The Dunning–Kruger effect and its implications for voting

The Dunning–Kruger effect

is a cognitive bias wherein relatively unskilled individuals suffer from illusory superiority, mistakenly assessing their ability to be much higher than is accurate.

The effect is named after David Dunning and Justin Kruger of the department of psychology at Cornell University who published a paper in 1999 which described a series of experiments which they conducted which demonstrated the effect.

In an interview, Dunning described his understanding of the effect as follows:

Dunning: [I]f you’re incompetent, you can’t know you’re incompetent.

[Interviewer:] Why not?

Dunning: If you knew it, you’d say, “Wait a minute. The decision I just made does not make much sense. I had better go and get some independent advice.” But when you’re incompetent, the skills you need to produce a right answer are exactly the skills you need to recognize what a right answer is. In logical reasoning, in parenting, in management, problem solving, the skills you use to produce the right answer are exactly the same skills you use to evaluate the answer. And so we went on to see if this could possibly be true in many other areas. And to our astonishment, it was very, very true.

In the discussion section of the paper itself the authors write:

When can the incompetent be expected to overestimate themselves because of their lack of skill? Although our data do not speak to this issue directly, we believe the answer depends on the domain under consideration. Some domains, like those examined in this article, are those in which knowledge about the domain confers competence in the domain. Individuals with a great understanding of the rules of grammar or inferential logic, for example, are by definition skilled linguists and logicians. In such domains, lack of skill implies both the inability to perform competently as well as the inability to recognize competence, and thus are also the domains in which the incompetent are likely to be unaware of their lack of skill.

In other domains, however, competence is not wholly dependent on knowledge or wisdom, but depends on other factors, such as physical skill. One need not look far to find individuals with an impressive understanding of the strategies and techniques of basketball, for instance, yet who could not “dunk” to save their lives. (These people are called coaches.) Similarly, art appraisers make a living evaluating fine calligraphy, but know they do not possess the steady hand and patient nature necessary to produce the work themselves. In such domains, those in which knowledge about the domain does not necessarily translate into competence in the domain, one can become acutely–even painfully–aware of the limits of one’s ability. In golf, for instance, one can know all about the fine points of course management, club selection, and effective “swing thoughts,” but one’s incompetence will become sorely obvious when, after watching one’s more able partner drive the ball 250 yards down the fairway, one proceeds to hit one’s own ball 150 yards down the fairway, 50 yards to the right, and onto the hood of that 1993 Ford Taurus.

Interestingly, as far as I could see, neither the paper nor followup discussions comment on the implications of these findings on elections. If voters, who are inevitably uninformed about most issues government deals with, do not recognize the fact that they are uninformed then they are easy prey for manipulation. If voting is more akin to logical reasoning than to playing golf, then no rational decision making can be hoped for on the part of the voters – not even rational uninformed decisions. Not only will the voters be unable to make informed decisions, they would be unable to make rational uninformed decisions since they are uninformed about their uninformed state. Voters would vote believing they understand matters they don’t and thinking their judgement is based on facts and reasoning when in reality it is based on sloganeering and superficialities.

Advertisements

32 Responses

  1. Yoram,

    I have long argued that the tendency of us humans to overestimate our own competence (75% of drivers might insist they are better than average, which is mathematically impossible), has important implications for electoral systems. However, you have over-stated the implications by asserting “If voting is more akin to logical reasoning than to playing golf, then no rational decision making can be hoped for on the part of the voters – not even rational uninformed decisions.” This is not a necessary conclusion, since, even if some unknown percentage of voters lack high competence (and don’t recognize it), it is logically possible that most voters (when aggregated together) could still have adequate competence to make rational decisions. It is fair to say that we should have no confidence that the mere fact that voters believe they are competent necessarily means that they ARE competent to vote rationally. The problem of rational ignorance is very real REGARDLESS of underlying competence.

    It is also important to look at the OTHER SIDE… those who seek office, as opposed to the voters. We should recognize that individuals who choose to offer themselves as candidates (under the BELIEF that they would be competent rulers — setting aside those who simply seek fame or power) may be completely incompetent as rulers, but be unable to recognize their incompetence. Many candidates AND VOTERS confuse competence at public relations and campaigning for competence at governing. Indeed, I believe the nihilism and higher than normal rates of psychopathy among candidates, means that they are actually LESS competent than average citizens (though I can’t prove this). This incompetence for governing could be akin to the golfer who should see that others can hit the ball further, but unfortunately the overwhelming influence of random variables and the lack of obvious GOOD golfers to compare with makes it more akin to the poor logician who CAN’T see his own inability by comparison.

    Like

  2. >The Dunning–Kruger effect is a cognitive bias wherein relatively unskilled individuals suffer from illusory superiority, mistakenly assessing their ability to be much higher than is accurate.

    If this means that voters are not even sufficiently competent to select political office holders then why do you think they are competent enough to make actual policy decisions? If anything this is an argument in favour of knowledge elites as it is aimed against “relatively unskilled individuals”, so I imagine Socrates would have approved.

    Like

  3. Hi Terry,

    > The problem of rational ignorance is very real REGARDLESS of underlying competence.

    My point is that the DK effect goes beyond rational ignorance (and as far as I am concerned, ignorance is all that “competence” means in this context).

    If voters were simply ignorant that would be bad enough. They would then be unable to make informed decisions, but they would be able to account for their ignorance and disregard misinformation and resist manipulation. They would focus on the little that they do know – direct observations from their own lives – and use that to make rational, if uninformed, decisions.

    If, however, voters are unable to account for their ignorance, they are easily manipulated by those who are able to feed them misinformation and make them think they know things they don’t. In this case, voters do not even make rational uninformed decisions. Instead they make completely irrational decisions.

    As for elected politicians – I don’t believe they are incompetent. They are simply self-serving, so that their level of competence hardly matters to the average citizen (it matters a great deal to the politicians themselves).

    Like

  4. Yoram,

    So ordinary people (aka “voters”) are incompetent, ignorant etc but politicians are not incompetent, yet you would would still rather be ruled by the incompetent, because the competent are self-serving and voters are too stupid to realise this. Do you rule out a priori the notion of a competent public-spirited person and if so why? The only plausible reason I can imagine is the Machiavellian view that it is in the nature of the grandi to oppress, i.e. that there is an intrinsic link between aristocracy and sociopathy. That is Arthur Robbins’ view, but then he’s a shrink — do you agree with him, or is the above true for structural reasons?

    Please enlighten us.

    Like

  5. *** Keith Sutherland writes “Do you rule out a priori the notion of a competent public-spirited person and if so why? The only plausible reason I can imagine is the Machiavellian view that it is in the nature of the grandi to oppress”
    *** Well, there can be some level of statistical truth in the mentioned “Machiavellian view”. But it is not the main point. The “political elite” is a specific elite, corresponding strongly to a specific competence (I distinguish from “general elites”, as the money elite; one may be rich without specific competence, as heir). The specific competence of the polyarchic “political elite” is the competence to climb in a political party, and/or to be elected and re-elected. That implies some level of intellectual and relational abilities, but, likewise, some specific behavior propensities. If a politician is too much public-spirited, he will make the policy choices he thinks better for the “public good”, resisting the lobbies, contradicting the uninformed citizens (and, except very specific situations, he is unable to convince them in serious deliberation), encountering forces which will defame him, ridicule him, story-telling him in biased ways. Statistically, he will lose elections, lose credibility in his party, lose money support, etc… and being ejected practically of the political elite, or put into its margins …. More than Machiavelli, it is Darwin. Will stay inside the political elite the morally corrupt, or, much more often maybe, the one who succeeds easily in persuading himself that the decisions good for his career coincide with the “public good”, at least in the long range. Sure, they can be exceptional persons or situations, but this is the trend. Some of the main points of the “specific competence” of the polyarchic political elite push it to issue political decisions along the “parallelogram of forces” in the society.
    *** Actually the political elites of our polyarchies are closely linked to the general elites (money elites and culture elites), and have akin spontaneous sensitivities; but even if we imagine a law with quota about such social criterions (as we see in many countries about gender), the systemic trend will be the same.
    *** If accused of a crime, especially of a crime sensitive for some lobby, I would prefer to be judged by an allotted jury than by an elected judge – probably this one will have higher level of general cognitive abilities, but I will trust the jurors to decide what is good for justice, without considering reelection, career, militant group or donors sensitivities, and I will hope a reduced influence of the lobbies, at least if the trial includes serious deliberation.

    Liked by 1 person

  6. Andre,

    I agree with your comments. I would make one reservation: I think the main obstacle the “public spirited” candidate faces is that of getting themselves and their agenda well known enough to even have their proposals evaluated by the voters. Once such a candidate breaks through this initial barrier, then they have a decent chance of garnering and maintaining the support of a non-negligible part of the public.

    Like

  7. Yoram,

    >Once such a candidate breaks through this initial barrier, then they have a decent chance of garnering and maintaining the support of a non-negligible part of the public.

    I agree. It would be a good idea to make that initial barrier as low as is practical. In a conventional elected assembly a threshold is necessary due to the problems associated with coalition building in fragmented legislatures. But if the power to vote on proposals were taken away from the elected officials and given to an allotted sampling there would be no need to form coalitions to pass legislation. The initial barrier for entry could be less than 1% nation-wide support without compromising the ability of the assembly to function. The ratio between the various parties would matter little (assuming reasonable rules for speaking time) and there would be no need for tiny parties to compromise their views or feel the need to frame their arguments in such a way to make them compatible with the views of their likely partners.

    Like

  8. Andre,

    Thanks for the clarification. I think we need to draw a firm (albeit analytical) distinction between system constraints and self-seeking (economic, sociopathic, or whatever) behaviour. The problem with the word “interests” is it straddles both aspects — it’s clearly in the interests of a public-spirited elected official to work with the system in order to achieve her benign policy objectives and it’s also in her interests to climb up the greasy pole of career advancement. Some of the more hysterical posts on this forum have privileged the latter (all politicians are venal creatures who are only interested in feathering their own nest) and this isn’t helped by Yoram’s use of scare quotes for a “public spirited” candidate. As you rightly point out these two aspects tend to overlap in practice, hence your use of the Darwinian metaphor — a theory whereby character is shaped by structural constraints — but for the sake of understanding we do need to maintain the analytical distinction, so let’s assume that the word “interests” refers to self- or group interests as opposed to the need to adapt to structural constraints.

    Yoram,

    >I think the main obstacle the “public spirited” candidate faces is that of getting themselves and their agenda well known enough to even have their proposals evaluated by the voters.

    As Naomi has pointed out, the separation of advocacy and voting should help resolve this problem (within a PR based electoral system).

    Like

  9. Naomi,

    IMO, the entry barriers currently common in electoral systems are due simply to them serving the interests of the large parties, not because they are conducive to good government. An extreme case, of course, is the FPTP system,

    In any case, the entry barrier is an inherent feature of the electoral system and even if the barrier is lowered as far as possible (i.e., to a single MP) it is still the main feature of the system and implies the Principle of Distinction, i.e., a non-democratic government. Such an institution should be abolished completely rather than tweaked in various ways.

    Like

  10. Yoram:

    The electoral system . . . implies the Principle of Distinction, i.e., a non-democratic government.

    Just because a body of [men] might not look like [America] (to paraphrase the 4th and 42nd presidents respectively), that doesn’t mean it cannot act like America. Democracy means the people have power, or that governance is in their interests and there is no intrinsic reason why this cannot be achieved by electoral means (whether or not its possible in practice being an entirely separate issue).

    Like

  11. Yoram,

    Larger parties do of course support electoral reforms aimed at limiting minor party participation, but enough countries have negligible thresholds for entry to where we can study the effects. Having a large number of minor kingmaker parties is deeply unpopular pretty much everywhere, just as American presidential elections would prove unpopular if the outcome was generally unpredictable based on the vote totals. It’s also very difficult for a highly fragmented parliament to function well. All the players want to prove they were worth the votes they received and tend to use every bit of leverage they have at every opportunity to get concessions. Brazil’s 28 party legislature is a fine case study in this if you are interested. Extreme unpopularity, dysfunction, and pork-dominated budgets are the result.

    If I remember correctly you expressed some measure of support for Bernie Sanders. He’s attained a very respectable share of the public’s support despite the aforementioned institutional limitations. If there were others, and if you could realistically expect to always have one or more elected advocates pushing your views and proposing legislation for votes in the sampling… wouldn’t that be just fine? Those espousing positions which can pass the allotted assembly can pull down enough votes to get a seat at the table. The difference between the general public’s support and the support in an allotted assembly drawn from the general public is not going to be a factor of fifty.

    I can respect the point of view that we should make the legislature look as much like the general public as possible and that we should give everyone an equal chance at participating in the political process at the highest level. However, the question of what happens at the edge of the statistical significance threshold is troubling. Keith’s fear of demagoguery—whether you agree with the concern or not—is an example of this. It’s not the only one. I also worry about the representation of local issues and the consequences of horse-trading below the significance threshold. At the same time I can’t see people giving up the ability to participate directly in the political process. People like participating—or at least they like having the ability to participate. Though perhaps this is due to a lack of trust in their elected officials.

    Having an aggregative process for the introduction of legislation and for debate before the sampling addresses these concerns. As long as the threshold for entry is low I don’t see any problems. If the threshold were high, or if you needed a majority of the elected advocates to make a proposal, then yes, there’d be a bottleneck which would limit the scope of the proposals reaching the sampling. But in the sort of arrangement Keith and I prefer, the only bottlenecks are the 1% (or less) needed for basic entry and the need for proposals to gain the support of the majority of the sampling.

    Like

  12. Naomi,

    > Extreme unpopularity

    A parliament would have to have negative approval numbers to be significantly less popular than the US Congress with its two parties. To provide a counter-example, Israel has been gradually increasing its parliamentary entry barrier and all the while confidence in its parliament has been eroding. I’d again suggest that if you are serious about your thesis that some electoral systems are noticeably better than others you should offer some definite ways by which such an association can be measured. Of course, picking some arbitrary examples and making assertions about them is not such a way. It would be even more interesting if, having specified such ways for measuring the claimed associations, you would also provide the data that does indeed show support for your thesis.

    > Bernie Sanders

    Sanders is obviously the exception. Both theory and evidence show that it is not parliamentary entry barriers that are the main reason this is the case. It is rather an inherent feature of electoralism.

    > However, the question of what happens at the edge of the statistical significance threshold is troubling.

    I don’t think so. I think this is a theoretical triviality that most people would care little about.

    > People like participating

    Elections are not a meaningful form of participation, as most people realize, despite the ceaseless indoctrination. A sortition-based system would provide much more democratic avenues of participation, since affecting the public’s opinion would be directly translated into affecting policy.

    > As long as the threshold for entry is low

    There is no way to make the threshold for “entry” low. Since serious discussion would have to be limited to a modest number of proposals, there would have to be some way to narrow down the field of proposals to this modest number. A threshold for entry to a short list is necessarily high. This is another manifestation of the Principle of Distinction. There is no way around it.

    Like

  13. Naomi,

    Your points are not unreasonable, but I have some other concerns with an elected body (even with low threshold proportional representation) being the source of proposals for a jury to act upon. I don’t like the idea of certain proposals (those coming from large parties) having an unfair advantage….simply because many jurors will take the mental shortcut of rubber-stamping proposals from a favored party, and not exercise independent judgement and due dilligence. Psychological research has shown that participants with a partisan preference strongly tend to endorse any idea they believe is supported by leaders of their party (the research is intriguing and stunning). Perhaps if the jurors were not told which party proposed a bill, there might be a work-around.

    By the way, as a personal aside, Bernie Sanders has been a friend of mine since the 1970s. After we both ran losing campaigns as minor party candidates, we were jointly elected to city government in Burlington, VT in 1981 (he as an independent and I as a Citizens Party candidate). I get my one second of fame when my photo appears with him in some of his background ads on TV.

    Like

  14. Yoram,
    For what it’s worth, it’s not my thesis. It’s conventional wisdom. Institutional design choices have practical consequences. Any book on electoral systems, even a really crappy one, is going to spend some time talking about the deleterious effects of fragmentation. Accountability is maximized when there is a single actor calling the shots who can be blamed or credited unambiguously. Representativity is maximized when the threshold for entry is low. Obviously, these are both desirable and mutually exclusive so a middle-of-the-road approach is almost certainly optimal for a conventional elected assembly. The US Congress has substantially more than two effective actors due to poor party discipline. While having many actors is necessary for a high degree of representivity, the two are obviously not the same. Congress has the worst of both worlds.

    >Sanders is obviously the exception. Both theory and evidence show that it is not parliamentary entry barriers that are the main reason this is the case. It is rather an inherent feature of electoralism.

    He’s no exception. Social democratic parties which push the same policies and use the same rhetoric as Sanders are not strangers to power in many European countries. I would point out that all of the examples of electoralism we have to work with involve the formation of majorities nominally supported by the voters. Broad compromise and coalition building can be pre-electoral or post-electoral but it’s always there in one form or another. The proposed system a very different beast.

    >I think this is a theoretical triviality that most people would care little about.

    Most people—myself included—care little for theory, but care greatly about outcomes. Which is what we are talking about, of course. Theory is the only tool we have to predict the behavior of untried systems. As we have no experience with full-blown allotted legislatures your own beliefs about them are entirely theoretical in nature as well. So we are comparing conflicting theoretical positions. Your position (if you’ll forgive me for couching it for you) is that if the interests of the members of sampling match the interests of the population at large then we can expect them to act in accordance with the interests of the public at large. This is perfectly reasonable. However, we need to to keep in mind that a sampling of a few hundred drawn from a national citizen pool is only representative over a narrow range of national-level issues. It would not be democratic to have an allotted legislature consisting of 10 people because representation in such a sampling would be random and the members could not be reliably expected to further the interests of the general population. Local issues, the members’ personal pet issues, or any other issue that falls below the statistical significance threshold is *exactly* the same sort of thing and will result in outcomes which are random, not democratic. If the members can cut deals involving such issues then all policy areas will be contaminated. We need to have clear segregation between advocates and voting members, with the advocates knowing as little about the individual voting members as possible. In this case, because the non-representative interests are random and unknown, the advocates would not be able to successfully introduce legislation to further those interests.

    >since affecting the public’s opinion would be directly translated into affecting policy.

    If this were the case, wealthy interests, the media, and so forth would have more influence than they do now. I imagine you expect them to have less.

    >A threshold for entry to a short list is necessarily high

    We both have short lists. To get entry into yours you need enough popular support for allies to be consistently drawn into the short list. The threshold is given by the size of the list in both cases. With an elected advocacy you get greater consistency and even-handed representation.

    Like

  15. Terry,
    A fair point. Some portion of a sampling will support a proposal introduced by their party just because they were the ones who introduced it, that’s true. But won’t another portion will oppose it for the same reason? I’m not familiar with the research in question, so I really don’t know. A proposal made by a minor party will start out with fewer supporters but also fewer opponents. Also, if the party system is highly fragmented even the big dogs won’t necessarily have all that much of an advantage. The largest party in Brazil holds 13% of the seats in the lower house, for example.

    Like

  16. Terry: >Many jurors will take the mental shortcut of rubber-stamping proposals from a favored party, and not exercise independent judgement and due dilligence.

    Political parties under the sort of proposals that Naomi and myself are making would be very different creatures from at present. They would have no decision or executive power and would merely be organised advocacy groups. Once the system was bedded in, allotted decision makers would understand that they are the judges and would be likely to be both diligent and independent in their judgment. I think the example of trial juries is relevant — although the influence of charismatic advocates cannot be overlooked, jurors seek to return a verdict on the basis of their consideration of the evidence and there is no particular reason to think that political juries would be any different. That was certainly the case with 4th century Athenian juries, and I don’t see why the reincarnation should be any different.

    Like

  17. Keith,
    I could certainly see parties breaking down along single policy domain lines. There’s no real advantage to voting for a major party that generally supports what you support. It’s not like they’re going to do any more good with one more seat. There is, however, something to be gained by voting for a minor party pushing a single under-represented issue. We could even give people multiple votes, so they could vote for their top several parties. People could vote for a general party to represent them in general (if desired) and a few single-issue parties simultaneously. Let voters pick a political salad bar of single-issue parties. There are other tricks we can play as well. We could cap the number of seats won by the largest parties. If it’s impossible to win more than 5% of the seats then any built-in advantage for large parties (due to party loyalty in the sampling) will probably be negligible.
    Even if there were only a hundred advocacy seats, the de facto threshold would probably end up being considerably lower than 1% due to a large chunk of votes going to parties with support levels exceeding the cap.

    Truth be told, I really have no problem with conventional parties. The best examples of electoral democracy in this world aren’t all that terrible, IMO. If the end result of our changes is not enough, then surely there will still be popular support for further reforms and better tools will be in place to explore them. Micro-parties will surely favor capping the biggest parties and giving voters multiple votes, if such things were really viable. Even under FPTP you have *some* micro-party representation and that’s all you need to get to a vote. There’s no need to shoot for perfection in one swoop.

    Like

  18. Naomi,

    > It’s conventional wisdom

    You’d think that since this is the case then it would be easy to find convincing empirical studies with evidence for this thesis. I am not aware of any, and it seems you are not aware of any either. I have a growing suspicion that there simply aren’t any.

    By the way, having some correlation between institutional parameters and quality of government is not unlikely. My claim is that both theory and historical evidence indicate that any parameter settings within the electoral framework will result in low quality government. Of course, there could still be systematic variation within the low quality zone.

    > The US Congress has substantially more than two effective actors

    Seems like an ad-hoc patch in the theory, which would be hard to operationalize. How would you be able to objectively measure the number of “effective actors”?

    > Social democratic parties which push the same policies and use the same rhetoric as Sanders are not strangers to power in many European countries.

    The label “social democratic” often is no more than a marketing ploy. I would be interested in evidence that Sanders-like rhetoric and policy proposals enjoy significant electoral representation in Europe. Certainly, the evidence is that policy on economic inequality bears no relation to public sentiment.

    > If the members can cut deals involving such issues then all policy areas will be contaminated.

    This is not a matter of statistical significance then but of logrolling. We have discussed such issues before. I think logrolling is a perfectly legitimate political activity and I see no reason to assume it will result in bad policy.

    >> since affecting the public’s opinion would be directly translated into affecting policy.
    > If this were the case, wealthy interests, the media, and so forth would have more influence than they do now. I imagine you expect them to have less.

    By “the public’s opinion”, I was not referring to what goes in the electoral system by the name “public opinion”, but rather to the public’s informed and considered opinion. While the former is easy prey to manipulation and determines electoral results, the latter is what in a democracy sets policy.

    > We both have short lists.

    I am not sure what you mean. The short list that matters is the short list of proposals that can actually be considered rationally and in an informed manner by policy makers. The threshold for entering a proposal into this short list will necessarily be high.

    Like

  19. Yoram,
    This is political science textbook-level conventional wisdom. I would be more than happy to recommend a few outstanding titles to you if you actually care. If pushing sortition is something you intend to do for the foreseeable future it might be a good idea to spend some time studying electoral systems and mainstream political science (particularly comparative institutions) in general. In any case, comparisons to elected governments are a non sequitur (at best) because we aren’t actually talking about an elected government. We’re talking about requiring a whopping 1% of the general public’s support to get access to the sampling. This is not a big deal. Anything that could get 50% in the sampling can get 1% in the general public. For an elite to monopolize power they’d need 99% of the vote, which is silly.

    >I would be interested in evidence that Sanders-like rhetoric and policy proposals enjoy significant electoral representation in Europe.

    The Wikipedia page on social democracy gives a good overview of their historical successes. You really don’t think these parties can reliably pull down a few percent of the vote? Come on now.

    >This is not a matter of statistical significance then but of logrolling.

    No, not exactly. As I’ve emphasized before, I’m also a big fan of log-rolling. The point is that it is undemocratic for a sampling to act outside the realm of its statistical significance. Can we agree to that much? In this case log-rolling ensures that no policy areas escape contamination.

    >The threshold for entering a proposal into this short list will necessarily be high.

    Yoram, we are both reducing the number of would-be participants to a similar number. You are using a lottery, I’m using a ballot box. It’s a similar degree of reduction. Your short list eliminates viewpoints at random. Major views are sure to be represented, as are a large number of very minor ones purely at random. Mine excludes those that can’t muster a trivial fraction of the vote while still giving them a proportional share of the leverage over the advocates who do make it onto the list.

    Like

  20. Naomi,

    > political science textbook-level conventional wisdom

    Unfortunately, this says little about its basis in fact. I am interested in evidence, not dogma.

    > We’re talking about requiring a whopping 1% of the general public’s support to get access to the sampling.

    Meaning you need to get the signature of 1 of every 100 people in the country. An impossible barrier for the average citizen. But the crucial point is that this is not an accident – it is essential. If getting proposals qualified for discussion were easy, there would be too many of them to discuss.

    > You really don’t think these parties can reliably pull down a few percent of the vote?

    I don’t think those parties are really presenting a democratic alternative to the elitist mainstream. The standard maneuver is for an elite group to use socialist rhetoric to gain electoral power and then collude with the establishment politics. Syriza is a prominent recent example. The time scale varies.

    > it is undemocratic for a sampling to act outside the realm of its statistical significance. Can we agree to that much?

    I don’t know what that means.

    > You are using a lottery, I’m using a ballot box.

    Yes – and that makes all the difference.

    > Mine excludes those that can’t muster a trivial fraction of the vote

    Where “trivial fraction” means somewhere between several tens of thousands of people to millions of people. This means exactly that elite proposals get a hearing while those of normal people, no matter how popular they are, don’t.

    By the way, it seems to me that the great emphasis that is being put here (and which is very common in the political reform space) on legislation is misguided. Laws are only what those who enforce them make of them. If the executive and the judicial bodies are not democratic then the power of legislation is very limited. It is a crucial part of a sortition-based system, for example, that the executive is hired, closely monitored and fired by an allotted body.

    Like

  21. Naomi,

    I’m strongly attracted to your flexible view of the political party of the future. If representative isegoria is the principle involved then there’s no reason for a priori constraints — your vision of multiple votes for small single-issue parties is very similar to my suggestion for the public votation to choose from a menu of successful petitions and citizen initiatives, the criterion being that all citizens should get to express their preferences, rather than just a randomly-selected handful. I also agree that Yoram would benefit from reading some political science textbooks rather than dismissing an entire field of knowledge as electoralist dogma. It’s also amusing that both Yoram and Terry view Bernie Sanders as the exception that proves the rule, as it looks to me more like a case of partisan preferences. As to whether or not Sanders’s policies would meet widespread approval in European polities, I can only point to the Jeremy Corbyn, who only really has the support of a relatively small number of hard-left political activists. It will be interesting to see the result of today’s by-election for a safe Labour seat.

    Like

  22. Yoram,
    Even if mainstream political science were just a bunch of dogmatic propaganda, you’d need to understand it if you want to dismiss it. Which means doing the background reading first. What you call collusion is, in truth, the simple result of facing hard practical realities. You need something to blame for the decline of the hard left and you chose elections. There’s absolutely, positively zero evidence to suggest an allotted government would not have done the same things in the same situations. You’ve filled the void of our practical experience with sortition with your hopes and dreams.

    Regarding the statistical significance issue, would you agree that it would be undemocratic for a sampling of 10 people to call the shots? Why? Because instead of interests being represented proportionally they would be represented at random. In a large sampling the major, national-level issues will be represented well and proportionally. The representation of views held by less than a percent or two of the public will be at random. There are about 20,000 towns in the US. Of course, many local issues will win the lottery and wield a few tenths of a percent of the political power. The most extreme example of non-representative interests winning significance would be the personal interests of the members. I care that my sister can pay her bills and live comfortably far more than I care about the vast majority of the issues I’d have to vote on in one of these samplings. All members would have interests that don’t merit anywhere near a whole seat. It’s just the way it goes. If we assume the members will act rationally to further their interests, as you do, we only get democratic results if the sampling can only act within the range of topics for which interests are represented proportionally though random sampling. If we have a division between voting members and advocates, preferably with a two-vote system, we get natural quarantining. The advocates won’t be aware of these non-representative details when they introduce legislation. They can only horse-trade over population characteristics likely to be represented in a statistical sampling. It’s a clean and robust solution.

    In your system, to gain representation a viewpoint either has to get lucky, and be represented for no rhyme or reason at all, or be shared by hundreds of thousands or millions of people. In my system, to gain representation a viewpoint has to be shared by hundreds of thousands or millions of people. The only difference is that in my system there needs to be some small degree of organization, enough to get tens of thousands of signatures to get on the ballot and be recognized by enough would-be supporters. This isn’t a big deal. Some of my favorite YouTubers have hundreds of thousands or millions of subscribers and probably agree with you on most policy points. Suggesting that the elite have 99% of the vote in their pocket is silly and you know it.

    Like

  23. Naomi,

    > Which means doing the background reading first

    I thank you for being concerned about my education, but it seems to me I understand the dogma just fine. The problem is not with understanding the claims, but with having no evidence to support them.

    > You need something to blame for the decline of the hard left and you chose elections.

    No, what I need is “something” to explain why popular positions do not get represented in government. Since elections are supposed to be the mechanism generating this representation, then clearly they are not working as it advertised.

    > You’ve filled the void of our practical experience with sortition with your hopes and dreams.

    It is true that the effects of sortition will not be known for certain until extensive application takes place and that therefore at this point we are dealing with educated guesses. This, of course, is true about any reform proposal that doesn’t suggest simply replicating an existing system. On the other hand, electoral and plebiscitary systems have been tried extensively and have generally produced poor results. I’d rather pursue non-existing systems based on an analysis of the expected results than give up and settle for the devil we know.

    > If we assume the members will act rationally to further their interests, as you do, we only get democratic results if the sampling can only act within the range of topics for which interests are represented proportionally though random sampling.

    Again, this is not a matter of statistical significance. This is a matter of whether the random sample adopts a view which puts its own interests above those of the non-allotted. This is what I called the level of background representativity. Yes, it is conceivable that such a phenomenon would exist, but (a) luckily it seems this not the case, (b) taking such a stance would risk retaliation by other (including future) allotted bodies, and (c) there is really no way around this issue. If the population is corrupt in the sense that any group of people with power immediately abuses that power, then all is lost. Certainly, having advocates would not help since they could either collude with the allotted, or, more likely, simply use their own privileged position to benefit themselves. By the way, if the advocates are themselves allotted or selected by an allotted body for short terms then there is no inherent problem with having advocates.

    > The only difference is that in my system there needs to be some small degree of organization, enough to get tens of thousands of signatures to get on the ballot and be recognized by enough would-be supporters. This isn’t a big deal.

    Of course it is. If it wasn’t there would be thousands of proposals on the California ballot every year. It is a big deal by design and it has to be a big deal – otherwise the whole process would become a farce.

    > Suggesting that the elite have 99% of the vote in their pocket is silly and you know it.

    The elite doesn’t have to have 99% of the vote in it pocket. It has to be able to collect signatures from 1% of the voters, while normal people can’t. By the way, collecting the signatures is just part of the process. Afterwards there is need for advertising money and mass media connections – again putting the elite at a huge advantage.

    Like

  24. [Ah – I forgot that you suggest having the final decision made by an allotted body rather than plebiscite. So supposedly “the advocates” will take care of the arguments for and against, so supposedly no need for advertising. But of course advertising could hardly hurt a proposal’s prospects. In any case, the matter of qualifying for the short list is unaffected.]

    Like

  25. Yoram,
    It strikes me as a matter of statistical significance. If the sampling numbered ten people, a 5% subdivision of the population would not generally be represented. And it shouldn’t, because that would give it disproportionate power. If the sampling numbered 300 people, a 5% subdivision would be properly represented but a 0.1% subdivision would not and should not. That 0.1% subdivision would occasionally gain disproportionate power and occasionally have no power at all. If the sampling numbered 10,000 a 0.1% subdivision would generally be represented but a 0.005% subdivision would not and should not. Of course that 0.005% subdivision would occasionally gain disproportionate power and occasionally have no power at all. Etc. The underlying problem that the resolution of a sample is limited.

    An elegant solution is to draw the voting sample after the proposal is introduced. Those who introduce proposals would know nothing of the individuals who vote on the proposals. They’d have to rely on attributes which are sure to be represented statistically. Collusion between advocates is not a danger if they are ideologically diverse, as would be the case with a very low electoral threshold, and if they can introduce legislation individually, rather than as a group. You are correct in that the advocates need not be elected. One could have allotted advocates and that would address the issue all the same. I believe it would be vastly better to have an elected advocacy, but that is an entirely different question.

    I would note that you don’t need signatures from 1% of people. You need 1% of people who cast ballots to put a checkmark by your name. Or less, if we have more than 100 advocates. Ballot access is a separate matter. Yes, it’s a concern and it absolutely has been used to keep minor parties from gaining a toehold in the past. I wouldn’t have a problem with citizen assemblies being used to approve ballot applications. Signature gathering works too and would be my first choice. The wild success of crowsourcing demonstrates that organizing like-minded regular people is not terribly difficult these days. A small and underrepresented subset of the population is not going to be thwarted by a modest signature requirement.

    Like

  26. > An elegant solution is to draw the voting sample after the proposal is introduced. Those who introduce proposals would know nothing of the individuals who vote on the proposals.

    As I wrote, I don’t think there is a problem in principle with such a procedure, as long as both groups – proposers and approvers – are representative, i.e., are allotted from the entire population and are given the resources and authority to make an informed and considered decision. On the other hand, selecting either group by mass voting is a non-starter. It is guaranteed to give elites disproportional power.

    Division of authority – in this way or in other ways – has advantages and disadvantages and therefore has to be designed iteratively and continuously rather than a-priori (as long as the basic principle of representativity is adhered to).

    Like

  27. Yoram:> If the population is corrupt in the sense that any group of people with power immediately abuses that power, then all is lost.

    That is the classical republican view, predicated on the notion of a virtuous citizenry. Modern republicanism (aka Madisonianism) assumes the need to design political institutions in such a way that they will function adequately in the absence of virtuous citizens. The “real” democracy that you advocate is suitable for a nation of gods rather than men (as Rousseau put it). A well-designed system of governance is mixed and requires parchment barriers, the principal ones being between the executive, judicial and legislative powers, the latter also being subdivided between advocates and judges, appointed by two entirely different methods of representation. Allotment is suitable for the latter, but the former requires election and Naomi’s suggestions as to how to modify the electoral principle to adequately reflect the views of minorities deserves serious consideration (and certainly should not be dismissed as “dogma”). Defining representative democracy as allotment (and thereby dismissing anything else as undemocratic) is a better candidate for this epithet as dogma is concerned with the way we use words as opposed to substantive issues.

    Like

  28. Naomi:

    >You need something to blame for the decline of the hard left and you chose elections. There’s absolutely, positively zero evidence to suggest an allotted government would not have done the same things in the same situations. You’ve filled the void of our practical experience with sortition with your hopes and dreams.

    This should serve as a salutary reminder to all of us to avoid the danger of wishful thinking, the principal failing of all utopian schemes.

    Like

  29. Keith’
    You wrote:
    >”A well-designed system of governance is mixed and requires parchment barriers, the principal ones being between the executive, judicial and legislative powers, the latter also being subdivided between advocates and judges, appointed by two entirely different methods of representation.”

    Having checks and balances is important, but I don’t think the traditional separation between legislative and executive is particularly helpful (there may be some practical reasons for it, but also lots of negative effects). The place where it DOES make sense (as you mention) is between Proposing and Deciding, Those who vote to adopt or reject should not suffer from either pride of authorship or jealousy of partisan power. To a SMALL extent the separate Court system nibbles around the edge of this, but unfortunately NONE of the modern “electoral democracies” (nor Yoram’s allotted legislature) enshrine this KEY separation of powers. As you have noted, Madison wrote about it, but he failed to work the concept into the Constitution at all.

    The main point that I disagree with you on is your insistence that the method of appointment of these separated authorities must be by “entirely different methods.” There is no logical reason for this, only tradition that sought to preserve some power for nobility, et. al. It is JUST as reasonable to use the same random selection method, so long as those selected form distinct and separate bodies. Like Yoram, I think using mass elections to select the proposers or agenda (mass election “votation” in referenda) undemocratically favors elites of high status, fame, and wealth, and undercuts democracy.

    Like

  30. Terry:> The main point that I disagree with you on is your insistence that the method of appointment of these separated authorities must be by “entirely different methods.” There is no logical reason for this, only tradition that sought to preserve some power for nobility, et. al.

    As a conservative I naturally have a lot more respect for tradition than a progressive like yourself. But this is for quasi-Darwinian reasons, similar to those recently pointed out by Andre, in that traditions tend to be based on arrangements that work in practice (although progressives will counter that it merely reflects military, political, financial and patriarchal power). It’s significant in this context that proposers and advocates in the Athenian demokratia were almost entirely drawn from an aristocratic background and, given the etymology of the term, it seems right that proposals should be made by the “best”. Yoram has recently acknowledged that hoi aristoi, unlike hoi polloi, do not lack competence, the problem is merely one of self-interest, so I would rely here on Harrington’s argument that the elite will have to make proposals that are in the general interest otherwise they will be rejected by the allotted assembly (this is standard neo-republican theory). And of course election, as has been frequently pointed out, is an aristocratic selection method, so it would seem to be the obvious way of determining the best proposals — especially in the light of Naomi’s innovative electoral scheme (and my own suggestion to supplement this with crowdsourcing and direct-democratic initiative).

    What I’m not aware of is any successful attempt to arrogate political proposal power to a statistically-representative group with no particular expertise in this field. Needless to say I accept the argument for cognitive diversity, but this can be tapped in other ways. And Naomi’s proposal is specifically designed to minimise the power of hegemonic political and financial elites, so I don’t really see why you would want to use random selection for a function that it is manifestly unsuited for. I think there is a tendency to view any form of election as business as usual, but Naomi and I both argue that it would take a much more benign form in the constitutional proposal that we are making. I can appreciate that it would be hard, psychologically, for you to acknowledge this, given your own provenance, but if Winston Churchill crossed the floor twice there’s no reason why you shouldn’t do the same! Bear in mind that, as the author of The Party’s Over I’ve also had to eat my words and I think we all owe Naomi a big debt in pointing out that we really don’t need to throw out the baby with the bathwater!

    Like

  31. Terry,

    PS I find Rousseau’s arguments for the separation of the legislative (moral) and executive (physical) power to be quite persuasive. What are the negative effects that you fear?

    Like

  32. Terry:> mass elections . . . favour elites of high status, fame, and wealth, and undercuts democracy.

    The word “elite” means “the choice or best of anything considered collectively, as of a group or class of persons”. We’re perfectly happy with this when it comes to athletes, musicians, businessmen, scientists etc. but seek to deny the epithet to those with proven political abilities. The Athenians did not think it undemocratic (notwithstanding the principle of ho boulomenos) for the vast majority of proposers and advocates to come from the ranks of the elites, because the final decision was left in the hands of the demos. At no stage did the Athenians use sortition to select political advisors, primarily for the reasons laid out by Socrates. It strikes me that a modern version of such a system, incorporating representative mechanisms throughout (on account of the problem of scale), would be as democratic as its Athenian template. Sortition, for the Athenians, was one mechanism amongst many, and they would have been puzzled by modern attempts to design a system of governance using a single principle (especially if they were aware of the implications of the Dunning-Kruger effect). The Stone injunction means that we cannot use the M word, but the only way I can make sense of your abhorrence for elites is in the light of your own political provenance (the old photo that I saw of you and Bernie Sanders did refer to you as a M******).

    Like

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: