The greens may be right. Or they may be wrong. Predicting the impacts of climate change is an uncertain business, and the more scientists understand, the more complex the effects appear. But a number of things are already apparent.
One is that global warming is unlike almost any other environmental problem. Its causes are international: the greenhouse gases, notably carbon dioxide, are the product of many human activities, from farming and forestry to motoring, and they are released by every country on earth. Its effects, too, are global. And its timescale is immensely long. While most kinds of pollution are felt within a few years - or at least a human lifetime - the greenhouse effect may last for several centuries. It is thus virtually irreversible.
Three years ago, at the Earth Summit in Rio de Janeiro, most of the world's governments signed a treaty agreeing to do something about climate change. From today until 7 April, the signatories hold their first formal meeting, in Berlin.
For those who recall the razzmatazz of Rio, there will be an air of fin de sicle disillusion. For it has become clear that few countries are likely to deliver even the modest promises they made three years ago.
The treaty vaguely bound signatories to stabilise greenhouse gases in the atmosphere at a level that would "prevent dangerous anthropogenic interference with the climate system". The rich countries volunteered to go further, and prepare plans to show how they might restore their output of greenhouse gases to 1990 levels.
In fact, few countries will carry out such plans. Britain may do so, thanks largely to electricity privatisation, which has encouraged huge shifts away from coal-fired to gas-fired generation: natural gas releases almost half as much carbon dioxide as coal when it burns. But generally, a decade of low energy prices has removed the incentive to economise on fossil fuels, and governments have had little success in restoring it by pushing through energy taxes.
In fact, even if the rich countries did meet their targets, the impact on the build-up of carbon dioxide would be small. Most of the growth in the demand for energy will come from developing countries. Their current energy use is small by first-world standards, but rising fast. By the middle of the next century, China's carbon-dioxide output will probably exceed that of the entire OECD.
Moreover, a growing proportion of the world's energy use over the next half century will involve burning coal, the most carbon-rich of all fuels. At present, coal is a less important source of global energy than oil. But it accounts for 70 per cent of the heat content of the world's fossil- fuel reserves. Moreover, about one-third of the world's known coal reserves are under China, and quite a lot under the former Soviet Union. It is hard to see how those relatively poor countries will be persuaded to leave their coal unburnt. Yet that is what must happen if the build-up of carbon dioxide in the atmosphere is to be stopped.
Part of the difficulty is that there is no easy technological fix that would allow coal to be burnt but release less carbon dioxide. The rich world scrubs sulphur dioxide from coal gases before they escape from factory chimneys, and has therefore achieved dramatic reductions in sulphur dioxide emissions and acid rain. But no satisfactory technology can yet strip the carbon dioxide from coal smoke. And many of the technologies that reduce the output of other energy-related pollutants (such as nitrous oxide) involve the burning of more fuel, not less.
In the long term, the only real option is to develop fuels that burn no carbon: nuclear, hydro, solar and so forth. The problem here is that some of these options (nuclear, hydro) impose damage on future generations that may be just as unwelcome as a warmer climate. Posterity might even prefer more storms to more nuclear waste. As for solar power, its future depends on its economics. Yet there is probably more chance that coal will become cheaper to burn (because it is already widely used commercially, and so its efficient use attracts commercial research) than that solar power will undercut it.
The upshot is that, for many countries, the costs of measures to slow down global warming will be prohibitively high. They will agree to undertake them only if the rest of the world can find some way to compensate them. But if, say, the rich world has to bribe China not to burn its coal, the cost to the rich countries of averting global warming will also be immense.
Will the rich countries be prepared to bear such a burden? The costs of measures that might slow down climate change are certain, and have to be incurred soon. The longer we postpone action, the more gas builds up in the atmosphere, and the more the world becomes locked in to high- energy technologies.
Yet the pay-off is far down the track, and its value depends on knowing how harmful global warming will be. Work by William Cline, a scrupulous and scientifically literate American economist, suggests that the benefits of taking action do not overtake the costs until about 2150. And Mr Cline sees global warming largely in terms of costs. Yet it is inconceivable that a change of such complexity will not bring gains (faster growth of crops and trees? warmer climates in disagreeably cold lands?) as well as losses.
Given the difficulties of doing something about climate change, should we try? Some measures are certainly worth taking because they make sense in their own right. Many countries subsidise energy consumption. Britain offers tax breaks on company cars. Germany subsidises coal. Many developing countries sell electricity for less than it costs to produce. Removing such subsidies would make the economy work more efficiently and benefit the environment, too.
Indeed, wise governments should go further, and deliberately shift the tax burden away from earning and saving (which should surely be encouraged, not penalised), towards energy consumption.
Beyond that, governments should do little. The most rational course is to adapt to climate change, when it happens. On past trends, most countries will by then be richer, and so better able to afford to build sea walls or develop drought-resistant species of plants. Money that might now be spent on curbing carbon-dioxide output can be invested instead, either in preventing other kinds of environmental damage or in creating productive assets that will generate future income to pay for adaptation. The richer countries are, the more likely they are to be able to protect their citizens from the impact of climate change.
Moreover, once climate change is clearly under way it will be more apparent, as it now is not - what needs to be done and where. Most of the decisions involved in adapting to change will be taken and paid for by the private sector rather than by government. Above all, adapting does not require the laborious negotiation of complex international treaties.
Adaptation is especially appropriate for poor countries, once they have taken all the low-cost and no-cost measures they can find. Given the scarcity of capital, it makes good sense for them to delay investing in expensive ways to curb carbon-dioxide output. Future economic growth is likely to make them rich enough to offset those effects of climate change that cannot be prevented.
Many people find such arguments unpalatable. To many environmentalists, the idea of adapting to climate change, instead of striving to minimise it, will sound wilfully irresponsible.
Yet the harsh reality is that plenty of other kinds of environmental damage deserve greater priority. Water pollution kills more people than global warming is likely to. Soil erosion leaves more people hungry. The loss of species that is now happening at a cataclysmic rate in many parts of the world is just as irreversible. Deforestation has locally (and, perhaps, internationally) more dramatic effects on climate. The world has only so much wealth to devote to solving environmental problems. Many of these deserve greater priority than global warming.
The writer is on the staff of the `Economist', and the author of `Costing the Earth'. Her latest book, `Green, Inc' will be published by Earthscan and Island Press in May.