The Perpetual Three-Dot Column
The Perpetual Three-Dot Column
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

by Jesse Walker

Wednesday, September 07, 2005
COMMAND AND CONTROL: I finished a Katrina column last night, and it should go up on the Reason site later today. One element of the story that space constrained me from emphasizing as much as I'd like is the folly of a command and control approach to disaster management, a point that was stressed to me by one of the founding fathers of disaster research,
E.L. Quarantelli. Because I think it's interesting, I'm going to publish that extract from our interview here:

Walker: How would you contrast the command and control model with what you think is needed?

Quarantelli: This is the notion that you do things from the top, that someone should be in charge, and so forth. Forget about what happened in New Orleans; just talk about disaster in general. The notion that at a time of a major disaster, someone could be at the top and straighten things out doesn't make a lot of sense, because one of the things lacking at a time of disaster is information. A lot of decentralized things need to be going on. In terms of prior planning and managing, you have to think of how lower levels will be involved.

The other thing, also, is you have to involve the public, the citizens. You don't plan for them, you plan with them -- which doesn't cross people in [the Department of Homeland Security]'s minds, apparently. That's a very important thing. This has nothing to do with democratic ideology or anything else of that kind. It has to do with the fact that if you get the perspective of the people from the bottom up, you get an important perspective, because that's the way they're going to think about the world. The notion you can come in and impose things from the top on citizens, much less on local emergency organizations, which of course are more knowledgeable about their own areas than anybody else around -- it's simply not gonna work.

The command and control model doesn't work in that kind of situation. It can work in the military, up to a certain extent, because it's set up hierarchically. But one of the reasons the military has always been kept out of disaster activity -- apart from the fact that they have a rather important mission, protecting the country -- is that they in terms of their operations can work well, but if they have to start to integrate with private organizations and religious groups and so forth, these groups are not used to being ordered around. The general can order his own troops around, but he's going to have difficulty telling the Salvation Army or the Red Cross or religious groups things....

Walker: So what does the decentralized model look like?

Quarantelli: It's more than simply decentralization. You do a risk analysis -- which apparently they did do down in New Orleans -- figure out what the problems are, and then say: OK, who has responsibility for what? If you're going to have large blocks of civilians or the general public involved, what do they know, what do they need to know, what do they think they know about what's going on? You have to take all those things into account, and then you feed them into an overall model -- I'm not talking about a detailed model, I'm talking about who does what.

At the height of the emergency, you have to think about decentralization. And that's not even peculiar to [disasters] -- for example, a police department necessarily will be very decentralized. You can't control every police car out there; you can't control every patrolman on the beat. A police department leaves a lot of discretion up to the car or to the local officer, because you can't simply control things from the top down. And this is an organization with a semi-military structure.

So that's what I mean by a decentralized model. You have to leave discretion up to the people who are on the scene, who have the problem facing them.

posted by Jesse 10:30 AM
. . .

. . .

For past entries, click here.

. . .