Famine in Somalia : warning against the warning!

Twenty years of war in central and southern Somalia have destroyed the country, and the drought now raging in the region only makes the immense difficulties of its people worse. Statements announcing an unprecedented famine in East Africa are not, however, very convincing. A series of nutrition surveys conducted under the auspices of the US Agency for International Development (USAID) in July gives us a clearer picture by distinguishing three different situations. First, there are pockets of famine in south-central Somalia characterized by high infant mortality (there is talk of some 30,000 deaths in the preceding three months). We should add that these “pockets” weren’t pinpointed geographically, no less visited, but rather deduced from the shape of the curves for mortality and morbidity (the number of people ill in a given population). Elsewhere in the south, there is “only” food insecurity, while the northern part of the country is enjoying normal harvests and healthy livestock, free from any food crisis at all. There is thus a heterogeneous situation, only partly described by the term “famine.”

But how, in fact, is famine defined? What constitutes a famine to the UN and the NGOs is the combined presence of the following three elements: a food shortage affecting 20% of households, over 30% of children with severe acute malnutrition, and an adult mortality rate exceeding the alert threshold of 2 deaths/10,000/day. Expressing the horror of a famine in terms of numerical indicators and quantitative thresholds may seem inhumane, or even shocking. Yet because declaring a humanitarian state of emergency has serious ramifications both politically and in terms of emergency aid, agreement on an at least somewhat objective definition is essential. In this as in other international aid sectors, meetings between practitioners and researchers and the lessons learned from experience have led to better practices. But this ability to objectify a crisis by quantifying it imposes its own constraints, without which the numbers themselves lose all meaning. For example, mortality calculations suppose knowledge of basic demographic data that are generally lacking in many poor countries. Likewise, extrapolating data collected in certain times and places is a risky enterprise. Equally questionable is the practice of expecting – as we do in the Horn of Africa – that a famine will spread out progressively from localized foci, like an infectious disease. More generally, and despite marked improvements in technology and know-how over the past two decades, we are still left speechless by how casually alarmist numbers are put forward, year after year, by the UN, and then immediately relayed by the media and NGOs – whether about the effects of the tsunami in Southeast Asia, cholera in Haiti, the swine flu worldwide, or the floods in Pakistan, to name just the most salient recent announcements. The combination of drought and large, armed conflict-driven population displacements can lead to famine, especially in displaced populations, which are always the most vulnerable. It was under these very conditions that the 1991-1992 famine in Somalia occurred. But what might happen is not necessarily what does happen, given the many other social, political, and economic factors that come into play in a subsistence crisis.

Millions and millions of people in East Africa live in extreme poverty without being subjected to the war and oppression experienced by those living in southern Somalia and Ethiopia's Ogaden region. And the presumed famine involves only some parts of the conflict-ridden regions. At the time of this writing, there are repeated calls for the urgent rescue of millions of people in danger of starving to death. While the televised reports bring the very real drama of displaced families home to us, they do not show the emaciated adults that are the hallmark of famines. And while an ounce of prevention is no doubt worth a pound of cure, you still need to know what you’re trying to prevent in order to decide how best to do it. Declaring a humanitarian state of emergency in which ongoing famine, the threat of epidemics, potential famine, material insecurity and malnutrition are indiscriminately mixed is no help at all.