THE case that the next doubling, the one we're now experiencing, might be the difficult one can begin as readily with the Stanford biologist Peter Vitousek as with anyone else. In 1986 Vitousek decided to calculate how much of the earth's "primary productivity" went to support human beings. He added together the grain we ate, the corn we fed our cows, and the forests we cut for timber and paper; he added the losses in food as we overgrazed grassland and turned it into desert. And when he was finished adding, the number he came up with was 38.8 percent. We use 38.8 percent of everything the world's plants don't need to keep themselves alive; directly or indirectly, we consume 38.8 percent of what it is possible to eat. "That's a relatively large number," Vitousek says. "It should give pause to people who think we are far from any limits." Though he never drops the measured tone of an academic, Vitousek speaks with considerable emphasis: "There's a sense among some economists that we're so far from any biophysical limits. I think that's not supported by the evidence."
For another antidote to the good cheer of someone like Julian Simon, sit down with the Cornell biologist David Pimentel. He believes that we're in big trouble. Odd facts stud his conversation—for example, a nice head of iceberg lettuce is 95 percent water and contains just fifty calories of energy, but it takes 400 calories of energy to grow that head of lettuce in California's Central Valley, and another 1,800 to ship it east. ("There's practically no nutrition in the damn stuff anyway," Pimentel says. "Cabbage is a lot better, and we can grow it in upstate New York.") Pimentel has devoted the past three decades to tracking the planet's capacity, and he believes that we're already too crowded—that the earth can support only two billion people over the long run at a middle-class standard of living, and that trying to support more is doing great damage. He has spent considerable time studying soil erosion, for instance. Every raindrop that hits exposed ground is like a small explosion, launching soil particles into the air. On a slope, more than half of the soil contained in those splashes is carried downhill. If crop residue—cornstalks, say—is left in the field after harvest, it helps to shield the soil: the raindrop doesn't hit as hard. But in the developing world, where firewood is scarce, peasants burn those cornstalks for cooking fuel. About 60 percent of crop residues in China and 90 percent in Bangladesh are removed and burned, Pimentel says. When planting season comes, dry soils simply blow away. "Our measuring stations pick up Chinese soil in the Hawaiian air when ploughing time comes,"he says. "Every year in Florida we pick up African soils in the wind when they start to plough."
The very things that made the Green Revolution so stunning—that made the last doubling possible—now cause trouble. Irrigation ditches, for instance, water 17 percent of all arable land and help to produce a third of all crops. But when flooded soils are baked by the sun, the water evaporates and the minerals in the irrigation water are deposited on the land. A hectare (2.47 acres) can accumulate two to five tons of salt annually, and eventually plants won't grow there. Maybe 10 percent of all irrigated land is affected.
Or think about fresh water for human use. Plenty of rain falls on the earth's surface, but most of it evaporates or roars down to the ocean in spring floods. According to Sandra Postel, the director of the Global Water Policy Project, we're left with about 12,500 cubic kilometers of accessible runoff, which would be enough for current demand except that it's not very well distributed around the globe. And we're not exactly conservationists—we use nearly seven times as much water as we used in 1900. Already 20 percent of the world's population lacks access to potable water, and fights over water divide many regions. Already the Colorado River usually dries out in the desert before it reaches the Sea of Cortez, making what the mid-century conservationist Aldo Leopold called a "milk and honey wilderness" into some of the nastiest country in North America. Already the Yellow River can run dry for as much as a third of the year. Already only two percent of the Nile's freshwater flow makes it to the ocean. And we need more water all the time. Producing a ton of grain consumes a thousand tons of water—that's how much the wheat plant breathes out as it grows. "We estimated that biotechnology might cut the amount of water a plant uses by ten percent," Pimentel says. "But plant physiologists tell us that's optimistic—they remind us that water's a pretty important part of photosynthesis. Maybe we can get five percent."
What these scientists are saying is simple: human ingenuity can turn sand into silicon chips, allowing the creation of millions of home pages on the utterly fascinating World Wide Web, but human ingenuity cannot forever turn dry sand into soil that will grow food. And there are signs that these skeptics are right—that we are approaching certain physical limits.
I said earlier that food production grew even faster than population after the Second World War. Year after year the yield of wheat and corn and rice rocketed up about three percent annually. It's a favorite statistic of the eternal optimists. In Julian Simon's book The Ultimate Resource (1981) charts show just how fast the growth was, and how it continually cut the cost of food. Simon wrote, "The obvious implication of this historical trend toward cheaper food—a trend that probably extends back to the beginning of agriculture—is that real prices for food will continue to drop.... It is a fact that portends more drops in price and even less scarcity in the future."
A few years after Simon's book was published, however, the data curve began to change. That rocketing growth in grain production ceased; now the gains were coming in tiny increments, too small to keep pace with population growth. The world reaped its largest harvest of grain per capita in 1984; since then the amount of corn and wheat and rice per person has fallen by six percent. Grain stockpiles have shrunk to less than two months' supply.
No one knows quite why. The collapse of the Soviet Union contributed to the trend—cooperative farms suddenly found the fertilizer supply shut off and spare parts for the tractor hard to come by. But there were other causes, too, all around the world—the salinization of irrigated fields, the erosion of topsoil, the conversion of prime farmland into residential areas, and all the other things that environmentalists had been warning about for years. It's possible that we'll still turn production around and start it rocketing again. Charles C. Mann, writing in Science, quotes experts who believe that in the future a "gigantic, multi-year, multi-billion-dollar scientific effort, a kind of agricultural 'person-on-the-moon project,'" might do the trick. The next great hope of the optimists is genetic engineering, and scientists have indeed managed to induce resistance to pests and disease in some plants. To get more yield, though, a cornstalk must be made to put out another ear, and conventional breeding may have exhausted the possibilities. There's a sense that we're running into walls.
We won't start producing less food. Wheat is not like oil, whose flow from the spigot will simply slow to a trickle one day. But we may be getting to the point where gains will be small and hard to come by. The spectacular increases may be behind us. One researcher told Mann, "Producing higher yields will no longer be like unveiling a new model of a car. We won't be pulling off the sheet and there it is, a two-fold yield increase." Instead the process will be "incremental, torturous, and slow." And there are five billion more of us to come.
So far we're still fed; gas is cheap at the pump; the supermarket grows ever larger. We've been warned again and again about approaching limits, and we've never quite reached them. So maybe—how tempting to believe it!—they don't really exist. For every Paul Ehrlich there's a man like Lawrence Summers, the former World Bank chief economist and current deputy secretary of the Treasury, who writes, "There are no ... limits to carrying capacity of the Earth that are likely to bind at any time in the foreseeable future." And we are talking about the future—nothing can be proved.
But we can calculate risks, figure the odds that each side may be right. Joel Cohen made the most thorough attempt to do so in How Many People Can the Earth Support? Cohen collected and examined every estimate of carrying capacity made in recent decades, from that of a Harvard oceanographer who thought in 1976 that we might have food enough for 40 billion people to that of a Brown University researcher who calculated in 1991 that we might be able to sustain 5.9 billion (our present population), but only if we were principally vegetarians. One study proposed that if photosynthesis was the limiting factor, the earth might support a trillion people; an Australian economist proved, in calculations a decade apart, that we could manage populations of 28 billion and 157 billion. None of the studies is wise enough to examine every variable, to reach by itself the "right" number. When Cohen compared the dozens of studies, however, he uncovered something pretty interesting: the median low value for the planet's carrying capacity was 7.7 billion people, and the median high value was 12 billion. That, of course, is just the range that the UN predicts we will inhabit by the middle of the next century. Cohen wrote,
The human population of the Earth now travels in the zone where a substantial fraction of scholars have estimated upper limits on human population size.... The possibility must be considered seriously that the number of people on the Earth has reached, or will reach within half a century, the maximum number the Earth can support in modes of life that we and our children and their children will choose to want.