Much of sortition literature focuses on the legislative branch. There are two reasons for this: from a theoretical point of view, the legislature is the symbolic repository of democratic sovereignty, and from a practical point of view, legislatures seem to be most in need of reform. While the latter point seems to me inarguable, the former is not, and in fact through much of pre-enlightenment history it was the executive—personified by the monarch—that held the nation’s sovereignty. Indeed, this was almost literal, as the sovereignty of the nation was symbolically vested in the crown and scepter carried by the monarch.
This neglect of the executive is also ahistorical. The Athenians considered random allocation of executive offices to be essential to democracy, and selection of magistrates in the Italian city-states had a heavy lot-based component as well. The logic of picking executive officers this way seems obvious. The great danger of the executive branch is that it can become a kind of open conspiracy. If everyone in the executive branch is appointed by a single person, and can be removed by that same person, then the entire executive is just the plaything of the apex officer. The distinction between presidents and kings here means little more than that presidents have only a few years to effect their plans. Selection by lot interrupts this conspiracy, and ensures some level of pluralism.
Depending on how it is done, however, we might pay a high price for choosing magistrates by lot. Particularly in large, advanced societies, the quality of officers matters greatly, and the pool of such officers must be highly professional. Just choosing by lot from the general population would be a disaster. Fortunately, the proposal-jury system solves this problem easily: Proposing bodies, analogous to today’s political parties, submit lists of what they believe are qualified executive officers, and the jury picks the best from those lists. I have described this system in greater detail in a previous post.
But how are these various officers to be incorporated into the executive? How are responsibilities to be allocated, when offices higher up the chain try to take over responsibilities from lower offices? How do we sort out the obvious conflicts that accompany organizing the executive branch as a pooled service?
A good place to start would be to ask how a well-designed hierarchy populated by good, honest administrators would work. Like all hierarchies, it would have “branch” offices that have some oversight authority over other offices, and it would have “leaf” offices that oversee no lower offices. So how would a branch office holder behave? First, there would be a clear delineation of each leaf office’s scope, and the branch office holder would delegate tasks to the appropriate leaf office in every case. Second, for issues that cut across multiple leaf offices, the branch office holder would make decisions about how to coordinate the various leaves. Importantly, however, they would still act through the leaf offices. In other words, the superior office would not cannibalize issues from lower offices, it would only coordinate and supervise.
There are also some issues that are specific to government. A corporation can appoint and remove office holders from within the hierarchy; that is, decisions about hiring and firing for lower offices can be made by higher office holders. That’s a problem if we want the executive branch to be pluralistic, because office holders will represent competing political ideologies. We must add the requirement that all decisions about appointment and removal must be made exogenously to the hierarchy itself. As discussed above, we already have a mechanism for this in the citizen jury.
It is now easy to lay out the principles of a coordination hierarchy.
- Each leaf office has a well-defined scope of authority
- Each branch office has two or more minor offices below it. They can be either leaf or branch offices.
- The scope of each branch office is defined entirely by the combined scope of its minor offices. That scope extends to any issue that crosses two or more of those minor offices.
- A branch office must act through its minor offices, and hence cannot take an entire issue over for itself.
- All office holders can only be appointed and removed by processes outside of the hierarchy.
One important feature here is that branch offices have no original responsibility for any issue. This is to prevent another way that a branch office could cannibalize its minor offices. Let’s say, for example, that there is an office to deal with worker safety, such as the head of the Occupational Safety and Health Administration (OSHA) in the U.S. Let’s say this is further divided into three areas: heavy industry, light industry, and agriculture. These three areas would have to be separate minor offices. If instead we created two minor offices for light industry and agriculture, and granted the branch office original authority over heavy industry, the branch office might be tempted to define more and more items as relating to heavy industry, so as to avoid having to work with minor offices that might be populated by members of a different political party. This tendency becomes more dangerous when dealing with offices that have immediate coercive power, such as law enforcement and military offices.
While this structure may seem to have little connection to sortition, in fact sortition, in the form of citizen juries, is absolutely essential. There will be thousands of independent executive offices in this system. The hierarchy will have many layers, as befits an operation of such size and scope. In fact, the system should be required to have a minimum number of layers (say, 4) separating the apex office from all leaf offices. Only a system of citizen juries has the bandwidth to select these officers in a democratically and epistemically satisfactory way.
This is, I believe, the only structure that can ensure both that the executive has the capacity to accomplish its tasks, but is also pluralistic and transparent in its operations. The danger we live with now is that the apex officer “goes rogue” and uses the executive as his (it is nearly always a he) personal power base. This happens so often, in fact, that it is really misleading to describe this as “going rogue”. In future posts, I will flesh out the system, especially with regard to conflict resolution, and deal with objections to this system, particularly from the anarchist perspective of why we even need an apex office to begin with. But compared to our current, dangerously unitary executive, this system cannot be seen as anything other than a huge improvement.