Richard A. Posner, a federal judge and author of "Catastrophe: Risk and Response," said tsunamis in the Indian Ocean had a low probability of occurring, but a high risk of damage if they do occur.

A disaster may occur only every 100 years and kill 40,000 people, Judge Posner said, but "one way to think about it is, that's an average of 400 people killed each year."

Sure, that's one way to look at it. Another way is that 40,000 people died in a matter of hours, many of whom might not have if a warning system was in place.

In any case, Posner's numbers are somewhat off:

The following destructive tsunamis are listed on a data base maintained by the Tsunami Laboratory, Institute of Computational Mathematics and Mathematical Geophysics (http://tsun.sscc.ru/tsulab/20041226tsun.htm)

1. 1797/02/10 Central part of the western Sumatra. The quake was most felt near Padang and in the area within +/-2 deg of equator. Padang was flooded by powerful waves. More then 300 fatalities.

2. 1833/11/24 South coast of the western Sumatra, estimated rupture from 1 S to 6 S latitude. Huge tidal wave flooded all southern part of the western Sumatra. Numerous victims.

3. 1843/01/05 Strong earthquake west of the central Sumatra. Terrible wave came from the south-east and flooded all the coast of the Nias Island. Many fatalities.

4. 1861/02/16 Exceptionally strong earthquake affected all the western coast of Sumatra. Several thousand fatalities.

5. 1883 Krakatau explosion 36,000 fatalities

So, even though the last major Indian Ocean tsunami was 121 years ago, these are not 100-year events, as Posner implies, since there were 5 of them in the course of 86 years, and now there have been 6 of them in the last 207 years.

Since we do not have a precise enough knowledge of how earthquakes work to predict exactly when they will occur and how strong they will be, to assume some kind of regularity over a period as tiny (geologically speaking) as 86 or 121 or 207 years is an extremely faulty assumption. Posner's calculus of risk is also faulty because it assumes that the longer the period of time between events, the less risk there is from the event, when exactly the opposite is true. We may not know exactly when a earthquake will happen, but we can observe the longer there is between them, the more the chance is that the next quake is going to be more rather than less severe, and the resulting tsunami will be more devastating.

I don't know quite how you figure out how damaging a quake or tsunami will be in order to determine what a rational approach to spending money on the problem would be, but then I don't pretend to be able to set up a "pragmatic" system based on cost/benefit analysis to determine what legal and political efforts are worthwhile and which are not justified, as Posner does. Still, it's obvious that the answer isn't simply to divide the number of people who die by the amount of time between events, because it entirely ignores the catastrophic nature of the event.

Suppose, for instance, that I was to give Posner the chance to choose between these two courses of action:

1. Every day for one hundred days I will to apply to his right arm, by pinching it, 1/100th of the energy that would have come from the impact of a bullet shot through that arm by a pistol.

2. I will shoot Mr. Posner in the right arm with a pistol.

Which choice, I wonder, would Posner prefer? Would he be as interested in preventing the first as he would most certainly be in preventing the second? If not, why, since they carry the same cumulative amount of energy?

The death of 400 people a year (actually, the figures work out to more like 600 to 800 people a year, once you factor in those who will die from diseases resulting from the tsunami) is certainly, in my opinion, a strong enough reason to take steps to help prevent something from happening, but I suppose one could legitimately disagree with that opinion. But it strikes me that it takes a particularly callous man to equate that with the death of over 60,000 people in a very short period of time, and the potential death of an equal number in the coming weeks -- not to mention the cost (certainly to be counted in billions) of the necessary relief effort.

Postscript: Jeffery Rosen skewers Posner in this review of his book on Slate:

Posner's thesis when discussing these emotionally laden subjects is as deadpan as his prose: "[T]he tools of economic analysis—in particular cost-benefit analysis—are indispensable to evaluating the possible responses to the catastrophic risks." Unfortunately, assigning precise numerical weights to the costs and benefits of preventing catastrophic risks is a daunting challenge, and Posner's attempts to do so are numbingly technical and ultimately unsatisfying. In the end, balancing liberty and security involves disputed questions of value rather than precisely quantifiable facts—questions that must be resolved not by experts but by politics. [...] [D]ifferences in risk perception can only be resolved through political negotiation. But democratic politics is an enterprise for which Posner has contempt.

Update: Then there's this statement (to be fair, not specifically about the devastation from the tsunami) by Posner, from his blog:

I do not think that foreign aid is a good use of public or private money. All the problems that foreign aid seeks to alleviate are within the power of the recipient countries to solve if they adopt sensible policies.

You know, if those people would just get off their behinds and get to work, they wouldn't have these problems.

Statements like his lend credence to Publius' summary of Posner's "law and economics" system:

Posner and his marching brooms adopt subjective values judgments and dress them up as objective empirical conclusions by using a facade of economic jargon. Like Bush (though in a much more subtle or even unconscious way), Posner is actually flipping empiricism on its head. His methodology (in practice) allows him to begin with a conclusion and subsequently justify it by playing around with the completely unquantifiable concepts of “costs” and “benefits.” Even if he doesn’t actually reach a conclusion, he is still articulating a process (or methodology) that does little more than to give people a way to mask values judgments as empirical conclusions by using economic jargon.

I'm put in mind of William Westmorland's statement, captured in the documentary "Hearts and Minds," that "Orientals don't place the same value on human life as we do," not because I want to accuse Posner of harboring that particular prejudice, but because it exposes the dangers of operating on assumptions which unwittingly reflect errant value judgements.


Update: As of around 9:30am Eastern, CNN International raised the death toll to 71,000, which means that it's rapidly approaching double the number that Posner was so apparently unconcerned about. I wonder if 80,000 deaths would be significant enough for him to think of this as an intrinsically important event? If not, at what number does the death toll start to impress him?

How does he factor in the 2 million in Thailand alone who have been left homeless by the tsunami? Obviously the estimated economic cost of $3.6 billion is easy enough to deal with, you just add it in and wham, bang a new cost/benefit calculation, how jolly!

Posner's viewpoint isn't "pragmatic" or "realistic," and by its cold-bloodedness it's not even entirely rational. What it is, is grossly inhumane.

[Note: More on the Sumatra earthquake in this post below.]