The French Citizen Convention on the Climate

Le Monde reported on August 26th (original in French):

The citizen convention on the climate: the allotment of 150 participants has commenced

This citizen convention, which is one of Emmanuel Macron’s responses to the “Gilets Jaunes” crisis, will propose measures to combat global warming.

The allotment of 150 Frenchpeople who will take part in the Citizen Convention on the Climate has begun on Monday, August 26th and will last until the end of September, before meeting for the first time at the beginning of October. This citizen convention, aimed by Emmanuel Macron as one of the rseponses to the grand debate that followed the “Gilets Jaunes” crisis, will propose measures to combat global warming, as France is far from meeting its obligations.

Telephone numbers are going to be automatically generated – 85% mobile numbers and 15% landline numbers – and about 25,000 people will be called, in order to select 150. Some criteria have been set in order to get the best representation: 52% women and 48% men, 6 age groups (starting at age 16), levels of educational attainment, a diversity of professions. Regional population will also be taken into account with 4 oversea representatives, as well as representation of urban centers, their surrounding areas and the rural areas.

Six three-day weekends

“The allotment of 150 citizens is a key step which should create a convention that is ‘France in miniature'”, emphasized the govenrnance committee, who co-presidents are the director-general of the Terra Nova foundation, Theirry Pech, and Laurence Tubiana, the former French negotiator at the COP21.

Six three-day weekend of work are planned, with a final day at the end of January 2020. “The 150 citizens will receive compensation based on the jury-member model [86 Euros per day and reimbursement of loss of income]. The cost of transportation, lodging and meals will also be reimbursed, as well as the cost of child care”, Pech and Tubiana said on Monday.

At the start of the discussions, “the government will publish a provisional schedule for implementing its proposals”. The citizens would then be able to “form a collective and public reaction to the government proposals” via the convention.

62 Responses

  1. Regional population will also be taken into account with 4 oversea representatives, as well as representation of urban centers, their surrounding areas and the rural areas.

    The only way that four persons could represent the entire French diaspora (x million?) would be if they selected them by majority vote (a similar argument would apply to divvying up the 150 “representatives” according to the other stratification criteria). Given that a large part of the reduction from 25,000 to 150 will be via the volunteer principle** (otherwise there would be no need for such a large initial sample), I’m simply appalled that this blatant attempt to pacify the Gilets Jaunes is taken seriously by anyone who understands the mathematics underlying the principle of statistical representation. This can only give sortition a bad name, hence my view that we should wash our hands of it and regroup under the stochation moniker.

    ** pace Arthur C. Clarke’s view that anyone who desires the job should automatically be disqualified from the lottery. The Songs of Distant Earth.

    Like

  2. *** The volunteer principle, at least in current French society, will lead towards a mini-populus not mirroring the populus.
    *** Such political use of sortition may give it a bad name.
    *** But, on the pther side, it means sortition is now in the general intellectual debate.

    Liked by 2 people

  3. NB “Overseas” is not related to French “diaspora”, French people in foreign countries, but the Overseas departments.

    Liked by 1 person

  4. Andre:> But, on the other side, it means sortition is now in the general intellectual debate.

    Yes but what’s the point if it becomes discredited in the process? That’s why I think we should give up on sortition in favour of your own term (stochation).

    >the Overseas departments.

    Sure, but how can they be represented by four persons selected at random?

    Like

  5. *** The “Convention citoyenne pour le climat” may belong, in the minds of France political leadership, to the” polyphonic democracy” (Rosanvallon) strategy directed against the “populist” threat. The oligarchizing elites think they are able to control the mini-populus process through:
    * the volunteer principle: they can hope most members of the small selected body will be people feeling “insiders” to the social and political system, therefore members of the elites or people under their influence
    * control of the agenda
    *control of the deliberation, specially control of informations, views, advices given to the body
    *control of the practical implementation.
    *** The choice of the subject – climate – is good for the deliberation control. If a mini-populus is convened for Brexit (an instance of binary choice), Keith Sutherland may say “equal time for LEAVE and REMAIN” and “equal time for the orators of both sides”. But climate is not a binary choice: any climate policy will be a mix of rational management of ressources, of energy austerity, of use of nuclear energy, of increase of “green” energies, with many degrees of freedom. The “equal time” principle cannot be used, and that gives much space for deliberation control.
    *** The multi-parameter character of the choice is likewise useful for the political leaders : they will be able to choose the proposals they like, specially proposals they are afraid are not very popular, and say “it is the People’s choice”.
    *** The “polyphonic” strategy is smart but dangerous. It puts the mini-populus idea in the center of the debate about sovereignty, an everlasting debate in France since the beheading of Louis XVI ! Ideas about “representative sample”, about the role of chance, about the volunteer principle, etc … will be put in the everyday political discourse. And we must remind the French taste for theory.
    *** Among the supporters of the Convention, there is a strange mix of people, some of them may be supposed of the Rosanvallon mind, some of them may be supposed leaning towards dêmokratia. Which side is strategically wrong? Maybe the French political leadership is right on the short run, and wrong in the long run.

    Like

  6. Andre,

    Exactly. As always you’ve “hit the nail on the head”.

    >Which side is strategically wrong? Maybe the French political leadership is right on the short run, and wrong in the long run.

    Right, and the danger is that sortition is discredited as a result.

    Like

  7. *** Keith Sutherland asks how four persons selected at random can « represent » the French overseas departments. Actually It is only a part of the stratification.
    *** And any stratification would increase the statistical representativity of an allotted sample of the French populus..
    *** The problem is that the stratification will be used to conceal.the socio-economic, socio-cultural and ideological discrepancy between the mini-populus and the populus it is said to mirror. The volunteer principle is enough to secure that the mirroring is a delusion, a deception – at least in the current French society.

    Like

  8. Andre,

    Agree completely. That’s why we need to distance ourselves from these bogus projects.

    Like

  9. Andre>> “climate is not a binary choice”

    I read that they want to put forward one or more specific proposals. So logically, the decision on each option is a binary alright.

    Keith>> “a large part of the reduction from 25,000 to 150 will be via the volunteer principle”

    My apologies but how you can still say that? As shown by the calculations last time, the best practice of two stages of randomness are a mathematical guarantee that the issues behind today’s “loud volunteer” problem will have next to no distorting impact in a proper voluntary citizen jury. I’d like to resolve this issue for once and for all and suggest a special procedure for the community here, using GILT’s epistemic decision process: a “structured” debate, falsifiable by a few predictions, concluded by a “systemic” vote.

    Anybody who would want to participate in such an orderly resolution process for this, please click “like”.

    Like

  10. Volunteers don’t have to be loud, but the evidence from existing projects is that only a tiny minority of those who receive the initial invitation take up the offer and voluntarism (IMO) is a highly significant population parameter wrt to the descriptive representativity of the sample vis a vis the target population.

    Liked by 2 people

  11. Guidance by IMO is problematic. Surely somebody at NewDemocracy or the Sortition Foundation must have measured this factor with their screener surveys? Descriptively, without influence on sortition obviously.

    Like

  12. To anyone who isn’t a sortition die-hard it’s plain bloody obvious. I mentioned my PhD on citizen assemblies to an engineer friend of mine and his immediate, and unprompted, reaction was that such a body couldn’t be representative as participation is voluntary. Are you seriously suggesting that an effective 96% refusal rate would not affect the ability of the sample to represent the target population? You can stratify for age, gender, education, etc but “voting with your feet” is (ex hypothesi) not possible to compensate for. My claim is that most ordinary people really can’t be bothered (“sortition takes too many evenings”, to paraphrase Oscar Wilde) and would much rather watch TV or go down the pub, and democracy means that the opinions and preferences of selfish, lazy and non-political people are just as important as those who do give a damn. As for the opinion of newDemocracy and the Sortition Foundation, are you seriously suggesting that organisations that view samples of 20-40 persons as “representative” have anything useful to tell us?

    Like

  13. Two topics here…

    1. hubertushofkirchner, I have never been able to understand how your GILT design would work in the real world. Can you refer me to an English language explanation online that uses either an actual example, or a hypothetical example, rather than the principles or theory?

    2. There are two level of volunteerism. In Athens citizens had to proactively put themselves forward to be in the pool. I’ll call that ACTIVE volunteerism.” The better approach is to randomly ask from a pool of all citizens to be in the final selection pool, as this brings in people who might be good to have but would never think to offer themselves. I’ll call this “PASSIVE volunteerism.” This second form is preferable, and in a model society, might include almost anybody, as a sense of civic responsibility would reign… (but we aren’t there now).

    But there are bigger philosophical questions. Democracy has never meant rule by absolutely every citizen. Many citizens did not WANT to participate and did not. In Athens the vast majority of citizens were off doing other things every time the Assembly met. But the key principle was that any citizen over 35 who WANTED to participate could be in the lottery pool to be nomothetai (law makers), and any citizen who WANTED could attend the assembly (unless they were late and there was no more room in the Pnyx). Athenian democracy was based on ACTIVE volunteerism, and that has ALWAYS been how democracy actually functioned. Can’t an even more inclusive methodology of PASSIVE volunteerism be a basis for a modern democracy? All citizens are in the grand pool to be invited to be in the smaller pool for a particular mini-public. Every citizen in this model has the right to decline to join the small pool. I personally prefer quasi-mandatory service in final decision-making mini-publics, but that is more for epistemic reasons, that principles of “democracy,” because democracy has always been voluntary. The idea that in an electoral democracy anyone who wishes can vote for one of the candidates on offer is a charade and not democracy, as Rousseau pointed out.

    Like

  14. Terry:> All citizens are in the grand pool to be invited to be in the smaller pool for a particular mini-public. Every citizen in this model has the right to decline to join the small pool.

    But the overwhelming majority won’t even get the miniscule chance to opt whether or not to serve, they will simply be disenfranchised. That’s why it’s essential to ensure that the sample is representative, so that it makes no difference whether or not any particular concrete individual is included or not. As the vast majority (96%) of citizens in modern states refuse the offer, then the sample must represent this particular population parameter as closely as any other (age, gender, etc etc). This has nothing to do with epistemic factors, simply democratic inclusion, warts and all.

    Like

  15. *** I said that « climate is not a binary choice » Hubertus answered : « I read that they want to put forward one or more specific proposals. So logically, the decision on each option is a binary alright. »
    *** But the different choices cannot be independent, you have to choose the best formula, the best mix. If you want not to increase the use of nuclear energy, or, following the « Greens », to end it, you will have to increase other items ; same reasoning if you don’t accept a very hard austerity (hard austerity = less heating, no air conditioning, few travels etc). I read different formulas proposed by different think tanks, it is not a binary choice.
    *** A multiple choice cannot be done by a sequence of binary choices, or, if it is the way, it is very easy to control the deliberation. For instance if the leadership is pro-nuclear, first we begin to vote about energy austerity – not too hard, please ; after, about green energies – not too expensive, please ; and in the end nuclear energy will be highly needed and will get a high coefficient. The opposite if the leadership is anti-nuclear.
    *** I remember reading four or five formulas which looked serious. A possibility would be to give to each of them an advocacy equal time. But how to select the serious formulas ? the one which will be deemed serious by some fraction of the Academy of Sciences ? And, afterwards, who will be the orators for each formula ?
    *** It is difficult to draw a process which will not be strongly suspected to be biased. But in a democracy-through-minipublics, we could imagine a minipopulus able to establish such a process. I doubt that our political system is able.

    Liked by 1 person

  16. *** Terry Bouricius wrote: “In Athens the vast majority of citizens were off doing other things every time the Assembly met.”
    *** Probably a majority, yes, but not the same majority, not a permanent majority. Mariners could not vote when on their ships; soldiers could not vote when in some far location; peasants with different crops could not vote at some times of their specific agricultural cycle; labourers did not go the assembly or a jury when a good job was for the day (they were not usually salarymen). But another day the same people could vote.
    *** The comparison with contemporary advanced societies is not valid. The job-related reasons not to participate will be much less strong – specially with telecommunications: even a US soldier in some warship in the Pacific could be a member of a minipopulus. Conversely a very strong factor of no-participation in a contemporary society is the feeling of being an outsider, not belonging to the system, and unability to imagine a role in sovereignty.
    *** I acknowledge that a smart drawing of sortition process may lessen this phenomenon. But I doubt strongly it can reduce it to a very small level. The result will be that the minipopulus will have a political sensibility different from the populus. That will appear, more or less quickly, and the democratic legitimacy of the minipopulus will crumble. This factor of discrepancy will be less easily considered that the deliberation control by elites, but anyway the fact of the discrepancy will appear. In contemporary societies, at least, the volunteer principle is enough to destroy the legitimacy of the minipopulus.

    Like

  17. Clearly any allotment process whose response rate is low is undemocratic. This is in fact a natural and easy way to screen allotment designs. If your design implies low response – as in fact all the allotted bodies in modern times have been – or relies on masking the low voluntary response rate by using coercion, then your design is a failure from the outset.

    Like

  18. Andre,

    > If a mini-populus is convened for Brexit (an instance of binary choice), Keith Sutherland may say “equal time for LEAVE and REMAIN” and “equal time for the orators of both sides”.

    Brexit is not a binary choice either because the leave-or-remain preference could depend on other choices that are available to government. For example, remain-with-austerity is a very different choice from remain-with-economic-stimulus and one may be superior to “leave” while the other is inferior. Having to first commit to leave or to remain without knowing the rest of the policy bundle coming with that choice is a fraud.

    Like

  19. Yoram:> Brexit is not a binary choice . . . Having to first commit to leave or to remain without knowing the rest of the policy bundle coming with that choice is a fraud.

    This bears a strong comparison with policy aggregation under electoralism. Presumably your assumption is that the preferences of a non-mandatory randomly-selected group will naturally converge on a coherent policy bundle, as the interests of the masses are closely aligned on all axes (economic, social etc), whereas the policies currently on offer cohere with the interests of the ruling class and the difference between parties is only (as you have put it) Tesco vs Sainsburys. Many people, however, voted for Brexit for social and cultural reasons and/or on principle (national sovereignty) and in the full knowledge that this could have malign economic consequences (at least in the short term). And the reason that Labour is ambivalent on Brexit is that EU membership places serious constraints on the ability of governments to adopt stimulus packages (in non-monetary areas). However it would be interesting to see the effects of Boris Johnson’s new policy of leave-with-economic-stimulus on the electorate (if he has the opportunity to offer it to voters).

    Like

  20. *** Yoram Gat says « Having to first commit to leave or to remain without knowing the rest of the policy bundle coming with that choice is a fraud ».
    *** Well, I agree that giving to the People true sovereign power about Brexit without true sovereign power about the other sides of policy may lead to tricking the People. It would not be a problem in an ortho-democracy, dêmokratia, but it is one in an hybrid model where some choices would be through minipopulus and some other through representative parliament.
    *** My idea was that Brexit was a choice of a superior order, because it is a choice about sovereignty itself. Is Britain a sovereign State, or a half-sovereign State ? In the first option, afterwards Britain will choose its policy.
    *** I understand another point of view : all is bundled, a British citizen may reject full sovereignty because, for instance, he is afraid that will give more probability to a policy he doesn’t like in some field ; therefore Brexit is not a binary issue.
    *** But this way of thinking leads to a basic problem. As all is bundled to some degree, we will conclude that a sovereign populus, or minipopulus, cannot decide on any specific issue, that he will have to choose between X global policies, each one including a set of choices about every issue. But examining a global policy is such a complex task that that will destroy one of the main superiorities of the minipopulus model over the presidential plebiscitary model, the ability for better information and deliberation.
    *** Personaly I think that the minipopulus democracy can work only by accepting some degree of unbundling ; some degree only, as I said it is not possible to separate the various parameters of the energy/climate issue ; but if we say « this issue must be bundled with all the other issues » we will logically come to a presidential plebiscitary model. In such a model a kind of judicial minipopulus could be useful to prevent the president to use his powers to stay indefinitely, but the choice of the president and the corresponding global policy would be a one-vote choice.

    Like

  21. Andre,

    I believe that in Yoram’s model for a long-serving full-mandate minipopulus “bundling” would be better described as sovereignty. The problem then is how to ensure that the decisions of the sovereign align with the preferences and beliefs of the vast majority of people who are excluded from the process (i.e. representative democracy rather than kleristocracy). He has in previous posts demonstrated alignment deductively (via a three-stage logical syllogism) but I think we need something a little more empirical than that.

    Like

  22. Andre,

    > *** But this way of thinking leads to a basic problem. As all is bundled to some degree, we will conclude that a sovereign populus, or minipopulus, cannot decide on any specific issue, that he will have to choose between X global policies, each one including a set of choices about every issue.

    The way to handle this is by having no one-time high-stakes decisions. Democratic decision-making must be an ongoing process so that decisions may be revised as the situation develops.

    Like

  23. Brexit is a one-time high-stakes choice, irrespective of both the constitution of the decision-making body and the overall political ecology, as you can’t be half pregnant.

    Like

  24. I’m currently observing a national level 4-day minipublic in Germany, and the organizers emphasized that is not representative, because of the response rate effect. Participants (not experts) in their turn pointed out that they were “representing” arguments and interests that would otherwise be crowded out by powerful lobbyists.

    Another issue being crowded out by the emphasis on representation, is that for the public, they will city participants as “people like us,” that is, non politicians and non-elites. There is a good chance that, if the process of the minipublic is transparent, the wider public will see the recommendations/findings of the Climate Assembly as the “reflected opinion” of “normal citizens.” What the public at large lacks is time and information; the minipublic, done right (with respect to information and facilitation), gives citizens time together and even-handed information to discuss what otherwise would never be.

    Moreover, this debate is an old one. If we conceive democracy as participatory, then we STILL need to make sure that minpublics are the STARTING point not the END POINT of generalized awareness and engagement. At the end of the day the overwhelming majority who are not chosen will still have to BUY and IMPLEMENT the results of the minipublic.

    But I agree the error would be in emphasizing “representativeness” in a descriptive sense when the strength of such a processes lies elsewhere.

    Like

  25. *** Yoram Gat gives as solution « The way to handle this is by having no one-time high-stakes decisions. Democratic decision-making must be an ongoing process so that decisions may be revised as the situation develops. »
    *** Good solution (good defence gainst the plebiscitary model), but only a partial one. First about issues like energy/climate, an « ongoing process » could easily trick the People, preventing him to see the multi-parameter problem. Second, as says Keith Sutherland, some high-stakes choices cannot be avoided. It is true for Brexit, but for instance about nuclear energy : you cannot build very expensive nuclear plants, and close them the following year.
    *** Yoram’s answer is interesting to consider in relationship with the « Convention citoyenne pour le climat », which is included in a polyarchic system where there will be no minipopulus ongoing process.
    His answer has some strength in a system where minipulus is the standard choice device but this strength will be minimal in an hybrid system.

    Liked by 1 person

  26. Ahmed,

    The problem with the focus on the representation of discourses by Dryzek and other “discursive” democrats is that ‘The devil has the best tunes’ (or at least a lot more of them) – Dryzek pointing out the plethora of different approaches of environmental activists, including wildlife management, conservation, preservation, reform environmentalism, deep ecology, environmental justice and ecofeminism. It’s all very well arguing that ‘for policy-making rationality all relevant discourses should get represented, regardless of how many people subscribe to each’ (Dryzek & Niemeyer, 2008, p. 482). However if Q methodology is to determine which discourses to include then each would have the same isegoric rights as the hegemonic perspective shared by most citizens (and not just powerful lobbyists for business corporations) that the earth is a resource intended primarily for the benefit of its human occupants.

    >There is a good chance that, if the process of the minipublic is transparent, the wider public will see the recommendations/findings of the Climate Assembly as the “reflected opinion” of “normal citizens.”

    Yes, but if participation is voluntary then this is nothing more than a confidence trick — the participants may look like “normal citizens” but will be heavily weighted towards activists (the only difference from elected politicians being that nobody chose them). This is why Extinction Rebellion is so keen on citizens’ assemblies.

    Ref
    ===
    Dryzek, J. S., & Niemeyer, S. J. (2008). Discursive representation. American Political Science Review, 102(4), 481-493.

    Like

  27. Andre:> [Yoram’s] answer has some strength in a system where minipopulus is the standard choice device but this strength will be minimal in an hybrid system.

    Yes indeed — the former is his proposal. And a unitary, voluntaristic, long-serving full-mandate one at that, with all the resultant problems for ongoing descriptive representativity.

    Like

  28. Andre,

    > for instance about nuclear energy : you cannot build very expensive nuclear plants, and close them the following year.

    Yes – some decisions are expensive to reverse or to modify and those making those decisions have to be aware of this fact. I would expect, for example, that if a nuclear power program would be adopted it would involve a gradual building of the generation capacity, one plant after another – with evidence of safety, pollution, etc. being reevaluated as more experience with the plants is being gathered.

    A plebiscite, on the other hand, would tend to be phrased as “in favor/against using nuclear power”, where a “yes” vote would commit the polity to a long-term hard-to-reverse trajectory toward nuclearization.

    > His answer has some strength in a system where minipulus is the standard choice device but this strength will be minimal in an hybrid system.

    That depends on the nature of the hybridization. It is indeed quite clear that introducing ad-hoc bodies as is the practice now provides very little democratization and is little more than manipulation by elites. However, moving certain decision-making powers to powerful permanent allotted bodies (e.g., anti-corruption bodies), does have a lot of promise for democratization, it seems to me, and can serve as a credible path toward a fully sortition-based system.

    Liked by 1 person

  29. *** I agree with Yoram Gat comment, it will depend on the specific case. But anyway the minipopulus must be really a minipopulus, mirroring the populus.
    *** In contemporary North-Atlantic societies, an allotted body of volunteers will be overpacked by two kinds of people; first people of higher classes, let’s say, the fifth superior fraction of the society, who feel “insider” and “legitimate to rule”; second, militant-minded people; with overlapping of the two categories. These people will feel “superior citizens” – we are more intelligent, more educated and /or more conscientized. Such a body will not be really a minipopulus, but a kind of aristocratic body.
    *** At the same time, even democrats will acknowledge that such a body has superiorities over the elected “representatives”, because they are independent of moneyed lobbies, and over “the public opinion” or over referendum voters, because of better deliberation.
    *** The blend of the two superiorities, the “aristocratic” one and the procedural one, is somewhat “vicious”, because it will allow to entertain the idea of aristocratic superiority without explicitly acknowledging it. It will have heavy consequences.
    *** Among the ordinary citizens, some will defer to the superior body judgment, whereas others, sensitive to the discrepancy with them, will stubbornly reject its leadership (aristocracy is difficult in modern societies).
    *** The aristocratic side of the pseudo-minipopulus and the conscience of the divide with at least a part of the common people will distort its deliberation and choices. In some ancient societies an aristocratic body could deliberate in a rational way, because it did not feel threatened, because ordinary people were only subjects. In modern societies it is no more possible. Any “aristocratic” body will develop some degree of mix of fear and contempt, will avoid choices giving more power to ordinary citizens, will avoid too much free debates which could give “bad ideas” to the ordinary citizens.
    *** We will have the present situation of conflict between elitism and populism, with the allotted bodies seen on the side of elitism. Such a phenomenon could block the road towards a democracy-through-minipopulus – a road which could seem opened by the banalization of political lot.

    Like

  30. Andre:> an allotted body of volunteers will . . . feel “superior citizens” – we are more intelligent, more educated and /or more conscientized. Such a body will not be really a minipopulus, but a kind of aristocratic body.

    That’s right, and it explains why minipopuli are currently in favour amongst those who are seeking to cancel Brexit and introduce unpopular policies on climate change.

    >We will have the present situation of conflict between elitism and populism, with the allotted bodies seen on the side of elitism. Such a phenomenon could block the road towards a democracy-through-minipopulus – a road which could seem opened by the banalization of political lot.

    Yes, that’s why we should denounce the initiatives of the Sortition Foundation, newDemocracy and the likes as counterproductive and likely to lead to elected populist tyrants, or even civil war. (Unfortunately this would also apply to the somewhat larger but voluntary, long-serving, full-mandate and unitary bodies advocated by Yoram.) Even if that’s a tad hyperbolic, we certainly don’t want to see sortition “banalized”.

    Like

  31. PS Andre’s remarks should be read in the context of Roslyn Fuller’s decision to lump sortitionists in with the anti-Democratic backlash (Jason Brennan et al) in her new book. Bear in mind that the the author is an advocate for Athenian rather than electoral democracy.

    Like

  32. Andre,

    > the minipopulus must be really a minipopulus, mirroring the populus.

    I completely agree. However, mandatory participation is counter-productive in getting to this goal. It will merely mask the fact that some of the allotted are marginalized rather than get everybody to participate equally.

    The design of the terms of service must be such that everybody feels well-empowered and well-rewarded so that they are motivated to participate actively.

    Forcing unmotivated people to show up just so that we can claim to have a representative sample is worse than useless. It is at least as destructive as having high rejection rate.

    Like

  33. Yoram:> get everybody to participate equally.

    Given your assumption that most randomly-selected people will agree to participate, what makes you assume that an equalisation of their speech acts (essential for representative fidelity) is possible? Some people are more interested in political issues than others, more eloquent, more persuasive and of higher perceived status. Many (most?) people would have nothing much to say and lack the confidence to say it. It’s true, of course, that the only activity that quasi-mandatory jurors can be expected to do is to listen to the arguments and vote, but there are very good reasons to believe that their votes will mirror the beliefs and preferences of those like themselves in the population that they represent “descriptively”.

    Like

  34. PS, I think we should focus on Andre’s claim that a voluntary minipopulus would be (and would be perceived as) an “aristocratic” body, especially as this directly contradicts the Aristotle/Montesquieu trope on the equation of sortition and democracy. He is right to say that “it will have heavy consequences”.

    Like

  35. As an indication of how confused debate is within the sortition movement, have a look at the definition on Participedia:

    Sortition is a sampling technique used to select a group of participants equally divided by gender and proportionally representative of age demographics, geographic location, and education level.

    No concern over the sample size or voluntary participation, and all the emphasis is on stratification by these 4 population parameters. Sortition is defined as “stratified random sampling” and the principal external link is to the Sortition Foundation. It should be noted that Dowlen’s book (The Political Potential of Sortition) is strongly opposed to stratification and Peter Stone (The Luck of the Draw) takes a similar line. But the lunatics appear to be taking over the asylum.

    https://participedia.net/method/5507

    Like

  36. Whether a high rate of declining lottery winners invalidates the democratic nature of the mini-public depends on the particulars of the society. I suspect Andre is right about a difference between those who agree to serve and those who decline in many modern societies. It matters tremendously WHY people decline. If people decline because they want to focus (for example) on their musical performing career or religious meditation on one hand, that may be very different than if it is because they can’t afford to take time off from their job, feel ignorant about policy matters, or socially awkward about it on the other. We don’t KNOW if those who respond positively to the lottery call are significantly different from those who decline. What little research related to this question (on people who vote and people who don’t register to vote at all) suggests the policy attitudes of voters and nonvoters are more similar than different. However, that is weak evidence since it examined their off the top of their head opinions, rather than their informed considered opinions (which would be developed in a mini-public). Some high quality polling, goes back to those randomly drawn respondents who declined to participate and offers them a substantial amount of money to finally participate … virtually not taking “no” for an answer. Real research could be conducted to find out if a high decline rate (but with stratification) ACTUALLY generates a significantly different mini-public than a high acceptance rate mini-public (perhaps through high pay). This research could be done using a mythical mini-public… we only need to identify volunteers and decliners.

    Like

  37. Terry,

    I don’t think it matters hugely whether it’s the stick or the carrot (or a combination of the two), so long as the vast majority of those allotted take up the invitation/summons. Given that the time and interest commitment for voting in an election and participating in a minipublic is so different, reflected in an order of magnitude difference in historic participation rates, it strikes me as prudent to assume this is a highly significant population parameter (for all the reasons that Andre has provided).

    Like

  38. Terry,

    > (but with stratification)

    Well – if you believe that takers and rejecters are similar, why would you need stratification? Or conversely, if you believe stratification is necessary, do you really claim that the rates of rejection are different between the strata but within each stratum takers and rejecters are similar? That seems like a horribly arbitrary and unlikely claim.

    No. To the extent that rejection is accepted – and this extent must be minimal – there must be no attempt to hide it behind some manipulation of the sampling procedure.

    Like

  39. Yoram,
    Stratification is used in case there is a big difference between young single mothers with little education or income on one hand and wealthy old men on the other. It may be that BETWEEN different stratas there is significant difference in uptake rates. We want to assure there are a proportionate number of various strata even if there turns out to be NO difference between the young single mother who accepts the lottery call and the young single mother who declines.

    Like

  40. Terry,

    Unless you believe in the “politics of presence” in the sense of the deliberate over-representation of minority beliefs and preferences, then the problems that you outline can be overcome by enlarging the size of the sample (along with quasi-mandatory participation). Anything else will be open to accusations of gerrymandering.

    Like

  41. Terry,

    As I wrote, the notion that rejection rates are different between strata but are the same for different groups within each stratum is incredible.

    For example, the notion that acceptance rate among young single mothers (is this really going to be a stratum?!) would not depend on, say, income or education, is, again, incredible.

    Like

  42. I tried to flesh out some strata with hypothetical detail… but my point is that (taking one example) it is possible that women who accept service and women who decline service might have very similar interests and policy preferences… BUT that women as a whole might be far more likely to decline than men. In that one metric, using stratification to raise the number of women to its proportionate share can only improve the representativenss of the body.

    Yoram, So is your notion that if 80% of lottery winners accept service, but that pool includes only a disproportionately small number of racial minority citizens (who are 10% of the population) because they were more likely to decline, that it is still a bad idea to use stratification sampling to restore this strata of society to its proportionate share of seats, because that only creates the illusion of representativeness?

    If that is your view, then you would favor that sortition ONLY be used in a fully egalitarian society that has already banished most discrimination and social status distinctions.

    Stratified sampling is intended to be a somewhat crude fix for an unequal society, to allow sortition to be useful now, rather than in a future utopia.

    Like

  43. Terry:> I tried to flesh out some strata with hypothetical detail

    Why worry about hypotheticals when you can have the real thing: large size and quasi-mandatory participation? Job done.

    Like

  44. Keith,
    As you know I favor large quasi-mandatory mini-publics for final decisions, but these are expensive and currently impossible to have (no legal authority for compulsion). I am willing to consider currently legally achievable and less expensive alternatives if they would be nearly as good. We need to FIND OUT (not assume) whether decliners and acceptors are significantly different in interests and preferences. IF they are similar, then a well-stratified smaller sample that makes a 90% majority decision would suffice.

    Like

  45. Terry,

    > I tried to flesh out some strata with hypothetical detail…

    But that is the point: if a high percentage of rejection is accepted, it is really incredible to believe that any kind of realistic stratification can capture the ways in which acceptance rates would be correlated with values and interests. Thus, accepting non-negligible rejection rates implies accepting misrepresentation. Maybe not as extreme as electoral misrepresentation, but misrepresentation nonetheless.

    > it is possible that women who accept service and women who decline service might have very similar interests and policy preferences…

    It is a theoretical possibility, not unlike the possibility that an elected official would have interests and values that are aligned with those of the population as a whole. It is not a possibility that you can realistically pin your hopes of a democratic government to.

    > In that one metric, using stratification to raise the number of women to its proportionate share can only improve the representativenss of the body.

    That’s very untrue. Such re-weighting could easily create much worse distortion than would be the case without re-weighting.

    It also creates an opportunity for manipulation because it would incentivize people to assert that they belong to strata whose rejection rate is high. For example, if re-weighting by race is used and it turns out that Blacks have relatively low acceptance rates, then for someone who is interested to become a member of the allotted body, asserting that one is Black would increase their chance of being selected.

    > Yoram, So is your notion […], because that only creates the illusion of representativeness?

    Yes.

    > If that is your view, then you would favor that sortition ONLY be used in a fully egalitarian society that has already banished most discrimination and social status distinctions.

    No. I would favor using sortition only in cases where low rejection rates can be attained. This is not utopian – it can be achieved in today’s society with a proper design of the sortition setup: a high-profile, high-powered body whose members are properly compensated for their efforts should enjoy low rejection rates.

    What low means exactly could be up for debate. For example, 80% acceptance rate is far from satisfactory, but is already much, much better than the 4% that is considered acceptable today.

    > Stratified sampling is intended to be a somewhat crude fix for an unequal society, to allow sortition to be useful now, rather than in a future utopia.

    First, it is important to drop the mis-usage of the term “stratified sampling” – stratified sampling is a specific procedure that has almost nothing to do with the re-weighting you are talking about. Using this term is either an honest mistake or an attempt to lend scientific credibility to what is in fact an act of manipulation by the elite that constitutes the allotted body.

    Second, re-weighting is not a crude fix for an unequal society, it is a way to mask a poor (possibly deliberately poor) design of the sortition setup. It brings us no nearer to a democracy and may very well drive us in the opposite direction.

    [ Edited to re-insert a few words that were dropped by the WP comment system. -YG ]

    Like

  46. Terry:> a well-stratified smaller sample that makes a 90% majority decision would suffice.

    Yoram:> a theoretical possibility, not unlike the possibility that [pigs might fly, or] an elected official would have interests and values that are aligned with those of the population as a whole.

    You guys appear to share the political anthropology of Marx, or even Machiavelli, which is of no relevance in the 21st century, where citizens of mature democracies are evenly divided on a range of overlapping axes. Do you seriously believe that these cleavages are manufactured by a sinister elite and will be overthrown as soon as the popolo, led by the kleristocratic vanguard, rise up and overthrow the grandi? It really is time to wake up, smell the coffee, and stop living in the nineteenth century.

    Yoram:> accepting non-negligible rejection rates implies accepting misrepresentation

    Absolutely, but it makes little difference as to whether the solution is the stick or the carrot. As an austere republican I prefer the former, whereas you’re content to (in the words of socialist ikon Aneurin Bevan) “stuff their mouths with gold”.

    Like

  47. Yoram,
    Can you explain how you think using strata allocation (assuring the sample is a close match along various demographic variables) is the same as “re-weighting?” In survey terminology re-weighting involves giving greater weight to the responses of individual respondents who are in a strata that was underrepresented in the survey sample. (If only 25% of the respondents were female, they count each response as two as a crude way to adjust for the fact that their sample was a poor match.) Nobody is proposing that for a mini-public. Is there another meaning for re-weighting?

    Like

  48. Hi Terry,

    What you call “strata allocation” does not give any specific individual added weight but it gives certain groups extra weight. The group of “accepters” in each stratum gets a weight which is proportional to the reciprocal of its proportion in the stratum.

    For example the accepters among a stratum with 50% acceptance get twice the decision making power of the accepters in a stratum with 100% acceptance rate.

    Like

  49. Yoram,
    I don’t understand your thinking. We want to accurately reflect the entire population, not accurately reflect only that segment of society that is likely to accept.

    Like

  50. Terry,

    > We want to accurately reflect the entire population, not accurately reflect only that segment of society that is likely to accept.

    I completely agree. However, the only way to accurately reflect the entire population is to get (almost) everybody to voluntarily accept. Everything else (including most definitely mandatory participation) is simply manipulation that masks the acceptance bias rather than fixes it (deliberately or not).

    Like

  51. Yoram/Terry/Andre

    I can see that there is a moral distinction between compulsion and bribery (philosophers might well argue the merits of either option), but why should that make a practical difference to the representativity of the sample? As far as Terry, Andre and I are concerned we don’t really mind whether it’s the stick or the carrot, so long as most people take up the invitation/summons, so we prefer the adjective “quasi-mandatory” (although if the stipend were too generous, this might well cause resentment amongst those citizens excluded by lot). We view acceptance as a statistically-significant population parameter — i.e. a personal characteristic that makes accepters different from refuseniks. Whether this personal characteristic is innate (very unlikely!) or generated by social, educational, cultural and economic conditions is irrelevant from the perspective of the representative fidelity of the sample to the whole citizen body. The sample must accurately represent existing social inequalities, otherwise it will rightly be viewed as aristocratic.

    Yoram appears to view the participation level in purely structural terms — if the body is constituted and (self-)organised in an egalitarian manner and has a powerful mandate, then most persons will accept and each will have an equal voice (this ignores a wealth of evidence from social psychology that would suggest power imbalances are inevitable). But if the problems are purely structural and allotted “delegates” are interchangeable, then why bother with random selection at all?

    Like

  52. Keith wrote that he Andre and I ” view acceptance as a statistically-significant population parameter.” Not so. I accept that it MAY be significant, but may also turn out to be trivial… the research has not been done. It might be significant if 99.9% declined, but may be de minimis if only 80% declined, or 50% declined.

    The problem with Yoram’s approach, is that I think it is unlikely that more than half of those called would ever willingly volunteer. Again, we don’t know, but evidence so far suggests this. Yoram hopes that if given vast amounts of power and good pay most people would WANT to serve. But I suspect the controlling factor is the enculturation of civic responsibility rather than power or money. As the power involved goes up, some people become more interested, but other people get more worried and scared of the responsibility. We don’t know which trend would be more substantial.

    There is an argument that can be made (I have hinted at it elsewhere) that democracy is always fundamentally voluntary (ho boulomenos –those who wish), and that so long as all barriers to participation are torn down, or adequately compensated for, that the random sample who accept service, regardless of the decline rate, is representative of the population that cares enough to participate … and thus is democratic. I don’t subscribe to that view, but have kept an open mind about it.

    Like

  53. > I think it is unlikely that more than half of those called would ever willingly volunteer. Again, we don’t know, but evidence so far suggests this.

    No – there is absolutely no evidence of this kind so far. The setups attempted so far are so far removed from any useful setup that they provide no useful evidence.

    I suspect that everything else aside, a salary at the 90th percentile of incomes and a guarantee of returning to the former workplace at the end of the service would by themselves raise acceptance rates to well above 50%. Add to that help with relocation, child care arrangements, real decision-making power, media attention, and an ethos that sees service as both a civic right and a civic duty and I suspect we would be talking about how to reduce rejection rates from 7% to 3%.

    Of course we will not know until we try. But what is clear is that:

    1. Declaring ahead of time that rejection rates will be high is a self-fulfilling prophecy that would be destructive for democracy,

    2. At any rejection rate – whether high or low – attempting to hide the real rejection rate (and the misrepresentation it creates) via manipulation of the appointment procedure or via coercion would be much worse than admitting that the rejection rate is what it is and taking all available steps to reduce it toward zero.

    Like

  54. Terry:> acceptance MAY be significant, but may also turn out to be trivial

    What we are interested in is the extent to which a population parameter is a token for a set of overlapping beliefs/preferences/interests. We don’t know that this applies to gender, age, ethnicity or education but most sortitionists insist that these population parameters should be represented accurately (50/50 men/women etc), but the research has not been done. The difference between accepters and refuseniks may or may not be even greater, so the precautionary principle would suggest quasi-mandatory participation as the default position.

    >democracy is always fundamentally voluntary (ho boulomenos –those who wish)

    Hansen’s view (1991, p. 72) is that ho boulomenos refers to “the originator of laws, decrees and public prosecutions” — i.e. isegoria not isonomia. The full quotation is “he of the Athenians who wishes from amongst those who may”, but in the kleristocratic utopia envisaged by some on this blog “those who may” are restricted to the tiny number of citizens who receive an invitation. Given this highly-restricted franchise — much smaller than England before the Great Reform Act — it is democratically essential to ensure that the ongoing representativity of the sample is ensured, hence the need for quasi-mandatory participation. Manin (1997, p. 87) insists that the historical role of the chosen representative was always a burden rather than a privilege (the king insisting on representatives on account of the QOT principle), so there is no reason why this should not apply to representatives selected by lot (the only difference being the balloting mechanism).

    Yoram:> I suspect we would be talking about how to reduce rejection rates from 7% to 3%.

    You may be right (although I doubt it) that participation rates are entirely down to institutional design and have nothing to do with structural inequalities that have become internalised within citizens. But we need to start from where we are, and the precautionary principle would suggest that the gap between 96% and 3% refuseniks is such that serious steps would need to be taken to ensure statistical representativity. You also suggest that the allotted body would self-organise in such a manner as to ensure equality between the perlocutionary effect of the speech acts of all members, but a couple of decades of experiments by deliberative democrats would indicate otherwise (I assume that you would rule out professional facilitators on the Quis Custodiet principle).

    Like

  55. Andre:> The “Convention citoyenne pour le climat” may belong, in the minds of France political leadership, to the” polyphonic democracy” (Rosanvallon) strategy directed against the “populist” threat. The oligarchizing elites think they are able to control the mini-populus process.

    The flip side of this is radical and neo-Marxist attempts to purloin mini-publics for their own agenda:

    The manifold approaches to radical democracy have always been intimately linked with a critique of existing liberal democratic institutions and demands for social transformations. Therefore, these conceptual debates lead to some practical questions: What are the means for social transformation proposed by agonistic and autonomist Marxist approaches to radical democracy? Are they compatible with the means for social transformation proposed by deliberative democracy? Can, for example, deliberative mini-publics be part of radical democratic transformations? This panel is also interested in the responses various radical democratic approaches can provide to cope with present and future challenges for democracy such as digitalization, populism, climate change, and mass migration.

    https://deliberativehub.wordpress.com/2019/10/07/cfp-for-a-psa-panel-quo-vadis-radical-democracy/

    Both of these strategies (Ronsanvallon and radical democracy) are seeking to subvert the representative potential of mini-publics, hence their eager acceptance of voluntarism.

    Like

  56. Keith
    *** Yes, the “radical democracy” interest towards minipublics may cover some far-leftish devious intent, with the usual means: non-representativity of the panel, control of information/deliberation, selection among the non-binding results.
    *** But it is possible that in some leftish circles, there is a sincere attraction towards the minipopulus model. I saw on Internet a Mediapart blog where Clément Sénéchal published again a former article against the minipopulus model. It must indicate some sincere interest for it in leftish circles (the same, when Lanthenas wrote against sortition, it indicated attraction among some Jacobin circles).
    *** Such an interest may be understood. The “council democracy” model was never very convincing. The “totalitarian” model of rule by an activist minority has now a very bad image. In many polyarchies, and in France anyway, far-left elements will have few hopes other than being gladflies, or, worse, auxiliaries helping the Establishment in some directions (ex against populism, including left-populism). Not very rewarding prospects. Thus the new model of democracy-through-minipopulus may have some attraction power.
    *** Now it will not be always easy to distinguish between those whose aim is democracy-through-minipopulus but with steps, those who favour an hybrid model, and those with devious uses of minipublics. It will be necessary to look closely.

    Liked by 1 person

  57. Andre:> It will be necessary to look closely.

    Yes, the devil is always in the detail (and he also has all the best tunes). I guess I’m just an old-fashioned liberal, in that I see the only role of minipublics being a way to establish the considered judgment of the demos, rather than some devious agenda of some elites (either political, financial or cultural).

    Liked by 1 person

  58. A late complement (I was sick)
    *** When mentioning a possible sincere interest, among some far-leftish, for the minipopulus idea, I considered mostly the negative factor: if we exclude the totalitarian and the polyarchic models, and actually all the former models, the democracy-through-minipopulus model, without historical past (the Athenian juries are prehistory), therefore without bad memories, is the one to be tried.
    *** “Without past”, that smacks of utopia, some will say. But this flavor of utopia may not be so negative for some far-left circles.
    *** We may add a positive strong factor. Many people in the radical left had to acknowledge the sad reality: the working classes they have been fighting for since almost centuries are, in the developed world, further than never to follow the revolutionary gospel. Some revolutionaries may give up, and transfer their revolutionary zeal to racial or sexual minorities. Others may look for a path towards “conscientization”, other than the activist preaching. Let’s imagine a democracy with wide use of minipopulus, for national legislation, for auditing and control, for local rule, for judicial decisions from divorces to constitutionality, where the low class citizens are induced, or conscripted, to spend at least 5 % of their time to deliberate and decide on serious issues. At the beginning they will feel humble, but on the long run they will feel able to be real rulers, to escape the traditional political fate of the lower classes, deference, apolitism, self-exclusion. They will be politically mature, and able to listen the radical left discourse.
    *** This is not a stupid prospect.

    Like

  59. […] As published earlier, the final sample is made by selecting from those who accept the invitation while matching the makeup of the French population in terms of gender, age groups, education attainment and place of residence. […]

    Like

  60. Here’s how to get the rejection rate way down, which is crucial to the acceptance of sortition: Let minipublics “gather” over the Internet, in a site accessible only to members and officials. Persons lacking a personal computer would be given one—or they would be given the use of one in a public building. Assemblies would occur in evenings and weekends. This would be much less inconvenient to initial contactees.

    (I assume a personal computer’s full screen will be more useful for such Citizen Assembly deliberations than the restricted screen of a smartphone. (But in poor countries smartphones may have to be accepted.))

    If there is felt to be a need to insulate proceedings from public view, or to protect members from being influenced by outsiders, a policeman or other official could visit members’ homes and ensure they were alone in a room while participating.

    And: avoiding physical get-togethers ought to defuse the operation of “crowd psychology” to some extent. But it would also be wise, I suspect, for members to be assigned to subgroups-only chatrooms, to create fire-breaks against the entire group being swept away by some bit of heated rhetoric or ill-considered brainwave.

    Like

  61. There are benefits and harms from both face-to-face, and “isolated” deliberation….that are distinctively related to the specific task of the allotted group.

    Some advocates of sortition are “deliberative democrats” who hope that the sharing of knowledge and personal narratives in face-to-face small groups can build empathy and aid the finding of new “win-win” policy options. But these sorts of groups are not expected to be accurately representative (only diverse, relatively impartial and roughly representative), and should never have final authority to adopt policies they craft.

    On the other hand, very large mini-publics, which are convened to hear the pro and con arguments for a proposed policy and make a final decision, would actually benefit from relative isolation to allow independent judgements to elicit the “wisdom of crowds” (reminiscent of the Condorcet jury theorem). A thousand members could all hear the same testimony, ask clarifying questions, and vote without ever meeting the other members of the mini-public. There are countless details to consider in how to run such an enterprise. If the concern is that isolated members might take their pay and not even listen to the presentations, to assure members are “doing the work” of attending to the presentations, there might be a short simple quiz about what was presented to assure only those who attended could vote (and get their pay), or whatever.

    Like

  62. Terry,

    That’s all true. The trouble is that deliberative democrats have recently started claiming (under the cover of stratification) that their small groups are “representative” in the latter sense. I do think we need to have entirely different terms for the two — sortition for the former and stochation for the latter, but bitterly resent that would mean that those of us advocating the wisdom of crowds would lose out on the work we have done over the last two decades getting sortition on the agenda. If you look at the comments at the end of the Spectator post on the climate change CA (thread just posted), you’ll see that the former will be seen as an affront to democratic equality — as they will enable governments to implement unpopular policies under the pretence that they are the will of the people.

    Like

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 )

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: