Yesterday I asked you to analyze a report presented at the Heartland Institute’s conference of global-warming skeptics. A lot of readers had the same reaction I did after I read the report and attended the conference yesterday: There are some interesting points here, but who knows? The skeptics point to some genuine discrepancies between the climate models and what’s actually happened; they’re probably right in criticizing the United Nations’ I.P.C.C. for not paying enough attention to the impact of solar variations on the Earth’s climate. But climate is so complicated, and cuts across so many scientific disciplines, that it’s impossible to know which discrepancies or which variables are really important.

Considering how many false alarms have been raised previously by scientists (the “population crisis,” the “energy crisis,” the “cancer epidemic” from synthetic chemicals), I wouldn’t be surprised if the predictions of global warming turn out to be wrong or greatly exaggerated. Scientists are prone to herd thinking — informational cascades— and this danger is particularly acute when they have to rely on so many people outside their field to assess a topic as large as climate change. So I’m glad to see contrarians raising awkward questions and pointing out weaknesses in predictions made with computer models. As S. Fred Singer, the editor of the skeptics’ report, said at the conference yesterday: “Models are very nice, but they’re not reality and they’re not evidence.”

But scientific uncertainty doesn’t necessarily imply doing nothing. Given the sum of the evidence today, I think the risk of global warming is sufficient to warrant buying insurance. The question then becomes how much insurance and what kind, and here I think the skeptics are especially useful in challenging what’s mistakenly called “the scientific consensus”: that if you believe global warming is a risk, you should be supporting drastic cuts in carbon emissions and expanded versions of the Kyoto Protocol.

That may be the policy urged by many scientists, particularly the most vocal ones in the climate-change debate, but it’s not a consensus based on climate research. It’s a conflation of two separate issues. You can fear global warming and favor adaptation policies rather than emissions cuts because you think adaptation is more cost-effective and technologically practical. You can fear global warming and be opposed to Kyoto-style cap-and-trade systems because you think they’re too expensive and too ineffective, as Kenneth Green argued yesterday at the skeptics’ conference yesterday in explaining why he favored a carbon tax instead.

Mr. Green, a resident scholar at the conservative American Enterprise Institute, took a little grief from fellow conservatives at the conference for his endorsement of a carbon tax (even though he took care to advocate a revenue-neutral tax whose proceeds would be rebated or offset by other tax cuts). His critics argued that even if a carbon tax may made theoretical sense, it would probably be implemented so poorly (by politicians who would use it for doling out favors and financing pet projects) that it would do more harm than good. One fellow panelist, Robert L. Bradley Jr., the president of the Institute for Energy Research in Houston, jokingly offered to buy Mr. Green dinner if he’d announce his support for a carbon tax in theory, just not in practice.

“I guess I’ll be buying my own dinner,” Mr. Green replied. While acknowledging the dangers of a badly implemented tax, he suggested that his fellow conservatives recognize it as the best alternative, politically and economically, to the other polices being debated in Washington: the billions in subsidies being handed out to solar-power and wind-power companies, the cap-and-trade systems being promoted on Capitol Hill and by the presidential candidates. As Mr. Green has pointed out here, setting up a trading system in carbon credits encourages corruption when it comes to setting caps on emissions and handing out credits to industries that emit carbon:

Companies have incentives both to overstate historical emissions, and to exaggerate the benefits of new technologies to generate bogus emissions that become ready cash. Experience in both the U.S. and Europe shows that firms usually get away with it: validating historic emissions is nearly impossible. And governments won’t look very hard–wanting to appear green, they have strong incentives to turn their eyes away from carbon credit malfeasance.

But these problems help explain why a cap-and-trade system is so popular in Washington: companies and lobbyists see it as a chance to make money, and legislators see a chance to appear virtuous without imposing any obvious costs on voters, as a carbon tax would. Given those political realities, Mr. Green isn’t expecting to see the tax enacted soon. But if Congress and the next president go for a cap-and-trade system instead, he predicted, it’s so sure to fail that they’ll end up considering a tax eventually.

(For more on the conference, see my colleague Andy Revkin’s article and his blog.)