The representativity of a random sample: the need for mandatory participation

The unexpected Conservative majority in the 2015 UK general election has led to considerable agonising in the polling industry. Why were the polls — which consistently predicted a hung parliament, or even a Labour victory — so wrong? A polling industry enquiry has come to some interim conclusions, here paraphrased by the BBC political editor, Laura Kuennsberg:

Pollsters didn’t ask enough of the right people how they planned to vote. Proportionately they asked too many likely Labour voters, and not enough likely Conservatives.

Nobody is suggesting that this bias was intentional — it was the accidental by-product of the polling methodology which, rather than drawing a random sample and then knocking on doors (the gold standard, but very expensive), relied heavily on telephone and online surveys. The problem with phone surveys, according to Martin Boon from ICM, is that out of 30,000 random numbers only around generally 2,000 agree to participate. This sample is anything but representative:

Adam Drummond, from Opinium, suggested that the people who agreed to be interviewed were so politically enthusiastic that the would even vote in elections for the European Parliament. . . This seems like a Groucho Marx problem — people answering the phone and agreeing to answer questions about their political opinions automatically means that they are too politically engaged to be representative.

Online polling is even worse:

Online polls are answered by a database of volunteers who have signed up to be on a panel and who the company knows a lot about. When the company is commissioned to do a poll it can be sure that the people it is asking have the same features as the whole population in — for example, the proportion of men and women, or the age profile or income distribution.

However what stratified sampling does not do is to control for the level of political engagement, so

The online polls came out with much the same inaccuracies as the phone ones. Does people’s willingness to be on a panel automatically make them unrepresentative?

If the 2015 election polling debacle is anything to go by, then the answer is a clear ‘yes’ — the methodology underestimated support for the Conservative Party, the suggestion being that conservatives (in the UK) are less politically engaged than progressives (this is less true in the US, on account of Tea Party, NRA and conservative evangelical Christian activists). A recent book manuscript submitted to my publishing company by the conservative political theorist Peter King, entitled The Grazing Herd: In Praise of Complacency argues that those of a conservative inclination (the ‘silent majority’) are least likely to blow their own trumpets. It’s quite hard to outline a positive theoretical foundation for conservatism as the devil has all the good tunes. Conservatives have always been referred to as members of the stupid party, so it’s tempting to remain in the closet (forgive this stream of mixed metaphors).

The ongoing debate in the polling industry is highly relevant to those of us advocating the adoption of random selection in political decision making. The selection process for the 2004 British Columbia constitutional forum invited 23,014 randomly-selected persons to participate, but only 964 (4% of the sample) turned up to the selection meeting. If the pollsters are right in their analysis, then this sample would be skewed in favour of activists and others with a high level of political engagement. Those of a conservative disposition would be underrepresented, so it is unsurprising that the proposals for constitutional change were rejected in the subsequent public referendum. If a sample is to be statistically accurate, then participation has to be mandatory, as in the case of jury service.

The same arguments apply to the speech acts of those included in the sample — they are likely to be biased in favour of those with a high level of political engagement, so the silent majority (by definition) would be underrepresented. As such the only activity that a randomly-selected sample can engage in without sacrificing its descriptive representativity is to listen to balanced advocacy and then determine the outcome via the secret vote.

62 Responses

  1. We need sound public opinion, the result of genuine deliberation about questions of policy. The present preoccupation with getting a verdict from people who have never though seiously about politics is absurd


  2. This is all true enough. However, those who participate in elections are also the more politically active. There is, in fact, no way around this problem of “non-representativeness” unless EVERYONE is FORCED to participate in whatever system we have or set up, which creates its own problems. Personally, it seems to me that random selection (especially stratified sampling) provides a BETTER range of perspectives than the “whoever comes” logic of elections (at least in the US).


  3. I agree that the only way to ensure an *initially* representative sampling is through mandatory participation. The flip side is some subset of the sampling will be pissed off at having been selected for service. I will not hazard a guess as to how this will affect the results, if at all.

    In any case, I’ve come to believe that the sampling needs to serve long enough to where the average member can be expected have a handle on all the issues that happen to be up in the air. This might mean terms too long for mandatory service. Maybe we could get good enough results with stratification. Perhaps much less than a year is good enough. I don’t know. If we can’t strike a reasonable balance I might have to go back to favoring bicameralism.


  4. John,

    Most trial jurors will not have thought seriously about the matters that they are charged with determining prior to their jury allotment and yet we consider them competent to judge the evidence presented to them by both parties to the dispute. If so, then why should that not be possible in the case of legislative trials?


    I thought your model was (effectively) bicameral, with the politicians providing the arguments and the allotted members determining the outcome? If so, and assuming issues of joined-up governance and fiscal responsibility are dealt with by other mechanisms (admittedly a big if), why do jurors need to have a handle on all the issues? If jury service is not mandatory, then the jury will not be a representative one. According to the (UK) Labour Party report released yesterday, the lack of representativity of the opinion-poll samplings cost them the 2015 election, so this has to be the sine qua non for sortition-based approaches to democratic lawmaking.


  5. Statistical representativity of the allotted body is a tool – not an objective. The objective is democracy – equality of political power. Mandatory service will not make political power more equal – it will simply hide any inequalities behind the formality of the statistical makeup of the body. It will also be very detrimental to the prestige of the body and to its morale. It is a horrible idea.

    The obvious solution is to make service attractive enough so only few will opt out. If service carries a nice salary, considerable prestige and true political power then it would be surprising if many would opt out. In any case, this is something that can be empirically verified. If a large percentage of those offered allotted slots turned them down, then something would be wrong and the causes would need to be investigated and fixed.


  6. Yoram,

    It doesn’t matter much whether it’s the stick or the carrot that works — so long as the vast majority of those allotted show up, bribery and flattery would do just fine. I agree that this issue can be resolved empirically.

    >The objective is democracy – equality of political power.

    Agreed, but the question is how to ensure the equality of those not included in the allotment. How is this possible without ensuring the fidelity of the statistical representation?


  7. >with the politicians providing the arguments and the allotted members determining the outcome?

    That’s what I refer to as unicameralism. There’s one assembly, one majority (the preferences of the sampling), with multiple types of members with different powers. There are plenty of mixed member assemblies in the modern world, both in terms of mode of selection and in powers. Even the US House of Representatives has non-voting members who can present arguments to the voting members. Mixed FPTP/list systems are common, as are parliamentary systems in developing countries where the government has the power to appoint a few members of the lower house to help shore up fragile majorities and bring in would-be ministers.

    In (what I refer to as) bicameralism the elected officials would form a separate majority which would collectively vote to introduce proposals for a vote by the sampling. Both the elected majority and the sampling would have to approve all legislation, not counting for the possibility of an initiative mechanism. In this case we can expect the proposals brought for a vote to generally make sense in context, save for the occasional oddball initiative proposal. The sampling serves as a check on the elected body.

    If, on the other hand, the proposals to reach a vote are diverse, then there’s no reason to believe proposals that can get a majority in a sampling will be compatible with each other. The role of making sense of the cacophony falls to the sampling. I don’t see how they can accomplish this if they don’t have a handle on everything going on. If they do, then everything should be fine. Their judgment on big-picture things should be no worse and no less representative than it is on individual items. But they would need an informed big-picture view for this to work out, I’d expect.


  8. Naomi,

    Thanks for the clarification. In the unicameral case my proposal is that the compatibility issue is a matter for the (appointed) executive. But I accept this is undemocratic, so perhaps your bicameral suggestion is the best option. It would also be less likely to put elected politicians’ noses out of joint and would fall in with the current plethora of proposals for an allotted second chamber. Sortition fundamentalists have no reason to dislike it any more than their current objections to the retention of election for the advocacy role. And even if they have to hold their noses, it would offer a very valuable foot in the door.


  9. My multi-body design tries to eat this cake and have it too, by having a Review mini-public with relatively long terms and thus likely lower participation rates (and representativity), and separate short-duration mini-publics that are convened to simply approve or reject a single bill proposed by the Review body, such that service on the jury would be closer to mandatory but less burdensome and thus more representative.

    While I don’t disagree that handsome compensation, honor, etc. are a good idea, there are still certain sorts of people who would consistently decline (introverts, very humble, young mothers, owners of startup businesses, et. al.)…and those people and their perspectives would be very beneficial to the diversity and quality of decision-making. I favor some sort of deferral option, where a person can defer service on a jury, but must eventually fulfill this civic obligation once called.

    A bigger question is whether democracy necessarily means representative of ALL the people, or if it is still democratic if some people (like the idiotes of Athens) are simply UNwilling to serve. Is it undemocratic if 1% refuse to serve? Is it only undemocratic when fewer than half will serve? I think of an isolated society of ship-wrecked survivors forming a democratic society. If a few of them willingly say they will simply go along with what everbody else decides… Is that democratic? It seems to me that the key is that ALL society-imposed obstacles to service must be ameliorated so that nobody declines due to their class, race, status, etc…. but I am unclear how to think about democracy if a huge percentage simply want no part of civics.


  10. Terry,

    The take-home message of this thread is that modern-day idiotes are likely to be supporters of the Conservative (aka stupid) Party, who are, on the whole, happy for things to muddle on as they are. As such an allotted parliament that was not based on mandatory participation would be as unrepresentative of the silent majority as was the opinion polling before the 2015 election. The same principle would apply if speech acts were included within the remit of the assembly (as the silent majority would not, by definition, have very much to say). This presents a serious problem for anyone who really believes in democracy. As to why Yoram believes this would be unproblematic we can only guess as he has not deigned to answer my question. I can only assume that his reasons are partisan (he is in favour of political institutions that are likely to promote radical and “progressive” policies [or “better” to use Andre’s terminology]).


  11. Keith,

    I am curious about your opinion of the shipwreck hypothetical society…If one person says “what ever” and happily defer to the will of the remaining members… is that still democracy? What if it is more than one? more than 50%? I don’t have a firm view on this myself.


  12. Terry,

    I don’t think it’s helpful to apply direct-democratic standards to large-scale (representative) democracy. In your example each castaway decides for herself whether or not to participate, but this would not be the case for the silent majority in a kleristocratic regime as their underrepresentation is a consequence of the decision of their proxies not to participate. So although your question is an interesting one, it’s orthogonal to this post, so maybe you should develop it fully and start a new thread. My point is that a non-mandatory kleristocratic regime would be undemocratic and, judging from the evidence cited in this post, would underrepresent the perspectives and interests of those of a conservative disposition.


  13. Hmm. Maybe this is not so problematic after all. Those who would choose to abstain in the population at-large are represented statistically by the subset of the sampling who chose to abstain. If you would not have abstained then you are part of the subset of the population which both shares your political views and who choose (or would have chosen) to participate. No one who would have chosen to participate loses representation due to the abstentions. The abstentions *are* representing the would-be abstainers by not participating.


  14. Naomi,

    That may be formally (mathematically) true but it is false from the perspective of substantive political representation. Under our present (electoral) arrangements people who are happy with the status quo can vote for conservative politicians who will represent their preferences and interests by resisting the initiatives of progressive reformers, and these people (ideotes) will be under-represented in a kleristocracy, as their proxies are more likely to abstain. As a a self-acknowledged conservative I would be deeply unhappy with such an undemocratic constitution.


  15. If a reform proposal is perceived as advantaging either the left or the right *over the current arrangement* it will rightly be opposed by something resembling half the country. Effort should be made to match the current balance. It’s more work to take part in a sampling (compared with voting in an election) but the payoff is vastly greater still. I’m not sure why conservatives would necessarily be at a disadvantage compared with the present system. After all, a desire to protect the status quo is what motivates them to get out and vote right now. Right? That motivation will still be there. But the effort-to-impact ratio would be greatly improved.

    On the other hand, mandatory participation would definitely give advantage to conservatives over the current system, at least at the moment in the UK, which would result in staunch opposition from the other half of the political spectrum. Both sides need to be on board.


  16. Naomi,

    If the selection criterion for the minipublic is accurate descriptive representation, and mandatory participation is necessary to achieve this, then there’s really nothing to argue about. Yoram denies that accurate descriptive representation is a sine qua non, but has failed to explain (non-tautologically) why, so I can only assume that his reasons are partisan, and your second paragraph adds support to that supposition.

    BTW mandatory participation would not advantage conservatives over the current system as they can currently elect conservative representatives (as they did in 2015, to the consternation of the pollsters). The problem is that a move to non-mandatory sortition would disadvantage them over the current system.


  17. Full participation would be greatly preferable, just as full participation in an election is preferable. In both cases low participation is (or certainly would be) de-legitimizing. I’m not convinced a comparable degree of abstention would prove more objectionable in a sampling given that those who abstain are representative of the subset of the population who would abstain from the process if given the choice.

    If I follow your argument, conservatives don’t want anything in particular, and so have less motivation to participate. If so… why participate now? Voting takes some measure of effort. Presumably the desire to protect the status quo provides enough motivation to take action. As you point out, the polls from the last election underestimated conservative turnout. The effort-to-impact ratio is much worse in a poll than it is in an election. Similarly, the effort-to-impact ratio is much better in a sampling than it is in an election. Why wouldn’t the same trend hold? The difference between voting for a representative and participating in a sampling is the greater effort required and the greater impact of one’s effort, which counter each other quite favorably. Why would different rules and trends apply?

    Mndatory participation in the US would almost certainly give advantage to the left at the moment. The most liberal voters are young adults who have very low turnout rates. The most conservative are retirees who are possibly the most reliable voting block.


  18. Naomi,

    Your point about those who decline to participate in a mini-public statistically representing (properly) the percentage and sort of people who would decline to participate if THEY had been drawn, is interesting. It takes me back to my shipwreck scenario. (and Keith, the principle IS the same regarless of whether we are talking about a direct democracy or a representative one)… How should we feel when a portion of the citizenry opts out (of direct democracy, voting in an election, or serving on a mini-public). Is Australia with compulsory voting on to something…is Australia more democratic as a result? In short is democracy …rule by the people,… or of the people who choose to participate?


  19. Naomi,

    I agree that electoral systems may overrepresent seniors and underrepresent young adults (and certain ethnic minorities) but, notwithstanding the scepticism of philosophers and neuroscientists regarding the possibility of counter-causal free will, every citizen has a right to vote and can (in principle) freely choose whether or not to exercise that right. Not so for the vast majority of citizens in a kleristocracy who would be entirely reliant on their allotted proxies. I am disenfranchised to the extent that proxies like me fail to perform their civic duty. We all (with the exception of Yoram) agree that accurate descriptive representativity is essential in an allotted legislature, hence the need for quasi-mandatory participation and the proscription of any speech acts that adversely affect the aggregate representativity of the group. Bear in mind that 96% of the initial BC sortition declined to participate and this has to have a marked effect on the representativity of the sample, even if it’s not possible to say exactly what the effect is.

    I think your analysis of conservative psychology is overly rationalistic. Speaking for myself I vote in elections out of a sense of civic obligation and slam the phone down as soon as I hear a voice at the other end asking me to participate in an opinion survey. I suspect that I’m not entirely atypical as far as regards older conservative voters (I’m 64).


    While the idiotes problem may be common to both direct and electoral democracy, different considerations apply to proxy-based systems, for the reasons given above.


  20. Keith,

    “Under our present (electoral) arrangements people who are happy with the status quo can vote for conservative politicians who will represent their preferences and interests by resisting the initiatives of progressive reformers, and these people (ideotes) will be under-represented in a kleristocracy, as their proxies are more likely to abstain. As a a self-acknowledged conservative I would be deeply unhappy with such an undemocratic constitution.”

    The free-choice system I’ve been proposing solves this problem by letting everyone decide whether to exert her personal portion of national sovereignity (Rousseau) by voting or rather by having an equivalent chance of being sorted.


  21. Arturo,

    The sort of person that I’m referring to would not want to be “sorted” as you put it. If they won the lottery then they would have to become directly involved in politics, whereas expressing a ballot preference involves a much lower cost in terms of time and commitment. Idiotes, by definition, do not want to get involved — that’s why they vote for the stupid party.


  22. Keith,

    The beauty of free-choice is that they don’t need to be allotted is they don’t want to — they can keep on voting and rest assured that the expected value or espérance E[x] for their voice to be heard in the mixed parliament is roughly the same, no matter if they choose to vote for a political party or to be allotted.


  23. Arturo
    Curiously enough, this would, if Keith’s view on them is correct, give conservatives disproportionate power. They would largely be represented by disciplined parties. A larger portion of liberal voters would be represented by diffuse assortment of allotted members who would be poorly positioned to exert leverage in an assembly. This basic phenomenon is best illustrated when parties on either the right or the left are more fragmented in a parliamentary system. It’s often easier to maintain a single party ideological minority government than to put together a four party ideological majority coalition.

    Mixed member systems need to be designed with great care to avoid introducing exactly this sort of bias. If you want a mix of allotted members and elected members in the same body doing the same job… that’s fine. It’s not my first choice. But certainly, it is something which could be defended well, though I can’t recall anyone here doing so. Everyone should be represented by the elected subset and everyone should be represented by the allotted members in parallel to avoid leaving the balance of power up to chance.


  24. Perhaps “chance” was the wrong word. It would have been better to say, “irrelevant details,” or something similar.


  25. Arturo,

    As Immanuel Kant put it: “this may be true in theory but it does not apply in practice”, for the reasons that Naomi outlined. Political science is an empirical discipline which is interested in what works, rather than philosophical abstractions. The UK “constitution”, which simply evolved over time as a result of purely contingent factors (such as the inability of the early Hanoverian monarchs to speak good English) has done a reasonable job in preserving our freedoms, despite having no regard for abstractions like free choice. So we should be wary about rationally derived (but untested) constitutional initiatives. I include my own work in this category, but hope that Fishkin’s two decades of experiments and the empirical findings on the “wisdom of crowds” would underwrite the theoretical claims.


  26. Naomi,
    No question there is some merit to your reasoning, more militare, that a small disciplined group would carry the day over a larger but disorganised opponent. Still, I feel somehow unconvinced. What is true for the battlefield, is not necessarily so for all and every human business. Nor am I sure that the example of four or more fragmented parties, each with its own leaders (psychopaths by definition — or rather by virtue of the “survival of the fittest” mechanism), its own internal struggles, its own loyalties to powerful backers, etc., applies here. Political parties have strong incentives not to compromise, lest them be seen as losing face, while professional politicians, in the rare cases in which they cannot be bribed, are easy to lure with the offering of an executive position… There are many ways in which an active party can sow division and prevent an alternative majority from forming. But if we talk about individual citizens, none of this matters anymore. I for one am not able to anticipate what the discussion and voting dynamics in such a mixed parliament would look like, but I have the feeling that we could be in for a surprise. At least in the eyes of the audience, the contrast between self-serving politicians and ordinary people without an external agenda would be starker than ever before.

    I want to make it very clear that free choice is here just a means to an end and not a petition of principle. Same with the concept of a mixed assembly: I am not at all interested in it per se, as in Naomi’s mention of a system where everyone is simultaneously represented by the elected and the allotted subsets. I have no use for that. To me, the mixed parliament is just a consequence of the fact that the only feasible way to introduce sortition is to base our demand on the individual right to political participation. An unexpected outcome if you want, neither welcome not unwelcomed, but certainly not a sought feature. I had no prejudgments when I began to consider the problem of representation. You see, I am not a scholar. I am a man of action. A grey, boring public servant, agreed, but a man of action at heart. And a practical one at that. To the nicest theoretical design, I would prefer any time the one with a higher probability to be accepted. And the elephant in the room, in this forum as well as in the French-speaking discussion, is how to convince half plus one of the electorate to vote in favour of an untested system you want to impose on all of them. My guess is that it would be a cold day in hell before you overcome the resistance of the powers that be — not by force, mind you, simply by ridiculing the idea through the mass-media under their control. This leaves us with only one solution: an opt-in system that moves the discussion from a perspective of majority consensus (we collectively agree to change the system) to one of respect of the minorities (as Rosseau put it: et d’où cent qui veulent un maitre ont-ils le droit de voter pour dix qui n’en veulent point?).


  27. Arturo,

    I don’t share your pessimism that hell will need to freeze over before the powers that be consider sortition. What will alienate them is the argument routinely heard on this forum that sortition will put them out of a job. The bicameral solution that Naomi and myself have discussed — an elected lower house and an allotted senate — would not undermine the role of elected politicians. Our other proposal — farming out controversial decisions to one-off allotted bodies — would also be likely to receive the support of elected politicians. One of the findings that Fishkin reports in his last book is the extent to which politicians viewed DP experiments as a way of legitimising their own role (I’m referring to the DPs in Italy and China in particular).


  28. To be even more clear: to me, sortition is just a means and not an end in itself. I don’t need sortition to decide on gay marriage (with all due respect to both gay and married folk), I need sortition to decide on bank bailouts. Why? Because elected politicians will always bailout the banks, and allotted citizens will not.

    Are you saying they will accept an allotted senate capable of effectively stopping the bank bailouts decided by an elected lower house? I don’t think so.


  29. Arturo,

    I don’t share your cynicism, believing that the bailout was on account of a genuine fear of the consequences of the failure of the financial system on whole economies (including the jobs, mortgages and savings of “the people”). In the UK the chief architect of the bank bailout was Labour PM Gordon Brown, who most sober political analysts (i.e. non-Corbynistas) would categorise as being on the side of the people, notwithstanding the strategic necessity of eating prawn cocktails in the City of London and schmoozing Rupert Murdoch. If you want to win elections you need to appear financially competent and to secure the support of Sun readers.

    This isn’t the place to discuss substantive policies (on gay marriage, bank bailouts or whatever), but I speculate that ordinary citizens, properly informed on the macroeconomic implications of a failure of the financial system, might well have chosen the same policies. Politicians have more to fear from going against the ignorant prejudices of voters than they have to fear from accepting the decisions of a well-informed deliberative minipublic. If the policy goes wrong they can shift the blame to the popular assembly (as did Pericles — he may have offered defective advice, but the demos freely chose to accept it). Looks like a win-win to me (given that the main thing a politician wants to do is to secure her future career)

    Of course if you (like many people on this forum) simply take it as axiomatic that politicians are blindly following the preferences and interests of fellow members of the “elite” then you would conclude otherwise.


  30. This is not a place to discuss substantive policies; let’s keep it that way. Suffice to say that I disagree with what you just said in toto.
    As an economist, I can tell you that social sciences’ only laboratory to verify, refute, or validate a hypothesis is History itself. No controlled conditions, no replication, no nothing. Just History as it happens to happen. Which means that we cannot go look into a parallel universe and see how things would had gone had we let the banks fail.
    The best we can do is look at how things went elsewhere. Like in, I don’t know, Iceland. Not that Icelandic politicians deliberately decided to let banks fail. Temperatures in hell will be very close to 0ºK before an elected politician lets a bank fail out of his own volition. It’s only that Icelandic banks were bigger than TBTF (too big to fail). They were much bigger than that. They were indeed TBTS (too big to save). And they could not be saved, and they failed, all three of them.
    According to most sober Sun readers, Icelanders took the wrong path and should be in pretty bad shape by now. Certainly worse that we are after having chosen the right path and bailed our banks out. But History is a bitch, and fact is, Icelanders are much, much, much better off than we are.


  31. Agree this is not the place to discuss substantive issues.


  32. Arturo,

    I think I see things much as you do. Power grants nothing willingly. Democracy would have to be fought for, and sortition, if it would be a meaningful tool for democracy rather than a charade, would have to be imposed on the ruling elites.

    A good way to proceed, I believe, is by pushing for allotted bodies which have an anti-corruption role – setting regulations controlling the conduct of elected politicians and putting those regulations into practice.


  33. Arturo:> elected politicians will always bailout the banks, and allotted citizens will not. . . . social sciences’ only laboratory to verify, refute, or validate a hypothesis is History itself.

    As a matter of historical record, in 1995 the British Government allowed Barings Bank to collapse. Barings was the second oldest merchant bank in the world and had very strong Establishment links and close ties with the royal family from the time of George V. In 2008 the US Government allowed Lehman Brothers (the 4th largest investment bank in the US) to collapse, so it would appear that hell is indeed shrouded in a thick coating of ice.


  34. Yoram:> Power grants nothing willingly. Democracy would have to be fought for, and sortition, if it would be a meaningful tool for democracy rather than a charade, would have to be imposed on the ruling elites.

    Given Arturo’s recent appeal to history, where’s your evidence that this is how democratisation has occurred in the past? Most historians take at face value Herodotus’s claim that the Athenian demokratia was the accidental by-product of a dispute between aristocratic factions — I can’t offhand think of anyone much apart from Peter Euben who buys into Ober’s claim that there was an “Athenian revolution”. And it was Marxist historians who first claimed that the modern revolutions (French, American and English civil war) were thoroughly bourgeois (a dispassionate analysis of the Russian revolution might well come to the same conclusion). Turning specifically to English history, does anyone still claim that the Peasant’s Revolt led to greater democratisation? The Whig movement for parliamentary reform suffered a serious setback as a result of the French Revolution and the successful passage of the Great Reform Act of 1832 was partly the result of the wish of disaffected Tories to counteract the threat to established religion from the Catholic Relief Act. The Chartist Movement (1836-1848) led to no further democratisation — the 1876 Reform Act being Disraeli’s (failed) attempt to seek electoral advantage by stealing the Liberals’ clothes. And women only received the vote as a consequence of their contribution to the war effort, not as a result of the Suffragette movement.

    In short, the historical record indicates that democratisation has been a result of political elites following their own interests and there is no reason to believe that the introduction of sortition will be any different. I look forward to evidence to the contrary from Yoram (and from Arturo pointing out why the Barings and Lehman examples don’t invalidate his claim that “temperatures in hell will be very close to 0ºK before an elected politician lets a bank fail out of his own volition”), but I’m not holding my breath. I’m also more than a little puzzled by Arturo’s startling claim that the leaders of political parties are “psychopaths by definition” — it looks like he’s been reading Arthur’s book.


  35. Keith,

    I will gladly withdraw my assertion that a politician will never let a bank fail (even if I still could cling to the clause “out of his own volition”).

    Frankly, I am not familiar with the case of Barings Bank. With regard to Lehman Brothers, it is undeniable that for some reason the US Treasury decided to let them fail. Their mistake, I suppose, was to neglect, when things started to go south, the insider relations that would have provided the required political clout. Competitors like Goldman Sachs (boy did they have political clout) were not necessarily unhappy to see them disappear. And the question of why the US Treasury did not let all other banks fail remains unanswered.

    In any case, let’s not lose sight of the fact that banks, as any other private company, should always be left to their own devices in a well-functioning market economy. To Schumpeter’s theoretical claims we can add the historical proof of Iceland recovering from the crisis faster than anyone else after having let her banks fail. In conclusion, there is no sound explanation for bank bailouts nowadays other than outright corruption.

    Now, regarding psychopathy, I am on my turn more than a little puzzled by you ignoring such an strong argument in favour of sortition. It is true that most people don’t know what psychopathy is, immediately thinking of serial killers; however, the actual prevalence of psychopathy is around 1% of the general population. In an allotted chamber, you could reasonably expect a 1% of psychopaths. There are of course no data on prevalence among politicians. Personally, I would expect no less than 75% of psychopaths among top-level elected politicians.

    I don’t think I have read Arthur’s book. Would you please point it to me? Recommended reading on this subject would be The Mask of Sanity (Cleckley, 1941) and Without conscience: The disturbing world of psychopaths among us (Hare, 1993). A good introductory article is to be found here:

    While psychosis is a psychic disability (loss of contact with reality), psychopathy can be defined as a “moral” disability. The muting of social emotions like shame or guilt due to a structural, possibly neurological lack of capacity to feel empathy towards fellow human beings (this trait is technically called “callous-unemotional”, or CU) is the proximate mechanism that enables psychopaths to pursue their self-centered goals. They naturally con and manipulate others because they inadvertently acquired, as a way to “belong” in society, the ability to mimic all those emotions they could not feel. They could not understand why other children smiled, laughed or wept, but learned very quickly to observe, detect and imitate all these behaviours as a response to the positive stimulus they received in exchange from both adults and peers. When they grow, they can use that skill to their advantage.

    As an aside, why do the 99% of us actually have some kind of empathy? Because it is a positive evolutionary trait: it increased the probability of survival of our species. Human groups with low empathy went extinct.

    Our society should at least be aware of this 1% of psychopaths without a conscience walking among us behind a mask of sanity. There is no way to identify them. Cleckley was the first to note that individuals with psychopathy are to be found in all cultures and races and are often raised in above average home environments. We can assume that many of the low-class psychopaths end in prison, but where do most high-class psychopaths end? My personal answer to this question would be: in the boards of directors and the councils of ministers.

    What skills are most needed to climb up the ladder of a corporation or a political party? I am afraid that honesty and integrity are a bit less useful than duplicity and treason and lying with a straight face. Incidentally, psychopaths master these better than anybody. As a logical consequence, the higher you climb up the ladder of power, the higher the prevalence of psychopathy. We all have just been witness to some meetings of major donors to the different candidates to this year’s US presidential elections. Would it be exaggerate to affirm that 90% of the gentlemen around that table were probably psychopaths? Only sortition will save us from them.


  36. Arturo,

    Thanks for that, my point was only that the reasons for bailing out the banks may well have been (in part) epistemic, as opposed to purely the conflicting interests of the elite(s) and the people. I think the relevant government officials (and some elected politicians) are well aware of Schumpeter’s theory, creative destruction, moral hazard etc and will have taken that into account (as would a well-briefed allotted forum), alongside considerations of the likely externalities of not bailing out the banks.

    Here’s my review of Arthur’s book. As a practicing psychiatrist he would most likely agree with your views on sociopathy (as its known in PC circles).


  37. Thank you for the reference, Keith.

    >Given that such psychopathic individuals – ‘a special subset of men’ – are fundamentally different from ‘us’, then the goal of democracy is not so much ‘power to the people’ as making sure that the bad guys don’t get hold of the reins. Rotation of office (and/or mass participation in government) is not so that we may all, as Aristotle put it, ‘rule and be ruled in turn’ but simply to reduce the likelihood of handing power to a psychopath.

    If I think about it, I find this argument more psicologicofor sortition and against elections as compelling as any and more than many.


  38. O dear, this is going from bad to worse. Rather than just arguing that elected politicians are self-interested we’re now claiming that they’re psychopaths — this sort of rhetoric is unlikely to help us gain friends and influence people. And why would the 4% of allotted persons who take up the sortition invitation be any different from the psychopaths who decide to offer themselves for election as, no doubt, their motive is a thirst for power? I can understand the argument for cognitive diversity via sortition (although I think there are better ways of securing it), but the strongest case for sortition in large modern states is descriptive representation. There is a danger that any sane person reading Arturo’s comments might think that psicologico is an appeal to parapsychology or other such flim-flam.


  39. Arturo,

    While I doubt the overwhelming rate of psychopathy (or the less extreme sociopathy) among elected officials you suggest, researchers HAVE confirmed that it is certainly MORE common among political leaders and CEOs than in the general population. But in addition to your and Arthur Robbins’ view about psychopaths rising to fill such positions, there is another psychological problem of elections… which is the ACQUIRED mental disturbance that David Owen (former Foreign Secretary and medical doctor) has proposed be added as a formal illness … the “Hubris Syndrome” that afflicts some people who get into positions of great power… Basically a medical take on the old adage that power tends to corrupt. He wrote a book about this with that title “The Hubris Syndrome: Bush, Blair and the intoxication of Power”
    It’s on Amazon here


  40. I would like to make two points.

    First, I still think the character of an opt-out body is an open question. Low turnout would favor the passionate, for sure, regardless of their political leanings, just as it does in elections. We don’t know what the turnout would be in practice. Last year my city (Atlanta) held a referendum on borrowing a quarter of a billion dollars to pay for infrastructure improvements. The turnout was about 7.5%. The BC citizen’s assembly was unusual, lacked of the power to decide, and was on a subject most people don’t care for. Given that the BC citizen’s assembly lacked the power to decide, those who favored the status quo knew they’d be served just fine by staying home and voting in the referendum. Furthermore, the proposal put forward by the assembly did manage to pull down a large majority in the first referendum. The second referendum went the other way, of course.

    Maximizing the participation rate is most certainly good, and making use of mandatory participation to get to nearly-full participation may well be justifiable, just as mandatory voting may well be justifiable. In that case, if it were needed, mandatory long terms may be okay too.

    Second, the corruption and “intoxication” that come from holding power are, as Terry said, acquired. They should be independent of the mode of selection and affect an allotted assembly no more or less.


  41. Keith,
    This is getting interesting.
    I think all three arguments mentioned by now are complementary. Descriptive representation answers the question of legitimacy: nobody voted them in, but they represent the people because they are the people. Cognitive diversity answers the question of competence or qualification: they don’t have to be the best and brightest individually because, taken as a group, they will perform better than an homogeneus group of middle-aged lawyers. What I will call now unflawed psychology, for lack of a better term, although we should start giving it some thinking, is not really an answer to a possible objection, as were the previous two. It is, in a way, much more than that — it is the reason as to why we need sortition to begin with. We live in a world in which the richest 1% has amassed, for the first time in human history, more wealth than the poorest 99%. As a consequence, 30,000 children die everyday of hunger and preventable diseases. We live in a world ruled by psychopats.

    I disagree with Naomi’s view that an allotted MP will be equally affected by the hubris syndrome — this is not as much a consequence of holding power as it is of having been elected, of feeling chosen. An allotted MP has no reason to feel superior just because the dice rolled her way.
    In any case, and with all due respect to Dawid Owen, the importance of this cannot be compared with the prevalence of psychopaty among 1% of the population. Please understand that I’m talking pretty hard science here: we may not know yet the exact physiological mechanism behind CU and the lack of empathy, but we do know for a fact that there are psychopats among us and that they are not like us. Indeed, they are much better than us at what it takes to get oneself elected.

    The problem of opt-out is best solved by strong incentives. My favourite is a lifelong pension, not necessarily too high but cumulable to any other source of income.


  42. Arturo
    I’m sure newly elected officials have a higher opinion of themselves than would newly drawn allotted assembly members. At least on average, that is. Different people handle power differently. I’m sure we’ve all seen examples of people in our own lives who became drunk off a truly trivial amount of power.


  43. I have been reading further in order to offer a better understanding of what psychopathy normally means for the roughly 1% of the population affected by it (nothing to do with the serial killers).
    Following Hare, who is the leading English-speaking authority, the four dimensions of psychopathy are: interpersonal (e.g., “cunning, manipulative”), affective (e.g., “callous, ruthless”), impulsive (e.g., “irresponsible, stimulus-seeking”), and antisocial (e.g., “greedy, exploitive”).
    Now, I am asking you: are psychopaths at a distinct advantage, first to climb the ladder of leadership in a political party, then to get elected?


  44. Arturo,

    Your distinction between 1) legitimacy, 2) cognitive diversity and 3) unflawed psychology is well made. My focus is on the first, Terry’s the second and yours is the third. I would argue that legitimacy (1) is the most important consideration in any constitutional regime and is the only one of the three that can be readily operationalised, in that the representativity can be tested by drawing parallel samples to deliberate on the same topic and seeing if they come to the same conclusion (i.e. representing “what everyone would have thought under good conditions”). But how can you tell if they are making (2) the right decisions (without the benefit of hindsight)? And how can you tell if the people being selected for office have (3) unflawed psychology? I think you are right to say that elected members will be more likely to be affected by hubris than allotted members, but if opting out is deemed acceptable then the sample will be highly atypical from a psychological perspective so will lose descriptive accuracy. So, although all three factors are important, I think legitimacy has to be the most important one. I also think that to claim that because someone is a member of the 1% they will necessarily act against the interests of the 99% is crude economic determinism. Most members of the UK Labour Party are London-based middle-class liberals, most of the traditional working class having defected to UKIP (and the SNP). And what about Bill Gates?

    >the four dimensions of psychopathy are: interpersonal (e.g., “cunning, manipulative”), affective (e.g., “callous, ruthless”), impulsive (e.g., “irresponsible, stimulus-seeking”), and antisocial (e.g., “greedy, exploitive”).

    While these may be characteristic of charismatic leaders, is it fair to tar all senior elected politicians with the same brush? I don’t think (say) Stanley Baldwin, Neville Chamberlain or Clement Attlee really fit that description.


    The hubris syndrome would apply more to government officers (presidents, prime ministers etc) rather than members of a legislature, whether elected or allotted.


  45. Arturo.

    I have known a number of people who were most likely sociopaths (or psychopaths) … and I can’t guess how many others who were skilful enough to hide this personality trait. Of those I am pretty certain were sociopaths or psychopaths, all but a few were politicians. I should explain that I served 20 years as an elected official so dealt closely with many politicians. HOWEVER, I am also quite certain that a vast majority of politicians I worked with DID have a conscience, felt guilt, and knew right from wrong and were NOT sociopaths….Of course, this was at a very low level of politics,– small municipal and small U.S. state legislature… and I expect the concentration of sociopaths becomes far more concentrated as one moves up to the Federal level.

    I feel this psychological concern may be a motivating reason for you, but unless these politicians honestly submit to evaluation (and why would they) it can’t be proven to be a problem. What CAN be proved is that elected legislatures are not descriptively representative of the population, and that their policies are frequently corrupt and certainly not optimal for society… Thus I think the “legitimacy” concern and “epistemic” concern are the issues that can motivate most people to consider reforms like sortition (even if you are correct and it happens to be true that the psychopathy danger is the most important).


  46. Terry,

    Yes, the descriptive representation argument (1) can be proved (both mathematically and experimentally) and the case for cognitive diversity (2) can be illustrated by the evidence cited by Page, Landemore etc (my only caveat being that there are better ways to encourage “outside the box” policy innovation than sortition). I don’t, however, agree that it can be “proved” that the policies of electoral legislatures are “certainly not optimal for society” on account of my scepticism over the claims of Gilens etc. The difference of opinion over what policies are optimal is the subject matter of politics, not epistemology. If such disputes could be resolved by (social) scientists then we would have no need for democracy by election or sortition, we could leave it to this latest breed of Platonic guardians to determine what policies are “optimal”.

    In the case of “unflawed” psychology (3) it is always tempting to attribute policy clashes to character flaws (take, for example, Yoram’s angry dismissal of “Sutherland” as a pathological liar). Liberals argue that public policy should be compassionate, whereas conservatives are more concerned with the unintended consequences (a conservative being a liberal who has been “mugged by reality”). In the case of the European refugee crisis liberals want to welcome refugees with open arms whereas conservatives are more concerned with the likely effect on social cohesion and indigenous culture (and robbing Syria of its brightest and best). I don’t think the latter priorities reflect the (sociopathic) character of conservatives, merely a difference of opinion on public-policy priorities. Although conservatives do tend to love their own kinder more than abstractions like humankind this is not an affective disorder, even though Kantian cosmopolitans might view it as a dereliction of the moral duty to love one’s neighbour (in a global world we are all “neighbours”) in a dispassionate way.

    Regarding the epistemic argument (2) I’m puzzled that anyone with a hard-left provenance should advocate sortition. Neo-Marxists (Gramsci, Frankfurt School, Eurocommunism, New Left, Althusser, Lukes etc) claim that the hegemony of mass culture is a product of indoctrination by the elite (knowledge production being also controlled by capitalism). If so then a randomly-selected microcosm will be equally hegemonic. Perhaps this is why Terry argues that policy forums should be composed on a voluntary basis as this will provide greater opportunity for activists and other subaltern voices (although, as Naomi pointed out, they are just as likely to be dominated by the Tea Party).


  47. With regard to terminology, I suggest that we use from now on the more neutral CU (callous-unemotional) instead or psychopath. We can then speak of a “diluted CU” of around 1% in the general population or a “condensed CU” in any delegated body in which the fact of being “cunning, manipulative”, “callous, ruthless”, “irresponsible, stimulus-seeking” and “greedy, exploitive”, that is, of being CU, increases one’s chances to be selected. Elections are an obvious case of a selecting mechanism producing a highly condensed CU, while sortition guarantees a diluted CU.

    In this perspective, the question of “CU concentration” (let’s forget about “unflawed psychology” – not a wise choice of words) is not a separate concern, but just a particular case of the descriptive and the epistemic questions.

    We can argue that the most important characteristic that has to be represented in the exact proportion within the minipopulus is not sex, age, race or wealth, but CU. The main problem with elected legislatures is not that they have too many middle-aged rich white men, but that they have way too many CU. And why is this so? Because none of those qualities helps a successful candidate as much as being CU has helped him in the first place.

    The epistemic consequences of a condensed CU parliament are quite obvious: no matter how clever they are, CUs are very unlikely to adopt policy decisions that benefit others because they lack the basic capacity to acknowledge other people’s needs (empathy). And the proof for this is all around us, if we only care to look for it: there is a huge amount of unnecessary pain in the world, and the reason for it is that power is being held by CUs everywhere.

    Finally, I’d want to clarify two misconceptions by Keith. The first is just a confusion between two different “one percents”. The richest 1% is just an arithmetic device. It happens to be important because, according to the latest report by Oxfam, when we split in two all the wealth in the world, half is owned by the richest 1% and the other half by the poorest 99%. This had never happened before in all human history. On the other hand, there is nothing arbitrary with the approximately 1% of the population suffering from a lack of empathy, what doctors used to call psychopaths and now call “callous-unemotional” (this trend started with regard to the youngest patients, once it was clear that all adult patients had suffered this condition from their earliest childhood). That percentage results from the medical consensus, although some authors defend higher rates of prevalence of up to 4%. In any case, there is no identification between these two subsets. Sure the prevalence of CU is much higher among the richest 1%, but certainly not all of them are CU.

    Much more doubtful is any kind of correlation between CU and political leanings. I at least will never say that conservatives are more likely to be CU. Sorry for not adding here to that comfortable feeling of victimism, Keith. In the example you bring, it is the “open borders” forces who call themselves in the left that are, willingly or unwillingly, being accomplices to what clearly is a geopolitical plot geared to increase islamophobia in Europe, at the cost of the personal suffering of most of the migrants, whose dreams will be shattered as soon as reality sets in (anyway, not a discussion that I’d like to open here).

    For the record, I’ve added a comment to the old thread discussing Arthur D. Robbins’ book, in which I insist in the point of view expressed here.


  48. Arturo,

    >according to the latest report by Oxfam, when we split in two all the wealth in the world, half is owned by the richest 1% and the other half by the poorest 99%

    If I owned a large comfortable house in the developing world and then sold it, the cash realised might, if I were lucky, buy me a phone box (or a shoe box) in Manhattan on Mayfair. So what? Most people would prefer living in a comfortable house. Similar considerations apply to income — the 1% would include those on a very modest income in the West, whereas the 99% would include many who are relatively rich in terms of local spending power. Oxfam is a campaigning and fund-raising organisation but, as an economist, you should realise that these headline-grabbing figures are of no value for sensible debate on political economy. As you rightly say (and then immediately ignore): “the richest 1% is just an arithmetic device”.

    You should also be careful before floating wild conspiracy theories (before saying you don’t want to discuss them) as this will reduce the credibility of your more sensible claims. Having already claimed [falsely] that no elected politician would let a bank fail, you now claim that the refugee crisis is the product of a “geopolitical plot geared to increase islamophobia in Europe”. Really? Who are the plotters — the evil Zionists and their evangelical Christian supporters? Leaving aside the need of the German government to steal the brightest and best from Syria to counteract their own declining population, the most significant pull factor for the refugee crisis was the outpouring of public empathy and compassion resulting from the sharing of a photograph of Alan Kurdi, a drowned three-year-old Syrian child, last year. As for the supposed lack of empathy in the political class, the decision by David Cameron and Nicolas Sarkozy to support the Libyan insurrection resulted from their compassionate calling to protect the citizens of Benghazi from an attempt by the Gaddafi regime to put down the insurrection (now regarded as a successful propaganda coup by the insurgents). This had the effect of emboldening the Syrian insurgency, leading to the grotesque chaos that we now see in the Middle East. An allotted parliament (if fully informed of the likely consequences of policy-making based on “compassion” and “empathy”), would be more likely to pursue the cold-hearted national interest than to indulge the messianic complex of charismatic politicians trying to save the world from the forces of evil (the allusion to Blair and Bush is intentional).


  49. Keith,

    >Oxfam is a campaigning and fund-raising organisation but, as an economist, you should realise that these headline-grabbing figures are of no value for sensible debate on political economy.

    Oxfam’s analysis was based on figures from the Credit Suisse Global Wealth Report. I hope Credit Suisse is up to your requirements for sensible debate.

    >If I owned a large comfortable house in the developing world and then sold it, the cash realised might, if I were lucky, buy me a phone box (or a shoe box) in Manhattan on Mayfair.

    In international comparisons of consumption or income it is common to convert currencies using purchasing power parity (PPP) exchange rates, which take account of local prices, especially for non-traded services. However, in all countries a large share of personal wealth is owned by households in the top few percentiles of the distribution, who tend to be internationally mobile and to move their assets across borders with significant frequency. For such people, the prevailing foreign currency rate is most relevant for international comparisons. So there is a stronger case for using official exchange rates in studies of global wealth.

    >Similar considerations apply to income — the 1% would include those on a very modest income in the West, whereas the 99% would include many who are relatively rich in terms of local spending power.

    According to Credit Suisse’s figures (and assuming you don’t dismiss them out of hand as headline-grabbing), the world population in 2015 was 7,264,391,000 of which 4,770,803,000 where adults. Global wealth was 250,145 USD billions. If half of that was owned by 47,708,030 people, their average net worth was 2,621,624 USD. I wouldn’t personally call that “a very modest income”, but of course I live on a salary paid by a European institution — not on one paid by an English university.

    >Leaving aside the need of the German government to steal the brightest and best from Syria to counteract their own declining population

    I humbly acknowledge not having any hard proof for the “conspiracy theory” I find more convincing. Secrets services are called secret because their dealings are kept, well, secret. These leaves us only with our own judgement to gauge the explanatory value of alternative theories. However, if you prefer to think that the German government fomented war in Syria to steal their best and brightest, I think this goes in pair with thinking that Cameron and Sarkozy destroyed Libya out of pure empathy. Please don’t be upset, but I insist that we leave geopolitics aside of this forum and focus instead on discussing sortition.


  50. Arturo,

    I’m no economist (unlike yourself) but my understanding is that salary (how much you get paid every month) and wealth (assets [land, houses, shares, motor cars etc]) were entirely different entities, yet you use the two terms as if they were synonymous. I don’t dispute Credit Suisse’s figures, I just pointed out that Oxfam is using them for polemical purposes (fund-raising campaigns). Thank you for confirming that USD exchange rates were utilised in calculating the total global wealth — from the point of view of lived experience a comfortable house in the developing world is a lot more “valuable” than a shoe-box in Manhattan, notwithstanding the views of the bean-counters.

    >If you prefer to think that the German government fomented war in Syria to steal their best and brightest, I think this goes in pair with thinking that Cameron and Sarkozy destroyed Libya out of pure empathy.

    I claimed (uncontroversially) that the German government saw economic benefit in the migrant crisis (not that it created it) and that the unintended consequence of actions based on empathy and compassion can be entirely malign, however much Cameron, Blair etc insist that they did the (morally) right thing. I’m not sure what geopolitical advantage the West saw in destroying Libya, but I agree with you that this is not the topic of this forum. However your claim that an allotted parliament would be more empathetic/compassionate, and the implication that this would be A Throroughly Good Thing certainly is.

    PS I wish I did have a salary from an English university, unfortunately I have to eke out a modest living operating printing machines.


  51. *** Common Lot Sortitionist praises (January 19) Keith Sutherland for underlining “the difference between juridical and political functions” of allotted juries. I think such words potentially misleading. Judicial power is a basic element of a political system, because actual law is what judges declare as law, and because a law which is not supported by serious sanctions may easily become only words.
    *** Right, Keith wrote specifically about juries whose task is only “determining guilt or innocence in the criminal justice system”. “Representative accuracy” here is not a strong requirement, he thinks, and small juries can be accepted.
    *** This is valid only when there is consensus among citizens about the political values and principles underlying the judicial decisions (consensus – let’s say when 95% citizens agree). Even speaking only of juries considering cases of criminal violence, it will not be always the case. In a State with death penalty, the citizenry may be strongly divided about the subject, and opponents to death penalty may be tempted to downgrade the culpability to avoid death. About non-violent crimes the division may be especially frequent, along class or ideological lines, and lead some to downgrading of the crime or to excessive doubt about it .
    We see patent division in Western contemporary “culture” concerning the complaints of women about sexual violence or harassment, with some considering that “word is not enough” and those considering that we must “respect and trust the victims”. Etc …
    *** The general trust about the police work, the general trust about witness or about scientific evidence, the high or low “repressive tendencies”, etc may vary strongly in some citizenries.
    *** Maybe these divisions are especially enhanced by the divisive political processes in our polyarchies, maybe they will be lesser in a future modern dêmokratia, but at least in our societies often we cannot suppose consensus and a 12 persons jury could often leave room to the luck factor.
    *** It seems that in the USA the law allows very wide challenging of jurors and that lawyers of both sides use it to get the kind of jury they like. I heard the joke that in UK a trial begins on convening the jury, whereas in USA the trial ends at this time. Such strategies would be pointless if Keith is right.
    *** Note that, in the cases we could trust a consensus about all the relevant parameters, it would be logical and more practical to limit jury service to retired citizens; for instance in their first year of retirement (in Athens, there was mandatory civic service as arbitrators for citizens in their 60th year)


  52. *** I agree with Arturo : a strong advantage of sortition is its efficiency against any concentration of “callous-unemotional” minds, concentration which is to be feared at least in the more powerful circles of polyarchies.
    *** But I think the idea must be considered more widely.
    *** First the “callous-unemotional” minds may be concentrated not only in the elected assemblies, but in the higher circles in many kinds of lobbies, which are the central actors in polyarchic systems.
    *** Second we must consider other kinds of dangerous “flawed minds”, as the paranoid ones, or those with tendency to see the world along one line – not only the outright fanatics, likewise people with soft minds but same tendency. Note that a mind may be flawed along one dimension and very good along other dimensions.
    *** Third, we must consider that the minds may be flawed before entering a group, but, likewise, through the specialization of the group. There is a risk of military men seeing the world only through military lens from the effects of initial selection, or self-selection but, likewise, because of “professional quirk” and excessive military centering. The same for many categories.
    *** Sortition among the whole citizenry is efficient to prevent concentration of flawed minds of any kind and from any cause. Likewise, it prevents concentration of better minds. But any device aiming to concentrate better minds (better along a chosen dimension) may actually concentrate flawed minds (along this dimension, or another one). Only sortition is safe.
    *** This argument is valid not only against polyarchic systems, but against totalitarian systems and authoritarian ones.
    *** Sortition gives the power to ordinary minds. In exceptional situations, ordinary minds may issue very bad decisions, even after due deliberation. But the risks are lesser than with systems giving much clout to statistically eccentric minds.


  53. Andre,

    I think your response to Common Lot Sortitionist refers to a different thread (“The difficulty of insuring accurate randomness”). But if you want a trial jury to be statistically representative then the same considerations would apply as in a legislative jury, i.e. it would need to be several hundred strong, be barred from jury-room deliberation and accept simple majority verdicts — as in the Athenian jury courts. I doubt whether that’s either practical or desirable, as the procedure adopted for the trial of Socrates would be unlikely to receive much modern support.


  54. Andre:> Sortition among the whole citizenry is efficient to prevent concentration of flawed minds of any kind and from any cause. . . Sortition gives the power to ordinary minds. In exceptional situations, ordinary minds may issue very bad decisions, even after due deliberation. But the risks are lesser than with systems giving much clout to statistically eccentric minds.

    Sure. The aggregate judgment of ordinary minds is the best way of spotting paranoia/psychopathy/groupthink, so long as you ensure the discursive input of “better” minds includes as wide a range of dimensions as possible (including the statistically eccentric). That way you get to have your cake and eat it.


  55. *** Keith says the Athenian procedure of very big juries for criminal judgment is neither praticable nor desirable. This can be debated, given the modern electronic devices, which allows much better processes, avoiding the simplistic deliberations and binary choices.
    *** But Keith does not answer to any of my arguments. I conclude he acknowledges that anytime there is no wide consensus along all the relevant dimensions, a twelve persons jury gives much room to luck. That could be accepted for practical reasons, but we must acknowledge it.
    *** And, as I said, if we affirm there is such a consensus, a jury recruited as the Athenian arbitrators would be good.


  56. *** Keith Sutherland wrote he is “puzzled that anyone with a hard-left provenance should advocate sortition. Neo-Marxists (…) claim that the hegemony of mass culture is a product of indoctrination by the elite (…). If so then a randomly-selected microcosm will be equally “under the sway of the hegemony.
    *** OK, the leftists who are extreme Gramscians will not support sortition. But others, with less extreme views about “hegemony”, may acknowledge the great strength of inculcation by the elite, but think that deliberation will lessen this strength. Elite ideologies will have less clout for votes by citizen juries after deep deliberation than for the mass electoral votes in our polyarchies.
    *** Furthermore, laws voted by minipublics, with some degree of “freedom”, may lessen the institutional grip of the elites on the political debates, and heighten the level of further “freedom” from hegemony.
    *** Along this line, pro-dêmokratia leftists will put great importance on the deliberation; on the conjunction of sortition and deliberation; sortition of minipublics being especially valued for allowing deep deliberation, usually impossible in general votes. And they will seek strong legislative power for the minipublics, expecting institutional reforms. Conversely some conservatives not deeply democrats but ready to allowing some role to minipublics will lean to limit deliberation and to rein in the role of minipublic through elected or “independent” institutions, to protect the hegemony.
    *** We must see that the insistence on deliberation is two sided. In conjunction with sovereign minipublics, it is democratic – and compatible with some kinds of hard-leftist sensitivities. Without sovereign minipublics and with deliberation referring mainly to debates in elite circles, it may be an alternative view of polyarchy, akin to the specific brand of left linked to the culture elite – as for Habermas as I understand.

    Liked by 1 person

  57. Andre,

    Although jurors will sometimes (illicitly) allow their own views on the nature of the offence and the likely punishment to colour their verdict, nevertheless they are charged with an epistemic task (deciding on whether or not the accused is guilty or innocent on the basis of the evidence outlined by the prosecution and defence and the law explained by the judge). It is an epistemic, not a moral decision which could (in principle) be taken by a computer and there is no need for the jury to be a portrait in miniature of the whole society in order to exercise this function. But the task of the legislative jury is entirely different — a decision on changing the law is a moral decision, so the jury has to accurately represent the morals and interests of the target population, hence the need for large juries.

    >a twelve persons jury gives much room to luck.

    Yes that’s true. From my own experience serving on a jury in a commercial fraud case (trading while knowingly insolvent), the problem wasn’t a moral one, although we all considered the accused to be something of a shyster. The problem was that nobody on the jury (apart from myself) had the experience of running a small company that was on the verge of bankruptcy, so were unable to put themselves in the shoes of the accused. This is necessary in the case of commercial fraud as the issue is whether or not the accused intended to defraud (when you’re about to go bust you do all sorts of desperate things that, with the benefit of hindsight, are misguided, but it’s easy to be wise after the event). I persuaded sufficient members of the jury that the answer was not guilty and as a result we were dismissed, leading to a new trial which returned a unanimous guilty verdict. So luck does come into it, but I don’t think the jury size would have made much difference as the time for individual speech acts in a jury of (say) 300 would need to be strictly limited, so the (majority) verdict would reflect the fact that only one or two of the jurors might have undergone the necessary life experience. So the verdicts of large juries would be more consistent, but that doesn’t mean that they would be more likely to be epistemically correct. In fact a large jury, swayed by a popular sentiment, would be more likely to put two fingers up to the law, hiding behind the anonymity of large numbers. This would be particularly true of guilty verdicts in capital punishment regimes — if you have one vote in twelve then you need to examine your conscience a lot harder than if you are just one in 501.


  58. Andre,

    >Elite ideologies will have less clout for votes by citizen juries after deep deliberation than for the mass electoral votes in our polyarchies.

    Is there any evidence to support this claim? I imagine Dryzek’s insistence on the equal representation of discourses is because he is sceptical regarding the power of deep deliberation to challenge hegemonic assumptions. You seem to be suggesting that the masses will spontaneously cast off their (mental) shackles as soon as they start to deliberate and I don’t see why that should be the case. If it doesn’t happen spontaneously in the saloon bar why should it happen in the deliberation chamber?

    >Furthermore, laws voted by minipublics, with some degree of “freedom”, may lessen the institutional grip of the elites on the political debates, and heighten the level of further “freedom” from hegemony.

    But if the masses are victims of false consciousness resulting from the existing hegemony, why will they accept these new laws? Although they may have been instituted by people who look like them, they certainly don’t think like them.

    >conservatives not deeply democrats but ready to allowing some role to minipublics will lean to limit deliberation and to rein in the role of minipublic through elected or “independent” institutions, to protect the hegemony.

    Yes, but conservatives like myself don’t view the existing hegemony (in so far as there is one in a multicultural pluralist society) as the result of false consciousness, we view it as the settled will of the people at that time. And this will be corrupted by a sortition process that permits opting out, as it will lead to the replacement of the will of the people with the will of activists and political anoraks (the 4% in the BC Convention figure). Rather than rule by the 4% of self-selecting busybodies I’d prefer to choose from those who offer themselves up for election, even if they include a disproportionate number drawn from the 1%.


  59. André,
    What makes CUs different and more dangerous than any other kind of “flawed minds” is not only the decisions they could take (i.e. the fact that they are “irresponsible, stimulus-seeking” or “greedy, exploitive”); those paranoids you mentioned as an example would undoubtedly choose wrong or unwise policies too. What makes CUs worse is the mere fact that they are better equipped than the rest of us to rise up to the higher offices (being as they are “cunning, manipulative” and “callous, ruthless”).
    A huge problem to be dealt with is, as you’ve rightly pointed out, the fact that CUs “may be concentrated not only in the elected assemblies, but in the higher circles in many kinds of lobbies, which are the central actors in polyarchic systems”. Being a public servant myself, I can confirm from experience that public administrations are no exceptions to the rule that there is a sizable proportion of CUs in high positions. When debating with Etienne Chouard, François Asselineau dismissed sortition by saying that experienced bureaucrats would make short work (ne feront qu’une bouchée) of allotted representatives. Cynics would affirm that it takes a CU to know and deal with another and that, after all, we are better served by having CUs as leaders since only them can keep other CUs at bay. Needless to say that I don’t share this defeatist view.


  60. *** Considering the postulate “Elite ideologies will have less clout for votes by citizen juries after deep deliberation than for the mass electoral votes in our polyarchies”, Keith Sutherland says (February 6) “Is there any evidence to support this claim?”.
    *** Good observation. We will have to look to empirical studies about “deliberative polls” and like. But note that this postulate was intended to explain how a hard-Gramscian may favor citizen juries. Personally I am very moderately Gramscian. I tend towards thinking that the efficiency of cultural hegemony is exaggerated, and that the pure fact of given sovereign power to citizen juries (with due deliberation procedures) will give them a strong degree of freedom from any “hegemony”. The degree of validity of this idea will be tested only when a modern dêmokratia will be established.


  61. Andre,

    I’ve never studied Gramsci but didn’t he believe there was a need for re-education to overcome false consciousness and that this was a (revolutionary vanguard) elite function, rather than an emergent property of deep deliberation? Personally I think cultural Marxism is BS**, in both its hard and moderate version. Returning now after a 30 year hiatus to departmental seminars where they still drone on about hegemony and subalterneity I wonder if I’ve fallen into a parallel universe that has remained the same since I left the university in the mid-1970s. Plus ça change . . .

    ** bullshit, rather than Bernie Sanders, that is.


  62. Keith Sutherland complains about the resilience of hard-Gramscian postulates in some academic circles. We may fear that , after the historical failure of Marxism-Leninism, these postulates got a new strength from a new source : hostility to modern dêmokratia. Some fragments of the culture elites may elaborate from Gramscian ideas a “leftist” brand of discourse supporting actually polyarchy.


Leave a Reply

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

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

Facebook photo

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

Connecting to %s

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

%d bloggers like this: