We’re done with the preface of Anarchy, State, and Utopia. On to chapter 1.

Nozick asks, “If the state did not exist would it be necessary to invent it? Would one be needed and would it have to be invented?” (1)

Nozick says the fundamental question for political philosophy is not about how states should be organized, but whether there should be a state at all. We’re used to living in nation-states, so we tend to act as if the nation-state were a natural, inevitable mode of organization. But of course it’s not. Political philosophers should not take it for granted—it’s our job to question what everyone else blindly assumes.

States or governments are pretty weird things, too. Following Kavka, let’s define the government of a society as a subset of that society which 1) claims a legitimate monopoly on the use of coercive force, which 2) claims the right to make and impose rules, and which 3) has the effective coercive power to enforce these claims. (Note that to say a government is legitimate is to say it is permissible for it to use force to enforce its rules.) Let’s add to this that when governments issue rules, they expect people under their jurisdiction to comply not out of fear, but out of a sense of duty. Governments see themselves as having authority—i.e., as being able to create moral obligations in their citizens when they create rules. For instance, Uncle Sam forbids me from smoking marijuana. Uncle Sam expects me to comply not merely to avoid punishment. Uncle Sam expects me to see myself as morally obligated to follow his commands.

If that’s what a state or government is, how could there even be such a thing?

One classic way to try to justify (or attack) the state is through state of nature theory. One tries to imagine what life would be like without government, and then one compares that to life with government. If things would be better with government than without, then that is presumptive grounds for favoring the state over anarchy. We might do the same thing to try to justify any other institution—if life without the institution is worse than life with it, the institution is pro tanto justified.

Part of the problem with these thought experiments, though, is getting the counterfactuals right. There are lots of ways a state of nature could go. Hobbes imagined the state of nature would be a total disaster, but perhaps he was wrong. To show that the state is better than the state of nature, it won’t suffice to compare an optimistic view of the state with a pessimistic view of the state of nature.

Nozick suggests that one might try a “minimax” criterion: Compare the most pessimistic version of the state of nature to the most pessimistic version of the state, then pick the better of the two worst cases. (This criterion is also often called “maximin” in philosophy. We can think of it as minimizing maximum risk or maximizing minimum gains.) Nozick isn’t so sure that this criterion would select the state over anarchy. The worst possible totalitarian regime is probably much worse than the worst anarchic mode of organization. (Google Holodomor, Great Leap Forward, or Khmer Rouge if you need some help imagining what bad states can do.)

The fundamental problem with state of nature theory is knowing which anarchy to compare to which state. Hobbes wanted compare the two most likely scenarios, but plenty of people think he was too pessimistic about anarchy or the state.

Also, philosophers aren’t just interested in asking which is the safest bet. After all, one goal of political philosophy is to identify what kind of society, if any, can be fully just. Many political philosophers, following Rawls, work mostly in ideal theory, where they imagine people to be motivated by a nearly perfect sense of justice. In ideal theory, it becomes especially difficult to establish that a state is preferable to anarchy. After all, if people have a perfect sense of justice, as Rawls often imagines them to have when defending his theory, then it is unclear why they need a monopolistic coercive authority to get along. (See here for an argument.) Rawls doesn’t even give it an honest try. In Theory of Justice, he does ask why we should have a state at all. But then he just compares the state in ideal theory to an anarchic society in non-ideal circumstances, and of course concludes the state is better. Oops.

Nozick has a clever way solution to the problem. His plan is to start with very optimistic assumptions about anarchy. He will imagine an anarchic society functioning as well as anarchists like Murray Rothbard or David Friedman think it would function. He then wants to show that a state would naturally arise in this scenario, and could do so without violating any of the libertarian rights Rothbard and Friedman think we have. As Nozick summarizes:

If one could show that the state would be superior even to this most favored situation of anarchy, the best that realistically can be hoped for, or would arise by a processing involving no morally impermissible steps, or would be an improvement if it arose, this would provide a rationale for the state’s existence; it would justify the state. (5)

So, if out of the best-case scenario for a realistic form of anarchy, a state would or could blamelessly arise, and if this would be an improvement, then the state is justified.

What exactly does it mean to say the state is justified, though? Of course he doesn’t mean that just any state anywhere is justified—at most some states are justified. When Nozick claims that state can be justified, that is consistent with holding that all extant states are unjustified. Does it mean we could just impose a minimal state on a well-functioning anarchic society, if we ever found one? Probably not. If X could or even is likely to arise out of Y through a series of blameless steps, and if Y is better than X, it doesn’t automatically follow that we can impose Y upon people who are in situation X. (E.g., suppose I could prove that nearly everyone who has a blue stripe Mesa Boogie Mark III will end up naturally wanting to buy a Mark V instead, and I could prove that the Mark V is superior to the Mark III. It wouldn’t follow that people with Mark IIIs are obligated to buy Mark Vs, nor would it follow that I could force them to buy Mark Vs right now.)

I have some views on what Nozick thinks it means to justify a state this way, but I’ll just leave it as a puzzle here.

Nozick says in a footnote that showing a state would naturally arise from anarchy is not quite enough to justify a state. After all, it might be that anarchy naturally “deteriorates” into the state. This would “resign us to [the state’s] existence” but wouldn’t quite justify the state. Instead, we need an account showing that even in the most optimistic but realistic version of anarchy, a state would or could naturally arise, through a series of morally permissible steps, and that this change would be an improvement.