Divination is perhaps one of the most intriguing forms of human knowledge practice and has often turned out to be a touchstone for anthropological theory. Given the complex and seemingly enigmatic character of divination practices in Africa and elsewhere, one might expect the nature of divinatory knowledge to be a pivotal theme in studies on the subject. Yet apart from Evans-Pritchard (1976 [1937]), who emphasized the internal rationality of divination as a means of decision finding, questions concerning the underlying epistemology of divinatory practice have only recently attracted serious anthropological attention. Until the early 1990s, most studies of divination in Africa and elsewhere focused either on the cultural-historical or the sociological properties of divinatory praxis. Studies of the first type mainly concentrated on the ethnographic and historical description of the textual (mythology associated with the origin of divination, sacred texts, divinatory manuals) and material (figurines, divination boards, astrological instruments, calculation tables) bases of the various forms of divination in historical and contemporary African societies (see, for example, the pioneering works on Ifá and related forms of divination by Trautmann Reference Trautmann1939; Maupoil 1988 [1943]; and Bascom 1969, 1980). While representing important contributions to our understanding of how divination manifests itself textually and/or materially in different cultural contexts, most works of this type offered little analysis of divinatory praxis itself, either in the form of case studies or in terms of symbolic or cultural analysis. Studies of the second, (structural-) functionalist type focused primarily on divination as a social practice, highlighting divination's significance as a central and often decisive instrument in the directing of (micro-)political processes such as the formation, maintenance and transformation of economic, political and parental power relations at local levels, and often included detailed case studies (Turner Reference Turner and Turner1975 [1961]; Werbner Reference Werbner1973; Mendonsa Reference Mendonsa1982; and, critically, de Boeck 1991).
From the 1990s onwards, however, as advocated in an earlier, seminal article by Devisch (1985), the literature on divination in sub-Saharan Africa has experienced an important shift in perspective and theoretical orientation away from ethnographic cataloguing and (external) social analysis, towards approaches exploring the internal, semiotic, semantic and praxeological dimensions of the divinatory process. Rather than denying the value of (structural-) functional approaches to divination for the analysis of social formation, transformation and micro-politics, these more internal approaches aim at understanding and defining the phenomenological and cultural properties that are specific to divination and that distinguish it from other, non-divinatory cultural practices of investigation and decision finding.
Following the programme outlined by Devisch, these recent investigations into the art of divination are less concerned with cultural origins, technical procedures, and questions of social functioning than with divination ‘as a system of knowledge in action’ (Peek Reference Peek1991), embodiment and world making (de Boeck and Devisch 1994), the performative qualities inherent in divinatory ritual and its objects (Pemberton Reference Pemberton III2000), and the relation between divination and healing (Winkelman and Peek 2004). Pursuing further these investigations, and especially their interest in the epistemological and performative dimensions of divination, in this article I try to understand the nature of the interpretative operations and kind of knowledge involved in Senegambian divinatory praxis. How does the diviner arrive at his divinatory pronouncements? What kind of technical and interpretative principles are involved in divinatory praxis, and what are the modes of knowledge the diviner draws upon and is able to activate in the course of a successful séance?
After the description and tentative analysis of the modes of knowledge underlying cowrie divination performances in Senegal and the Gambia, the question will be asked whether this focus on the thought processes involved in divination, and on the diviner as the bearer of specialist knowledge, is not also problematic. Does it not overlook what may be the primary reason for the undiminished significance of divination in a multitude of cultural contexts today? Is this not its consultational rather than its cognitive quality – its capacity as a cultural institution to address, deal with, and counter existential personal situations, ranging from illness and marital conflicts to unemployment and migration, in a perhaps unique way? In a further analytical move, the question of the nature of divinatory knowledge will then be reconsidered in relation to its own epistemology, according to which the source of the divinatory enunciation is not located in the person of the diviner but in the divinatory apparatus itself.
INTRODUCING SENEGAMBIAN COWRIE DIVINATION
In the Senegambian context, divination is practised in many different forms ranging from cowrie divination and Islamic geomancy to divination by the use of prayer beads, the casting of roots and dream divination. When I first witnessed a session of cowrie shell divination, I was impressed by the atmosphere of the event, its sincerity, rhythm, and steady speed. The white shells fall on the hard surface of a woven plastic mat with a light and characteristic rustle, every new cast throwing a new kaleidoscopic image before the eyes. A second, parallel rhythm is set by the diviner's voice, counting some of the shells in Wolof according to principles unknown to the uninstructed observer. Each count is followed by interpretations, instructions, sometimes questions, soliciting the client's response to the messages of the cast shells. The diviner pronounces what he sees, moves on in his interpretation, elaborates upon specific points, and responds to the reactions of his or her client by recasting the cowries for further detailing. Every gesture, every word forms yet another element in the diviner's search for the nature and development of the client's predicament. Every interpretation further adds to the diviner's analysis of his client's situation, indicating the crucial aspects of the issues at stake, and eventually leading to the formulation of the necessary ritual prescriptions.
For the outside observer several questions arise from such a first witnessing of a cowrie divination session. Considering the fact that, in Mandinka and Wolof, divination is literally referred to as an act of ‘viewing’, ‘looking at’ or ‘looking for’ (jubeeroo in Mandinka, seet in Wolof), the first question is: what is there to be seen in the seeming disorder of the cast shells? What kind of signs or messages appear in the constellation of the divinatory paraphernalia that allow the diviner to discern the concerns of his client? Where does the diviner's gaze first take hold? Is there a method or technical basis to cowrie (and other forms of) divination that can be described, studied, and maybe even learned? Or does the diviner's clairvoyant capacity elude an outward description from the very beginning? Are there actually things to be seen, or can the signs and messages of the cast shells only be perceived in an altered state of mind?
Despite frequent references to spirit entities (Mandinka jinnool; singular, jinno) and esoteric initiation procedures, cowrie divination, like Islamic geomancy, rests upon a technical basis that can be learned and acquired. This was the emphatic view of Samba Nguer, a Wolof-speaking diviner from inland Gambia, now based in Serekunda.Footnote 1 At the beginning of our cooperation he insisted that his instructions would enable me to understand most of what the cowries would tell me, without going through any kind of initiation, trance experience, or other ritual measure aiming at establishing a relation with the spirit realm, apart from the initial ritual preparation of the set of shells he was going to give me at the end of my apprenticeship.Footnote 2 And even where diviners do refer to contacts with jinn or other non-human spirit entities as mediators, to extrasensory capacities, or to a highly specialized knowledge (Mandinka londoo) of (ritual) secrets (kungloo) revealed through dreams or acquired from other specialists at high cost and/or through long periods of apprenticeship, the technical basis outlined by Samba Nguer is accepted by them almost without exception. Specific positions or configurations of shells within a particular layout are perceived as signs (Mandinka tamansee) indicating the course of development of specific aspects of the subject's immediate social and economic environment, and as a basis for the understanding of the inquiring subject's personal state of mind and emotional condition.
One of the most basic distinctions necessary for the reading of the different positions in cowrie divination is that between shells falling with their concave, smooth and often glossy side upwards, displaying the indented slit-opening typical of all molluscs of the Cypraeidae family, and those that fall with their other side uppermost – the side that was originally a smooth hump, but which has been broken open, leaving only an outer-ring.Footnote 3 The former were referred to by Samba Nguer either as fermee (closed, from the French fermé) or as dafa dep (lying face down), while the latter were referred to as ubeku (open). In many layouts, the closed shells were conceived of as symbolizing male persons while open ones were seen as relating to women. Samba Nguer applied this distinction at the beginning of the session when identifying, through several preliminary casts, the shells upon which he would then ask his client to pronounce his intention or concern. In the case of male clients, he would select four closed shells pointing towards the client, while with female clients he would choose the same number of open shells pointing in the client's direction.Footnote 4 In other situations, however, open and closed shells would be distinguished without being interpreted according to their male and female connotations. Thus Samba Nguer might count the closed shells in certain layouts in order to identify the number of objects that should be distributed as sarax (or sadaa in Mandinka), one of the principal ritual remedies recommended by Senegambian diviners: it consists in the voluntary donation or offering of objects ranging from sugar cubes and candles, to cloth or food. In these instances, the indication of sarax would be the only interpretative value of the closed shells and their otherwise assumed male connotation would not come into play.
Many of the significant positions or signs are formed by not more than two shells. This means not only that great attention has to be paid to the exact position of the various shells within a layout, but also that a single layout can easily contain two or three signs that can be, will be and often must be read in relation to each other. An important example for a position containing only two single shells consists in two either open (female) or closed (male) cowries, lying side by side but pointing in opposite directions. This position is interpreted as a sign of long life (gudd fan) and good health (wër). If one of the shells lies with its open side upwards, the same position signifies a state of hesitation, uncertainty, ambivalence, and indecision called xel ñaar ([being of] two minds), xel bu werente (an arguing mind), nakhar (something disliked or unpleasant), or sikisaka (zigzag). If in the same position both cowries point in the same direction, this indicates a calm and untroubled mind (xel mu dal). Two open shells pointing towards each other with their ‘back’ (taat, literally the bottom) indicate a good marriage (sey bu nex). If, in the same position, the two cowries seem to move away from each other, the position is called sey butas, ‘a dispersed marriage’, indicating divorce. Both positions are related to the position of ngoro, ‘engagement’, formed by an open and a closed shell pointing at each other with their back sides in a straight line: this indicates concern for a person of the other sex in the form of a love affair or an existing intention to make the first arrangements towards marriage. Metonymically, the same position also often indicates a conflictual relationship with one's in-laws as those towards which one acquires obligations through marriage. Another important position is that of a closed shell partly covering another, open shell: this indicates the state of being ill (feybar).5
All the above-mentioned positions of cowrie shells occur regularly during divination sessions and can be interpreted by the diviner. The direction and spatial position of single shells or specific clusters of shells indicate how far and in what way different positions refer to the consulting subject himself or may be related to other persons within the subject's environment. Additional information can be derived from counting the number of open and closed shells in a particular layout. Distinguishing again between the female and male connotations of open and closed shells, with female clients, Samba Nguer counted the shells that fall with their perforated open side pointing up, while with male clients he counted those with the open side down. An even number of shells (matna, literally ‘it is enough’) indicates positive developments, while an uneven number (dafa mankee, literally ‘there is a lack’, from the French manquer) indicates obstacles or difficulties in the predicted course of events.
Considering the large number of different positions that the cast shells can form, and remembering Samba Nguer's emphasis upon the necessity of learning how to recognize and interpret them, technically cowrie shell divination can be described as a process of grasping the meaning of significant patterns and constellations of shells within the larger layout in which they appear. The interpretative process concentrates on the identification of relevant topics (social relationships and conflicts, personal projects such as marriage, work or migration) and conditions (such as unrest, ambivalence or illness). This process of identifying the issues most urgent and relevant for his client is complemented by a parallel interpretative move of identifying the specific value or tendency of the issues indicated by the divinatory signs. By translating the signs found in the divinatory layout into statements about the client's personal situation and/or affliction, the diviner is able to gain insight and to develop a far-reaching understanding of the specific case.6
FROM TECHNICAL KNOWLEDGE AND LINEAR INTERPRETATION TO FORCE AND INTUITION
In practice, one of the main difficulties in cowrie shell divination seems to be deciding which of the many positions appearing within the complex constellations of shells resulting from each cast has to be considered meaningful and which can be neglected. Another difficulty that I sensed consisted in the fact that many almost identical positions in cowrie divination can be interpreted in different ways: a circle of shells appearing in a particular layout, for instance, can be conceived of as the house or compound (kër) of the person, indicating the involvement of the family and relatives living with the consulter; in another consultation/layout a similar circle of shells could indicate the ill-intended speech of others (catt) likely to necessitate ritual protection, or a place or job opportunity within an institution or company (plas). What this seems to indicate is, on the one hand, that the meaning of a particular layout or sign is not only derived from the positions and constellations themselves but also from the context of the consultation: that is, from the diviner's understanding of his client's situation arrived at prior to or during the consultation, as well as from his general insight into the material and psycho-social conditions of his clients. On the other hand, however, when a particular layout or sign has several different or ambivalent meanings, it was clear in most of the consultations I witnessed that the ability of the diviner to actually identify the intentional concern of the client, to assess the possible developments of the issues at stake, and to prescribe the necessary ritual remedies depended not only upon referential knowledge but also on other, non-inferential forms of knowing drawing neither on the explicit value of the signs appearing in the shells nor on the diviner's knowledge of his client's plans, situation or affliction.
Mandinka-speakers refer to this general ability of the diviner to read the divinatory signs and to come to know and reveal what could not be known from the consultational context itself as ‘having force’ (ka semboo soto). On the one hand, this expression simply transfers the phenomenon of physical force to the domains of knowledge and understanding. On the other hand, however, it is important to realize that such an expression must also be understood in a more direct sense as denoting a force that, although perhaps not physically tangible, is nevertheless conceived of as a real quality, manifesting itself with every successful divinatory performance. The difficulty here, of course, lies in the fact that while from the perspective of the client a quality such as semboo is perceived as deriving from the authority of the diviner's interpretation and performance, a notion such as ‘force’ is difficult to describe and grasp in itself. This is especially true because – unlike in other cultural contexts where the diviner's revelatory abilities are, for instance, compared to the ability of the hunting dog to sniff out what cannot be seen with the human eye (de Boeck and Devisch 1994 and Devisch 1999: 93–116 provide a detailed discussion of the multiple symbolic and cosmological dimensions at play in Luunda basket divination and mediumistic divination among the Yaka in south-west Congo) – the notion of semboo is generally not described in more detail and thus offers few additional hints as to the nature or character of the diviner's ability to divine.
Pragmatically, the ability of the diviner to know how to interpret certain signs in a specific situation without having extra-contextual knowledge about his client can perhaps best be attributed to an intuitional insight into the meaning of the divinatory patterns of the cast shells. It is an intuition that is gradually developed through the practitioner's growing experience and his increasing trust in his own immediate apprehension of the nature of a specific case. Of course, such a pragmatic construction of the force of the diviner as divinatory intuition just replaces one unfamiliar notion (force) with another, maybe more familiar one (intuition), without actually being able to reach to the core or possible origin of the phenomenon, let alone arrive at some kind of scientific explanation of it.Footnote 7 However, the description of the interpretative process as being based on both inferential (contextual knowledge, reading of signs, interpreting the client's individual reactions during consultation, etcetera) as well as more intuitional ways of knowing is useful for at least two reasons. First, by indicating that cowrie divination (like most other forms of divination) draws upon a technological basis and method that can be learned and acquired, one avoids an overly mystifying reading of divination as if it were a way of knowing incomparable to other, ‘normal’ modes of cognition. Second, by stating that the insight developed by the diviner during the consultation can be reduced neither to the result of a schematic application of its underlying method, nor to prior contextual knowledge about his client, one avoids the over-rationalization of the divinatory art.
Although the above offers a useful starting point in obtaining some idea of what divinatory praxis is, I would argue at the same time that the exclusive description of divination in terms of its formal principles and how the diviner is able to apply these principles, however nuanced and detailed, must always remain incomplete. It overlooks the fact that what is most relevant about divination for the persons involved in the divinatory encounter are not the (extra-)cognitive qualities of the diviner's performance but its consultational quality. Divination, in other words, is subjectively significant primarily because it responds to the inquiry of the client or patient about his most urgent personal intentional concern (Mandinka nganiyo) or need (hajoo), and allows for the identification of the ritual remedies necessary for its solution. The significance of the divinatory performance thus depends primarily upon whether the diviner is able to locate, articulate, and bring into view the subject's intention, and to concretize those aspects of the subject's concern that were unclear and unsure for the client at the moment he or she decided to take recourse to divinatory consultation, and which were thus central to the expectation of the subject when entering into the divinatory encounter. It is this subject- or Daseins-relatedness of the divinatory event, its subject-related specificity, that makes the enunciation immediately significant for the inquiring subject.Footnote 8 Consequently, from the perspective of the client, how a specific technique has to be used, how the divinatory instrument is read, and whether this interpretative process depends upon technical, intuitional, or other forms of knowledge is almost irrelevant as long as the diviner is able to locate his client's concern, to analyse his situation, and to point out the necessary ritual remedies. Consequently, while an approach to divination focusing on divinatory method and the person of the diviner as bearer of a specific corpus of knowledge helps to understand certain aspects of divination as a(n) (extra-)cognitive epistemic activity, such an approach almost inevitably falls short of understanding how divination actually affects the subject in his or her personal situation as a consultational and potentially therapeutic encounter. Another difficulty in relation to the analytical focus on the diviner and the question of the nature of divinatory cognition arises when one looks at the fact that within the divinatory discourse it is not the diviner who is considered to be the source of the divinatory enunciation but the divinatory apparatus itself.
THE QUESTION OF THE SOURCE OF DIVINATORY KNOWLEDGE: THE DIVINATORY APPARATUS AS ENUNCIATING SUBJECT
According to Senegambian diviners, the real source of the divinatory enunciation lies not within the person of the diviner but in the divinatory apparatus itself, that is, depending upon the method employed, the cast shells, the geomantic signs, the roots, etcetera. This authoring role of the divinatory apparatus becomes explicit in the way diviners pronounce the results of the divinatory procedure. Rather than presenting these results as their own insight, divinatory statements are regularly introduced with the phrase ‘it says’ (Mandinka a koo) or ‘there it is’ (a fele) indicating that the pronounced findings are not the diviner's private opinion or invention but revelations originating in the divinatory instrument itself. This locating of the source or subject of the divinatory enunciation in the divinatory apparatus (or, depending upon the underlying epistemological assumptions, in the divine, spiritual, or ancestral entities considered to be expressing themselves in the respective material, human or animal medium), is not limited to the Senegambian context but has also been noted in other divinatory traditions (see for instance Zempléni 1982 and 1995 for Senufo divination and Devisch 1991: 129 for Yaka divination). Despite its perhaps universal character, this feature of divination has received little attention in anthropological analysis. Offering a detailed treatment of this peculiar feature of divination, Andras Zempléni considers ‘the human speaker's evacuation as the subject of the (divinatory) enunciation’ to be divination's main pragmatic property (Zempléni 1995: 241–2). According to his analysis, the significance of the removal of the diviner as the subject of the divinatory enunciation as it is expressed through the diviner's explicit assertions, specific ritual acts establishing ancestral contact or, in mediumistic divinatory traditions, a state of trance, lies in the fact that it ‘induce(s) the impression of impersonal veracity of the mantic statements’ as well as ‘the specifically “divinatory” relationship or situation that anybody distinguishes unhesitatingly from other types of ritual or everyday relationships or situations’ (Zempléni 1995: 242).
Although capturing an important psychological aspect of how divination is typically experienced, Zempléni's consideration of this particular feature of divination as ‘induc[ing] the impression of impersonal veracity’ (my emphasis), could at the same time be perceived as problematic in so far as it may be read as reducing the phenomenon of the removal of the diviner as subject of the divinatory enunciation to a merely psychological mechanism without any further intrinsic epistemological or ontological truth. It would be a rhetorical device designed to obfuscate the fact that in reality it is the diviner (with all his possible shortcomings, intentions or hidden agendas), not the oracle or divinatory apparatus itself, who must be considered the real source of the enunciation. The various endogenous epistemologies which do attribute the main authorship in divination to the divinatory apparatus would be reduced to being nothing more than a response to the postulated psychological necessity of camouflaging the authoring role of the diviner in order to make divinatory procedure more convincing. The question that remains here is that of the possibility of an alternative reading of this phenomenon. What kind of reading or theoretical perspective would allow us to consider the removal of the diviner as subject of the divinatory enunciation in its own terms, that is, as a true etiological statement describing the source of knowledge activated and tapped into by divinatory procedure? How is it possible to describe the role and location of knowledge in divinatory praxis as being independent of the cognitive abilities of the diviner as knowing subject? How can one grasp the meaning and knowledge-producing quality of the divinatory procedure or instrument itself?
DIVINATION AS ACTIVE REPOSITORY OF EXISTENTIAL INSIGHT
Drawing on earlier work, I have argued above that, from the perspective of the client, the significance of the divinatory enunciation lies primarily in its ability to respond to his or her most urgent concerns and to indicate the necessary ritual remedies. From a subject-oriented perspective, divination can thus be perceived as a subject-related lifeworld-poiesis, a process of gradually making visible those areas of life most relevant for the individual subject in his or her current situation. If the divinatory signs themselves are to be considered the source of this process and not the person of the diviner, it must be possible to locate the knowledge or understanding of the issues at stake in the system of signs itself. In other words, the knowledge being tapped into by the diviner must somehow be located outside of his personal knowledge, predating and outliving the specific moment of the divinatory consultation. At first sight, the only case in which such conditions seem to exist is that of strictly text-based divination, that is, forms of divination where mantic procedure is used to identify specific passages in a text or book which are then read as statements referring to the client in his or her specific situation.
In the Senegambian context, several forms of text-based divination exist. One form consists in the location of a specific letter in the Qur'anic text by the calculation of prayer beads. The divinatory meaning of the respective letter is then looked up in a divinatory chart attributing different divinatory values to the different letters of the Arabic alphabet (tasabayo la jubeeroo). In another form of text-based divination, the numerical value resulting from the counting of the beads is linked to one of the sixteen houses characteristic of Islamic geomancy and interpreted with the help of a text explaining the divinatory meaning of each house. And also in geomancy itself, some diviners rely upon texts explaining the divinatory meaning of the geomantic signs, their houses, and the possible combinations in often great detail. In all these cases, the divinatory enunciation is derived through the combination of mantic procedure and a fixed catalogue of divinatory statements contained in a divinatory chart or text and thus largely independent of the diviner's knowledge of the specific signs. In these cases, the diviner's activity thus seems to fit very closely with the endogenous conceptualization of divination as a process in which the divinatory enunciation is not authored but only mediated by the diviner.
The situation seems to change when the diviner does not rely upon an actual text or chart in order to interpret the appearing signs but draws on his own, acquired and internalized knowledge of these interpretative schemes. It could be argued, however, that this situation is not so different to text-based divination in so far as the diviner still relies upon a corpus of interpretative schemes, the only difference being that this corpus has now been submitted to memory and thus no longer needs to be contained in a material copy of the respective text. This understanding of these apparently non-textual forms of divination as quasi text-based could be criticized in so far as it seems to overlook that, when a corpus of interpretive knowledge is submitted to memory, this corpus may start to change as it is no longer fixed in the form of an unchangeable written text. In response to such a critique, however, it could be argued that even if an interpretive corpus starts to change as soon as it loses its written form and becomes subjected to different consultational situations (which is not entirely evident in so far as orally transmitted and memorized bodies of knowledge often show surprising degrees of continuity and may be regarded as impersonal, fixed and unchangeable by the practioners and clients themselves), this does not detract from the fact that the diviner reads and interprets the appearing signs according to an inherited corpus of rules rather than inventing their meaning intentionally or arbitrarily.9
A good example for such a quasi-text-based form of divination is described by Richard Werbner in his detailed account of Tswapong divination, where reference to a memorized body of archaic verse is crucial (Werbner Reference Werbner and Werbner1989). What makes his account particularly interesting for the present discussion of the location of knowledge in divinatory process is his attention to the mediating role of the diviner and the complexity of the interpretative process resulting from a combination of a memorized body of archaic verse, the reading of the position of several types of lots, the diviner's contextual knowledge of his client's life history, and the client's responses to the findings expressed by the diviner. While it is clear from his analysis that in such partly text-based types of divination the actual divinatory enunciation is not fully autonomous but necessarily mediated by the diviner's interpretation, his selection of verse, his experience in reading the position of the lots, and his individual sensitivity concerning the client's situation, it is also clear that the diviner relies upon the actualization or setting in motion of a set of interpretative categories and principles that underlie and structure the divinatory interpretative process from the onset. In other words, even where individual interpretation becomes crucial, the outcome of the consultation is as much structured by the underlying corpus of interpretatory rules as it is mediated by the diviner's own interpretative efforts.
The situation seems to become even more complicated, however, in cases where the temporal succession of the execution of mantic procedures and the application of interpretative categories is reversed. In the Senegambian context, this is for instance the case with divination by the use of a divination mat (Mandinka basoo nding la jubeeroo). Here, the diviner proceeds by asking series of questions concerning the situation of his client which are then answered by the movements of the divination mat which the diviner keeps as motionless as possible between the thumb and index finger of his right hand but which occasionally moves and opens up, indicating a positive answer to the question posed or pointing toward the positive development of the issues concerned. Perhaps more than the earlier mentioned techniques, this form of divination seems to invite the scepticism of critical observers in so far as the diviner may appear to be able to manipulate the mat in such a way that it may open up when he himself wants to answer a certain question in the positive rather than waiting for the impartial answer of the mat itself.Footnote 10 However, as long as one excludes the possibility of intentional manipulation, the divinatory inquiry in these forms of divination also relies upon a text-like corpus of knowledge and depends upon responses that can only be acquired through divinatory procedure: a text-like corpus of knowledge consisting in a catalogue of questions concerning the areas of life most likely to necessitate divinatory consultation, and the responses of the divining mat in order to be able to identify the concerns of his client within the unavoidably much larger field of potential areas of concerns contained in the used catalogue of questions. In this regard, here too the enunciating subject could be said to be primarily constituted by the divinatory apparatus itself, not the diviner.
But how is this possible? How can the shells, geomantic signs, or the movement of a divining mat make meaningful statements about a client's specific situation? A partial answer to this problem comes into view if we consider the text-like corpus of knowledge underlying divinatory procedure not only as a means to rearticulate the topography of the socio-cultural lifeworld of the consulting subject but also in itself as an expression of the lifeworld it is meant to express. Divinatory knowledge systems have been formed over time and constitute a repository of insights into the existential dimensions of life in a specific socio-cultural environment. They do so in a similar, but perhaps more direct and explicit way than other cultural forms such as mythology, narrative structures, ritual symbolism or the symbolic dimensions of material culture (all of which may to different degrees simultaneously be reflected and enacted in the divinatory procedure itself). It is because of this lifeworld-relatedness, or, more precisely, because of this originating out of the multiple lived-through subjective realities forming the cultural lifeworld in a specific context, that responses to existential situations produced through divinatory procedure are meaningful. Looked at in such a way it becomes clear that divinatory enunciations are experienced as meaningful not just because of being perceived as the outcome of a procedure involving divine or spiritual forces (although theological and/or etiological assumptions do of course contribute to the way the divinatory enunciation will be experienced in the individual case). Rather it is because every possible answer formulated within this cultural system of existential analysis is already meaningful in itself as part of a corpus of insights concerning the nature of human existence produced through a long and complex process of sedimentation of experiences. A crucial aspect here is to realize that this process of sedimentation is, in the same way as the lifeworld itself, simultaneously historic and dynamic. That is, it is both a product of the past and past experiences as well as a continuous process of integrating new aspects of reality taken up and dealt with in divinatory praxis, recreating and reshaping the (experience of the) lifeworld at any given moment in time.
In this context, it is interesting to note that the continuous nature of this process of sedimentation of meaning is, in some divinatory traditions, not only the result of a diviner's (or his predecessors’) ongoing consultational praxis but may also be linked to the personal experiences of the diviner in relation to the acquisition of the different elements making up the divinatory apparatus. Writing about the meaning of the divinatory lots in Tswapong divination, Werbner emphasizes that ‘the diviner is expected to accumulate personal wisdom from the varied experiences of life … objectified in the lots he adds to the ensemble from time to time, such as the bone of a lion whose attack he survived or the shell he stumbled upon during an arduous journey’. He further emphasizes that due to this relation between the lots, their divinatory meaning, and the diviner's personal experience, a divinatory lot ‘is not only a symbol in the restricted sense of a vehicle for the conventional association of meaning’ but it is also ‘a means for the embodiment of, and thus the recall and reflection about, personal perceptions from the diviner's own life history’. Therefore, he concludes, ‘the discourse of divination resonates with experience, with echoes of the passions and suffering of a lifetime’ (Werbner Reference Werbner and Werbner1989: 32–4). In further developing Werbner's analysis, I would argue that even where a divinatory symbol or interpretative category appears to be only ‘a vehicle for the conventional association of meaning’ unrelated to the diviner's own personal experience, it always retains at least part of the direct significance of the experience it represents. This is precisely because the experiences referred to in divination are not idiosyncratic but, due to the process of sedimentation described above, part of a larger and shared cultural repository of experience, meaning, reflection and insight of often great historical depth, informing and structuring the way life is experienced in a specific socio-cultural setting (as well as the way these experiences can be articulated) from the outset.11
Looked at in such a way, it becomes clear why and in how far the divinatory process could be considered as being autonomous from the expertise of the individual diviner. It is an historically grown but dynamic system or repository of existential insight. Being activated through mantic procedures such as the calculation of geomantic signs or the casting of cowrie shells, it produces answers in ever-changing consultational situations. The capacity of meaning production contained in such a system necessarily transcends the knowledge and control of the individual practitioner. This becomes even more evident if one considers not only the force of the divinatory enunciation itself but also the force and efficaciousness of the ritual actions prescribed through and following the divinatory consultation. It is through these ritual actions that the divinatory encounter becomes meaningful even where, for instance, divinatory techniques are applied in seemingly schematic or even deficient ways. Even where calculations seem to contain faults or the diviner's knowledge of the interpretative catalogue is incomplete, the enunciation and ritual recommendations will be experienced as meaningful as long as they relate to the subject's concerns and help him or her to deal with whatever situation caused him to seek out divinatory consultation. In this regard, the proposed perspective on divination as existential repository of insight and meaning production may also offer an explanation as to why, on the one hand, divination seems to function even cross-culturally, while, on the other hand, it often seems to be confined to a specific socio-cultural environment outside of which it loses much of its appeal and power. As a field of existential understanding and analysis, a specific divinatory tradition forms a specific hermeneutic horizon, a demarcation of an area of existential concerns and understanding. As long as the individual subject's concerns fall within this area, the divinatory encounter is likely to be able to grasp the subject's situation and to unfold its meaningfulness. Where this is not the case, the enunciation will lose much of its power as the person concerned will be unable to relate to it in a meaningful way.
CONCLUSION
The insistence of divinatory epistemology on the authoring role of the divinatory apparatus shows that the question of knowledge and knowledge practices does not exhaust itself in the question of the knowing subject. Of course, divination and other rituals do not work by themselves but are necessarily mediated by individual specialists; it is because of this that no consultation or ritual event is like the other, and that different practitioners may vary in the exact procedures they employ, eventually integrating new elements and generating new forms. In this regard, divination and other ritual knowledge practices cannot be fully autonomous in the way they unfold their meaning. However, the question of the enunciating subject in divinatory discourse reminds us that the diviner, therapist or counsellor is as much written by the art he practises as he is its author. And this may be true not just for ritual practices but for other fields as well.
The complexity and richness of divinatory praxis poses multiple challenges to any theory of knowledge. At the same time, or perhaps precisely due to the challenges it poses to anthropological theory, the study of divination forces and allows us, in the best phenomenological fashion, to bracket preconceived notions and to take a closer look at things themselves. The fact that divination only unfolds its meaningfulness in an interpersonal consultational setting, for instance, reminds us that divination is, above all, a crucial means to come to terms with difficulties and situations that cannot easily be solved otherwise, even if it constitutes only one of many steps in a much longer process of dealing with the situation at hand. Thus divination is not just an intellectual practice or object but a consultational practice dealing with existential situations in ever-changing contexts. In other words, divinatory practices, and perhaps knowledge and knowledge practices more generally, imply more than cognitive operations and positions. The intersubjective consultational setting in which these practices unfold is not accidental but essential to the form, content and sociocultural significance of these practices. Any reflection has to start from here.
ACKNOWLEDGEMENTS
I am grateful to Trevor Marchand and Kai Kresse for their thoughts and comments on the present text. I thank both of them as well as Jan Jansen for inviting me to the panel ‘Expertise and the Transmission of Knowledge’ at the 2005 AEGIS conference in London, where an early version of this article was first presented. For revisions, I benefited from the comments of two anonymous reviewers and a close reading by the editor of Africa herself. The text incorporates parts of and also extends earlier work in Graw 2005. I gratefully acknowledge the financing of this research by the Fund for Scientific Research – Flanders (FWO).
