Law and Social Order - NYU

Law and Social Order - NYU

PI 11-3 3 Philosophical Issues, 11 Social, Political, and Legal Philosophy, 2001 Law and Social Order Russell Hardin Introduction There have been ...

119KB Sizes 1 Downloads 16 Views

PI 11-3

3

Philosophical Issues, 11 Social, Political, and Legal Philosophy, 2001

Law and Social Order Russell Hardin

Introduction There have been many debates about the relationships, if any, between law and morality: the positivists, such as John Austin ~1954! and Hans Kelsen ~1967! vs. virtually everyone else, especially natural law theorists; H. L. A. Hart ~1958! vs. Lon Fuller ~1958, 1969!; Hart ~1963! vs. Lord Devlin ~1959!; and recently Richard Posner ~1999! vs. Ronald Dworkin and many others ~Posner et al., 1998!. Most of the debate in each case has been about the moral content of the law, often about whether there is a necessary or minimal moral content of the law. In the most recent debates, the question is whether judges and others in the legal system would be able to do their jobs better if they were educated in academic moral theory. This issue presumably turns on whether laws are thought to have a moral content or are thought to be guided by a morality external to the law or at least whether a common law or appellate judge should bring moral considerations into deciding cases, perhaps in order to bring law into line with moral views. I wish to argue that actual legal systems in reasonably successful societies have a clear moral principle behind at least much of their law. That moral principle is mutual advantage, which, at a

62

Russell Hardin

minimum, includes social order. There may be many laws within a general legal system that have other moral bases or that arguably have no moral basis at all. If certain laws have a moral basis, that is merely a historical fact about what motivated legislators or common law judges to give those laws their content. But the central program of law to create social order is itself commonly guided by mutual advantage. This can happen by intent of those who devise the law or it can happen through a more nearly evolutionary or quasi-common law development that is itself not under the clear control of designers with intentions. It would be perverse to say that, if there is a legal system, then it must have been designed to be mutually advantageous. But it is not so perverse to say that, if a legal system has succeeded in effecting social order, then it must substantially serve mutual advantage. Mutual advantage, which is simply ordinal utilitarianism, is roughly Paretian. A policy serves mutual advantage if it benefits some and harms none—this is the Pareto principle in economics. At a more nearly systemic level, a legal system serves mutual advantage if, ex ante, it secures social order and the opportunity for prosperity. This is the core claim of Thomas Hobbes in his defense of rule by a sovereign with the power to secure order. Because he lacked a strategic insight that David Hume thoroughly grasped, Hobbes supposed that any sovereign must be supported without any threat of attempting reform against that sovereign’s rule. He seems to have believed that order must be intentionally created and managed. Hume’s insight is that we are capable of spontaneous coordination that can secure political stability even when there is no absolute power over us. Hume called such coordination over any particular matter a convention. Government helps us to organize and accomplish many things, but conventional devices help us to accomplish many other things that are not formally regulated—including, most importantly, government itself. The content of law is commonly quite specific and it is often backed by powerful sanctions that make us comply with it. Prior to such content, however, a legal and enforcement system must be created to give the content and to back the sanctions. Hence, the explanation of the working of law in detail and the explanation of the fact of a legal system are fundamentally different. We can sensibly say that what makes a particular law right in some sense is that it is promulgated by a legal system. To say that a particular legal system is right or good, however, is essentially to say that it serves us well. 1

3.

Law and Social Order

63

If a legal system roughly meets the Pareto criterion, we can say that the system itself is efficient relative to no system at all, as Hobbes argued must be true for virtually any system that secures order. If this is a reason for having a legal system, then we might suppose that the same criterion should often apply to the details of the system, that is, to the actual content of various laws. In general, however, it is wrong to apply a systemic principle to choices under a system justified by that principle. For example, we might give a utilitarian justification for a particular public institution but not be able give a direct utilitarian justification of specific actions required of the role holders in the institution. This is the argument of John Rawls ~1955! against a commonplace criticism of utilitarianism: that it can sometimes commend that officials act unjustly in order to reduce overall harms to citizens. Rawls argued that institutions that are designed to achieve utilitarian outcomes might require role behaviors that are not directly governed by a principle of utility maximization. For example, a sheriff should obey the law and should not ignore the requirements of the law in order to achieve higher utility. Nevertheless, although it is not always correct to argue from the same principle at both the institutional and the individual level, it might be correct to defend both a particular legal system and the laws and other details of it on the grounds that they serve mutual advantage—that is to say, that they are Pareto efficient in this sense. Before turning to issues of morality in the law, I will briefly discuss the structure of much of modern argument in the law, the two-stage structure of social and legal order, the nature of mutual advantage, and the related nature of efficiency arguments in the law. The dominant tradition of legal theory until recently has been legal positivism, which is driven by concern with definition, a concern that continues to haunt scholars today who would not wish to be labeled positivists. Prior to the positivist sidetrack into definitionalism, the focus of debate in legal theory was more nearly on explanation, about making law work. Jeremy Bentham and John Austin are transitional figures between Hobbes, who focused on making social order work, and the twentieth-century positivists, who have tied law into definitional knots. In value theory, Hobbes was arguably the first great mutual advantage theorist and, implicitly, an ordinal utilitarian. He was only implicitly an ordinalist because the issue of cardinal, additive welfare was not well framed before Bentham, so that Hobbes had no reason to argue against it. Hobbes’s mutual advantage

64

Russell Hardin

was a grand version of efficiency at the whole-society level. In the twentieth century, efficiency in the law is more commonly a concern at the intermediate level between specific actions or choices and legal rules or laws. After a long and misconceived detour into cardinal utility theory in the nineteenth century, Vilfredo Pareto restored the focus on ordinal efficiency, which has massively influenced contemporary law and economics through the Coase theorem of Ronald Coase ~1988! and, at the case level, the Hand rule of Judge Learned Hand. Once these general theoretical issues are made clear, I will turn to various arguments about the role of morality and moral theory in the law. My general argument is that the one general moral principle in law is the background principle of mutual advantage, as in Hobbes and Coase. Moral claims that are far more explicit and more pervasive than this principle are not persuasively requisite for law to work. Indeed, one can make a claim that is nearly the opposite of the usual claim for a necessary connection between law and morality. Any legal system that is heavily infused with morality will work well only if its subjects share that morality.

I. Two-Stage Theory of Social Order In any account other than a simplistic natural law theory, justification of law must inherently be in two stages. In the first stage, we follow mutual advantage to design and put a government and legal system in place or to make substantial changes in it; and in the second stage, we follow the dictates of the state and its legal system. Such a two-stage theory of legal justification is a commonplace in utilitarianism ~Rawls 1955, Hardin 1988!, in legal positivism ~Hart 1961!, and in constitutional theory ~Hardin 1999, chap. 3! Even within a legal system, there are often institutional choices that then determine specific policies or actions. For example, we justify an action, such as establishing a legal rule, given the prior justification of the system for establishing such a rule. This is a distinction Joseph Raz and others see between making a new law and applying an existing one. This is the inherent nature of institutions that are to serve particular purposes. Hobbes gives only a holistic justification of a legal system, and his argument does not immediately apply to particular laws. I have a reason for wanting a legal system in general, a reason I share with virtually everyone. But we do not necessarily share a common reason for choosing particular laws. Jurisprudential de-

3.

Law and Social Order

65

bates about the law or particular laws within a legal system have no status in debates about whether to have law ~Fuller 1969, 138!. For example, Kelsen’s basic norm ~Raz 1979, 122–127! must be a non-positive norm from which positive norms are derived. Indeed, constitutional principles are conventions outside or over the law. Hart’s rule of recognition—a customary rule—is external to the legal system and justifies actions within that system ~Hart 1961, 100–110!. Fuller says praise and blame for actions of an agency ~for example, such allocative agencies as the now defunct Civil Aeronautics Board! are misplaced when directed at individual agents of the agency; the praise or blame should be “directed to the aptness of the institutional design of the agency to perform the task assigned to it” ~Fuller 1969, 174!. Similarly, we may characterize many of Dworkin’s ~1996! issues as part of his implicit two-stage theory of law. There are constitutional moments to establish general principles and institutional structures and then, under the constitution, there are quotidian or policy moments. Hart and Fuller supposed their views apply to law as such. One might often read Dworkin’s views as applicable only to the US Constitution and law under it. Although his integrity of a legal system seems quite generally applicable to any legal system, freedom, liberty, and equal treatment need not be.

II. Hobbesian Social Order I will briefly spell out Hobbes’s theory of social order, because that theory is the background of subsequent discussions here. It is not important here whether my account of Hobbes is correct— those who think I have Hobbes wrong can read my account of his theory as my theory of social order. What leads to his theory are several background issues of what the world and the people in it are like. Among these are the rough equality of ability ~Hobbes 1968, 183!, the state of war in which we find ourselves if we have no powerful government, and the consequent uncertain payoff from investment ~186! and lack of security. There are three causes of quarrel: competition, anticipation of attack ~out of diffidence or fear of equality!, and glory seeking ~185!. What makes sense in general for maintaining individual interests is peace. Passions that incline us to peace are fear of death, desire of such things as are necessary to commodious living, and a hope by our industry to obtain them ~188!.

66

Russell Hardin

Hence, our basic problem is the need for regulation of competition and exchange in daily life. Hobbes supposes that this can basically be accomplished by coordinating on a sovereign. What the sovereign oversees and what leads us to require a sovereign are interactions that are not simply matters of coordination but that involve elements of conflict, as exchange does. Hobbes understands that exchange is not of “equal values” ~Hobbes 1968, 208! but benefits both ~“the just value is that which they be contented to give”!. Hence, exchange is superadditive in that, after completing an exchange, we are both better off. It can be represented game theoretically as a prisoner’s dilemma ~Hardin 1982!. Because it increases value to both ~or all! parties, exchange is a source or generator of value. It is not merely a libertarian issue of fulfilling my desires and yours, but of our making ourselves better off. That is why someone committed to the use of law to support morality might insist that law should not be used to govern contracts that do not, in the view of their morality, enhance value, as in contracts for prostitution, gambling, and so forth. Hobbes’s view of exchange is contrary to that of Aristotle and medieval Catholic doctrine, in which exchange is of supposedly equal values. 2 If exchange is of equal values, then it must be subjectively a matter of indifference, in which case it is not clear why we would bother to exchange. The value of X to me is subjective ~meaning that the value I assign to it need not be the value you assign to it! because it depends, among other things, on what I already have. Moreover, Aristotelian exchange is not superadditive so that, game theoretically, it would be zerosum ~or constant sum!. Ordinarily, zerosum games are of pure conflict so that whatever value you gain from our interaction I must lose, while Aristotelian exchange would entail moving from one outcome with its utility values to another outcome with identical utility values for each party, with neither gains nor losses to either party. We can say that Hobbes presents two theories, one about the creation of government and the other about the maintenance of government. Both of these are coordination theories, but the second is relevant to the real world of his and all times, while the first is essentially fictional ~Hardin 1999, 145–150!. In the two-stage theory of social order, the central problem of the state of nature is that, in it, there are many potential prisoner’s dilemmas. In these, there is always the possibility that one of the players will cheat the other by taking what the other has without surrendering anything in exchange. But neither the whole state of nature nor its resolution is a prisoner’s dilemma; and

3.

Law and Social Order

67

contract cannot resolve it because there is no sovereign available there to enforce the contract. In these conditions, notions of right and wrong, justice and injustice have no place—these are qualities that relate to men in society ~Hobbes 1968, 188!. We need government to get us out of or to keep us out of such a disastrous state of nature. Hobbes could not handle two problems that were urgent in his time: religious commitments and glory-seeking. Religious fanatics preferred disorder to any order that did not establish their views in power to govern the society; and ~generally aristocratic! gloryseekers wanted disorder that would give them opportunity to achieve glory through combat. Hence, Hobbes’s coordination on a sovereign was not strictly a mutual advantage move in his time, because it ran against what these two groups wanted. Two centuries after Hobbes’s Leviathan, Hume resolved a lingering problem of Hobbes’s analysis, although without reference to Hobbes per se. Hume supposed that we can spontaneously coordinate for order and that we can more or less spontaneously sustain that coordination for order, so that we do not require a regime of draconian enforcement ~Hardin 1993!. Strangely, Hobbes himself would have had implicitly to assume that such coordination could work for the establishment of government through contracting if the social contract argument could be made to work. He also assumed that the strategic structure of social order once a government is in place is that of a coordination game ~all prefer to go along with any extant government rather than to attempt to change it—if they understand the nature of the problem as Hobbes did!. Hence, maintenance of a government, whether one established by contract or by imposition by a foreign power or a revolutionary party, is also sustainable by spontaneous coordination. Hobbes’s overall argument for government was that it serves mutual advantage to have one and to keep one in place once it is in place.

III. Mutual Advantage and Efficiency If social order is mutually advantageous, then by definition it strategically coordinates us. As Henry Hart supposes, law is a facility enabling individuals to live a satisfactory life in common ~Fuller 1969: 223!. The purpose is something like Hobbes’s mutual advantage. In comparison to other major legal theorists, Hobbes was much more nearly a sociologist. He was concerned with what could

68

Russell Hardin

make order work to our general benefit, both to help us survive and also to help us prosper once we survived. Note that Hobbes’s device for making us prosper was a bit like the device of nuclear deterrence in our time until recently. Each side’s capacity for retaliation from beyond the grave reduced the incentive of the other side to make a preemptive attack. Persons cannot have a literal capacity for retaliation from beyond the grave, but they have a near equivalent from a state that acts as their agent of retaliation. Fuller is essentially a Hobbesian interested in what will make the law work. He is not a Hobbesian in the specific details of his sociology. In particular, he is much more sanguine about our prospects than Hobbes was. But his value theory is Hobbes’s value theory: mutual advantage. He shares Hobbes’s view that we need law for our own benefit. He also supposes that we cannot even live up to any moral principles we might have in a social vacuum or in a war of all against all ~Hobbes 1968, 215–217; Fuller 1969, 205!. Hence he is centrally concerned with the facilitative branches of the law and with the coordination function of the law ~Fuller 1981!. The concern with mutual advantage is a concern with efficiency in an ordinal sense. Such efficiency is the core of ordinal utilitarianism ~Hardin 1988!. The idea of efficiency has had a long developmental history that is instructive for our understanding of mutual advantage in the law. We could begin an account of its rise and articulation with Hobbes. In brief, we can say that Hobbes’s theory of social order and the notion of efficiency are closely tied together and the modern views of efficiency of Vilfredo Pareto and Ronald Coase complement Hobbes’s view. I will not rehearse the full history here ~see Hardin 1996!, but will only note highlights. A vague ordinalism reigned from Hobbes to Hume and Smith. But it was sometimes distorted by the assumption that value is in objects, so that it is objective. Jeremy Bentham assumed an objective utility in this sense and supposed that it is cardinal, so that we could add values of various objects or consumptions to obtain a measure of our total utility. Pareto broke this view completely with his metaphysical assertion that interpersonal comparison is meaningless. Since Pareto’s time, utility is seen as subjective, ordinal, and not interpersonally comparable. Contemporary discussion assumes a rough general form for the values that can be counted: they must be capable of trading for one another. There can be no absolute values included. Every value must be in a form that can be taken as more or as less

3.

Law and Social Order

69

depending on how it ‘trades off’ with other values. Hobbes, Pareto, and Coase attempt to avoid one kind of tradeoff—that between individual satisfactions. Hobbes fails for the cases of religious fanatics and aristocratic glory seekers. Pareto resolutely refuses to fail even at the cost of being radically indeterminate. Coase comes close to succeeding, although he has some difficulty for changes in rules or regimes when there are intrusive transaction costs or for changes whose effects we cannot know very well in advance. This is the major problem for such a value theory: how to add or judge across persons. Hobbes evidently was as antipathetic to interpersonal comparisons as was Pareto later. He was concerned only with improvements to each and every person’s existence, not with some additive notion of overall improvement. As noted, he also held the vision of exchange that came to dominate after Pareto. Exchange between two people is always Pareto improving if it is voluntary. Voluntary exchange works to mutual advantage. For example, in the 2-person static Pareto diagram, the only way to move to the frontier is through exchange, because that diagram represents the possibilities given what is already available— that is, without further production. Hobbes’s view has become the modern economist’s view as well as the libertarian’s view. Hobbes differed somewhat from the modern view in his concern with efficiency when there are massive knowledge constraints ~for example, he supposed that, at least in his time, we could not even know of significant differences between monarchy and aristocracy or democracy!. We may call this Hobbesian efficiency: It specifies merely a crude direction of improvement to mutual advantage rather than a precise direction, and it specifies only that we move some substantial distance without any claim to reach a frontier of possibilities ~since we cannot know where such a frontier might be!. In this, Hobbes shares the vision of John Dewey and other pragmatists. Dewey ~1960! says the most that we can expect of our efforts is that they bring improvement in our state of affairs, not that they bring us to some well defined final goal. This is also Fuller’s ~1969, 10–12! view. We need not know the morally best state of affairs in order to judge one state better than another. “And it is on this common sense view that we build our institutions and practices” ~32!. Legal positivists, such as Hart, Hobbes, and others, hold that legal duties are positive. They somehow derive from social relations of reciprocity, and so forth. We want greater welfare; we achieve it by creating institutions that regulate behavior and make

70

Russell Hardin

it a matter of legal duty. This is what Fuller ~1981; see also Fuller 1969, 9! describes elsewhere as the “coordination function of law.” Law serves mutual advantage. This is analogous to the account of Hobbes given above. Fuller is concerned not with whether you and I make an exchange but with whether, if we wish to exchange or we stand to gain from doing so, the law will facilitate our doing so. Contract law, tort law, and much else are, in this sense, among the facilitative branches of the law. ~Fuller argues that laws which are not facilitative but that attempt to block voluntary mutual choices, such as laws against crimes without victims, fail because their morality as law is wrong. Contrariwise, law can be very well used to enforce morality in promise-keeping or contract.! Fuller argues that there is much we need to know that is not in the law. The law is limited in what it can include. For example, it cannot give us the meanings of the words it uses to stipulate legal requirements. Such knowledge may be conventional or “customary.” For example, our knowledge of what it means to violate a relatively terse contract is likely to be customary in this sense. But we did not agree to or legislate the customary meanings— why then should they be binding? Only because it is mutually advantageous to us that they be binding, because there is no other way to go. If we agree with Fuller on this issue, then we evidently must grant that there is an inherent morality of law. But note that Fuller’s general argument, as is true of Hobbes’s argument, applies essentially at the whole-system level, not at the level of choosing particular laws, which are apt to have losers as well as gainers. Hence, the inherent morality is merely that of the mutual advantage of having a legal system, not of particular laws ~Fuller 1969, 138!. Typically, there are numerous, often very different moves that we can make that are all efficient in comparison to the status quo. We can, therefore, choose an overall efficient organization of the society that would not be the same as the organization that would have followed from manifold moves over piecemeal matters even though every one of those moves might have been individually efficient. A legal system that is geared for efficient outcomes need not include laws that themselves would all produce efficient outcomes piecemeal. Therefore, although one might argue that the principle of efficiency in some guise should be the moral foundation for law, it does not follow that it must be the moral foundation for all particular laws. Much of the time, however, it may be the right principle despite its being only intermediary for wel-

3.

Law and Social Order

71

fare. That makes generally good sense given that law is itself only intermediary for good life.

IV. Legal Theory Much of the modern concern with what might be called the morality of law is a residue of historical arguments for and beliefs in natural law, which is both a moral and legal theoretic tradition. Classicists recur to the arguments of Antigone and Aristotle that there is a natural law that is more or less everywhere right for every society. Catholics recur to medieval Catholic scholars who argued that there is a true law from god that we can either infer directly in specific instances of its application or in general as the background of a legal order. In the strongest natural law positions, it is supposed that individuals should follow natural law against the actual positive laws of a legal system when these conflict. Natural law theorists generally suppose there must be objectively correct moral-legal principles discoverable by reason. Legal validity is therefore merely moral validity, which is everywhere the same. Legislated law and judicial precedent might be useful in helping us to deduce what the true law is, but they have no moral weight in their own right otherwise. I will not address traditional natural law principles or claims here except in passing when addressing contemporary scholars who assert supposed principles of natural law. There is a third position that is labeled natural law by some writers. If there is some common human nature, we might immediately suppose that efficacy requires some fit with that nature; if that common or universal nature includes normative predispositions, efficacy might even require relevant normative content in the law. De facto, this is Fuller’s position, which leads him to call himself a natural law theorist. While this position is often explicit or implicit in legal theory today, the Aristotelian and Catholic visions are taken by many legal theorists to be so transparently wrong as not to require any argument. Although he spoke of natural law, Hobbes broke with the natural law tradition, whose terminology he used only to undergird his own positive law views. He was almost Cartesian in inferring his natural laws, or “lawes of nature.” 3 But his natural law arguments are very different from those of the natural law tradition. He does not say what is a priori right at all times independently of what is the positive law but only what would be right if it were

72

Russell Hardin

the law. His laws of nature are laws in the sense of theorems, not of commands of a sovereign ~Hobbes 1968, 217!. These laws of nature include, for example, the law to perform covenants ~contracts for exchange over time! made ~201!. They do not bind rationally—in foro interno—as David Gauthier ~1986!, Alan Gewirth ~1978!, and various others suppose their deduced moral laws do bind as rationally incumbent on any individual. Rather, they bind in foro externo ~Hobbes 1968, 215!, which is to say, only when they are enforced by some power external to the individual. For Hobbes, law is not a means of securing rights. Rather, law defines positive rights, which are the only rights worth discussing. But law is a means. It is a means of securing welfare. The welfare consists in survival beyond the likely lifetime one would have without law; success in social interactions, such as exchanges, that would be virtually impossible without law; and the benefits of working for one’s own interest. It is in the nature of the tasks of law that they can best be accomplished collectively. Hence, law is a collective good. But the good that comes from it is individualized and fully accountable at the level of the individuals whose welfare is served by it. Natural law arguments are commonly brought to bear at the level of specific cases or policies, such as policies on abortion and genetic engineering. In this form of argument, they are therefore not part of a two-stage theory of law, because natural law reduces directly to the single stage of applied laws, which are supposedly knowable independently of the nature of a legal system or its institutions. That Hobbes is not a genuine natural law theorist follows immediately from the fact that he requires a twostage theory, with the elevation of a sovereign at the first stage and the application of laws at the second stage. We can deduce the content of laws of nature but not any motivation independent of the threat of external force to follow them. For example, the bonds of a covenant have their strength from fear of some evil consequence upon the rupture ~Hobbes 1968, 192!. Hence, the laws are binding only for rational reasons. A covenant not to defend ourselves is void because we cannot rationally decide to let even the state kill us ~199!. Our duty to the state is not one of moral obligation, but is merely that of one who is obliged by the power of the state to abide by its laws. We might call this view, which is caricatured by Hart ~1961, 18–20!, the “gunman theory” of law. The big change from natural law to positive law perspectives is the positivists’ rejection of any claim that there is a necessary

3.

Law and Social Order

73

connection between law and morality ~Raz 1979, 38!. Moreover, there is no substantive “natural” good as would seem to be required for natural law theories or even, arguably, for the theories of Hobbes and Fuller, although most of us would accept Hobbes’s central values of survival and prosperity. The positive law view raises the is-ought problem, an issue in the so-called fact-value distinction that pervades discussions of law and its justification. We cannot derive an “ought” from an “is”—hence, evidently, we cannot say from the fact of law somewhere, such as the US, that there is anything right about it. This is a basic principle of positive law theorists. Hobbes makes an initial claim of value: self interest, especially in survival. One might conceive this as part of the nature of humans, in which case the value is merely functional. But one might also think it generally good that human interests be served. Fuller ~1981! is, in effect, Hobbesian here. He thinks law should serve interests, it should be facilitative. This is not a conceptual or definitional “should.” Sometimes, it sounds like a moral “should,” but it is only a causal “should.” That is to say, law should be facilitative or it will fail to work. Fuller supposes we can conclude from the fact that law works that it must serve people’s interests ~facilitatively!. This is a social-psychological, not a normative or conceptual, claim. Unfortunately, in populations that split into harshly conflicting groups, what is facilitative for one group may be destructive for another ~consider the US; Iraq with its religious factions; some of the residual nations from the former Soviet Union; Yugoslavia; and the case that drove much of the debate between Hart and Fuller, Nazi Germany!. To have moral force, Fuller’s view requires a principle of universalization. Coke’s objection to repugnancy ~Fuller 1969, 100! is really about non-discrimination, or equal treatment. Non-discrimination is not inherently required by law. Or, if it is, then the US Constitution and the law under it, which have massively discriminated over, for example, race and gender, cannot be law. Much of legal theory is dominated by conceptual analysis and definition with little or no attention to explanation. Austin, Kelsen, and Hart were centrally concerned with definition, with the “province of jurisprudence,” the “pure theory of law,” or the “concept of law.” This is the core of their positivism: to define law by what people do. The focus on the definition of law raises the prior issue of deciding whether what people do in this case is law. For example, is a system that is secret from the people to whom it is applied a system of law?

74

Russell Hardin

Fuller was concerned not only with definition but also with explanation, with the “efficacy of a system of legal rules.” Fuller ~1969, 49! says Austin failed to distinguish these two matters, and this failure was the basic defect of his analysis. Hart distinguishes them occasionally, but he insists on keeping them separate and on giving his attention almost wholly to conceptual analysis. His main exception is his sociological claim that, without the voluntary cooperation of many, “the coercive power of law and government cannot be established” ~Hart 1961, 201!. In this respect, Fuller is much more fully Hobbesian than are the main legal positivists of the past two centuries. Indeed, as does Hobbes, Fuller focuses more generally on the social science of the law than on conceptual issues. The differences between Fuller and Hart are that Fuller addresses the more important issue ~how law can be made to work vs. Hart’s concern with the definition of law! while Hart writes with much greater clarity. Fuller was a strong voice for the Hobbesian vision in our time. But his was a muddled voice, as for example when he spoke for natural law in a way that can be aptly characterized as confusing. His misconceived declaration for natural law was particularly provocative to the positivist school of contemporary legal theorists. Part of the intellectual agenda of legal positivism has been a variant of that of logical positivism in the sciences. The chief concerns have been the separation of normative and positive considerations in explanations of the law and its workings and the careful stipulation of how we are to know when we have a case of whatever is at issue. The first of these concerns is one that most philosophers since Hume have shared and that I will generally assume. The second concern is operationist. As physicists suppose they can know what is the meaning of an electron or a meter only by stipulating how to operationalize their observation, so the legal positivists have been at great pains to give operational definitions of the central concepts of law. Unfortunately, positivists such as Austin, Kelsen, and Hart have begun with operationalization of the largest, most complex, most inclusive social concepts: “law” and “legal system.” Modern social scientists commonly join forces with the legal positivists when they insist that we start any inquiry by first defining what we mean by “institution,” “government,” or “power.” Against this move, one might imagine Ernst Mach, the physicist and early logical positivist, starting with the operationalization of “nature” instead of mass or distance. But that would have

3.

Law and Social Order

75

been metaphysics, not physics. And it would have played little role in the life of an actual physicist concerned with real, perhaps newly discovered phenomena. Kelsen, Hart, Raz, and, to a lesser extent, Austin have been unduly concerned with the relatively metaphysical issue of what law is, the concept of law, the province of law. They are definitionalists. Fuller ~1969, 157! criticizes Hart for conceiving law entirely in terms of its formal source—for example, King Rex—rather than as a complex undertaking capable of various degrees of success. Although he was an important part of the tradition that led through Austin to Hart, Hobbes was very different. His chapters 13 through 15 in the Leviathan are plausibly the most richly packed justification of government and law ever written ~Hobbes 1968, 183–217!. The central focus of these few pages is the sociological possibilities of law and order. Hobbes started not from the top down with a definition of law but from the bottom up with a psychology of the person in social interactions. He shared with the positivists a relative disinterest in the content of the law. But he dramatically differed from them in his focus on its workability. He shared none of the positivists’ concern with the “identity” of law or of a system of law, with such questions as whether some particular set of institutions is a system of law. At first glance, one might think the positivists’ overwhelming worry with such questions is merely analogous to the physicists’ worry whether what they are seeing in their observations is the effect of an electron rather than of something else. But even a moment’s thought suggests that the parallel is faintly ridiculous. For the physicist, the operational definition of an electron is useful only as part of a theory of how it works, what it does. That was Hobbes’s concern with law, a concern the twentieth-century positivists have largely ignored. The legacy of definitionalism infects the work of contemporary legal scholars who do not share the utilitarianism of Austin and Hart. For example, Joseph Raz writes of “the authority of law,” by which he means its source. Hence, he is with the analytical legal positivists in his concern with who can make law. He speaks of the “set of all the pure statements referring to one legal system” ~1979, 80!. Mach would bridle at such metaphysics. The core of a positivist position should be that what works works. For Kelsen, a legal system “exists if it is effective and this does not entail acceptance as morally just” ~Raz 1979, 141!. If Hobbes and Fuller are ~positively! right about human nature, then part of what works for us is what is good for us—or at least what

76

Russell Hardin

serves our interests. Hence, the demand that it work might require that law have some moral core. But this is a sociological debate about how we are constituted and how we are likely to react to various devices of regulation. It is an issue on which the legal theorists have had better or worse insights, but none of them is seriously concerned with sociological analysis, even though Hobbes, Austin, and others have made substantial sociological claims. In one of the major debates over the morality of law, Devlin ~1959! has an essentially sociological view about the effect of certain “immoral” practices on the stability of society and its legal system. Hence, he is interested in explaining the effectiveness of a legal system and not merely in giving its definition. But his views on legalizing homosexuality and its supposedly catastrophic effects on his society are little better than an amateur’s prejudices, a relatively ignorant amateur’s prejudices. Hart ~1963, 50! slyly says that “it is not at all clear that for @Devlin# the statement that immorality jeopardizes or weakens society is a statement of empirical fact.” Sociologists can do little more than formulate Devlin’s question. They can certainly not answer it with any confidence or agreement. Fuller ~1969, 133! argues that legal suppression of homosexuality will fail. A subjective assessment of the difficulties and disruptiveness of enforcing laws on homosexuality and on various vices, such as gambling, prostitution, and drinking alcoholic beverages suggest that Fuller and Hart have the better case.

V. Moral Theory in the Law Posner and his recent critics address the question whether academic moral theory would be useful to judges in helping them rule in specific cases. Posner ~1998, 1999! holds that laws cannot and should not be grounded in what he calls academic moralism. Of course, any particular law could be grounded in academic moralism, but the whole system of law could not be because no academic moral theory is adequate to the task. From the perspective of a two-stage theory of legal justification, one could ask whether moral theory can or should play a role in understanding the principles behind the law itself. The answer is clearly yes. But Posner’s focus is on the role of judges dealing with specific cases, not on justifying a legal system ~he is not explicit about this, but the position seems to be clear in a sense analogous to that in which Dworkin holds that certain principles are instinct in the Consti-

3.

Law and Social Order

77

tution!. Understanding the general principles behind the law need commonly play no role in guiding judges in their decisions. Indeed, this is virtually an instant implication of the two-stage theory. Whereas Posner seems to keep his arguments focused on the lower level issue, some of his critics slip between the justification of the legal system itself and the justification of decisions in particular cases within the system. Variants of academic moralism include Kantianism, virtue theory, rights theories, and utilitarianism. Kantian moral theory is so thoroughly anti-consequentialist that it arguably cannot have a major role in guiding a general legal theory, but it is regularly invoked and Kant also wrote extensively on law. Arguably, however, his moral theory does not inform his legal theory, whereas it is his moral theory that often informs contemporary claims for morality in the law. It is often brought to bear at the level of specific cases rather than at the systemic level. Utilitarianism, which is causal social science plus central concern with welfare ~consequentialism!, and legal positivism have had a long and close association from Jeremy Bentham and Austin through Hart. Indeed, Austin ~1954! supposed that good law has a utilitarian moral basis. Virtue theory, which is usually functionalist, was for many centuries closely associated with natural law, although it is not independently associated with a major school of contemporary legal theory. There are many rights theories, including those of Dworkin, Gewirth ~1978!, libertarians, and others. These have been the chief alternative to utilitarian justifications in law in recent decades, although their lack of a coherent set of principles often makes them seem ad hoc. ~Hart @1983# saw utilitarianism as the standard position in Anglo-American legal theory for nearly two centuries until the rights revolution after about 1970.! Their popularity may be fading. Dworkin himself has more recently argued from principles that he supposes are “instinct” in the very idea of law itself. So, for example, he argues that the idea of integrity is “instinct in the concept of law itself” ~Dworkin 1996, 83!. Sometimes, particular laws are grounded in overtly moral principles, although it would often be wrong to claim that the principles are consistent from one law or body of law to another within the same legal system. Civil law commonly coordinates and makes more efficient. To a substantial extent, criminal law does as well, although superficially most observers are apt to suppose that any particular criminal statute or prohibition is itself directly moral and likely morally motivated. It is a bit too easy to reach such a conclusion in many contexts, because what is commonly outlawed

78

Russell Hardin

is virtually defined as immoral. For example, although it is not always either morally or legally wrong to kill someone, it is legally wrong to murder someone. But this is just to say that “murder” is legally defined as wrongful killing. Our view of what counts as wrongful killing has changed over time. Killing in a duel was once a matter of great honor in many societies in which today it would be murder. Debates about dueling were commonly about the stupidity and waste of the practice. The most effective move against it was arguably not to call it immoral but to ridicule it, because ridicule eviscerated the sense of honor that backed the convention of dueling ~Hardin 1995, 98– 100; Bacon 1614!. Similarly, so-called crimes of passion—killing one’s spouse and the spouse’s paramour when caught in flagrante— have been a matter of honor in many societies, including Texas within recent memory. Posner has three theses on the relation of academic moralism to law. These are, first, that there are no moral universals. One might ask, does this mean merely that there is no actual agreement across times and places? And one might think that this claim trades on the act versus consequences distinction. There might be no universal moral rules on rightness or wrongness of types of acts even though there might be a universal principle of judgment of consequences. Second, much of what is of moral concern can be handled without moral theory, as in a judge’s decisions. Even a casual reading of important recent decisions suggests that Posner’s point here is descriptively compelling. Judges do not often invoke moral theories. However, if what is of interest is “what works,” Posner must show that reading academic philosophy does not work for judges. But reading economic theory or having an economic rule of thumb ~such as the Learned Hand rule! works. Why not moral theory? Economic theory is useful because it helps to handle consequences. The only major moral theory that focuses on consequences is utilitarianism, and it depends heavily on economic analysis for many issues. It is not clear that moral theories not focused on consequences have resources for assisting judges. Finally, Posner claims that academic moralism cannot succeed, because the discourse of moral theory is interminable because indeterminate ~Posner 1998, 1802!. Posner’s claims may be too strong or even outrageous by some lights, but they mostly have the strength that we know what it is that we agree or disagree with. But some of the criticisms of his views are too high-blown to be very grounded—for example, those

3.

Law and Social Order

79

of Anthony Kronman and Martha Nussbaum ~Posner et al. 1998!. There are strings of words that look like sentences but that lack the content. It would be hard to say what it would take to show that they are either right or wrong. Kronman ~Posner et al. 1998, 1758! says that reason “promotes the flexibility of attitude and approach that every effort of cultural preservation requires”; and he speaks of our character as a people ~1763!. Critics of Posner’s complaint that moral theory is of little use in the law often invoke “moral reason,” as though to say that it is something other than reason tout court. For example, Kronman ~Posner et al. 1998, 1764! says Posner has it backwards when he says that the role judges play leaves especially little room for moral reason. 4 ~Posner’s complaint is generally against the use of moral theory, not moral reason.! What is added when the qualifier “moral” modifies reason? We could speak of reasoning in a moral theory, which is presumably nothing more than reasoning that begins from moral claims and reaches conclusions from these. Or we might speak of reasoning to a moral theory or principle, which is clearly something more than mere reasoning because it actually generates moral principles or rules. Natural law theorists, intuitionist moral theorists, and many others, such as Gewirth ~1978!, have claimed to reach moral conclusions directly from reason. The central problems or disputes of moral philosophy are cases of disagreement between moral theories ~for example, as in the case of abortion!. 5 If reason is supposed to lead us to moral principles in such cases, then it evidently fails us in the sense that it leads some of us to one set of conclusions and others to a contrary set of principles. This is Posner’s complaint, that when there is a moral issue, moral theories cannot settle it for a court. He says that judges do legal reasoning, which is not much different from ordinary reasoning about nonmoral issues; rarely do they engage in moral reasoning and, when they do, it is without the help of academic moral philosophy ~Posner 1998b, 1811!. Hence, problems of reasoning to a moral theory or principle vitiate any claim for the use of moral reason. Indeed, one might suppose that a case could be made for a reason-value distinction that is analogous to the fact-value distinction. Posner argues that judges don’t need moral theory, indeed, that they can’t make use of it ~Posner 1999, 131, 141!. His critics do not actually address this point with cases to show how judges can or do use moral theory successfully. He argues further that such political “innovations as republicanism, the separation of powers,

80

Russell Hardin

the system of checks and balances, and the secularization of politics can be...detached from their philosophical aegis and evaluated without regard to philosophical principles” ~Posner 1998b, 1810!. He supposes that the central issue for legal scholars is how to instruct a judge and that metaphorical and vague talk about moral reason cannot do this.

VI. Concluding Remarks We can say that law as mere principle or set of behavioral rules would have no content. Its use typically involves implicit evaluations, which must often have moral overtones. This fact may suggest that there is an even stronger connection to morality, but there need not be. Still, if law is to work it must serve people well. This is the minimal moral content of law not by definition but by causal requirement if law is to work well. Of course, one might still suppose that a legal system would work best if its subjects shared many important values. One might suppose that we must have common values if we can see the same meanings in what are essentially vague laws. Much of our written law is in reality unwritten; the law in the books requires an understanding of the shared assumptions that enter into making and interpreting it ~Fuller 1969, 232!. Indeed, custom is explicitly made a standard of decision in the Uniform Commercial Code ~234!. Customary law helps fill the gaps that will always be perceived in any body of enacted law. But shared meanings are not shared values, as any logical positivist would have recognized from the beginning. For some objective matters in the law, the conventional understanding can be quite clear and definitive; for many more nearly subjective matters, such as quality of workmanship, the understanding may be much looser and it may be much harder to agree on what “the” convention is. In sum, a grand principle of efficiency—overall mutual advantage from having social order—stands behind a successful legal system. A simpler principle of piecemeal efficiency might be thought derivatively to apply to particular laws. However, no such inference can be drawn from the higher order grand principle. Although efficient piecemeal moves often do lead to overall efficient states, inefficient piecemeal moves do not rule out overall efficiency. Moreover, if law is restricted to the governance of dyadic relationships, these can be piecemeal efficient even while they lead to overall inefficient outcomes. When we have done our best with

3.

Law and Social Order

81

dyadic moves, we may stand in need of government or other collective action to move us to a state in which we are all still better off. It is striking that the principle of efficiency—from Hobbes to contemporary law and economics—has come to stimulate a theory that is much fuller and more articulate than any alternative. Indeed, most alternatives seem to be the ad hoc creations of individual scholars. Arguments in law and economics ~Calabresi 1985; Polinsky 1989; Posner 1986! are generally relevant to considerations at the margin of a background of general arrangements. Hobbes addressed this background and not the marginal issues. From roughly Hume and Adam Smith forward, most discussion has been at the margin and, indeed, marginal analysis revolutionized economic theory in the late nineteenth century. John Rawls ~1971! attempts to establish principles for the background of general arrangements. Henry Sidgwick ~perhaps the greatest utilitarian! argued that we don’t know enough to start from fundamentals, we can only talk intelligently about improvements. Oddly, Hobbes seemingly agreed in that he thought we should live with the regime of law that we already have rather than engage in revolutionary action to change it. I have not considered any alternative moral principles for the law with anything more than passing comments here. In large part, that is because many of the critics of the mutual advantage vision and its variants do not present any alternative and do not appear to have one ~see, for example, Coleman 1988; see further, Hardin 1992! or have positions that are idiosyncratic and not widely enough adopted by others to be very articulate or comprehensive ~for example, Dworkin 1996!. Hence, extended criticisms of other principles would be premature here. In considering any alternative, however, one might ask first: Must every philosophy of law be consequentialist? We would not have law except to achieve or have effect on something, or on various things. Hence, law is inherently consequentialist, although the consequences at issue could be quite diverse, including, for example, autonomy, welfare, and religious commitment. ~Jonathan Baron @1998# makes the same claim about public policy.! Even when law is used to enhance autonomy or religious commitments, however, it must generally also maintain social order or it cannot achieve these other goals. Social order and even more extensive mutual advantage are not much, perhaps, in comparison to an articulate moral theory, but we must almost all want them. Social order, a minimal form of mutual advantage, is a sine

82

Russell Hardin

qua non of almost all else; and more extensive mutual advantage is the equivalent of individual welfare generalized to cover all. I can achieve prosperity best if my society achieves it in general. Of course, the occasional Mobutu can do best as a parasite on the whole society, amassing great personal wealth by impoverishing most others ~at some risk of assassination or overthrow!, but few of us can expect to prosper best while most others languish. Our prospects are causally connected. As noted earlier, Hobbes cogently supposed that survival and welfare are to be achieved collectively. Raz ~1986! wants liberty or autonomy to have a role in his account roughly like that of mutual advantage in the account here. He recognizes the essentially consequentialist nature of law, and therefore adopts a consequentialist principle of autonomy in the sense that he wants laws to have the effect of enhancing autonomy as measured somehow in the aggregate. As moral theories go, this one is much less well worked out than is utilitarianism, and the nature of autonomy as a value is much less articulated than is the nature of economic value theory ~Hardin 1992, 382–4!. Robert Nozick ~1974! criticized what he called a consequentialism of rights, that is, the trading off of some rights fulfillments for others. That is, of course, a large part of what a working legal system does. Indeed, one might even claim that this is a principle reason for having a legal system. For Nozick, rights were prior to law. For legal positivists, rights are the construct of law, so that this claim is true essentially by definition in the positivist vision. In any system of law, however, it seems implausible that a complex set of rights could be applied without occasional—even frequent—need to trade one against another, either my right to X against your right to X or one kind of right versus another kind, such as the partially conflicting rights of the accused and of the victim ~Posner 1999, 160; see also Hardin 1988, 117–126!. Consider an essentially sociological claim that is made in many forms by many legal theorists of quite different views. Hart ~1961, 201! says that law works because of the willing compliance of most people, without which the coercive power of law and government cannot be established and the state would not have the capacity to deal with even the few miscreants. Fuller ~1969, 219! says that the functioning of a legal system depends on a cooperative effort, as, for example, it obviously does in traffic regulation ~220!. The central sociological fact here might mislead us into thinking that somehow people share a normative commitment to the law and that it works therefore.

3.

Law and Social Order

83

Fuller’s example of compliance with traffic regulations highlights the central issue, which is that law coordinates us, commonly in ways that we would choose for ourselves if only we could do so without a central regulator. That there is a lot of willing obedience to the law ~virtually any law or system of law anywhere! does not entail any moral commitment beyond that of a commitment to any order that coordinates us in each living his or her own life. We might happen to share some moral vision of what that life should be, as deeply religious communities commonly do, but we need not have any vision in common that goes beyond mere coordination on social order. In contemporary pluralist societies, we might share strong moral views within particular communities but not across multiple communities, all of which are sustained by a single legal system. Hart ~1963, 51! supposes it an “acceptable proposition that some shared morality is essential to the existence of any society.” He has no way to show that this is true and it seems plausible that some societies have existed without any shared morality beyond what could count as egoistic concern with coordination on social order. Notes 1. We might quibble over legal systems that are essentially religious, such as Shariya in Islam. The arguments here do not depend on whether every legal system fits my account, but one might relatively easily characterize even an ostensibly religious system as one that serves mutual advantage for a particular society—insofar as its populace shares the religious views. 2. Aristotle’s view seems to turn on seeing value as objectively inherent in the objects exchanged, not determined by the use we would make of the objects and hence their subjective worth to us. Hobbes is one of the first major thinkers to adopt the subjective theory of value, a theory that did not finally become firmly established in economic writings until the nineteenth century. 3. One could suppose that Hobbes argued from natural law because that was the going theory of law in his time. Similarly, Hume argued as though he were a virtue theorist because that was the going theory of morality in his time. 4. Kronman ~1988, 1759! further says that, “just as a good habit cannot be created by reason, it cannot be undone by reason either”; and that “only reason can reconstruct the individual who no longer feels the force of an earlier, unthinking loyalty to a particular community either to that community or to any other” ~1761!. 5. Some might wish to say that, in the case of abortion, the disagreement is between religious and secular moral views.

84

Russell Hardin

References Austin, John. [email protected]# 1954! The Province of Jurisprudence Determined. New York: Noonday. Bacon, Francis. ~1614! The Charge of Sir Francis Bacon Knight, His Majesties Attourney generall, touching Duells, upon an information in the Star-chamber against Priest and Wright. London; New York: Da Capo Press, facsimile reprint, 1968. Baron, Jonathan. ~1998! Judgment Misguided: Intuition and Error in Public Decision Making. New York: Oxford University Press. Calabresi, Guido. ~1985! Ideals, Beliefs, Attitudes, and the Law. Syracuse, NY: Syracuse University Press. Coase, R. H. [email protected]# 1988! “The Problem of Social Cost.” In Coase, The Firm, the Market and the Law. Chicago: University of Chicago Press: 95–156. Coleman, Jules L. ~1988! Markets, Morals and the Law. Cambridge: Cambridge University Press. Devlin, Patrick @later Lord#. ~1959! The Enforcement of Morals. Oxford: Oxford University Press. Dewey, John. [email protected]# 1960! The Quest for Certainty: A Study of the Relation of Knowledge and Action. New York: G. P. Putnam’s Sons. Dworkin, Ronald. ~1996! Freedom’s Law: The Moral Reading of the American Constitution. Cambridge, MA: Harvard University Press. Fuller, Lon L. ~1958! “Positivism and Fidelity to Law.” Harvard Law Review 71: 630–672. _ . [email protected]# 1969! The Morality of Law. New Haven, CT: Yale University Press, revised edition. _ . ~1981! “Human Interaction and the Law.” In Fuller, The Principles of Social Order, ed. by Kenneth I. Winston. Durham, NC: Duke University Press: 211–246. Gauthier, David. ~1986! Morals by Agreement. Oxford: Oxford University Press. Gewirth, Alan. ~1978! Reason and Morality. Chicago: University of Chicago Press. Hardin, Russell. ~1982! “Exchange Theory on Strategic Bases,” Social Science Information 2: 251–272. _ . ~1988! Morality within the Limits of Reason. Chicago: University of Chicago Press. _ . ~1991! “Hobbesian Political Order.” Political Theory 19 ~May!: 156–180. _ . ~1992! “The Morality of Law and Economics.” Law and Philosophy 11 ~November!: 331–384. _ . ~1993! “From Power to Order, From Hobbes to Hume.” Journal of Political Philosophy 1 ~March!: 195–207. _ . ~1996! “Magic on the Frontier: The Norm of Efficiency.” University of Pennsylvania Law Review 144 ~May!: 1987–2020. Hart, H. L. A. ~1958! “Positivism and the Separation of Law and Morals.” Harvard Law Review 71: 593–629. _ . ~1961! The Concept of Law. Oxford: Oxford University Press. _ . ~1963! Law, Liberty and Morality. Stanford, CA: Stanford University Press.

3.

Law and Social Order

85

_ . ~1979! “Between Utility and Rights.” In Hart, Essays in Jurisprudence and Philosophy. Oxford: Oxford University Press, 1983, pp. 198–222. Hobbes, Thomas. [email protected]# 1968! Leviathan. Harmondsworth: Penguin, ed. by C. B. Macpherson. Hume, David. [email protected]–40# 1978! A Treatise of Human Nature. Oxford: Oxford University Press, ed. L. A. Selby-Bigge and P. H. Nidditch, 2nd edition. Kelsen, Hans. [email protected]# 1967! Pure Theory of Law. Berkeley, CA: University of California Press. Locke, John. [email protected]# 1950! A Letter Concerning Toleration. Indianapolis, Ind.: Bobbs-Merrill. Nozick, Robert. ~1974! Anarchy, the State, and Utopia. New York: Basic Books. Okun, Arthur M. ~1975! Equality and Efficiency: The Big Tradeoff. Washington: Brookings Institution. Polinsky, A. Mitchell. ~1989! An Introduction to Law and Economics. Boston: Little, Brown, second edition. Posner, Richard A. ~1986! Economic Analysis of Law. Boston: Little, Brown, 3rd edition. _ . ~1998a! “The Problematics of Moral and Legal Theory.” Harvard Law Review 111 ~May, no. 7!: 1637–1717. _ . ~1998b! “Reply to Critics of The Problematics of Moral and Legal Theory.” Harvard Law Review 111 ~May, no. 7!: 1796–1823. _ . ~1999! The Problematics of Moral and Legal Theory. Cambridge, MA: Harvard University Press. Posner, Richard A., Ronald Dworkin, Charles Fried, Anthony T. Kronman, John T. Noonan, Jr., and Martha C. Nussbaum. ~1998! “Symposium on Richard Posner’s Oliver Wendell Holmes lectures at Harvard.” Harvard Law Review 111 ~May, no. 7!: 1637–1823. Rawls, John. ~1955! “Two Concepts of Rules.” Philosophical Review 64 ~1955!: 3–32. _ . ~1971! A Theory of Justice. Cambridge, MA: Harvard University Press. Raz, Joseph. ~1979! The Authority of Law. Oxford: Oxford University Press. _ . ~1986! The Morality of Freedom. Oxford: Oxford University Press.