Hart's Concept of Law Lecture Four 2001

I turn now to Fragmentation and Dependent Norms:-

(1) "Extreme" Line: this would deny that even the rules of criminal law are "laws"; i.e. there is no law prohibiting murder, there is merely a directive to officials:- "If X is, then [you ought to] apply sanction Y" (where X= definitional elements of murder and Y= the legally determined sanction for murder). Thus the common sense view that the criminal law is directed to ordinary citizens in rejected (although the content of criminal law may yet be a source, inferentially, of guidance to ordinary citizens). On this "extreme" view laws are understood as conditional directives to officials to apply sanctions ... 64. "Law is the primary norm which stipulates the sanction"(Hans Kelsen, General Theory of Law and State, p. 60). The secondary norm is implied from it and is superfluous in a strict description of the law - "law from its own point of view" (cf Einstein 216) - though it may be utilised as an "auxiliary concept", helpful but not necessary in an exposition of the law. Thus 65 "By greater and greater elaboration of the if-clauses, legal rules of every type, including rules conferring and defining the manner of exercise of private and public powers, can be restated in this conditional form".(Hart, op.cit., p.35). This is important. Hart actually admits that the Kelsenian programme is possible. The Extreme Line bids us strip away the obscuring substance to reveal the underlying form. Hart's appreciation of this approach does not exhibit any great enthusiasm for it and is couched in the "vocabulary of prejudice": 66. "This is a formidable and interesting theory purporting to disclose the true, uniform nature of law latent beneath a variety if common forms and expressions which obscure it". (Hart, op.cit., p.36).

Now this theory involves a shift from the simple view of orders backed by threats addressed to the populace at large (i.e. Austin's position) to a different idea of law as directives to officials to apply sanctions (i.e. the views of Kelsen, and Ross). On this view it is unnecessary that every law be "backed by a sanction"; it is only necessary that every genuine law shall stipulate or direct a sanction. Perhaps a legal order is best understood as an order to an official of the law. It is not necessary that an official who disregards such a directive be subject to a sanction for that directive to be law. And this is just as well for it is logically impossible for every directive to be backed by a sanction ... infinite regress, "Who guards the guards?" [Quis custodes, custodiet? No one's found the answer yet". Kelsen offers a knock-down argument against Austin's notion that every law is a command backed by a sanction 67 "... if it is necessary to guarantee the efficacy of a norm prescribing a certain behaviour by another norm prescribing a sanction in the case the former is not obeyed, a never-ending series of sanctions, a regressus ad infinitum, is inevitable" (Kelsen, General Theory of Law and State, p 28)

(2) "Moderate" Line: the original conception of law, as orders backed by threats, which is intuitively acceptable is preserved at least for these rules that "on a common sense view" [p 37] refer primarily to the conduct of ordinary citizens and not merely to officials. Thus the D-rules of criminal law are laws already as they stand. But P-rules of civil law are merely fragments of laws of which one is to ask of them, What person is liable to a sanction if he or she does not comply. This alludes to Bentham's "Fragmentation" solution. Hart comments, 68. "Both versions of this theory attempt to reduce apparently distinct varieties of legal rule ro a single form alleged to convey the quintessence of law". (Hart, op.cit., p.38). Hart's objections to both versions are summarised in the Handout:-

ad II (a) Fragmentation would fail if sanctionless laws are conceivable. But this is never argued for in the Concept. Indeed it is only by fragmentation that one can abstract a so-called law without a sanction; e.g. "Steal Not !" is a fragment of or dependent upon the complete law "If someone steals he ought to be punished". But even if sanctionless laws are possible, this would not count against the uniformity thesis. The thesis that all "laws" are of the same logical type is not the thesis that all laws are coercive.

(b) Fragmentation and Dependent Norms; i.e. the views of Bentham and Kelsen, would distort the social function. But this is not an argument against the possibility of presenting all law in a body of propositions of the same logical type; it is merely an argument against the desirability of doing so. Thus, in a sense, Hart admits that all law can be presented in a body of uniform propositions but that, in order to highlight different social functions, it should be presented in propositions of different types. But the same logical form can subserve a multitude of social functions; since logical form does not reveal social function, it cannot distort it either.

Hart's main objection is as follows:- 69. "... they all purchase the pleasing uniformity of pattern to which they reduce all laws at too high a price: that of distorting the different social functions which different types of legal rules perform". (ibid). This concedes that the unifying procedure is possible and insists only that it is not desirable. But some theorists might differ from Hart on whether there is "distortion" at all or if there it such theorists might think it a price worth paying for such an elegant and insightful general conception of "law". Hart develops his "distortion" point as follows:- First, he rightly observes that 70 "... the members of society are left to discover the rules and conform their behaviour to them: in this sense they 'apply' the rules to themselves ...". (ibid). But this, contrary to Hart's argument would suggest that the rules about the conduct of ordinary citizens are derivative from rules about the conduct of officials. And it would call for a "legal science"; since the law does not directly tell citizens what they ought to do, they cannot know this without some intellectual guidance and assistance.

Secondly, Hart holds that a fine and a tax involve different ideas, and that the "logical uniformity of rules" thesis distorts this. At a common sense level there is a distinction which turns upon individual and/or collective value judgments. The distinction to which Hart draws attention is a moral one - "fines" are reactions to wrong conduct; "taxes" are responses to acceptable conduct. Fines involve compulsion whereas taxes involve consent. But this is hardly persuasive. Not all individuals regard conduct which is the condition of a fine as wrong; nor do all individuals consent to taxation. And Hart, himself, admits that ... 71 "A punishment for a crime, such as a fine, is not the same as a tax on a course of conduct, though both involve directions to officials to inflict the same money loss". (Hart, op.cit., p.39). For example, what is the difference, if any, say, between perhaps modest fines for possession or use of cannabis and heavy taxation on tobacco?

Thirdly, Hart argues that the law should not only be concerned with the "bad man" but should seek to instruct the "puzzled man" or the "ignorant man". The allusion is the Holmes (see 20) But the object of a logical analysis is to reduce the legally relevant material to a simplified and unified form. Both Ross and Hobbes have apposite comments about the nature of law: 72. "A parliament is not an information bureau, but a central organ for social direction". "Alf Ross, On Law and Justice, p.8) and 73. "And first it is manifest, that law in general, is not counsel, but command". (Hobbes, Leviathan, Chapter XXVI). However, having analysed the nature of law, legal science may try to state what the law requires of or permits to the ordinary citizen, in so far as that is contained in the content of the law. Legal science cannot be more precise than its subject matter and if or where the law vests wide discretion in an official, there may be no such content to describe. But note, too, that this secondary stage whereby the general propositions of legal science are translated into citizen-friendly propositions about what the law requires of or permits to citizens is arguable a matter of legal practice as much as a matter of legal science. In other words, it may not be the function or responsibility of legal science to assist the puzzled or the ignorant who may be better served by their local legal practitioners.

Fourthly, Hart finds some "justification" unsurprisingly in "common usage" ... 74. "... power-conferring rules are thought of, spoken of, and used in social life differently from rules which impose duties, and they are valued for different reasons. What other tests for difference in character could there be?" (Hart, op.cit., p.41). Well that rhetorical question certainly can be answered and there are tests other than correspondence to actual usage by which intellectual theories may be assessed ! Nonetheless the way we use rules in discourse, reasoning and action may provide insights into their character.

Fifthly, Hart is ambivalent. His criticism borders on a contradiction. He asserts at one point that a coercive definition of law excludes rules conferring public and private powers, but his objection is cancelled by an admission that they are not excluded by all coercive definitions (e.g. Kelsen's in particular). So Hart's argument changes. Such definitions are now declared inadequate because they do not distinguish sufficiently, if at all, criminal law from civil law, fines from taxes, or P-rules from D-rules. But in so far as a definition (of law) purports to be general or generic, Hart's objection cannot be valid. It is as if a definition of "horse" was to be held inadequate because it fails to distinguish Shetland Ponies from Clydesdales ! Hart's objection is that a generic definition does not accurately describe how laws operate; i.e. their social function; But neither Hart nor anyone else would seriously argue that each and every law has the same social function. Questions about the function of law are central both in jurisprudence and in the sociology of law but they cannot be answered, definitively, if at all, a priori through conceptual debates. Because a definition of law must be neutral as to social function, no such definition can distort social function. The genius of the Roman jurists exploited and exposed the separation of form and function. For example a slave's peculium was turned into the basis of limited liability trading. I conclude therefore that Hart's central criticism of "unifying theories" fails and I suspect that it is the fruits of methodological syncretism, i.e the expectation that empirical sociological questions about the function can be answered by the logical analysis of concepts.

Hart makes two other noteworthy points in this chapter:

(1) The coercive model is inadequate because laws can bind those who make them and this conflicts with the vertical notion of commander and subject. However this point counts only against Austin and not against Kelsen ... 75. "But it is possible that a norm be created by the same individuals who are bound by this norm". (Hans Kelsen, General Theory of Law and State, p.36). Thus the theories of both Kelsen and Hart, unlike that of Austin, allow of the possibility of a democratic legal system; i.e one in which laws are made by those subject to them rather than laid down by a political superior. One telling, in my view fatal, criticism of Austin's command theory is that it cannot without extensive modification, if at all, accommodate democratic law. You may remember that I introduced one test of the adequacy or acceptability of a theory of law earlier in these lectures. I said that one test of a theory of law is that it should help un "understand" law. Another test of legal theory is that it should both (1) fit the facts and (2) be normatively appealing. In other words it is legitimate to ask of a theory of law if it necessitates something morally monstrous, or excludes something morally desirable. In any event, if the objective is to reach some general truths about law, it cannot be a recommendation of a theory that it wholly excludes the possibility of democratic law. It seems to me that a theory which holds that only non-democratic arrangements can count as "law" is a theory to be rejected on moral grounds.

(2) The second noteworthy point in this chapter is that statute law at least resembles a coercive order since it is a deliberate datable act, but Custom, though it plays a modest role in our law is unlike this. But both Austin and Kelsen found ways of reconciling the apparently difficult nature of custom within theory definitions of law and whereas Austin's account in terms of the fiction "What the Sovereign permits the Sovereign commands" may be an implausible post hoc rationalisation, to shore up an unconvincing theory, Kelsen's account of custom as a constitutionally authorised source of law may seem acceptable, or at least as acceptable as any treatment of the subject offered by Hart and when we come to Simpson's criticisms (towards the end of this series of lectures) we will see that the possibility of Hart's own theory satisfactorily accommodating non-statutory law is at least open to serious question.


Chapter IV: Normativity, Habits, Rules, and Constitutional Law. The central point of this chapter is that we need to substitute for sovereigns and habits of obedience the idea of a fundamental rule. That is we must move from a descriptive or factual account to a normative account of law: 76. "... the persistence of laws ... is something which cannot be made intelligible in terms of the simple scheme which conceives of laws as orders given by a person habitually obeyed". (Hart, op.cit., p.60) and 77. "... it [persistence] involves the substitution, for the too simple notion of the habit of obedience to a sovereign person, the notion of currently accepted fundamental rules specifying a class or line of persons whose word is to constitute a standard of behaviour for the society, i.e. who have the right to legislate". (Hart, op.cit., p.61).

Only in this way, says Hart, can we adequately account for the continuity of law and the persistence of law. Thus we have a critique of Austin's concept of Sovereignty as a basis of law; it would involve an interregnum until the "habit of obedience" built up again. A habit is unreflective, effortless and ingrained in character, though it also entails that behaviour has been adopted for some considerable time and is likely to be repeated in the future. The habit of obedience is a personal relationship between subject and sovereign ... 78. "... all that is required from the community to constitute Rex the sovereign are the personal acts of obedience on the part of the population. Each of them needs, for his part, only obey; and, so long as obedience is regularly forthcoming, no one in the community need have or express any views as to whether his own or others' obedience to Rex is in any sense right, proper, or legitimately demanded". (Hart, op.cit., pp.51-52).

Hart postulates a simple community with a sovereign who receives habitual obedience. If RexI were to die, there would be no immediate likelihood that Rex II will be obeyed; no habit of obedience to his orders has yet developed. Hart is concerned here with the continuity of a legal system rather than with the nature or law. There must be, says Hart, more than a mere habit of obedience. There must be recognised rules under which the right to issue orders may be transmitted. Thus ... 79. "If there is to be this right and this presumption at the moment of succession there must, during the reign of the earlier legislator, have been somewhere in the society a general social practice more complex than any that can be described in terms of habit of obedience: there must have been an acceptance of the rule under which the new legislator is entitled to succeed". (Hart, op.cit., p.54).

Here Hart introduces a difference between Austin's "habit" and his own "acceptance". The latter implies a reflective critical attitude; the right to rule comes from this latter. Austin was concerned with the fact of habitual obedience; Hart with the quality of such obedience; only if it is reflective and critical does it amount to "acceptance"; only then is the right to make rules established; i.e. though habit cannot produce an ought, apparently acceptance can. Of course, Austin was not unaware of the need for rules by which Sovereignty could be transmitted ... 80. "In order that a supreme government may possess much stability, and that the society wherein it is supreme may enjoy much tranquillity, the persons who take the sovereignty in the way of succession, must take or acquire by a given generic mode, or by given generic modes. Or... they must take by reason of their answering to a given generic description...". (John Austin, op.cit., p.152). Austin, however, held that rules of this type were positive morality merely; Hart insists that they are legal rules; i.e. Constitutional Law.

Hart reiterates his distinction between habits and rules:

1. Habit:- enough if the behaviour of a group converges; deviation need not be a matter for criticism. Rule:- deviation generally regarded as lapses or faults open to criticism; and they meet with pressure for conformity.

2. Where there are rules, they justify criticism.

3. The internal aspect of rules: A general habit is just a fact about the behaviour of most of the group; it is not maintained, nor consciously taught. But a social rule implies that some at least look upon the behaviour as a standard to be followed and maintained; thus rules have an internal aspect which habits lack. But this internal aspect is not a psychological phenomenon; the idea of "feeling bound" is neither sufficient nor necessary for the existence of social rules ... 81. "There is no contradiction in saying that people accept certain rules but experience no such feelings of compulsion. What is necessary is that there should be a critical reflective attitude to certain patterns of behaviour as a common standard and that this should display itself in criticism (including self-criticism), demands for conformity, and in acknowledgements that such criticism and demands are justified, all of which find their characteristic expression in the normative terminology of 'ought', 'must', and 'should', 'right' and 'wrong'. These are the crucial features which distinguish social rules from mere group habits". (Hart, op.cit., p.56).

It is in this light that we are to understand the rule by virtue of which Rex II has a right to legislate; i.e. Rex I and subsequently Rex II are rulers by virtue of 82. "... acknowledgements that obedience is something to which he has a right by virtue of his qualification under a general rule". (Hart, op.cit., p.57). Thus, for Hart, it is "acceptance" which constitutes a sovereign's right to rule; but, of course ... 83. "... acceptance of a rule by a society at one moment does not guarantee its continued existence". (Hart, op.cit., p.58) This may appear to contradict the explanation which Hart seeks to give to for the continuity of law . Still, says Hart, 84. "... in principle the matter is clear. The statement that a new legislator has a right to legislate pre-supposes the existence, in the social group, of the rule under which he has this right". (Hart, op.cit., p.58)

Hart summarises ... 85. "Even in we concede that a person such as Rex, whose general orders are habitually obeyed, may be called a legislator and his orders laws, habits of obedience to each of a succession of such legislators are not enough to account for the right of a successor to succeed and for the consequent continuity in legislative power. First, because habits are not 'normative': they cannot confer rights or authority on anyone. Secondly, because habits of obedience to one individual cannot, though accepted rules can, refer to a class or line of future successive legislators as well as to the current legislator..." (ibid). Yet, as Hart admits, it would be absurd to think of the mass of the population in a modern and complex state as having a clear understanding of the rules qualifying a changing body of persons as a legislature; i.e. our old friend "the educated man" has lost all trace of his earlier "ease and confidence" by now ! 86. "To speak of the populace 'accepting' these rules, in the same way as the members of some small tribe might accept the rule giving authority to its successive chiefs, would involve putting into the heads of ordinary citizens an understanding of constitutional matters which they might not have. We would only require such an understanding of the officials or experts of the system". (Hart, op.cit., p.59). Thus Hart's position reduces to the view that officials accept these rules and ordinary citizens acquiesce ... 87. "The ordinary citizen manifests his acceptance largely by acquiescence in the results of these official operations". (Hart, op.cit., p.60). So (1) the ordinary citizen is less of an authority than he appeared to be in Chapter I; and (2) Because a distinction falls to be made even on Hart's own account between officials and citizens, it seems that his criticism of Kelsen and of Ross who conceive of law as directives to officials may be less valid than he believes.

I have three points to make about Hart's views as developed in this chapter. First, a point to which I often return in these and other lectures, Hart is not nearly critical enough of Austin's notion of sovereignty. Secondly, it is arguably unfair to criticise Austin for any failure to include rules of Constitutional law within his scheme of things when Austin expressly, and as we have concluded, rightly on his own premises, treated constitutional law as an exception to his theory and beyond his definition of law. As we have seen, Austin was well aware of the desirability of rules of succession. However, he classified such rules as positive morality merely and not as law. Thirdly, Hart's account of official acceptance and citizen acquiescence may obscure an important distinction between types of government: 88. "There are two kinds of kingdom, one of which is a lordship called in Latin dominium regale and the other is called dominium politicum et regale. And they differ in that the first king may rule his people by such laws as he himself makes. And therefore he may set upon them tallages and other impositions, such as he himself wills without their consent. The second King may not rule his people by laws other than those to which they consent. And therefore he may set upon them no impositions without their own consent". (Sir John Fortescue, "Royal and Political Dominium", On the Governance of England, c. 1473, cited in Lively & Lively, Democracy in Britain: A Reader, p 37). This distinction parallels that between divinely appointed rulers (such as the view James VI and I had of himself : 89 "The state of Monarchy is the supremest thing upon earth: for Kings are not only God's lieutenants upon earth, and sit upon God's throne, but even by God himself they are called Gods". - Lively & Lively, 38-9) and rulers by acclamation 90 "Yet Robert [our most valiant Sovereign, King, and Lord, King Robert], himself, should he turn aside from the task that he has begun, and yield Scotland or us to the English King and people, we should cast out as the enemy of us all, as subverter of our rights and of his own, and should choose another king to defend our freedom: for so long as a hundred of us are left alive, we will yield in no least way to English dominion. We fight not for glory nor for wealth nor honours; but only and alone we fight for freedom, which no good man surrenders but with his life". Declaration of Arbroath, 1320; it parallels, too, a distinction between the Scottish notion of conditional sovereignty and the English notion of absolute sovereignty.

Chapter V: The Fresh Start; primary and Secondary Rules; the D-Rule Society I commence my review of this chapter with another bon mot from Gellner's ample stock 91. "There are two main forms of philosophical arrogance: one is to claim to encapsulate all past thinkers in one's thoughts: the other is to claim to be initiating an entirely new departure". (Ernest Gellner, Words and Things, p.137). Next I draw attention to an important point made by Kelsen which may call into question the "freshness" of Hart's approach: 92."It is impossible to grasp the nature of law if we limit our attention to the single isolated rule. The relations which link together the particular rules of a legal order are also essential to the nature of law". (Hans Kelsen, General Theory of Law and State, p.3).

However, Kelsen, himself is dubious about the utility of the concept of a rule as indeed Hart was at an earlier stage of his discourse. In the following passage, Kelsen raises two objections: 93."If law is characterised as 'rules', it must be stressed that legal rules essentially differ from other rules and in particular from those which are presented as laws of nature (in the sense of physics). Whereas laws of nature are statements about the actual course of events, legal rules are prescriptions for the behaviour of men. Laws of nature are rules which describe how natural evens actually occur and why these events occur; that is to say, what are their causes. Rules of law refer only to human behaviour. They state how men ought to behave, and say nothing about the actual behaviour of men and of the causes thereof. In order to prevent misunderstandings (as to the nature of law) it is therefore better in this context not to use the term 'rule', but to characterise law as norms.

Another reason why the designation of law as 'rule' is misleading is that the word 'rule' carries the connotation of something 'general'. A 'rule' does not refer to a single non-recurring event but to a whole class of similar events. The importance of a rule is that a phenomenon of a certain kind occurs - or ought to occur - always, or almost always when conditions of a certain kind are fulfilled. In fact, law is often explained as 'general rules'. Austin draws an explicit distinction between 'laws' and 'particular commands': where a command, he says, 'obliges generally to acts or forbearances of a class, a command is a law or rule. But where it obliges to a specific act or forbearance... a command is occasional or particular'. Having identified 'law' and 'rule', we can, of course, recognise as law only general norms. But there is no doubt that law does not consist of general norms only. Law includes individual norms; i.e. norms which determine the behaviour of one individual in one non-recurring situation and which, therefore, are valid only for one particular case, and may be obeyed or applied only once. Such norms are 'law' because they are parts of the legal order as a whole in exactly the same sense as those general norms on the basis of which they have been created. Examples of such particular norms are the decisions of courts as far as their binding force is limited to the particular case in hand. Suppose that a judge orders a debtor, A, to return $1,000 to his creditor, B. By expressly or tacitly threatening A with a civil sanction in case of non-payment the judge here 'commands' A to pay $1,000 to B. The decision of the judge is a legal norm in the same sense and for the same reasons as the general principle that if someone does not return a loan then a civil sanction ought to be inflicted upon him on the motion of the creditor. The 'binding force' or 'validity' of law is intrinsically related, not to its possibly general character, but only to its character as a norm. Since, by its nature, law is norm, there is no reason why only general norms should be considered law. If, in other respects, individual norms present the essential characteristics of law, they too must be recognised as law". (Hans Kelsen, General Theory of Law and State, pp.37-38).

Kelsen has made two important points here : (1) the notion of a "rule" is misleading because it may obscure the prescriptive element; (2) Even if the prescriptive element is stressed, the notion of a "rule" is inadequate as a "key" to legal science since it is irreducibly general whereas legal norms and decisions may be particular. Hart, of course, believes the concept of a rule to be central and the main defect of Austin's account is, for him, its inability to generate normative concepts in general and the vital concept of a rule in particular: 94. "... the ideas of orders, obedience, habits and threats, do not include, and cannot by their combination yield, the idea of a rule, without which we cannot hope to elucidate even the most elementary forms of law". (Hart, op.cit., p.78). So Kelsen and Hart agree that Austin's account is inadequate because it is not normative but they differ as to what normative concept best captures the nature of law. For Kelsen , the answer is "norm"; for Hart, despite some tergiversation, it is "rules", and moreover, different types of rules.

Types of rules:-

1. Primary Rules: Human beings are required to do or abstain from certain actions, whether they wish to or not; i.e. duty imposing rules, "D-rules";

2. Secondary Rules: are 95. "... in a sense parasitic". (Hart, op.cit., p.79) which again renders the thesis that P-rules are separate from D-rules implausible. These provide that human beings may introduce new rules of the primary type, extinguish or modify old ones, etc. Such rules confer public or private powers; i.e. "P-rules". And ... 96. "... in the combination of these two types of rule there lies what Austin wrongly claimed to have found in the notion of coercive orders, namely, 'the key to the science of jurisprudence'". (ibid). This union of elements is central because of ... 97. "... their explanatory power in elucidating the concepts that constitute the framework of legal thought". (ibid).

Now this is a very large claim. To fulfill it, Hart needs to show that "the concepts which constitute the framework of legal thought" can be related to his union of rules. Such concepts must include at least "rights", "duties", "legal personality", "property", "crime", "delict", "punishment" or "sanctions", and many more. To demonstrate that the union of two types of rules can explain all these concepts, Hart ought to produce a series of conceptual analyses. But in fact we get very little analytical jurisprudence in the remainder of the Concept; we do get a lot of linguistic or philosophical analysis, and some sociology, neither of which can demonstrate Hart's thesis that the conceptual framework of legal thought can be best elucidated in terms of a union of two types of rules. We should get ... 98. "... a new account of law in terms of the interplay of primary and secondary rules". (Hart, op.cit., p.80).

Now we do get an analysis of "obligation" which, rather than showing a relationship between this concept and Hart's "key", merely asserts that the existence of a rule is the normal, tacit, background to statements about obligation. Hart allows that even the Austinian position took off from a valid insight, viz. Where there is law, then human conduct is in some sense non-optional. But a mistake was made. The non-optional nature of some forms of conduct was linked to coercive orders; under the rubric that in a situation of the gunman type the bank teller was obliged to hand over the money. But, says Hart, this overlooks that ... 99. "There is a difference, yet to be explained, between the assertion that someone was obliged to do something and the assertion that he had an obligation to do it". (ibid).