Robert Nozick, my new favorite libertarian.

Sometime last summer I announced that I was reading Robert Nozick’s Anarchy, State, and Utopia, and that I’d try to post evaluations as I went. Well, due to a string of events I won’t relate here, I put the book aside for a few months, but picked it back up last week and I’ve now finished it. I’m so glad I did: if there are other books of political philosophy out there that are as delightful to read as this one, I’d like to get my hands on them. Nozick peppers his argument with stray questions and thought experiments, any one of which could provide enough material for a late-night philosophy argument. On top of that, he admits his doubts and points out weaknesses in his own argument, which is a pretty admirable way of writing. So even though I wasn’t a libertarian when I started the book and I’m not a libertarian now, I can say it’s well worth reading.

For those of you that aren’t familiar with Anarchy, State, and Utopia, it’s Nozick’s 1974 work of libertarian political philosophy, written in part to respond to John Rawls’s A Theory of Justice. Nozick shores up and extend lines of argument from John Locke’s Second Treatise of Government (1690), arguing that nothing more than a minimal state — i.e., one which does nothing more than protect its citizens from murder, theft, fraud, etc. — can be justified. One great virtue of the book, at least from the perspective of a non-libertarian, is that Nozick is much more interested in the argument than the application, and he never lets a policy preference get in the way of structural clarity. This isn’t to say that he doesn’t have policy preferences, only that the book is not about policy, and indeed, as I’ll argue below, probably doesn’t do enough to get us from philosophy to policy.

Anarchy, State, and Utopia is divided into three sections, and this division is reflected in the title. The first section answers an anarchist objection to Locke’s justification of a minimal state, the second extends the argument of the first by considering and rejecting possible justifications for a more-than-minimal state, and the third uses utopian considerations to offer a separate and independent state-of-nature argument for a minimal state. A state-of-nature argument, if you want Nozick’s take on it, strives to “explain fundamentally the political in terms of the nonpolitical” (6). Since we can’t find an example of a nonpolitical society in the world, we imagine such a society as best we can, and try to trace out how the political could emerge. (Is it necessary or merely contingent that no nonpolitical societies exist?)

A very crude summary of Locke’s state-of-nature argument: Most people generally do all right without a government, contracting with each other for goods and services. But those who make fraudulent contracts or steal or murder will cause an massive problem: each individual will have to enforce her own property rights, and this will take time. And then there will be disagreements over whether rights have been violated, and what kind of retribution is owed to whom, and hassle upon hassle will ensue. Finding this situation difficult, people in a state of nature will establish a social contract for a limited government. That’s just the nutshell version of Locke, which avoids most of the best arguments and all the interesting ambiguities in the Second Treatise.

The first section of Nozick’s book fills a gap in Locke’s argument (as construed by Nozick). Anarchists object to Locke that moving from a state of nature to even a minimal government will violate someone’s rights. For the minimal state doesn’t let people opt out and return to enforcing their own rights, and the natural-rights anarchist sees a grave injustice here that derails the Lockean argument. If the minimal state must violate people’s rights in order to come into being, how can it be morally justifiable?

Nozick responds to this objection by providing an account of how a minimal state (or something so near as to make no difference) could arise without these violations. It quickly appears that in a state of nature people will form associations for protection. Because there is an advantage to belonging to the strongest protective organization, there will be strong incentives for everyone to join the same group, so that there is one dominant protective association in an area. Such an agency will, for a fee, enforce agreed-upon rules of justice for its clients. But the dominant association is not a state until it has the power to prohibit other people or groups from enforcing their own rules of justice. It’s here that Nozick offers an intricate and fascinating discussion about when it’s justifiable to prohibit risky behavior. At some point, I will go through this argument again and take careful notes. The bottom line is that private enforcement of justice is risky enough for the dominant protective agency to be justified in prohibiting it — without adding any special “political” rights! That is, Nozick shows how the rights of individual clients of the dominant agency justify the agency in becoming a state-like entity.

But I don’t want to spend too much time on this first part of the book. Whether or not government can be justified is not, for me, a live question. I’m just not tempted by anarchism. So while Nozick’s line of reasoning in section one was interesting and enjoyable in itself, his conclusion was not surprising or challenging. It’s the second part of the book that takes on issues I care about.

Nozick spends a whole chapter arguing against the idea that distributive justice requires a more-than-minimal state. Nozick’s own view is that the entitlement theory of justice is the correct one. In an entitlement theory, there is a principle of acquisition, a principle of transfer, and a principle of rectification (151). Entitlement arises by proper acquisition or proper transfer, and if what people have is what they are entitled to have, then justice is satisfied, no matter what the distribution looks like across society. If proper acquisitions and voluntary transfers lead to inequality, nothing more can be said.

One might hold, in opposition to entitlement theories, that resources should be distributed across society by some pattern. Perhaps, when we look at society, everyone should have the same thing. Or perhaps a person’s share of resources should be proportional to his or her moral character. Maybe distribution should be in accordance with IQ. Nozick follows other libertarians in asking what happens when the pattern is realized. When everything’s distributed just right — what then? People’s voluntary actions will disturb the pattern: they’ll want to pay athletes or rock stars inordinate amounts of money. Does the state restrict their action? Liberty, says Nozick, will upset such patterns. If a patterned distribution is to be maintained, the state will have to severely restrict people’s liberty, and most of us will find this intolerable.

Nozick spends the second half of the chapter on distributive justice directly engaging John Rawls’s difference principle, but not having read Rawls I’m going to skip out on evaluating that section. It’s enough to say that Nozick finds Rawls’s conception of justice too flawed to justify a more-than-minimal state.

So Nozick would have us believe that non-entitlement theories of justice are wrong, and that the entitlement theory can’t justify a more-than-minimal state because it prohibits redistribution. Or that’s where things seem to be heading until the close of the chapter, when he admits:

“Perhaps it is best to view some patterned principles of justice as rough rules of thumb meant to approximate the general results of applying the principle of rectification of injustice … one cannot use the analysis presented and theory presented here to condemn any particular scheme of transfer payments, unless it is clear that no considerations of rectification of injustice could apply to justify it.” (231, italics in original)

This admission is huge — you can’t apply Nozick’s argument to the modern United States! Given the myriad ways in which the government has already (always already?) shaped land use and loan structures, can we ever know that the principle of rectification doesn’t need to be applied? But more on that below.

There is a grab-bag chapter on other assorted arguments for an expanded state, the most interesting section of which was the final one. There Nozick observes that even though many proponents of redistribution want to help the worst off, the middle class benefits much more from government interventions. He conjectures that this happens because the richest voting coalitions use such programs to buy the allegiance of the middle coalitions. The poorest coalitions, if they wanted the support of the middle classes, would have to match such programs even as they took enough money for themselves, and this costs more for them. Nozick wonders whether this conjecture has any predictive power; I for one would like to know if anyone followed up on this.

Finally, at the end of the section on the state, Nozick has a chapter called “Demoktesis,” which was far and away my favorite chapter of the book. The challenge Nozick sets for himself is this: is there a way to extend the state-of-nature account so that the minimal state yields to something much more like our own society? I would have assumed, given all that’s come before in the book, that we’d have to stop at the minimal state. But Nozick does manage to think of a way that a society could make this transition without violating anyone’s rights, and the line of reasoning he takes is simply ingenious. I never would have thought of it in a million years.

The final chapter of the book is the section on utopia. Nozick here provides a very clever independent justification for a minimal state, or rather a minimal state framework, and he does it without leaning so heavily on the notion of individual rights. The argument’s kind of complicated, so I’d recommend that you check out the book if you’re interested, but what’s interesting is that the his utopian society is not one where everyone lives by libertarian principles. It’s one where people accept a minimal state as a framework within which they try out their own non-libertarian communities. In other words, it’s a utopia that communitarians might love. Or perhaps not — Nozick runs into some familiar difficulties about conflict the rights of children and conflict between communities. These problems are familiar to anyone who’s thought hard about trying to situate non-liberal communities within a liberal framework. Won’t the neutral minimal state have to rule out many kinds of communities?

So goes the main thrust of the Anarchy, State, and Utopia. I’ve had to leave out any number of the fascinating counterexamples, conjectures, and asides that make the book so gratifying. I’ll close by explaining how I can like this book so much and still not be a libertarian.

When I started the book, I wanted to disagree with Nozick’s undefended assertion that, as the first line of the book has it, “individuals have rights, and there are things that no person or group may do to them (without violating their rights” (ix). I’ve been suspicious of the concept of rights for a while now; I think our moral lives are much more fully described in terms of obligations, and I’ve read enough virtue ethics to suspect that these obligations are derived from virtues. I can’t go along with the view that says any action is permitted as long as it doesn’t violate someone else’s rights. But is that what Nozick’s asking me to do? If the virtues are at all coherent and the basic obligations that spring from them are roughly consistent from person to person, then would I end up with something close enough to individual rights to support Nozick’s argument? Or would my softer take on rights undermine Nozick’s use of rights as moral side constraints, and therefore the rest of the book? When I re-read the book, I will be paying careful attention to this issue.

A better reason not to be a libertarian is the problem of the principle of rectification, discussed in the chapter on distributive justice. The entitlement theory of justice, Nozick notes, is a purely historical one. All that matters for any particular holding (property, etc.) is that it was properly acquired and then properly transferred. But if we’re talking about land, how can we say or know that any of it was properly acquired? How far back do we have to go? What does the entitlement theory say for my hometown of Charlotte, North Carolina, where the city government intervened in the composition of neighborhoods, the rules of ownership, and the value of property in different neighborhoods? Where city leaders used federal urban renewal money to raze the African-American business district in the 1960s, with little compensation for the displaced residents? What does the principle of rectification require? A busing policy for schools? Reparations? When conjoined with a survey of historical injustices in acquisition and transfer, the entitlement theory of justice might point away from the kind of society libertarians want to see. And to his credit, Nozick pointed this out in the book.

The final point I want to make comes from my reading of Aristotle and Alasdair MacIntyre. When I started drafting this post, it had been at least a few years since I read the section of After Virtue where MacIntyre discusses Rawls and Nozick. It was a little disturbing to find that MacIntyre deals a bit more forcefully with both of the concerns I discussed in the two paragraphs above this one, making me wonder how After Virtue shaped my reading of Nozick. But flipping through MacIntyre again reminded me of how powerful the concept of desert — that is, deserving something — is in Aristotelian philosophy. Nozick discards the notion of distributing according to what people deserve as a form of patterning. MacIntyre argues that in cases where it is unclear who is entitled to what, we have to resort to questions of who deserves what. (See 153 and 246-252 in After Virtue.) When we don’t have the historical facts that the entitlement theory needs, Nozick’s approach to justice will leave us in the lurch.

Anarchy, State, and Utopia is a very good book. It’s got three surpassingly clever solutions to philosophical problems — the argument that a dominant protective agency in a state of nature is justified in preventing private enforcement of justice, the explanation of how a more-than-minimal state can arise without violating anyone’s rights, and the derivation of the framework for utopia. Nozick’s book made me think harder about my own muddled political philosophy than any libertarian has managed to do in a long time, and I’d certainly recommend it to anyone who has been jaded by exposure to too many politically zealous free-marketers.