¹ The Concept of Law, by Hart, H. L. A., Professor of Jurisprudence in the University of Oxford. Clarendon Law Series. Oxford: Clarendon Press, 1961, x and 263 pp., 21s.Google Scholar

² See “The Concept of a Lawyer's Jurisprudence” in C.L.J., Vol. II, No. 2, 1952 at p. 229, continued in No. 3, 1953 at p. 404.

³ Dr. Kantorowicz claims to have invented this term, see The Definition of Law, 1958Google Scholar, Chap. 1, note 8. It denotes an attitude which may be expressed as follows: “Any question posed by any science as to the meaning of a term can be answered only if the intention is to ask what in this particuiar science ought to be understood by this particular term … (p. 5) … the jurist, the sociologist, the political scientist and the social philosopher must each ask ‘what ought to be understood by law?’ if he wishes the definition to be useful for jurisprudence, sociology, political science, or social philosophy” (p. 6).

⁴ In Prof. Hart's terminology “primary rules are concerned with the actions that individuals must or must not do … secondary rules are all concerned with the primary rules themselves. They specify the ways in which the primary rules may be conclusively ascertained, introduced, eliminated, varied, and the fact of their violation conclusively determined” (p. 92). In a society lacking a legislature, courts or officials of any kind, Prof. Hart would refer to the social structure as one of primary rules of obligation (p. 89). Such a society, he suggests, would suffer from the defects of the uncertainty, the static quality and the inefficiency of a régime of social custom. To remedy these defects there come into existence rules of recognition (providing criteria for the validity of other rules), rules of change and rules of adjudication. In “the structure which has resulted from the combination of primary rules of obligation with the secondary rules of recognition, change and adjudication” Prof. Hart finds “the heart of a legal system” (p. 95). “For the introduction into society of rules enabling legislators to change and add to the rules of duty, and judges to determine when the rules of duty have been broken …may fairly be considered as the step from the pre-legal into the legal world” (p. 41).

⁵ It may be helpful, at this point, to anticipate a little. In comparing the concept of law, as elucidated by Prof. Hart, with the concept of law constructed by the writer, it must be remembered that it is not, of course, the same phenomenon, law, which is being differently conceived—any more than it is the same year, 1963, which is so designated according to the Julian and the Georgian reckonings. It is simply that the exploration of social phenomena is being undertaken with the aid of different basic jurisprudential concepts and thus a different terminology. Briefly Prof. Hart conceives law as a system of rules, operating effectively in action—a concrete structure of society; the writer conceives a legal system as a system of action, operating in general accordance with norms—an analytically distinguishable structure of society. For the difference between these two types of social structure, see Marion, J. Levy, Jr. The Structure of Society, Princeton University Press, 1952, pp. 35, 36.Google Scholar Prof. Hart's initially linguistic approach precludes him from using the un-English word “norm”—but his “legal rules,” in what he calls their “internal aspect,” are accepted behaviour patterns not essentially different in structure from the writer's “legal norms.” These are expressed in statements, made from within legal systems, in the general form “in any situation of the type X the courts ought to apply a sanction of the type Y.” The “external” or “predictive” aspect of Prof. Hart's legal rules, on the other hand, is that uniformity of conduct which grounds and justifies the predictions of judicial behaviour which, in the writer's terminology, state “legal rules.” Here the general form of expression is “in any situation of type X the court will probably apply a sanction of type Y“— an “external” statement in essentially the form in which all scientific law is expressed. The writer's legal norms and legal rules, then, correspond closely to Prof. Hart's legal rules in their internal and external aspects, and are expressed in the two different forms of statement, “internal” and “external,” the distinction between which is regarded by Prof. Hart as being crucial. That the writer conceives substantive norms as addressed primarily to the courts is balanced by the fact that Prof. Hart so conceives the secondary rules, in accordance with which primary rules of obligation are made—although he conceives these secondary rules as requring the courts to apply, not sanctions in certain circumstances, but the primary rules which provide such sanctions for those circumstances.

The writer's “predictive theory” in which a legal system is conceived primarily as a structure of action, is not, however, motivated by “the conviction that only thus can metaphysical interpretation be avoided …” (p. 101). It is motivated, as Prof. Hart's concept of law is motivated, by the hypothesis that this provides the most fruitful approach to the understanding of the lawyer's specific experience—or “a most powerful tool for the analysis of much that has puzzled both the jurist and the political scientist” (p. 95).

⁶ It will be assumed that, even when it is by a natural process of growth that one kind of thing is marked off from another by a separate word, “language does not give name to pre-existing things or concepts so much as it articulates the world of our experience” (Gombrich, E. H., Art and Illusion (London: Phaidon Press, 1960), p. 90).Google Scholar For the difference between verbal realism and the conceptual pragmatism here advocated see Kantorowicz, The Definition of Law (1958), Chap. I.

⁷ Definition, Oxford, 1952.Google Scholar A “lexical” or “dictionary” definition gives rules of correct usage. A “stipulative” definition is a proposal to use a word in a certain way. A “real” or Aristotelian definition involves verbal realism and the theory of distinct “natural” kinds. Prof. Hart usually appears to be seeking a “real” definition of law by linguistic methods. See note 13, infra.

⁸ The relation of law to morality and coercion may be seen, according to Prof. Hart, in the truism that both law and morality must afford minimum protection to person, property and good faith by the enforcement of sanctions. This assertion is said to “disclose the core of good sense in the doctrine of Natural Law” (p. 194), and sets out the conditions under which social life is possible for men who are observed to be approximately equal in strength, limited in altruism, vulnerable to attack, and are facing limited resources with limited understanding and strength of will.

⁹ Attempts to explain the “nature” of law at the same time as defining the term “law” have resulted in making it seldom possible to say exactly what is being meant by “law,” or, in other words, what it is whose “nature” is being explained. Intelligible explanation of law or of anything else seems to presuppose precise identification or definition—in a separate operation. “Define first, explain afterwards” would seem to be a useful operational maxim, if misunderstandings and arguments at cross-purposes are to be avoided.

¹⁰ There is a sense in which “familiarity with the law” is not an illusion—learned counsel certainly possess it. But it seems legitimate, in the context, to treat the expression as equivalent to “familiarity with law,” or, adopting the previously used expression, “familiarity with the phenomena to which the word ‘law’ commonly refers.” After all it is the concept “law” and the question “what is law?” which is under consideration,—not the very different concept “the law.” It should occasion little surprise that those who have “familiarity with the law” should be puzzled by “law,” any more than that it should be surprising that those who have “familiarity with the time” should be puzzled by “time.” And surely by “ability to recognise examples” Professor Hart means “examples of law” or “examples of the phenomena to which the word ‘law’ commonly refers“—rather than “examples of the law”?

¹¹ Philosophy and Language (Oxford, 1960) at p. 24.Google Scholar

¹² There are passages on pp. 24 and 25 of Prof. Ayer's inaugural lecture, op. cit. supra, which are of interest for the light they throw on the techniques, often adopted by Prof. Hart, of using language for taking “a new look at the facts.” This is said to be not so much a question of asking what makes it correct to use certain expressions and concepts, or what are the typical circumstances in which we should use them, as rather a question of asking, given that we do use these expressions and concepts, what are the circumstances that would make the statements in which we use them true? The procedure described of “…starting with an identification of the facts and then seeing how they can best be described” rather than “starting with the words and then looking for the facts to which to fit them” (p. 25) is precisely the order of inquiry advocated for jurisprudence by the writer. The question is how to identify “facts” as “law.” Must this be done solely through and by means of the existing use of the word “law,” or “an examination of the standard uses of the relevant expressions” (Hart, p. vi);—or is it possible to make a fresh start by considering the role or function which the word or concept “law” performs? Prof. Hart recognises that “the standard terminology of legal and political thought, developed in the shadow of a misleading theory, is apt to over-simplify and obscure the facts” (p. 109). Yet when Prof. Hart tries to “take off the spectacles constituted by this terminology and look at the facts” (p. 109) he seems to find them already articulated by the concept we are trying to get behind—“it becomes apparent that a legal system …” (p. 109). Is it not possible to take a fresh look at the general field of experience traditionally articulated by the concept of law, independently of such articulation? The writer has characterised this field as that of “the lawyer's specific experience ”—and finds it centred on a court or system of courts.

¹³ Consider the ambiguous opening sentence of the Preface in which Prof. Hart states his aim “ … to further the understanding of law, coercion and morality as different but related social phenomena”—and how this differs from “ … to further the understanding of three different but related social phenomena, which we shall call law, coercion, and morality.” On p. 2 we are told that answers to the persistent question “What is law?” are given by serious thinkers as “the outcome of prolonged reflection” —not on how the word “law” is used or might be used—but “on law.” On p. 6 we are told with confidence that “the most prominent general features of law at all times and places is. …” And at length at p. 95, “ … it is plain that we have here … the heart of a legal system.”

The whole basis of Prof. Hart's criticism of Austin is less that Austin's theory of law has no counterpart in reality than that law is different from what Austin says, “ … we have seen that … the simple model of law as the sovereign's coercive orders failed to reproduce some of the salient features of a legal system” (p. 77). The difficulties of legal theory, as they seem to be seen by Prof. Hart, are not indeed due to the fact that a concept of a law adequate to the needs of an empirical jurisprudence has proved elusive, but arise because “certain features of law … have proved most perplexing and have both provoked and eluded the search for definition” (p. 79). “It is,” however, “a virtue of the idea which we have taken as central that it permits us to see the multiple relationships between law, coercion and morality for what they are …” (p. 208).

¹⁴ Prof. Hart's apparently triple reliance on truth, linguistic usage and theoretical and practical convenience as standards for differentiating types of phenomena may be again seen in his treatment of morality. To assert that legal rules only require “external” behaviour while morals require only good will, does, he says, “contain a hint of the truth” (p. 168). But he adopts a broad view of morality “not merely because the weight of usage of the word ‘moral’ favours the broader meaning, but because to take the narrower restricted view … would force us to divide in a very unrealistic manner elements in a social structure which function in an identical manner” (p. 177).

It may be suggested that while a simple “analysis” (p. 151) or “elucidation” (p. 151) may be appropriate to a concept grounded in truth or linguistic usage, in the case of a concept grounded in theoretical and practical convenience it must be counted as a happy chance if an a priori “construction” of the concept proves unnecessary.

¹⁵ If a concept is being avowedly constructed on grounds of theoretical convenience it is surely unnecessary to resort to any simple version of Natural Law in order to justify the conception of municipal law in terms of sanctions. Yet Prof. Hart seems to regard such resort as necessary in order to escape “two unsuitable alternatives which are often taken as exhaustive; on the one hand that of saying that this is required by ‘the’ meaning of the words ‘law’ or ‘legal system’ and on the other that of saying that it is ‘just a fact’ that most legal systems do provide for sanctions” (p. 195). Surely sanctions are a “natural necessity” (p. 195) to municipal law only when this is conceived in roughly the same way as Prof. Hart conceives it—in accordance with the fourth alternative ground for his conception, theoretical convenience.

¹⁶ Little, certainly, is said to establish the superior convenience of Prof. Hart's classification over the still broader “functional” concepts of law adopted by anthropologists like Malinowski and his followers, Radcliffe-Brown, A. R. (in his preface to African Political Systems. Edited by M., Fortes and Evans-Pritchard, E. E., London, 1940)Google Scholar and Max, Gluckman (in The Judicial Process Among the Barotse of Northern Rhodesia, Manchester, 1955).Google Scholar See also Mair, L. in Primitive Government (Penguin Books, 1962) at pp. 18 and 19.Google Scholar Whatever may be the case regarding the general understanding of man in society it seems clear, however, that for understanding the specific institutions of the modern western state Prof. Hart's classification is the more appropriate. A concept of law like Radcliffe-Brown's, which did not entail courts or judges, would certainly prove confusing to the modern legal system” (p. 17), rather than as theoretical sociologist, that one wishes to isolate for academic study “the specific method of social control to be seen in a system of primary and secondary rules.” There is something to be said for frankly constructing a “lawyer's” concept of law, as the basis of a “lawyer's jurisprudence.”

¹⁷ Praise of Law: The Origin of Legal Philosophy and the Greeks. By Jaeger, Werner. Interpretations of Modern Legal Philosophies. Edited by Paul, Sayre (O.U.P., 1947) at p. 361.Google Scholar As it was the content of this vóμoς rather than the content of sovereign will which, under the influence of Greek philosophy, the Roman juriste discussed in their philosophy of “law,” it is not altogether surprising that the incomprehending Austin could say of them “it is hardly conceivable how men of such admirable discernment should have displayed such contemptible imbecility” (Jurisprudence, 1885, at p. 217).

¹⁸ , St. Augustine I, De libero ArbitrioGoogle Scholar, 5.

¹⁹ Digest i.4.1.

²⁰ The Antigone of Sophocles, Fr. 345 (Dindorf); translated by E. Barker.

²¹ An Introduction to the Philosophy of Law, Pound, Roscoe (revised ed., Yale University Press, 1954) at p. 25.Google Scholar

²² It may be instructive to apply here Prof. Hart's “crucial” distinctions between “two different kinds of statement … ‘internal’ and ‘external,’ and which can both be made whenever social rules are observed” (p. vi). On this basis Prof. Hart's concept of law might be said to incorporate an “external” statement of the nature of law. If we now consider “internal” statements as to the nature of law which might characteristically be attributed to English and American judges respectively, would we expect any difference to emerge between these internal statements which might be attributable to the fact that the ultimate rule of recognition in the one system is, according to Prof. Hart, “What the Queen in Parliament enacts is law” (p. 104), and in the other “the ultimate criterion of legal validity explicitly incorporates principles of justice or substantive moral values” (p. 199). If the American judicial (or academic) concept of law should prove to be rather more “laden with the theory of Natural Law” than its English counterpart the difference between the concepts may be more readily explicable as due to local differences of function than to any puzzlement of fact. And we may have here the key to typical differences between “external” statements as to the nature of law emanating from the two common law countries. It is natural to clothe the universal in national costume.

²³ See the writer's article “Propositions about Law” in [1951] C.L.J., p. 31.

²⁴ In his otherwise admirable critical notice in Mind, Vol. LXXI No. 283, July 1962 at pp. 398–399, L. J. Cohen does not seem to realise the possibility of elucidating specifically legal concepts by means of specially defined terms or “ideal types” even when these have the same designation as the concepts being elucidated. Thus there is no reason why the ordinary legal usage of “power” should not be elucidated by means of a juristic definition of “power,” e.g., by showing when the former did and did not correspond with the latter. This misdirected criticism, based on linguistic usage, is perhaps again due to Prof. Hart's failure to insist on the status of his concept as a convenient tool of thought.

²⁵ “Only when there are certain types of laws giving men jurisdiction to try cases … do they constitute a court” (p. 5). It will later be shown, however, that this statement in the text is misleading in the absence of an important qualification.

²⁶ See also note 5, supra.

²⁷ Acceptance as a common standard involves “a critical reflective attitude to certain patterns of behaviour … and that this should display itself in criticism (including self-criticism), demands for conformity, and in acknowledgments that such criticism and demands are justified, all of which find their characteristic expression in the normative terminology of ‘ought,’ ‘must,’ ‘should,’ ‘right’ and ‘wrong’ ” (p. 56).

²⁸ One may be said to maintain an attitude of acceptance towards a particular course of conduct, or, generally, towards courses of conduct formulated or to be formulated in a certain way. In a simple structure of rules of primary obligation the first type of acceptance only is involved. A rule of recognition implies both the first and the second. The first acceptance—of the way of formulating further courses of action—may, however, be said to be directly given, while the acceptance given in advance to courses of action to be formulated in that way—the content of rules of primary obligation—may be said to have been given to those courses of action, not directly, but at one remove. We have here an example of the logic of authority, of delegation, or of ought propositions, on which Hans Kelsen so heavily relies. To accept as authoritative a way of making rules entails the acceptance as authoritative of all rules so made. In this way the acceptance “as valid” of Kelsen's grundnarm entails the validation of all rules made in accordance with it—and the acceptance by the courts of the rule of recognition binds them to apply rules made in accordance with it. Kelsen qualified this acceptance as an acceptance “as valid” in an effort to avoid the linguistic mistake of analysing “positives Recht” into simple facts. He might have been better advised to have broken with linguistic usage. For “as valid” Prof. Hart simply substitutes “as a standard,” see note 27, supra, and is enabled thereby to dispense with the Neo-Kantian notion of validity or das Sollen.

²⁹ The alternative is that they are law by virtue of the rule of recognition, which is law by virtue of acceptance by the courts they constitute. Thus the circularity in definition involved is just as real if not quite so obvious.

³⁰ It has been assumed above that Prof. Hart envisages that just aa the enactment of the rule of recognition as a statute will make no difference to its legal status, so too the enactment of the rules of change and adjudication will make no difference to their similar status. It may well be asked what would be the position if an attempt was made by statute to change these rules whose legal status depends on the fact of being accepted standards. There would arise, of course, a possibility of conflict between two sets of rules of different status. In the stable legal system, however, any possibility of conflict would promptly be removed by the conformity of practice to the statutory enactment, unless the courts decided that, for one reason or another, the legislature had exceeded the powers which the courts were prepared to allow. The conflict might then pass from the stage of one between rules of different status to one between courts and legislature for the allegiance of the citizen.

³¹ The definition proposed by the writer in C.L.J., Vol. EC, No. 3, 1953 atp. 410 is as follows: “A court is defined as a tribunal (a person or definite group of persons) which regularly and effectively decides issues by authorising or refusing to authorise the application of force to persons and which supports its decisions by reference to a known and limited number of agencies which it recognises as norm creating and accepts as authoritative in a definite hierarchy of precedences. Sources of legal norms are defined as such accepted agencies, and legal norms are defined as guides to action deriving from them.” Implicit in this is a definition of law as a system of norms followed and/or enforced in courts. This definition will not, however, here further be elaborated.

³² Op. cit., C.L.J. Vol. II, No. 2, 1952 at p. 229Google Scholar, continued in No. 3, 1953 at p. 404.

³³ Op. cit. at p. 7.

³⁴ Thus discussion still flourishes about the exact legislative power of Queen in Parliament. It may yet be a cause for congratulation that the Privy Council is not regarded as being bound by its own decisions—and that it was, in any case, only a dictum when the court expressed the view that the imperial parliament, despite the express words of s. 4 of the Statute of Westminster, 1931, was incapable of surrendering the power to legislate for the Dominion of Canada (British Coal Corporation v. The King [1935]Google Scholar A.C. 500, 520). As the court went on to remark that this was a position of “abstract law” which had “no relation to realities” it may be assumed that they thought the situation demanded some other solution—the ultimate rule of recognition some other content.

³⁵ If so it would only be another example of the way theories of law operate in their “guidance” aspects, or as “judicial” or “internal” concepts of law. Prof. Hart has shown how German judicial thinking has been affected by the concept of law as part of morality (p. 204) and English judicial thinking by the concept of law as coercion or sovereign will (p. 145). The incapacity to bind itself now attributed to Queen in Parliament, for instance, owes much to the supposed necessity of giving effect to the Austinian conception of a sovereign power. There is no reason to doubt that Prof. Hart's elucidation of the relation of law to his third puzzling category “rules”—its dependence on a rale of recognition—if accepted, would be equally influential.

³⁶ That the spectacle of a constitution conferring authority to say what a constitution is should appear paradoxical is possibly due to a false analogy suggested by the concept of law as a system of rules of primary obligation logically dependent on a rule of recognition. The false analogy is with the deductive logic of propositions whereby particular rules are deduced from general principles, to whose content or authority the deductions cannot add. In deductive reasoning the authority for propositions descends in a Bingle order. But in the logical structure of legal rules or norms there is no reason why authority should not move in a circular direction. As in social life authority rises by acceptance from below before it descends by delegation from above; so that a constitution which gives the electorate power to amend it by a referendum, or a court power to interpret it, seems both natural andjust. The reason for these free movements in law and in social life is exactly the same, i.e., that authority—whether it be of person, source, or rule—is simply the obverse of attitudes of acceptance by fundamentally the same parties as are subject to it. The free movement of authority may be relied on by a court seeking a technique for creative work at the heart of a legal system. But a better technique has been suggested in the text.

³⁷ It is not clear why at this point Professor Hart appeals to the criterion of truth, when it appears to be statements of law within legal systems which are under discussion.

³⁸ [1950] 2 K.B. 368.

³⁹ It may be instructive to consider how close the Supreme Court of South Africa came to this predicament, and by what means it was avoided, in Harris and Others v. Donges and Another [1952]Google Scholar 1 T.L.R. 1245, repudiating its own opinion in Ndlwana v. Hofmeyer (1937)Google Scholar A.D. 229; in Minister of the Interior and Another v. Harris and Others (1952) 4 S.A.L.R. 769Google Scholar, and in the last case of the series, granting victory to the South African Government. Collins v. Ministry of Interior, 1957 (1) S.A. 552Google Scholar (A.D.).

⁴⁰ Introduction to Year Books of Edward II, Selden Society Year Book Series, Vol. I at pp. xviii, xix.

⁴¹ Prof. Hart recognises this point, but not always all its implications. No analysis is given of the development, as distinct from the existence, of standards of conduct in a social group. On p. 90 he envisages a primitive group living by primary rules of obligation which do not “form a system, but will simply be a set of separate standards.” But if the group is living as a group at all their standards cannot be “separate.” There may be no perceptible logical relation between its rules, but these must inevitably reflect a more or less integrated structure or system of values. The practical modes of relating values are surely as important to the understanding of legal material as the logical modes of relating rules or norms. If law is conceived as a system of action incorporating a system of norms attention is directed to the interplay of values and rules or norms—to the way in which the pursuit of values leads to the deposit of norms and norms create new values. Legal reasoning, in the absence of absolutely binding authority, is seen to be less a matter of logic than a balancing or weighing of interests or social values—a summing-up of pros and cons in order to allow the realisation of the most highly ranking or weightiest values or the greatest sum total of positive values involved in a particular situation. The exception to the general rule, so distasteful to the analytical jurist, is seen, for instance, to protect the higher value—unless of course the doctrine be invoked that hard cases make bad law and the exception, together with the higher value, be suppressed.

⁴² See , Parsons and , Shils (editors), Toward a General Theory of Action (Harvard University Press, 1951)Google Scholar and , ParsonsThe Social System (Tavistock Publications, Ltd., London, 1952).Google Scholar Action theory, as developed in these works, distinguishes personality systems, social systems and systems of cultural norms, with different units and different modes of integration, as the special provinces respectively of psychology, sociology and anthropology. It has been evolved by a group of workers in these fields as a contribution towards facilitating their co-operation and cross-fertilisation by the establishment and use of a common conceptual scheme. The thesis being maintained is that jurisprudence should not restrict itself to the study of law as the logical structure of a system of cultural norms, and that this end is best furthered by adopting as its basic concept a definition of law in terms of action.

⁴³ See note 2, supra.

⁴⁴ In all these cases the reaction of courts to human action will probably not be what legally it ought to be.

⁴⁵ In these cases it is human action which, rightly or wrongly, is not what it legally ought to be—or is not what it is, or was, presupposed to be by legal institutions which are accordingly failing in their intended functions.

⁴⁶ There might be some advantages, too, for sociology. In his Law and Sociology (The Free Press of Glencoe, 1962), the editor, William, M. EvanGoogle Scholar, observes that in this interdisciplinary field at least three basic questions remain to be answered:

1. What empirical data being accumulated in law could be illuminated by sociological concepts, and vice versa?

2. What concepts and hypotheses developed in law open up new problems and stimulate research in sociology and vice versa?

3. Can the two disciplines bring their respective theoretical frameworks to bear on the investigation of the same empirical problems (p. 9)?

It is hoped that the concept of law as a system of action may contribute at least to the clarification of some of these questions, and prove, especially in relation to the second question, to be an “image of the legal system that has much to offer the sociologist who will work with it” (p. 186). To divert attention from questions like “What is the nature of law?” to questions like “What are the conditions for the existence and persistence of a functioning system of predictable court reactions?” would alone be a service. There could be few simpler pointers to the need for legal education, judicial independence and, above all, the earning of respect for “law” as such.

See also Selznick, Philip “The Sociology of Law” in Sociology Today (New York: Basic Books, 1959).Google Scholar