[governance] Muti-stakeholder Group structure (some ideas)

Dan Krimm dan at musicunbound.com
Fri Jun 1 04:36:10 EDT 2007

At 8:22 AM +0200 6/1/07, Bertrand de La Chapelle wrote:

>The founding principle/spirit of a multi-stakeholder governance process is
>therefore something like : "Any person has the right to participate in the
>governance of the issues of interest or concern to him/her. Specific
>processes established to facilitate the elaboration, adoption or
>implementation of regimes must guarantee transparency, inclusion and
>I stop on this post to address the other issues, but understand this is
>just a stage in the discussion and am willing to engage. I suggest we
>start a separate thread on this notion of stakeholders. Long overdue and
>As I've had the opportunity to mention in other posts, participants in
>multi-stakeholder processes do not and should not "represent" people or
>organizations in the traditional sense of representative democracy,
>meaning taking decisions in their place. They represent viewpoints, the
>diversity of viewpoints. The purpose of a multi-stakeholder deliberation,
>in my view, is to make sure that all facets of a given issue (technical,
>social, economic and policy) are taken into account in the discussion from
>the onset, before rushing towards the drafting of a "solution". As we can
>witness in the ICANN whois debate, involvement of all categories of actors
>is critical to understand completely an issue.
>Therefore, the question should never be : "how many divisions ?" (ie how
>many members does this person "represent"). Because we do not talk about
>voting here, but about thorough examination of issues, discussion,
>democracy through deliberation. And therefore the right question is : does
>this person help understand a specific dimension of the issue or the
>position and interests of a given group of actors, does this person
>contribute constructively to a better common understanding ? The primary
>goal is consensus building in the analysis of an issue, not weighted
>voting. The question is participation, not representation. A single
>individual with good ideas is much more important to these processes than
>the "representative" of an organization claiming millions of members who
>have never heard of the positions he/she is taking publicly in a given


As a (still) newcomer to the Internet Governance community, I'm trying to
digest these ideas so as to create clarity in my own understanding.  I
understand that "multi-stakeholder" systems are in some sense a new beast
in the governance world (though some international organizations would seem
to present an initial precedent to refer to).

Nevertheless, I am concerned about the fundamental incommensurability
between "voting/representation" models and "multi-stakeholder/consensus"
models in your ideas, and I am interested in considering whether this
disconnect can somehow be connected.  The metaphor that comes to my mind is
the "wave-particle duality" issue that emerged in 19th-century Newtonian
physics and was resolved in quantum physics.

In the case of ICANN, for example, there are voting systems combined into
what are presented as "multi-stakeholder" groups, such as the GNSO Council
with six stakeholder constituencies each with well-defined fixed voting
power.  There is well-known criticism of this setup that I need not
reiterate.  The GNSO Whois Working Group is more purely a consensus-driven
process, where voting has been actively discouraged, which is probably a
more compelling model than the Council.  And so far as I know, ICANN has no
genuine representational voting system, given that even the Board is not
set up with any sort of widespread elections by a represented population.
It's a private organization, after all, even with all of the supporting and
advisory groups it has created around itself.

If the idea of a multi-stakeholder/consensus structure is to genuinely
abandon the idea of voting, a real issue is how do you really define
consensus?  Ultimately, there must be some quantitative measure if
"degrees" of consensus are somehow intended to be considered, as they are
in the Whois Working Group, though perhaps that is simply to identify
priorities as to where to focus attention.

Consensus in your ideal (non-voting) vision ultimately seems to entail a
binary proposition, like being pregnant: either it's true or it's false for
any single issue, but there is no "in-between" or "degree of closeness" for
consensus without stepping into the world of quantitative (i.e.,
voting-type) measurements.  A problem with a "pure consensus" model of
governance is that there will inevitably be residual disagreement on
certain issues (not all conflicts between individual and collective
interests, or competing interests in general, can ultimately be guaranteed
to be resolvable), and decisions may simply have to be made even in the
absence of consensus, so how does one make such decisions?  It isn't fair
to say "just don't decide" because that *is* a decision in its own right
(in the policy profession we are taught to regard "status quo" as an
explicit policy choice to be considered equally among all other policy

While I am not about to suggest that it is *impossible* to resolve this
conundrum, I must say that I feel that the initial attempts are not
successful in "squaring the circle" just yet.

I'm sorry that I can't contribute a more positive suggestion as to how this
resolution could be achieved, but for now I only see the obstacles.  It may
be that a true resolution is not possible, and perhaps only some sort of
hybrid model will get closer to the goal, where consensus is used where it
makes sense, and voting is used where it makes sense.  But then, the voting
part *has* to involve some sort of genuine representational process, and
with 6 billion stakeholders on the planet that representational process
must be fairly weighty in nature and magnitude.  (And of course with 6
billion stakeholders, not "every" stakeholder can participate in a
consensus process, even if "any" stakeholder can do so.  How many
participants is "enough" in a consensus process?  How do you know that
you've covered all the diverse bases you can, or that a synthesis of
participating interests is indeed generally fair?)

Another possible strategy is to identify consensus where it is possible,
and then pass on the remaining disputes to an external representational
process.  Narrow policy scope to gain narrow consensus, and recuse from the
sticky policy issues that can only be decided with a fair representational
process.  Maybe it's okay to "pass the buck" in such circumstances.
Consensus as a first priority, representation as a fall-back.

If you have more detail regarding your ideas for resolving this duality
into an integrated synthesis, I for one would be tremendously interested to
hear them.  It seems to me that what you are calling for is nothing short
of a "quantum revolution" out of the world of "Newtonian" governance models.

Not a small proposition at all.  If it is resolvable, it will likely take
time and perhaps some genius level of inspiration, perhaps collaboration
will help.  In the meantime, we have pressing issues that need to be
*decided* in a *finite* time frame (like: yesterday...), and I'm not sure
that the multi-stakeholder/consensus process is up to the task in practical
terms across the full range of applications.  And I'm not sure that the
status quo in all cases is an acceptable policy choice, or that we can wait
for the conceptual breakthrough that will make everything work miraculously.

You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:

More information about the Governance mailing list