| iBRIEF / Media & Communications | Cite as 2001 Duke L. & Tech. Rev. 0001 |
| 2/28/2001 |
AN INTERVIEW WITH MICHAEL FROOMKIN
¶ A. Michael Froomkin1 is an Administrative Law and Internet Law scholar from the University of Miami School of Law and a vigorous critic of the Internet Corporation for Assigned Names and Numbers (ICANN). He is the author of a controversial new law review article, Wrong Turn in Cyberspace: Using ICANN to Route Around the APA and the Constitution, 50 DUKE L.J. 17 (Oct. 2000), available at http://www.law.duke.edu/journals/dlj/. In his new article, Professor Froomkin argues that ICANN's relationship with the Department of Commerce is illegal. We interviewed Professor Froomkin via e-mail about his new article and about other recent ICANN-related events, such as ICANN's plan to assign new generic top-level domains (gTLDs).2
************************************************************************ ¶ Professor Froomkin, you argue in your article, Wrong Turn in Cyberspace, that ICANN's relationship with the Department of Commerce is either unconstitutional or in violation of federal statutes. Can you briefly explain for us why you say that?
¶ Hey, unfair question! The article took 168 pages for a reason. ;>
¶ To vastly oversimplify, the argument is:
¶ 1. ICANN is not just a standards organization. It makes social policy, e.g., the UDRP [the Uniform Dispute Resolution Policy, under which trademark owners who claim another party's domain name registration is improper can force the domain name owner into arbitration].
¶
2. As such, it is
either a state actor or private. If it is a state actor, then the
Department of Commerce may not use it to avoid the rulemaking requirements
of the APA [the Administrative Procedures Act, which sets out procedures
that all administrative agencies must follow]. If it is private,
then the Department of Commerce's delegation of policymaking powers
to a private group is too broad here to be constitutional, violating
the doctrine in Carter Coal (and re-stated in more recent
Texas Supreme Court cases) that limits the giving of public power
to private groups-primarily on due process grounds.
¶ It sounds as if government can delegate some policymaking powers to private groups, as long as the delegation doesn't rise to the level it did in Carter Coal. Where do you draw this line? Why does ICANN fall on the wrong side of it?
¶ The two clearest cases are (1) the UDRP, and (2) any future decision to prefer one new gTLD [generic top-level domain, such as ".com" or ".edu." Top-level domains (TLDs) can also be country codes (ccTLDs), such as ".uk" or ".us."] over another application that also meets minimum neutral technical qualifications. There are several others, I think, but these are the ones that I consider slam-dunk cases.
¶ I think they would say two things. Wrong on the facts (they argue that ICANN is really just a standards organization, doesn't make "policy") and it is a mere contractor, hence the APA shouldn't apply to it. And wrong on the law: the Carter Coal doctrine is dead. I argue at some length in the article that the facts do not support this attempted rebuttal. I honestly think that part of my paper is very strong.
¶ The legal questions are perhaps a little closer; for example the case for finding ICANN to be a state actor although strong is not at all a certainty, especially given the current trend in the case law against finding private bodies to be state actors without overwhelming evidence.
¶ That takes us to the Carter Coal doctrine. Anyone arguing for the revival or re-vivification of a pre-New Deal doctrine, much less one somewhat incorrectly called a "non-delegation" doctrine, better be prepared to lift some heavy baggage. Although Carter Coal has never been overruled, it is hardly the most vital case in the books. So here too, I'd have to recognize that although I believe my argument is correct, one would have to advise a client that there was risk a court would not agree.
¶ This is explained in the article: the recent cases explain why the doctrine is still needed and important, and import some substantive tests from the academic literature. But you have to slog to nearly the end to find that part.... ¶ There has been a lot of press lately about ICANN's review of proposals for creating new gTLDs. Assuming that new gTLDs are technically necessary, doesn't ICANN need to find a method for det
ermining which ones are best? Is this technology, or policy?
¶ I'm afraid I think the question is misguided. There is no technological limit on, say, thousands of new gTLDs. Even if you take the (debated) view that registries should be tested for technical competence (rather than letting the market sort things out), that doesn't justify an artificially low limit on new gTLDs. And picking the best out of many technically qualified proposals is clearly "policy" once the technical bar is out of the way. Choosing between .xxx and .kids and .shop on social utility is 100% policy. ¶ As you mentioned, ICANN claims to be primarily a group that sets technical standards, and that sets policy only where such policy is necessary to its primary mission. What do you think of this argument? If ICANN is going to pursue its technical mission, doesn't it have to set some policy?
¶ No. Certainly not on the scale it is doing so. One could set technical standards easily without a UDRP-just leave it to the courts. One could define technical requirements for new gTLDs without arbitrarily capping them at ten or fewer (the tech limit is orders of magnitude higher, tens of thousands at least, maybe a million or more). ¶
One argument for ICANN
goes like this -- the US government created the Domain Name System,
or DNS, voluntarily handed it over to private parties, and private
parties pay for it. Why should the public have a say in running
it? Does the public really have any claim to "ownership" of the
DNS?
¶ The parties who run ICANN don't pay for the DNS. They don't pay for the Internet-just some parts of it. And the government hasn't "handed it over" to ICANN or to private parties-it has just handed *some* day-to-day control on a short leash. I agree that if it were legal for the US to give away its interest in the DNS, then once it had done so there would no longer be either constitutional or APA issues. (The General Accounting Office, interestingly, suggested that it wouldn't be legal for the government to just give this away without an act of Congress authorizing it; I find the question wonderfully complicated, because-as I explain in the article-in order to answer it you have to characterize the legal nature of the government's interest in the DNS, and that is surprisingly difficult.) ¶ The issue of ICANN's representativeness (or lack thereof) seems to come up often. Elsewhere, you have said that, "I agree that a purely technical body wouldn't need to worry about representing end users directly." Given your argument that ICANN is NOT a purely technical body, do you think it should represent end users directly, or indirectly (through the Department of Commerce)?
¶ Ideally, US policy decisions should be made via constitutional procedures, such as notice-and-comment rulemaking by federal agencies. ICANN's current electoral processes have enough question marks that it would not be responsible to depend on them. The difficulty with representation via Department of Commerce, of course, is that (at best) only US citizens are represented. That's why I proposed that Department of Commerce should share out decision-making power as much as possible with foreign groups, in a sort of parallel processing. ¶ Assuming ICANN should be more of a democratic body, what's a good model? The United Nations? An administrative agency, like the Federal Communications Commission? An international organization, like the World Intellectual Property Organization?
¶ I am not sure the assumption is correct. The best ICANN would be a true technical body; those don't need to be democratic. The best insulation for a wire is not a subject well suited to a ballot.
¶ If policy is to be made, I argue in the article that an international organization is a bad idea. They are not at all democratic.
¶ The best approach I can think of is to get a lot of different types of bodies into the act, so that policy doesn't have a single point of failure. ¶ If ICANN's relationship with the Department of Commerce was found to be illegal, what would happen?
¶ The Department of Commerce would have to start regulating directly instead of via its proxy. I propose in the article how it might do that-by bringing in "policy partners" to diversify the decision making, yet ensure that no two entities attempt to launch the same TLD. ¶ Do you see any lawsuits challenging the constitutionality of ICANN of the horizon? If so, who would be the plaintiffs? ¶ No comment. ¶ You have written extensively on ICANN's changes to its bylaws. What changes are you most concerned about? ¶ A. The frequency of change itself is a serious problem. Deals are not stable. ¶ B. The multiple changes on elections of at-large directors, and on the retention of un-elected directors. ¶ C. I'm also concerned about the routine ignoring of the bylaws that have not yet been gutted. For example, ICANN has secret committees it doesn't even admit exist. It doesn't publish minutes for Board committees, e.g., the Conflicts Committee, although this is required. It has yet to name a Reconsideration Committee, after TWO YEARS, although this was supposed to be a major check on the Board. ¶ What do you think about ICANN's recent at-large elections? Is this a move in the right direction?
¶ * Many people who tried to register couldn't.
¶ * The registration software was not put out to tender, is not open source, is kept secret by ICANN, and was apparently written by one of the opponents of ICANN elections.
¶ * Of the people who did register, ICANN determined that a large number were not real registrants, using what metric we do not know.
¶ * Of the people whose registrations ICANN accepted, many didn't vote.
¶ * We don't know anything about who the voters were because ICANN keeps the list secret. (Registrants cannot even opt out of this secrecy!)
¶ * ICANN nominated so many people itself that there were very few slots left for member-nominated candidates-only two in Europe, for example.
¶ * Member-nominated candidates trounced the ICANN ones in Europe and the US, but the ICANN-nominated candidate dominated by an even larger margin in Asia/Pacific. So few people voted in Africa and Latin America that one hesitates to say anything about the results other than that they were small.
¶ * The people elected are smart and diverse.
¶
I think the jury
is still out here.
¶ Earlier this year, the Second Circuit determined that Name.Space, a company that had accepted registrations in 530 new gTLD's such as ".forpresident" and ".microsoft.free.zone," could not bring a monopoly claim against Network Solutions because Network Solutions' conduct (refusing registration of any new TLDs) was compelled by its contract with the government. Would ICANN, as a government contractor, have similar conduct-based immunity against an antitrust suit? Is this a concern of yours?
¶ Mark Lemley and I are going to be addressing the anti-trust issues in a future article. The antitrust issue that most interests me right now is that, since ICANN claims to be private, it follows that the bodies that lobby it do not have Noer-Pennington immunity. If they think about it hard enough, they really might rather that ICANN were public....
¶ That's an awfully good question. My family asks it all the time. However unjust and unfair ICANN may have been, no one has died, no one has starved, no one has been unjustly jailed, and while I could paint you scenarios in which some of these things did happen as a result of ICANN decisions, I frankly think those scenarios are quite wild and unlikely. ICANN just isn't as important as global warming.
¶ Nevertheless, I care (too much!) for a number of reasons:
¶ First, I've been on the Internet just long enough to have participated in the relatively naive utopianism that characterized its formative years. I still harbor some (diminishing) hopes that this tool can be used to increase human freedom and self-realization.
¶ Second, there is a substantial chance that ICANN will last, and that it will be a precedent-setting organization. There seem to be substantial costs to getting this one wrong.
¶ Third, ICANN now threatens to have spill-over effects, as its proponents tout it as a model for other problems.
¶ Fourth, the other subject I teach is Administrative Law. In that mode, I wrote an article a few years ago about government corporations; I remain fascinated by odd, experimental, modes of organizing the public sphere. Alas, so many experiments go wrong...
¶ And, fifth, and by far most decisive, as Saul Alinsky taught us, "you start where you are." I always imagined that one of the things I would want to do in a teaching career is some form of public service. I imagined it would be pro bono cases. But, by strange accident, I ended up here. Three or four years ago, I was not a trademark lawyer. I had only a nascent interest in "Internet Governance," although I'd written a lot about Internet-related regulation. But I knew a lot about arbitration law, having practiced it before entering teaching, and I had spent several years in Europe and speak good French; these are assets when dealing with international bodies.
¶ So when a friend called up and twisted my arm to allow my name to go forward as the sole civil liberties representative to be added to WIPO's [the World Intellectual Property Organization's] first panel, I let him talk me into it, because I truly believed that of the people WIPO would accept (i.e., those who lacked a paper trail demonstrating either knowledge of the law, or opposition to WIPO's intellectual property-maximalist agenda), I was the one who could most effectively advance a civil liberties agenda.
¶ By the time that process finished, I found I was something of an expert, or could at least fake it convincingly; ICANN had in any case become "my" issue. Again, when ICANN railroaded the UDRP, I found myself being added to a drafting committee for balance. And, now that the UDRP is going on, I cannot help but see that it is quite often mis-applied; some of those cases may be rough justice, but many are injustice. The injustices sometimes may be only financial, but oftentimes they are the stomping of the weak who dare to embarrass the powerful (e.g., companysucks.com cases). That isn't right, and it is a wrong that our institutions, both governmental and not, should be so complicit in this.
Footnotes
1. Michael Froomkin received a M.Phil. degree from Cambridge University in 1984, and a J.D. from Yale Law School in 1987. He clerked on the District of Columbia Circuit and the Northern District of Illinois. Professor Froomkin joined the University of Miami faculty after working in the London office of Wilmer, Cutler & Pickering. He teaches courses in Internet Law, Electronic Commerce, Jurisprudence, Administrative Law, and Constitutional Law. Professor Froomkin is also a director of diputes.org, an ICANN-accredited dispute resolution provider, and a co-founder of ICANN Watch, an organization dedicated to raising awareness of ICANN's activities. His personal webpage is at http://personal.law.miami.edu/~froomkin/.
2. ICANN's report recommending seven new gTLDs was released on November 10, 2000. A full copy of the report is available at http://www.icann.org/tlds/report (last visited Feb. 6, 2001). On January 12, 2001, the US House of Representatives' Committee on Energy and Commerce announced that it was preparing to hold a hearing on the process by which ICANN selected these new gTLDs. The hearing is planned for February, 2001. See Chairman Tauzin Calls for Hearings on ICANN, News Release, Committee on Energy and Commerce, at http://www.house.gov/commerce/news107/01122001.htm (last modified January 12, 2001).