Society without a State
[Murray Rothbard delivered this talk 32 years ago today at the American Society for Political and Legal Philosophy (ASPLP), Washington, DC: December 28, 1974. It was first published in The Libertarian Forum, volume 7.1, January 1975, available in PDF.]
In attempting to outline how a "society without a state" — that is, an anarchist society — might function successfully, I would first like to defuse two common but mistaken criticisms of this approach. First, is the argument that in providing for such defense or protection services as courts, police, or even law itself, I am simply smuggling the state back into society in another form, and that therefore the system I am both analyzing and advocating is not "really" anarchism. This sort of criticism can only involve us in an endless and arid dispute over semantics. Let me say from the beginning that I define the state as that institution which possesses one or both (almost always both) of the following properties: (1) it acquires its income by the physical coercion known as "taxation"; and (2) it asserts and usually obtains a coerced monopoly of the provision of defense service (police and courts) over a given territorial area. An institution not possessing either of these properties is not and cannot be, in accordance with my definition, a state. On the other hand, I define anarchist society as one where there is no legal possibility for coercive aggression against the person or property of an individual. Anarchists oppose the state because it has its very being in such aggression, namely, the expropriation of private property through taxation, the coercive exclusion of other providers of defense service from its territory, and all of the other depredations and coercions that are built upon these twin foci of invasions of individual rights.
Nor is our definition of the state arbitrary, for these two characteristics have been possessed by what is generally acknowledged to be states throughout recorded history. The state, by its use of physical coercion, has arrogated to itself a compulsory monopoly of defense services over its territorial jurisdiction. But it is certainly conceptually possible for such services to be supplied by private, non-state institutions, and indeed such services have historically been supplied by other organizations than the state. To be opposed to the state is then not necessarily to be opposed to services that have often been linked with it; to be opposed to the state does not necessarily imply that we must be opposed to police protection, courts, arbitration, the minting of money, postal service, or roads and highways. Some anarchists have indeed been opposed to police and to all physical coercion in defense of person and property, but this is not inherent in and is fundamentally irrelevant to the anarchist position, which is precisely marked by opposition to all physical coercion invasive of, or aggressing against, person and property.
The crucial role of taxation may be seen in the fact that the state is the only institution or organization in society which regularly and systematically acquires its income through the use of physical coercion. All other individuals or organizations acquire their income voluntarily, either (1) through the voluntary sale of goods and services to consumers on the market, or (2) through voluntary gifts or donations by members or other donors. If I cease or refrain from purchasing Wheaties on the market, the Wheaties producers do not come after me with a gun or the threat of imprisonment to force me to purchase; if I fail to join the American Philosophical Association, the association may not force me to join or prevent me from giving up my membership. Only the state can do so; only the state can confiscate my property or put me in jail if I do not pay its tax tribute. Therefore, only the state regularly exists and has its very being by means of coercive depredations on private property.
Neither is it legitimate to challenge this sort of analysis by claiming that in some other sense, the purchase of Wheaties or membership in the APA is in some way "coercive." Anyone who is still unhappy with this use of the term "coercion" can simply eliminate the word from this discussion and substitute for it "physical violence or the threat thereof," with the only loss being in literary style rather than in the substance of the argument. What anarchism proposes to do, then, is to abolish the state, that is, to abolish the regularized institution of aggressive coercion.
It need hardly be added that the state habitually builds upon its coercive source of income by adding a host of other aggressions upon society, ranging from economic controls to the prohibition of pornography to the compelling of religious observance to the mass murder of civilians in organized warfare. In short, the state, in the words of Albert Jay Nock, "claims and exercises a monopoly of crime" over its territorial area.
The second criticism I would like to defuse before beginning the main body of the paper is the common charge that anarchists "assume that all people are good" and that without the state no crime would be committed. In short, that anarchism assumes that with the abolition of the state a New Anarchist Man will emerge, cooperative, humane, and benevolent, so that no problem of crime will then plague the society. I confess that I do not understand the basis for this charge. Whatever other schools of anarchism profess — and I do not believe that they are open to the charge — I certainly do not adopt this view. I assume with most observers that mankind is a mixture of good and evil, of cooperative and criminal tendencies. In my view, the anarchist society is one which maximizes the tendencies for the good and the cooperative, while it minimizes both the opportunity and the moral legitimacy of the evil and the criminal. If the anarchist view is correct and the state is indeed the great legalized and socially legitimated channel for all manner of antisocial crime — theft, oppression, mass murder — on a massive scale, then surely the abolition of such an engine of crime can do nothing but favor the good in man and discourage the bad.
A further point: in a profound sense, no social system, whether anarchist or statist, can work at all unless most people are "good" in the sense that they are not all hell-bent upon assaulting and robbing their neighbors. If everyone were so disposed, no amount of protection, whether state or private, could succeed in staving off chaos. Furthermore, the more that people are disposed to be peaceful and not aggress against their neighbors, the more successfully any social system will work, and the fewer resources will need to be devoted to police protection. The anarchist view holds that, given the "nature of man," given the degree of goodness or badness at any point in time, anarchism will maximize the opportunities for the good and minimize the channels for the bad. The rest depends on the values held by the individual members of society. The only further point that need be made is that by eliminating the living example and the social legitimacy of the massive legalized crime of the state, anarchism will to a large extent promote peaceful values in the minds of the public.
We cannot of course deal here with the numerous arguments in favor of anarchism or against the state, moral, political, and economic. Nor can we take up the various goods and services now provided by the state and show how private individuals and groups will be able to supply them far more efficiently on the free market. Here we can only deal with perhaps the most difficult area, the area where it is almost universally assumed that the state must exist and act, even if it is only a "necessary evil" instead of a positive good: the vital realm of defense or protection of person and property against aggression. Surely, it is universally asserted, the state is at least vitally necessary to provide police protection, the judicial resolution of disputes and enforcement of contracts, and the creation of the law itself that is to be enforced. My contention is that all of these admittedly necessary services of protection can be satisfactorily and efficiently supplied by private persons and institutions on the free market.
One important caveat before we begin the body of this paper: new proposals such as anarchism are almost always gauged against the implicit assumption that the present, or statist system works to perfection. Any lacunae or difficulties with the picture of the anarchist society are considered net liabilities, and enough to dismiss anarchism out of hand. It is, in short, implicitly assumed that the state is doing its self-assumed job of protecting person and property to perfection. We cannot here go into the reasons why the state is bound to suffer inherently from grave flaws and inefficiencies in such a task. All we need do now is to point to the black and unprecedented record of the state through history: no combination of private marauders can possibly begin to match the state's unremitting record of theft, confiscation, oppression, and mass murder. No collection of Mafia or private bank robbers can begin to compare with all the Hiroshimas, Dresdens, and Lidices and their analogues through the history of mankind.
This point can be made more philosophically: it is illegitimate to compare the merits of anarchism and statism by starting with the present system as the implicit given and then critically examining only the anarchist alternative. What we must do is to begin at the zero point and then critically examine both suggested alternatives. Suppose, for example, that we were all suddenly dropped down on the earth de novo and that we were all then confronted with the question of what societal arrangements to adopt. And suppose then that someone suggested: "We are all bound to suffer from those of us who wish to aggress against their fellow men. Let us then solve this problem of crime by handing all of our weapons to the Jones family, over there, by giving all of our ultimate power to settle disputes to that family. In that way, with their monopoly of coercion and of ultimate decision making, the Jones family will be able to protect each of us from each other." I submit that this proposal would get very short shrift, except perhaps from the Jones family themselves. And yet this is precisely the common argument for the existence of the state. When we start from the zero point, as in the case of the Jones family, the question of "who will guard the guardians?" becomes not simply an abiding lacuna in the theory of the state but an overwhelming barrier to its existence.
A final caveat: the anarchist is always at a disadvantage in attempting to forecast the shape of the future anarchist society. For it is impossible for observers to predict voluntary social arrangements, including the provision of goods and services, on the free market. Suppose, for example, that this were the year 1874 and that someone predicted that eventually there would be a radio-manufacturing industry. To be able to make such a forecast successfully, does he have to be challenged to state immediately how many radio manufacturers there would be a century hence, how big they would be, where they would be located, what technology and marketing techniques they would use, and so on? Obviously, such a challenge would make no sense, and in a profound sense the same is true of those who demand a precise portrayal of the pattern of protection activities on the market. Anarchism advocates the dissolution of the state into social and market arrangements, and these arrangements are far more flexible and less predictable than political institutions. The most that we can do, then, is to offer broad guidelines and perspectives on the shape of a projected anarchist society.
One important point to make here is that the advance of modern technology makes anarchistic arrangements increasingly feasible. Take, for example, the case of lighthouses, where it is often charged that it is unfeasible for private lighthouse operators to row out to each ship to charge it for use of the light. Apart from the fact that this argument ignores the successful existence of private lighthouses in earlier days, as in England in the eighteenth century, another vital consideration is that modern electronic technology makes charging each ship for the light far more feasible. Thus, the ship would have to have paid for an electronically controlled beam which could then be automatically turned on for those ships which had paid for the service.
Let us turn now to the problem of how disputes — in particular disputes over alleged violations of person and property — would be resolved in an anarchist society. First, it should be noted that all disputes involve two parties: the plaintiff, the alleged victim of the crime or tort and the defendant, the alleged aggressor. In many cases of broken contract, of course, each of the two parties alleging that the other is the culprit is at the same time a plaintiff and a defendant.
An important point to remember is that any society, be it statist or anarchist, has to have some way of resolving disputes that will gain a majority consensus in society. There would be no need for courts or arbitrators if everyone were omniscient and knew instantaneously which persons were guilty of any given crime or violation of contract. Since none of us is omniscient, there has to be some method of deciding who is the criminal or lawbreaker which will gain legitimacy; in short, whose decision will be accepted by the great majority of the public.
In the first place, a dispute may be resolved voluntarily between the two parties themselves, either unaided or with the help of a third mediator. This poses no problem, and will automatically be accepted by society at large. It is so accepted even now, much less in a society imbued with the anarchistic values of peaceful cooperation and agreement. Secondly and similarly, the two parties, unable to reach agreement, may decide to submit voluntarily to the decision of an arbitrator. This agreement may arise either after a dispute has arisen, or be provided for in advance in the original contract. Again, there is no problem in such an arrangement gaining legitimacy. Even in the present statist era, the notorious inefficiency and coercive and cumbersome procedures of the politically run government courts has led increasing numbers of citizens to turn to voluntary and expert arbitration for a speedy and harmonious settling of disputes.
Thus, William C. Wooldridge has written that
Arbitration has grown to proportions that make the courts a secondary recourse in many areas and completely superfluous in others. The ancient fear of the courts that arbitration would "oust" them of their jurisdiction has been fulfilled with a vengeance the common-law judges probably never anticipated. Insurance companies adjust over fifty thousand claims a year among themselves through arbitration, and the American Arbitration Association (AAA), with headquarters in New York and twenty-five regional offices across the country, last year conducted over twenty-two thousand arbitrations. Its twenty-three thousand associates available to serve as arbitrators may outnumber the total number of judicial personnel … in the United States…. Add to this the unknown number of individuals who arbitrate disputes within particular industries or in particular localities, without formal AAA affiliation, and the quantitatively secondary role of official courts begins to be apparent.
Wooldridge adds the important point that, in addition to the speed of arbitration procedures vis-à-vis the courts, the arbitrators can proceed as experts in disregard of the official government law; in a profound sense, then, they serve to create a voluntary body of private law. "In other words," states Wooldridge, "the system of extralegal, voluntary courts has progressed hand in hand with a body of private law; the rules of the state are circumvented by the same process that circumvents the forums established for the settlement of disputes over those rules…. In short, a private agreement between two people, a bilateral "law," has supplanted the official law. The writ of the sovereign has cease to run, and for it is substituted a rule tacitly or explicitly agreed to by the parties. Wooldridge concludes that "if an arbitrator can choose to ignore a penal damage rule or the statute of limitations applicable to the claim before him (and it is generally conceded that he has that power), arbitration can be viewed as a practically revolutionary instrument for self-liberation from the law…."
It may be objected that arbitration only works successfully because the courts enforce the award of the arbitrator. Wooldridge points out, however, that arbitration was unenforceable in the American courts before 1920, but that this did not prevent voluntary arbitration from being successful and expanding in the United States and in England. He points, furthermore, to the successful operations of merchant courts since the Middle Ages, those courts which successfully developed the entire body of the law merchant. None of those courts possessed the power of enforcement. He might have added the private courts of shippers which developed the body of admiralty law in a similar way.
How then did these private, "anarchistic," and voluntary courts ensure the acceptance of their decisions? By the method of social ostracism, and by the refusal to deal any further with the offending merchant. This method of voluntary "enforcement," indeed proved highly successful. Wooldridge writes that "the merchants' courts were voluntary, and if a man ignored their judgment, he could not be sent to jail…. Nevertheless, it is apparent that … [their] decisions were generally respected even by the losers; otherwise people would never have used them in the first place…. Merchants made their courts work simply by agreeing to abide by the results. The merchant who broke the understanding would not be sent to jail, to be sure, but neither would he long continue to be a merchant, for the compliance exacted by his fellows … proved if anything more effective than physical coercion." Nor did this voluntary method fail to work in modern times. Wooldridge writes that it was precisely in the years before 1920, when arbitration awards could not be enforced in the courts,
that arbitration caught on and developed a following in the American mercantile community. Its popularity, gained at a time when abiding by an agreement to arbitrate had to be as voluntary as the agreement itself, casts doubt on whether legal coercion was an essential adjunct to the settlement of most disputes. Cases of refusal to abide by an arbitrator's award were rare; one founder of the American Arbitration Association could not recall a single example. Like their medieval forerunners, merchants in the Americas did not have to rely on any sanctions other than those they could collectively impose on each other. One who refused to pay up might find access to his association's tribunal cut off in the future, or his name released to the membership of his trade association; these penalties were far more fearsome than the cost of the award with which he disagreed. Voluntary and private adjudications were voluntarily and privately adhered to, if not out of honor, out of the self-interest of businessmen who knew that the arbitral mode of dispute settlement would cease to be available to them very quickly if they ignored an award.
It should also be pointed out that modern technology makes even more feasible the collection and dissemination of information about people's credit ratings and records of keeping or violating their contracts or arbitration agreements. Presumably, an anarchist society would see the expansion of this sort of dissemination of data and thereby facilitate the ostracism or boycotting of contract and arbitration violators.
How would arbitrators be selected in an anarchist society? In the same way as they are chosen now, and as they were chosen in the days of strictly voluntary arbitration: the arbitrators with the best reputation for efficiency and probity would be chosen by the various parties on the market. As in other processes of the market, the arbitrators with the best record in settling disputes will come to gain an increasing amount of business, and those with poor records will no longer enjoy clients and will have to shift to another line of endeavor. Here it must be emphasized that parties in dispute will seek out those arbitrators with the best reputation for both expertise and impartiality and that inefficient or biased arbitrators will rapidly have to find another occupation.
Thus, the Tannehills emphasize:
the advocates of government see initiated force (the legal force of government) as the only solution to social disputes. According to them, if everyone in society were not forced to use the same court system … disputes would be insoluble. Apparently it doesn't occur to them that disputing parties are capable of freely choosing their own arbiters…. they have not realized that disputants would, in fact, be far better off if they could choose among competing arbitration agencies so that they could reap the benefits of competition and specialization. It should be obvious that a court system which has a monopoly guaranteed by the force of statutory law will not give as good quality service as will free-market arbitration agencies which must compete for their customers….
Perhaps the least tenable argument for government arbitration of disputes is the one which holds that governmental judges are more impartial because they operate outside the market and so have no vested interests…. Owing political allegiance to government is certainly no guarantee of impartiality! A governmental judge is always impelled to be partial — in favor of the government, from whom he gets his pay and his power! On the other hand, an arbiter who sells his services in a free market knows that he must be as scrupulously honest, fair, and impartial as possible or no pair of disputants will buy his services to arbitrate their dispute. A free-market arbiter depends for his livelihood on his skill and fairness at settling disputes. A governmental judge depends on political pull.
If desired, furthermore, the contracting parties could provide in advance for a series of arbitrators:
It would be more economical and in most cases quite sufficient to have only one arbitration agency to hear the case. But if the parties felt that a further appeal might be necessary and were willing to risk the extra expense, they could provide for a succession of two or even more arbitration agencies. The names of these agencies would be written into the contract in order from the "first court of appeal" to the "last court of appeal." It would be neither necessary nor desirable to have one single, final court of appeal for every person in the society, as we have today in the United States Supreme Court.
Arbitration, then, poses little difficulty for a portrayal of the free society. But what of torts or crimes of aggression where there has been no contract? Or suppose that the breaker of a contract defies the arbitration award? Is ostracism enough? In short, how can courts develop in the free-market anarchist society which will have the power to enforce judgments against criminals or contract breakers?
In the wide sense, defense service consists of guards or police who use force in defending person and property against attack, and judges or courts whose role is to use socially accepted procedures to determine who the criminals or tortfeasors are, as well as to enforce judicial awards, such as damages or the keeping of contracts. On the free market, many scenarios are possible on the relationship between the private courts and the police; they may be "vertically integrated," for example, or their services may be supplied by separate firms. Furthermore, it seems likely that police service will be supplied by insurance companies who will provide crime insurance to their clients. In that case, insurance companies will pay off the victims of crime or the breaking of contracts or arbitration awards and then pursue the aggressors in court to recoup their losses. There is a natural market connection between insurance companies and defense service, since they need pay out less benefits in proportion as they are able to keep down the rate of crime.
Courts might either charge fees for their services, with the losers of cases obliged to pay court costs, or else they may subsist on monthly or yearly premiums by their clients, who may be either individuals or the police or insurance agencies. Suppose, for example, that Smith is an aggrieved party, either because he has been assaulted or robbed, or because an arbitration award in his favor has not been honored. Smith believes that Jones is the party guilty of the crime. Smith then goes to a court, Court A, of which he is a client, and brings charges against Jones as a defendant. In my view, the hallmark of an anarchist society is one where no man may legally compel someone who is not a convicted criminal to do anything, since that would be aggression against an innocent man's person or property. Therefore, Court A can only invite rather than subpoena Jones to attend his trial. Of course, if Jones refused to appear or send a representative, his side of the case will not be heard. The trial of Jones proceeds. Suppose that Court A finds Jones innocent. In my view, part of the generally accepted law code of the anarchist society (on which see further below) is that this must end the matter unless Smith can prove charges of gross incompetence or bias on the part of the court.
Suppose, next, that Court A finds Jones guilty. Jones might accept the verdict, because he too is a client of the same court, because he knows he is guilty, or for some other reason. In that case, Court A proceeds to exercise judgment against Jones. Neither of these instances poses very difficult problems for our picture of the anarchist society. But suppose, instead, that Jones contests the decision; he then goes to his court, Court B, and the case is retried there. Suppose that Court B, too, finds Jones guilty. Again, it seems to me that the accepted law code of the anarchist society will assert that this ends the matter; both parties have had their say in courts which each has selected, and the decision for guilt is unanimous.
Suppose, however, the most difficult case: that Court B finds Jones innocent. The two courts, each subscribed to by one of the two parties, have split their verdicts. In that case, the two courts will submit the case to an appeals court, or arbitrator, which the two courts agree upon. There seems to be no real difficulty about the concept of an appeals court. As in the case of arbitration contracts, it seems very likely that the various private courts in the society will have prior agreements to submit their disputes to a particular appeals court. How will the appeals judges be chosen? Again, as in the case of arbitrators or of the first judges on the free market, they will be chosen for their expertise and their reputation for efficiency, honesty, and integrity. Obviously, appeals judges who are inefficient or biased will scarcely be chosen by courts who will have a dispute. The point here is that there is no need for a legally established or institutionalized single, monopoly appeals court system, as states now provide. There is no reason why there cannot arise a multitude of efficient and honest appeals judges who will be selected by the disputant courts, just as there are numerous private arbitrators on the market today. The appeals court renders its decision, and the courts proceed to enforce it if, in our example, Jones is considered guilty — unless, of course, Jones can prove bias in some other court proceedings.
No society can have unlimited judicial appeals, for in that case there would be no point to having judges or courts at all. Therefore, every society, whether statist or anarchist, will have to have some socially accepted cutoff point for trials and appeals. My suggestion is the rule that the agreement of any two courts, be decisive. "Two" is not an arbitrary figure, for it reflects the fact that there are two parties, the plaintiff and the defendant, to any alleged crime or contract dispute.
If the courts are to be empowered to enforce decision against guilty parties, does this not bring back the state in another form and thereby negate anarchism? No, for at the beginning of this paper I explicitly defined anarchism in such a way as not to rule out the use of defensive force — force in defense of person and property — by privately supported agencies. In the same way, it is not bringing back the state to allow persons to use force to defend themselves against aggression, or to hire guards or police agencies to defend them.
It should be noted, however, that in the anarchist society there will be no "district attorney" to press charges on behalf of "society." Only the victims will press charges as the plaintiffs. If, then, these victims should happen to be absolute pacifists who are opposed even to defensive force, then they will simply not press charges in the courts or otherwise retaliate against those who have aggressed against them. In a free society that would be their right. If the victim should suffer from murder, then his heir would have the right to press the charges.
What of the Hatfield-and-McCoy problem? Suppose that a Hatfield kills a McCoy, and that McCoy's heir does not belong to a private insurance, police agency, or court, and decides to retaliate himself? Since under anarchism there can be no coercion of the noncriminal, McCoy would have the perfect right to do so. No one may be compelled to bring his case to a court. Indeed, since the right to hire police or courts flows from the right of self-defense against aggression, it would be inconsistent and in contradiction to the very basis of the free society to institute such compulsion.
Suppose, then, that the surviving McCoy finds what he believes to be the guilty Hatfield and kills him in turn? What then? This is fine, except that McCoy may have to worry about charges being brought against him by a surviving Hatfield. Here it must be emphasized that in the law of the anarchist society based on defense against aggression, the courts would not be able to proceed against McCoy if in fact he killed the right Hatfield. His problem would arise if the courts should find that he made a grievous mistake and killed the wrong man; in that case, he in turn would be found guilty of murder. Surely, in most instances, individuals will wish to obviate such problems by taking their case to a court and thereby gain social acceptability for their defensive retaliation — not for the act of retaliation but for the correctness of deciding who the criminal in any given case might be. The purpose of the judicial process, indeed, is to find a way of general agreement on who might be the criminal or contract breaker in any given case. The judicial process is not a good in itself; thus, in the case of an assassination, such as Jack Ruby's murder of Lee Harvey Oswald, on public television, there is no need for a complex judicial process, since the name of the murderer is evident to all.
Will not the possibility exist of a private court that may turn venal and dishonest, or of a private police force that turns criminal and extorts money by coercion? Of course such an event may occur, given the propensities of human nature. Anarchism is not a moral cure-all. But the important point is that market forces exist to place severe checks on such possibilities, especially in contrast to a society where a state exists. For, in the first place, judges, like arbitrators, will prosper on the market in proportion to their reputation for efficiency and impartiality. Secondly, on the free market important checks and balances exist against venal courts or criminal police forces. Namely, that there are competing courts and police agencies to whom victims may turn for redress. If the "Prudential Police Agency" should turn outlaw and extract revenue from victims by coercion, the latter would have the option of turning to the "Mutual" or "Equitable" Police Agency for defense and for pressing charges against Prudential. These are the genuine "checks and balances" of the free market, genuine in contrast to the phony check and balances of a state system, where all the alleged "balancing" agencies are in the hands of one monopoly government. Indeed, given the monopoly "protection service" of a state, what is there to prevent a state from using its monopoly channels of coercion to extort money from the public? What are the checks and limits of the state? None, except for the extremely difficult course of revolution against a power with all of the guns in its hands. In fact, the state provides an easy, legitimated channel for crime and aggression, since it has its very being in the crime of tax theft, and the coerced monopoly of "protection." It is the state, indeed, that functions as a mighty "protection racket" on a giant and massive scale. It is the state that says: "Pay us for your 'protection' or else." In the light of the massive and inherent activities of the state, the danger of a "protection racket" emerging from one or more private police agencies is relatively small indeed.
Moreover, it must be emphasized that a crucial element in the power of the state is its legitimacy in the eyes of the majority of the public, the fact that after centuries of propaganda, the depredations of the state are looked upon rather as benevolent services. Taxation is generally not seen as theft, nor war as mass murder, nor conscription as slavery. Should a private police agency turn outlaw, should "Prudential" become a protection racket, it would then lack the social legitimacy which the state has managed to accrue to itself over the centuries. "Prudential" would be seen by all as bandits, rather than as legitimate or divinely appointed "sovereigns" bent on promoting the "common good" or the "general welfare." And lacking such legitimacy, "Prudential" would have to face the wrath of the public and the defense and retaliation of the other private defense agencies, the police and courts, on the free market. Given these inherent checks and limits, a successful transformation from a free society to bandit rule becomes most unlikely. Indeed, historically, it has been very difficult for a state to arise to supplant a stateless society; usually, it has come about through external conquest rather than by evolution from within a society.
Within the anarchist camp, there has been much dispute on whether the private courts would have to be bound by a basic, common law code. Ingenious attempts have been made to work out a system where the laws or standards of decision-making by the courts would differ completely from one to another. But in my view all would have to abide by the basic law code, in particular, prohibition of aggression against person and property, in order to fulfill our definition of anarchism as a system which provides no legal sanction for such aggression. Suppose, for example, that one group of people in society holds that all redheads are demons who deserve to be shot on sight. Suppose that Jones, one of this group, shoots Smith, a redhead. Suppose that Smith or his heir presses charges in a court, but that Jones's court, in philosophic agreement with Jones, finds him innocent therefore. It seems to me that in order to be considered legitimate, any court would have to follow the basic libertarian law code of the inviolate right of person and property. For otherwise, courts might legally subscribe to a code which sanctions such aggression in various cases, and which to that extent would violate the definition of anarchism and introduce, if not the state, then a strong element of statishness or legalized aggression into the society.
But again I see no insuperable difficulties here. For in that case, anarchists, in agitating for their creed, will simply include in their agitation the idea of a general libertarian law code as part and parcel of the anarchist creed of abolition of legalized aggression against person or property in the society.
In contrast to the general law code, other aspects of court decisions could legitimately vary in accordance with the market or the wishes of the clients; for example, the language the cases will be conducted in, the number of judges to be involved, and so on.
There are other problems of the basic law code which there is no time to go into here: for example, the definition of just property titles or the question of legitimate punishment of convicted offenders — though the latter problem of course exists in statist legal systems as well. The basic point, however, is that the state is not needed to arrive at legal principles or their elaboration: indeed, much of the common law, the law merchant, admiralty law, and private law in general, grew up apart from the state, by judges not making the law but finding it on the basis of agreed-upon principles derived either from custom or reason. The idea that the state is needed to make law is as much a myth as that the state is needed to supply postal or police services.
Enough has been said here, I believe, to indicate that an anarchist system for settling disputes would be both viable and self-subsistent: that once adopted, it could work and continue indefinitely. How to arrive at that system is of course a very different problem, but certainly at the very least it will not likely come about unless people are convinced of its workability, are convinced, in short, that the state is not a necessary evil.
This article was first published in The Libertarian Forum, volume 7.1, January 1975, available in PDF. It is based on a talk delivered at the American Society for Political and Legal Philosophy (ASPLP), Washington, DC: December 28, 1974.
 William C. Wooldrdige, Uncle Sam, the Monopoly Man (New Rochelle, New York: Arlington House, 1970), p. 101.
 Ibid., pp. 103–104.
 Ibid., pp. 95–96.
 Ibid., pp. 100–101.
 Ibid., p. 68.
 E.g., David Friedman, The Machinery of Freedom (New York: Harper and Row, 1973).
 Thus, see Bruno Leoni, Freedom and the Law (Princeton, New Jersey: D. Van Nostrand Co., 1961).