MicSem Publications
Cultural Patterns in Trukese Sucide
by Francis X. Hezel, SJ
1984 Suicide
Truk, an island group with a population of 40,000 and situated in the geographical center of the Trust Territory of the Pacific Islands, has been experiencing what may be called an epidemic of suicide in recent years. Suicide in Micronesia first began to attract attention in 1976 following a sharp increase in the number of deaths and the discovery that suicide had become the leading cause of death among young males of the 15-30 age cohort. My publications on suicide in Micronesia (Hezel 1976; 1977), which represented the first, crude attempt to analyze the circumstances and causes of self-destruction by Micronesian youth, were soon followed by further research. Rubinstein began a study of the problem in I 979 while a post-doctoral research fellow at the University of Hawaii, and has published several articles on Micronesian suicides (Rubinstein 1980, 1981a, 1981b, 1982, and 1983). In April 1983 he initiated a three year NIMH-funded ethnographic study of a single Trukese community to isolate factors leading to stress and suicide.
My early efforts to publicize the problem of youth suicide in Truk attracted much more attention from outsiders than from members of the Trukese community, since the latter regarded suicide as a problem that was endemic to their society (White 1982: 42-4). Their contention was that Trukese, like many of their Pacific neighbors, have always committed suicide, so there was no reason to become alarmed at what was simply the manifestation of an old cultural trait. There is some evidence to support this sentiment. Protestant missionaries in Truk around the turn of the century recorded several hangings, two of them by wives of their own native teachers who were thought to be "homesick" (ABCFM: Logan to Smith, 3 Mar 1899; Logan 1888: 45-6). At the mission boarding school on Moen there were six hangings within a period of two years, four of them unsuccessful and two ending in death. The school directress found it necessary to train a squad of boys to cut down victims in future emergencies (ABCFM: Logan to sister, IO Mar 1899). The story is much the same in the other island groups of Micronesia. The Augustinian Recoleto priests Ibanez del Carmen and Resano (1976) recorded eighteen suicides in Guam during the 35-year period, 1857-1891, while American Board missionaries working on Ponape during the same period testified to spates of suicide from time to time (ABCFM: Sturges, 26 Oct 1858). Wilkes (1845), the commander of the celebrated US naval exploratory expedition to the Pacific in the early 1840s, noted the frequency of suicide attempts among the people of the Gilberts. His (Wilkes 1845 v5:107) observations on this phenomenon, which ring as true today as they did then, are worth quoting:
(Gilbertese) are said to be subject to despondency and sullenness, that sometimes causes them to commit suicide…. To terminate their lives they always resort to hanging on a tree. The motive of this act is generally the treatment they have received, or offence taken at the conduct of some person, whom affection or fear renders them unwilling to injure, the mortification and grief produced thereby leads them at last to suicide, which is considered by them as a remedy for their evils, as well as a severe revenge upon those who had ill-treated them.
Indisputably, then, suicide is endemic to Truk, as it is to the rest of Micronesia and quite possibly much of the Pacific beyond.
Yet, there is strong evidence that suicide in Truk, as in other parts of Micronesia, has an epidemic quality as well. The data collected on suicides prior to 1970, although not absolutely conclusive, suggest strongly that the suicide rate has risen over the past twenty years, but we do not know to what extent this apparent epidemic increase is owing to the rapid cultural change that Truk has been undergoing during the same period. For the present, we can do no more than hypothesize on the part that social change plays in the high suicide figures of late; any conclusions must await further research on social factors in the community.
The aim of this article is to describe the pattern of suicide in Truk as it has emerged in recent years, to present a basic typology of suicide, and to indicate its cultural significance. Overall, I hope to show how suicide serves as a culturally patterned response in Trukese society. An understanding of the pattern of Trukese suicides could very well have relevance far beyond Truk. The afffinity between suicides throughout Micronesia has already been noted. Moreover, the startling climb in suicides in Western Samoa shows striking similarities to that in Truk; the age-sex group affected is the same, as is the etiology and circumstances of death. A clear view of suicide in Truk, then, may further our understanding of the cultural patterns of suicide elsewhere in the Pacific and assist us in making a response to this problem.
GENERAL CHARACTERISTICS OF TRUK SUICIDE
Collection of data on Truk suicides has been carried on more or less continuously since late 1976 by the Micronesian Seminar, a church-sponsored research institute in the Caroline and Marshall Islands. Donald Rubinstein (see Bibliography) has contributed greatly to this effort by conducting an epidemiological survey in 1979-1981 under the auspices of the University of Hawaii and in collaboration with the Micronesian Seminar. Police reports, hospital records, death certificates, and pertinent church records were exarnined, as were the case materials and papers of psychiatrists at the Trust Territory Department of Health Services in Saipan. Since public records in Micronesia are always spotty, and a sizable number of deaths in outlying areas are never reported to government authorities, we regularly canvassed various segments of the population for information on recent suicides in their area. Periodic checks for names and dates of suicides were followed up with longer interviews with friends and relatives of suicide victims, and often by still further interviews to cross-check the reliability of the information received. Hence, the average case file is based on reports by three or four different informants, with the information often received at different times. These techniques insured that our data were much more thorough and dependable than would have been the case if we were forced to rely on offficial records alone.
From 1971 to 1983 inclusive there have been 129 recorded suicides in Truk, for an average of ten a year, or a rate of about 30 per 100,000 (Table I). Almost half of the suicides took place during the four-year period between 1979 and 1982, a period that showed an average of over fifteen suicides a year. This number is all the more impressive when we consider that there were only thirteen recorded suicides for the entire ten-year period of 1961-1970, the years immediately preceding those covered by this study. Suicide for the 1960s is almost certainly under-reported since accurate data collection began only in I976 and we can expect informants' recall about earlier events to diminish. Nonetheless, it is most probable that the true suicide rate for the 1960s was nowhere near as high as that of the past decade.
Table 1: Trukese Suicides by Year and Sex
| Male | Female | Total | |
| 1971 | 4 | 0 | 4 |
| 1972 | 4 | 1 | 5 |
| 1973 | 3 | 0 | 3 |
| 1974 | 8 | 0 | 8 |
| 1975 | 15 | 0 | 15 |
| 1976 | 5 | 0 | 5 |
| 1977 | 10 | 0 | 10 |
| 1978 | 6 | 1 | 7 |
| 1979 | 15 | 2 | 17 |
| 1980 | 16 | 1 | 17 |
| 1981 | 13 | 3 | 16 |
| 1982 | 11 | 1 | 12 |
| 1983 | 9 | 1 | 10 |
| Total | 119 | 10 | 129 |
Suicides in Truk occur predominantly among young males between the ages of 15 and 30. Ninety-three per cent of the victims are males (although females are better represented in attempts than in completed suicides); about 70 per cent of the successful suicides fall within that age bracket, and another 10 per cent take place in the 10-15 age cohort (Table 2). The median age for victims is nineteen. This is not surprising because in Micronesian societies it is the young who are subject to the greatest stress, and men more so than women since, among other things, their roles are much more fluid and subject to change than are women's. The rate for Trukese males of this age group (15-30) is a shocking 200 per 100,000, about ten times the suicide rate for American youths of the same age.
The method of suicide used in the overwhelming majority of cases is hanging, but the victim often simply leans into the rope until he passes out and dies of anoxia. There are a few cases of death by ingestion of a toxic substance such as kerosene, clorox, poison or an overdose of drugs; a method preferred by women. In more recent years, there have been a few cases of death by firearms or explosives, but these are still quite uncommon.
TABLE 2: Trukese Suicides by Age and Sex
| Male | Female | Total | |
| 10-14 | 11 | 0 | 11 |
| 15-19 | 37 | 4 | 41 |
| 20-24 | 35 | 2 | 37 |
| 25-29 | 17 | 0 | 17 |
| 30-34 | 4 | 2 | 6 |
| 35-39 | 2 | 0 | 2 |
| 40-44 | 2 | 1 | 3 |
| 45-49 | 2 | 0 | 2 |
| 50-54 | 2 | 0 | 2 |
| 55-59 | 0 | 0 | 0 |
| 60+ | 7 | 1 | 8 |
| Total | 119 | 10 | 129 |
Suicide is frequently accompanied by heavy consumption of alcohol; about half of the victims were drinking just before their death. Still, it can be argued that alcohol use is as much the result of their previous decision to die as it actually conditions this choice in other words, young people drink in order to die as well as die because they drink. Heavy alcohol consumption is often used by Trukese as a means of expressing sentiments and performing actions that they would otherwise lack the boldness and resoluteness to express. We might add that increasingly in the last few years marijuana use occurs before the suicide.
Suicides in Truk generally are triggered by an apparently trivial disagreement or quarrel, as when a young man is scolded or refused some minor request by his parents or older sibling. One 16-year old boy hanged himself when his parents refused to buy him a new shirt for Christmas; another young man of 23 killed himself after his older sister would not give him the yeast he demanded for a drinking party; a 17-year old boy, active in church groups, hanged himself after being scolded by his older brother for singing too loudly in the house. Most often the suicide is precipitated by conflict with a parent or older relative, but in a few cases a spouse has been the cause of a suicide, as when a 60-year old man hanged himself after complaining that his wife was neglecting to care for him.
The suicide act is usually explained by Trukese themselves as motivated by anger, particularly towards those to whom negative feelings should not be shown; but the emotions of shame and perhaps frustration are involved as well. There are other cases that can best be explained by fear, even panic, while a few rare cases display a surprising altruism. In any event, very few of the suicides in Truk are rooted in anything that we would call a serious mental disorder; there is almost none of the depression that so often precedes suicide in Western countries, and psychosis plays a part in no more than a handful of suicides.
Another significant difference between Trukese suicide and that in the Western world lies in what occasions this decisive act. Trukese suicides are never brought on by failure in business or school, sudden loss of a job or social position, or other blows to one's sense of achievement, as is commonly the case in Western and some Asian societies. Instead, Trukese suicides are occasioned by what is perceived to be disruption in a significant interpersonal relationship, generally with a family member or members holding authority over the victim. This reflects a value hierarchy among the Trukese (and other Micronesians) that is at great variance with that in most Western countries, with Trukese according such enormous importance to the quality of intra-familial relationships that they would prefer to destroy themselves than to see these bonds ruptured.
SUICIDE: IMPULSIVE OR REASONED RESPONSE?
To speak of the rupture of the familial bond because the father refuses to buy his son a shirt or give him a dollar to play pool would seem to be ludicrous. Could the young suicide victim have really thought that he was being definitely rejected by his parents? This gave rise to the question that we faced in attempting to analyze our case data: Is the typical act of Trukese suicide an impulsive gesture with little regard for social reality, or is it a response to what might be called, at least from the victim's perspective, an unhappy and perhaps even oppressive family situation? At first, all the evidence seemed to favor maintaining that the act was an impulsive one, perhaps even pathologically so. The trivial nature of the precipitating incident appeared thoroughly disproportionate to the act that followed it. Moreover, there were usually no readily perceptible attempts to communicate warnings to parents or other offending parties before the suicide. Hence, suicide looked like the response of a young man to a single isolated reproof, something that was almost by definition attributable to impulsivity. If this were the case, then questions naturally would have to be raised about the psychological stability of the agent. Murphy (1982) argues that the suicides might be the product of a narcissistic personality, a type that might have proliferated in the islands because of pathogenic child-rearing techniques among some women (Murphy personal communication; and 1982:167-8). Others looked for other kinds of personality disorders to explain what seems to be an irrational and unintelligible act of the worst sort.
But Micronesians themselves strongly assert that behind most of these single trivial incidents that immediately preceded suicides lies a tale of long-standing family tensions and conflicts. As we accumulated more data on our cases, evidence began mounting that there was indeed a history of troubled family relations behind many of the suicides. This is best illustrated through an actual example and the different interpretations that we put on it as additional information became available to us. We use the case of Pedro, a 23-year old married man who hanged himself near his house in a moderately populated village.
The first version of the suicide stated that Pedro's wife had run off a short time before and that he hanged himself because he could not persuade her to return. This account strained our credulity even then, because a Trukese will almost never kill himself for the loss of a lover or spouse.
In the second account Pedro had gone to his parents to ask for their assistance in getting his wife to return, but they refused to help. In this explanation Pedro's anger was directed not towards his wife, but towards his parents for their failure to support him when he called on them. This is consistent with the cultural pattern of suicides in Truk, but Pedro's suicide would still have to be regarded as impulsive.
As new information came to light, it became clear that Pedro had been criticized by his parents for several months because he insisted on devoting a major portion of time to food-gathering for other members of the extended family. In an act to protest against what he saw as his parents' insensitivity, Pedro himself sent away his wife, a woman originally chosen for him by his parents. This act can be seen as a strong signal to his parents to convey to them how seriously he was offended by their behavior. It was a few weeks after this warning that Pedro finally took his life.
Although we lack precise statistical information on this point, it appears that many of the suicide victims have experienced stormy relations with their family for months or even years. Pedro had made an earlier suicide attempt six or seven years before his death, suggesting long-standing conflict with his family. A 16year old who hanged himself after his grandmother warned him not to use his father's possessions had been arguing with her virtually from the day his grandmother assumed care for him a year before. An even younger boy hanged himself after his father ordered him to do something that he did not want to do, but the father had provoked growing resentment in his son through his insults and beatings. An 18-year old girl took an overdose of Darvon after being scolded by her older sister, but she had also been hurt by the separation of her parents and the remarriage of her mother some months before. Another victim, an 18-year old boy who wrote a letter just before his death stating that he was tired of living, had been unhappy for some time with his mother's new spouse and the man's treatment of him and his brothers. In all these cases, the rebuff that triggers the suicide is the most visible because it is the last of the quarrels that have occurred over a fairly long period of time. It simply provides the occasion for the victim's angry outburst against all the misunderstandings and reproofs directed at him.
This is not to deny, however, that an element of impulsivity exists in all suicides, and that in some this element looms quite large. Impulsive suicides do happen, especially in the case of young people in their early or middle teens, but they are not as frequent as a quick survey of the case data might lead one to think. Given the very young age of most victims who take their lives over a single incident, and the importance of perceived family support for the Trukese youth, we need not have recourse to pathology to explain even the more precipitous suicides in this culture.
ANGRY SUICIDES: THE TRUKESE PATTERN
Of the 159 Truk suicides on which we have amassed case data, 229 occurred in the period of 1971-1983, the time frame that we are considering in this article. There is adequate information on the immediate cause of suicide in 96 of the 129 cases to enable us to use them in our typology of Trukese suicide. The depth of our data on these cases varies considerably, and it is always possible that new information on some of the cases may necessitate a revised interpretation, as was illustrated in the case of Pedro. Nonetheless, it is very unlikely that additional case information will invalidate either the major categories of suicide proposed below or the main conclusions that follow.
There is a strong pattern that emerges in the vast majority of Trukese suicides; it appears in 74 of the 96 cases with sufficient detail to form conclusions on the reason for suicide (Table 3). This pattern, which forms the basis of the description given above, follows these lines; the victim, usually a young man and often intoxicated, hangs himself after he is scolded, refused a request, or otherwise rebuffed by parents or an older sibling. The motive that is almost universally attributed to the victim by family and friends, as well as by those who have survived suicide attempts themselves, is anger rather than shame or any other emotion. This pattern, it should be noted, is so well recognized by Trukese themselves that any suicide which can not be explained in this way evokes a reaction of surprise and concern from local people. If there is anything that can be regarded as a folk explanation of suicide, this is it.
TABLE 3: Trukese Suicides by Type and Sex
| Male | Female | Total | |
| Anger suicides | 66 | 8 | 74 |
| Shame suicides | 14 | 1 | 15 |
| Psychotic suicides | 6 | 1 | 7 |
| Undetermined | 33 | 0 | 33 |
| Total | 119 | 10 | 129 |
The anger that is expressed through the suicide is what American mental health personnel in Micronesia rightly have termed "retroflective anger"–anger turned inward by the victims upon themselves. Where they may have erred, however, was in assuming that this was a pathological mode of expression instead of an exigency arising from a culture that prohibits the venting of strong negative feelings towards parents and older kin. In Truk, one does not display such feelings towards one who is superior in terms of age and social status, particularly towards one with a special claim to love and respect besides. Certain liberties may be taken when a person is intoxicated–or at least is acknowledged as such by others, whatever his physiological condition at the time–but even here there are recognized limits to the nature of the display that is tolerated. Drunken youths often smash windows and do damage to property in their rage, but they may not with impunity insult, much less do physical violence to, parents or older brothers. The danger that alcohol presents to the angry young man is that as he begins to vent his grievances up to the point that he is permitted to do so, the feelings intensify rather than abate and the young man is driven to some other, more extreme, form of expression. This escalation will almost invariably lead him to harm himself rather than those who are the object of his anger, a course that often leads to self-destruction.
Cultural restrictions on the expression of anger are extremely important in understanding Trukese suicide and its motives. For a young man to commit suicide because of a quarrel with a younger brother is virtually unheard of in Truk, because the young man could deal with the problem much more directly by simply thrashing his younger brother or verbally abusing him. Men would normally deal with recalcitrant wives in the same way, although there are a few exceptional cases (six in all) in which the husband has reputedly killed himself following an argument with a wife. Interestingly, we might add that there is no recorded case of a wife committing suicide because of a disagreement with her husband. A common enough cause of suicide is the refusal of the victim's parents to approve his choice of a marriage partner; but this is motivated by anger at his parents rather than the thought of a life apart from the young lady in question. A young man would never entertain the thought of suicide if his marriage plans were frustrated by the girl's family; he would adopt less introverted means of displaying his displeasure.
In the eyes of Trukese, there are very few significant others whose rejection, real or imagined, would prompt a person to commit suicide. For the young man they would include classificatory older brothers and parents in his direct line or perhaps once removed. Other distant members of the family normally could not precipitate a suicide by scolding or rebuking him. This is all the more true of those not related by blood or affinal ties, even though they hold a high status in the community. Hence, the school teacher, pastor, village chief, or mayor could not impel a young man to suicide no matter how severely they chastised him.
It is tempting for a Western observer to carry the idea of retroflective anger a step further and to interpret suicide as a vengeful, defiant strike by the victim against his family. In this view, the offended young man, who lacks any other means of expressing his anger at his parents, lashes out at them by the very act of self-destruction. His strategy would be to get at them by hurting himself. Yet, the twenty Trukese survivors of suicide attempts we interviewed seemed genuinely shocked at the attribution of a vindictive motive to their act and vigorously denied any such revenge motive. The family and friends of those who have taken their lives also seem to reject this interpretation, not just by their denial of its truth, but by their unfeigned show of grief and love at the funeral of the victim. This would be difficult for them to do if they regarded the suicide as a decisive retaliatory act directed against them. Accordingly, any psychological profile of the suicidal act that took serious account of Trukese beliefs on the matter would have to see the suicide as initially motivated by anger, but avoid seeing the act as a vengeful response aimed at the parents.
We are often misled in our attempt to make sense of Trukese suicide by the word "anger." Although Trukese themselves use the single word song to describe all kinds of anger, we might do well to nuance our understanding of the emotion that comes into play in suicide. The refusal or rebuff of parents will evoke feelings that cover a fairly large range of the emotional spectrum; shame at being taken so lightly by those who have a special obligation towards him, frustration at not receiving the treatment that he felt was due him, dejection at the rift that has grown between himself and his parents, and anger at the injustice of it all. There is invariably a good measure of self-pity operative as well. The anger that Trukese speak of in connection with suicide, then, is not so much the fury of raised fists and clenched teeth; it is a more peevish and plaintive sort of anger, as anyone knows who has witnessed an exchange between family members.
SUICIDE AS 'AMWUNUMWUN'
In a situation like that which usually precedes suicide, one in which a person is hurt and angered by someone he loves and respects, Trukese commonly use a strategy that they call amwunumwun. The refusal of a boy to eat when his parents have offended him is an example of amwunumwun; another is the openly promiscuous conduct of a girl who, in anger at her parents' conduct, "goes to the beach" to signal her feelings. A young man who wishes to protest some demand of his father–such as, in one actual case, his request to enjoy sexual relations with his son's wife may stab himself in the arm or inflict other injury upon himself. Amwunumwun, therefore, is a strategy of withdrawal or self-abasement used to show those one must both love and obey that one is hurt by them. The act of amwunumwun is intended not principally to inflict revenge–although it would be naive to maintain that there is nothing of this in the act–but to dramatize one's anger, frustration and sorrow in the hope that the present unhappy situation will soon be remedied. If the one who employs amwunumwun is trying to shame the one who has offended him it is always with the intention of showing the offending party the sad state into which their relationship has fallen so that he will take steps to restore it to what it once was or should have been.
Suicide, in the overwhelming majority of Trukese cases, and quite possibly those in other parts of Micronesia as well, must be understood as a kind of amwunumwun. Indeed, it is the extreme form of amwunumwun since it means inflicting the ultimate harm upon oneself in order to compel the parents or others to recognize the harm they have done and to repair it. This type of suicide can be, paradoxically, a gesture of both despair and hope at one and the same time. To the extent that the suicide represents a considered judgment rather than simply an impulsive act, it is a sign of despair at the possibility of retrieving a central relationship here and now. On the other hand, the suicide also exhibits some optimism that the relationship, even if doomed as long as the victim lives, might be salvaged on the other side of the grave. We might note in passing that this strategy does not imply a belief in the after-life (although virtually all Trukese are at least nominally Christian); it merely rests upon the survival of one's memory among the living. The victim's life may have been a constant struggle with his family while he was with them, but his removal from the family through death will change everything and guarantee that he is restored to his former rightful place in their eyes. Far from being a vindictive act in which the victim takes his last revenge, suicide presupposes a continuing regard for the family and an acknowledgment of the preciousness of family bonds.
This interpretation may appear both gratuitous and fanciful, but there are some peculiar notes to Trukese suicide that support it. In the first place, a significant number of the youths who have taken their own lives were quiet and dutiful sons who have not been known as trouble-makers in the community and who have had no record of police arrests. Although no statistical data are yet available, the biographical information hitherto collected indicates that perhaps 20 to 25 per cent of the victims are what informants would call "good boys." Whatever their number may be, it is clear that suicide is by no means the exclusive device of deviants. Moreover, young suicide victims, whether they are well behaved or not, often manifest what might appear to be a surprising concern for their place in the family after their death. Pedro gave explicit directions that he was to be buried with his head facing the road; a few hours before he went out to hang himself, he made an imprint of his foot in the wet cement on the floor of the lineage house they were building, and he wrote his name above the footprint with the remark, "They will remember me by this always." Another young man, without any trace of facetiousness, told his grandmother before his suicide to remember him the next day as the family ate sardines (at the funeral feast). Two boys, on different occasions and for different reasons, hanged themselves in such a way as to pay tribute to their respective bond brothers, each of whom was himself a victim of suicide. One of them hanged himself on the grave of his "brother" the evening before the big memorial celebration dictated by Trukese custom, and the other was found dead with a photo of his "brother" in his hands exactly a year and a half after the latter's suicide. Each of the two was virtually assured of winning a permanent place in the affection of his adoptive family by advertising his love for the dead son and by timing this display for maximum effect; henceforth, they must have hoped, it would be impossible for the family to recall their dead son without also fondly remembering their dead son's devoted bond brother.
The suicidal form of amwunumwun, as we have described it above, is employed, in part, to achieve a reconciliation between the victim and his family after the death of the former. There are, however, at least two cases in which the suicide was used as a means of reconciling not principally the victim to his family, but two quarreling members of the victim's family with one another. A few years ago a 22-year old boy, disturbed at the chronic political disputes between his father and his paternal uncles, hanged himself in protest. Two years later, an 18-year old youth who was living in another part of Micronesia, committed suicide when trouble broke out between his father and his paternal grandfather and they ignored his repeated pleas to settle their differences. In both incidents, the victims might well have taken personal offense at the unseemly dissension in their family and the unwillingness of all parties to find a peaceful solution. This seems more likely than to interpret these suicides as entirely altruistic. Nonetheless, the victims probably consciously sought to use their death to reconcile the quarreling factions, for funerals furnish an excellent mechanism for resolving such conflicts because of the long and open exchanges between family members that occur at these times.
While focusing on the positive elements of suicidal amwunumwun, we should not forget that this is essentially an act of anger. Although it may imply devotion and love in the best traditional sense, it is also an act of self-destruction. As such, it should be the last desperate recourse in the repertoire of the young man who is hurt by his family. Yet, other milder forms of amwunumwun seem to be increasingly ignored by today's youth in favor of immediate recourse to suicide. Why this escalation has occurred is a question that cannot be answered here, but can only be answered with further research.
SHAME SUICIDES: REVERSE 'AMWUNUMWUN'
A secondary pattern of suicide, that appears in 15 of the 96 cases, represents the obverse side of the amwunumwun type. Here again the etiology can be traced to the perception, whether well founded or not, that a critical relationship is threatened. In this pattern, however, the victim does not see himself as the offended party, as in the amwunumwun form of suicide, but as the offender. His suicide, which is motivated principally by fear or shame rather than anger, is a response to the realization that he has done something that will have driven a wedge between himself and his family. The victim's strategy, in keeping with the common tendency throughout Micronesia and Polynesia, is to withdraw from a conflict-laden situation. This means of resolution is especially favored when the conflict occurs between members of the same family.
A striking example of this pattern occurred some years ago when a young man in his mid-20s shot himself in the presence of his older brother because he suspected that a crime he had committed some months before had been discovered by his older brother. The timing of the suicide is significant here, it took place long after the original crime but only minutes after the victim heard his brother being informed of his misdeeds. Another young man, who beat up his wife and mother-in-law when they embarrassed him by telling him in public that they did not have enough money for their movie tickets, hanged himself soon afterwards in shame at what he had done. Another young man, this one in his late teens, was so ashamed of the foul language he had shouted out on the road the night before while he was drunk that he instructed his mother to apologize to anyone he may have offended and hanged himself. A 24-year old youth also hanged himself under similar circumstances when his friends told him after a drinking party on the previous day he had cursed his brother-in-law.
Perhaps the clearest illustrations, however, are those that involve unacceptable sexual conduct. There are four persons, all of them in their late teens or early twenties, who hanged themselves shortly after their incestuous love affairs came to light. Each of them was involved in a sexual relationship with a first cousin, and each was presumably ashamed at the public disgrace for them and their families that would follow the disclosure of their relationship. Some years ago, a married woman who had fallen in love and run off with her daughter's fiance, committed suicide rather than face the shame of confronting the daughter that she had injured. In a more recent case that followed the same lines, a young man who had been having sexual relations with his older brother's wife killed himself as soon as he had reason to suspect that his brother knew about the affair.
To attribute these suicides to a sense of personal shame is far too broad an explanation. Trukese can be shamed in the classroom by their teachers, in their work situation by their employers, or in public meetings by another villager, but they will not commit suicide in response to this shame. What is at issue here is not the disapproval that they can expect from the community for their wrongdoing, but the effect of this disapproval on their family. The young Trukese can live with the fact that he is the subject of public opprobrium in the community, but he cannot at all live easily with the feeling that he is the cause of disgrace for his family. The young man who is known to have been having a sexual relationship with a woman classified as his "sister" has to contend with the damage that he has done to his family's reputation in the eyes of the community, apart from the friction within family circles that his liaison might have touched off. But beyond this, he has reason to fear that he will not be trusted again by the family and that his relationship with one or several of his close relatives will be seriously harmed by the episode.
Under these circumstances, the offender might well choose to remove himself from the scene of the disgrace as a means of repairing to some extent the damage he has done to his family. This is a common theme in the interpretations that Trukese place on such suicides as these. The victim probably sees his suicide less as an atonement for his sins than as a way of ridding the family of a presence that, in his judgment, would be a constant irritant. With his death he can hope that relations within the family would again become tranquil and that outsiders would have far less reason to continue deriding the family for his misdeeds.
Within this pattern there is a striking subgroup of suicides by very young boys that deserves some comment. A 16-year old boy who had given his uncle's tape recorder to another youth to fix, and could not get the machine back despite repeated requests, hanged himself for fear of his uncle's wrath. Two 12-year old boys, both from the same island, committed suicide two years apart for fear of the punishment they would suffer; one boy had vandalized a classroom with a group of friends, and the other had hit an older relative on the head with a srone. An even younger boy hanged himself after being teased by schoolmates about an incestuous relationship in his family. Five of the suicide victims were I6 or younger. These especially young and inexperienced boys may have acted out of panic at the consequences–beatings or scolding or both–that they imagined would befall them for what they had done. Of importance here, however, is that among the consequences they presumably feared was disruption in normal relations between them and their father, uncle or other close relative. As impulsive and fear-motivated as these suicides might appear to be, they probably belong in this type.
PSYCHOTIC SUICIDES
There is one last type of Trukese suicide committed by a small and easily identified group of mentally disturbed people. Within the last four years, there have been seven suicides by young people who have been certified as psychotic by state mental health authorities. Most of these victims were diagnosed as schizophrenics, with five of them receiving antipsychotic medication sometime in the few years prior to death. The symptoms that they showed were generally the classical symptoms for this disorder; auditory hallucinations, dissociation in speech, withdrawal and often acute paranoia. One other suicide victim was said to be suffering from clinical depression for the year before his death. Each of these victims was recognized by the community as mentally aberrant, and their suicides were attributed to their mental condition.
The bio-data collected on these seven persons support the judgment of the community that their death was due to the mental affliction from which they suffered. Virtually all of them had shown suicidal ideation and most had made previous attempts to kill themselves, one of them on six occasions. Three of them had become angry at members of their family before their deaths–one at his older brother because the latter did not allow him to sleep with his wife, another at his older brother for more prosaic reasons, and the third (a man in his 60s) at his family for not taking better care of him in the hospital. Although the suicides of these three followed the amwunumwun pattern, they and the other victims in this group were said to have showed, even by Trukese standards, hypersensitivity in their judgments towards close family members. This undoubtedly stems from the paranoia that was a nearly universal symptom in their cases.
One of the victims, a bright and articulate young man of 30 suffering from schizophrenia, kept a diary in which he recorded the acute sufferings that emanated from his mental condition. This and the fragments of information that we have on one or two others leave no doubt that their own suicides were an attempt, during a lucid interval, to find rest from the increasingly unbearable burden of insanity. In the case of the young diary-keeper this tranquility was purchased in a particularly gruesome manner; after failing to commit suicide by bashing in his head with a steel pipe, he jumped out of the boat that was taking him to the hospital and managed to lodge his body under a reef, drowning himself.
The etiology of these psychosis-related suicides is altogether different from that of the major patterns of suicide elucidated above, even though some of the cases show strong similarities to these patterns. Psychosis is a growing problem in its own right throughout Micronesia and deserves to be studied separately. This grouping, statistically small as it is, can serve to show us that mental disorders account for a very minor percentage of the total suicides in Truk.
MODELING
These patterns of suicide, pronounced as they are, can provide us with a better understanding of the cultural context in which Trukese suicide occurs and help us identify the social meanings that it bears. Nonetheless, we must be wary of investing suicide with an overly rational explanation and positing too conscious a link between means and end, for there will always remain an irrational, even an antirational, element in all suicide, whatever the culture in which it occurs. This is particularly true of Truk, where suicides have usually been unplanned and highly impulsive. Any attempt to explain Trukese suicide, therefore, must take into account modeling and other factors that may be relevant to the impulsive nature of the act.
Clustering, both geographical and chronological, is a significant feature of Trukese sulcide. In the ten-month period between April I982 and February I983, for example, there were six suicides on Fefan, an island that until recently has had a surprisingly low rate of suicide. Yet these six suicides represent almost half of the total for Truk during this period. Each summer between I98I and I983 rhere has been an outbreak of suicides on Moen Island, the district center for Truk State, with three or four suicides occurring during each of these epidemics. In some instances the clustering is especially tight and modeling is evident from the circumstances of the suicides. Such was the case in a single village on Toloas Island in December I 975 when two suicides and one unsuccessful attempt occurred within five days. After a young man committed suicide, a friend of his hanged himself over the former's grave and a brother-in-law touched by his relative's death, tried to do the same a day later. Much the same thing happened four years later when another spate of two suicides and a suicide attempt occurred in the same village within four days, with the attempter hanging himself at the grave of a friend who had taken his life the day before. On Tol, a few months later, a suicide was followed by the attempts of two other members of the family to kill themselves, although these attempts can be traced to a family quarrel that broke out shortly after the funeral.
The frequent clustering found in Trukese suicides suggests the importance of modeling in precipitating suicide. What is true of Truk, incidentally, seems ro be equally true of Yap, Ponape and the Marshall Islands, if not the rest of Micronesia as well. Hence, it appears that exemplary behavior is another critical factor in the etiology of suicide that warrants further study.
CONCLUSION
By way of summary, we have seen that a perceived threat to a central family relationship is the key element in virtually all the suicides that occur in Truk. A blow to one of these essential relationships can bring on deep dejection and a resolve to end one's life. The most common pattern by far is that in which the victim is offended by a rebuff or refusal offered by a close relative–or, in the words of Wilkes (1845: v107) again, "offence taken at the conduct of some person, whom affection or fear renders them unwilling to injure." In anger the victim then uses suicide as a form of amwunumwun to declare his innocence and signal his hope that he will be remembered fondly after his death. A secondary and much less common pattern is that in which the victim has done some damage to his own name and that of his family and has jeopardized his relations with members of the family. In this pattern, as in the first, his death is intended as a way of partially redressing the wrong and restoring the proper relationship. In addition, there are a few psychosis-related suicides whose features bear some similarities to the two major suicide patterns. Beyond this, the importance of modeling cannot be neglected in any attempt to study the causes of suicide in Truk.
How has modernization affected the suicide rate in Truk? It appears that there has been a significant increase in the number of suicides since at least the mid 1970s, but our earlier data are not reliable enough at this point to make more than an assertion of probability. Even if the rate has risen in the last two decades, there remains the problem of determining whether this is a cyclic phenomenon, as some Trukese assert, or an increase that is unique to our day. We know from the data already in hand that there is no simple and direct relationship between modernization and suicide on Truk. There is no correlation between suicide rates and either urbanization of geographical areas or acculturation of segments of the population; indeed, some of the highest rates are found in periurban areas rather than the port towns themselves, and the rate among those who have never gone abroad to school is higher than among those who have. Yet, this should not be too surprising in view of the fact that Trukese suicide is, as we have tried to show here, a culturally patterned response to certain conflict situations. Accordingly, it is reasonable to presume that those who choose to utilize this response should retain, to a considerable degree, their traditional values and lifestyle. Suicide victims in Truk almost without exception accept the centrality of the traditional family relationships, and most of them have dramatically signified their espousal of the cultural strategy of amwunumwun through the act itself.
Despite the traditionalist cast to suicide in Truk, the common assumption that there is some link between high contemporary rates of suicide and recent modernization very likely has some truth to it. The past rwenty years of intensive change in Micronesia have eroded some of the cultural institutions that formerly aided in the socialization process, and could also have contributed to increasing tensions between the young man and his parents, while weakening the social identity of young people. The net effect of this change may have been to render the period of youth between 15 and 30, already a tension-laden stage of life, even more perilous. Yet these are merely hypothetical assertions at this point and will require validation through the research that is now being condllcted on suicide in Truk.
BIBLIOGRAPHY
American Board of Commissioners for Foreign Missions [ABCFM].
1852-1909 Letters and Papers of the American Board of Commissioners for Foreign Missions: Mission to Micronesia, 1852-1909. 19 Vols. Harvard University, Cambridge.
Hezel, F. X. 1976 Tragic End for Troubled Youth. Micronesian Reporter 24:8-13.
Ibanez del Carmen, A., and F. Resano.1976 Chronicle of the Mariana Islands. Translated and edited by M. Driver. University of Guam, Agana.
Logan, M. 1888 Last Words and Work of Reverend Robert W. Logan. Boston:ABCFM.
Murphy, H. B. M. 1982 Comparative Psychiatry. New York.
Rubinstein, D. H.1980 Micronesia's Troubled Generation. Glimpses of Micronesia and the Western Pacific 20:71-75.
White, A. N. 1982 Culture, Development and Mental Health in the American Pacific: Observations from Field Studies in Micronesia and American Samoa. Report prepared for Pacific/ Asian Mental Health Research Project, University of Hawaii.
Wilkes, C. 1845 Narrative of the United States Exploring Expedition During the Years 1838-1842. 5 Vols. Philadelphia.
Note: I gratefully acknowledge the assistance of Donald Rubinstein throughout our research and particularly the contribution he has made to the development of this paper. I am also grateful to Geoffrey White of the Culture and Communication Institute, East-West Center, for his support of our project and his critical comments on the first draft of this article.