“Gadè, deceptions and lies told by the ill :
the Caribbean sociocultural construction of truth
in patient-healer encounters”.
Un article publié dans la revue Anthropology & Medicine, vol. 9, No. 2, 2002, pp. 175-188.
- Introduction : beyond romanticism
- The patient-healer encounter
- The role of lying in client-healer interactions
- The lie in the context of a language-game that makes sense of reality through narratives
- Language and idioms
- The lying-game and acting on the social stage
- Lying and the moral community
A constructivist approach in medical anthropology suggests that the boundary between lies and truth in sickness narratives is thin. Based on fieldwork in the French (Martinique) and English (Saint-Lucia) Carribbean with gadé and quimboiseurs (local folk healers), this paper addresses the gap between naïve romanticism and radical cynicism in the anthropological analysis of patient-healer encounters, Is the sick person lying when she accuses evil spirits for her behaviour or sickness ? Is the quimboiseur who is building a meaningful explanation or diagnosis simply a liar taking advantage of his client's credulity ? The challenge for anthropology is not to determine whether or not a person is lying when attributing their ill fortune to witchcraft. Instead, in this paper, the author approaches lying as a language-game played by both patients and folk healers. Concepts of lying as games, tactical lies, pragmatic creativity, and constructive lies are introduced here as a perspective for a reconsideration of lying as a pertinent research object.
Writings by anthropologists on ethnomedicines and folk healers are often tinged with a sort of naïve romanticism. A focus on the analysis of healing rituals, ethnomedical knowledge and magico-religious beliefs as coherent symbolic systems has taken precedence over critical reflections on the uses and abuses of the magical or religious power of healers. The illness narratives of sick persons, are submitted to hermeneutics and interpretativist analysis, while discrediting any questions regarding the veracity of the content of these accounts. Anthropologists, with reason, do not set out to separate truth from lies in illness narratives, to uncover charlatan healers or analyse the real or placebo medical effectiveness of healing practices. This concern for anthropological correctness can be explained, in part, by the negative, moralising image associated with lies and deceit, and by a preoccupation for a methodological cultural relativism that prohibits any value judgements not based on an analysis of the viewpoint of those being studied. Anthropology must avoid the pitfalls of a totalistic positivist deconstruction. Nonetheless, the study of lying and deception by the social sciences has demonstrated the importance of the role they play in the intersubjective construction of reality (Barnes, 1994 ; Duncan & Weston-Smith, 1979 ; Saarni & Lewis, 1993). In a deliberately provocative approach, I will suggest some analytical tools that may guide a study of the role of lying and deception in ethnomedicine. The patient-healer encounter can in no way be reduced to mere lying and deception, although anthropology should not deny that they play a certain role in diagnosis and therapy. However, the starting point must necessarily be a redefinition of truth as a strategic cultural construction rather than an empirically established fact. I suggest that we consider the encounter between the ill person and the traditional healer as a meeting of two discourses that more or less consciously integrate truths and lies, franchise and deception for the production of a dynamic truth. This process, we suggest, relies in part on a consensual lying-game and on tactical lies. It will be illustrated using examples from gadé practices on the Caribbean island of Saint-Lucia and quimboiseurs in Martinique.
The patient-healer encounter
In Creole societies of Saint-Lucia and Martinique, the explanation of illness and misfortune is always based to some extent on a reference to magico-religious forces. Disease, behavioural problems, depressive episodes, outbreaks of violence, as well as marital and professional problems are frequently explained by the action of a work that is done on the ill person. A jealous or envious person who wishes to cause harm will consult a gadè (Saint-Lucia and Martinique) or a quimboiseur (Martinique) to cast an evil spell on the ill person. Thereafter, that person will be defined as the victim of witchcraft. The victims then consult either a gadè or a quimboiseur to be released from the spell. In such cases, they play the reverse role of traditional healers. The magico-religious world to which these healers refer does not involve the same degree of complexity and structure as the world of Haitian voodoo houngans or other African ethnomedicines, for example. The evil spirits considered responsible for health problems do not have specific personalities, nor do they belong to any pantheon of divinities. Sometimes they are associated with demons or with Satan. Likewise, the origin of the supernatural forces invoked by the gadè or the quimboiseur to undo evil spells is not specified. The ill person's trust in the gadè is not based on any presumed mastery of a determined corpus of occult knowledge that the gadè is considered to possess via a complex process of handing down knowledge from one generation to the next. It is based on the credibility that he or she has acquired through previous practice, following a natural gift, a revelation or apprenticeship with a master. The following case is an illustration of this practice.
A young Saint-Lucian  man has symptoms of depression and various behavioural disturbances. He takes hard drugs, which lead to episodes of violence and ensuing problems in his family relations and friendships. With the support of relatives and friends, he manages to find a multitude of small jobs. But after a few weeks or months of work, his employers inevitably fire him. He believes that he is the victim of 'work' done on him by someone who wishes him harm. He and his grandfather agree that he should consult a gadè. After listening to the young man's account of his misfortunes and his interpretation of them, the gadè agrees that the man is under an evil spell. The gadè is famous in the region for his powers, although no one really knows where he comes from. The gadè suggests that he use three sorts of oils : one on his body, the other to be mixed in with his bath water , and a third that must be sprayed on his body. At each visit, the patient is seated in the centre of a bath containing a secret mixture and washes himself with his underwear, which he has worn for three consecutive days. The gadè wrings out his underwear three times and drinks the liquid collected in a bowl. The ritual is completed by drinking various secret concoctions, by rubbing a Vaseline-based mixture into the client's hair and by a short Catholic-inspired prayer to God. Each visit costs approximately US $100. Each renewal of the prescription for the oils costs US $100 including a substantial amount for the parchment paper on which the prescription must be written. Today the young man says he is cured and has recently been given a job. Throughout the interview, the young man appears to be aware of the fact that his drug addiction is partly responsible for his problems. The fact that he suggests to the gadè that witchcraft is the main cause of his problems is more a tactical selection of information than an outright lie. Indeed, he refers to the pressure his mother and grandfather have been exerting on him to consult a gadè. He accepts the diagnosis that those in his social circle have made of his condition. On the other hand, interviews with the gadè reveal that he himself is well aware of the social and personal causes of his clients' health problems and other misfortunes.
There is not enough space here to describe the many other examples. However, it is worth giving two short illustrations. A young woman in Martinique reports to a quimboiseur (local folk healer) that the voices she hears and the violent outbreaks that plague her family are not related to a depressive state connected to marital and professional problems, but to an evil spell. A man accused of being a rapist after trying to break into the private house of a young woman by night says he was not in control of his mind since somebody turned him into a dorlys . The quimboiseurs consulted by these victims of witchcraft confirm to their clients (and also to the doctors and lawyers), based on their experience and knowledge, that in fact they were not responsible for their acts, their minds being altered by the work of an evil spirit sent to them by a jealous person. Are the sick persons lying in accusing evil spirits for their hallucinations and psychotic crises ? Are the quimboiseurs, who constructed or confirmed these meaningful explanations or diagnoses, liars taking advantage of their clients' credulity ? As anthropologists know, understanding such situations is much more complex. However, interpretativist-constructivist approaches in medical anthropology suggest that the boundary between lies and truth in such narratives is thin. In fact, in the encounter between the healer and the sick person, lies and deception may play a role that is not to be neglected. These may include :
- • Lies told by the ill concerning the true causes of their illness. By blaming witchcraft for their woes, the ills can shift responsibility away from their own weaknesses, addictions, alcoholism, and uncontrolled sexuality, and their share of responsibility for the diverse misfortunes (marital, professional, etc.) that bring them to consult a quimboiseur or gadè.
- • Lies told by healers concerning the true source of their powers, their training, their revelations, their contacts with supernatural forces, their previous successes and treatments with other clients, and the limits of their powers.
Obviously, I am not suggesting that traditional medical systems and traditional healing practices should be reinterpreted as based on pure lies and deceptions. Anthropologists know well that the social and cultural context of the encounter between sick persons and healers, and their symbolic universe, is complex. We know that help-seeking behaviour is driven by a true belief in the existence and efficacy of supernatural mediation. The high price paid by the patients confirms that they do somewhat believe in the healing encounter. Moreover, illness behaviour cannot be explained without reference to the local culture as a coherent cultural system into which healer and ill person are enculturated. However, if we limit our understanding to this culturalist interpretation and to an analysis of idealised symbolic systems, we are prey to a naive and romantic interpretation that disregards the complexity of the relations between truth and lies, between deception and empirically based healing practices. One major challenge for anthropology is to move beyond such a dichotomic, manichean reading of the trust relationship that develops between the ill person and the healer. What I suggest here is that we are dealing with a dynamic process of communication and a cultural construction of shared, negotiated and legitimised truth.
In the modem pluralist medical context now prevailing in Saint-Lucia and Martinique, everyone has some basic medical knowledge. They also have access to biomedical services and are not restricted to a single, homogeneous, closed ethnomedical system. Without going into the debate on how people make strategic choices between the various health care systems available to them (biomedicine, obeah man, quimboiseurs, bush doctors., Pentecostal healing services, and so on), we can suggest that the ill person, if not lying stricto sensu, may sometimes adapt the content of his or her illness narrative to the nature of each healer consulted. An analysis of the discourse on therapeutic paths confirms that people from the Caribbean region select what they tell the healer in question, omitting certain details, emphasising others, and so on. Conversely, pastors, quimboiseurs, and physicians adjust their discourse to fit into the patient's beliefs. Thus, patients and healers leave out certain facts or comments, focusing instead on other issues and interpretations. Lies therefore take on a variety of forms. But what is the nature of lies ?
The role of lying in client-healer interactions
Among the different taxonomies of the lie, Saarni and Lewis (1993) distinguish three types of deception based on the state of the deceiver's awareness of their deception as well as the subject of the deceiver's action. These are :
- (1) ordinary deception toward others committed with self-awareness. Here deception is designed to consciously fool others or to hide personal actions from them ;
- (2) deception toward others that requires some degree of self-deception. Here deception can be involuntary or unconscious ; and
- (3) self-deception even in the absence of another, that is, the need for illusion (Saarni & Lewis, 1993, p. 9).
In the case of the relationship between an ill person and a healer, I believe that we are dealing with a combination of the three types of lying-deception. Illness narratives include statements that are most often partly true and partly false. Insofar as truth and truthfulness are not context-free notions, lying cannot simply be the opposite of telling the truth. As concerns the client, we are dealing less with pure inventions of symptoms, causes or circumstances than a tactical selection of what to remember or forget, to say or leave unspoken, and what contextual elements, hypotheses and interpretations to provide. In this context, anthropology is less concerned by truth and falsehood that hinges on issues of the epistemological foundations of statements in a positivist perspective than with truthfulness and deception that belong, for Bok (1978), to the moral domain of intention. Following Barnes, we define intending to deceive not simply as intending to communicate a falsehood but "intending to cause a dupe to adopt an understanding of the state of the world and/or of the mind of the liar that the liar believes to be false" (Barnes, 1994, p. 11). In such a context, readers should not ask what the young Saint-Lucian man himself consciously recognises as true or false in his narrative. This is not a pertinent question for anthropology. However, two conditions are required for moving beyond a positivist agenda of separating truth from untruth, reality from deception. First, we must transcend a strictly interpretative anthropology preoccupied with the symbolic content of exchanges between the ill person and the healer. We will come back to this in the next section. Second, we must analyze communication dynamics. Truth is not static, but dynamic. As Eck (1970) suggests, truth should be viewed as a dynamic process supported by a "pragmatic creativity".
Anthropologists should also examine the adaptiveness of liars using deception and falsification functionally and strategically for personal as well as for collective goals. From that perspective, lying can be viewed as a tool to strengthen the patient's belief in the healing power of the gadè or quimboiseur. Although lying may be false, it is also tactical. Lies about the sources and limits of healing powers can be interpreted as a tactic used by healers to strengthen their control over the community of the ill. Likewise, lies are used as tactical tools by the ill to mask the real (personal) sources of their misfortunes or illnesses. Lying becomes a tactic in constructing the morality of both healers and patients. As George Simmel (1906, in Petitat, 1998) reminds us, one of the conditions of communication, indeed of mutual understanding, is a "communicative reduction" whereby individuals select what is to be said and what will remain unspoken. Thus, the social construction of truth refers to a dialectical interactive process between knowing and not knowing.
Generally, definitions of lies implicitly suggest an interaction between a liar and a dupe. However, in the patient-healer encounter, both interlocutors are to some extent both liar and dupe, even if no one is systematically lying or being completely deceived. Here, the goal is not to mislead the recipient of the message but instead, to guide his diagnosis toward a cause that lies outside the ill person. The goal is not to harm a third party but instead to enhance the dupe's interest in his or her case and to steer the diagnosis in a direction that will shift responsibility away from the victim. The lie will be benevolent, but in this case, for the liar. The goal of the sick is not to deceive the healer. Instead, it is to guide his diagnosis and therapy towards the "work" that is being done on him by someone and to rid him of his guilt. A patient's ultimate goal when integrating parts of "social lies" into his narrative could be to justify a therapeutic prescription that removes responsibility from him by legitimising the identification of causes that lie outside him, for example, by confirming that he is indeed the victim of witchcraft, an evil spell ordered by a jealous, envious or vengeful enemy who wishes him harm. The language-game therefore operates on the basis of local idioms of illness explanation.
This approach to lying does not, however, imply a radical relativism, nor a nihilism that denies any relation to truth. From a constructivist perspective, truth must be viewed as a socially, culturally and historically determined construct. What is problematic in truth is not its existence ; it is its claim to universality and the absolute. What is of interest, then, is not to determine whether the sick person and the healer are telling the plain truth. Neither is it to evaluate, through scientific measures and experimentation, whether their claims to heal or to have been healed are empirically based. The challenge instead is to identify the mechanism of truth (or deceit) construction.
The lie in the context of a language-game
that makes sense of reality through narratives
Lying, as well as deception and secrets, is inseparable from the conventional rules that govern social relationships. It is part of a game of veiling/unveiling, secrets/confessions, truth/lies that refers to what Petitat (1998) calls a systematic process of virtual reversibility between the hidden and the revealed. Anthropologists must, then, be aware of a kind of language-game played by both the victim and the quimboiseur. The victim as well as the quimboiseur are using these folk beliefs to build a narrative that makes sense for them and for those close to them. A first challenge, therefore, is to uncover the sociocultural process of meaning construction. The concept of language-game in the philosophical pragmatist tradition (Austin, 1970 ; Wittgenstein, 1953) can help to explain the connivance and complicity between the sick and the healer. What interests us is how illness narratives are co-constructed through a process managed by the ill person and the healer. Our focus is the cohabitation of truth and lies in the "clinical" encounter, or the strategic selection of certain facts, certain contextual elements, and certain plausible explanations of the illness, to the exclusion of others. Being careful not to fall into a semiotic analysis of lying through the properties of the meanings as such, we must also place the study of lying and deception in the domain of pragmatics, that of patient-healer negotiation for services and that of a quest for legitimacy and morality (Brodwin, 1996). Katon and Kleinman (1981) have pointed out the importance of communication in the physician-patient relationship. To convey the cognitive-interpretative dimensions of the patient-healer encounter, they developed a "transactional model" that emphasises the negotiation process between two divergent explanatory models of illness : the medical-clinical model and the patient-popular model. Integrating lying and deception in such a transactional model, as applied to Creole Caribbean societies, would suggest that what is negotiated is not a compromise between the healer's truth and the patient's, since those truths are defined in the context of different medical systems. Here, patient and healer belong to the same ethnomedical system. What is negotiated, then, among other things, are maybe the limits of lies and deception, that is, the limits of complicity. Thus, the challenge for anthropologists becomes the analysis of the negotiation process of truth and lies, and the unravelling of narrative construction mechanisms and the personal interests both parties have in that narrative construct.
Wittgenstein's later philosophy (1953) rejected the referential theories of meaning that portray meaning as something that is fixed by a referent, and that treat language and meaning as if they were separate from human action. Language and meaning are inseparable from human action ; meaning can only be an experiential phenomenon (Wittgenstein, 1953, p. 23) in no way reducible to a fixed connection to a referent. Meaning evolves through "language-games" as ventures in meaning creation. Meaning, and here, the consensual meaning developed by patient and healer, must be seen as an experiential phenomenon, as a production of the use of language in lived situations. To consider that language becomes an instrument of meaning production in context suggests the need for a reinterpretation of the concepts of lies and deception, no longer in terms of immutable relations between signifier and signified, but in terms of the elements of language that play a central role in the communication dynamic.
To take up Jankélévitch's analysis (1979), this language (lying)-game may be, above all, an irony-game. As concerns the patient, we are dealing less with a lie (whose objective clearly is to fool, to distort reality to mislead the interlocutor) than with irony, which, as a false lie, is intended to be understood and deciphered by the healer. The ironic person is not lying ; he feigns ignorance of what his interlocutor clearly knows. He speaks as if his interlocutor in no way doubts what he is saying, knowing full well that his interlocutor is not taken in. In short, irony involves a strong dose of complicity between the ill person and the healer. It does not call into question the sincerity of the interlocutors. Indeed, for the healer, the objective is not to lie or mislead directly ; rather, the healer lets his patients return home with their illusions intact regarding his alleged healing powers.
Neither the healer nor the ill person ultimately aims to deceive. No one wants their lies to be believed in full. For example, in the example cited above, the young drug addict's objective is not to deceive the gadè when he puts forward the hypothesis of witchcraft ; he is simply inviting the healer to decipher his message, which is that he is seeking to be relieved of responsibility and guilt. The gadè does not expect the young man to believe entirely in the direct power of oil baths and parchment paper. Both actors are sincere accomplices. Their goal, and the rule of the lying-game, is not truth by veracity, plausibility, and truthfulness. Patient-healer encounters are based on strategic misinterpretations, overinterpretations and a variety of omissions or deformations of truth, which can be seen as lies of good faith or "lying truths" (Duncan & Weston-Smith, 1979). The healer's irony is intended to veil, but is inspired by altruistic motives. It assumes that the ill person will be capable of lifting the veil. As actors in a ritualised encounter, they are both involved in a game of self-deception leading to a reciprocal positive feedback relation. To see the ill-person-healer encounter as a deception-game may explain why sick persons continue to have confidence in traditional healers and trust them even if they have often lied to them about the causes or the nature of their sickness. As for actors in a play, the issue here is not the truth in the text, but the game (the acting) that gives credibility and legitimacy to both patient and healer. Just as the actors' skill at acting gives a play its credibility as a legitimate and valued art form, patienthealer interactions confirm the legitimacy of ethnomedical knowledge and practices. What is at stake in this deception-game therefore goes far beyond the patient-healer relationship : it is the reproduction of the local ethnomedical system.
We can also hypothesise that ill person and healer, in some ways, are lying to themselves as well as to each other and to the community. The language-game in which they are engaged is a game of self-deception played in order to protect themselves against the disappointment of their respective relative incompetence in managing their health and personal lives or in the face of the limits of their therapeutic powers. We cannot lie to others without to some extent lying to ourselves. The construction of deception always entails self-deception. If truth is a sociocultural construct among a community of locutors, and lying an act of language between at least two persons, lying to oneself as a form of self-deception can be defined as an inner connivance, a social interaction with oneself or an internal audience, as part of a sort of reflexive dialogue with oneself. The discourse constructed in a patient-healer encounter is characterized by an interactive sense of deceit. In fact, each person experiences a tolerance toward illusionary consciousness and self-deception.
As interesting as it may be, this image of the language-game also has its limits. As in a poker-game, the language-game has its rules, based on customs, both stated and unstated. The language-game of lying needs to be learned (Wittgenstein, 1953, pp. 198, 199). However, the rules of the game, as well as the process of interpreting signs (speech, acts, behaviour, etc.) of both healer and ill person, are cultural and emotional constructs. The lying-game is governed by expressive as well as by instrumental reasons (Barnes, 1994, p. 5). It is also governed by social norms (that define when and to what extent one can lie, whether or not ridding oneself of guilt is normal, how much emphasis to place on the lie without jeopardising the interlocutor's complicity) and cultural idioms that frame the content of the discourse on the nature of the illness, the physical and mental expression of symptoms, and the imagined causes (e.g. marital conflict, property disputes with neighbours, jealousy, etc.).
Language and idioms
The shared knowledge of local idioms of illness expression, interpretation and explanation enable both healer and ill person to decipher the hidden meaning behind the said and the non-said. A lie, as well as a truth, must first be examined as a sociocultural construction framed by the local culture before being analysed pragmatically as a means to serve social and individual interests. While lying and deception can be understood as a component of a language-game, they are part of language first before being a game. So, another challenge for anthropologists could be to uncover the process of meaning (and lying) construction in specific cultural contexts. In any cultural context, this lying-game is founded on a series of idioms or cultural forms that frame the language used to describe the problem that is the object of the consultation, its manifestations or explanations (Massé, 1999). Thus, each culture has its own idioms for identifying illness and ill fortune (names, labels given to the problem). These idioms refer to culturally accepted modes (physical, verbal, emotional, behavioural, etc.) for expressing the problem and idioms of explanation that frame the identification of plausible causes put forward when a diagnosis is being made and relevant treatments are being chosen. In the cultural context of Creole Caribbean societies, for example (Mass&, 2001), the idioms of explanation are :
- • The devein, a sort of chronic bad luck that follows an individual for a long period of his or her life and become a specific case of a logic of unhappiness. It refers to a chronic fatalism that requires the intervention of an outside supernatural force responsible for the series of misfortunes.
- • This devein is almost systematically reinterpreted as the work of "someone who wants to harm you", most often out of envy or jealousy. jealousy, viewed as a sort of "sorcerer's brother" that motivates neighbours to cast evil spells, can be caused by an infinite variety of situations : jealousy of a rival, in love or in business ; rivalry in situations of social competition ; envy towards anyone who has managed to climb the social ladder.
- • Spirits that refer to entities without their own identities but that can be manipulated by a gadè or quimboiseur for the purpose of either good or evil.
- • But the central idiom around which all other idioms of explanation gravitate is the quimbois, which refers both to a magic or evil spell that affects a victim, and to the object (miniature casket, piece of clothing, etc.) that serves as the medium for the spell.
These misfortunes are understood as consequences of a work that is causing harm to the person consulting the quimboiseur. This is where the search for relief from guilt and responsibility take root, a theme that invariably emerges in the illness narratives that patients tells the quimboiseur, gadè, Pentecostal pastor, charismatic priest or physician. Notions of truth, lying, deception or secret cannot be understood without a continual reference to the cultural forms that define the causes of ill fortune. It is around such idioms that the discourse on the causes of ill fortune is constructed. It is such cultural forms that set the stage for the lying language-game and the sociocultural, but also intersubjective, construction of truth and deception.
The lying-game and acting on the social stage
Life is a theatre, says Goffman (1959), but not in a restricted sense. Using this acting metaphor with caution, we must keep in mind that on the social stage, the roles of patient and healer leave room for improvisation and the interference of people from the community of spectators. What is more, in the clinical encounter, patient and healer play both actor and spectator on the ethnomedical stage. In real life, "the part one individual plays is tailored to the parts played by the others present, and yet these others also constitute the audience" (Goffman, 1959, p. xi). According to Petitat, both interact in the context of a sort of modus vivendi according to which the spectator must respect the definition that the actor gives of the situation, but only up to a certain point. Likewise, the actor must give a coherent, plausible and respectful definition of the situation to the spectator (Petitat, 1998, p. 166). Clinical encounters may be seen as an interactional game-space that excludes both complete authenticity and complete falsehood and in which the roles are largely predefined. Lies and deceit are to a significant extent expected, both by the healer and by the ill person. Gadè and quimboiseur anticipate the motives of the consultation and the discourse of accusation, jealousy and interference by evil supernatural powers. They also anticipate their own diagnosis and even the structure of the prescription, leaving room for some improvisation in the details of the therapy. Clinical encounters are not the time to denounce the incoherence or contradictions in the illness narratives of the sick, nor the apparently unrealistic, whimsical prescriptions of the healer. Both parties must respect the culturally constructed limits of credibility in accounts of illness and therapeutic proposals. It is more important to appear to comply with each other's discourses than to sort through what is true and false.
To say that the lying-game is played on the social stage does not mean that what is at stake is an invented illness. The disease or misfortune reported by the ill person can be real and entail great suffering. As in a poker-game, the money at stake is real. It simply reminds us that the language as well as the "game" is never context-free. Here lying is a matter of bringing certain truths and interpretations of events to light, at the expense of others, in specific social and cultural contexts.
Lying and the moral community
Patients and healers interrelate in the context of a given moral community. From an ethical perspective, such a community refers to all those whose relations are structured by rights and obligations and who are subject to normative restrictions that direct the behaviour of moral agents (Goffi, 2001, pp. 213-216). These norms of conduct and obligations are defined by a given community. It is this same community that defines the boundaries between truth and falsehood. Similarly, the patient's trust in the gadè's or the quimboiseur's word emanates not only from prior experiences with healing practice, but also from the credibility and legitimacy other members of the community give to their healing ministry. In short, the lying-game has a communal dimension.
Some see lying as a threat to the community. Others see the lying-game as a tool of community reinforcement and cohesion. On the negative side, for example, while recognising that lying can be "an essential lubricant in the gears of private life", some believe that it can become a serious threat to public life, undermining the foundation for trust in social institutions (Pratte, 1997). For Wallach, lying can do "damage to society as a whole by decreasing the level of trust in the community" (Wallach, 1998, p. 1). It risks harming social trust and creating disappointment in significant community leaders. But the threat is real only if we adopt a positivist definition of truth in terms of its bearing to "reality". Where truth is seen as a sociocultural construction, and lying as a constructive game, a positive perspective reveals lying and deception as tools to reinforce the moral community and as components of an enterprise in building a moral community. The lying-game can then be seen as a communication setting for communal solidarity building. Lying and deception-games fulfill a basic function of reinforcing group solidarity. Basso has suggested that the Kalapao of the Amazon "understand deception to be a fundamental mode of insight and understanding in human thought" (Basso, 1987, p. 351). In his view, it creates not only deficits and threats to community, but also opportunities and potential for solidarity.
This alternative, positive interpretation of lying must be repositioned in the context of the ill person's limited rationality and uncertainty. Patients usually do not really understand what is happening to them. They are searching, if not for certainty, at least for explanations put forward by credible leaders, and that will have credibility in the eyes of the community. This uncertainty is shared by healers, who are equally insecure about the limits of their powers. The deeper meaning behind lying and the deception-game is probably to manage this uncertainty by staging it in the context of an interaction between patient and healer, but an interaction framed by local cultural idioms and legitimised by community solidarity in the sociocultural game rules.
In the context of the encounter between an ill person and a healer, lies must not be rejected as invalid data that should be excluded from analysis, as biases in the description of "real" ethnomedical practices (Bleek, 1987 ; Salamone, 1977). Let us be clear : it would be simplistic, even ridiculous, to maintain that lies and deception constitute the essence of consultation practices. However, anthropologists must seriously question to what extent lies and deceit are integral aspects of such encounters in the modem Caribbean context.
Are magico-religious beliefs in therapeutic powers a sort of consensual and mutual deception, a deception-game for which the culture defines the rules by which healers and sick persons voluntarily play ? The answer is partly yes, partly no. This formulation is overly simplistic. While Martiniquais and Saint-Lucians do not blindly believe in the influence of supernatural forces that either heal or make ill, they do suspect that such forces exist and can have an effect on them. They have a bias in favour of traditional healing practices and interpretations. These beliefs are part of the cultural heritage that suddenly and uncontrollably rises to the surface in times of crisis. Their existence is implicitly confirmed through the language-game played on the "clinical" stage, as well as on the social stage, through a kind of theatrical understanding.
Patient-healer relationship is not between equal partners. We must agree that they are not on the same stand. Hierarchy and asymmetric power relationship is a core dimension of healer-patient encounter. However, anthropologists must also be careful not to mechanically transfer to the ethnomedical domain the Nietzschean interpretation of religion as "the holy lie" invented by priests to deceive the laity and justify their hegemonic power over dupes, and this, for two reasons. First, in Caribbean patient-traditional-healer encounters, both parties are involved in a symmetric two-way truth/lying-game. Second, anthropologists must be careful not to overinterpret encounters between healers and the ill in terms of an asymmetric power relationship involving an usurper who takes advantage of the credulity of an innocent patient. Such an interpretation of the meaning of the sick-healer relationship should itself be treated as a case of scientific lying (which is beyond the scope of this article).
But in the context of the relativist anthropological credo, can we ask how far we can go in interpreting healing rituals as a staging of deception and dissembling, as nourishing a sort of lying-game, whose rules are shared by both the healer and the ill person, without being accused of cultural and moral imperialism ? Where do we draw the line between naïve romanticism and cynical etic interpretation leading to a symbolic interpretative violence of other cultures ? Medical anthropologists have traditionally been reluctant to address the issue of lying and deception in ethnomedical practices. In spite of the suggestion made many years ago by Kundstadter (1980) and Lieban (1990) that a comparative cross-cultural "ethnoethics" of folk healers be developed, anthropological publications are still more focused on in-depth descriptions of formal ethnomedical beliefs and rituals. All too often, ethnomedicines are analysed as idealised symbolic systems, free from ambiguities, dissimulations, or deception. The point of this article is to emphasise that a constructivist approach to truth, lying and deception frees these concepts from their positivist connotations. Concepts such as constructive lying, lying-game, and constructive lies release lying from the stigma that medical anthropologists have implicitly attached to it. Lying and deception must be rehabilitated in the eyes of anthropologists.
AUSTIN, J.L. 1970. How to Do Things with Words, 2nd ed. Oxford : Clarendon Press.
BARNES, J.A. 1994. A Pack of Lies : Toward a Sociology of Lying. Cambridge : Cambridge University Press.
BASSO, E.B. 1987. In Favor of Deceit : A Study of Tricksters in an Amazonian Society. Tucson : University of Arizona Press.
BLEEK, W. 1987. Lying informants : a fieldwork experience from Ghana. Population and Development Review, 13, 314-322.
BOK, S. 1978. Lying : Moral Choice in Public and Private Life. New York : Pantheon Books.
BRODWIN, P. 1996. Medicine and Morality in Haiti. The Contest for Healing Power. New York : Cambridge University Press.
DUNCAN, R. & WESTON-SMITH, M. eds. 1979. Lying Truths : A Critical Scrutiny of Current Beliefs and Conventions. Oxford : Pergamon Press.
ECK, M. 1970. Lies and Truth. New York : Macmillan.
GOFFI, J.-Y. 2001. Communauté morale. In HOTTOIS, G. & MISSA, J.-N. eds. Nouvelle Encyclopédie de Bioéthique. Bruxelles : DeBoeck Université.
GOFFMAN, E. 1959. The Presentation of Self in Everyday Life. New York : Doubleday.
JANKÉLÉVITCH, V. 1979. L'ironie. Paris : Flammarion.
KATON, W. & KLEINMAN, A. 1981. Doctor-patient negotiation and other social science strategies in patient care. In EISENBERG, L. & KLEINMAN, A. eds. The Relevance of Social Science for Medicine. Dordrecht & Boston : Reidel.
KUNDSTADTER, P. 1980. Medical ethics in cross-cultural and multi-cultural perspective. Social Science and Medicine, 14B, 289-296.
LIEBAN, R.W. 1990. Medical anthropology and the comparative study of medical ethics. In WEISZ, G. ed. Social Science Perspective on Medical Ethics. Dordrecht & Boston : Kluwer.
MASSÉ, R. 1999. Les conditions d'une anthropologie sémiotique de la détresse psychologique. Recherche Simiotique/Semiotic Inquiry, 19, 39-62.
MASSÉ, R. 2001. Pour une ethnoépidémiologie critique de la détresse psychologique à la Martinique. Sciences Sociales et Santé, 19, 45-73.
PETITAT, A. 1998. Secret et formes sociales. Paris : PUF, Sociologie d'aujourd'hui.
PRATTE, A. 1997. Le syndrome de Pinocchio : Essai sur le mensonge en politique. Montréal : Éditions Boréal.
SAARNI, C. & LEWIS, M. 1993. Deceit and illusion in human affairs. In LEWIS, M. & SASRNI, C. eds. Lying and Deception in Everyday Life. New York & London : Guilford Press.
SALAMONE, F. 1977. The methodological significance of the lying informant. Anthropological Quaterly, 5, 50, 117-124.
SIMMEL, G. 1906. The sociology of secrecy and of secret societies. American Journal of Sociology, XI, 442-498.
WALLACH, J. 1998. Lying, an ethical culture perspective. Available online at : www:nysec.org/essays/lying.html .
WITTENGENSTEIN, L. 1953. Philosophical Investigations. Oxford : Basic Blackwell.
* Correspondence to : Raymond Massé, Département d'anthropologie, Université Laval, Québec G1K 7P4, Canada. Tel. : 1 418 656 2131 (poste 3942). Fax : 1 418 656 2831. Courriel : email@example.com.
 I have been conducting fieldwork in Martinique since 1975. Data reported here come from two research projects sponsored by the Conseil de la Recherche en Sciences Sociales du Canada and from many fieldworks conducted between 1995 and 2002 in Martinique, Saint-Lucia and Barbados. Six graduate students have also done intensive fieldwork in the Caribbean in the context of those projects. The analyses presented are parts of that research program concerned with the interaction between ethnomedicines, biomedicine and churches' healing activities.
 The therapeutic use of baths is very widespread in Creole Caribbean cultures. "Bains marrés" are used to free individuals of evil spirits.
 A sort of supernatural being that can possess an individual and make him or her invisible or minuscule so that he or she can enter into houses through a keyhole. It is in the habit of raping women in their sleep. The women apparently do not realize they have been raped until they awake and notice the traces of their orgasm.