Many thanks to Kevin Vallier (here), Chris Morris (here and here), Bas van der Vossen (here and here), and Massimo Renzo (here) for their very thoughtful and interesting posts. I have some replies to Kevin and Massimo below. Replies to Chris and Bas will follow in a separate post.
Eight comments on Kevin’s post:
1. As I understand it, the notion of “political authority” involves placing one agent above other agents. It involves postulating a special moral status for a particular agent, “the state,” such that (a) other agents have duties to obey the state in a wide range of circumstances in which they would not owe obedience to any other agent, and (b) the state is entitled to do things that would be clearly wrong for any other agent. This violation of the ideal of moral equality calls for an explanation. What is so special about this one agent?
No parallel question naturally arises about property rights. In claiming that there are property rights, one is not claiming that there is any agent who is special in those ways. When I say that I own a car, I am not placing myself above all other agents, and it is not at all natural to respond, “What’s so special about you?”
Now, you might say that I am claiming a special status, namely that only I and no one else am entitled to decide how the car is used. That I may use coercion to prevent others from taking the car. That others must obey my wishes with regard to how the car is and isn’t used. But this would be a pretty contrived way of arguing that I have a “special moral status,” since other people have exactly parallel rights with respect to their own property.
2. I certainly agree with Kevin that none of the philosophical theories of authority can be used to defend property rights, any more than they can be used to defend the authority of the state.
3. It isn’t my job to give a theory of property rights, for two reasons: first, because it is a reasonable thing to take for granted in this context. Second, because my main theses probably hold even if there are no property rights.
On the first point: there is only so much you can do in one book. (Literally – the publisher had a word limit. Various people have listed things that I “should have” discussed. To fit all of them in, I would pretty much have had to delete all the existing chapters.) To make progress in political philosophy, one has to be able to take some norms for granted. Similarly, I didn’t give an account of why violence is bad, or why human life is valuable, or indeed why we should believe in any norms at all.
4. Now, if the notion of property were highly controversial, or something that only right-wing ideologues accepted, then it would be a problematic starting point. But that is not the case. Only a tiny, extremist minority (perhaps overrepresented in the academy) reject property. This doesn’t prove that those people are wrong. But it does make it reasonable for me to address myself to those with the consensus view.
Kevin raises the case of a person who steals from a very rich person to prevent herself from suffering “a life of serious material deprivation.” This characterization is too vague and abstract for me to evaluate the action. But no doubt there is a range of cases in which ordinary people would disagree about the permissibility of theft.
Kevin seems to take this problem as a challenge to the notion of property as such, but I don’t think so. To believe in rights of some kind, one need not be an absolutist. Clearly, in common sense morality, property rights are not absolute. As I explain on p. 94 in the book, in some circumstances, a person may violate another’s property rights to prevent something much worse from happening. Now, exactly where the dividing line is – which harms one may avert by violating another’s property rights – is controversial. But I don’t see why I would have to take a stand on where that line is.
5. Perhaps the problem is that an anarcho-capitalist society might fail to correctly identify the line. True enough. But I don’t see any more reason to think that a state would correctly identify the line, so I don’t see a reason so far for preferring government over anarchy. The question is not whether anarchy is perfect but whether it is better than government.
6. But now suppose that extremist minority is right and there are in fact no property rights. This wouldn’t explain how any state has legitimate authority, nor would it undermine any of my arguments against theories of authority in part 1 of the book. So part 1 of the book would still be correct.
What about part 2, which defends anarcho-capitalism as a desirable social order? Anarcho-capitalism would perhaps be seriously flawed, insofar as it is a type of social order that enforces property rights. However, it’s not really clear that the anti-property theorist should prefer some form of government, since every government in the world also enforces a scheme of property rights. Whether anarcho-capitalism would be better or worse depends upon what the anti-property theorist thinks the correct social order would be – I can’t usefully address that without knowing more about what the propertyless society would be like.
7. In truth, Kevin read the manuscript before publication, and he told me then something like what he said here in his section IV on “justificatory liberalism.” At the time, I didn’t understand his point. After reading his latest comments, I still don’t. I (still) don’t see the difference between the “justificatory liberal” approach to which Kevin alludes and the version(s) of “hypothetical contract theory” that I addressed in sections 3.3-3.4.
This is from Kevin’s characterization of the justificatory liberal approach:
[T]hey only appeal to hypothetical consent stories as a heuristic for what people have most reason to do. […] [H]ypothetical models are evidence of what rules are most fair or just […].
But that’s precisely the sort of view I meant to address in sections 3.3-3.4 of the book. My guess is that Kevin was thrown off by the fact that I labeled this a form of “hypothetical social contract theory,” whereas he would not have called it that. Could the whole problem be that I mislabeled the view?
To remind readers of what I said about that sort of view: typically, it is not a sufficient justification for coercing someone into accepting some arrangement, to point out (even if this is true) that the person had most reason to accept the arrangement. It does not suffice that the arrangement have been fair, reasonable, and just. That doesn’t make coercion permissible. I gave the example of a potential employer who makes you a completely fair, reasonable, and favorable employment offer. You’d have to be irrational to reject it. Nevertheless, you reject the offer. May the employer now force you to accept it? Certainly not. Here, we see that the appeal to what you had reason to accept really cuts almost no ice at all.
Now, if the “justificatory liberal” has some explanation for why this appeal would cut more ice when used on behalf of the state, I’m going to need to hear more about that. Kevin has said, understandably, that he can’t work out the justificatory liberal view here now. But I’m going to need to hear at least a bit more to be convinced that I didn’t already refute the view.
8. One last thing that I’m going to need to hear more about. Kevin laments that I have ignored the empirical evidence showing “that people agree on general principles”. Okay, Kevin doesn’t have to reproduce all the evidence that he’s talking about here. For now, I just want someone to tell me one example of a useful political principle that all reasonable people agree on.
Once someone has done that, then we can talk about whether that principle can be used to justify the state.
Eight comments on Massimo Renzo’s post:
1. Massimo mentions a potential defense of government based on coordination problems (in the game theoretic sense). Roughly, it goes like this:
a. For us to live together peacefully and justly, we need an agreed-upon set of rules of conduct.
b. Coordination cannot be achieved through each person relying on their individual judgment, independent of others. This is because (a) human beings will disagree on what the best rules are, and (b) there may not even be a best set of rules. Furthermore, it must be assumed that this situation will continue to obtain in any realistic human society.
c. There is no other way to achieve coordination without a state.
d. Therefore, we need a state.
I have two main problems with this line of thought: first, premise (c) is false; second, even if it is true, conclusion (d) does not establish political authority.
2. Why is (c) false? There are many cases where coordination is achieved without anything like state intervention, even though there is no uniquely best alternative. Start with a small example. When I was in high school, I took a class where I learned to touch type. I learned on an IBM Selectric typewriter. Since then, for the last 25 years, this skill has never failed me. Every time I see a keyboard, no matter who made it, I can sit down and start typing, and I don’t have to check where the keys are located. The same experience is shared by every typist. What benevolent governmental statute established this coordination among all the world’s typists, typing class teachers, and keyboard manufacturers?
Of course, there is no such statute. It is perfectly legal to make a keyboard with keys in funny locations. No one is exerting coercion to maintain this coordination.
Ah, then it must be that there is some one arrangement of keys that everyone can see to be objectively correct? Again, no. Some arrangements are better than others, but there is no one layout that everyone would arrive at, if designing a keyboard independently.
So what maintains the coordination? Well, it wouldn’t pay to be the iconoclastic keyboard manufacturer who puts keys in a different arrangement from everyone else. Nor would it pay to be the iconoclastic typing teacher who teaches students a different layout from the one used on most keyboards. The convention is maintained by the fact that individuals want to be in step with others. How did the convention get established in the first place? The first typewriter manufacturer chose a layout. Subsequent makers followed that layout because they wanted to coordinate with existing makers. Typing teachers teach that layout because it is found on most typewriters.
That is one example. A more momentous example is provided by natural language. Members of a given society tend to make similar noises with their mouths, and to assign similar meanings to these noises. This represents a vast and extremely complex coordination. This is not because a coercive monopoly punishes people who invent their own words or who use words with nonstandard meanings. Nor is it because everyone can intuitively identify a single, objectively correct language. It is because people want to coordinate with each other – in this case, because we want to understand and be understood.
There are many more cases. Advocates of the Coordination Argument need to explain specifically why they think coercion is required to establish coordination.
3. But now suppose you’re unconvinced. For some reason, you still think that we need a state in order to coordinate. This still doesn’t establish political authority. Let’s focus in particular on the content-independence condition for political legitimacy. This is the condition that the state is entitled to make and enforce laws, even when the laws are useless or wrong (hereafter, I’ll call these “bad laws”). Here is Massimo:
I believe that authorities must be “accorded some leeway in the form of a content-independent entitlement to make rules as long as its rules are not too unreasonable” […]. Authorities perform the valuable functions that justify their existence […] despite the fact that they are not perfect. No authority is infallible. This is why infallibility should not be adopted as the standard of their justification.
There seem to be two points here: (a) the state can perform its functions, even if it sometimes enforces bad laws; (b) every state is fallible – that is, it sometimes enforces bad laws.
4. About point (a): it sounds as though the argument is that the state will still be desirable overall, even if it sometimes enforces bad laws. But how would this even appear to support the desired conclusion, that it is ethically permissible for the state to enforce bad laws? The reasoning seems analogous to the following: suppose Starbucks is a good company, which brings a lot of value to society. Suppose also that, even if Starbucks were to fire one employee on the basis of that employee’s race, it would still be, overall, a good company that brings a lot of value to society. Could we infer from this that it is ethically permissible for Starbucks to fire that employee?
Perhaps some work is supposed to be done by the assumption that the state is not only beneficial but necessary for any decent society. Okay. Suppose that farmer Al Falfa is the only person who knows how to grow food. His farm is necessary for his society to survive. Suppose also that, if Al were to steal a lollipop from a little girl, Al would still perform his necessary function of growing food. Does it follow that it is permissible for Al to steal the lollipop?
5. What about point (b)? Again, I can’t see how this could support the desired conclusion. From the fact that every state sometimes does x, it just does not follow that x is permissible.
Take another analogy. Suppose it’s true that not only all states, but all human beings are fallible. Suppose this is true specifically in the sense that all human beings at some point in their lives do something wrong. Does it follow that it is permissible to perform wrongful actions?
6. What about points (a) and (b) in conjunction? Okay, assume that all human beings are fallible (in the sense of sometimes committing wrongs) and that Al is the only one who can grow food. It does not follow that it is permissible for Al to commit wrongs.
7. Massimo’s last remark in the quoted passage – “infallibility should not be adopted as the standard of their justification” – suggests that he was not actually arguing for content-independence. Perhaps he is merely arguing that a state can be justified even if it sometimes enforces bad laws – that is, the existence of that state may be justified. That is quite a different thing from saying that those actions by the state (namely, its enforcement of bad laws) are permissible – which is what content independence commits one to.
8. In the book, I gave the following analogy (p. 175): It would be unreasonable to expect that any large corporation (which exists for many years) will never do anything wrong. But it does not follow from this that it is permissible for the corporation to commit wrongs. Similarly, even if we agree that it is unreasonable to expect that a state will never enforce a bad law, it does not follow that it is permissible for the state to enforce bad laws.
Massimo discusses this argument of mine. As far as I understand it, his response is something like this: “Oh yes, I agree. The corporation isn’t entitled to perform wrongs. And similarly, I agree that the state is not entitled to commit profound injustices. (But they’re still entitled to commit small to moderate injustices.)”
This reply would make sense if my claim about the example was (only) that it is wrong for a corporation to commit profound injustices. But that isn’t my claim. It is wrong for a corporation to commit any injustices, even small ones. Similarly, it’s wrong for the state to commit even small injustices.
Now return to the case of Al. Stealing a lollipop is not a profound injustice. It is a small injustice. Nevertheless, it is impermissible. This is not changed by the fact that all human beings are fallible (in the sense that they sometimes commit small injustices), nor the fact that Al continues to perform his necessary task of growing food even after stealing the lollipop.
These examples are merely for illustrative purposes. My main point is that the argument for content-independence is simply a non sequitur. “All states do x” and “the state can still perform its functions if it does x” just obviously do not establish “x is permissible.”
In sum: Coercion is not required to establish coordination. And even if it were, this would not establish content-independent authority.