Biomedicine is inherently a mid-level enterprise, somewhere between the local and the global. I want to understand what happens when standards of care developed in one metropolitan context are developed in another? How do clinicians improvise in order to provide care? In this case, what they are doing is applying a certain ad hoc version of oncology. The same set of tools is not available as it is where the evidence-based work is being produced.

In terms of what ethnography has to offer, maybe two observations. First, ethnography can be an early warning system. The concerns of my friends in Botswana are different from those of the international organizations. Cancer, suicide and road accidents are written out of global public health. If ethnography is structured to be open enough to listen, it can form an early warning system by recognizing that people on the ground recognize what’s troubling them. Second, ethnography can thicken the instrumental case study vignette and this can be a powerful tool for motivating policy. This is very different than setting up a hypothesis and going to test it. There is this image in the global health literature that Africans only suffer, and this is profoundly false. People are too often flattened out and made two dimensional. We need to accept the awareness of people on the ground, the humor, the recognition of the absurdity of the power differentials, without making them enactors of a dramatic gesture of resistance. It is not always heroic.

What are vignettes really for? They have an emotional power that is not achievable in any other form. They work with our emotions. There is a tension between the vignette and statistics, but the statistics are actually extremely important for you. The numbers were shocking. How are they blotted out? Suicide is invisible unless you do the statistical work on it. Epidemiological statistics can lead to ethnographic discovery as well as the other way around.

Your paper raises the important questions of what is the meaning of global health at the local level and how goals can be disconnected from local priorities. Could you say more about race and more about history?

Statistics are great things, but you can get statistics without doing experimental trials. I am thinking about the power of the way that we write. I read your story about the lack of pain relief as a story about torture, a torture economy formed under colonial inequality. But when you think of it in global health terms, I am not sure that palliative care will ever be on the table because it is ultimately about bioethics and questions of triage. African-American patients often don’t get pain relief in America. The racial aspect of palliative care is important, so are the racialized practices of medicine even in the United States.

Can you envision a way to get beyond the disease-specific, vertical way of thinking about medical problems? I thought it was interesting that you worked in a hospital. Hospitals are also departmentalized and people there complain that doctors are only interested in certain problems, not in pain. Is this something that is just innate to biomedicine? The issue that you are raising is bigger than cancer and pain.

You write that cost matters, but that cost doesn’t explain everything. For example, how you can irradiate a patient but not deal with her terminal palliative care? What causes this inability to see and feel and how can we turn this into a topic of research? Would an ethnographer ever be allowed to stay close to where decisions are being made? And is the decision indeed ever formally made, or does it simply emerge?

Pain, palliative care, bioethics, and laughter. I understand why you invoking Elaine Scarry, but her examples were of torture; so I liked your move back to Talal Asad and the notion that pain is relational. We may also think about Lévi-Strauss’ work on how pain is metaphorized. The notion of sentinel effects is really wonderful. If only we could form a language to describe what people are talking about on the ground. Statistics are an important tool. They shape hopes, help to identify trends, and are instrumentalized in a variety of ways. I was really intrigued by the lone doctor in the cancer ward and I would like to know more about the trajectories and identifications of central European medics working in southern Africa today. Just like the IRBs, bioethics (as dominated by philosophers of the British analytical school and lawyers) is one of those areas that has to be unpacked and reworked. We can’t allow them to continue to dominate.

On the issue of statistics and evidence, states have to make a formal request to the International Narcotics Control Board for morphine. There is no market. There is only government responsibility for the amount of morphine requested by a country. At Human Rights Watch we have calculated how many patients the amount of morphine requested by particular states would serve. Then we’ve created numbers for how many people are in severe pain due to HIV or cancer, and compared the two. Burkina Faso, for example, requested enough morphine for only eight people a year. That means that each year more than 25,000 with severe pain from cancer or AIDS or car accidents in Burkina Faso get nothing.

We have addressed the issue of pain and torture in two ways. Legally, the frame we have used is that the failure to provide access to morphine can rise to the level of torture when there is willful intent. A more direct frame, though, is comparing the words of patients in severe pain with the testimonies of torture victims. They both say they’ll do anything to make the pain stop. For those being tortured that may mean that they’ll sign a false confession. For those who are sick, unfortunately, it may mean that they kill themselves – even when the pain is treatable, and their illness may be curable. There are issues and problems in advancing this comparison because it would essentially mean the complicity of doctors in torture (setting them up as the enemy here). And that isn’t necessarily effective in creating change.

On the importance of numbers in making cancer a visible problem. The over-reliance on the DALYs (disability-adjusted life years) might be one of the reasons why cancers still rank fairly low in the developing world. Anthropologist should be involved in how the DALYs are created in the first place. The measures we have now rely on crazy comparisons, like, is it worse to have one arm or have diabetes? And pain often doesn’t get figured in.

At times pain comes across as a biomedical fact, and I wondered if you could give to pain the same complexity and subtlety that you give laughter? If pain can carry pride by being stoic in the hospital, within particular relations of authority, then pain can be open to these other meanings.