Section One: The Private Production of Defense

1. The Private Production of Defense

AMONG THE MOST POPULAR AND CONSEQUENTIAL BELIEFS OF OUR age is the belief in collective security. Nothing less significant than the legitimacy of the modern state rests on this belief.

I will demonstrate that the idea of collective security is a myth that provides no justification for the modern state, and that all security is and must be private. Yet, before coming to the conclusion let me begin with the problem. First, I will present a two-step reconstruction of the myth of collective security, and at each step raise a few theoretical concerns.

The myth of collective security can also be called the Hobbesian myth. Thomas Hobbes, and countless political philosophers and economists after him, argued that in the state of nature, men would constantly be at each others’ throats. Homo homini lupus est. Put in modern jargon, in the state of nature a permanent underproduction of security would prevail. Each individual, left to his own devices and provisions, would spend too little on his own defense, and hence, permanent interpersonal warfare would result. The solution to this presumably intolerable situation, according to Hobbes and his followers, is the institution of a state. In order to institute peaceful cooperation among themselves, two individuals, A and B, require a third independent party, S, as ultimate judge and peacemaker. However, this third party, S, is not just another individual, and the good provided by s, that of security, is not just another “private” good. Rather, s is a sovereign and has as such two unique powers. On the one hand, s can insist that his subjects, A and B, not seek protection from anyone but him; that is, s is a compulsory territorial monopolist of protection. On the other hand, s can determine unilaterally how much A and B must spend on their own security; that is, s has the power to impose taxes in order to provide security “collectively.”

In commenting on this argument, there is little use in quarreling over whether man is as bad and wolf-like as Hobbes supposes, except to note that Hobbes’s thesis obviously cannot mean that man is driven only and exclusively by aggressive instincts. If this were the case, mankind would have died out long ago. The fact that he did not demonstrates that man also possesses reason and is capable of constraining his natural impulses. The quarrel is only with the Hobbesian solution. Given man’s nature as a rational animal, is the proposed solution to the problem of insecurity an improvement? Can the institution of a state reduce aggressive behavior and promote peaceful cooperation, and thus provide for better private security and protection? The difficulties with Hobbes’s argument are obvious. For one, regardless of how bad men are, s—whether king, dictator, or elected president—is still one of them. Man’s nature is not transformed upon becoming s. Yet how can there be better protection for A and B, if s must tax them in order to provide it? is there not a contradiction within the very construction of s as an expropriating property protector? in fact, is this not exactly what is also—and more appropriately—referred to as a protection racket? To besure, S will make peace between A and B but only so that he himself in turn can rob both of them more profitably. Surely S is better protected, but the more he is protected, the less A and B are protected from attacks by s. Collective security, it would seem, is not better than private security. Rather, it is the private security of the state, S, achieved through the expropriation, i.e., the economic disarmament, of its subjects. Further, statists from Thomas Hobbes to James Buchanan have argued that a protective state S would come about as the result of some sort of “constitutional” contract.¹ Yet, who in his right mind would agree to a contract that allowed one’s protector to determine unilaterally—and irrevocably—the sum that the protected must pay for his protection; and the fact is, no one ever has!²

Let me interrupt my discussion here, and return to the reconstruction of the Hobbesian myth. Once it is assumed that in order to institute peaceful cooperation between A and B it is necessary to have a state S, a two-fold conclusion follows. If more than one state exists, S1, S2, S3, then, just as there can presumably be no peace among A and B without S, so there can be no peace between the states S1, S2, and S3 as long as they remain in a state of nature (i.e., a state of anarchy) with regard to each other. Consequently, in order to achieve universal peace, political centralization, unification, and ultimately the establishment of a single world government are necessary.

Commenting on this argument, it is first useful to indicate what can be taken as non-controversial. To begin with, the argument is correct, as far as it goes. If the premise is correct, then the consequence spelled out follows. The empirical assumptions involved in the Hobbesian account appear at first glance to be borne out by the facts, as well. It is true that states are constantly at war with each other, and a historical tendency toward political centralization and global rule does indeed appear to be occurring. Quarrels arise only with the explanation of this fact and tendency, and the classification of a single unified world state as an improvement in the provision of private security and protection. First, there appears to be an empirical anomaly for which the Hobbesian argument cannot account. The reason for the warring among different states s1, s2, and s3, according to Hobbes, is that they are in a state of anarchy vis-à-vis each other. However, before the arrival of a single world state not only are s1, s2, and s3 in a state of anarchy relative to each other but in fact every subject of one state is in a state of anarchy vis-à-vis every subject of any other state. Accordingly, there should exist just as much war and aggression between the private citizens of various states as between different states. Empirically, however, this is not so. The private dealings between foreigners appear to be significantly less war-like than the dealings between different governments. Nor does this seem to be surprising. After all, a state agent s, in contrast to every one of its subjects, can rely on domestic taxation in the conduct of his foreign affairs. Given his natural human aggressiveness, however pronounced it may initially be, is it not obvious that s will be more brazen and aggressive in his conduct toward foreigners if he can externalize the cost of such behavior onto others? Surely, I am willing to take greater risks and engage in more provocation and aggression if I can make others pay for it. And surely there is a tendency of one state—one protection racket—to want to expand its territorial protection monopoly at the expense of other states and thus bring about, as the ultimate result of interstate competition, world government.³ But how is this an improvement in the provision of private security and protection? The opposite seems to be the case. The world state is the winner of all wars and the last surviving protection racket. Doesn’t this make it particularly dangerous? And will not the physical power of any single world government be overwhelming as compared to that of any one of its individual subjects?

¹ James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962); James M. Buchanan, The Limits of Liberty (Chicago: University of Chicago Press, 1975); for a critique see Murray N. Rothbard, “Buchanan and Tullock’s Calculus of Consent,” in idem, The Logic of Action, vol. 2, Applications and Criticism from the Austrian School (Cheltenham, U.K.: Edward Elgar, 1995); idem, “The Myth of Neutral Taxation,” in ibid.; Hans-Hermann Hoppe, The Economics and Ethics of Private Property (Boston: Kluwer, 1993), chap. 1.

² See on this particular point, Lysander Spooner, No Treason: The Constitution of No Authority (Larkspur, Colo.: Pine Tree Press, 1996).

³ See Hans-Hermann Hoppe, “The Trouble With Classical Liberalism,” Rothbard-Rockwell Report 9, no. 4 (1998).