Walter Lippmann wrote a haunting paragraph, which I’ll paraphrase thusly: Man’s problems are complex. Man’s capacity is limited. So how is Man to master his problems? That is the conundrum of the age.
The answer (obviously—or not so obviously) is to cut Man’s problems down to manageable sizes and designate task groups to deal with each. Divide and conquer, in other words.
In governance, this cutting down implies topical specialization of the governing entity. In other words, it implies many (say two dozen) topic-focused mini-, or Demi-, legislatures at the state and national levels, corresponding to the existing congressional committees at those levels. For example, there would be a Demi-legislature for topics such as health, education, welfare, commerce, labor, transportation, communication, the environment, justice, the interior, etc. Specialized Proxy Electorates would oversee each specialized Demi Legislator.
This topic-specialized, semi-elected, long-serving, small-sized arrangement is not open to the criticisms below of a “citizens jury,” (which many sortition fans endorse), which is unspecialized, randomly selected, new-to-the-job (inexperienced), and blob-sized.
… the differences between a jury system and government by lottery are profound. A jury consists of only 12 people. These 12 are chosen rather carefully…. The questions they must decide are rather limited —generally only a single question of right or wrong in a specific instance, and within the framework of a well articulated, body of law and precedent—and in this decision they are guided by a judge, who explains carefully what they can and cannot consider…. This is qualitatively different from throwing hundreds of people randomly chosen into a room, with huge numbers of issues …. —Malcolm Margolin, quoted in Ernest Callenbach & Michael Phillips, A Citizen Legislature, 1985, p. 77-78.
Back in 1985 there was no internet, which is what enables the “Demi” arrangement I propose. In other words, back then citizen-deciders had to be physically assembled, which couldn’t be imposed on them every month or two, and couldn’t be imposed on the much larger number of them in total that Demiocracy requires. Only via the internet, appropriately employed, can Demos adequately be on top of things.
Details:
• At the local and county levels, Proxy Electorates could oversee multiple topical areas. (At the local level, maybe all of them.) So, fewer topic-focused legislative groups would be needed there.
• Ranked Choice Voting (or “Approval Voting”) would be employed to elect the winning legislator. This method is especially useful when there are a large number of candidates, as there sometimes would be. It is also useful when the leading two candidates are too divisive to get a majority, but a trailing third candidate is the second choice of the majority.
It is a maxim, that those whom everybody allows the second place, have an undoubted title to the first. — Jonathan Swift, A Tale of a Tub, Dedication.
• Alternatively, five (say) Proxy Electorates might vote as one entity to elect five legislators, using a proportional representation method. This would better mirror the range of opinions in the electorates. A few jurisdictions at a minimum should use this method, for experimental purposes.
• There would be anywhere from 5 to 15 legislators per Demi-legislature; 15 at higher-levels, and for more important topics.
• Legislators could serve on more than one Demi-legislature, if elected by a separate Proxy Electorate. This would give them enough backup job security to risk casting principled but unpopular votes in at least one of their legislatures.
Where else … is the individual expected to sacrifice all—including his career—for the national good? —J.F. Kennedy [& Ted Sorensen], Profiles in Courage, 1956.
• Campaigners for state and national legislatures must ordinarily have served first at the next lower level.
• There would be a coordinating Omni-legislature and Omni-PE to oversee, veto, or modify measures passed by Demi-legislatures. Its members would come from the overseen Demi-legislatures and Demi-PEs. It could impose budget-limits on Demi-legislatures’ proposed spending. And it could amend bills, or reject them.
• Probably, for major decisions, there should be a cooling-off period between argumentation and voting/balloting; and between the first and final votes on bills, to allow for sober second thoughts.
• Two-house legislatures would need a similar back-and-forth negotiating process, possibly with the senior house having the last word.
• Each Proxy Electorate would funnel the winners of its internal balloting into a higher-level PE specializing in the same topic, or a somewhat similar one. The resulting structure would resemble a hierarchical pyramid.
• PEs would use the services of a Secretariat, which would assist more than one IVE. It would provide moderators, researchers, secretaries, police, and investigators. (PEs could choose to employ independent providers of some of those functions instead or as well.) Elections would be evenly distributed throughout the year, to avoid overloading the Secretariat.
• PEs might, in time, elect not just legislators, but executives (mayors, governors, and presidents) too. And possibly their department / cabinet heads as well, or instead. (Especially if the chief executives are chosen by the mass electorate.)
• PEs should allow individual supporters of candidates to speak on their behalf. I hope this would sometimes result in “the office seeking the (reluctant) man.”
• PEs might allow political parties and interest groups to endorse candidates, supply them with campaign material, and allow the candidates to boast of these endorsements. Experience will determine if this should be continued or not.
• Proxy Electors will be insulated from outside influences. Their identities will be concealed. Any attempt to contact them will be severely punished. Electors who fail to report attempted contacts will be dismissed and punished. The Secretariat will employ persons who will attempt to make contact, in order to motivate electors to report contact-attempts.
• Proxies will be well-paid, possibly very well paid, to blunt the appeal of bribery, and to motivate them to do their duty.
One purpose of the lot component and of randomized drawing is to encourage Proxies to be somewhat independent of the majority—because majority disapproval can’t prevent a dissenter’s name from being drawn, only decrease its likelihood.

Leave a comment