Equality by lot

The Climate Convention: Technocratic illusions and pseudo-direct democracy

31 Responses

1. 

   rcaze, on November 15, 2019 at 11:42 am said:

   “How are the 150 citizens who are supposed to embody the people as they face the climatic challenge supposed to reach reasoned conclusions after a few weekends of work whereas expert researchers spend years in order to understand the complexities of energetic and ecological issues?” Let me disagree vigorously. Nobody asks them to find new solutions. There are already PLENTY of them proposed by thousand of expert. We ask them to select among these proposals. You might like the nuclear option but more and more people think the drastically reduce energy consumption may be the way (no ads would be a drastic measure that only citizen could propose).

   Let me try an example relating to the citizen jury in Ireland on abortion. Nobody asks them new abortion or contraception methods. Politics did not want to touch the subject but citizens did.

   LikeLike
2. 

   keithsutherland, on November 15, 2019 at 1:13 pm said:

   Excellent piece Yoram, thanks for translating and posting it. I would endorse all his conclusions — in fact I have an article under review with Res Publica that comes to very similar conclusions.

   LikeLiked by 1 person
3. 

   keithsutherland, on November 15, 2019 at 1:15 pm said:

   rcaze: for a critique of the Irish abortion CA, see https://equalitybylot.com/2019/11/04/a-citizens-assembly-on-climate-change-is-the-cowards-way-out/

   LikeLike
4. 

   Yoram Gat, on November 16, 2019 at 9:57 am said:

   It’s a mix of valid points and invalid ones. Here is my break down:

   1. The Convention is being created by the government for its own purposes and this gives it significant control over its procedures and therefore over its outcomes – valid point.

   2. “Contempt for intermediary bodies”: this is to some extent false – since intermediary bodies have (or at least could have) an influence over people’s views as expressed in their actions in the Convention. To the extent this is true (the intermediary bodies do not have official power in the process) this is a good thing, since those bodies tend to be non-democratic.

   3. “The procedures of selection and operation of this body are highly suspect” – valid. This is a corollary of point #1.

   4. Various statements about sampling: meaningless gibberish. (Of course, as mentioned in point #3, the selection procedure to the body is opaque and unrepresentative. This has nothing to do with the confused comments made about sample size, quotas, etc.)

   5. “The interactions and deliberations of a heterogeneous group with where opinion leaders may appear leaves the door open to different forms of manipulation of the results” – invalid. The fact that those designing the procedures have a lot of room for manipulation is undeniable and important (point #1). However this is not connected to the question of “interactions and deliberations”. This would be true in any decision making setting (e.g., if no deliberations take place).

   6. “It is very unlikely that reasoned conclusions can be reached after a few week-ends of talks and debates”, “the required solutions demand a different discursive mode than that of a [few weekends’ worth of] debate” – valid points.

   7. “Discussing opposing points of view is not really democratic unless the opposing views are organized by authorized representatives of different parties” – invalid. “Authorized representatives” presumably means “elected”, which translates elite domination.

   8. “Only a class society can be a free society, conflicts between organizations can only be resolved by despotism” – useless Marxist[?] sloganeering.

   9. Potential for manipulation in favor of nuclear energy: valid – a corollary of point #1.

   LikeLiked by 1 person
5. 

   keithsutherland, on November 17, 2019 at 12:57 pm said:

   Yoram,

   Any serious social science, statistics or polling researcher knows that the representativity of a sample with respect to a population from which it is drawn is only achieved by drawing a truly random sample of several thousand individuals. Or, at a minimum, when using quotas, to achieve bigger but still acceptable margins of error, no less than 800 to 1,000 people are required. Here they are using insufficient pseudo-quotas aiming to mix deliberative democracy with opinion polling.

   Given the context and intended readership of the article (a newspaper, rather than a statistics journal), why do you consider this “meaningless gibberish”?

   >“Authorized representatives” presumably means “elected”, which translates elite domination.

   That is only true if there is one homogeneous elite; not so if there is competition between elites, each seeking to attract popular support. Why do you dismiss this as Tesco vs Sainsbury?

   I told Prof. Salvador that you had translated and reposted his article to EbL and this was his response:

   Cher collègue, Je vous remercie de votre appréciation. Cordialement.

   LikeLike
6. 

   Yoram Gat, on November 17, 2019 at 6:09 pm said:

   Thank you for contacting Prof. Salvador and relaying his response.

   (I’ll avoid rehashing questions we have discussed many times before.)

   LikeLike
7. 

   keithsutherland, on November 17, 2019 at 6:15 pm said:

   I would be satisfied with a response on Prof. Juan’s statistical observations, and repeat the question:

   >Given the context and intended readership of the article (a newspaper, rather than a statistics journal), why do you consider this “meaningless gibberish”?

   LikeLike
8. 

   Yoram Gat, on November 17, 2019 at 6:26 pm said:

   As the famous expression goes, these sentences are not even wrong. They have no coherent meaning.

   If someone (“a serious social science, statistics or polling researcher”?) is able turn these sentences into claims that can be discussed rationally, maybe a discussion can be had. As they stand, this is impossible.

   LikeLike
9. 

   keithsutherland, on November 17, 2019 at 6:36 pm said:

   Well, you’re the statistician, but they make sense to a layman like myself, my understanding being that there is a stochastic relationship between sample size and decision threshold that can be calculated for a genuine random sample. This is how I put it in my PhD thesis:

   Two factors need to be considered when it comes to calculating the necessary sample size and decision threshold for a proportionately representative body. A ‘confidence interval’ (margin of error) is a range of values that is likely to contain an unknown population parameter. If you draw a random sample many times, a certain percentage of the confidence intervals will contain the population mean. This percentage is the confidence level. For example, suppose all possible samples were selected from the same population, and a confidence interval were computed for each sample, a 95% confidence level implies that 95% of the confidence intervals would include the population parameter.

   Given that it would be essential for a legislative minidemos to be an accurate portrait-in-miniature of the target population, the most demanding confidence level of 99.9% is assumed in the following calculations of sample size and decision threshold for a target population the size of the UK electorate (37,831,600) or US electorate (235,248,000):

   Margin of error Decision threshold Sample size
   2% 52/48 6,766
   5% 55/45 1,083
   10% 60/40 271
   Table 8.1: Margin of error, decision threshold and sample size

   The thesis was accepted by my examiners (including Peter Stone), but are you suggesting this is also “meaningless gibberish”?

   LikeLike
10. 

    Yoram Gat, on November 17, 2019 at 8:21 pm said:

    Unlike the sentences in the article, this excerpt has the merit that it makes meaningful statistical statements, that are moreover true.

    Of course, the idea that your little table at the bottom can serve as a useful basis for political science theory is false, but that is a different issue. Again, since we’ve been over this many times, I will not bother rehashing this issue.

    LikeLike
11. 

    keithsutherland, on November 18, 2019 at 10:59 am said:

    Yoram,

    Thanks for confirming that my presentation of the statistical principles involved is both meaningful and true. If so, then I think all Prof. Juan (or the newspaper sub-editor) is guilty of is journalistic shorthand.

    >the idea that your little table at the bottom can serve as a useful basis for political science theory is false

    That’s strange, because all I did was crunch the numbers (using the online calculator at at https://www.mccallum-layton.co.uk/tools/statistic-calculators/sample-size-calculator/).** Given that we both advocate “statistical” representation, why do you not consider this a “useful” basis for political science theory? Are you using “statistical” as a metaphor for your own proposal or are you taking slogans about “the few and the many”, “99%-1%” literally? This is a serious question, as it’s hard to understand how the diversity of large multicultural states can be accurately represented without numbers in the order of magnitude suggested by Prof. Juan (unless you assume, for doctrinaire reasons, a simple distinction between the interests of the elite and the masses).

    ** other tools give broadly similar results, for example https://www.surveysystem.com/sscalc.htm

    LikeLike
12. 

    tbouricius, on November 18, 2019 at 10:55 pm said:

    Keith,
    Your stated “assumption” that a mini-public needs to match the population at a 99.9% confidence level is inappropriate. We are comparing the democratic accuracy of a mini-public with the democratic accuracy of an elected chamber (even on policy preference, rather than demographics). Setting aside the fundamental reality that even the exact same individuals in a body would inevitably have different preferences on different days… pretending preferences were fixed things, a 99.9% accuracy is unrealistic. Do you think and legislative chamber matches the preferences of the electorate (if well informed on the issue) with 99.9% accuracy? with 50% accuracy, with 10% accuracy (maybe). ANY representative democracy is about being free from domination and “good enough” as well as better than the electoral alternative.

    LikeLiked by 1 person
13. 

    keithsutherland, on November 19, 2019 at 9:03 am said:

    Terry,

    The very demanding confidence level (99.9%) pertains to high-stakes decisions like Brexit. If the level were relaxed then the losing side would claim that it was a statistical aberration. Even if it were reduced to (say) 95% this would still require sample sizes in the order of magnitude specified by Prof. Juan (derided by Yoram as “meaningless gibberish”). These numbers are far in excess of the top limit for a deliberative group — this is why he calls out the danger of conflating deliberative democracy and polling. In my latest paper I argue that it is this conflation that causes Surowiecki to dismiss Fishkin’s project as “quixotic”.

    > ANY representative democracy is about being free from domination and “good enough” as well as better than the electoral alternative.

    “Freedom from domination” is straight from the neo-republican lexicon and puts you in the company of the recent wave of anti-democratic writers (Brennan, Somin, Caplan etc). By all means argue for republican or epistemic governance but please don’t claim that this has anything to do with democracy. From the perspective of being “good enough”, democracy means that the people (or an accurately representative sample thereof) has power. This is true, warts and all, as the majority are free to dominate whoever they choose, hence the need for constitutional safeguards (that have nothing to do with democracy).

    LikeLike
14. 

    Alex Kovner, on November 20, 2019 at 4:33 am said:

    The statistical discussion seems beside the point. Confidence intervals only have meaning if the CA is given a highly structure set of proposals to choose from. For example, the CA could be given a proposal from each of the main political parties in France about how to solve the problem, then vote on which it preferred.

    I’m not advocating that solution, by the way, just pointing out that statistics don’t mean much for a CA that is designed to come up with proposals, rather than make decisions. And why is that anyway? Why are all CAs given these open-ended mandates to make suggestions? Anyone who follows politics at all knows that any suggestions it makes will just be used as political Molotov cocktails.

    Instead, why not have the CA actually decide something? The Climate Convention seems to want to balance environment and social justice, and came about after anger over a carbon tax. How about a revenue neutral carbon tax, where the proceeds of the tax go to a subsidy for low wage work? A citizen assembly would then be called to set the rate of the tax, and a new assembly could be called perhaps every two years to adjust that rate.

    This would upend politics as we know it. Instead of having the CA make suggestions, and wait for the politicians to act, we can have the politicians make suggestions, and have the people (through a CA) act. It also answers some questions about the quality of deliberation. Coming up with a single number (the carbon tax rate, in my example) is much easier than generating a comprehensive solution to climate change.

    LikeLike
15. 

    keithsutherland, on November 20, 2019 at 8:58 am said:

    Alex:> This would upend politics as we know it. Instead of having the CA make suggestions, and wait for the politicians to act, we can have the politicians make suggestions, and have the people (through a CA) act.

    Yes, such a public process would not be open to covert manipulation (unlike “generating a comprehensive solution to climate change”). And from a Bayesian perspective, the agents (new-style political parties) that came up with the (epistemically) best proposals would be more likely to have their future proposals accepted, and the charlatans/populists would be sealing their own fate. Needless to say the statistical criteria that we are discussing would apply to such a citizens jury.

    LikeLike
16. 

    tbouricius, on November 20, 2019 at 4:44 pm said:

    I agree that making authoritative decisions (such as yes/no, or a tax rate, or to dismiss an executive) is the next essential step in sortitional democracy. The reason most citizen assemblies have NOT been this, but have instead sought to develop widely acceptable proposals, is that these are merely recommendations to either elected governments or to referendums. Because their decisions are not final, people are less concerned about how exactly they match the population as long as they are NOT dominated by special interests or elites. This is not to assert that such advisory Citizens Assemblies are necessarily a bad design… they are simply giving advice to the wrong final authority… that advice needs to go to a large mini-public.

    LikeLiked by 1 person
17. 

    keithsutherland, on November 20, 2019 at 5:01 pm said:

    Terry:> Because their decisions are not final, people are less concerned about how exactly they match the population as long as they are not dominated by special interests or elites.

    How can you tell, and who are these unconcerned people? We all agree that an average refusal rate of 95% generates a distinctly biased sample and, as Andre insists, elites come in different shapes and forms. There is a real danger of unrepresentative “random” sampling discrediting the whole sortition movement and, as Alex has pointed out, their recommendations can be used as political Molotov cocktails — as I’m sure the organisers of the Climate Convention propose doing (Prof. Juan’s concern is that it will be dominated by the nuclear industry). The advantage of election is that people get to choose between those competing to dominate them and this is a valuable constraint that should not be rejected too hastily.

    LikeLike
18. 

    Alex Kovner, on November 20, 2019 at 10:24 pm said:

    Terry:> I agree that making authoritative decisions (such as yes/no, or a tax rate, or to dismiss an executive) is the next essential step in sortitional democracy.

    One important consideration is whether the decision space is continuous or discontinuous. Yes/no decisions are discontinuous, and therefore tend to propagate a winner-take-all mentality. Parameters such as tax rates and the minimum wage can be a continuous function of citizen preferences. I have new blog post (https://alexkovner.com/2019/11/20/sortitionists-think-small/) describing how this would work.

    Focusing on continuous quantities could go a long way toward establishing the legitimacy of citizens’ assemblies.

    LikeLike
19. 

    tbouricius, on November 21, 2019 at 3:03 am said:

    Keith,
    I want to correct your statement above that
    >”We all agree that an average refusal rate of 95% generates a distinctly biased sample”
    What we actually agree about is that this MAY generate a sample that in some way is different than the population. Or it may not. Or the difference may be substantial but materially insignificant. We only have speculation.

    In my opinion, ideally, a lottery call should not tell those invited the specific topic they will be deliberating about… thus seeking people willing to serve their community, rather than people with a prior strong opinion on a particular topic. Alternatively, one proposal in Poland is that when doing a stratified sampling, in addition to matching age, income, education, etc., one trait that would be part of the stratification would be attitude about the topic. So if it the issue were climate change adaptation, the prospective members would be asked whether they consider climate change to be an important issue, or not, with the final draw matching the general population on this metric as well (to avoid the excess of “activist” sorts.)

    LikeLike
20. 

    keithsutherland, on November 21, 2019 at 9:41 am said:

    Terry,

    It’s true of course that we don’t know whether voluntarism is a statistically significant population parameter with respect to political preferences but your wish to seek people “willing to serve their community” would suggest that it is. You suggested recently that you would be happy for your civic-minded neighbour to make laws on your behalf and I retorted that such people might well assume that their fellow citizens are as well-disposed at themselves and make laws that are less well aligned to the “crooked timbers” of humankind. The precautionary principle would suggest we treat voluntarism as significant so, given your republican (small r) inclinations, why are you unhappy about viewing participation as a binding civic obligation?

    LikeLike
21. 

    tbouricius, on November 21, 2019 at 2:18 pm said:

    Keith,
    This is an example of a tendency of yours that so frustrates Yoram…. assigning beliefs to others…. Just because I used the term “free from domination” in talking about the integrity of a representative democratic body, and one thread of (small r) republican philosophy uses the term “freedom from domination,” that does not make me an advocate of republicanism. Also, as you know, I DO favor quasi-mandatory (civic obligation) service in final decision-making mini-publics. However the heavy workload and extended service of other bodies (such as agenda setting, proposal refining and review, oversight, etc.) is incompatible with mandatory service… here we seek impartiality and freedom from control by elites. (are you satisfied that I used the word control instead of domination?)
    Perhaps one problem we are having is that you only see one legitimate role for sortition (final yes no decisions without active deliberation) and are viewing everything through that microscope.

    LikeLike
22. 

    keithsutherland, on November 21, 2019 at 2:57 pm said:

    Terry,

    Yes, it’s true I see only one role for sortition as a democratic tool. Like it or not “freedom from domination” is the key principle of neo-republicanism, and I genuinely believe voluntary sortition-based agenda-setting committees simply trade in one form of elite for another (the only difference being that nobody gets to choose aleatory elites). Democracy means that the people (or a representative sub-set) have power, and the small, long-serving voluntary committees championed by deliberative and epistemic democrats do not square up to this definition, for reasons outlined by Prof. Juan in his article. (Domination and control are effectively synonymous.)

    I’m genuinely puzzled as why you want to do everything by sortition — this is entirely undemocratic.

    LikeLike
23. 

    Alex Kovner, on November 21, 2019 at 8:09 pm said:

    Keith and Terry,

    Far be it from me to break up a good spat, but I’m not sure how far apart you really are. The volunteers who would work in the agenda setting and proposal bodies will become professionals simply by the length of their service. The differences between political elites and the rest of us are not really so big, and are mostly related to how we spend our days.

    Terry, it sounds like you have two sorts of citizen bodies: mandatory decision bodies, and voluntary proposal bodies. Would you object if the voluntary bodies were subject to some form of rating by the mandatory bodies? And if that rating became the basis of their right to propose? This would prevent your voluntary assemblies from spinning off into a self-propagating elite.

    LikeLike
24. 

    tbouricius, on November 22, 2019 at 2:42 am said:

    Alex,
    Rather than explain it in a comment on a post, I’ll refer you to my 2013 paper https://www.publicdeliberation.net/jpd/vol9/iss1/art11/
    My intent is to overcome Michels “iron law of oligarchy.” I use a division of labor among mostly randomly selected groups appropriate to each task.
    In brief, a randomly selected agenda council with constant rotation (each person serving perhaps a year), selects an agenda and issues a call for proposals. They draft no proposals and vote on no proposals themselves. Any group of citizens can respond to the call for proposals from the agenda council. They know, however, that what they produce will ultimately have to pass a large random policy jury (so they will have an incentive to avoid extremist proposals that can’t get passed). A separate Review mini-public (serving for a year or two) for each policy area, reviews all proposals in their field, combines, rejects, amends, etc. to generate a final draft that goes to a large very short duration and very large mini-public that only deals with one bill, attends to pro and con arguments and votes without debate (to avoid informational and reputational cascades, etc.).

    Elections are anathema because they fail due to rational ignorance, empower elites and package together policy positions so that no voter can find a good representative on all matters. Those elected are completely unlike the people as a whole, with inflated self-importance, and with pre-decided positions that make them unfit for deliberation or drafting policy or decision making.

    LikeLiked by 1 person
25. 

    keithsutherland, on November 22, 2019 at 1:01 pm said:

    Alex,

    Could you clarify how the voluntary proposing bodies come into existence and are dissolved? My impression from reading your book is that they are elected. I think this is important, in that it’s hard to envisage a democracy in which all citizens (rather than the tiny number selected by lot) should not be able to express their (unconsidered) preferences.

    Terry:> My intent is to overcome Michels’ “iron law of oligarchy.”

    Alex makes the case in his book that oligarchy is the appropriate mechanism for voluntary proposing bodies (aka parties). Democracy arising from the competition between them both for unconsidered votes and the considered preferences of the disposing jurga. This fits nicely with the argumentative theory of reasoning, whereby the confirmation bias is of value for proposers (but should be avoided in bodies that only dispose). The trouble with deliberative democracy is that it conflates the two.

    Here’s the link to the book: https://alexkovner.com/2019/11/12/the-jurga-system/

    LikeLike
26. 

    Alex Kovner, on November 23, 2019 at 5:44 pm said:

    Keith:> Could you clarify how the voluntary proposing bodies come into existence and are dissolved?

    I have a blog post on that (https://alexkovner.com/2019/03/15/the-political-class-is-useful-for-something/). Basically, the proposing bodies are just agencies run by a board, similar to a corporation. Their right to propose rises and falls based on some measure of their popular support. In the blog post, I propose a rolling election to determine popular support. This avoids the pain of having an election at a particular time. But their are other ways: we could just create a large popular assembly to rank each of the proposing bodies, and get the score from that. As always, we must balance epistemic quality against perceived democratic legitimacy.

    These bodies are similar to Terry’s Interest Panels, except they are governed in accordance with the Iron Law, rather than as a way around it. I acquiesce to the Iron Law simply as a way to govern an agency, not because it has any inherent virtue. In my view, the Interest Panels are likely to act as oligarchies internally no matter how they are populated. Either that or they will become dysfunctional.

    I also replace the last place proposing body on a periodic basis to get new blood into the system. There are a number of ways to select the board for the replacement agency; I don’t have a strong opinion on the best way. Perhaps signature gathering, or perhaps just randomly from groups of citizens who apply.

    Another twist to consider is the possibility of eliminating one of the proposing bodies at random, perhaps weighted by score. Those with a higher score have a better chance at survival, but are not totally immune. This helps prevent the proposing agencies from becoming a self-propagating elite.

    One nice feature of this is that the amount of “new blood” that is introduced into the system is simply a parameter. Not enough new blood? Increase the frequency of replacement. Instability due to unstable proposing bodies? Decrease the frequency. Simple.

    LikeLike
27. 

    Alex Kovner, on November 23, 2019 at 6:04 pm said:

    I read your paper, and there is a lot in there to like. In particular you define Policy Juries as bodies that vote on one piece of legislation by secret ballot, guided by the principles of “The Wisdom of Crowds”. This is very similar to my notion of a jurga. I add the notion of mandatory participation (unless I missed it in the paper) as well as compensation (which you do include).

    The other body that is similar to what I envision is the Interest Panels, similar to proposing agencies for me. I differ, however, in their internal governance: as I see it, proposers are just agents, and should be held accountable to a score that reflects popular support, but otherwise left to govern themselves in accordance with the Iron Law. I rely on giving jurgas—Policy Juries—multiple choices from agencies that depend on popular support for their success. This leads to policy pluralism, resolved by citizen panels, which to me is heart of the matter.

    LikeLike
28. 

    keithsutherland, on November 23, 2019 at 6:36 pm said:

    Alex, I assume that’s a reply to Terry.

    LikeLike
29. 

    keithsutherland, on November 23, 2019 at 6:43 pm said:

    Alex:> I propose a rolling election to determine popular support [or] we could just create a large popular assembly to rank each of the proposing bodies, and get the score from that. As always, we must balance epistemic quality against perceived democratic legitimacy.

    Agree. The latter suggestion has epistemic merit, but does mean that the vast majority of citizens will have no direct input to the democratic process, and I can’t see that being accepted (at least not until the jurga principle is put to the test).

    LikeLike
30. 2019 review – sortition-related events | Equality by lot, on December 14, 2019 at 10:31 am said:

    […] ways to address the climate crisis. This body is relatively high profile and received attention by various writers. A similar body is being demanded in the UK by the Extinction Rebellion […]

    LikeLike
31. The French Citizen Climate Convention: a provisional analysis | Equality by lot, on November 25, 2020 at 7:34 am said:

    […] 4. Since the entire affair was organized by experts of various stripes and since, inevitably, experts were invited to speak before the CCC, suspicions about the influence of the experts on the proposals of the body naturally emerged. These too were widely voiced by critics. […]

    LikeLike

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