It's a soggy Monday night but the pews in one of Ottawa's most spacious churches are overflowing with believers. "We have tried to assume the position of the gods," the angry man at the lectern thunders, "without the knowledge to manage our ecological footprint." No, the speaker is not a preacher, at least not a preacher of the conventional sort. He is David Suzuki, scientist, environmentalist, icon. The natural systems that sustain us are infinitely complex, he tells the worshipful audience. We are only beginning to understand them and we cannot possibly predict what effect the actions and technologies of almost seven billion people will have on them. We must be humble. We must be cautious and reverent. "We don't know enough to take the place of the gods," he proclaims. It's a familiar theme, which is appropriate because Suzuki, at 74, is summing up his life's work -- his "legacy," as he puts it in the title of his new book. Suzuki delivers another familiar theme this night. He illustrates it with a thought experiment. Imagine a test tube filled with food. That's the Earth, he says. Now introduce a single bacterium to that test tube and let it grow exponentially. In the first minute, one bacterium becomes two bacteria. In the second minute, two become four. Four become eight. Eight become 16. If it takes one hour for the bacteria to multiply until they fill the entire test tube and there's no more food -- and the bacteria all die -- when will the test tube be exactly half full of food and half full of bacteria? In the 59th minute. Which is strange because at that moment things look fine. But the very next minute, catastrophe strikes. "Every scientist I talk to agrees with me," Suzuki declares, "that we're already past the 59th minute." We must drastically change the way we live, immediately, or we are doomed. Neither of these themes is unique to Suzuki. Indeed, they are standard fare among environmentalists. And therein lies a little-recognized paradox. I recently wrote a book called Future Babble (to be released Oct. 12), which is about expert predictions, why they fail, and why we believe them anyway. The experience of sifting through heaps and heaps of failed predictions has made me quite sympathetic to Suzuki's first theme of humility. We truly are awful at foreseeing what is to come. And there's little reason to think we'll get much better. Indeed, key properties of complex systems make prediction inherently impractical or even impossible. We really should be humble. And cautious. But how can a humble and cautious man say we are "past the 59th minute"? To know that, one must fully understand all those complex natural systems -- to say nothing of social systems -- and be able to see how they will develop in the future. Monday night, Suzuki said this is now possible thanks to "scientists and supercomputers" -- the same scientists with supercomputers who "don't know enough to take the place of the gods." This stunning contradiction shows up most clearly when environmentalists talk about climate change. On the one hand, greens oppose geo-engineering schemes -- deliberate attempts to alter the atmosphere to counteract the effects of climate change -- on the grounds that we cannot possibly predict the consequences of our actions. But they also treat forecasts of what will happen if humanity doesn't curtail carbon dioxide emissions as perfectly reliable glimpses of the future. That makes no sense. Either we can reliably predict the effect human actions have on climate and the natural world or we cannot. Which is it? Environmental activism is steeped in this contradiction. A 1992 statement by leading scientists warned of "unpredictable collapses of critical biological systems" and insisted -- rightly, I believe --that this uncertainty strengthens the case for action. But then the statement declared humanity had "no more than one or a few decades" to make the necessary changes. If the collapses are unpredictable, how could they possibly know how much time we have to avert them? This isn't humility. It's hubris. To see how foolish it is, consider the history of similar forecasts. The famous and revered 1972 report "Limits to Growth" -- the first to dazzle the world with "scientists and supercomputers" -- opened with a statement from UN secretary general U Thant that the world has "perhaps 10 years left" to stave off calamity. The report itself offered computer-generated forecasts that supported this claim: Radical changes implemented in the 1970s would avert disaster, the forecasts showed, but it would be too late if the same changes were not implemented until the year 2000. Either that report was wrong then, or Suzuki is wrong now. There's no way around it. But Suzuki seems not to care for historical references like this. He even lionizes the ecologist Paul Ehrlich, the author of 1968 smash The Population Bomb, who made a long list of predictions that failed, using essentially the same analysis as Suzuki. In 1974, for example, Ehrlich argued that growing resource shortages would make Americans and others drastically poorer in the years ahead, putting an end to consumerism and the conventional economics of growth. "We are facing, within the next three decades, the disintegration of nation-states infected with growthmania," Ehrlich wrote. Thus, it was a little odd to listen on Monday as David Suzuki railed against the consumerism and economic growth which Ehrlich said would be swept away long ago. And then there were the two books which introduced a young Paul Ehrlich to ecology. Both William Vogt's Road to Survival and Fairfield Osborn's Our Plundered Planet were published in 1948. Both were influential best-sellers. Both made arguments similar to Ehrlich's and Suzuki's. And both were stuffed with predictions that flopped: "The most critical danger is that we shall not realize how short we are of that one unrenewable resource -- time," wrote Vogt. "If we wait until next year, or the next decade, to push our search for a solution, then our fate may well be sealed." We never learn. Earlier this year, scientists delivered the startling news that despite increases in sea level caused by climate change, only four out of 27 small Pacific islands surveyed were smaller than they had been in the 1950s. The remaining 23 had either stayed the same size or grown, thanks to offsetting accumulations of coral debris. Water rises; islands get smaller. It seems like the simplest thing in the world to predict. Nature constantly surprises us, Suzuki said Monday night. "I hope we can learn some humility." Indeed.