Most people take the terms order, law, and government to be coextensive. Without government, there would be no law. Without law, there would be no social order. In fact, however, the three concepts are distinct.
Law may be defined as that institution or set of institutions in a given society that adjudicates conflicting claims and secures compliance in a formal, systematic, and orderly way. Law thus defined is one species of social order, but not the whole of it; there are also less formal mechanisms for maintaining social order. Indeed, the vast bulk of cooperation in society in fact depends on informal order rather than on law.
Law may be subdivided into voluntary and coercive law, depending on the means whereby compliance is secured. Voluntary law, as the name implies, relies solely on voluntary means, such as social pressure, boycotts, and the like, in order to secure compliance with the results of adjudication. Coercive law, on the other hand, relies at least in part on force and threats of force.
Coercive law in turn may be further subdivided into monocentric and polycentric coercive law. Under monocentric coercive law, there is a single institution that claims, and in large part achieves, a coercive monopoly on the use of force to adjudicate claims and secure compliance in a given territorial area. This institution is called a government, and everyone other than the government and its agents is forbidden to adjudicate by force. Under polycentric law, by contrast, no one agency claims or possesses such a monopoly.
An anarchist, then, is not someone who rejects order or law or even coercive law, but rather one who rejects government. The anarchist argues that informal order, voluntary law, and polycentric coercive law are sufficient to maintain social cooperation; the advocate of government argues that monocentric coercive law is needed in addition, and indeed typically maintains that the amount of social order that can be maintained through non-governmental sources alone is quite small.
Yet a great deal of social order is maintained through informal means alone. In Order Without Law: How Neighbours Settle Disputes, economist Robert Ellickson has shown how disputes over land use are frequently resolved informally, without recourse to official adjudication, and certainly without recourse to legal statutes (the relevant statutes being generally unknown to the disputing parties in any case). More broadly, Robert Axelrod in The Evolution of Cooperation has explained why cooperation is generally a successful strategy and thus why it tends to be “selected for” by the market, as cooperative relationships emerge and grow spontaneously without being directed by any authority.
When there is a need for the more formal mechanism of law, this law may be voluntary rather than coercive. An example of voluntary law is the Law Merchant, a system of commercial law that emerged in the late Middle Ages in response to the need for a common set of standards to govern international trade. The merchants, fed up with the excessive rigidity of governmental laws regulating commerce, and frustrated by the lack of uniformity among the commercial codes of different nations, simply formed their own Europe-wide system of courts and legal codes. For enforcement, the Law Merchant relied not on state-imposed penalties but on credit reports; those who refused to abide by the system’s rules and decisions would have a hard time finding other merchants willing to deal with them. (The case of the Law Merchant shows that systems of private law need not depend on kinship or other local ties for their success.)
When law is coercive, it need not be monocentric. For example, under early Anglo-Saxon law, Kings made foreign policy only; domestic policy was left to local courts called Moots, which simply enforced agreed-upon local customs. Neither Kings nor Moots had any power of domestic enforcement; it was up to individuals to enforce the law by private coercion. Such individuals generally formed associations called borhs, pledging security for one another’s reliability; even here, much enforcement was through social sanction (being denied membership in a borh) rather than coercion.
Thus private law, whether strictly voluntary or also coercive, has proven itself historically as an effective provider of social order. But the anarchist’s point is not simply that monocentric law is not necessary in order to maintain social order, but more fundamentally that introducing monocentrism into the picture actually decreases social order.
Advocates of government assume that non-governmental mechanisms for achieving order will be ineffective because of public-goods problems — specifically, the problem that unless people are forced to cooperate, each person will have an incentive to free-ride on the cooperation of others without cooperating himself. This argument is often taken to show the necessity of government.
But if market solutions are beset by perverse incentives caused by public-goods problems, governmental solutions are likewise beset by perverse incentives caused by public-choice problems: Monopolies that collect revenues by force are not accountable to their clients, and state officials need not bear the financial cost of their decisions; inefficiency is the inevitable result. Since both systems involve perverse incentives, the important question is: Which system is better at overcoming such incentives?
And here the answer is clear. Under a market system, entrepreneurs stand to reap financial rewards by figuring out ways to supply “public” goods while excluding free riders. Thus the system that creates the perverse incentives also creates the very incentives to overcome them. That’s why every so-called “public” good has been supplied privately at one time or another in history. Governments, by contrast, must by definition forbid competition. Thus governments, unlike markets, have no way of solving their incentive problems. We would be well-advised, then, to buy our law on the market rather than from the state.
The purpose of a legal system is to provide a systematic, orderly, and predictable mechanism for resolving disagreements. In order to do its job, any such system must perform three closely connected, but nevertheless distinct, functions: Adjudication, legislation, and execution.
The judicial function is the core of any legal system. In its judicial function, a legal system adjudicates disputes, issuing a decision as to how the disagreement should be settled. The other two functions are merely adjuncts to this central function.
The purpose of the legislative function is to determine the rules that will govern the process of adjudication. Legislation tells judicial function how to adjudicate. The legislative process may be distinct from the judicial process, as when the Congress passes laws and the Supreme Court then applies them; or the two processes may coincide, as when a common-law body of legislation arises through a series of judicial precedents.
Finally, the purpose of the executive function is to ensure, first, that the disputing parties submit to adjudication in the first place, and second, that they actually comply with the settlement eventually reached through the judicial process. In its executive function the legal system may rely on coercive force, voluntary social sanctions, or some combination of the two. The executive function gives a legal system its “teeth,” providing incentives for peaceful behaviour; both domestic law enforcement and national defence fall under the executive function.
With regard to these various functions, there are three primary ways in which a legal system may be constituted:
Various combinations of these are possible, since there are legal systems under which some functions are monopolised while others are not. For example, in the Icelandic Free Commonwealth, the legislative function was monopolised by the All-Thing (althingi), or General Assembly; the judicial function was shared between the Thing courts and the private sector; and the executive function was privatised entirely. This is why the legal system of the Icelandic Free Commonwealth cannot easily be classified either as a pure government or as a pure anarchy.
Most of us have been taught to regard Constitutionalism as the best of the three options. Concentrating the three functions in a single agency avoids the chaos allegedly endemic to Anarchism; while assigning the three functions to distinct sub-agencies within the monopoly agency allows the three branches (legislative, executive, and judicial) to serve as checks on one another’s excesses, thus avoiding the potential for abuse and tyranny inherent in Absolutism. This is the “separation of powers” doctrine built into the U. S. Constitution.
In practice, however, Constitutionalism has proved only marginally better than Absolutism, because there has been sufficient convergence of interests among the three branches that, despite occasional squabbles over details, each branch has been complicit with the others in expanding the power of the central government. Separation of powers, like federalism and elective democracy, merely simulates market competition, within a fundamentally monopolistic context.
In his libertarian classic Two Treatises of Government, the 17th-century English philosopher John Locke offered one of the most famous cases ever made for the monopolisation of the three functions of government. Locke believes that all human beings are naturally equal, so that in their natural state each person has as much right as any other to exercise the various functions of law:
“Man, being born, as has been proved, with a title to perfect freedom and an uncontrolled enjoyment of all the rights and privileges of the Law of Nature, equally with any other man, or number of men in the world, hath by nature a power not only to preserve his property — that is, his life, liberty, and estate, against the injuries and attempts of other men, but to judge of and punish the breaches of that law in others, as he is persuaded the offence deserves . . . each being, where there is no other, judge for himself and executioner . . .” (II. vii. 87.)
This egalitarian distribution of political authority, Locke argues, is required by justice unless individuals voluntarily relinquish their authority to a government. However, Locke thinks that people living in a state of anarchy will find it rational to set up a government in order to gain greater security:
“If man in the State of Nature be so free as has been said, if he be absolute lord of his own person and possessions, equal to the greatest and subject to nobody, why will he part with his freedom, this empire, and subject himself to the dominion and control of an other power? To which it is obvious to answer, that though in the State of Nature he hath such a right, yet the enjoyment of it is very uncertain and constantly exposed to the invasion of others; for all being kings as much as he, every man his equal, and the greater part no strict observers of equity and justice, the enjoyment of the property he has in this state is very unsafe, very insecure. This makes him willing to quit this condition which, however free, is full of fears and continual dangers; and it is not without reason that he seeks out and is willing to join in society with others who are already united, or have a mind to unite for the mutual preservation of their lives, liberties and estates, which I call by the general name — property. The great and chief end, therefore, of men uniting into commonwealths, and putting themselves under government, is the preservation of their property; to which in the State of Nature there are many things wanting.” (II. ix. 123-124.)
Locke then goes on to list what he sees as the three principal defects of the state of natural anarchy. Although he does not point this out explicitly, the three defects appear to correspond to the three functions of law that I have been discussing, and I have labeled them accordingly:
[The Legislative Defect.] “Firstly, there wants an established, settled, known law, received and allowed by common consent to be the standard of right and wrong, and the common measure to decide all controversies between them. For though the Law of Nature be plain and intelligible to all rational creatures, yet men, being biased by their interest, as well as ignorant for want of study of it, are not apt to allow of it as a law binding them in the application of it to their particular cases.
[The Judicial Defect.] Secondly, in the State of Nature there wants a known and indifferent judge, with authority to determine all differences according to the established law. For every one in that state being both judge and executioner of the Law of Nature, men being partial to themselves, passion and revenge is very apt to carry them too far, and with too much heat in their own cases, as well as negligence and unconcernedness, make them too remiss in other men’s.
[The Executive Defect.] Thirdly, in the State of Nature there often wants power to back and support the sentence when right, and to give it due execution. They who by any injustice offended will seldom fail where they are able by force to make good their injustice. Such resistance many times makes the punishment dangerous, and frequently destructive to those who attempt it.” (II. ix. 124-126.)
Locke concludes that these three defects may be remedied by centralising the legislative, judicial, and executive functions in a constitutional government.
I think Locke’s arguments for a monocentric legal system contain a serious confusion: The confusion between the absence of government and the absence of law. Locke’s arguments are good arguments for a formal, organised legal system; but Locke mistakenly assumes that such a system requires a governmental monopoly. The majority of legal systems throughout history, however, have been polycentric rather than monocentric. Locke did not have the benefit of our historical knowledge however; nor, despite his brilliance, was he able to imagine on his own a legal system that was not a government. The actual history of stateless legal orders shows that they do not noticeably suffer from any of the three defects Locke lists; on the contrary, those defects are far more prevalent under governmental law.
Consider first the judicial defect: The worry that, in the absence of common authority, each individual would have to act as a judge in his or her own case, with all the problems of bias and partiality that entails. Locke is correct in thinking that submitting disputes to impartial third-party arbitration is generally preferable to acting as one’s own judge and jury (except, of course, in emergency cases in which one must act quickly and no such impartial judge is available). But such third-party judges will always be available, whether or not there is a government. There is a widespread tendency to suppose that if something is not supplied by the government, it cannot be supplied at all; I call this “the invisibility of the market.” (The problem with invisible hands is that you need libertarian lenses in order to see them — whereas everyone can see the visible hand of government.) Polycentric legal systems have always had plenty of third-party judges, from the relatively formal Moots of early Anglo-Saxon law (in which disputants were judged by their peers on the basis of local custom) to the relatively informal arrangements of the American frontier (in which each disputant would pick an arbiter, the two arbiters together would pick a third, and the judgment of the three together would be binding). History shows that stateless legal orders tend to create powerful incentives for people to submit their disputes to arbitration wherever possible, in order to avoid the appearance of being an aggressor (and thus the target of defensive coercion oneself). Anarchy does not suffer from Locke’s judicial defect.
But government does. In any dispute between a citizen and the state, the state must by necessity act as a judge in its own case — since, as a monopoly, it can recognise no judicial authority but its own. Hence governments by their nature must be subject to the judicial defect. Constitutionalism is supposed to remedy this defect by separating the judicial branch from the executive and legislative branches, so as to prevent the judging agency from being a party to the dispute. But what if the citizen’s quarrel is with the judicial branch itself? In any case, even if the quarrel is solely with the legislative or executive branch, it would be naive to assume that the judicial branch of a monopoly will be unsullied by the interests of the other branches. No one with a complaint against the marketing division of General Motors would be satisfied to have the case adjudicated by the legal division of General Motors! The solution to the judicial defect, then, is not a monocentric judiciary, but a polycentric one.
Next, consider the legislative defect: The worry that without government there will be no generally known and agreed-upon body of law. Why not? We should rather expect markets to converge on a relatively uniform set of laws for the same reason that they tend to converge on a single currency: Customer demand. The late medieval private system of mercantile law known as the Law Merchant (lex mercatoria), for example, offered a more unified body of law than did the governmental systems with which it competed.
This should be no surprise. Why are there no triangular credit cards? The reason is not government regulation, but rather that — given our current system that relies on rectangular cards — no one would accept it (unless the government made them accept it, thus preventing the market drive toward uniformity). Similar reasons explain why the market no longer carries both VHS and Betamax video cartridges, but only VHS; the market creates uniformity when customers need it, and diversity when they need that instead. It’s a good thing that video cassettes come with lots of different kinds of movies, and so the market ensures this; it would be a bad thing if video cassettes came in fifty different shapes and sizes, and so the market prevents this.
Indeed, it is not polycentric legal systems, but rather monocentric ones, that suffer from the legislative defect, since a mountain of bureaucratic regulations that no one can read is in effect equivalent to an absence of generally known law. Under a private legal system, changes in law occur as a response to customer needs, and so the body of law is less likely to metastasise to such unwieldy proportions. The solution to the legislative defect is not to monopolise legislation, but rather to privatise it.
Finally, consider the executive defect: The worry that without government there would be insufficient power on the part of private individuals to enforce the law. It is true that under anarchy each individual has the right to exercise the executive function on his or her own, but it does not follow that law enforcers will in practice be solitary and unaided. On the contrary, voluntary associations of enforcers typically emerge — as in the case of the thief-takers’ associations of early 19th-century England, or the vigilance committees of the old American frontier. Hollywood movies have accustomed us to think of the latter associations as unruly lynch mobs, and have depicted the frontier as nightmarishly violent; in historical fact, the level of criminal violence in frontier society was far lower than in our own, and the protective associations were, for the most part, reliable organisations that gave their defendants fair trials (at which defendants were often acquitted — not the mark of a kangaroo court). Indeed, the whole notion of an organised police force is a relatively modern concept; police were extremely rare throughout ancient, medieval, and modern history, until about the mid-19th century. (Indeed, even the notion of a distinct governmental military is fairly unusual historically; in most societies, both law enforcement and national defence have been the job of the armed citizenry.)
If there is an executive defect, it applies not to private law but to public law, in which individuals typically lack the power to withstand the arbitrary caprice of the state. Against one marauding band one can form one’s defensive band; but who can resist the overwhelming force of an organised government? Let the victims of Warsaw, Tiananmen, or Waco judge whether the centralisation of law enforcement enhanced the security of their lives, liberties, and estates.
Abuse of power by law enforcers is in fact much easier to keep in check under the discipline of a competitive market system. The LAPD would have gone bankrupt overnight after the Rodney King beating if it had been a private security force with competitors in the same territory; but as matters stood, despite the public outcry, the LAPD’s “clients” had nowhere else to go, and so the LAPD’s incentive to reform its behaviour is much weaker.
In short, then, the three defects Locke cites as objections to anarchy are in fact much more effective objections to government. None of the three functions of government — executive, legislative, or judicial — should be assigned to an exclusive monopoly. In the words of F. A. Hayek: “Law is too important a matter to be left in the hands of government.”
In one of Plato’s philosophical dialogues, called the Minos, Socrates asks an unnamed comrade for a definition of law. The comrade complies, offering the following definition: “Law is what is legislated.” But Socrates objects: Just as sight isn’t what we see, but rather that by which we see, so likewise law is not what is legislated but that by which we legislate. The comrade accepts this criticism and retracts his definition. This might surprise us: Surely law is the product of legislation, not vice versa. But in saying that law is that by which we legislate, Socrates is in fact appealing to a very old and deep-seated idea, as we shall see.
The comrade’s second definition is this: “Law is the judgment of the state.” But through repeated questioning Socrates quickly proves that this definition clashes with other things the comrade believes; thus the comrade is committed to an inconsistent triad of beliefs:
If the comrade accepts any two, he must reject a third.
Clearly, Socrates’ comrade is attracted both to a positivist conception of law (according to which law is whatever the government says, be it just or unjust) and to a moralised conception of law (according to which law is inherently just); and Socrates exploits this tension.
So Socrates suggests a revision: “Law is the correct judgment of the state.” Thus only those judgments of the state that are correct count as genuine laws. This may seem odd to us; when state decrees are incorrect, we tend to say that they are bad laws or unjust laws, not that they aren’t laws at all. Being a law is a purely descriptive fact with no evaluative weight: Anything the legislature cooks up, whether good or bad, is ipso facto a law.
Why would anyone think otherwise? Well, consider the distinction between power and authority. What’s the difference between a command issued by a legislature, and a command issued by a mugger with a gun? Both have the power to enforce their demands; but the legislature, unlike the mugger, is presumed to have authority. Yet the legislature’s authority is conditional; if Congress were to pass a bill outlawing Methodism, it would be overstepping its constitutional authority, and so its decree would not have the force of law. But if the Congress derives its authority from the Constitution, where does the Constitution get its authority? At this point we can only conclude that the Constitution’s authority, if any, must be moral in character, deriving from natural justice. Only something with intrinsic normative weight could serve as the Unauthorised Authoriser that transforms all lesser decrees into laws.
But Socrates needn’t be relying purely on an argument of this sort. He also has a weighty historical tradition on his side. Socrates’ conception of law is arguably the dominant one historically, and our positivistic one a mere anomaly; the concept of law as an objective standard to be declared or discovered (rather than created) by legislators was the dominant notion both in legal practice and in legal philosophy throughout most of history — called rta or dharma in India, ma’at in Egypt, and torah in Judea. That’s why Socrates can speak uncontroversially of law as not what is legislated but that by which we legislate. It was a standard principle of jurisprudence for the next two millennia that lex injusta non est lex: An unjust law is not a law. Not until the Enlightenment did the notion of Natural Law degenerate from its original notion, a constraint on what law was, to a mere constraint on what law ought to be.
Today’s positivistic conception of law is thus really something of a historical aberration; though it seems to have had some currency in ancient Greece as well, as is shown by the comrade’s resistance, as well as by the fact that the Greek word nomos means both “law” and “convention.” (A similar tension between positivist and moralised conceptions of law is found in the Greek statesman Pericles’ confused responses to Alcibiades’ Socratic questioning in Xenophon’s Recollections of Socrates. Perhaps the fact that Athens was a democracy, and the average Athenian was constantly engaged in passing and repealing laws, served to weaken the traditional moralised conception of law.)
Socrates argues that only decrees based on knowledge of objective justice and injustice can count as true laws; he adds that all states legislate the just, but they make mistakes about what in fact is just. Socrates’ point here is reminiscent of an argument by David Lyons that legal interpretation presupposes moral theory:
“Imagine that you and I disagree about the substantive requirements of social justice. We then differ as to how the concept of justice applies; we differ, that is, about the principles of justice. This is possible if the concept of justice admits of different interpretations, or competing conceptions . . . Now consider a constitutional example . . . a court applying the just compensation clause would not necessarily decide a case as the original authors would have done . . . Instead, a court would understand the Constitution to mean precisely what it says and thus to require just compensation. A court would need to defend a particular conception of just compensation . . . against the most plausible alternatives . . . Contested concepts do not seem confined to morality and law. Their properties are at any rate similar to those of concepts referring to natural substances or phenomena, such as water and heat. On a plausible understanding of the development of science, for example, the caloric and kinetic theories of heat are (or at one time were) competing conceptions of the concept heat . . . If, as most people would agree, ‘heat’ refers to a determinate physical phenomenon, there can be, in principle, a best theory of heat. This implies that there can be a best conception of a contested concept. This suggests, in turn, that contested concepts in the Constitution might have best interpretations . . . Now if the idea that the Constitution includes contested concepts is correct, then to apply the Constitution in terms of their best interpretation is, in effect, to apply doctrines whose application is called for by the original Constitution. But, just as interpretation of the concept heat requires more than mere reflection, any interpretation of this type inevitable draws upon resources that are neither implicit in the text nor purely linguistic. It . . . requires that courts applying ‘vague clauses’ of the Constitution interpret ‘contested concepts,’ which requires reasoning about moral or political principles.” (David Lyons, “Constitutional Interpretation and Original Meaning.” Social Philosophy & Policy IV, pp. 85-99.)
If the law says that government employees must be paid in gold, then they may not be paid in iron pyrites, since iron pyrites is not in fact gold, even if those who wrote the law were ignorant of the difference. If the law says that fishermen may not hunt mammals, then in fact the law says they may not hunt dolphins, even if the lawmakers had thought dolphins were fish. Likewise, if the law says that involuntary servitude is forbidden, then the government may not conscript soldiers, since military conscription is in fact involuntary servitude, even if those who wrote the law did not recognise this.
Professor Lyons’ point is that precisely the same argument applies to moral terms: If the Constitution demands just compensation for victims of eminent domain, then such victims must receive whatever is actually just, not what the framers thought was just, since the Constitution says to give “just compensation” rather than saying to give “what we consider just compensation.” (The 19th-century abolitionist Lysander Spooner used similar arguments in his Unconstitutionality of Slavery, claiming that slavery was outlawed by various clauses in the Constitution even if the authors of those clauses had no such intention, because such phrases as “republican form of government” and “against domestic violence,” when interpreted in accordance with the correct moral and political theory, ruled out slavery.)
The conclusion of Plato’s Minos, then, might be described as follows: All states legislate both the concept of justice, and particular conceptions of it. Insofar as they legislate the concept, they all legislate the same thing, and these legislations are genuine laws. Insofar as they legislate different conceptions, their decrees (or most of them) are not genuine laws, and their legislators are simply proving themselves to be ignorant of what the law truly requires.
In Part I of this series of articles on the nature of law, I defined law as “that institution or set of institutions in a given society that adjudicates conflicting claims and secures compliance in a formal, systematic, and orderly way.” It should now be clear that I was there defining positive law, not Law in the traditional strict sense discussed here. One of my principal aims in Parts I and II was to argue in favour of a specific kind of positive legal system — market anarchism — as both morally and practically superior to other systems. My conclusion there might now be rephrased as follows: Market anarchism is the variety of positive law most in accordance with Law in the proper sense.
But what is the precise relation between positive law and Law proper? To that question I now turn.
My account of the traditional conception of Law proper might suggest that the content of this Law is entirely independent of human will. Some legal philosophers in this tradition have indeed thought this. Lysander Spooner, for example, insists that human legislation can neither add to nor remove from the true Law a single provision.
The more common view historically, however, has been that of the great medieval philosopher Thomas Aquinas. Aquinas held that the content of true Law included not only Natural Law — that is, the principles of justice requisite to genuine human well-being, and inherent in human nature as created by God — but also Human Law. By Human Law Aquinas does not mean what I have been calling positive law. His idea is rather the following:
Some of the provisions of Natural Law, while absolute and binding, are often lacking in specificity. For example, it might be a provision of Natural Law that cars going in opposite directions on a highway should drive on opposite sides of the highway — but the Natural Law might be silent on the question of whether cars should drive on the left or on the right. Any decision on this latter question is a matter of indifference, from the standpoint of Natural Law, and may be left up to human convention. All the Natural Law requires is that there be some decision on the matter, and that whichever convention is adopted should then be obeyed. Thus if a particular nation adopts the rule of driving on the right, this latter provision then acquires the force of Law, and so is morally binding. The rule “Drive on the right” is not part of the unchanging Natural Law, but is rather a provision of mutable Human Law. Medieval jurists spoke of such rules as reducing (that is, as making more specific) the provisions of Natural Law; but they denied that Human Law could ever contradict the Natural Law. Law in the strict sense, then, covers both Natural Law and Human Law, the latter being subordinate to the former; but Human Law is narrower than positive law, since only those provisions of positive law that are consistent with justice are to be counted as Human Law. The legislator may have some creative freedom, but only within the bounds of the Natural Law, and it is his or her task to discover those bounds, not to stipulate them by fiat.
I have spoken of the standard to which legislation must answer as Natural Law — a set of immutable moral principles that transcend human will. Such was indeed the view of Plato, Aristotle, Cicero, Aquinas, and indeed most legal philosophers throughout history. (For some representative quotations, see the documentary evidence on pp. 17-18.) But legal historians point to what might seem a different conception of true Law: The practice, in most pre-modern societies, of regarding traditional custom as the supreme standard of Law. (Again, see pp. 17-18.) The task of the legislator, in such societies, is seen as the attempt to discover, state, and apply the already existing practices of the tribe or nation — what the British jurists call the “custom of the country” — and not to appeal to some abstract standard of transcendent justice such as Natural Law.
This conflict is largely illusory, however. For we must recall that (Spooner to the contrary not withstanding) the true Law comprises not only Natural Law but also Human Law — and Human Law may be enacted not only by an official legislature but likewise, and with equal (if not greater) authority, by spontaneously evolving custom. Indeed, such customary law is probably a more reliable method for “reducing” the Natural Law, because a spontaneously evolved and voluntarily maintained custom is more likely to promote mutual advantage than a decree devised and imposed by a small group in power.
On similar grounds it has been argued, by F. A. Hayek and Bruno Leoni among others, that a common-law system, in which legislation arises through judicial precedent, is superior to a system in which judges and courts simply apply legislation created by a separate legislature. (The American system is a mixture of these two.) One advantage of the common-law system of judge-made law is that a judge cannot simply start legislating about anything that strikes his fancy, but must respond to particular claims brought by particular people, and so the system of precedents that evolves has been shaped by the needs of individuals.
Such a common-law system works best, however, if there are competing courts and competing jurisdictions, so that courts that make bad decisions will lose out over time to those with better judgment. Under a centralised judiciary with restricted choice in jurisdictions, many of the advantages of common law are lost — though even here there is a sort of competitive element, insofar as different precedents may be thought of as competing against one another.
A common-law system will not work well if courts ignore precedents altogether; at that point a judge simply becomes a mini-legislator, rejecting the wisdom embodied in earlier judicial experience. On the other hand, a common-law system will also fail to work well if it adheres too closely to precedent; for if judicial entrepreneurs refuse to innovate or to introduce competing precedents, the invisible hand has nothing to work with. Human Law, unlike Natural Law, is supposed to be flexible, adapting itself to changing circumstances. Tom Bethell offers the Islamic legal system as an example of a common-law system that degenerated when it lost its flexibility, thus freezing the once dynamic and progressive Islamic civilisation into medieval rigidity:
“. . . the decline of Islam began approximately in the 15th century . . . Gradually, Islamic law was ‘frozen,’ so that the interpreters of the law could no longer apply their independent reasoning to it. They were obliged to live with the interpretation that had been reached when the ‘freeze’ took place. This event is known to shari’a (religious law) scholars as ‘the closing of the gate of itjihad’ — itjihad meaning ‘the struggle for understanding,’ or more simply the use of reason. It was replaced by taqlid, the submissive acceptance of earlier interpretation. Continued interpretation ceased because it was said to show disrespect for earlier jurists. Taqlid brought with it serious problems . . . Some Islamic scholars . . . believe that the closing of the gate was a major cause of the decline of Islam . . . With independent thought no longer desired, law in the Muslim world became dominated by people of a subservient disposition who were attracted to the service of power . . . Taha al Alwani denounces the fallen state of the Muslim world . . . ‘Muslims and non-Muslims alike are amazed that one of history’s most advanced civilisations could fall into such a state of overwhelming wretchedness, ignorance, backwardness and overall decline . . .’ He believes that the ingrained deference to authority and the discouragement of reason that began with the ‘closing of the gate’ is an important part of the explanation.” (Tom Bethell, “The Mother of All Rights,” Reason 25 (April 1994), p. 45.)
In his classic manifesto On Liberty, John Stuart Mill pointed out the benefits of intellectual competition in reaching the truth. It is by precisely this method that we have achieved the staggering scientific progress of the past four centuries. A judicial system that likewise incorporates the principle of competition — neither forswearing the vast information embedded in the market process, nor prostrating itself before it in such a way as to preclude entrepreneurial innovation — is more likely than any other to succeed in discovering and effectively applying the principles of Natural Law.
In previous instalments of this series, I have referred to Natural Law as the transcendent standard to which manmade laws must correspond in order to be legitimate. But is there such a thing as Natural Law? Are we justified in appealing to such a concept? Or is it hopelessly outdated, an unscientific remnant of a medieval way of thinking?
Traditionally, Natural Law is called “natural” for two reasons. First, Natural Law is distinguished from conventional law; in other words, Natural Law does not depend on or derive from manmade institutions and customs. (If it did, it would not be able to serve as a standard by which to judge manmade law.) Second, Natural Law is distinguished from supernatural law; in other words, Natural Law is accessible to human reason rather than requiring divine revelation. (Historically, Natural Law theorists have disagreed with one another about whether Natural Law derives its authority from God’s commands; but even those who have held — wrongly, in my view — that Natural Law does indeed depend on divine commands have nevertheless insisted that Natural Law represents that portion of God’s commands that we could figure out for ourselves as being rational and reasonable, through our own unaided intellect, without appeal to scripture or other forms of revelation.)1
But the very features of Natural Law that make it attractive — its independence of human customs and its accessibility to reason — are also the features that make it controversial.
How can there be a law that doesn’t rest on any legal institutions or practices? What is it grounded on instead? In other words, what is the metaphysical basis of Natural Law?
Likewise, how can a moral standard be ascertained by human reasoning? How could we ever acquire objective knowledge of what is right and what is wrong? In other words, what is the epistemological basis of Natural Law?
Without some answers to these questions — or at least, without some hope that they can in principle be answered — any political theory that appeals to Natural Law is going to be on shaky ground.
A full-scale defence of Natural Law theory, however, is a task beyond the scope of this article; so I will confine myself to responding to some of the most common objections I’ve encountered within the libertarian community to the notion of Natural Law (and the associated concept of natural rights).
But first let me make a point about the burden of proof. Most critics of Natural Law assume that the burden of proof lies with the proponent of Natural Law — presumably because they see Natural Law as something bizarre and implausible, something one couldn’t sensibly believe unless there were a knock-down argument for it. But in fact, to believe in Natural Law is simply to believe that there are moral standards that transcend the practices and customs of any given community — that there are rational grounds for condemning the Nazi regime as immoral, that it is possible to be justified in so condemning it, even if we assume that what the Nazis did was perfectly in accordance with the values of Nazi culture. When we condemn Nazism, we don’t ordinarily take ourselves to be expressing a purely personal, subjective preference, like the preference for chocolate over vanilla; rather, our ordinary practices of praising and condemning seem to implicitly assume that there are objective moral standards, i.e., that there is a Natural Law to which manmade laws are answerable.
Now of course the fact that ordinary practices implicitly assume something is no guarantee that what they assume is true. But such a fact does seem to shift the burden of proof.
Consider: The fact that it seems to me that I am sitting in front of my computer typing these words doesn’t guarantee that I really am doing so; I might be dreaming, or hallucinating, or I might be trapped in an incredibly realistic virtual-reality program. Now a philosopher like Descartes would say that I have the burden of proving that I’m not dreaming, hallucinating, etc. — that I have to be able to rule those alternatives out before I can be justified in thinking I really am here, awake and typing.
But if Descartes were right — if we couldn’t be justified in believing anything unless we first ruled out all possibility of error — then we would never be justified in believing anything, since whatever evidence we pointed to in order to prove our starting beliefs reliable would in turn have to be justified by appeal to further evidence and so on ad infinitum. And if that were so, then we couldn’t be justified in holding the
belief that started us down this infinite regress in the first place — namely, the belief that in order to be justified in believing anything we must first rule out all possibility of error. So Cartesian skepticism ultimately undermines itself: If everything should be doubted, then the claim that everything should be doubted is itself one of the things that should be doubted — and once we go doubting that, we lose our original reason for doubting everything else.2
What that means is that we are, after all, justified in accepting the way things initially seem to us as a true picture of the world, despite the possibility that those beliefs are mistaken. Now that doesn’t mean we’re justified in clinging to our beliefs with blind faith, defying all evidence to the contrary. But it does mean that those who oppose these ordinary beliefs are the ones who have the burden of proof; we’re justified in accepting our initial beliefs as true until we find convincing evidence that they’re false. This must be so, because the contrary position, as we’ve seen, is rationally incoherent. So if our ordinary practice of moral judgment commits us to believing in Natural Law, then Natural Law is part of our picture of the world, and we’re justified in accepting it until someone gives us good reason to reject it. The burden of proof thus rests with the opponents of Natural Law.
That is not to say that I think there is no positive case to be made for Natural Law. On the contrary, much of my own philosophical research is devoted to making such a case, relying on the insights of the Aristotelean tradition combined with the philosophical discoveries of the last thirty years. My point is simply that the justifiability of accepting Natural Law as part of one’s picture of the universe does not require that the positive case for Natural Law be established first.
Now let’s turn to some of those common objections to Natural Law theory.
One objection one sometimes comes across in libertarian circles is that Natural Law, and in particular natural rights (the rights we have under Natural Law), are useless. A Natural Law against murder or theft will not protect us from murderers and thieves; a natural right to life will not turn a mugger’s knife-blade or repel an assassin’s bullet; a natural right to property is not as useful as high walls and sturdy locks.
One version of this criticism is put forward by L. A. Rollins in his pamphlet The Myth of Natural Rights (Port Townsend: Loompanics, 1983). Rollins asks:
“How many Jewish lives [under the Nazis] were saved by their natural right to life? The answer, of course, is: Zero . . . If all Jews of Nazi-occupied Europe had a natural right to life and, yet, the Nazi regime was able to kill six million of them, then clearly natural rights are of no value whatever as protective devices. A bullet-proof vest may protect a person against being shot, but a natural right has never stopped a single slug.” (Rollins, pp. 40-41.)
“Another natural rights mythologiser is Eric Mack who says, ‘Lockean rights alone provide the moral philosophical barrier against the State’s encroachment upon Society.’ But a ‘moral philosophical barrier’ is merely a metaphorical barrier, and it will no more prevent the State’s encroachment upon ‘Society’ than a moral philosophical shield will stop a physical arrow from piercing your body.
But if natural rights are merely fake or metaphorical rights, what then are real rights? Real rights are those rights actually conferred and enforced by the laws of a State or the customs of a social group.” (Rollins, p. 2.)
What are we to make of this criticism? Well, let’s draw some distinctions.
In ordinary speech we often switch without noticing it between different senses of “rights.” For example, we might say in one breath that citizens in China have no right to free speech — and then say in another breath that Chinese citizens’ right to free speech is being violated. Logically, this seems to make no sense; you can’t violate a right your victims don’t even have. (No one would say, for example, that my right to rule North America is being violated, because nobody thinks I have such a right in the first place.) But our ordinary speech makes more sense once we realise that the term “rights” is being used in more than one sense, so that the kind of right that’s being violated in China is a different sort form the kind of right the Chinese don’t have.
First, we can distinguish between “rights” in the normative sense and “rights” in the descriptive sense. Normative facts are facts about what people ought to do; descriptive facts are facts about what people actually do.
In turn, we can distinguish two sub-varieties of descriptive rights: Legal rights and de facto rights.
This gives us a three-way distinction:
Going back to my China example, when someone switches between saying that the Chinese have no right to free speech and saying that their right to free speech is being violated, he probably means one of the following things:
(I don’t know enough about Chinese law to know whether (1) or (2) is closer to the truth, though (3) would be true in either case.)
Now we can see where Rollins’ critique has gone wrong. Rollins is thinking of natural rights as if they were a special kind of legal right — a right legislated by God or Nature rather than by the state. Given that assumption, what he says makes sense: Legal rights are of little value unless they are also de facto rights. (When Rollins refers to “real rights” as “those rights actually conferred and enforced by the laws of a State or the customs of a social group,” he clearly has in mind de facto rights.) Just as it does me no good to have a legal right on paper that the state pays lip service to in theory but systematically ignores in practice, so it does me no good to have a natural right inscribed in the Law of Nature if no one is willing or able to enforce that right.
But this is the wrong way to think about natural rights. A natural right isn’t a legal right, it’s a normative right. To claim that natural rights don’t protect anything is to miss the point; natural rights are supposed to receive protection, not to provide it. Likewise, the function of Natural Law is not to protect any claims, but rather to tell us which claims deserve protection. As normative concepts, natural rights provide guidance for people’s conduct. Blaming natural rights for not protecting us is like blaming a cookbook for not making dinner. Cookbooks don’t make dinner for us; their purpose is to teach us how to make dinner for ourselves. Likewise, Natural Law doesn’t lead our lives for us; its purpose is to guide us in the living of our own lives.3
So if natural rights don’t protect us, that’s no indictment of Natural Law theory. In fact, however — even though this is not their essential function — natural rights can and do sometimes provide people with de facto protection. In discussing the Holocaust, Rollins takes it as obvious that the Jews’ natural rights didn’t save any of them. But is this true? All over Nazi-occupied Europe, thousands of Jewish lives were saved by brave and committed people who were motivated by their recognition of the Jews’ rights to life and liberty — rights whose authority transcended the dictates of the Nazi state. In fulfilling their primary normative function of guiding the choices of the rescuers, the Jews’ natural rights thereby indirectly did what Rollins says natural rights cannot do — they saved the Jews’ lives.
Now Rollins would no doubt respond that these Jews were saved not by natural rights but by their rescuers’ belief in natural rights. Well, suppose I’m walking along absent-mindedly and I’m about to step inadvertently into a pit of deadly scorpions, when Rollins suddenly shouts “watch out!” I hear his warning shout, and stop just in time. Now if I said that his warning shout had saved my life, would Rollins object that this is wrong, that it’s only my perception of a warning shout that saved my life? In such a case this would be an idle quibble, because although my salvation was caused by my perception of the warning shout, that perception of the warning shout was in turn caused by the warning shout itself; so either one can be credited as causally responsible for my escaping the scorpions.
But Rollins would presumably insist that the Holocaust rescuer case is different, because although the Jews’ salvation was caused by the rescuers’ belief in natural rights, the rescuers’ belief in natural rights was not caused by natural rights themselves. Here I must disagree, though; I don’t see why the rescuers’ belief in natural rights couldn’t be the result of their having correctly recognised and identified the fact of the Jews’ natural rights, just as my avoiding the scorpion pit was the result of my having correctly recognised and identified the fact of Rollins’ warning shout.
The only answer Rollins can give is that the rescuers can’t have recognised and identified the fact of natural rights because there is no such fact; but in that case Rollins’ argument for the uselessness of natural rights begs the question against his opponents by presupposing that natural rights don’t exist. (After all, it’s easy enough to prove something useless if you first presuppose that it doesn’t exist!)
A variation on the natural-rights-don’t-protect argument is Rich Hammer’s article “Might Makes Right: An Observation and a Tool” (Formulations, Vol. III, No. 1 (Autumn 1995)). Rich argues that the rights we have are the ones we are able to secure by force:
“As we humans live, we constantly propose new rights and test old rights. What determines which rights survive this continual struggle? Force. Those rights that survive are those backed up by the greatest force — by which I mean both ability and willingness to police . . . In the long run, the amount of force which people can bring to bear to defend any right depends upon how much that right helps those people survive in their environment. This limits the extent to which humans can invent rights to serve their whims.” (“Might Makes Right,” p. 14.)
When I read a passage like this, my first question is whether the rights being talked about are normative rights, legal rights, or de facto rights. If Rich is talking about de facto rights only, then I don’t think I have any disagreement with what Rich says, at least if “force” is defined broadly enough (e.g., does the ability to motivate people through persuasive argument to respect certain rights count as effective policing of those rights?).
Most of the arguments Rich goes on to give do seem to be intended to apply specifically to de facto rights (and also, to some extent, to legal rights). For example, Rich offers the following challenge to his readers:
“Here I ask you to refute the thesis with a counterexample. If the thesis is wrong, then you can show me an example of a right which has survived even though a contrary claim was supported by greater willingness and ability to use force.” (“Might Makes Right,” p. 15.)
This request for a counterexample seems to presuppose that it is descriptive rights that are in question. For of course no natural-rights libertarian will have any trouble coming up with examples of normative rights that we still retain despite the ability to back them up by force — because normative rights are rights that we retain even when we’re deprived of whatever it is the rights are rights to. Indeed, that’s the crucial difference between normative rights and de facto rights. If you steal my jacket, I lose my de facto right to my jacket, because my claim to that jacket has not been respected; but I retain my normative right to the jacket, so long as it remains true that my claim to the jacket ought to be respected. So when Rich assumes it will be tough to come up with examples of rights we retain despite a contrary claim’s being supported by greater force, it seems he must be thinking about de facto rights, not normative rights.
But this can’t be the whole story. For Rich thinks his position is going to be a controversial and unwelcome one:
“I fear that the thesis, which can be paraphrased ‘might makes right,’ will upset some fellow libertarians who believe that rights come from other sources.
Let me make it clear that I am not saying that I want might to make right. In many instances this thesis runs contrary to the values by which I live. But I observe that the thesis makes sense, like it or not.” (“Might Makes Right,” p. 14.)
Who are these libertarians who are going to be upset by Rich’s thesis? Libertarians who believe that de facto rights come from sources other than superior might? I’m not sure there are any such libertarians. And I strongly suspect that Rich instead has in mind libertarians who believe in normative rights that they regard as deriving from sources other than superior might — sources like Natural Law, for example. But why would such libertarians be upset by Rich’s thesis, if it is only a thesis about the source of de facto rights? Clearly, Rich thinks his thesis has implications — unfavourable implications — for the theories of natural-rights libertarians. And that means that Rich’s thesis is more than just a thesis about de facto rights; it has something to say about normative rights as well.
But what? There seem to be two salient possibilities. Either a) Rich is saying that superior might is the source not only of de facto rights but also of normative rights; or else b) Rich is saying that there are no normative rights, that de facto rights are the only rights there are.
I’m inclined to doubt that (a) is what Rich means. If (a) were Rich’s thesis, then he would be committed to endorsing and approving of whatever de facto rights actually end up getting favoured by superior might. Now, to be sure, Rich does argue that the results of superior might will generally tend to be beneficent, at least in the long run; but he also says that there are cases in which the might-makes-right thesis “runs counter to the values by which I live”; by saying this, Rich seems to be denying that in every case he will automatically regard as valuable whatever set of arrangements wins out.4
Thus I think the most likely interpretation is (b): Rich believes that there simply are no rights over and above de facto rights — that his occasional aversion to the results of force is simply a matter of personal preference.
This interpretation is reinforced by the following passage:
“. . . be on the lookout for the distinction between rights which are merely claimed and rights which are backed by force.
Through tricks of language, wishes often advance in status to rights. But one point of my writing this paper is to help us see the difference between wishes and rights . . . In the country in which I live, most members of the population seem to believe that they have a right to share in the fruits of other people’s labour, just so long as that sharing is passed by the legislature. And . . . they do in fact have that right, since it is backed with willingness and ability to prevail in use of force. Of course I favour the alternate claim, to keep all the fruits of my own labour, but this claim diminishes to the status of a wish; it lacks force.” (“Might Makes Right,” p. 14.)
If I understand him properly, Rich seems to be saying that any claim that is not backed up by sufficient force — that is, any claim that fails to be a de facto right — is nothing more than a wish, a subjective preference. And that in turn would seem to mean that there are no objective grounds for preferring one claim over another, no such things as claims that ought to be respected and claims that ought to be denied. In other words, there are no normative rights, and there is no Natural Law.
Now of course this sort of moral skepticism might be true. But I can’t see that Rich’s article gives us reason to think it is true. I find quite convincing Rich’s arguments for the claim that de facto rights are made by might; but those arguments do not seem to rule out the possibility of normative rights that do not depend on might for their validity.
Another objection that’s a bit harder to get a handle on is the complaint that there’s something spooky and mysterious about Natural Law and natural rights. In his pamphlet Natural Law: Or Don’t Put a Rubber on Your Willy (Port Townsend: Loompanics, 1987), libertarian science-fiction writer Robert Anton Wilson characterises natural rights theory as the view that “some sort of metaphysical entity called a ‘right’ resides in a human being like a ‘ghost’ residing in a haunted house.” (p. 4.) The implication is that natural rights are too weird to be believable.
Like Rollins, Wilson seems to want to treat natural rights as if they are supposed to be descriptive facts. But natural rights are normative facts. To say that Wilson has a natural right to be treated in manner X is not to say that there’s some kind of invisible sprite lurking somewhere inside Wilson’s body. Rather, what it means is this:
In other words, natural rights consist in facts about what people ought to do, how people ought to treat each other. Now people do sometimes talk as though (1) and (2) are true because Wilson has a right to be treated in manner X, as though the right were something separate, over and above facts (1) and (2). But this strikes me as no more than a figure of speech. Strictly speaking, there is nothing more to Wilson’s having a right to be treated in manner X than the existence of the moral obligations and moral permissions listed in (1) and (2). So natural rights can be dismissed as objectionably spooky and mysterious only if the entire notion of an objective morality — of there being facts about what people ought and ought not to do — is also dismissed as objectionably spooky and mysterious.
This is indeed the position Rollins takes:
“If you want someone to do something which he has no personal reason for doing, but you are unable or unwilling (perhaps afraid) to use real coercion to get him to do it, then you can try to get him to do it by means of metaphorical or fake coercion. You can tell him it’s his duty to do it. You can tell him he ‘must’ do it. Why? Simply because he must. And if he is gullible enough to believe that he must do as you tell him, simply because he must, then you control that two-legged sheep by means of the metaphorical or fake coercion of duty . . . Morality . . . is a myth invented to promote the interests / desires / purposes of the inventors. Morality is a device for controlling the gullible with words. ‘You “must not” commit murder!’ Why not? ‘Because murder is “wrong!” Murder is “immoral!”’ Bunk! Murder may be impractical or excessively risky or just not worth the trouble. There are all sorts of reasons why I might refrain from committing murder even when I would like to do it. But murder is not ‘wrong.’ Murder is not ‘immoral.’ And the same goes for rape, robbery, assault, battery, burglary, buggery, bestiality, incest, treason, torturing children, suicide, cannibalism, cannabisism, etc . . . Nothing is sacred. Nothing is ‘entitled to reverence.’ Nothing is ‘inviolable.’” (Rollins, p. 8-9, 19.)
Rollins is more willing than most critics of Natural Law to face the logical consequences of his position. But if morality is merely a tool for manipulating other people into doing what one wants, one wonders why people ever wrestle privately with moral dilemmas, why they ever find themselves compelled by conscience to do something that is unwelcome not only to themselves but to those around them.
More importantly, though, the question is why we should accept Rollins’ claim that nothing is right or wrong and nothing is entitled to reverence. These are extraordinary claims, claims that run contrary to our ordinary beliefs and practices, and so the burden of proof rests with the person making such claims.
Natural Law theorists may not have the burden of proof; but it’s still a fair question to ask what kind of facts normative facts could be, what basis in reality they could have. This is a question to which different Natural Law theorists provide different answers. In my philosophical work I’m attempting to develop an answer of my own; my position is not fully worked out yet, but what follows is a thumbnail sketch of the kind of approach I find most plausible:
On the other hand, if we insist on renouncing violence even when peaceful relations are not an option — that is, if we refuse to defend ourselves from aggression — then we are declining to extend our lives even when we could do so without decreasing the humanity of our lives. Thus, while human beings are under an obligation to respect one another’s autonomy, they are not under any obligation to refrain from forcibly defending their own autonomy.7 (Indeed, they may even be obligated to defend themselves, since we have other ends (such as self-preservation) which become imperative for us when they do not conflict with higher goals.) But this means that every human has an obligation to refrain from invading the freedom of every other human, and that it is permissible for the latter to defend this freedom by force against incursions from the former. In other words, every human being has a right to freedom — a natural right, one that derives from the Natural Law specifying our natural ends.
I do not expect the nine steps I’ve just set down to persuade anyone; what I’ve just offered is not an argument but an outline for an argument, and each step would have to be filled in with a lot more detail and backed up by further arguments in order to be convincing. Indeed, this project is one I’m pursuing in my own philosophical research. The point of setting down these nine steps here is simply to show what kind of metaphysical basis I think can be given for Natural Law (and in particular to show that no supernatural basis is required).
In addition, I should stress that it is not primarily on the basis of the nine steps I’ve just outlined that I believe in libertarian natural rights. I feel a good deal more certain of the existence of libertarian natural rights than I do of my ability to ground this nine-step argument. The purpose of such an argument, if it can be made to work, is to explain why we have the rights we do, not to justify our belief in them (though on the other hand, the process of working through and developing such an argument naturally induces modifications in the details of the natural-rights claims that I think are justified).
In earlier centuries, Natural Law theorists drew a useful distinction between Natural Law’s principium essendi and its principium cognoscendi. The principium essendi of X is the basis for X’s being so; the principium cognoscendi of X is the basis for knowing that X is so.
For example, sandalwood has a distinctive smell by which it can be identified; so that smell is a principium cognoscendi of sandalwood. But that smell is not what makes sandalwood what it is; it is not sandalwood’s principium essendi. The principium essendi of sandalwood is presumably its biochemical microstructure; but the presence of the distinctive sandalwood smell is a reliable indicator of the presence of that biochemical microstructure. A principium essendi can also serve as a principium cognoscendi; that is, we can identify sandalwood by its biochemical microstructure as well as by its smell. But not every principium cognoscendi is also a principium essendi.
The purpose of the research program described in my nine-step outline is to discover the principium essendi of Natural Law and natural rights. But I do not think the success of such a program is required in order for us to be able to say what natural rights we have or what Natural Law requires of us. As we shall shortly see, there are many epistemic avenues to moral truth; the principium essendi of Natural Law, whatever it turns out to be, is only one of many principia cognoscendi.
One of the most common objections to Natural Law is that it is not open to scientific test. Wilson, for example, insists that he is open-minded and willing to accept Natural Law if only it can be provided with a scientific basis:
“. . . all I am asking is that somebody should [produce] a shred or a hint of an adumbration of a shadow of a ghost of something like scientific or experimental evidence in place of the metaphysical, and meaningless, verbalisms Natural Law cultists habitually use. Until they produce some such sensory-sensual space-time evidence, I still say: Not proven.” (Wilson, p. 37.)
But Wilson is not terribly optimistic about the prospects for finding such a scientific basis; by its very nature as a moral code, Natural Law “is not subject to experiment; experiment, and refutation by experiment, are simply not relevant to it.” (p. 14.) Noting that the Ayatollah Khomeini thinks Natural Law authorises divorce in some cases while the Pope does not, Wilson remarks:
“I still don’t have a clue as to a scientific test to determine which of these vehement and dogmatic old men might actually know what Natural Law is, or how to be sure they aren’t just calling their own prejudices Natural Law . . . there is no experiential-experimental way to judge among any of them . . .” (pp. 35-37.)
In other words, Wilson’s objection is that normative statements are not testable, and so there are no rational grounds for deciding whether they are true or false.
But this is a mistake. Normative statements — moral judgments — are as open to being tested as any other kind of judgment. For normative statements entail empirical statements, and if the empirical statements in question are falsified by sensory observation, then the normative statements that entail them are likewise falsified.8
Consider the following two normative statements: “Adolf Eichmann is a virtuous person” and “A virtuous person would never participate in genocide.” These two normative statements, taken together, entail an empirical statement, namely, “Adolf Eichmann did not participate in genocide.” This is a statement that is open to empirical test; the clearest falsification would be one’s own eyewitness observation of Eichmann participating in genocide, but barring that, we can still have convincing evidence that Eichmann did indeed participate in genocide. And once the empirical conclusion has been falsified in this way, we can infer that at least one of our normative premises must be wrong. (From “If P & Q, then R” and “Not R,” the inference “Not both P & Q” logically follows.) So the results of empirical investigation can indeed require us to revise our moral beliefs; in short, normative statements are indeed testable.
Now it might be objected that all this test shows is that at least one of our normative premises must be wrong, but it doesn’t tell us which normative premises to reject.
This is true. But the same criticism applies equally well to any application of the scientific method. Suppose I want to test the proposition that water boils at 100° C. So I heat some water, and when it starts to boil I stick in a sturdy thermometer and see what reading I get. Now suppose the thermometer reads 96° C. What should I conclude? Well, I could regard the assumption that water boils at 100° C as having been disproven. But this is not my only option. It is also open to me to hold on to that assumption and instead reject some auxiliary assumption — e.g., my assumption that this stuff is really water, or my assumption that the thermometer has been labeled correctly, or even my assumption that I am awake rather than dreaming.
One can never test any belief in isolation; one can only test groups of beliefs. In natural science as in ethics, empirical tests can expose an inconsistency in one’s total belief-set, but they cannot tell one which belief(s) to reject. How we resolve the inconsistency will depend on which beliefs we find most plausible, how committed we are to them, how many of our other beliefs depend on them, and so forth. In this regard, ethics is no worse off than natural science.
The implication I would want to draw from this is “so much the better for ethics!” But some will instead want to conclude: “So much the worse for natural science!” If no belief can ever be tested in isolation — if all our conclusions, in science as well as in ethics, depend on personal and inevitably impressionistic judgments of relative plausibility — then isn’t it impossible for any belief to be justified? Instead of upgrading our assessment of moral reasoning to place it on a par with the objectivity of scientific reasoning, why shouldn’t we downgrade our assessment of scientific reasoning to place it on a par with the subjectivity of moral reasoning?
Well, one reason not to do so is that this would amount to the kind of global skepticism that we’ve already seen to be self-defeating. If the skeptic wants to claim that the standard scientific method does not yield justified beliefs, then the skeptic has set the standards of justification so high that it is very hard to see how the skeptical thesis itself could meet those standards. And if it cannot, then the skeptic has given us no reason to accept his claim that the standards should be set so high. We do not have to build our system of beliefs on a bedrock foundation of self-evident truth before we are justified in accepting those beliefs as provisional starting-points. Our current beliefs deserve the benefit of the doubt until we find some positive reason to suspect them; we have to start where we are, not somewhere else. The structure of a belief-set is not hierarchical, like a skyscraper with each floor resting on the floor below it, all the way down to the ground; it’s more like a spiderweb, a network of interrelated, mutually supporting judgments varying in strength and centrality. In epistemology (the theory of knowledge), this spiderweb model of justification is known as coherentism, while its skyscraper rival is called foundationalism.
Thus far I’ve been talking about testing normative beliefs by seeing whether they conflict with empirical observations. But if coherentism is correct, we can also test normative beliefs by seeing whether they conflict with each other. And we can even test descriptive beliefs by seeing whether they conflict with normative beliefs. On the skyscraper model, higher-level beliefs can be revised in the light of changes in lower-level beliefs, but never vice versa; the arrow of justification points in one direction only. But according to coherentism, any belief is in principle open to revision if it clashes with a sufficient number of other beliefs, of whatever kind. Which beliefs we should keep and which ones we should toss out will depend on how central the beliefs in question are to our overall picture of the world.
Most people, for example, have a lot more invested in the judgment that genocide is immoral than they do in any particular view about the status of normative judgments; so if someone like Rollins comes up with a theory about the status of normative judgments that implies that genocide is not immoral after all, the rational response is to hold on to one’s condemnation of genocide and reject Rollins’ theory — unless Rollins can show that his theory rests on judgments that are more central to our belief structure than our belief that genocide is immoral. There is no fundamental difference between moral reasoning and the experimental method of natural science; both involve what Plato and Aristotle call dialectic, or what John Rawls calls the method of reflective equilibrium: Tracing the implications of our beliefs and attempting to eliminate inconsistencies among them. (And performing experiments is simply a way of adding new beliefs to our total belief-set — and using those new beliefs to test the old ones.)
At this point the following objection might be raised: In the case of a disagreement between two different descriptive theories, there is a possibility of resolving the dispute by performing experiments. Perhaps, as the coherentist claims, experiments are just a way of acquiring new beliefs, but at least they cause the disputants to acquire the same new beliefs, thus bringing the two belief-sets into greater alignment. But there seems to be no analogous way to resolve disputes over different interpretations of Natural Law. For example, Stephen O’Keefe writes in his preface to Rollins’ book:
“Any superficial political conversation with libertarians will sooner or later touch on the subject of rights . . . If the discussion gets serious, though, libertarians find their solid ground of rights quickly disintegrating into quicksand. One deadly question usually puts an end to the rights nonsense: Why is the communist claim that people have a right to live off the labour of others less valid than libertarian rights? The libertarian must then find an authority behind his or her authority, and there is no rational place to turn. They can appeal to God, or nature, or human nature as ordaining their brand of rights, but the communists can do the same.”
But I think this difference between scientific method and moral reasoning is overstated. Even people who differ violently on various moral issues generally have many more beliefs (both descriptive and normative) in common, and it is often possible to mine the vast area of common agreement for premises with which to resolve the disputes. Thought-experiments play a role in moral reasoning similar to the role that actual experiments (and thought-experiments too, for that matter) play in natural science.
For example, someone who believes that we should always do whatever maximises social utility may have second thoughts when asked to imagine a case in which a doctor secretly kills a healthy patient in order to redistribute the patient’s organs to five sick patients who will die unless they receive organ transplants as soon as possible. If we agree that the doctor’s action maximises social utility, but we nevertheless find ourselves inclined to evaluate the doctor’s action as wrong, then the thought-experiment has resulted in new beliefs that conflict with our older belief that whatever maximises social utility is okay. Thus moral thought-experiments can also serve the function of bringing divergent belief-systems into alignment.
An important function of new data — whether acquired through sensory experience or through philosophical reflection — is to introduce inconsistency into a previously consistent belief-set, thus prompting a revision in belief.
Of course, someone might choose to reject the new data rather than revise old beliefs; and sometimes (e.g., in the case of hallucinations and the like) this can even be the rational option. Once again, what we accept and what we reject will depend on the number of beliefs at issue and the weight or plausibility we assign to those beliefs. So the attempt to resolve inconsistencies among one’s beliefs may not necessarily bring one’s belief-set into greater consilience with those of others.
In the moral case, for example, Rollins, a self-proclaimed “amoralist,” chooses to hold on to what most would view as a highly implausible belief — the belief that there is nothing wrong with “murder, rape, robbery, or torturing children” — and to reject more plausible beliefs whenever they come into conflict with that one. But this is no proof that moral reasoning is useless in reaching agreement, because the same phenomenon can show up in natural science — as in the case of creationists who cling so stubbornly to the belief that the universe is only a few thousand years old that they reject countervailing evidence (whether astronomical, geological, or paleontological) as fake clues planted by God to test our faith.
In ethics as in natural science, dialectic is a powerful tool for reaching agreement, but in neither case does it offer any guarantee of convincing people like amoralists and creationists, who, when confronted with inconsistencies in their belief-set, insist on resolving these by keeping the less plausible beliefs and rejecting the more plausible ones. (Of course both the amoralist and the creationist will protest that the characterisation I’ve just given of their positions depends on my personal perspective as to what is or is not plausible. Well, sure. My personal perspective is the only place I have to stand.)
Wilson (in Natural Law) is skeptical about the degree of similarity between ethical disagreement and scientific disagreement:
“The suspicion that what is called ‘Natural Law’ may consist of personal prejudice with an inflated metaphysical label pinned on it grows more insidious as one contemplates the fantastic amount of disagreement about virtually everything among the various advocates of ‘Natural Law.’
Rothbard tells us that this means nothing, because there are disagreements among physicists, too: But I find this analogy totally unconvincing . . . In the area of Natural Law and metaphysical ‘morality’ in general, there is no shred of . . . agreement about how to ask meaningful questions (questions that can be experimentally or experientially answered)9 or even about what form a meaningful (answerable) question would have to take. There is no pragmatic agreement about how to get the results you want. There is no agreement about what models contain information and what models contain only empty verbalism. There is, above all, no agreement about what can be known specifically and what can only be guessed at or left unanswered . . .
Some states and nations believe in capital punishment; others do not. Pacifists are against killing anybody, but not all pacifists are vegetarians. Some quasi-vegetarians will not eat the higher mammals but will eat fish. Pure vegetarians kill vegetables to eat. And so on . . .
To compare this ontological spaghetti with the highly technical disagreements in physics seems to me like comparing ten drunks smashing each other in a saloon with the difference in tempo and mood between ten conductors of a Beethoven symphony.” (Wilson, pp. 33-36.)
It is probably true that there is more disagreement in the natural sciences than there is in ethics.10 But are the natural sciences the best comparison class? Ethics surely has more in common with the social sciences than with the natural sciences; and in the social sciences — e.g., economics, sociology, psychology — the extent of disagreement is notorious. Consider the differences between, say, the Keynesian, monetarist, econometric, public-choice, Marxist, Georgist, Austrian, and mainstream neoclassical approaches to economic theory. Here we find not only a torrent of disagreements about specific policy issues such as whether or not a given policy will or will not increase inflation, unemployment, economic growth, etc., but also precious little agreement about “how to ask meaningful questions,” or “what form a meaningful (answerable) question would have to take,” or “what models contain information and what models contain only empty verbalism,” or “what can be known specifically and what can only be guessed at or left unanswered.”
Should economic method be inductive or a priori? Should it aim at prediction or at explanation? Should it employ a subjective or an objective conception of economic value? How useful are mathematical models when applied to human behaviour? How many simplifying assumptions can we make about the motivations of economic agents before our models cease to be useful in elucidating social reality?
These are questions on which the economic field is not even close to reaching a consensus. Yet, as a libertarian, Wilson would probably be unwilling to conclude that all economic theories are equally valid and that none is better grounded than any of its rivals, or that there is no fact of the matter as to whether a given policy would cause a rise or a drop in unemployment. I would bet that despite the lack of consensus among economists, Wilson probably believes in some kind of economic truth.11 So why should an equivalent level of disagreement in ethics make us skeptical about the possibility of ethical truth?
There’s no great mystery about why agreement is harder to reach in ethics and the social sciences than it is in the natural sciences. For one thing, the subject matter (human activity) is both more complex and less susceptible to mathematical analysis, thus making theoretical modelling and controlled experimentation inherently more difficult. For another, researchers are likely to bring more prejudice, self-interest, and ideological baggage with them to issues in ethics and social science than to issues in natural science, thus making the problem of bias more pervasive. It is complexity and bias, not inherent subjectivity, that make moral disagreement so intractable.
I’ve been arguing that normative beliefs can be justified. Now someone might grant this, but still deny that our moral beliefs can count as knowledge. At one time it was fashionable in philosophical circles to define knowledge as justified true belief, but nowadays philosophers recognise that a belief can be both true and justified, and yet be such that few would be willing to call it knowledge.
The paradigm case is when a justified true belief is based on a justified false belief. Suppose I believe that alligators are mammals. Suppose further that I have good reasons for my mistaken belief; the encyclopaedia I looked in contained a misprint, the biologist I consulted lied to me, and so on. So I’m justified in believing, falsely, that alligators are mammals. Since I know that all mammals are vertebrates, I’m justified in concluding, on the basis of my false belief that alligators are mammals, that alligators are also vertebrates. Now it just so happens that alligators actually are vertebrates, although my reasons for believing this truth are mistaken. So I have a justified true belief that alligators are vertebrates, but most people would be reluctant to say that I know that alligators are vertebrates, and the source of their reluctance is the fact that the connection between the belief’s being true and my being justified in believing it seems so chancy and accidental. Hence most philosophers conclude that some sort of reliability condition, showing how our beliefs track truth, must be added in order for justified true belief to count as knowledge.
It seems to follow that even if a) I believe that people have a right to freedom, and b) my belief is true, and c) I am justified in holding it, I don’t count as knowing that people have a right to freedom unless I believe this because it is true. But, the objection runs, we can causally interact only with descriptive facts, not with normative facts; therefore, normative beliefs can never satisfy the reliability condition, and so can never count as knowledge.
Briefly, my response to this objection is threefold:
While I’m on the subject, I think the coherentist approach to moral argument that I’ve been defending can shed some light on a topic of common discussion among libertarians — namely, whether libertarianism should be based a) on the consequentialist argument that we should allow people to be free because doing so will have beneficial social consequences, or instead b) on the deontological argument that we should allow people to be free because doing so is mandated by our moral obligation to respect other people as ends in themselves.13 (Generally it is only the deontological libertarians who employ the language of Natural Law, but historically there have been both consequentialist and deontological versions of Natural Law theory; if you believe in a higher moral standard, independent of convention but accessible to reason, to which manmade laws are properly answerable, then you are a believer in Natural Law, even if your higher moral standard is simply social welfare.)
Sometimes all the dispute between consequentialist and deontological libertarians amounts to is simply a debate over the best way to present libertarianism when trying to convince non-libertarians. In that case I think the debate is a somewhat silly one; for reasons I’ll soon get into, most people will be unwilling to accept as socially beneficial a system they think is unjust, and vice versa, so neither the consequentialist nor the deontological argument can stand very well alone. And in any case, since there are plenty of good consequentialist arguments for libertarianism and plenty of good deontological ones, why not use all the ammunition in our arsenal?
But more often the disagreement is not about how to package libertarianism when selling it to the infidels, but rather about which set of reasons — the consequentialist or the deontological ones — constitutes the deepest truth about why libertarianism is the right system. For example, deontological libertarians often say that although it’s a lucky break for us that libertarianism is socially beneficial, we would still be obligated to respect libertarian rights even if it turned out that doing so would lead to social chaos and misery; and consequentialist libertarians make similar remarks on the other side. In other words, each side of this debate is officially committed to the view that the other side’s reasons are irrelevant to the justification of libertarianism.
Yet, interestingly enough, although deontological libertarians don’t think it matters that libertarianism is socially beneficial, they all seem to think that in fact it is beneficial. And likewise, although consequentialist libertarians don’t think it matters that libertarianism expresses respect for persons, they all seem to think that in fact it does express respect for persons.
If deontological libertarians were to become convinced that libertarian policies would actually cause social chaos and misery, I suspect that most of them would find their faith in libertarianism shaken. Consequentialist libertarians, recognising this, often accuse the deontologists of hypocrisy, claiming that under their deontological veneer they are crypto-consequentialists. (I recall reading a lengthy debate on this topic in Liberty magazine during its first few issues.) But this accusation is a two-edged sword, since if consequentialist libertarians were to become convinced that libertarian policies in fact express contempt for persons, I imagine their faith would be shaken too.
So what’s going on here? Well, suppose I believe that water is H2O. Then that belief commits me to thinking that if there were no such thing as H2O, there would also be no such thing as water (since they’re the same thing). However, if I were to become convinced that the atomic theory of matter is wrong — if I were to come to believe that there are no hydrogen and oxygen atoms, and thus no H2O — I would not conclude that there is no water. Instead, I would revise my belief that water is H2O.
I have a particular theory about what the principium essendi of water is; I think it’s H2O. And that commits me to the belief “If there were no H2O, there would be no water.” But that statement does not commit me to the belief “If I didn’t believe in H2O, I wouldn’t believe in water.” H2O is not my primary principium cognoscendi of water; I ordinarily identify water by its appearance, potability, boiling and freezing points, etc., not by its molecular composition. So if I were to learn that H2O is nonexistent, but my ordinary principia cognoscendi still indicated the presence of water, the most plausible way of resolving the inconsistency would be to reject my theory about what water’s principium essendi is, rather than giving up my belief in the existence of water.
The same point applies to the dispute over the basis of libertarianism. The disagreement is about the principium essendi of libertarianism’s validity; consequentialist libertarians think the principium essendi is social welfare, while the deontological libertarians think it’s respect for persons. However, libertarians, like most people (myself included), tend to think that social welfare and respect for persons go together, at least roughly; that is, they think that a system that respect persons is likely to be socially beneficial, and vice versa, so that each trait can serve as a reliable (though not exceptionless) indicator of the other’s presence. Given that belief, those who regard social welfare as the principium essendi of rightness will tend to treat respect for persons as at least a principium cognoscendi of rightness, just as those who regard respect for persons as the principium essendi of rightness will tend to treat social welfare as a principium cognoscendi.
The debate about whether social welfare or respect for persons is the principium essendi of libertarianism’s validity is an important one (and it’s no secret that I’m in the respect camp); but I think its participants have sometimes misconstrued what their positions commit them to. Recall the H2O case. Those who believe that respect for persons is libertarianism’s principium essendi are indeed committed to the belief “If libertarianism were not socially beneficial, it would still be morally obligatory.” But many of them have made the mistake, as I see it, of thinking that this belief commits them to the further belief “If we ceased to believe that libertarianism is socially beneficial, we would still regard it as morally obligatory.” (And likewise, mutatis mutandis, for the consequentialists.) This further belief is rarely true, nor should it be; both consequentialist and deontological considerations are crucial for the justification of libertarianism, even if one is more fundamental than the other when it comes to explaining why libertarianism is the correct position.
A final objection I want to consider is that Natural Law is an unnecessary hypothesis, because moral evaluations can be explained as a product of evolution, rather than as a response to objective moral truth.14 In a recent article, Rich Hammer writes:
“If beauty is beauty, you might think that [a cockroach] and I would fight over the same lady . . . [But we] are each programmed to seek females with whom our genes might, well, carry on . . . [Because we recognise this] we do not fall into bitter dispute because we disagree about which lady is more beautiful . . . But we do get into a tiff sometimes when our other senses, especially our senses of right, recommend different rules of conduct . . . Maybe this sense of right, which causes me to form opinions about how I should regulate my actions in order to consider the needs of others, has been programmed into me, just like my sense of beauty. Maybe my genes have figured out that they have a better chance of surviving if human individuals are programmed to hunger for rules of behaviour which favor cooperation over conflict.” (“The Sense of Right and a Man-to-Man Talk With Archy About Women,” Formulations, Vol. IV, No. 1 (Autumn 1996), p. 37.)
But I have some questions about this analogy. In Rich’s story, it’s no surprise that human and cockroaches disagree in their judgments of beauty, because biological needs have programmed them to have different responses — and so we should take a similar attitude toward moral disagreements. This last inference, about morality, is what puzzles me.15 Moral disagreements don’t occur between humans and cockroaches; they occur among humans — members of the same species, products of the same evolutionary process. So an explanation of our moral judgments that appeals only to evolutionary considerations is necessarily going to be incomplete.
So evolutionary explanations of moral disagreement seem unpromising. Evolutionary explanations of moral agreement are on firmer ground. But even here there is room for skepticism. It’s often thought that if the Darwinian theory of evolution through natural selection is correct, then any central or important feature of human beings must have an evolutionary function. But this isn’t true. Consider the ability to solve mathematical equations. This is an important and valuable skill, and arguably has survival value; but was it selected for because of its survival value? I doubt it. Evolutionary pressure did select for something, but what it selected for was reason — i.e., a generic capacity for figuring things out — and our more specialised capacity to solve mathematical equations is a byproduct of that more generic capacity, rather than something that was selected for directly.
So if human beings generally have a tendency to assent, upon reflection, to the proposition that 374 times 98 equals 36652, that’s not because the belief that 374 times 98 equals 36652 has any particular survival value; rather, it’s because we have a generic capacity to figure things out (a capacity that does have survival value), and when we apply that capacity to the problem of what 374 times 98 equals, we come up with 36652 because we are able to figure out that 36652 is the actual right answer.
Likewise, then, it is possible that our capacity for moral reasoning, like our capacity for mathematical reasoning, is the byproduct of our general ability to reason, rather than something for which natural selection is directly responsible. In other words, if people have a tendency to hold certain normative beliefs, it might be because they have used their rational capacities to figure out that certain things are right and others wrong.
Now I certainly don’t mean to deny that evolutionary considerations of the sort Robert Axelrod appeals to in his book The Evolution of Cooperation play an important role in explaining why we tend to favour “rules of behaviour which favour cooperation over conflict.” I wholeheartedly endorse this basic point. But these basic cooperative impulses are not specific enough, by themselves, to ground the full spectrum of our normative attitudes.
Consider the following pattern of moral reasoning:
My present concern is not with whether this is a good or bad argument. The point is that it is a typical, and easily understandable, mode of reasoning. Even those who disagree with the argument can easily see the point of it.
Now suppose that we have a natural tendency to believe (1), and that this tendency was selected for by evolution, because creatures who kill their own kind have a harder time building cooperative networks and so are disadvantaged in the struggle for survival.
Suppose also, on the other hand, that we have no particular tendency to believe (3), and that the absence of such a tendency is also the product of evolution, because before the development of agriculture, people who were squeamish about eating animals tended to die out before they had a chance to reproduce and pass on their genes.16
We can assume, then, that our early ancestors had no qualms about eating animals, and did not feel any tension between their acceptance of (1) and their rejection of (3). But the exercise of reason can prompt people to notice the tension, and to resolve it by embracing (3). (I am not saying that this is the only way to resolve the tension, only that it is one salient and intelligible way.) This is one of the modes through which people come by their moral beliefs, and it is a mode to which evolutionary considerations are only peripherally relevant.
We may think of our evolutionarily-implanted normative impulses as playing a role in moral reasoning analogous to the role that sensory experience plays in scientific reasoning. The data of the senses are one of the most important sources of our beliefs about how the universe works. But we are not confined to the sensory level. Our capacity for reason drives us to try to build up a conceptual picture of the universe that makes sense; and although we rely heavily on sensory data in building that picture, if we have to sacrifice some sensory data in order to achieve a scientific picture that makes a little more sense — if we have to decide that, despite initial appearances, the earth isn’t flat, the sun doesn’t circle it, and tables aren’t continuously solid all the way down — then some of what the senses tell us may have to be scrapped for the sake of a more intellectually satisfying theory.
Likewise, our evolutionarily-implanted moral impulses are one of the most important sources of our beliefs about how we ought to live. But we are not confined to the instinctual level. Our capacity for reason drives us to try to build up a conceptual picture of right and wrong that makes sense; and although we rely heavily on innate impulses in building that picture, if we have to disregard some of our innate impulses in order to achieve a moral picture that makes a little more sense — if we have to decide that, despite our initial impulses, we shouldn’t kill animals for food — then some of what our moral instincts tell us may have to be scrapped for the sake of a more intellectually satisfying ethic. Once again, a purely evolutionary account of our sense of morality, however illuminating, will be importantly incomplete.
In any case, the ratio of learned behaviour to instinctual behaviour is higher in humans than in any other known organism. So it’s not surprising that many defenders of the evolutionary objection to Natural Law have chosen to focus on cultural evolution rather than natural evolution. As this version of the objection has it, our moral attitudes are by and large the result not of natural selection acting on species, but of natural selection acting on ways of doing things. Cultural practices that promote their society’s survival tend to survive themselves, both because the society where they originated survives and keeps those practices, and because other societies notice their success and start imitating them. Harmful social practices, by contrast, tend to undermine a society’s chances of survival; the society is more likely to perish, and other societies are more likely to avoid the practice because failed societies have less prestige and so attract fewer imitators. Thus the harmful practice dies out.
I think there is a core of profound truth to this argument. It exemplifies the classical liberal insight — developed in different ways by writers like John Stuart Mill, Michael Polanyi, Friedrich Hayek, and Bruno Leoni — that competition is above all a discovery process. Still, the argument has its limits. To borrow a comparison from David Ramsay Steele:17 It is true that organisms with beneficial parasites are more likely to survive than organisms with harmful parasites, but it would be rash to conclude from this that existing parasites are likely to be beneficial. The fact that a given society has survived is no proof that any particular practice of that society is beneficial.
This caveat applies to any evolutionary approach, whether biological or cultural; but cultural evolution in particular faces special problems as an explanatory factor. In biological evolution, mutations arise slowly and incrementally; no species sprouts wings or antlers overnight. Thus, when we see organisms with wings or antlers we can be sure that these features have developed over many thousands of generations, and so the hypothesis that these features are beneficial, or at least not inimical, is a salient one. But in cultural evolution, mutations — i.e., new ideas and practices, or what Richard Dawkins calls memes — are often the product of human thought and can emerge fully developed in a single generation (examples: Islam, the U. S. Constitution, the paper clip), and so the presence of a meme is very weak evidence that it’s been reliably selected for by evolutionary pressures.
Worse yet, because memes, unlike genes, can reproduce via imitation, a particular meme can spread and survive even if it kills off its host group. The fact that a meme is good at ensuring its own survival is no guarantee that it will be equally effective at ensuring the survival of groups who adopt it.
For example, as the Roman Empire grew more centralised and authoritarian, it so weakened its economic and cultural base that it essentially self-destructed, unable to fend off marauding tribes that in earlier years it could have crushed without blinking. Yet the fall of stagnant, ossified, hierarchical Rome did not put an end to the Roman centralist meme, which continued to attract admirers and imitators over the centuries. Having destroyed its original host, the imperial virus propagated, infecting countless societies from the Byzantine Empire to the Thousand-Year Reich, killing them off in turn.
When we read the 14th-century Italian poet Dante singing the praises of world government in his treatise De Monarchia, looking specifically to Rome as his model, or treating the assassination of Caesar, in his famous Inferno, as a crime comparable in seriousness to the betrayal and crucifixion of Christ,18 we recognise that the staying power of a meme may have little to do with its success in promoting the survival of societies that adopt it. And a glance at our own sprawling reproduction — both architectural19 and political — of ancient Rome in full imperial splendour on the banks of the Potomac bodes ill for the future of the United States.
This distinction between the success of memes and the success of societies that adopt those memes provides one possible response to a worry posed by Rich Hammer that if we make anything other than evolutionary success our normative standard, we run the risk of endangering our own welfare:
“Rights can be viewed as ways to economise, ways to save the cost of battle . . . Rights guide behaviour within a dominant community. Among a group of people who have won, and who are in process of harvesting (or looting), rights limit counterproductive struggle within the group. Rights guide each individual member of the group to seek to satisfy his wants by harvesting from outside the group rather than from another member within the group . . . If you believe the evolutionary account of formation of life, then you may observe that we, presently surviving humans, find ourselves here as the present culmination of a long history of evolutionary struggle. And if you believe my thesis, that there is a competitive survival-of-the-fittest among systems of rights, then you may observe that we, in Western Civilisation, find ourselves here, in a position which seems to dominate other cultures, because we are the beneficiaries of evolutionary struggle and selection of rights . . . As I am presenting it, rights minimise violence and bloodshed among us humans who dominate the ecosystem in which we live. To argue that rights have a different basis argues, I believe, either against our dominance or for more violence and bloodshed.” (“Might Makes Right,” pp. 15-16.)
But of course rules that encourage “harvesting from outside the group,” thus enabling a society to “dominate other cultures,” will successfully “minimise violence and bloodshed” only within the group. Successful societies have a long history of exploiting and even exterminating those who are outside the group; witness the treatment that Africans, Asians, and Native Americans have received at the hands of Western colonialist powers. Rich notes (p. 16) that powerful cultures can afford to be more generous in granting rights to their weaker neighbours. True enough, and sometimes they do. But powerful cultures can also afford to enslave or murder their weaker neighbours without fear of reprisal, and sometimes they do that instead. Civilisation is largely a process of increasing people’s options (advances in technology and advances in political freedom can both be seen in this light); but unfortunately, one of the things one is better able to do once one’s options have increased, is to decrease one’s neighbours’ options.
But, leaving aside the issue of violence toward outsiders, is it at least true that dominant societies manage to minimise violence and coercion within the group? Not necessarily. Once a given society achieves a position of dominance over other cultures, it tends to squelch the competitive process that brought it to power (by conquering the competitors); and once competitive pressure is diminished, the presumption that the dominant society’s practices enjoy the continued blessing of evolutionary selection must inevitable be weakened.
Such a society’s status is rather like that of a business enterprise whose efficiency and innovation earns it success on the free market — but which then uses its newfound resources, the fruits of its competitive success, to lobby government for laws insulating it from competition. Once such laws are passed, the company’s incentives change, and it grows inefficient and lazy because it can now afford to. It would then be a mistake to assume that the company’s continued dominance makes its top-down management structure, unimaginative product design, and lack of responsiveness to customers a useful model for would-be entrepreneurs to imitate.
In short, a society’s dominance does not guarantee, and may even undercut, its efficiency in any particular area, including the minimisation of violence and bloodshed. Indeed, the following pattern is a common one throughout history:
Holding Group B in subjection is an expensive proposition; it requires conscription, tax hikes, and perhaps a military-industrial complex, all burdens that will end up being shouldered by the population of Group A. Keeping an eye on potential troublemakers from Group B requires a system of surveillance and documentation that the rulers of Group A can later use against their own citizens. Those within Group A who criticise the treatment of Group B threaten A’s dominance over B and may find themselves subject to censorship. Free economic transactions between members of A and members of B may result in improvements of B’s economic status that empower it to start resisting A’s authority, so the freedom of A’s members to deal with B’s members will also need to be curtailed. And so on.
In the ancient world, Sparta and Rome provide paradigmatic examples of this dynamic at work. Both began as vigorous, progressive centres of trade and culture, but the need to maintain control over subject populations (the Helots, in the case of Sparta; the Empire, in the case of Rome) turned Sparta into a grim military collective and Rome into a bureaucratic, dictatorial police state.
But there are examples closer to home as well. Consider the case of the American Civil War. For centuries, white settlers had been using the expanded options bequeathed to them by the progress of Western civilisation to hold blacks in servitude. Then the American Revolution brought a dramatic increase of freedom to whites throughout the colonies. Northern whites, still riding the wave of revolutionary libertarian fervour, actually used their newly expanded options to increase the options of blacks, by enacting a series of laws leading ultimately to the abolition of slavery in the North. But in the more agrarian South, where slavery was more deeply entrenched, whites were less attracted to the cause of the emancipation (though they often paid it lip service).
Later economic and political developments cemented Southern whites’ attachment to slavery still more firmly. Specifically, Eli Whitney and Katharine Greene’s invention of the cotton gin made plantation farming more profitable, while the Constitution’s three-fifths compromise (treating each slave as three-fifths of a person for purposes of representation) gave slave states a disproportionate voting bloc in Congress, and thus an added incentive to continue slavery. In order to take advantage of the expanded economic options offered by the cotton gin and the expanded political options offered by the three-fifths compromise, whites in the slave states needed to make sure that blacks’ options remained severely limited.
But to maintain the slave system, the South had to retreat from the libertarian principles of Jefferson and the revolution. Southern governments found it necessary to impose greater and greater restrictions on the civil and economic liberties of whites in order to keep blacks in subjection. Many states made it illegal for slaveowners to free their slaves; and there was soon no freedom of speech or press for whites who advocated abolition. In some cases, speaking against slavery was punishable by death.
Once secession finally came and the Confederacy was established, suppression of white freedoms grew even greater, as the central government, in the name of military necessity, extended its controls over every aspect of life. Internal passports were required for travel, traditional civil rights like habeas corpus were suspended, currency was devalued, and most sectors of the economy were nationalised. In their desperate quest to maintain their control over blacks, Southern whites found themselves compelled to establish an authoritarian political order that ended up claiming their own freedom as well.
This retreat from the principles of the American Revolution in political practice was accompanied by a parallel deterioration in political theory as well. During the 1810s and 1820s, the great intellectual spokesman for the South — the defender of agrarian interests against Federalist neo-mercantilist regulation — was John Taylor of Caroline (author of Arator, Tyranny Unmasked, and An Inquiry into the Principles of Government), whose political outlook was deeply Jeffersonian and libertarian — with the predictable exception of a massive blind spot about slavery. Taylor refused to face the tension between the principles of the Declaration of Independence and the institution of slavery; but later Southern intellectuals would face that tension — and resolve it in the wrong direction.
In the 1830s and 1840s, the ideological champion of Southern interests was not John Taylor but John C. Calhoun (author of A Disquisition on Government and A Disquisition on the Constitution). To his credit, Calhoun was a fierce opponent of centralised power, and came up with some rather ingenious ideas for curbing its growth (e.g., veto rights for minority factions); to this extent, Calhoun stood squarely in the Jeffersonian tradition. But the need to avoid that tradition’s radical implications for the legitimacy of slavery drove Calhoun to repudiate the principles of ‘76. Human rights, Calhoun maintained, rest on legal custom, not on the Laws of Nature — and the exercise of political authority does not depend for its legitimacy on the consent of the governed, but is a natural and inevitable feature of the human condition. By tossing the Declaration of Independence out the window, Calhoun was able to develop a Southern political ideology that could accommodate the institution of slavery. (Blacks were not one of the minority factions to whom Calhoun contemplated offering veto rights!)
The process of decay did not stop there. In the 1850s, the new ideological spokesman for the South was the arch-communitarian George Fitzhugh (author of Cannibals All! or Slaves Without Masters and Sociology for the South, or the Failure of Free Society). In Fitzhugh’s system, the need to justify slavery resulted in a full-scale assault on the Jeffersonian tradition in all its aspects; every vestige of libertarianism was methodically uprooted. Combining the right-wing nostalgia for an idyllic traditionalist feudal past and the left-wing hunger for a scientifically organised socialist future, Fitzhugh championed the Society of Status — an organic, hierarchical view of society in which every person has an assigned social role that carries with it both compulsory duties of obedience to one’s superiors and a guarantee of support, security, and paternalistic oversight from those same superiors. Black slavery, in Fitzhugh’s vision, was just a special case of the general principle that no person, black or white, is entitled to be the master of his or her own destiny.
Not all defenders of slavery accepted Fitzhugh’s philosophy, of course; but the general way of thinking which his works represented was becoming pervasive in Southern society. By 1862, the Confederate journal De Bow’s Review was trumpeting the slogan “The State is everything, the individual nothing.” (Some of the people who wear the Confederate flag on their jackets might want to think that one over.) The need of the Southern white culture to maintain dominance over its black population had led it to adopt principles which ended up threatening the freedom of its own white members.
It was not inevitable that Southern whites would choose to close their eyes to the injustice of slavery. That was their choice to make, and they made it. What was inevitable, or close to inevitable, was that this choice, once made, would have costly consequences — that it would have a corrupting influence on both their institutions and their ideals. When we find ourselves in a position of dominance over others, we cannot afford to excuse our authority on the grounds that the struggle for survival has favoured us. We cannot afford to follow Calhoun and Fitzhugh in rejecting the Natural Law that all human beings are entitled to equal respect, regardless of who has been dealt the winning hand. For if we do, we run the risk of destroying not only their freedom but, in the long run, our own.
I don’t mean to be giving the Union a free ride here. In the Civil War, both the North and the South decisively turned their backs on the ideals for which the American Revolution had been fought.20 The North’s drive to subjugate the South had an effect on the North analogous to the effect the South’s drive to preserve slavery had on the South. More authority was centralised in Washington; civil liberties were routinely violated; income taxation and Federally administered conscription were introduced; and an ominous cult of national unity spread through the American consciousness. The result was a Federal government with vast new powers — a fledgling Leviathan that quickly proved too tasty a treat not to be captured by the corporate élite. And so we are left, at the end of the twentieth century, with a burgeoning American police state whose primary victims, ironically, are the very blacks whose liberation was supposed to be the moral justification of Union victory.
The moral of this long historical digression is that when a society acquires a dominant position, the prospects for freedom can sometimes become not less but more precarious, first for the society’s neighbours and second (as a result of the need to keep those neighbours in subjection) for the society’s own members. Hence we are trusting in a weak reed if we rely on the process of cultural evolution to secure freedom for ourselves or our neighbours. If we want the meme of liberty to prevail, we must take the initiative and work to promulgate it, taking as our guide the polestar of Natural Law.