Open letter: Sortition as a tool of democracy

Dear Mr. Scialabba,

I am writing to you following your article “Plutocratic vistas: America’s crisis of democracy”. I am a committed sortition supporter and advocate and a member of a group of like-minded people. We have a blog – Equality by Lot ( – devoted to discussing and promoting sortition as a tool of democracy.

I liked your article a great deal. Articles discussing sortition in one way or another appear occasionally in the mainstream press (you can find a running record of such articles on Equality by Lot – the most prominent of these is Joe Klein’s 2010 Time article ”How Can a Democracy Solve Tough Problems?”). I think yours was substantially different.

Most of the mainstream articles that deal with sortition present it as some sort of a curiosity or a possible add-on to the existing system that could be used when the powers-that-be find it necessary or convenient. This presentation is probably attributable to some extent to conservatism on the part of the writers and to some extent to the background for the discussion, which is very often the work of Prof. James Fishkin who presents his “Deliberative Polls” as more of an ad-hoc decision-making device of convenience rather than a device for fundamentally democratize public policy making.

Your article, on the other hand, starts at the root: a complete rejection of elections as a democratic device. This leads naturally to a search for an alternative. The alternative that you offer – sortition – is therefore conceived in your presentation as a radical democratic solution to a problem inherent in the current system, which, dogma and slogans notwithstanding, is oligarchical. Appropriately, then, your proposal is radical – not an ad hoc panel with limited powers, but the use of sortition for the selection of one of the most high-powered bodies in the world.

I would also add that I think your awareness of Callenbach and Phillips’s proposal is, by itself, a major point of difference from the standard mainstream treatment. The standard treatment would be quite unlikely to have enough of background work behind it to discover relevant prior work in the form of a rather obscure decades-old book on the topic.

My purpose in writing this letter, beyond nodding my agreement and inviting you to join the ongoing discussion at Equality-by-Lot, is to urge you not to allow this article to become one-time action. Fleeting attention is appropriate for those who see sortition as an adornment for the existing system. The position you presented justifies, I think, maintaining a focus on sortition. If it is really the only known democratic device for selecting decision makers, then how could we justify not doing anything in our power in order to push sortition to the forefront of the arena of public political discussion?

I am committed to doing just that, and I would presume to ask you to use your prominence as an intellectual to do the same.

Warm regards,
Yoram Gat

Geroge Scialabba responded briefly:

Dear Yoram,

Thanks for your letter. I’m not sure whether I’ll write again about sortition in the future, but it happens I reviewed the Callenbach/Phillips book when it first appeared:



22 Responses

  1. >Most of the mainstream articles that deal with sortition present it as some sort of a curiosity or a possible add-on to the existing system . . . This presentation is probably attributable to some extent to conservatism on the part of the writers.

    An alternative explanation is that not everyone would agree with you that sortition is the only known democratic device. Not all advocates of a mixed constitution (in which sortition would necessarily play a leading role) can be dismissed as conservatives.


  2. The point is not that the writers do not endorse sortition as a substitute for elections, but that they never consider the idea.


  3. Exactly so, and all the more reason to nudge people into first considering the value and utility of sortition, rather than insisting a priori that it’s the only democratic show in town. This is true both for pragmatic reasons (people are unlikely to endorse the wholesale adoption of something they haven’t even previously considered) and also because a rudimentary examination of the nature of political representation shows that sortition could only ever be one element in a mixed constitution. The mistake that Callenbach and Philips make is to assume that because an allotted chamber looks [like America] it would automatically act in the same way. Not so, as speech acts are the province of individuals, hence the reason why we will always need election or direct democratic initiatives to fulfil the active role of political office. And that’s just the legislative branch, the case for sortition as the sole principle for appointing government executives is, as Socrates indicated, without merit.


  4. Sortition is used today in every court room, how do you think they pick a Jury. They use that because it is the only fair way. The word Ballot, Comes from how the Greeks picked there Council, Bal-lot Ball Lot. They wrote their names on a ball and threw it into a tub, and the Gods chose the ones, so they thought.


  5. No-one disputes that sortition is the only fair way to select a jury (judicial or political), although the religious explanation of the Greek ballot has long been discredited. However, returning to the jury analogy, the role of the jury is to decide the outcome of the trial; juries do not play an active advocacy role, neither do they initiate the prosecution. In the Callenbach and Philips (C&P) model, the same tiny group of citizens would be judge, jury and executioner. This is not democracy, as the descriptive representation mandate of a sortive group only applies in aggregate rather than to the volitions and speech acts of the individual members. The C&P constitutional proposal would be better described as a klerotocracy, a variant of oligarchy.

    The trial jury analogy is a good illustration of why a fair and effective political constitution would need to be mixed, rather than determined by any single principle. The legislative agenda (by analogy the decision to prosecute) would be determined by either government officials, elections or direct democracy, in which all citizens participate; advocacy would correspond to the expert knowledge aspect of the deliberative poll and the final deliberations and outcome would be in the hands of the allotted jury. Such a system would include a mixture of direct democracy, elections and sortition, along with a government appointed on merit alone, but held to account by a randomly-selected citizen assembly.


  6. A minor historical/etymological note for accuracy’s sake…
    “Ballot” is Italian in derivation (pallotte is the diminutive of palla), referring to voting by putting small balls in a container, and “lot” is Germanic in origin. It is mere coincidence that the letters “lot” are in both. The Greeks did not use balls in any of their voting or lottery processes. The lottery was conducted using metal or wood tabs inscribed with the citizen’s name inserted into slot of a kleroterion machine. Voting was conducted in three ways…raising of hands, or by dropping brass wheel like disks with a central axle (hollow axle or solid indicating yes or no), or by scratching a name on a pottery shard.


  7. It’s interesting to hear that the word lot is of Germanic origin, considering that the system of electoral representation is also supposed to hail “from the woods of Germania”!


  8. Keith,

    Yeah, but in the original Germanic usage it was a gambling term for the precursor to dice… “throwing lots.”



  9. Another semantic question. Keith, neither my online dictionary nor googling finds ‘klerotocracy’. Just what variant of oligarchy is it?


  10. This is another of my neologisms. The point that I’m trying to make is that an allotted body with full powers (policy generation, advocacy and decision making) is another case of rule by the few, the only difference being that the few are selected randomly (rather than by wealth, reputation, birth or force). The modern argument for allotment depends on the resulting body being a descriptively-accurate representation of the whole citizenry (a portait in miniature), but this only applies at the aggregate level, so only refers to collective behaviour, such as voting, rather than the speech acts of individual members. (The parallel with the individual respondents in an opinion poll is an exact one in this respect — nobody attaches any significance to the views of individual respondents, only the final tally). Such a body may well ‘look like America’ (in Bill Clinton’s words) but this doesn’t mean it will automatically act like America.

    Peter Stone makes this point nicely in his introduction to the Imprint Academic edition of Callenbach and Philips:

    “The Representative House apparently doesn’t have to do anything specific in order to do its job. It just has to exist, and somehow whatever it winds up doing will be right, precisely because it descriptively represents the people as a whole.” (p.14)

    Of course as soon as such a body does start to do something (e.g. making legislative proposals), the speech acts involved pertain to individuals and have no descriptive-representation mandate, especially as those individuals may well be a small minority of the whole assembly (ie a microcosm of a microcosm). This was certainly the case with Athenian democracy. Thus an allotted assembly with full powers would be oligarchical rather than, as Yoram (and, apparently, Sciallaba) insists, democratic. I invented the word “klerotocracy” to make the point that there is no automatic connection between sortition as a selection mechanism and democracy.


  11. Since this is the blog of the ‘kleroterians’ I did guess that ‘klerotocracy’ meant ‘rule by the randomly selected’. But I was perplexed that that would be what you meant since the blog promotes rule by the sortitioned.

    Your further (and, I realize, repeated) explanation regarding the influence of ‘speech acts’ is the clearest I have read. Or maybe it’s just that repeating your point finally gets through to me. You are talking about the power of demagoguery coupled with the tyranny of the majority. Correct?

    I too had understood that could be the Achilles heel of a Citizen Legislature. I have felt that in the U.S. the tripartite balance of powers — particularly the Supreme Court’s — would protect against legislative abuse. And I also hold that only until there is a truly representative legislature will be ever be able to say ‘we have what we deserve’. Until then, it’s ‘us’ (the ‘ordinary citizens’) versus ‘them’ (those psychologically inclined and financially bankrolled to campaign).


  12. A typo in my previous post. Should be “…only until there is a truly representative legislature will WE ever be able…”


  13. Demagogues are really just at the extreme end of the spectrum. My point is that (apart from voting, marching in step, singing from the same hymn sheet etc) any action of a member of a collective only represents themselves. So if the collective is to continue to descriptively represent the whole population, each speech act needs to be replicated by everyone else. This is both impossible (not everyone has the same intelligence, vocal charisma and social standing) and would be extremely time-consuming to implement. It would also be profoundly silly. But if you believe in democratic representation that is what you would have to do to legitimise a sortition-only constitutional settlement.

    The only way it makes sense for an allotted assembly to have an active role is, as you rightly say, to resort to the “us and them” social-stratification trope. But this is just a piece of cod-Marxist nonsense that bears no resemblance to the actual distribution of power in complex modern societies. The point that you omitted in your description of “them” is that they are only present to the extent that they have received the votes of “us”. I fully accept the distorting effects of campaign funding and power-seeking psychological aberrations, but we need to address these problems directly (campaign funding is not such a big problem in the UK) rather than adopting the Sciallaba/C&P/Gap case for sortition as the only show in town. Sortition could only ever be part of a mixed constitution, for the reasons that I’ve attempted to demonstrate.


  14. PS Another reason for scepticism regarding the sortition-only scenario is the need to protect the interests of the unborn. Although proponents of the “us and them” model like to suggest that environmental rapine and the accumulation of unsustainable levels of borrowing is the result of a conspiracy between international capital, bankers and politicians (aka the “ruling class”), it’s equally plausible that politicians are simply responding to electoral pressure for ever-increasing consumption and economic growth. It’s not at all obvious that sampling public opinion in an alternative way (sortition) would magically resolve what must be the biggest problem in contemporary political economy. Although Jon Roland’s proposals in this respect have been dismissed as the rantings of a “right-wing conservative” (and, to make it worse, from Texas) at least he has been addressing the issue — all we get from the left is the standard black-box/magic-bullet response: “leave it all to sortition”. This, to put it mildly, is naive and optimistic, even Rousseau understood that unbridled popular sovereignty would require painful (and entirely illiberal) levels of social engineering, combined with a powerful executive.


  15. Keith,

    It doesn’t require “magic” for a group of randomly people, who study an issue, to come to a different, and more long-term vision of “the good” than when the general public is presented with opportunistically framed choices by candidates who want to smear their opponent and claim to have a no-pain solution to some problem. I genuinely believe (as Fishkin’s minimal studies have suggested) that random people who are given time and motivation to understand an issue will do a better job at decision making than uninformed people in a referendum, or politicians feeling the need to pander.

    Terry Bouricius


  16. The Texas utilities DPs certainly indicate that, when presented with two options and adequately informed, the group voted for the higher-cost eco solution; it will be interesting to see what the outcome of the Japanese nuclear power DP is. As to whether a full-power AC would privilege longterm over current interests we can only conjecture, but I don’t think there are any obvious reasons to assume the former (particularly as it would involve a significant drop in current consumption). But we don’t have any evidence one way or the other, so you have to resort to your own intuitions on human nature (a very unreliable tool).


  17. Keith,

    I agree in general. Although I reference Fishkin’s deliberative polls, his China budget deliberation experiment, and the recent Australian energy panel, showing some shift towards the long-term good over current good, this is hardly proof…Since in each case they weren’t making any ACTUAL decision, just registering an opinion, it gives people psychological pleasure to support what seems more noble…but we can’t be certain that would carry over if given the ultimate decision. I tend to think so, but as you say, that’s an unreliable tool.


  18. Keith, one more point about your concern about the influence of ‘speech acts’ by the more adroit. Why does this concern an allotted assembly any more than it already does for an elected one? Yes, of course, the rhetorically gifted will have more voice than the silent.

    And on the issue of whether a deliberative sortitioned group will be smarter than an elected one … I think it can be said that the sortitioned group will, yes, be smarter — and I’d even go so far as to say kinder. Reason being: 1.) it will be more diverse, especially more psychologically diverse…and the studies with which I am familiar say that diversity is a qualitative guarantor; 2.) a legislature that will perforce and finally have someone in it ‘like me’ (i.e., someone who will NOT submit himself to stand for election…which is to say most citizens) can not help but be perceived as more legitimate than those who, as you admit, are chosen through the distortions of elections.


  19. @keithsutherland :

    > The modern argument for allotment depends on the resulting body being a descriptively-accurate representation of the whole citizenry (a portait in miniature),

    Allotment/sortitition does not imply representation.

    The primary utility of sortition is impartiality.

    See “Political Potential of Sortition”, Dowlen pg 24, 90, 191-2, 200.


  20. David: Sorry I missed your last comment. The difference between an elected and allotted assembly is that in the former you get to choose your representative and may well decide to favour the rhetorically gifted. In the latter case if “your” representative is rhetorically-challenged then that’s just tough luck. I agree that cognitive diversity is a strong argument for allotment, but not sure about the kindness issue. Indeed you could argue that many (or even most) of those who put themselves up for election are motivated (at least initially) by altruism.

    Anonymous: Indeed Dowlen (and Stone) argue that impartiality is the primary utility of sortition. Yves Sintomer, however, has argued that there are at least four lottery principles, including rotation, statistical representation and impartiality (I forget what the fourth one is). In ancient Athens rotation was most important, in renaissance Italy impartiality was most important but, for the age of representative government, descriptive representation is the salient principle. If this were not the case then we would be happy drawing lots for just the handful of people needed to fill the vacancies for government officials. This would be impartial, but there would be no way of ensuring that the decisions that these individuals took represented the wishes of the whole population in any meaningful sense.


  21. @keithsutherland :

    Good points.

    I believe the following system would satisfy the sortition, rotation & representation criteria:

    The current parliament is filled randomly every week.

    An order of business could be drafted by the civil service & decided upon on (in a weekly session) by the parliament, on say a quarterly basis.

    Thereafter, each policy brief (drafted by the civil service) would be decided upon on a weekly basis using a preferencial vote; ie rank proposed solutions in preference order.

    In the UK for example, the House of Commons has 650 seats. Given a population of 65 million, 650 seats represents a sample with ~98% confidence level, 5% confidence interval.

    Hope to discuss more in Dublin – possibly over a Guinness.


  22. Look forward to it!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: