University of Western Sydney Law Review
Rachael M. Patterson[*]
Is there a prima facie moral obligation to obey the law? It was once generally agreed that the answer to this was yes – that, independent of the moral desirability or merit of the actual rules provided for by the law, there was a reason to obey the law simply because it was the law. From the early 1970s, however, this view was challenged when legal and moral philosophers began to argue against the existence of a general obligation to obey the law. Legal positivist Joseph Raz, for example, denied the existence of any such obligation and claimed that it is 'at best a mere shadow of other moral duties'. 
The challenge presented by those refuting the moral authority of law did not, however, go unanswered. In response to Raz, for example, natural law theorist John Finnis provides an explanation and defence of legal authority and obligation in terms of what is needed to secure the human good. One result of this debate is that any consensus that might have been as to the moral authoritativeness of the law is now gone. Another is that the issue remains contentious and continues to be of interest. This paper examines one aspect of this debate, namely that between Raz and Finnis over the moral authority (or lack there of) attached to the law as a result of its instrumentality in solving social coordination problems.
Both Raz and Finnis agree that everyone has a moral reason to further, or at least not to frustrate, the achievement of certain social goals. They also agree that the law is instrumental in the achievement of these goals through the setting up and maintenance of schemes of social action. Raz and Finnis disagree, however, on the implications that this has for the moral authority of law.
Finnis argues that what is instrumental in securing a morally obligatory goal must itself be morally obligatory unless there exists some other instrument that is equal or better.  On this basis Finnis treats the law as authoritative. He accepts that there is a moral obligation to obey the law since, for Finnis, the law is distinguished as the most efficacious process for coordinating social action.
Raz, on the other hand, denies that the instrumentality of law in facilitating social schemes of cooperation gives rise to a moral obligation to obey the law.  Raz argues that the moral reasons to cooperate with certain social schemes derive entirely from the practice of cooperation and not from how the scheme originated or is maintained. That the law is instrumental in the institution of cooperation is irrelevant and is of no moral significance greater than if the scheme had been facilitated by some other means such as propaganda, private or public exhortations, or spontaneous formation.
For there to be an obligation to obey the law, Raz argues that there must be a reason to do what the law requires simply because the law requires it. This reason or set of reasons must be of sufficient generality that the obligation to obey applies to all laws, to all subjects of the law and in every circumstance. Furthermore, it must also be a peremptory reason. Raz denies that such a non-redundant, general and peremptory reason or set of reasons to obey the law exists and argues that there is neither a prima facie nor an absolute obligation to obey the law.
In response, Finnis states that Raz 'overlooks the diversity of views about social problems'. Finnis argues that in order to coordinate social activity necessary to the pursuit of the common good or the fulfilment of morally obligatory social goals (coordination problems), unanimity of action is required. This is because for most, although Finnis admits not all coordination problems, there are a number of solutions that are available, reasonable and appropriate, but none of which will be sufficient to resolve the problem unless adopted to the exclusion of all alternatives. As an example, Finnis refers his readers to the problem of river pollution as raised by Raz, and lists four available solutions. For any of the listed solutions to be effective they require coordination such that large numbers of people act or refrain from action in accordance with the stipulations of the chosen solution. That is, there must be a degree of unanimity of action in compliance with the chosen scheme in order for it to be successful.
Unanimity of action is achieved either when there is consensus on the desirability of a selected scheme of action such that compliance is universal (Finnis refers to this simply as ‘unanimity’), or through authority. However, ‘unanimity’ in practice is not possible within a political community. This is because in any given political community there are a plurality of goods, persons and perceptions and, as such, it is impossible to obtain consensus among members of a community on all the relevant issues needed for a meeting of minds on the desirability of one particular solution for any particular social problem. Why? Because the common good is complex. The common good 'excludes no aspect of individual wellbeing and is potentially affected by every aspect of every life plan' and ultimately involves 'the participation by a multiplicity of persons in a manifold of goods'. In addition to its complexity, the common good is also open-ended for it can never be fully realised and there will always be some coordination problem to be remedied. Most communities are made up of members who possess a degree of autonomy but require coordination of social action in order to obtain some common human goods. These members have different and varying perceptions of what goods contribute and belong to the common good, the relative importance of these goods, and the most suitable means of obtaining them. Furthermore, not all members of the community act with intelligence and commitment to the common good at all times. Given such plurality and complexity, it is difficult for a community to collectively determine a course of action to solve any particular coordination problem or foster any particular good, and even more difficult for it to instigate and maintain the decided solution when those who favour alternate courses of action do not have sufficient reason to comply. 
When authority is relied upon to act as a substitute for ‘unanimity’ in determining solutions to practical coordination problems, uniformity of action results from a quasi-unanimous acknowledgment of and acquiescence to the process used to select one scheme of cooperation from another. Although there may be a lack of consensus about the suitability of the eventual chosen scheme of action, cooperation still takes place because the selection is treated as authoritative.  Finnis writes:
'Unanimity in action must result from the quasi-unanimity about the desirability of conforming to the authoritatively selected scheme'.
Of course, if ‘unanimity’ in practice is impossible, so too is unanimity on the authoritativeness of any solution selection process. This is explicitly acknowledged by Finnis in Natural Law and Natural Rights. In this text, Finnis writes that the unanimity of judgment required for a decision or process to be authoritative and to facilitate social coordination is not easy to come by. Finnis argues that the degree of consensus needed is a qualified ‘quasi-unanimity’ or ‘virtual unanimity’. Finnis does not explore the empirical question of exactly what degree of consensus must be obtained for a solution selection process to be considered authoritative but accepts that whatever the degree might be, it exists in the case of law.
Finnis takes it as a given that there is in practice a ‘quasi-unanimous’ acceptance of the solutions to coordination problems offered by law. In contrast to other means of social coordination, the law is treated as authoritative. Finnis cites in example that a person will cooperate with or acquiesce to a scheme of action selected by law even though that person thinks the scheme is mistaken and knows that it may be damaging to his or herself in some way. On the other hand, a person will not cooperate (or is much less likely to) in such instances when the scheme has been designated by one of the alternative selection processes such as propaganda, private or public exhortations or spontaneous formation.
Finnis explains why the law is treated as authoritative using three concepts: efficiency, salience and shared interest. The law is treated as authoritative, not because of the quality of the solutions offered by the law, but because the law is salient.  This salience is attributable to certain features of the law that create a ‘shared interest’ in the 'regular, impartial upholding of the law itself' thereby providing individuals with moral reason to 'to collaborate with the law’s co-ordination solutions' and to regard the law as authoritative. These features are also the very things that make the law an incredibly apt and efficient means of identifying and solving coordination problems. 
What does Finnis mean by salience? Finnis adopts and applies the game-theoretical concept of salience. Game-theoretical coordination problems, Finnis explains, are solved through the salience of one solution. For any given problem, there may be several possible solutions and each of the players may possess a different preference. However, one solution is chosen because for some reason it is conspicuous. This allows the solution to serve as a focal point on which the interests of the players can converge and thus facilitates cooperation. The selection of the salient solution has nothing to do with the merits of the solution itself.
Finnis raises a hypothetical to illustrate this. A salient solution for where a group of people unfamiliar to New York might meet in the city would be New York Central Station even though this solution may not be as suitable as many of the other available options given that there are certain drawbacks to meeting in the station. Finnis does not explain, however, why meeting in New York Central Station would be a conspicuous solution in such circumstances. It may be because each member of the group has heard of Grand Central Station and it would not be difficult to find out how to get there. The well-known character of the station is therefore one aspect over which the interests of the members of the party merge since each member has an interest in meeting in a place that they know of and that they can get to. Because the interests converge and agreement over the location becomes possible the solution is an obvious one and becomes prominent in the consciousness of the individuals and the group. It might be argued that the reason the station solution is prominent or salient in this case is because the solution itself, or rather one aspect of the solution (that the location is well known,) has more ‘merit’ than the locations of the other solutions. However, the solution becomes conspicuous because everyone is easily able to agree over the fittingness of the location. It is the fact of agreement (that the parties do and can agree) rather than the merits of the location itself that elevates and brings to attention this particular solution.
Finnis explains that the salience of one solution may be due to a number of factors, including past behaviour (precedent), convention, agreement or decree and occurs without any consensus as to the superiority of the salient solution over the alternatives. Likewise, in the case of law there are a number of factors that contribute to its salience as a means of solving coordination problems and which have little to do with the merit of the particular solutions offered by law. These factors are predominantly values of the Rule of Law, which not only promote and establish the law’s salience, but also make the law a highly effective and handy instrument of social coordination in its own right. 
A legal system that exemplifies the Rule of Law should have rules that are prospective, clear, coherent, stable and that are possible to comply with. The solutions to coordination problems promulgated by such rules will be clear and practicable because the directions contained within the rules can be understood and complied with. Because laws are prospective they foster predictability and precision, and they can provide for future and recurrent coordination problems. Solutions selected by the law in response to social problems are provided in a relatively speedy and prompt manner. The existence of interrelated institutions and laws (where the rules define, constitute and regulate the institutions, while the institutions create, define, interpret and apply the rules) means that the law provides an already existing and obvious process by which coordination solutions can be determined. This allows efficiency so that time is not lost in determining how one ought to go about establishing a process to select a solution. A further consideration is that the Rule of Law stipulates that rules or laws should be promulgated. This means that the solutions provided by law are widely known and can be implemented, that the solutions are salient because they are available and that the law itself, as opposed to merely the solutions, is salient.
The values of the Rule of Law promote and equip the law as an obvious and easily settled upon tool for coordinating social action and solving coordination problems. As such the law’s salience means it is able to serve as a focal point on which the interests of individual members in a community can converge to allow cooperation.
In addition to efficiency, certain features of the law contribute to its salience by motivating compliance with social schemes of action coordinated by the law. They motivate compliance by providing individuals with an interest in participation by virtue of the fact that it is the law and not some other process that has facilitated the cooperation. That is, they provide an interest in the regular, impartial upholding of the law itself. Finnis calls such an interest a ‘shared interest’, and argues that it provides people with a moral reason to collaborate with the law’s co-ordination solutions even though there might be disagreement about the form of cooperation or detriment as a result. This interest is contained in the procedural and formative features of the law that are derived from the values of the Rule of Law, and by the 'good of there being and continuing to be ... a fair method of relating benefits to burdens and persons to persons over an immense, wide complex and lasting ... set of persons and transactions' .
Finnis argues that the values of the Rule of Law (and the law’s institutions for devising and maintaining solutions) secure fairness because the solutions are general, stable and practicable, and are imposed in a manner that minimizes arbitrariness or self-interested or partisan deviance. Further, the law presents itself as a seamless web from which a wide range of benefits and burdens are offered to the citizen. People who receive the benefits from the law also participate in the burdens that it imposes and cannot pick and choose among the law’s prescriptions and stipulations. In this, there is a fair means of relating benefits to burdens within the community. As a result, all the subjects of the law are put in a like case because they are linked to one another by a network of protections and benefits in which they all share but that are achieved through the imposition of restraints and burdens on all. Finnis argues that only the law is able to do this.
Thus, Finnis argues, the law fosters fairness (a.) as a process of coordination, (b.) through general qualities common to all coordination solutions it provides and (c.) through the fair relation of benefits to burdens it achieves throughout the complex matrix of societal members and activities. Finnis argues that anyone who values fairness has an incentive to introduce norms that protect and foster fairness. This includes the existence of norms both in the coordination of social action (the process itself) and in the solutions opted for (the outcome of the process). Hence, due to an interest in fairness, a person can acknowledge the moral authority of law as a decision-making procedure even when it does not maximise the satisfaction of that person’s preferences.
It is can be difficult to obtain a clear and precise understanding of Finnis’ explanation as to how and why certain features of the law lend to its aptness at coordinating social action and establish its moral authority. However, this is not to say that it is impossible to understand Finnis or that Finnis is wrong. In a number of instances, Raz misunderstands Finnis. He thinks that Finnis argues that because the law presents itself as a seamless web such that its subjects are not allowed to pick and choose then this is how we ought to take it. Raz correctly asserts that one cannot prove the existence of an obligation by providing a proof that assumes the existence of the very obligation one is trying to prove, but he incorrectly claims that this is what Finnis does. The crux of Finnis’ argument is not that the law is authoritative because it is treated as authoritative or that the law is authoritative because it presents itself as authoritative, rather, the law is authoritative because it achieves for society common goods that are morally obligatory and does so in a way that will significantly factor in the practical reasoning of the individuals within a community.
It is true that with regard to the ‘authority of rulers’ Finnis claims that because most people will acquiesce to someone’s exercise of authority then this justifies the exercise of authority subject to two riders. These two conditions are (a) that the ruler’s rules promote the common good and (b) that while consent is not the same as acquiescence nor is it necessary, consent will provide a good rule of thumb in deciding who has authority. Finnis, however, does admit that this argument is 'only presumptively sufficient to justify the claim to and recognition of authority'. Concerning the authority of law, as opposed to a ruler’s authority, Finnis offers us an altered and more substantive justification of the obligation to obey.
One of the main arguments raised by Raz in response to Finnis is that when an argument for the moral authority of law is based on natural law reasoning (that there is an obligation to obey when the laws of a legal system are essentially just or because all legal systems are morally good) then an individual’s reasoning as to whether or not to obey the law is affected only to the extent that his or her disobedience will undermine the law. There are many instances in which disobedience to the law will go unnoticed - it will have no detrimental or undermining effects at all on either fairness or the law itself. Breach of the law will in such instances be insignificant. Raz lists as examples undetectable traffic and tax offences. In cases such as these, Raz states that the argument of setting a bad example fails to apply and as such it fails to satisfy the requirement of generality.
It is on this basis that Raz also rests his objection to arguments derived from considerations of fairness (that it is unfair to enjoy the benefits derived from the law without sharing in its burdens). Raz argues that it is not unfair to make innocuous violations of the law that are harmful to no one. That law’s object in securing fairness is not frustrated in some instances of breach means that there is no general peremptory reason to obey the law. In addition to this Raz also maintains that the argument is of dubious validity when one has had no choice but to accept the benefits. Most people do not make a voluntary and conscious decision to commit to the law on the understanding that they accept both the benefits and burdens imposed by the law.
Finnis responds to Raz by arguing that the point of law is to maintain real fairness between members of the community and that this object is not affected by innocuous or covert breaches of the law. A person who breaches the law so as to avoid unfairness in the outcome of the breach will still act contrary to the requirements of fairness contained within the legal process of social coordination. Refusal to conform to the stipulations of law even though the intended breach may be undetectable is a failure on the part of the individual to participate in a fair manner with and for the common good. The law recognises possible forms of association that meet at least some of the requirements of fairness and this contributes to the well being of the members. This is by definition the common good itself (the participation by a multiplicity of persons in a manifold of goods). Given this Finnis correctly states that 'the law creates a prima facie generic moral obligation and thus has prima facie and general moral authority'.
What Raz fails to challenge is Finnis’ account of fairness and the common good (both in and of themselves and their relationship to practical reasoning) upon which rest the success of Finnis’ defence and the inadequacy of Raz’s critique. This is the real point of difference between the two. Finnis rejects consequential and utilitarian approaches to fairness, practical reasonableness and in determining moral duty. Raz, on the other hand, assumes, although not explicitly, an emaciated account of fairness and public reasonableness.
For Finnis, fairness, practical reasonableness and philia politike have intrinsic as opposed to merely instrumental value. Practical reasonableness as a basic aspect of the human good and fairness as a requirement of practical reason are 'not like the definitive objectives of particular objects but are participated in'. An important implication of this for Finnis’ argument on the moral authority of law is that the achievement of human good is itself valuable. Finnis considers it a mistake to consider the achievement of the common good or human goods as a process or recipe akin to that of cooking a pancake, which may be merely a means to an end and which has only instrumental worth. In the case of the common good, the end is not external to the process required to produce. By definition the common good requires the participation by a multiplicity of persons in a manifold of goods. That is, the common good includes the very means adopted in order to achieve it. This is why fairness within the process of coordination as well as in the outcome can be valuable to an individual and factor in their practical reasoning. It is also why the law itself (within a system embodying the Rule of Law) has intrinsic value. One can judge the legal order itself to be a good. For this reason there is a sufficient shared interest in and reason for cooperation with the law even when it might be breached without detection.
For an individual to act contrary to the requirements of law and fairness would be to act contrary to practical reasonableness which is 'reasonableness in deciding, in adopting commitments, in choosing and executing projects, and in acting in general'. Finnis lists practical reasonableness as a basic good and notes that there is no one first principle or requirement of practical reasoning but that there are many. One requirement of practical reasonableness is that in acting and in reasoning how one ought to act (practical reasoning), one should account for the promotion of the common good of the community. Another requirement is that of impartiality towards human goods and persons so as to avoid bias in one’s practical reasoning. This is because basic goods should not be arbitrarily devalued or overvalued. In relation to human flourishing each is fundamental and they are incommensurable. As such there is no objective hierarchy between them and one should not be considered merely instrumental to the pursuit of another. Unless discriminated between on the basis of a subjective scale of assessment (for example, according to one’s taste or time constraints etc) it is irrational to arbitrarily discount or devalue a basic good in one’s practical reasoning. In regard to impartiality towards human beings, although reasonableness allows for self-preference in one’s practical reasoning, this is limited to the extent that it is necessary for one’s own participation in the basic goods. Practical reasonableness requires that one’s reasoning be free from egotistical or group bias.
A practically reasonable rationale means that one has reason to choose procedural fairness over disregard for fairness and the law and to choose the common good over self-interested self-gain. It enables one to recognise the intrinsic and structural role law plays in facilitating fairness and social coordination. For social coordination and the achievement of morally obligatory social goals to occur there must exist within society members who have as their internal viewpoint the practically reasonable rationale. Such members have as their concern the function of law as an answer to real social problems because practical reasonableness involves a regard for the promotion of the common good. Raz, on the other hand, excludes from his analysis the significance or necessity of such an internal viewpoint and this is in part why he fails to adequately critique Finnis.
Finnis demonstrates the necessity and significance of having practically reasonable members of society with reference to game theory and social choice theory. One difficulty in social coordination that has been identified by social choice theory is the occurrence of intransitives. Arrow’s theorem, which Finnis considers to be the pinnacle result of social choice theorizing, is that the attempt to amalgamate the preferences of individuals in order to decide upon a course of action can result in an intransitive outcome such that no salient or meaningful solution is identifiable. The problem of intransitivity applies not only in social choice (in determining what society should do) but also in individual choice since it may arise whenever one has several interests or standards of evaluation or criteria of choice and is choosing between more than two alternatives. As such, individuals when faced with a multitude of possible goods or solutions to select from may not arrive at a meaningful or salient preference of solution. Finnis argues that Arrow’s theorem causes considerable limitation on the usefulness of game theory to legal theory, and refers, in particular to the game theoretical Co-ordination Problem and the Prisoner’s Dilemma.
The game theoretical Co-ordination Problem assumes that players have ranked preferences and that these rankings are complete and transitive. Arrow’s theorem indicates, however, that both individual preferences and group preferences will be intransitive. Therefore, it is possible that in reality (as opposed to the abstract application of game theory), there will be no convergence of interests, that is, no one salient solution because no meaningful solution is identifiable or because the players prefer no solution to any of the available solutions.  This can be avoided with the introduction of enforceable norms such as those requiring fairness so that there is a shared interest in achieving and promoting fairness. However, fairness in outcome is not necessarily guaranteed even when fairness is preferred or when fairness is procedural. The resolution of coordination problems requires a concept of fairness in which fairness in coordination itself (the process) is valued in addition to merely fairness in the outcome.
Finnis applies a similar argument to the Prisoner’s Dilemma. In this Dilemma the effort of each individual to maximise his or her individual gain means that each individual is in the end worse off than if they had acted in cooperation. All cooperative arrangements, however, are unstable because each individual has incentive to renege on their cooperation for greater personal gain. In the real world, stability of solution is achievable given that individuals can hold a concern for the well-being of other individuals, prefer a fair outcome and employ practical reasonableness in selecting this outcome. Elaborating on this, Finnis writes that:
In the real world no one’s preferences and choices need be determined by the emaciated rationality of self interest. Participants in the real world may prefer fairness to a marginal or even to a substantial increment in the advancement of their own interests; and if each participant is willing to guide his choice by considerations of fairness, the dilemma can be resolved and the higher equilibrium attained. Players with a preference for a fair outcome to the game do have incentive to introduce laws or other norms ... as a path to... fairness in the playing and outcome of the game itself.
It has been suggested that Finnis’ approach to the obligation to be law-abiding might be taken as suggesting that that there is always a moral obligation to obey the law, regardless of individual calculations of whether in a particular case obedience or disobedience would better promote the common good. Raz also criticised Finnis on the basis that there are considerable risks in an indiscriminate acceptance of the law and blind obedience to it. Finnis, however, does accept that in some instances one will better secure fairness in breaching the law rather than in compliance with the law. This does not mean, however, that there is no general obligation to obey. The authority can be overridden by countervailing reasons but this is not to say that the obligation never existed in the first place. Of course, it should also be noted that the law that Finnis defends as being morally authoritative is the legal system that embodies the Rule of Law. By Finnis’ definition this necessarily entails a legal system that is in 'good shape' (just and good). Finnis admits that not every law within such a system will be just but argues that overall, the law is. His claim is not that every law is just or that every solution offered by the law as a means of resolving some coordination issue will, on its merits, be the best solution available. Rather he asserts that a legal system embodying the Rule of Law will, on the whole, be just because it secures a fair distribution of benefits and burdens among citizens, and presents general and practicable solutions to coordination problems in a manner that minimizes arbitrariness or self-interest.
Finnis argues that the law is instrumental in securing morally obligatory goals and that, as the most efficacious process for coordinating social action, it is itself morally obligatory. Finnis claims that the values of the Rule of Law promote and equip the law as an obvious and easily settled upon means of coordinating social action. That is, they make the law salient. This allows it to serve as a focal point over which the interests of individual members in a community converge. The salience of the law is also attributable to the fact that citizens have a shared interest in the upholding of the law. This is because it fosters fairness by bringing about a fair relation of benefits and burdens among societal members and through the general qualities common to all coordination solutions that it provides. Finnis argues that due to an interest in fairness, an individual can acknowledge the moral authority of law as a decision-making procedure even when it does not maximise the satisfaction of his or her preferences. This, he states, distinguishes the law from all other means of furthering social cooperation.
Raz does not set out to conclusively prove that there is no moral obligation to obey the law. Rather he attempts to disprove those arguments that claim to establish the moral authority of law. In response to the claims made by Finnis, Raz argues that an individual’s reasoning as to whether or not to obey the law will be influenced only to the extent that his or her disobedience might undermine the law. Given that it is not unfair to make innocuous violations of the law that do not harm anyone and which do not undermine the law, there is no general peremptory reason to obey the law.
Although at first blush Raz’s argument appears persuasive, its does not undermine or discredit Finnis’ account. Finnis argues that the achievement of fairness is itself valuable and that the law, in promoting fairness, is therefore of intrinsic worth. As such a person who breaches the law in such a way that no one suffers will still act in a way that is contrary to fairness. The value to be found in achievement of fairness means that there is, as Finnis claims, at least a prima facie obligation to obey the law. Raz’s failure to challenge Finnis’ understanding of the common good (upon which rests his account of fairness) means that he is unable to undermine Finnis’ argument in support of the moral authority of law.
[*] BSc (Hon), LLB (UNSW), Solicitor (Sydney), Masters candidate (School of Philosophy, ACU); Lecturer in Law, University of Western Sydney. This paper was written as a result of a research scholarship awarded by the James Madison Program in American Ideals and Institutions, Princeton University. I would like to thank Robert P. George and the James Madison Program for their generous assistance and support.
 J Raz, The Authority of Law: Essays on Law and Morality, Oxford, Clarendon Press, 1970, 233.
 See R Sartorius, 'Political Authority and Political Obligation' (1981) 67 Virginia Law Review 3, AJ Simmons, Moral Principles and Political Obligation, Princeton University Press, 1979, ME Smith, 'Is There a Prima Facie Obligation to Obey the Law?' (1973) 82 Yale Law Journal 950, and AD Woozley, Law and Obedience: The Arguments of Plato's Crito, University of North Carolina Press, 1979.
 Raz, n 1.
 See WA Edmundson, The Duty to Obey the Law: Selected Philosophical Readings, Rowman & Littlefield Publishers, 1999, RP Lawry, 'Ethics in the Shadow of the Law: The Political Obligation of a Citizen' (2002) 52 Case Western Reserve University 655, and SH Tan, 'Validity and Obligation in Natural Law Theory: Does Finnis Come too Close to Positivism?' (2002) 15 Regent University Law Review 195.
 J Raz, 'The Obligation to Obey: Revision and Traditionalism' (1984) 1 Notre Dame Journal of Law, Ethics & Public Policy, 248 and J Finnis, 'The Authority of Law in the Predicament of Contemporary Social Theory' (1984) 1 Notre Dame Journal of Law, Ethics & Public Policy 115, 117.
 J Finnis, Natural Law and Natural Rights, Oxford, Clarendon Press, 1980, 117.
 Raz, n 1.
 Raz, n 5 at 248.
 Ibid 233.
 Ibid 234.
 Ibid 235. A peremptory reason is a dependant (first order) reason that is also an exclusionary (second order) reason where what is excluded are the antecedently existing reasons on which the dependant reasons depend. Finnis agrees with Raz’s definition of a preemptory reason and with the necessity of such a reason in proving the moral authority of law. Finnis writes that 'A person treats something as authoritative when he treats it as, in Joseph Raz’s useful terminology, an exclusionary reason, i.e. a reason for judging or acting in the absence of understood reasons, or for disregarding at least some reasons which are understood and relevant and would in the absence of the exclusionary reason have sufficed to justify proceedings in some other way' (Finnis, n 6 at 234).
 Finnis, ‘Law as Co-ordination’ (1989) 2 Ratio Juris 97, 101.
 Finnis, n 5 at 133.
 Finnis, n 6 at 232.
 Finnis, n 5 at 134: Finnis notes that the listed solutions are not exhaustive. The solutions include S1- no waste disposal allowed, S2- unlimited waste disposal, S3-unlimited waste disposal for those willing to pay, and S4- limited waste disposal.
 Finnis notes that for S2 this requirement of unanimity of action applies to a limited extent.
 Finnis, n 5 at 135.
 Finnis, n 6 at 233 and Finnis, n 5 at 122.
 Finnis, n 6 at 233.
 Finnis, n 14 at 10. For a more detailed analysis of the theoretical difficulties in coordinating social action see pages 10 to 12 of this essay.
 Finnis, n 5 at 135 and Finnis, n 6 at 248.
 Finnis, n 5 at 135.
 Finnis, n 6 at 249.
 Ibid 249.
 Finnis, n 5 at 135.
 Ibid 136 and Finnis, n 14 at 101.
 Finnis, n 14 at 102.
 Finnis, n 5 at 136.
 The English Oxford Dictionary defines salient as applied to immaterial things as 'standing out from the rest; prominent, conspicuous' and as 'standing out or prominent in consciousness' when employed in a psychological sense.
 Finnis, n 14 at 101.
 Ibid 98: Game theoretical co-ordination problems occur when players have some shared interest that can be obtained only through coordination but when there is more than one means of coordinating activity in order to obtain this interest and when preferences as to which means is the best varies among the players.
 Finnis, n 5 at 136.
 Finnis, n 5, 270.
 Ibid 269: The fourth formal feature of a legal system is that past acts (of enactment, adjudication or the exercise of public and private powers) are treated as giving sufficient and exclusionary reason for acting now in a way then provided for.
 Ibid 268.
 Ibid 270.
 Finnis, n 5 at 136.
 Finnis, n 14 at 102.
 Finnis, n 5 at 135.
 Finnis, n 14 at 101.
 Ibid. An analysis as to the empirical correctness of this claim is beyond the scope of this paper.
 Ibid 127.
 Raz, n 5 at 150.
 Finnis, n 6 at 250.
 Raz, n 5 at 240 and 238.
 Raz, n 1 at 151.
 Ibid. In regard to this objection by Raz, it is enough to note that Finnis does not share the approach adopted by Herbert Hart. While Hart may have based his reasoning in support of a moral obligation to obey the law in the idea of fair-play, Finnis does not. On page 251 of Natural Law and Natural Rights Finnis writes that 'consent as distinct from acquiescence is not needed to justify or legitimate the authority of rulers'.
 Finnis, n 14 at 102.
 Ibid. This paper does not consider the accuracy of Finnis’ account of fairness, practical reasonableness or the common good. Rather what is examined is whether, given the assumptions made by Finnis, Raz is successful in undermining Finnis’ argument in favour of the moral authority of law.
 Finnis, n 5 at 129.
 Ibid 120: Political friendship or good neighbourliness.
 Finnis, n 6 at 104.
 Finnis, n 5 at 121.
 This is important for Finnis given that in reality although the procedural aspects of law will meet some of the requirements of fairness, there is no guarantee that the outcome will be fair.
 Finnis, n 6 at 12.
 Finnis, n 6 at 88 and Finnis, n 5 at 121.
 Finnis, n 6 at 125.
 Ibid 105-106.
 Ibid 92.
 Ibid 14.
 Finnis, n 5 at 124: An example of an intransitive outcome is given by Finnis as being when outcome A beats outcome B and outcome B beats outcome C but outcome C beats outcome A so that there is no meaningful identification of the preferred social choice.
 Ibid 125. Other difficulties arise when the method used to amalgamate individual preferences is a system of voting. Social choice theory has shown, for example, that when all issues are determined by majority vote it is possible for the majority of voters to be in the minority on the majority of issues and for the voters directly affected by the vote to always be in the minority. Furthermore, for any system of voting it is impossible to determine whether the outcome is a result of a meaningful amalgamation of preferences or due to manipulation of the voting process by strategic voting. Finnis argues that the upshot of these findings is that in the real world, legal theory must make its way in a world where fair procedures cannot guarantee fair outcomes.
 Ibid 124.
 Ibid 125. In some instances, regardless of intransitivity issues, some people may simply prefer no solution despite the different solutions available. Even when a preference is adopted, recalcitrance is always possible since individual preferences and cooperation between individuals are unstable. Individual preferences can change owing to the variety and complexity of human goods and, as indicated by game theory, for the sake of further individual gain.
 Ibid 126.
 Ibid 131.
 Finnis, n 14 at 100.
 Finnis, n 5 at 132.
 Ibid 125.
 Ibid 126.
 Ibid 127.
 Sober P, ‘Legal theory and the problem of definition’ (1983) 50 University of Chicago Law Review, 1170.
 Raz, n 1 at 151.
 Finnis, n 6 at 354-362.
 Ibid 270.
 Raz, n 2 at 233.