What is humanitarian medicine? Medicine that targets populations that have been marginalised, are suffering the effects of a crisis or have been deprived of access to care. Medicine that is practised for its own sake, with no other aim than to make itself useful. Most members of an organisation working in this area would no doubt acknowledge this definition, the conditions in which it is practised having been described in the previous pages, though with no claim to exhaustiveness. Explaining the issues and political constraints it faces seemed to me preferable than drawing up a catalogue, at the risk of leaving out certain aspects and operating methods. Questioning its limitations, at the expense of quelling rash enthusiasm, seemed to me more enlightening than hammering down its principles or reiterating its ideals.

The categorisation into “exceptional situations” and “ordinary situations” may be debatable but is justified by the differences in status and the time frame over which actions are implemented. It does accentuate, however, perhaps excessively, the degree of consistency that prevails in the first type of interventions, which are free from the degree of arbitrariness that leads to the second. Any catastrophe, human or natural, calls for action to be taken, if only on an exploratory basis. But how is it possible to distinguish between countries and populations that are all broadly the same as each other and characterised by the same degree of wretchedness? Why for example, focus on tuberculosis rather than road accidents, the boat people in the China Sea rather than those in the Gulf of Mexico or the Mediterranean, children but rarely old people, and so on? The answer lies outside the situations themselves. It lies in the circumstances, the differing interests and preferences of decision-makers, as well as in the history and the culture of institutions. NGOs are often criticised for this lack of consistency, with some justification, but it would be wrong to counter it with a supposed consistency based on the concept of exception.

In the first place, because the usefulness of the aid provided is an essential element in assessing the operation, and usefulness needs to prevail over legitimate concerns around institutional consistency. Let us recall that usefulness should not be taken for granted and that there also needs to be agreement on the parameters used for its assessment.

Secondly, because humanitarian action, medical or otherwise, can only be fragmentary and therefore somewhat arbitrary. Those in charge are constantly making choices and therefore eliminating certain options: they cannot imagine, even in the long term, providing general health or social care. Except to claim for themselves a status as an international public service, which some NGOs seem to do implicitly by setting themselves up as the spokespeople for an imaginary third world. Humanitarian medical assistance is not some kind of embryonic Universal Health Insurance on a global scale.

I cannot conclude without making some reference to the question of organisation, the structure needed for effective action. The head office of a humanitarian NGO is not unlike a business and its organisation chart, even when it operates within the voluntary sector, traditionally represents the different skills that make it possible to run, fund, supply and monitor operations. This is not the place to go into detail about the different forms such a system can take but, in addition to its operational necessity, its importance needs to be underlined for two different reasons.

Remote follow-up is essential to assess how operations are going in an often changing context, as the ability to have a distant look and the perspective it offers are complementary to the views and analysis of the teams in the field. Combining these positions and helping them to interact would avoid both overreacting to unexpected changes or intercurring events and trivialising them. Such a shift would, of course, not be enough in itself to counter any untimely outburst or flawed routine, but it would limit the risks.

At another level, another reason to raise this question relates to the determinants of action. Over the last 20 years, hundreds of millions of dollars have been poured into the humanitarian movement, as a result of the massive involvement of governments and international organisations. Humanitarian aid has moved – in part – from a demand-based economy, focused on needs, to a supply-based economy, focused on resources: how many aid programmes – distributing basic products, renovating buildings, medical aid – have been run in the last few years, not because there was a real need, but because there was an available budget? Unlike United Nations agencies, NGOs have the ability to refuse such requests. It is always the case that any institution is tempted to grow for its own sake and to confuse its desire to exist with the reasons for its existence, and no budgetary control process can shield it from such a temptation. It can only be held in check by internal debate, on condition that such debate is not excluded in advance in the name of the nobility and urgency of the work to be done. It is important that in organisations where immediate action is the priority, deliberation and contradiction are not seen as a waste of time. The messages put out by most aid organisations on the urgency of the situation after the tsunami are a useful reminder that there can be a huge gap between the desire to please donors and providing meaningful help.