H. L. A. Hart (1907-1992)

Oxford University Press 1993 Utilitas Vol. 5, No. 2 November 1993, 145–156

JOSEPH RAZ Oxford University

Herbert Lionel Adolphus Hart was a prominent member of the postwar movement which became known as Oxford Philosophy. He led the revival of interest in the philosophy of law in the English-speaking world and his theories became the standard liberal view of the nature of law and its relation to morality which other theories either incorporate or strive to refute.

After graduating from New College, Oxford (1929), where he read Classics, Hart was called to the Bar (1932) and practised for eight years as a Chancery barrister. During the war he served in military intelligence. In 1945 he was elected to a fellowship at New College as a tutor in philosophy. From 1952 to 1968 Hart was Professor of Jurisprudence at Oxford University, and this period saw the publication of his most important works. In 1968 he resigned his chair and became a senior research fellow at University College, Oxford. From 1972 until his retirement in 1978 he was Principal of Brasenose College, Oxford.

Apart from his writing and teaching Hart pursued his interest in the work of Jeremy Bentham by participating in editing several volumes in the new edition of Bentham’s Collected Works. He also served as a member of the UK Monopolies Commission and chaired an Oxford University inquiry into relations between junior and senior members of the university instituted in the wake of the student unrest in the 1960s. Hart was elected to the British Academy in 1962 and received numerous honorary degrees.

Hart noted that much of the writing of legal philosophers was apparently concerned with the definition of a small number of key notions, such as ‘law’, ‘rights’, ‘duties’, ‘legal persons’. Many philosophical battles were fought over the adequacy of such definitions. Hart regarded such warfare as unproductive for two reasons. First, the traditional mode of definition— per genus et differentiam spedficam—is inappropriate for the tasks of a philosophical inquiry into the nature of very general and abstract notions. The definition of such notions can only be carried out in terms which are as puzzling, and in the very same respects, as the notions which they define. One may perhaps define ‘rights’ in terms of entitlements and ‘law’ in terms of rules but ‘rules’ and ‘entitlements’ are just as puzzling, and puzzling in the same way as ‘law’ and ‘rights’. The inadequacy of such definitions led Hart to recommend contextual explanations in which what is explained is not a word (e.g. ‘a right’) but the word in context (e.g. ‘John has a right to have this book’). The explanation is not of the word but of the sentence in which it occurs, and can take the form of providing a synonymous sentence in which it does not occur or (since this is often impossible) specifying the truth conditions of the explained sentence. The nature and importance of contextual explanations were first explored in this century by Bertrand Russell. But Hart rightly pointed out that Bentham had anticipated Russell’s work by over a century.

Hart’s second objection to the jurisprudential preoccupation with definitions was that it misrepresented the focus of philosophical inquiry. Definitions are instructions for the use of words. But those who seek philosophical definitions of ‘law’ and other jurisprudential notions do not require such instruction. ‘Law’ is a familiar word which we use unerringly without resort to any philosophical definition. The purpose of jurisprudence is not to instruct us in the use of ‘law’ but to explore the law’s essential relations to morality, force and society. The task of jurisprudence is misrepresented when it is conceived as a search for definition. It is in fact an exploration of the nature of an important social institution. Of his central work, The Concept of Law, Hart said that it is an essay in descriptive sociology.

Hart’s criticism of the jurisprudential preoccupation with definitions, which was the major theme of his inaugural lecture in Oxford in 1953, reveals much about the foundations of his work in legal philosophy. As his advocacy of contextual explanations shows, he was particularly anxious to apply to the analysis of law philosophical methods developed in the analytical school during the first half of this century and those which were being explored in Oxford after the war. Perhaps of even greater importance than his contribution to the solution of many jurisprudential problems was his success in reuniting English-language jurisprudence with mainstream philosophy. During the previous century and a half, the growth of departments of law in English and American universities had led jurisprudence to become the preserve of academic lawyers. Hart reversed the trend. He turned jurisprudence back into legal philosophy, as it had been of old, and helped to make it a bridge between students of law, politics, and philosophy.

But while introducing sophisticated methods of linguistic analysis into jurisprudence, Hart was steering away from the view of philosophy as concerned with language as such and of jurisprudence as the analysis of key legal terms. His interest was in the analysis of law as a social institution. Since law is a cultural phenomenon and is shaped by language, an understanding of language and its use contributes to the understanding of the social institution, the law, but it is not an end in itself. The nature of his interest led Hart to a more systematic inquiry than was common in Oxford at the time. The Concept of Law was among the first theory-constructing works in social philosophy in English after the war, despite its author’s mistrust of grand theories.

The claim that ‘law’, ‘rights’ and ‘duties’ cannot be explained by their genus reflects the anti-reductivist stance that Hart assumed from the first. One cannot adequately explain the law as a species of command or in any other terms which will eliminate its normative character. Explaining the normative character of law is explaining the way in which it can be a source of rights and duties and the sense in which it imposes requirements for action. Hart rejected theories of law which deny its normativity, i.e. which regard statements of rights and duties not as concerned with what people ought to or may do but as factual, sociological or psychological statements about the behaviour of officials and the like. The employment of the method of linguistic analysis led him to regard the use of normative language as the key to understanding the normativity of law. This linguistic turn has two crucial consequences. First, the most general terms and expressions used in legal discourse (‘rights’, ‘duties’, ‘rules’, ‘property’, ‘agreement’ etc.) are not specifically legal. They are the common currency of much normative discourse generally. Therefore, the explanation of the normativity of law is not just a matter of explaining what, if any, moral force it has. It is an explanation of normativity generally and on that basis an exploration of the specific characteristics of morality, law, etiquette and other special normative spheres. Once more we see here how philosophy of law has come to rest, in Hart’s hands, in the midst of general philosophical theories.

The second consequence of the linguistic turn is that the normativity of law is explained as it affects those who regard themselves as bound by the law. It is one of Hart’s main themes that when a person says ‘I have a duty . . .’ or ‘You have a duty . . .’ he is expressing his endorsement of a standard of behaviour, holding it as a guide for conduct. It is this fact about normative discourse which explains its specific character and prevents its interpretation in terms of sociological or psychological generalizations or predictions. By his insistence on the importance of this insight Hart aligned himself with that school of thought which regards the explanation of all human behaviour, both private and social, as bound to give pride of place to the agent’s point of view. An explanation of an action has to include an explanation of the way the agent perceived his situation and his reason to act as he did. This approach is coupled, in Hart’s writings, with a rejection of a uniform notion of causality as the basis of all explanations of actions and of events. In particular, saying that something made, or caused, a person to act in a certain way does not normally presuppose a generalization of a constant conjunction (i.e. whenever an event of a certain character occurs a certain class of agents perform acts of a certain kind) nor does it presuppose the counterfactual that the agent would not have undertaken the act but for the event. It is often simply a statement that the event was the agent’s reason for his action.

Hart’s insistence on the irreducibility of the normative led him to interpret normative utterances as sui generis. Analytical philosophers are usually divided into cognitivists and non-cognitivists, depending on whether or not they regard normative statements as either true or false. If Hart must be classified as one or the other, then it is better to classify him with the non-cognitivists. But he charted his own way between the two, and developed a distinctive view which combines cognitivist and non-cognitivist elements. Statements of rules, duties, and rights are true or false, but the conditions which render them true or false do not exhaust their meaning and do not account for their normative character. The truth-conditions of such statements are the existence of certain social practices (see below). A simple moral statement such as ‘parents have a duty to look after their children’ is true if there is (in the community to which the speaker belongs) a practice that parents have such a duty, that is, roughly speaking, if most parents do so and are consciously disposed to do so. But the statement means more than that. It also expresses its speaker’s endorsement of this rule, his willingness to be guided, and to require others to be guided, by it. This second non-cognitivist component expresses the normative element in the statement.

The way the cognitive and non-cognitive elements combine in statements of rights, duties, and rules can be seen most clearly by comparing them to statements which display only one or the other of these features. One may express views belonging to one’s ‘critical morality’ as Hart calls it, i.e. views about what moral practices the society should have but does not and how people ought to behave according to them. It would be a misuse of language to talk of rights and duties in such a context. One has a duty to be a vegetarian only if the community has a moral practice requiring this. But a person may say ‘one ought to be a vegetarian’ even in a meat-eating community, for saying this merely expresses his endorsement of the standard as appropriate for the society without stating that it is already practised in it. Here the utterance lacks the cognitive element present in statements of duties, rights, and rules. On the other hand, there is the possibility of making statements from an external point of view, i.e. statements about the existing practices of a community which do not express endorsement of those practices. For example, saying that members of the community accept a duty to look after their children has the same truth-conditions as the statement from the internal point of view that parents have a duty to look after their children, but lacks its normative force, for it lacks the expression of the speaker’s endorsement of that form of behaviour as a guide to oneself and others.

Hart’s initial conception of the non-cognitivist element in normative statements was derived from J. L. Austin’s theory of performatives. Later his position drifted away from that of Austin. He continued to regard the performance of legal acts such as the making of a contract or a will as best explained by Austin’s theory of performatives, but ordinary normative statements he saw as simply expressing an endorsement of a standard in a sense which, though never fully analysed, was meant to differ both from the interpretation of normative statements as expressing emotions and exhorting to similar emotions and their interpretation as commands or prescriptions. Believing oneself to be bound, accepting a certain standard by which to guide one’s behaviour, is not having a sense of inner compulsion, nor being in any emotional state. Nor does the making of normative statements amount to commanding others or oneself (if that makes sense) to act. Normative statements can be used as exhortations, recommendations, criticisms, etc. But the only normative element common to all their uses is that they express the speaker’s endorsement of a standard as a guide for action.

Morality for Hart is primarily social. It consists of social rules of conduct largely (at least if we regard rules of duty as the core of morality) concerned with securing conditions necessary for maintaining social life. Critical morality concerns the evaluation of social morality and beyond both moralities individuals may adopt personal goals and rules of conduct. Social rules are for Hart no more than regularities of behaviour which are by and large observed in the community, and which members of the community think ought to be observed, where those facts are widely known in the community. His non-cognitivist position led Hart to emphasize not people’s views as to what ought to be done, but their willingness to criticize and put pressure on those who fail to do so. His is, therefore, a sanction theory of rules and duties. The existence of social rules does not depend on the existence of institutionalized sanctions characteristic of the law. But it does depend on diffuse critical reactions which Hart treated as nebulous sanctions rather than as merely the expression of people’s judgements about their duties and their and others’ conformity to them.

Hart was relatively unconcerned to separate social morality from other social rules. But he was very much concerned with separating legal rules from other social rules (moral or otherwise). The law is characterized by a combination of primary rules of conduct with secondary rules of three kinds. Rules of change specify ways in which legal rules can be made, repealed or modified. Rules of adjudication provide for authoritative ways of settling disputes concerning the application of legal rules to specific situations. Finally, every legal system contains one rule of recognition which requires its officials, and primarily its courts, to apply rules identified by criteria set out in the rule of recognition. The existence of rules of change and of adjudication makes the legal system into a self-regulating system. It is the mark of the law that it provides means by which it can be changed and a machinery for its own application and enforcement. The rule of recognition turns the law into a separate, identifiable system of rules.

A legal system consists of a rule of recognition and all the rules which it requires officials to act on and to apply. Social rules, or other standards which the officials are not required by the rule of recognition to apply, are not legal rules, at least not of that legal system. The rule of recognition itself is a social rule. Its existence and content can be established as a matter of fact by investigating the customs and practices of the officials. The rule of recognition is accepted and practised by the officials. They not merely conform to it but accept it as a guide to their own behaviour. In most countries the same is true of large sections of the population as well. But the minimal conceptual requirement for the existence of a legal system is that its officials accept and follow its rule of recognition and that the population by and large conforms with the law. For a legal system to exist, only the officials need be guided by it. The rest of the population often accepts the law as a guide to their behaviour as well. But the minimal conceptual requirement is satisfied so long as by and large they do not break the law.

Establishing the contents of a legal system is therefore a two stage process. First, one establishes the contents of the rule of recognition by a factual inquiry into the practices of the courts and other officials. Secondly, one establishes which rules meet the criteria laid down in the rule of recognition, that is, which are the rules the officials are bound to apply, according to the rule of recognition. Hart marked the difference between the two stages by saying that rules of law are valid if, according to the rule of recognition, the officials are bound to apply them. The rule of recognition itself exists as a social practice, but cannot be said to be valid.

That every legal system has a rule of recognition does not mean that establishing its content is easy. Rules are invariably vague and opentextured and their core meaning is surrounded by an area of borderline undetermined cases. The same is true of rules of recognition. The social practices which constitute them may not allow a decisive answer as to their application to certain cases. Hart mentioned as an example the controversy in English law concerning the power of parliament to pass legislation binding on future parliaments. There is no complete and correct legal answer to the question, to what extent can parliament bind its successors. The practice of the officials on this is too inconclusive and confused.

Vagueness and open texture are endemic to language. Legal rules, being formulated in language, are therefore indeterminate at the edges. Courts and other officials entrusted with the task of applying and enforcing legal rules often find that due to vagueness and open texture the meaning of these rules is uncertain. In such cases, the rules do not provide clear guidance to the courts, and the courts have to use their own discretion to decide what to do. Not in every case do they have discretion. Those who suppose otherwise fallaciously argue either that language has no fixed meaning or that obligations exist only when there is a formal machinery for supervising their enforcement. Both arguments fail, the first because rules have a certain core meaning, and the second because the fact that there is no appeal from decisions of the highest courts does not establish that they are not subject to duties. Furthermore, in cases where the courts have to exercise discretion because of the indeterminacy of the rules, their discretion is both limited and broadly guided by the rules. But the rules, in such cases, fail to point uniquely to one outcome as the only legally acceptable one. While discretion is hardly ever absolute it is pervasive. To deny that sometimes cases are decided by judicial discretion is either to overlook the indeterminacy of language or to deny that the law consists only of rules identified by criteria laid down in the rule of recognition.

Since the law consists of rules recognized by a rule of recognition which need be no more than the practice of officials, it can diverge both from social morality and from critical morality. Social rules which courts are not required by the rule of recognition to apply are not legal rules. Legal rules, while being applied by the courts, need not be socially accepted, indeed many of them may be generally disregarded. Naturally, legal rules need not always conform to everyone’s views of critical morality. It is important, however, to realize that when Hart claims that the law’s existence rests on its acceptance as a guide to behaviour, at least by the officials and usually by many more, he is not committed to the view that those who accept the law morally approve of it. They may accept it and guide their behaviour by it for fear of sanctions or loss of prestige or to earn money or for many other prudential as well as for moral reasons.

Even though the existence of law is rooted in social fact, and though law may diverge both from social and from critical morality, Hart maintained that given the basic facts of human existence there is a necessary overlap between social morality and law in all societies. The facts that Hart alluded to are that people are vulnerable to attacks by others and are approximately equal in strength, at least to the degree that even the strongest cannot dominate the others without cooperation from at least some of them, that many resources are limited and that both altruism, understanding and strength of will are limited. These facts establish minimum conditions for social cooperation, and without such voluntary cooperation survival is at risk for it cannot be assured for long by coercion. Only societies which recognize rules curtailing the use of violence, rules of property to enable security in possessions, and rules of voluntary undertakings can secure a sufficient degree of voluntary social cooperation. Therefore, given the need to survive and the perennial conditions of human existence, all stable societies recognize such rules both in their social morality and in their law. This necessary overlap between law and morality is, however, too limited to endow law with moral authority or to guarantee the overall moral decency of the law. The degree to which any legal system reflects moral values beyond the minimal degree specified depends on its particular historical circumstances.

Many writers who, like Hart, deny the intrinsic moral authority of law tend to emphasize that law rests on force and to highlight the role of sanctions. While Hart did make his analysis of legal duties depend on the existence of formal or informal sanctions, he deviated from many of his predecessors in two crucial respects. First, as we saw above, he was at pains to emphasize that ultimately law rests on the cooperation of at least part of the population in holding themselves bound to be guided by it. Second, the law not only imposes duties. It also confers powers and grants rights to people. Duty-imposing rules constrain individuals and require them to behave in a certain way, whether they like it or not. Power-conferring rules, which, Hart insisted, are a class unto themselves, help individuals to realize their purposes. Private law powers such as the ability to marry, make a will, and buy and sell property, facilitate individuals’ efforts to arrange their own affairs as they wish. Individual legal rights are of many different kinds. But they all manifest, albeit in different forms, one core idea: the law’s respect for the choice of the individual in the matter of his right. For example, one central kind of right, often called a claimright, consists in control of another’s duty. To say that one person has a legal right that another shall behave in a certain way is, in the typical case, to say that there are legal rules which impose on that other person a duty so to act, and grant the right-holder control over that duty through the dual power on the one hand to waive his right and release the other person from his duty and on the other hand to enforce the right by court action for breach of duty. Moreover, such powers amount to a right only if their possessor is at liberty to use them at will, i.e. if he is under no duty to use them or to refrain from doing so. A claim-right is in effect a special case of a power-right which involves the possession of a legal power coupled with a liberty to use it at will. The power to make a will is an example. A liberty-right is simply the liberty to act in a certain way or not to do so, that is it consists in the absence of a legal duty either to refrain from the act or to perform it. In all these cases others are under duties which restrict their freedom to interfere with the right-holder’s exercise of his right. Regarding all rights, the law shows its respect for the individual’s choice either by facilitating its implementation or at least by not interfering with it.

Hart’s writings on the nature of rules, duties, rights and the law were innovative. Through them he influenced those pioneers in modern times who explored and advocated explanations of human actions and of social processes which give primacy to the agent’s reasons for their own actions. He also contributed to the emergence of the theory of action (and related concepts such as decision, intention, the will, negligence, recklessness) as a major branch of analytical philosophy. On these issues he followed Wittgenstein and Ryle in rejecting the view that mental phenomena such as desires, wishes, intentions, decisions are discrete mental events which cause one to move one’s limbs or body in a certain way. But on the whole he was a follower and not a leader in philosophy of mind. Not so in moral theory. It is true that Hart did not develop a general moral theory of his own, but he wrote much about moral issues especially those concerned with the law. His sympathies were broadly utilitarian but in his writings he hardly ever espoused narrow utilitarian positions. His interest, and his greatest influence, were in analysing and defending liberal causes. During the first half of the twentieth century, English moral philosophy was rather uninspiring. The emphasis was on moral epistemology and the analysis of moral language. The results were not very impressive and by the late fifties a sense of impasse was growing. Hart’s essays combining conceptual analysis with a detailed lucid argument about concrete moral issues which have clear implications for legal and social policies were an inspiring model of an alternative style in moral philosophy and helped to launch the wave of writing on specific moral issues which swelled in the 1960s.

Hart wrote extensively on punishment and criminal responsibility. The general internal purpose of the Criminal Law is to announce to society that certain actions are not to be done and to secure that fewer of them are done. To justify any system of criminal law one has to show that there is good reason for doing so regarding the forms of conduct it defines as offences. The general justifying aim of punishment is therefore to reduce the incidence of crime. The severity of punishment is also primarily to be determined by the goal of reducing crime, be it through deterrence of potential offenders, or of the punished person, or through his reform. However, considerations of justice impose two major restrictions on the pursuit of the general justifying aim of punishment. First, ‘Each individual person is to be protected against the claim of the rest for the highest possible measure of security, happiness or welfare which could be got at his expense by condemning him for a breach of the rules and punishing him. For this a moral licence is required in the form of proof that the person punished broke the law by an action which was the outcome of his free choice’. Since the justifying aim of punishment is the protection of society from harm, individuals should not be made to suffer unless they caused harm and were given a fair opportunity to choose between obeying the law and paying the penalty. This is required by fairness and it also increases individual freedom compared with systems based on manipulation or the ‘treatment’ of anyone with anti-social tendencies. Hart wrote extensively against tendencies to steer the criminal law in such directions, because of the limitation of individual freedom they involve. Hart’s concern with liberty led him to prefer punishment after an offence to trying to cure people of criminal tendencies before the offence. He therefore regarded reform as a secondary aim of punishment since it primarily treats criminal tendencies and interferes with individual freedom of choice.

The second set of considerations of fairness requires that different kinds of offences of different gravity should not be punished with equal severity. This is a requirement of fairness, of treating like cases alike. Similarly, justice requires that those who have special difficulties in keeping the law which they have broken should be punished less. The first set of considerations of justice establishes the need for excusing conditions such as mistake of fact and insanity. The second leads to a gradation of penalties relative to the gravity of the offence and to pleas for mitigation of sentence.

Hart became the leading philosophical spokesman in the campaign to liberalize the law which gathered force in the late 1950s and 1960s. The philosophical debate revolved round the question of the legal enforcement of morality. Hart espoused Mill’s Harm Principle claiming that the only purpose for which people may be coerced or punished by law is to prevent harm to individuals. One may not restrict a person’s freedom simply to stop him from committing a morally wrong act. In one important respect Hart deviates from Mill. For Mill the only legitimate purpose of the law was to prevent harm being caused to people without their consent. Hart allows that sometimes it is justified to protect people from harm they consented to. He does not share Mill’s belief in the balanced and independent judgement of individuals. They may have to be defended against their own judgements when those are swayed by propaganda, weakened by temptation, based on ignorance and the like. But even in such cases, the law may not force virtue on a person. It may only protect him from harm brought by his own hands or by others with his consent.

Harm included physical, economic and psychological harm. Hart allowed laws restricting, for example, an indecent display in public, if the harm such a display causes is greater than the pleasure it brings. But he denied the right of the law to proscribe harmless immoral behaviour in private, nor would he allow such a restriction of liberty to gratify feelings of hatred towards wrongdoers or disgust at the knowledge that wrongdoing is privately perpetrated. He rejected punishment as a symbol or expression of moral condemnation. His position was based on a deep respect for the liberty of the individual, on the value he found in a morality embraced freely out of conviction and not one inculcated by fear, and on the belief that morality should be open to free critical evaluation through public debate and that it should not be insulated from change by the use of force by public authorities.

Hart’s work was for his age and for the future. For his time he was one of the leading spokesmen of the liberal spirit and thought which led to the liberating and envigorating changes of the 1960s in Britain and throughout the Western world, with their attendant social consequences. In particular he fought to abolish the death penalty, to decriminalize harmless conduct, and to end the persecution of people because of their sexual preferences. At a time when ethical skepticism and subjectivism were rife he taught us the beginning of a way of thinking clearly and rationally about moral issues which affect social policy. And he harnessed his intellect to the cause of the liberal and liberating reform of stifled and stifling institutions. His (somewhat ambiguous) sympathy for and admiration of Bentham arose not only because Bentham’s important philosophical and jurisprudential writings needed rescuing from obscurity, but also because Bentham’s passion for reform and his conception of reform appealed to him. He was drawn by the way Bentham’s reforming instincts were fed by his vigorous intellect, the way his intellectual rigour and clarity combined with irreverence for obfuscating traditions and institutions. Bentham provided the ethical theory which inspired the reform movement of the Radicals of the nineteenth century, and I believe that for Hart it showed how radicalism need not be based on Marxist assumptions.

For the future Hart has bridged the gap between philosophy and jurisprudence that opened—in the English-speaking world—with the work of John Austin. Hart rightly saw in Austin a second-rate mind regurgitating Bentham’s insights into the nature of law, giving them a simple and tidy shape through his somewhat dogmatic and inflexible treatment. Austin’s tremendous influence on legal education and legal thought in England and the Commonwealth was in part due to the fact that he served as the poor man’s philosopher, releasing legal scholars from the need to struggle with the richer and subtler works of Aristotle, Aquinas, Hobbes, Hume, Kant or Hegel. But jurisprudence languishes when it is studied independently of general philosophy. Hart rescued jurisprudence in English by re-establishing its lifeline to general philosophy. Arguably apart from his work (together with Honore) on causation he has not contributed greatly to general philosophy. But his writings in philosophy of law join those of Hobbes and Bentham as the major contributions in English to that subject.