e all know that the genocide in Rwanda was the result of politicians' fanning ethnic rivalries between Hutus and Tutsis. The evidence is pretty clear that such manipulation led to the holocaust that claimed 14 percent of the country's population, the overwhelming majority of them Tutsis. Oddly, though, social scientists working after the horrors subsided in northwestern Rwanda found that at least 5 percent of the residents there had been massacred -- even though there were no Tutsis. Why were the Hutus slaughtering each other? Probably, says Jared Diamond, because the area contained 2,040 people per square mile, twice the population density even of famously crowded places like the Netherlands. The average "farmer" worked a parcel 0.07 acre in size; by 1990, 40 percent of the population was living on less than 1,600 calories a day, which is otherwise known as famine. A pair of Belgian economists whom Diamond quotes say the outbreak of fighting "provided a unique opportunity to settle scores, or to reshuffle land properties, even among Hutu villagers.…It is not rare, even today, to hear Rwandans argue that a war is necessary to wipe out an excess of population and to bring numbers into line with the available land resources." In other words, a classic Malthusian trap.

But we all know that Malthus was wrong, don't we? The gloomy Victorian economist reckoned that population growth would outrun food supply, and hence crisis would ensue. In fact, so far just the opposite has happened. Even as populations have soared, total food production has grown even faster. With the exception of misgoverned, horribly crowded places like Rwanda, famine has become less common, not more.

In a larger sense, however, what one might call the Malthusian idea -- that humans demand more of the planet than it can provide -- has been at the unspoken core of environmental thinking for many years. Each decade that passes without catastrophe seems to undermine the idea; we appear to be so smart that we can invent our way out of whatever box canyon we stumble into. On the other hand, each decade sees us climb farther out on what just might be a limb; the stakes keep going up as we build the human enterprise larger and larger. From simply running out of water and oil, to complications such as global warming that arise when we use those resources that remain, to the furor over technologies such as genetic engineering, it's the question behind all the other questions: How exactly will we keep this up? Or, more precisely, can we?

And it's the question that Jared Diamond engages with wide-ranging erudition in this huge new book. Having won a Pulitzer in 1998 for Guns, Germs, and Steel, which explained why some civilizations rose to greater heights than others, he now takes on the equally interesting question of why some civilizations collapsed. Though he is careful, even in the case of Rwanda, to avoid environmental determinism, he nonetheless lays out a matrix of 12 key problems that he thinks can threaten entire civilizations, markers that indicate societal sustainability.

Some of these deadly dozen have been with us for millennia. Citing examples from the Maya to the island South Pacific, Diamond writes that societies may go under because they destroy forests, wild food sources, wetlands, or soil. In fact, deforestation and soil erosion were factors in almost every collapse he describes. There is, for instance, the story of the Greenland Norse, who cut their grasslands for turf to build homes, only to watch strong winds blow away the underlying soil, reducing the carrying capacity for cattle below what the settlers needed to survive. It is not comforting, then, to read that the spread of deserts in China has been so rapid in recent years that as many as 20,000 villages have been abandoned.