What prior assumptions might a psychiatrist bring into the room when the patient is an asylum-seeker? First, as a member of the public he or she is likely to have a view on asylum-seeking as a sociological phenomenon – up to 5 million have sought asylum in Western Europe in the past decade – and on how ‘deserving’ the average case is. After all, two adversarially opposed constructions of asylum-seeking have predominated. Governments, and the conservative social sectors, have stressed the prevalence of ‘bogus’ applications by people who are essentially economic migrants, portraying them as wily, determined and tough rather than as having suffered. On the other side are the agencies and interests who support asylum-seekers, and the liberal and radical social sectors. They portray asylum-seekers as people who had no choice but to run from their countries, innocent of any thought but to escape further persecution and the risk of death. This portrayal invokes the image of suffering and vulnerability rather than resilience and agency. The reality is the muddied, uneven terrain that lies between these two entrenched positions. Many asylum-seekers do not have stories that easily fit the definition of a political refugee in the 1951 United Nations Convention. Even those with the clearest-cut cases – such as those with a credible history of torture – usually cannot prove it; few are vulnerable in any medically attestable sense and, however much they have suffered, they continue to make choices and actively engage with their situations. All asylum-seekers are looking for a better life for themselves and their children.

Behind all this stands the implacably widening gulf between the socio-economic fortunes of the Western world and the rest of the planet. In most of Africa, South America and Asia, living standards, including indices such as access to schools and public health as well as jobs, have fallen strikingly in the past 20 years in relation to what had been achieved by the 1960s and 1970s. Structurally inequitable societies whose elites are prepared to use force to defend what they have (often with Western collusion) are as much the norm as the exception. The Western-led economic order dictates the terms of trade to third world nations, with the poorest not being the priority. The developmentalist ideal that the fortunes of the world's poor would gradually converge on those of the better-off has died. The Western political order is focused on what has been called ‘the management of difference’, an ominous phrase. One element of this is raising the height of the fences around the wealthy Western societies to keep unwanted outsiders out (Reference DuffieldDuffield, 1996). The asylum-seeker is the beggar at the rich man's gate.

The other background factor that can influence how a psychiatrist manages an asylum-seeker or refugee is his or her assumptions about the personal legacy of, say, experiences of atrocity or torture. Do these self-evidently make victims candidates for a diagnosis of post-traumatic stress disorder (PTSD) and a psychological intervention? Elsewhere, I have critiqued the problems associated with the application of PTSD on a supposedly universalistic basis to non-Western patients – and even to Western ones (Reference SummerfieldSummerfield,1999, Reference Summerfield2001). PTSD has become a catch-all diagnosis and signifier, yet its criteria are frequently not what refugees consider significant about their predicament. Even when cross-cultural factors are not present, PTSD has low specificity, distinguishing poorly between the physiology of normal distress and the physiology of pathological distress. A finding that 43% of Cambodian refugees had PTSD 12 years after settling in the USA (cited by Reference TribeTribe, 2002, this issue) says little about how they had fared and what they most needed. During the 1990s, I saw over 800 asylum-seekers and refugees with a history of exposure to political persecution and torture, referred to me for psychiatric assessment: remarkably few had significant psychopathology.

As Tribe notes, presentations in a physical idiom, in the absence of diagnosable organic disease, are common in general practitioner surgeries. To label all this ‘psychosomatic’ is too crude: somatic idioms of distress are located in multiple systems of meaning serving diverse personal and sociocultural functions (Reference Kirmayer and YoungKirmayer & Young, 1998). Asylum-seekers may be as interested in advocacy as in treatment, seeing the doctor (as British patients may do too) as a gatekeeper to scarce social resources such as public housing.

Those with pre-flight contact with health services for psychological reasons seem overrepresented among psychiatric referrals in the UK. Properly resourced and supervised interpreter services are needed. We need more sophistication in psychiatric encounters across cultural boundaries and an awareness that the degree of ‘fit’ is likely to be poor between National Health Service mental health services and presentations by asylum-seekers from cultures where Western psychiatry and the detached introspection of talk therapy are alien. ‘Culturally sensitive’ psychiatry can mean using an interpreter not to elicit the model of illness the patient holds, but rather to garner answers that permit pigeon-holing in pre-set biomedical categories (Reference WattersWatters, 2001). DSM–IV (American Psychiatric Association, 1994) and ICD–10 (World Health Organization, 1992) are not, as some imagine, atheoretical and purely descriptive nosologies. They are Western cultural documents, carrying particular ontological notions of what constitutes a real disorder; epistemological ideas about what counts as scientific evidence; and methodological ideas as to how research should be conducted (Reference Mezzich, Kirmayer and KleinmanMezzich et al, 1999). PTSD is a Western culture-bound syndrome, but I would argue that all of psychiatry is culture-bound: even presentations by patients with organic mental disorders are shaped by local points of view and lifeworlds.

One danger of overemphasising the medical approach to refugees from war or atrocity is that still-evolving concerns and understandings are reduced to a unitary concept, ‘trauma’, neglecting the role of social factors in exile. The Gorst-Unsworth & Goldenberg (1998) paper cited by tribe, found that depressed mood in Iraqi asylum-seekers in London was more closely related to the presence or absence of current social supports than to a history of torture. Eastmond (1998) found that survivors of Bosnian concentration camps living in Sweden did better when offered work-training than when offered psychological services. At 1 year the majority of the second group were on indefinite sick leave. The one surely indisputable fact in the literature of involuntary migration is that people do well, or not, as a function of their capacity to rebuild social capital and meaningful ways of life. Work is central to this.