Volcano icon

Tragedy and Trauma
in Tuvalu

The Australasian Journal of Disaster
and Trauma Studies
ISSN:  1174-4707
Volume : 2000-2

Tragedy and Trauma in Tuvalu

A.J.W. Taylor, Emeritus Professor, School of Psychology, Victoria University of Wellington, PO Box 600, Wellington, New Zealand. Email: Tony.Taylor@vuw.ac.nz
Keywords: school fire, trauma and ghosts, cross-cultural factors

AJW Taylor PhD DHC

Emeritus Professor
School of Psychology
Victoria University of Wellington
PO Box 600 Wellington
New Zealand


The key components of project that was designed to assess and reduce the trauma that prevailed after a tragic fire in a dormitory at a Secondary School in the small isolated Pacific Island State of Tuvalu are presented here, and indications are given of the outcome. They pose questions of a clinical, conceptual, and organisational kind that might arise in other cross-cultural settings.

Tragedy and Trauma in Tuvalu


At about 11 p.m. on Thursday 8th March 2000 a fire broke out in one of the dormitories for girls at the mixed-sex Motufoua Secondary School on the island of Vaitupu in Tuvalu. It was thought to have started from a naked candle that a girl held under her blanket as she was doing extra revision for a class examination that was to be held the next day. The conflagration spread rapidly and caused the Matron and 18 of the residents to be burned to death. Eighteen other girls escaped the inferno, only three of whom suffered burns. Its traumatic effects on the total Tuvalu population of about 10 000 was widespread, and said to be in proportion to the immediate loss of either 8 000 New Zealanders or 25 000 Australians in any one calamity in their respective countries. The response of the Tuvalu Government was to establish a Commission of Inquiry into the causes of the fire, and to seek the help of a trauma consultant from New Zealand in dealing with the human aspects of the aftermath.

The invitation came my way through the NZ Ministry of Foreign Affairs and Trade, together with the latest bulletin on the extent of the tragedy and a series of economic and social studies to provide the initial situational context. En route in Suva the regional office of the NZ High Commission updated the information, and the Tuvalu High Commission elaborated the country's cultural organisation and religious framework. On the spot in Funafuti, the Capital Island of Tuvalu, the relevant Ministries and Departments of Government opened the road ahead – not the least by providing a rental motor scooter to give independence of movement. They also arranged for a large patrol boat to be available for visiting the island of Vaitupu some 70 nautical miles away on which the tragedy occurred.

This paper sets out some of the issues to which the assignment gave attention. It begins with a brief description of the geographical setting, history, and economy of Tuvalu, declares the methodology adopted, and describes the modus operandi that followed. Then it covers the outreach of a scratch team of local health professionals to potential casualties, documents the response in terms of the symptomatology presented, and describes the interventions undertaken. Finally it sketches the plans that were laid for follow-up work, and touches on a few fundamental issues that the Church and the community were invited to consider concerning the appraisal of ghosts and the care of the dead, and the segregation of adolescent girls. It accepts that indigenous people have the inherent right to determine their own destinies. It also acknowledges that even if certain adjustments might seem to be indicated for bringing relief to one part of a specific cultural complex, their enactment might create unexpected problems in another that could perhaps be greater than those they were designed to remedy.

Geographical setting, history, and economy

Tuvalu is an independent nation of about 10 000 people that lies just below the equator in the western Pacific. It is situated to the south of Kiribati, to the northwest of Samoa, to the north of Fiji, and to the east of the Solomon Islands. It occupies a chain of nine tiny low-lying coral atoll islands that stretch over some 560 kilometers in a northwest to southeast direction. It has a total land area of 24.4 sq km, with the largest island (Vaitupu) having 4.9 sq km and the smallest (Niulakita) having 0.4 sq km. It has a tropical climate with uniform temperature ranging between 25 and 32 degrees Celsius, and droughts of up to three months in the summer, particularly in the northern islands. But it also has high humidity and a heavy rainfall that averages 3 064 mm per annum.

The Tuvaluans are believed to have stemmed from the Lapitans that originated in Southeast Asia about 5 000 years ago before spreading to Melanesia and settling in Samoa and Tonga in Polynesia. From there a contingent ventured to the present string of islands around 2 000 BP (O'Brien, 1983). They have a distinct language that has affinity with Samoan and Tongan, but many also have English from being taught under the British educational system.

The country's formal association with Britain began around 1819 when an American sea captain De Peysler named Funafuti the Ellice Island after the British merchant who financed the voyage and owned the cargo. In 1841 another merchant-venturer called Wilkes extended the name to cover the whole group of islands (Kofe, 1983a). In 1892 in the era of territorial expansion, Britain linked the islands formally with the Melanesian Gilbert Islands as a Protectorate, and 24 years later it proclaimed them a Colony. Then in 1926 Britain gave the Dominion of New Zealand responsibility for their administration, and in 1978 it relinquished the constitutional ties to make Tuvalu an independent State. However, this year (2000) Tuvalu sought and gained full membership of the British Commonwealth of Nations.

By mutual agreement New Zealand retained a strong link with the country, and it fostered a number of developmental programmes (New Zealand Official Development Assistance, 1999). Tuvalu welcomed such assistance because it had and still has a precarious economy that Nero (1997, pp. 441-2) described cryptically as conforming to the MIRAB model for being dependent on migration, remittances, and aid that sustains a bureaucracy. It derives income from a margin of returns from fluctuating offshore fishing catches, remittance money from Tuvaluans working abroad (mostly seamen), the sale of spare international telephone-line capacity, and from a substantial franchise agreement for the use of its Internet address that happens to have international marketing appeal. But it has set an example to other developing countries with its initiative in establishing a national trust fund that has prospered (Tuvalu Trust Fund 10th Anniversary Profile 1987-1997). Yet its isolation, extremely small size, dispersed population, and susceptibility to seasonal hurricanes, tidal waves, and storm surges limited its ability to produce and trade in goods and services. For this reason in 1986 the United Nations Organisation granted Tuvalu the status of a 'Least Developed Country' from which substantial economic benefits flowed.

But a current report shows that the country has other problems to face that also affect its economy. These include high population growth rates, deficiencies in environmental education and awareness of environmental issues, a decline in traditional resource management practices and production systems, the unsustainable use of natural resources, and problems with waste management and pollution control (Tuvalu 1997 Economic Report, 1998, p. 90). Yet the same document reporting those facts went on to acknowledge the 'innovative capacity…and strength of islands communities…(as) evident in active participation, a shared cultural outlook, religious unity, and the imperative of shared subsistence resources'.

Such positive attitudinal, behavioural, and unifying qualities had to be the focus of anyone trying to assess the response of the community to the present tragedy it faced. Obviously they would have formed part of the prevailing cultural remedies for restoring personal and community equilibrium after any catastrophe, and should be fostered in any proposed programme of interventions (Wen-Shing & Jing, 1979: Herrara, Lawson, & Sramek, 1999). They would have been shaped by the bitter experience of calamities of different kinds in the distant past, and they would have been the first lines of defence against the effects of any subsequent catastrophe. In the last 50 years the population had to face an increasing frequency of tropical cyclones, with a base of 11 in the 30 year period from 1940 to 1970, 15 in the 15 year period from 1972 to 1987, and six in the three years between 1990 and 1993. Two of these cyclones developed into hurricanes i.e. Hurricanes Bebe in October 1972 and Ofa in February 1990. But in addition, the population of the low-lying country had also to contend with the threat of the sea level rising from the effects of global warming. One official report (A situational analysis of children and women in Tuvalu 1996, 1997, p.15) lamented that the 'Tuvalu's fragile land formation as an atoll, makes recovery after a natural disaster … seem a lifetime process of rehabilitation'.

In terms of fatalities, the recent school dormitory fire with 19 victims was the biggest calamity that Tuvalu had to endure since the loss of 20 through a cholera epidemic in 1985. Previously it suffered the slaughter of an unknown number of its people among the 100 Pacific Islanders in August 1945 that had been taken by the Japanese to Ocean Island to build their fortifications (Firth, 1997). Then further back in the 1860's it suffered the abduction of 443 males by slave-traders for the mines of Peru (Maude, 1981, pp. 74-82). In the light of such experience an outside observer could be forgiven for thinking that Tuvalu was a country that had received more than its share of depredation and misfortune.


In the fulfilment of the assignment a blend of conventional clinical, applied, and qualitative methods was used, the components and combination of which required cross-cultural validation. Clinical interviews, with the help of interpreters when necessary, seemed appropriate for the elicitation of symptoms and signs of grief and trauma and their appropriate classification in the range of anxiety disorders. In determining this matter the DSM IV (American Psychiatric Association, 1994, pp. 424-432, pp. 684-685) was preferred to the ICD 10 alternative (World Health Organisation, 1990) because it offered a more systematic guide for categorising adverse reactions. However in practice it was not straightforward to compare the transient state of Acute Stress Disorder (ASD) with the more chronic condition of Posttraumatic Stress Disorder (PTSD), because the textbook did not present the categories in a way that made for easy comparison (Diagram 1). In applying them it also became necessary to question the stipulated and somewhat arbitrary time limits of the different diagnoses when the requisite symptoms were reported but were diminishing in their effects. As a consequence, the professional defence of 'clinical judgement' (DSM 1994, p. xxiii) was invoked to determine diagnoses when the presentation of symptoms fell just short of meeting the full set of criteria.

Diagram 1: showing comparison between ASD & PTSD symptom categories

A. The catastrophic event involving a response of intense fear helplessness or horror (plus fatigue) A. DITTO
B. Three or more dissociative symptoms of numbing, loss of awareness of surroundings, derealization, amnesia B: (C) DITTO plus estrangement from others, restrictedrange of feelings, considers future foreshortened , and includes symptoms of persistent avoidance below in the count
C. One or more persistent re-experience of images thoughts, dreams, illusions, flashbacks, episodes of reliving the experience, distress on exposures to reminders C: (B, D) DITTO - but persistent avoidance of three or more stimulii associated with the trauma and symptoms of numbing of general responsiveness
D. Marked avoidance of stimuli that arouse recollections, thoughts, feelings conversations, activities, places, people etc. D: (E ) - persistence of two or more symptoms of hyperarousal, sleep problems, anger, poor concentration, hypervigilance, exaggerated startle response
E. Marked symptoms of anxiety or increased arousal - restlessness E: (G) : - duration of symptoms is more than one month, and acute if >3 months chronic > 3 months, delayed onset > 6 months
F.Causes clinically significant distress or impairment either in social, occupational, or other important areas of functioning such as seeking help1 F: DITTO :
G. Duration 2 days - 4 weeks  
H. Not due to the direct physiological effect of a substance nor related to any other current or pre-existing medical condition  

# Letters in brackets in PTSD column refer to comparable blocks in ASD column - to make for easier cross-reference.

When turning from clinical causation to intervention, reservations had to be kept in mind concerning the efficacy of the procedures wherever they were applied, because as Doherty (1999) warned, imposing a 'one-size fits all model might cause more harm and ill feeling than good'. Consequently it seemed appropriate to utilise the range of conventional clinical interventions through individual and group counselling cautiously, with or without medication as prescribed in a few cases by medical members of the team. But in exercising such caution, Seligman's (1995) findings gave confidence with regard to the efficacy of short-term psychotherapy, as did Sherman's (1999) relating to the specific treatments for PTSD. Then Rose & Bisson's (1998) reservations about the use of critical incident stress debriefing for emergency workers had to be heeded - although for logistic reasons in Tuvalu the practical involvement of the Police and Fire Services in tackling the fire and handling the bodies was limited.

The applied method was dictated by the circumstances, because there were no facilities to support any experimental studies of stress - the condition being described as the substantial imbalance between demands that are made on individuals and their capacity to respond. In following that track the method adopted was in keeping with the emerging field of qualitative analysis (Hayes, 1998) in which rigor rather than rigor mortis was desired in the pursuit of answers to real life crises. It respected Aristotle's dictum that 'it is the mark of an instructed mind to rest satisfied with the degree of precision which the nature of the subject admits and not to seek exactness when only an approximation of the truth is possible'. In my case it stemmed from training and practice in social work and professional psychology, and to a long-standing clinical association with medicos in psychiatry, neurology, and endocrinology. The combination of such tutelage did much to induce me to avoid the narrower, safer, and more fashionable line that might have led to scientific respectability but practical sterility. It led me to promote a dynamic transactional model of human behaviour such as that exemplified by McFarlane & Yahuda (1996, diag. p.157) that might also meet Phillips' (1993) plea for a psychological science that was closely coupled to practice.

The question arose in my mind as to whether, like the early ethnopsychologists, I might be able to obtain the help of Tuvaluans on the spot to guide me through uncharted cultural waters. I knew that such guidance would be needed, because the concepts of psychological trauma and reactions were largely creations of the western world that could only be extended tentatively elsewhere. As deVries (1996) says, 'the very notion that PTSD occurs as a normal response to an abnormal condition implies that ordinarily people can have control over fate, is decidedly unrealistic, because there are cultures in which fatalism, witchcraft, and the violation of tapu prevail as determinants of distress'. But even when such other cultural views of causation prevail, there could be a complex variety of contributing factors to be considered that includes the conflict of values, social change, acculturation, life events, goal striving and attainment, role discrimination, and role-conflict (Marsalla, 1979: Paton, 1992).

But no matter how compelling the case for a clinician to identify and work within a relevant framework of cultural forces, it has to be said that a robust definition of the term 'culture' has yet to appear. The uninitiated might cross the threshold safely with a consideration of the anthropological and sociological viewpoints that Munch & Smelser (1992) set out. Personally, I clung to the early definition advanced by Tylor of culture being 'that complex whole which includes knowledge, belief, art, morals, law, custom, and any other capabilities that people acquire as members of society' (cited in Smelser, 1992) - while accepting that none of these components was impervious to change.

Modus operandi

On arrival in Funafuti my first step was to make formal introductions to senior personnel of the Ministries and Departments of Education and Health, and then to visit the staff at the Princess Margaret Hospital. There I was delighted to find that a small composite crew of medicos had organised itself to attend to the trauma presented by the immediate survivors. It had treated the three girls that had been referred with mild burns and had made plans to assess the psychological needs of the other 15 that were with them in the same dormitory as those that died. The crew consisted of Dr Pukena Boreham, a Tuvaluan paediatrician, Dr Stephen Homasi, a Tuvaluan public health specialist, and Dr Isaac Marks, a Burmese surgeon: all of which had long-since forgotten the preliminary orientation to psychiatry that they had in their medical training but were keen to restore. The very name they had chosen, The Motufuoa Psycho Social Revival Mission, showed that the group respected the importance of non-physical factors in health and wellbeing (cf. Stewart-Brown, 1998). There was a fourth member - Dr Vilikesa Rambukawaqa, a Fijian hospital manager - but he had to return to Suva on compassionate grounds on the very plane by which I arrived. There was nobody in the position of a social worker or psychologist on hand, but by good fortune I came across Lise Talia, a Tuvaluan teacher in Funafuti that had just returned from a two-year post-graduate course in guidance counselling and I was able to recruit her to the team.

The willing medicos had gone so far as to fashion a standard questionnaire for determining mental status along the lines suggested by Hahn, Albers, & Reist (1997), and they welcomed the handy GHQ 21, the Hopkins 20, and the Horowitz IES scale that I had brought for registering specific symptomatology. Quite rightly it thought the subtle differences in the adjectives used in the McKay Stress/Arousal Checklist would create problems of reliability even for respondents that were conversant with English as a second language. But in the end they had to proceed without the additional questionnaires because subsequently the specimens were lost in the local photocopying centre!

Then before extending its reach beyond the survivors from the one dormitory in which the fire occurred, the team held a seminar session to discuss the normal expectations of everyday life that are shattered by disasters (cf. Maslow, 1970) and the classification of disaster and casualties (Taylor, 1999). Then the discussion turned to cover the phasic reactions of grief (Kubler-Ross, 1975), the General Adaptation Syndrome (Selye, 1959), and the specific diagnostic options and differential diagnoses as per DSM IV (1994). It touched on memory for traumatic events, the unblocking of which was calculated to bring declarative memory into working memory (van der Kolk, 1996) and to reduce the levels of neuro-endocrinological excitation in the limbic system (Markowitsch, Kessler, Van Der Ven, Weber-Luxenberger, Albers, & Heiss, 1998). It made reference to the value of fostering the placebo effect at the expense of its opposite, the nocebo (see references in Benson & Stark, 1996, pp. 314-316), and to the possibility that tragedy might even induce positive outcomes (Linley, 2000). Finally it harkened to the warning of the authorities like Gordon and Wraith (1993) that in some instances adverse reactions would emerge slowly and that their diminution could be far from straightforward.

Outreach and interventions

Once on an agreed conceptual path, the augmented hospital team made arrangements for assessing anyone in the community that felt unduly troubled by the effects of the inferno, whether as relatives of the deceased, people involved in body-handling, the pupils still at school, or their parents. It decided to keep the different age-groups somewhat separate in the hope of inducing fuller participation in each. Then the team approached of the Paramount Chiefs and Presidents of the Island Councils to gain their support, obtain access to their village maneapas for meeting respondents, and have them advertise the team's plans over the local radio station. It promoted the cause further by meeting the clergy of different denominations and by accepting invitations from primary school teachers to talk about the kind of problems they might expect in their classrooms from the younger members of bereaved families.

Having attracted the clientele, my job was to introduce the team, explain its purpose, and give information in translatable bytes about the nature of trauma and the value to those affected of 'talking it out'. After that the gatherings broke into more manageable groups to allow everyone sufficient time to describe their personal reactions to the tragic fire, to reveal their underlying fears, and to disclose the symptoms they might have developed. There the doctors helped the speakers as individuals to continue the management of their grief and trauma, and they encouraged the groups as a whole to strengthen their supportive networks. They took the names and ages of those that attended, and noted pertinent contributions for further consideration and follow-up purposes if necessary. In a few cases they prescribed medication for those that were suffering unduly with hypertension and sleeplessness, and referred those with more marked psychological reactions to a group that I ran with the Student Counsellor. At the end of each session the team came together to recalibrate their clinical judgements before returning to hear the continuing stream of anguish and sorrow. Here the collective expertise and good humour of the group did much to restore the professional resilience of its members.

The response

The turnout was very high, with some 298 people between the ages of 14 and 74 presenting themselves on the two Islands of Funafuti and Vaitupu in which about 70% of the population lived (Table 1). It would have been higher had the men not assumed that the sessions were intended primarily for women! But as it was, the number represented 7.4% of the combined over 14 population living on the largest islands. The majority was grieving for the loss of one or more family members, because under the Polynesian kinship system, relatives of the same generation as the parents were regarded culturally as parents no matter how distant biologically their personal position in the extended lineage might be (Danielsson, 1956, p. 89). Similarly relatives of the same generation as the grandparents were all regarded as grandparents, and so on for relatives of the same generation as brothers and sisters. Hence the majority of the population in Tuvalu was related to the deceased, and the personal grieving was widespread.

Table 1: showing the number of Tuvaluans meeting government mission clinical team at week 4 after the event, and their diagnoses.

Number presenting
Total needing treatment/ monitoring

ASD - Acute Stress Disorder (DSM IV)
PTSD - Post Traumatic Stress Disorder (DSM IV)
OTHER - Anxiety/grief reactions
(*) indicates improving

One hundred and twenty six of the respondents were students, 36 were either teachers or other staff members of the school, and 136 were either grieving relatives of the deceased or they were worried about the response of the survivors of the tragedy. Those attending the sessions listened attentively to speakers around the circle until their turn came to express themselves, because they were from an oral tradition in which the spoken rather than the written word commanded attention and carried weight. Both the young and the old described their reactions, their perceptions, and their search for meaning to account for it. The only difference between the generations was that developmentally the younger participants were less articulate in identifying their feelings than were the older and they required the opportunity to express themselves non-verbally.

Overall the Tuvalu people displayed a typical range of traumatic reactions to the tragedy. They were not primarily of the somatic kind as was said to be typical of Fijians (private communication Dr Henry Aghanwa Senior Lecturer in Psychiatry at the Fiji School of Medicine, 18 April 2000). Nor were they as unique to their culture like those that have long been identified as being particular elsewhere (cf. Honigman, 1954). For almost everyone the search seemed much as Frankl (1984) indicated was the case for individuals wrestling with the experience of extreme trauma, and as MacFarlane and van der Kolk (1996) suggested was typical for cultural groups. Most of the survivors sought to regain a measure of personal dignity and control over their environment, and then progressively to reconstitute their familiar cultural, social, and economic institutions and activities in a way that made sense to them (cf. Summerfield, 2000). There were two exceptions – one with bland indifference and the other with complete insensitivity – and the clinical team could only speculate about their motivation for turning up to the sessions.

As to be expected, the 18 survivors that escaped the blazing dormitory had symptoms of avoidance, intrusive imagery, episodes of reliving the experience, and sleep disturbance. They had very mixed feelings about having survived while their friends in the same dormitory had been burned alive. In effecting their escape, they skirted the pile of bodies at the foot of the locked door that normally was the one for their use, and they dodged internal obstructions to reach and break through another outside door that was for the exclusive use of the matron.

On the night of the tragedy the staff of the school attended to the needs of the girls that escaped the inferno. It tried to douse the embers of the burning dormitory and then later to protect the site from marauding dogs. Villagers raced along to help, and at first light the next day they extracted gently the charred remains of the victims. Some made the coffins, dug a mass grave in a central position in the school grounds, and made preparations for the burial service that followed with due ceremony that evening. The site was dictated by circumstance, because the rapid decomposition of the bodies in the tropical heat necessitated a speedier burial than that planned to take place in the village cemetery further away.

At the time of the fire the majority of the other pupils rushed to the scene, having been aroused by the alarm of the staff and the noise of the burning timbers. Although some were organised briefly into a bucket brigade to bring water from the ocean nearby, the fire raced beyond control. Thereafter they remained as silent sentinels at a safe distance from the site until the fire died away and darkness returned. They had a most troublesome night, but they returned the next day to witness the recovery of the bodies and to follow the proceedings. Some of the senior pupils took part in wrapping the dead in traditional sleeping mats before the coffins were made, and a few acted as pall-bearers later in the day.

The active participation seemed to help many of the school pupils come to terms with the trauma of the event - as much as Milne (1979) had found with the survivors of Cyclone Tracy and Hoiberg and McCaughey (1982) with their injured sailors after a collision at sea. They remained on the spot with their group, and they remained together for about four weeks afterwards while the school was closed and they were directed to stay either in their own homes or with their extended families on the capital island Funafuti. The close association gave them the opportunity to regain their emotional security while among their friends and under watchful supervision in the fold of their immediate families - except for those that lived on islands other than Funafuti.

During the coping period some of the pupils regressed to the point of wanting to sleep beside their parents and of returning to more childish games. Some were reluctant to remain at home alone during the day, and many found their trauma reactivated by the pervasive daily smell of meat being grilled on open fires. They were also more afraid of going outdoors in the darkness to the toilet than before, because of the ghosts of the deceased that were said to abound at night. But after two weeks at home the children were mooching about with too much time on their hands and they could have done with some organised activity. The keenest among them were also concerned about making up for lost time in readiness for their school-leaving exams and scholarship competitions. After a month most were ready to return to school to take part in the three-day memorial games that the headmaster had planned for the reopening: some were reluctant because their fearful associations were too strong.

Unlike their children of secondary school age, the parents did not know the reality of the traumatic event because they had not been there during the fire and they had only their imagination to torment them about it. Emotionally the adults were impelled to try to envisage the last moments of the victims that died. Several likened the experience to seeing horror videos of death and destruction that were often shown on the local television network. One man tortured himself with the thought that his daughter would have been crying out for him when he was not there to help her. He became moronic, and the day after the event boys from his village found him wandering senseless like an automaton and they led him home by the hand. The longest period he had slept in the weeks before we saw him was two hours.

Other adults were still quite distraught weeks after the event. They were tearful, suffered sleep loss, and had appetite disturbance. Two were suicidal because they thought they had nothing left to live for. One man was homicidal with anger against the authorities for not having protected his daughter and for not allowing parents to attend the burial service on the island. Another man found it difficult to use a nearby road along the edges of which he could discern fragments of rubble from the demolished dormitory that evoked memories of the fire. A woman had obliged her family to shift away to cramped quarters in order to avoid exposing her to daily reminders of the scene.

Several adults had difficulty in picking up the threads of their daily life, and some were reluctant to go out alone to fish in the shallows. Like the school pupils many were afraid of malevolent forces that were said to roam after dark. One man was convinced that he had heard the noise of a ghost to which his neighbour had drawn his attention, and to protect himself he had rearranged the sleeping positions of his family on their mats around him as he had when he was a child. Another had brought members of his extended family in to his house at night to provide the extra security he needed. For some adults the tragedy reactivated earlier trauma, and it compounded pre-existing family problems. A woman trembled as she expressed her fervent hope that her deceased daughter might still come home to her, like the son that she lost at sea a few years before. A man reacted badly with guilt because his relationships at home had been far from satisfactory and he had not been as comforting a father as he should have been.

In diagnostic terms 33 (11.1%) of those presenting were judged to have had the classic clusters of intrusive and avoidance symptoms and signs of hyperarousal that fitted the diagnosis of Acute Stress Disorder (ASD), but they declared their symptoms to be improving. This group included 14 of the pupils that had been either in the dormitory or in close proximity to the victims, six of the staff, and 13 of the relatives. Thirty-one (10.4%) had the more persistent symptoms of Posttraumatic Stress Disorder (PTSD), and 13 (4%) had classifiable DSM IV symptoms of anxiety, grief reaction, or psychosomatic disorder (Table 1). Subsequently arrangements were made for the 44 with the more troublesome symptoms to be given closer attention and the progress of the 33 others to be monitored.

The church and care of the dead

The vast majority of respondents were devout in their Christian beliefs, and it is idle to speculate on what the outcome might have been otherwise. Most were trying to overcome their grief by accepting that the fatalities were part of God's benign plan for the improvement of the universe. Some were less accepting, particularly if their daughters had been regular worshippers and keen students of the Bible. A few cast their thoughts about in an attempt to find some mortal person or organisation to blame for the tragedy - one even to the point of blaming a magician from the visiting circus whom he thought had control of the powers of evil.

To my relief the Ministers were supportive of their flock rather than condemnatory - as I had discovered had been the case with the clergy after the 1997 cyclone on Manihiki in the Cook Islands (Taylor, 1999). On that occasion they obliged the survivors to search their memories for the moral transgressions that had brought the natural disaster about, without suggesting that global warming and the El Nino effect might have been alternative and more plausible explanations for le force majeure. However because of the fear of ghosts, spirits, and apparitions emanating from the dead that were expressed in Tuvalu, and the pervasive influence of the church there, I asked the clergy whether they might consider addressing the matter in their preaching and pastoral work.

I was not reluctant to make the request, because I was committed to the World Health Organisation's definition of health that had an underpinning of moral and social values (see Ottawa Charter for Health Promotion 1986 update December 1997). I became more certain of doing so later when I discovered that the clergy in some of the Pacific nations had taken an inspirational stand on social issues (Finau, 1980: Lini, 1980: Mavor, 1980). My request might have taxed the clergy, because it raised matters on which the authorities on remedial liturgy (Henderson, 1994) and on exorcism (McNutt, 1995: Pearson, 1995: Medline and PsycLit, accessed 8 August 2000) were somewhat silent. However, Grant (1999) and Smale (2000) did make a welcome start recently by addressing the more general relationship between religion and disasters.

My own inquiries showed that in Tuvalu the ghosts of the departed were said to arise in the darkness under the command of the devil on the third day after burial, eternally to haunt and chastise people passing nearby. According to Kofe (1983b) the supernatural beliefs about ghosts and spirits emanated in pre-Christian times, and subsequently they were overlaid by the teachings of the missionaries. But obviously they remained far from a spent force in the culture. The present-day adults said that their grandparents had introduced the beliefs to them when they were very small. But while such childhood training must have had some adaptive value for facing the realities of life, in the present circumstance that seemed not to have been the case because apart from inducing extraneous anxiety, the fear of ghosts arising from the mass grave in the centre of the campus might have presented a major obstacle to the reopening and even the continuation of the school. To counter this I suggested that the authorities might try to create a more positive relationship between the living and the dead by building a memorial church close-by that would feature firmly in the school's daily activities.

In the villages by day there was obviously some accommodation in the community between the two states of life and death - otherwise there would not have been graves of family members to be seen alongside a number of the homes. But the same relationship with the dead did not obtain by night. Incidentally, while personally not being partial to frequenting cemeteries, nor being an 'ethnoarchaeologist' with a special interest in the macabre (cf. Pearson, 1999), I could only wonder why supernatural forces, particularly those that were said to emanate from dead children, were always perceived to be malevolent.

While on this sombre topic, the thought crossed my mind that because of the country's high and increasing population density rates (A situational analysis of children and women in Tuvalu, 1996, pp. 7-10), Tuvaluans might soon need to reconsider their funerary customs. In the past, with 2 000 years of occupancy and the scarcity of land, they must have had some alternative to burials in the ground, because there was not the proliferation of graves to be seen as would otherwise be expected. Yet I was told that unlike elsewhere in Polynesia (cf. Mackenzie, 1930, ch.7) other methods such as mummification, burial at sea, and cremation had never been used in Tuvalu.

Protection of adolescent girls

Usually a consideration of the causes of catastrophe falls outside the province of health professionals. In the case of fire, it is left for engineers to consider such matters as the observance of building codes, safety precautions, evacuation drills, and blaze extinguishing, and for the Police to search for evidence of malicious intent. But in the present case in which adolescent girls were held securely in their dormitory at night with locked doors and barred windows, it seemed important to raise questions about the justification for the practice and the need for it to continue.

From discussions with knowledgeable informants it seems that it is the well established practice of many boarding schools in the South Pacific to lock up the girls up at night and to control and monitor their heterosexual interactions by a fairly strict code of rules by day. The practice is justified on the grounds of preventing pregnancy in girls of school age, yet its success is difficult to determine because the specific data seem not to have been gathered, and that which approximates is inconclusive. For example, in Tuvalu the data on the unwanted pregnancy rate of unmarried women in the wider 15-24 age group, albeit with a suggestion of under-reporting, appears to be among the lowest of the Pacific Island nations (State of the Pacific Youth Report. 1998, p.43, table 9). But another report commented that although the number of teenage mothers was low, 'premarital teenage childbearing is a serious problem facing young women in Funafuti' Chandra (1999a). However the same researcher went on to compare the responses of 196 never-married women between the ages of 15 and 24 with those of a wider age range of 291 males, and she reported that 'in keeping with traditional and cultural values most women were not sexually active (because) virginity is still a very important and valued possession. There was much shame and stigma associated with women who became sexually active or had premature births. Such relationships brought shame and degradation to the family and especially to the parents…' (Chandra, 1999b, pp.12-13)

But whether school-age pregnancy is or is not currently a real problem, a superficial acquaintance with history shows that the sexual attitudes and behaviour of the South Seas Islanders were not always a matter of concern. From the 17th century the relaxed sexual mores of the South Seas Islanders captured the attention of expeditioners after their long voyages of deprivation. Later they caught the imagination of the European school of philosophers, painters, and writers that pined for the glory of lost arcadia (Sutton, 1995). Then as far as Tuvalu was concerned, came two Samoan Pastors 'who succeeded in putting the people under their complete and dominating influence…(and began a tradition that) destroyed much of the wealth of the culture' (Taafaki, 1976).

The particular clergy had been trained in Apia by the London Missionary Society, and despite any other benefits they might have brought, they had somewhat inflexible attitudes with regard to sexual behaviour (Moorehead, 1966). They went on to establish single-sex schools, and one of their successors who ran the school for boys on Vaitupu, was a strict disciplinarian named Kennedy that became a District Officer and ruled his 'Kingdom' with a very firm hand (Teo, 1983). Frank Pasifika, who had been under Kennedy at the school in the 1920's and then an associate of his in the colonial administration, recalled him as prohibiting even friendly boy/girl relationships. If a boy was seen speaking to a girl 'the reward would be half a dozen thrashes with a cricket back, which gave you a bruised backside for several days' (Pasefika, 1990, p.20). Evidently the young unmarried men had a particular protective role in their own homes concerning females, but not elsewhere, because through observing the ancient custom of moeototolo, or night creeping, they were entitled to pay surreptitious nocturnal visits to the homes of young women in which they were interested. According to Kofe (1983a), in the early days the secretive practice so offended one Pastor Tema that he had the village police in Funafuti steal into the houses 'to see if all was as it should be'.

But evidently the 'merry old days of heathendom' survived the repressive measures that were imposed, because in the 1930's they attracted the strictures of Arthur Grimble. He was another District Officer, immortalised by his own memoirs (Grimble, 1952) but described by a local as a 'conceited and ambitious man who ... believed he knew better than anyone else what was best for his subjects' (Teo, 1983). Shortly afterwards the anthropologist Danielsson (1956, caption to photo 11) remarked that 'the girls, afraid of ghosts in the palm trees, often receive their male friends at home. The huts are quite open, of which the 'night creepers' take advantage'. But after World War 2, the prospect if not the reality of an earthly paradise in which free love abounded on sun-drenched tropical islands continued to attract writers to the South Seas. A few decades later Field and Field (1972) renewed the theme, and more recently the travel writer Theroux (1992) expressed his disappointment in not being able to confirm it after making an extensive sortie in the region.

However, the present excursion into history is not to plead for a return to less inhibited times, but to show that cultural practices are not immutable. It is for the Tuvaluans themselves to consider whether they should or should not change their protective attitudes towards their adolescent females and change the forms that give them expression. Already since the fire at the secondary school they have demolished the remains of the burned dormitory building, removed the bars on the windows of the remaining dormitories, and erected a perimeter fence around the dormitories in which the girls slept. It is up to them to consider whether they might develop and implement a policy that places more emphasis on internal controls and less on external measures to achieve the desired ends. Were they to want to do that, they would need to open the topic of sex and sexuality to give male and female adolescents a better understanding of both subjects and to make them accept more responsibility for the consequences of their behaviour. It has to be said that the issue is a constant that faces every community throughout the world.

Concluding comments

In retrospect it can be claimed that in carrying out the assignment the essential elements for planned interventions following a school disaster were followed (Nader & Pynoos, 2000). The comment can also be made that despite the widespread assumption that any psychological intervention is effective only if made within the first 72 hours of a traumatic event, in the present case the lapse of three to four weeks was seen as possibly beneficial. Although the delay was incurred more by circumstance than choice, it enabled casualties to use whatever personal, social, or cultural systems with which they were familiar to address their reactions, and it left those with a residue of symptoms less questionably for professional attention.

Finally, it is too early formally to evaluate the interventions that the team used to help those in need. But informally the community was so appreciative of its efforts that at short notice it held a feast of honour for those involved. There followed a drumming and dancing competition between two villages that far surpassed any physical programme for stress reduction that that Davis Eshelman and McKay (1995) or any of the other authorities might have prescribed.


A situational analysis of children and women in Tuvalu. (1996). Government of Tuvalu & UNICEF/Pacific. Suva.

Andrews, G., Crino, R., Hunt, C., Lampe, L., & Page, A. (1994). The treatment of anxiety disorders. Cambridge University Press.

Bell, B., Wiseman, G., & Hughes, T. (n.d.). Tuvalu Trust Fund: 10th Anniversary profile: 1987-1997. GP Print: Wellington, NZ. pp. 23.

Benson, H. & Stark, M. (1996). Timeless healing: The power and biology of belief. London: Simon & Schuster.

Chandra, D. (1999a). Knowledge, attitude and behaviour of young women (15-24 year olds) towards contraceptive methods, reproductive and sexual health and teenage pregnancy: Country project report for Tuvalu. Suva: U.S.P. International Planned Parenthood Association of Oceania. pp.33.

Chandra, D. (1999b). Male knowledge, attitude and practice of family planning: Country project report for Tuvalu. Suva: U.S.P. International Planned Parenthood Association, Oceania. pp.28.

Daniellsson, B. (1956). Love in the South Seas. London: Allen & Unwin.

Davis, M., Eshelman, E.R., & McKay, M.(1995). The relaxation and stress reduction workbook. (4th edn.) Oakland CA: New Harbinger Publications.

DeVries, KM.W. (1996). Trauma in cultural perspective. In B.A. van der Kolk, A.C. McFarlane, & L. Weisaeth. (eds.). (1996). Traumatic stress: The effects of overwhelming experience on mind, body, and society. (ch.17). New York: The Guilford Press.

Diagnostic and statistical manual IV. (1994). Washington DC: American Psychiatric Association.

Doherty, G.W. (1999). Cross-cultural counseling in disaster settings. Australasian Journal of Disaster & Trauma Studies, 2, pp.20. http://www.massey.ac.nz/~trauma/issues/1999-2/doherty.htm (accessed 13/05/00).

Fields, J. & Fields, D. (1972). South Pacific. Wellington: Reid.

Finau, P. (1980). Prayer, persuasion and politics. In S. Tupouniua, R. Crocombe, & C. Slatter. (eds.). The Pacific Way: Social issues in national development. (pp. 165-175). Fiji: South Pacific Social Sciences Association.

Firth, S. (1997). The war in the Pacific. In D. Denoon, S.Firth, J.Linneken, M.Meleissea, & K.Nero. (eds.). The Cambridge History of the Pacific Islands. (ch. 9). Cambridge: University Press.

Frankl, V. (1984). Logotherapy: Man's search for meaning. New York: Washington Square Press.

Gordon, R. & Wraith, R. (1993). Responses of childen and adolescents to disaster. In J.P.Wilson & B.Raphael, (eds.). International Handbook of traumatic stress syndromes. (ch. 48). London: Plenum Press.

Grant, R. (1999). Spirituality and trauma: An essay. TRAUMATOLOGYe, 5:1, E2. pp.4. http://www.fsu.edu/~trauma/T-088.html accessed 24/06/99.

Grimble, A. (1952). A pattern of islands. London: John Murray.

Hahn, R.H., Albers, L.J. & Reist, C. (1997). Current clinical strategies. Laguna Hills CAL: CCS Publishing Co.

Hayes, N. (Ed.). (1998). Doing qualitative analysis in psychology. Hove, East Sussex: Psychology Press.

Henderson, J.F. (1994). Liturgies of lament. Chicago: Archdiocese Liturgy Training Publications.

Herrara, J.M., Lawson, W.B., & Sramek, J.J. (1999). Cross cultural psychiatry. Chichester: John Wiley.

Hoiberg, A. & McCaughey, B.G. (1982). Collision at sea: The traumatic aftereffects. Report 81/39. San Diego: Naval Medical Research & Development Command. pp. 25.

Honigman, J.J. (1954). Culture and personality. New York: Harper & Brothers.

Kofe, L. (1983a). Palagi and pastors. In H. Laracy, M.T. Crocombe, T. Isala, K.H.Thaman, & A.Ravuvu. (eds.). Tuvalu: A history. (ch.15). Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Kofe, L. (1983b). Old-time religion. In H. Laracy, M.T. Crocombe, T. Isala, K.H.Thaman, & A.Ravuvu. (eds.). Tuvalu: A history. (ch.3). Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Kubler-Ross, E. (1975). Death: The final stage of growth. Englewood Cliffs NJ: Prentice-Hall.

Laracy, H., Crocombe, M.T., Isala, T. Thaman, K.H., & Ravuvu, A. (eds.). (1983). Tuvalu: A history. Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Lini, W. (1980). Should the church play politics? In S. Tupouniua, R. Crocombe, & C. Slatter. (eds.). The Pacific Way: Social issues in national development. (pp. 176-177). Fiji: South Pacific Social Sciences Association.

Linley, A. (2000). Transforming psychology: The example of trauma. Psychologist, 13, 7, 353-355.

Linnekin, J. (1997). Contending approaches. In D. Denoon, S.Firth, J.Linneken, M.Meleissea, & K.Nero. (eds.). The Cambridge History of the Pacific Islands. (ch. 1). Cambridge: University Press.

Mackenzie, D. (1930). South Seas: Myths and legends. (1930 reprinted 1996). London: Random House.

MacNutt, F. (1995). Deliverance from evil spirits: A practical manual. Grand Rapids MICH: Chosen Books.

Markowitsch, H.J., Kessler, J., Van Der Ven, C., Webber-Luxenberger, G., Albers, M., & Heiss, W-D. (1998). Psychic trauma causing grossly reduced brain metabolism and cognitive deterioration. Neuropschologia, 36, 1, 77-82.

Mavor, J. (1980). Is the church obsolete? In S. Tupouniua, R. Crocombe, & C. Slatter. (eds.). The Pacific Way: Social issues in national development. (pp. 178-182). Fiji: South Pacific Social Sciences Association.

McFarlane, A.C. & van der Kolk, B.A. (1996). Trauma and challenge to society. In B.A. van der Kolk, A.C. McFarlane, & L. Weisaeth. (eds.). Traumatic stress: The effects of overwhelming experience on mind, body, and society. (ch.2). New York: The Guilford Press.

McFarlane, A.C. & Yehuda, R. Resilience, vulnerability, and the course of posttraumatic reactions. In B.A. van der Kolk, A.C. McFarlane, & L. Weisaeth. (eds.). (1996). Traumatic stress: The effects of overwhelming experience on mind, body, and society. (ch.8). New York: The Guilford Press.

Marsalla, A.J. (1979). Cross-cultural studies of mental disorders. In A.J.Marsalla, R.J.Sharp, & T.J.Ciborowski. (eds.). (1979). Perspectives on cross-cultural psychology. (ch.11). New York: Academic Press.

Maslow, A. (1970). Motivation and personality. New York: Harper & Row.

Maude, H.E. (1981). Slavers in paradise:The Peruvian slave trade in Polynesia,1862-1864. Canberra: Australian National University Press.

Milne, G. (1979). Cyclone Tracy: Psychological and social consequences. In J.L.Reid. (ed.). Planning for people in natural disaster. (pp. 116-123). Queensland: University Press.

Munch, R., & Smelser, N.J. (eds.). (1992). Theory of culture. Berkeley: University of California Press.

Nader, K & Pynoos, (eds.).(2000). School disaster: Planning and initial interventions. pp.17. http://www.sourcemaine.com/gift/nader.html accessed 17/8/00.

Nero, K. (1997). The end of insularity. In D. Denoon, S.Firth, J.Linneken, M.Meleissea, & K.Nero. (eds.). The Cambridge History of the Pacific Islands. (ch. 13).Cambridge: University Press.

New Zealand Official Development Assistance. (1999). Official Review 1999. Wellington: Ministry of Foreign Affairs & Trade.

O'Brien, T. (1983). Genesis. In . In H.Laracy, M.T.Crocombe, T.Isala, K.H. Thaman, & A.Ravuvu. (eds.). (1983). Tuvalu: A history. (ch.1). Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Ottatwa Charter for Health Promotion. (1986 update December 1997). pp.4. http://www.who.dk/

Pasefika, F. (1990). The autobigraphy of Frank Pasefika. Tuvalu: University of the South Pacific Extension Centre.

Paton, D. (1992). International disasters: Issues in the management of relief workers. Disaster Management, 4, 4, 183-190.

Pearson, M.A. (1995).Christian healing: A practical and comprehensive guide. (2nd edn.). Grand Rapids MICH: Chosen Books.

Pearson, M.P. (1999). The archaeology of death and burial. London: Sutton Publishing.

Phillips, B.M. (1993). Role of the practitioner in applying science to practice. In J.A. Mindell. (ed.). Issues in clinical psychology. (pp. 29-35). Maddison NJ: Brown & Benchmark.

Rose, S., & Bisson, J. (1998). Brief early psychological interventions following trauma: A systematic review of the literature. Journal of Traumatic Stress, 11, 4, 697-710.

Seligman, M.P. (1995). The effectiveness of psychotherapy: The Consumer Report Study. American Psychologist, 50, 12, 965-974.

Selye, H. (1956). The stress of life. New York: McGraw Hill.

Sherman, J.J. (1999). Effects of psychotherapeutic treatments for PTSD: A meta-analysis of controlled clinical trials. Journal of Traumatic Stress, 11, 3, 413-435.

Smale, R. (2000). The Port Arthur anniversary services: Post-disaster rituals and symbols. Australian Journal of Emergency Management, 15, 1,2-5.

Smelser, N.J. (1992). Culture: Coherent or incoherent. In R.Munch & N.J. Smelser, N.J. (eds.). Theory of culture. (ch.1). Berkeley: University of California Press.

State of the Pacific Youth Report. (1998). UNICEF, Fiji.

Stewart-Brown, S. (1998). Editorial: Emotional wellbeing and its relation to health. British Medical Journal, 317, 1608-1609.

Summerfield, D. (2000). War and mental health: A brief overview. British Medical Journal, 321, 232-235.

Taafaki. P. (1976). The old order. In H.Laracy, M.T.Crocombe, T.Isala, K.H. Thaman, & A.Ravuvu. (eds.). (1983). Tuvalu: A history. Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Taylor, A.J.W. (1999). Towards the classification of disasters and victims. TRAUMATOLOGYe, 5, 2. http://www.fsu.edu/~trauma/ accessed 19/09/99.

Theroux, P. (1992). The happy isles of Oceania: Paddling the Pacific. New York: Putnam's Sons.

Teo, N.P. (1983). Colonial rule. . In H.Laracy, M.T.Crocombe, T.Isala, K.H. Thaman, & A.Ravuvu. (eds.). (1983). Tuvalu: A history. (ch.17). Suva, Fiji: University of the South Pacific and the Government of Tuvalu.

Tuvalu 1997 Economic Report. (1998). Manila: Asian Development Bank.

Tuvalu Trust Fund 10th Anniversary Profile 1987-1997. Wellington: GP Print. pp.23.

van der Kolk, B. (1996). Trauma and memory. In B.A. van der Kolk, A.C. McFarlane, & L. Weisaeth. (eds.). (1996). Traumatic stress: The effects of overwhelming experience on mind, body, and society. (ch.12). New York: The Guilford Press.

Wen-Shing, T., & Jing H. (1979). Culture and psychotherapy. In A.J.Marsalla, R.J.Sharp, & T.J.Ciborowski. (eds.). (1979). Perspectives on cross-cultural psychology. (ch.15). New York: Academic Press.

World Health Organisation. (1993). International classification of diseases. 10th revision (ICD-10). Geneva: authors.


A.J.W. Taylor © 2000. The author assigns to the Australasian Journal of Disaster and Trauma Studies at Massey University a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Massey University to publish this document in full on the World Wide Web and for the document to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the author.

| Home | Current | Back Issues | Reports | Conferences | Books | Links | Information |

Comments to
Massey University, New Zealand
URL: http://www.massey.ac.nz/~trauma/

Last changed October 12, 2000
Copyright © 2000 Massey University