Archive of Articles



Join the

Archive of Free Life


A Plague on Both Your Statist Houses


Why Libertarian Restitution Beats State-Retribution and State-Leniency[1]

Charles Murray describes himself as a libertarian, most notably in his short book, What it Means to be a Libertarian[2].  He might more accurately have described himself as having libertarian tendencies. My reading of “Simple Justice” is that the views it espouses are far more traditionalist than libertarian. Neither traditionalist state-retribution nor modernist state-leniency is libertarian. Nor does either provide as just or efficient a response to crime as does libertarian restitution, including restitutive retribution. Here, I shall respond directly only to Murray’s views, rather than also deal with state-leniency.  This is because I accept Murray’s thesis, without endorsing his specific arguments for it, that state-leniency is disastrous as a response to crimes against persons and their justly acquired property.

It is shocking and disgusting to see states today give violators of persons and property the upper hand, while they commit their crimes, throughout the judicial procedure after apprehension and during their trials, and in their final sentencing upon being convicted. The offensiveness of this country’s criminal justice system is compounded by the gross inefficiency of state policing here. However, to agree with Murray about the injustice and inefficiency of the current way of dealing with crime is about as far as a libertarian can really go. In commenting on Murray’s paper, I shall outline a radical and genuinely progressive libertarian option. In so doing, I recognise, and make no apology for the fact, that I stand at the extreme end of the libertarian spectrum.

Who needs the state?

According to those who occupy my preferred end of the libertarian spectrum, states serve no useful purposes, including the maintenance of law and order, that could not be achieved more effectively and justly by private and purely voluntary agencies and associations, created and maintained out of the uncoerced actions of ordinary private individuals acting only from self-interest and the dictates of their consciences.[3] Throughout, I use ‘libertarian’ in this extreme sense, although there are also minimal-state libertarians. I shall begin by briefly outlining my own libertarian conception of crime and of the way in which it should be treated, with which even most radical libertarians may disagree, and without offering very much by way of clarification or criticism of it.[4]

The origins of law may be traced to anarchically evolved, and ever-evolving, enforceable rules of conduct specifying how people must behave to avoid aggressing against the persons or justly acquired property of others.[5] Acts that intentionally or recklessly aggress against the persons or property of others—and which thereby tend to add indignity and fear—constitute the only ‘crimes’ there are in the libertarian sense of that word. That there are no real victimless crimes—for instance, producing and selling state-banned medications or recreational drugs—is a key libertarian tenet Murray fails even to mention.

In the libertarian view, there is no necessary connection between law and crime, on the one hand, and what a state decides to command or forbid by way of conduct, on the other. A state may forbid conduct not at all criminal in the libertarian sense, and it may permit conduct that is criminal. Indeed, from the libertarian perspective, states themselves notoriously authorise and engage in forms of criminal conduct, most notably taxation (systematic extortion) and aggressive war (mass murder). Note that this evolved-law thesis asserts not merely that such state ‘law’ and activity is immoral, but that it is not really law or legal at all. A type of conduct no more becomes legal or illegal simply because the state says so, than it becomes moral or immoral because it does. Nor does anyone’s command become law simply because he has the power to enforce it.

Wherever a genuine crime has taken place, in the libertarian sense of the term, then there is some victim of it whose person or justly acquired property has been proactively imposed on in some way by another person and who, in consequence, enjoys against that perpetrator of the crime a just claim to full restitution for the disvalue sustained as a result of its perpetration—in principle, at least, for it might not always be practical or possible to extract it. Thus, libertarian criminal law and civil law overlap. All crimes require restitution, but not all legal restitution is owed because of a crime. Perpetrators of crime owe their victims as restitution more than whatever would fully compensate for whatever proximate damage or loss their victims suffer to their person or property, including any feelings of shock or fear they suffer as a result of these crimes. In addition, perpetrators of crime owe their victims restitution for the additional risk to which they put them that they might not be able to recover any restitution because their assailant manages to escape conviction. I call this latter variable the ‘risk-multiplier’.

The risk-multiplier

For example, if there is only a one in ten chance perpetrators of a given kind of crime are apprehended and convicted, then full restitution to victims of it involves their perpetrators having to cede to their victims something of equal value to the value of the proximate loss each suffers to person or property, multiplied by ten. This is precisely what victims require to receive from their assailants to take account of the risk that was imposed on them.[6] Only to require criminals to make restitution for whatever proximate damage they cause their victims would mean they are allowed to impose on their victims, without having to make any restitution for it, the often far greater disvalue they cause their victims by the risk that they might escape.[7]

Consider a different kind of case from the risk-multiplier, but one that clarifies the disvalue of imposed risks as such. Suppose someone imposes on your head a game of Russian roulette in which the gun does not fire, although there was a one-in-six chance it might have done. Surely you are owed not just for any fear, etc., to which you were subjected, which, in this instance, forms the proximate damage you were made to suffer, and which might have been relatively small had the episode occurred very suddenly or even without your knowing at the time. You are surely also owed an additional amount, probably much higher and possibly infinite, that it would be reasonable of you to demand from anyone who sought to impose that risk on you. Similarly, someone caught and convicted of a crime should not be let off having to provide his victim with restitution for the risk he had imposed on him that he might, ex ante, have got away with his crime.[8]

Insuring against crime

However, the full debts perpetrators of crime owe their victims in restitution for their crimes are debts the victims might have chosen, in anticipation they might become victims, to ‘sell’ on to insurance companies through taking out policies against any losses sustained by becoming a victim of such (or any) crimes. Victims might also be able literally to sell the restitution owed them after they have fallen victim to a crime. In either case, victims of crime would acquire against their insurers a claim for a sum that would fully compensate them for any proximate disvalue suffered, but which takes no, or only partial, account of the risk-multiplier. Whether or by how much it did would depend on the precise terms of their contract.

Those who take out such policies might be able to guarantee they receive compensation should they ever fall a victim of crime, even if its perpetrators are never detected. Meanwhile, their insurance companies will have acquired from their clients a claim, should they become victims of crime, against their assailants for recovery of the full debt they owe their clients. This full debt they owe includes what is generated by the risk-multiplier. The difference between what companies pay out to their clients in compensation for becoming victims of crime and what the companies thereby become owed by its perpetrators provides them with the inducement to take over these debts that criminals, in the first instance, owe their victims.

Competing private agencies are far more likely to be able to catch and prosecute genuine criminals without becoming corrupted in the process than are state institutions, which maintain monopolies in this domain by aggressive violence. Moreover, competing private insurance companies are more likely than states to ensure that victims of crime receive quick and adequate compensation. The large amounts owing as a result of the risk-multiplier might be thought to create the moral hazard of inviting fabrication of evidence, whether by individuals or institutions. However, it must be remembered the risk-multiplier also applies in cases of any large sums fraudulently claimed. In addition, most claims will be sold on to insurance companies which stand to lose all custom if found fraudulent in this way, if not simply wiped out immediately by having to pay any risk-multiplier debt.

Restitutive retribution

If mere financial compensation were the only form in which restitution could be demanded by victims, people who wished to commit crime would effectively be able to purchase a licence to do so. Should victims prefer, they should be able to obtain ‘restitutive retribution’.[9] This is exacted by criminals being made to suffer as much personal injury or pain as they caused their victims magnified by any risk-multiplier. If you twist my arm, as though it were your property to use as you wish and in doing so break it, you thereby cede me a reciprocal right to break yours, or else for me to have it broken by an agent acting for me. This might look like retribution pure and simple. Where is the literal restitution, or restoration, in my breaking your arm? However, suppose the restitution owing me in monetary terms is £100,000. Should I prefer to take some fraction of that sum in the form of some reciprocal treatment of you, then that is simply how I choose to spend that much of the restitution I am owed. Alternatively, I might prefer to take all my compensation in money and buy a car instead, but that would not ‘restore’ my arm to not being broken either. You cannot complain that I am proactively imposing on you or imposing to a greater degree than you had imposed on me. Proactive impositions and reactions in excess of the risk-multiplier are all that this libertarian theory disallows.

From a libertarian perspective, therefore, the key deficiency in Murray’s account is the false dichotomy it poses between, on the one hand, retribution in the sense of punishment and, on the other, leniency. The libertarian position, by contrast, is one that embodies restitution for crime, where this is understood as criminals having to repay their victims amounts equal in value to whatever overall losses they are caused—although, as we have seen, victims may choose to obtain restitution in a retributive way. I should add that Murray also fails to distinguish, and then reconcile, deontological and consequentialist arguments for retribution.

In none of the seven hypothetical scenarios that Murray offers to test the moral proclivities of his readers (pp. 7-8) does he include among the possible options an explicitly libertarian response. If, in all the relevant hypothetical scenarios save the last, restitution is put in place of punishment, a libertarian can happily answer ‘3’ to all the questions that Murray asks about them. The final hypothetical scenario concerns the legitimacy of forcibly injecting a criminal-suspect with a truth drug. In this case, to force a suspect not yet found guilty to take such a drug, without at least his having previously entered into some contractual obligation to submit to one upon suspicion, is itself a case of proactively imposing upon someone, and thus a crime in the libertarian sense. Moreover, when libertarian restitution is substituted for punishment, it becomes difficult to see why opting for ‘3’ in any of the other cases qualifies as being “tough”, the adjective Murray uses to describe the attitudes of those likely to choose that option. What is so tough about thinking victims of crime qualify for receiving from their assailants full restitution and ought to receive it, if often only indirectly via insurance companies?

Is the state a community?

Murray claims “[t]he primal function of a system of justice is to depersonalise revenge.... [T]he individual will take his complaint to the community. In return, the community will exact the appropriate retribution; partly on behalf of the wronged individual, but also to express the community’s moral values” (pp. 18-19).  What is said here seems wrong on many levels. Justice does not have an ‘essence’ or “primary function” that simply needs to be cited to succeed thereby in refuting all competing conceptions of justice. If retribution is superior to restitution, Murray needs to argue for that thesis. Individual victims of crime may need the support of others, but why should they be entitled to receive it from “the community”? Murray appears to use this term as a euphemism for the state. However, whereas the state is an organisation (and, in the eyes of libertarians, a criminal one), a community is not. Nor is a community a moral agent, so it has no “moral values.” Only individuals have these. Why cannot private agencies be able to assist wronged persons better than states, as has been argued by many libertarian theorists, not least by Bruce Benson?[10]

On behalf of his position, Murray cites the Kantian thought-experiment that asks whether a murderer should be executed if his execution served no purpose other than ‘pure justice’ (p. 19). Kant and Murray say he should be. Libertarians say the correct answer is to be found in the victim’s legal defence contract or in his will, or in his known or likely opinion or in the decision of his heirs or other relevantly assigned persons, although I doubt many would want to let the murderer off. It is not up to “the community”—that is, the state—to decide.

Murray is similarly wrong when later on he explicitly states that victims “do not have the moral right to abrogate the community’s obligation to punish wrong behaviour” (p. 20). In the event a victim of some crime genuinely wishes to receive no restitution from his assailant, then, assuming there has been no intimidation of the victim by the criminal etc., that should be his or her choice, however foolish most other people might find it. In a sense, the victim retroactively consents to undergoing whatever the criminal has inflicted on him. In these circumstances, whoever exacts “retribution” on behalf of that victim or “the community” initiates a crime against the aggressor who has been forgiven by his victim. Such injustices are the sorts of thing that typically occur when statists attempt to take the law into their own hands in the name of “the community” or society. From a libertarian point of view, however, there is nothing unjust in people choosing to ban, boycott or berate anyone for any reason at all, provided in so acting they proceed in accordance with private property rules. Hence, provided they conform with these rules, people may take such action against anyone whose behaviour they regard as despicable, although whoever it is might have escaped and be able to escape successful prosecution for acting as he has done. For instance, many might for such a reason choose to ban someone from their private property and policing companies might even refuse to protect such a person.

Murray explains the “core tenets” of retributive justice as follows:

The necessary and sufficient justification for punishing criminals is that they did something for which they deserve punishment. 'Something' refers to the behaviours that society has defined as offences. 'Deserve' means that the offenders are culpable—morally responsible. Society not only has the right but the duty to punish culpable offenders. (p. 20. Emphases in original.)

Again, from a libertarian perspective, what Murray claims here is open to all sorts of question. What right has “society” to define what does and does not count as an offence, when all that is here meant by “society” is some state run according to the rules of elected oligarchs? It is, objectively, an offence, as the opposite of a defence, for anyone knowingly to impose proactively on the person or justly acquired property of someone else. If people merely defend themselves or their property against such impositions, they are not guilty of any offences against anyone. The state itself commits crimes when it attempts to impose on people things that conflict with protecting persons and their property. How and why should anyone be “culpable” if they seek to evade such arbitrary impositions? The state is not “society”, a term which denotes the free and spontaneous association of people. Nor has a state the right to punish anyone, even if a victim wants it to do so. For the opportunity-cost of its so doing is to exclude the possibility of the superior market system that would operate without the state’s extortion of resources through taxation and inflation of the money supply, the two principal sources of the state’s revenue.

Who needs judges?

Why should it be supposed, as Murray appears to, that, in all criminal cases, there is need of “jurors” or “judges”?  Murray only supposes this because he is thinking entirely within the traditionalist statist framework of law and order. It is hard in advance to know what different methods of securing and administering criminal justice would evolve were only the market allowed to operate here. On-the-spot payments for relatively minor crimes, as even the present British government has recently suggested for shoplifting although not as restitution, need not be either inefficient or an easy option, especially given the risk-multiplier element.

In defending the admissibility in court of the past criminal record of an accused on trial, something with which I cannot disagree, Murray interestingly suggests that “[d]ivinely accurate retributive justice would not punish for the one burglary out of dozens when the burglar got caught, but for the aggregate harm that the burglar has done” (p. 25). This is effectively what criminals are being asked to provide as restitution when what they are computed as owing takes into account the risk-multiplier. Its extraction would feel to the criminal like he was being punished for all the times he was not caught as well. Moreover, what the criminal will be deemed to owe for his crime will, through using the risk-multiplier, often be a lot more severe than state punishment currently is. Only the exaction of this form of restitution maximises the chances that “crime does not pay” which is Murray’s expressed desire in his final end-note. In principle, it will, typically, not be worth committing any crime because its potential benefits will be at least negated by its potential losses, and any other efforts and expenses will make it even less attractive. We should certainly see the crime level drop back again, and to far lower levels than obtained even in the 1950s to which Murray likes to hark back. We would only see the risk-multiplier fall if proportionally more criminals were brought to book.

Like many traditionalists, Murray is keen on prison. He writes, “[i]n modern England, the only authentic punishment for modern felonies is imprisonment” (p. 25). Prison is indeed a serious punishment. But it is both Draconian and unnecessarily expensive for the most part, while being too lenient in extreme cases. Unless someone poses so great a risk to others that he is likely to do more damage than he could ever pay in restitution, or else he refuses to pay restitution (non-contractual bankruptcy cannot be an option), there is no need for his incarceration. Such extreme cases are relatively few and far between and will be all the more so once criminals see that full risk-multiplier restitution will be enforced. In any case, since, in a libertarian world, prisoners will be obliged to pay their way in prison, being obliged to work there if they want to be fed, there is no need to worry about the expense of maintaining them whilst incarcerated in private prisons. However, for many lesser criminals, mere electronic tagging would at most be necessary or else some other, more inventive, option that only competition is likely to evolve efficiently. And these would provide more cost-effective and more humane alternatives to prison.

In the small minority of cases in which huge debts are owing that are unlikely to be paid by ordinary work, then extreme measures must be taken to recover them. These would still not be punishment but remain the enforcement of restitution. What I am proposing might sound harsh. But the only alternative is to allow the guilty to get away with their crimes against the innocent, which is surely harsher and completely unjust. Sometimes, a crime will be too great for full restitution to be possible, either in terms of property damage (malicious computer viruses often cause this) or personal damage, even a single murder, let alone bombing innocent civilians for political reasons. In these cases, we shall at least have done the best we can.

Overall, Murray’s traditionalist-retribution might be less bad than is the existing modernist-leniency that offers even less by way of just and efficient deterrence. However, in arguing for it, Murray entirely overlooks a third option more just, progressive and efficient than either. This is the way of dealing with crime through enforcing libertarian-restitution as the appropriate response to it. Murray must have read enough libertarian literature to be aware of this third option. It is a pity he chose not to consider it in his essay.[11]


[1] Being one of the commentaries from Simple Justice by Charles Murray with commentaries, Civitas 2005.  

[2] New York: Broadway Books, 1997.  

[3] The two rightly celebrated introductions to this ideology are David D. Friedman, The Machinery of Freedom: Guide to Radical Capitalism (1973; 2nd ed., La Salle, Ill.: Open Court, 1989) and Murray N. Rothbard, For a New Liberty: The Libertarian Manifesto (1973; rev. ed., New York: Macmillan Co., 1978).  

[4] For a more detailed and philosophical exposition of the position taken here see J. C. Lester, ‘Libertarian Rectification: Restitution, Retribution, and the Risk-Multiplier’, Journal of Value Inquiry 34, no. 2-3 (2000): 287-297, or J. C. Lester, Escape From Leviathan: Liberty, Welfare and Anarchy Reconciled (Basingstoke: Macmillan; New York: St Martin's Press, 2000), 108-120.  

[5] F. A. Hayek has famously distinguished spontaneously evolved law from state legislation in his Law, Legislation and Liberty (London: Routledge & Kegan Paul, [1973-79], 1982). But as a classical liberal, rather than a libertarian, he thinks that state legislation can be a useful supplement. See also Bruno Leoni’s Freedom and the Law (1961; expanded 3rd ed., Indianapolis: Liberty Fund, 1991). 

[6] All victims of this type of crime have a claim of this sort, but if more than one in ten aggressors starts being caught then the risk-multiplier eventually comes down in proportion.  

[7] I suspect there are difficulties with my current formulation of the risk-multiplier but my intuition is that some consistent version of it is possible and correct.  

[8] During the course of a crime, the risk-multiplier restitution that would be owed if the aggressor escapes means the victim can retaliate up to that value. It will be very approximate at the time, of course, but it means that the victim has clear leeway to be more violent than the criminal (even, in restitutive retribution, as the criminal is fleeing: this might be seen as a more just version of the “outlaw” view that Murray defends).  

[9] Here I agree with Bruce L. Benson, and disagree with some other libertarians, that there is room in libertarian restitution for retribution (i.e., that restitution may be taken in the form of retribution). See Bruce Lowell Benson, ‘Restitution in Theory and Practice’, Journal of Libertarian Studies 12, no. 1 (1996): 75–97.  

[10] For instance, see Bruce L. Benson, The Enterprise of Law: Justice without the State (San Francisco: Pacific Research Institute for Public Policy, 1990).  

[11] This essay is far clearer than it otherwise would have been thanks to critical responses from Mark Brady, David Conway, David Goldstone and David McDonagh.



UP HOME  © Libertarian Alliance  2005

web la-articles


Top 50 books of all time : by Old Hickory:-
  "I have limited the selection to the books I have read. I keep to the norm of not recommending to others books I have yet to read. Clearly, books I have not read by now suggests a judgement of some sort."

Read Online

Further reading: