Emotions generated by the brutal and murderous character of a natural disaster are not calmed by the feeling that the victims bear a certain responsibility for the origins of the drama-whether because they are engaged in conflict (war), or poor sanitary or economic practices (epidemics, famines). Claiming, on the public scene, to be an actor responding to a natural disaster, is a way of linking one's image with that of the ideal victim. Governments are particularly fond, moreover, of events such as these-natural disasters that exonerate authorities, in advance, of accusations that they themselves might be implicated in its origins.

Superficial preparations and responses by the authorities in place can, at times, be made an issue of, but governments cannot be reproached for not controlling the tectonic plates of the earth's crust. We might well celebrate such an enthusiastic embrace of emergency aid-if the expertise international aid brings were truly adequate to meet the needs of a population impacted by a natural disaster.

But the operating methods and culture of international emergency aid institutions are most often the product of experience gained in time of war. In armed conflicts an international actor's legitimacy flows from its non-participation in hostilities. Techniques of assistance are compatible with a war context, in which planning anywhere beyond the short-term is impossible. The presence of a foreign actor as a non-participant in the hostilities offering its services to all the combatants is an undeniable advantage when it comes to securing their consent for deploying aid. But, from the perspective of practical continuity, aid groups are accustomed to managing within a temporary framework that lasts as long as the war goes on. In natural disaster situations, the temporary is not bound to last.

Needs for durable reconstruction emerge as soon as the disaster recedes. A review of operations over the last decade confirms this diagnosis, from Hurricane Mitch (Honduras, 1998) to Haiti's earthquake (2010), not to mention the Asian tsunami (2004). In such situations-common assumptions notwithstanding-the aid actually received by disaster victims is in large part indigenous.

The international aid operation in Haiti confirms the above-made point concerning structural unsuitability, and two issues have illustrated it from the very first weeks of the intervention: the injured and shelters. In Port-au-Prince the treatment of the injured was a demonstration, at one and the same time, of the uncontestable support provided by foreign teams as well as all their limitations. The first step in aiding the survivors was removing the injured from the rubble and transporting them to the treatment centers. After a brief period of shock, the number of initiatives for removing the survivors from the ruins increased rapidly. No precise enumeration will do justice to the efforts of the city's population, but it is obvious that tens of thousands of injured persons benefited from the concern of their families, neighbors, Red Cross volunteers, and simple passers-by.

Dozens of foreign teams (1,400 people) coming from around the world in specially chartered planes participated in removals necessitating specialized equipment, making possible the digging out of roughly 150 people. These international teams often gave priority to nationals of the countries that had sent them there, and the personnel of international organizations present in the capitol.

During the next step-attending and treating the injured in health centers-the international teams' contribution is undeniable given how overwhelmed local capacity was by the extremely high numbers and severity of injuries. Yet, despite the obvious plus-value of international medical support, the limitations of foreign aid groups became very quickly apparent. Their contribution to the initial phases of surgical treatment was crucial, but barely had this stage concluded when the majority of them packed their bags for home. Yet thousands of injured will need care for a number of months before they regain any degree of independence.

Moreover, one surgical prodecure, amputation, became a subject of public controversy. For medical teams, the correlation to war surgery came most naturally to mind at the time. Some surgeons, calling on experience acquired on the battlefields, may have thought that the decision to amputate was justified not only when the prognosis for survival so indicated in the short term, but also because, in the middle-term, a simplified post-operatory follow-up is more consistent with the level of care provided in the country. The comparison with war surgery in a city at peace was incongruous- no doubt linked to the details of the context. Most of the expatriate surgical teams did not plan for a trip long enough, in any case, to postpone and be responsible for the follow-up to their decisions over the time period this would have required.

Debate over the justifications for amputating a limb is not a new thing, and has long provoked agitation in exceptional situations inside the small world of surgery. The choice to amputate amounts to a gamble on the future that will always draw criticism, either from optimists who favor a prudent, conservative course that allows for time to wait and see, or from pessimists, seduced by the prospect of a solution that is immediate, radical and more appropriate when the future is unpredictable. The arguments pro and con involve a two-part assessment: what are the foreseeable developments, first in the individual medical case, and then in the society in which it will have to be treated?

Much as we might want to avoid this dilemma by declaring that, in principle, any therapeutic proposal is open to debate and that the patient must be given the final say, the possibility that an injured person will be making a calm, well-informed decision is never as likely as it is desirable. It is impossible to evaluate the frequency of abusive amputations in Haiti. But it is clear that a morbid fear of amputation is now piled atop the fear of falling concrete. In the minds of many of the injured in Port-au-Prince, leaving home -that had become a source of danger- to go to the hospital, meant exposing oneself to another danger: having a limb amputated without the chance to give one's consent to surgery carrying permanent consequences.

The gap between the characteristics of the medical assistance provided (rhythm, type, volume) and the needs of the population -immediately apparent when the injured are treated-widens when the focus shifts from the individual to the collective level-from lavishing surgical care on one person, to arranging necessary shelter for hundreds of thousands of survivors. The problem isn't that there aren't shelters available yet, a few weeks after the earthquake. The history of large operations of international assistance shows that shelter cannot be allocated on such a large scale in such a short time, coming from such a distance.

By contrast, the absence, several weeks after the earthquake, of a complete, coherent plan of action in this area is a more interesting point for discussion. The first impulse of observers is to point to the insufficient coordination of aid. But it isn't hard to understand that, by definition, a great disaster in its initial phase floods the usual capacities for coordination. In any case the coordination of aid has made great advances over the past fifteen years. It has just recently been redefined and strengthened as a result of the latest UN reform at the start of the 00's. Still, the temptation is great to blame weaknesses in coordination for the unavailability of shelter.

The merit of that explanation is that it leaves pushes to the margins a more unsettling line of inquiry: does international aid expertise square with the social, economic, and political realities of the country? The legitimate authority, the Haitian government, has few resources to act and a reduced capacity to absorb them if they are suddenly offered. The major foreign actors, Brazil, the UN, and the United States, have significant resources at their disposal but don't have sovereignty, an indispensable element for transforming those resources into rapid and effective aid for the population. To provide shelter, one must ordinarily know where to set it up. The difficulties are numerous in this respect: areas are littered with rubble; many people have since been living, on their own initiative, in public or private sites; the identity of the landowner is sometimes unclear; and, because urban tensions were so high prior to the earthquake, some re-planning of areas is desirable.

International aid groups can't help but feel uneasiness in the face of this kind of complexity. Their professional capital has accumulated over time from their experiences treating refugees outside of borders, or displaced persons fleeing war and its consequences inside their own country. These displaced groups receive shelter assistance once they have been confined within a dedicated space: camps. In the absence of camps, international aid institutions involve themselves very little in questions of shelter. The object that embodies their expertise in this area is the plastic tarp: a core of black polyethylene woven into a laminated sheet, with a surface of between 20 and 28 square meters and a durability of about two years in mild climate. The plastic tarp is a compromise: an economical solution that provokes less hostility amongst the people of host countries because the quality of the shelter used by displaced people or refugees is inferior to that of residents. The tarp's fragility seems to say that the recent arrival of a group often deemed undesirable won't be prolonged into too long a stay. In Port-au Prince the survivors aren't displaced; still less are they refugees. They are the inhabitants of the capitol of a country at peace. In the aftermath of an earthquake the new geography of the city offers less space for housing in camps. Disaster victims require a tent in order to deal with several constraints: to shield themselves from the threat of falling concrete, regain a closed living space, and retain their mobility. Aid groups are of another opinion entirely; tents are expensive, heavy, fragile, and it is difficult to obtain thousands of units on an emergency basis. In addition, tents-especially their current versions, which don't need to be fixed to the ground in order to stand up-are troubling because they are simply and quickly deployed, and this makes them handy for occupying private and public spaces without prior authorization. In reality, distributing emergency shelter among thousands of inhabitants of Port-au-Prince raises questions that cannot be solved by simply improving the work foreign aid groups are doing or its coordination.

Haiti is a moment of truth for emergency international aid. Surgery and shelter are not the only examples of aid provision that is poorly adapted to the realities of the country. Access to water is one of the rare facets of aid that haven't been disparaged, but then, this activity relies in large part on the national company in charge of furnishing potable water, and private operators, that already worked effectively before the earthquake. Up to this point whatever food has been distributed has reached only a portion of the population, and the rations often provoke irritation because their composition (rice and oil) is perceived, by a city-dwelling population, as too-lacking in variety. Tens of thousands of latrines will be necessary, but, as in the example of shelter, the first step is to decide on sites where they should be set up. The difficulties of aid do not show up in minor matters. Water, food, shelter, sanitation, and treatment are the bases of a solid emergency aid operation.

When disasters occur in countries where the State can provide the bulk of the effort, as it did at the time of the 2004 tsunami in Asia, the relative ineffectiveness of international actors is little noticed. In Haiti, it is likely that the weakness of public authorities will induce international groups to deal with their own limitations and also, perhaps, a public that will remember their promises. Still, this overall picture doesn't inevitably lead to pessimism as to the role international aid can play. A clear-eyed view of solutions that are not really solutions is the precondition to identifying aid policies that can be relevant.

With its calm and its solidarity, Haitian society has shown a maturity that finds no counterpart in the state of its public institutions. The primary request addressed by the inhabitants of Port-au-Prince to international aid workers is for a job, not charity. The effectiveness of aid will depend on its capacity to understand and support these social dynamics, which are the foundations of day-to-day survival and cannot flourish without international financial support. The fireman must quickly give way to the banker-provided that the latter gives a very wide birth to any sort of extreme, neo-liberal approach to reconstruction and economic development-the track record the 1990's showed that it excluded services essential to the survival of too-large portions of the population.