ICANN's Global Elections: On the Internet, For the Internet

A Study of the ICANN At-Large Elections by
Common Cause and the Center for Democracy and Technology

March 2000

3. Concerns and Issues Regarding Current Plan for the At-Large Elections

• 3.1 The Current Plan and Issues Raised
• 3.2 Major Concerns

3.1 The Current Plan and Issues Raised

Nearly every interested party with whom we spoke expressed skepticism or alarm about some aspect of the current election system, and with the ability of ICANN to implement a legitimate and fair election by this summer.

We catalog below the major concerns about the present proposal for the At-Large election. Most concerns stem from the tensions between the several divergent goals for the election, such as representing the diverse range of stakeholders, selecting a board which is technically qualified to achieve ICANN's narrow mission, avoiding capture by interest groups, and operating fairly and accessibly on a global scale.

Many concerns are legitimate, to greater or lesser degrees, even as some are odds with each other. For this reason, the concerns do not lend themselves to a single "silver bullet" solution that would address and resolve them all. There is a virtual consensus that the existing plan for an At-Large election is deeply flawed. There is far from a consensus on the right solution.

[Top] [3.1 The Current Plan and Issues Raised] [3.2 Major Concerns]

3.2 Major Concerns

There are a wealth of criticisms made of the current model for the At-Large election process. While we do not suggest here that all the criticisms are valid, we note recurring themes, for they suggest the range of problems that members of the ICANN community have identified with the current plan:

a. The Election Will Not Be Meaningful if the Role and Mission of the ICANN Board Are Not Clear - A key concern is that ICANN's mission is not well-defined, well-understood, or well-constrained. On its face, this raises a problem with analyzing the election: the At-Large election for the board of a narrow technical organization is appropriately quite different from the election for members of a broad Internet policy-making body for the world.

Others believe that an election cannot be meaningful if the electorate does not understand the mission or purpose of the organization in which it is participating.

As noted above, there remains much confusion about what a "narrow technical management mission" for ICANN really means. Without real limits on ICANN's authority that can be both relied on and enforced by the electorate, many fear that ICANN will inevitably be pulled in the direction of greater policy making. Well-intentioned efforts by ICANN to gather increased public membership might attract many with agendas unrelated to the sound functioning of the Internet. Many fear that such groups could use the elections as a way to select board members who will give in to pressure to expand ICANN's mission or adopt policies without sufficient consensus.

b. The Election Process Will Not Engage a Sufficiently Representative and Informed Electorate - A key threshold concern is the ability of ICANN to engage the broad, diverse, and informed electorate needed to produce a representative election that confers the consent of the governed. This is particularly important because in order to provide legitimacy and avoid capture, the voting membership must include a substantial number of participants, from across a diversity of regions and interest groups, who are informed and interested in the issues confronting ICANN.

The current election plan contemplates a "sunrise" for the election based on achieving an electorate of at least 5,000 members. On the one hand, many commentators express skepticism about ICANN's ability to reach even this level of informed members, given the relative obscurity of ICANN, the narrowness of its technical management mission, and the indirect nature of the voting system. [5] International participants raised particular concerns about the challenges of informing and recruiting members in developing nations.

On the other hand, many have expressed doubts that a mere 5,000 members can adequately represent the diversity of interests in ICANN, and that the size of such a small membership - in the thousands, relative to a potential voting population in the tens of millions - creates tremendous opportunities for capture. The ability of ICANN to engage an appropriately large, diverse, and informed electorate thus remains a key unknown and fundamental concern.

c. The Electorate Is Not Properly Defined, Either Broadly Enough or Narrowly Enough - We have heard numerous complaints that the interests of many parties - particularly individuals, small businesses, and non-commercial groups, and perhaps certain geographic constituencies - are not now sufficiently represented in the Supporting Organizations. They believe the At-Large election should provide those under-represented interests with a voice in choosing the board.

For some, obtaining the consent of the governed should lead to the broadest possible view of ICANN's electorate. Many have argued that better efforts must be made to make voting simple, cheap, and available globally. Minimal membership criteria, simple electronic voting, and broad outreach - the plans for which are only now being developed - are viewed as essential to an inclusive election.

For others, however, this concern translates into a desire to limit the electorate. Many have expressed a view that only stakeholders directly impacted by ICANN's work (such as domain name holders), or those who have made some affirmative demonstration of interest (such as by paying a membership fee), ought to have a vote. They fear that the "consent of the governed" cannot be achieved if the interests of those only remotely affected by ICANN are allowed to dominate the interests of those who are, for example, bound to follow ICANN's rules. They note that there is also a tension between the goal of easy access to the ballot box on the one hand, and the desire to minimize fraud on the other. Limiting the electorate would make it easier to deal with these tensions.

d. The Election Process Is Vulnerable to "Capture" - There is a substantial risk that large institutions, factions, interest groups, corporations, or even governments will mobilize efforts to "capture" a disproportionate interest in ICANN by obtaining a larger voice in the election of board members than appropriate. The fear is that the current model - even taking into consideration the proposed indirect election - does not adequately guard against this type of capture.

There appears to be an almost universal fear of capture within the ICANN community, with each interest group convinced that other interest groups might control disproportionate numbers of board members. The fear of capture is exacerbated by the relatively small number of people expected to vote (in the thousands), the large number of potential voters (in the tens of millions), and the unprecedented nature of the election which makes it difficult to predict exactly who will vote.[6]

This sort of capture can take many forms. For example, a large corporation with an economic stake in the DNS could mobilize tens of thousands of employees and shareholders to vote, conceivably installing a number of ICANN Board members who support the company's positions. A national or local government could mobilize its citizenry to vote to install board members that favor regional interests. Or a political group could mobilize voters to elect board members committed to using ICANN to further a political agenda unrelated to technical management issues.[7]

Paradoxically, the more limited and "technical" the mission of ICANN appears, the more likely it is that only truly interested members will join - but also the more likely it is that a small, well-organized, and well-motivated group could gain a substantial voice in ICANN and promote its own narrow agenda, perhaps one outside the scope of ICANN's mission.

e. An Indirect Election Discourages and Disenfranchises Voters - The two-tier membership structure currently proposed - where a public electorate votes for members of an At-Large Council which in turn elects the At-Large Directors of ICANN's Board - has been widely criticized. A chief concern is that members will be driven away by the complexity of the system and the lack of clarity over what Council members will do or who they will vote for when selecting board members. Critics bitterly complain that such a system disenfranchises members, and provides little incentive to vote and little on which to campaign. [8]

Others are concerned that indirect elections strip members of certain legal rights they would otherwise have as "statutory" members of ICANN under California law. Some believe such rights, including the ability to bring derivative suits or impose certain reporting requirements on ICANN, are an essential check. Others, including ICANN staff, argue that ICANN cannot afford the risk of disruption from sweeping legal remedies. [9]

Uncertainty about the shape of an At-Large council election adds to these concerns. As yet, there are few rules about how the council will vote. Winner-take-all majority rule voting in the At-Large Council, for example, could allow a group capturing just half the council seats to control all nine At-Large seats on the board. Moreover, failure to tie a council member to a particular platform or "slate," as in the U.S. Electoral College, removes accountability and leaves individual members further from the process of electing board directors.

f. The Election Process Will Not Produce High Quality Board Members - For many, a measure for viewing the success of the election process is its ability to put forward and elect good board members, capable of leading ICANN in its mission. Many have voiced concern that an open election allows for the selection of council members and directors who might have little direct qualification for the task of governing ICANN.

Although some argue that the indirect election is a method of minimizing the likelihood of unqualified directors, others criticize the absence of any sort of nomination process for failing to provide a threshold filter on the suitability of board members.

g. The Election Process Will Not Produce Representation of Diverse Viewpoints - Much criticism has been leveled at the fact that there is no sure mechanism in the current election system for guaranteeing voices representative of the broad range of stakeholders involved. It is true that the current system does mandate a measure of geographic diversity. But some have noted that diversity cuts along many different, non-geographic lines, and have argued for multiple dimensions of representation.

h. The Two-Stage Election Disfavors Minority Viewpoints - Not only is the election to be indirect, but it is to be fragmented as well. The current plan calls for election of one-third of the At-Large Council and board members first, and then a further election several months later. The well-intentioned motive for this division of the process is to experiment with the election, and allow adjustments to be made. But some of the election experts noted that such a system leaves many viewpoints competing for the few truly at-large seats. The smaller the number of seats, the less likely it is that minority voices will be included. The concern is that the current plan makes it difficult for minority viewpoints to be represented on either the council or the board.

i . There Are Inadequate Safeguards to Ensure Fair Elections, and Guard Against Fraud and Corruption - There is widespread agreement on the need for further efforts to assure that the elections are fair, both in the sense that votes are counted fairly and that adequate notice and opportunity to participate are provided. Rules must be established and applied even-handedly and should not intentionally disadvantage any particular group. Many fear that the online election mechanisms being put in place can do little to prevent fraud, and that further work must be done to authenticate votes and audit the election. Others have noted that some form of corruption or the appearance of corruption could stem from unbounded expenditures on behalf of a candidate, and the difficulty of establishing and enforcing campaigning rules across national boundaries.

j. It Is Doubtful That ICANN Can Conduct Fair and Effective Elections by September 30, 2000 - There is widespread and intense skepticism of ICANN's ability to implement credible, publicly legitimate At-Large elections by September 30, 2000. As many participants pointed out, ICANN just recently opened up its servers for registration; it has not selected a voting system for the elections; it has not explained adequately to potential candidates the roles and responsibilities of the position for which they will run; it has not set up procedures through which candidates will come to appear on the ballot; it does not have systems in place to audit the electorate and prevent fraud.

Given all the decisions that must be made and the tremendous amount of work that must be done, the elections experts we consulted, who have a wealth of experience in building elections from the ground up, indicated that these tasks cannot realistically be done properly in the short timeframe created by the September deadline.

k. ICANN Should Recognize the Experimental Nature of These Unprecedented Elections - Many of those with whom we spoke admitted that even if ICANN established an election process that conceptually met their every wish, they still would not be completely confident that the resulting election would be fair and effective. While participants generally want the best possible election system according to their goals for it, many mentioned that ICANN must recognize up front that these elections are experimental by their nature. There are too many unknowns - particularly regarding the interest of the potential electorate and the future development of the Internet - to conceive of these elections in any other way.

[Top] [3.1 The Current Plan and Issues Raised] [3.2 Major Concerns]

[Next: 4. Options for Change]

[Return to Table of Contents]

5: The Industry Standard reported that on February 25, 2000, ICANN had received 2,599 applications from people in North America, 656 from Europe, 315 from the Asia-Pacific region, 66 from Latin America and the Caribbean and 38 from Africa. (Full article available at http://www.thestandard.com/article/display/0,1151,12433,00.htm)

6: An excellent overview of the capture issue is included in Jonathan Zittrain's testimony before the U.S. House of Representatives, available online at http://cyber.harvard.edu

7: The concern was also raised that those most likely to mobilize large numbers of members are the same organized and motivated groups that are already heavily represented in the Supporting Organization structure.

8: Many concerned cited the recent and somewhat similar DNSO's 21-member Names Council election of board members, which many believe was characterized by a great deal of gamesmanship and political deal-making outside of the public eye.

9: In August 1999, an official report on statutory membership was issued by the ICANN staff. (Full text available at http://www.icann.org/santiago/membership-analysis.htm).

Free Speech | Data Privacy | Government Surveillance | Cryptography | Domain Names | International | Bandwidth | Security | Internet Standards, Technology and Policy Project | Terrorism | Authentication | Right to Know | Spam

Our Mission / Get Involved / Staff / Publications / Links / Search CDT / Jobs / Action! Previous Headlines | Legislative Tracking | CDT's Privacy Policy