Mitt Romney used to unequivocally believe that the world is warming:

"I believe that climate change is occurring — the reduction in the size of global ice caps is hard to ignore."

Now straightforward climate science skeptic Rick Perry has overtaken him, and Romney is less sure:

"Do I think the world's getting hotter? Yeah, I don't know that but I think that it is."

This hedge is worth pondering in light of a debate I've participated in over science, empiricism and climate change. Kevin Williamson and James Manzi of National Review have argued that Perry's views on climate science don't matter, since the important question is not the scientific fact of climate change but whether we should adopt a policy response. Here's Williamson:

The real question about global warming isn’t whether one computer simulation or another is the better indicator of what our climate will be like a century hence, it is whether such policies as envisioned by the environmentalist-anti-capitalist green coalition are wise.

And here's Manzi's more measured concurrence:

[S]cientific findings in some area are used to justify some related political or moral opinion. Key examples are exactly the topics you touch upon: global warming and evolution. In one example, the indisputable scientific finding that CO2 molecules redirect infrared radiation is used to argue that “science says” we must implement a massive global program of emissions mitigation, when in fact, the argument for this depends upon all kinds of beliefs about the growth of the global economy, Chinese politics, technological developments and so on for something like the next couple of hundred years.

If the "real question is only what policy response should we adopt," then why would Romney, already burdened with a reputation for flip-flopping, hedge his views on that question? It doesn't matter as long as he maintain his opposition to any policy response designed to minimize carbon emissions.

The answer, of course, is that it does matter. The position that climate change is real but not worth the cost of addressing at least exists within the realm of empiricism. Manzi is right that science can't provide us with the answers about the policy we should adopt. But accepting the science is an necessary premise to even entering the cost-benefit discussion. If you think climate science is a fraud, then there's no point in contemplating a policy response, since there is no cost to unlimited carbon emissions.

If you accept climate science, then you accept that there is some cost to carbon emissions. If the projected impact of climate change were to worsen, or if the relative cost of reducing carbon emissions were to fall, then your position might change. Perry's climate skeptic stance is important because it signals that no particular data will ever cause him to change his mind. That's a much more credible promise to conservatives than Romney's previous position, which is why Romney is altering his position.

In any case, the broader import of Perry's skepticism of climate science, along with evolution, is what it tells us about his intellectual style. That he disbelieves evolution, and believes climate science is a fraud cooked by by corrupt grant-seeking scientists, tells us a great deal about how he approaches a range of policy questions. Williamson and Manzi keep avoiding this question but it's pretty vital.