|Multi-Stakeholder Public Policy Governance and its Application to the Internet Governance Forum|
It next falls to consider various consensual decision-making procedures that preserve the deliberative quality demanded by deliberative democracy. These procedures are not to be considered at a macroscopic level as in the preceding subsection of this thesis—that is, focusing on the design of political institutions required to support consensus decision-making—but rather at the level of process: how the pursuit of consensus can best be measured, facilitated and managed within any facilitative institutional framework.
Although these questions overlap with those already dealt with in the discussion of deliberative democracy, by stepping outside that paradigm we may find new and more specific insights developed in the practice of other consensual decision-making processes, that are also compatible with the pursuit of consensus within a deliberative democratic framework. Conversely, some other pitfalls to avoid may also be revealed.
Just as there are a variety of implementations of deliberative democracy such as the 21st Century Town Meeting, Deliberative Poll and citizen’s jury, so too there are various implementations of consensus-based processes; indeed, there is much overlap between the two. Three examples applicable in an offline setting, that are compatible with democratic deliberation yet not explicitly derived from liberal democratic theory, will be briefly outlined here: Formal Consensus, Saint and Lawson’s private sector model of consensus, and the Consensus Workshop.
Formal Consensus is a consensus-based process developed by Butler and Rothstein for decision-making within a civil society organisation, which is compatible with the ideals of deliberative democracy as applied in such a setting, but without sharing the same roots in liberal theory. It was formerly known as “Secular Consensus,” to differentiate it from the process used by the Quakers at their meetings, with which it shares a number of similarities.
The process begins with an introductory phase in which the facilitator clarifies the process to be used, presents the proposal or issue, and facilitates the resolution of any questions put by the group for the limited purpose of clarifying the proposal or issue presented. Discussion of the proposal then takes place, and if it appears that it has the general approval of the group, a call for consensus is made immediately.
If it does not, then the process enters the next phase in which all concerns the group has with the proposal are listed, and related concerns are grouped together. This sets the stage for the following phase in which the concerns (or groups of concerns) are discussed in turn by the group with the objective of resolving them. Once all concerns have been resolved, a call for consensus is made.
If consensus still cannot be reached, then the group has three choices: to declare the proposal blocked, or for the objectors to stand aside and allow the decision to be adopted with their concerns noted, or to send the proposal to a committee which can endeavour to generate additional options to be brought back to the larger group at a later time.
Saint and Lawson have also developed a formal method for consensus-based decision-making, very similar to that of Formal Consensus, but from a private sector perspective. The four stages of the process they put forward are as follows:
Preconsensus (determining the group’s membership and agreeing on its purpose, values and authority, settling on a definition of consensus, and setting standards for interpersonal behaviour);
Understanding the proposal (stating and clarifying the proposal, stating any objections or concerns, and making an initial call for consensus);
Resolving concerns (an iterative process based around discussion of the each of the expressed concerns with reference back to the group’s purpose and values, followed by further calls for consensus); and
Closing options (which include extending the time for discussion and perhaps engaging a mediator to assist the group, withdrawing the proposal or the outstanding objections, allowing the proposal to be agreed by a supermajority, or even excluding the minority from the group).
Similar again but with a slightly different focus is the Consensus Workshop, one of a package of complementary techniques for participatory decision-making called the Technology of Participation or ToP®, developed by the Institute of Cultural Affairs (ICA). A Consensus Workshop breaks the process of seeking consensus into five stages:
Contexting (preparing for discussion by ensuring that all are aware of the purpose of the workshop and the process to be followed);
Brainstorming (generating new ideas to address the issue or dispute under discussion, without yet passing judgment on any of them);
Ordering (identifying relationships amongst the ideas and grouping them into related clusters—this is normally done by rearranging cards on which the ideas are written, but a flip-chart can also be used);
Naming (giving each cluster a name, and then discussing them in turn beginning with the largest cluster); and
Evaluating (once agreement is reached, the resolution is confirmed, and the significance and implications of the consensus and the next steps to be taken are discussed).
None of these three processes are prescriptive of the exact manner in which discussion of the proposal or objections must proceed, leaving the facilitator with some discretion in this regard. However there are a variety of common techniques upon which the facilitator can draw for various purposes.
For example, to determine who is entitled to speak, the facilitator’s choices range from the formal such as the use of Robert’s Rules of Order (although taking care that its strict application does not provoke division rather than fostering cooperation and open discussion), through to the informal such as allowing participants to engage in uninterrupted storytelling, depending on the size and composition of the group.
To provide a quick, non-binding overview of the group’s overall position on an issue, straw polling can be conducted, using methods ranging from the conventional such as a show of hands or the use of coloured cards (holding up green to indicate agreement, red to object, and yellow to abstain or stand aside), through to the unconventional such as the use of humming to indicate agreement. This latter option, which has been used within the IETF, carries the benefit that the group can discern whether there is broad or narrow agreement upon a proposal, without the need for a vote or the identification of specific objectors.
Finally to creatively overcome objections and generate solutions, techniques ranging from the relatively unstructured such as brainstorming or mind mapping, through to more highly structured tools such as Dialogue Mapping, policy Delphi and the Nominal Group Technique can be employed. As noted above, the Consensus Workshop also incorporates a structured brainstorming session, which draws from the Delphi technique.
There is again much overlap between the tools available to facilitate online deliberation and those for online consensus, especially since the tools most often used for either purpose, such as discussion boards and mailing lists, are generic in design.
For example, the coloured cards technique for straw polling has been implemented by an open source software project called Monit simply by means of its developers’ mailing list, to which votes of +1, 0 or -1 may be posted as the equivalents of green, yellow and red cards respectively.
There are however also a few tools that have been developed specifically for the purposes of facilitating online consensus. These can be divided in turn into those that implement the equivalents of offline techniques, and those unique and perhaps specific to the online environment.
In the former category, there are online implementations of each of the three structured techniques for consensus decision-making noted at the end of the discussion of offline consensus above. Dialogue Mapping, which is a method of visually representing deliberative processes, can be implemented using a free product called Compendium or a proprietary one called Decision Explorer. Policy Delphi, which is a moderated, consensus-oriented decision-making method based upon the use of questionnaires, is implemented by the DEMOS (Delphi Mediation Online System) Project. Finally the Nominal Group Technique is a method of discussion and brainstorming in which participants individually rank the group’s ideas, and those rankings are aggregated with the aim of isolating an alternative that meets with the group’s consensus. This has also successfully been implemented online, in a form that incorporates both synchronous and asynchronous participation.
An example of a technique for facilitating online consensus that is unique to the online environment is Consensus Polling. As presently implemented, Consensus Polling takes place in three main stages:
Framing the issue (settling upon an agreed form of the question or problem to be addressed);
Concerns and interests (in a forum where participants discuss the interests they wish to be met by whatever solution is adopted, and the concerns they wish it to address); and
Background information (the compilation of a comprehensive and accurate archive of relevant background information, which can inform the listing of concerns and interests);
Solution pieces (various possible approaches for addressing the issue in question that are evaluated against all of the concerns and interests that were isolated in the preceding stage); and
Solution (assembling these into a composite solution upon which consensus can be reached).
At each stage, a “Yes meter” is used which records each participant’s state of agreement with the current articulation of the solution being considered in that stage: either “yes, or “not yet.” Consensus can be deemed to be reached when all reach “yes,” or when some other high standard of agreement, such as 90%, is reached. Since participants can change between “yes” and “not yet” at any time, there must also be a defined period for which a consensus must be maintained, and a defined minimum number of participants. These variables are all settled upon before the poll begins.
The ICANN community has established an experimental Consensus Poll on the question, “What should ICANN policy with respect to new TLDs be?”, utilising a customised version of MediaWiki.