Thursday, December 27, 2007

What is Democracy?

I think it is possible for majority rule to be, in an important sense, undemocratic. Imagine a society split 60/40 into two comprehensive factions, such that people in the same faction always vote together, and against the other faction. In such a situation, I think it would be misleading to describe a system of majority rule as 'democratic'. It is not the people (generally) who rule here; what we have instead is a mere oligarchy, however large: 'rule by the majority faction'.

What more is required for democracy, then? Total consensus is an unrealistic ideal, and democracy still ought to be possible in the face of robust disagreement. I'd suggest that we instead understand 'rule by the people' to mean that everyone is able to make a meaningful contribution to the collective decision-making process, over time. The votes of a permanent minority are pointless, as they never have a chance of making a difference. But in a more flexible political culture, "the majority" is constantly in flux. Each person will be in the majority on some issues (and in the minority for others), so their will is at least sometimes heeded. In this sense, they all contribute to the state's decision-making. Even if they do not always get their way, there is still a meaningful sense in which we can describe this as a polity ruled by all (diachronically).

However, it is consistent with such 'diachronic democracy' that everyone be completely dogmatic. At least there are no stable factions, and thus no consistently oppressed (or effectively disenfranchised) subclass of the citizenry. But it would still be the case that for each particular decision, those in the minority were simply disregarded, their "contributions" effectively nullified. The system is effectively a rotating oligarchy, where everyone gets to take a turn.

This raises the question: is synchronic democracy - rule by all at once - possible? I want to suggest that it is possible, so long as the political system is sensitive to and responsive to reasons that any may put forward. In the absence of faction and dogmatism -- better, in the presence of civic respect -- even those who are initially in the minority have a real chance of affecting the outcome, by convincing others of the virtues of their position. Since the outcome is influenced (ideally: determined) by the strength of reasons, and these reasons may be contributed by anyone, it follows that any can make a meaningful contribution. 'Democracy' in the fullest sense is thus realizable in the form of deliberative democracy.


  1. I have the impression that the best way to define democracy is not through a prescription for the decision-making process, but rather as a kind of behavior that a people (including its government) may or may not exhibit -or may exhibit in different degrees. The characterization of that behavior may be discussed, but some minimum requirements would be (i.m.h.o.)

    - main orientations of policy according to the will of the majority
    - respect for minorities
    - not only tolerance, but actual valoration of dissent and criticism
    - no power without accountability

    I don't think that such things can be hard-wired into a constitution, and on the other hand, if they are accepted by a large proportion of a population, then more than one constitutional agreement would do.

  2. Why would you expect democracy to be more likely to be "responsive to reasons" than other systems of government? Assuming (as seems the case to me) that the choices are between deliberative oligarchy, non-deliberative oligarchy, and non-deliberative democracy, how do you rank-order your preferences?

    If what you really want is "responsiveness to reasons" why not demand that directly? If you want democracy, then why not use the traditional definition, in terms of suffrage, rather than making up a new definition that corresponds to everything you see as good at once?

  3. I like this train of thought. One worry, though, is that different people may have different standards of rationality. In that case, is there a rational way of resolving differences between conflicting standards of rationality? (E.g., say that there is conflict between a Kantian and a utilitarian, construing this conflict not as moral, but as rational, e.g., a deontological mode of reasoning contrasted with means-end reasoning.)

    Another related worry: suppose that one standard of rationality is correct, and that is utility-maximization (paired with a measure of collective welfare). If deliberation can be reduced to this standard, possibly the only benefit of discussion is to make collective choice consistent (e.g., ensuring a transitive rank-ordering of preferences). The selection of best means to achieving this end would be an engineering problem best left to technocrats. It's hard to see what's democratic about this, it may just as well be oligarchic, which is the problem I think Michael was getting at above. Perhaps there's a role for deliberative democracy (as opposed to voting) in avoiding collective preferences that are cyclical, but this means that deliberation should be something more than utility-maximization, involving deliberation about ends, and not just means to ends.

    A second related worry: wouldn't deliberative democracy privilege us philosophers over, say, artists or comedians? Then how is it different from an oligarchy? The worry is that favoring one standard of rationality, even one rational way of adjudicating between different standards of rationality, would end up favoring one group over others. Philosophers may not mind this, but perhaps not. In Madagascar, the traditional decision-making procedure is to let the person with the funniest/wittiest remark dictate policy. This presupposes a standard of rationality that favors quick-witted comedians over philosophers.

    It seems to me that deliberative "democracy" is the way to go. But then the requisite notion of deliberation should be made precise, and once it is made precise it seems somewhat likely that one group will be favored over others by nature or training. In other words, deliberation seems to face the same problem you observe for majority vote: it seems to entail a move away from democracy to oligarchy. (It will be very interesting to see if you can spell out a notion of deliberation that is democratic.)

  4. Thanks for the comments!

    Dileffante - I'm sympathetic to that, at least insofar as the civic virtue required for deliberation cannot be "hard-wired" into the process, as you say. But do these factors determine what kind of political system we have, or just whether it is a good (or legitimate) instance of its kind?

    Michael - I'm not here aiming for a "new definition", but a better understanding of the traditional definition, i.e. "rule by the people", and what this amounts to. I argued in the main post that universal suffrage is not sufficient for rule by all, as the case of permanent factions illustrates. So far this is mere conceptual analysis.

    I think the best way to carve up the different forms of government is not merely in terms of who votes. (That isn't a dimension much worth tracking.) A more philosophically interesting dimension, I suggested, concerns who can make a meaningful contribution to the political process. This may be done without voting at all (e.g. if you have an opportunity to persuade the oligarchs).

    Now, because what I really want is "responsiveness to reasons", I do demand that directly: that is my criterion of political legitimacy.

    So I always prefer (rationally competent) deliberative systems to non-deliberative ones, and the extent of suffrage does not especially concern me (though broader suffrage may generally be preferable, insofar as this offers protection against exploitation).

    Boram - Great questions. I'm assuming that there's just one maximally coherent conception of rationality that applies universally. (Otherwise we're screwed!)

    I do think that we should deliberate about ends and not just means. Once we've settled on a goal, it could make sense to delegate the implementation to (widely trusted) technocrats. I don't think that's undemocratic.

    I also don't think it's a problem if some groups (the better educated, more rational, etc.) play a larger role in virtue of being more able. It's bad to arbitrarily privilege certain groups, of course. But this isn't arbitrary. It simply reflects the fact that some people are naturally more able to contribute than others. Nothing would be gained by handicapping the more able, I take it. But perhaps more could be done to help the less able to contribute -- and I do think that educational assistance, etc., should be available for just this reason.

    Assuming some rational inequality remains, however, I don't see this as 'oligarchical' because there are no external barriers (and some external aids) to meaningful participation.

  5. [Shifted from the other thread - RC.]

    two (related) issues

    1) transaction costs. in a big country a large amount of people will have opinions but consider the transaction costs greater than their statistical influence - those costs increase the more complex their involvement needs to be.
    2) some people have fun participating. Those groups systematically gain more influence. rather like those with an urge to march down the street with signs have more influence - that urge (or skill) might not be directly related to being good at anything.

    should one force participation (like forced voting?) or is there a risk participation will plummet? Does that matter (i.e. is it the opportunity or the actual participation that matters)?

  6. I think it's the opportunity that matters. If most people are sufficiently satisfied with how things are run that they don't feel any need to try to change things, then that seems fine. (It's another matter, however, if their reason for abstaining is an expectation that no-one would listen to them anyway. I would count that as a form of non-opportunity.)

    In cases where balanced representation is important, say to avoid distortion, I think statistical microcosms (e.g. Citizens Juries, Deliberative Polls, etc.) are the way to go. Jury duty is forced participation of a sort, I guess, but very low-cost in virtue of being so small.

  7. Richard,

    Thanks for your helpful response. I'm not persuaded by it, but it does lay out your position clearly.

    It still seems to me possible that an enthusiastic supporter of majority vote can appeal to roughly the same sort of appeals you make in favor of deliberation as a decision making procedure. For instance, it can be said that the minority who vote do make a "meaningful contribution", in the sense of expressing their voice, raising awareness of their concerns, and so on, but just not in the sense of having any impact on the final result of the vote. Just like those on the losing side of deliberation, or those who do not subscribe to a favored standard of rationality: they too make "meaningful contributions" in the same futile way. And the points you make in favor of rational superiority can be made, too, in favor of numerical superiority.

    I am sympathetic to deliberative democracy. But it seems that my disagreement with your post and comments boils down to my doubts regarding your statement: "I'm assuming that there's just one maximally coherent conception of rationality that applies universally. (Otherwise we're screwed!)"

    I doubt that there is just one maximally coherent conception of rationality. (Think of different standards of rationality that have been promoted by philosophers, and even by a single philosopher, e.g., Sidgwick.) And even if there were, I doubt that it would be powerful enough to provide uniquely rational solutions to problems. (Think for instance of the problem of multiple equilibria in game theory.) The virtue of majority vote is that it does provide unique solutions in most cases, absent ties and too-close-to-call's. But I doubt that we're screwed.

    In raising these skeptical worries I am to some extent following Hume: human reason is weak and feeble, and nature provides solutions to problems that reason is poor at solving. I reject his fideism about nature, but I accept his skepticism about reason to some extent. Rational deliberation isn't the almighty problem-solver that philosophers sometimes make it out to be. But it is powerful enough to suggest non-rational solutions to problems.

    Majority vote is a non-rational solution (to what sort of problem I won't mention here out of laziness, though it is an interesting and important problem that should be mentioned). It won't do though, for the problems you point out. A viable non-rational solution seems to me to involve a reorganization of society into smaller decentralized units, which are voluntary associations with their own favored standards of rationality or lack of standards, problem-solving methods, etc. Each voluntary association will be focused on dealing with local problems, with how to distribute the cooperative surplus among the members who have contracted their way into the association, and so on. This will allow for more direct, participatory forms of democracy.

    I should stop rambling here :) This is a topic that I'm writing on now, and I find your post thought-provoking.

  8. Sounds like you are an anarchist.
    To me the anarchist solution seems to create problems it is supposed to solve.

    To deal more directly with some of the issues

    1) Being a stakeholder in an issue is not always - or even usually, something you could term to be clearly voluntary. For example it assumes that my parents, my children (and others tied to me) share a philosophy. If not I suffer a tyranny that is probably much deeper than I would ever experience under the current system where I can appeal to the laws of the country which are set to suit the whole country.

    2) I have a stake in the Amazon rainforest and in the use of CFC in a factory in Qingdao. How can I get my voice heard - or does it just not get heard?
    3) Setting up laws is something that requires some skill at designing a system and also philosophical intelligence. a random small group would tend to create bad law (e.g. poor laws regarding getting and trusting evidence) or grossly inefficient law (e.g. taking everything to the village council) where laws became complex.
    4) Without a strong state how does one maintain the system? it would be hard to stop some communities slowly putting in place laws that compromised the freedom of association or the rights of other communities, or gaining power over each other (e.g. via trade).
    5) Justice - what if a person is 'convicted' of committing a crime in one area then goes to another. Is he immune or does the new community assume his guilt? Or do they go through complicated procedure of calling witnesses from far away to their village and holding him in the meantime?
    If he is found guilty what does one do? Expel him? (Expulsion is far cheaper and probably as effective for protecting the community)
    6) if a copmmnity becomes entirely disfunctional (let us say it is like a street gang training antisocialbehaviour). Is the cancer ignored or does one revoke their freedom of association?

  9. Genius, I'm guessing your comments are directed towards me. Is my position anarchist? My professor for the democracy seminar I took back in 2002, Len Krimerman, will be pleased to hear that I am an anarchist. He tried to persuade Rawls that anarchy is the way to go, apparently w/o success.

    You raise good questions, though I doubt that other forms of govt. or social organization face challenges any less difficult. Here are my responses (or just tentative starting points for full answers to your problems).

    (1) I do not see how this is a special problem with the kind of social reorganization I propose. Children are usually indoctrinated in the ways of their parents. Perhaps your worry is that since this form of "tyranny" is generally the case, that poses a special problem for the voluntary associations I propose (i.e., how "voluntary" can they be?)

    It's easy to fantasize from an armchair, but it's also a lot of fun. I envision a very different system of schooling for children, one that is apt for making them autonomous individuals, preparing them for later participation in voluntary associations. We may classify these associations broadly as culturally, economically, politically, and intellectually oriented, allowing these orientations to overlap in particular associations. Then I imagine a curriculum for teenagers, where they decide to participate as apprentices in the day-to-day activities of the associations of their choosing. (E.g., the curriculum may require that the student participate in three different culturally oriented associations, each for a couple of months, say. The student may then choose to live in an Amish community, then a Buddhist community, then a Muslim community. This will also have the virtue of inculcating appreciation for different religions and cultures, or at least tolerance--it's hard to hate a different religion when you've met with its sincere practitioners and immersed yourself in their lives for a while.) The various voluntary associations will also welcome this type of schooling as a way of getting new members into their associations. If they oppose it, I suppose it will be for the wrong reasons, e.g., wanting to control the lives of their young.

    (2) This is a problem for democratic societies nowadays as well: they are mostly concerned only about national or local issues, so that environmental and humanitarian issues whose importance has no borders often escape the attention of national and local govts. One may press the worry that a reorganization of society into smaller voluntary groups would tend to focus attention only on local issues, and not on global problems. That is indeed worrisome. To address this, among politically oriented associations let's say that there are environmental groups, charities, and so on, and others that encourage economic sanctions against uncooperative groups, imposed by other communities. Also, the flourishing of small-scale local economies would mitigate the harmful effects of globalization.

    (3) We may translate Mill's ideas about "the marketplace of ideas" and "experiments in living", from the level of individuals to that of small communities. Better designed communities will flourish, as more members will flock to them or their designs more frequently copied. Poorly designed communities will go out of existence. Here I am talking about political and economic designs.

    This is also how inquiry in the social sciences is best conducted, on the scale of small voluntary associations. The problem with testing hypotheses in the social sciences is that such tests involve social engineering on a massive scale that can have devastating effects, and are often immoral. So such direct tests are avoided, and we resort to indirect support for theses that are underdetermined by evidence, or simply stick to the status quo.

    (4) The solution to this problem will be a matter of finding an adequate system of checks and balances, or of countervailing powers. For instance, there will be a need for enforcement mechanisms that protect against violation of rights between communities, and institutional checks against the abuse of power by these mechanisms.

    (5) The challenge you raise here seems to apply to any case where a criminal in one jurisdiction flees to another. It doesn't seem to be a special complication for my proposal. Below I address a related complication on justice.

    Different economically oriented communities will have different principles of economic justice (e.g., property rights), and will be enforced on terms that community members voluntarily accept. Generally people will be allowed to become members in one economically oriented community at a time, to avoid problems such as tax evasion if multiple memberships are allowed. Opting out of one economic community and joining another will also be allowed under certain constraints, e.g., to exclude the possibility of free-riding, etc.

    Basic human/political rights will be a different matter--my guess is that it will be rational to want the same basic rights. This will be stated in an constitution that everyone can agree on, say something like the U.N. charter, but in addition outlining some rights and duties for communities as well (see below).

    (6) I suppose communities can be treated as legal entities with their own rights and duties. Communities that violate the rights and liberties of other communities will be treated in ways analogous to individuals who violate the rights and liberties of other individuals.

  10. Boram - "It still seems to me possible that an enthusiastic supporter of majority vote can appeal to roughly the same sort of appeals you make in favor of deliberation as a decision making procedure."

    My thought was that an overridden vote is "futile" in a stronger sense than an appreciated but ultimately rejected argument is. If the factions are fixed and dogmatic, then the minority voter has no chance to influence the outcome. They are completely ignored by the majority who wield power. This is not so in case of genuine deliberation. Even those who ultimately lose the argument are at least seriously listened to, and given the opportunity to sway others' minds and thus influence the outcome. Some will be better at this than others, true, but it's not a foregone conclusion that the educated elite will win no matter what, as it is with the majority faction.

  11. Richard, in response to your last comment, it's not clear to me that an overriden vote is more futile, so to speak, than a rejected argument.

    Consider voting for candidates. There are three candidates, one of whom is clearly in the minority (say Ralph Nader). The minority vote can sway the ultimate outcome in favor of one or the other candidate. To generalize, it is false that the minority voter has no chance to influence the outcome.

  12. Sorry I wasn't clear. Suppose that those who were voting for Ralph Nader would have otherwise voted for Al Gore. Then the minority vote in this case (for Ralph Nader) can determine the election outcome (Bush winning over Gore), by drawing voters away from one of the two popular candidates. This sort of situation can be generalized to any issue put to the vote, on which opinion is split three or more ways.

  13. To be clear I favor a world with a single government - and where that government has considerable power as well as considerable checks and balances on that power.

    (1) I don't think it is a special problem it is just a bigger problem the less diverse your major law making body becomes.

    for any system (including this education strategy and your NGO proposal or the system of free association itself) to be reasonably permanent it must have a overarching structure - that structure needs to resemble a government it also need to have the economic and military muscle to protect itself.

    (2) ifthere was a single government that would internalize much of the externalities. Any step in that direction helps.

    (3) the Chinese system and the NAZI system and to an extent the early communist system flourish. I think flourishing in terms of power is only loosely related to the general publics view of flourishing in general.

    (5) my ideal scenario solves the problem any step n that direction reduces it.


  14. Boram - right, I was talking about the specific situation where one faction has a clear majority. This is enough to show that voting does not guarantee that one has an effective input (even though there are undoubtedly some situations where one's vote could be effective).

  15. Derek writes: "Yeah, I just guess the idea of a dysfunctional democracy isn't conceptually incoherent. Likewise, the idea of an oligarchy that makes up 60% of the population of a country just seems like a misuse of the word.

    Why not just identify two different political virtues: participation by the people at large, and a general culture of open-mindedness and intellectual curiosity. What theoretical/moral/political payout comes with calling both of these virtues 'democracy', and anything that lacks at least one of the virtues an 'oligarchy'? To be honest, the distinction right now looks entirely rhetorical.

    Two thoughts:

    (1) I just don't see any principled reason to treat really big oligarchies as a fundamentally different kind of government. That's not a difference that makes a difference. If democracy -- "rule by the people" -- is suppose to capture anything theoretically interesting, it had better not just be "rule by a larger faction than the alternatives".

    (2) I'm not sure why the 'rhetorical' element is objectionable. Practical reasons reinforce, rather than undermine, the case for thinking about democracy as I suggest. Besides, insofar as 'democracy' is a moralized concept, its meaning is partly dependent upon the normative facts about political legitimacy, etc. That is, I take myself to be spelling out conceptual commitments that we all already have.

  16. Richard wrote:

    <(1) I just don't see any principled reason to treat really big oligarchies as a fundamentally different kind of government. That's not a difference that makes a difference. If democracy -- "rule by the people" -- is suppose to capture anything theoretically interesting, it had better not just be "rule by a larger faction than the alternatives".

    I don't see defining democracy as "good government" or "good decision-making government" as much more theoretically interesting. In any case, I didn't suggest that democracy was "rule by a larger faction than the alternative". I said that rule by 60% of a country's population was not an oligarchy under any normal use of that word. Democracy requires, minimally, participation from a wide subset of society. How wide? Don't know, exactly. But pretty wide.

    <(2) I'm not sure why the 'rhetorical' element is objectionable. Practical reasons reinforce, rather than undermine, the case for thinking about democracy as I suggest. Besides, insofar as 'democracy' is a moralized concept, its meaning is partly dependent upon the normative facts about political legitimacy, etc. That is, I take myself to be spelling out conceptual commitments that we all already have.

    Yeah, I don't have those commitments. I mean, statements like: "Democracy unmediated by courts and a constitution would quickly destroy individual rights" or "Democracy, without the right sort of institutional and cultural safeguards, encourages ethnic and class warfare" seem conceptually coherent. People who've said things like this may be wrong, but if they are that's an empirical matter. It's not a conceptual impossibility.

    I agree that democracry figures in legitimacy, but that could work any number of ways without implying the commitments you suggest. Democracy could be one thing among many that contribute to the legitimacy of a government. Democracy could be a necessary condition on legitimate government. Only if it's a sufficient condition would that imply the conceptual commitments you're spelling out. But why would we have thought it was a sufficient condition?

  17. I agree that your example statements are not incoherent on the face of it. The second, especially, seems to invoke the idea of mere majoritarian voting, which is certainly one sense of the term 'democracy'. But I am trying to explicate the theoretically interesting notion of rule by (all) the people, and I explained in the main post why a kind of reasons-responsiveness might be thought to be relevant here. I don't see that you've responded to anything I said in that regard. (Note that I am not defining democracy as "good government"[?], but rather, a government where "any can make a meaningful contribution", i.e. "have a real chance of affecting the outcome.")

  18. Actually, it might help to clarify that on my view the formal requirements for democracy could survive substantive normative inversion, e.g. a state that was responsive to bad reasons. Everyone could potentially wield influence by coming up with especially poor arguments. It would not be a good government by any stretch of the imagination, but it would be democratic!

  19. Richard,

    Why would a society that's responsive to rational arguments be one in which every citizen had a greater chance of making a difference in the government's decisions? One way I could influence society's choices is by coming up with a great argument or bit of evidence that it should choose a certain way. I might also influence it through an ad campaign. Both seem equally unlikely to have any effect if the society is large and complicated enough. But if all that's required is the possiblity of effecting change in order for someone to be 'ruling' in the sense relevant for democracy, wouldn't everyone have that possibility no matter what? Even in totalitarian states there's a chance that something I do will get the government to do what I would like.

  20. Right, it's not enough to have some vanishingly slim chance of influence. Rather, there should be a sufficiently good chance (never mind the exact details) that by presenting a "great argument" you will influence the outcome.

    So it sounds like what you're really suggesting here is that "if the society is large and complicated enough", it is unlikely to be sufficiently reasons-responsive to qualify as democratic in my sense.

    (Aside: a society might also be democratic in virtue of, say, responsiveness to rhetoric rather than reasons. In that case, anyone who came up with a sufficiently spiffy sound-bite might be able to sway the outcome. Like the case of anti-reasons, this shows that democracy is not always good.)


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)