washingtonpost.com
Follow the Leader, or Think Like a Starfish?

Monday, January 1, 2007

"The Starfish and the Spider: The Unstoppable Power of Leaderless Organizations" describes the oft-overlooked reason Skype, Craigslist and al-Qaeda have prospered under assault from the seemingly more powerful.

The book's authors, high-tech entrepreneurs Rod A. Beckstrom and Ori Brafman, say these groups' decentralized structure makes them more nimble, whether in the marketplace or on the battlefield. User-driven, they distribute decision making among all members, who do not need word from headquarters to function. If a portion of the organization is defeated, the whole survives and recovers, as a starfish regrows an arm if it is severed.

More commonplace in industry and government are "spider" organizations: They function within a rigid, centralized structure, with the leader calling the shots. One good blow to the head cripples or kills a spider.

The spiders are the establishment: the telephone companies undercut by Skype's free Internet calling; newspaper classified sections threatened by the free, local classified listings on Craigslist; and the U.S. government apparatus struggling with al-Qaeda.

The trick to beating a starfish, the authors say, is to become -- or at least think -- like one. We spoke with Beckstrom about how the book's theories might be used to create a more efficient government, and a better strategy in Iraq.

-- Elizabeth Williamson

Q You say that when our founding fathers sculpted our Constitution, they put the government in the "sweet spot," between centralized and decentralized. Are we still there?

AWe've drifted strongly back toward centralization over time as a country, and of course we wobble back and forth a little bit. One of the biggest examples was after 9/11, when we took all the different police forces and intelligence forces and put them all under Homeland Security. That was a major centralization move, and typical: When a fairly centralized player gets attacked by a decentralized force, like al-Qaeda, the first reaction is to centralize further, and that's usually a strategic mistake.

So how do we get back into the sweet spot?

One way is to push responsibility back to the state governments. In some areas you can decentralize by outsourcing services further. One of the ultimate moves in terms of combating terrorism is to have the government use more Special Operations forces, which tend to be more decentralized, working in small teams that in general are given a high level of autonomy. . . . I gave a presentation at Stanford in 2004 to 50 CEOs from around the world. One CEO took it back to a head of state in a Middle Eastern country to the top levels of government. Based on it they decided to start their own local special operations in a selected city, and found it to be much more effective than their traditional, centralized counterterrorism operation -- at a very small fraction of the cost.

The people living in any community have the best sense of what is really going on in that community. They have local intelligence. The best information is at the edge of a network . . . where people are bringing what they want into the network and taking out what they want, without any centralized control.

That sounds chaotic. Is it?

Things can be chaotic when we decentralize to an extreme. As I asked someone in that country, how do you know the counterterrorism cells aren't using their power for the wrong purposes? Their response was "We've put in checks and balances to monitor and review their operations the same way you guys do in the U.S." When we create these powerful decentralized networks, we have to manage them by monitoring and measuring activity and what's taking place in the networks.

A simple example comes from YouTube. YouTube is able to keep a record of every person who logs in and contributes. One of the huge strategic advantages that YouTube has versus ABC, CBS or NBC, all centralized networks, [is] . . . they know who exactly is watching what, when, because they have our IP address.

How would we apply these principles to the federal government?

Let's go to the State Department, responsible for our formal diplomacy. The current administration has said we want businesses to do a lot of this bridging work for us. [State] recently took a group [of industry representatives] to Lebanon and said, "We need your help to rebuild this country." Instead of having the government itself directly fund or manage reconstruction . . . they're trying to engage businesses to do that. That's just one small example of the government trying to decentralize its approach.

We can also take this same issue of diplomacy in another related direction: There's an explosion now in blogs and online communities to support peace dialogue and bridges across countries. That's leading to more person-to-person connections that are rewiring the world.

How could we apply the principles to the Iraq war strategy?

This is an extremely fractured war with a lot of different fronts, with a lot of different groups. Taking out the "boss" is not going to help.

The most important thing for us to remember is that decentralized terrorist networks are driven by ideology. Ideologies are not only the glue that binds them; they coalesce a social energy. One of the most important ideology factories in the world are schools.

As a nation we have not engaged in supporting good education across the Middle East. That is probably the most important thing we can contribute to . . . increase employment, promote more critical analysis and reasoning among the population, and to teach what we might consider modern social values.

Another thing we can do is focus on a simple ideological principle called respect. Many Muslims do not feel respected by Americans or by the Western press, and they feel that their own cultures are diminished. When cultures feel insulted, people can become radicalized.

Finally, we can centralize our opponents and decentralize our own activity.

Trying to build a coalition in the neighborhood is very important: Saudi Arabia, Iran, Syria and Israel. Obviously it's not easy because of Muslim countries' attitude to having Israel at the table, but they are the other huge stakeholder and power in the region.

You make it sound so easy.

The key issue is your strategy. If it's wrong, it does not matter how well you execute, you will still fail. I think it's easier to develop a better strategy than what we've had. . . . I also believe it will still be a challenge to execute the strategy and this will be a long-term endeavor.

View all comments that have been posted about this article.

© 2007 The Washington Post Company