washingtonpost.com  > Print Edition > Sunday Sections > Book World
Global Issues

Worst-Case Scenarios

Reviewed by Graham Allison
Sunday, January 9, 2005; Page BW05

CATASTROPHE

Risk and Response

By Richard A. Posner. Oxford Univ. 322 pp. $28


Some 250 million years ago, an asteroid or comet struck Earth, causing mass extinction. (Nasa Illustration)

Horrific as the tsunamis that ravaged South Asia were, Richard A. Posner is worried about calamities many thousand times worse. Finding the puzzles presented by his day job on the U.S. Court of Appeals for the Seventh Circuit insufficiently challenging, Posner has seized upon the risk of mega-catastrophes, particularly events that threaten to extinguish all human life on Earth. His central question in Catastrophe is: How should society respond to the possibility of global warming, nuclear terrorist attacks, planet-obliterating asteroids and similar disasters?

Posner groups catastrophes into four types: natural catastrophes (e.g., a massive asteroid collision); catastrophic accidents caused by unfettered scientific exploration (e.g., a "strangelet" scenario in which hyper-dense quarks produced in a physicist's lab compress the Earth in seconds); unintended byproducts of technological progress (e.g., abrupt climate change); and intentional catastrophes (nuclear terrorism, bioterrorism or even nuclear winter). Catastrophe is worth the price of the book simply for Posner's lively and readable summary of the apocalyptic dystopias that serious scientists judge to be possible. (For a similarly sobering account from one of the world's premier astrophysicists, see Martin Rees's 2003 book Our Final Hour.)

But how can sophisticated citizens think productively about such staggering possibilities? Indeed, is it even worth worrying about such appalling but unlikely tragedies? While the first two chapters of Catastrophe present fascinating information for general readers, chapters three and four will likely leave statistic-phobes cold. Posner advocates using basic cost-benefit analysis -- a widely applied technique developed from economics for carefully weighing pros and cons -- not only to structure the questions, but also to derive some answers. Moreover, based on those answers, Posner recommends that Washington act now. Reviewing the best scientific estimates of likelihoods and consequences, he concludes that the U.S. government should seriously consider prohibiting some types of scientific research, creating new international oversight agencies and imposing blanket restrictions on the study of dangerous subjects by all students from suspect nations. No orthodox University of Chicago conservative he.

The cost-benefit framework Posner recommends is good as far as it goes. Unfortunately, it doesn't go very far in dealing with this class of problems. Students of cost-benefit analysis have long recognized this tool's limits in dealing with extremes. In the classic social science formula, risk equals probability times consequences. Since a species-extinguishing event like Earth colliding with a huge asteroid would be infinitely bad, even the slightest probability of such an event occurring would swamp any cost-benefit calculus.

Posner acknowledges these theoretical difficulties but attempts to finesse them. He does so by making heroic but arbitrary assumptions that leave his conclusions unpersuasive.

If cost-benefit analysis cannot provide convincing answers to questions about what our society should be doing about potential mega-catastrophes, where else can we look? My best suggestion is that we draw lessons from recent practice. The most instructive case is the Cold War threat of total annihilation in nuclear war. For decades, American policymakers faced a grim reality: Soviet nuclear bombs that could destroy American cities and kill millions. This possibility drove policymakers to a categorical imperative: Do everything feasible to prevent such a war. President Reagan expressed it best in his oft-quoted bumper sticker: "A nuclear war cannot be won and must therefore never be fought." To prevent that, Republican and Democratic administrations, with broad public support, spent more than $2 trillion developing and deploying missiles, nuclear weapons, satellite-alert systems, elaborate command-and-control systems -- essentially everything technically feasible to prevent an unthinkable outcome.

In that effort, American leaders recognized the danger that they might compromise higher values. (Remember the old debate about "Better Red than Dead"?) But President Kennedy rightly rejected that choice, arguing that America's goal was "not peace at the expense of freedom, but both peace and freedom." America's Cold War strategy combined several elements: strength to deter any conceivable Soviet attack; imaginativeness in a global battle for hearts and minds, which was aimed at undermining the appeal of Moscow's Marxist ideology; and a readiness to cooperate with Soviet leaders in ending the Cold War with not a bang but a whimper.

This precedent provides clues for addressing Posner's catastrophic challenges, perhaps most importantly nuclear terrorism. In the first debate of the 2004 presidential campaign, both President Bush and Sen. John Kerry identified nuclear terrorism as "the single most serious threat to the national security of the United States." What should the U.S. government now do to combat that threat? Take a page from the Cold War's imperative by doing everything possible that does not otherwise compromise a higher value or interest -- and do it as quickly as possible, without assuming we have the luxury of time. That would mean locking down all nuclear weapons and materials as securely as gold in Fort Knox; stopping additional states like Iran from acquiring the capacity to make nuclear bombs; and preventing North Korea from becoming a "Nukes 'R' Us."

While the application of cost-benefit analysis Posner so ably champions cannot provide answers to the species-threatening specters he discusses, there are scores of less catastrophic, more urgent and more likely threats -- the recent tsunamis, for example -- that cry out for clarification by applying this methodology. It is now clear that investments costing less than one percent of the price of the current relief efforts in Asia could have provided effective warnings, saving thousands of lives. The larger challenge that Posner and others of his tribe should address is how to narrow the gap between what cost-benefit logic compels and what the U.S. government is doing. •

Graham Allison is director of the Belfer Center for Science and International Affairs at Harvard University's Kennedy School of Government. His new book is "Nuclear Terrorism: The Ultimate Preventable Catastrophe."


© 2005 The Washington Post Company