Against consensus, for dissensus

SANYO DIGITAL CAMERA[tl;dr for this post: consensus is often a fake consensus, not just because we’re doing it wrong but because we are trying too hard to make it happen and that creates certain pressures, social, animal or otherwise. Sometimes we just don’t agree, and we need to admit that and leave room for disagreement.]

I wanted to summarise and develop some arguments made elsewhere on consensus decision-making. The longer I have thought about it the more I feel that there are better alternatives to consensus decison-making.

Even the evangelist of consensus decision-making, David Graeber, admits in The Democracy Project that the first 20th Century attempt to adopt consensus decision-making (in the 70s feminist movement) failed. His claim is that Occupy got right things that the earlier attempt got wrong. Yet if you read around you find that there is little consensus on the value of consensus at Occupy Wall Street.

Now you can argue that consensus is going to be difficult because we are not used to it. We are not used to speaking for ourselves or acting together. It’s true. Maybe that is the reason it is so hard. But there may be other problems:

1. One weakness of consensus decision-making is already known by those who use it: certain people who are confident, good at speaking, used to speaking, often dominate. We can compensate for that with careful facilitation, but we still have to make a special effort to stop it being swung by the loudest voices. Often, let’s be honest, the efforts fail.

2. The next problem is that the majority can be dominated by the minority. Hardly an improvement on the minority being dominated by the majority. There are ways to deal with this (though I have seen meetings that failed to deal with it). We can allow a ‘stand-aside’ option when things really reach an impasse – this relies on those in disagreement being willing to stand aside. Or we can go for a Consensus-1 or Consensus-2 option, in which we reach consensus with all but one or two people and that’s okay. But we still have a system where a few people (those resisting consensus) can dominate the discussion.

3. What about if someone disagrees with the way the consensus is going, but they either don’t feel strongly enough to try to block it or they don’t feel confident speaking up? This is a very common problem I think. The social pressure on these people not to break the consensus is immense. In the name of banishing oppression we have created an almost irresistable social pressure to keep your mouth shut, particularly if you disagree without feeling strongly about it. We can try to create an attitude of openness to all disagreement but the reality is that, in the face of social pressure felt differently by each person, we try and often fail. And so people don’t speak.

4. Now let’s consider something that consensus decision-making really struggles with: sometimes people’s views simply cannot be brought together. People are different. They see the world in different ways. Perhaps because of their experiences, perhaps because of their position in society – and perhaps you don’t know the reason. But it is dangerous to deny that real differences exist. Try putting a banker and me in the same room to discuss economic policy if you think all viewpoints can be brought together. I suspect most of the examples of functional large-scale consensus decision-making come from more communal societies than ours, where people start from a position of less divergent opinions.

5. Another problem is quite fundamental to the way humans work. If you examine the ‘how to’ documents for consensus decision-making, you get the impression that if consensus isn’t happening, you should just work harder to get at what people really think, to bring out their underlying motivations. But this ignores the realities of how people interact. Often people don’t even understand their own motivations; the idea they can or will want to always expose their thoughts to others is rather idealistic in the bad sense of the word, i.e. totally unworkable. Think about really egotistical people you know. Will they, when pressed, finally admit to their egotism? The chances are they won’t even recognise it, and if they do, they’ll go to great lengths to hide it. But even with less egotistical people, it’s strange to assume they can always understand their own emotional reactions which go to form their stance on a decision – people undergo years of therapy to get at that and you want it in a meeting of a few hours? We can’t design a decision-making system that’s dependent on the untrue notion that if everyone just works a little harder we can all be transparent to each other.

6. Another more subtle problem I feel is that there is a hidden individualistic streak within consensus decision-making. In theory consensus should involve being prepared to make compromises but it is not always interpreted that way. Quite the opposite: for many people it seems to give them space to think ‘I can be part of a group while never compromising on my demands.’ In a society that propagandises the virtues of individualism (even if frequently failing to practice them) it is easy to assume that compromise is oppressive – but it isn’t necessarily oppressive if we choose it ourselves. And for collective action to happen, particularly on a mass scale, we have to get used to compromises. It simply isn’t possible to get millions or even thousands of people all wanting exactly the same thing – or only for brief moments. Good consensus working recognises that if we are to work together, we are going to have to learn to put aside our own opinions sometimes – perhaps only temporarily, perhaps for a long time – in order to agree on actions. But I think it’s important that when we do so, we are able to express our dissent, and that it be clear that agreement is partial and contingent – because that’s what real, messy collectives involve. Consensus decision-making purports to be about collectivity but in my experience it can often hinder collectivity in practice by over-emphasising the individual, allowing people to think that any decision they disagree with is oppressive and to be resisted.

7. There is also a problem with being honest with each other about the process and our part in it. let’s go back to the idea that sometimes people’s viewpoints are genuinely different. If they are too different to bring together, are we going to pretend otherwise? Or perhaps we’ll find we have tried to ensure our views are all heard by forging them into a consensus, but then lying in bed later we realise that the ‘consensus’ isn’t actually what we want. Perhaps everyone will have that realisation (The Abilene paradox is the name for when a group of people collectively decide on a course of action that is counter to the preferences of many of the individuals in the group). At what point would it be best to raise this doubt, having spent hours hammering out a consensus? What is the right moment to reveal that our viewpoint breaks the consensus we ourselves helped create? And if we dare to do that, what if consensus is never acheived? Are we never going to admit that we can’t get there?

8. Then there is the issue of meetings, how long they take, and whether they become an endurance test. My suspicion, having spent a lot of time around groups that use consensus, is that consensus often doesn’t exist. They are fake consensus groups. You can hear it in the mutterings after the meeting, or the sighs during it. But in the end you just want the meeting to end, particularly after it hits the third hour, and sod it, why not just go along with it? At the very heart of your process is the notion that all views will go into the consensus, but now you’re sick of the meeting, you’re hungry, you’re tired of arguing, and you want to go home. I’m not sure this is even the exception. I suspect that those who love meetings often win out, and most consensus isn’t real.

9. Finally we need to look at how decion-making structure affects a group or institution in the long term. It seems to me that the harder it is to make a decision, the more (small c) conservative the group will be. To fail to make a decision maintains the status quo. To get change requires positive agreement. Requiring consensus will, in many circumstances, dramatically raise the bar in terms of how difficult it is to get a decision. The effect of this over the long term will be to make it difficult to institute changes within the group or changes in behaviour.

So, recognising all these flaws, is consensus decision-making so fantastic and radical that we should wear ourselves out trying to make it work? If you do it properly it won’t wear you out, says David Graeber, but I want consensus on that from a meeting of a hundred people who’ve used the process for a year.

If you can make consensus decision-making work for you, great, but it just isn’t so wonderful that the rest of us should kill ourselves trying to make it work. I think it probably works best in small groups of people with a fairly constant membership. I think personal relationships can give a depth to consensus decision-making that almost make it live up to the radicalness ascribed to it. But I know it rarely works in large groups, and in groups where people are constant passing in and out.

I used to be in a group that encouraged dissensus rather than consensus. That is, we admitted that sometimes people have differing views and we should let the conflicts between them emerge and play out, potentially over long periods. The actual decision-making process went like this: we would make an initial attempt to reach consensus. This would enable everyone to have a chance to speak and discuss the issues. Then, if it was apparent we didn’t have consensus, and if all the points of view had been expressed, we would vote in an open vote (A two-thirds majority may be reasonable, though note the possible conservative tendency in point 8 above). The vote shows who is for or against a particular course, and the positions of opponents are known. They aren’t being ignored. The group has tried and failed to take their point of view on board – perhaps for good reasons, perhaps for bad reasons – but the disagreement cannot be brushed over and it will not disappear. That’s a good thing, because the disagreement is real, and consensus would not be.

This method gave us the advantages of consensus through the attempt to reach consensus initially. And there are advantages to consensus – it creates a process in which we try to work collectively towards a position we can share, with everyone valued within that. But I feel that departing from pure consensus allows us to be more honest with each other – we never had to be terrified of breaking the consensus and accidentally extending a meeting already at 3 hours to 5 hours. Dissensus also – and this is an important point – worked, by which I mean we reached decisions without constantly wearing ourselves out. If your meetings are too long, people will stop going. Meetings are always, however we try to lighten them with good friendships and cake, going to be something of a necessary evil. Most people can’t take too much boredom, or friction, or sitting still, or arguing. If you can, good on you. But you will be the remnant who end up pushing decisions when others are so tired of the meeting they’ll just shut up in order to end it. And how democratic is that?

As an aside, within dissensus decision-making, as in consensus, there also remains the option (though we didn’t use it often) of those who agreed on a course of action taking it, while those who disagreed stepping out. It’s something that can be done where the group itself isn’t bothered by some members taking that action. As soon as it requires, for example, use of shared and limited resources, it doesn’t work. I hesitate to recommend it as the ideal path because, as I said above, I think it awakens our inner individualist again – Ah! we think, Here’s a way I will never have to compromise on what I want! – and that is dangerous, because we need to compromise or we will never organise together. But it will sometimes make sense to do this.

So here’s a more complete depiction of dissensus decision-making procedure:

1. A proposal is made, the group tries to reach consensus. Modifications may be made to the proposal. Consensus may be reached, or it may be decided that no consensus is necessary in this case – a few members of the group can implement the decision without bothering the others.
2. If, after reasonable attempts by all sides to accommodate each other have been made, no consensus is reached, a vote is taken using the agreed voting method. The facilitator will decide the right moment for the vote. The right moment can be the subject of debate, and people may raise objections, but if you have a facilitator who can be held accountable then they need to be trusted to make the final decision. Making a group decision on how to make a group decision when in the middle of a contentious meeting is not a viable option.
3. The group considers how those who lost the vote can be accommodated – this may have been dealt with already in the stage 1 discussions but it is often worth raising again after the vote.
4. The debate between differing views continues over the long term, and decisions can always be re-considered if appropriate.

An important message here is that we do not need to reject all voting as wrong. I understand that representation doesn’t work. I understand that majority voting alone is oppressive. But I think that consensus decision-making has equal and sometimes worse problems. It is too unwieldy to take decisions effectively, allows egotistical people and meeting-lovers to dominate, and ultimately it can make us lie to ourselves and to each other. In the long run, to oppose the current global order, I think we need to build large-scale local, national and international organisations. We simply won’t do that with consensus decision-making.

To illustrate the difference between consensus and dissensus, let’s imagine a group is trying to reach consensus on whether we should make decisions by consensus. Sounds silly? It’s actually kind of logical. If you can’t reach consensus on that, there’s bugger all else you’ll be able to reach consensus on. Problem is, I’m in this group. I’m against consensus, for the reasons I have outlined above, and I won’t allow consensus to be reached on this, because I feel the viability and self-honesty of the group is at stake. In fact, I will agree to consensus decision-making over my dead body, because I actually care about the group being effective. Stalemate.

There is a way out of the stalemate. Someone could propose that the group operate on Consensus-1 instead. Problem solved. But would you really feel the problem had been resolved? Doesn’t it just leave something simmering away under the surface? What is the price of ignoring the dissensus? What if one member of the group changes their mind next week? Would they dare mention it? Would they rock the boat? Would they force it to a vote again? And if they did, would someone simply propose Consensus-2 instead? And would that make the problem go away?

One of the advantages of dissensus decision-making, and to me a signal of its viability, is that it has the ability to contain its own opposition without tying itself in knots of dishonesty. Sometimes people simply don’t agree, and we might not all be agreed on that, but that’s fine. At least we all know where we really stand, and are not under pressure to conform to a consensus the reality of which we can never feel sure of.


  1. Sam

    Interesting post. I’m broadly agreed with you on the problems of consensus decision making, though I have very minimal experience of it. Firstly, I was wondering whether you would condone a minimum level of support in dissensus voting, I.e 60-40 in favour?

    Secondly, I was wondering what your thoughts are on upward scaling of such processes, or on the most effective way to run larger-scale democratic decision making. Obviously it’s not ideal to be scaling up, but very little is ideal at the moment.

  2. Sam

    Also, you say representation doesn’t work. No arguments about the dominant form, but what do you think of the limited form used by the Pirate Party in which members can either vote directly or invest their voting power in other members, and withdraw it at will? I’d say that’s about the best form of rep dem I’ve seen. It could potentially be complicated in various ways too. For instance, your voting intentions could be specified on broad or individual issues such that your chosen representative is unable to draw on your vote to go against your wishes.

  3. preorg

    Thanks for the comments. Yes, I can see some value in going for a 60-40 or two-thirds minimum rule when voting, though it could create a slight conservative bias through protecting the status quo. But 50/50 splits are pretty awkward and there’s a feeling of unfairness when someone wins with a 51/49 majority. I remember thinking in the run-up to the Iraq war when the polls were split about 50/50 that 50% was not enough to take a nation to war on.

    I think scaling up is important and another reason I don’t think it worth putting huge energy into the consensus model. It definitely won’t scale up – the more people involved, the more horrendous it becomes to manage. The best method I can think of is using dissensus decision-making but in a bottom-up hierarchy of councils. So each group at the bottom would send a recallable delegate to the next level council. Again this wouldn’t really function with consensus – delegates may well be under instruction to pursue a particular line.

    A system where someone is carrying out the expressed intentions of those below them I would call a delegate system to distinguish it from representation as we know it. I think the kind of internet-based voting systems that allow that are an improvement on the current system. My main problem with some of them is they struggle with the learning and discussion processes that are necessary for informed decision-making, so there can be a risk of people just pooling their ignorance/prejudice through direct voting, rather than having a chance to learn from each other. And I’m not sure that problem will ever be solved purely online – there’s no substitute for face to face discussion sometimes. I can imagine some ways to deal with this – for example, you only get to vote online if you have turned up to a locally-held debate on the issue.

    1. sam

      I’m agreed on the problems of simple direct democracy and having used a range of online message boards, I’d agree that they’re no substitute for face to face discussion.

      That said, the realities of scaling up are such that you can expect the majority of the population to ‘not turn up’ even at a local level. I don’t think that’s entirely a bad thing either. I think it’s mistaken, and perhaps unreasonable, to expect everyone to make that kind of regular commitment. One solution, as you say, is to require attendance before assigning voting powers. However I fear this would lead to there being a clique who regularly attend and who would likely be seen as distant and exclusive by those who don’t attend, as well as by those who try to go once or twice.
      But, as you say, we don’t want uninformed voters. One option is to webcast meetings (also allowing web-conferencing for those with accessibility problems) and to have summaries and reports uploaded for public viewing. Perhaps online debate could also be built around this.
      One model which has been tried in some forms of deliberative democracy is to have the deliberative group produce an information leaflet to be sent out to all voters.
      Essentially, there are risks both in closing the vote down to those present at a meeting and to opening it up to those uninformed. I don’t think these problems have a perfect solution.
      Other compromises available: everyone has a vote, but those attending meetings have 10 votes – or something similar. Another compromise, in line with the Pirates, would be to allow non-attendees to invest their vote in an attendee – perhaps once they have at least ticked a box saying that they’ve watched a video/report/read the minutes etc.


Leave a Reply to preorg Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>