I like you as a voter

Ever since Socrates

It is a long standing tradition to deride sortition for putting in power unqualified people. The critics of sortition interviewed by Kevin Hartnett carry this tradition to the present.

Whether it is because the average person is incorrigibly incompetent, or just because they are inexperienced, the bottom line is the same: you just cannot hand power to the average person and expect good government. Socrates put it this way:

[N]o one would care to apply [sortition] in selecting a pilot or a flute-player or in any similar case, where a mistake would be far less disastrous than in matters political.

The straightforward argument is that the unqualified would simply make poor decisions:

There are ways in which we want our elected officials to look like us and then there are other ways in which we want them to be better than us. We actively try to select for some skills and talents when we choose politicians. (Susan Stokes, professor of political science at Yale University)

A more sophisticated version of this argument is that the unqualified would be manipulated by someone more competent or more experienced – lobbyists, bureaucrats, demagogues – into serving their purposes:

Basically what would happen is that regular staff and the president would become more powerful. You really do need to know something to pass legislation. By the time these [lottocratic legislators] learn where the bathrooms are, they’d have to leave. (Bruce Cain, professor of political science at Stanford University)

Some of my best friends are average people

Socrates may have had in mind something along the lines of Plato’s guardianship as an alternative to sortition. This would be akin to suggesting today that government should be handled by academics. Whatever are the drawbacks of such a proposition, it is at least consistent as an outright rejection of democracy.

Those who prefer elections to sortition, as modern sensibilities require, are in a more delicate position than Socrates since they do not reject popular judgement entirely. Supporters of elections take the position that while the average person is able to judge the abilities of a government official (prospectively or at least retrospectively), he or she would not be able to carry out the functions of the official themselves.

The truth of such claims is certainly very far from clear. It seems much more likely to me that the opposite is true. I would argue that any person would find it very difficult, working with the limited and distorted information available to the public, and with the amount of attention that a rational person would invest in the matter, to tell whether a person would perform well, or has been performing well, as a government official. At the same time, I would argue, the average person would be able, given adequate resources, opportunity and motivation, to perform reasonably well as a government official.

But the major problem with the position of election advocates is that it is inconsistent even if it is assumed that their factual claims are true. The point is that the allotted chamber does not have to function as a substitute for the elected chamber. If the allotted chamber chooses to do so, it can limit its role to selecting government officials and in this way simply function as a substitute for the electorate. If an average person can, as a voter, make a reasonably good selection of government officials then the members of the allotted chamber should be able to make a selection that is at least as good. In fact, with the added resources available to them and with the added incentive of having significant impact on the final outcome (due to being one of a small group of electors instead of being one in a group of millions of voters), the selections of the allotted delegates should be much more well-informed and well-considered than the selections of the electorate as a whole.

Therefore, even if – as sortition critics argue – the average person is incompetent at governing, and their ability is somehow completely limited to selection of government officials (a limitation that seems very unlikely) this would not constitute a valid argument against sortition. It would simply imply that allotted delegates should limit their activity to selecting government officials rather than taking any more hands-on roles.

In order to prefer elections to sortition, it would be necessary to argue that while most people are able to select well as voters, those same people are somehow unable to make good choices as members of an allotted chamber.

I don’t doubt that such claims would be forthcoming, since the conclusion that sortition is to be avoided is a foregone conclusion for some, but at least let’s have those claims made explicitly rather than implied.

16 Responses

  1. Some might argue that humans evolved an ability to judge the trustworthiness of others, but NOT the ability to judge tax policy. The flaw with this argument, of course, is that that ability is tied to personal contact at a tribal scale in which it evolved, not media created images designed to capitalize on those natural proclivities in order to deceive. And of course, just as humans have evolved an imperfect ability to detect untrustworthiness in others, we have also developed (in a sort of evolutionary arms race) a natural and learned skill for deceiving others about our own trustworthiness. It may be that judging non-human “tax facts” is easier for us than judging human characteristics BECAUSE we didn’t evolve knee-jerk predispositions about them.


  2. I just today happened to meet an old friend who just completed two years as staff director of a Congressman. He lamented how carefully he had had to learn to speak … in order for his Congressman to be re-elected (and re-elected and re-…etc.). / Emblematic, he said, regarding the ‘fiscal cliff’ decision: “…only 5 or 6 people are making the decision” // And of course we’re not thinking about choosing a pilot or a surgeon by lot … those are executive functions (despite the Greeks having done so!)


  3. > Some might argue that humans evolved an ability to judge the trustworthiness of others, but NOT the ability to judge tax policy.

    Well, my point here is that even if we accept this idea it still doesn’t make a valid argument against sortition.


  4. I very much agree with your analysis. I would stress that claiming that we are not able to make good political decisions but somehow are good at choosing who will make good decisions for us is a weird claim indeed, with some serious consequences. It means that it isn’t in any way undemocratic if a representative makes choices totally different from what he promised during the electoral campaign, after all we aren’t in a position to judge political policy, we did not choose them because of what decisions we thought they would make, but then based on what *are* we supposed to decide our vote? Further, it would mean that the ability of politicians to make decisions against the will of the citizen without facing any consequence is a desirable property of democracy, and it will be used only for the good of the people. Seriously.

    If we really have this magic ability of knowing who to trust, if selected by sortition we could just call the person we know we can trust blindly and follow its advice. Or as Yoram says let the allotted body select the decision makers.

    Now, I do agree that on average we are not very good at making decisions, and that this is a big weakness of democracy (and thus of sortition). Only it applies just as much to electoral democracies, so it is in no way an argument to favor the current system as opposed to sortition. *Unless* one argues that currently citizen aren’t given much of a choice, and elections are a just a farce, real decisions are taken elsewhere, which I must say probably is not that from the truth.

    Of course I’m preaching to the choir here, but I think it is an incredibly important point. Either electoral systems suffer just as much from incompetence or they are not democracies at all, and only the interests of a few matter. In either case serious political reform is due, and sortition might well have great potential of being part of the solution. In my view this and the notion of statistical representation are at the core of why sortition is so attractive.

    The incompetence argument is the single most common criticism to sortition, unfortunately the answer might be a little non obvious, but I think there arguments against the criticism are pretty solid once they are understood.


  5. [ This comment refers to a comment that was moved: https://equalitybylot.wordpress.com/2012/12/07/the-veto-nonsense-and-unanimity ]

    I’m not really sure what exactly you are trying to say. I think the premise, with which I agree, is that under majority rule a majority of fools can easily overrun a minority of well-informed. And certainly if you want the opinion of the very very best, they will by definition be a minority. So you will have to be content with hoping they will be able to convince the others of their motives.

    I’m however quite confused of how changing the majority rule to require other percentages is supposed be of any help. Requiring veto and in general requiring majorities greater than 50% has the only consequence of change more difficult, whether it is good or bad change, which in some situations might be a good thing but in general is not really desired of a body that should make decisions.

    As for the boat example, imagine there is a hole, and you require the crew to vote whether to fix it. I’m pretty sure you don’t want veto right here.

    750 votes necessary for proposals that leave things alone

    uhm… and what exactly is supposed to happen when the proposal to not change anything doesn’t pass?


  6. Fela,

    > Either electoral systems suffer just as much from incompetence or they are not democracies at all, and only the interests of a few matter.

    The latter position is the Schumpeterian (or elitist) theory of electoralism (or of democracy as he called it). It seems to me that it was the standard scholarly view in the third quarter of the 20th century. At that time it was seen as not only being descriptive but also normative.

    Over time (particularly after the civil rights struggles of the 60’s) popular opinion became less tolerant of political elitism, and as a result the idea that the situation described by the Schumpeterian theory is desirable fell out of fashion in academic circles. Following this normative re-evaluation, the theory was abandoned as being descriptive as well, since it is inconvenient for the establishment to admit that the existing system is normatively problematic.

    Today, it is essentially absent from mainstream discussion and we are back to using what Schumpeter called the classical democratic dogma (more accurately, the classical electoral dogma).


  7. >In order to prefer elections to sortition, it would be necessary to argue that while most people are able to select well as voters, those same people are somehow unable to make good choices as members of an allotted chamber.

    But there is a “democratic” factor involved, in addition to epistemic considerations such as competence and making good choices — namely that the allotted chamber would decide in the same way as everyone else would. This is the argument for voting under conditions of universal suffrage so, if the argument for sortition is to carry, it is also dependent on the interchangeability hypothesis* as well as on epistemic arguments. This would suggest certain constraints on what an allotted chamber could do. Selecting government officials would not contravene the interchangeability hypothesis so long as the longlist was externally provided, not so if it depended on the speech acts of individual members.

    * The interchangeability hypothesis claims that any number of allotted assemblies would deliver the same output, irrespective of which empirical individuals are selected. The hypothesis presupposes identical inputs, thereby ruling out the speech acts of individual members.


  8. Keith, if I understand you correctly you are saying that an allotted body selecting government officials should be considered democratic only if different allotted bodies would select the same officials. I guess that this can never be 100% true, but arguably appropriate conditions would make the choices fairly similar and interchangeable. In any case it seems to me that the result would still be a lot better than in the case of classical elections, where the results largely depends on factors totally out of control of the citizen.

    It is debatable what “appropriate conditions” exactly means, but an external long list as you propose and conditions to avoid too much influence from single individuals could be good ideas…


  9. Keith,
    Your “interchangeability theorem” doesn’t work for ELECTED legislatures. An election held one day later or earlier could generate a body that makes opposite decisions. Even the SAME legislature can make opposite decisions from one day to the next.

    It seems to me that this standard is too high. Often there are a few thousand democratically acceptable decisions that might be made on a single matter, and many millions of democratically unacceptable ones. There are also perhaps hundreds of epistemically good options and millions of bad ones. We are looking for the sweet spot of making decisions that fall in the overlap of these two criteria, which a diverse allotted chamber using expert advice seems most likely to achieve.


  10. As I have already pointed out, Keith’s arbitrary demand that an allotted body must make exactly the same decisions over multiple allotments is a senseless formalism. What is important is that body’s decisions serve the public.


  11. I’ve never claimed that the interchangeability theorem applies to elected legislatures, as the mandate of a successful candidate in a preference election is reliant on the number of votes received. No such claim can be made for an allotted assembly, hence the recourse to statistical representativity. Bear in mind that I’m pursuing the strong claim that the descriptive mandate embodies the informed consent of the whole citizenry, hence the need to ensure consistency between each sample.

    Yoram’s criterion — decisions that “serve the public” — would appear to be epistemic rather than a concern for democracy as a normative goal. If this is the case then how does one determine which decisions serve the public and which ones don’t, and how is it possible to argue that such decisions embody the informed consent of the whole citizen body? The interchangeability hypothesis is a way of putting the second claim to the test — if the decision outcome is the same irrespective of which citizens are allotted, then informed consent can be strongly implied (as the decision would be the same, irrespective of whether or not we each attend in person).

    As for the first (epistemic) claim, it’s only really possible to find out with hindsight which decisions do and which do not serve the public. I agree with Terry that a diverse allotted chamber using expert advice is the best way of getting the right answer (epistemically speaking) but if we have any regard for democratic consent we would need to ensure that the policy agenda and expert advice is consistent, irrespective of which individuals fall into the sample and this would inhibit individual speech acts (other than asking questions).

    I’m reading Bob Goodin’s 2008 book and take heart from his claims that both in the case of the Bloomfield Track citizen jury and Fishkin’s DPs the information phase is far more important than the interactive deliberation in generating optimally informed outcomes. This would suggest that the interchangeability requirement is not such a “senseless formalism” after all, as balanced information and “deliberation within” is perfectly adequate. Andy Dobson is working this up into a book which is a plea for deliberative democracy to shift its focus from ideal speech acts to encouraging the art of listening.


  12. > If this is the case then how does one determine which decisions serve the public

    Have an allotted body make this determination.


  13. Yoram,

    An allotted body may attempt a) to discern what is in the public interest or b) individual members may choose to privilege their own interests. In all likelihood there would be a combination of a) and b)

    a) Discerning what is in the public interest
    There is no way of knowing what “works” (epistemically) until the decision has been made and the consequences experienced. As most allotted members will not have any particular expertise in the area under consideration it is likely that they will be influenced by those members who claim (rightly or wrongly) to have the requisite expertise and/or social or professional status. It’s also possible (or even likely) that the deliberations of a group of inexpert legislators will fall prey to groupthink, particularly if the choice of information and expert advocacy is under the group’s own determination. Bear in mind that the necessary conditions for the epistemic success of the Wisdom of Crowds/Condorcet Jury Theorem is cognitive and information diversity AND independence. The latter will be corrupted by the above factors, therefore the design of allotted assemblies has to be such as to minimise these influences and to ensure information diversity. According to Goodin (2008), the most important factor is balanced information, not deliberation and this presupposes externally-provided resources.

    b) Allotted members deciding in their own interests
    Although allotted members are exhorted to legislate for the general good, there is no particular reason to think that they will successfully bracket out their own individual interests. In an assembly with voting powers only this is unproblematic as the statistical aggregate of the votes will mirror the aggregated interests of the larger community. Not so if the allotted mandate includes individual speech acts as this will mean that the will of those citizens with the best ability to articulate their own interests will predominate. Some may well decide that it’s in their interests to accept inducements from lobbyists to propose and/or advocate policies that are in the lobbyist’s interest and this will further distort the process.

    Thus deciding what decisions serve the public is a non-trivial problem that will not be resolved automatically by the allotment process. There really is no magic bullet with which to shoot down the “senseless formalisms” of those who seek to take seriously the problems of allotted assembly design.


    Robert E. Goodin (2008), Innovating Democracy: Democratic Theory and Practice After the Deliberative Turn (OUP).


  14. […] argument regarding the lack of technical expertise of the allotted is a standard – in fact, the standard – argument against sortition and is easily refuted. But in any case, it is hard to see how […]


  15. […] In fact, analysis shows that the claim that this is the crucial difference between the positions is rather questionable. In fact, it is elections that imply that the elite and the population are substantively similar, […]


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 )

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: