Deadly Words: Witchcraft in the Bocage 
by Jeanne Favret-Saada, translated by C. Cullen.
Cambridge, 271 pp., £17.50, December 1980, 0 521 22317 2
Show More
Show More

Jeanne Favret-Saada’s book, Deadly Words, subtitled ‘Witchcraft in the Bocage’, deals with a subject of abiding fascination. While there are always people who readily admit to practising good or ‘white’ witchcraft for the benefit of their fellow mortals, witchcraft is usually defined negatively as the ability to cause others harm by the use of one’s psychic power. For historians and social anthropologists, most of whom do not themselves believe in the reality of witchcraft (good or bad), the problem is to decide how to interpret such beliefs when they occur in the cultures and civilisations we study. Until quite recently, the florid ‘witchcraft crazes’ of earlier centuries in Europe have generally perplexed and embarrassed historians, provoking extreme explanations for such ostensibly bizarre beliefs. For Professor Hugh Trevor-Roper, writing in 1967, this involved an appeal to the atmospherics of the precarious Alpine settings in which some of our ancestors lived. Their curious witchcraft fantasies, he suggested, might be traced to the swirling mists and rarified air of stormy mountain peaks. Social anthropologists are by no means averse to the attractions of theories based on environmental determinism. But since the publication, over forty years ago, of Evans-Pritchard’s Witchcraft, Oracles and Magic among the Azande, they have been in the fortunate position of possessing a key to the understanding of these beliefs which makes such reductionist theories redundant. Indeed, since this book was published, the problem has become not so much, ‘Why do some people believe in witchcraft?’ but rather: ‘Why doesn’t everybody?’

When Evans-Pritchard lived amongst and explored the mystical ideas of this Central African people, he discovered that their witchcraft beliefs constituted a subtle theory of misfortune which made excellent sense in the local cultural setting. How things went wrong was explained and understood in common-sense, pragmatic terms. Witchcraft was invoked as a supplementary factor to explain why one person, rather than others equally at risk, turned out to be the victim. Anyone’s crops might fail, anyone’s family might fall ill: why, therefore, did not such misfortunes affect everybody equally? Why, moreover, were some people more successful than others when they manifestly didn’t work any harder? This recalcitrant ‘Why me?’ question of the particularity and selectivity of ill and good fortune was what Zande set out to solve in terms of witchcraft, or, in the case of their ruling élite, in terms of the superior power of magical spells and potions. The degree to which witchcraft (or sorcery) was considered to be implicated in any misfortune depended also, of course, on how directly one was involved. The victim and his immediate circle emphasised this malign mystical presence, which, for those less directly affected, was hardly relevant. Bystanders thus offered straightforward explanations for what the victim plaintively ascribed to witchcraft.

The witches responsible for these afflictions (including death, which, like St Augustine, the Zande viewed as unnatural) were detected by oracles, diviners and ‘witch-doctors’, whose diagnostic and therapeutic powers as dispensers of anti-witchcraft magic Zande ascribed to their intimate knowledge of witchcraft. The ‘witch’ so discovered was typically a personal rival and enemy, witchcraft being synonymous with jealousy, envy, spite and hatred. The person whom the bewitched accused was equally likely to accuse him in turn, the danger here being qui accuse s’accuse. Zande witchcraft was thus not only a philosophy of misfortune (supplementing rather than supplanting empirical causation), but also a theory of interpersonal relations. For the bewitched victim, the accused witch was a conscious agent, deliberately seeking the downfall of his opponent. From the viewpoint of the accused, however, witchcraft was an unconscious force which, contrary to one’s best intentions, might inadvertently cause trouble between friends. In the case of death, the bereaved practised lethal vengeance-magic against the suspected witch, who, in Evans-Pritchard’s colonial days among the Zande, could not be openly accused. The colonial authorities regarded witchcraft as uncivilised superstition and, treating accusations of witchcraft as inhuman witch-hunts, outlawed them. The effect of this well-intentioned intervention was to increase the number of undetected witches, making witchcraft more rather than less prevalent. From the Zande point of view, their European administrators appeared to be protecting witches – almost, indeed, treating them as an endangered species! This suggested complicity between Europeans and witches.

These insights provide the point of departure for all subsequent anthropological studies of witchcraft and have been assimilated in the more recent works by historians like Keith Thomas and Alan MacFarlane. No apology is needed for their rehearsal here since, despite (or perhaps because of) their classic status, they are regularly misrepresented even by anthropologists who should know better. Jeanne Favret-Saada’s foray into contemporary witchcraft in the Bocage in France is a case in point. The author, an anthropologist at the University of Nanterre in Paris, has been working in the area since 1969 and spent over thirty months ‘in the field’. Coming originally from Tunisia, Madame Favret-Saada possesses impeccable credentials for approaching the French natives with precisely that mixture of detachment and attachment which constitutes the classic anthropological stance in studying ‘other cultures’. Her account of rustic French beliefs in witchcraft, in which she became ‘caught’, starts by tilting at a few local windmills in the best anthropological tradition. Folklorists, psychiatrists, and sensationalist journalists who rashly dubbed her the ‘Witch of the CNRS’ (this being the Centre Nationale de la Recherche Scientifique), are all soundly castigated for their patronising and ill-informed attitudes towards the Bocage villagers, whose beliefs in witchcraft they treat as primitive superstition.

No anthropologist following in the footsteps of Evans-Pritchard would want to disassociate himself from this spirited defence of people who believe in witchcraft. They will, however, be surprised to read that the Zande ‘in all circumstances only has the choice between witchcraft and sorcery’, and that the countryman who ‘knows perfectly well that there are explanations of another kind’ is placed in a different situation from the Zande. As I have indicated above, Evans-Pritchard went out of his way to demonstrate that, for the Zande, witchcraft and sorcery were options within a gamut of explanatory beliefs in which empirical causality played a major role. Moreover, the treatment of witchcraft as unacceptable superstition by the Colonial administration made the situation more, rather than less, similar to that in the Bocage. Favret-Saada’s point here is that, as part of a wider scientific society, the people of the Bocage are embarrassed to confess their faith in superstitious witchcraft, which only serves to confirm Parisian stereotypes of their primitiveness. Hence witchcraft in the Bocage is shrouded in secrecy. Thus, as soon as she started her inquiries, Favret-Saada found that those she spoke to (cumbersomely referred to as ‘interlocutors’) assumed she had some special interest in the topic: that she was either bewitched, or an ‘unwitcher’, or witch. Her informants would not allow her a ‘neutral’ position: they assumed that she either possessed knowledge (of bewitching and unwitching – for the two come to the same thing in the end) or needed help.

It seems, therefore, that, however irritating, the journalistic appellation ‘Witch of the CNRS’ is not so wide of the mark. In fact, Madame Favret-Saada makes much of what, rather naively, she sees as unusual and difficult in her situation as an ethnographer forced to adopt a local role in order to elicit information. She thus inadvertently carries a torch for the participant part of the ‘participant-observation’ shibboleth of modern social anthropological fieldwork. Although we do not have independent evidence on how the Zande regarded Evans-Pritchard, it is, of course, quite common for European anthropologists studying witchcraft in Africa and elsewhere to find themselves suspected of being witches. Indeed, in many parts of colonial Africa Europeans generally were regarded – partly because of their seemingly effortless superiority – as witches and cannibals (closely related inverted activities).

Blissfully ignorant of this, having discovered that she was ‘caught’ in the Bocage web of occult intrigue, Madam Favret-Saada decided to become an apprentice ‘unwitcher’ by joining the séance circle of a local witch-doctor, By a process familiar in what anthropologists elsewhere call ‘cults of affliction’ (where illness is the entry ticket to therapy that becomes religion), the bewitched here are primarily treated by recruitment into the consoling ranks of the ‘unwitchers’, whose main task is to fight and overcome witches. With hesitation and trepidation, Madame Favret-Saada thus makes her début as a kind of Carlos Casteneda of the Bocage. To a greater extent than Castaneda, while taking ‘magic force seriously’ and not just as ‘someone else’s belief’, she has at least managed to present an ostensibly detached account of her findings and experiences. On the other hand, Castaneda writes with beguiling clarity and vividness, and lacks the verbose, jargon-ridden pretentiousness which makes Madame Favret-Saada’s book such heavy going. Stylistically, it reads as though she is suffering from excessive exposure to pseudo-semiotica.

It is presented as a serious contribution to that ‘anthropologist’s paradise’ (as the late Maurice Freedman once called it), the study of witchcraft. As such, what has it got to tell us? The answer, I fear, is not a great deal – and certainly not worth translating into English. Her book is mainly the record of a number of accounts, given by different villagers (‘interlocutors’) at different times and in different contexts, of misfortunes and afflictions affecting people and animals which are ultimately ascribed, by some, to witchcraft. In one instance, the victim is dying of cancer. In another, he is afflicted with what the author puzzlingly sometimes describes as ‘impotence’ and sometimes as premature ejaculation. Other cases in this farming community predictably concern the well-being of livestock and land. There is no systematic presentation of case material, and no attempt to indicate how representative are the incidents she long-windedly discusses. However, she does not hesitate to propose a general scenario, and it is a tantalisingly interesting one.

‘Most witchcraft stories ... originate at the particularly dangerous moment in which, in the space of a few moments, a son buries his father, takes over the tenancy in his own name, gets into debt for a quarter of a century with the credit banks, and takes a wife to help him in his tasks.’ The implication, which is not systematically pursued, is that such dangerous transitions to power and property provide the seeds for subsequent envy and conflict in the retrospective perspective of witchcraft accusation. In fact, the author maintains, ‘witchcraft is always a matter of dual relationship between two families only.’ Like the Zande, the immediate circle of the bewitched victim takes witchcraft much more seriously than others not immediately affected, who tend to scoff at the charge. Unlike the Zande, the official ideology (of wider French culture) discounts witchcraft, and so those who believe themselves bewitched have to be more circumspect in their remarks and accusations than the Zande. In the Bocage, villagers laugh when a suspected witch falls ill, and at a known (or suspected) witch’s funeral. Ultimately, the witch and his or her victim are locked in mortal combat in a situation where ‘there is no room for two.’ A pervasive theme throughout the book is thus, inevitably, envy and its expression in covetous looks (the evil eye to which the author’s Tunisian background might have been expected to alert her earlier), as well as in their verbal expression in the ‘deadly words’ of the title. Unfortunately, Madame Favret-Saada relegates discussion of these crucial themes of Bocage witchcraft to the closing pages of her book.

The final chapter, pompously styled ‘Mid-Way Speculations’, purports to reconstruct the ‘conceptual scheme underlying the representation the bewitched use to describe what they are caught in’. Here we are finally confronted with the proposition that there is ambivalence not only between the categories witch and unwitcher (= witch-doctor) but also between the categories of bewitched and witch. Most of those who complain of being bewitched, the author concludes, have usually themselves been once accused of witchcraft. This leads the author to recognise ‘the system’s fundamental contradiction’: that those who have themselves been wrongly identified as witches nevertheless believe that other people have bewitched them. She appears to argue that the secrecy which surrounds witchcraft in the Bocage, in contrast to the Zande, and the vague and limited character of the accusations (although her data seem conflicting here), prevent people realising that ‘ “the witch” is only a logical prop, in a system enabling the repetition of biological misfortunes to be understood and dealt with.’ This contradiction between the perspectives of accuser and accused is, of course, precisely what, as we have seen, leads the Zande to their composite view of witchcraft as both conscious and unconscious, deliberate and accidental. Again, Madame Favret-Saada has evidently not studied Evans-Pritchard’s text with much attention.

Although she invites the comparison, it would be unfair to criticise her for not providing for the Bocage what Evans-Pritchard did for the Zande. One can, however, legitimately object to her manifest ignorance of crucial points in the Zande study and of subsequent developments in the anthropology of witchcraft relevant to her data. The distillation of her attempt to seize the essence of Bocage ideas on witchcraft is presented as a system where ‘life is thought of as a full sack that may empty, or as an enclosed field that may open; death is thought of as the end result of suction by a vacuum, is the active principle which alone causes the force to circulate.’ Her Bocage hosts are given a patronising pat on the head for this display of ‘philosophical talent’.

If Madame Favret-Saada were more familiar with anthropological writing on witchcraft, she would have discovered that over fifteen years ago the American anthropologist George Foster coined the useful expression ‘limited good’ for just such a situation. Here a fixed store of goods, with no opportunity for expanding resources, with in Favret-Saada’s terms ‘no room for two’, creates competitive conditions conducive to witchcraft beliefs of the Bocage type. The concept of witchcraft as ‘negative charisma’, which has also been around for a while, might have enabled the author to develop a more incisive discussion of the power struggle between witches (with negative charisma), bewitched and unwitchers (with positive charisma) than she achieves in her laboured concluding pages and cabalistic diagrams. Even without this knowledge of the anthropological literature on witchcraft, she could have found apt parallels for Bocage concepts of positive and negative magic ‘force’ in Mary Baker Eddy’s Christian Science, or, for that matter, in the magically charged ‘points’ of Haitian voodoo. Again, greater familiarity with the literature on transcultural psychiatry would have enabled her to set her criticisms of ethnocentric psychiatry in a wider perspective. She would, presumably, have relished the irony of the warning issued by a leading American authority, Ari Kiev, concerning the introduction of psychodynamic psychiatry in cultures where people believe in witches. This, Kiev feared, might simply confirm native beliefs (which, being based on mutual animosity, are essentially psychodynamic) in witchcraft. With such concepts as psychosomatic illness, and our question-begging ‘accident-proneness’, Western science makes some partial attempt to respond to the daunting ‘Why me?’ question so freely posed by the Zande. Other than that, we simply have to acknowledge that life is unfair, and abandon the implicit optimism of the Zande and the people of the Bocage.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 3 No. 15 · 20 August 1981

SIR: I.M. Lewis misses the point about Mme Favret-Saada’s Witchcraft in the Bocage (LRB, 18 June). The book is not about ‘objective’ knowledge of what may be witchcraft. The stake is truth – against scientific objectivity. Mme Favret-Saada discards all the usual advantages of the ethnographer: the ‘savages’ are not her ‘objects’. She does not ‘observe’: she enters their world, not to spy as a stranger, but bona fide. There is no methodological barrier between her and them. She does not describe them: she recounts an adventure with other people, who are very strange but equal.

The methodological barrier is between her and her colleagues. Between her and I.M. Lewis. His reaction to the book proves it.

Octave Mannoni
Paris XVI

Vol. 3 No. 19 · 15 October 1981

SIR: Octave Mannoni (Letters, 20 August) appears to misunderstand my review of Jeanne Favret-Saada’s book. Deadly Words: Witchcraft in the Bocage. I had hoped I had made it clear that Mme Favret-Saada was indeed writing very much in the spirit of a sorcerer’s apprentice in the Bocage. Hence, of course, my comparison with Carlos Castaneda’s work. Unlike the latter, however, Mme Favret-Saada also makes frequent reference to the standard anthropological literature on witchcraft, and writes, so the blurb informs us, as a professional anthropologist teaching the subject at Nanterre. Much of my criticism was thus directed at the author’s unsatisfactory handling of this classic material, and at her apparent unfamiliarity with relevant research and concepts, which makes her Bocage data nothing like as ‘strange’ or exotic as she and Mannoni claim.

I am also a little puzzled by the claim to novelty made for an anthropological approach which treats the people whose culture the anthropologist studies as ‘strange but equal’. This seems old hat to me. The classic French ethnographic tradition, associated with the name of its great champion, Marcel Griaule, insisted on the anthropologist being ‘initiated’ (sometimes literally and physically) into the culture he or she was studying. Of course, the extent to which anthropologists of this or other schools actually succeed in ‘going native’ always remains an intriguing question. Mme Favret-Saada’s pretentious – and condescending – account of her experiences (for Mannoni an ‘adventure’) makes it fairly clear what she feels about the people of the Bocage. I would not wish to question the value of this as a sort of anthropologist’s personal testimony. But it would perhaps be of even greater interest to know what the people of the Bocage made of their anthropologist.

I.M. Lewis
London School of Economics

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences