A. Scott Crawford made some good points in his comment on Shannon’s “Critical Failure” post. A lot of commentators seem to assume that governments can just order people to leave and that they will do it. A lot of them will — almost everyone in New Orleans left — but some will not, no matter what the govt does. Even if the NO municipality or LA state govt had provided transportation out of town, some people would have stayed. In hindsight, evacuation looks like the obvious choice, but before the storm arrives you can’t know. There have been plenty of false alarms in the past, and evacuation is costly.
Initiating a mandatory mass evacuation is a bit like ordering a military mobilization in a small country like Israel: you have to be damn sure the risk you are responding to is extreme, because the cost is huge: you are essentially shutting down the society for days at least. And the cost isn’t just monetary: evacuations, like hurricanes, kill people, both directly (via auto accidents, heart attacks, etc.) and, in the event the predicted deluge doesn’t materialize, by convincing some residents that next time they should stay in town and ride out the storm. And for some citizens, like the guy pictured in the NYT article guarding his auto-repair shop, fleeing means that they probably go out of business, because in a place like NO they will be cleaned out by looters. How can you force such people to leave on the strength of a fallible weather forecast? By the time the storm is close enough for the danger to be certain, it is probably going to be too late to implement a successful mass-evacuation. This strikes me as an insoluble problem.
I suspect that the best that NO could have done in this case would have been, as many people have suggested, to offer transport out of town to people who lacked the means to leave on their own. But this is all hindsight. It’s impossible to know how many people would have been saved if the City had actually done it. Even with the best of planning, it seems likely that many citizens would have chosen to stay behind and would have been killed.
Not all problems are soluble, not all risks can be hedged at acceptable cost. NO’s central problem was its location. Levee reinforcement, evacuation plans and other risk-abatement measures would have helped, but everyone always knew that a direct hit by a big enough storm would be a catastrophe. Accepting that possibility was an unavoidable part of living there, just as accepting the possibility of earthquakes is part of living in California.
Perhaps, in the long run, the evacuation model of hurricane civil defense, which depends critically on competent local government (which, it almost goes without saying, NO has not had) will be supplanted by emphasis on improved building codes and other technological measures that might eventually minimize the need for evacuations, and by increased emphasis on citizen self-reliance during emergencies. Perhaps, but NO has a dysfunctional third-worldish political culture that may make it difficult to implement reforms of any type. The best that we can hope for may be that New Orleans will be rebuilt with more-robust technology, and that Americans who live along our hurricane-prone coasts will take storm safety more seriously for the next generation or so.
UPDATE: Here’s a take on the current situation that I think makes a lot of sense.
UPDATE2: I combined and edited the 2nd and 3rd sentences of the last paragraph 9/8/2005.
(A related post is here.)