‘Traditional’ culture and refugee welfare in north-west Thailand

The effects of displacement on culture can have significant impacts on the psychological and physical welfare of individual refugees and on the social dynamics within a refugee population.

Yet, refugees and relief agencies alike often underestimate or feel too overworked to incorporate the importance of cultural factors in assistance programmes. Potential cultural conflicts between refugee communities, host communities and relief agencies are of course important. Less often recognised, however, is the importance of cultural variation and tension within the refugee community.

This article argues that if relief agencies develop a greater awareness of cultural patterns and potential cultural conflict within as well as between communities, their assistance programmes may be more effectively and appropriately designed and implemented. The casestudy discussed here focuses on the perspective of one group within a refugee population. There is not space to explore fully the perspectives of other members of that refugee population or of the relief agency, except where they impact upon the refugees concerned. My aim, however, is not primarily to criticise the NGO but rather to highlight the experience of one group of refugees. This is not an evaluation of one situation, but a description of a process. It is partly a case-study of the significance of cultural factors in the refugee experience, and partly an attempt to address an inequality in the extent to which different perspectives (those of different sectors within a refugee population, and those of relief agencies) get aired.


This article is based on anthropological field research, conducted by the author at the request of the NGO concerned during the course of wider field research conducted in 1996-7 and 1998, with Karenni refugees living in camps on the Burmese border, in Thailand’s northwestern province of Mae Hong Son. Karenni people have been fleeing from Karenni (Kayah) State in eastern Burma and seeking refuge on the Thai side of the border for some years, the first significant numbers arriving in 1989. The main NGO working with the Karenni is a medical agency, responsible among other things for the training of staff for camp clinics.

In early 1996, the total Karenni refugee population was about 5,500. By the end of 1997, it had doubled to 11,000. It now stands at over 16,500.(1) The dramatic increase between 1996 and 1997 resulted from the arrival of new refugees in one of the Karenni camps in and after June 1996, because of ‘village relocations’ inside Karenni State, enforced by the Burmese army from 31 May 1996 onwards. Conditions in this camp deteriorated rapidly and great demands were placed upon space, on existing residents and the Camp Committee, on NGO staff and on Karenni medics. New refugees were arriving in poor physical states, as a result of which dysentery and malaria morbidity and mortality rates within the camp markedly increased. By late 1996, conditions had stabilised but to date smaller numbers of people, some of whom have been in hiding in the jungle for over two years, have continued to arrive.

Diversity within the refugee community

Experiences at the hands of the Burmese army and subsequent displacement to refugee camps have thrown together people who all originate in Karenni State but differ markedly from each other. 'Karenni' itself is an umbrella term under which a number of different ethno-linguistically self-defined groups gather. Most originate in Karenni State but otherwise show great diversity in ethnicity and language, socio-economic and educational backgrounds, religion, political awareness, and the experience of displacement itself. Displacement has not only brought greater contact with non-Karenni outsiders and the wider world but has also thrown together diverse groups who previously had less contact with, and influence upon, each other. Consequently, the effects of some Karenni groups upon others are sometimes greater than the effects of any single outside force.

The recent, post-1996 arrivals are ethnically Kayah, the majority Karenni ethnic group. The majority of the pre-existing refugee populations in two of the three main Karenni refugee camps are also ethnically Kayah. There are, however, some important differences between these groups. Unlike many of the pre-existing refugee population, most recent arrivals cannot speak Burmese and in their villages had no access to health clinics or schools. Before crossing the border, most had apparently rarely, if ever, seen motor vehicles or foreigners. Their villages are remote, permanent hill settlements, sustained by subsistence agriculture. Most have not converted to Christianity and instead follow traditional Kayah religion and curative practices. Village-based spheres of activity and contact are small, and travel to local towns and markets is infrequent. Travel farther afield is rare indeed. Unlike the longer-staying refugees, they also (with the exception of some men) have little conception of the pan-Karenni nationalism behind the ongoing conflict between the Karenni and Burmese armies. Furthermore, many women in particular had not previously seen other Kayah women, such as those among the longer-staying refugees, who do not dress as traditionally as they do.

The importance of culture: traditional Kayah female dress

In their own eyes as well as in those of the mostly Christian, pre-existing refugees, these recent arrivals are ‘traditional’ Kayah who ‘continue to do as our grandmothers and grandfathers did’ while their ethnic cousins apparently do not. For insiders and outsiders alike, the most obvious emblem of this ‘traditional’ identity is women’s clothing.

Almost all pre-existing Karenni refugee women, ethnically Kayah or otherwise, wear a sarong (reaching to mid-calf) and T-shirt or traditional tunic. It is considered improper and unfeminine to show any leg above mid-calf, or to show the chest area (except when breast-feeding). By contrast, all recently arrived Kayah women wear, or wore on arrival, a short skirt-cloth exposing knees and lower thighs, a breast-cloth exposing the back and often one breast, a head-cloth and various ornaments, including silver ear-plugs, and many rings around the knees. The skirt- and head-cloths are always home-made, using home-grown, naturally dyed, home-spun cotton.

A traditional Kayah woman sees this dress as an extension of herself, marking her not only as Kayah but also as a woman of certain age and marital status. It is also a source of pride and marker of identity for all traditional Kayah, male and female. However, the circumstances in which they had to leave home meant that few women could bring with them spare clothes and/or cotton with which to make more. They had only the clothes they wore and, if those clothes became too worn or dirty, once in the camp the only alternative was to start wearing the ubiquitous sarong and T-shirt.

To change dress in this way, however, is distressing. Both male and female recent arrivals think traditionally dressed women very beautiful. Most importantly, immediately on arrival traditional dress, an obvious marker of difference in a situation of sudden exposure to people who do not look or act similarly, was the only visible evidence of what and who they were and had been. All new arrivals, of either sex and whatever age, hoped women would be able to continue wearing traditional clothes but were worried this would become impossible in the camp, without cotton and without money to buy it. In 1996 especially, the inability of women to continue producing textile items in the camp directly caused much group and personal distress.

These anxieties were real enough but they were also due to dress becoming a focus for wider stresses resulting from displacement. In part, this was because the production of traditional textiles is as important as the textiles themselves: the process of weaving, like the process of farming, is as important to the integrity of Kayah culture as are its end-products. Suddenly being unable either to weave or farm, was a stressful experience that exacerbated the trauma of violent displacement itself.

Inter-community dynamics: impacts of host communities and relief agencies

The majority still wear traditional dress but there have been numerous cases of women abandoning it and since 1996 the rate of change has increased. Women concerned feel they had no real choice in the matter and invariably are unhappy in their new sarong and T-shirt. Others’ attitudes also play an important part. For example, some women changed after illness and subsequent referral to Mae Hong Son hospital. Sometimes there were practical reasons, such as illness causing weight loss and leg-rings consequently falling off (if a part of traditional dress is removed - even unintentionally, as here - the rest is also removed); but more disturbing were claims that ‘doctors in the hospital don’t like our clothes’. Certainly, the lack of adequate breast coverage offended Thai cultural norms and contributed to ill feeling.

Such attitudes were difficult for new arrivals to comprehend. Women felt confused and unhappy, uncertain of what was wrong. The medical NGO, while it certainly did not actively perpetuate negative attitudes, neither fully realised their impact nor actively sought to contradict them.

Indeed, in trying to address the conflict between traditional Kayah and Thai ideas of decency, the NGO inadvertently exacerbated the women’s confusion. Rather than trying to discuss the matter with the women and with Thai hospital staff, they were complicit in a process whereby it was suggested to women being referred to hospital that, for their own sake, it would be easier if they were to change their dress before leaving the camp. This process was initiated by camp clinic staff, themselves members of the longer-staying, less traditional refugee population; that is, NGO staff did not personally suggest changing clothes but nevertheless did not engage with the clinic staff’s suggestions.

At that time, the NGO was visiting the camp daily and driving sick women (and others) to hospital. Most of those women were in distressed states, not only because of their illnesses and recent experiences of violence and displacement, but also because for the first time, they found themselves in both a motor vehicle and a new sort of dress (the sarong of which they found difficult to keep up). Furthermore, parents reported that when children were ill enough to require hospital referral, both NGO staff and refugee clinic workers advised that it would be easier for the father to accompany it. Such advice was geared towards minimising offence to the host community. This was understandable and expedient, but it had less impact in easing the situation of newly arrived refugees.

Also problematic was a later weaving project. In 1996, given the cultural importance of traditional female dress and the anxiety caused by women’s inability to continue producing it, I recommended that the relief agency consider facilitating a refugee-run weaving project amongst the recent Kayah arrivals. The NGO subsequently did indeed facilitate a weaving project but among members of the longer-staying community in another camp. While this was not necessarily a deliberate decision directly to substitute a project among longer-stayers for one among new arrivals (and from the NGO’s perspective there may have seemed no connection), it was nonetheless so interpreted by and distressing for recent arrivals, particularly as they had discussed a possible weaving project of their own with NGO staff and myself. While no promises had been made to new arrivals, they felt they had been passed over and let down. The project that did get set up was certainly uncomplicated to design and implement, as NGO staff already had reasonable knowledge of and contacts within longer-staying refugee community structures. It would have been more problematic for the NGO to design a weaving project in conjunction with recent arrivals given insufficient knowledge of the new refugees' community structures and consequent difficulties in developing good working relationships with key individuals within those structures. It appeared to the new arrivals that the NGO did not fully appreciate either the cultural significance of traditional dress and weaving, or the significance of new arrivals’ reliance on their own distinct community structures. While the NGO did not conduct an evaluation of this project-versus-no-project situation, informally they later appeared to share this conclusion.(2)

Intra-community dynamics: attitudes of other refugees

Impacts on culture equal to or greater than those of the relief agency and the host community were also generated within the refugee population itself. The pre-existing, less traditional refugee community’s attitudes to the new arrivals were continually reinforced by traditional female dress(3). Most had seen traditionally-dressed Kayah women before, but not in such large numbers. For less traditional refugees, this dress smacked of backwardness, lack of education and an un-Christian immodesty. Longer-stayers’ talk about the new arrivals focused on the impropriety of traditional dress, and on what they supposed was the new refugees’ ignorance of basic hygiene. (4)Such talk became problematic as it filtered through to interactions between the two groups.

Interactions with refugee clinic staff (members of the longer-staying community, relatively well-educated and almost all Christian) were particularly influential. Even the smallest remark or unintentional hint of disapproval from these individuals could be picked up by new refugees (already in a fragile state) and cause distress. The power of these individuals inadvertently to cause distress was not surprising, given their elevated role as teachers and specialists within the community.

Most significantly from the perspective of relief agencies, these negative refugee-refugee interactions also demonstrated one way in which approaches to cultural factors can affect the success of welfare programmes. The young woman whose baby was acutely ill with dysentery but who tearfully refused to return to the clinic because its refugee staff had told her it was her own fault for being dirty, was a particularly poignant example of the negative impact of insensitivity.


Relief agencies play a significant part in such situations. It is to the credit of the agency concerned here that they requested the assistance of an anthropologist in understanding more about the new arrivals. Furthermore, subsequent to my research the NGO did attempt to address clinic staff’s attitudes. Nonetheless, while the agency was certainly not to blame for all the clothing-associated tensions, some of its actions exacerbated the situation. It could also have engaged more extensively with the new arrivals than it did. The situation at the local hospital, for example, and the attitudes of refugee clinic staff might have been significantly altered by a strategy of discussing issues of cultural difference and perhaps by facilitating an awareness-building programme of discursive contact between new and old refugees.

In essence, there was insufficient allowance for (i) the significance of cultural differences within the refugee community and (ii) the importance of culture not only in the pre-displacement past but also in the new refugees' experiences of arrival in a refugee camp and associated contacts with non-traditional Karenni, Thais and expatriate relief workers. There was (and is) slowness to explore social structures within the new refugee community or, at least to engage with them to the same degree as the pre-existing population in the design and implementation of relief programmes. The effect of this was exacerbated by its contrast with (i) the relief agency's long-standing, good understanding of and engagement with longer-staying refugees and (ii) an understandable concern not to offend the host community and thus compromise working relationships developed over a long period.

Among the longer-staying refugee community, the agency works closely with those in key positions; such individuals are always, in the context of their community, highly respected, relatively well-educated, and usually Christian. Their influence would in any event have been significant but it was further strengthened and perpetuated by the agency's reliance on them to act as go-betweens, and by the agency not fully acknowledging the potential impact of longer-stayers' negative attitudes towards the new arrivals. Here, as in the agency's concern that the Thai host community should not be offended, the agency and its partners were effectively sensitive to each other at the expense - albeit unintentional - of the less well understood new arrivals.

Certainly, agencies have to take into account political and practical considerations as well as cultural ones, but they are also often in a position to facilitate mutual understanding, both between members of the refugee and host communities with whom they work closely, and within complex refugee populations. Furthermore, in examples such as the one presented here, for new arrivals and other relatively vulnerable sectors of the refugee population ‘culture’ is often of more immediate concern than political aspects of refugee-agency-host relationships. Refugee populations are not necessarily homogenous in either culture or needs, and cultural differences within a refugee population can cause distress as much as can differences between it and the host community and relief workers. In complex populations, especially – as here - where power, influence and mutual refugee-NGO understanding are not equally distributed, it is not sufficient to rely on good working relationships with representatives of only some sectors of the population.

In sum, cultural elements such as women’s clothing may seem, on the surface, tangential to the stark reality of being a refugee but in fact, they are integral to it and to associated welfare issues. As such, they should have significant influence on the design and implementation of relief programmes. Equally, increased cultural knowledge and sensitivity on the part of relief agencies can go some way towards minimising further distress (and its negative impacts on physical and psychological health) caused by cultural conflict, be it between refugees and outsiders, or within the refugee population itself.


Sandra Dudley is completing her DPhil in Social Anthropology at Oxford University.


  1. Source: local NGO.
  2. Agency area coordinator, personal communication, 1998.
  3. inserting this last clause makes up for earlier loss of second sentence of previous paragraph
  4. I think this clause has become redundant

Thanks to Edith Bowles; Kerry Demusz; Helen Dalton; the Emslie Horniman Fund of the Royal Anthropological Institute; Old Member's Scholarship and further grants from Jesus College at Oxford University; the Cha and Peter Lienhardt Funds at Oxford University; and the Evans Fund, Cambridge University.

Parts of this paper first appeared in S Dudley 1997 ‘Recent arrivals in Karenni Camp 2: an ethnographic report’, an unpublished report prepared for IRC in Thailand, and in S Dudley 1998, ‘Aspects of research with Karenni refugees in Thailand’, Bulletin of the International Committee on Urgent Anthropological and Ethnological Research (UNESCO), 39: 165-84.

For further information on Burma, Burmese refugees and human rights abuses, visit:







Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at www.fmreview.org/copyright.