Mainstream debate about global warming has moved on from the question of what needs to be done to when it needs to be done. There is broad agreement that heat-trapping emissions need to be cut. That is prompted by the science. But there is furious argument over when to start doing it. That is an especially tricky one because it's where the science ends and politics and economics take over.
And while science may be imprecise and incomplete, it is usually a better guide than politics, which Ambrose Bierce defined in his 1911 classic The Devil's Dictionary as "a strife of interests masquerading as a contest of principles". And physical science is generally more dependable than economics, the so-called dismal science. As George Bernard Shaw said: "If all economists were laid end to end, they would not reach a conclusion."
Yet these are the disciplines, or perhaps indisciplines, that societies are obliged to work with. So what do they tell us in answering this question - should a country act unilaterally and cut its carbon emissions even if other countries do not?
Modern history gives us two models for weighing the merit of unilateral action in a highly interdependent world.
One is from the world of military strategy and the other is from trade policy. In the case of military strategy, the Cold War posed the question of whether one of the superpowers should start cutting its nuclear arsenal even if the other did not.
The argument in favour of unilateral disarmament said that doing so would break the cycle of confrontation and escalation, ease tensions, save money, and move the superpowers to a better relationship.
But reality showed otherwise. Unilateral disarmament turned out to be the equivalent of surrender. The Soviet Union abandoned the arms race and lost its empire. The lesson here is clear - unilateral action is a bad idea.
In the case of trade, one of the big questions of the last century was whether a country should cut its protective tariffs even if other countries did not.
We shouldn't cut the tariffs on our car industry, for example, unless other countries do. Because if we did lower our protective barriers unilaterally, other countries will flood our market and ruin our industry. To this day, global trade negotiations are based on this concept of reciprocity - I will if you will.
But in the real world, experience showed that it doesn't work that way. Even if other countries keep their protective barriers in place, the country that liberalises its trade reaps big benefits of efficiency and competitiveness. So the experience from trade is that unilateral action is generally a good idea.
How do these examples help us understand whether to start cutting carbon emissions unilaterally?
The eminent economist Ross Garnaut, the man appointed by the Rudd Government to report on the economics of climate change, put it to me this way: "With trade liberalisation, you are going to benefit whatever the rest of the world does. With unilateral disarmament, you open yourself to a security risk, but at least you save money in the national budget.
"With climate change, it's even worse than unilateral disarmament - if you are the only country to move, there is only cost."
If Australia were to impose a significant price on the right to emit carbon and other countries did not, companies here would be put at a disadvantage.
Some, like the electricity companies, are stuck and would have to sit tight and put up with it. But the ones that sell so-called tradeable goods could just take their factories overseas, and their jobs with them.
And if Australia managed to curb its greenhouse emissions as a result, so what? Accounting for all of 1.5 per cent of global carbon output, it would make no discernable difference whatsoever. Australia would have punished itself economically for no good purpose environmentally.
In his report to the Federal Government released last week, Garnaut likens this to the prisoner's dilemma.
Consider: Two people are arrested for a crime. They are questioned separately. Each may give evidence against the other, or say nothing. If both say nothing, both go free for lack of evidence. But if one rats on the other, he goes free and the other is severely punished. If both give evidence, both are severely punished.
Not knowing or trusting what the other may do, the best individual - or local - strategy is to give evidence against your friend. In climate change, this translates into a strategy of allowing other countries to act while you do not. Or as Garnaut puts it: "With climate change, a country can be rewarded for cheating."
But the best overall - or global - strategy for the prisoners is for them to each trust the other and for neither to give evidence. This way, both go free. Or, in climate change terms, for all nations to agree to act at broadly the same time in roughly the same manner.
This was the aspiration that Kevin Rudd voiced at the G8 meeting in Japan this week, urging the developing countries and the developed to act in unison. But the response from China and India starkly revealed that, even with the US signing up to a communiqué that supports the idea of cutting emissions in half by 2050, the two biggest emerging economies are not interested in acting at the moment.
So does that mean Rudd should take Brendan Nelson's advice and not even start the emissions trading scheme he has pledged to begin from 2010?
Rudd's thinking is that he will proceed with the program. How is this good politics? With the Greens and the environmental movement clamouring for more action from Rudd, and the Liberal Party urging less, Rudd will position himself in between these voices of the Left and the Right at what he calls the responsible reforming centre of Australian politics.
The emissions trading system can start off assigning a low price to the right to emit a tonne of carbon. The Government can set out on a gently-sloping upward trajectory of carbon-curbing. This would keep all Australia's options open as international negotiations proceed.
The Australian Business Roundtable on Climate Change, including firms such as BP, Westpac, Origin Energy and Visy, has urged the Government to act swiftly because "the longer we delay acting, the more expensive it becomes for business and for the wider Australian economy".
Similarly, an impressive group of 1700 US scientists and economists, including six Nobel laureates, has told the US Government that "the longer we wait, the harder and more costly it will be" to curb carbon emissions.
"The most risky thing we can do is nothing," they wrote.
This is true, but, as Garnaut points out, "only if other countries act eventually". And if not, we should forget unilaterally cutting carbon emissions and start unilaterally building seawalls.
Peter Hartcher is the Herald's political editor.
