Equality by lot

Austria’s Climate Citizen Council: Broken from the Get-go

79 Responses

1. 

   paulnollen, on January 4, 2022 at 4:47 pm said:

   In our code of good practice (developed here on the forum) we propose a system for appointing the “experts” to hear (better proposals are always welcome) https://www.academia.edu/42201754/Code_of_Good_Practice_for_allotted_mini_publics_involved_with_legislation

   LikeLiked by 1 person
2. 

   keithsutherland, on January 4, 2022 at 5:34 pm said:

   it is a Citizen Jury’s sole prerogative to determine – already as part of a strictly neutral process – from where it gets its expertise and who is to testify. Needless to say, not just activists but potential experts of both sides with equal say should be contacted and invited to apply as witnesses.

   Unfortunately the latter does not necessarily follow from the former –hence Hubertus’s resort to the normative (“should”). This is particularly problematic if the citizen jury is voluntary, given that (on average) over 95% of those selected by lot decline the invitation. That’s why Alex and I insist that citizen juries need to be combined with advocates selected by the Superminority Principle, there being no other way to ensure that “both sides” are equally represented. I agree that it is essential for us to alert the public to the ever increasing abuse of sortition in the public sphere.

   LikeLike
3. 

   keithsutherland, on January 4, 2022 at 6:49 pm said:

   To make my point more clearly, why is it assumed that a randomly-selected group of citizens would be able to identify expert witnesses in a balanced and well-informed manner? Laypersons know little more on a most topics than they come across via their (increasingly polarized) media sources, so there would be a good chance they would arrogate the decision to the permanent support staff, lobbyists or the few jury members who happened to have some knowledge or contacts on the particular issue. This would be a very poor way of ensuring balanced advocacy, and there are good reasons to believe that a voluntary body would be heavily slanted towards activists who would choose expert witnesses in a partisan way.

   LikeLiked by 2 people
4. 

   Hubertus Hofkirchner, on January 4, 2022 at 7:40 pm said:

   paul: we propose a system for appointing the “experts” … better proposals are always welcome

   Paul, looking at section 6 of your Code (“Expert hearings”), it is definitely superior to the Minister’s charade. The first paragraph correctly summarises the very real potential for distortion and outright manipulation. And yes, those testimonies should be public, unless the assembly decides otherwise.
   As you asked for improvement proposals, I will take the liberty:
   We should not assign expert selection power to universities, this would break the principle of bottom-up power, and it is a back-door for elitism. The expert selection process must be unter full control of the citizen assembly in a rigoros initial evaluation process. The assembly decides on applications of experts who wish to testify and it is the assembly to send out invitations for such applications. The “pro/progressive” and “con/conservative” side* of a proposal must get an equal weight of witness statements. Last sentence: Instead of a full random sequence, you may wish to consider alternating pro/con presentations, starting with a coin toss.

   *) Terms used in the sortitionist sense, not meaning left/right parties.

   Keith: This is particularly problematic if the citizen jury is voluntary, given that (on average) over 95% of those selected by lot decline the invitation.

   Keith, briefly: practitioners (including you and I) who are fully conscious of this huge problem (international climate juries so far being a case in point) do have methods to avoid or largely mitigate this problem.

   Keith: … citizen juries need to be combined with advocates selected by the Superminority Principle

   Let us distinguish between: organisers, proposers, advocates/experts, citizens (in a stratified random sample). A supermajority principle for expert appointments would be problematic in practice, as the ex-ante distribution of opinions/values is rarely 50/50 which would bias the selection. Simplifying for brevity, I recommend a “systemic” vote on the list of expert applications, in parallel sub-samples if needed, maximising audience coverage with a TURF-analysis for each of the assembly’s separate pro/con/undecided segments.

   LikeLike
5. 

   tbouricius, on January 5, 2022 at 1:40 am said:

   Hubertus,
   Firstly a small explanation… Keith’s notion is a neologism “superMINORITY” process, where minority parties in proportion to their votes also get to put forth a share of proposals (or in this case witnesses)… not some traditional superMAJORITY concept.

   However, it is wrong to think of the realm of expertise as existing in a “both sides” or progressive and conservative dichotomy. How would the “two sides get defined? If dealing with space, could one side be who insist the earth is flat, and they get equal time?

   The best solution is to have a separate mini-public be in charge of designing the system for selecting expert witnesses for subsequent citizen assemblies. This should not be done by the mini-public directly addressing the issue, as it is a monumental task requiring a large amount of time. The idea is that they would design a system for selecting witnesses on ANY issue in the future (not making direct witness selection). They should not know what issues are coming up. This “veil of ignorance” gives them an incentive to design an optimal system (like the two children dividing a cake… one divides and one decides who gets which piece… the one who divides does not know who will get which piece and thus is motivated to be as fair as possible.) I agree that merely using university degrees, or government bureaucrats to make the witness plan is a disaster… but also letting the jury addressing a particular issue select their witnesses is likewise a mistake. It must be a prior jury that is removed from the particulars and independent.

   LikeLike
6. 

   Hubertus Hofkirchner, on January 5, 2022 at 8:33 am said:

   Terry: How would the “two sides get defined

   Thanks for asking. In fact it is a key problem with many well-intended recent citizen juries, that the task posed to citizens is creation (“Citizens come up with the solution”) instead tasking them properly, with decisions between proposals. This nonsense has been critisised since Ancient Greece, if you remember Socrates’s flutist. Properly defined decisions must always be formulated as agreement and disagreement (or undecided) with a proposed policy or resolution. Once this is achieved, experts and citizens alike can clearly define which side they are on at a certain point in time. This quasi-clarity yields remarkable cognitive advantages in practical processes.

   Terry; … have a separate mini-public be in charge of designing the system for selecting expert witnesses …

   Sorry but this is just absurd, and a perfect example for what’s wrong with many (most) of today’s citizen juries, as I just explained above. Such a design is obviously a highly complex matter, requiring deep understanding of several highly specialised scientific disciplines which is not sitting in a random sample. Experts, such as on this forum, come up with proposals for expert selection designs which we believe in, THEN only the citizen sample can and shall decide on a proposed expert selection design.

   Terry: It must be a prior jury that is removed from the particulars and independent.

   I see no reason for doubling the effort involved. Any citizen jury is perfectly able to select their own experts if we follow the procedure I described above, it takes but two hours for each citizen. Remember it is them who need the information which experts may have to impart.

   LikeLike
7. 

   Hubertus Hofkirchner, on January 5, 2022 at 11:45 am said:

   “superMINORITY” process, where minority parties in proportion to their votes also get to put forth a share of proposals (or in this case witnesses)

   I see, the term “superminority” is actually taken for sthg else. Anyway, I found a piece by Alex K. on that alternative superminority principle.

   Unfortunately, it does not ensure a fair witness structure, so I cannot recommend it.

   Proof:
   Assume:
   Jury: 6 members, a straw poll gives us 4 Pro some populistic proposal, 2 Con.
   Experts: 4 applications, 2 Pro (A, B) , 2 Con (C, D)
   Jury has time to hear 2 witnesses.
   Jury votes:
   1. A
   2. A
   3. B
   4. B
   5. C
   6. D

   If we were to use the proposed procedure for two expert spots, it needs a “superminority” of 33%. Only Pro witnesses would get to testify which obviously is undesirable.

   A systemic vote as described above avoids unbalanced hearings, it is under full control of the jury and no other power.

   LikeLike
8. 

   keithsutherland, on January 5, 2022 at 1:12 pm said:

   Hubertus:> Properly defined decisions must always be formulated as agreement and disagreement (or undecided) with a proposed policy or resolution.

   Agree. I’ll invite Alex to explain how the Superminority Principle achieves this (drawing his attention to your refutation).

   Terry:> This “veil of ignorance” gives them an incentive to design an optimal system.

   Rawls’s use of this phrase is normative — it was the key principle of his theory of justice. But the task of your foundational committee is epistemic (identifying the experts on all aspects of public policy). Given that the committee will be dealing with unknown unknowns, their task strikes me as impossible in principle.

   Given that the referent throughout is to “juries”, then why not adopt the practice of the law courts? The prosecution will select experts that advance its own case and the defence will do the same, leaving the jurors to decide who is the most credible. Nobody is pretending that the expert witnesses are impartial, and the procedure accords perfectly with the Argumentative Theory of Reasoning which Helene Landemore has adopted to support her epistemic case for deliberative democracy.

   LikeLiked by 1 person
9. 

   Alex Kovner, on January 5, 2022 at 3:10 pm said:

   Hubertus:> Unfortunately, it does not ensure a fair witness structure, so I cannot recommend it.

   Proof:…

   I think there’s a bit of misunderstanding about the superminority principle here. It is used by the permanent legislature (parliament) to produce a set of options; it is never used by the jury. As a result, the advocates or witnesses chosen would reflect the ideological diversity of the country as contained in parliament.

   I think you make a good point, however, that two advocates is too small a number. Three advocates would be better, even for a single up-or-down proposal. This implies a threshold of 25% + 1. That way the jury would see “left, right, and center”, however those terms are understood in parliament.

   Here is how it would work. Take the National Council of Austria, which according to Wikipedia is divided as follows:

   ÖVP (71)
   Greens (26)
   SPÖ (40)
   FPÖ (30)
   NEOS (15)
   Independents (1)

   With 183 members, the threshold (25% + 1) is 46. ÖVP can meed the threshold alone, but the other parties would have to form alliances to make it. I know nothing about Austrian politics, so I won’t speculate on possible combinations, but if only two groups meet the threshold, then the threshold might be lowered; a threshold of 20% + 1 (37 members) allows SPÖ to meet the threshold without another party.

   The point is, any group that meets the threshold gets to appoint witnesses or advocates, or whatever is specified by the jury process. The only requirement is that any privilege is available to each group equally.

   Hope this helps,

   alex

   LikeLiked by 1 person
10. 

    Hubertus Hofkirchner, on January 5, 2022 at 4:09 pm said:

    To get back on topic, I note that the forum responded unified in its firm rejection of a government minister appointing the experts for a citizen jury, as just happened in Austria.

    So, can we conclude that the Austrian Climate Assembly is broken from the start? (And we have not even discussed the suspicious appointment of organisers.)

    PS: Our verdict is not contingent on how exactly the citizens are to appoint experts. This debate could be another thread. Just to summarise the opinions, so far:

    – Parity voting by the jury itself
    – A second expert selection jury
    – A set of universities or high schools
    – Superminority voting* by the jury itself

    *) Included in the list even though Alex explained:

    … misunderstanding about the superminority principle here. It is used by the permanent legislature (parliament) to produce a set of options; it is never used by the jury.

    but Keith did in fact propose it:

    Alex and I insist that citizen juries need to be combined with advocates selected by the Superminority Principle

    LikeLiked by 3 people
11. 

    keithsutherland, on January 5, 2022 at 5:37 pm said:

    Hubertus,

    By “combined” I meant that in our proposal the decision-making jury is one part of a bicameral legislature in which advocates are elected by the superminority principle.

    LikeLiked by 1 person
12. 

    Liam Jones, on January 5, 2022 at 8:31 pm said:

    Hubertus,

    Thanks for drawing our attention to this. I suppose it’s a given that the people in power will try to co-opt whatever democratic mechanisms they find: It gives them a cloak of legitimacy while leaving them in charge. We can expect to see more stunts like this in the future.

    On a more general note:

    I suspect expertise is sortition’s biggest vulnerability. If you’re doing sortition properly (that is, randomly and not with stratified samples), it’s the easiest input to manipulate.

    Choice of experts is an ugly problem. I don’t think there’s any clean solution. If you leave it up to the citizen sample, they might be ignorant or misinformed. But if you leave it up to any other organisation, you’re giving that organisation power to control the results. If you insist on an adversarial model, you give power to those who want to obscure the scientific consensus for their own benefit (as tobacco companies did regarding carcinogens), and if you don’t, you risk false certainty.

    That said, electoral systems suffer from this weakness even more, so sortition still wins the title of least-bad.

    Overall, I’d be inclined to let the sortition body choose its experts, but also allow suggestions from outside — ideally another sortition body devoted to the task. All in all, I think it’s better to open the field (bandwidth permitting, of course).

    LikeLiked by 2 people
13. 

    tbouricius, on January 6, 2022 at 1:47 am said:

    Final decisions should be made by a jury as yes or no on a proposal. That we agree upon. Some think a citizens’ Assembly can also (0r alternatively) wade through a variety of ideas with witnesses to craft a final proposal, while others reject that idea, preferring to leave proposal generation to either partisans, or variously defined experts or bureaucrats. My notion is that the best option is to let any group who wants to draft proposals, and have a citizens’ assembly with an extended mandate listen to testimony on all sides to craft the final proposal that goes to the ultimate jury. EVERY stage needs to be overseen by a mini-public, including designing the process for selecting expert witnesses. That task is so complex that it is unreasonable to hope that a short duration jury can both decide a good procedure for selecting witnesses AND dig into the meat of the particular policy at hand. Thus, my plan for separate mini-publics for separate pieces of the overall project.

    LikeLike
14. 

    keithsutherland, on January 6, 2022 at 8:14 am said:

    Terry:> My notion is that the best option is to let any group who wants to draft proposals.

    Broadly speaking that is the superminority principle, the only caveat being that the group needs to have attracted a pre-specified level of public support before putting the proposal to the minipublic. Using a self-selecting minipublic as the filter would not pass Dahl’s stricture that the demos should have exclusive control of the proposal function.

    Liam:> If you insist on an adversarial model, you give power to those who want to obscure the scientific consensus for their own benefit (as tobacco companies did regarding carcinogens), and if you don’t, you risk false certainty.

    The superminority principle distributes advocacy power to all groups that have sufficient public support. If there were a scientific consensus on a particular issue (say global warming or vaccines) then the citizen jury would get to listen to opposing voices before deciding which was most convincing. I doubt whether lobbying by the tobacco industry would have succeeded. As you point out, if you drop the adversarial model (as classical deliberative democrats propose) then you run the risk of false certainty. There is an old adage that legislation based on a consensus decision is invariably wrong.

    LikeLiked by 1 person
15. 

    paulnollen, on January 6, 2022 at 8:45 am said:

    The proposal in the “Code of good practice” is covering some of the comments (page 10/34) :

    Therefore we suggest that universities and high schools, who are involved in the program “Educating for democracy” [ 1 3 part 3 page 1 2 3 ] (or any similar program) can apply to take part in this phase of the decision making. For this choices we also make use of sortition in order to avoid personal preferences and manipulation. A small Jury can oversee and report publicly this part of the events.

    Several universities/high schools needs to work together (at least three) in order to select a balanced panel of experts in the appropriate specialties. The final selection is done by sortition by an allotted Jury, taking into account the availability of the experts when the presentation is scheduled. Those experts are under the contractual obligation to deliver to the Jury a presentation in accordance to previously published work, listed in the contract.
    The universities/high schools involved (or even better if possible, other ones selected by lot) evaluate the presentation with the submitted publications.
    The presentations are live recorded and are part of the published general documentation, together with the evaluation, and eventually any answers to questions of the Jury.
    It is to avoid that experts are nominated who have a to clear political commitment (elected representative, prominent member of a political party think tank, .. ) [ 8 page 1 0 ]
    .
    It is clear that violations of the contract can have judicial consequences.
    The input from separate special task forces or mini-publics [ 8 p 17 ] [ 1 4 ] is to be regarded as separate expert input.
    The sequence of the expert presentations has to be random

    LikeLiked by 1 person
16. 

    paulnollen, on January 6, 2022 at 8:50 am said:

    I found it especially useful that young studens learn about the diversity of thoughts and the evaluation of different arguments. This way they also learn how “our democracy”, and sortition, works. Even if they are not chosen they can work out their own input and submit it to the citizens assemblee.

    LikeLike
17. 

    Hubertus Hofkirchner, on January 6, 2022 at 9:42 am said:

    Liam.> Overall, I’d be inclined to let the sortition body choose its experts, but also allow suggestions from outside — ideally another sortition body devoted to the task.

    The first sentence is certainly correct. However, there are important benefits if the jury itself selects its own experts, initially and as they go.

    I summarise:

    1. First, a proper citizen jury needs neutral and competent organisers. Therefore in a first step, the organisers need a vote of confidence from the citizen jury, with a supermajority as there should not be the shadow of a doubt.

    2. After confirmation, the organisers issue an open call for experts to apply as witnesses for or against a proposal. There should also be a reasonable honorarium.

    3. After hearing the proposal, the citizens can request the organisers to invite specific experts to apply directly (adding to step 2)

    4. The experts file their applications in an online form, summarising their position, their side (pro/con), their credentials, including a short video.

    5. The citizens vote their (initial) side and vote on randomly assigned applications of any side, in sufficient number to reach a 95% confidence, using the “lemon” filtering technique (for highest efficiency). The ranked list of experts for each side is compiled only from the votes of each side plus undecideds.

    Citizens benefit from an overview of both side’s arguments, they have full control of who is selected, both sides get equal voice. The time spent with this neutral process is minimal, it can be done online.

    And government manipulation and corruption is blocked.

    For the avoidance of doubt, the effort is so low that a second round of expert invitations can be triggered (at the sole discretion of the citizens) if they deem the first round insufficient.

    PS: If anybody wants the needed software for this efficient procedure, feel free to contact me at Prediki Prediction Markets.

    LikeLike
18. 

    paulnollen, on January 6, 2022 at 10:26 am said:

    The Hubertus proposal is covered in the Code by : The input from separate special task forces or mini-publics [ 8 p 17 ] [ 1 4 ] is to be regarded as separate expert input.

    LikeLike
19. 

    Alex Kovner, on January 6, 2022 at 4:40 pm said:

    Liam Jones:> Choice of experts is an ugly problem. …

    There are no experts, from a structural point of view. The only distinction is between advocates and jurors. Advocates often have knowledge, but they are assumed to be partial, and jurors often lack prior knowledge, but can be impartial.

    “Experts” are just a species of advocates. The problem is you are trying to find “impartial experts”. This is an empty category.

    LikeLiked by 1 person
20. 

    Alex Kovner, on January 6, 2022 at 4:44 pm said:

    terry:> Final decisions should be made by a jury as yes or no on a proposal. That we agree upon.

    Speaking on my own behalf from the superminority perspective, the jury should rank a number of options in order of preference. There should be at least 3 options.

    terry:> That task is so complex…

    Indeed it is. Simplicity is not a luxury.

    LikeLike
21. 

    Alex Kovner, on January 6, 2022 at 6:21 pm said:

    A lot of the discussion here seems to revolve around the idea of finding “impartial experts”; i.e. people who possess clear knowledge but whose views are somehow purely in the public interest. However, if such people existed, and could be identified, then there would be no need for a citizen jury. We would have our guardian council.

    I think that is the motivation for some when it comes to CAs. They see “experts” as true and righteous, and they believe that putting such unimpeachable rectitude before ordinary citizens will allow them to circumvent the evil of elected politicians. Hence the focus on universities.

    This is fantasy, and it is a fundamentally misguided model. There are no “impartial experts”; every expert is part of a society of experts through training and habit. As the famous Adam Smith quote goes, “People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public,…”. This applies to expert societies as well as to trade groups. It’s not evil, it’s just human nature.

    Now, my own opinion is that universities are more publicly spirited than major corporations. But that’s just a personal opinion. I express that opinion as a voter, and I would express it as a juror if I had the chance. But as a structuralist, I must view universities and corporations as exactly the same: organizations of self-selected people who have a large informational asymmetry on certain topics, and strong biases on those topics.

    The participants of this forum (and people involved in sortition everywhere) need to decide whether they are structuralists, or whether they are trying to win specific political battles. If you are a structuralist, you cannot view universities as any different from corporations, or labor unions, or Beyoncé fan groups for that matter.

    Please, stop trying to find the impartial expert. I beg you. Knowledge and bias are innately connected. Stop looking for pink unicorns.

    If it’s not the goal to make value judgments about self-selected groups, then what is the goal? The goal is to ensure that selected advocates represent a broad spectrum of the knowledge class on a particular issue (by “knowledge class” I’m not assessing truth value, just self-selection and information asymmetry). This is a distributional requirement. Structuralists only care about the distribution of advocates in a political/ideological sense. It’s the jury’s job to figure out which advocates to believe.

    Some people pushing these climate change CAs don’t want corporations to have a chance to speak. But, whether you like it or not, corporations (and capitalism as a political view) have substantial support in the general public; in fact, my guess is they have more support than climate activists have. If you want to prevent an ideological position that has significant representation in the public from addressing the jury, then you don’t trust the jury. Then what is the point of sortition?

    Ultimately there is a gut check here. What if citizen juries consistently make decisions that are contrary to your own passionately held political views? Can you accept that?

    LikeLike
22. 

    keithsutherland, on January 6, 2022 at 6:36 pm said:

    Alex, it’s great to have you back with your laser-sharp clarity — especially as you’re suffering from Covid-19 at the moment!

    LikeLiked by 1 person
23. 

    Yoram Gat, on January 6, 2022 at 7:30 pm said:

    Liam,

    > Choice of experts is an ugly problem. I don’t think there’s any clean solution. If you leave it up to the citizen sample, they might be ignorant or misinformed.

    I don’t see why the problem of who to consult with is more difficult (or less difficult) than other problems facing the allotted body. Many factors in the way the body works could affect its effectiveness and need to be done carefully and democratically. In all such cases, the only democratic way to handle things is to let an allotted body (either the same one or another) decide matters (even if in all such cases, the body may make mistakes – everybody can make mistakes). It is clear that any intervention by non-allotted elements that is not managed by the allotted would damage the ability of the allotted body to produce democratic decisions.

    LikeLiked by 1 person
24. 

    Liam Jones, on January 6, 2022 at 9:04 pm said:

    Hubertus,

    “The first sentence is certainly correct. However, there are important benefits if the jury itself selects its own experts, initially and as they go.”

    I agree. The people who are engaged in the decision are well place to know what sort of information they need to decide. But if there is a position they are ignorant of, they might never see it by selecting their own experts.

    The system you propose fixes that issue, because the experts put themselves forward. It seems workable at first glance, though I don’t like pro/con part — the space of possible positions isn’t binary. I’d want some way for them to declare their position if they have one, but in a more complex way than that.

    Incidentally, regarding your earlier disagreement with Terry: If I’ve understood multibody sortition correctly, your proposal would be perfectly compatible. Its just that there would be another sortition body that would be able to change the rules of expert selection if it saw the need. I think that’s worth doing; it’s better to have the rules under democratic control than remain as something absolute and unchangeable.

    LikeLike
25. 

    Alex Kovner, on January 6, 2022 at 10:05 pm said:

    Liam Jones:> The people who are engaged in the decision are well place to know what sort of information they need to decide.

    No, they are not. The jury is chosen to be impartial, which requires a certain amount of ignorance. Most jury members will not know where to go for information. That’s the value of the jury: if they did have it all mapped out ahead of time, they wouldn’t be impartial.

    I think you are underestimating how difficult impartiality is to maintain. It’s not just the selection process. If jurors start arguing with each other, they get into personal rivalries and impartiality is lost. Do they choose by majority vote? If so, who decides which experts to vote on? You are now reproducing the dynamics of traditional legislatures which are full of pointless infighting.

    LikeLike
26. 

    tbouricius, on January 6, 2022 at 10:52 pm said:

    I agree with Alex that a jury would NOT be competent to select experts or other advisors. They only know of those that they have seen on TV who have been selected by various (often partisan) news media. There are often entire domains of knowledge that we as lay people do not even know exists (we may not even know the most important questions to ask). The democratic oversight of the flow of information to juries is critical. Alex’s approach is to rely on elections to select parties that will control the process prior to the final decision by a jury. I think that is very problematic, as it elevates rational ignorance, public relations manipulation and campaign skill to the fore rather than any beneficial expertise. Letting special purpose mini-publics design and evolve the information control process is essential for democracy.

    LikeLiked by 1 person
27. 

    Alex Kovner, on January 6, 2022 at 11:02 pm said:

    terry:> Alex’s approach is to rely on elections to select parties that will control the process

    I have no attachment to elections or parties. I am simply making better use of what we already have. Elected legislatures aren’t going anywhere anytime soon.

    I’m more than happy to replace elected legislatures with something better. My only requirement is that the experts and advisors must distributionally represent the spectrum of political opinion within the country as a whole.

    LikeLike
28. 

    Liam Jones, on January 6, 2022 at 11:31 pm said:

    Yoram,

    I confess I don’t have an argument as to why expert selection is comparatively worse that other issues in sortition. It’s more that when thinking about the other issues, they relatively tractable, whereas with expert selection there seem to be deep problems with every approach.

    It might be a matter of perspective: Part of my background is in philosophy of science, so I’m already somewhat familiar with how muddy the terrain is. And connected to that is that it’s not just a sortition problem but an epistemology problem. Every society faces it in some form.

    One reason it’s difficult comes down to recursion: If you’re ignorant about something, you need someone who has more knowledge. But if you’re ignorant, how do you know who to ask, and how do you tell the difference between a charlatan and an expert? (This is the point Terry is making above, but he doesn’t extend it to the general case.)

    We do solve that issue every day in practice, because the recursion isn’t deadly. We can build on the knowledge we have, we can do some exploratory research, we can fall back on heuristics, and so on. But all of these solutions are partial. They can fail. And, in some cases, they can be *made* to fail with the right institutional setup.

    “(even if in all such cases, the body may make mistakes – everybody can make mistakes)”

    From this, I think we mostly agree. Certainly giving away the choice of expert selection to some elite body undermines sortion’s democratic potential. It’s just that I’m triple underlining the “everybody can make mistakes” part.

    LikeLiked by 2 people
29. 

    Liam Jones, on January 6, 2022 at 11:44 pm said:

    Terry,

    “There are often entire domains of knowledge that we as lay people do not even know exists (we may not even know the most important questions to ask).”

    This is true, but I think it’s relatively (bearing in mind all my earlier worries) solvable.

    You only need one person to know where to look, and they can inform the rest. As the sortition body grows, it becomes more likely that someone will know where to look. Quite a lot of people are quite well informed about something, even if they’re ignorant about other things.

    If you add to that another body that can suggest experts (even, as Hubertus suggests, the experts themselves) you reduce the space of ignorance even more. You don’t need to remove the power of choice from the jury to do this. As I said before, it’s better to open up the field, bandwidth permitting.

    (And I should qualify all of the above by saying it applies to solving the unknown unknowns. When it comes to the choice between bad experts and good experts, things get difficult again.)

    LikeLiked by 2 people
30. 

    Alex Kovner, on January 6, 2022 at 11:47 pm said:

    Liam Jones:> …whereas with expert selection there seem to be deep problems with every approach.

    When every approach seems to fail, this usually indicates a fundamental misunderstanding of the problem.

    LikeLike
31. 

    Alex Kovner, on January 7, 2022 at 3:50 am said:

    Liam Jones:> You only need one person to know where to look, and they can inform the rest

    This is a disaster. The one person who “knows where to look” might be a conspiracy theorist who takes everyone else with him. The wisdom of crowds requires intellectual independence within the crowd. You’re negating that here.

    LikeLiked by 2 people
32. 

    Hubertus Hofkirchner, on January 7, 2022 at 8:38 am said:

    Alex,

    The jury is chosen to be impartial

    You correctly argued that experts cannot be impartial. The same impossibility concerns the jury, for exactly the same reasons. I know some (all?) here have their pet procedures, but that’s a striking logical dissonance of your argument, thus a fatal flaw in whatever design you have in mind.

    A good design must be aware that jury members will have an initial opinion, based on unknown incomplete beliefs of theirs, which again may be correct or incorrect. Thus, they are perfectly able to invite experts which would testify their belief. And, in the “open” bottom-up procedure proposed, experts can apply freely without invitation, and also alert others whom they believe worthwhile, so it is up to the experts to convince the jury that their testimony is worth their time. There is nothing “hard” about this procedureI, at all.

    PS: The various other approaches by some here did not “fail, yet” – although some are logically inconsistent (Alex’s) or elitist (Paul’s) from the outset. It takes the practical experience with the approaches, and then the study of success or failure of each approach, to determine what works and what not.

    LikeLike
33. 

    keithsutherland, on January 7, 2022 at 12:32 pm said:

    Hubertus,

    The impartiality of a large jury where each decides independently is an aggregate function, so there can be no comparison with the speech acts of an individual advocate or expert witness.

    LikeLiked by 2 people
34. 

    paulnollen, on January 7, 2022 at 3:21 pm said:

    I made a summary of most of the proposals that I think can find (some) agreement here.

    We should not assign expert selection power to universities, this would break the principle of bottom-up power, and it is a back-door for elitism. The expert selection process must be unter full control of the citizen assembly in a rigoros initial evaluation process. The assembly decides on applications of experts who wish to testify and it is the assembly to send out invitations for such applications.

    That’s why Alex and I insist that citizen juries need to be combined with advocates selected by the Superminority Principle, 

    Simplifying for brevity, I recommend a “systemic” vote on the list of expert applications, in parallel sub-samples if needed, maximising audience coverage with a TURF-analysis for each of the assembly’s separate pro/con/undecided segments.

    The best solution is to have a separate mini-public be in charge of designing the system for selecting expert witnesses for subsequent citizen assemblies. 

    THEN only the citizen sample can and shall decide on a proposed expert selection design.

    Overall, I’d be inclined to let the sortition body choose its experts, but also allow suggestions from outside — ideally another sortition body devoted to the task. All in all, I think it’s better to open the field (bandwidth permitting, of course).

    EVERY stage needs to be overseen by a mini-public, including designing the process for selecting expert witnesses

    As far as I can see they are all covered (may be not in detail but the interpretation allows it) in the “Code of good practice” where I (and some others) tryed to think of all the proposals and experiences done already (Ireland proposals, BC electoral system; Iceland constitution, Hélène Landemore cosntsitution of Mars, Terrill Bouricius multi body system and so on).

    Please correct me if I am wrong.

    LikeLike
35. 

    keithsutherland, on January 7, 2022 at 3:31 pm said:

    It strikes me that there is a lot of hand-waving going on here (aka pious wishful thinking). What exactly does it mean for a minipublic to design anything — especially a future-proof constitutional abstraction?

    LikeLiked by 1 person
36. 

    tbouricius, on January 7, 2022 at 4:16 pm said:

    Two points:

    1. As Alex wrote: “The one person who “knows where to look” might be a conspiracy theorist who takes everyone else with him. The wisdom of crowds requires intellectual independence within the crowd.”
    It is this independence that we get by separating a mini-public that adopts an expert selection process from a mini-public with an issue at hand. If I know my jury is deciding on a carbon tax, and I initially support the concept, I am likely to know or find experts to ask to reinforce my views. However, if I am working with a diverse group designing a process for future mini-publics to get expert advice, and I don’t know what topic is at hand, I have an incentive to design a system that will be sure to give that future body a wide range of perspectives, perhaps with some diversity of qualifications, and random selection from volunteer experts, etc.

    2. While the super-minority concept might be appropriate for limiting the number of proposals to a short but diverse set of option, it doesn’t work for selecting experts. Having five proposals, where one is completely different than the other four encourages listeners to examine the unique one with real attention. But if I have five witnesses, with 4 all agreeing on the main points, I am more likely to dismiss the odd ball. If anything, the scientific method encourages SEEKING OUT disproportionate input from those who might falsify early assumptions.

    LikeLike
37. 

    keithsutherland, on January 7, 2022 at 4:44 pm said:

    Terry,

    Scientific practice (an intensely competitive process) involves something not dissimilar to the superminority principle. Each party seeks to garner evidence to persuade the jury of their peers. Anyone who has experience of submitting journal papers for peer review will attest that referees are anything but impartial. It’s a messy business, but seems to work out in the long run.

    >a system that will be sure to give that future body a wide range of perspectives, perhaps with some diversity of qualifications, and random selection from volunteer experts, etc.

    What would that mean in practice and how would the jury not be overwhelmed by experts all contradicting each other? I’m minded of Harry Truman’s preference for one-armed economists.

    LikeLike
38. 

    Alex Kovner, on January 7, 2022 at 5:17 pm said:

    terry:> If anything, the scientific method encourages SEEKING OUT disproportionate input from those who might falsify early assumptions.

    I think we should keep in mind that we are not creating a committee to come to a scientifically correct conclusion. We are selecting advocates to make arguments on policy decisions.

    Policy decisions are about weighing harms and benefits. That’s why I think this focus on finding true “experts” is misguided. The focus on climate-based CAs over the last few years (including this one) is doing damage to sortition. Climate activists who don’t care a whit about sortition think they can use it to crack the skulls of politicians whose campaign funds are full of oil money.

    Or, conversely, some politicians (Macron) think they can use it to deflect attention from their own political weaknesses. Whatever the motive, the recurrent use of CAs only for climate policy is creating the impression that CAs are like a panel of scientific inquiry. Nothing could be further from the truth.

    Maybe we should ask, what if the procedures we are suggesting were applied to a CA whose purpose was to determine sentences for violent crimes? or to deal with the boundaries of religious liberty/exceptionalism? These questions are much less scientific and much more obviously values based.

    People in this forum care about political reform first. I think we’re getting sucked into a discussion that is less about sortition and more about the particulars of climate politics.

    LikeLike
39. 

    Hubertus Hofkirchner, on January 7, 2022 at 5:25 pm said:

    Gentlemen, I am mindful that we have not yet discussed the minister’s suspicious appointment of the single moderators/organisers applying after an “EU-wide tender”. Opinions?

    What can we do if we disagree with this kind of selection and want to stop this blatant abuse of CAs?

    PS: I have a plan, obviously, but will hold back until there are a few comments.

    LikeLike
40. 

    Alex Kovner, on January 7, 2022 at 5:48 pm said:

    Hubertus,

    Can you provide links to news items? Some of us do not follow Austrian politics.

    LikeLike
41. 

    Hubertus Hofkirchner, on January 7, 2022 at 6:00 pm said:

    Alex, the key document, not yet picked up by Austrian news, is the parliamentary protocol:

    Click to access imfname_1021573.pdf

    Otherwise the CA is not exactly click bait (yet), only 1 news article by the #7 ranked print newspaper, “Kurier” after the minister’s announcement:
    https://kurier.at/politik/inland/100-buerger-sollen-die-klimapolitik-vorgeben/401859422

    Essentially it rehashes the minister’s blurb without any analysis or understanding of the issues this entails.

    LikeLiked by 3 people
42. 

    Liam Jones, on January 10, 2022 at 10:42 am said:

    Alex,

    That’s quite a lot to respond to. I’ll try and get through it all. First, let me clear up a couple of your minor mistakes:

    This is a disaster. The one person who “knows where to look” might be a conspiracy theorist who takes everyone else with him.

    You’re correct. However, if you trouble yourself to read the last paragraph of the post you’re responding to, you’ll notice I qualified my argument: “When it comes to the choice between bad experts and good experts, things get difficult again.”

    And if you read my first post in the thread, you might notice I made a similar point: “If you leave it up to the citizen sample, they might be ignorant or misinformed.”

    So you’re trying to argue against me by making a point that I’ve already made multiple times, in my first comment and *in the very comment you’re responding to*.

    But that’s not all. Because yes, while one person might invite a conspiracy theorist, another might invite a debunker. By opening up the field as I’ve suggested, we tend to get advocates for differing positions making their case. Even by the standards of advocacy you mention above, this works.

    You fail to mention that both these problems – not knowing who to ask and distinguishing bad experts from good experts – aren’t just properties of citizen juries or assemblies. They apply in general to all social information gathering, because we all start from a position of ignorance.

    When every approach seems to fail, this usually indicates a fundamental misunderstanding of the problem.

    This is an oddly petulant attempt at a gotcha. If we’d clashed before, I might understand it, but this is the very first time I’ve spoken to you. Anyway, even on its own terms, it doesn’t work.

    I did not say “seems to fail”. I said “has deep problems.” A problem is not a failure. A solution that fails is by definition unworkable, whereas a solution with problems can still be enacted. This difference is crucial.

    Come on, dude. If you can’t even accurately repeat what I’ve said *when the words are right there in front of you*, what hope do you have of offering a remotely decent criticism? Are you just going to beg and accuse people of misunderstanding until they capitulate?

    Oh, and – I’ve found that if your think your solution has no problems, that usually indicates it’s an unworkably optimistic utopian solution.

    All of above is just the sideshow. On to the main event:

    In your initial response, you say:

    The problem is you are trying to find “impartial experts”.

    And later:

    I think you are underestimating how difficult impartiality is to maintain.

    Am I? Because that’s news to me. When I was writing those words, I didn’t think once about impartiality. In my first post I was concerned about problems in a more general sense. Mostly about how decision making requires reasonably if not perfectly accurate information, and about sortition can be manipulated by outside forces.

    If you look closely, you might notice that I didn’t actually use the word impartial at all. In fact, out of curiosity, I did a phrase search for “partial” on this comment thread. It occurs precisely once before your accusation, in a comment by Keith.

    So, what you’ve offered us is a long and impassioned speech, full of paroxysms of pulpit-thumping and passionate pleading against a position I haven’t even thought about – against a position that, as far as I can tell, no one here has taken. That’s quite a feat for someone who accuses others of looking for pink unicorns.

    Nevertheless, though you’ve failed to argue against anything but your own imagination, you have put forward a few positive claims, so let’s examine those.

    (And to clarify here, I’m not talking about the superminority proposal as such, which fails for many more reasons than I list below, but the underlying abstract schema you’re using.)

    As I understand it, you suggest a system divided between partial advocates-not-experts on the one hand and an impartial jury on the other. Your advocates are assumed to be necessarily partial because they’re experts (but also you claim that experts don’t exist from a structuralist perspective, which contradicts that). Your jury is assumed to be ignorant and therefore impartial. The consequence is that the jury decisions will therefore be partial.

    Barring the question of experts who sometimes exist and sometimes don’t, is that an accurate summary of your position? If not, please correct me.

    So let’s go through all the ways in which this schema fails.

    First, we need to confront that contradiction. You say that experts don’t exist from a structuralist perspective (whatever that is). If that is the case, then the claim that experts are necessarily partial doesn’t work. Now, it might be that the latter statement is only from a non-structuralist perspective. But since you’ve gone to great lengths to demand we take the structuralist perspective (though you’ve neglected to explain or justify this), it seems the latter statement is something you think we ought to avoid.

    Further, it’s not really clear what you mean by the structuralist perspective. The most charitable interpretation I can come up with is that on the level of social structures, one can’t in principle distinguish between good experts and bad experts; the professional networks of a doctor and a homeopath can in principle look identical. That’s correct (and it’s part of why the problem of expert selection is so difficult), but it’s not the whole story. Because objectively, there is a difference, and that difference is important to the functioning of the sortition body.

    Moreover, your own later statements counter this charitable interpretation: “universities and corporations are exactly the same” from a structuralist perspective. But this doesn’t make sense. Even on the level of social structure, universities and corporations *are* objectively different and behave in different ways. Obviously by abstracting away all the differences, you can make anything look like anything else. But it’s not informative to do so. A button mushroom and a death cap are both fungal fruiting bodies, but that’s definitely not an equivalence you want to use when you’re making dinner. If your theory fails to pick up objective and relevant distinctions that exist in reality, it’s your theory that’s at fault, not reality.

    Second: This model of partial advocates and impartial jurists is a cartoonish oversimplification of the world. (I suppose that’s why Keith finds it appealing in terms of clarity.) You say simplicity isn’t a luxury, but then set about simplifying the world into a series of artificial dichotomies jammed together to justify your model.

    Your reason for wanting advocates and not experts is that experts are never impartial. This may be true (although the only justification you manage to offer for this is an Adam Smith quote that doesn’t even mention experts and the same old appeal to “just human nature” every ideologue uses to cover conclusions they’ve pulled out of their arse). But even if it is, it doesn’t justify wanting advocates. Because even if expertise always entails advocacy, the two are different things. To make a decision, we generally want information. Expertise implies someone can provide information, but advocacy does not. If I ask a doctor for health advice, it’s because of his knowledge and training, not because of what he believes. Even if I want a second opinion, I generally ask another doctor and not some bloke down the pub. The latter might have some strong opinions – he might be an advocate – but that’s not what I need for health advice.

    On the other side, you declare that the jurists are impartial because they’re ignorant. This is poor reasoning because while expertise may prevent someone from being impartial, its absence doesn’t guarantee that they are impartial. But also, as Hubertus points out above, it’s also just manifestly false. People have all manner of strongly held opinions about all manner of things. They’re influenced by the media, by their own experiences, by each other. They’re not going to leave all this at the door. And, a point that hasn’t been made yet, a great many ordinary people are experts of a kind, in that they tend to know a great deal about their careers, avocations and interests. That’s not going to go away either. (On this point, you seem to support some of Keith’s blurting about independent decisions, but there is no reason to believe aggregation or deciding independently of others can turns partiality into impartiality. This is a fever dream of a political alchemist.)

    Third, this entire structure is built upon the ideal of trying to achieve impartiality. But why? You’ve made all manner of demands, laid down stringent requirements in the pursuit of this goal, so it’s worth asking.

    Yes, impartiality is a nice-sounding word, full of associations of humanism and properly-carried-out justice and such. But all that, by itself, is just fluff. What does it mean in substantive terms? Impartiality might make sense in terms of dispute resolution (you don’t want a someone’s best friend judging who’s right in a dispute with a stranger), but political decision making is an entirely different context.

    You’ve offered us an implicit definition when arguing against impartial experts: “People … whose views are somehow purely in the public interest.”

    If this is what you want from impartiality, you’re chasing a bigger pink unicorn than anyone here. This is more fluff. There are no views purely in the public interest. It’s not even clear what the public interest really is in ant significant sense. There are individuals, with their own views and their own interests, including a sense of community and always influencing each other, yes, but none of that amounts to a public interest. We can aggregate their views by voting or by other means, but that’s not the same thing.

    But even if this definition did make sense, there’s nothing in it that implies impartiality can be created by a bunch of partial people voting. There’s nothing in the definition that implies people influencing each other breaks it. Even in view of the definition you offer, all your demands are arbitrary.

    Ultimately, you’ve jammed together a hodgepodge of arbitrary and absurd demands, linked only by unjustified assertions, in the service of an imaginary ideal.

    I’m not beholden to your phantasms. I’m not interested in magically impartial juries, a shy and ghostly “public interest” that for some reasons only appears when ignorant people don’t talk to each other, or what the tooth fairy wants. I’m interested in how people can govern themselves without being ordered around or controlled by rulers. Which is to say, I’m interested in democracy.

    LikeLiked by 1 person
43. 

    Liam Jones, on January 10, 2022 at 10:55 am said:

    Hubertus,

    I’m sorry for hijacking your blog with that monster comment.

    Regarding the matter at hand, I think that despite all the arguing, we’re all agreed that government shouldn’t interfere with the working of a sortition body.

    I can’t claim any familiarity with the situation, so all that follows is very speculative:

    Going by what you’ve said, there already seems to be a broad awareness of corruption in Austria. Could that be useful? Is there potential for say, a petition or an open signed letter condemning the interference and saying that sortition should always be free of political interference? That statement should be simple and intuitive enough to appeal not just to the sortition geeks (who are relatively few) but to many ordinary people who are opposed to corruption in general. And should it take off, it will be a useful precedent for later battles. If you’re really lucky, you might even get the assembly itself to reject this interference.

    LikeLiked by 2 people
44. 

    keithsutherland, on January 10, 2022 at 1:18 pm said:

    Liam,

    The superminority principle (SP) is that parties will choose their own expert witnesses and that jurors will independently decide which is most persuasive, leading to aggregate impartiality. The lower the SP threshold, the greater the array of advocates and witnesses. This perspective aligns with the argumentative turn in cognitive science (Mercier and Landemore, 2012) which overturns the classical view that rationality is the pursuit of objective truth.

    Ref
    ==
    Mercier, H., & Landemore, H. (2012). Reasoning is for arguing: Understanding the successes and failures of deliberation. Political Psychology, 33(2), 243-258.

    LikeLike
45. 

    Liam Jones, on January 10, 2022 at 3:58 pm said:

    Keith,

    Repeating your assertions does not make them any more persuasive.

    LikeLiked by 1 person
46. 

    keithsutherland, on January 10, 2022 at 4:44 pm said:

    I was simply clarifying our position, as you didn’t seem to understand it (have you read Alex’s original post?). As to whether you find the SP attractive or not, the ball is in your court.

    LikeLike
47. 

    Liam Jones, on January 11, 2022 at 11:19 pm said:

    have you read Alex’s original post?

    I have, yes. More than once.

    In my post above, I restricted my attention to Alex’s comments. Were I to go through all the failings of superminority (mostly due to its reliance on electoral politics), we’d be here all week.

    (Another characteristic of the utopian mindset, I suppose, is to interpret all disagreement as misunderstanding, because obviously anyone who understood your System of Perfection would do nothing but praise it.)

    LikeLiked by 1 person
48. 

    keithsutherland, on January 12, 2022 at 12:55 pm said:

    Liam,

    Given that the Superminority Principle (SP) argues for little more than a modest reduction in the policy proposal threshold (from 51%), I’m surprised that you view this as “utopian”. Pragmatic conservatism and utopianism are generally seen as diametrically opposed. Our SP is not a System of Perfection, just a way of making politics more democratic.

    LikeLike
49. 

    Alex Kovner, on January 13, 2022 at 3:20 am said:

    Liam:> Were I to go through all the failings of superminority…

    Ok, no point in throwing mud. Let’s just accept the fact that we are on different planets. I won’t comment on your posts again.

    LikeLike
50. 

    keithsutherland, on January 13, 2022 at 7:13 am said:

    This outburst of mudslinging raises some important points that are pertinent to the design of deliberative assemblies, but this merits its own post (to folllow).

    LikeLike
51. 

    Hubertus Hofkirchner, on January 13, 2022 at 8:27 am said:

    I know it’s hard not to refer to one’s own ideas and I have made this mistake myself often enough but this recent “mudslinging” is triggered by what can be seen as pushy marketing behaviour.

    I for one find it quite frustrating when – instead of a focused argument on the specific issue which a blog’s author brought up – references to full pet projects are thrown in (be it Paul’s “Code” or Keith’s “SP”) .

    Likewise “You misunderstand …” (insert pet project) is an illicit killer argument. Why should anyone here have to read a whole work when its author seems unable, unwilling or unmotivated to comment concisely with what is necessary and sufficient to a specific issue at hand?

    I humbly suggest that we all refrain from such references in our debates and stick to specific, issue-related arguments when discussing.

    LikeLiked by 1 person
52. 

    keithsutherland, on January 13, 2022 at 9:12 am said:

    Fair point Hubertus, but given that what is at stake is who should have the right to appoint expert witnesses, it’s hard to see how that can be resolved without reference to democratic principles. I don’t think a reference to another blog post can be dismissed as “marketing”, it’s just indicative of a fundamental disagreement regarding the role of election and sortition in the implementation of democracy. As to who is “right” about this, I don’t anticipate agreement between the protagonists any time soon, if ever!

    LikeLike
53. 

    Hubertus Hofkirchner, on January 13, 2022 at 9:15 am said:

    Keith> My intention was not to dismiss, it was to accuse. Now we need a judge.

    LikeLike
54. 

    keithsutherland, on January 13, 2022 at 9:19 am said:

    Yes indeed, but it should be a jury, as we are partisans.

    LikeLike
55. 

    keithsutherland, on January 13, 2022 at 9:50 am said:

    PS I don’t think it’s possible to have an informed view on any of the specific issues without a broader understanding on the topic (I still think Pitkin’s book should be required reading). It would be a bit like thinking you understand quantum theory on the basis of reading an article in Scientific American.

    LikeLike
56. 

    Hubertus Hofkirchner, on January 13, 2022 at 1:11 pm said:

    Keith:> Reference to Pitkin or any other authorities or classics is not the issue. Pushy marketing of own writings is.

    LikeLike
57. 

    keithsutherland, on January 13, 2022 at 1:18 pm said:

    But the PS is all Alex’s work — I only reference it because it fits so well with my own perspective on the (limited) role of sortition. Although we hope to bring out a joint book, the only collaboration we have had to date is an unpublished conference paper (that I didn’t cite). I imagine Liam would have been equally rude about Alex and myself if we had never heard of each other.

    LikeLike
58. 

    keithsutherland, on January 13, 2022 at 1:20 pm said:

    Of course Alex isn’t an authority and his blog is hardly a classic monograph, but I hope we are not averse to innovations in this field.

    LikeLike
59. 

    paulnollen, on January 13, 2022 at 1:36 pm said:

    Thanks Hubertus

    LikeLiked by 1 person
60. 

    Liam Jones, on January 13, 2022 at 5:51 pm said:

    Ok, no point in throwing mud.

    This outburst of mudslinging

    How interesting. What transgression have I committed to earn this accusation? Going by Alex’s quote, I said I had a lot of criticisms of the superminority principle. And I made a gesture about how Keith’s way of speaking looks rather utopian.

    Truly horrifying stuff, I know.

    But if this so dreadful and vulgar, what can we say of the primary accuser, Alex, and his fawning sycophant, Keith? The former has offered accusation of “looking for pink unicorns”, and the latter regularly comes up with underhanded insinuations. Mild, yes, but no better than a prod about utopian thinking. They’re both hypocrites. Judging by their posts, neither of them have any real concern for politeness or engaging the argument rather than the person.

    So what’s behind this accusation? I can’t say for certain, but I have my suspicions.

    Notice that giant post I made above in reply to Alex? It’s long because it’s full of reasoning. I go into to depth because I want to show everyone why I think he’s mistaken. And because I want to give him (and everyone else) the chance to engage with my arguments.

    Now look at Alex’s reply. It’s not full of reasoning. Neither is Keith’s. They consist of nothing but empty assertions, repetition, and accusations. This is important.

    So let’s talk about reasoning.

    Reasoned argument, notoriously, doesn’t lead us to unfailingly to an objective truth. (Except, sometimes, maybe, in mathematics.) Of course not. There are too many value judgements, empirical uncertainties, and background assumptions. But it can clarify disagreements. It can help us deepen our understanding. It can lead us to consider difficulties we hadn’t before. It can show us our assumptions. And, occasionally, when pride and ego don’t get in the way, it can even convince us. This is all in line with that Mercier & Landemore paper Keith shared earlier.

    And what is a reasoned argument, really? It’s just explaining why you believe what you do. It’s saying “You’re wrong because x, y and x …” instead of just “You’re wrong.” And if someone else does the same, it’s giving them to courtesy of reading their argument and trying to show why it doesn’t work.

    Does being mean corrode reasoned argument? A little bit, sure. It might hurt some feelings, which makes open exchange a bit more difficult. But, crucially, it doesn’t stop the process of argument itself. You can explain your reasons and throw an insult in there too. There’s more to human interaction than the cold blades of logic, after all.

    What absolutely does kill debate, however, is refusing to engage. If someone says “I’m right because of x, y, and z,” and you ignore x, y and z, the debate goes no further. No progress can be made. And if you don’t engage, you don’t have to think.

    Of course, not by itself is rather obvious. So the people who do this often try to distract from what they’re doing. They might say, “You don’t understand.” At which point the debate stops. They might repeat their previous claims. At which point the debate stops. They might say, “You’re slinging mud” At which points – guess what? – the debate stops. That way, they can make it look like they’re having a discussion, when really they’re just avoiding having to actually defend their ideas.

    There are all manner of things Alex could have said in response to my last post. He could have clarified what he meant by impartiality. He could have explained why it’s important. He could have laid out precisely how aggregating opinions of juries generates this outcome. But he did none of these things. Instead, he accused me of throwing mud. And Keith, like a slobbering faithful hound, picked up that diversionary tactic and ran with it. He’s even seen it as a justification for his own views. Talk about confirmation bias.

    Now, I do play hardball. I do like to throw out a sharp-edged comment here and there if my opposition isn’t playing nice either. And when I do it, I do it openly. And because I’m not a crybully like Keith, I don’t pretend I’m the completely innocent victim. But I also lay out my reasons for believing what I do. Alex says he won’t reply to my posts anymore, which is rather sad, really, because every argument I make here is an invitation for you to challenge me, and show me what I’ve missed. We might not ever reach agreement, but we can still enrich our understanding.

    Because we’re not on different planets. We’re on the same planet, of the same species, of pretty much the same cultural background. Communication, though imperfect, is possible if we choose it.

    Now, personally, I’d much rather hear what Hubertus has to say about how to stop the elite manipulating sortition. I’d much rather rise to Yoram’s challenge about democratic schools in the other blog. But instead I feel obliged to explain all this because these two pipsqueaks would rather call me a meanie than justify their claims.

    LikeLiked by 1 person
61. 

    keithsutherland, on January 13, 2022 at 7:01 pm said:

    The interesting point about this farrago is what it says regarding the (im)possibility of deliberative democracy. But I explain that in my forthcoming post Dikastic Thorubos. Watch this space.

    PS being referred to as “Alex’s slobbering hound” reminds me of one of our authors who gave a review of one of his books pride of place on his personal website. The book, The Snake That Swallowed its Tail was dismissed as “a brainless slab of Leftist bigotry”. And I like to quote Sir Bernard Crick’s comment on my first book: The Rape of the Constitution? — “Incremental totalitarianism indeed! Such partisan exaggeration defames the memory of the dead.” So I’m afraid Liam’s harrumphing is just water off a duck’s back, but I do worry that this “hardball” turns off potential readers of this blog.

    LikeLike
62. 

    Liam Jones, on January 13, 2022 at 8:18 pm said:

    A perfect demonstration of the pathology I talked about, Keith. An entire paragraph devoted to an offhand swipe, and not a single word to explain where I’m mistaken.

    LikeLiked by 1 person
63. 

    Yoram Gat, on January 13, 2022 at 8:48 pm said:

    I of course agree completely with Liam. It is important to note – as I have many times (only to be of course ignored by Sutherland) – that Sutherland’s pretense that the reason for the sharp tones in the exchanges with him is about substantive disagreements is false, a lie.

    The reason for the sharp tone is that Sutherland shows – as Liam notes – absolutely no respect for his interlocutors or for any minimal standard of rational discussion. Sutherland is simply a shameless liar – repeating his baseless assertions over and over with minor variations, ignoring any counterarguments, and making absolutely no effort at intellectual honesty. Substantive differences I have with many of the commenters and posters here. This is very different from confronting a habitual liar.

    LikeLiked by 1 person
64. 

    Hubertus Hofkirchner, on January 13, 2022 at 9:08 pm said:

    Meanwhile some unbelievable data about Austrian Climate Assembly, just for the record:

    2,000 Austrian random citizens were contacted by the Austrian Statistics office. So far so good.

    But now this: Only if they were interested in the climate topic and able to participate in the assembly, they ware invited to fill in a questionnaire and go into the lottery. Only 145 citizens did.

    Then the government also imposed a Corona-vaccination obligation (or previous infection) on them. General vaccination rate stands at 75% so that gives as a lottery of 108 people to win 100 seats.

    Not to forget, they also need a few spare citizens as some may drop out.

    Now that’s some lottery, right?

    LikeLiked by 1 person
65. 

    keithsutherland, on January 14, 2022 at 6:34 am said:

    Hubertus,

    Thanks for brining this to light — an outrageously anti-democratic development that risks causing serious damage to the sortition movement.

    LikeLike
66. 

    Hubertus Hofkirchner, on January 14, 2022 at 8:00 am said:

    Thanks Keith, may I quote you on this?

    LikeLike
67. 

    keithsutherland, on January 14, 2022 at 11:18 am said:

    Please do!

    LikeLike
68. 

    paulnollen, on January 14, 2022 at 2:53 pm said:

    The key question for me is in what circumstances can we allow a diversion from the optimum. The Bertelsmann Stiftung answered (to the application of my pet project, the evaluation grid): ” The question should be answered and justified on a project-specific basis. ”
    Fine, but no justification is given until now. Nevertheless I think there are examples where voluntary and non representative (statistically and descriptive) are acceptable. Maybe we must discuss them here somewhere. I think about the military example in France studied by Dimitri Courant. It is clear that a voluntary sortition of the population (eligible to vote for example) gives a total different outcome than a compulsory participation. And small samples are not representative statistically and descriptive. But when doesn’t it need to be? Is the military example a good one where it is not necessary?

    LikeLike
69. 

    keithsutherland, on January 14, 2022 at 3:18 pm said:

    The problem is we use a single term (sortition) to refer to any lottery, irrespective of its intended use. And this has serious consequences with respect to the legitimacy of representative claims. Presumably the Austrian government claims that their sampling process is democratically representative on the strength of the original 20,000 lots drawn. But this is clearly not true, as indicated by Hubertus’ comment from yesterday. Jane Mansbridge has called the Deliberative Poll the “gold standard” of random selection, so we might be better to focus on that. Although different projects will require different standards, while we are trying to get the sortition process accepted it might be better to go for gold, rather than let the currency of the sortition movement to be contaminated by a variety of different standards for different applications. I don’t think the nuance will hold up, given the level of scepticism and ignorance that we are facing. Once the gold standard established then maybe we should introduce different terms to refer to different applications.

    LikeLike
70. 

    paulnollen, on January 14, 2022 at 5:20 pm said:

    So far the only thing we can do (and must do in my view) is insist on the application of the “Bertelsmann Stiftung” answer: “The question should be answered and justified on a project-specific basis. ”
    I think at least we can write the right questions. I started with the “size of the conference room” as a justification.

    << Hello xxx

    I read with great interest your paper xxx

    But sadly, I could not find the answers I am looking for.

    Many initiatives who are using sortition (and “manipulated” or “corrected” sortition * ) in some way, are claiming to be “representative” or a “mirror of society” and so on, but they lack to provide any explanation why this is the case. They come up with 15 citizens, 45, 100, 500 or whatever fits in the conference room available I suppose. Sometimes they claim “statistical representativeness” with 1000 to 1200 participants (or more) but that seems not always necessary. If that number is to big to fit in the conference room, I suppose you can take any approriate number that fits and proclaim that it is maybe not (statistical) “representative” or “descriptive representative” , but it has “maximum diversity”, and it is, nevertheless, a “mini public” mirroring society as a whole and, as a conclusion, is “representative”. Of course you can use some questionnaires and computer algorithms to help you out, or some “selection” or whatever fits you. But we don’t have to worry, it has always “good results” (normative I suppose).

    That lack of motivation was the main reason we developed our own “Evaluation grid” (and Code of Good Practice) and we start using it as the paper in attachment shows.

    We hope that our evaluation is not perceived as criticism only but as a quest for answers.

    * Through a mix of random selection and targeted recruitment, 1000 citizens were invited to participate in a deliberative event in Brussels. http://www.g1000.org/en/method_phase_2.php

    <<

    Of course we can ask similar questions for the other criteria, the use of volunteers, the appointment of the experts, the rules for the surrounding commitees and private companies, and so on. We have to put some presure on them for answers.
    (for Hubertus : I also mentioned a computer program for compensation of the code)

    LikeLike
71. 

    keithsutherland, on January 14, 2022 at 6:43 pm said:

    Surely to merit the term “minipublic” or “minipopulus” the sample should act in the way that the target population would act under the same conditions? Obviously that’s impossible to know (otherwise there would be no need for sampling), but it is possible to compare two samples of the same population to test the hypothesis. If the samples didn’t cohere then there would clearly be a need to alter the selection criteria (size, mandation, deliberative style etc), but these are empirical considerations (“hacks” in Nick’s parlance). And it has the merit of being simple and easily grasped.

    >I suppose you can take any appropriate number that fits and proclaim that it is maybe not (statistical) “representative” or “descriptive representative” , but it has “maximum diversity”, and it is, nevertheless, a “mini public” mirroring society as a whole and, as a conclusion, is “representative”.

    How can it “mirror” society and yet not “describe” it (statistically)? Just “looking like America” would be a shallow confidence trick — especially as the “diversity” criteria are whatever crude social metrics happen to be in fashion. Note also that the notion of statistical representation is only of interest in relation to democratic majoritarianism (or whatever the decision rule), whereas “diversity” is little more than a New Left shibboleth.

    LikeLike
72. 

    Liam Jones, on January 14, 2022 at 8:35 pm said:

    Paul,

    I think the big worry here is we’re facing the sortition analogue of voter suppression. Who are these people who aren’t answering? Single mothers with no one to
    provide childcare? Shift workers who can’t spare the time? Carers for elderly family? People with social anxiety who would like to contribute but don’t feel up to it? If so, it’s a very serious problem. Just like voter suppression, this effect will tend to harm disadvantaged people the most. And so it plays into the hands of the powerful who want a cloak of democratic legitimacy.

    Of course we can try and fix that with sampling. But that’s always going to be a crude fix, because the filters are still in place. Let’s say we stratify to get the right proportion of people of colour. Those selected might still be wealthier on average. So we have to fix that too. If stratify for gender balance, then we need to deal with transgender people and nonbinary people. Unless we can account ahead of time for every possible way people might be prevented from joining, we end up with a source of systemic bias. And what’s worse, the sort of traits (whatever they may be) that prevent someone from joining an assembly will also tend to erase them from the statistics that stratification needs.

    To a degree we have to work with a project-specific basis. While sortition remains a side project of governments, it won’t command the resources to get a proper sample. And a stratified sample might be better than a bunch of bureaucrats and minor politicians making the decision. But it’s always going to be worse than a proper sample, and in ways that can be dangerous.

    LikeLike
73. 

    keithsutherland, on January 15, 2022 at 7:57 am said:

    Liam,

    You are right to point out the flaws with stratified sampling, as there is no way of correcting for an unknown unknown. If we are going to take sortition seriously then we need to be clear that participation is a civic obligation, rather than a political right. In Classical Athens participation in democratic politics was a corollary of taking up arms to defend the polis. This form of civic virtue was lost a long time ago and won’t return any time soon (this is the weakest aspect of Jeff Miller’s new book). It’s interesting to note that although voter suppression is the (unacknowledged) goal of many Republican legislatures, it’s conservatives like me who are arguing the case for quasi-mandatory participation in allotted legislatures, as we don’t want to see them dominated by activists (as in the Austrian example).

    LikeLike
74. 

    paulnollen, on January 15, 2022 at 9:50 am said:

    Let’s look at the G1000 (Belgium) publications:

    >>
    A myriad of techniques:
    There is not just one unique form of deliberative democracy. There is a multitude of methods and techniques that can be used on a small or large scale. For a review, see: The Deliberative Democracy Handbook: Strategies for Effective Civic Engagement in the Twenty-First Century van John Gastil en Peter Levine (Jossey-Bass, 2005).

    The G1000 mainly uses the method known as the construction of a ‘mini-public’. Instead of relying on existing institutions or civil society organisations, a new group of citizen is assembled. The group has to remain small enough to allow for deliberation but has to be large enough to comprise a large and diverse set of opinions on the matter at hand.

    Is important to stress that in the case of the G1000 participants were randomly selected citizens. In many other cases citizens are involved on the basis of self-selection (they declare to want to participate in the deliberative event) or because they have some expertise, or because they are stakeholders. Through random selection the G1000 wanted to reach out to broader sections of the population, but that does not mean that deliberative democracy always has to rely on random participation. Instances of self-selection of participants have proved useful, for example, in the successful Participatory Budgeting processes that started in 1989 in Porto Alegre in Brazil (and that have spread to many other Latin-American cities) or in the Chicago Alternative Policing Strategy that has been credited for a spectacular reduction in crime in that city in the 1990s.

    Below are some well known techniques that are based on the idea of a ‘mini-public’ and that try to attract participants from as diverse backgrounds as possible:

    Citizens’ Jury : Has been used for the first time in the US in 1974. A citizen jury consists of a group of 18 to 24 randomly selected citizens. They have to formulate a policy advice after a relatively brief period (4 to 7 days) of testimonies of experts on the matter.

    Consensus Conference : Has been developed by the Danish Board of Technology in 1987. It consists of a group of about 15 people that gather for two weekends to decide on the agenda for a public forum that in turn lasts another four days.

    Deliberative Poll : A method introduced in 1988 by James S. Fishkin. A randomly selected group of 250 to 500 citizens listen to a number of experts. Following this they deliberate in groups of 15 and subsequently re-engage in a debate with the experts. Before and after this process surveys are used to measure and map the change in attitudes among the participants.

    The ‘21st Century Town Meeting’: Developed by AmericaSpeaks. Usually this method makes use of a large meeting of about 500 to 5000 participants. Deliberation and discussion happens in groups of 10 to 12 people. The information from the groups is collected by a network of computers and is communicated back to all the participants
    <>Through a mix of random selection and targeted recruitment, 1000 citizens were invited to participate in a deliberative event in Brussels. <<

    From the beginning they use "self selection" AND "random sampling". In order to compensate for the flaws in their system they add 10 % of the participants by "selective recruitement" . Ofcourse this "selective recruitement" is not necessary if your sample is "representative" for the targeted population.

    LikeLike
75. 

    keithsutherland, on January 15, 2022 at 10:05 am said:

    Paul,

    I think the goal of this forum is to make democracy deliberative, rather than to democratise deliberation (the referent of the former is the entire demos, rather than group per se). If so this will rule out many (most?) of your alternatives, especially given the need for both technical/theoretical and perceived democratic legitimacy.

    LikeLike
76. 

    paulnollen, on January 15, 2022 at 10:24 am said:

    My feeling is that there is an evolution towards a “sortition technocracy” that is even further away from democracy and legitimacy then the “electoral aristocratic” system. They are only hiding behind a “sortition” element in the procedure to disguise the technocratic nature of their project.
    (William Henry Smyth, a California engineer, is usually credited with inventing the word technocracy in 1919 to describe “the rule of the people made effective through the agency of their servants, the scientists and engineers”, )

    LikeLiked by 1 person
77. 

    Yoram Gat, on January 15, 2022 at 11:04 am said:

    For the allotted to be self-controlled rather than controlled by the organizers, the system must be self-designed, in the sense that allotted bodies are empowered to redesign any aspect of the system. (Some basic principles – like the one just asserted – can be constitutionally mandated so that they may not be changed without a referendum, but those principles will necessarily be abstract and their interpretation must, again, be in the hands of allotted bodies.)

    As long as some important parts of the system are set a-priori “from the outside” (i.e., by some non-democratic mechanism) the system as a whole would be suspect of being controlled from the outside (and reasonably so).

    LikeLike
78. 

    Hubertus Hofkirchner, on January 17, 2022 at 2:19 pm said:

    Further information emerges: Just the above “random” selection process for citizens by the Statistics Office cost 41.295 Euros, this involved sending 2000 mass mail letters and handling 145 responses. I wonder what Brett Hennig’s Sortition Foundation has to say on this.
    charges.

    The cost for the organisation will be subject of an upcoming parliamentary enquiry.

    @Yoram: “the allotted to be self-controlled rather than controlled by the organizers”

    I agree in general with Yoram’s statement but fear that a “constitutional mandate” is quite unrealistic in the near and medium term. Maybe in 30 years out.

    Maybe, because I am not really sure a constitutional rule – other than in the most general form – is possible or desirable.

    As our discussion shows once again, while this forum agrees on the separation of expert selection from government influence, there are several possible ways to organise such a process. My procedure with the steps listed above is informed by 20 years of research into collective intelligence, but I am fully aware that maybe Paul or Alex’s could perform better in practice. (Low chance though.) Point is: we will only know after trying, studying and comparing.

    Hence, it is more realistic and consistent with an open innovation approach that sortition parties be formed, design such systems and improve them over time, using expert knowledge and the wisdom of crowds as appropriate.

    The legitimacy of each of these parties and their proposed systems and methods will be determined democratically by their respective vote shares at general elections, thus promoting progress with a healthy competition of sortition methodologies.

    LikeLike
79. 

    keithsutherland, on January 17, 2022 at 2:31 pm said:

    >maybe Paul or Alex’s [expert selection proposal] could perform better in practice. (Low chance though.)

    Bear in mind that Alex’s approach involves little more than a change to the proposal threshold of existing legislatures. And the SP is a democratic rather than an epistemic innovation, in that there is no distinction made between advocates and experts (or, more accurately, proposing parties may recruit any “expert” witnesses that they think will support their argument). If there is any epistemic value to the SP, it will be over the long run, as parties proposing unworkable legislation will be de-selected.

    LikeLiked by 1 person

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