Hart, Radbruch and the Necessary Connection Between Law and Morals

Abstract

Legal positivism maintains a distinction between law as it is and law as it ought to be. In other words, for positivists, a law can be legally valid even if it is immoral. H. L. A. Hart hoped to defend legal positivism against natural law. This paper analyses Hart’s criticism of Gustav Radbruch, a natural lawyer, before suggesting that Hart’s account of legal positivism gives rise to a logical problem. It is concluded that this problem leaves logical space for a theory of natural law based on moral authority rather than legal validity.

Introduction

Legal positivism maintains a distinction between law as it is and law as it ought to be.¹ In other words, for positivists, a law can be legally valid even if it is immoral. H. L. A. Hart hoped to defend this distinction.² In attempting to do so, he was critical of Gustav Radbruch, who claimed that the laws of the Nazis were legally invalid,³ because they were morally evil.⁴

The purpose of this paper is to analyse Hart’s criticism of Radbruch, and to contrast the two scholars’ approaches to morally evil laws. In attempting to achieve this purpose, the body of this paper has two parts. First, Hart’s criticism of Radbruch is considered. Second, it is argued that Hart’s approach contains a logical problem that may have relevance for the debate between positivists and natural lawyers. It is concluded that this problem leaves logical space for a theory of natural law based on moral authority rather than legal validity.

Argument

Hart’s Criticism of Radbruch

Hart’s purpose was to defend the distinction between law as it is and law as it ought to be.⁵ In attempting to achieve his purpose, he considered a ‘passionate appeal’ based on ‘a terrible experience’.⁶ The appeal was Radbruch’s, and the experience was of Germans under the laws of the Nazis.

Prior to the Nazis coming to power, Radbruch had himself been a positivist.⁷ Following his conversion, however, he developed the following formula:

Preference is given to the positive law, duly enacted and secured by state power as it is, even when it is unjust and fails to benefit the people, unless its conflict with justice reaches so intolerable a level that the statute becomes, in effect, ‘false law’ and must therefore yield to justice. It is impossible to draw a sharper line between cases of statutory non-law and statutes that are valid despite their flaws. One line of distinction, however, can be drawn with utmost clarity: Where there is not even an attempt at justice, where equality, the core of justice, is deliberately betrayed in the issuance of positive law, then the statute is not merely ‘false law’, it lacks completely the very nature of law. For law, including positive law, cannot be otherwise defined than as a system and an institution whose very meaning is to serve justice.⁸

This formula has two aspects.⁹ The first concerns an intolerable departure, and the second concerns a deliberate betrayal.¹⁰ The post war German courts held that it was sufficient to apply the first aspect of the formula in deciding whether a law was legally valid.¹¹ Perhaps this was because the judges had more difficulty determining others’ deliberate betrayal than their own level of intolerance. Alternatively, they may have thought that justice was objectively measurable, such that a radical departure from it could be said to be intolerable.¹²

However that may be, for Radbruch, the distinction between law and morals, typified in the phrase ‘Gesetz als Gesetz’, led to quietism regarding morally evil laws, both amongst the general population and the legal profession.¹³ For this reason, Radbruch decided that fundamental moral principles were a part of the nature of law, and that no law, no matter how well it conformed to formal criteria for legal validity, could be law if it was against such principles.¹⁴

Hart, on the other hand, said that there was ‘extraordinary naïveté’ in the belief that positivism had made the Germans obsequious in the face of such laws.¹⁵ Indeed, according to Hart, the problem lay not with positivism, but with the Germans.¹⁶ After all, said Hart, positivism coincided with ‘the most enlightened liberal attitudes’ in England.¹⁷ Further, according to Hart, Radbruch’s rejection of positivism showed that he had ‘only half digested the spiritual message of liberalism’.¹⁸ After all, said Hart, just because a law is legally valid does not mean that it ought to be obeyed.¹⁹ Further, he said, if a law can be disobeyed, then there is nothing to be gained by refusing to regard it as legally valid.²⁰

Hart’s comments about Radbruch, in particular, and Germans, in general, are argumenta ad hominem. Nevertheless, there is force in the idea that a legally valid law need not be obeyed, and that if a law can be disobeyed, then there is nothing to be gained by refusing to regard it as legally valid. In this connection, it will be apposite to consider the case of the ‘grudge informer’.

The case involved a woman who, under the Nazis’ anti-sedition laws, in order to get rid of her husband, had denounced him for speaking ill of Hitler.²¹ Her husband was arrested and sentenced to death, although his sentence was commuted, and he was sent to the front instead.²² The woman argued that she was innocent, because she had acted in accordance with validly enacted laws.²³ However, according to Hart, the court found that the laws had been legally invalid, because they had been ‘contrary to the sound conscience and sense of justice of all decent human beings’.²⁴ As a consequence, the woman’s defence was rejected, and she was convicted of rechtswidrige Freiheitsberaubung (‘illegally depriving a person of his freedom’) under the Penal Code of 1871.²⁵

The case is an example of how Radbruch’s formula was applied by German courts, in cases where war criminals, spies and informers claimed to have been acting lawfully, because they had been acting under legally valid laws.²⁶ Such cases met with almost universal approval.²⁷ For Hart, however, such approval was hysterical.²⁸

As Hart observed, there were at least two other options available to the courts.²⁹ First, accused persons could have gone unpunished.³⁰ Second, they could have been punished under retrospective legislation.³¹ Hart considered the second approach to be less offensive than the first, and certainly less offensive than Radbruch’s approach.³² For Hart, punishing people under retrospective legislation would have made it clear that a choice was being made between two evils.³³ That is, between leaving criminals unpunished, on the one hand, and overriding valid law with retrospective legislation, on the other hand.³⁴ This approach, according to Hart, would have had the merit of an honest declaration that a moral principle had been sacrificed.³⁵ In other words, the ‘very precious principle of morality’ that ‘retrospective criminal legislation and punishment’ ought not to occur.³⁶

Thus, for Hart, the application of Radbruch’s formula obscured the moral choice being made.³⁷ However, in characterizing the choice as one between leaving criminals unpunished, on the one hand, and overriding valid law with retrospective legislation, on the other hand, Hart was begging the question against Radbruch. After all, whether that was the choice was precisely the point at issue. If, for example, the choice was between not punishing criminals and upholding moral principles more generally, then the application of Radbruch’s formula obscured nothing.

Curiously, Hart also criticised Radbruch’s formula on the basis that it rests on a ‘disputable philosophy’.³⁸ This is curious, because positivism is also a disputable philosophy. Hence the philosophical dispute in which Hart was participating. In any event, the point is not whether Radbruch’s philosophy is disputable, but whether it is true. Or, for redundancy theorists of truth, whether it is persuasive.

That said, as indicated earlier, there is force in Hart’s suggestion that nothing is gained in taking morally evil laws to be legally invalid, since not all valid laws need to be obeyed. In the next section, however, it will be shown that Hart’s approach gives rise to a logical problem. The same reasoning, though, will suggest a different approach to that offered by Radbruch.

Hart’s Logical Problem

In arguing against Radbruch, Hart’s purpose was to defend the distinction between law as it is and law as it ought to be.³⁹ Unfortunately, he conflated this distinction with the ostensibly similar, but logically incompatible, proposition that there is no necessary connection between law and morals.⁴⁰

From the distinction between law as it is and law as it ought to be, one may infer that there is something the law ought to be. By way of explanation, if there is something the law is on the one hand, and something the law ought to be on the other hand, such that one can be distinct from the other, it follows that there is something the law ought to be. This is because such relations depend for their existence on the existence of their relata.⁴¹ In the present connection, this means that the relation of being distinct from, that Hart held existed between law as it is and law as it ought to be,⁴² relies for its existence on the existence of both law as it is and law as it ought to be. It then follows, by the logical rule of simplification,⁴³ that there exists something the law ought to be.

Of course, someone might claim that the distinction between law as it is and law as it ought to be does not imply the existence of something the law ought to be any more than the alleged distinction between horses and unicorns implies the existence of unicorns. In reply to such an objection, one may say that if the existence of such a relation depends on the existence of its relata, it follows that if unicorns do not exist, then no relation between horses and unicorns exists. The claim that horses are distinct from unicorns would be meaningless, and anyone who thought otherwise would be like the person, imagined by Fraser MacBride, who thought that Canberra was north of Lobart.⁴⁴

Alternatively, someone could maintain that the distinction between law as it is and law as it ought to be is similarly meaningless, because there is nothing the law ought to be. However, this strategy would render meaningless the central idea behind legal positivism. Another option would be to adopt the position of philosophers who argue that non-existent things can have being, ⁴⁵ and that the existence of a relation depends upon the being, and not the existence, of its relata.⁴⁶ In which case, the distinction between law as it is and law as it ought to be would entail the being, but not the existence, of something the law ought to be.

While this strategy would prevent one from inferring that there exists something the law ought to be from Hart’s distinction, it would still allow one to infer that there is something the law ought to be, where the words ‘there is’ are held to connote being, but not existence.⁴⁷ If it is not the case that there exists something the law ought to be, then this would seem to entail the truth of moral anti-realism. Regardless, it is consistent with antirealism that moral norms about the law exist, for instance, as the products of intersubjective agreement through rational discourse.⁴⁸ However that may be, the words ‘there is’ will be used here to connote being, but not necessarily existence, in order not to beg any metaphysical questions. It is in this sense that Hart’s distinction will be said to entail that there is something the law ought to be.

It also follows that there is something the law ought to be from Hart’s view that the distinction between law and morals allows that while laws ought generally to be obeyed, some laws may be ‘too evil to be obeyed’.⁴⁹ If a law ought to be obeyed, then either it is as it ought to be, or it is not so far removed from what it ought to be that it ought not to be obeyed. Conversely, if a law ought not to be obeyed, then it is not as it ought to be. Thus, regardless of whether a law ought to be obeyed, or ought not to be obeyed, there is something it ought to be.

Of course, someone might reject Hart’s assertion that the distinction between law as it is and law as it ought to be does not imply that all laws ought to be obeyed. However, this would mean accepting that the distinction does imply that all laws ought to be obeyed, which would be difficult to argue.

It seems reasonable to suppose, therefore, that if there is something a law is, then there is something it ought to be. A statement of this kind is called a ‘conditional’.⁵⁰ Conditionals have antecedents and consequents.⁵¹ In the conditional under consideration the antecedent is ‘there is something a law is’ and the consequent is ‘there is something it ought to be’. Antecedents are sufficient conditions for consequents being true, and consequents are necessary conditions for antecedents being true.⁵² For example, in the conditional ‘if Wadsworth is a horse, then Wadsworth is a mammal’, being a horse is a sufficient condition for being a mammal, and being a mammal is a necessary condition for being a horse. Since in the conditional ‘if there is something a law is, then there is something it ought to be’ the consequent is the necessary condition of the antecedent, it follows that a necessary condition for being a law is that there is something it ought to be. Thus, pace Hart, there is a necessary connection between law and morals.

It may be objected that Hart did not conflate two distinct, logically incompatible, propositions, because the propositions are equivalent. Indeed, he seems to suggest that the proposition that ‘there is a distinction between law as it is and law as it ought to be’ is equivalent to the proposition that ‘there is no necessary connection between law and morals or law as it is and ought to be’.⁵³ Indeed, Benjamin C. Zipursky has observed that ‘Hart’s casual flip from one [proposition] to the other indicates that he treated them as equivalent’.⁵⁴ Further, Hart’s positivism is often identified with the thesis that there is no necessary connection between law and morals.⁵⁵ The two propositions, however, are clearly different. To give an analogy, the proposition that there is a distinction between being a horse and being a mammal – since, for instance, horses are only a subset of mammals – is not equivalent to the proposition that there is no necessary connection between being a horse and being a mammal, because being a mammal is a necessary condition for being a horse.

Alternatively, it may be objected, as John Gardner has,⁵⁶ that Hart’s apparent support of the proposition that there is no necessary connection between law and morals was merely a ‘bungled’ attempt to express ‘(LP*)’,⁵⁷ which is the proposition that:

In any legal system, whether a given norm is legally valid, and hence whether it forms part of the law of that system, depends on its sources, not its merits (where its merits, in the relevant sense, include the merits of its sources).⁵⁸

Gardner claims to have distilled this proposition not only from the writings of Hart, but also from those of Bentham and Austin. In addition, Gardner claims that it is a myth that legal positivists support the proposition that there is no necessary connection between law and morals.⁵⁹ More precisely, according to Gardner, the proposition:

[I]s absurd and no legal philosopher of note has ever endorsed it as it stands. After all, there is a necessary connection between law and morality if law and morality are necessarily alike in any way.⁶⁰

In reply to Gardner’s assertion that Hart merely ‘bungled’ an attempt to express (LP*), one may simply note, for example, that in The Concept of Law Hart characterised natural law as the view that there is a necessary connection between law and morals,⁶¹ and then spent much of that book, including an entire chapter,⁶² arguing against the proposition that there is any relevant necessary connection between law and morals.⁶³ It is hard to accept that this was merely a ‘bungled’ attempt to express (LP*), and that Hart was not at least somewhat committed to the proposition that there is no necessary connection between law and morals.

Moreover, Gardner’s assertion that ‘there is a necessary connection between law and morality if law and morality are necessarily alike in any way’ would empty the concept of necessary connection of any meaningful content. It would mean, for instance, that there is a necessary connection between any two propositions, however absurdly unrelated, simply because all propositions are necessarily alike in having the necessary property of being a proposition. For example, there would be a necessary connection between the proposition that Socrates is a philosopher and the proposition that unicorns eat naught but rainbow-coloured fairy floss. Thus, Gardner’s characterisation of a necessary connection must be rejected, and philosophers should not regard the proposition that there is no necessary connection between law and morals as absurd simply because of this characterisation. Further, unlike Gardner’s claim that law and morals ‘are necessarily alike in both necessarily comprising some valid norms’,⁶⁴ the necessary connection presently being considered is not trivial, and would seem to provide a good reason to reject legal positivism.

Another possible objection is that Hart took the proposition that there is no necessary connection between law and morals to be equivalent to the proposition that ‘it is in no sense a necessary truth that laws reproduce or satisfy certain demands of morality’,⁶⁵ and that no one has shown the latter proposition to be false. The obvious response would be to agree that it is clearly not a necessary truth that laws are moral, and any suggestion that natural lawyers would argue otherwise is a straw man. After all, if Leibniz was right, a proposition is necessarily true if and only if it is true in all possible worlds, and there are clearly immoral laws in the actual world.⁶⁶ Thus, for instance, while the proposition that ‘bachelors are unmarried’ may be an a priori necessary truth,⁶⁷ and the proposition that ‘gold is the element with atomic number 79’ may be an a posteriori necessary truth,⁶⁸ the proposition that ‘laws are moral’ is not a necessary truth in either sense.

Be that as it may, it may be a necessary truth that laws ought to be moral, in the limited sense that there is no possible world in which there are rational beings who have laws to govern them, but in which it is not the case that laws ought to be moral. Hart, of course, argued that legal and moral prohibitions against murder, for example, were only contingently true, because there are possible worlds in which human beings have physically evolved to be invulnerable to murder.⁶⁹ But even in such a possible world, it is difficult to imagine that it is not the case that laws ought to be moral, since even such beings might be susceptible to harm from which the law ought to protect them. For instance, such beings might be psychologically harmed by the unjust deprivation of their liberty, or a lack of social order. Thus, the proposition that some laws might be unnecessary in some possible worlds does not negate the proposition that such laws that do exist ought to be moral.

In the context of this paper, however, it is important to note that Hart argued that morality is in no way a relevant necessary condition for legal validity,⁷⁰ as if this were also equivalent to the proposition that there is no necessary connection between law and morals. To be clear, although it is being argued that a necessary condition for a rule being legally valid is that there is something it ought to be, it is not being argued that a law is legally invalid if it is not as it ought to be. This is because it does not follow from the proposition that a necessary condition for being a law is that there is a way it ought to be that a necessary condition for being a law is that it is in fact the way it ought to be. All that is being claimed is that it follows from Hart’s philosophy that a necessary condition for being a law is that there is something it ought to be, and that this is a relevant kind of necessary connection in deciding whether a law is morally authoritative.

Thus, the objection that a law that governs the behaviour of motorists, for example, ‘does not even have to be morally permissible’,⁷¹ is counterintuitive if one is talking about the moral authority rather than the legal validity of the law. It would be even more counterintuitive if rather than requiring motorists to drive on a certain side of the road, for instance, it required them to drive in the middle of the road. Moreover, if a law required motorists to drive on a particular side of the road, then it would not be ‘too evil to be obeyed’, and thus, as Hart would have it,⁷² would be exactly as it ought to be. At this point, it should be clear that there may be more than one way for a law to be as it ought to be.

It might also be objected that the claim that there is something the law ought to be entails the truth of moral realism, which is the metaphysical thesis that morality is grounded in the nature of things, rather than in intersubjective agreement.⁷³ Implicit in this objection is the view that a philosophical commitment to moral realism is undesirable, because, for example, as Habermas would have it, it is contrary to post-metaphysical thinking.⁷⁴ In responding to such an objection, one may make the following points. First, and most importantly, the argument that the position presented here is false, because it entails the truth of moral realism, is merely argumentum ad baculum, and, as such, may be dismissed on this ground alone. Second, as discussed above, the claim that there is something the law ought to be is consistent with moral anti-realism. If, of course, moral realism were true, which one may argue is plausible,⁷⁵ then the proposition that there is something the law ought to be might carry more weight, but the proposition would not entail realism about morals. In any event, moral realism is consistent with legal positivism,⁷⁶ since moral facts, in and of themselves, cannot render positive laws legally invalid.⁷⁷

Another possible objection is that whether a law is morally authoritative is settled, at least in part, by whether it is legally valid, since legal validity is itself a morally important consideration. However, the proposition that a necessary condition for a rule being legally valid is that there is something it ought to be seems to put the question of moral authority prior to the question of legal validity. In replying to this possible objection, it should be remarked that although it is being argued that a necessary condition for a rule being legally valid is that there is something it ought to be, it is not being argued that a necessary condition for a rule being legally valid is that it is as it ought to be, or, in deciding whether a law is morally authoritative, that one cannot take legal validity into account.

Finally, it may be objected that the ‘ought’ in the proposition that there is a distinction between law as it is and law as it ought to be is a legal ‘ought’, and not a moral ‘ought’. To be sure, Hart did argue for the existence of a legal ‘ought’.⁷⁸ In this case, however, it would mean that the distinction between law and morals would vanish from the proposition that there is a distinction between law as it is and law as it ought to be, which, of course, would undermine the central claim of positivism. Alternatively, it could be objected that there is no relevant necessary connection between law and morals. After all, to say that a necessary condition for being a law is that there is something it ought to be is not to say that a necessary condition for being a law is that it is the way it ought to be. Thus, a law may be legally valid even if it is not as it ought to be. This, of course, has already been granted. In any event, the argument here is concerned with moral authority and not legal validity.

The analysis above suggests that the distinction between law as it is and law as it ought to be is consistent with a theory of natural law that relies on moral authority rather than legal validity. So long as, of course, natural law can be seen as the view that the law ought to be moral. After all, if the law ought to be moral, then one may have good reason not to accept morally evil laws as morally authoritative. That such a position is consistent with the distinction between law as it is and law as it ought to be is a problem for Hart’s approach. At best, it suggests a dialectical stalemate between Hart and natural lawyers, who may, like John Finnis, hold that particularly unjust laws may lack moral authority, and therefore may not be binding on citizens, qua moral agents, despite being legally valid.⁷⁹ The purpose here is not to defend so called ‘weak’ theories of natural law, such as those proposed by Finnis, but only to suggest that they may be persuasive insofar as they take morally evil laws to be legally valid, but not morally authoritative.

So long as legal positivism rejects the proposition that there is no necessary connection between law and morals, as suggested by contemporary positivists, such as Gardner,⁸⁰ then it may seem no less persuasive than weak natural law. After all, Hart’s positivism allows that morally evil laws need not be obeyed,⁸¹ and that moral considerations may be ‘included within the grounds for establishing the existence and content of valid, positive laws’.⁸² Thus, Hart’s theory has been classified as a variety of inclusive legal positivism.⁸³ An inclusive legal positivist could accept that certain necessary connections obtain between law and morals – for example, that they are necessarily alike in some way – without surrendering positivism altogether. However, the claim that a logically necessary condition for a rule being legally valid is that there is something it ought morally to be – i.e., that morality is in an important sense a conditio sine qua non of legal validity – is not obviously reconcilable with legal positivism. Hence, the relevance of the present argument to the contemporary debate between positivists and natural lawyers.

In considering Hart’s logical problem, a different approach suggests itself. More precisely, one that would navigate between the approaches of Hart and Radbruch. It would accept the distinction between law as it is and law as it ought to be, and, therefore, pace Radbruch, that a morally evil law can be legally valid. However, it would rest on the understanding that, pace Hart, a necessary condition for a rule being legally valid is that there is something it ought to be.

Summary

The purpose of this paper has been to analyse Hart’s criticism of Radbruch, and to contrast the two scholars’ approaches to morally evil laws. In attempting to achieve this purpose, the body of this paper has had two parts. First, Hart’s criticism of Radbruch was considered. Second, and more importantly, it was argued that Hart’s approach contains a logical problem that may have relevance for the debate between positivists and natural lawyers.

Conclusion

In arguing against Radbruch, Hart conflated the distinction between law as it is and law as it ought to be with the proposition that there is no necessary connection between law and morals. This is important, because from the distinction between law as it is and law as it ought to be, as well as from the proposition that the distinction allows that laws ought generally to be obeyed, but that some laws may be too evil to be obeyed, one may infer the negation of the proposition that there is no necessary connection between law and morals. More precisely, one may infer from Hart’s writing that a necessary condition for being a law is that there is something it ought to be, and this is consistent with a theory of natural law that relies on moral authority rather than legal validity.