Victory in democratic systems seems straightforward: get a majority and rule unopposed. To combat the obvious dangers of this, various “checks and balances” exist which create multiple power centers, each of which exercises an effective veto over some aspects of the others. This strategy goes some way to preventing autocracy, but it can also make society ungovernable as checks and balances become roadblocks and bunkers. Ironically, this can make autocracy more likely, as voters turn to the executive branch to break the logjam.
But how did we get into this position in the first place? Why is majority rule so dangerous? There are two problematic aspects of legislative assemblies: 1) they make their own rules, and 2) the victory condition is likewise internal. We have come to accept this in politics as a matter of habit, but we would reject it as corrupt in almost any other context. In sports, teams must play scheduled games no matter what; they cannot cancel games just because a player is injured. When companies bid for a project, the final choice should represent the will of an external player, not some sort of internal discussion among the bidders.
Our failure to define rules and success criteria for assemblies that are exogenous to those assemblies is, in my view, a fatal flaw. No assembly that sets its own rules and winning conditions can be viewed as democratic, no matter how its members are chosen. That’s where the citizen jury comes in. The assembly should not be viewed as a body for passing laws, but for proposing laws. The citizen jury takes the options and renders a final verdict. The citizen jury determines the winner. This is exactly what juries do in court: they listen to arguments from paid advocates (lawyers) as well as testimony from witnesses. They then synthesize all that information into a final decision. A lawyer can use all the tricks she knows, but in the end, the jury can still act as it sees fit.
Once we have the victory condition settled, making the rules is not so hard. We want multiple outputs from the assembly, so we lower the threshold of passage. Since the 50% + 1 rule guarantees a single output, a threshold of ⅓ + 1 gives us two outputs, a threshold of ¼ + 1 gives us three, etc. In addition, we need to set the legislative agenda, and we must ensure that the legislative agenda is binding: it cannot be blocked or delayed by any action of the legislature. The agenda can be set in a number of ways. The simplest is to make setting the agenda the first item in any legislative session. The citizen jury picks its favorite from among the options sent to it by the assembly, and that’s the agenda. But there are other ways. For emergency items, the assembly could add to the agenda on a majority basis. One can also imagine signature drives, or perhaps even a separate political body (elected or chosen by lot) to determine the agenda.
So much of the current sortition literature is about replacing elected politicians with randomly selected citizens in the assembly. But what if the assembly itself is flawed? Throw 500 people into a room and say, “solve our problems”? Ridiculous! It doesn’t matter how you select those folks. The assembly has no agency, it’s just a bunch of people. Factionalism arises to supply the missing agency, not because elected politicians are a distinct species from ordinary citizens.
Let’s instead set people up for success. Give members of the assembly (however they are selected) clear, externally defined rules, and a clear, externally controlled success criterion. The citizen jury does this: a group of randomly chosen citizens, selected after all options have been finalized, whose only task is to choose the best option from a short list. This in turn allows the assembly to ditch the ridiculous pretense that it acts as one. It can instead pass multiple solutions to every question, and let the jury decided. This is Democracy for the win.