Robert Nozick was, with John Rawls, one of the two most important and influential political philosophers in the Anglo-American analytic tradition. Where Rawls's work systematized the egalitarian liberalism of the political Left, Nozick did the same for the free-market libertarianism of the Right. A thinker with wide-ranging interests, he also made a number of significant contributions to areas of philosophy outside political theory, most notably in epistemology and the metaphysics of personal identity.
Most critics of the libertarian minimal state don't complain that it allows for too much government; they say that it allows for far too little. In particular, they claim that a more-than-minimal state is necessary in order to fulfill the requirements of distributive justice. The state, it is held (by, for instance, Rawls and his followers), simply must engage in redistributive taxation in order to ensure that a fair distribution of wealth and income obtains in the society it governs. Nozick's answer to this objection constitutes his "entitlement theory" of justice.
Talk about "distributive justice" is inherently misleading, Nozick argues, in that it seems to imply that there is some central authority who "distributes" to individuals shares of wealth and income that pre-exist the distribution, as if they had appeared like "manna from heaven." Of course this is not really the way such shares come into existence, or come to be "distributed," at all; in fact they come to be, and come to be held by the individuals who hold them, only through the scattered efforts and transactions of these innumerable individuals themselves, and these individuals' efforts and transactions give them a moral claim over these shares. Talk about the "distribution of wealth" covers this up, and unjustifiably biases most discussions of distributive justice in a socialist or egalitarian liberal direction.
A more adequate theory of justice would in Nozick's view enumerate three principles of justice in holdings. The first would be a principle of justice in acquisition, that is, the appropriation of natural resources that no one has ever owned before. The best-known such principle, some version of which Nozick seems to endorse, is the one enshrined in Locke's theory of property, according to which a person (being a self-owner) owns his labor, and by "mixing his labor" with a previously unowned part of the natural world (e.g. by whittling a stick found in a forest into a spear) thereby comes to own it. The second principle would be a principle of justice in transfer, governing the manner in which one might justly come to own something previously owned by another. Here Nozick endorses the principle that a transfer of holdings is just if and only if it is voluntary, a principle that would seem to follow from respect for a person's right to use the fruits of the exercise of his self-owned talents, abilities, and labor as he sees fit. The final principle would be a principle of justice in rectification, governing the proper means of setting right past injustices in acquisition and transfer.
Anyone who got what he has in a manner consistent with these three principles would, Nozick says, accordingly be entitled to it – for, his having abided by these principles, no one has any grounds for complaint against him. This gives us Nozick's entitlement theory of distributive justice: a distribution of wealth obtaining in a society as a whole is a just distribution if everyone in that society is entitled to what he has, i.e. has gotten his holdings in accordance with the principles of acquisition, transfer, and rectification. And it is therefore just however equal or unequal it happens to be, and indeed however "fair" or "unfair" it might seem intuitively to be. Standard theories of distributive justice, Nozick says, are either ahistorical "end-state" or "end-result" theories, requiring that the distribution of wealth in a society have a certain structure, e.g. an egalitarian structure (regardless of how the distribution came about or how people got what they have); or they are historical theories requiring that the distribution fit a certain pattern reflecting such historical circumstances as who worked the hardest or who deserves the most. The entitlement theory of justice is historical yet unpatterned: The justice of a distribution is indeed determined by certain historical circumstances (contrary to end-state theories), but it has nothing to do with fitting any pattern guaranteeing that those who worked the hardest or are most deserving have the most shares. What matters is only that people get what they have in a manner consistent with the three principles of justice in holdings, and this is fully compatible with some people having much more than others, unlucky hard workers having less than lazier but luckier ones, morally repulsive individuals having higher incomes than saints, and so forth.
Nozick illustrates and defends the entitlement theory in a famous thought-experiment involving the basketball player Wilt Chamberlain. Imagine a society in which the distribution of wealth fits a particular structure or pattern favored by a non-entitlement conception of justice – suppose, to keep things simple, that it is an equal distribution, and call it D1. Nozick's opponent must of course grant that this distribution is just, since Nozick has allowed the opponent himself to determine it. Now suppose that among the members of this society is Wilt Chamberlain, and that he has as a condition of his contract with his team that he will play only if each person coming to see the game puts twenty-five cents into a special box at the gate of the sports arena, the contents of which will go to him. Suppose further that over the course of the season, one million fans decide to pay the twenty-five cents to watch him play. The result will be a new distribution, D2, in which Chamberlain now has $250,000, much more than anyone else – a distribution which thereby breaks the original pattern established in D1. Now, is D2 just? Is Chamberlain entitled to his money? The answer to these questions, Nozick says, is clearly "Yes." For everyone in D1 was, by hypothesis, entitled to what he had; there is no injustice in the starting point that led up to D2. Moreover, everyone who gave up twenty-five cents in the transition from D1 to D2 did so voluntarily, and thus has no grounds for complaint; and those who did not want to pay to see Chamberlain play still have their twenty-five cents, so they have no grounds for complaint either. But then no one has any grounds for a complaint of injustice; and thus there is no injustice.
What this shows, in Nozick's view, is that all non-entitlement theories of justice are false. For all such theories claim that it is a necessary condition for a distribution's being just that it have a certain structure or fit a certain pattern; but the Wilt Chamberlain example (which can be reformulated so that D1 is, instead of an egalitarian distribution, a distribution according to hard work, desert, or whatever) shows that a distribution (such as D2) can be just even if it doesn't have a particular structure or pattern.
Moreover, the example shows that "liberty upsets patterns," that allowing individuals freely to use their holdings as they choose will inevitably destroy any distribution advocated by non-entitlement theories, whether they be socialist, egalitarian liberal, or some other theory of distribution. And the corollary of this is that patterns destroy liberty, that attempts to enforce a particular distributional pattern or structure over time will necessarily involve intolerable levels of coercion, forbidding individuals from using the fruits of their talents, abilities, and labor as they see fit. As Nozick puts it, "the socialist society would have to forbid capitalist acts between consenting adults." This is not merely a regrettable side-effect of the quest to attain a just distribution of wealth; it is a positive injustice, for it violates the principle of self-ownership.
Distributive justice, properly understood, thus does not require a redistribution of wealth; indeed, it forbids such a redistribution. Accordingly, the minimal state, far from being inconsistent with the demands of distributive justice, is in fact the only sure means of securing those demands.