Is it me, Mr. Risk, or are disasters visiting us more regularly? The glib answer is: Blame the media. You're spending too much time watching the Weather Channel or YouTube. Or maybe it's just lousy luck: a gigantic earthquake, tsunami and nuclear meltdown in Japan, Aussie floods, an oil spill in the Gulf of Mexico, a killer hurricane inundating New Orleans, snowstorms, forest fires, pestilence and a dearth of polar bears. Meanwhile, revolutions race through the Middle East like a chicken virus. Jihadism meets financial crises. It's true, electronic media splashes every disaster across flat screens. We live-blog revolutions, follow Tweets from disaster zones, watch a tsunami toss Toyotas like children's blocks via cellphone Internet video. And yet, bathed in media, we miss a deeper truth. Earthquakes and tsunamis cannot be controlled, only mitigated, and have occurred since tectonic plates were invented. They were less a problem when there were fewer of us, though if you got hit by the bubonic plague you wouldn't care. But building nuclear plants in quake zones or populating coastlines represent human choice; so does global warming. We economize, seek efficiencies, launch vast constructions, real and abstract, to the heavens. The more complex our systems -- nukes, derivatives -- the more complex our disasters. They are products of modernity and growth.

The Japanese have a far more sophisticated economy than, say, the Haitians. When an earthquake hits Haiti, the entire society folds like a cardboard box. That won't happen in Japan. The Japanese know about earthquakes and tsunamis. They are affluent, organized, altruistic; they have building codes; they trust each other. Their disaster isn't one of underdevelopment, it's one of hyperdevelopment: A nuclear plant shuts down, backup generators fail, coolant boils off, fuel rods heat up, gas is released: boom. Who would have guessed? (Well, someone did, as someone always does. But who was listening?) Now consider the collapse of the Japanese economic bubble. Its consequences have lasted two decades. Who knows how that economic meltdown interacts in space, time and demography with practices at Tepco's nukes, which waited quietly for pressure to build along silently shifting plates. Ba-boom.

Until we learn to muzzle Mother Nature -- and the very notion reeks of hubris -- we must regulate ourselves. This has proved to be difficult, in part because, human nature being what it is, we figure odds of local disaster are long. So we shrug and live for now, holding cocktail parties on dunes, putting off levee repair, anchoring chateaus on muddy hillsides. Knowing this, how do we minimize the damage? One expert on this subject is Charles Perrow, an emeritus Yale University sociologist who has written books with cheerful titles like "The Next Catastrophe" and "Normal Accidents." Perrow stresses the link between complexity and catastrophe; he juggles redundancies, interdependencies, coupled, decoupled, complex and linear systems. Disasters, both human and natural, are inevitable. Given that, he argues that coping requires not centralized bureaucracies of overseers and systems, but decentralized networks of smaller, more local, less hierarchical units. "Networks are decentralized with minimal concentrations of destructive energy and economic power," he wrote in "The Next Catastrophe." "They are efficient, reliable, and adaptive, which minimizes the dangers of organizational failures." In networks such as the Internet and the electricity grid, consolidating factors always threaten, "but they can be resisted." Politics, he wrote wistfully in 2007, can produce proper regulation. Back then he also worried about Japan's nuclear industry.

Economic crises are rarely considered disasters in the same class as earthquakes or nuclear meltdowns; Perrow himself doesn't discuss market breakdowns. But they are related. There is a kind of inevitability to market breakdowns. Market failures result from organizational failures and are transmitted along complex systems. Our recent crisis featured institutions that had consolidated; too-big-to-fail is a symptom of the risk inherent in consolidation. (Perrow is a big fan of active antitrust and skeptical about declarations of economies of scale.) So is the answer to break up the banks and create smaller, flexible, decentralized players? It's an attractive idea -- though one Washington has so far rejected. And yet, "consolidation" cannot simply be banished. Liquid markets are, at best, highly networked, particularly in our age of electronic communications. They can also be flexible, decentralized, smart; it's not hard to see what excited the efficient-market crowd. But as we know, markets are vulnerable to bubbles, in which many players succumb to the same consolidated view. This explains how a system that seems so decentralized and diverse, so riven with redundant pathways, can prove to be such a superhighway of tightly coupled woe. It suggests how hubris spawns ever-more baroque disasters. And it's proof that we still have a lot to learn about our relations with Mother Nature.