This article describes and analyzes some 40 case reports of contemporary spirit possession in Chuuk. The possession-trance complex of today shows continuity with distinct features of the now defunct status of the medium (waanaanu): trance, calling the spirits of deceased kin, and spirits descending and addressing the assembled kinfolk through the changed voice of the host. But recent spirit possession is not possession-trance on demand; it is spontaneous and sudden, arising from the felt stress of the individual. Family problems are the focus of today's cases. We view the possession-trance complex as culturally sanctioned communication of prohibited expression about family and kinship problems. As a response to stress, cases of female possession appear positively functional when compared with male drinking. Contemporary Chuuk possession-trance is, therefore, a remarkable adaptation of former ways applied to new situations, functioning as a kind of family therapy.

One cannot live for any time in Chuuk (formerly Truk), the most populous state of the Federated States of Micronesia, without hearing tales of spirit possession. These possession stories, in turn, often include descriptions of trancelike states. This article examines contemporary possession-trance as behavior that is rooted in a traditional institution of Chuuk society, but it is behavior that has been modified over time to serve new purposes. If the spirit world of Chuuk has changed over the past century, so too has the way in which the people of Chuuk interact with it. Against this background, we describe how the experience of possession-trance itself has been transformed into a mechanism that serves essential functions in Chuuk society today.

Most of the data for this analysis come from a 1989 collection of case reports made by Cathy Hung, then a Peace Corps Volunteer in Chuuk. Within 2 months she collected through interviews with Chuuk women accounts of 57 spirit encounters, more than 40 of which could be called spirit possession. Had it not been for limitations of time and access to informants, she could easily have collected twice this number of reports; we have collected over a dozen more since her work. It is apparent to us that, despite Cathy Hung's dedicated efforts, we have done no more than skim the surface of a deep pool of such cases in Chuuk. All but two or three of the possession case reports involved females. An undeniable bias toward women characterized the collection techniques because the investigator was herself a woman and so had greater access to female informants. On the other hand, the choice of a woman to do the interviewing was prompted by the strong association the people of Chuuk make between spirit possession and women. A deliberate effort on our part to elicit case reports of male possession may have yielded a few more instances of episodes among men, but the strong gender imbalance almost certainly would have remained.

The compiled case reports on possession-trance are admittedly shallow. Usually there was only one informant for each incident and he or she was asked to provide information on an event that in some instances occurred years earlier. Our own interviews confirmed that contemporary Chuuk inhabitants who are not close kin of the trancer tend to recite a skimpy pattern of detail. Consequently the description is often sparse and lacks the rich detail that would be more desirable. Where we have later built up enough interviews to generate a detailed case study, we are wary about the possibility of revealing the identity of the family. Nonetheless, what the case files lack in texture they compensate for in quantity. And it is precisely the quantity of cases that reveals the highly patterned behavior and local interpretations of possession-trance. These patterns in turn describe important functions in contemporary Chuuk society.

The additional reports we collected demonstrate the same patterns found in Cathy Hung's reports. Sometimes we interviewed the same trancers or onlookers in cases described by Cathy Hung. Unfortunately, we had the serendipitous opportunity to witness only one possession-trance episode. Basque anthropologist Beatriz Moral, however, who saw multiple episodes, confirmed the authenticity of the reports we received (personal communication, January 5, 1994, and July 20, 1994). We base our work, however, mostly on reports, not observations.

Chuuk, one of four states in the Federated States of Micronesia, is situated in the geographic center of Micronesia and includes a large lagoon containing a dozen inhabited high islands, with outlying atolls to the east and west. Contemporary Chuuk offers an interesting amalgam of traditional island folkways and foreign influences. It is populated by approximately 50,000 people and, to judge by the stories these people tell today, at least as many spirits. In 1992, 97 percent of the 47,871 people were Christian; at least half of the work force in any region were in the cash-wage economy; and most of the eligible youth were enrolled in school (OPS, 1992, pp. 70, 80-90). Most of the individuals described here are from the area of the main lagoon, although several families came originally from outlying groups, such as the Mortlocks.

The spirit beliefs that early writers recorded over a century ago have proved remarkably resistant to change, despite the conversion to Christianity, formal Western-style education, and a monetized economy (see Kubary, 1878-1879, p. 23). These spirits resist easy classification, as anthropologists have frequently noted (Alkire, 1989; Lessa, 1987, p. 498), but they are part of a cosmos that Chuuk inhabitants, like most other Micronesians, still see as populated by gods, spirits, and living humans (Bollig, 1927/ 1967, p. 39). The gods and the spirits (whether human or nonhuman) are known by the single term enu (Mahony, 1969, p. 133).

A word is in order on the quality of the older ethnographic sources, especially those of Jan Kubary (1878-1879), Max Girschner (1912-1913), Laurentius Bollig (1927/1967), and later, Frank Mahony (1969). Kubary, a Polish-born naturalist and longtime resident of Micronesia, had a keen eye for ethnographic detail and apparently a talent for learning languages. His descriptions of Chuuk possession-trance episodes come mostly from a 3-month stay in the Mortlocks in 1877. Max Girschner was a colonial government physician during the German period before World War 1. Like Kubary, he was headquartered in Pohnpei and conducted interviews with Chuuk inhabitants from the Mortlocks. His detailed ethnography, published in 1912-1913, describes possession-trance behavior that he apparently observed. Laurentius Bollig, on the other hand, was a German Capuchin missionary assigned first to Pohnpei in 1912, then reassigned to Chuuk from 1913 to 1919 (Hezel, 1991, pp. 115, 185). His ethnography, published in 1927, reveals the same linguistic interest and detail as found in Kubary and Girschner and demonstrates the lexical continuity between the possession-trance complex before World War I and the contemporary reports. The 1969 date for Frank J. Mahony's doctoral dissertation belies his long and deep involvement with Chuuk culture. Having worked as a Trust Territory official in the period just after World War II, he was able to draw on this background to write one of the few monographs (alas, an unpublished PhD dissertation) on Micronesian medical anthropology. His work bridges the testimonies of Kubary, Girschner, and Bollig and the contemporary reports, and it places the possession-trance complex within the Chuuk categories of spirits and medicine. We should add parenthetically the 1915 handwritten German manuscript "Some Remarks About the Religious Views of Our Islanders," by an anonymous Liebenzeller (.7) missionary. This work corroborates the possession-trance description given by writers from the same period. It is difficult to assess the reliability of these sources. All that can be said is that their ethnographic detail is considerable and they do not contradict one another. Only Mahony was a trained ethnographer.

THE CHUUK SPIRIT WORLD

Certain spirits were believed to inhabit the sky and they have long been cult figures. The trinity of spirits regarded as the oldest and most powerful were Anulap (Great Spirit), the uncreated spirit who lived remote from mortals and their affairs; Lukeilang (Middle of Heaven), son of Anulap and lord of all in the realm of spirits and mortals; and Olofat, the eldest son of Lukeilang, a trickster god and the subject of a cycle of popular tales. But these gods were the stuff of myth and had little to do with human affairs; only Inemes, goddess of love (Bollig, 1927/1967, pp. 39-40), still regularly intervenes in human affairs. Spirits of old and today, on the other hand, fall into the two broad categories of nature spirits and spirits of the dead. The nature spirits are those identified with places or events: mountain-dwelling spirits, woodland sprites, and ghosts of the sea. The origin of this group of spirits is lost in the mists of the past, although some Chuuk people knowledgeable in traditional lore maintain that these spirits were once human spirits. Some are still believed to inhabit a particular place, a mountaintop, a spring, or the sea–like the sea spirits (oos) that come to shore to sleep and attack unwary humans, or the lagoon spirits (chenukken). Other spirits seem to wander freely. Some were associated with life crises: For example Bollig (p. 12), writing about experiences from 1913, described kir, a nasty spirit that is still believed to dry up the breasts of nursing mothers. One of the best known and most feared of these spirits was Anumwaresi, the spirit of the rainbow, which is still thought to afflict pregnant women (Anonymous, 1915, p. 4). The innumerable nature spirits, wrote Bollig (p. 12), could be classified only in the most general way, but they share as a common trait their potential danger to people.

The focus of this study is on spirits of the dead (enu, or enu aramas; sometimes soope; Bollig, 1927/1967, pp. 21-22), particularly of ancestors, which intervene more directly through the communication of possession-trance. Formerly, inhabitants believed there were two spirits in each individual: the good spirit and the bad spirit. At death the bad spirit hovered around the grave of the deceased, sometimes taking the form of a fruit bat or another animal and terrorizing those who lived nearby. The bad spirit attacked people and bit them, the effect of which was to darken the victim's mind or cause temporary insanity (Mahony, 1969, pp. 134-135). But any harm the bad spirit could do was limited because the bad spirit remained in the immediate vicinity of the grave, stalked only at night, and could be frightened off by light. If the spirit became too annoying, a sorcerer was called to drive it into the earth to rejoin the body of the deceased (Bollig, pp. 20-22). The good spirit, which had the power to roam more freely than the bad spirit before it eventually drifted up to its abode in the sky, could be a source of blessings for its former family. This spirit, if well disposed toward the family, revealed to its kinfolk a wealth of valuable information on such things as new kinds of medicine, ideas for dances, and the location of good fishing grounds. On the other hand, the good spirit became a scourge for the family unless it was treated with proper respect and placated with gift offerings (Mahony, p. 137). When angry with the family, the good spirit caused serious illness and, oddly enough, was more of a threat than the bad spirit. We will later elaborate on the ambivalent nature of good spirits, because it was with the good spirit that Chuuk families attempted, in years past, to establish communication after the death of one of their members.

Thus, Chuuk Islanders today assume the existence of nature and ancestor spirits and their intervention in human events. Although many people might disavow a belief in spirits when asked directly, especially by a church minister, their behavior indicates otherwise. They are uncomfortable walking alone in the woods in the evening and avoid places associated with the more notorious nature spirits. People continue to use traditional medicine (safei) to ward off malevolent spirits even as they still believe in the efficacy of love potions (omwmwung) to win the affections of those whom they find sexually attractive.

Although belief in spirits exists in Chuuk today, it is limited compared with the past. Islanders no longer make offerings or prayers of propitiation to the spirits, neither to nature spirits nor to the spirits of the dead. In this way, most have satisfactorily integrated their belief in the spirit world with their Christian faith. Christian teaching, after all, recognizes the reality of a spirit world of its own populated by demons and angels. The panoply of spirits that shares the universe with Chuuk Islanders is regarded as a potent but lower order of beings subordinate to the Christian god. Hence, the spirit world of Chuuk might be seen not as a rival of Chuuk Christian beliefs, but as a substratum of this professed Christianity, even when it is devoutly embraced. Indeed, the Christian god is often called upon to counter the destructive influence of these spirits.

SPIRIT ENCOUNTERS AND CONTEMPORARY POSSESSION-TRANCE

There is no doubt, then, that Chuuk Islanders today see the influence of the spirits in their lives–in misfortune, in accidents, and even in the suicides of kin. People continue to report direct encounters with the spirits, that is, seeing ghosts, or being bitten by ghosts, or experiencing possession events through the spirits of deceased kin. But people do not trouble themselves with fine distinctions in their typology of spirit encounters. It is often difficult to know whether they considered the individual "possessed"–that is, whether the spirit in question has actually occupied the body of the victim– or whether the spirit acted as an external force on the person. The Chuuk term awarawar enu, "the coming of the spirit," suggests simply an encounter with a spirit, while the word waanaanu, "the vehicle of the spirit," implies actual possession. The people of Chuuk often use the terms interchangeably and few attempt to distinguish them.

The following four case descriptions show the range of behavior and belief associated with spirit encounters. The first two illustrate spirit encounters; the second two describe cases of spirit possession.

Not many years ago a woman in her early 30s, with a group of other women from her village, was net fishing in the shallow water just offshore. As she was hauling in her section of the long fishing net, she became entangled in it. While struggling to free herself, she saw in the deep water a big, naked black man advancing toward her. Upon freeing herself from the net, she ran toward the house screaming in panic. An adult male relative of the woman intercepted her on the way, but the woman threw him to the ground with apparent ease. While the woman cowered in the house over the next several days, an expert in medicine (sousafei) was summoned from another island to treat her. After applying his remedies and relieving the woman's distress, he explained that the woman had been afflicted by Soumwerikes, the ghost that troubles women who have recently given birth. The woman had just delivered her sixth child a few weeks before the incident.

Another story is told of a 16-year-old girl, from a different island in Chuuk, who was pregnant and soon due to deliver. One night as she was preparing to go to sleep, she saw a stout man, clad entirely in yellow, who lasciviously clutched at her dress. She fought off the man with remarkable strength while the household looked on in astonishment at her wild thrashing. Members of her family saw no assailant, but they noticed that the girl seemed to be pulled toward the door and resisted violently, screaming all the while. She calmed down later that evening, but more episodes occurred during the next few weeks. The family called the Catholic priest to assist them, and the incidents stopped when the girl delivered. This girl, like the woman in the previous case description, was unmarried when she became pregnant.

Both of these women were said to be agitated by an encounter with a spirit, and the Chuuk informants were quick to attribute their abnormal behavior to the work of the spirits. However, they could not say whether the spirit was actually "on" the individuals. The informant for the second case inclined to the view that the young girl did not experience true possession because there was no change in her voice.

Quite different are the next two case descriptions. In both of these the individuals were thought to have been possessed by the spirit of a relative.

Marie (pseudonym) was about 40 years old when the first incident happened. On this occasion Marie's brother was drunk in the house, so her husband and daughter asked him why he brought alcohol into the house. The brother answered that he would not talk to them but would talk only to Marie's daughter's husband, who was out at that time. The brother then stomped out of the house, and Marie started crying as she sat on the living room floor. During the night the ghost of Marie's mother came upon her and remained until morning. Marie's voice changed to that of her deceased mother and the mother's spirit told her son–the brother who had left the house–to hold her hand because it hurt (the mother had died of an illness that involved pain in her right arm). The ghost showed everyone that she believed in her son and trusted him because she asked him alone to relieve the pain by holding her hand. The informant thought this meant that the ghost was slighting the daughter, Marie. The ghost said she loved and pitied her son. She said she was angry with Marie and scolded her for being selfish and without love for the brother. The mother's ghost called in her sister (the aunt of Marie) and her brother.

Marie's family wanted to take her to the hospital, but she refused and went to another island. There she experienced a possession episode every night for a week, although during the daytime she was her normal self. At night she would awake shouting and crying, talking and acting like her mother. She once threatened to pick out and eat her son's eyeballs and even started to hurt the boy physically. The family had to pull her away from him. Sometimes she was unusually strong and wanted to run around outside. She would clutch her throat as though the ghost were choking her.

The family called in local people who knew Chuuk medicine. They told her to stay in one house and not move, but after she experienced possession in that house she wanted to move. The family used holy water and Lourdes water, but this calmed her down for only a short time. The Chuuk medicine experts then took her to a stream to bathe. Again she was temporarily better, but the family continued to think that she might die.

The informant also believed Marie would die, suffering a painful death. He believed the ghost was real and explained the situation in terms of family problems:

Marie's husband is having a difficult time with all this and is trying to find the right medicine people for Marie. Maybe there is love in their marriage, but they don't talk to each other about their feelings. Also, Marie had a sexual relationship with a cousin. She liked him. That was when Marie's husband and brother started getting drunk and causing problems.

The next case description is the story of a woman we will call Anna. She was about 25 years old when she went to visit her adopted father, who was seriously ill. When she lay down by her ailing father, she began to moan and cry. Then her voice changed to that of her father-in-law, who was still alive. When her husband started to ask her questions, she used bad language. The voice from Anna told everyone to leave the sick man because he hated them. Observers said she shouted loudly and was so strong that it was difficult to restrain her. Neighbors brought Chuuk medicine that they rubbed on her arms; the family sprinkled holy water on her face and body. The informant, however, thought that Anna was "lying."

These four case descriptions indicate the range of detail in the accounts and the difficulty in distinguishing what are only spirit encounters from spirit possession. Accounts are often too vague or incomplete to permit assigning them to one or another category with confidence, but accounts do reveal recurrent behavior indicative of spirit possession: changes in voice, unusual displays of strength, and the use of language prohibited between kin. Both Chuuk medicine and Catholic holy water were used to calm the person, implying that the onlookers saw the incident as an illness episode or at least as behavior that is potentially dangerous. The last case (Anna) is an anomaly because it is the only one we know of in which the possessing spirit was that of a living person. Both of the possession cases described here evidence another important feature: that is, the family as the focus of the episode, as the intended audience of the spirit's talk, and as the party intervening to resolve the situation. Finally, the reports about Marie and Anna, like so many others, indicate that what the people of Chuuk find significant is that the spirits speak about problematic family relationships.

The case descriptions of Marie and Anna reveal the patterns characteristic of most other possession descriptions. Most of those who experience possession are female; two of our informants thought male possession to be highly exceptional. A possession episode frequently begins with a severe pain, often a persistent headache or abdominal pain, or with a bout of illness. One woman's episode was preceded by a sore throat and laryngitis; another started with a pain in the chest that continued despite all the remedies she tried. The individual regularly then takes to a sleeping mat as the rest of the family gathers around her to look after her in her illness.

Next, the person typically experiences a violent bodily upheaval, perhaps yelling and screaming, sobbing hysterically, flailing her arms at those around her, or convulsing and shaking in what resembles a seizure. This often takes the form of an assault on those who are in the household–one girl suddenly grabbed her grandmother and began to throttle her; another picked up a broom and struck an older relative standing nearby. One woman (Marie) suddenly began beating her young son, all the while threatening to pluck out his eyeballs and eat them. Sometimes the individuals experiencing possession chased people around the house, throwing things at them and grappling with them, occasionally even wrestling them to the ground. Repeatedly informants commented on the remarkable strength of the afflicted woman, who was often capable of besting a much heavier grown man. But it is also noteworthy that the violence exhibited by those experiencing possession was controlled violence, much as a young man on a drunken rampage

in Chuuk will carefully moderate his violence, despite appearances to the contrary. The relatives of an afflicted woman may be scratched and sore from restraining her, but there is no instance to our knowledge of anyone having received a serious injury from someone who was experiencing a possession episode.

Individuals experiencing possession demonstrated other sorts of unusual behavior, some of it flagrantly transgressing the code of propriety in Chuuk society. Much of the time the afflicted persons simply moaned unintelligibly or babbled nonsense, but at times there was a sharp edge to their speech. Some wantonly insulted those around them and even used vulgarity that women of Chuuk under ordinary circumstances would never dream of speaking in the presence of their brothers and other male relatives. At times there was a sexual tenor to their remarks, as when they described to the full circle of relatives the sexual fantasies they experienced. One woman hitched up her skirt and scratched her pubic area while her family tried to cover her and pull away her hand. Another young woman, after telling some older relatives how ugly they were, made sexual overtures to preteenaged boys standing nearby.

In roughly one third of the case reports we have collected, informants recalled that persons experiencing possession demonstrated a strange prescience. One woman, for example, seemed to know who was about to come into the room and often announced the person's arrival before the person appeared. Others were sometimes able to point out exactly where lost or hidden objects were to be found. One young girl gave detailed instructions on where a pistol belonging to a recently deceased relative might be found. Before his death the man had concealed the pistol in a house in a distant village; it was found in the exact spot the girl indicated. Many years ago another woman in a possessed state predicted that a group of men who had gone to a distant island to work on a plantation would be back very soon. They did return a week later. The same evening that the woman had made her prediction, as her kinfolk later learned, the men were boarding the ship that would bring them back to Chuuk. One woman allegedly told her family that a boy who had been missing for a few days had hanged himself in the bush, and she told her family where they might find his body. At times the clairvoyance takes threatening forms. Women experiencing a possession episode have been known to predict deaths of relatives and to announce publicly the intimate details of the personal life of family members, including clandestine sexual relationships. For this reason the family members of someone who is experiencing possession are usually guarded in their dealings with her and do everything possible to avoid provoking an angry outburst, another indication of the ambivalent nature of even the good spirits.

Although one of the most dramatic and telling signs of possession is the change in voice that occurs when the "spirit" begins speaking, the transformation involves more than this. A woman in a possessed state usually takes on the very persona of the spirit that is upon her, and the timbre and tone of her voice and her speech patterns become that of the dead person. A corresponding change in gestures, facial expressions, and other mannerisms also occurs. When one stooped, elderly woman began speaking in the voice of her brother, she walked around the house ramrod straight–the first time anyone could remember her doing so–and began smoking, just as her brother had done during his lifetime. In the instance of Marie, described above, in imitation of her deceased mother's painful hand, Marie held her own hand as if to favor it and asked her "son" to massage it to relieve the "pain." Another woman stroked her throat while speaking, just as her mother had done during her lifetime, and conversed in the same soft, strained voice her mother had used.

At times persons in a possessed state may speak as more than one spirit during an episode. Recently a young girl, in a virtuoso performance of characterization, took on the speech and mannerisms of four different spirits during a single episode. There are other instances in which the afflicted person adopted the personae of two or three spirits in an evening.

Although the change in persona is usually brief, lasting only a few minutes at a time, the entire episode has a much longer duration. Individual episodes may range from a day or two to 2 weeks, with a few continuing sporadically for 2 or 3 months. One elderly woman was said to have experienced possession from the time of her paralysis through the last year of her life. Sometimes it is difficult to tell precisely when one episode ends and another begins, for some of the afflicted individuals have such a long and continuous history of possession experiences that they can be considered to be in a chronically possessed state. Two younger women have had a series of incidents that began in the early 1980s and continued intermittently until 1990. One informant, whose sister-in-law first experienced possession in 1980 at the age of 15, said that in recent years the afflicted woman has been exhibiting the symptoms of possession every 3 months.

Although we found definite patterns in the behavior of individuals who experienced possession, we were unable to discover similarly clear, distinct patterns in their backgrounds–except that they were almost exclusively women. The ages of the women appear to cluster; at least 20 were in their teens or twenties at the time of their first episode, but most information on ages is only an approximation. The ages range from one instance of a 7-year-old to an elderly woman whose first possession episode occurred when she was already 71. Chuuk spirit possession today occurs overwhelmingly among females, mostly young females.

The socioeconomic status of individuals experiencing spirit possession ranges from low to high. The profile includes persons with little or no formal education, university students, those with no experience away from Chuuk, and those who have spent time working or studying in Hawai'i or Guam. No correlation appeared between the strength of Christian beliefs and the incidence of possession; episodes occurred in the households of Protestant ministers and Roman Catholic catechists. Nor did any dear pattern emerge regarding marriage–about half were single and half married, and several were divorced persons. Although biographical data are far too incomplete to demonstrate significant statistical associations, when these limited data are viewed alongside information about kinship relations, a clearer association appears that links possession cases to some sort of family problem.

At least eight of the possession experiences were occasioned by a death in the family–for example, the suicide of a relative (4 cases), the recent death of a cousin, a father's death, the funeral of a sister, and the drowning of an uncle. Deaths and funerals were associated in former times with possession by relatives. Bollig (1927/1967, p. 16) encountered this cultural pattern around 1912; Goodenough (1963, pp. 132-133) recorded the link during his fieldwork after World War II:

The deceased's family and near kin watch the grave for four successive nights in order to give his good soul an opportunity to possess one of them…. If the good soul has not possessed a kinsman in the four days preceding its ascent, then no more will be heard from it. By possessing a kinsman, however, it serves notice of intention to maintain contact with its surviving relatives. The person possessed becomes a medium for the good soul, which in turn now belongs to a special category of beneficent spirit. Symptoms of possession are moaning, violent shaking, and speaking in a special tone of voice, which signifies that a spirit rather than the medium is really doing the talking.

Other occasions noted in the case histories centered around life crises such as pregnancy, recent or illegitimate childbirth, parents' disapproval of a boyfriend, or a recent marriage. Other occasioning events were as vague as "bad family feelings" or as specific as a family feud over landholdings. The link between contemporary Chuuk possession-trance episodes and kinship is unmistakable. Most often the possessing spirit, the person experiencing spirit possession, and the audience addressed by the spirit were all close kin. The most common outcome of the episode was family reconciliation or at least a family meeting. A reminder here is that most of the reports we received were already culturally processed, and remembered details were probably selected to match the cultural expectations. If the Islanders thought of possession-trance episodes as family affairs, chances are that strangers coming and going and other culturally extraneous detail would not be remembered and reported.

Even the casual observer of Chuuk society would not be surprised at the close connection between spirit possession and kinship concerns. Kinship still plays an important role in Chuuk society, as Hezel (1985, 1987, 1989) has maintained in his study of social problems in Chuuk and elsewhere in Micronesia. Goodenough (1963, p. 138) saw this a generation ago: "It is impossible to live independent of one's kinsman without loss of almost all of one's rights, privileges, and immunities within the social order." Earlier, Gladwin and Sarason (1953, p. 145) had noted that the security of close kin relations is a double-edged sword for the Chuuk inhabitants, that is, close kin relations can also be a source of anxiety: "While the lineage or other kin group provides a large degree of economic and undoubtedly psychological security for the individual, the possibility of rejection by the members of such groups must be a source of serious anxiety."

During the period between Goodenough's (1963) and Gladwin and Sarason's (1953) writings and those of Hezel (1985, 1987, 1989), the shift from a subsistence to a cash economy has altered kinship relations, attenuating those of lineages and placing different burdens on parents. Although the economically independent nuclear family of postindustrial societies is not characteristic of Chuuk today, the movement to a cash-based economy has placed new strains and demands on the roles of mother and father (see Hezel, 1989).

Despite the deep historical roots for the association between kinship and possession as a culturally valued event, spirit possession is still perceived by Chuuk Islanders as dangerous or at least potentially so, and they lose no time in seeking treatment for an afflicted person. Spirit possession is considered dangerous for several reasons. First, it moves both the entranced person and the onlookers into the spirit world. True, there are good spirits, but Bollig (1927/1967) noted long ago, and Gladwin and Sarason (1953) more recently, that people are just plain afraid of ghosts or spirits. Moreover, the nature and identity of the spirit may not be known. Is this a bad or helpful spirit? Is this merely an encounter where the spirit bites and causes injury, or is it one that will result in illness or even death? Chuuk Islanders attribute all sorts of misfortune to spirit intervention and see any encounter with the spirit world, even with good spirits, as ambiguous at the least and therefore potentially dangerous. Second, the wild actions and words of the afflicted person threaten injury to the one exhibiting the symptoms and to those around her. Recall the case where the mother threatened to gouge out the eyeballs of her son. Third, women experiencing possession often defy the cultural canons that govern social conduct in Chuuk. Women flaunt their sexuality in the presence of male relatives, they voice publicly what ought to go unspoken, and they flail verbally and often physically at those to whom they are expected to show respectful restraint. In short, possession episodes are potentially dangerous because they unleash the spirit world and threaten violence to physical well-being and social equilibrium.

It is not surprising then that Chuuk Islanders continue to view possession-trance as within the realm of illness and medicine. In earlier times the Chuuk concept of illness was broad and embraced what we might call the physical, psychological, and social dimensions of life. The people of Chuuk looked to the spirit world as both cause and cure for disturbances to physical well-being, healthy psychological functioning, and good social relations. Today they still use local medicine to find out which spirit is afflicting the person, why the person is so afflicted, and what treatment will cure the problem. We know that the spirits, communicating through possession, were once a source of new medicines (Bollig, 1927/1967; Mahony, 1969), and we have interviewed one medicine expert who claimed that she received new medicine through spirits in a dream. And just as the Islanders have managed to integrate Christianity and spirit beliefs, so too have they integrated Chuuk medicine and hospital medicine (Mahony, pp. 34-35). Each type of medicine has its place.

Because possession continues to be seen as illness, the afflicted individuals usually return to their sleeping mats in a weakened condition and are treated by the family as ill. They typically then turn to healers for medicine that combines local plants, roots, and leaves, accompanied by chants. Although the practice of chanting has declined (we could elicit very few examples, perhaps because the chants are treasured family secrets), villagers still recognize local specialists or experts (sousafei) in medicine for specific problems. According to our reports and interviews, however, the family and onlookers more commonly seek treatment by turning to Christianity, praying over the afflicted family member or sprinkling holy water on her. If the episode continues for more than a day or two, someone usually summons a Catholic priest to assist the afflicted person, even if the family is Protestant. Part of the attraction of Catholic services no doubt lies in the ritual, often conducted with litanies, candles, religious medals or relics, and abundant ablutions with holy water. There are instances, though, when Catholic families have turned to Protestant faith healers for treatment, especially when the healers are known to use chants and more theatrical techniques. The Christian ritual, which is often a combination of a natural ingredient like water and chanting or prayers, is similar to the treatment provided by the sousafei. Inhabitants may even think of the ritual as within their own category of medicine; certainly, as Mahony commented (1969, pp. 137, 143-144), illness, spirits, and possession are intimately related.

The early reports of Kubary (1878-1879), Bollig (1927/1967), and Girschner (1912-1913) agree on the essentials of spirit possession, which normally occurred in the lineage meeting house in which was hung a model double-hulled canoe (naan) that served as a vehicle for the descent of the spirit. Smeared with fragrant perfume, a spirit medium, known as a sou awarawar or waatawa, was seated in the midst of the family, and as the family chanted the name of the dead relative with whom they wished to speak, the medium "mumbles to himself, begins to moan, breaks several times with his mouth wide open, and then lapses into convulsive trembling" (Anonymous, 1915, p. 8). Often his head jerked back and forth, his neck bent "under the weight of the spirit" (Bollig, 1927/1967, p. 61) and his hands began to quiver. As the trembling became more violent, the family, recognizing that the medium was in a state of possession, began asking questions of the spirit. Although the voice of the medium might not change, his response was spoken in a special "spirit language" that had to be interpreted by someone especially knowledgeable in this form of speech (Girschner, p.190).

HISTORICAL TRANSFORMATION OF CHUUK POSSESSION–TRANCE

The spirit possession described in the older literature, although generally similar in form to the type of possession occurring today, was significantly different. First, possession was once purposeful: It resulted from a deliberate attempt to communicate with the spirits of the dead (nature spirits and gods could not be summoned). When serious illness befell someone, the family would typically gather to summon the spirit of one of their deceased members in the hope that the spirit would reveal the nature of the illness. At other times of family crisis–during a famine or confrontation with another family, for instance–the lineage group might call on a spirit to provide the crucial information needed to survive unharmed. The spirits were also consulted on the whereabouts of missing valuables or about what would happen in the future. Shortly after the death of someone in the family, the relatives would communicate with the spirit in the hope of gaining access to knowledge–the knowledge hidden from the living–that might have value for them.

Second, in former times the medium in spirit possession was, ideally at least, someone from within the inner family group who showed a sensitivity to the movements of the spirits. The chief requirement for mediums, in the words of a German doctor who observed possession incidents during the first decade of this century, was that they "have a nervous sensibility, increased by continuous auto-suggestion, so that they can easily go into a hypnotic state" (Girschner, 1912-1913, p. 190). Anyone might be "chosen" as a medium–man, woman, or child. Bollig (1927/1967, p. 67) recorded that those whom the family regarded as its leading candidates were placed upon mats and told to relax; eventually one of them would suddenly jump up with a shout, signifying to all that he or she was the one selected by the spirit as its medium. If Bollig's report (p.70) of 90 mediums on Fefan Island is to be believed, there was no dearth of individuals able to serve as spirit mediums in former times. Almost every lineage would have tried to cultivate its own medium, and many lineages apparently had more than one person able to communicate with the spirits. Finally, although females could and did perform this function, the literature describes only cases of male mediums.

Spirit possession in Chuuk, then, appears to have undergone a substantial transformation in this century. The possession incidents described in the early literature were intentional, were aimed at obtaining important information from ancestors, and were mediated by males or females. (We admit to the possibility, however, that early writers missed or failed to record the more spontaneous and involuntary type of possession-trance.) Given the available evidence, the possession incidents recorded today are different. They are involuntary, occur mostly at times of family stress when so perceived by the afflicted individual, and involve mostly women. It is difficult to date this change with precision, but we must assume from the case reports we collected that the newer form of possession became increasingly common after World War II. That all four cases of unambiguously intentional possession in our records date back to the 1930s or 1940s supports this assumption. The older form of intentional possession seems to have gradually declined during this century, with incidents of this kind occurring only rarely in the postwar years. The model double-hulled canoe (naan) that was once suspended from the Chuuk meeting house to facilitate spirit communication could no longer be found after World War II, although it was remembered by older informants (LeBar, 1964, p.181-183). By the late 1960s or early 1970s, as spirit calling became rare, the type of involuntary possession that is the subject of this article was commonplace.

One of the major reasons for the disappearance of voluntary spirit possession was the growing influence of Christianity in Chuuk, as can be seen in Bollig's descriptions (1927/1967). What might be called "possession on demand" was discountenanced by a society that had enthusiastically embraced Christianity. Anything that might be interpreted as spirit worship, including offerings to the spirits or supplication for assistance, was bound to be judged by missionaries and early native pastors as incompatible with Christian practice. The dividing line between "heathen" and Christian was drawn across such practices as voluntary spirit possession and the active solicitation of help from the spirit world. It was one thing to acknowledge the power of the spirit world and distinctly another to seek help actively from this realm. Chuuk Christians might retain a fearful respect for the pre-Christian spirit world and attempt to counter the ever-present threat of these beings, normally by an appeal to the higher powers that Christianity put at their disposal, but they would not initiate contact with the spirits to secure their assistance. Christianity, therefore, did not eradicate the belief in spirits among Chuuk converts any more than it did among the first century converts in Asia Minor; it simply eliminated, over time, the propitiation of these spirits.

Seen from a slightly different perspective, the historical transformation of possession in Chuuk has been a change in status. The earlier literature attests to the existence of a highly institutionalized form of possession-trance in the person of the waanaanu, a recognized status like that of a medium. The medium as a distinct person and recognized position has disappeared, but the behavior continues, as does some recognition of that behavior as "possession." Formerly, spirit possession was the mark of a distinct person, the waanaanu, or medium; today it is a type of behavior, an illness.

In summary, Chuuk spirit possession manifests beliefs about the spirit world, especially beliefs about the intervention of spirits of the dead in the affairs of living kinfolk. The precipitating events are most often family related problems, but family problems as they are felt and experienced by the individual who experiences possession. Although possession-trance has deep historical roots and was once a sought-after experience, today it is viewed as a potentially dangerous illness calling for the medicines that the spirits themselves reveal. Formerly, possession-trance was on demand in service to the family and occurred in the person of a medium, the waanaanu; today the possession-trance is involuntary, spontaneous, and although it still occurs in service to the family, it is triggered by the family problems as felt by the individual who experiences possession.

POSSESSION-TRANCE: ALTERED STATES OF CONSCIOUSNESS

The Chuuk phrase for what we have called spirit possession is emen soope a teeta woon, meaning "a spirit is on" the person. The most important indicator of possession is the person's speech, by which the spirit manifests itself. The heart of the matter for the Chuuk Islanders is the spirit being "on" the person (which we gloss as "possession"), not the ecstatic behavior; the behavioral manifestations, such as shaking or gestures of wild abandon, are only symptoms of spirit possession. Perhaps it would be more precise to say that the Chuuk concept of possession focuses on the spirit speaking.

Anthropological studies of possession, on the other hand, have focused on the distinction between the possessed person's behavior and ordinary, daily action, labeling the former behavior trance, hysteria, dissociation, or altered state of consciousness. Some writers find it helpful to distinguish between trance and possession, hence the term "possession-trance." Bourguignon (1968, 1974, 1991) has stressed the crucial distinction between the phenomenon itself and the interpretation of the phenomenon. One person may look at the wild behavior of Chuuk trance and call it possession; another may call it hysteria. In this article we prefer a classification for the phenomenon that is relatively neutral and that carries a minimum of interpretive baggage. We have adopted the popular term, trance, or its technical twin, altered state of consciousness (ASC), as best identifying the phenomenon with a minimum of interpretation. As a working definition of ASC, we offer Ludwig's description (1968, pp. 69-70):
 

those mental states, induced by various physiological, psychological, or pharmacological maneuvers or agents, which can be recognized subjectively by the individual himself (or by an objective observer of the individual) as representing a sufficient deviation, in terms of subjective experience or psychological functioning, from certain general norms as determined by the subjective experience and psychological functioning of that individual during alert, waking consciousness. This sufficient deviation may be represented by a greater preoccupation with internal sensations or mental processes than is usual, by changes in the formal characteristics of thought, and by impairment of reality-testing to various degrees.

Several implications of this description are worthy of comment. The behavior that can induce this state and characterize its onset ranges from sleeplike states to hyperactivity. In Chuuk the onset of ASC is often marked by hyperactivity: shouting, running, or unusual displays of strength; but other trances begin when the person is Iying in bed or sleeping. The pattern can thus be one of hyperactivity followed by calm or quiet followed by a phase of intense activity. In cases of the calm-then-hyperactivity pattern, we can only speculate that the triggering mechanism may be a combination of prolonged meditation and inactivity, as Henney (1968) found among the trancers of St. Vincent in the West Indies. There is no evidence that Chuuk trance is brought on by liquor or drugs. Whatever the cause, the Chuuk cases display an increase in activity, frequently with heightened emotional behavior. Ludwig (pp. 72-73, 76) specifically cites spirit possession states, depersonalization, and hysterical conversion reactions as examples of ASC characterized by increased activity. He also notes that in revivalist religious cults having possession beliefs, the behavioral manifestations include jerks, laughter, rolling, barking, spinning, shaking, quaking, convulsions, whirling and howling, running and leaping, and shouting and crying. Prince (1968, pp. 121, 127) finds similar manifestations in ASC, such as a brief period of collapse or inhibition at the onset of the episode followed by a period of hyperactivity with tremors of head and limbs, even jerks. The return to normal, alert consciousness is followed by sleep or exhaustion and amnesia regarding events during the episode. Among the Chuuk cases, only one person reported remembering the events; eight others had no recollection of what had occurred; all other case reports omitted mention of memory. Both Bourguignon (1991) and Ludwig (p. 83) regard personality changes and even identity change as characteristic of ASC. The changes in voice mark the identity shifts in the Chuuk cases, at least in the recent cases.

The Chuuk cases, old and new, thus fit into a widely observed behavioral pattern described as an altered state of consciousness. ASC is a convenient, nonjudgmental characterization of the phenomenon. Possession is not the phenomenon; possession is an interpretation of the phenomenon, hence the designation of possession-trance. Another advantage in using ASC as a label for the phenomenon is that detailed descriptions of ASC, such as that offered by Ludwig (1968), suggest reasons for why possession interpretations are mobilized. ASC is often characterized by shifts in personality and identity and a lessening of the boundaries between self and others. Because possession belief is already an available choice within Chuuk culture, the interpretation of the cases as possession is even clearer. The person experiencing possession consciously or unconsciously has selected a form of behavior in which she can transcend the limitations of her own personality and self. The behavioral manifestations of ASC are culturally learned and thus culturally specific to a given society. The remarkable similarity exhibited in the trance behavior of the Chuuk cases confirms that the patterns of action and speech are learned behavior, even though the participants are outside normal consciousness. Part of the similarity may also be due to the selective, culturally patterned remembering of the informants. In any event, that the trancers' friends and relatives witnessing the behavior, and the informants, have a shared understanding of its meaning also suggests that the behavior is culturally learned and shaped. The ASC thus conveys meaning in a culturally encoded act of ritual communication that other members of the culture understand.

FAMILY CONFLICTS AND CULTURAL COMMUNICATION

For all their reliance on the possession interpretation, island informants were well aware of another, more "this worldly" dimension to spirit possession. One informant pointedly mentioned that, apart from the few cases in which the victim violated a strict taboo, spirit possession was usually brought about by family problems. When told of an incident of possession, people routinely will ask what precipitated it. The first suspicion is that a family problem, perhaps a serious quarrel or tensions that have smoldered for several months, occasioned the possession episode. For example, one young girl who had experienced episodes was understood to enter a possessed state just an hour or two after a noisy drunken fight had broken out in her house. Other kinds of stress may also bring on possession-trance, such as the death of a family member or a dispute about family land. In general, the data indicate that the person nearly always has experienced considerable stress associated with friction in the family. Sometimes that individual plays a direct part as a protagonist in the dispute and is cast in the role of tormented observer. Overall, there is little doubt that stress is a major factor contributing to the possession-trance episodes. We propose that possession-trance is an outcome of that stress, functioning as a socially sanctioned release of culturally inhibited expression, perhaps even as a redirection of aggression.

The connection among culturally inhibited expression, possession-trance, and culturally accepted or tolerated patterns of expression has been widely documented across cultures (Bourguignon, 1968; Dobbin, 1978; Foulks, 1972; Kleinman, 1980; Metraux, 1972; Parsons, 1985; Wedenoja, 1990). In Chuuk society, the tendency toward inhibition and repression of emotional expression has been noted, as in Gladwin and Sarason's description (1953, p. 227):
 

What the Trukese individual thinks and feels is not likely to reach overt expression in a spontaneous or direct manner. This would be especially true in the expression of strong feeling, particularly aggressive or hostile feeling. One might almost say of the Trukese that one of their rules of behavior is that what is strongly felt should not be expressed. If we bear in mind that in the face of novel, conflictful, or problem-solving situations the Trukese tend to react passively and dependently rather than aggressively or actively, and add to this their strong inhibitory tendencies, our conclusion would be that the Trukese outward passivity is a kind of learned defense against strong aggressive tendencies. In fact, one might say that perhaps one of the most crucial problems which faces the Trukese is that of expression of aggression.

Elsewhere Gladwin and Sarason (1953, p. 241) concluded that aggression rarely reaches the level of direct aggression, but that it is disguised in such indirect forms as gossip and sorcery. They also found situations within family and close kin relations that were characterized by a high potential for hostility but where it was impossible to express the felt aggression in a direct manner. For Gladwin and Sarason the kin group was the "security system" of the individual Islander.

Both Spiro (1950, pp. 105-106) and Lutz (1988, p. 194 inter alia) recorded a similarly strong cultural prohibition of direct expressions of hostility and aggression on nearby Ifalik. And Black (1985, p. 272), describing the folk psychology of Tobi Islanders, associates the prohibition of direct hostility with the nature of small-island life:
 

Tobian everyday social life is pleasant in the extreme. People are highly skilled at constructing pleasant and rewarding interactions out of which can come the cooperative behavior on which life on this island, as it is presently constituted, depends. This pleasant tone is the product of strict adherence to the social norm that demands it while prohibiting direct expressions of hostility. It also depends, in this extremely small-scale society, on the intimate knowledge available to each person about the biography and personal attributes (especially those which are best called "foibles") of every other person…. The disputes and divisions are associated with a good deal of interpersonal hostility and other negative feelings, many of which arise from the inevitable frictions of a life lived in constant and inescapable intimacy.

Like Black, Spiro (1972, pp. 480-481) found the ethic of nonaggression to be a functional adaptation to life on small islands, but he also linked the strong sanctions on Ifalik against aggression with culturally sanctioned beliefs in ghosts and possession. Possession beliefs thus served to channel culturally prohibited expression. Elsewhere, Spiro (1950, pp. 103-104; 1965, p. 480) writes that little overt aggression is seen in interpersonal relationships among adults, but he also cautioned that this does not mean there are no aggressive impulses; these impulses are either few or repressed. Spiro has indicated that the resultant anxieties and repressed aggression would become unbearable were there not some socially sanctioned alternative to overt and direct aggression. The Ifalik belief in spirits or ghosts that are thought to cause evil and illness–including possession–acts as a balancing mechanism for redirecting antisocial drives (Spiro, 1965, p. 479).

This belief makes it possible for the Ifalik to displace their own aggressive thoughts away from themselves and against a common enemy, the spirits (Spiro, 1965, p.482). Spirit belief and possession interpretation thus become acceptable outlets for forbidden aggression.

This interpretation would seem easily applicable to Chuuk, where similar spirit and possession beliefs provide meaningful outlets within a similarly inhibitory culture. But the transfer is not that simple. As Lutz (1988, p. 195) observes, Spiro's approach may be too mechanical: The "person becomes not an agent but a passive reactor to internal emotional forces." True, Spiro's model does not explain individual agency, that is, why some individuals use possession-trance and others do not. On the other hand, possession-trance involves by definition the passivity of the individual who experiences possession and the activity of the spirit. Lutz also finds Spiro's interpretation overly individualistic in depicting Ifalik fears as primarily of their own violence rather than that of others. She argues that the causes of Ifalik fear are broader than their own selves and that belief in the spirits and spirit possession is a way of defining the problem of violence for oneself and for others.

The combined insights of Spiro, Black, and Lutz throw light on Chuuk spirit possession. The relationship between what is repressed and what is expressed is not automatic; a symbolic communication occurs among the entranced individual, the possessing spirits, and the observing kin and friends. The Chuuk cases are, as Lutz (1988) found on Ifalik, talk about what is right and wrong, especially regarding proper relationships among kin. What Lutz (p. 95) wrote about Ifalik might apply equally well to Chuuk:
 

As Spiro also remarked, the spirits personify danger, give it a face. By so doing, they allow for more complex explanation and rehearsal of danger and community response to it. The drama of the spirits on Ifaluk communicates that the anger of others, whether spirit or human, should produce fear, not angry counterattack. This, the drama says, is our path to safely, at least for now, at least here.

One longtime resident of Chuuk claimed that the possession-trance episodes were just ways for young girls to get attention. The episodes are that, for they did generate considerable attention, but they also offered advice and potential solutions to family disputes. Although the trance behaviors appear bizarre even to inhabitants, the communication is not gibberish; they are instead goal oriented and thus may serve a useful function (see Van der Walde, 1968, p. 67).

Chuuk possession-trance sometimes promotes greater domestic tranquility–siblings are reconciled, kin admitted to neglect of other kin, the neglected are helped. But the data do not indicate any outcome, positive or negative, for many cases–hence the impossibility of determining the success rate. Perhaps the mere outburst and release are therapeutic in themselves. The case histories strongly suggest that the person who experiences possession must offer verifiable predictions, advice, and demands about the network of kin relationships for the episode to be considered authentic, that is, adaptive, within the Chuuk ethos (on maladaptive expressions of ASC, see Prince, 1968, p. 86).

Contemporary Chuuk possession-trance works not only as a culturally approved outlet for prohibited expression but also as a new way of handling problems of individualized and personal stress caused by family relationships. The mobilization of possession-trance to handle family problems shows remarkable continuity with the past, but the focus on the personal stress of the one who is experiencing possession is a new feature. In the older cases described by Kubary (1878-1879), Girschner (1912-1913), and Bollig (1927/1967), the possession-trance episodes served family or lineage interests by offering solutions to a crisis facing the group, such as a famine, bad weather, or a family squabble. The waanaanu of old was clearly in the service of the group. Contemporary possession-trance, on the other hand, springs from the feelings or perceived stresses of the individual trancer who draws on the model of the now defunct waanaanu as a coping mechanism. The Chuuk cases demonstrate the extent to which the perception and evaluation of stressful stimuli are culturally shaped, as is the selection of possession-trance as a coping mechanism (see Kleinman, 1980, pp. 146-148). Possession-trance as a coping mechanism is universally available but culturally legitimized only in certain societies or contexts. In Chuuk it continues to be culturally sanctioned even though it is considered by Chuuk observers to be dangerous and to require attention. By interpreting trance as spirit possession, the people of Chuuk legitimate this behavior as a sanctioned alternative. Calling in a priest or minister–who sprinkles holy water, performs exorcisms, or attributes the trance to the work of Satan–also serves to validate their belief in spirit possession.

Why do family problems seem to be such a powerful source of personal stress today, unlike at the turn of the century? The obvious answer–albeit one difficult to prove–is that it is because of the impact of rapid social change on the family and lineage (see Kleinman, 1986, pp. 107, 125, 140ff.). Studies of alcohol use and epidemic suicide in Chuuk also point to postwar social change, especially in socialization and family roles, as the cause of social problems (Hezel, 1989; Marshall, 1979; Rubinstein, 1995). The words of the recently spirit-possessed women offer additional clues in this direction. While in trance, the women spoke with a problem-solving authority that formerly might have been held by a lineage elder or even a chief, and in so doing they effectively substituted for absent lineage authority. Moreover, if we accept the thesis that aggressive drunken comportment is the young men's outlet for defunct male roles (Marshall, 1979), then female possession-trance may be seen as a corrective to the family disruptions caused by widespread alcohol abuse by both the youth and adults in Chuuk. While Chuuk males have adopted an imported resource, alcohol, to deal with role expectations, females have drawn from resources of old to handle new problems.

SOCIAL DRAMA AND THE RESOLUTION OF CONFLICT

The significance of the spirit communication must be understood in the context of the actors, gestures, voice changes, and action that create a cultural drama. The principal dramatis personae include the entranced individual who has suffered some stressful personal crisis, the ancestor spirit speaking with the voice of authority, and the family or friends who witness and respond to the possession-trance. The episodes are processual, developing in stages; they work changes in social relations and effect a resolution of the social conflict or problem (see Turner, 1974, pp. 37-38).

In a typical scenario of Chuuk possession-trance, stress caused by mounting family turmoil increases to a point at which one individual in the family explodes in a barrage of words and wild activity. As the trance behavior continues, the family ties to learn the identity of the spirit and determine whether it is good or bad. Possession-trance draws attention to the plight of the afflicted woman and brings the family together at her bedside. Her words carry a special significance and authority. The sorts of messages communicated by the enhanced person to the family are usually a plea, or a command, to care for the neglected members of the family so as to resolve whatever conflict is dividing the family. The process often ends with a family meeting, with family members crying, sobbing, and promising to change their ways. The pattern of these dramas conforms generally to Turner's (1974, pp. 37-38) proposed stages of ritual social drama: first a break in social relations followed by mounting crisis, then regressive action and reintegration. Precipitating stimuli are events such as fighting in the family, alcohol problems of a relative, or neglect of some family member, or in Kleinman's terminology, the external "stressors" (see Kleinman, 1980, 146-148) cause the break and trigger the possession-trance. The trance event itself begs regressive action, for the spirit voice brings into the open the perceived injustice and demands redress. Reintegration occurs after the possession-trance episode: Relatives cry, plea for forgiveness, or call meetings to reconcile differences. Possession-trance breaks the status quo and creates a new set of relationships (Turner & Turner, 1978, p. 249). As a cultural drama, then, Chuuk possession-trance serves not only as a cathartic release of stress caused by family problems but also as an attempt to redress the existing problematic family situation.

CONCLUSIONS AND DISCUSSION

If there is one theme that unites both the Chuuk understanding of possession-trance and that of outside analysts, it is communication. Even the earliest reports of Micronesian religious beliefs and practices pointed to the role of communication. Father Juan Cantova (1722) found that the Carolinians paid more attention to their ancestral spirits than to the gods and that they used mediums to communicate their needs to the ancestors. Contemporary Chuuk inhabitants also focus on communication: The spirit speaks with a changed but calm and recognizable voice. Observers–especially those with a psychological bent–might infer that the possessed host is communicating her innermost, subconscious thought and that the spirits are a means or vehicle for externalizing an interior state. Both understandings have recourse to spirit belief, but the people of Chuuk approach the spirit communication somewhat differently. For them the human host is, as the waanaanu was of old, a vehicle between the spirits of old and the family of today; for a psychologically-oriented analyst like Spiro, the spirits are the vehicle for repressed communication.

We have seen that the spirit communication is goal oriented–it moves toward a solution to family crises. The often bizarre opening events of a possession-trance episode belie the effectiveness of the communication and language used. For Chuuk participants, the event may indeed be craziness or illness, but only the ongoing drama will tell whether this is illness or important communication. The people of Chuuk appreciate the effectiveness of the communication, such as when a sister was possessed by her deceased mother: The voice was that of the mother, and the mother's spirit lectured her son about his alcohol abuse and child neglect. The mother did have grounds for chiding her son; words from her were proper and effective in a way they could not have been if coming from the sister. Within Chuuk culture, a sister dare not speak to her brother so directly and aggressively. The spirit communication is culturally patterned as an outlet for repressed expression and is thus comprehensible to the Chuuk people. Communication through possession-trance in Chuuk fits a worldwide pattern of socially acceptable expression for socially prohibited or repressed feelings.

Chuuk possession-trance is comprehensible communication because it flows from traditional Chuuk spirit and possession beliefs. This communication does not use high ritual as much as it does drama to convey the message of social tension. The change of persona in the person experiencing possession legitimizes and sanctions a message that consciously or unconsciously is perceived as forbidden communication through normal means of discourse.

A deeply social and moral dimension attaches to the words of the spirits. The trancers speak about what is good and bad in concrete situations, although sometimes from selfish, self-serving motives. They provide relief from the stress and friction arising from family relationships. The possession-trance satisfies more than just the individual's needs and wishes. It addresses group needs, especially those of kin.

The communication theme of Chuuk possession-trance produces what might be called a kind of family therapy. The enhanced individual role plays, and a drama of family problems unfolds during and after the possession episode. The outcome is frequently a family discussion or meeting, a confession of mistakes and wrongs. The Chuuk experience of possession belief and trance behavior works like the impartial therapist. Of course, Chuuk possession-trance as family therapy is only an analogy, but the basis for such a comparison is the focus on family: The possessing spirits are frequently those of the ancestors, the spirit words are about family, and the family often discusses the problems that surfaced during the episode. Contemporary Chuuk women exhibit a behavioral complex that is both ancient and modern, one that is remarkably similar to what happened in the meeting houses a century ago and one that also takes place today in a therapist's office.

The most obvious feature of contemporary Chuuk possession-trance is also its most puzzling aspect–that it is a woman's role. The cross-cultural data show possession-trance to be a predominantly female role across the globe (Bourguignon, 1968, 1991), so the mere fact that Chuuk women experience possession might be expected. The question is why should a male or female status of former times become a female gender role today. Another question is what precisely does this communicate about Chuuk gender roles. At first glance, two answers seem plausible. First, the mostly female trancers could be said to constitute a protest movement against male-generated problems such as alcohol abuse and male dominance. Second, female possession could be the functional equivalent of male drinking. There is little doubt that male authority roles dominate in contemporary Chuuk culture, but the experiences of female possession-trance do not coalesce into an anti-male cult or movement, such as the Zar cult identified by Lewis (1971) in Ethiopia. Nor is it a protest movement against colonial and outside influences, as in Palau's Modekngei (see Barnett, 1949; Yaoch, 1966). Participants in the Chuuk possession-trance episodes and the medicines they use demonstrate a curious coexistence with Christianity, not a break with it or wholesale readaptation of its teachings. Our observations do not support the idea that Chuuk possession-trance is a protest movement or cult, but the accounts do point to something less obvious, to a more subtle and indirect note of protest against the helplessness of female gender roles. Mahony (1969, p. 246) was the first observer to comment on the strong association among the Chuuk spirit world, spirit-caused illness, and women:
 

It is rather interesting to note how many of these [spirit powers] are concerned with restraining, controlling and channeling the behavior and activities of women, either by threats to their own health, or to the health of their children. The spirits seem to be supporting established social authority which, of course, has always been in the hands of the men.

We suggest that in a cultural system where the spirits communicate the social structure, women are consciously or subconsciously using the spirits to break through the social structure. Turner (1974, pp. 172-173), as noted earlier, sees ritual drama, especially that involving trance and spirit possession, as breaking through the status quo in a type of antistructure. Paradoxically in Chuuk as elsewhere, the trancers use traditional beliefs about spirit possession to defy traditions of proper communication between the sexes. This is not so much a protest as it is a statement of affliction and pain (cf. Lewis, 1971, pp. 66-99). The Chuuk examples, however culturally sanctioned, do not rise to the level of an organized cult or movement of the afflicted and oppressed, as Lanternari (1965) describes in Religions of the Oppressed. Although it is culturally appropriate that possession-trance be selected as a way of expressing the affliction and stress, we cannot explain precisely why only women select a role that was earlier taken by both women and men. Perhaps a due lies in the way males express their stress.

Could possession-trance serve for females what the consumption of alcohol serves for males? This too seems plausible at first glance. The people of Chuuk do make the comparison: Drunks are possessed by alcoholic spirits. Marshall (1979, pp. 51, 54, 127) put it more poetically: The possessing spirits are bottled, imported, and purchased. At a deeper level, such a comparison would hold that just as the role of the male is to project an image of bravado and violence, so the woman must project her central role in the matrilineage. The male gets drunk to express a warrior's machismo (Marshall,1979, p.99); the female experiences spirit possession to express family and domestic needs. The logic of comparing male drinking with female possession is attractive, but the analogy fails. Male drinking is a regular social routine; possession-trance is not everyday behavior. The prevalence of male drinking is far more widespread in the population compared with female spirit possession (which is less than 100 cases, granted the limitations of Cathy Hung's convenience sample from the Chuuk Lagoon and our interviews). If Marshall is correct, males drink to fulfill the lost status and role of warrior. Female possession, on the other hand, is not a replacement role even though the domestic stress on contemporary females may be new, different, or greater. Perhaps a case could be made for the function of male drinking as "letting off steam," but male drinking certainly does not have the therapeutic potential in family disputes that possession-trance does. On the contrary, male alcohol abuse is often the precipitating event that leads to a possession-trance episode.

In short, possession-trance has become a female gender role. As such, possession-trance responds, in part at least, to the situation of Chuuk females in a society that favors male authority. Like male drinking, female possession-trance also responds to personal stress, but unlike males, the stress females respond to arises from family and kinship problems. In one sense, female possession-trance operates like the male alcohol complex by allowing for a period of time-out from the daily cultural rules. In another sense, it is a special time-in for communicating what would otherwise be impossible. Selection of possession-trance is particularly appropriate because trance is a time-out, or liminal, period between what has been and what will be. The contemporary use of possession-trance shows distinct continuity with the old Chuuk religion, especially in the person of the entranced medium, the waanaanu. This speaks of the resilience and adaptability of Chuuk culture. Both the waanaanu of old and contemporary possession-trance are in the service of the lineage or family, but the contemporary cases are spontaneous, not on demand, and stimulated by personal stress.

Why do certain women of Chuuk, however subconsciously, select possession-trance? At a societal level, the selection is possible because it is culturally available. At a gender level, the selection is today more appropriate for females. But at the individual level, neither data nor explanatory models answer why a given female experiences possession-trance. The case histories do not give enough information to answer this question. Many of the subjects are described as normal, healthy, and functioning well within Chuuk society. Their entranced behavior itself may be considered an illness or a potentially dangerous condition, but there is no correlation between inadequate or neurotic personalities and the incidence of possession-trance.

The explanation we offer here is only part of the answer. Chuuk possession-trance has survived the onslaught of intense modernization not only because it is culturally available and culturally understood, but also because it works. It does communicate family problems as felt by individuals. Although we have described the historical shift from possession-trance as a mostly male (waanaanu) experience to a mostly female experience, we have not adequately explained it. We are tantalized by the possibility that the female gender role of keeper of the matrilineage is now being served by possession-trance. We suspect that more specific data on the kinds of family and lineage problems associated with possession-trance will show us a clearer link between possession-trance as a medium of communication and Chuuk gender roles.
 

References

Alkire, W. H. 1989 Land, sea, gender, and ghosts on Woleai-Lamotrek. In M. Marshall & J. Caughey (Eds.), Culture, kin, and cognition in Oceania Essays in honor of Ward H. Goodenough (pp. 79-94). Washington, DC: American Anthropological Association.

Anonymous 1915 Some remarks about the religious views of our Islanders. (Comments recorded by an unidentified missionary). Available at the University of Guam, R. F. Kennedy Memorial Library, Micronesian Resource File.

Barnett, H. 1949 Palauan society: A study of contemporary native life in the Palau Islands. Eugene: University of Oregon Publications.

Black, P. 1985 Ghosts, gossip, and suicide: Meaning and action in Tobian folk psychology. In G. White & J. Kirkpatrick (Eds.), Person, self and experience: Exploring Pacific ethnopsychologies (pp. 245-296). Berkeley University of California Press.

Bollig, L. 1967 The inhabitants of the Truk Islands Religion, life and a short grammar of a Micronesian people. New Haven, CT: Human Relations Area Files. (Original work published 1927)

Bourguignon, E. 1968 World distribution and patterns of possession states. In R. Prince (Ed.), Trance and possession states (pp. 3-34). Montreal: R. M. Bucke Memorial Society.

Bourguignon, E. 1974 Psychological anthropology. In J. J. Honigmann (Ed.), Handbook of social and cultural anthropology (pp. 1073-1118). New York: Rand McNally.

Bourguignon, E. 1991 Possession. Prospect Heights, IL: Waveland Press.

Cantova, J. A. 1722 Unpublished letter, dated March 20, 1772, from Fr. John Anthony Cantova of the Society of Jesus to Rev. Fr. William Daubenton of the same Society. Available at the University of Guam, R. F. Kennedy Memorial Library, Micronesian Resource File.

Dobbin, J. D. 1978 Religion and cultural identity. Caribbean Issues 4, 71-84.

Dobbin, J. D. 1986 The jombee dance of Montserrat. Columbus: Ohio State University Press.

Foulks, E. F. 1972 The Arctic hysterias. Washington, DC: American Anthropological Association.

Girschner, M. 1912-13. The Caroline island Namoluk and its inhabitants. Available at Micronesian Seminar, Pohnpei, FSM.

Gladwin, T., & Sarason, S. B. 1953 Truk: Man in paradise (Viking Fund Publications in Anthropology, No. 20). New York: Wenner-Gren Foundation for Anthropological Research.

Goodenough, W. H. 1963 Cooperation in change: An anthropological approach to community development. New York: Russell Sage Foundation.

Henney, J. H. 1968 Spirit possession, belief and trance behavior in a religious group in St. Vincent, British West Indies. Unpublished doctoral dissertation, Ohio State University, Columbus.

Hezel, F. X. 1985 The anthropologist and social problems in Micronesia. In Reflections on Micronesia: Collected papers of Father Francis X. Hezel, SJ (Working Papers Series, Pacific Island Studies, pp. 101-112). Honolulu: University of Hawai'i.

Hezel, F. X. 1987 Truk suicide epidemic and social change. Human Organization 4, 283-291.

Hezel, F. X. 1989 Suicide and the Micronesian family. The Contemporary Pacific 1, 43-74.

Hezel, F. X. 1991 The Catholic Church in Micronesia Chicago: Micronesian Seminar, Loyola University Press.

Kleinman, A. 1980 Patients and healers in the context of culture. Berkeley: University of California Press.

Kleinman, A. 1986 Social origins of distress and disease. New Haven: Yale University Press.

Kubary, J. S. 1878-79 The inhabitants of the Mortlock Islands. Available at Micronesian Seminar, Pohnpei, FSM.

Lanternari, V. 1965 The religions of the oppressed. New York: Mentor Books.

LeBar, F. M. 1964 The material culture of Truk (Yale University Publications in Anthropology, No. 68). New Haven: Yale University Press.

Lessa, W. A. 1987 Micronesian religions: An overview. In M. Eliade (Ed.), The encyclopedia of religion (Vol. 9, pp. 498-505). New York: Macmillan.

Lewis, l. M. 1971 Ecstatic religion. Harmondsworth, Middlesex: Penguin Books.

Ludwig, A. M. 1968 Altered states of consciousness. In R. Prince (Ed.), Trance and possession states (pp. 69-96). Montreal: R. M. Bucke Memorial Society.

Lutz, C. 1988 Unnatural emotions: Everyday sentiments on a Micronesian atoll and their challenge to Western theory. Chicago: University of Chicago Press.

Mahony, F. J. 1969 A Trukese theory of medicine. Unpublished doctoral dissertation, Stanford University, Stanford, CA .

Marshall, M. 1979 Weekend warriors: Alcohol in a Micronesian culture. Mountain View, CA: Mayfield Publishing.

Metraux, A. 1972 Voodoo in Haiti. New York: Schocken Books.

OPS/Office of Planning and Statistics. 1992 Chuuk State 1989 census of population and housing. Pohnpei, FSM.

Parsons, C. D. (Ed.). 1985 Healing practices in the South Pacific. Honolulu: University of Hawai'i Press.

Prince, R. (Ed.). 1968 Trance and possession states. Montreal: R. M. Bucke Memorial Society.

Rubinstein, D. H. 1995 Love and suffering: Adolescent socialization and suicide in Micronesia. The Contemporary Pacific, 7, 26-53.

Spiro, M. E. 1950 The problem of aggression in a South Sea culture. Unpublished doctoral dissertation, Northwestern University, Evanston, IL.

Spiro, M. E. 1965 Religious systems as culturally constituted defense mechanisms. In M. E. Spiro (Ed.), Context and meaning in cultural anthropology: Papers in honor of A. L Hallowell (pp. 100-113). Glencoe, IL: Free Press.

Spiro, M. E. 1972 Ghosts, Ifaluk, and teleological functionalism. In W. A. Lessa & E. Z. Vogt (Eds.), Reader in comparative religion (pp. 479-482). New York: Harper & Row.

Turner, V. W. 1974 Dramas, fields and metaphors. Ithaca, NY: Cornell University Press.

Turner, V. W., & Turner, E. 1978 Image and symbol in Christian culture. New York: Columbia University Press.

Van der Walde, P. H. 1968 Trance states and ego psychology. In R. Prince (Ed.), Trance and possession states (pp. 57-68). Montreal: R. M. Bucke Memorial Society.

Wedenoja, W. 1990 Ritual trance and catharsis: A psychobiological and evolutionary perspective. In D. K. Jordan & M. J. Swartz (Eds.), Personality and the cultural construction of society (pp. 275-307). Tuscaloosa: The University of Alabama Press.

Yaoch, F. K. 1966 A reflection on Modekngei: The Palauan pagan religion. Available at the University of Guam, R. F. Kennedy Memorial Library, Micronesian Resource File.