In the early morning, speeding between green shimmering fields of paddy, a line of soldiers comes into view. They are strung along the road, languidly prodding piles of rubbish with metal tridents, checking for claymore mines that may have been planted overnight. Arriving at the rural hospital, I settle down to supervise the young medical officer running the new mental health clinic. I concentrate on developing her clinical skills. I give her plenty of feedback, highlighting things done well, areas to work on, that sort of thing. We have discussions about assessment issues, risk, appropriate treatment strategies, family involvement, working with local belief systems, thinking critically. I also supervise her in training the community workers who attend with the patient and family; it is almost unheard of for a patient to come alone.

Six weeks in, I am only just beginning to get my head round the multilayered complexities of rural east coast life. Something I would have found completely beyond me had I been delivering ‘how to do it’ seminars on a flying visit to Colombo. The most important yet most elusive layer is that of ethnicity, the basis for the currently escalating yet undeclared civil war. Patients in the tsunami-affected coastal clinics where I work are Tamil or Muslim, never Sinhalese, reflecting the ethnic mix of the local communities. Towns and clinics are often dominated by a single ethnic group and, although unpopular among health professionals, there is talk of segregated services. There are also ethnic differences in help-seeking behaviour and in the nature of presenting problems. Muslims, for example, seem more likely to seek medical help but less likely to present with problems related to substance misuse. The ethnicity of the doctors themselves can also be an issue. Sinhalese doctors are reluctant to work on the east coast. Ten days ago an intruder entered the doctors’ residence of a nearby district hospital and threatened two Sinhalese medical officers with death if they continued to work locally. They departed early next morning, leaving a distressed Tamil junior medical officer to run the hospital single-handedly. Some doctors are also said to be reluctant to be supervised by colleagues from certain ethnic groups, despite superficial bonhomie. Evidence of the unease and distrust stirred up by the resurgence of troubles. Finally, there is the prickly issue of whether the allocation of INGO (international nongovernment organisation) resources among ethnic groups is equitable, in which I am inevitably seen as a player. This is difficult since, irrespective of the facts, what is at stake is the ‘perception’ of resource allocation, and this of course has become part of the war of spin waged by all sides.

Other layers are perhaps easier to read. For example, the organisational culture of the health system is very hierarchical, and you must respect and work with this power structure if you want to get things done. Local discourses on INGO activities form a further layer, often hard to access, and complicated by the fact that the aid community is seen as an expression of foreign power and power inevitably invokes resistance. One view often expressed by locals, for example, is that INGOs don’t consult enough, follow a biased Western donor agenda, and fail to meaningfully invest in the long-term development of local skills.

Meanwhile, the outside world bumps against us. The World Cup is everywhere. Then just north of the ancient capital city of Anuradhapura a claymore mine kills 64 bus passengers, and on the day of the funerals the town where I’m living, Ampara, is sealed off by roadblocks of giant water pipes and burning tyres. Crowds of young men march in protest against this latest atrocity, bearing white flags of mourning, black of protest. The INGO community stays put in their compounds. But by evening, some shops have reopened, and the young men are back on the football ground. Later they will watch the big game in Germany. Meanwhile there are unconfirmed reports of a driver being dragged from his car while attempting to leave town earlier in the day.

Two-thirds of the one hundred assessments I have observed so far are of people with either schizophrenia or depression, and less than one in ten has a mental health problem linked to the tsunami or the ethnic troubles. One of these is a person with schizophrenia whose home and possessions were destroyed by the tsunami, and whose family was dispersed. He was lost to treatment and relapsed. The disruption of normal family and community support systems in disaster areas is particularly hard on those with severe mental illness, who may be cast adrift. But these are problems caused by social damage, rather than the direct effects of trauma, reinforcing yet again the importance of restoring normal social structures such as schools, jobs and communities, as well as services to support those with severe mental illness.

Being away from home, I find myself inventing new rituals to punctuate the day: early morning purchases of buffalo curd; visits to the market, where razor-edged cleavers effortlessly slice 12 inches of plump red tuna; visits to the Sea Breeze, a rooftop café stranded on a beach of rubble, counting the two hundred colourful new fishing boats riding at anchor; and watching the rubble slowly being carted away for recycling into new causeways and road foundations. But another sort of punctuation begins the week. First I get a text from the security tree, then confirmation on the BBC World Service. A suicide bomber on a motorbike has killed the Deputy Chief of Staff of the Army, Major General Parami Kulatunga, just south of Colombo. Not good for trust between communities in the east.