The San Francisco ICANN meeting has been completely dominated by GAC-Board negotiations over the new top level domain program. The underlying theme of those negotiations is the role of nation-states in the governance of Internet names and addresses. In the past few days I’ve learned a lot about the perspectives of various parties on this standoff, including people in civil society, the At Large Advisory Committee and the Noncommercial Users Constituency, business people, trademark lawyers, and government officials.

I come away more convinced than ever that we are not getting it right, this relationship between traditional nation-states and ICANN. At the same time, I see that most people involved, especially governments, fail to grasp that. The answer is not, as everyone seems to assume, more discussions and negotiations. The collisions that took place in Brussels last month will be replayed endlessly unless we diagnose the problem better.

It was nice to have Ira Magaziner in San Francisco to remind us of why we created ICANN in the first place. As he put it in his opening speech, “the slow and bureaucratic processes of government and multilateral government bodies are not the best way to coordinate a fast-moving, creative, chaotic medium like the Internet.  They move too slowly.  They’re too risk-averse.  They officially represent only governments and not other constituencies.” All true but this is not an entirely correct diagnosis. The real problem with governments is not a simple matter of speed. How could anything be slower to produce results than ICANN’s own new gTLD process, which in some respects has been going on since year 2000?

The real problem with governments is more fundamental. It is 1) the mismatch between the need for global coordination of the name space and the fragmented jurisdictional boundaries of nearly 200 territorial states; and 2) the inability of governments to participate flexibly in bottom up policy development working groups with other stakeholders.

Magaziner’s presence in San Francisco notwithstanding, it is astounding how few people seem to understand and accept the radical implications of the original approach to ICANN. It becomes clear that many people here want to have their cake and eat it, too. The GAC and its apologists seem to want to retain a private sector-based global governance institution and at the same time allow nation-states to exercise, or pretend to exercise, the same kind of policy control over global internet resources that they have traditionally exercised at the national level. They just don’t get it: that isn’t possible.

Even the US government is deeply confused about this. There was a poignant contrast between the reflective speech by Ira Magaziner and the one that followed from the current head of the Commerce Department’s NTIA, Larry Strickling. To his credit, Strickling presented an unusually direct and forceful presentation of the U.S. Commerce Department’s position. It was especially refreshing to hear Mr. Strickling openly discuss the “political sustainability” of ICANN. Before him that has always been the unmentioned elephant in the room. But there are many flaws and contradictions in the new U.S. position. In the rest of this post I concentrate on analyzing Strickling’s speech.

Strickling has taken a hands-on approach to ICANN. He played a guiding role in formulating the Affirmation of Commitments and participated directly in the Accountability and Transparency Review Team. He has claimed responsibility for the U.S. government’s sudden about-face on using “morality and public order” as the grounds for governmental objections to new TLDs and for the U.S. decision to push for allowing governments to make objections “for any reason” instead.

Unlike his predecessors, Strickling no longer describes ICANN as “private sector-led.” He describes it as a “multi-stakeholder model.” This is a shift. The ideology of “multi-stakeholderism,” a mouthful which I prefer to shorten to “MuSH” (MultiStakeHolder), brings governments more prominently into the picture. But it does so in a confused, often dysfunctional way. In contrast to the Magaziner approach, which tries to create a space for global governance outside of national governments through privatization and contract, MuSH pretends that governments are a “stakeholder group” just like civil society and business. In other words, MuSH transmutes national governments, which are themselves aggregations of a myriad of national-level “stakeholders,” into a single super-stakeholder. It ignores the fact that they are lawmakers with coercive powers and dissolves the 200+ distinct territorial jurisdictions and polities into a single group. What a massive conceptual error! For one thing, if governments are stakeholders they would have to participate on the same terms as conditions as the rest of us in the bottom up processes.

Of course, there are good reasons why Strickling now emphasizes “multistakeholderism.” Strickling wants to keep the rest of the world’s government happy with this unilateral creation of the U.S. Talking MuSH is more palatable to them than using more accurate terms such as “privatization” or “non-governmental governance.” Those terms make it clear that ICANN is a substitute for, and in some sense a competitor to, sovereign authority. Strickling fears that the world’s governments will feel threatened or disempowered by ICANN and will not tolerate it. In one-on-one conversations U.S. government people must admit that there is no chance the UN or ITU could “take over” or destroy ICANN over the U.S.’s wishes. But they are tired of dealing with the pressure.

Strickling’s speech recognized clearly one of the weakness of current model. GAC policy advice, he noted, comes at the end of an arduous process through which the other stakeholders have debated, made compromises and reached a consensus. Thus, GAC intervention is made in ignorance of most of the information and ideas exchanged by the other “stakeholders,” and acts as a kind of veto or re-booting of the process. But Strickling’s diagnosis, which suggests a remedy of more continuous dialogue, is only partially correct. The more fundamental problem – which was noted repeatedly in San Francisco by governmental and nongovernmental actors alike – is that governmental representatives are incapable of participating in ongoing working groups. The mid-level bureaucrats who populate the GAC lack the expertise to contribute much to ICANN’s policy processes; even when they do have the needed expertise they lack the flexibility to bargain and negotiate policy positions. They must go to their superiors in a hierarchical chain of command before they can accept or approve a position. That kind of back and forth doesn’t mesh well with a nongovernmental bottom up process, where individuals have the agency to make their own decisions, compromises and adjustments. Indeed, it completely negates it, for if people go through this arduous process only to be forced to do it again to satisfy slow-moving governments, why bother in the first place?

Strickling’s speech also argues that the ICANN Board should rely entirely on consensus policies and refrain from picking winners and losers in the policy process. Strickling urged the Board to insist on obtaining consensus for any policy before it reaches the board, and told the Board to refrain from substituting its own judgment for that of “the community” by selecting among policy alternatives when there is no consensus. I was shaking my head during this part. What an incredible thing for him to be saying! The U.S. Commerce Department, more than any other party, has intervened in the ICANN process to try to choose policy winners in the absence of consensus. Whenever there is a contentious policy, such as Whois or new gTLDs, the Commerce Department can be relied upon to side with trademark owners and law enforcement agencies, especially when they fail to get community consensus for their favored policies. Commerce has repeatedly ignored or overruled civil libertarians, industry groups and other interests when it wants to pick those “winners.”

Aside from the hypocrisy, Strickling’s position is naïve. Any policy that has negative economic effects on a particular interest group will not gain consensus when the prospective losers can prevent any policy from being implemented simply by withholding their agreement. Consensus is a sure-fire way to freeze in place the status quo. We learned that during the Whois process and again during the vertical integration debates. As a veteran of the FCC process, Strickling must know this. If we had require the agreement of AT&T or any other incumbent telephone monopoly to move from regulated monopoly to competition in telecommunications, it never would have happened. So what is accomplished by insisting that consensus be achieved when conditions dictate that it is impossible to have consensus? The message can only be that ICANN should do nothing. Why have it, then?

It’s time for everyone involved to recognize that ICANN is a global regulatory authority, not a Quaker meeting. Insofar as it is an effective and fair regulator, its policies will achieve consensus only rarely. Consensus decision making was never an essential component of the ICANN model in the first place. It was just a misguided transference from the IETF culture, promoted by techies who don’t understand political economy, and amplified by U.S. government officials who wanted to pretend that they weren’t creating a global internet governance authority.

So the underlying roots of the GAC-Board conundrum will not be resolved at this meeting, regardless of how many items on the GAC scorecard are ticked off. Still, one does not get from this meeting any sense that ICANN as an organization is threatened or about to collapse because of that. Far from it, it seems as vibrant as ever. The amount of swag in our conference bags continues to grow. Instead of working on the real GAC-Board problem, there seems to be a dogged commitment to endlessly expanding “communication” between the contentious parties. Most of that communication merely repeats the same arguments about the name space that we have been having for 10 years. We need a better approach. It’s time for governments to re-educate themselves about the nature of global networking and Internet governance, and for Internet people to understand better what drives governments. A good place to start is right here.