The Uses and Abuses of Sortition

Given that sortition is finally beginning to be taken seriously by politicians, academics and the mainstream media, some of us on this forum have expressed concerns about potential abuse. André has drawn our attention to the risk of politicians and public intellectuals using sortition to provide a patina of legitimacy to undemocratic practices — examples include Emanuel Macron’s ‘Great Debate’ — and there has been the usual concerns about the rich ‘n powerful using sortition to paper over the cracks of the electoralist oligarchy.

My own concerns are over the willingness of sortition advocates to assume that small stratified samples, in which participation is entirely voluntary, can represent the beliefs and preferences of all citizens. Leaving aside the size issue (most statisticians insist on a minimum of several hundred or even 1,000 for a representative sample) my principal concern is that the voluntary principle will significantly over-represent activists, “progressives” and those who want to change things, as oppose to the ‘silent majority’. The decision of the 2004 British Columbia Constitutional Assembly to change the voting system was overturned in the subsequent referendum, but this might well have been anticipated as only 4% of the original random sample opted for selection, thereby generating an unrepresentative sample (Warren and Pearse, 2008). My assumption here is that the decision not to participate might well be a sign of a conservative (small ‘c’) disposition.

Extinction Rebellion (XR) is a UK climate-change group which has gained a lot of publicity recently on account of its civil disobedience campaign, which brought much of  central London to a standstill in April. Their goal is zero carbon emissions by 2025, which would mean the banning of air transport and the removal of 38 million cars (both petrol and diesel) from the roads. In addition, 26 million gas boilers would need to be disconnected in six years.

XR claims that the failure of mainstream politicians to sign up to this radical agenda is because they are in hock to capitalist interests, although it’s hard to imagine an election manifesto including such policies gaining much popular support. This being the case XR  are enthusiasts for the replacement of elections by a Citizens Assembly selected by lot. Rupert Read, one of their key spokesmen, was a Green Party councillor in Norwich, but was beaten for a place in the EU elections by UKIP in 2009 and 2014 (he also failed to be elected as an MP in 2009 and 2015). As a result he is now agrees with XR that ‘government must create and be led by the decisions of a Citizens’ Assembly on climate and ecological justice’ (Wilson & Walton, 2019, p.40).

XR are somewhat vague about how such a body should be constituted:

We would take, I think it’s around a thousand people by sortition, by you know selection, like you would for a jury. And they would look at the reality of the situation, and if they don’t know it, they’ll be shocked and frightened. And then they’ll be presented with potential solutions. (‘Welcome to the Rebellion’)

Whilst the sample size suggested is commendable, there is no mention as to whether participation will be quasi-mandatory (as with jury service) or whether there will be any attempt to offer balanced information and advocacy. The latter seems rather unlikely according to another key XR spokesman, Roger Hallam:

We are setting up assemblies where ordinary people can decide whether thy want their children to be delivered to their deaths by the rich and powerful. (Roger Hallam, 20 February 2019)

More specifically:

When we have forced governments around the world to meet our demands your local Citizens Assembly advised by natural and social scientists who TellTheTruth will determine the role of capitalism, i.e. the role of infinite growth and infinite inequality. (Extinction Rebellion, Twitter, 13 May 2019, my emphasis).

Note that the scientists selected as advisers are those who ‘tell the truth’ about global capitalism. Extinction Rebellion is a project of the left-wing campaign groups Compassionate Revolutionaries and Rising Up!, who have acknowledged that they were ‘birthed’ in the Occupy Movement of 2011-2012. On 18 April, XR tweeted the following message (subsequently deleted):

This movement is the best chance we have of bringing down capitalism. (quoted in The Times, 18 April 2019).

Plus ça change:

In place of the Marxist belief that the capitalist economic system must inevitably collapse because of its own internal contradictions, so this movement holds that capitalism must end because its own perpetuation necessitates destroying and depleting the resources it depends upon. . . At its core, Extinction Rebellion is an anti-capitalist movement that envisages no possible accommodation with a free market economy. (Wilson & Walton, 2019, pp.31, 35, 43)

This is unsurprising, given that it was the deliberative democracy movement that spawned the recent focus on sortition and that ‘deliberative democracy, when properly conceived, is the rightful heir of the early Frankfurt School [of cultural Marxism]’ (Scheuerman, 2006, pp. 86). More ominously:

While for the most part it appears that Extinction Rebellion demands a national Citizens Assembly which would specifically focus on issues of environmental “justice”, there is no reason to think that this would be the limit of either these activists’ ambitions, nor the ambitions of those from a wide variety of campaign groups of the left which have demanded the establishment of Citizens Assemblies for numerous other issues. (Wilson & Walton, 2019, p. 43).

This being the case, it’s vital that forums like EbL focus on the need for the impartial selection and ongoing representativity of decision-making bodies selected by lot and for the information and advocacy guiding their deliberations to be well balanced.


Extinction Rebellion, Twitter, 13 May 2019

Roger Hallam (Extinction Rebellion) addresses Amnesty International, 20 April 2019.

Scheuerman, W. E. (2006). Critical Theory Beyond Habermas. In B. Honig, J. S. Dryzek & A. Phillips (Eds.), The Oxford Handbook of Political Theory (pp. 84-105). Oxford: Oxford University Press.

The Times. 18 April 2019, ‘Extinction Rebellion: Protesters set sights on Heathrow’

Warren, M. E., & Pearse, H. (2008). Designing Deliberative Democracy. Cambridge: Cambridge University Press.

‘Welcome to the Rebellion’ — Extinction Rebellion, YouTube, 20 April 2019.

Wilson, T., and Walton, R. (2019), Extremism Rebellion (London: Policy Exchange)

15 Responses

  1. I agree with your final point. I’m more relaxed than you about the nutcases proposing sortition because, ultimately anyone proposing sortition if they’re doing it in a bona fide way is proposing to give away political power.

    Of course they may cling to fantasies about ‘the people’ in the assembly will decide, but the strategy for real believers of sortition is clear which is to insist on acceptable standards of selection and of presentation of evidence – as you point out.

    I’ve also said on this blog previously that my own way around some of the other concerns you raise regarding the potential non-representative nature of a body for lack of size or lack of compulsion is to increase the size of the majority required to have much influence – as we do with juries.

    I’m comfortable with your own concerns about these issues in the long-term, but I can’t imagine getting far in the short term without being prepared to countenance such compromises. The great benefit of accepting non-compulsory attendance is that it allows an activism of sortition even where the powers that be want nothing to do with sortition. It can be funded and organised quite independently of government.

    Liked by 1 person

  2. Food for thought. Has there ever been a CJ that has been given a final say, whose decisions will be accepted in advance? I think not.

    So is it not curious that a law-court jury has the power to decide guilty or not, yet Citizen’s Juries are only ever ‘advisory’? C’mon Sortitionistas, let’s start advocating for CJs with power!


  3. Nick:> The great benefit of accepting non-compulsory attendance is that it allows an activism of sortition even where the powers that be want nothing to do with sortition.

    The problem is the slippery and mendacious use of language activists. In an interview in today’s Sunday Times, Rupert Read claims that XR’s vision is “to establish citizens’ assemblies conscripted [my emphasis] from the public — a bit like super-juries” but then goes on to reference the example of the Irish abortion assembly where participation was entirely voluntary. Citizens’ assemblies instituted by activist organisations are just as likely to corrupt the representation process as those instituted by government.

    Liked by 1 person

  4. Conall:> Has there ever been a CJ that has been given a final say, whose decisions will be accepted in advance? I think not.

    The only example I know of is the Zegou DP in the People’s Republic of China, in which the local communist party dutifully enacted the preferences of the jury (Fishkin, 2009, pp. 106-111). Needless to say the XR movement will insist that the decisions of their citizens’ assembly are enacted, hence the need for sortition theorists and statisticians to specify the precise conditions for a random sample to accurately reflect the informed preferences of the target population.

    Liked by 1 person

  5. A companion op-ed by Dominic Lawson in the Sunday Times is of interest:

    My hero of the week is Brian Salt. This 73-year-old Bristolian had become increasingly enraged by the roadblock set up by Extinction Rebellion on a street he used to drive his camper van to pick up a fellow pensioner, Lynne Mattin, and drop her off at a nearby firm where she had an evening shift. So on Wednesday he tore down the Extinction Rebellion roadblock. . .

    Don’t worry — Extinction Rebellion has the solution [to the problem of how to achieve “total buy-in for a policy of self-inflicted mass immiseration”]: citizens’ assemblies, rather than parliament. These, it says, “would be a game changer for the climate”. Really? If such assemblies were as representative of the people as Extinction Rebellion claims, they would be more likely to include people like Brian Salt and Lynne Mattin than [XR founders] Gail Bradbrook and Roger Hallam.

    Lawson’s claim presupposes quasi-mandatory participation, as the two pensioners are more likely to be among the 96% who refuse the allotment invitation whereas activists like Bradbrook and Hallam would certainly accept.


  6. > the information and advocacy guiding their deliberations to be well balanced

    By which, as always, Sutherland, and the rest of the establishment elitists, mean: “the range of allowable positions must be determined by the elites” through some combination of their electoral power, media power and privileged status.

    Somehow, disproportional power of activists is of great concern and yet disproportional power of the elites is just fine.

    Liked by 1 person

  7. Yoram,

    The difference is that the elites have been chosen by electors and/or newspaper subscribers. Nobody is claiming the electoral process is perfect (we all await Alex’s forthcoming monograph on how to improve representative isegoria), but citizens have not chosen the personnel, ideology or policies of XR. Although standing as a candidate in a strong green constituency Rupert Read has lost half a dozen elections (mostly under PR), hence his enthusiasm for citizens’ assemblies. Our job is to make explicit the criteria that will ensure the ongoing representativity of bodies selected by lot.

    Liked by 1 person

  8. I am puzzled with the reply of Terry in .

    Terry >2. Random selection can be combined with stratification in a legitimate way to improve representativeness when mandatory service is not an option (at least if the number of demographic traits is not unmanageable). It does nop harm to keep randomly drawing until an equal number of each agree to serve. Voluntary service is always less ideal, but until laws are passed making service mandatory we will need to rely ona) removing as many obstacles to service as possible, b) providing substantial inducements, and c) some stratification.<
    As far as I understand this is only valid if the same size is maintained
    The principal reasons for using stratified random sampling rather than simple random sampling include:
    1. Stratification may produce a smaller error of estimation than would be produced by a simple random sample of the SAME SIZE. This result is particularly true if measurements within strata are very homogeneous.
    This means that if I take a simple random sample of 471 people out of the population of Belgium (11,26 million) I have 5 % error rate and 97 confidence level according to . Let's suppose this is acceptable for our purpose.

    Suppose that we want a geographic stratification, North (Dutch speaking) 6.477.804, South (French speaking) 3.602.216 and Brussels (capital) 1.187.890.
    What do we need in order to maintain the margin of error and confidence level in our strata? If we need, as indicated in the course, to maintain the same size in each strata, then we need 471 x 3 people in total (the calculation shows us that error rate and confidence level stays the same for these numbers of population). Or do you mean that you can divide 471 in proportional numbers (about 235, 120, 47) and still have the same error rate and confidence level in each strata?

    And how do we proceed with the selection? Suppose we take a lotto drum and we put a numbered ball in it for each citizen (to reply at Yoram's question) and we draw from the whole list until each strata has his quota, balls who do not fit in a strata are put back in the drum, or do we put only the balls in the drum that belong to a strata? I think both systems work but are they both acceptable (by statisticians?) and are they both delivering the same error rate and confidence level?


  9. Paul,

    Yes – the only good use of stratification is to reduce the sampling error. It cannot reliably address issues like self-selection (i.e., people turning down offers to be on the allotted body).

    As for the technical issue: If you have a good way to do simple random sampling (SRS), then applying the same system for stratified sampling is straightforward.

    However, it is not easy to set up a publicly verifiable SRS mechanism. If someone says to me that they have a drum with millions of balls and the chance of drawing each ball is the same, it would be very difficult to verify that this is indeed the case. Moreover, it is not enough to convince a single person that this is the case – you would have to convince millions of people that this is the case. Not an easy task.


  10. Conallboyle asked:
    >”Has there ever been a CJ that has been given a final say, whose decisions will be accepted in advance?”

    I agree that final authority is a crucial step, but we are getting closer. In nearly every implementation done by the newDemocracy Foundation, they require the relevant government authority agree BEFOREHAND to follow the decision of the assembly. There are typically constitutional limitations that prohibit a sitting government from delegating final authority to another body. I think newDemocracy uses the phrase the minipublic’s decision must be “consequential” rather than “final.” In Belgium, and Madrid the new laws creating randomly selected agenda setting minipublics with authority to call policy minipublics into existence, that call is “final” (requiring no ratification by the government), but the policy decisions of the called minipublic is again advisory. In South Australia, when the minipublic looking at siting a nuclear waste dump went against the prime minister hopes, he tried to do an end run by rejecting the minipublic decision and going the referendum route, but was effectively stopped by the moral authority of the minipublic’s considered decision. Again, not officially decisive, but getting close.

    As a side note, when Keith mentions that the referendum on the election reform proposal initiated by the British Columbia Citizens’ Assembly, and says the voters rejected it… that is misleading. It was supported by a strong majority of 57.7%. However the Liberal government had imposed a 60% winning threshold for the referendum to become law (even though the Liberal government had won LESS than 57.7% of the vote itself when it came to power.)

    Liked by 1 person

  11. Keith makes a valid point about the possibility that people who agree to serve if called, may reflect a bias in favor of ACTIVISTS. I agree that ideally, quasi-mandatory service is the goal. That can only be achieved after lots of experience with solution allows it to be incorporated into law… so if we wait for THAT standard we will never go anywhere with sortition. We need a strategy in the interim to build acceptance of sortition. However, rather than only comparing voluntary service with the ideal, we need to also compare it with the existing system. Elections are probably even MORE biased in favor of ACTIVISTS. Indeed this bias is double layered… both the candidates who offer themselves are biased in this way almost 100%, and the voters who choose to go to the polls and cast a vote, are also steeply biased in this way. This “bias” was ALSO the case in the original Greek democracies, where participation was generally voluntary (though there was a period when they would round up citizens from the agora to reach the quorum threshold in the Assembly.) Every democracy has been fundamentally VOLUNTARY, so this weakness is hardly “undemocratic.” It is likely that a voluntary minipublic will be far LESS biased than an elected government (and can avoid the corrupting dynamics of campaigns, contributions, elite manipulation, etc.).

    However, we should examine means of minimizing this bias in a lottery process. Substantial pay, child care, honor, promoting a sense of civic responsibility, even social shaming at not showing up, etc. are possibilities. Before imposing a super-majority requirement on decisions of the mini-public as Nicholas offers (which has anti-democratic aspects and favors the status quo), we might look first at imposing an acceptance rate requirement for the assembly to be formed (perhaps a 50% acceptance rate?).

    Liked by 2 people

  12. Terry:> We need a strategy in the interim to build acceptance of sortition

    The recent debate has been on citizens’ assemblies making decisions with huge consequences. Unlike the BC case the only reason that the Brexit referendum did not require a supermajority (the norm for constitutional change) was that nobody believed there would be a majority to leave. If a government were to set up a citizen assembly on an issue with massive consequences (Brexit, global warming etc) then it would need to apply de jure rather than de facto egalitarian standards. In theory anyone can stand for election and everyone is free to vote as they see fit, and everyone’s vote in election or a referendum counts exactly the same. We all know that a host of factors introduce biases but that doesn’t affect the formal equality.

    If one were to take an equally formal approach to minipublics then one would have to start by acknowledging the stochastic nature of the representation principle (unlike in elections) and then introduce the necessary measures (large sample size, independence, quasi-mandatory participation, balanced advocacy, silent deliberation etc) that ensure the formal equality of all citizens engendered by the representation. The government would need to commission a study of these factors from statisticians, opinion pollsters etc. and design an assembly according to the agreed principles (50% acceptance rate would be far too low).

    If the citizen assembly proposal from XR were implemented without insisting on these formal principles then sortition would come into disrepute. For example, this is XR spokesman Rupert Read’s take on balanced advocacy (he’s clearly not a fan of the advocatus diaboli principle):

    Last year [Read] was instrumental in the BBC changing its policy of “balancing” scientific statements about climate change after he refused to debate with a climate-change denier. BBC staff subsequently received a memo saying: “To achieve impartiality, you do not need to include outright deniers . . . in the same way you would not have someone denying that Manchester United won 2-0 last Saturday.”


  13. >Keith: Unlike the BC case the only reason that the Brexit referendum did not require a supermajority (the norm for constitutional change) <

    I would add that supermajority for constitutional change by referendum is not the case in Switserland. Furthermore supermajorities do not belong to the code of good practice for referendums :


    III. Specific rules

    7. Quorum

    It is advisable not to provide for:

    a. a turn-out quorum (threshold, minimum percentage), because it assimilates voters who abstain to those who vote no;

    b. an approval quorum (approval by a minimum percentage of registered voters), since it risks involving a difficult political situation if the draft is adopted by a simple majority lower than the necessary threshold1

    7. Quorum

    et page 23:

    50. Based on its experience in the area of referendums, the Venice Commission has decided to recommend that no provision be made for rules on quorums.

    51. A turn-out quorum (minimum percentage) means that it is in the interests of a proposal’s opponents to abstain rather than to vote against it. For example, if 48% of electors are in favour of a proposal, 5% are against it and 47% intend to abstain, the 5% of opponents need only desert the ballot box in order to impose their viewpoint, even though they are very much in the minority. In addition, their absence from the campaign is liable to increase the number of abstentions and thus the likelihood that the quorum will not be reached. Encouraging either abstention or the imposition of a minority viewpoint is not healthy for democracy (point III.7.a). Moreover, there is a great temptation to falsify the turn-out rate in the face of weak opposition.

    52. An approval quorum (acceptance by a minimum percentage of registered voters) may also be inconclusive. It may be so high as to make change excessively difficult. If a text is approved – even by a substantial margin – by a majority of voters without the quorum being reached, the political situation becomes extremely awkward, as the majority will feel that they have been deprived of victory without an adequate reason; the risk of the turn-out rate being falsified is the same as for a turn-out quorum.

    supermajorities are leading to a 'minortity rule'.


  14. > Terry : We need a strategy in the interim to build acceptance of sortition. <
    I think we have to proceed in two ways, the pragmatic way with concessions (even to much) and the optimal way.
    It will never be finished, even today we are discussing about improvement of electoral and referendum systems.

    This means that we must have the courage to discuss and propose improvements of ongoing experiments and applications without rejecting them entirely.


  15. […] sortition-based decision making bodies. Among other issues, the question of stratification got some attention. It turns out that Sortition Foundation, which is engaged in such activities, has a document (PDF) […]


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 )

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: