More on the Nozickian Minimal State
by Randy E. Barnett
February 20th, 2008
Julian Sanchez thinks that Nozick did make a descriptive argument for why a “monopolistic” dominant protection agency would develop. He may be right and I thank him for his response. But Nozick’s claim in the passages highlighted by Julian is far from transparent. In the first passage he references, for example, Nozick describes the natural evolution of what he calls a “federal judicial system.” In the Journal of Libertarian Studies article, “The Invisible Hand Strikes Back,” Roy Childs notes the inaccuracy of this characterization in a passage worth quoting at length:
According to Professor Nozick, then, if competing protection associations make arrangements between themselves to settle disputes, we have a type of “federal judicial system,” a variant of government. This is surely metaphorical and unjustified. Surely, if we take all the protective devices in use in a given society and lump them together, then the total has what some might call a “monopoly” on protection. Similarly, all farmers taken collectively have a “monopoly” on growing food. But this is tautological.
The real point which Professor Nozick wishes to make is that if either of these alternative courses result, then we have a “legal system” resulting. Now, no one has ever denied that there would indeed be a “legal system” under anarchism. Many prominent anarchists have claimed that they advocate that structures and processes (even content, in some cases), be separated from the state, and the state abolished entirely. If one is going to term any “legal system” in this broad sense a “state,” then there is little point in pursuing the matter.
Discussion may proceed along more productive lines if we distinguish between two radically different types of legal systems: a “market legal system” and a “state legal system.” A “market legal system” could be designated as a system of rules and enforcement procedures which arises from the processes of the market economy: competition, bargaining, legal decisions, and so forth; a legal system whose order is “spontaneous” in the Hayekian sense. A “state legal system” on the other hand, could be designated as a system of rules and enforcement procedures which are designed by the state apparatus, as a result of political procedures, and imposed by force upon the rest of society.
In a society with a “market legal system,” the shape of the legal system is determined by the processes set in motion by the actions of a number of independent agencies whose plans may conflict, and therefore cause some adjustment in the means-ends structure of themselves and others. Independent agencies, then, can make agreements, reach decisions, set precedents, bargain and so forth, producing a legal “order” which is not designated by anyone. The resulting system is not a “federal system” in the traditional manner: we may have ad hoc decisions for individual disputes, procedures agreed upon in advance, such as drawing the names of arbitrators out of a hat, alternating arbitrators chosen by each agency, ad infinitum. We need not suppose that any permanent, distinct, appeals system has been erected. (If one had, it would not change our essential argument.) Anarchism, then, can have a legal system, a “market legal system” as opposed to a “state legal system.” The analogy is to the distinction drawn between state-economic systems and unhampered market economic systems. Both are systems, but not of the same sort; they are built on different principles of organization and are the manifestations of different processes altogether. What we shall conclude, then, is that if the third of Professor Nozick’s three alternatives results, then there will not be a state apparatus as the result.
Also, while it does seem that Nozick attempted a descriptive prediction of the rise of a dominant protection agency (DPA), the DPA is not a “monopolistic” minimal state even on Nozick’s account. His move from the DPA to a monopoly minimal state involved a normative claim about the right of the DPA to protect its clients from the procedures of rivals that it deems to be too risky provided that what he calls the principle of compensation is respected. This moral argument, I believe, is the main thrust of his challenge to a polycentric legal order, and four replies to it appear in the first issue of the Journal of Libertarian Studies. (BTW, at the conclusion of his piece, Childs offers a reverse dynamic by which the dominant protection agency can be undermined by competitors respecting the same principles Nozick invokes.)
In the end, however, I have no stake in the claim that Nozick never offered a prediction of how a minimal state would evolve from a polycentric constitutional order. Julian could be right that he did and I could be wrong that he did not.
