Twitter

Saturday, November 23, 2002

A few days ago Damien linked approvingly to an article in the National Review called Taking Environmentalists Seriously. He summarises the authors' argument as follows:

"Part of the reason why people are inconsistent in their positions is that they use different heuristics when faced with different problems, even if they are of a similar nature or the stakes are comparable. This is especially true when it comes to policy, when people are perfectly willing to tolerate some risks (say, a 1-in-1 million chance of dying in a terrorist attack) and not others (a 1-in-1 million chance of getting cancer from a pesticide-sprayed apple, say.) The outcomes are comparable (death in both cases) but one is treated as being worse, or more serious, than the other. Policymaking would be a lot sounder if more realistic attitudes towards costs, benefits and risks were taken into account."

This post, I suspect, is obliquely aimed at me, in light of the little tussle I've had with Damien recently over the Nariva exploration issue.

His point is a solid one: it's merely common sense to consider the costs as well as the benefits of any potential policy or action. But I must (humbly) note that there's a flaw in the reasoning behind the example Damien uses (death due to terrorist attack vs. death due to pesticide-induced cancer). If these hypothetical risks are of the same degree, Damien says, we should treat them as equally serious. But he does not notice their crucial difference in kind. In the case of pesticide-induced cancer, policymakers could act to directly eliminate the cause of the risk: the production & use of the toxic substance in question could be banned, making the risk of death due to that substance effectively zero. The same is not true of death due to terrorist attack, which, as just about every US government official has warned over the last year, is a risk that by its nature cannot be eliminated, though it could potentially be reduced.

Possibility vs. impossibility: a categorical difference. This too must be taken into account.

No comments: