Non-governmental Uses for Sortition

From "The Trouble With Elections: Everything We Thought We Knew About Democracy is Wrong," Chapter 15.1

Up to this point I have focused primarily on sortition as it applies to governmental units, whether national, state or municipal. However, there are many sorts of non-governmental associations that nominally operate (or seek to operate) democratically. These would-be democratic organizations include unions, consumer cooperatives, housing associations, credit unions, worker-owned businesses, political parties, and even global non-governmental organizations (NGOs). Even virtual associations such as Internet platforms could function democratically. There are also countless organizations that do not currently seek to operate democratically, but perhaps should, including schools, charities, and even for-profit business firms. These would-be, and could-be, democratic associations have a pretty poor track record in terms of achieving equality, participation, and self-rule. Many of the failings that afflict nominal governmental democracies are also visited upon non-governmental democracies. Some problems, such as expensive election campaigns and gerrymandering may not apply, but the more fundamental problems, such as rational ignorance and elite domination are rampant.

Very small organizations may function using some variant of direct democracy. Think of the stereotypical worker-owned business that has frequent all-member collective meetings. These micro-democracies have their own problems, including time inefficiency. But the tougher issue is that this direct democracy isn’t scalable to huge organizations. With more and more members, at some point, virtually every democratic organization resorts to some sort of representative system by necessity. Such representative systems almost universally rely on elections. Elections are the only familiar tool for representative selection. The jury model of sortition never enters the minds of members as an alternative to consider. In this chapter we will explore the advantages of sortition. Sortition can allow non-governmental associations to solve many of their democratic deficits. It is generally easier to make big reforms within private associations than within governments. But even some private associations may be chartered under state statutes that mandate the use of elections — so some research needs to be done before launching any specific reform effort.  By experimenting with sortition in non-governmental associations, democratic reformers can amass a wealth of real-world experience that will also be useful for subsequent governmental reforms. Trial and error is an essential part of perfecting democratic procedures and design.

When is Sortition Appropriate

It is important to note that sortition is not always the appropriate democratic tool. Dissatisfied members of an association have the classic choice of “voice or exit” — speak up or quit. Sortition is most useful for larger organizations with defined memberships, where exit by individuals would be a personal hardship. Governments meet this criterion, as do associations dealing with necessities like housing, or employment. Simon Pek has argued that sortition might be a good tool for reviving democracy within labor unions . While a person might technically be able to leave their union or job by quitting and looking for new employment, and a condominium owner might be able to leave a condo association by selling their unit and moving, the effort entailed makes this an impractical hardship. Exercising voice through some democratic process is fitting. In contrast, voluntary advocacy organizations with more porous membership boundaries, (such as political parties or movement organizations seeking to make changes in society), do not satisfy this criterion. Members of such organizations have a wide range of commitment levels, with many members being peripheral. Members who disagree with the direction of their organization can easily quit. They can even help form a new organization more in keeping with their preferences. Such associations might prefer the principle that the more one contributes through personal effort to the project of the association the more say one should get. Members who attend a lot of membership meetings and do lots of volunteer work, get to have more say about how the organization is run. This is an alternative to the democratic principle of political equality. 

Deciding when sortition is a good fit is not always obvious. In an example at the intersection of governmental and association representation, the leftist Mexican political party, the National Regeneration Movement (Morena) has used random selection to nominate some of its candidates.  In 2015, 2018 and 2021 Morena used a lottery among a pool of party members who have been advanced by district party assemblies to nominate a portion of its candidates in the party list proportional representation election for both chambers of the national legislature. This is an unusual application of a sortition, both because of the restricted pool from which nominees are drawn and because those nominated then need to pass through a filter of election. This electoral filter, however, is very different than in the single-seat constituencies we are accustomed to (and which Mexico also uses for most of the seats in its legislature). The winning seats elected by proportional representation are filled from a party list in proportion to the national vote for each party.  This means that the individual candidate-based “principle of distinction” as described by Bernard Manin (and discussed in chapter 7) we are accustomed to in single-seat winner-take-all elections is less dominant. However, this lottery nomination process was essentially symbolic, as it only filled out the bottom of the party list, making it unlikely any of the lottery designated nominees would actually take office. Even if the full slate were nominated in this manner, since those names put forward by local units would likely a “certain sort” of people, this lottery process would serve more of an anti-corruption function (to avoid obligations and deal-making with party leaders) than a democratic function.

Platform Democracy

In addition to long existing member associations that seek to govern themselves democratically, there is a new sort of association made possible by peer-to-peer creations on the Internet. As an example, suppose some large Internet entity such as Uber or facebook were owned by its users in a cooperative, rather than owned by stock investors. The term “platform co-op” can be used to describe such organizations. Many platforms can develop with minimal upfront capital. The paradigm by which the founders eventually get to cash in with a public stock offering is not the only way such Internet firms could mature. A democratic co-op option is possible. But how could an organization with members who may be spread across the globe, most of whom have minimal or no direct knowledge or interaction with other members beyond that mediated through the particular application, run the organization as a democracy? How could they hold elections in which members are well informed about issues and candidates (when even high stakes governmental elections fail this standard)? Certainly the Internet could allow all members who wished to vote  directly on policy in a sort of indirect-direct democracy, but the problem of rational ignorance would still dominate. Elections are only a responsible governance tool if the voters understand what they are voting on. Rather than throwing one’s hands up and concluding that a global democracy of a business or other sort of association with many millions of members is just too impractical, we need to consider sortition. 

Imagine a cooperatively-owned ride sharing service. It might be owned by the drivers, or by the riders, or both. Let’s take an example owned by the drivers. A feature of a lottery, is that although for typical associations it is standard to give an equal probability for each member to be drawn in a governance lottery, it is also possible to tailor it to unique circumstances. We could assign drivers proportionate chances in the lottery based on how many hours per month they drive, for example. If a driver governing assembly is drawn, it could fairly reflect the interests of the drivers as a whole, including a balanced share of full time and part time drivers. These drivers might get a healthy salary to devote themselves to the issues of governance for a set period, in lieu of their regular driving time. But with regular rotation, these directors would know they were making decisions that would affect them once they returned to being drivers. There are numerous possible variations on this theme, in terms of what tasks mini-publics might be charged with, their terms of office, their size, etc. A worker-owned business might decide to use a stratified random sample to assure each department gets an appropriate number of members (such as from the sales department, factory floor, human resources, etc.)

Sortition could even be a useful tool for other sorts of business governance. The problem of rational ignorance and ill-informed members of a traditional stock-ownership corporation being asked to elect a board of directors could be addressed in a similar manner as the co-op model described in a previous chapter. A “representative” subset of shareholders randomly drawn (whether based on equality or based on the number of shares owned) could constitute a  hiring body that becomes well-informed and crafts a desired board of directors. While the term “democratic” is inappropriate in situations in which there is intentionally no equality among members, sortition can still play a useful role. While I am not advocating this reform (more akin to the aristocratic use of sortition in Renaissance Italian city states, than democracy), the fact that this tool hasn’t even been considered is striking, but not surprising. Existing elites hold their current positions thanks to the status quo system of selection, and it would take an incredibly well organized shareholder revolt to change it. Sortition might better be used within a capitalist firm in conjunction with something like the German codetermination (mitbestimmung) laws, which require large investor-owned firms to include worker representatives on their boards of directors.

Public oversight role for sortition

The anti-corruption potential for sortition suggests new sorts of public oversight mini-publics. There are situations where some activity has a significant impact on the public, but where the public has no, or ineffective input. Here I am discussing quasi-governmental functions, which are not currently handled by existing governments, but might be well suited for sortition oversight. I will not present any fully formed proposals here, but merely suggest the sorts of things sortition could be used to address.

Some cities that have had serious problems with relationships between police departments and the community, have appointed community oversight boards. Rather than having elected officials appoint a group of self-styled “community leaders,” a community panel that is fully representative of the community could be drawn by lottery. The goal would not be to assemble leaders or experts, but rather facilitate the expression of community priorities and values. Any such oversight body obviously needs its own staff and control of information flows independent of the politicians and the entity being monitored. The body would also need to have real power over the organization they were overseeing. 

Some examples of possible uses will help. As technology advances, we are entering an age in which aerial drones and public video cameras, linked to facial recognition software and identity databases could make policing much more effective, but also subject to abuse and loss of privacy.¹ If such technology is to be deployed, it makes sense to assure that it remains permanently under democratic control of the populace and that no government officials or police department can abuse it. The jury model is a natural fit here. In ancient Athens, public officials were subject to review by an auditing jury, as a further protection against corruption. A  constantly refreshed body of ordinary citizens supported by their own staff could prevent or punish abuse today as well. Citizens gathered into a jury have an interest in fighting corruption such as bribery. One by one, citizens often have an immediate short-term interest in participating in corruption, such as seeing their own individual bribe be successful (to speed the issuance of some personally needed permit, etc.). But as a group (on a jury), they have a more powerful democratic community interest in punishing and preventing bribery and corruption as a whole.

We have barely begun to scratch the surface of all the places where sortition could improve public life.

1. There are already companies seeking to contract with municipalities to provide constant aerial drone video recording of entire cities, allowing police to review video to determine, for example, where each car that was near a crime incident came from, and where it went after the crime.