I think this argument can be understood by comparing it to the individual case and looking at it as a claim about political groups. The norm (default position) in an individual case is that the individual rules herself. We need an extra argument to justify others exercising authority over that individual, (for example, we have such an argument in the case of the parenting of children). Similarly, in a group situation, collective decision-making among members is the default position. I don’t think this argument makes any claims about the details: unanimous consent vs majority rule vs elected representatives, all of these fit the default of a single group exercising authority over itself. If society is broken into two or more political groups (epistocrats and commoners; nobles and serfs; men and women), we would need an additional argument to show why one group has authority over the other. Democracy does not divide society into any groups where one rules over another but retains the default position of a single group.

Suppose there are no qualified objections to having a common rule of conduct on some matter. Now we need a decision procedure. There are reasonable objections to both epistocracy and democracy. But ALL coercive power has to be justified by the special “no qualified objections” standard, and in the epistocratic set-up, there is extra coercive authority, over and above that implicit in there being a collective rule all have to follow. Not only is there a common rule, but some get input into what this rule will be, while others don’t. Given that there are no qualified objections to having a common rule, any deviation from equal say in determining what this rule will be should also have to meet the “no qualified objections” requirement, because not only are we each subject to all, some are subject to others. So what we have here is a presumption in favour of equal authority given that there is no qualified objection to lodging the authority over the conduct in question with the collectivity. I don’t think this is begging the question, because the argument can proceed from first premises about the justification of coercion (thinking of Gaus’s arguments about public justification here). But it does rely on the premise that in the absence of conclusive justification (i.e. if there are qualified objections), the collectivity cannot impose a common rule of conduct. Depending how one spells out the qualified objection criterion, this starting point may severely limit the scope of collective authority and undermine the possibility of a democratic society, even if democracy wins within the limited sphere permitted for collective decision-making.

Thanks for your thoughtful comments. First, I should say that you are right - you caught me again being a little sloppy in my summary. I should not have said ‘political decisions’. David refers to the QAC (at least in these early chapters) as a test that alleged relations of authority or legitimacy must meet, not a test that each political decision must meet. Maybe it was something like a Freudian slip on my part since I do believe that all our political decisions ought to be held to something this standard, but I didn’t mean to attribute this position to David.

Two further quick comments. I’m not too persuaded by your suggestions as to how qualified objections might still arise in a society with epistocratic educational institutions. Your suggestions rely on non-ideal assumptions (e.g. a society’s unjust past or present class system, or sexism, or the worry that the rulers might act unjustly). While these are valid concerns once we move to non-ideal theory, they don’t provide any principled reasons to oppose such epistocratic institutions.

Finally, I happen to think David’s characterization of at least some deliberative democrats is fair. Some deliberative democrats, particularly of the Habermasian variety, have been heavily critical of Rawls precisely because Rawls’ political liberalism places pre-political normative constraints on the process of deliberation - crudely speaking, it makes certain liberal principles prior to democratic autonomy. These deliberative democrats argue that the process cannot be constrained by any prior normative principles other than those necessary to the process itself. Habermas’s ideal speech situation does not produce any principles that constrain real political discussions since no one person can know what would be agreed in the ideal speech situation. The speech situation just offers the ideal model for our own deliberative institutions.

1) “it is very easy to predict that a system of majority rule is just a mechanism of handing authority to the majority religious group within that society”

That’s something I’m very worried about. If majority rule amounts to permanent rule of one group over another, then I don’t think there’s anything democratic about it. Moreover, I think you could strengthen the point by noting that ‘the wise’ need not denote a fixed group either, but could be people who vary case by case – for we could submit medical matters to an epistocracy of doctors, town planning decisions to an epistocracy of town planners, etc. (Of course, budget trade-offs between hospitals and town planning might call for a different set of epistemic qualifications, but that would be yet a third group of experts, if there were any experts in such). So, I’m similarly slightly puzzled as to the status of universal, equal enfranchisement as a sort of default, although I suppose one could justify it contractually – e.g. as reasonable agreement, what all would accept behind the veil of ignorance, minimax relative concession, etc.

2) The path-dependence of our lack of political experts.

Another very good point, recalling all the training Plato prescribes for his Guardians. I think one could still raise qualified objections, however (drawing on some of what David has to say in ch.11). We hear plenty of controversy about educational achievement gaps according to sex, race, class, etc – not to mention supposed public school favoritism in Oxbridge. Presumably, any supposed meritocratic standard to select the ‘brightest and best’ for this political education would be open to similar controversy and qualified objection. Moreover, you might also argue that such a division will create divided ‘class’ (ruler vs. non-ruler) interests in society. The political experts need not, like Plato’s Guardians, be supremely virtuous or denied private benefits. Thus one may worry that they’ll use their knowledge to do what’s best for them rather than all – and, again, whether or not this is true, it seems a qualified objection.

Like Jonathan, I’d like to hear more from David, or anyone else, on these points. I also had two of my own worries, sparked by Jonathan’s summary, however. I’m not sure whether these are problems with the book, Jonathan’s summary, or my understanding, so any clarification would be enlightening.

3) “political decisions must be acceptable to all qualified points of view” [Jonathan’s paragraph 3]

There’s more to come about qualified acceptability in the next chapter, but if this is right (and I don’t currently have my text in front of me) I’d misunderstood. I thought the qualified acceptability test only applied to something like ‘constitutional essentials’. That seems to be how it’s applied – democracy or majority rule must be acceptable to all qualified viewpoints. Is it (also) the case that democratic decision outcomes must be similarly acceptable to all? And does this amount to allowing potential for significant checks on democratic rule, e.g. supreme court review?

4) “Estlund notes that both social choice theory and deliberative democracy tend to embrace the thesis that there are no prior normative standards by which our political processes can be evaluated - each claims to offer a purely procedural approach to political decision-making”

Does anyone seriously entertain the notion that any political decision is as good as any other? That had, for example, the Nazis put the Holocaust to a vote it wouldn’t matter whether or not it went ahead, provided that what was done reflected a majority of preferences? If anyone did, I’m not sure what I’d have to say to them. What I think is going on, however, is that many writers sharply distinguish good government (realizing substantive principles of justice) from democratic government (respecting the preferences of the people). If we care only about the former, then we could accept Platonic Guardians if we thought their rule would ensure the best results. Now, David wants to challenge this, effectively arguing that no specific epistocratic arrangements could pass the qualified acceptability test. (It’s all very well saying we want to be ruled by those who know the truth about justice, but no help in practice if we can’t agree who they are). He thinks that part of the justification of democracy has to appeal to the fact that it’s epistemically best out of those procedures that are acceptable to all qualified points of view – and this is what differentiates it from the also procedurally fair possibility of deciding randomly. Nonetheless, surely it’s misleading to characterize his opponents (the social choice/deliberative democrat* people) as denying that there are normative standards prior to the democratic procedure? I’d have thought they only want to deny the relevance of such to democracy.

*I admit, what I’ve said here may be more my reconstruction though. I don’t know much about Habermas, but some of what I have heard suggests he might want to say norms are produced by the constructive procedure of the ideal speech situation – but in so far as this differs from an actual democratic decision situation, doesn’t it still impose prior normative standards on the latter?