Adventures in consensus

Some people like to use consensus decision-making processes. Some like it so much that they have declared that all other decision-making techniques are oppressive. When people are first confronted with the idea of needing unanimous support to approve of decisions, hypothetical problems immediately occur to them. What if you can’t get unanimous agreement? What do you do then?

I’ve yet to actually see anyone overcome this problem. The answer that is usually given is that you just don’t go forward with it if there isn’t unanimous agreement. But this answer assumes there is an agreeable baseline for what constitutes not going forward. It requires a theory of the default, which is impossible to construct. At minimum, it is impossible to construct it without disagreement, and that disagreement then recreates the whole unanimous consent problem once again.

Consider the following real-life example. At an occupy encampment, a proposal is made to ban smoking inside the perimeter of the encampment. As many left subcultures are big on smoking cigarettes, this generates considerable opposition. Unanimous consent is not achieved, not even close, and blocks are made.

What follows from what just happened? Well, you could say that what follows is that there will be no action to ban smoking inside the perimeter. But wait a minute. There was never any proposal to allow smoking inside the perimeter either. Such a proposal would also have failed to get unanimous consent and elicited blocks.

So there is not consent to allow or ban smoking in the area. How do you proceed? Here is where the theory of the default is necessary. You’d have to have a theory of what the default rule is, i.e. the rules governing what is allowed when no proposal has been passed one way or another. What is the default rule though? Is it to allow or to ban?

I think one natural reaction is to say that you allow it. Everything is allowed by default before it is banned through a proposal passed by consent. This is the most autonomy-promoting or whatever. But this can’t be true. Surely, it is not the case that by default I can go around the encampment punching people in the face until we have a proposal banning it.

So now you have to modify the default rule. Instead of everything being allowed by default, the new rule is that everything is allowed except if what you are doing physically harms others. This ensures that I cannot just punch people by default, but it throws into doubt what the default position on smoking should be. Does smoking harm others? It seems that second-hand smoke does harm others. So shouldn’t the default rule have been that you cannot smoke in camp, and the only way to allow it therefore would be through a proposal passing with consensus?

This approach to the default seems alright, but it problematically turns on empirical assessments. We know now through studies that second-hand smoke harms bystanders. But what if we did not know that for sure? What if some folks just thought it harmed others, but there was not definitive evidence? How would you proceed then? This theory of the default will not be able to function because there will be disagreement about what harms others and there will not always be evidence to settle that disagreement. So you find yourself in another consensus bind.

More than that, even when you have evidence, members of the group might deny that it qualifies as evidence. All sorts of people reject scientific research for instance, which would include, I’d presume, epidemiological studies. The left in particular has its fair share of New Age types who have beliefs about harms that are not verifiable scientifically. What do you do if one of them says wearing blue in the camp harms them by throwing off their chakras, and neither a proposal to allow nor ban wearing blue can get consensus?

This is just the tip of the iceberg of course. There are more interesting questions in how exactly you could use this process politically, especially in deciding who belongs to the community of people from whom consent is required for approval. For instance, when you decide or don’t decide to build more housing in your area, that affects tons of people who are not in your area, e.g. those who would end up moving there to fill the new housing at some future date. Do they have to be included? If yes, how could you even possibly identify who they are in advance? It’d be impossible.

For the most part, I find it really hard to get that enthused about process stuff altogether. In a universe of physical scarcity like ours, it is impossible to avoid some kind of force on others absent unanimous consent from all people about all things at all times. Once you admit that bar is foreclosed by the mechanics of a universe with matter scarcity, it seems like substance shows itself to be of way more importance than process.