In reaction to the contemporary crisis of electoral democracy (marked by decreasing turnouts to elections, marginal party affiliation and general distrust towards politicians), there is a growing interest (from scholars, activists and politicians) in the idea of using random selection for selecting representatives, as was the case in ancient Greece and several Italian Republics in the Middle Ages.
Random selection cannot replace elections altogether. Despite all their shortcomings, elections are a very important democratic tool, in particular because they offer a space for wide participation to self-government, whereas active participation through random selection is limited to a very small section of the population. What is more, randomly selected representatives are not accountable to a constituency. This can have interesting effects in shifting decisions from the preferences of the median voter and reducing the short-termism inherent to elections. Yet it also dangerously impairs legitimacy if a government is not accountable.
For these reasons, the most plausible idea is to have only one chamber of representatives selected by lot (the other and the government remaining elected). What I want to do in this post is to identify the main challenges that such proposal faces. And I count on you to tell me whether you think that there are decisive or not, and if there are other important challenges that I’ve failed to consider.
The first challenge is epistemic: lay citizens lack the expertise to judge important political issues. Obviously, they would need time to grasp all the stakes and understand the technicalities of the questions they would be asked to consider. Yet the task might not be insuperable if they are asked to give an advice only on some issues considered as particularly important. What is more, selected representatives might show more humility and eagerness to learn than elected representatives: they know that their selection cannot be attributed to their merits, and this creates a better incentive to listen to experts. Another frequent argument is that the diversity of the perspectives gathered through random selection might overall improve the epistemic quality of decisions, whereas elected politicians often come from the same background and can show a tendency to think alike and in a distorted way on several issues.
The second challenge is motivational: who would accept to serve in such assembly given the public exposure and demandingness of the task? Unless participation is mandatory (which would be excessively coercive), there will probably be a selection bias. As a result, the chamber would not be statistically representative of the population at large. Yet this challenge is not decisive, in my view, because the increased diversity allowed by sortition would be enough to outperform elections in terms of social biases. What is more, participation can be incentivized with a high wage, a guarantee to get one’s job back, or through secrecy.
The third challenge is independence: how to avoid the corruption or intimidation of selected representatives? Why should this worry us more than in the case of elections? Because party discipline in voting has this desirable side effect of making corruption more difficult: on most issues you will have to buy the whole party’s allegiance. With sortition, if votes are public, it would be easier to identify the pivotal representatives and try to buy their allegiance. Yet sortition also has a major advantage regarding independence: people do not need to raise funds for a campaign and are thus not accountable to campaign contributors. In order to minimize the other risks of corruption, secrecy is once again a potential solution. You could either hide the very identity of the SR or let them deliberate and make all their decisions behind closed doors. Other possibilities are a very high wage, a denunciation reward for selected representatives reporting attempts at corruption or an oath taking with high sanctions in case of breach.
The last challenge is the lack of accountability mentioned in the introduction. Because they would not be chosen by the people and not face the prospect of reelection, selected representatives would have more or less free rein (within constitutional constraints). Would there not be a risk of misbehavior (and an increased risk of corruption)? My view is that this challenge justifies giving only an advisory function to this new chamber. This would reduce the stakes and the temptation to misbehave. For the rest, we might count on public exposure and social pressure to incentivize appropriate behavior. This creates a trade-off between publicity and secrecy (the latter being valuable for increasing participation and reducing corruption), yet I do not see it as insuperable.