Extreme Consensus

August 16, 2001, September 8, 2001, & October 3, 2002 (edited January 10, 2004)

I have been interested in consensus for a while, but I have had concerns about some ways I have seen it practiced. Formal consensus procedures start with proposals which are modified or put aside according to concerns raised by the group. Ideally, a proposal is only accepted once all concerns have been resolved. In rare cases it is considered acceptable if one person "stands aside" rather than blocking consensus with their concerns.

In one group, the minutes show that many (maybe most) calls for consensus ended with a majority of members "standing aside". I can only guess that they understood that phrase as meaning "abstaining". They also listed those not standing aside as "voting yes". The minutes also showed that attendance at the meetings was low.

In many formal consensus procedures there are provisions for calling a vote if consensus is blocked on an urgent matter. But isn't consensus most important when it is hard to achieve?

So I wonder why consensus has failed for these groups, with many members either not participating or blocking urgent decisions. Are they deciding things that are not of consequence to the entire group? Are they trying to reach consensus on bathroom tile patterns? Do they simply disagree on the charter of the group, or are they proposing infringing on private decisions?

Here are some of my thoughts about consensus. First, consensus is not necessary for the details of all proposals. For example, the group should come to a consensus about the need for new bathroom tiles, but the pattern for those tiles can be chosen some other way, either by voting or by appointing a designer. The crucial distinction here is between concerns and opinions. Consensus should be used to resolve all of the group's concerns, but negotiating all the different opinions in the group is pointless. If someone really can't live with pink tiles, that should be treated as a concern. But if someone simply prefers blue over beige, that is an opinion that does not need to be resolved with consensus.

Now, once we accept the distinction between concerns and opinions, we have to be careful to recognize opinion issues when they arise and not try to use consensus for them. If we train ourselves to not raise our opinions as concerns, but we allow opinions to be presented as proposals, then the first opinion to be proposed will "win", even if most members hold another opinion, because those other opinions won't be raised as concerns. Instead, when an opinion-based proposal is presented, a concern should be raised, not about the opinion itself, but about the fact that the proposal does not account for other opinions. Then this concern can be resolved by listing all the different opinions in the proposal, to be decided by vote (or some other mechanism) after the proposal is accepted by consensus. In summary, voting is more appropriate for matters of opinion, but the proposal for a vote should still be decided by consensus.

This suggests something to look out for if the group reaches an impasse. Is the current proposal based on an opinion not held by the entire group? Can the basic concern motivating the proposal be factored out, leaving the opinion factor to be decided by vote (or not decided at all)?

Something else to watch out for is a proposal which infringes on anyone's rights. An organization must have clear limits on its decision making, so as not to impinge on the private decisions of its members. The group's charter specifies the scope of common activity and the principles guiding that activity, and members join for just that scope and those principles. Maybe the one person blocking consensus is the only one whose rights would be violated by the proposal; then the group must not circumvent that block. In other words, not all activities in a group should be decided by the group at all; the charter should make clear what kinds of decisions will be delegated to the group, leaving the rest private.

There are times when someone may not agree with a proposal but is not able to state a clear concern. Maybe the person is just shy, or maybe the objection is more intuitive than reasoned. Either way, such persons must be given a chance to register their disagreement without being pressured to express it as a concern in front of the whole group. Maybe the group should take a break for small group discussions, or maybe the proposal can be sent to committee for another week. That person's nagging feeling about the proposal could signal some hidden drawback that no one else has recognized yet.

It is tempting to wonder whether a parliamentary system could be more appropriate for larger organizations. However, consider a federal structure for the organization. This brings each meeting down to a reasonable size for consensus. Further, the charter of the larger organization will clearly delegate responsibilities considered common to the full body while leaving most decisions to the smaller groups, in the same way that a smaller group leaves private decisions to its members. The same techniques given above can then be used to determine when a vote is appropriate or when consensus is required.

Another question is how decisions can be reversed. In a parliamentary process, when only a majority is required to make a decision, once there is no longer a majority in support of a decision, there is automatically a majority opposed to the decision, and it can potentially be reversed. However, in a formal consensus process, if a new concern is raised so that there is no longer consensus on an established decision, that does not automatically create consensus to reverse the decision. In other words, in parliamentary process the disappearance of the original conditions for a decision is enough to reverse that decision, but in formal consensus process it is not.

It becomes easier to resolve this issue if we realize that each proposal begins with a concern. Seen this way, formal consensus is from the start all about resolving concerns. Once a decision is made, action will likely be taken on it; and then any proposal to reverse the decision will raise new concerns about interrupting activities already underway. So a new concern about an old decision does not automatically reverse that decision because it does not necessarily outweigh the concerns about disrupting ongoing activities. Instead, any proposal coming from a new concern has to address all of these additional concerns as well.

Thinking about proposals as coming from concerns, and about consensus as basically concern-oriented, suggests another class of proposals for which consensus is not necessary. If a proposal for group action does not require everyone to participate, and does not infringe on the rights of those who choose not to participate, then there is no need to get the entire group to agree. Such proposals are not so much concern-oriented as action- or idea-oriented.

This attitude can also help deal with the frustration which can occur when "open" groups try to make decisions by consensus. Often there will be a few people at a large open meeting who have different intentions than those who called the meeting. If the meeting is structured as a formal consensus process, these few people could easily block consensus, frustrating the will of the majority in attendance. But at this point there is really no reason to pursue consensus because there is not yet any common basis for the group. Such a meeting is really trying to find a common basis for organizing. Once a basis is found that many people in the group agree to, those who choose to participate can proceed to establish the principles on which common concerns will be built; while those who disagree are not infringed at all by simply not participating in a group where they don't really belong in the first place.

Coming back around to the beginning, these observations suggest an approach to eliminating the "stand-aside" issue. If someone agrees to stand aside, they are saying that they won't participate but they aren't opposed to others participating. But this indicates that the proposal wasn't really appropriate for consensus in the first place. Better than having those who disagree stand aside would be to come to a formal consensus on a slightly different question—whether it is okay for some to do the activity while others do not. Then no one has to stand aside, but each person can decide individually whether to participate.

The preceding observations suggest a process I will call "extreme consensus". It starts with the assumption that all calls for consensus must be unanimous, and pursues challenges from there.

"On Conflict and Consensus", a guide to formal consensus by C. T. Butler and Amy Rothstein, is a great introduction for anyone considering consensus. I have some concerns, however, about their treatment of the relationship between "the group" and "the person". They write, for example, "The individual is responsible for expressing concerns; the group is responsible for resolving them. The group decides whether a concern is legitimate; the individual decides whether to block or stand aside." My concern is that they leave the question of just how "the group" makes decisions ambiguous. Formally, "the individual" decides whether to block, but here they suggest that it is really not up to the individual, and the group gets to decide. But this negates the formality of the process by abandoning the one potential criterion for consensus, unanimity.

Instead, I would like to understand the call for consensus as follows. Each person individually should consider two questions for each concern which has been expressed: First, is the concern valid? Then, has the concern been resolved? If any person believes the answer to the first is yes and the answer to the second is no for some concern, then consensus has not been reached. At this point it does not matter who originally expressed a concern; when consensus is called, each person should consider every concern. The person who originally expressed a concern cannot "stand aside" if others in the group believe the concern is valid. Likewise, as long as anyone believes the concern is valid it must be resolved by the group.

Understood this way, each concern, once expressed, does belong to the group rather than the individual. The group must decide first whether the concern is valid and then determine when it has been resolved. But in extreme consensus, the only way for the group to make such decisions is by unanimity of each person's rational appraisal of the question.

In this model it is still true that not all concerns which are expressed must be resolved; if no one considers a concern to be valid, it does not need to be resolved. This is far from a "stand-aside", however, as that would indicate that someone believes a concern is valid and unresolved but does not want to block consensus. In extreme consensus, this is not acceptable; formally, as long as any person believes that any concern is valid and unresolved, consensus has not been reached.

Some ideas from On Conflict and Consensus are meaningless in extreme consensus. They write that a person must "persuade" the group that a concern is valid; but formally, "the group" is already persuaded if that one person believes the concern is valid. Along the same lines, they write that during a stand-aside "the person and the group are agreeing to disagree"; but formally, "the group" can't possibly disagree with one of its members. Individuals within a group may disagree, but it is meaningless to talk of "the group" disagreeing or being persuaded.

Considering the preceding discussion of extreme consensus, it occurs to me that the case of one person blocking consensus on bad faith terms was not addressed. A person may have joined an open group with the intent of subverting it, or perhaps just did not understand the group's charter. In such a case this person may block consensus based on concerns which are not actually consistent with the group's charter. This is very frustrating to the other members of the group and can be the driver for abandoning the consensus process with voting or pressure to stand aside. But in extreme consensus, these remedies are unacceptable breaches of the formal process. Instead, the group must recognize these cases as pathological and expect some occasional disruption of consensus meetings while addressing the issue through arbitration outside of the formal consensus meeting. If the person in question is truly out of place, the group needs a formal way to ask or force that person to leave the group, rather than some ad hoc passive-aggressive disregard for particular invalid concerns during a meeting. §

Sampson Synergetics

Copyright 2001, 2002, & 2004 by Justin T. Sampson