What is sortition?

Sortition – government by jury

In our society elections and democracy are considered inseparable. In fact, this connection is far from clear. The ancient Greeks, for example, thought that elections are part and parcel of an oligarchy1. It was oligarchical Sparta, rather than democratic Athens, that elected its government.

The Athenians had a very different system: political offices were distributed using a lottery. The lottery method – known as Sortition – could be implemented here. If Congresspeople were drawn at random from the U.S. citizenry, Congress would not be an elite body made predominantly of rich, male, white, old lawyers2. Rather, it would look like a statistical sample of the people: it would contain 50% women, 28% hispanics and blacks, rich and poor, young and old, straight and gay, and very few lawyers.

More information on the sortition system can be found on Wikipedia at http://en.wikipedia.org/wiki/Sortition, and other online resources. One such resource is A Citizen Legislature by Ernest Callenbach and Michael Phillips – a book with a specific proposal for using sortition to select the U.S. House of Representatives. The book is available at http://www.context.org/ICLIB/IC11/Calnbach.htm.

This blog, Equality-by-Lot, is devoted to discussion of issues associated with sortition and with the promotion of sortition as a tool of democracy.

[1] “[T]he appointment of magistrates by lot is thought to be democratic, and the election of them oligarchical, democratic again when there is no property qualification, oligarchical when there is.” Aristotle, Politics, book IV, 9.

[2] Of the 535 members of the 112th Congress there were 75 (14%) blacks and hispanics, 91 (17%) women, and 222 (41%) lawyers. The average age in Congress was 62, vs. 37 in the population. Sources: http://www.senate.gov/reference/resources/pdf/R41647.pdf, 2009 population data – Statistical abstract of the U.S., 2011, Table 6.

37 Responses

  1. I do agree way to many attorneys, time to get back to we the people, not the we the select few attorneys that ignore all laws as we are above them.


  2. > time to get back to we the people

    Was there ever a time in which Congress was “we the people”? That would go against the very nature of elections.


  3. I have started my first crowd funding project.

    Washington State Citizen’s Advisory Legislature 2015

    I propose paying one hundred random Washington State eligible voters to create their own model budget for state budget. Anyone interested.




  4. “straight and gay” Why does sexual preference have to be a part of every discussion now? This is just nasty.

    Liked by 1 person

  5. Nasty in what way? Representation of oppressed groups according to their proportion in the population is a desirable outcome of sortition.


  6. In some societies, say in the upper house, there is a constitutional guarantee for certain groups or minority sectors of the population to be represented. That is a good idea, IMO. That significantly undercuts lobbying, so that it is out in the open.


  7. […] of the dice (although if you do really like the idea of democracy/governing via lottery, there is a process called sortition that you may want to familiarise yourself with.) Such approaches totally defeat the purpose of […]


  8. […] So what if we took that idea seriously? Could we replace some of our elections with selection by lot, also known as “sortition?” […]


  9. […] the most effective proportional representation system is in fact not an election system. Sortition is the name given to a system where representatives are chosen by random lots from the eli…. In fact, we already use this system to choose representative […]


  10. […] a number of ancient democratic notions and techniques that do seem highly attractive: the use of sortition, for instance – a random method of polling by lottery that aimed to produce a representative […]


  11. […] What is sortition? […]


  12. […] a number of ancient democratic notions and techniques that do seem highly attractive: the use of sortition, for instance – a random method of polling by lottery that aimed to produce a representative […]


  13. […] revolt is at hand, a need for change does grow. Will it come from on high, will it rise from […]


  14. […] revolt is at hand,a need for change does grow.Will it come from on high,will it rise from […]


  15. […] fans of sortition – the selection of citizen’s juries or mini-publics by lottery – there is certainly plenty to […]


  16. Who are the non-white and non-Western supporters and scholars of sortition?


  17. > Who are the non-white and non-Western supporters and scholars of sortition?

    The person who wrote the earliest modern polemic for sortition is CLR James, who was an Afro-Trinidadian.

    That said, I am aware of no prominent non-Western-White sortition activist. There aren’t that many Western-White sortition advocates either…


  18. Thank you, Yoram Gat. This is great. There must be others throughout history. The trick is finding them all.


  19. […] An enduring problem is that our democracies are not really democracies. They are oligarchies masquerading as democracies. Any system which has elections as the centrepiece of its popular participation is inherently flawed and easily corruptible. The Classical Athenians knew this, which is why they preferred lotteries to elections. […]


  20. Some people consider sortition as one form of direct democracy while others classify sortition as something distinct from direct democracy (that sortition is not a form of direct demoracy). What are the objections to considering sortition as a form of direct democracy?


  21. Yes – I think Chouard and other participation-minded people (some of whom are committed democrats like Chouard, and some of whom are elitist who are really looking for ways to shore up the existing system rather than to fundamentally change it) tend to try to include sortition under the “direct democracy” banner, which they consider an inherently desirable banner to wave.

    At some superficial level this is just about definitions and one can define “direct democracy” as one wants. However, there is a deeper, even fundamental, issue at stake. It is important to realize is that “direct democracy” (as it is usually perceived and discussed) and mass-participation are not democratic. They are mass-political arrangements which have a shallow democratic appeal but in fact are inherently oligarchical.

    This understanding is important both because it is part of the understanding of why elections have to be rejected and because it is a guidance toward what alternatives to elections are in fact democratic. This understanding must be part of the sortitionist agenda. It is therefore important to distinguish between sortition-based government and “direct democracy”, rather than to try to blend them together (either ideologically or institutionally).


  22. Thank you, your description and links on the “direct democracy” question are incredibly helpful! I agree that the terms ultimately are indeterminate and are primarily important in regards to their pragmatic effect. Sortition as a form of “direct democracy” may cause confusion; it may cause one to think that self-selecting town halls, referenda, ballot initiatives, recall, and sortition can all be used to similar effect, that they’re all similarly democratic when they’re not.

    When talking to others, what term do you use to describe sortition? The term “sortition” itself seems perfectly fine for academic and other similar discussions, but it seems to be unknown to people at large, seems to be uninspiring, esoteric, and you can’t tell what sortition may be about just from looking at the term — unlike “direct democracy” which you can at least guess at. It seems even many political scientists don’t know what “sortition” means. The verb “to sort” has the meaning “to distribute by lot” (OED), but this mean is “obsolete” and “rare” (OED). English speakers are not going to make the connection between “sort-” in “sortition” and “sort” in “tirage au sort.”

    Is there a term that can effectively convey the idea of “sortition” in a clear and inspiring way? Other terms like demarchy, lottocracy, stochocracy, klerotocracy seem to have similar downsides for wide use due to their being esoteric, uninspiring, and unclear to people at large, with some of those terms being difficult to pronounce.

    There’s a proliferation of terms in the form of “adjective + democracy”: allotted democracy, participatory democracy, deliberative democracy, deep democracy, real democracy, pure democracy. All of those (except “allotted democracy” which has the problem of being unclear) seem to have similar problems as direct democracy: they’re ambiguous and could cause one to think a range of mechanisms that include a number of inherently oligarchic mechanisms are equally democratic.

    There are also terms like James Fishkin’s “Deliberative Polling®” and terms like citizens’ jury/assembly, citizens jury/assembly, citizen jury/assembly. Fishkin claims a trademark on the first term. “Citizen” may be a problematic term, particularly with rampant nativism and xenophobia, and it is conceptually linked to a problematic limited political sphere; Carol Gould prefers the term “membership” (which may also be problematic) to “citizenship” as she advocates for democracy in the political, economic, and other realms.

    Pragmatically, “citizens’ jury” does seem to make some intuitive sense to people because people are aware of the institution of the jury (although sometimes thinking of it negatively). However, “citizens’ jury” does not seem to be as inspiring as “democracy” or “direct democracy”; “citizens’ jury/assembly” sounds more technical, bureaucratic, tedious, and potentially unappealing. Everyone seems to love democracy (at least the term), but not everyone loves juries/assemblies.

    There is also the term “mini-public.” This does not seem to be catching on much, though.

    It seems the best term would be whatever one catches on the most and is most effective for sortition. And it seems useful to try multiple terms simultaneously. Are there other terms that are being used? Is there a need to invent a new term(s)?

    Another option seems to be to try to reclaim “democracy” or “direct democracy” or other such widespread popular terms for sortition and correctly brand other systems as oligarchic.

    What do you find to be the most effective approach in describing “sortition” to others?

    Liked by 1 person

  23. (Also, of course, some of the terms mentioned above, such as ‘Deliberative Polling®,” have major differences with sortition as discussed here. I’m just trying to brainstorm.)


  24. Hi Jonathan,

    I agree with your points.

    I am in the “reclaim ‘democracy'” camp. In my mind, all the “adjective-democracy” is largely a distraction – often deliberate distraction. It allows the user to avoid dealing with what “democracy” (un-adjectived) is and at the same time borrow the good reputation associated with democracy.

    Democracy is a situation of political equality – equal impact on policy, equal representation of values and interests in policy. This is a definition most people will accept, although some would try to tack on various additional conditions. Some of these conditions (like freedom of speech) can be derived from the definition above, while others (like rights to private property) should be rejected (i.e., whether you like those or not, they are not part of what democracy means).

    My view is that we should insist on the political equality definition serving as a criterion for judging whether a government system (existing or proposed) is democratic. This is the starting point for arguing that elections-based systems are rather transparently non-democratic. This of course goes against lifelong indoctrination for most people in the West. But this indoctrination must be overcome, and these days it is becoming easier to do so, as people are growing increasingly disillusioned.

    As for explaining what sortition is about, I use “representation through statistical sampling”, “government by statistical sample” and “government by jury”.

    The name of an allotted body should depend on its function. I use “allotted parliament” or “allotted chamber” or “allotted congress” when talking about the top level government body (corresponding to the elected parliament and elected congress in the Western system).

    For an issue-specific body the name would be different, of course. I avoid “assembly” because in my mind it implies a mass body, which, for the reasons I described in the posts I linked to, means it is not democratic. “Jury” conveys the idea of statistical sampling but has the drawback of implying a short service term, which should certainly not be part of the message. Maybe “citizens’ council” (like the Athenian allotted council [Boule])?


  25. Thanks, this is very helpful. And great points: “often deliberate distraction” / “some of whom are elitist who are really looking for ways to shore up the existing system rather than to fundamentally change it.”

    Liked by 1 person

  26. Hi Yoram (or others with expertise in this),

    A friend who is a medical researcher made this comment: “From my very superficial understanding of Iceland politics and from purely the perspective of scientific evidence based studies, I think 950 is too small a test group and any results derived from that may be considered underpowered and can’t possibly have an accurate presentation of Iceland as whole.”

    How would you respond to that? I understand how to calculate sample size for things like public opinion polling. It’s very clear and well-established scientifically that you don’t need a very big sample to accurately represent even millions of people.

    But I’m not familiar with other methods of calculating sample size, like these from the National Institute of Health for calculating the sample size for clinical trials: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2876926/

    Would a sortition body be less like a public opinion poll and more like a clinical trial? And if so, from “the perspective of scientific evidence based studies” should a different sample size calculation be used in place of the one used for things like public opinion polling? I’m unfamiliar with the latter, but would imagine that even using the latter methodologies you would still end up with a rather small sample size for even very large populations (and of course in the Iceland example the total population is relatively small at 323,000, and the relevant population — the adult subset of that — is even smaller).

    Thank you for any help or links you may be able to provide. I’m not finding much on this precise question.


  27. Also, this comment mentioned above illustrates the error that is often made of critiquing sortition in comparison to an ideal world. The relevant comparison isn’t sortition to an ideal world but sortition to the actually existing world (which is a very low bar to meet). As we know, for the Iceland example, the Icelandic unicameral Althingi only has 63 elected members. It’s immediately obvious that a randomly selected body of 950 adults is far more representative than a body of 63 elected members. But I still want to come up with a response to that ideal world question above. Thank your for any thoughts.


  28. Hi Jonathan,

    First, I completely agree with your last point – the current system sets the bar for governance quality at a rather low level and this fact should be kept firmly in mind when considering alternatives.

    As for representativeness as a function of sample size: this greatly depends on the intended function. Therefore, absolute claims, such as “950 is too small”, are essentially meaningless. We really have to define what it is that we expect an allotted body to achieve before we make any judgement as to its desired size.

    Of course, calculating probabilities of deviations of the sample from the population is straightforward. The problem is that this is just one aspect of the issue.

    It is crucial to realize that a body can just as easily be too big to be democratic as it can be too small to be representative. For example, a 950 person body is too big to perform many of the needed functions of a parliament, while for various functions a 10-person body can easily be representative.

    Making a full account of the considerations for the sizes of different bodies would be a non-trivial task – one that I would like to undertake but haven’t done yet, and one that I am not aware anybody has. I do have one post that touches upon this issue and which I think provides an important upper bound for the size of a body that would have functions similar to those that an elected parliament has. See: https://equalitybylot.wordpress.com/2012/02/28/dunbars-number.


  29. Thanks Yoram! This is very helpful, as always!

    I’m going to this conference at McGill this Thursday on sortition: http://csdc-cecd.ca/workshop-representation-bicameralism-sortition-application-canadian-senate/


  30. You’re very welcome.

    If there is anything interesting to report from the workshop, or if you have work of your own that is related to sortition, I’d be happy to set you up as a contributor to the blog and we can have a discussion.

    Also, if you get a chance to mention Equality-by-Lot to the workshop attendees, they may find it interesting and we can grow our community.


  31. I’d like to be added as a contributor. I’ll report on the workshop. And I’ll post on my research on sortition. Also, as you probably saw, one of your contributors, Peter Stone, is presenting at this workshop. Thanks!


  32. Great – I just sent you an invitation.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: