r958] FIDELITY TO LAW is not the gunman situation writ large. 8 There is no need to dwell here on the inadequacies of the command theory, since Professor hart has already revealed its defects more clearly and succinctly than I could. His conclusion is that the foundation of a legal system is not coercive power, but certain"fundamental accepted rules specifying the essential lawmaking procedures. 9 When I reached this point in his essay, I felt certain that Pro fessor Hart was about to acknowledge an important qualification on his thesis. I confidently expected that he would go on to say something like this: I have insisted throughout on the importance of keeping sharp the distinction between law and morality. The question may now be raised, therefore, as to the nature of these fundamental rules that furnish the framework within which the making of law takes place. On the one hand, they seem to be rules, not of law, but of morality. They derive their efficacy from a general acceptance, which in turn rests ultimately on a per ception that they are right and necessary. They can hardly be to be law in the sense of an authoritative pronouncement, since their function is to state when a pronouncement is authorit- tive. On the other hand, in the daily functioning of the legal sys tem they are often treated and applied much as ordinary rules of law are. Here, then, we must confess there is something that can be called a" merger"of law and morality, and to which the term intersection"is scarcely appropriate Instead of pursuing some such course of thought, to my surprise I found Professor Hart leaving completely untouched the nature of the fundamental rules that make law itself possible, and turn- g his attention instead to what he considers a confusion of thought on the part of the critics of positivism. Leaving out of account his discussion of analytical jurisprud ence argument mething as follows: Two views are associated with the names of Bentham and austin. One is the command theory of law, the other is an insistence on the separation of law and mo- ality. Critics of these writers came in time to perceive -"dimly Professor Hart says -that the command theory is untenable By a loose association of ideas they wrongly supposed that in ad- lancing reasons for rejecting the command theory they had also refuted the view that law and morality must be sharply separated This was a"natural mistake, but plainly a mistake just the same 8 Hart, supra note 2, at 603 g Ibid
FIDELITY TO LAW is not the gunman situation writ large." 8 There is no need to dwell here on the inadequacies of the command theory, since Professor Hart has already revealed its defects more clearly and succinctly than I could. His conclusion is that the foundation of a legal system is not coercive power, but certain "fundamental accepted rules specifying the essential lawmaking procedures." 9 When I reached this point in his essay, I felt certain that Professor Hart was about to acknowledge an important qualification on his thesis. I confidently expected that he would go on to say something like this: I have insisted throughout on the importance of keeping sharp the distinction between law and morality. The question may now be raised, therefore, as to the nature of these fundamental rules that furnish the framework within which the making of law takes place. On the one hand, they seem to be rules, not of law, but of morality. They derive their efficacy from a general acceptance, which in turn rests ultimately on a perception that they are right and necessary. They can hardly be said to be law in the sense of an authoritative pronouncement, since their function is to state when a pronouncement is authoritative. On the other hand, in the daily functioning of the legal system they are often treated and applied much as ordinary rules of law are. Here, then, we must confess there is something that can be called a "merger" of law and morality, and to which the term "intersection" is scarcely appropriate. Instead of pursuing some such course of thought, to my surprise I found Professor Hart leaving completely untouched the nature of the fundamental rules that make law itself possible, and turning his attention instead to what he considers a confusion of thought on the part of the critics of positivism. Leaving out of account his discussion of analytical jurisprudence, his argument runs something as follows: Two views are associated with the names of Bentham and Austin. One is the command theory of law, the other is an insistence on the separation of law and morality. Critics of these writers came in time to perceive - "dimly" Professor Hart says -that the command theory is untenable. By a loose association of ideas they wrongly supposed that in advancing reasons for rejecting the command theory they had also refuted the view that law and morality must be sharply separated. This was a "natural mistake," but plainly a mistake just the same. 8 Hart, supra note 2, at 603. 9 Ibid. i958] 639
HARVARD LAW REVIEW [ Vol.7I I do not think any mistake is committed in believing that Bentham and Austin's error in formulating improperly and too simply the problem of the relation of law and morals was part of a larger error that led to the command theory of law. I think the connection between these two errors can be made clear if we ask ourselves what would have happened to Austin's system of thought if he had abandoned the command theory One who reads Austin's Lectures v and VI cannot help being impressed by the way he hangs doggedly to the command theory, in spite of the fact that every pull of his own keen mind as toward abandoning it In the case of a sovereign monarch law is what the monarch commands. But what shall we say of the“ laws"of succession which tell who the“ lawful” monarch is? it is of the essence of a command that it be addressed by a superior to an inferior, yet in the case of a"sovereign many say, a parliament, the sovereign seems to command itself since a member of parliament may be convicted under a law he him self drafted and voted for. The sovereign must be unlimited in gal power, for who could adjudicate the legal bounds of a su- preme lawmaking power? Yet a"sovereign many"must accept the limitation of rules before it can make law at all. Such a body can gain the power to issue commands only by acting in a"cor- porate capacity"; this it can do only by proceeding"agreeably to the modes and forms""established and accepted for the making of law. Judges exercise a power delegated to them by the su- preme lawmaking power, and are commissioned to carry out its "direct or circuitous commands. Yet in a federal system it is the courts which must resolve conflicts of competence between the federation and its components All of these problems Austin sees with varying degrees of plicitness, and he struggles mightily with them. Over and over again he teeters on the edge of an abandonment of the command theory in favor of what Professor Hart has described as a vic that discerns the foundations of a legal order in"certain funda mental accepted rules specifying the essential lawmaking pro- cedures. "Yet he never takes the plunge. He does not take it be- cause he had a sure insight that it would forfeit the black-and- white distinction between law and morality that was the whole object of his Lectures-indeed, one may say, the enduring ob- 10 I AUsTIN, LECTURES ON JURISPRUDENCE I67-34I(5th ed. I885)
HARVARD LAW REVIEW I do not think any mistake is committed in believing that Bentham and Austin's error in formulating improperly and too simply the problem of the relation of law and morals was part of a larger error that led to the command theory of law. I think the connection between these two errors can be made clear if we ask ourselves what would have happened to Austin's system of thought if he had abandoned the command theory. One who reads Austin's Lectures V and VI 10 cannot help being impressed by the way he hangs doggedly to the command theory, in spite of the fact that every pull of his own keen mind was toward abandoning it. In the case of a sovereign monarch, law is what the monarch commands. But what shall we say of the "laws" of succession which tell who the "lawful" monarch is? It is of the essence of a command that it be addressed by a superior to an inferior, yet in the case of a "sovereign many," say, a parliament, the sovereign seems to command itself since a member of parliament may be convicted under a law he himself drafted and voted for. The sovereign must be unlimited in legal power, for who could adjudicate the legal bounds of a supreme lawmaking power? Yet a "sovereign many" must accept the limitation of rules before it can make law at all. Such a body can gain the power to issue commands only by acting in a "corporate capacity"; this it can do only by proceeding "agreeably to the modes and forms" established and accepted for the making of law. Judges exercise a power delegated to them by the supreme lawmaking power, and are commissioned to carry out its "direct or circuitous commands." Yet in a federal system it is the courts which must resolve conflicts of competence between the federation and its components. All of these problems Austin sees with varying degrees of explicitness, and he struggles mightily with them. Over and over again he teeters on the edge of an abandonment of the command theory in favor of what Professor Hart has described as a view that discerns the foundations of a legal order in "certain fundamental accepted rules specifying the essential lawmaking procedures." Yet he never takes the plunge. He does not take it because he had a sure insight that it would forfeit the black-andwhite distinction between law and morality that was the whole object of his Lectures - indeed, one may say, the enduring ob- 10 I AUSTIN, LECTURES ON JURISPRUDENCE I67-34I (5th ed. I885). 640 [Vol. 7
FIDELITY TO LAW 41 ject of a dedicated life For if law is made possible by "fundamen tal accepted rules'-which for Austin must be rules, not of law but of positive morality - what are we to say of the rules that the lawmaking power enacts to regulate its own lawmaking? We have election laws, laws allocating legislative representation to specific geographic areas, rules of parliamentary procedure for the qualification of voters, and many other laws and of similar nature. These do not remain fixed, and all of them shape in varying degrees the lawmaking process. Yet how are we to distinguish between those basic rules that owe their validity to acceptance, and those which are properly rules of law, valid even when men generally consider them to be evil or ill-advised? In other words, how are we to define the words < fundamental and "essential" in Professor Hart's own formulation: 'certain fundamental accepted rules specifying the essential lawmaking The solution for this problem in Kelsen's the instructive Kelsen does in fact take the plunge over which Austin hesitated too long. Kelsen realizes that before we can distinguish between phat is law and what is not, there must be an acceptance of some basic procedure by which law is made. In any legal system there must be some fundamental rule that points unambiguously te the source from which laws must come in order to be laws. Thi rule Kelsen called"the basic norm. In his own words, The basic norm is not valid because it has been created in a certain way but its validity is assumed by virtue of its content. It is valid then. like a norm of natural lay The idea of a pure positive law. like that of natural law, has its limitations. 11 It will be noted that Kelsen speaks, not as Professor Hart does offundamental rules that regulate the making of law, but of a single rule or norm. Of course, there is no such single rule in any modern society. The notion of the basic norm is admittedly a symbol, not a fact. It is a symbol that embodies the positivist quest for some clear and unambiguous test of law, for some cle sharp line that will divide the rules which owe their validity to their source and those which owe their validity to acceptance and intrinsic appeal. The difficulties Austin avoided by sticking with the command theory, Kelsen avoids by a fiction which sim plifies reality into a form that can be absorbed by positivis i KELSEN, GENERAL THEORY OF LAW AND STATE 4oI (3d ed. I949)
FIDELITY TO LAW ject of a dedicated life. For if law is made possible by "fundamental accepted rules" - which for Austin must be rules, not of law, but of positive morality - what are we to say of the rules that the lawmaking power enacts to regulate its own lawmaking? We have election laws, laws allocating legislative representation to specific geographic areas, rules of parliamentary procedure, rules for the qualification of voters, and many other laws and rules of similar nature. These do not remain fixed, and all of them shape in varying degrees the lawmaking process. Yet how are we to distinguish between those basic rules that owe their validity to acceptance, and those which are properly rules of law, valid even when men generally consider them to be evil or ill-advised? In other words, how are we to define the words "fundamental" and "essential" in Professor Hart's own formulation: "certain fundamental accepted rules specifying the essential lawmaking procedure"? The solution for this problem in Kelsen's theory is instructive. Kelsen does in fact take the plunge over which Austin hesitated too long. Kelsen realizes that before we can distinguish between what is law and what is not, there must be an acceptance of some basic procedure by which law is made. In any legal system there must be some fundamental rule that points unambiguously to the source from which laws must come in order to be laws. This rule Kelsen called "the basic norm." In his own words, The basic norm is not valid because it has been created in a certain way, but its validity is assumed by virtue of its content. It is valid, then, like a norm of natural law .... The idea of a pure positive law, like that of natural law, has its limitations.1 It will be noted that Kelsen speaks, not as Professor Hart does, of "fundamental rules" that regulate the making of law, but of a single rule or norm. Of course, there is no such single rule in any modern society. The notion of the basic norm is admittedly a symbol, not a fact. It is a symbol that embodies the positivist quest for some clear and unambiguous test of law, for some clean, sharp line that will divide the rules which owe their validity to their source and those which owe their validity to acceptance and intrinsic appeal. The difficulties Austin avoided by sticking with the command theory, Kelsen avoids by a fiction which simplifies reality into a form that can be absorbed by positivism. 11 KELSEN, GENERAL THEORY OF LAW AND STATE 40I (3d ed. I949). I958] 641
HARVARD LAW REVIEW [Vol. 7I A full exploration of all the problems that result when we recognize that law becomes possible only by virtue of rules that ire not law, would require drawing into consideration the effect of the presence or absence of a written constitution. Such a consti tution in some ways simplifies the problems I have been dis cussing, and in some ways complicates them. In so far as a writ ten constitution defines basic lawmaking procedure, it may re- move the perplexities that arise when a parliament in effect de fines itself. At the same time, a legislature operating under a writ ten constitution may enact statutes that profoundly affect the law aking procedure and its predictable outcome. If these statutes are drafted with sufficient cunning, they may remain within the frame of the constitution and yet undermine the institutions was intended to establish. If the"court-packing,proposal of the 'thirties does not illustrate this danger unequivocally, it at least suggests that the fear of it is not fanciful. No written consti- tution can be self-executing. To be effective it requires not merely the respectful deference we show for ordinary legal enact- ments, but that willing convergence of effort we give to moral principles in which we have an active belief. One may properly work to amend a constitution, but so long as it remains unamended one must work with it, not against it or around it. All this amounts to saying that to be effective a written constitution must be ac- cepted, at least provisionally, not just as law, but as good law What have these considerations to do with the ideal of fide law? I think they have a great deal to do with it, and that they reveal the essential incapacity of the positivistic view to serve that ideal effectively. For I believe that a realization of this ideal is something for which we must plan, and that is precisely what positivism refuses to do. Let me illustrate what I mean by planning for a realization of the ideal of fidelity to law. Suppose we are drafting a written constitution for a country just emerging from a period of violence and disorder in which any thread of legal continuity with previous governments has been brok Obviously such a constitution cannot lift itself unaided into legality; it cannot be law simply because it says it is. We should keep in mind that the efficacy of our work will depend upon general acceptance and that to make his acceptance secure there must be a general belief that the con- stitution itself is necessary, right, and good. The provisions of the constitution should, therefore, be kept simple and understandable
HARVARD LAW REVIEW A full exploration of all the problems that result when we recognize that law becomes possible only by virtue of rules that are not law, would require drawing into consideration the effect of the presence or absence of a written constitution. Such a constitution in some ways simplifies the problems I have been discussing, and in some ways complicates them. In so far as a written constitution defines basic lawmaking procedure, it may remove the perplexities that arise when a parliament in effect defines itself. At the same time, a legislature operating under a written constitution may enact statutes that profoundly affect the lawmaking procedure and its predictable outcome. If these statutes are drafted with sufficient cunning, they may remain within the frame of the constitution and yet undermine the institutions it was intended to establish. If the "court-packing" proposal of the 'thirties does not illustrate this danger unequivocally, it at least suggests that the fear of it is not fanciful. No written constitution can be self-executing. To be effective it requires not merely the respectful deference we show for ordinary legal enactments, but that willing convergence of effort we give to moral principles in which we have an active belief. One may properly work to amend a constitution, but so long as it remains unamended one must work with it, not against it or around it. All this amounts to saying that to be effective a written constitution must be accepted, at least provisionally, not just as law, but as good law. What have these considerations to do with the ideal of fidelity to law? I think they have a great deal to do with it, and that they reveal the essential incapacity of the positivistic view to serve that ideal effectively. For I believe that a realization of this ideal is something for which we must plan, and that is precisely what positivism refuses to do. Let me illustrate what I mean by planning for a realization of the ideal of fidelity to law. Suppose we are drafting a written constitution for a country just emerging from a period of violence and disorder in which any thread of legal continuity with previous governments has been broken. Obviously such a constitution cannot lift itself unaided into legality; it cannot be law simply because it says it is. We should keep in mind that the efficacy of our work will depend upon general acceptance and that to make this acceptance secure there must be a general belief that the constitution itself is necessary, right, and good. The provisions of the constitution should, therefore, be kept simple and understandable, 642 [Vol. 7
r958] FIDELITY TO LAW 64 not only in language, but also in purpose. Preambles and other explanations of what is being sought, which would be objec tionable in an ordinary statute, may find an appropriate place in our constitution. We should think of our constitution as es tablishing a basic procedural framework for future governmental ction in the enactment and administration of laws. Substantive limitations on the power of government should be kept to a mini mum and should generally be confined to those for which a need can be generally appreciated. In so far as possible, substantive aims should be achieved procedurally, on the principle that if men are compelled to act in the right way, they will generally do the right things These considerations seem to have been widely ignored in the constitutions that have come into existence since World War II Not uncommonly these constitutions incorporate a host of eco- nomic and political measures of the type one would ordinarily associate with statutory law. It is hardly likely that these meas ures have been written into the constitution because they represent aims that are generally shared. One suspects that the reason fc their inclusion is precisely the opposite, namely, a fear that they would not be able to survive the vicissitudes of an ordinary exer cise of parliamentary power. Thus, the divisions of opinion that are a normal accompaniment of lawmaking are written into the document that makes law itself possible. This is obviously a procedure that contains serious dangers for a future realization of the ideal of fidelity to law. I have ventured these remarks on the making of constitutions not because I think they can claim any special profundity, but because i wished to illustrate what I mean by planning the con ditions that will make it possible to realize the ideal of fidelity to law. Even within the limits of my modest purpose, what I have said may be clearly wrong. If so, it would not be for me to say whether I am also wrong clearly. I will, however, venture to assert that if I am wrong, I am wrong significantly. What disturbs e about the school of legal positivism is that it not only refuses to deal with problems of the sort i have just discussed, but bans them on principle from the province of legal philosophy In its concern to assign the right labels to the things men do, this school seems to lose all interest in asking whether men are doing the right things
FIDELITY TO LAW not only in language, but also in purpose. Preambles and other explanations of what is being sought, which would be objectionable in an ordinary statute, may find an appropriate place in our constitution. We should think of our constitution as establishing a basic procedural framework for future governmental action in the enactment and administration of laws. Substantive limitations on the power of government should be kept to a minimum and should generally be confined to those for which a need can be generally appreciated. In so far as possible, substantive aims should be achieved procedurally, on the principle that if men are compelled to act in the right way, they will generally do the right things. These considerations seem to have been widely ignored in the constitutions that have come into existence since World War II. Not uncommonly these constitutions incorporate a host of economic and political measures of the type one would ordinarily associate with statutory law. It is hardly likely that these measures have been written into the constitution because they represent aims that are generally shared. One suspects that the reason for their inclusion is precisely the opposite, namely, a fear that they would not be able to survive the vicissitudes of an ordinary exercise of parliamentary power. Thus, the divisions of opinion that are a normal accompaniment of lawmaking are written into the document that makes law itself possible. This is obviously a procedure that contains serious dangers for a future realization of the ideal of fidelity to law. I have ventured these remarks on the making of constitutions not because I think they can claim any special profundity, but because I wished to illustrate what I mean by planning the conditions that will make it possible to realize the ideal of fidelity to law. Even within the limits of my modest purpose, what I have said may be clearly wrong. If so, it would not be for me to say whether I am also wrong clearly. I will, however, venture to assert that if I am wrong, I am wrong significantly. What disturbs me about the school of legal positivism is that it not only refuses to deal with problems of the sort I have just discussed, but bans them on principle from the province of legal philosophy. In its concern to assign the right labels to the things men do, this school seems to lose all interest in asking whether men are doing the right things. I958] 643