There are many ways that groups of people make decisions. Some of them are conspicuous: we talk about “collaboration” or “coming to consensus”, for example. Other cases are more subtle: sometimes decisions are “handed down from above”; the price of some good reaches an equilibrium in a market; a conversation shifts to a new topic. In each of these cases, a stable outcome is reached after being influenced by many individuals. Even when Jones, CEO of Company X, dreams up a policy that his middle management has to implement without consulting them first, the fact that they do implement it involves many individual decisions not to resist the new policy, to grumble or shrug to coworkers instead, and so forth. If middle management pushed back violently enough, we would not say that any decision had been reached.
Probably the most conspicuous form of distributed decision-making that we regularly come across is voting. No decision-making process has been more touted as the paradigm of fairness: everyone gets her say, and everyone’s say counts exactly the same. Insofar as this is true, and leaving aside the contingent matter of what is (or isn’t) being voted on, voting is unquestionably fair. At any rate, I do not want to voice any qualms about it here.
I want to urge instead that voting might be so conspicuous because, as decision-making processes go, it is a fairly crude device. Decisions arrived at by vote are binary: they cannot solve any problems that cannot be phrased as a yes or no question. (Compare the process of shifting a conversation to a new topic. Imagine if, each time a group of people were talking, they had to stop and enumerate all the possibilities for new topics when they wanted one, and vote yes or no on each one. No one would ever speak to anyone else.)
Imagine that making a decision is like re-routing a river. In a vote, each person gets a single stone with which to try to adjust the flow of the water. If everyone places their stone along one bank, the river will bend in a new direction around them. This is unlikely, however, especially if everyone is allowed to place their stone simultaneously and independently, according to their own desires and interests. Some stones will be placed directly opposite others, negating the effects of both. The river may flow through the entire decision-making process without changing direction much at all; it might only speed up or slow down a little.
Many of the more subtle decision-making processes in our lives are not like this: we do not make our choice of where to place our stone independently. Some people place their stones first, and perhaps they explain their reasoning to others; each person that goes in succession to place a stone can respond to the earlier placements of others. If the collective goal of re-routing the river in a certain direction is made explicit, it becomes much easier to achieve in this way. If there is no explicit collective goal, there will often be an implicit or unconscious one that develops as later stone-placers try to read off the intentions of earlier stone-placers from the stones already in the river. Even if there is no explicit goal at the outset, the outcome of the stone placing is likely to be much more interesting here, because it is much less likely to be a zero-sum game: when individual decisions can be made in a way that responds to the decisions of others, they will tend not to simply cancel out.
In this way, it’s possible for almost any collective decision-making process to be more efficient than voting.
We should not hold, then, that a decision-making process is necessarily good simply because it’s fair, or that a different process is bad because it’s unfair. (Ask yourself if it is unfair when someone else begins talking about something you hadn’t thought of.) There is much waste to be avoided, I think, if we look to our more mundane lives for models of how to reach consensus on a larger scale.