• Choose a category
• All
• Classic-Fiction

By Root 905 0

that a specific location will be impacted by a natural disaster at a given point in time is usually independent of whether it was impacted by a disaster at a previous point—even for hazards that follow known cycles such as earthquakes. Yet, residents and communities are notorious for believing that low-probability events are highly temporally correlated, as when the survivors of a 1-in-100-year flood who quickly resettle in the flood plain claim that they are “safe” for another 99 years. But the opposite—and equally wrong—conjecture often arises as well; an obvious example is the insurance company that dramatically raises its premiums after a busy hurricane season (if allowed by regulation) in the belief that this signals a long-term trend.

THE COUNTERPRODUCTIVE EFFECTS OF KNOWLEDGE

The above accounts still leave us with one remaining puzzle. In principle, the better we are at recalling the past and the more advanced our scientific knowledge about hazards and mitigation, the lower our losses should be from them. But the reality seems to be the opposite. For example, while hurricanes have plagued communities along the United States’ Atlantic and Gulf coasts for decades, the inflation-adjusted losses imposed by recent storms dwarf those imposed by past storms. The reason is simple: It is not necessarily because all the recent storms have been more intense than those in the past but, rather, because we have been increasingly willing to put valuable property at risk. Hence, we are presented with something of a paradox of knowledge: The more that we know about hazards and how to survive them, the more, it seems, we are willing to place ourselves and our property in harm’s way. When it comes to protecting against natural hazards, more knowledge—perhaps because it makes us more comfortable—can be a dangerous thing.

A good example of this paradox of knowledge is the fate of the town of Indianola, Texas. In 1875, Indianola was a thriving Gulf Coast town of 5,000 that was home to the second-most important port in Texas, behind Galveston. In September of that year, however, Indianola was destroyed by a major hurricane and the accompanying tidal surge. The community rapidly recovered from this disaster and rebuilt—only to see the town destroyed yet again by a hurricane in 1886. Lacking any knowledge of hurricane climatology, and knowing little about how such a disaster might be prevented from happening yet again, the surviving residents reached the only sensible conclusion they could: They moved elsewhere.

In contrast, the far more advanced knowledge that we have today about hurricane risks and mitigation tends to produce something of the opposite reaction among communities and individuals: The belief that storms and other hazards are both survivable and predictable has induced aggregate risk taking, with the dominant migration pattern being to coastal zones that would have been deemed too unsafe for sensible development in earlier, less knowledgeable times.

What explains this effect? Past research on risk taking and learning in other domains suggests three complementary mechanisms. The first is that the tendency for knowledge to foster risk taking can be seen as the mirror image of the well-known tendency for individuals to avoid risky decisions when the odds of different outcomes are unknown or ambiguous—as Ayse Öncüler describes in this book. The more precisely that we know the odds of a gamble we are facing—such as the odds that our home will be hit by a hurricane or an earthquake in a given year—the more amenable we may be to the prospect of playing that gamble.

A second, complementary, mechanism is the effect of controllability: our willingness to be more open to taking on risks when we believe—however wrongly—that we have control over the outcome. Just as people are often willing to wager more in a dice game when they think they can control the roll, greater confidence in our ability to control losses from hazards by engineering risk reduction measures (even if we don’t actually follow through) may contribute to an