Again, I’m gratified to see such a clear and accurate summary of the chapter. Let me start with Simon’s excellent questions, drawing connections to Jonathan’s where appropriate.
Simon’s question (1) is whether the system of justice that arises in Juristic Prejuria has some unique claim to obedience or whether other schemes might also arise and place similar claims on people’s obedience. This question grants, at least for the sake of argument, that the case for the jury system’s original authority goes through. It then asks, since it goes through for this arrangement, why wouldn’t it go through for multiple parallel arrangements? So the specific challenge in (1) is not to the argument for the original authority of the jury system. But if parallel arrangements can make an equal claim to authority then the authority of the jury system seems to be limited by those other possibilities. It will only work as an analogy for the authority of a whole political system if it grounds unique authority.
I wonder if the normative consent account of authority offers an answer in the following way. If you grant for the sake of argument that people would be obligated to consent to the authority of the jury system, then we can look together at what that consent would amount to. I say that it is helpful to see it as a promise to obey; I mean that this is another way of describing the exact same thing as consent to the system’s authority. So, if I promise to obey this jury system, and now some other system emerges that proposes authority over me with regard to the same cases as the first system, it looks like my first promise renders the question moot. Since I’ve promised to obey the first one, normally I may not promise to obey the second one in any case where they conflict. So uniqueness emerges from the way that my promise to one system locks up my duties, keeping them out of reach of subsequent systems.
The issue is more difficult if parallel systems present themselves to me simultaneously. If each of them has an equal claim to my promise to obey, how can I be obligated to promise obedience to any one of them? It won’t be enough to say that I am obligated to promise to one or another. The reason is that, assuming the different systems will issue different directives, the idea of normative consent won’t determine which system of directives I’m bound to. I’m bound as if I had consented because I was required to consent. But to which system? At this point I just concede that I have not tried to answer the uniqueness problem when there are parallel simultaneous systems each of which has as good a claim as the other on my duty to promise to obey. The problem doesn’t look insuperable. It might be solved by bringing in some duty to commit to the one that other people are already committed to, or are likely to commit to. But I leave the matter here. Keep in mind, though, that this issue is not one that limits the authority of the jury system. Whichever of the parallel systems is owed commitment, its authority over me would be unique owing to the way my promise ties up my duties, as described above.
Question (2) concerns my attempt to show that the juries in Juristic Prejuria, when they acquit someone, do not only cause it to be the case that private punishment is wrong, but they count as forbidding it. In light of what Simon says about this and my treatment of the case of the dictator’s child, I’m not sure I know how to argue that the jury’s acquittal is literally an act of authority, even as it results in obligations not to privately punish. (That example was mentioned at p. 118. By the way, note 4 on p. 285 gives the page for the example as “p. xxx.” Darn.) But, now I wonder, why did it matter to me whether I could account for this? The jury (system) plainly does command the jailer to lock the guy up, which fits the account of what authority is. The issue is what we can get in the case of democratic politics from the analogy with the jury. Suppose I’m forced to say that the analogy only supports the idea that the properly democratic state is authoritative when it commands. Just as in the jury case, there might be lots of associated moral requirements and prohibitions that flow from the authoritative commands of the state. But it doesn’t seem to costly if we have to say that they are not, literally, cases of authority because they are not actually commanded or forbidden by the state. This looks like a palatable bullet.
Simon’s question (3) is a version of the direct authority objection to my argument for normative consent. There’s a pretty thorough discussion of that in my lengthy exchange with David L. about Chapter 7, so I’ll leave that aside here. Jonathan, in comment (1) also presses a version of the direct authority objection. One brief comment about what he says: As I argue at p. 155-56, it is not generally the case that doing my fair share in the most efficient collective scheme for solving a problem is the most efficient way for me to contribute to a solution. So a direct duty to help in the most efficient way won’t apparently yield authority. If you change the natural duty to require doing my fair share of the most efficient scheme, my reply would be what I say to the direct authority objection generally.
Simon’s question (4) is partly whether I can only establish that there is a duty to promise to do something when there is already, independently, a duty to do it. But the examples are not offered as support for a duty to promise. They have the very limited purpose of trying to show that in some cases there is value to action’s being promised that is separate from the value of the action itself.
The other part of his question is whether a duty to promise to obey is so different from my examples that they are little support. But in light of my first point I’m not sure how this objection is meant to go. I claim that just as there is great value in doctors not only acting in certain ways but also in their promising to do so, there is great value in people not only obeying their local district but also in their promising to do so. If there is, then I can say that my promising would be a contribution to the valuable condition of people having generally promised to obey (the “general commitment task”). That’s all I’m trying to establish with those examples.
In his question (5) Simon writes, “…if the duty to promise to obey generates a duty to obey, then the costs of the latter have to be counted somewhere…” This issue is significant, I argue, because if the costs of promising to obey some scheme are much lower than the costs of actually obeying the scheme, then there is less chance that the excessive cost will cancel an obligation to promise. Simon correctly suggests that the duty to promise generates a duty to obey. He says the costs of the latter must be counted somewhere. By “latter” he means either the duty or the obedience. In any case, he seems to mean that the costs of obedience must be figured in the costs of promising to obey. My contention in that passage is that this is not so. I really make two points there, and I don’t clearly distinguish them.
One is that it seems wrong to say that the cost of a duty to do x includes the cost of doing x. The reason is that, the duty notwithstanding, I might never do x. It doesn’t cost me $100 to promise to pay $100, as shown by the fact that promised money often doesn’t actually change hands. While I think this is correct, I don’t claim to know how cost considerations are to be brought into the question whether promising something is excessively costly.
The other is that even supposing (for some reason that is unclear given my first point) that if one has a duty to obey one will obey, the cost of putting oneself under the duty must be the expected cost of obedience. So, even if I will incur an obligation, say, to go to war and risk my life if so ordered, the expected costs is that possibility modified (actually multiplied) by the probability that I will actually be so ordered. This point will often show the cost of incurring the duty to be less than would otherwise be thought. So when someone obediently goes off to war and is killed, the loss of his life cannot retrospectively be entered as one of the costs of promising to obey. (This point seems to be grasped by the many thousands of volunteers in the U.S. armed forces.) The cost of promising, in that case, is much lower than the cost of obeying.