I thought of ICANN yesterday when reading about the devolution of the Iraqi Governing Council, which managed to unite for just a moment to approve a constitution with about the half-life of lutetium. ICANN and the IGC: two institutions put in charge of ill-behaved constituencies and stuck in chronic failure mode. Could anything be learned by examining them at arm’s length? Indeed, different as they are, their histories contain several common elements.

• They were artificially created. In the case of ICANN, the U.S. Commerce Department implemented a hastily assembled and bizarrely structured proposal for a sui generis body in order to put a quick end to the uncertainty surrounding the Domain Name System. In the case of the IGC, the United States and its coalition partners had to make a desperate bid for legitimacy where they had no natural base of support in a country they were invading, partly because of suppression by the dictator Hussein and partly because of the sanctions and other unpopular international anti-Iraqi activities.
• They had no chance to establish legitimacy among the ruled. ICANN never even managed to establish what its constituency was, and ultimately suffocated its abortive attempt at democracy because two of its most vocal critics were elected to its board. The IGC suffers from the twin blows of being appointed by invaders and seen as short-timers.
• Their goals are contradictory. ICANN is supposed to promote competition in domain names and to protect trademark interests, the latter goal being most easily satisfied by preserving an artificial scarcity in domain names. The IGC is supposed to enable elections while protecting principles that would probably be overturned by elections, such as minority rights, equality for women, and privatization of oil and other industries.
• They need to sweet-talk their patron. ICANN goes before the Department of Commerce every year or two with promises to live up to fine language about representativeness in the Memorandum of Understanding it signed with the department. The IGC has to persuade the United States government that it is achieving stability—along with other American goals—in order to acheive the one goal on which nearly all the IGC participants agree, getting the military out.
• They struggle with constituencies that insist on working outside the system and that challenge their authority. Recent lawsuits involving Verisign and others, along with the challenge that the U.N. gave to ICANN at the World Summit on the Information Society, show that the boundaries of ICANN authority are completely obscure. The IGC, of course, is challenged by very frightening and very bloody violence on a daily basis.

The essential defect underlying all these problems is that each system deals with constituencies that disagree deeply on where the systems are heading.

ICANN will not become effective because there is no clear definition in the area of names and numbers about what effectiveness is, despite claims of interest among all parties for reliability, competition, and so forth. What is a good domain name system? One where names clearly indicate the weight and authority of the owner, as in a trademark? Or where anybody can have a desired domain name and where the names are catchy and evocative?

The IGC will not become effective because, even though most Iraqis seem to value the same things people value everywhere (security in their homes, control over the forces that determine their lives, cultural preservation, and so on) too many forces pull them in different directions. On the role of women, the degree of decentralization for Kurds, and foreign ownership of the economy, just to name three huge questions, they can’t find a common ground—at least a common ground that their occupiers would accept.

Meanwhile, each institution muddles along while accumulating bad decisions and a history that causes observers to hold their noses. We may be stuck with them, though. Perhaps, this weary observer thinks, it’s time to leave them alone and see whether they can limp along to their finish lines—whatever those may be.