Financing the Non-Coercive State

by Roderick T. Long

[This essay was submitted for the Institute for Humane Studies’ Donald Bogie Philosophy Prize Competition (22 December 1987), which it won. Obviously this was written before I became an anarchist; see the comments I’ve added at the end. I have a vague memory that my first contact with either Spencer MacCallum or Carl Watner (or possibly both) resulted from this essay, but I can’t recall the details.]

Few aspects of the libertarian vision of a non-coercive society strike non-libertarians as more preposterously utopian than the dictum that “taxation is theft.” How could society run, how could government operate, without tax revenue? Libertarians have been quick to suggest non-coercive methods of paying for governmental or quasi-governmental protection. I shall argue that several of the most popular such suggestions, while perhaps practicable, are unacceptable from a moral point of view. I shall also, however, outline some ways in which government might legitimately be financed non-coercively after all.

Privatisation is usually a libertarian’s favourite solution to problems of funding. It’s not surprising, then, that the doctrine of anarcho-capitalism has gained so much favour in the libertarian community. For anarcho-capitalism proposes the privatisation of the government in toto; all the services traditionally provided by government, including police protection and court systems, would be offered by private forms competing on the free market.

The debate between anarchist and minarchist libertarians is a lengthy one, and will not be resolved here. But I do want to call attention to what seems to me an undesirable consequence of the anarcho-capitalist system, one I haven’t seen discussed elsewhere.

Nearly all laws necessarily contain an arbitrary element. For example, suppose that it is right that there be a copyright law. The world is an anarcho-capitalist one; so the Bifrost Protection Agency establishes the law that copyright should extend fifty years past the holder’s death, while the Bodhisattva Protection Agency instead establishes the law that copyright should extend only forty-five years past the holder’s death. If either one of these laws is justified, the other one is too, since the requirements of morality are not fine enough to allow us to discriminate between the two options. (Nor need this be an epistemological problem; there may simply be no fact of the matter as to which law is better.) At this point, some arbitrariness is permissible: a protection agency may justifiably choose either rule. Thus far, then, neither Bifrost nor Bodhisattva appears to have done anything wrong.

But note the consequence. Suppose Fred sells copies of Beatrice’s work forty-eight years after Beatrice’s death. If Beatrice is a client of Bodhisattva, Fred has done nothing wrong; but if Beatrice is a client of Bifrost, Bifrost may justifiably punish him. Yet this violates the principle that all persons are equal before the law. It is unjust to institute a law that makes it a punishable offense to do to one person something it is permissible to do to another person. Now neither Bifrost nor Bodhisattva has instituted such a rule individually. But since (let’s assume) both protection agencies operate in the same geographical area, the social system of that area in effect embodies an unjust practice. In order to rectify this injustice, it would be necessary to force all the protection agencies in given area to have mutually consistent laws; so a minimal state would be required in order to enforce the standardisation of permissible moral arbitrariness (since it is only the non-uniformity of such arbitrariness that generates the injustice).1

Our problem, then, is the financing of the minimal state. The minarchist solution most commonly offered is that government, like any other business, should charge for its services. In the case of public goods like national defense, they could be at least partially paid for out of the revenues from private goods like contract protection. Tibor Machan explains:

... government provides both [private and public] goods, and payment for the former would also serve to fund the latter. ... this is similar to the way in which Coca Cola buyers pay for the overhead and security provisions at the bottling plants .... Every valid contract imposes a burden on the legal system and its administrators, for the “machinery” for interpreting and enforcing contracts, should disputes arise, must be in place. So providing this protection requires expenditures on the part of government. A system of contract fees, collected at the time of signing or registering of contracts ... would provide funding for this government activity. ... Like contract protection, other governmental services are deliverable to individuals, so fees for the services rendered could be established. Among such potentially individualized services are securing political justice and defending private homes and businesses ....2

The law enforcers, like other labourers, should charge for their services may, from a libertarian standpoint, seem unexceptionable. But recall that, unlike private protection agencies, these law enforcers are the agents of a minimal state.

Let’s suppose that this state holds a monopoly on the legitimate use of force in a particular geographical region. The services it provides it will then prohibit anyone else from providing. I submit that in such a situation the minimal state has no moral right to charge for its services. One may legitimately sell only what one has property in, be it a good or a service; and property (at least on standard libertarian understandings of that concept) involves the right of use and disposal. Thus, whatever one has a right to sell one has a right to withhold. (It is on this ground that libertarians object to the concept of “forced sellings,” as in eminent domain, anti-discrimination laws, etc.) If, then, the minimal state has a right to charge for (i.e., sell) its services, it has equally the right to withhold them.

But we have assumed that this state also prevents its citizens from taking the law into their own hands. Thus, if Beatrice robs Fred, Fred cannot seek redress outside the proper governmental (this is prohibited), but he may not be able to seek redress from the government either (since it may choose to withhold its services). If the minimal state has property in its services, it follows that Fred may justifiably be made entirely helpless, legally prohibited from seeking any form of protection whatsoever. But that is absurd. Once a minimal state claims a monopoly on the use of force, then, it can no longer claim property in its services, but must provide them free of charge to all its citizens.3

The withholding scenario becomes more than hypothetical in Machan’s plan for financing contract enforcement. Ayn Rand is still more explicit:

Suppose that the government were to protect – i.e., to recognize as legally valid and enforceable – only those contracts which had been insured by the payment, to the government, of a premium in the amount of a legally fixed percentage of the sums involved in the contractual transaction. Such an insurance would not be compulsory; there would be no legal penalty imposed on those who did not choose to take it – they would be free to make verbal agreements or to sign uninsured contracts, if they so wished. The only consequence would be that such agreements or contracts would not be legally enforceable; if they were broken, the injured party would not be able to seek redress in a court of law.4

Suppose Beatrice breaks her uninsured contract with Fred. Beatrice has violated Fred’s rights. (It cannot be the case that only the breaking of insured contracts is a violation of rights; that would make contractual obligation logically posterior to government, whereas on the (Lockean-minarchist) libertarian view the enforcement of contractual obligation is (part of) the justification of government’s existence, and so such obligation must be logically prior to government.) But although Fred’s rights have been violated, he is barred from seeking redress. In the Rand-Machan system, Fred now may not receive legal protection even if he pays for it;5 but he also may not employ extralegal protection.6 Perhaps Fred has been imprudent in failing to insure his contract with Beatrice; but that does not change the fact that his rights have objectively been violated, and that he is legally barred from seeking any redress, whether through legal or through extralegal channels, for such violation.

We’ve seen, then, that a government which holds a monopoly on the legitimate use of force in a given area may not justifiably charge for its protection services. If the only libertarian alternative to monopoly government were the anarcho-capitalist vision of competing protection agencies, the notion of fees-for-protection would have been disposed of, since the anarcho-capitalist system has been shown to be objectionably inegalitarian. But there is an intermediate possibility: as mentioned earlier, a non-Weberian minimal state might allow private protection agencies but insist on being the ultimate arbiter in any disputes among them. Might such a state legitimately charge for its services?

It could certainly charge for any services to which it permitted competition it would most likely7 be outcompeted by agencies with smaller overhead expenses). But in its essential functions as arbiter and as standardiser of laws, it would necessarily be a monopoly and so could not charge for these services. How, then, are these to be funded?

Since the libertarian state is to be non-coercive, it must depend for its revenue on voluntary contributions. Since these contributions cannot be made in exchange for governmental services, they must be made either a) in exchange for some other good or service, or b) unconditionally. Let’s look at the second way first.

The notion of government’s relying in charity for its revenue may initially seem unrealistic. Would you voluntarily send money to Washington? Few people would pay their taxes if doing so were optional. But let’s recall that a libertarian society would differ in a number of important ways from our present one.

First, government in a libertarian society would be much cheaper. All government activities that involve the initiation of coercion would be banned. Since these constitute the overwhelming majority of the government’s present activities, most of the government’s current expenditures would be eliminated. Thus, while paying the bills of the welfare-warfare state out of charitable contributions would be impracticable, doing so for the night-watchman state might not be.

Second, a government that depends on voluntary contributions rather than taxation will have more incentive to maintain economy in its expenditures. For example, if the Pentagon had to “hold a bake sale to buy a bomber,” it would not spend thousands of dollars for a screwdriver. Thus, a non-coercive government’s protection services are likely to be cheaper than even the same services provided by a government that knows it can always arbitrarily increase its revenue by forcibly extorting money from its citizens.

Third, if we [are to] accept the claims of libertarian economists, a libertarian society free from burdensome economic regulation would be much more productive and prosperous than our own. Thus, not only would there be less government to pay for, there would be more money with which to pay for it; the low cost of the minimal state might [be] paid out of the productive surplus with plenty to spare.

Fourth, a libertarian government would be a popular charity, even among its more statist citizens. Statists might want the government to do more, but they certainly wouldn’t want it to do less, and so would pay to maintain its existence (unless they thought they could overthrow it, but I’m supposing our hypothetical minimal state is a relatively stable Western-style democratic polity with broad popular support). The general populace would be more willing to contribute to the basic security functions of the state than they might be to fund imperialistic adventuring, socialistic redistribution, or paternalistic moral nosiness.8 In a libertarian society, the current widespread resentment and distrust of government would presumably vanish, and everyone would recognize its activities as beneficial and worth contributing to. (There might be trouble in collecting from anarchists, but even they would contribute if they feared collapse of the government would lead to a statist takeover rather than to anarchy.)9

As for the “public goods” argument that so many citizens will free-ride on the expected contributions of others that the government will have to close down, this ignores the fact that that for many the knowledge of contributing to a cause they value is a positive private good. (If this were not so, there would be no successful charities.) People often place a high value on feeling generous or doing a good deed, and so might contribute out of patriotism; “we can easily imagine some citizens contributing to the provision of the public good as a matter of principle.”10 Moreover, large foundations and corporations might contribute heavily to government revenue in order to enhance their public image, just as they now do to other charities and public projects.

Nor need charity be the minimal state’s only source of revenue. While the government may not charge for its protection services (or at least not for those whose private provision it prohibits), there is no reason that it should not be allowed to provide, and charge for, non-protection services. Libertarians have traditionally opposed governmental participation in private enterprise (e.g., the Postal Service); but it’s not clear that they need to do so, so long as such governmental enterprise is not funded by coercive taxation and does not prohibit competition.11 Indeed, many libertarians have advocated government lotteries as peaceful alternatives to taxation; and lotteries don’t seem to be more peculiarly governmental than other sorts of private enterprise.

Governmental business ventures do, however, seem to be at a disadvantage in the marketplace relative to their competitors, since the governmental widget factory has all the overhead costs of a private widget factory plus the expensive protection services (e.g., police, national defense) it was created to fund. But the disadvantage is not quite as great as it seems, since the governmental widget factory unlike its private competitor, is a non-profit organisation, and thus can divert more of its revenue into overhead. The difference between the cost of government and the private business’ profits represents the government’s disadvantage in the market. It may be sizeable, but there are several ways of lessening it.

First, the charity factor may kick in. It might be regarded as patriotic or public-spirited to pay somewhat higher prices for government goods, given that the extra cost helps defray government expenditures. (And it’s not as though this required complete altruism; the continued existence of government is presumably in the self-interest of the consumer.) Second, the most natural services for the government to provide might be ones whose overhead is part of its normal functioning; the minimal state might charge for the use of government research libraries, or rent commercial space on test flights of military spacecraft, or provide firefighting services through the emergency police network, thus gaining some advantage over its competitors in these fields.

So far I’ve been assuming that a libertarian government ’s only source of revenue will be voluntary contributions in one form or another. But need this be the case? Many libertarians regard criminals as a possible source of involuntary contributions to government. Since criminals are the ones who create the costs of government in the first place (without them, no protection services would be needed), it seems appropriate that they should be required to pay part of the bill.

Certainly it is legitimate to require that a convicted criminal at least partially compensate the government for the cost of apprehending, trying, and perhaps imprisoning him (assuming the government carried these procedures out in an economical manner).12 This cost to government is one imposed by the criminal just as much as the cost suffered by the victim is so. The government is also a “victim” of the criminal’s actions, because the criminal, by committing a crime, has unjustly imposed a cost (the cost of bringing him to justice) on the government.

It might be retorted that the government assumed this cost voluntarily; but that is true only in a way. Suppose a kidnapping victim, struggling to escape from his bonds, accidentally cuts himself. The victim bore this cost “voluntarily”; he could have stayed put. But suffering this cost, while not necessary absolutely, was necessary in order to pursue a course of action that the victim had a right to pursue. Thus, the criminal has imposed a conditional cost on the victim. The victim did not accept this conditional cost voluntarily; it has been imposed on him by his kidnapper. Similarly, the criminal, by committing a crime, has imposed on the government the conditional cost of whatever it takes to bring him to justice. Since this conditional cost was imposed unjustly (no one should have to pay to bring the criminal to justice; he should bring himself to justice), the criminal may be required to compensate the government for it.

But it’s sometimes argued that we may impose on the criminal still further monetary penalties, beyond mere compensation, in the forms of fines and the like. Isn’t being forced to help fund the government a fitting punishment for criminals?

On the contrary, retributive punishment – any punishment that goes beyond the rectification of injustice – is inconsistent with the spirit (at least) of libertarianism. Libertarians accept Spencer’s Law of Equal Freedom: “Every man has freedom to do all that he wills, provided he infringes not the equal freedom of any other man.”13 The only permissible limit to one person’s freedom is the freedom of another. Demanding that the criminal compensate his victim is permissible, since restitution takes from the criminal just what is needed (and no more) to rectify the loss of freedom he imposed on the victim. Likewise, imprisoning dangerous criminals permissible, since such criminals pose a direct threat to society, and this threat is canceled by incarceration. (So imprisonment is actually a form of compensation.) But if I limit your freedom by more than is required to cancel your excessive limitation of my freedom, I have violated the Law of Equal Freedom, and my action is unjust. Thus, compensation is the only permissible form of punishment; the state may not raise money by imposing fines that do not reflect reasonable compensation to some victim of the criminal’s actions. (Deterrence, the practice of punishing one criminal in order to prevent depredations by other criminals, is not acceptable. We may limit the criminal’s freedom in order to cancel his limitation of others’ freedom, but we may not limit his freedom simply in order to cancel some generalised limitation of others’ freedom.)14

Nonetheless, the government may be able to extract money from the criminal that does not correspond directly to the compensation of individuals. If there exist such things as“crimes against society,” then perhaps society as a whole may be regarded as a victim of the criminal’s actions, and the government may demand restitution in the name of society. So we must answer two questions: Are there crimes against society? And if so, who may collect restitution on society’s behalf?

J. Roger Lee argues that libertarianism may indeed countenance the notion of crimes against society:

The civil liberty of free association to initiate proposals of cooperation is something that we earn by consistently acing in certain ways, avoiding fraud and the initiation of force, by conforming to the rational expectation frameworks which are associated with the situations into which we put ourselves. It is right that this is so, because our having the civil liberty of free association demands of others that they assume that we will behave like that. When we breach this responsibility to earn this treatment from others, we have no claim on them to allow us to freely associate with them. ... [It is sometimes argued] that only the victims of crime have a right to redress. Since the populace at large is not victimized by crime, the reasoning goes, the government, acting for all the people, has no right to claim redress from the criminal. [But the] populace at large is, in fact, victimized by crime. ... a crime undercuts the rational expectation framework which each citizen finds to be of value and which governments are set up to preserve. In fact, each citizen is harmed by the undermining of the rational expectation framework. Citizens are harmed in their role as citizens. It is a civic-social harm.15

Is this harm one for which society may be compensated monetarily?

It is not. Lee’s argument works in the context of his discussion, but it establishes less than it might appear to. Lee is granting for the sake of argument that only victims (and not third parties) have the right to punish criminals.16 He argues that society in general is harmed by crimes because the “rational expectations framework” is undermined. This seems to be a fancy way of saying that criminals pose a threat to society. Lee claims that the appropriate response to such a threat is imprisonment. (It’s a bit more complicated than that, but I won’t go into the details of his theory.) He does not, however, appear to regard the threat as a violation of anone’s rights (only as a threat of such violation). In imprisoning the criminal, we simply fail to grant him what Lee calls the “strong [non-judicial] presumption of innocence,” something nobody has an automatic right to.17 Thus, the government has no right to extract compensation in the form of money rather than imprisonment, since imprisonment does not represent compensation in the first place.

I’ve argued above, though, that imprisonment is a kind of compensation. So my view, unlike Lee’s, seems to allow the possibility that the threat posed by criminals is a monetarily compensable crime against society. But that is not so. Imprisonment compensates not for actual crime but for threatened crime. (If the criminal actually makes a threat, that is an actual crime; but merely being likely to commit a crime is not itself a crime.) Prevention (via imprisonment) is an appropriate compensation, or cancelling-out, of a threatened crime; but there is no justification for treating threatened crimes as actual by demanding non-preventive compensation for them. (Nor would this be fair to the potential victims, for whom prevention an compensation are hardly interchangeable goods.) In short, then, the fact that a crime is a threat to society does not make it a crime against society, and government cannot legitimately draw revenue from its compensation.

But this does not mean that there are no monetarily compensable crimes against society. If Fred has an item of property, he has a right to trade that item to any person who consents to the transaction. Conversely, any person in the world has ipso facto a right to enter into such a transaction with Fred, so long as he consents. But now suppose I steal Fred’s property. I have violated Fred’s right to engage in commercial transactions over that property with others, should they have chosen to do so; but I have equally violated the right of everyone in the world to engage in commercial transactions with Fred over that property, should he have chosen to do so.18 Even if Fred had no intention of selling the property, and nobody wanted to buy it in any case, it remains true that everyone in the world had a right to the following conditional: “If Fred should wish to sell his property to me, I may buy it from him.”19 By stealing Fred&146;s property, I have made it the case that even if Fred wanted to sell it, nobody would be able to buy it. Thus, in addition to compensating Fred for his loss, I must compensate mankind for all those foregone [sic, for “forgone”; to forgo is to go without, whereas to forego is to go before] transactions.

Since these transactions were purely hypothetical, not anything anyone was counting on, the compensation owed to any one person will be fairly minimal;20 but it will be real nonetheless. For example, suppose I end up owing a millionth of a nickel (I choose this figure randomly) to each person on Earth for the foregoing [sic] of their hypothetical transactions. Assuming a total population of five billion, I end up owing to society at large the far from evanescent [sic, for “negligible,” I guess] sum of $250.

How can the size of the thief’s debt to society be determined? Well, his debt to any one person is obviously less than his debt to the victim, whom he has hared directly. But since there are so many indirect victims, this leaves open the possibility that the thief’s debt to society might far outweigh his debt to the victim. But this is counterintuitive. Clearly the person he stole from is the primary victim of his actions. If the debt to society exceeded the debt to the individual, that would mean that the harm to society was the worst aspect of the theft. But the harm to the individual is surely the worst aspect. So the thief’s total debt to society must be less than the compensation he owes to the victim. It would be reasonable, then, to legislate that society receive some fixed percentage of an amount equal to whatever amount is awarded to the victim.

It might be objected that such a law would entail that the maximum amount owed to any one individual will decrease as the total population of the world increases. Isn’t it implausible to suppose that the amount of compensation to which I am entitled should depend on how many other people there are? Not in this case. The more people there are in the world, the less likely the victim was to enter into a transaction with me specifically; so the value of my lost opportunity is, roughly, inversely proportional to the total population.

We’ve established that there are, indeed, monetarily compensable crimes against society. But I have yet to show that the government is entitled to collect such compensation in order to fund its own activities. Popular usage to the contrary notwithstanding, government and society, as Paine reminds us, are far from synonymous.

Let’s consider how the compensation owed to society might best be collected. It is obviously impracticable to make the appropriate awards to each creditor. We cannot handout a check to every living person on Earth; and even if we could, what could they do with a check for a millionth of a nickel? The only way to award the compensation is to donate it to some public good that benefits everyone in the world. Since governments must be neutral among the projects of private citizens, it cannot award the compensation or charity or the like. The obvious beneficiary is the government itself.

A libertarian government is beneficial to everyone in the world.21 It is obviously a benefit to its citizen, insofar as it protects them from force and fraud. But it is also a benefit to citizens of other nations, since it represents a possible place of refuge from (actual or potential) tyranny in their own countries. To use Bentham’s phrase, it is “an asylum to all, a prison to none.” Thus, for any crime which involve the coerced foregoing [sic] of transaction opportunities by society at large, a libertarian government may award to itself a fixed percentage of an amount equal to the compensation awarded to the victim.

It might be feared that under such a system the government would be motivated to award excessive compensation to victims in order to obtain its proportional share. But this problem could be avoided by stipulating that the victim receive full compensation before the government gets anything. (This should be stipulated in any case, since the victim’s claim to compensation for clear and direct harm is more pressing than society’s claim to compensation for indirect and somewhat recherché harm.)22 The greater the amount awarded to the victim, the longer the government must wait to collect its share (and the less likely it is to get it); and the smaller the amount awarded to the victim, the less the government gets. Thus, the twin pressures of greed and time-preference should prevent both excess and deficiency in the awarding of compensation.

Libertarians have traditionally (and rightly) been suspicious of the notion of “crimes against society.” The notion seems a collectivist one; it distracts attention from the actual victim, suggesting that the wrongness of the action lies not in the injury done to individuals but in the disobedience to authority. It thus seems unpleasantly reminiscent of the doctrine that the King is the primary injured party in any trial. “Society” appears to be a hypostatised abstraction, invoked as a cover for the interests of those in power.

But, as we’ve seen, the notion of crimes against society need not carry collectivist implications. In the compensation system outlined above, society is understood to be nothing over and above the sum total of individuals; and the criminal has violated their rights as individuals, not as members of a group. The government is allowed to identify its interests with society’s only so long as it remains strictly libertarian. Moreover, the direct victim is not shortchanged, but is acknowledged to be the primary injured party, to whom restitution must be made before state and society may make any claims on the criminal.

I’ve argued that some of the most popular libertarian proposals for the financing of governmental or quasi-governmental protection – anarcho-capitalism, fees for monopoly services, punitive fines for criminals – are morally unacceptable. However, as we’ve seen, there are several promising alternatives to such measures: charity, state-run businesses, mandatory compensation for the cost of law enforcement, and mandatory compensation for crimes against society. Given the comparatively low cost of maintaining the minimal state, I see no reason to suppose that these sources of revenue would not be sufficient to finance government in a tax-free world.


1 Such a minimal state need not hold a monopoly of force, so long as it is the final arbiter in all disputes involving force. Thus, it fails to be a state according to strict Weberian definition; but so do most actual states, insofar as they permit private bodyguards and security forces, self-defense and citizen’s arrests, etc. (I am grateful to David Schmidtz for pointing this out to me) So perhaps the Weberian definition should be revised so as to speak of a monopoly of the power to authorise the use of force. This language is used on p. 50 of Randy E. Barnett, “Pursuing Justice in a Free Society: Part One – Power vs. Liberty,” Criminal Justice Ethics (Summer/Fall 1985), pp. 50-72.

2 Tibor R. Machan, “Dissolving the Problem of Public Goods: Financing Government without Coercive Measures,” pp. 204-205; in Tibor R. Machan, ed., The Libertarian Reader (Totowa: Rowman & Littlefield, 1982), pp. 201-208.

3 Government employees, of course, still have property in their labour, which they sell to the government. But if they withhold their services, they effectively resign from the government, and may not continue to use governmental force to restrain rivals.

4 Ayn Rand, “Government Financing in a Free Society,” pp. 116-117; in Ayn Rand, The Virtue of Selfishness: A New Concept of Egoism (New York: New American Library, 1970), pp. 116-120.

5 The Rand-Machan system requires that the premium be paid at the time that the contract is signed. If a citizen were allowed to wait until the contract is violated and then pay the premium, nobody would pay ahead of time and much less revenue would be generated. The same principle applies in ordinary insurance today.

6 Rand accepts the classical Weberian characterisation of the state as a monopoly of force (Ayn Rand, “The Nature of Government,” in The Virtue of Selfishness, pp. 107-115); and Machan likewise endorses the prohibition of extralegal protection: “To prohibit the provision of [contract enforcement] apart from the legal system is tantamount to prohibiting vigilante groups, lynching, and similar paralegal [sic] processes which always involve third parties whose rights are seriously endangered without the full protection of due process of law.” (Machan, p. 206.)

7 But not necessarily, as we’ll see.

8 “Men would pay voluntarily for insurance protecting their contracts. But they would not pay voluntarily for insurance against the danger of aggression by Cambodia.” (Rand, “Government Financing in a Free Society,” p. 118.)

9 In any case, any government that cannot win enough voluntary support from its citizens to stay in business (assuming there is no economic crisis) is not satisfying the people and probably deserves to perish. Besides, there is, alas, little danger as yet that anarchism will be popular enough to cause a problem; it’s the statists we need to worry about. And if the anarchists do take over, well, we could do worse.

10 Machan, p. 203.

11 There is a possible danger of conflict of interest, but this could be avoided by constitutional safeguards to guarantee that the salaries of government employees in charge of regulating the private manufacture of widgets not be tied to the success of the government’s widget-producing company. (I mean “regulating” in the libertarian sense of outlawing fraud, pollution, and the like.) Thus, the power and the incentive to put competitors out of business would not coincide in the same officials.

12 Such a requirement would actually be more beneficial to criminals than is the present system. In order to collect its compensation (as well as whatever compensation is owed to the victim), the government would ordinarily have to allow criminals, including prisoners, to earn money; and in order to ensure that the prisoner has an incentive to work for that money, the government would have to refrain from taking all the prisoner’s earnings. (Forcing the criminal to work, besides guaranteeing inefficient output, is a form of slavery; and enslavement as a punishment for crime is a barbaric practice that has no place in a civilised society.) So prisoners would be allowed to provide for their future and to become productive members of society. (We might see fewer prison riots if that were so.) A libertarian society would also have smaller and fewer prisons to maintain, since a) fewer actions would be illegal, and b) of those, only dangerous cases would be imprisoned, mandatory compensation being a preferable solution in the majority of instances.

13 Herbert Spencer, Social Statics: The Conditions Essential to Human Happiness Specified, and the First of Them Developed (New York: Robert Schalkenbach Foundation, 1970), p. 95. Spencer’s Law bears a surprising resemblance to Rawls’ decidedly non-libertarian First Principle of Justice. (John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971), p. 60.) This is not the place, however, to examine Rawls’ and Spencer’s conflicting understandings of the concept of equal freedom.

14 Here lies the crucial difference between Spencer’s libertarian Law of Equal Freedom and J.S.Mill’s collectivist variant with which it is often confused: “the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. ... the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others.” (John Stuart Mill, On Liberty, p. 73; in H.B. Acton, ed., Utilitarianism, On Liberty, Representative Government (London: J.M. Dent & Sons Ltd, 1972).) Mill’s version, unlike Spencer’s, contains no stipulation that the harm which a citizen may be required to compensate must be harm traceable to him.

15 J. Roger Lee, “The Arrest and Punishment of Criminals: Justification and Limitations,” pp.89-90, 96; in Machan, The Libertarian Reader, pp. 86-97.

16 Such a view is, incidentally, mistaken. Suppose I steal five dollars from a pacifist, who denies my right to the money but refuses to use force to get it back. A policeman comes along, wrests the five dollars from me, and returns it to its owner. On what grounds can I protest? The five dollars is not mine – I have no right to it – so the policeman has not violated my rights in taking it from me. Has he violated the pacifist’s rights? The pacifist may protest that since the five dollars is his property, he has a right to determine what shall be done with it, and he did not grant the policeman the right to forcibly return it to him. However, the pacifist does not clearly have the right to demand that the money remain in my possession. We’re assuming that the pacifist hasn’t given the money to me; he still regards it as his property, and I hold that property against his will. Thus, my possession of his property is unjust; and the policeman is under no obligation to respect the pacifist’s wish that an unjust situation continue.

17 The strong presumption of innocence is “an assumption that, by and large, people are safe. We tolerate anybody to mingle freely with us in public places, to approach us with an aim to proposing deals .... we assume that everyone will observe the minimal conditions for worthwhile association in society.” (Lee, p92.) Since, Lee tells us, someone can (e.g., when accused of a crime) lose the strong presumption of innocence, and so be liable to imprisonment, while retaining the judicial presumption of innocence, losing the strong presumption of innocence cannot itself be a violation of anyone’s rights. So Lee’s argument is misleading; he seems to be arguing that crimes “harm” (violate the rights of) society and so may be punished by society, but his actual claim is that punishment does not require anyone’s rights to have been violated.

18 This is similar to the way in which restrictions on immigration violate the rights of native citizens who might want to employ aliens or invite them as houseguests.

19 Although Jerry Falwell does not value and would not exercise his right to publish pornography, a law that prohibited his doing so would nonetheless violate his rights.

20 Anyone who was actually intending to buy Fred’s property and so has suffered a loss as a result of its theft would deserve a higher level of compensation.

21 It may not be subjectively beneficial, since a sincere statist may prefer to give up his own freedom in order to live in a statist utopia. There is probably no public good that is subjectively beneficial to everyone on Earth. But a libertarian government is an objective public good, since it is in everyone’s objective interest to have his or her rights protected.

22 Likewise, the government must wait to be compensated for court costs, etc.; for until it has ensured the compensation of the victim, it hasn’t done its job and doesn’t deserve to be paid.

Retractationes [2010]:

Okay, so what’s wrong with my 1987 argument against anarchism?

Well, first, it’s no longer clear to me that any injustice is involved in the application of different but equally acceptable standards in the same territory. I mean, if the standards really are equally acceptable, why doesn’t that fact foreclose any further problem? Moreover, what’s so special about it being the same territory? Why isn’t it just as unjust if two adjacent territories choose different but equally acceptable standards? In other words, why isn’t my argument for government an argument for world government?

Second, even if the coexistence of different standards were an injustice, why suppose that a state would be needed to correct the injustice – or would be especially good at doing so? (The nonstate Law Merchant, after all, provided needed uniformity when the government courts proved unwilling or unable to do so.)

I have other problems with the piece. My quaint fantasy in footnote 11 about how to prevent conflicts of interests ignores the likelihood of logrolling; and my suggestion that the government make itself the proxy for criminals’ restitution to society is a public-choice nightmare. (The naïve suggestions I offer for fending off incentive problems ignore the fact that the interpreter of my proposed restrictions would be the government itself.) My attribution to the minimal state of “research libraries” and “military spacecraft” makes me wonder just what I meant by “minimal”; my example of copyright is ill-chosen (but I still believed in IP back then); and my use of the terms “punishment,” “compensation,” and “capitalism” now strikes me as muddled.

On the other hand, the piece does make a good case against certain standard minarchist positions; and at least I’d evidently abandoned Isabel Paterson’s confused liberty-requires-political-structure-and-anarchy-has-none argument that had previously been my main brief against anarchism. Moreover, judging from what I say in footnote 9, my anti-anarchism obviously wasn’t very fervent. (When I eventually did become an anarchist three years later, both my mother and my former girlfriend, on being told the news, each said “I thought you already were one.”)

Back to Juvenilia and Other Offenses