The Question To Be Faced Is One of Fact: H.L.A. Hart’s Legal Theory Through His View of International Law

H.L.A. Hart says that The Concept of Law is focused on municipal or domestic law because that is the “central case” ¹ for the usage of the word ‘law.’ At the beginning of the book he states that “at various points in this book the reader will find discussions of the borderline cases where legal theorists have felt doubts about the application of the expression ‘law’ or ‘legal system,’ but the suggested resolution of these doubts, which he will also find here, is only a secondary concern of the book.” ² Yet among those borderline cases there is one that is rather intriguing, since Hart closely discusses a particular instance of them: it is international law, to which he devotes an entire chapter—the final one—of The Concept of Law. My goal in this article is therefore to make clear why the ‘resolution’ of the borderline case of international law is not entirely ‘secondary’ to Hart’s overall project in The Concept of Law and, in so doing, to show that Chapter X is not as unhappy as many think it is.

Indeed, the literature on Hart’s conception of international law is mostly negative (section I). But if we abstract from the content of this conception and take a methodological stance looking at how Hart treats the topic (section II), Chapter X appears in a different light: not as an ill-conceived attempt to articulate a general account of international law, but rather as a piece of conceptual analysis applied to a non-central case of law. Thus, as we will see (section III), Chapter X offers an alternative viewpoint on the way Hart practises conceptual analysis—a viewpoint other than the one normally focused on paradigmatic cases of ‘law.’ Through this perspective it is possible to better assess Hart’s account of international law with all its strengths and shortcomings.

I. The Negative View of Scholars

International legal scholarship has not paid much attention to Hart’s discussion of international law (IL), while legal philosophy is quite critical of the Hartian conception of IL, regarding it as one of the least successful parts of his thought. This is not surprising with respect to international legal scholars. Indeed, Hart did not seem to be much concerned with IL. His writings on this topic are few ³ and even the Postscript, which re-examines most of the principles and arguments of The Concept of Law, does not mention IL at all. In addition, in Hart’s view municipal law and IL differ so much that they cannot be equated to each other. Compared with municipal law, IL is so simple and elementary that, for Hart, one may even come to question its legal status—indeed, the whole of Chapter X revolves around that very question. But should one doubt whether IL is really law, it is understandable to find Hart quite unhelpful within IL research. ⁴ Thus, it is hard to engage in discussion with an international lawyer when we have an unsophisticated account of IL like Hart’s. ⁵

What is more surprising is the position of legal philosophers, particularly those influenced by Hart. First, they seem to have become interested in Hart’s conception of IL only a short time ago. For example, Neil MacCormick, in his seminal work on Hart’s thought, covers the subject only on the surface. ⁶ According to Jeremy Waldron, “legal philosophers in Hart’s tradition mostly neglect the subject of international law altogether,” probably owing to the weak significance of IL for the founder of that tradition. ⁷ More importantly, their opinions of Hart’s account of IL are rather unfavourable. Basically, there are two criticisms made of Hart. The first might be labelled ‘external’ because it highlights a gap found between the reality of IL and the Hartian account of it; the second is more ‘internal’ and refers to the adequacy of Hart’s concepts in relation to IL. ⁸

Hart asserts that IL is different from municipal law because it lacks secondary norms. Critics, however, believe that this assertion is false. ⁹ Today IL is too developed to sustain a contention like that, ¹⁰ but what is worse, is that it was an excess even in the fifties to hold that IL does not possess any sort of rules of recognition, rules of change, or rules of adjudication. ¹¹ To be sure, no one argues that Hart shares Austin’s view with respect to the legal status of IL ¹² and, admittedly, no one equates secondary rules in the international realm to those in the municipal realm. ¹³ But—and here is the second criticism—the problem with Hart is that he nonetheless looks at IL through the eyes of municipal law. He does not see secondary rules in the realm of IL because these rules are part of a notion of a legal system, which is narrowly tailored to state law.

Accordingly, the concept of a secondary rule is utterly state-centred, too biased towards the role law plays in the municipal realm, as if one may speak of legislators’ and judges’ power only in relation to municipal law. ¹⁴ Thus—the criticism goes—to admit the legal status of IL while depicting it as “a set of separate rules” ¹⁵ without any unifying feature ends up obscuring the reality of IL. It would be better to devise a broader notion of legal system, one able to encompass the ‘secondarity’ of power-conferring rules in IL. ¹⁶

No doubt, as we will see in section III below, those opinions have good reasons on their side, but they leave unanswered a basic question: Why did Hart have such an impoverished view of IL? ¹⁷ Indeed, the critics give the impression that the best choice would have been to exclude Chapter X from The Concept of Law altogether. If so, then while The Concept of Law would have been less general than Hart claimed, it would have gained much more in terms of explanatory power. In that way, Chapter X turns out to be a sort of foreign body when compared to the theoretical mastery of the rest of the book. And that impression gets stronger when some of them declare that international relations in the fifties did not justify a conception of IL like Hart’s.

II. Hart’s Rationale for Chapter X

However, the critical conclusion just expressed may depend on a too narrow perspective on Hart as an IL scholar—one too focused on the content of his conception of IL. It may, therefore, be helpful to reconsider this criticism by adopting a methodological point of view—as we can see if the rationale for Chapter X is brought to light. By making clear the reasons behind Chapter X we may discover that the apparent difference between IL and municipal law is an intentional choice by Hart, coming from reasons internal to his theory of law and revolving around the way in which he understands conceptual analysis. This method provides deep insights into state law which nonetheless do not entirely fit IL. On this approach, Chapter X basically aims to apply the same method at the international level and to reveal the specificity of IL despite the fact that both IL and state law are linguistically ‘legal.’

Indeed, Chapter X does not flesh out a broad conception of IL but instead seeks to demonstrate essentially one single thesis: that IL is really law, even if it is not a legal system, i.e., even if it has no secondary rules at all. Hart himself makes explicit the reasons for this choice at the beginning of Chapter X. Previous chapters show that the concept of a union of primary and secondary rules plays a central role to understand what law is, but Hart expressly warns against the use of this concept as a definition of law. His intention is not to describe (or even dictate) the rules governing the proper use of word ‘law.’ He is not only sceptical about the definition in the case of law, ¹⁸ but his primary aim is to elucidate the concept of it, i.e., to explain the linguistic usage without accepting it uncritically and by making clear the substantive reasons supporting that usage. ¹⁹

A case examined in the book where Hart finds such a method useful is the problem of unjust laws. ²⁰ Ordinary language holds them as laws despite their immoral content, but according to Hart it would be simplistic to do so, because the theoretical and practical advantages of conforming to ordinary language would be ignored. ²¹ Consequently, an in-depth analysis aimed at elucidating the concept finds good reasons for the current linguistic usage along with a broader notion of law, encompassing morally iniquitous laws. ²²

While IL is a case where this approach is equally helpful, it raises the opposite problem. “It is consistent with the usage of the last 150 years to use the expression ‘law’ here,” ²³ but, again, Hart does not intend to follow it uncritically. “As in the German case, we shall ask whether the common wider usage that speaks of ‘international law’ is likely to obstruct any practical or theoretical aim” ²⁴ and, in so doing, Hart finds serious substantive doubts against that common wider usage supporting the legality of IL—doubts bringing it closer to “that simple form of social structure, consisting only of primary rules of obligation, which, when we find it among societies of individuals, we are accustomed to contrast with a developed legal system.” ²⁵

Now, the crucial problem is: What is the origin of those doubts? How can they call into question a linguistic convention that takes the legal nature of IL for granted? A reasonable response might draw on John Austin’s theory of law, but it would not be convincing. Austin’s arguments against the legality of IL are essentially twofold: first, that IL conflicts with a necessary feature of statehood, namely the sovereignty; and second, that IL lacks enforcement institutions, so those who deviate from a rule are unlikely to suffer for it. ²⁶ But Hart notices that both arguments are nothing more than corollaries of arguments found faulty before in the book: “We have seen in the early chapters of this book how bad a guide this seductive notion [of sovereignty] is to the structure of a municipal legal system.” ²⁷ And the lack of organized sanctions “is tacitly to accept the analysis of obligation contained in the theory that law is essentially a matter of orders backed by threats. This theory, as we have seen, identifies ‘having an obligation’ or ‘being bound’ with ‘likely to suffer the sanction or punishment threatened for disobedience.’” ²⁸ So why write a whole chapter—as a finale—when it would have been enough merely to reshape old arguments?

In fact, one person who casts doubt on the legality of IL is Hart himself; and he points at his own idea of law as the union of primary and secondary rules as the main source of those doubts.

In Chapter IX Hart states that “reflection on some very obvious generalizations—indeed truisms—concerning human nature and the world in which men live, show that as long as these hold good, there are certain rules of conduct which any social organization must contain if it is to be viable.” ²⁹ It is the first step to the well-known “minimum content of Natural Law” claim, ³⁰ which comprises (primary) rules restricting the use of violence to human beings or things and (secondary) rules organizing sanctions, so that “those who would voluntarily obey shall not be sacrificed to those who would not.” ³¹ In Chapter X, Hart makes the following statement:

The sceptic may point out that there are in a municipal system, as we have ourselves stressed, certain provisions which are justifiably called necessary; among these are primary rules of obligation, prohibiting the free use of violence, and rules providing for the official use of force as a sanction for these and other rules. If such rules and organized sanctions supporting them are in this sense necessary for municipal law, are they not equally so for international law? ³²

That is the plausible reason why Hart discusses the case of IL and merely to demonstrate that it is law even without secondary rules—a reason that says a lot about Hart’s method.

First, it proves the importance of the ‘minimum content of Natural Law’ claim in Hart’s ‘toolbox.’ The borderline nature of IL is not conveyed by language, which regards IL as law simpliciter. Hart could stop here and rely on this linguistic convention, but this would mean repudiating the type of analysis he espouses. That analysis demonstrates that law cannot have just any content whatsoever and IL is a case showing that, if we aim to grasp law conceptually, “minimum content” matters much more than language because it brings into light some substantive doubts on the legal status of IL that language conceals.

Secondly, such a reason is completely internal to Hart’s theory. In his opinion, theorising about law in general is not just a matter of linguistic analysis or empirical inquiry, but it is necessary to take into account a “third category of statements: those the truth of which is contingent on human beings and the world they live in retaining the salient characteristics which they have.” ³³

This “third category of statements” reveals that legal practice would be nonsensical if it did not have a coercive system. This is a conclusion which elucidates state law well, but proves to be problematic for that practice which, even if it is not a legal system, is linguistically labelled as ‘international law.’ Thus, Chapter X represents the effort to keep that argument from turning to conclusions that are not only opposed to linguistic conventions about IL, ³⁴ but that Hart himself is not willing to admit substantively. And unsurprisingly Hart prevents this use not from a linguistic point of view, but by re-shaping the ‘minimum content of Natural Law’ claim, so to say, ad usum juris gentium.

In the international domain one of the truisms, viz., approximate equality, is different from that of the minimum content related to municipal law. There the use of violence is not restricted to individuals, but takes place between complex organizations that are states. This fact implies not only that such use must be public (contrary to the violence between individuals, which can be also a private matter) but also that states can be invariably unequal in strength. ³⁵ Accordingly, a further fundamental truism of the minimum content concerning municipal law is subject to change: the “limited understanding and strength of will.” ³⁶ Usually “all are tempted at times to prefer their own immediate interests and, in the absence of a special organization for their detection and punishment, many would succumb to the temptation.” ³⁷ However, due to the lack of approximate equality in the international context, a coercive system is by no means a necessity. On the contrary: threatened force not only may have no deterrence against very powerful states, but may be hazardous even for a strong state, because it may prove to be ineffective or it may even have adverse effects. ³⁸

This account of the rationale for Chapter X may also explain why Hart discusses a doubt which is, again, substantive, and not linguistic in character: the alleged moral nature of IL. ³⁹ As before, even in this case the source of doubts appears to be outside The Concept of Law, ⁴⁰ but actually comes from some contentions previously advanced in the book concerning (state) law and morality that might be used to classify IL rules as moral ones. And, as before, even in this case the solution proposed by Hart is not so much linguistic, but rather based on an attitude open to the reality of IL as it is. ⁴¹

Since, for Hart, the key to the science of jurisprudence ⁴² lies in the union of primary and secondary rules, he seems to suggest throughout The Concept of Law that a society without rules of recognition, change and adjudication has an undefined nature, swinging between morality, pre-legality, and a mixed condition. ⁴³ As a result, in Chapter X he has to demonstrate that even a set of primary rules, such as those of IL, can be entirely legal. When there are no secondary rules, a rule can be said to exist only through that practice called ‘rule-acceptance’—a practice that is frequently moral in character. ⁴⁴ So Hart points out that, with respect to IL, that practice is mainly legal instead. It can be motivated “by calculations of long-term interest, or by the wish to continue a tradition or by disinterested concern for others;” ⁴⁵ regarding the pressure for conformity “often no mention is made of moral right or wrong, good or bad;” ⁴⁶ rules of IL “are often morally quite indifferent,” ⁴⁷ while moral rules cannot but be important. Finally, since IL and morality cannot be modified through a rule of change, the “immunity from deliberate change” ⁴⁸ proper to morality is (and must be) necessary, while in IL “the lack of a legislature is just a lack which many think of as a defect one day to be repaired.” ⁴⁹

III. Hart’s Conception of International Law and its Methodological Background

Hart’s conception of IL is a good standpoint from which to shed a different light not only on the solutions Hart gives on the “persistent questions” ⁵⁰ of the theory of law, but also on the way he reaches those solutions. ⁵¹ Chapter X is indeed one of the few chapters in The Concept of Law containing methodological remarks.

We can say that Hart’s conceptual analysis, if considered in the light of that conception, appears less language-dependent than it actually is and more responsive to the ‘reality’ conveyed by language. In Hart’s view, if one puts aside the simple truisms underlying the use of the word ‘law,’ the concept of it cannot really be grasped and analysis leads to a matter “concerning the proprieties of linguistic usage,” ⁵² without sharpening “‘our perception of phenomena.’” ⁵³ Multiple practises are commonly called legal, some of them show similarities in structure, function, and content, and analysis, if aimed to be conceptually deep, cannot ignore “the core of good sense” ⁵⁴ in the ‘minimum content’ claim. To be sure, this claim should be not meant as a method properly understood: the simple truisms “are for sociology or psychology like other sciences to establish by the methods of generalization and theory, resting on observation and, where possible, on experiment.” ⁵⁵ But without this “good sense,” conceptual analysis would be trapped between linguistic enquiry and empirical research. As result, those similarities might deceive, some relevant problems in legal theory would be left undetermined and in any case cannot be solved through language only as, for example, the problem of coercive system. “There are no settled principles forbidding the use of the word ‘law’ of systems where there are no centrally organized sanctions, and there is good reason (though no compulsion) for using the expression ‘international law’ of a system, which has none.” ⁵⁶ However, although both IL and state law are linguistically referred to as ‘law,’ “we do need to distinguish the place that sanctions must have within a municipal system, if it is to serve the minimum purposes of beings constituted as men are.” ⁵⁷

This approach is particularly appropriate in the international domain because here the linguistic usage is even more misleading. This usage has no doubt about the legal status of IL, but in fact many are sceptical about the binding force of it. Thus, either we passively rest on that usage or at most we can decide “whether we should observe the existing convention or depart from it,” ⁵⁸ but in both cases we are not making “explicit … and examining the principles that have in fact guided the existing usage.” ⁵⁹ And exploring those principles causes Hart to go beyond “definitions and ordinary statements of fact” and to reveal that, despite the current usage of word ‘law’ common to municipal law and IL, there are two realities that are rather different from each other: IL does not enable secondary rules and, on the contrary, highly discourages them because “to initiate a war is, even for the strongest power, to risk much for an outcome which is rarely predictable with reasonable confidence.” ⁶⁰

It is this attentiveness to reality that explains why Hart openly opposes those who go in search of a basic norm or a rule of recognition for IL. This occurs either when theorists remain prisoners of their own theories, “as if we were to insist that a naked savage must really be dressed in some invisible variety of modern dress,” ⁶¹ or when they stop at external resemblances between state law and IL, “in their anxiety to defend against the sceptic the title of international law to be called ‘law.’” ⁶² But as Hart states, “once we emancipate ourselves from the assumption that international law must contain a basic rule, the question to be faced is one of fact” ⁶³ and the facts say that those resemblances “at present seem thin and even delusive.” ⁶⁴

Nevertheless, IL can be said to be law just as much as state law. First, despite of the lack of a coercive system, properly understood rules exist, as they possess what Hart labels the “internal aspect.” ⁶⁵ Secondly, the pressure for conformity has a mostly legal nature, as “what predominate in the arguments, often technical, which states address to each other over disputed matters of international law, are references to precedents, treaties, and juristic writings.” ⁶⁶ Two things emerge. First, IL is law even if it is not a legal system. And second, this conclusion cannot be gained through mere conceptual analysis or theorization, no matter how refined that analysis or theorization is, of lexical recurrences, but can only be arrived at by arguing in terms of ‘minimum content.’ ⁶⁷

We can now turn back to Hart’s account of IL with a less critical eye. No doubt, he has good reasons for restricting the notion of a legal system to state law. True, IL and state law exhibit analogies of function and content, Hart says, ⁶⁸ and after World War II the United Nations seems to have strengthened the structure of IL. ⁶⁹ But to conclude based on those resemblances the systematic nature of IL is to ignore that “the simple truisms which hold good for individuals do not hold good for states, and the factual background to international law is so different from that of municipal law.” ⁷⁰

The opposite view fails to grasp the methodological reasons behind this position and conceives of Hart’s conceptual analysis more as linguistic than as dependent on the ‘minimum content’ claim. ⁷¹ A notion of a legal system embracing both IL and municipal law would be a purely linguistic solution, and that is not something that Hart is interested in. ⁷² In addition it would have a weak explanatory power, since it would gloss over the substantive differences between international and national domains. ⁷³ And certainly conceptual analysis comes up short if the problem is only to decide whether to give IL the name ‘legal system.’ ⁷⁴

However, it is worth noticing that this approach, to the extent that it offers a deep insight on IL, ends up with a sort of split image about what the word ‘law’ applies to. On one side there is the central case of the concept of law as applied in state law. On the other side, there is a further instantiation in IL and primitive orders of that concept. There are even some commonalities between them: pressure for conformity based mostly on the use of coercion; a rule-acceptance usually not morally grounded; and the lower importance of a great deal of legal rules. Yet, since the union of primary and secondary rules represents, for Hart, the only central case in the usage of word ‘law,’ those commonalities are not enough to give IL an undisputed legal status. ⁷⁵ Such status is unsettled and prone to be confused with morality, from which IL needs to be constantly distinguished, even if it is linguistically deemed as legal. ⁷⁶

In other words, the relevance Hart gives to a genre of statements “besides definitions and ordinary statements of fact” ⁷⁷ does not lead to a balanced account of the simple truisms at the roots of social life, but to a bias in favour of those proper to state law. ⁷⁸ Both municipal law and IL have their own minimum content, but only that content related to the municipal realm seems to possess an unquestionable legality. ⁷⁹ That related to the IL is not conceptually self-standing and differences seem to amount to deficiencies of IL in respect of municipal law rather than variants of an overall view of law. ⁸⁰ The result is that state law certainly falls into the core of settled meaning of the word ‘law’ and the idea of state law is well expressed in terms of union of primary and secondary norms. On the other hand, IL slips into a (not linguistic, but substantive) penumbral area, and is theoretically coupled with the law of primitive societies, blurring its complexity and structure. Law par excellence is, then, only state law: it turns out to be the touchstone of the legality of any other rule practice, which, however, can never be on a level with it. ⁸¹

From this point of view, while those criticisms summarized in section I are not groundless, they do need to be better targeted. Hart’s theory of law is flawed, but not because it denies the systematic nature of IL—municipal law only can be regarded as a genuine legal system. Hart’s theory of law is flawed because he interprets the ‘minimum content’ claim in a state-centric fashion—a claim which nevertheless offers a deep insight into law and strengthens conceptual analysis, as seen before. ⁸² In other words, Hart’s state-centrism does not affect the concept of legal system, but that of law, which fails to explain to the same extent any rule practice held to be, in ordinary language, ‘legal.’ Law cannot be defined per genus et differentiam; it could be better analyzed on the basis of the well-known idea of “family resemblances” ⁸³ —some of them are the commonalities seen before between IL and municipal law—but in this ‘family’ IL seems to have only a minor role to play. Hart does not allow for the idea of a multifaceted legality where any instantiation of law deserves equal consideration and could be deemed to be as central as any other. ⁸⁴ The essence of legality lies ultimately in the notion of ‘legal system,’ which means that state law is the only central case of ‘law’: sets of separate rules may be legal—as the case of IL—but, since they are not systematic, they do not possess a fully developed legality. As result, the place of IL in Hart’s legal theory is nothing more than a sort of by-product arising from the weight Hart attaches to state law and such a bias, applied to IL, ends up distorting it. IL has such an uncertain legal status that it could even be viewed as a kind of (interstate) morality. It appears to have no structure, which is surely not a union of primary and secondary rules, yet it is an excess to speak of “a simple regime of primary or customary norms.” ⁸⁵ That is an image of IL that no international lawyer should accept.