Nozick begins Section 2 of Chapter 7 by presenting John Rawls’s influential theory of distributive justice. (If you are not familiar with Rawls’s work or need a refresher, check out this summary here, via Larry Solum’s Legal Theory Blog.) By way of introduction (p. 183), he tells us right off the bat that Rawls’s theory is “powerful, deep, subtle, wide-ranging, [and] systematic” but that he (Nozick) will focus on his disagreements with Rawls. Next, Nozick draws a sharp contrast between two ideal types (in the Weberian sense) of societies: (1) no cooperation and (2) general cooperation.
In one ideal type of society there is absolutely no cooperation among its members (p. 184): “Let us imagine n individuals who do not cooperate together and who each live solely by their own efforts.” This Hobbesian or individualistic society will produce a collective payoff of “S”, where S is the sum total of goods and services produced by its individual members. In such a Hobbesian society there is no need for a sophisticated theory of distributive justice, since each person simply receives “what he gets unaided by his own efforts.” (Nozick, 1974, p. 185.) The other ideal type of society, by contrast, is one of widespread and general cooperation: everyone cooperates with each other for the greater good or mutual advantage of all members of the society. By cooperating with each other, this idyllic society thus produces a much larger collective payoff of “T”, where T is the sum total of goods and services collectively produced.
To his credit, Nozick concedes that T > S. In other words, a cooperative society will always produce far more goods and services than an individualistic society. But a cooperative society’s material abundance generates a new problem: how are the spoils to be divided among its members? Here, Nozick makes the following brilliant insight and identifies a huge blind spot in Rawls’s theory of justice: whatever theory of distributive justice we prefer (Rawlsian, libertarian, etc.), should we apply our theory to T or just to T – S?
As Nozick notes (p. 184), Rawls does not distinguish between T and T – S. Instead, Rawls assumes that the entirety of T is up for distributive grabs! Perhaps such an assumption would be justified if no individual’s contribution to the joint social payoff T can be isolated or disentangled. Yet according to Nozick, Rawls’s own theory of justice does, in fact, distinguish between winners and losers (p. 188): “For Rawls goes out of his way to argue that inequalities [of wealth] are justified if they serve to raise the position of the worst-off group in the society.” Of course, no one knows the identities of the winners and losers in Rawls’s hypothetical original position, when the principles of distributive justice are negotiated from behind a veil of ignorance, but as Nozick notes, the fact remains that once the veil of ignorance is lifted, there will be winners and losers.
For our part, we would award round 1 to Rawls, not to Nozick, because in reality, S will be so small as to approach zero. In a purely individualistic or Hobbesian society with zero cooperation among its members, the best each person could hope for is bare subsistence. Nevertheless, Nozick will introduce a third type of society in the next subsection of Chapter 7 of ASU (“Terms of Cooperation and the Difference Principle”), one involving limited cooperation. We will review this third possibility in our next blog post.