[Simon has advised me to submit this as a “post” rather than as a “comment.” As I write, there are 19 comments. I’ll insert a comment to mark the point at which I posted these remarks]
This might be a good time to jump in and respond to a few of the points that have been raised in the discussion of Chapter 1. Let me first say that Jonathan’s summary is excellent, and gets this off to a great start.
Obviously, Chapter 1 is a condensed run-through of many of the themes and arguments of the book. So I think the best plan is for me to respond only rather quickly on issues that will come up again in more detail later in the book. Some of the points raised this week don’t get any further attention in the book, so I’ll say what I can now. There are just too many points to respond to. If I skip something that any of you thinks is especially important, feel free to push me on it and I’ll take it up if I can.
I might also say that I would ask people not to quote what I say here in published work without checking with me, a courtesy I’ll extend to you all as well. (This raises interesting questions, of course, about what, these days, counts as a publication.) Maybe it goes without saying, but I am not putting the time and thought into this that I would if I regarded it as part of my published output. It wouldn’t be possible to do that and keep rolling in a timely way. It’s like a conversation at a conference. I (or you) might well trip up, or contradict myself, maybe fixing it a few days later, maybe not, etc. It’s not that I live in fear of having my mistakes exposed publicly. I just think that if we treat blogs and related things as part of the published literature we are going to wreak a lot of havoc unnecessarily. This issue could trigger a whole thread of its own, but we don’t want to get off on that tangent. Enough said. Back to democracy.
A preliminary thing: is the theory meant to “bless” all democratic arrangements? It is not. My thesis is only that there is no adequate theory of political authority and legitimacy that can (in those two senses) justify political arrangements that are not recognizably democratic. So the logical form of the claim is “if not democratic then not justified.” If democracy is defined in a broad manner to include anything with free and fair elections, or some such thing, then I would not assert the contrapositive “if democratic then justified.” My preferred definition of democracy (at, for example, p. 38) is less institutionally specific: “the collective authorization of laws and policies by the people subject to them.” In at least one place I believe I add “by voting,” but I’m uncertain whether I want that included. If collective authorization is possible without voting (I suspect it is) then I would still regard the procedure as democratic. The presence of “authorization” in the definition is meant to be a moral element, so it’s unlike many operational definitions. I won’t defend the definition here, but I hope this clarifies things a bit.
Jonathan points out that on some political questions there seems to be no correct answer, in which case an epistemic approach makes little sense. I do think this is an important line of challenge, especially when it’s generalized in the following way: In those non-epistemic cases we still think there is a moral basis for democratic procedures. But if there’s a non-epistemic basis there, why not think there’s a non-epistemic basis even in cases where there are right answers? The challenge is compounded by the Waldronian point that since there’s (reasonable?) disagreement about so many substantive matters, it is philosophically attractive to keep the justification of democracy entirely procedural. The view I call “fair proceduralism” tries to do just that.
Fair proceduralism is a major character in the narrative of this book and so I won’t try to fully address it here. So just a few points. First, for what it’s worth, I think that the number of political issues on which there is no right answer as to which decisions would be right or best is smaller than is often assumed. Few things really come down to matters of taste or brute preference, even though it is possible for that to happen. Should we have a playground or a sculpture in the park? Different people have different preferences, certainly, but there is more to it than that. Just try putting the issue on a ballot and attending to the political deliberation and mobilization that will surely take place. There will be perfectly sensible claims about whether there are enough playgrounds in this area of town, whether the city is as supportive of the arts as it should be, etc. OK, you might say, but how about whether it should be a sculpture by one artist or another? Yes, it could come down to mere preference. My view about those cases is that if they are, in a generally acceptable way, acknowledged to be matters of mere preference then there is a political question about how to resolve them, and that question might have correct or incorrect answers. So, I think that “meta-question” is properly decided democratically and I would just wheel in the whole epistemic proceduralism story here as elsewhere. Maybe it will be decided (democratically) to decide preference-issues by a vote, or maybe the decision will be to decide them in some other way. The legitimacy of the chosen method would arise in the normal epistemic proceduralist way at the meta-level, and would infuse the actual first order decision that eventually gets made.
Later in the book (especially in Chapters 4 and 5) I consider the idea of avoiding substantive standards altogether in the way that fair proceduralism tries to do. I try to clearly define the idea of a substance-free procedure, and then argue that no democratic theory actually follows through with purity. I’ll leave those questions for later.
A related point that has come up is whether my view ascribes only instrumental value to democracy or also intrinsic value. There’s no simple answer. The view is not a simple instrumentalist or even rule-instrumentalist view, largely because of the very prominent role of the general acceptability requirement. The whole theory operates under the umbrella of that Rawlsian principle, which is not itself instrumentally grounded in any way. But under that umbrella, instrumental considerations play a role. We need (so I claim) the epistemically best political decision procedure so far as that case can be made to all qualified points of view. That crucial proviso means that we are not looking for the epistemically best procedure. If we were, then we would be led toward epistocracy and away from democracy. The general acceptability requirement is a moral principle I draw on in order to morally explain the superiority of democracy over epistocracy, even if epistocracy is epistemically better. The general acceptability requirement is not instrumentalist in any way, and so the resulting account of democratic authority and legitimacy has a large “intrinsic” element, along with the embedded instrumental epistemic element.
When I was first developing the idea of epistemic proceduralism (in my paper “Beyond Fairness and Deliberation…”) I said we needed more than procedural fairness. That implied that along with the epistemic element there was an intrinsic value to procedural fairness. So that’s a further question about the role of intrinsic vs. instrumental value in the theory, and it has come up in the comments here. As I say in the book (somewhere) I now think that while procedural fairness does have some intrinsic value, my theory does not make any appeal to it at all. I’ll stop at that, since this will come up again in later chapters.
The large role played by the general acceptability requirement means that it is incumbent on me to say what I can to defend that principle against objections. I do that (among other things) in Chapter 3, so again I’ll postpone lengthy discussion of that. I should say, however, that while it is tempting to think of the principle as a kind of “internalism,” as Simon usefully puts it, the principle as I understand it is not internalist. Simon’s thought is that the principle, in requiring acceptability to all reasonable points of view, seems to require that political arrangements track the actual attitudes of the people in a certain way. It seems to require a kind of endorsement by reasonable citizens. Actual people could block the justification by failing to have the right mental states.
That is not how I understand the principle of general acceptability (nor Rawls’s very similar “liberal principle of legitimacy”). The reason is that whether or not an arrangement is acceptable to all reasonable points of view is not an empirical question, but a moral one. Nothing about anyone’s actual mental states bears on the matter. The range of reasonable views is determined not by the range of actual views, but by the range of reasonableness itself—a purely moral question. It’s similar in moral contractualism. Scanlon, for example, asks whether a set of rules is subject to any reasonable objection. But this is not a question about whether anyone actually objects, and so it is not in that respect an internalist idea. (Internalism means different things in different places, but I’m pretty sure Simon has in mind an analogy with views about practical reason and maybe also epistemology that link justification to certain actual mental states of actual people.)
Let me make a few points about normative consent, leaving further discussion to our later treatment of the chapter that lays it out more fully. Normative consent is my name for an approach to political obligation that says, roughly, you are obligated to obey a command if it would be morally wrong to refuse to consent to a new proposal of such authority. It’s a kind of hypothetical consent view, but I think a novel one: you are obligated because you would have consented to such new authority if you were not immoral.
I thank Bill for passing on Joe Adams’s comment. Joe points out that once we move to the paradigm of normative consent it is not at all obvious that we need to appeal to epistemic considerations in the way I do. We might be able to appeal to familiar considerations such as receipt of benefits in a fair scheme, just for one example. In one respect that is a friendly point, and in another respect it is a challenge to me. The friendly point is one I make myself near the end of Chapter 7 (pp. 131-135). Stepping back from my concerns with democracy and epistemic proceduralism, normative consent is an umbrella under which one or another of the traditional approaches to political obligation might fall, albeit in an altered form. So it could turn out that the moral relevance of receipt of benefits in a fair scheme consists in the fact that it would be morally wrong not to consent to the authority of such an arrangement.
But Joe’s point is also meant as a challenge: why do I need all this troublesome epistemic stuff? The normative consent approach could make it all unnecessary. Maybe it could be shown that it would be wrong not to consent to authority of arrangements from which you fairly benefit, or some such thing. I take up the relation of my view to the “fair play” approach in Chapter 8. Let me just say here that I don’t believe that authority depends on the receipt of benefits to the obligated agent. The capacity of a political system to deliver morally good or correct laws and policies would be a great value, even though it might never benefit you. You might be benefitting from a less reliable political system in which your taxes are unjustly low, your crimes are unjustly ignored, and so on. On my approach, the value of a more accurate system grounds its authority over me nevertheless. We might put it this way: in order to show that I would be wrong not to consent to the authority, over me, of a system of laws and policies, it is not necessary to show that I would myself benefit from the system in any way. It suffices to show instead that the system has a certain tendency to produce substantively just decisions (where this showing does not contradict any qualified (or “reasonable”) point of view. That is what makes it wrong not to consent to the system’s authority, and that in turn is the basis of its authority. I try, in Chapter 8, to explain how the epistemic approach uniquely meets up with the normative consent approach.
In comment 4, Jonathan is right to say that the view of authority resembles certain kinds of appeal to natural duty. However, the normative consent apparatus incorporates a quasi-voluntarist element that is no part of simple appeals to the great good of justice (or whatever), so it’s really a very different kind of view. On the other hand, it is only quasi-voluntarist, not voluntarist. Normative consent is not a species of consent. Obligations to obey can, on my view, simply befall you quite apart from anything you do or believe (just as we all agree many other kinds of moral obligations can). But the effort to support a claim of authority is aided, I think, by being able to say, “maybe you didn’t consent but you would have if you weren’t a jerk.” That seems like a morally significant reply to “I never consented.” There’s a wrinkle about the fact that you might not have been offered the chance to consent, but I argue in that chapter that it is easily ironed out.
As for the several remarks on what I say about “utopophobia,” I’m keenly interested, and promise to engage them fully when we get to that chapter (the final chapter). I think Andrew is correct to wonder whether political philosophy would have a subject matter at all if we operated at too ideal a level. And yet justice (or authority, or legitimacy) is not automatically met by whatever behaviors or institutions we can safely predict. We might be able to safely predict injustice, etc. So we must idealize in certain ways. My research has turned to these questions since I finished the book, and so I’m looking forward to that part of our discussion.
I’m sorry I haven’t replied to every significant question that’s been raised in the comments. As I say, you should feel free to press me, now or later, on any that you think are especially important.
(It might be best if only comments specifically on this post are submitted here, while other comments on the thread about Chapter 1 continue over on that post. But I’m open to suggestions.)