Anthony de Jasay: The Private Enforcement of Public Rules, June 2006


A man travelling at night in the New York subway is menacingly surrounded by five young men brandishing long screwdrivers. He shoots and wounds all five, one grievously. His plea of self-defence is vitiated by his disproportionate use of force. Convicted and sentenced to a long prison term, he fights a protracted legal battle financed by donations, and is finally acquitted on appeal.

In a French school, a 12-year old boy misbehaves, disrupts the class and ends up kicking everyone in sight, including the teacher. The assistant head takes him to his office and spanks him. He is dismissed on the spot by the head of the school, and is prosecuted under the act forbidding corporal punishment. He is acquitted because he did not humiliate the boy (as he would have done if he had spanked his bottom in front of the class).
After the July 2005 terrorist attacks on London Transport, police shoot dead an innocent foreigner, mistaking him for a terrorist. In the ensuing storm of indignation, the police are severely blamed for not ascertaining the intentions of their victim. The article of The Economist discussing the matter is entitled ‘Excuse Me, But Are You A Suicide Bomber?’


2.1 Rule-Enforcement as a Public Good

It is a well-understood problem of public goods that because it is very difficult or morally repugnant to exclude any member of the public from enjoying such goods if they are provided at all, each individual member will choose to enjoy them without contributing to their cost. In other words, he will choose to be a free rider. Obviously, if everyone tries to be a free rider, no contribution is made to the cost and the public good cannot be provided at all. Logically, therefore, everybody cannot be a free rider. Some – at least a critical minimum number – must contribute for the public good to be produced so that all, contributors and free riders alike, can enjoy it.

In a world where each is motivated by the good of all, presumably all would contribute. The public goods problem would not arise. In a world where each pursued mostly his own interest, two solutions would be open.

One is for an authority with sufficient backup force – the state – to compel some or all individuals to contribute by taxing them on some basis. The angelic version of democratic theory supposes that all individuals positively wish to be so compelled, because compulsion ensures that all will contribute equitably. Adult versions of democratic theory tell us that a majority of people will accept compulsion provided the major part of taxation falls on the minority, so that the majority gets at least a little bit of a free ride.

The other solution is the state-of-nature one, where the state does not intervene (or there is no state) and any contribution to producing a public good is voluntary. Received wisdom from mainstream economics holds that voluntary contributions would be irrational, hence the public good either could not be produced at all, or only to a sub-optimal extent. Optimality criteria are contestable, but this is not the place to explore that particular controversy. For the present purpose, what we need is to relax the received wisdom about voluntary contribution being necessarily inadequate or nil. It is the wrong kind of a priori reasoning to deny the possibility that some individuals attach a high enough ‘utility’ to a public good to contribute to its cost rather than allow it to fail altogether, given the requisite probability that enough other individuals will contribute, too for the same reason. The secret of this voluntary solution is the reliance of each contributor on a reciprocal probability that each accords to others behaving the right way – a probability that serves as the foundation of a coordination equilibrium or convention.

Instead of any public good in general, consider next the particular public good of the enforcement of beneficial rules, notably rules against torts and rules of civility. It is true of many such rules that if they are upheld, a temptation is provided to breach them or at least not to contribute to their enforcement, but to free-ride instead. Hence the need for enforcement by punishment of some kind or the credible threat thereof. We can consider that the rule is self-enforcing, i.e. an equilibrium, if the conditions for generating voluntary contributions to the cost and effort of enforcement are fulfilled.

One such condition is that the contributor should derive a benefit from punishing or deterring the breach of a particular rule that would cause him direct and certain (rather than uncertain future) damage. In such cases the rule plays the role of a private good of his own. Deterring harm to one’s life, limb or property by building a reputation for swift revenge, frightening a bad debtor into repaying his debt, or teaching a lesson to a bad neighbour who is making a nuisance of himself, are cases in point where bearing the enforcement cost brings a direct personal benefit. This alone may serve as a sufficient condition for binging it forth.

Separately and additionally, the potential contributor may support enforcement for the same reason that he might support any public good in general, namely that the uncertain but probable stream of future benefits from upholding a good rule in coordination with others outweighs its cost.

Formally, an individual facing a state-of-nature public goods problem has two options, to contribute and not to contribute to its cost. Contribution can have two outcomes, successful production of the public good in conjunction with others, or failure to produce it. Non-contribution may permit a free ride if the public good is produced by others, or failure to produce it. The fully rational choice depends on the probabilities and ‘utilities’ of the four alternative outcomes. The probabilities, in turn, depend on the expected behaviour of others.

I am far from suggesting, however, that in practice such subjective probability estimates are in fact made and calculations are really based on them. Actual behaviour is more likely to be rough-and-ready calculation and adjustment to what others close by are doing. It is clear enough, though, that voluntary contribution to rule-enforcement is a logically coherent alternative that can be fully consistent with rationality. That it can under the right circumstances be a realistic outcome is attested by the fact that as far as we can read history, rules enforced by spontaneous civic actions have largely preceded specialised rule-enforcers such as the state.

2.2. Monopoly of Enforcement: the State Favours Free Riding

Looking at the sufficient conditions for rule-enforcement by voluntary effort in the state of nature furnishes, as a matter of course, some insight into the consequences of state intervention.

A widely used definition of the state attributes to it the monopoly of the legitimate use of force. This definition is vacuous, for under it force is legitimate if and only if the state possesses it, hence the condition that it has a monopoly of it is satisfied by definition. (It is like defining Nestlé as the monopoly producer of Nestlé chocolate). The state is the state by virtue of its monopoly (Nestlé is Nestlé by virtue of being the producer of Nestlé chocolate). A less vacuous definition would have the state (1) possessing enough force to deter a coalition of equal or greater force being formed in civil society, and (2) using its force to dispossess actors in civil society of such force as they may still hold and punishing them for using it.

If the state bears the cost of public goods in general, and of rule-enforcement in particular, by levying taxes it relieves individuals of the need to make choices in the matter. Since they must pay taxes anyway, and public goods are provided anyway, all is left to them is gratefully to accept the comfort of free-riding, the sole option the state can leave open if it wishes to affirm its monopoly of rule-enforcement.

By dispossessing its subjects of the means of threatening or using force (except such tamed means as firearms licensed by the police), and by punishing unlicensed private enforcement (except under carefully defined restrictive conditions) the state ostensibly relieves its subjects of a burden. It also assumes a responsibility which it is intrinsically ill-suited to discharge.

2.3 Why Seek a Monopoly?

Manifold reasons move the state to establish and defend its monopoly. The most evident ones serve its own survival as an institution, the consolidation of its discretionary power over and above the power it must exert merely to ‘stay in power’, and the growth of its authority over and above that conferred by its possession of a great concentration of material force.

When rule enforcement was a diffuse, decentralised function, non-corporal punishment took the form of fines for the benefit of victims and plaintiffs, and judgments were either free or remunerated by fees. Kings saw obvious advantage in diverting this income stream to themselves. Stripping civil society of rule-enforcing functions also stripped it of much of the justification for possessing arms and organised forms of exerting force. This has permitted a substantial reduction in the force the state needed to protect itself within civil society, or alternatively a great expansion of the area over which it could exert its will by relying on a given endowment of force. Last but not least, by effectively playing the role of sole teller and interpreter of the law, sole judge above the parties, and sole enforcer, the state gained a moral authority that had previously existed only dispersed in society among the elites.

Though they may be more open to argument, other kinds of reasons for the state’s monopoly may weigh more heavily than the ‘opportunistic’ ones noted above. The major one can be traced back to the ingrained fairness principle that like cases must be treated alike, and they will probably not be treated alike if a party may be the judge in his own cause and execute his own judgment. The judiciary and the police, as parts of the state and in some, albeit a little obscure, sense standing above civil society have a better chance of being impartial in judgment and even-handed in execution.

There is much to be said about the rights and wrongs of this argument, but one cannot deny that it is generally held with great conviction and is inculcated in people from an early age by the education system which identifies respect for the law with respect for the state as the sole legitimate law-enforcer. I am not concerned in this paper with the validity of the ‘party and judge’ argument beyond noting the weight it carries, but rather with the ultimate effect I believe it has upon the very possibility of reasonably effective rule-enforcement. However, for those who feel no reservations about the authority of the state and the duty of political obedience, it appears truly anomalous that private individuals should act as their own policemen and their own judges. Not only is it in the institutional interest of the state and its police and judiciary agencies to foster this belief; it is certainly also the deeply held conviction of what one may loosely call the political class.

Taken together, these reasons seem amply sufficient, and have proved amply sufficient over the last two or three centuries of Western history, to induce the state to assert its monopoly of rule-enforcement. It has sought to restrict the capacity of civil society to assume such tasks, discouraged its last vestiges by punishing private initiatives, and has done much to cause the rule-enforcing ability of civil society to atrophy by disuse. There is now by and large a clear-cut division of labour. Rule-enforcement is the exclusive responsibility and privilege of the specialized agencies of the state. With but a mild exaggeration, we might say that private persons and informal groups are breaking the rules when they try to uphold them.


3.1. Restraining the Monopolist

In medieval Europe there was some local rule-enforcement by elements of civil society at the grassroots level, but there were also specialised agencies fulfilling both police and judicial functions in competition with each other. Disputes could be brought to, and remedy solicited from feudal lords, towns, ecclesiastical and royal authorities. By and large, royal agencies drove out the competition and the state’s monopoly has become increasingly effective in Europe from the 17. century onwards It was to be expected that establishment of the monopoly should soon be followed by demands for restraining and regulating it. This has duly happened. It was inevitable for states depending to a greater or lesser extent on the consent of their subjects to meet these popular demands, the more so as regulation of the monopoly has pari passu enhanced its moral authority and confidence in its moderation and impartiality.

Restraint of the monopoly, becoming more and more strict and elaborate under pressure from the rising clamour of ‘rightsism’ has many minor and two major consequences. One of the latter impinges on the police, the other on the judiciary.

Before it was restrained, the duty and the prerogative of the police included the maintenance of public order and respect for the rule system by using such force as proved necessary. Theoretically, this could authorise it to shoot rioters with live ammunition or torture suspected criminals and terrorists to extort information about accomplices, though such recourse to force has never been frequent in the western world. A probably more important police prerogative was to settle disputes on the spot and administer summary justice and immediate punishment for commonplace breaches of the rules. Obviously, this carried a risk of police error, abuse, brutality and arbitrariness. It is fair to add that in less civilised countries where police power is circumscribed in constitutional rhetoric but in practice is unrestrained, the conduct of the police often verges on the intolerable.

Opposition to these practices became near-unanimous with the rise of ‘rightsism’. It was never properly understood that the risk involved in police powers great or small cannot be abolished by restraining them, but can only be transformed and transferred, often from the innocent victims of police abuse to the no less innocent victims of rule-breakers let loose upon them by the lack of police powers. Which set of innocents should be sacrificed to the other is a painful matter of arbitrary judgment coloured by one’s taste for one kind of society rather than another. Be that as it may, the end result of restrictively defining police duty and prerogative was that the chief function of the police became, not to apprehend and punish misbehaviour, but to deliver presumed rule-breakers to the judiciary together with the evidence supporting their indictment.

In sum, one is inclined to conclude that monopoly of the police function leads to a dilemma: it is far too dangerous to let it have a free hand, yet it loses a large part of its usefulness when its hand is tied behind its back.

The other major consequence of restraining the monopoly of the state affects the judiciary, which has adjusted itself, and lent a helping hand, to the progressive deformation of due process.

It is a morally quite appealing principle that all things being equal, it is a greater wrong to condemn the innocent than to let the guilty go unpunished. Under the impulse of ‘rightsism’ the first half of this principle came to overshadow the second; it became all-important, and worth paying almost any cost, not to condemn the innocent, and never mind if as a result more of the guilty go unpunished than would otherwise be the case This has led to an accelerating development of the prerogatives of the defence and a corresponding restriction of the powers of the prosecution in criminal cases. It has, in particular, led to a luxuriant growth of appealed judgments, in part because a multiplication of the admissible grounds for appeal and in part because the cost of the appeals process, no matter how frivolous the grounds, came to be mostly a charge on the public purse. Appeal became a ‘heads I win, tails they lose’ game, a meal ticket for lawyers and a major reason for the slowness of justice.

In civil justice, the principle of ‘better spare the innocent than punish the guilty’ exerted itself in a different but no less portentous fashion. Broadly speaking, courts came gradually to give less weight to the arguments for the plaintiff and more for the defendant in breach-of-contract cases, allowing various extenuating circumstances to void obligations – a tendency that has greatly contributed to the decline of the freedom of contract. By the same token, it weakened confidence in the binding force of contracts. At the same time, judgments in disputes between corporations and individuals, notably in tort cases, came to tilt in favour of individual plaintiffs looking as if they were inspired by a wish to correct for the disparity in the parties’ size and wealth. If justice was to be a state monopoly, at least let no one say that it did not protect the weak against the strong – as if doing this, instead of protecting the just against the unjust, were the task of justice.

3.2. Punishment Must Not Hurt

Punishment administered by the state must fulfil two parallel functions. First, it must demonstrate to society at large that justice is being done. Second, it must deter breaches of the rules. Beyond a basic level where they can be complementary, the two functions soon become rivals; promoting one will involve demoting the other.
It is obvious that the deterrent effect of punishment becomes stronger as it grows swifter, more brutal and merciless. Opponents of capital punishment obsessively argue that it has no discernible deterrent effect on capital crime. Since the problem does not lend itself to controlled experiments where potential murderers are exposed both to a death or a life sentence and react by committing or not committing the contemplated murder, I think it is a matter of common sense to suppose that death deters more than life, even if it is life in prison. However, the real point is that a modern-day death sentence becomes executory only after an appeals process that is notoriously long and usually extends over two decades. There are perhaps strong reasons why this is so, but it is liable to wipe out the deterrent effect of death and assimilates the death sentence to a long prison sentence.

The swiftness of punishment is arguably the most important factor in deterring all breaches of the rules, be they minor misdemeanours or grave felonies. The most likely-looking hypothesis explaining it is that the various types of rule-breakers all tend to discount the future at a higher rate than rule-abiding people, and/or they systematically under-estimate the probability of getting caught and being convicted. If this hypothesis is near the truth, it underlines the efficacy of summary punishment on the spot, and the softening effect of a delayed judicial process.

Summary justice, however, almost necessarily involves corporal punishment. Maximising deterrence, as I suggested at the head of this section, calls for punishment that is not only swift – a rap on the knuckles – but also brutal. It must hurt physically and psychically. Spanking the disruptive pupil may teach him a lesson, but spanking him in front of the whole class will teach something to the class as well. It will also belittle him in his peers’ eyes. Vandals, leaders of youth gangs and neighbourhood bullies may suffer loss of prestige if put in the pillory.

Finally, punishment will lose much of its deterrent effect if it is alleviated on grounds of divided responsibility. The classic case is for the offender to be treated more lightly because responsibility for his offence is shared by the society that ill-treated him, failed to educate him and has offered him no decent future. Obviously, however, if punishment serves not so much to deter, but to show that justice is being done and the offender is made to ‘pay his debt to society’ it need not be swift, it must not be brutal, and it should be merciful or at least humane. Summary justice must be replaced by due process, Corporal punishment must be abolished altogether as a violation of human rights, schoolchildren must neither be spanked nor otherwise humiliated and rioters dispersed without doing them bodily harm. Prisons must be salubrious, prisoners must have exercise, entertainment and education, and not be made to do forced labour.
The same social forces that seek to impose restraint upon the state for fear of its monopoly of force, also work to make punishment increasingly symbolic, painless and consistent with the humane treatment that befits a civilised society. The price paid for this laudable development is the progressive blunting of the edge of deterrence. If rules are respected without effective enforcement involving punishment that hurts, well and good. Taking a lucky path may lead a society to this happy condition. The society we now live in has clearly not taken this path.


4.1. Restore Competition?

Section 3.2 was intended to persuade the reader that state monopoly of rule-enforcement leads to soft, sluggish and ineffective punishment. As a consequence, rules will be poorly enforced and public order and the security of person and property undermined.

It might be thought that a possible remedy lies in the formal abolition of the state monopoly and the legalisation of do-it-yourself policing, judging and punishing by individuals and ad hoc groupings at the grassroots level. There would then be competition between private and public rule-enforcement. The pressure of opinion for taming and restraining the state monopoly would presumably ease off. A spontaneous division of labour might evolve, private enforcement by vigilantes taking on the vandals and the street gangs that spoil the daily life of ordinary residents, while more complex tasks, such as crime detection, would be assumed by the state. There might even be statutory dividing lines, capital punishment and imprisonment remaining a state prerogative, while private enforcers might be given freedom to deal with petty offenders by using strong-arm methods, perhaps including the imposition of forced labour in local infrastructure projects. Popular imagination and initiative could probably relied upon to devise easily applied varieties of punishment.

The principal, and very likely decisive, objection against this solution would be the indignation it would provoke in a large section of opinion, and in the politically most articulate part at that. Formally to authorise private individuals and groups to employ violence at their own discretion would seem to be unthinkable: after all, one could not pass laws legalising lawlessness. At all events, it is hard to see the state legislating for its own withering away.

A little less unrealistic solution, given a measure of good luck, may emerge spontaneously. The mechanism cold be driven by the progressive worsening of the state’s performance in the maintenance of public order and personal security. In recent decades, several Western European states have tried to reverse this rather threatening evolution by pouring money into the police and judiciary services. The effort may have slowed down the worsening of the security situation and the disruption in parts of the school system, but has not reversed it.

It is clearly in the interest of the governors not to allow this deterioration to go on indefinitely, for it brings ever nearer the point at which the patience of the governed snaps and political obedience becomes precarious. One way for the state to ward off this danger might be, not to abolish the monopoly of enforcement, but to turn a blind eye when exasperated individuals and their ad hoc coalitions resort to self-help and organise themselves for the private enforcement of basic rules. Once such a movement takes hold – and it may take fortuitous events to help it to take hold – it would have some chance of being carried wider and farther by its own momentum. There would be no need for a formal division of areas of responsibility between the state’s services and private endeavours, nor for any explicit acknowledgment that the state is turning a blind eye. Success would be more likely if the rise of private enforcement happened by tacit occupation of the ground by private initiatives and by its tacit acceptance by the former monopolist.

It must, of course, be borne in mind that private enforcement of public rules requires some private muscle, some private willingness to bear the inconvenience and risk of self-help and reciprocal aid, and no doubt also certain social skills in carrying out enforcement and imposing punishment. During the long years of an effective state monopoly, the muscles have atrophied and the social skills have been largely forgotten. The beginnings of evolution toward a public cum private enforcement system might therefore be shaky, but performance could be expected steadily to improve for the obverse of the very same reasons that brought about the worsening performance of the monopoly.

4.2 Fiat Mundus, Pereat Justitia

There could be charges that the present essay, particularly Section 4.1, is Utopian. Such charges are mostly well founded, but not completely so. It is certainly not Utopian readily to admit, as this essay does, that when teachers freely use the cane, angry householders break the bones of vandals and rowdies with baseball bats, thieves and defaulters are stripped of their ill-gotten gains and cheats are repaid in the same coin, there is ample scope for hasty judgments turning out to be unjust. Innocent people may in fact fall victim to rough-and-ready procedure. Such things would not happen in Utopia, but they very likely would in the world of private rule enforcement depicted here.

In this sense, the relevant charge is not that the mixed public-private system is Utopian, but that it is excessively, almost cynically realist. It is tantamount to inverting the allegiance paid to justice in fiat justitia, pereat mundus – let there be justice even if the world were to perish. Instead, its oath becomes fiat mundus, pereat justitia – let justice perish so the world may live on.

Unlike the charges of Utopian imagination that are partly or mostly valid, this charge of cynical pragmatism is false. It may be the case that under monopolistic enforcement, few innocents are wrongly condemned and punished, while under private enforcement many such would be unjustly sacrificed to the cause of more effective protection of the rules and of those the rules protect. From this, however, it does not follow that the monopolist safeguards justice, while the private competitor trades it off for a more orderly world. Both systems must sacrifice justice, and both will create innocent victims.

The victims of private enforcement get unjustly punished and may be presumed guilty though, had they been presumed innocent, they could have cleared themselves. Arguably, the monopolist makes few such mistakes. What it does do, though, is to create vast numbers of innocent victims among ordinary people. The degradation of schools, places of public entertainment, entire streets and neighbourhoods and the bullying and depredations of those whom weak rule-enforcement encourages to ride high, hurts mainly the helpless poor.

Which of the two alternative sets of innocent victims is more important and worth saving at the expense of the other is a question that has no morally defensible answer. Given, however, that the world is what it is and both sets cannot be saved, choosing to save one of the two is not a question of justice but one of making unavoidable choices among rival values.

June 2006