modern, we might be able to trace both their cultural and their extra-cultural histories. It is taken for granted here that: (1) Religion is an all-encompassing system through which the manifold semantics of human culture and man's experiences can be comprehended. Hence, any religious system tends to be fully comprehending and its categories are those of human experience. (2) Religion is a modeling system; thus, there is no direct correspondence between it and categories of human experience since modeling does not imply mere reflection.

Melford E. Spiro has defined religion in such a way that leaves many questions open for the semiotician: "I shall define 'religion' as an 'institution consisting of culturally patterned interaction with culturally postulated superhuman beings'" (1969:96). It should be noted, however, that although Spiro proceeds to briefly explain what is meant by interaction, institution, and superhuman beings on a universal level, he presents them in a manner that is predominantly Christian. Unfortunately, it seems to be a general rule that anything that we try to explain is assigned to the concepts that we already have. Thus, unavoidably, our explanations appear to be the intersections of our conceptualizations with those of another.

The first draft of this article was delivered as a lecture at the Research Center of King's College, University of Cambridge; the final version was completed at Indiana University, Bloomington, for the Pilot Program in Semiotics in March 1976. I would like to thank Professor Edmund R. Leach and Thomas A. Sebeokfor the opportunity of staying at these two universities.

Piatigorsky has written a remarkable passage on this problem, in trying to explain what a strange world Buddhism is to one not acquainted with it.

The approach explicated by Piatigorsky seems to be the only realistic one. Sociologists and, in particular, those studying religion rightly complain that the interaction of both observer and observed—in this case, society or religious categories—presents an uneasy situation (Goody, 1961:156-57; Goody even reports P. Winch's opinion that "all sociology is impossible since the observer can never get outside the conceptual apparatus of his own society nor, conversely, inside that of any other"). It follows, then, that if we succeed in creating a metatheory of religious thinking and put as a first step the questions of a particular religion, such an approach may help in developing a further synthesis. Creating a metatheory within semiotic vistas accentuates not so much what we should understand in a religion (Spiro, 1969:92-94, may be right when he insists that superhuman beings are accounted for even in Theravada Buddhism, where, unlike ordinary human beings, Buddha acquires the power of attaining Enlightenment) but what the particular religion we are studying communicates either to its practitioners or to us. A semiotic metatheory does not include discussions of whether—to take a most controversial issue—the sacred or the profane is present and, if so, in what forms; rather, it examines which features and which categories are structurally relevant within a religion, which of them signify and what they signify within that religion. Religion is treated, then, as a self-sufficient system for interpreting man and his world within a socio-communicative matrix. Adding the dimension of communication does not change our understanding of religious phenomena. The theoretical framework proposed by Clifford Geertz is, after all, very close to the semiotic conception of religion and culture—so much so that he defines culture as "a system of inherited conceptions by means of which men communicate, perpetuate, and develop their knowledge about and attitudes toward life" (Geertz, 1969:3). Geertz, then, has cast his definition of religion in accordance with this conception of culture:

I will comment, below, on this conception, which is a succinct formulation of some fundamental principles—although I do not find all the implications that Geertz discusses acceptable.

First, let us look at the concept of motivation. Motivations for Geertz are of two kinds: motives and moods. The formal difference between them is that moods (for example, a reverential', 'solemn, or worshipful' mood) occur under particular circumstances; they are made meaningful' with reference to the conditions from which they are thought to spring. Motives are vectorial qualities described as liabilities that perform particular classes of act or have particular classes of feeling; they are 'made meaningful' with regard to the ends toward which they are thought to conduce (Geertz, 1969:11-12). Motivations are to my mind the essential operations that man performs across the continuum of meaning—i.e., the perceptible universe on which he tries to apply the matrix of order. That religious problems are always problems of meaning, as postulated by Weber, is quite true, but semioticians approach meaning in a more technical sense by looking for meaningful operations over a semanticized universe. The universe itself is, as it were, felt and perceived as coextensive and commensurable to the human discourse on it. Universe is discourse. Again, it may be said that this discourse does not need to be specifically religious since there are many varieties of discourse on mans universe; as Geertz points out, scientific, esthetic, or just commonsensical discourse challenges religious discourse. Rather than try to define what distinguishes religious discourse from other types of discourse, I will analyze one aspect of religious discourse —that of motivational patterns as complexes of semantic rules and operations over a meaningful whole.

One may wonder why this particular aspect of religious discourse is singled out for analysis. It is an area proper for semiotic analysis since such an analysis would insist on the manipulation of meanings by a religious subject, on the techniques of performing operations, and on the types of operations fulfilled. It seems to me that by emphasizing this point, one approaches more the essence of individual involvement in the religious understanding of the universe, since meaningful operations not only have their own meaning but also provide us with a knowledge of what a person wants when he or she faces the choice of being religious. What I have in mind is that accounting for religious experiences implies more of subjective psychology than of dealing with cultural patterns that are necessarily highly socialized and, as the semiotic approach deems it, quite mandatory. This generalization, of course, may appear questionable; its immediate justification is that I will try to delineate the most general features common to religions in societies where all of its members are involved in religious life and in those in which one can choose whether to become involved in religion or not.

One important feature of religion is that it not operate directly with worldly matters—although it was noted above that the entire universe is permeated with religious discourse. Another proper feature of religious discourse is that it coexist with mythological discourse; in fact, it may be difficult to separate the two. It is therefore advanced here that religious discourse contains a set of semantic rules and operations to be performed on mythological discourse. Looking at the relationship of these two types of discourse, one can find characteristics of both the rules and their operations. Mythological discourse intercedes between worldly discourse—that is, if it can be assumed that the world communicates with us—and religious discourse.

If one compares the meaning of operations produced within a religion and within a mythology, these two complexes of organized, patterned meaning would be defined as having somewhat different relations with each other than what Geertz, for one, postulates for religion and ritual performance. Ritual performance often tends to be considered as an illustration of religious principles, belonging more to the domain of mythology than of religion; this is especially so since religion appears to be more abstract than either mythology or ritual. It is obvious that the three are somehow related, but can we say, with Geertz, that "by inducing a set of moods and motivations—an ethos—and defining an image of cosmic order—a world-view—by means of a single set of symbols, the performance makes the model for and model of aspects of religious belief mere transpositions of one another" (Geertz, 1969:34)? Here I maintain that mere transposition or, in other terms, transcoding, is not the case universally and cannot be viewed as a permanent, inherent feature of the interrelation between religion and the two domains.

Thus coming back to motivational patterns and to the question of their working within both mythology and religion, we will now give some examples from which we will later draw conclusions.

As far as scholars have been able to reconstruct, Vedic religion is not identical with Vedic mythology; however, it can be stated that the motivational patterns within both are similar. This means that what is induced, advised, or recommended by the religious ideology as taught by Vedic seers is what mythology—which in this particular case is shaped by religion—enhances and intensifies. If Vedic religion is predominantly a cosmogony (Kuiper, 1975:107; Ogibenin, 1973) and a cosmogonical conception of all-pervading character, it may be said that the operations supposedly carried within such a conceptual framework have been the very ones that are fulfilled the moment that religious understanding of phenomena is abandoned and their mythological interpretation is taken up. Motivation in Vedic religion, if concisely formulated for the present purposes, consists of stressing the overall importance of two essential and opposite poles, which have been expressed in a multitude of ways—now as "the being, that which is" (sat) vs. "the non-being" (asat), "gods" vs. "demons" (asuras), Order (Rta) vs."chaos" (anrta), and so on. The whole series of bipolar concepts is valued according to the elementary binary design in which a positive term is opposed to its corresponding negative term. This patterning is, however, intermediary between the concept that the universe was first undifferentiated and the notion of Rta, the "Order" and "Truth" on which the welfare of the world depends. The basic philosophy reflected in these concepts is followed by the mythology of the Vedic texts, which, in fact, is ruled by them. It may appear quite trivial that a mythology that operates with less abstract material is modeled on the premises of a set of more abstract ideas. It is, however, valuable to take advantage of this in order to see the basic congruence between religious presuppositions and assumptions and the body of mythological signs.

Vedic mythology makes extensive use of the same patterning of motivations that is most probably clothed in a series of binary oppositions that are structurally similar to those that have been mentioned above. A created universe is thus valued positively as against an uncreated one, i.e., the realm of chaos. The Vedic universe is in fact considered to be the universe in which normal human existence can be carried on, while the non-Vedic universe—the realm of the foes of the Vedic man—is doomed. Vedic man's microcosm is opposed to non-Vedic man's microcosm. As in religious thinking, mythological patterning includes the concept of an all-encompassing totality: in fact, the god Visnu appears throughout the Vedic texts as one who transcends the cosmic antithesis in which sat and asat have been reconciled in the synthesis of an all-embracing entity (Kuiper, 1975:118). In some respects, if viewed in their roles as cosmogonie agents, all the other Vedic deities only reiterate the deeds that are supposed to lead to the establishment of a livable world—although these deeds are presented in various forms corresponding to aspects that are necessary for producing such a world. The main hypothesis that is being formulated here is that Vedic poets, authors of the mythological hymns, conspicuously attack the problems of environment and its conceptualization in a one-sided manner: they elaborately treat the world which they desire while the one which they reject and which would be that of their god's enemies is either given little note or not described at all. All we can know of this adversary world is that it is the inverse of, and opposed to, the Vedic world. It could be, as U. Masing (1973) suggested, that such dichotomies were only the mode of description used by the Vedic authors and that their adepts were aware of such a model and deliberately arranged their ideas on those levels. Masing (1973:18) admits though that even if "they could express their thoughts by means of some other sublanguage also, the structure of their system would . . . remain the same." The principle of bipolarity is pervasive, leading one to infer that here the laws of Vedic religious thinking were also determinant. As in the case of reiterating the basic structural oppositions that have been instrumental for the conception of the world, the principle has been used anew for discriminating the actions of the gods within the descriptions of Vedic mans existence; here, only one pole has been evaluated while the value of the other pole had to be automatically inferred.

The mythology of evil in the Veda is not as abundantly elaborated upon as in Tantric mythology or in Dante's Inferno. Again, all that we are allowed to know about evil is that it had to be defeated by Vedic deities so that the desired goods could be obtained within the Vedic universe. If, then, one admits that this one-sided extolling of the Vedic universe may have originated at the very outset of Vedic religious thinking and mythological speculation (since the initial dichotomizing seems to be a product of the imposition of a matrix of ordering in contradistinction to that of disordering), it may help explain why Visnu, for one, has been chosen to impersonate the concept of totality. This totality, referred to also as a "third heaven" or the "highest heaven" in the Rgveda, encompasses not only chaos but also ordered cosmos; not only gods as positive agents but also, at the same time, the potential gods— the Asuras—who have been discarded to the netherworld (Kuiper, 1975:114-18). If Kuiper's explanation of the basic concept of Vedic religion is accepted, it is worthwhile to ask why it is not the god Varuna, the former Asura, who still preserves his connections with the asuras since he is their jailer, as Mahãbhãrata narrates it, and who accordingly is an ambiguous figure. One may think that, in accordance with the dominant semantic operation, Vedic religion prescribes the promotion of only a positively evaluated personage of the mythology to the rank of a symbol of totality, even though an ambiguous personage possesses all the requisite qualities when viewed from the standpoint of their significant values within the structure of the mythology.

This semantic operation may be considered dominant, thereby commanding other structurally relevant operations within the mythology. Dichotomizing and positively valuing one element of the dichotomy are most likely to be connected with the fact that whatever was subjected to this operation was designed to saturate the meaningful field of mythology. In other words, once positive meanings were proclaimed as valued, the sets of meaningful elements in general increased. Nonpositive meanings were not described at all; rather, as mentioned above, they have been designated as functions of positive meanings, i.e., brought within the signified and sensible field through the operation of relating them to the set of positive meanings.

Mythology is always concrete. If the religious system is really one that provides rules for handling the meanings and values within a mythology, this should hold even in the case of a more complex mythology. It is not implied here that Vedic mythology is a simple system; we only mean that the rules and the structures on which these rules operate are more conspicuous than in, say, Purãnic mythology. It is still possible, however, to elicit some overtly formulated rules and operations by comparing the doctrinal teachings of the Purãnas with their mythological presentation.

It is known that in such religious works as the Bhãgavata Purãna the most important tenet that is taught and preached is bhakti, the devotion to the Lord Krsna or Visnu. The purpose of bhakti is to destroy man's attachment to worldly values; bhakti alone is likely to bring salvation to an individual, but such salvation is mainly understood to be total concentration on the person of Bhagavãn (The Blessed Lord). As the texts assembled by Thomas J. Hopkins (1971; Hopkins's article also interprets other aspects of the same teachings) show, the doctrine makes constant use of two levels and two sets of meaningful elements. The two sets obviously correspond to the levels in such a way that the level of selfless devotion to the Lord reworks the elements of secular life and transforms them into the elements of devotional and religious life. It is explicitly demonstrated that attachment may be of two kinds : attachment to the world of senses, and attachment and devotion to Bhagavãn. Uncontrolled thinking is conducive to increasing one's attachment to the world, whereas thinking under the control of bhakthyoga helps one to concentrate on Bhagavãn. Moreover, almost any human act that is felt to be alien to the ritualism of the true devotee of Krsna—even those that are accomplished as religious acts within Vedic ritualism—can be classified in view of its value toward increasing devotion to Bhagavãn. As written in the Bhãgavata Purãna:

In a more sophisticated manner, the same Purãna rejects acts that are not aimed at Bhagavân's devotion even though Bhagavãn himself is embodied in everything, including human acts:

It appears that there are unmarked acts vs. marked ones. The former belong to the level of worldly life, where devotion to Bhagavãn is not practiced, while the latter are the same elements of worldly life (which include rituals of Vedic origin) that are introduced within the realm of devotional life. The category of the Lord Bhagavãn, if seen as a unit of religious structure, appears to be a cover category in which both marked and unmarked acts and elements merge. Mythology may well distinguish between both: bhakti, treated as an imperative goal within such a system of values and goals, may thus eliminate minute distinctions.

Religious doctrine may also formulate principles that can be illustratively revealed by specially arranged texts that portray these abstract principles. These texts may, however, contradict or use a given technique that responds to the doctrine. This is perhaps where the elegant ingenuity of Buddhist parables lies. Let us look now at two such parables.

When King Milinda asked the Reverend Nãgasena to explain one of the subtle problems of Buddhist doctrine—that of the discontinuity of personal identity, which is based on the negation of the self—the monk answered by telling two famous parables, "Embryo and child" and "Lamp and flame." The former illustrates the absence of continuity in the development of a human being, which takes place regardless of prevailing conditions; the latter describes the continually changing nature of a flame. The account that is offered is as follows.

Seen as a short narrative, the first parable relates that (I) the king's development from a being "once young, tender, weak, lying on .. . back" to an adult proves that the king could not always have been the same person. This is further confirmed by the king himself: "He that was young, tender, weak, lying on his back, was one person; I, big as I am now, am a different person" (quoted from Burlingame, 1922:204-205); (II) as Nãgasena suggests, the corollary is that the king never had a mother, a father, or a teacher, that he never acquired knowledge of the arts and crafts, and so forth. This last point is illustrated by another parable inserted within the first; this second parable relates that (a) it is possibly true that (i) the mother of a fetus in the first stage of development is one person; that (ii) the mother of the fetus in the second stage is another; and so forth. (b) It ends by asking whether the mother of the little child is one person, and the mother of a grown man is another person. The next point is (III) of the first parable; here analogous questions are asked—whether the person who is learning the arts and crafts is different from a person who has already acquired them, and whether the person who performs evil deeds is different from a person whose hands and feet are cut off. (IV) The king reverses the question; Nãgasena answers that he himself is obviously not the same person now that he is an adult, as compared to when he "was once young, tender, weak, lying . . . back," since only the dependence on his body would explain that all these states are embraced in one. Nãgasena thus reiterates point (I) of the story.

In the second story, questions are again asked in an analogous manner: (I) Is the flame that (i) burns in the first watch the same as (ii) the flame that burns in the middle watch; and (iii) is the flame that burns in the middle watch the same as (iv) the flame that burns in the last watch? (II) Since the lamp remains the same during all three watches, it is (III) stated that the lamp was only the cause of the flame that burned all night.

It should be noted that both parables are coupled in the collection "Milindapanha" because of their overt structural similarity. Moreover, the common summary given to these two parables by Nãgasena emphasizes the doctrinal point that they illustrate: "there is an uninterrupted succession of mental and physical states. One state ceases to exist and another comes to exist. The succession is such that there is, as it were, none that precedes, none that follows. Thus it is neither that same person nor yet a different person which goes to the final summation of consciousness" (Burlingame, 1922:205).

The structuring of both parables is purposely analytic; this is shown by the segmentation indicated by the Roman numerals and the small Latin letters. The analytic nature of the exposition is shaped after the model of discontinuous reality in which everything is subjected to momentary changes. Remarkably, the parable "Embryo and child"—after stating that the king never had a mother, a father, or a teacher; that he never acquired the arts and crafts; and so forth—draws explicitly only on the example of the mother denied. Other items (father, teacher, acquisition of arts and crafts, etc.) are left for the attentive listener to analyze, and this is encouraged by both the topic under discussion and by the structuring of the parable.

It can be said that the narrative code of the parables corresponds to the dominant philosophic categories of Buddhist doctrine and, at the same time, disavows it. Its correspondence—which is in fact the correspondence of the narrative metalanguage to the semantic system of Buddhist religion—is shown by the emphasis of the iconic, which dissects the items that Nãgasena calls forth; its contradiction seems to follow from the special attitude of Buddhism toward word and sign. The discussion of this problem by Piatigorsky proves quite clearly that in Buddhism words like I, you, person may have been used only very carefully, as purely objectified signs (or in linguistic terminology, as "markers") of the precarious and ever changing because of the nature of the states of the speakers (Piatigorsky, 1975:9-13ff.). Nãgasena could only be too well aware of the basically approximate character of his demonstration.

We have shown here that fundamental semantic properties and characteristics that can be called the "semantic dominants" of religious thinking determine the operations that are produced on the level of mythological structure. In fact, motivations—the motives and moods of Geertz—may be either identical or dissimilar in both frameworks. It is, however, very likely that the system of religious thought precedes that of mythology with regard to the hierarchy of operations fulfilled.

The other implications that Geertz raises are beyond the scope of this paper. I was particularly interested by his "powerful, pervasive, and long-lasting moods and motivations in men," the properties of which appear to fluctuate the most.