For Emile Durkheim, to explain our social institutions through individual psychology was generally regarded as illegitimate. It has, however, often been assumed by psychologists and psychiatrists, particularly in the case of religious doctrine and institutions. However, the people actually concerned, our religious informants, might sometimes themselves volunteer interpretations which are psychologically functional for the origin and maintenance of certain cultural facts. This is particularly so when they are faced with a 'modern' worldview. The instance of the Druze belief in bodily reincarnation after death is considered in the context of the recent civil war in the Lebanon.

The relations between the social and psychological sciences have never resulted in an agreed use of individual (psychological) interpretation in explaining or understanding social institutions, including standardized beliefs. Generally, social anthropologists have not followed Weber in employing interpretation of individual motivations and emotions as a possible social or historical datum. Emile Durkheim famously declared that social facts are independent of individual psychology: "there is between psychology and sociology the same break in continuity as there is between biology and the physical and chemical sciences. Consequently, every time a social phenomenon is directly explained by a psychological phenomena, we may rest assured that the explanation is false" (Durkheim, 1982; p. 129). In particular he emphasized that no widespread social phenomenon such as 'religion' or 'culture' (or, in his words, 'collective representations') could be attributed to the psychology of the individual, notably in his well known study of suicide, whose methodology still influences social science (Durkheim, 1952). Social institutions>, he suggests, consist of "certain ways of acting ... imposed or at least suggested from outside the individual and ... added on to his own nature" (Durkheim, 1982; p. 248, emphasis in the original). One must look for the social in the social.

Despite his later use of what are essentially sociophysiological hypotheses (Durkheim, 1915), he did not alter his opinion, and his methodological emphasis continues as fairly normative in the social sciences, despite a recent interest in questions of individual agency and subjectivity, hermeneutics and pragmatic relativism in relation to the 'facts' of social life (Littlewood, 1995; Obeyesekere, 1981). Outside the minor disciplines of psychohistory and psychobiography-both heavily influenced by Freud, who was apparently not familiar with Durkheim-it is generally considered unacceptable to relate a shared culture back to individual cognitions, affect and experience (Littlewood, 1993). This remains generally true of medical anthropology and medical sociology where, although obviously local health related practices may be expressly set up to serve certain recognized psychological or medical needs, more widespread social institutions such as religious or marriage systems are not directly related to particular individual psychological experience.

If social scientists have sometimes assumed that for times of rapid social change, they might justifiably have resort to psychological explanations more than for more settled periods (ibid.), there remained a more general problem of the role of psychology in seeking to understand culture. Durkheim's pupil, Mauss, followed him in 1934 to describe "psychological facts as connecting cogs and not as causes, except in moments of creation or reform" (Mauss, 1979; p. 121). Whilst the social anthropologist Malinowski (1944) argued that any culture must satisfy certain "basic [physiological needs", he gave a minor place to the possibility of psychological needs. Turner (1967), by contrast, proposed that the components of any successful culture must have both a 'public' (Durkheimian, social) utility at the same time as they are psychologically meaningful and satisfying to the individual: indeed, for the psychologist it is difficult to see how they could not. Social anthropologists and sociologists, however, have always been vague about individual motivation and the communication of cultural knowledge, simplistically assuming that something called 'socialization' makes all individuals in a society think and behave in the same way (Boyer, 1994). A more recent solution argues that individual agency and experience just seem to act in a different 'direction' to social structuring (Bourdieu, 1977) and that we continually have to place one against the other in particular situations (Littlewood, 1993, 1998).

By contrast, the idea that institutions, including religion, exist for a specific collective purpose-solidarity or cohesion-is today regarded as frankly banal. That social institutions subserve particular social needs is regarded as self-evident-and thus not requiring particular comment. But when the reasons given by the people themselves are couched in such socially functional terms they are generally regarded as irrelevant and as self-serving.

Despite Durkheim's strictures, it is commonplace for cultural psychiatrists and psychologists to maintain a psychological function for religious doctrine: i.e. that certain `collective representations' exist essentially to subserve individual psychological needs (e.g. Kiev, 1972; Murphy, 1982; Tseng & McDermott, 1981). Certain shared religious practices-confession, shamanism, possession, glossolalia, and certain types of religious 'consolation'-are seen as 'intended' to satisfy a person's distress or existential doubt. Indeed religious practice in general is seen as 'therapeutic', whilst allowing that certain religious practices may be taken as 'pathological'. This has occurred within a context of widespread interest in the mental health advantages of religious practice: advantages intended or otherwise. The instance considered here is a set of explanations about reincarnation after death among the followers of the Druze religion in the Lebanon which might seem to have an obvious psychological value, a point which is maintained by the people themselves.

The Druze

The Druze are a heterodox movement in Ishmaeli ('Sevener') Shia Islam, originating during the Fatimid caliphate in Cairo (Abu-Izzeddin, 1984; Betts, 1988; Firro, 1992). The capricious and antinomian Caliph al-Hakim bi amr-Allah, said by some followers to be the Mahdi or even God, disappeared suddenly in the year 1021 C.E. after which his followers entered a period of proselytization. In a period of persecution by normative Islam, the sect's major religious books were written and the faithful departed north for the relative isolation of, first Syria, and then the Shuf mountains of what is now South Lebanon, coming under Ottoman rule in the sixteenth century and a French mandate in the twentieth (Salibi, 1988). Both polygamy and slavery had been abolished by the Druze in the eleventh century, but their universalist and millennial aspirations seemed to have declined by the time of European intervention in the Lebanon and are currently not prominent (Abu-Izzeddin, 1984).

The call to convert (da'wa) into the group ended in 1042, and membership since then has been hereditary: no new members are accepted or can marry in. The Druze now number about 900 000 Arabic speakers in the Lebanon, Syria, Jordan and Israel, with a considerable diaspora in Europe, South America and the USA. Under the Lebanese National Pact of the 1940s, they are recognized as a permanent political and religious constituency in the country with guaranteed governmental representation: an agreement which has survived the recent civil war and Syrian intervention.

The community divide themselves into the uninitiated majority (juhhal) and the enlightened (uqqal or sheikhs) who have access to the six holy books: the latter, both men and women, comprise about 10% and have somewhat stricter obligations (no alcohol, no tobacco, strictly no adultery) and a traditional costume of white fez, dark gown and knee-- breeches for men, and they meet regularly on Thursday evenings in the khalwa to discuss the religious texts and matters of welfare and social interest to the community. Initiation into sheikh-hood typically takes place in one's 50s but may be earlier among the more earnest members. Druze religious beliefs have traditionally been regarded as esoteric if not secret, but in conversations with both juhhal and uqqal I found no evidence of evasive answers : Druze are nominally monotheistic (tawhid) although not Islamic, with a long history of protective dissimulation (taqiyya), and have collected together a set of ideas which can loosely be termed neo-Platonic (and which in the past have led them to being confused with Masonic and other Hermetic Christian philosophies (e.g. De Nerval, 1984)), without a clear notion of a judgment to come or of predestination, no practice of ritual circumcision nor adherence to the five fundamental tenets of Islam and, in their newer forms, sympathy for what might be called 'New Age' beliefs common in the contemporary West2. Druze now have monogampous extended households which still place emphasis on female sexual honour (ird) but with considerably more female autonomy than in most Muslim societies, with preferential parallel cousin marriage, and their social organization is very similar to that of rural Maronite Christians, Sunni and Shia Muslims in the Lebanon (Salibi, 1988): local residence in kin-related villages with loyalty to a feudal chief (bey) (see Gilenan, 1996, for a Sunni instance).

of referring everyday even and peasants to religious dogma, all Druze agree that accepted reincarnation. This idea seems to have been present from their formation is crucial to Druze eleventh century (Makarem, 1974). All Druze I spoke moment of death-doctors, businessmen and peasants-accepted reincarnation. This a new-born baby: "You brea seems to have your last, then present second breath is their first breath on in the eleventh century (Makarem, 1974). All souls were created at the beginning of death time, and the is automatically reincarnated as a new-born baby: "You breathe body at bir last, then the body being compared to a shirt second breath you discard. Unlike Hindu or Buddhis the first breath of subsequent lives are not seen aby". All souls were created at the beginning of time, and the individual soul enters the body at on action in past lives orth, the body being compared to ar direction (cf. Daie, Wit which you discard. Unlike Hindu or Buddhist ideas, & Rabinowitz, 1992). The subsequent lives by experience quite vare not seen as so dependent on action in past lives a poor person, as healthy and as sick. The main purpose seems to be for direction (cf. Daie, Wit to Mark, & Rabinowitz, 1992). The soul range of human experience; where quite variously all ends is now none too certain: most person full unity with the deity. Male Druze reincarnate and as a poor person, as male Druze, female as sick. The main purpose seems to be for it to gain the full range of human experience; where it all ends is not reincarnate as non-Druze (but some say too cey might appear in: most say in full unity with the deity. Male Druze reincarnate as may be present in those who are apparently nonGenerally Druze do not reincarnate is according to local ideas non-Druze (but some say they might appear in India and China, Christians), although Druze souls members may be present in those who are of being Druze; the soul cantly not be split and the Druze 3. Thers problem' (an increasing population needs more souls) is according to local ideas a Druze community in India and China, although its members may be unaware of be souls (and also non-observant Druze in the soul cannot be split and the ,numbers problem' (an increasing population needs more souls) are similarly reincarnated among non-Druze, say some, generally also in unknown number of 'surplus' Druze souls (and also non-obser-specific want Druze teaching on reincarnation is echoed among the diaspora). Non-Dria, although these souls are closer to orthodox Islam. Western advocates of reincarnated among non-Druze, say some, generally also in a gender-specific way. Druze teaching on reincarnation is echoed various instances ing the Lebanon (Stevenson, 1980), and Daie et al. (1992) Syria, a clinical instance among these are closer to orthodox Islam. Western advocates of reincarnation

Memories of a past life are usually triggered by some partial reminder. Fous instances in the Lebanon (Stevenson, 1980), and Daie et al. (1992) Naseem saw his father clinical instance among throat and, bursting in tears, recalled that he himself had been a soldier who had attacked a Christian village with 16 others, ally triggered by some partial reminder. Four-year-old Naseem saw his father cut a sheep's throat and, bursting into tears, recalled that he himself had been a soldier who had attacked a Christian village with 16 others, all of whom were captured and had their throats cut 3 days later. (In this case he was later found to have been born 3 months before the actual raid and was not able to recognize anybody in his previous life village, so he is not generally regarded as being that particular reincarnation.) Hareem's father Faris (Case 3 below) tells of how he had gone to a local well with his own father as a boy, and then recalled how in a previous life as his father's boss he had drowned in the well: his last memory was lying in the water with a rescue ladder on top of him. Faris' father had then taken him to the presumptive past-life family where he recognized photographs of his past-life wife and her relatives. He was recognized by his past-life daughter who still visits him, but not by his past-life uncle who is the head of the household (apparently because he then tried to help the daughter claim some property in Beirut).

Knowledge of past lives is normally limited to the one immediately before the current life. Generally, human souls have no such knowledge of a past life unless they die violently: the reason given being that the 'person' is not easily reconciled to a sudden death and is taken protesting from their previous body. This memory possessed by the soul is especially significant in the surge of reincarnation beliefs in the aftermath of the Lebanese civil war of the 1980s, of which I shall give some illustrations. Most Druze, like other Lebanese, are currently engaged in reconstruction after the war: few families have been left untouched by the war killings, massacres of civilians and other atrocities of that period (see Fisk, 1992, for a detailed account).

Reincarnation narratives

Case 1

Alaa S. is now 12 years old, the son of a plumber. At the age of 2 years, he demonstrated a familiarity with both car engines and guns, including habitually lying under a favourite car exhaust, which his family could not readily explain. As he learned to talk, he used phrases usually only spoken by old people and sheikhs. At 6 years old he led his father to a shop some miles away where in a previous life he used to run a car-exhaust business. People working there told me he knows the details of the business and can describe workpeople and incidents in the workplace which he could not have known about by general conversation. He asked about holes in the garage floor which were filled in 7 years before. His death in an earlier life is recalled by them: he was a sheikh, one of six Druze militiamen in a car who were taking another soldier to hospital when a rocket hit the car and all were killed. In his current life he does indeed seem very familiar with car engines and trade details of replaceable parts such as tyres and exhausts for which he does not seem to have had any special training: "I look at a car and what's in it; it's knowledge that came with me. Nobody taught me". The boy does not seem to have gained financially by being identified as a reincarnation of the dead man, but his family regularly visits his past life relatives, and a friendship has grown up between them.

Case 2

Abeer S. is now 21 years old, the daughter of a life insurance salesman. At about the age of 4 or 5 years, her parents became concerned about her because she spent a large part of the day by herself moping and crying. She had one older brother who was handicapped, whom she had to look after and who died when she was 18: there are no other siblings. A couple of years later, her grandfather saw her crying again and asked her why. She told him she missed her previous-life family and he insisted that her parents located them. She went with her father in his car, and turned up at a village some 20 miles away (familiar to him from his work, apparently not to her in this life) and she recognized the house of a previous-life aunt and the neighbourhood. Her past-life father told me she had not hesitated in the recognition; and he himself had known at once that she was the reincarnation of his daughter who had died some years previously at the age of 14 after a mortar shell hit the car in which she was travelling (when shopping with her cousin who had also died). He told me that since his daughter's death he had kept an eye open for her reincarnation, for he missed her. When Abeer appeared at the age of 10 he knew immediately that she was the reincarnation of his daughter. His wife backed him up but other daughters of his had agreed somewhat more slowly and (reluctantly?) that she was their sister. Relations between the two families are now very close, involving mutual exchanges of gifts and regular visiting, although Abeer's pre-reincarnation family is much wealthier and is considering putting her through college which she could not otherwise afford. Her present-life parents express no doubts about the relationship, pointing out that her 'new' parents have an equal share in her. Relations between the two families appear very cordial. Abeer was only born 2 months after the death of her previous life incarnation, but Druze allow for a short 'detour' between lives. I took her to the site of her earlier death where she recalled standing by the side of the car when the shell hit. She met again the men who had retrieved her body, one of whom she said she had known in her earlier life.

Case 3

Haneen A. is a 10-year-old schoolchild in Eghmid who recalls a past life as a school teacher and mother of five, Safa, who was killed in her back garden in Bchamoun (10 miles away) by a shell from the US battleship New Jersey . In 1984 it was shelling the Shuf mountain villages at a point in the civil war when the Druze were opposed to the Maronite Christians who then had American support (Fisk, 1992). Her elderly husband, Ajaj, was in the garden putting out a fire caused by the shelling and as she went to call him to come in out of danger, she was struck by a shell splinter in her neck and was killed. She recalls her body being licked after death by the household dog, of which she is still frightened because he had 'eaten' her. She then remembers being carried away by some soldiers but nothing more. Haneen now recalls the names of her two pre-reincarnation sons, Riyadh and Nedr, and her three daughters Shahira, Adla and Nebila, after whom she had spontaneously named her dolls, and details of the house. Her present-life parents report that at the age of 2 1/2 years she had talked of having a son Riyadh and of living in a village, Bchamoun, about 10 miles away. Haneen's `this-life' father, Faris, had also been reincarnated and told her stories about it. She was identified by the now grown-up daughter from her past life, Shahira, who had visited her after hearing rumours of a reincarnation, and recognized her because of her immediate familiarity with the names of the family. Haneen was immediately taken back to visit her past-life family where they say she successfully found her way around the home and identified objects and people, although being frightened of the garden where the shell had landed. Her past-life son, Riyadh, immediately recognized her as his mother and hugged her. Relations between the two families are cordial, although her current-life father is a bus driver whilst her past-life son is a wealthy currency dealer. She expresses sadness at the recent illness of her past-life husband, Ajaj, who is now an old man, attends like a wife to his plate at mealtimes and calls him by his first name, and as a young girl once slapped the face of her adult 'grandson' for talking rudely to his 'grandmother'. When Adla, one of her past-life daughters, appeared wearing her own mother's (Safa's) clothes, Haneen asked if they were her (Adla's) own, and Adla said 'yes': Haneen then called her a liar. Riyadh told me that Haneen talks like an adult and to outsiders appears a strange mixture of adult and child. She appears very confident in her role as an elderly woman, taking on the gestures and mannerisms of an adult. Riyadh confirms she has the same personality as the dead woman, proud and moral, and that she worries about Riyadh and Shahira as if they were her children. Her present-life father, Faris, says she is now more settled having discovered her original family.

Bereavement: the local view and psychological interpretation

Less than 10% of Druze seem to have firmly accepted memories of a past life, although all assume that they have been reincarnated 5. They reluctantly admit that the rediscovered past-life family is often wealthier than the present-life family but say that this is no motive for asserting a particular reincarnation. Recognition as a reincarnated person confers no accepted special status (inheritance, etc.) and appears to operate socially through affective bonds and gifts alone. In many instances, the bereaved family had searched for the reincarnated soul of the dead person and met the claimant to reincarnation 'half-way', the criteria for authenticity in such cases being fairly low, generally being an apparent ability to recognize relatives. Evidence for and against the reincarnation of a lost relative is debated vigorously and I came across various 'failed claimants' (who are regarded as innocently mistaken) as Naseem above. Hussan, a 10year-old boy from a broken home, remembers that in a past life he was a soldier called Raja who was shot in the ear and died. But Raja (who lived in a neighbouring village) was actually still alive and it was his brother Murfa who died. A friend of Murfa also died at the same time and had already been accepted as having been reincarnated in Hussar's village. Hussan is now recognized as having been mistaken but nobody attributes mercenary motives to him (Raja's family again were relatively wealthy compared with Hussar's, whose father had no job).

What actually constitutes a 'soul' and its continuation across different lives is uncertain. Some argue for common features of personality (such as courage, nurturance or generosity), particularly in socially acknowledged instances, but many say the soul has no recognizable human characteristics at all. Most reincarnations are instantaneous but the Druze were skilled at avoiding questions about the correlation of battle deaths and new pregnancies (which would be necessary to make available new human bodies).

In relation to the personalities of the children who claim (or are recognized) to be specific reincarnations, there appear to be no abnormal features, except for a slight sense of dissatisfaction; but we cannot exclude this from the normative expectations of everybody being reincarnated. This would agree with Daie et al. (1992). It may be significant that the period of volunteering reincarnation narratives-typically between 3 and 6 years of age-corresponds to the 'family romance' period identified by psychoanalysts when the child is apparently prone to fantasize that they are really the children of more glamorous or famous parents. In one case reported by Stedman (1998), a 3-year-old Druze village boy claimed he was the reincarnation of a well-known footballer and singer, whose own family accepted the relationship. It is not too difficult to realize how general expectations, reinforced by grieving relatives, confirm the child in their identification. In some cases, such as Haneen (Case 3), the identification is primarily made by the bereaved relative. As Stevenson (1980, p. 13) comments: "Druze parents nearly always encourage their children to tell them whatever they can remember of their previous lives".

When asked directly about any personal value for the belief, Druze unanimously agree it has had a validity in having bound together as a single community people who were dispersed over four different countries with diverse public policies of their governments. Druze in Israel have been at war with those in Syria (Israeli Druze join the Israeli armed forces).6 To acknowledge some function for customary ideas in no way questions the belief itself and I met no Druze who doubted the likelihood of reincarnation. Many mentioned that the idea usefully led to fearlessness in battle and quoted a Druze battle cry "Tonight my mother's womb" (I'll be killed in battle today, but I shall be instantly reincarnated), which may go some way to explain the ferociousness of the Druze in war (Fisk, 1992).

The most evident locally stated utility of the belief in reincarnation (given that bereaved relatives tend to seek out the soul of the one who has died) appears as part of a process of bereavement. Druze spoke of reincarnation beliefs as a support for the bereaved person. Though few of the personal characteristics of the dead person are believed to continue in their soul, enough survives to mitigate the loss of a person's identity. The social recognition of what constitutes death differs widely (Eisenbruch, 1984b; Hertz, 1960) whilst the actual type of mourning in Islamic societies may change fairly fast depending on political and cultural changes (Good & Good, 1988). A problem for considering the therapeutic value of cultural systems for grieving is that we cannot start from any `natural baseline' uncontaminated with culture (Eisenbruch, 1984a, b). As Phillips (1999, p. 6) notes, "It is possible that we have no idea what secular grief is; what grief unsanctioned by an apparently coherent symbolic system would feel like" (see Wieseltier, 1998). Psychological theory, deriving from Euro-American experience, argues that first the reality of the loss has to be accepted, then the emotional pain is `worked through', with an eventual detachment from the deceased (Parkes, 1972; Pincus, 1976). Druze ideas would seem to contradict the first stage. The cross-cultural evidence however argues that denial of the actual death, if supported by a systematized set of values, may be adaptive for the individual. Al-Adawi, Burjorjee, and Al-Issa (1997) describe the mu ghayeb concept in (Arab) Oman in which an assumption that the dead is only ensorcered and will return alive usefully seems to deny the loss altogether, in a similar way to zombification ideas in Haiti (Littlewood & Douyon, 1997). Assumptions of reincarnation, though equally contrary to Islamic tradition, are not such a radical denial of the death itself. The individual has died in a real physical sense and yet the notion of death itself is attenuated compared with other societies in which death is the immediate end of the individual.

Mourning among the Druze is restricted to a short period and overt grieving discouraged as I witnessed at funerals (also see Firro, 1992; Stevenson, 1980). Thus the psychological substitution of the dead relative by the hypothetical reincarnee takes place at a fairly early stage in the grieving process. The obvious sort of psychological interpretation would be that having reincarnation beliefs to account for violent deaths in the civil war usefully diminishes socially wasteful mourning in the context of a central religious tenet which already provides solidarity and unity for a community under immediate threat.

That Druze give psychologically functional reasons of this sort for their reincarnation beliefs-that it serves to alleviate grieving-in no way alters the validity of the belief itself. But nor does it necessarily provide a model for us to decide the origin and individual articulation of the phenomenon: Durkheim (1982, p. 171) expressly ruled out agents' explanations for their own actions, a view with which 'structural' anthropologists and sociologists concur. We are enmeshed in a web of culture which we can never comprehend objectively. If we do follow the local informants, however, and argue for some sort of psychological purpose for reincarnation beliefs, where does this leave Durkheim's view that religious institutions cannot be explained by individual psychology? Various possibilities include:

(1) The Druze tradition of reincarnation has only become 'psychological' in response to a modern world view (individualization and psychologization) and, particularly, the immediate context of my investigation and its reception by the Druze.

(2) Satisfying psychological needs is not primarily related to an institution but it may later serve to maintain it by some process of socialt selection.

(3) Psychological need satisfaction is an occasional and insignificant by-product of social institutions.

(4) We are unduly influenced by the explicit local views as to what constitutes a psychological> explanation.

(5) Durkheim is mistaken when he maintains of a person that "beliefs and practices of his religious life [j if they existed before he did, it follows that they exist outside him" (Durkheim, 1982, pp. 50-51).

The evidence offered here does not easily help us answer this difficult epistemological question one way or another, but I would support (2) as the most promising. It does, however, call attention to a too-easy assumption that social institutions are to be explained through their satisfying individual psychological needs.

1 On a brief field-trip in 1997/98 to make a film on Druze religion. Interviews were carried out in homes and workplaces, at funerals and during religious meetings (khalwa) in the Shuf mountains south of Beirut. I owe a particular debt to Jad Al Younis for local Druze knowledge and for interpreting from the Arabic, and especially to Chris Ledger, without whose preparatory work this study would have been impossible.

2 I talked with the son of the paramount Druze religious figure (sheikh al-aql), Sheikh Bahjat, who publishes a religious magazine which combines Druze religion with an interest in pyramidology, flying saucers and Teutonic mythology. His father maintained to me that there is no esoteric Druze religious doctrine at all, for all religions aim at the same truth, yet one should not seek for a Creator outside oneself. This set of 'liberal' and near-pantheistic ideas is somewhat at variance with the past Druze tradition of not sharing their religious ideology with outsiders (De Nerval, 1984), and the fierce solidarity of the Druze community during the recent civil war in the Lebanon (Fisk, 1992). 1 was openly invited to attend the regular meetings of the uqqal with my interpreter (a Druze man and later a Sunni woman). It is likely that the Druze tradition of dissimulating their beliefs in the past was to avoid orthodox (Sunni and Shia) persecution: as they put it, "a Muslim outside and Druze 'at home (Dale et aL, 1992).

3 The possibility was often raised that I myself was a Druze 'inside' because of my interest in the community, and after I became ill Sheikh Bahjat's son wondered whether if I had died my soul would have become incarnated in a recognized Druze. It might be noteworthy that, in the nineteenth century, the Druze accorded a special religious status to the English (Inchbold, 1906, p. 83), perhaps a consequence of the 1830 Druze-British alliance against the French-backed Maronite Christians. There have been various speculative theories for Druze origin, giving them an English or Frankish (French) crusader ancestry.

4 My companion Roy Stenman has reported this instance at greater length (1998).

5 On going to a local school and asking an assembled class of 12-14-year-olds, about 10% agreed they had accepted memories of a previous reincarnation.

6 Another instance, Adham B., at the age of 3 years reported a past life as Ra'ad Ismail (the same name as a schoolfriend of his). This past life R'ad was apparently the pilot of a Syrian air force plane which crashed in the Barouk Mountains in 1984, together with Ra'ad's father as navigator. Plans to visit Syria with Adham were prevented by the refusal of a visa from the Syrian authorities.

Received 1 December 1999; revised version received 2 October 2000

References

Abu-Izzeddin, N. M. (1984). The Druzes: A new study of their history, faith and society. Leiden: E. J. Brill. At-Adawi, S., Burjorjee, R., & Al-Issa, I. (1997). Mu-Ghayeb: a culture-specific response to bereavement in

Oman. International Journal of Social Psychiatry, 43, 144-151. Betts, R. B. (1988). The Druze. New Haven: Yale University Press.

Bourdieu, P (1977). Outline ofa theory of practice (Orig. pub. 1972), Cambridge: Cambridge University Press. Boyer, P. (1994). The naturalness of religious ideas: A cognitive theory of religion. Berkeley: University of California Press.

Daie, N., Witztum, E., Mark, M., & Rabinowitz, S. (1992). The belief in the transmigration of souls: Psychotherapy of a Druze patient with severe anxiety reaction. British Journal of Medical Psychology, 65, 119-130.

De Nerval, G. (1984).Journey to the Orient (Orig. pub. 1853). London: Owen.

Durkheim, E. (1915). The elementary forms of the religious life (Orig. pub. 1912). London: Allen and Unwin. Durkheim, E. (1952). Suicide (Orig. pub. 1897). London: Routledge and Kegan Paul.

Durkheim, E. (1982). The rules of sociological method and selected texts on sociology and its method (Orig. pub. 1895). London: Macmillan.

Eisenbruch, M. (1984a). Cross-cultural aspects of bereavement: 1. A conceptual framework for comparative analysis. Culture, Medicine and Psychiatry, 8, 283-309.

Eisenbruch, M. (1984b). Cross-cultural aspects of bereavement: 2. Ethnic and cultural variations in the development of bereavement practices. Culture, Medicine and Psychiatry, 8, 315-347.

Firro, K. M. (1992). A history of the Druzes. Leiden: E. J. Brill.

Fisk, R. (1992). Pity the nation: Lebanon at war. Oxford: Oxford University Press.

Gilsenan, M. (1996). Lords of the Lebanese marches: Violence and narrative in an Arab society. London: Tauris. Good, M-J., & Good, B. (1988). Ritual, the state and the transformation of emotional discourse in Iranian society. Culture, Medicine and Psychiatry, 12, 43-63.

Hertz, R. (1960). Death and the right hand (orig. pub. 1907). London: Cohen and West. Inchbold, A. C. (1906). Under the Syrian sun. London: Hutchinson.

Kiev, A. (1972). Transcultural psychiatry. Harmondsworth: Penguin.

Littlewood, R. (1993). Pathology and identity. Cambridge: Cambridge University Press.

Littlewood, R. (1995). Psychopathology and personal agency: modernity, culture change and eating disorders in South Asian societies. British Journal of Medical Psychology, 68, 45-63.

Littlewood, R. (1998). Reason and necessity in the specification of the multiple self. Royal Anthropological Institute Occasional Paper No. 43. London: RAI.

Littlewood, R., & Douyon, C. (1997). Clinical findings in three cases of zombification. Lancet, 350, 1094-1096. Makarem, S. N. (1974). The Druze faith. New York: Caravan.

Malinowski, B. (1944). A scientific theory of culture. Chapel Hill: North Carolina University Press.

Mauss, M. (1979). Body techniques (Orig. pub. 1934). In M. Mauss, Sociology and psychology. London: Routledge and Kegan Paul.

Murphy, H. B. M. (1982). Comparative psychiatry. Berlin: Springer.

Obeyesekere, G. (1981). Medusa's hair: An essay on personal symbols and religious experience. Chicago: University of Chicago Press.

Parkes, C. M. (1972). Bereavement: Studies of grief in adult life. Harmondsworth: Penguin. Phillips, A. (1999). Commanded to mourn. London Review of Books, 18 February, 3-6. Pincus, L, (1976). Death and the family: The importance of mourning. London: Faber and Faber.

Salibi, K. (1988). A house of many mansions: the history of the Lebanon reconsidered. Berkeley: University of California Press.

Stenman, R. (1998). Lebanese research among the Druze. Reincarnation International, 15, 11-21. Stevenson, I. (1980). Cases of the reincarnation type: 3. Lebanon and Turkey. New York: Alloh. Tseng, W-S., & McDermott, J. F. (1981). Culture, mind and therapy. New York: Brunner/Mazel. Turner, V (1967). The forest of symbols: Aspects of Ndembu ritual. Ithaca: Cornell University Press. Wieseltier, L. (1998). Kaddish. New York: Knopf

Roland Littlewood*

Departments of Anthropology and Psychiatry, University College London, UK

*Requests for reprints should be addressed to Roland Littlewood, Departments of Anthropology and Psychiatry, University College London, Gower Street, London WClE 6BT, UK.