In his wide ranging and carefully argued book The Long Arc of Legality David Dyzenhaus offers a new general theory of law.Footnote1 The book is a major contribution to jurisprudence. I can only offer here a very partial assessment of this immensely rich work. I focus on its most original contribution, the restatement of the question of ‘what is law’ as a question of personal morality. I do not comment on the book's many other contributions, for example, the fascinating discussion of the rule of law in historical legal orders or a discussion of international law and its legitimacy. I discuss only the general proposition defended by the book, namely that a legal order is one ‘in which the law has de jure authority’ (2).
Whereas many legal theories look at law from the point of view of its effectiveness in creating an environment of stability and social peace, Dyzenhaus looks at it as a personal guide to action and proposes that we cannot understand law theoretically without understanding the moral claims it makes on us. The law's authority exists, Dyzenhaus says, when ‘the officials who speak in its name can answer satisfactorily the “legal subject's question”: “But, how can that be law for me?”’ (2). This simple question is remarkably powerful and complex.
It is a question asked by the subject of law to the office holder who comes to impose a course of conduct on other persons. Dyzenhaus looks at this hierarchical relationship mainly, but not exclusively, from the point of view of the ruled, who needs to consider his options: obeying, hiding or, perhaps, rebelling. Of course, the same question can be asked from the point of view of the ruler: ‘How can I be doing this to you?’ or something like it. The burden of responsibility is equally real for the person who is acting in order to affect the lives of others.
The key idea is that if the subject is satisfied that this can, indeed, be law because it has the required authority, then it can then guide him or her as a reason. Dyzenhaus looks at this guidance function from a first-person perspective, so that law is a reason for me to do something. Law is seen as relevant to practical deliberation precisely because it has some practical weight, in all likelihood a very substantial practical weight. Moreover, implicit in the ‘subject's question’ is a test of universality. If I consider this to be a law for me so that I consider it a reason for my actions, then similarly you will consider this to be a law for you, for the same reasons, and you will know that I consider it a law for me for the same reasons and so on. The purpose of justification will not be satisfied, if a person has been uniquely benefited by a particular ruling, for example, because a judge misunderstood the evidence. The test for the authority of law must be something we can all reasonably accept. The grounds of law are public reasons. In this sense, the law's justification may well build a moral community, where rulers and ruled deliberate together on the basis of premises they can all share, even if they disagree on some outcomes. Dyzenhaus believes this is possible under an appropriate constitutional ideal and calls the resulting community a ‘jural community’.
These arguments take jurisprudence into a new direction. What are the public reasons we can share? Where does the practical weight of law come from? Is it real? Concerned with such questions, the theory of law turns to ethics and political theory, not to history and sociology. It may not be necessary, I don't think, to show exactly how the law relates to other reasons that apply to the subject in all their complexity, for example, his or her commitment to his or her children. But an account of law as a reason with real authority must be able to show that law is in principle a good reason and in many cases a reason with priority over other reasons. The subject's question invites, therefore, real deliberation about what we ought to do. In Dyzenhaus’ account, this is not purely philosophical challenge. It is also an urgent question of practice. It does not emerge from speculative interest but from lived experience. We ask the question ‘Can that be a law for me?’ every time we encounter the legal system of our place and time and certainly when we work as one of its servants. Any legal authority must be able to answer the question in the affirmative, although not every one will be able to do so.
A third theory
This introductory summary may suggest that Dyzenhaus rejects legal positivism in favour of a fully moral theory of law. This is not exactly the case. Dyzenhaus argues that the traditional contrast between legal positivism and natural law, which in his view has reached a ‘standoff’ (1), is misleading. There is room, he argues, for a third theory of law. Such a theory would accept the basic premises of both theories, namely that law is some kind of fact and that law is also a reason, a matter of moral authority. Can this work? At first sight this seems implausible.
The key to the argument is the premise that legal practices change our moral rights and duties. The morality of legality for him is not an abstract set of principles emerging from the pure exercise of reason. Morality, for Dyzenhaus, is a product of historical and political events. He calls this view ‘pragmatist’, and invokes the work of C. S. Peirce in its support (368). He also relies on some of the political writings of Bernard Williams, although his focus is not so much concerned with culture or tradition, or similar ideas that have animated Williams’ ethical writings, but with legal procedure.
Dyzenhaus presents the third theory of law as an amalgamation of the theories offered by Hobbes, Kelsen and Hart. He starts with Hobbes’ theory of the state (chapter 2) continues with Hart's theory of law (chapter 3) and ends with Kelsen's theories of democracy, law and the international society (chapters 4 and 5). It is Hobbes’ take on authority and Kelsen's argument on constitutionalism that inform the book's most substantial arguments. Hart's arguments are somewhat peripheral, compared to those of the other two. Indeed, it is not immediately obvious why the book spends so much time on Hart. Dyzenhaus enlists Hart as a reluctant moral theorist and argues that Hart follows in the footsteps of Hobbes in showing how positive law embodies a political community's ‘moral conscience’. It is a very unusual way of reading The Concept of Law and in my view a rather implausible one. Dyzenhaus has read Hart very carefully and reveals some very interesting things Hart said. But utlimately, at least in my view, this stretches Hart's theory too much. Hart is fully committed to a view of law as fact. His occasional concessions to seeing law as a reason do not fit with his overall picture. But this is a minor objection. The point of the book is not to offer an accurate account of other people's theories, but to enlist the most important or eloquent works of our tradition in illuminating the idea of law by showing those texts in the best possible light. The book is not a commentary on other scholars but a completely new theory about legality, authority and legitimacy.
Nevertheless, because the book engages so closely with the writings of Hobbes, Kelsen and Hart and because it goes back and forth from one theory to the next, it makes heavy demands on its readers, not all of whom will have Dyzenhaus’ masterful grasp of jurisprudence, political theory, European history or constitutional law in various jurisdictions. Yet, this complex presentation repays close study and achieves its aims with both elegance and rigour. The thesis is original as well as attractive: law exists whenever political power is channelled through legitimate processes of legality. Is it persuasive?
Authority
Authority is the book's key idea. Dyzenhaus argues that a theory of the law's authority does three things. First, it ‘explains authority as compliance with fundamental principles of legality’ (2). Second, it ‘gives to officials a role in interpreting enacted law in the light of … principles [of legality]’ (2). Third, a ‘constitutional idea’ sets out the precise tests according to which the authority of the modern legal state is based on the ‘voluntary interaction of the very individuals who are subject to that authority’ (3). For Dyzenhaus authority is built on a constitutional idea that justifies public power when used under an institutional framework to each one of its subjects. The constitution is a complex idea which is to be taken both as a ‘fact’ and as an ‘achievement’ (149 ff.).
The point is developed through a detailed discussion of Hobbes and especially of the way in which ‘natural laws’ constrain sovereignty. Dyzenhaus quotes Hobbes as subjecting law to a test of morality, not merely of effectiveness or power: ‘Law in Generall, is not Counsell, but Command; nor a Command of any man to any man; but only of him whose Command is addressed to one formerly obliged to obey him’ (90). Dyzenhaus reads Hobbes as saying that the authority of law is not based exclusively on coercion but on the ‘social contract between the invdidivuals in the state of nature whose agreement constitutes the state and authorizes the sovereign to act in their name’ (90). The meaning of ‘guidance’ here is one based on reason, not fear of sanction or other consequences.
In Hobbes’ argument, the individual conscience is in effect replaced by the public conscience of the Commonwealth, which ultimately binds not because of the force that backs it, but because of the rational argument that justifies it. A sovereign power that satisfies the tests of the social contract provides sufficient reason for compliance. The emergence of a civil society from the chaos of the state of nature follows a rational progression. Under this view, which Dyzenhaus in principle accepts as a compelling account of the authority of the state, a citizen will recognise that ‘the enacted laws of a civil society are the repository of the society's values which the individuals in that society must take as moral’ so that ‘an individual who is subject to the law of a sovereign should see that “the law is the publique Conscience, by which he hath already undertaken to be guided”’ (91). The idea that the law is an embodiment of a society's values and functions in the way of a ‘public conscience’ is a constant theme in the book.
There are four basic elements to this reading of Hobbes. First, sovereignty is constituted by the processes of constitutionalism, so that ‘the Hobbesian sovereign is a legally constituted sovereign’ (99). Second, the legal conditions of sovereignty include ‘natural lawes’ of reason (100). Third, laws are made by an act of sovereignty only when that act meets constitutional tests of validity which are both procedural and substantive (the ‘validity proviso’) (101–09). Fourth, the validity proviso is also dependent on the ‘legality’ proviso, or the requirement that a legal order requires ‘not only a Declaration of the Law, but also sufficient signes of the Author, and Authority’ (109). Dyzenhaus sets this out in a detailed section which is both exegetical of Hobbes’ writings and an original argument about the analytical sense of ‘legality’ (115–20). Hobbes’ sovereignty is thus morally constituted, in a way that distinguishes it radically from the idea of sovereignty outlined by Bentham and Austin.
Dyzenhaus argues that Hobbes’ views on the law of nature amount to a framework of liberty under ‘public laws’. In this sense, Hobbes’ Leviathan sees law as constitutive of civil liberty:
Restraint … without the which there cannot possibly be any Peace. And Law was brought into the world for nothing else, but to limit the natural Liberty of particular men, in such manner, as they might not hurt, but assist one another, and joyn together against a common Enemy. (119)
Legality
What does ‘legality’ mean? At this point, Dyzenhaus moves away from Hobbes and the social contract tradition. He believes that Hobbes’ argument is correct but incomplete. He seeks further detail in the various tests of effectiveness suggested by Hart and Kelsen. Dyzenhaus argues that a legal order has authority not at the stage when it meets some abstract tests of a social contract theory, but when it also enjoys the real acceptance of those subject to its institutions, so that the law's authority is ‘a hybrid of actual and hypothetical acceptance’ (185). The test ‘depends on the fact of widespread acceptance’ provided that this acceptance relies on legal institutions. Consensus is necessary, but not sufficient.
So, Dyzenhaus argues, ‘the order of the modern legal state is designed to elicit the consent of those subject to its law’ by providing them with the assurance to individual legal subject that the laws ‘make sense to them as serving their interest as equal before the law and the law can make such sense because of its interpretability of the order's fundamental legal commitments’ (185–86). Dyzenhaus argues that laws that are not interpretable in light of fundamental principles of legality lose their claim to authority because they do not bear the marks of having been enacted with legal right and are therefore ‘illegal laws’ (3).
Of course, the very expression ‘illegal laws’ reveals the tension inherent in any argument about law as a moral idea. A standard cannot be constitutive and evaluative of the same thing. How can something be and not be a law at the same time? When the powerful produce statutes and court decisions that accord with the ideals of legality the two standards are harmonious. Law has authority. What happens if reality and ideal begin to diverge? A bad law, we want to say, may not be a law at all. Yet it must be a law if we consider it a bad version of itself. But this is paradoxical. As we shall see, Dyzenhaus’ discussion of this paradox is not simple.
The starting point for any theory of legality as a moral idea must be a clear distinction between legitimate and illegitimate institutions. The relevant standard concerns a certain procedure that makes institutions legitimate. Dyzenhaus argues, relying on the social contract tradition, that institutions are legitimate when they ensure that the legal subjects’ interests in liberty and equality are properly served so that they could plausibly consent to them. This basic idea is explored in detail in a final chapter, entitled ‘Legality's Promise’, which brings together the book's conclusions.
It is important to observe here that most political philosophers that have discussed the social contract have proceeded from a different route, namely ideas of the protection of private property and contract. It is these horizontal relations that lie at the heart of the civil condition in the works of Locke and Kant, for example. Dyzenhaus, by contrast, focuses on relations between individuals and the state. He looks especially at the available opportunities people have under public law to contest the decisions that have gone against them (what in England we call ‘judicial review’). This is an important departure from the tradition, however. The moral content of the civil condition for Locke is the justice it achieves in the relations among citizens, not just the relationship it establishes between citizen and the state. Locke is concerned with the fair terms that must obtain among equal citizens so that they can relate to one another on the basis of justice. Dyzenhaus’ account of equality and liberty is focused on political relations, something which enables him to rely, somewhat more narrowly, on political/legal procedures alone as a key to legitimacy.
Dyzenhaus explains that he does not wish to rely on the general and abstract arguments by Locke or Kant, because his account of political morality is ‘pragmatist’ and that pragmatism is ‘the conception of morality one should adopt if one wishes to explain law in the register of authority’ (354). He has little time for the more abstract arguments that we find in the social contract tradition or its contemporary versions. This seems to me a significant – and regrettable – omission, since a great deal of what Rawls or Dworkin have written could have been of assistance to the discussion of legitimacy. The book, however, does not seek this assistance and charts its own course.
Dyzenhaus argues that an ideal conception of legality is available for us in Kelsen and Hart. If these authors are read carefully, he argues, they guide us into the true meaning of ‘legality’ as a moral ideal. Dyzenhaus looks at these theories in much detail, identifying moral dimensions that have been lost to less attentive readers. For example, Dyzenhaus argues that the appropriate moral content of law and legality is evident in, among other places, Kelsen's theory of the legal order as a hierarchy of norms and Hart's idea of practice-based acceptance (418). True legal rights and duties arise when the requirements of legality are well respected. Only then can the legal order answer sufficiently the subject's question: ‘Can that be law for me?’.
The subject's question
What does the subject wish to know by way of the ‘subject's question’? As I already suggested above, in Dyzenhaus’ argument the question ‘can that be law for me?’ operates as a reminder of moral responsibility. Someone is seeking a justification for a coercive imposition that will burden him, or for an action that will create an imposition on others. This is what Dyzenhaus has in mind as the meaning of the question, but it is not the only possible meaning. There is another, perhaps more likely to arise, version of the ‘subject's question’, which follows a legal positivist understanding of the legal order. For this reading, no moral responsibility need to be involved. If one defines law only as a political fact, something someone has done, then the ‘subject's question’ is just a request for information. Under a legal positivist theory, says Bentham's command theory, law is an event arising from some distinct actions of the sovereign. Under Hart's theory, a rule is a legal rule just if it has been brought about in compliance with a rule of recognition. Under these conceptions of law, the question of whether or rule or decision ‘can be a law for me’ is just a question whether something has happened, if the relevant events A, B, C have actually taken place. No justification is needed.
If these things have happened, then this can indeed be a law for me and you – or, more accurately, is a law for me and you. If Bentham's sovereign has spoken in the right way, through the words of a King or the publications of a Parliament, then there is a law for you and for everyone else. If the terms of the rule of recognition have been met, then we have a law. These are straightforward answers but they are not the answers Dyzenhaus has in mind. He seeks justification, not information.
This point shows us that the ‘subject's question’ does not work as an argument. It merely illustrates a deeper point, whose validity must lie elsewhere. The point is that law is a moral idea. How can we show that? Dyzenhaus does not address this very extensively. One may say that the question is fundamental, it concerns issues of ‘ontology’ or metaphysics, or ‘conceptual truths’ about ‘law’. I am not sure this is true. We can show that law is a moral idea, if we show that the factual answers to the ‘subject's question’ offered by Bentham and Hart are unsatisfactory. Are they? Dyzenhaus offers a simple such argument. He understands the idea of law as necessarily implicated in a moral relationship of imposition, a relationship of domination of one person to another. Dyzenhaus sees the moral relevance of this relationship – not in the start of the book, by certainly by its closing chapters – and writes that the question ‘But how can that be law for me?’ is a question ‘invited by the institutional structure of the legal order’ (356). This must refer to the inescapable reality of the hierarchical relationship between ruler and ruled. The moral salience arises from the structure of this situation. Law must be justified by reasons, because domination calls for justification. How could be otherwise? It is a moral challenge both for the person dominating and for the person being dominated. Because the subject's question is a moral question, it requires a moral answer. Factual answers fail to respond to it.
A self-reflecting agent cannot turn his gaze away from the act of coercive imposition. If we tried to ignore it, we would be abdicating our moral responsibility and undermining our identity as persons. We would have adopted the point of view of the immoralist who does not care, who is guided not by reason but by occasional impulses or emotions.
It is important to note here that Dyzenhaus does not address, at least not directly, Hart's own answer to the ‘subject's question’. He discusses Hart's essay on ‘Commands and Authoritative Legal Reasons’ in the Essays on Bentham at one point in the book.Footnote2 He notes that although Hart had great difficulty accommodating law as a reason he was vividly aware of the problems this caused for the internal coherence of his theory of law. Hart famously spoke of an ‘internal’ point of view and of the fact that valid law was some ground of criticism in case of failure to comply. Dyzenhaus reports that Hart understood that maintaining a ‘two-systems’ view of law and morality led to the ‘paradoxical’ and even confused conclusion that ‘judicial statements of a subject's legal duties need have nothing directly to do with the subject's reasons for action’ (371). This would mean that some true duties, namely legal duties, would not be true reasons. Dyzenhaus argues that Hart's view can be ‘rescued’ from these problems by adopting a ‘pragmatic’ view of morality (371) or a view that he develops later in the chapter of ‘social morality’ (372–82) that allows law to truly become some kind of ‘public conscience’ (383). I don't think this is possible.
Dyzenhaus does not notice that Hart allows for the position of the immoralist, by adopting a permissive account of the required attitudes of the legal official. Hart does not agree that the relevant point of view is Hobbes’ ‘Just Man’, the man that wishes to act well and considers law to be one of the reasons that apply to him. Hart presents the perspective of the judge as something much broader. It includes the deeply egotistical man who pursues his or her inclinations without any plan or reason. Hart writes that the authority of law depends purely on the idea that judges have some comprehensible motivation, whatever that may be, moral or immoral:
individual judges may explain or justify their acceptance of the legislator's enactments by saying that they simply wish to continue in an established practice or that they had sworn on taking office to continue it or that they had tacitly agreed to do so by accepting the office of judge. All this would be compatible with judges either having no belief at all concerning the moral legitimacy of the legislature or their beliving that it had none.Footnote3
Although Hart does not say so directly, he accepts that a judge may well treat the litigants as mere means to his private ends. The immoralist judge will look at the law merely as a fact and issue orders and impose coercion without considering the effects they will have on other people. In this sense, at its most basic, the Hartian view does not see law as an authoritative reason, to be considered alongside other reasons, but only as a relevant event, a tool we make use of to achieve some other ends. Since law is not a reason, it can be deployed without trouble by the person who does not act on reasons. Hart's theory is thus quite different from that of Kelsen's, for whom a judge or lawyer merely ‘presupposes’ the validity of law. This act of ‘presupposition’ shows that the lawyer is here fully aware of the role of law as a reason for action.
Hart's theory, by contrast, is clear that the tests for law are factual, not practical. Something is law because something happened and that is all. Of course, as Hart himself observed, this view sits uneasily with another observation that Hart made about law, namely that law is to be assessed from an ‘internal point of view’. This position creates a tension at the heart of legal positivism. It seems that Hart's theory both takes law, at times, to be a reason (from an ‘internal point of view’) and also, at other times, not to be a reason at all (because the ‘internal’ point of view is supposedly factual and not moral). I believe that the effect of Dyzenhaus’ argument is that Hart's theory must be abandoned as incoherent and paradoxical. Dyzenhaus takes a more generous view. Yet, his attempt to rescue Hart seems to me both futile and unnecessary. Indeed, one of the most interesting aspects of the book is the reconstruction of Kelsen's pure theory of law as a theory of law as a reason for action and the suggestion that the idea of the ‘presupposition’ of the ‘Basic Norm’ that grounds the validity of law is an attempt to answer, in a way, the ‘subject's question’ (see, e.g., 171–72).
The priority of law
A very important idea that Dyzenhaus develops is that of the priority of law over other reasons. When understood as a moral question, the ‘subject's question’ is extremely difficult to answer. How can any coercive imposition be justified morally? The difficulty has persuaded many theorists that law has no authority whatsoever. Joseph Raz, for example, argues that there is no obligation to obey the law because it is the law – whereas there may be reasons to obey because of its good or beneficial content.Footnote4 Such content-specific accounts, or ‘success-accounts’, of political obligation are indeed the most obvious response to the ‘subject's question’. One could respond by saying a parliament's statute or a court's ruling can be justified only if they fully accord with the inclinations or preferences of the person burdened by it, so that they promote his or her interests. If one is a Christian, for example, only a law inspired by Christian doctrine – in the right way – will accord with one's inclinations and preferences. A socialist law can be law to a socialist, a liberal to a liberal. And so on.
But this is precisely the problem that the law wishes to solve. The members of any community will have radically different standpoints to which they will be reasonably committed. There is no way that a legal order could serve all of their true interests or satisfy their wishes all of them at once. Social cooperation cannot survive such fragmentation.
The problem is not just the fragmentation of commitments itself, but also the relative strength of the reasons that any person considers in principle binding on him. Law must enjoy a certain priority over all other reasons under a simple duty of citizenship. Let us return to our socialist. Assume he is living in a conservative state full of inequality. He may recognise that law is some kind of prima facie reason for him, but not a strong enough reason to motivate compliance against his deep commitment to equality. Why should the socialist answer the ‘subject's question’ in the affirmative in this case, when the laws are deeply unequal in a way he deeply resents? The same applies to the Christian. How can that person give up the deepest elements of his faith in order to comply with the desires of a secular majority? The examples can easily proliferate. What can there be about law that makes it so important?
The idea that law enjoys such a priority as a reason, that it can be a reason that overrides not only personal preferences but also deeply held convictions is extremely difficult to substantiate. At the same time, it is central to the theory of law as a moral idea. Dyzenhaus acknowledges this problem and, once again, adopts the answer offered by Hobbes. As is well known Hobbes thought that when you bring your personal conscience to the public domain, you are bound to destroy the commonwealth. Cooperation under an idea of sovereignty aims to prevent this from happening by silencing somehow the plurality of individual views or pushing them to the private domain. Hobbes outlines a set of political institutions that when in place provide us with general reason for compliance. This is why Hobbes says that law replaces private conscience with a ‘public conscience’. It is not force that achieves this, but force coupled with persuasion about the need to enter a civil condition.
Dyzenhaus makes the same argument. He says that legality creates a higher level reason for us, which rises above our immediate preferences and takes priority over them. Whenever the law is being determined by an institutional framework that meets certain tests of ‘legality’, it becomes authoritative not on the basis of its contents, but on the basis of the procedure that produced it. It is the procedure that creates the overriding reason. The fact that such a procedure is in place answers the ‘subject's question’ to any reasonable persons’ satisfaction. In this sense, political legitimacy is even more important than your faith.
So just like Fuller argued for some kind of ‘internal’ morality of law, Dyzenhaus argues for a similarly authoritative weight of ‘legality’ that focuses on procedure, predictability, reciprocity and accountability. In the course of this argument, Dyzenhaus cites this interesting passage from Hart's The Concept of Law, which shows support for Hobbes’ procedural argument (374):
It seems clear that the sacrifice of personal interest which such rules demand is the price that must be paid in a world such as ours for living with others, and the protection they afford is the minimum which, for beings such as ourselves, makes living with others worth while … These simple facts constitute … a core of indisputable truth in the doctrines of Natural Law.Footnote5
Dyzenhaus goes on to explain that legality justifies sacrificing our personal interest because of the value of the constitutional procedure that makes us equal citizens. He defines a ‘legal subject’ as someone that occupies an equal place in a ‘jural community’:
The legal subject is a construction whose personality informs the whole design of legal order. Legal subjects are the individuals in the jural community whom particular laws must address by offering reasons for action which make sense to them as free and equal before the law. The process of converting public policy to legal form has the political point of making it possible for the justification of that policy to be undertaken in just this fashion. (404–05)
So Dyzenhaus ultimately distances himself from legal positivism and rejects Hart's ‘static’ view of law and his strict ‘separation’ thesis and argues:
Put positively, one must accept that principles of legality or Recht are intrinsic to legal order, that they are moral as well as legal and that for legal order to be maintained in good shape, officials must manifest their fidelity to law by interpreting the positive law in light of such principles. (378)
In what sense are the principles of legality ‘legal’ as well as ‘moral’? They have to be exclusively moral in their capacity as reasons. Isn't there only one domain of practical rationality in which reasons make sense as grounds for justification, namely the domain of morality? Legality can only make sense as a moral test. If the principles of legality are a test for the existence of law, they cannot be themselves ‘legal’. And if legality is a constitutive test for law, how can it work to maintain law in a ‘good shape’? This sets the bar too low, by suggesting that legality is merely an evaluative test, not a constitutive test. The book's argument has been that legality makes the law. There is no law without it, not even law in a ‘bad shape’. This sounds like the legal positivist way of speaking, because it takes respect for the rule of law to be something qualifying the law's existence, not defining that existence.
Political realism
These ambiguities are not the result of some kind of oversight. They are part of Dyzenhaus’ subtle argument throughout The Arc of Legality to the effect that law is both a moral reason and a political event. The principles of legality do seem to be applying to themselves. This is why, I believe, the theory is seen as a third theory of law and not a full-blown moral theory, in the way of Dworkin or Kant. For Dyzenhaus, legality is a reason for the existence of law, but it is also a contingent creation from law's practice.
The way these ideas are combined becomes much clearer in the section ‘The Autonomy of the Legal’, where Dyzenhaus relies on some arguments about legitimacy made by Bernard Williams.Footnote6 I think that the idea of ‘autonomy’ here is invoked as autonomy from moral abstractions and generalisations. Dyzenhaus relies on some arguments made by Bernard Williams’ about ‘realism’ in political theory and about the fact that legitimacy ‘is a matter of the actual acceptance of a political order's authority by those subject to it’ (393). Acceptance, of course, is just a fact or event. Here the argument endorses some of the legal positivist emphasis on historical fact. It follows, for Dyzenhaus, that in the ‘realist’ argument pursued by Bernard Williams:
what counts as legitimate in a particular political order is relative to what is accepted as a matter of fact in that order, not what some theory has determined as the moral standards which any order has to meet to count as legitimate; for example, and as many liberals would argue, that its laws respect some predetermined list of individual rights. (393)
Dyzenhaus’ scepticism about the role of ‘critical’ or abstract morality is a central element of his argument. He notes that
positivistic legal theory helps to show that the critical theory principle can be given an institutional expression which both explains the role of moral principle in a political realism and why the line between acceptable and in-fact-accepted reasons should be blurred. (395)
What are we to make of this argument? The key point is that political events shape the relevant sense of legitimacy. This is where Dyzenhaus and Williams depart from Rawls and Dworkin (and presumably Locke and Kant). The point is not that acceptance or obedience by a community is one necessary ground for legitimacy. All theorists accept this. Dyzenhaus goes a few steps further. He says that acceptance shapes the test. The test is, partly theoretical but also partly social. This is a far more controversial premise.
One can see its attraction. We can think of instances where a shifting account of legitimacy makes sense in the way that some de facto allegiance creates new understandings of de jure authority. A case like that may be the institutions of the United Kingdom. Here an unwritten constitution has changed over time, on the basis of important political events, for example, the adoption of a secret ballot for parliamentary elections with the Ballot Act 1872 or the expansion of the suffrage by various acts in the nineteenth century, or accession to the European Community in 1973. With its many ambiguities (and with the incongruous idea of the ‘Crown’ at its heart), the unwritten constitution of the United Kingdom would not be something that one would put forward as a model of constitutionalism. Yet, its strength lies in the fact that it changes with the times. This has enabled the British nations to cooperate in an atmosphere of stability and trust. That kind of stability is of course one argument for the legitimacy of the flawed institutions in place. So what counts as a legitimate institution in Britain has changed because of the historical record.
This story, however, does not suggest that the tests for legitimacy have changed. The example is equally consistent with the view that the tests of legitimacy have remained constant. Even when one proceeds from a ‘non realist’ account of legitimacy as a purely abstract or philosophical ideal, the relevant tests can be met in many different ways. Abstract principles are open ended. Equality and liberty mean different things today compared to even ten years ago. Just think of the recognition of equality in marriage and adoption. It follows that equally legitimate political institutions can have different historical manifestations. A legitimate constitution may introduce, for example, a presidential system on the basis of a written constitution in France, but a parliamentary system on the basis of unwritten law in Britain. The reason for these differences lies in the political history of these places. Different political communities have made different choices. But this does not mean that we are deploying different tests for political legitimacy in each case, one set for France and another for Britain. The tests are the same, yet they require that we pay attention and give due recognition to history and practice. Public justification relies partly on history so that the same tests of public justification produce different results.Footnote7 In France Thomas Paine's ideas won, in Britain Edmund Burke's ideas did. This is not ‘realism’, but the very ordinary application of an abstract standard to real life.Footnote8
It seems to me more plausible therefore, that the reason why convergence in conduct contributes to legitimacy is not, as Dyzenhaus says, a recognition of the power of facts over the role of ‘critical morality’, but a distinctive moral test that assesses political institutions on the basis of their actual stability. The fact that particular historical practice has weight, is a general and abstract political principle.
In political communities that need to build some effective cooperative principles, legitimacy tracks common practice. The liberal political philosophers do not dispute this. John Rawls has written for example that the ‘liberal principle of legitimacy’ requires that
Our exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in the light of principles and ideals acceptable to their common human reason.Footnote9
Abstraction and context
Τhe last section of the book is called ‘The Individual Legal Conscience’ and returns to a broad discussion of the ‘subject's question’. The argument appears to qualify somewhat the earlier endorsement of ‘political realism’. Here Dyzenhaus returns to the perspective of the acting person. He starts his discussion by drawing attention to Hobbes’ example of the person, he calls him the ‘Foole’, who ‘asserts that even in an established civil society he is entitled to disobey the law when he calculates that disobedience will better serve his interests’ (412–13). Following Hobbes, he rejects the ‘Foole's’ attitude and endorses an account of the ‘Just Man’ who ‘has a conception of the justice of the law which permits him to criticize it from the internal point of view, from the perspective of the law's own standards’ so that only when ‘law by and large meets those standards that the Just Man will think that he should obey the law despite the fact that he disagrees with its moral content’ (413). The Just Man does two things: first accepts that he is responsible for his decisions and, second, considers that in some ways the institutions and procedures of law – because they meet the tests of legitimacy outlined by legality – constitute a general reason for action, so that particular laws may well take priority over our private views.
Dyzenhaus claims that both ‘legal subjects and legal officials’ must share this ‘frame of mind’, i.e., of the ‘just man’, recognising the authority of both domestic and foreign law. The idea of the ‘individual legal conscience’ means that ‘in the last resort individuals must decide for themselves whether to obey the law … ’ (417). In effect, Dyzenhaus presents here the legal point of view as a directly ethical one. The legal subject is engaged in genuine reflection on what he ought to do, thinking of his or her responsibility to other agents. We may recall here that this is something that Hobbes had called ‘private conscience’ and considered it very dangerous for the life of the commonwealth. This is because when one is reflecting about their own responsibility, they may well reject the choices that the political institutions have made for us. Our conscience may rebel. There is no guarantee that one's rational reflection will give its consent to the choices of any office holders. As we saw Hobbes saw in this the seeds of revolution, but he did not believe it can be eliminated. Instead, he offered ultimate authority of the sovereign as the most persuasive argument, in light of the natural laws of reason. The subject is thus persuaded to comply by both force and reason.
Dyzenhaus’ attitude is similar to Hobbes but not identical to it. He responds to the dangers of conscience by grounding the ideal of law on a rational argument about equality and liberty. As we saw above, he deploys the idea of the equality of all citizens as a requirement for the formation of a true ‘jural community’, one where ‘subjects are able to get adequate answers to the question “But, how can that be law for me?” because the answers do not undermine their status as equal before the law’ (417). This is, fundamentally, how the subject's question is effectively answered. The equality of human beings is the key ingredient of legitimate institutions.
This argument, however, is fundamentally based on abstract ‘critical morality’, namely the equality of all human beings and the challenge of showing equal concern and respect for them. It is not based on political ‘realism’. Although Dyzenhaus does not comment on this feature of his answer, the principles of equality and liberty introduce an element of universality in his own argument. They are abstract arguments, not lessons from cultural practice. Dyzenhaus calls this an ‘inner morality’, which is exhibited by a legal order (and is explored in detail in chapter 5). This may well be true. But a further point remains also true, namely that it is the same morality whether we call it ‘inner’ or ‘external’. It still binds us in exactly the same way.
One may recall that Dyzenhaus has already relied on Hobbes’ account of political liberty and natural rights which creates a legitimate ‘civil condition’ where persons have given up their wild natural liberty for the sake of civic liberty:
For Hobbes, then, the point of individuals consenting to lie under the condition of an all-powerful sovereign is to move from the natural condition in which the only freedom they can enjoy is a worthless pre-political freedom to the civil condition in which they enjoy civil liberty. That point requires not only that the sovereign has a monopoly on coercive power but also that when it exercises that power, it does so through law – by putting in place an order of public laws. … The law also constitutes the spaces of civil liberty, which are unavailable in the absence of law, and which are put at individuals’ discretion, thus permitting peaceful interaction. (119)
These points seem to compete with the idea that legality is determined by a community's ‘public conscience’ under a conception of ‘political realism’. There is a tension here between the two arguments: that the terms of legality are the results of obedience under a shared ‘public conscience’, on the one hand, and the argument that they are determined by the laws of nature, on the other. It is possible that a shared public conscience will come into conflict with the laws of nature. What happens then? I believe the book's argument (in Chapter 5) is that the legal order crumbles or changes radically. This means that the work is being done by the principles themselves, not their manifestation as a form of ‘public consciense’.
Dyzenhaus appears ambivalent about this when he says – at the start of the book – that for the Hobbesian natural law theory the relevant moral principles ‘are not to be found in some natural order of things which tends towards some kind of telos or end, as thinkers from Aristotle through Aquinas to John Finnis have supposed’ but they ‘are entirely internal to a legal order in that no legal order worthy of the name could fail to instantiate them’ (16–17). When Dyzenhaus sets out to explain the law's authority in his final chapter he does not recall Hobbes’ abstract ‘laws of nature’ but relies on the very different argument that the tests emerge from the social manifestation of underlying values through ‘public conscience’:
I have thus argued in this book that the law of a modern legal state does amount to a public conscience, for it contains the fund of the society's values which have been transformed into law. On this view, the law is a part of our moral fabric. (421)
The key to this disagreement is the relevance of abstract principle. I think that Dyzenhaus is still fundamentally committed to such abstract principles. As the book so clearly shows, legal structures often fail to express true moral values, or Hobbes’ ‘laws of nature’. Chapter 5 offers a detailed account of the troubles of the South African legal system as well as the collapse of the Nazi legal system. Indeed, some of the most powerful pages of the Dyzenhaus’ book recount how some lawyers have struggled with their conscience practising law in a highly unfavourable environment. In what sense can we then say that these places had formed a proper ‘public concsience’? Dyzenhaus does not suggest that they had. Obedience to a defective legal order does not change the nature of the principles of legality.
One of the most telling examples of the book is the experience of Ernst Fraenkel, the Jewish lawyer who practised in Nazi-era German courts in the 1930s and wrote a masterful account of the law of dictatorship.Footnote12 Another one was the human rights lawyer Michael Sfard who practised law in the Occupied Territories and wrote about his own experiences.Footnote13 Elegantly retold by Dyzenhaus, these powerful stories of courageous lawyering exhibit the many ways in which those who participate in the machinery of the legal system confront their own ethical responsibility in the course of their practice. This is the subject's question in action, so to speak. Neither Fraenkel nor Sfard would have accepted that they had an obligation to respect any discriminatory rules purely because the majority had said so, or that racism could be a ‘public conscience’ of their states. They often had to accept that the legal process provided important opportunities to their clients, knowing that other parts of the institutional framework were obscenely unjust. They did not believe, though, that a ‘public conscience’ resolves the matter.
So the best way to read the book, in my view, is as a refinement of Hobbes’ argument about law. Hobbes said that only a legal order that respects the ‘laws of nature’ enjoys any authority at all. I think Dyzenhaus agrees. Power – or practice – is necessary but not sufficient. But this does not seem to me a ‘third theory’ of law, as Dyzenhaus says at the start of the book, but a fully moral theory, relying on a conception of critical morality coupled with attention to the principles of political legitimacy. Dyzenhaus moderates that point when he says that instead of abstract principles he opts for a ‘pragmatist’ account of moral inquiry (368). Peirce's pragmatism for Dyzenhaus entails that law is a kind of ‘laboratory’ of morality, which emerges through practice. So he argues that ‘much of our compulsory public morality – the morality we feel is settled and important enough that it was made into law – is no more than the judgments which have survived the tribunal of experience and inquiry’ (368). This metaphor, however, cannot work in the way Dyzenhaus occasionally speaks of it.
The idea of a ‘laboratory’ seems to be the wrong metaphor for ‘political realism’. In a scientific laboratory, we conduct experiments that are subject to the truths of the natural world. An experiment that we conduct a hundred times, will fail a hundred times, if it does not track the laws of natural causality. The truth of the laboratory is entirely independent from our practices. Usage or custom or ‘culture’ in repeating our experiment change nothing. There is no room for ‘realism’ here. If this is so, then the idea of a laboratory points to the opposite direction: namely, that by way of trial and error political institutions are seeking to arrive at general truths that are independent of our practices. We seek to agree on a reasonable approximation of legitimacy, constantly revising and improving our institutions in light of our ideals. This brings us back to Hobbes’ true ‘laws of nature’, who remain our guides. In my reading, therefore, the best reading of the book's argument – and perhaps more truly consistent with Dyzenhaus’ theoretical commitments – is that Hobbes’ argument about the critical rationale of the civil condition is the most central argument of The Arc of Legality.
Jurisprudence as moral reflection
This is the book's most powerful point: when we ask ‘what is law?’ we are asking a moral question. We are taking the point of view of moral responsibility or moral appraisal of the institution that makes claims on us. This is why the best form of the question is the subject's question: ‘How Can that be a Law for me?’ We can look at this question either from the point of view of the person creating the law or from the point of view of the subject that receives it. It makes little difference. The answer will be along the same lines. Just like a thoughtful person would not wish to impose unjust constraints on fellow human beings, similarly a thoughtful person would not accept becoming the supine victim of injustice.
So any person who engages with the law will feel some self-reproach and guilt if he forces on others or allows onto himself laws that are evil, cruel or unfair. The relevant question can therefore be expanded: ‘How can this be a Law for me and you given that we are fellow members of a community that must find a way to cooperate in a shared world?’ Any laws that cannot provide a satisfactory answer to that old question, will constitute violations of what we owe to each other. The question arises not only in the work of judges, but also in the work of legislators and officers of the executive and indeed in the life of every engaged citizen. Unjustified impositions will be a distinct moral wrong that cannot be ignored by any self-reflective person in whatever capacity.
A person's self-awareness will demand clear answers on the scope of our responsibility towards his fellow citizens, especially when legal institutions give us the power to impose coercive constraints on others. Are we to treat those people with casual indifference, as Hart's immoralist judge seems to accept? I very much doubt that this is consistent with the self-identification of judges as people of conscience. Our identity as reflecting agents is not abandoned when we put on a judicial robe or when we work as legislators or civil servants. This does not mean that we are constantly engaged in this kind of reflection. Many of our actions are unthinking or unreflective, precisely because they have become second nature to us. But they are second nature because they are consistent with our practical identity – as citizens, lawyers, officials, dedicated servants of the law, rigorous scholars or something like it. Practitioners are aware of the weight of the institutions under which they work. This creates, I suppose, what Brian Simpson called a ‘caste’ of lawyers that develop legal practice under the weight of these practical identities.Footnote14 In our own times the authority of legal institutions has been enhanced by democratic and egalitarian institutions and respect for human rights and social justice, practices which now create even deeper reasons for the legitimacy of law and for political obligation. Dyzenhaus accurately calls this a ‘constitutional ideal’ and organises his account of legality around it. This ethical perspective is a golden thread running through the whole of the book and animating its most powerful passages.
By setting the stage for law by way of a first-person question of moral responsibility Dyzenhaus has framed the key questions of the philosophy of law in an entirely novel way, rejecting the point of view of much contemporary jurisprudence in the style of ‘legal positivism’ (although this approach is not novel at all when seen in light of the writings of Plato, Cicero, or Grotius). When he puts the subject's question ‘How can this be law for me?’ at the heart of legal philosophy he restores its place to a critical tradition of reflection about human affairs with ambitions of universality.Footnote15 Rather than offering descriptions or pieces of information as Hart and his followers seem to do, this new type of jurisprudence addresses real questions of responsibility for our actions. It responds to the difficult and profound questions raised by lawyers caught up in moral dilemmas, such as Ernst Fraenkel and Michael Sfard. These questions are not peripheral to jurisprudence, as the legal positivists might suggest. They are central to law's nature because they invite us to specify the grounds of the law's authority. Dyzenhaus’ masterful book shows clearly that the question of the nature of law calls for reflection on the most fundamental questions of human association. Jurisprudence has the task of helping us understand how social life is possible for free and equal persons.
Acknowledgements
I am grateful to the Faculty of Law of the University of Oxford and to Mansfield College, Oxford for sabbatical leave in Michaelmas Term 2022 that allowed me to complete this essay. I am also grateful to NYU Abu Dhabi where I was a visiting professor while working on this essay. I benefited greatly from discussing these ideas with Julie Dickson and our students in our postgraduate seminars in jurisprudence and political theory at Oxford. I also had the good fortune of discussing the book with David Dyzenhaus and hearing him discuss it at the School of Law at New York University in November 2022 and at the Jurisprudence Discussion Group at Oxford in February 2023. The usual disclaimer applies.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 David Dyzenhaus, The Long Arc of Legalty: Hobbes, Kelsen, Hart (Cambridge University Press 2022). All references in brackets are to pages of the book.
2 See H L A Hart, Essays on Bentham: Jurisprudence and Political Theory (Clarendon Press 1982) 243–68.
3 Hart (n 2) 265.
4 See Joseph Raz, ‘The Obligation to Obey: Revision and Tradition’ in Joseph Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford University Press 1995) 341–54.
5 The reference is from Hart, The Concept of Law, 181.
6 The relevant essay is Bernard Williams, ‘Realism and Moralism in Political Theory’ in Bernard Williams, In the Beginning Was the Deed: Realism and Moralism in Political Argument (Princeton University Press 2005) 1.
7 I explain this point in greater detail with reference to the idea of ‘public justification’ in Pavlos Eleftheriadis, Legal Rights (Oxford University Press 2008) 56–60.
8 This is in effect Dworkin's response to Williams’ and Berlin's historicism about political ideals, which he contrasts to interpretivism. In Justice for Hedgehogs Dworkin writes that history ‘is no privileged guide to [the] best interpretation’ of political concepts (350). I therefore disagree with Thomas Bustamante who believes that Dyzenhaus’ argument in The Long Arc of Legality is very close to Dworkin, because Dworkin and Williams are similarly ‘pragmatist’. See Thomas Bustamante, ‘Interpretive Authority and the Kelsenian Quest for Legality’ July 5, 2022, Jurisprudence JOTWELL, <https://juris.jotwell.com/interpretive-authority-and-the-kelsenian-quest-for-legality/>.
9 John Rawls, Political Liberalism (Columbia University Press 1993) 137.
10 For another argument in this direction, which Dyzenhaus does not discuss, see Arthur Isak Applbaum, Legitimacy: The Right to Rule in a Wanton World (Harvard University Press 2019). Applbaum summarises his position as follows:
Legitimacy and justice are neither identical in concept nor in extension, though they clearly overlap. The subject of justice is the distribution of rights, duties and advantages of social institutions. The subject of legitimacy is the specification of the normative power to impose normative liabilities, or, more colloquially, the right to rule. (256)
11 See Javier Gallego, ‘Legal Positivism’s Internal Morality’ (2022) 42 Oxford Journal of Legal Studies <https://doi.org/10.1093/ojls/gqac030>.
12 See Ernst Fraenkel, The Dual State: A Contribution to the Theory of Dictatorship (Oxford University Press 2017).
13 See Michael Sfard, The Wall and the Gate: Israel, Palestine and the Battle for Human Rights (Metropolitan Books 2018).
14 A W B Simpson, ‘The Common Law and Legal Theory’ in A W B Simpson (ed), Oxford Essays in Jurisprudence (Oxford University Press 1973) 77–99. Simpson writes that in his view the (common) law:
consists of a body of practices observed and ideas received by a caste of lawyers, these ideas being used by them as providing guidance in what is conceived to be the rational determination of disputes litigated before them, or by them on behalf of clients, and in other contexts. (ibid 94)
15 I defended a very similar argument about the nature of jurisprudence in Pavlos Eleftheriadis, Legal Rights (Oxford University Press 2008).