In his refutation of the famous libertarian arguments of Robert Nozick’s Anarchy, State, and Utopia, socialist thinker G. A. Cohen showed the absurdity of thinking that we had to accept an unequal society in order to preserve liberty.
Robert Nozick’s Anarchy, State, and Utopia (1974) is one of the books about political philosophy most frequently assigned to college students. Fifty years after its publication, I’d be surprised if you could walk into a reasonably well-stocked Barnes & Noble anywhere in the English-speaking world without finding a copy.
This level of success is, in some ways, well-deserved. Nozick is an excellent writer, combining stylistic flair with a high level of rigor in most of his arguments. It’s a pleasure to read, even if (like me) you deeply abhor his conclusions.
Nozick was, when he wrote it, a hardcore libertarian. He thought any distribution of wealth whatsoever was acceptable as long as it came from letting the chips fall where they may in a free market. The only justifiable state would be a “night-watchman state” that would be “limited to protecting persons against murder, assault, theft, fraud, and so forth.”
If inequality becomes so severe that poor people are starving to death in the streets, it may be admirable for the rich to voluntarily help them, but taxing the rich to fund a welfare state would be an unacceptable violation of their property rights. And nationalizing the businesses currently owned by the rich and turning them over to the democratic management of workers or larger communities, as socialists propose, would certainly be out of the question. However pleasant a more equal society might sound, Nozick thought that there was no way to achieve and then preserve such a society without illegitimate violations of liberty.
As it happens, Nozick himself had backed away from this extreme position by the end of the 1980s, although his second thoughts have never received nearly as much attention as Anarchy, State, and Utopia. And no one casually browsing the shelves at Barnes & Noble is likely to run into the work of the philosopher who took apart Nozick’s arguments point-by-point in the 1970s — the Canadian-born “analytical Marxist” thinker G. A. Cohen.
The Wilt Chamberlain Case
One of the most memorable arguments in Anarchy, State, and Utopia concerns the basketball superstar Wilt Chamberlain. Nozick’s Wilt Chamberlain thought experiment is supposed to show that a just distribution of goods is one that comes about in the right way, rather than one that conforms to a certain pattern, like egalitarianism (the idea that, in some important sense, distribution should be at least roughly equal) or even sufficientarianism (the idea that, while it doesn’t matter how big the gap is between the top and bottom end of a distribution, the bottom should be set high enough that the people at the bottom can have a reasonable and dignified life).
Nozick thinks he can refute all such ideas, with the following thought experiment:
Let distribution D1 be whatever you think the best pattern of wealth distribution looks like. If you’re a strict egalitarian, imagine that everyone has an equal share. If you accept some looser quasi-egalitarian principle, like John Rawls’s view that inequalities can only be acceptable if they work to the benefit of the worst off by creating incentives that lead people to do things that benefit everyone, assume that D1 embodies all the nuances of your view.
Next imagine that Wilt Chamberlain refuses to play basketball unless everyone who attends one of his games drops twenty-five cents in a special box with his name on it. Chamberlain is so popular and beloved that a million people attend his games, each cheerfully dropping a quarter in the box. He now has $250,000 more than whatever he had in D1. (According to this handy inflation calculator, a quarter of a million dollars when Nozick’s book came out is well over a million and a half dollars today.) We’ve moved from D1 to an at least somewhat less egalitarian D2.
But, Nozick asks, how can the proponent of a “pattern-based” view of distributive justice coherently object?
There is no question about whether each of the people was entitled to the control over the resources they held in D1; because that was the distribution . . . that (for the purposes of argument) we assumed was acceptable. Each of these persons chose to give twenty-five cents of their money to Chamberlain. They could have spent it on going to the movies, or on candy bars, or on copies of Dissent magazine, or of Monthly Review.
(Dissent and Monthly Review were the highest-profile socialist magazines at the time Nozick was writing. Presumably, if he’d been writing in 2024 instead of 1974, that little jab would have been at the expense of Jacobin.) In any case, Nozick says that, in light of these facts (that everyone in D1 had as much wealth as the theorists he’s arguing against think they should have had, that this wealth was then theirs to do with what they wanted, and that this rather than movie tickets or copies of socialist magazines is what they wanted to spend it on), the idea that the transition from D1 to D2 involves the introduction of some sort of injustice is absurd. Who has legitimate grounds for complaint? Not the people who voluntarily gave their quarters to Chamberlain, not the people who decided not to (after all, they still have their initial shares, so what do they have to complain about?), and certainly not Chamberlain himself.
But now you have at least one wealthy man. Presumably, this isn’t a one-off event! Starting from D1, new fortunes can emerge through similarly innocuous mechanisms. Will you stop Chamberlain and similar beneficiaries of innocent wealth accumulation from voluntarily passing their fortunes to their heirs? If not, before you know it you’ve got intergenerational class divisions. Will you stop Chamberlain from using some of this money to buy a factory and hire willing workers? If not, then even if D1 was a socialist society where the means of production were collectively owned and democratically run by the public at large, capitalism will be reintroduced in D2.
The lesson Nozick draws is that the only way to maintain a particular pattern of distribution over time is to forbid certain kinds of “capitalistic acts between consenting adults.” Conversely, any kind of robustly maintained liberty “upsets patterns.”
How Patterns Preserve Liberty
The range of economic views undergraduates are most likely to be presented with in most introductory surveys of political philosophy in the “analytic tradition” dominant in contemporary Anglo-American philosophy ranges from Nozickian libertarianism to the philosophy of John Rawls, who’s usually understood as a defender of modified welfare state capitalism. (Later in life, Rawls explicitly rejected this reading, but most people who read extracts of his A Theory of Justice don’t know that.) This is a map of the terrain students are being introduced to that entirely leaves out socialist and Marxist alternatives.
Similarly, someone browsing the philosophy shelves at Barnes & Noble will be likely to run into both A Theory of Justice and Anarchy, State, and Utopia, but they’re unlikely to find any of the books written by analytical Marxist G. A. Cohen. That’s a shame for many reasons, not least that one of the best responses to Nozick’s Chamberlain argument was written by Cohen. Cohen’s intervention is well-known and respected by academic philosophers, but it’s always been less well-known to the general public than Nozick’s book.
In his 1977 paper, “Robert Nozick and Wilt Chamberlain: How Patterns Preserve Liberty,” Cohen addresses the Chamberlain argument point-by-point. Take, for example, the crucial principle Nozick is appealing to in the Chamberlain story. “Whatever arises from a just situation as a result of fully voluntary transactions on the part of all legitimately concerned persons is itself just.” Without this crucial assumption, Nozick can’t trap the advocates of D1 into “therefore” having to accept the justice of D2. As Cohen says, Nozick is so utterly convinced of this principle that he assumes that it “must be accepted by people attached to a doctrine of justice which in other respects differs from his own.” But we should examine that assumption.
A standard way to test whether we should accept such a principle, Cohen points out, would be to look for possible scenarios that conform to the principle but that are nevertheless manifestly unjust. And it’s very easy to find one. The “strongest counterexample” would be slavery. If “voluntary self-enslavement is possible” but slavery is unjust (even when it originates in someone voluntarily deciding to sell themselves into slavery), Nozick’s principle can’t be assumed to be entirely correct.
The problem is that Nozick anticipates this objection — and bites the bullet. He says that while of course hereditary slavery is wrong, people have a right, in the (perhaps exceedingly rare) cases where someone would freely make this choice, to enter into even a contract making them someone’s slave for life. In light of Nozick’s acceptance of even the most extreme disparities of wealth (so long as they come about in the right way), though, it’s far from clear that this would be particularly rare in his ideal society.
Things get even grimmer when we remember that the role of his night-watchman state includes enforcing contracts and protecting property-holders against “theft.” When we put all this together, it’s hard to avoid the conclusion that in Nozick’s libertarian utopia, not only would economically desperate people have the option of literally selling themselves (perhaps to save their families from starvation), but when such people later had second thoughts, the night-watchman state would be in the business of catching runaway slaves.
At this point, it would be tempting to say that, in biting a bullet of this size, Nozick has refuted himself and no further external refutation is necessary. Cohen doesn’t take this route. The point about slavery is just the opening move in his paper. He moves on to examine, for example, the claim that, because the transfers to Chamberlain were voluntary and those who opted out of them still have their shares, no one has any grounds to object.
In defending the justice of the Chamberlain transaction, Nozick glances at the position of persons not directly party to it: “After someone transfers something to Wilt Chamberlain, third parties still have their legitimate shares; their shares are not changed.” This is false, in one relevant sense. For a person’s effective share depends on what he can do with what he has, and that depends not only on how much he has but on what others have and on how what others have is distributed. If it is distributed equally among them he will often be better placed than if some have especially large shares.
There is, moreover, the question of those third parties who have yet to be born, who Nozick leaves entirely out of the Chamberlain parable. They too have an interest in growing up in a more just and equal society. Nozick dismisses any concerns about the wealth of others as “envy,” but what Cohen relentlessly emphasizes is that inequalities in wealth beyond a certain point can’t be meaningfully separated from inequalities in power. In every nominally democratic society that’s ever existed, extreme material inequality has translated itself into unequal influence on the political process. That much is obvious.
What Cohen emphasizes, though, is not unequal power in this secondary sense but the unequal economic power at the heart of the capitalist mode of production. Most of the working-age population under capitalism has no realistic choice except to sell their working hours to capitalists. They then spend half of their waking hours, most days of the week, following orders from an unelected boss, and producing wealth over whose eventual distribution they have little say.
Cohen writes:
One difference between a capitalist state and a slave state is that the natural right not to be subordinate in the manner of a slave is a civil right in liberal capitalism. The law excludes formation of a set of persons legally obliged to work for other persons. That status being forbidden, everyone is entitled to work for no one. But the power corresponding to this right is differentially enjoyed. Some can live without subordinating themselves, but most cannot.
Wilt Chamberlain is a strategically chosen example because, looking at the case with the habits of mind we have all acquired growing up in what’s already a highly inegalitarian society, we have no particular reason to object to a basketball player we enjoy climbing to the top of that society (or at least far nearer to the top than the great bulk of us will ever reach). If we’re going to have a society in which there is a category of rich and powerful, then who “better and more innocently” could join their ranks?
But the deeper question is whether we want society to be divided between a working-class majority and a minority with wealth that can be translated into economic power over the rest of us. If most of us are forced to subordinate ourselves, we are therefore less free. Maintaining a relatively egalitarian pattern of distribution, whether by phasing money out entirely in some extremely advanced stage of the socialist future, or just by taxing the Chamberlains of the world until the inequality between them and the rest of us isn’t enough to start building up intergenerational fortunes or privately buying up means of production, is justified. That’s true even if (at least for the sake of argument) it infringes to some extent on the liberty of those having their wealth redistributed away — precisely in order to prevent far greater infringements on the liberty of the majority of the population.
By the way, Cohen points out in passing, if you really “think it obvious” that Wilt Chamberlain really wouldn’t continue to play if he weren’t allowed to build up great wealth, you’ve merely shown that you misunderstand “human nature, or basketball, or both.”
Liberty and Socialism
Nozick objects to an egalitarian position of that sort on two grounds.
First, even in an extreme situation where someone is literally left with the options to either work for a capitalist “or starve” (an extreme situation that, by the way, would become much less rare in the absence of the welfare state 1970s Nozick would abolish), he denies that they’ve thereby been forced to work for a capitalist or starve. That’s because Nozick defines coercion in such a way that someone is only being coerced if their moral rights are violated. As Cohen points out elsewhere, this is an absurd abuse of language. Taken seriously, it would suggest that “if a criminal’s imprisonment is morally justified, he is then not forced to be in prison.”
Second, Nozick says that liberty is such an absolute value that it’s not acceptable to intentionally engage in even slight violations of the rights of some in order to protect against far greater violations of the liberty of others. He thinks this has something to do with the “separateness of persons.” We can’t sacrifice the good of some for the greater good of society, because there is no “social entity” that “undergoes a sacrifice for its own good” but only “different individual people” with “their own individual lives.”
Cohen simply notes in response that nothing about the tradeoff in question requires us to appeal to a “social entity.” We can simply think that forcing the working-class majority into a lifelong position of subordination is so much worse for those individuals than the comparatively trivial violation of the liberty of capitalists involved in wealth redistribution that this isn’t a difficult choice. We can, indeed, meditate on the separateness of persons and thereby realize it would be so bad to force anyone to spend the one life they have this way that anyone who thinks it can be justified needs a much better argument than anything Nozick offers.
Cohen concludes that “‘libertarian’ capitalism sacrifices liberty to capitalism, a truth its advocates are able to deny only because they are prepared to abuse the language of freedom.” That damning summation is well-earned, and fifty years later Cohen’s insight continues to be depressingly relevant in contexts far removed from academic philosophy.
This post was originally published on Jacobin.