Skip to navigation – Site map

HomeNuméros52Symposium on the making of consti...Legal constitutionalism and the I...

Symposium on the making of constitutional democracy

Legal constitutionalism and the Ius/Lex distinction

Mathieu Carpentier

Abstract

This paper takes a critical look at Paolo Sandro's distinction between ius and lex, which, according to him, is essential to modern constitutionalism. After explaining why he thinks this distinction is important, I will explain why I think it is problematic in its own right, and furthermore, why it undermines several parts of Paolo’s main tenets. I will show that his main worry, namely, the question of “how are we to explain the existence of illegal/unlawful law?”, can be addressed and answered without having recourse to the ius/lex distinction. The somewhat banal adage "lex superior derogat inferiori" is much better suited to that task.

Top of page

Full text

1Paolo’s book is masterful in many respects, and there is much in it with which I agree. For instance, chapter 4 on discretion is, in my opinion, a tour de force. Obviously, this paper would not be of any interest if it just pointed out the many areas of agreement between Paolo and me. So, instead, this paper will address some issues that Paolo and I staunchly disagree about, and this will allow us to push an ongoing discussion further. Ultimately, I think these issues are unnecessary complications brought about by Paolo’s eagerness to rattle and renew our common frameworks, and I think they are severable from the main points of his book.

2 Paolo writes that the distinction between lex and ius is “the duality that makes constitutionalism possible” (Sandro 2022: 50). It must be an important distinction then! However, the distinction is introduced in chapter 2, where it is discussed at length, and then it just disappears from the remainder of the book, only to make a timid reappearance at the end of chapter 6 (Sandro 2022: 245) where it surprisingly (somewhat) undermines the book’s main thesis. Ius and lex, it appears, are treacherous friends.

3 I will proceed in three parts. After explaining why Paolo thinks this distinction is important, I will show why I think it is problematic in its own right, and furthermore, why it undermines several parts of Paolo’s main tenets. Instead, I will show that Paolo’s main “Ferrajolist” worry, namely, the question of “how are we to explain the existence of illegal/unlawful law?”, can be answered quite simply: Lex superior derogat inferiori. (There will be a lot of Latin in this paper, and I apologize in advance).

1 The distinction: a quick précis

4One can sum up the ius/lex distinction, as it is introduced by Paolo, as follows. If law is understood as the mere product of political power, then it is impossible to understand how that power can be limited by law. It may be limited by a set of very different factors (self-restraint, religious factors, political expediency etc.), but law cannot be one of them. If law is to limit the ability of political organs to enact law, then it must be of a different kind than the law it limits. Hence the distinction between lex, which is the product of political power and ius, a type of positive law that stems not from political will but from the customs discovered by jurists by the use of reflexive reason.

  • 1 These cursory remarks cannot do justice to Vermeule’s common good constitutionalism. Suffice it to (...)

5It’s important to stress that ius is positive law, even if it is (at least prima facie) not posited. Paolo’s insistence on this positive character is important, since he wants to avoid the association with the recent resurgence of natural law theorizing about the ius/lex distinction. As Paolo himself notices (Sandro 2022: 52), arch-natural lawyer John Finnis (2011: 228) briefly but sharply criticized people like Michel Villey (see infra) who drew too sharp a distinction between lex and ius. But nowadays, natural lawyers are rediscovering this distinction as part of the highly contested new “common good constitutionalism”. Adrian Vermeule thus writes: “Lex is the enacted positive law, such as statute. Ius is the overall body of law generally, including and subsuming lex but transcending it, and containing general principles of jurisprudence and legal justice” (Vermeule 2022: 4). Vermeule insists, somewhat mysteriously, that ius civile (municipal law) is “positive law without jurisprudential positivism” (Vermeule 2022: 18). However, he claims that positive law as it is “can only be interpretated in light of principles of political morality that are themselves part of the law” (Vermeule 2022: 6). Ius civile is determined by ius naturale through and through. Based on this understanding, Vermeule suggests that constitutional provisions should be interpreted not only as an instance of lex, but with regard to its background moral and political commitment, that is the principles of legal justice that are part of ius naturale1.

  • 2 As we shall see shortly, the English language doesn’t have a distinction between Gesetz/loi/legge/l (...)

6Paolo acknowledges that many authors in the Anglo-Saxon world2 (such as Hart 2012: 208) tend to reduce ius to a moralized concept of law (Sandro 2022: 53). But Sandro disagrees: ius and lex ought to be understood “as two different independent types, or bodies, of law, whose constitutive difference lies in their distinct sources”. Ius (at least in the sense of ius civile) is positive law through and through. It is grounded not in moral facts, nor in moral laws. Rather, it is a mere social fact, but this social fact (or set of facts) is the complex combination of various lawyers and judges working in coordination. So, there are two “types” or “bodies” of law, lex and ius. Paolo is adamant that we should distinguish between them, and I think there are two main reasons. One is jurisprudential and the other is a more contingent, political reason.

  • 3 On Ferrajoli, see notably, Moreso 2021 and Chiassoni 2011.

7The first reason is that Paolo is fascinated by the problem of unlawful law, which is part of the background of Ferrajoli’s garantiste conception of constitutionalism3 (Ferrajoli 1989). Modern constitutionalism brings a series of warrants against misrule with the use of one specific technique, namely, making law produced by the political organ (typically the legislature) invalid. Ferrajoli (2007: 530-542) and Paolo both wonder: how can a law be unlawful? The former answers by distinguishing between law in force and valid law. Paolo elaborated and improved this distinction in a very important paper, “Unlocking Legal Validity” (Sandro 2018). The ius/lex distinction appears to be yet another refinement on his earlier theory, since it brings the idea of a certain type of law not amenable to political fiat to the fore: “it is only with the emergence of a new type of positive law (besides law as the expression of political authority) that it becomes possible, conceptually to limit law by law” (Sandro 2022: 53). If law is only lex, Paolo argues, there is no other way of limiting law but to limit it from the outside (e.g., by moral norms); it is only if there is another, kind of law, different than lex, that lex can be limited and checked by law.

  • 4 Thomas Bonham v College of Physicians [1610] 77 Eng. Rep. 638.
  • 5 R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22.
  • 6 R (Miller) v The Prime Minister and Cherry v Advocate General for Scotland [2019] UKSC 41.

8The second reason is Paolo’s strategic insistence that constitutional codification is but a stage of constitutionalism and that there is no intrinsic difference, in that respect, between codified and non-codified constitutions. Codification is just a way of fixating the content of ius at a moment in time. This insistence is, as I said, strategic, and it is political. The point is that in a country such as the UK, common law provides a constitutional framework that the “lions under the throne” (Sedley 2015) can use — and have used — against first the King and then Parliament, from, say, Dr Bohnam’s case4 to the Supreme Court’s treatment of ouster clauses in e.g., Privacy international.5 Paolo writes: “the common law of the land that has developed in England since the twelfth century represents – or so it can be argued – a direct evolution of the Roman ius model” (Sandro 2022: 11). This is a hotly debated subject and I will not delve into it in much detail here. Suffice it to say that Paolo’s brand of common law constitutionalism clashes with many dogmas, such as parliamentary sovereignty, and that the extent to which courts in the UK can be put to the task of checking the actions of political organs is a matter of deep controversy (especially after the Miller 2 case in 20196). It could also be argued that by making the common law the paradigm of constitutionalism in the UK, Sandro to some extent erases Scots Law, but again, that is really beyond the reach of this paper.

2 A problematic distinction

9There is no denying that, both as a terminological matter and as a legal and conceptual one, ius and lex mean two different things. Any French or German reader would readily agree that Gesetz and loi mean something different than Recht and droit, and to a very large extent this distinction maps onto the lex/ius distinction. Paolo is also right that the relationship between ius and lex is not — or at least not exclusively — a relation between a sum total and one of its components (lex being a kind of ius, or a source of ius). Indeed, the exact locus of the distinction is somewhat ambiguous. Although in certain contexts it makes sense to say that lex is one of the sources of ius, in other contexts it does not. What makes the distinction difficult to translate into English is that although lex can be translated as statute and ius as law in certain contexts (statute being —uncontroversially I think— a source of law), it’s not so simple in others. Sometimes, lex doesn’t mean a (statute-like) norm, but the law, such as in the expression “in the name of the law” (“Im Namen des Gesetzes”, “au nom de la loi”), in which lex means nomos, the very concept of a general, authoritative legal norm. Thus, Article 6 of the French Declaration of 1789 reads: “The law (la Loi) is the expression of the general will”, “la loi” goes well beyond mere statutes: it refers to the very idea of a general legal norm, laid down in a top-down manner by any authority legally authorized to represent the people and express their will.

10Meanwhile, ius often means law in general, but it can also do so in a moralized sense (pace Paolo): in French, droit is infused (albeit certainly mistakenly) with a somewhat undertheorized notion of justice. When we say that something is contrary to “le Droit”, we do not just mean that a law (a lex) prohibits it. Besides Ius/droit/Recht can also mean “right”, hence the famous mistake of translating Rechtsphilosophie in the Hegelian context as Philosophy of Right.

11 I am not sure that what precedes will make much sense to a native English speaker, but do not think it would be disavowed by a French or a German jurist. It shows that the ius/lex distinction is real, but also that it is complex, if not outright messy. In what follows I will try to show that, in part due to this messiness, the ius/lex distinction cannot serve the function nor fulfill the purposes that Paolo assigns to it.

2.1 Problems with the Roman law pedigree of the distinction

12I am not a legal historian, and this is meant to be a short paper. So, I apologize in advance for the oversimplifications that follow.

13According to Paolo, ius emerged as a distinct body of law, different and separate from law as the expression of the will of political authorities. Drawing on Schiavone (2012), Paolo locates the emergence of ius at the end of the Roman Republic, when a small group of jurists interpreted and systematized the ancient mos into a body of workable rules. Ius would, then, compete with lex, and ultimately limit lex, as the proper way to regulate the social order.

14However, a closer reading of Schiavone’s book yields a somewhat more nuanced and complex picture. As Schiavone shows, ius actually predates lex, and its origins can be found in the early years of Rome’s foundation. Contrary to the Greek notion of nomos, which was the standard way of ordering the social group in Greek polities, in the early years of the Roman monarchy the Roman legal system did not take the form of top-down, publicly promulgated rules. Instead, it was a religious matter (see Magdelain 1986), placed in the trusted hands of the pontifices, the religious authorities (Schiavone 2012: 76 ff): they would frequently be asked questions by the patres — the heads of the affluent and powerful patrician families — on what ius required in such-and-such a case, and the pontifices’ answers (responsa) were authoritative. The archaic ius, such as it appeared and consolidated itself in the last century of the Roman monarchy (7th-6th century BC), was not a coherent set of publicly laid-down general rules but rather a set of private answers to private legal problems, provided on a case-by-case basis. As such, they dealt exclusively with private law matters.

  • 7 The existence, as well as the nature and content, of leges regiae, which predated the Republic, is (...)

15The first recorded leges appeared with the Republic7 at the very end of the 6th century: the first recorded lex is the lex Junia, which provided for the deposition of the last Tarquinius king. Leges were primarily enacted by the comitiae, the people’s assemblies, and although they were top-down laid-down acts of political will, they remained quite sparse throughout the Republic: there is only a hundred or so statutes recorded between 509 and 30 BCE. The first written lex that we have written traces of is the Law of the Twelve Tables, promulgated in 449 BCE. The Twelve Tables were a fundamental turn in many respects. As Schiavone shows, they were an attempt to put the secret, aristocratic and religious content of the pontifical responsa into a set of coherent, publicly promulgated rules: “from a secret ius to an officially proclaimed lex” (Schiavone 2012: 93). There is no denying that the Twelve Tables marked a paradigm shift, not only because of the specific mode of law-issuance it entailed but also because lex was one of the tools of a shared democratic aspiration (Ducos 1984: 83 ff).

16Ius was not dead as a way of law-making, though. And as Paolo rightly points out, it regained its strength throughout the Republic in an ever more secularized form, albeit still a tool at the hands of the patrician nobility (Schiavone, 2012: 136). It evolved into a set of rules, worked out by jurisconsults (such as the Mucii Scaevolae) and judges (praetores), whose edicts went beyond the mere case at hand and applied to all citizens. By the end of the Republic, ius was a wholly systematized body of rules, regulating most aspects of the lives of Roman citizens.

  • 8 Besides, as Magdelain notes, this (allegedly) non-derogable ius was also comprised of… previous leg (...)
  • 9 This is the formula that Cicero mentions in the Pro Caecina. The exact formula may be different, an (...)

17Does that mean that Paolo is right that ius was thought of as a check on the legislative powers of the political authority (that is, mainly the comitiae)? It is quite doubtful. Not only were they “distinct but not diametrically opposed” (Schiavone, 2012: 135), but they did not have the same object. Ius was concerned exclusively with private law issues. Some leges, beginning with the Twelve Tables, did bear on private law, but they were the exception. Most of them were dealing with public law issues, such as the powers of consuls, the relationships between plebeians and patricians, the powers of the Senate and the comitiae, etc. It is therefore inaccurate to make ius the ancestor of constitutionalism, as Paolo does, since ius (as a specific body of law) was not at all concerned with constitutional rules. Paolo is right that in the early years of the Republic there was a conflict between ius and lex, but it resulted in a division of labour between ius and lex: ius was not a limit on lex, but rather a set of rules adopted in another way than leges were and relative to another domain. Even though in the early Republic leges would typically have sections stating that they ought to be construed in a way consistent with previously established ius8 (“si quid ius non esset rogarier, eius ea lege nihilum rogatum9), the fact remains that leges that would encroach on ius’ domain or that would even be inconsistent with ius were not eo facto unlawful (Ducos 1984: 220). It is therefore misleading to claim that the ius/lex distinction makes constitutionalism possible, since 1) the distinction between ius and lex was domain-dependent, 2) a lex encroaching on ius would not be invalid, and 3) (what would today be called) constitutional rules were primarily leges.

18Besides, throughout the imperial period, the distinction became more and more fluid, as the legislative powers of the people were transferred into the hands of the Senate and the emperor’s decrees (constitutiones) evolved as an autonomous source of law. By the time the Corpus iuris civilis was compiled, the idea that ius and lex were two different bodies of law, or even two different “sources of law”, had long been abandoned.

19Rather, “ius meant roughly three different things. The first meaning of ius was what we call law, or “le droit”, in general (making it quite clumsy to designate it as a “body/source of law”, that is, of itself!) Even though Livy was clearly indulging in a bit of wishful thinking in the early years of Augustus’ reign when he wrote that the Twelve Tables where “the source of all public and private law” (fons omnis publici privatique iuris) (Livy, History of Rome, III, 34), the idea that lex was a “source” of ius was clearly right, even though the Roman meaning of “fons iuris” was very different from our modern notion of formal sources of law. Gaius thus writes: “Constant autem iura populi Romani ex legibus, plebiscitis, senatusconsultis, constitutionibus principum, edictis eorum qui ius edicendi habent, responsis prudentium” — i.e., “Roman law consists of statutes (leges), plebiscites, senatusconsults, constitutions of the emperors, edicts of magistrates authorized to issue them, and opinions of jurists” (Gaius, Institutes, I.1.2). Ius (law) is a set of iura (laws) of which leges are one important species. See also e.g., Digest, 1.3.3.: iura constitui opportet in his quae (…) ut plurimum accidunt” (“one ought to establish laws (iura) only with regard to cases which occur frequently”). And Digest 1.3.9 clearly shows that political authorities can enact ius:Non ambigitur senatum ius facere posse”: there is no doubt that the Senate (as political an authority as there is, especially during the imperial period!) can make ius.

20The second meaning, certainly the clearest legacy of the original concept of ius, was the notion of ius as a specific science or expertise. The famous sentence at the very beginning of the Digest, “ius est ars aequi et boni” (Digest, I.1.1), refers to ius as an art, that is, as the science of what is just and unjust, to which jurists, lawyers and judges alike partake.

21The third meaning is ius as right, albeit in the sense of authorization or (conferred) power (such as in the praetor’s ius edicendi), rather than in our modern sense of subjective rights.

22 To sum up: although Paolo is right that ius and lex initially meant two different “bodies of law” regulating two different aspects of social life, the distinction did not mean that ius acted as a check or a limit on the lawmaking powers of political authorities, the way constitutional rules would. A lex that would contradict some part of ius would not be eo facto unlawful or invalid. And the ius/lex distinction evolved in such a way as to make the distinction practically irrelevant, except for the moralized idea of ius as the art of justice.

2.2 Problems with the jurisprudential relevance of the distinction

23Now, let us put aside issues with Roman legal history. Is the ius/lex distinction (even stipulatively defined) relevant for jurisprudential enquiries into the nature of law? I do not think so.

24Some legal philosophers have tried to give jurisprudential significance to the distinction. For instance, the French legal philosopher (and all-around eccentric) Michel Villey, a staunch defender of the ius/lex distinction, claimed that ius, not lex, was the core of law’s nature. He argued that ius, as such, was not a matter of rules or norms at all. The consequence was that law was not a matter of rules or norms. The modern emphasis on la loi, Gesetz, norm, and the like was a distortion of law, of the true nature of law, of le droit (see Villey 2006: 196 ff).

25In Villey’s view, ius is not a matter of normative prescription; it is not a matter of deontic propositions; it is a mere matter of objective fact. Drawing on the famous formula by Ulpian (Digest 1.1.3) that justice is “ius suum cuique tribuere” (to give to each their own), Villey argues that ius means a just state of affairs, for instance, the just state of affairs that Pomponius has a debt of 1000 euros vis-à-vis Titus. The relevant point according to Villey is not that Pomponius ought to pay 1000 euros, or that there is a norm to that effect. The point is that Titus has a thing, which is a claim of debt against Pomponius, and if Pomponius disagrees, a judge will say, “no, Titus is the creditor”. This is a descriptive sentence, not a prescriptive one. Whereas lex is a norm, ius is a fact. It’s a thing, or rather a relation between things (debts, claims, properties, etc.) and persons. Ius is the just ordering of relations between things and persons, and whether x is ius or not is just as truth-apt as the proposition “it rains”. In that perspective, lex can, albeit very marginally, modify the parameters of what is just, just as other inputs, such as the jurisprudents’ writings and the pretorian edicts usually do, but the operation of law is ultimately a matter of fact, not a normative matter. Per Villey, “all the things that we call law [le droit] preexist the written rules which express them” (Villey 2009: 93). Villey’s theory is very strange (I will not elaborate further), but it is at least consistent in the sense that it draws a sharp ontological distinction between ius and lex.

  • 10 For a defense of the idea of legal scholarship as a source of law, see, notably, Peczenik 2005: 17 (...)

26I cannot see anything similar in Paolo’s distinction (and that is for the best, if I may add). If we are not to draw an ontological dichotomy between ius and lex as sharp as Villey draws — a dichotomy that is not germane to our intuitions about the way legal systems currently operate —, then one is left with the mere distinction between two bodies, or two sources, of law. Paolo is clearly right about lex being a distinct body of law that is dependent on the will of political authorities. But I do not see why we should lump together all the other sources of law (notably custom, precedent, and perhaps legal scholarship10) under the appellation “ius” just because at some point during the late Roman Republic ius meant something such as a distinct body or source of law. And even if one uses ius with this specific meaning of a distinct source of law, there is still no explanation for why one body of law (ius) would necessarily act as a limit on the other (lex), making it unlawful to enact a norm belonging to one set that would be repugnant to the other. One would have to understand ius as a kind of ius cogens by which the lawmaker (that is the lex-maker) has to abide on pain of nullity. But I certainly do not see why it should necessarily be so, especially as ius was not thus understood by the Romans themselves.

27Indeed, it makes much more sense to understand ius the way imperial jurisconsults did, as (roughly) a synonym for “law in general” and/or a legal system: ius civile, ius gentium etc. In this respect ius is a set of legal norms (iura). Lex can be then understood in two different ways.

28Lex in the narrow sense means a specific source of law. As such, it refers to legislation in general, that is, any legislated norm, which includes not only primary legislation passed in Parliament but also constitutional legislation, delegated legislation, executive decrees, etc. In that narrow sense lex is one important source of law, and Paolo is clearly right when he suggests that lex is an important source of law, but by no means the only one. If we admit that sources of law are social facts, which the rule of recognition picks out as the criteria of validity for legal norms, then lex in the narrow sense means one specific type of source, that is one specific type of social facts, alongside other kinds of sources (custom, treaties, precedent, and so forth).

  • 11 Military and Paramilitary Activities in and Against Nicaragua, Nicaragua v United States, Merits, J (...)

29By extension or metonymy, however, leges and iura can be used interchangeably as meaning legal norms (or laws) in general, regardless of their sources. Such is the way lex is used in many Latin legal adages. A good example is Lex posterior or Lex specialis. When we say that the special lex (that is, “the special legal norm”, or “the special law”) derogates from the general one, the specific source of the lex in question is not quite relevant. One example among many others: in the Nicaragua v. US case in 1986, the International Court of Justice stated that “treaty rules being lex specialis, it would not be appropriate that a State should bring a claim based on a customary-law rule if it has by treaty already provided means for settlement of such a claim.”11 Neither treaties nor customary laws are leges in the narrow sense, but they definitely are in the broader, metonymic sense of lex as legal norm.

30So, either lex means legislation, and it is just one source of law among many others; or lex means legal norm, whatever its source. In both cases ius simply means law (in the sense of “legal system”, or “le droit”). The jurisprudential relevance of the distinction is quite limited, and we can use the usual categories of jurisprudential talk: law (le droit), legal system, laws, legal norms, sources of law, etc. Of course, the difference between various sources of law, and the centrality of legislation in modern legal systems is of great jurisprudential importance. But one does not need the ius/lex distinction to conceptualize it.

2.3 Problems for legal constitutionalism

31Paolo aptly identifies the crux of legal constitutionalism: how to make certain laws unlawful and how to understand the possible unlawfulness of law? He thinks that the answer is to be found in the ius/lex distinction: ius would be understood as the original matrix of the limits that can be legally imposed on the law-making powers of legal authorities, which is what a constitution is about. Even if I were to grant that such a distinction corresponds to Roman law usage (which I dispute), and even if I were to concede that it has jurisprudential relevance (which is highly debatable), I still think the ius/lex distinction, such as Paolo conceptualizes it, would not be fit to the task Paolo wishes to assign to it.

32First, the distinction is much too simplistic. On the one hand, you have lex as the product of the political will, and on the other hand, you have ius, a body of law acting as a limit on lex. There is unlawful law when and if (and only if) lex contravenes to ius. But as such, this distinction, as an explanation of unlawful law, cannot account for multi-layered legal hierarchies. To put it quite simply (trivially, even), unlawful law happens all the time and across the normative board, not just at the constitutional level. Constitutional systems usually create multi-layered legal hierarchies, not just a two-term relationship between one inferior and one superior set of legal norms. This is typically the case when e.g., an administrative decree clashes with a statute. And in certain countries, such as France, constitutionalism was first a kind of administrative law constitutionalism. The main guarantees of rights against the executive power were enforced by the administrative courts applying statutory law. For a long time in France, since the courts were not empowered to review the constitutionality of statutes, statutes themselves provided the benchmark of rights protection. Yet the Conseil d’Etat “discovered” general principles of law (PGD), which do smack of ius in Paolo’s sense. But the point was that statute was paramount and that statutes could indeed displace PGDs as they saw fit. Things are obviously different nowadays, and there is something like a constitutional review of statutes (such as it is exercised by the Conseil constitutionnel), but statutes still enshrine many fundamental rights guarantees against administrative misrule. Still, if I follow Paolo, a statute is lex when we compare it to the constitution (which is ius), but becomes an instance of ius when compared to an administrative decree or bylaw. Isn’t it more apposite to say that multi-layered constitutionalism creates a hierarchy of leges? If so, the ius/lex distinction obviously cannot be the key to understanding the process of unlawful law.

  • 12 Could ius describe only those parts of the constitution that are not amendable, such as supra-const (...)

33Second, Paolo writes that ius doesn’t have “its source in the will of the sovereign” (Sandro 2022: 51). But the whole theory of pouvoir constituant, of constituent power from Sieyes on, is that constitutions are the product of a sovereign political decision. Of course, Paolo rightly criticizes the notion of popular sovereignty (Sandro 2022: 61-62), but that’s beyond the point for the present purposes. My point is that, sovereign or not, constitution-making is a wholly political process — even in the context of a non-codified constitution. And of course, ordinary political organs play a huge part in amending the constitution, even in the most entrenched constitutions. So, it seems to me that even the most fundamental norms (Sandro 2022: 58) are still amenable to political fiat, even if such an event is constrained by a cumbersome amendment process.12 I agree with Paolo that constitutionalism doesn’t require an entrenched constitution, but rather that “the exercise of political power through law is limited juridically” (Sandro 2022: 59). However, this is neither here nor there insofar as the ius/lex distinction is concerned, since the very source of this “juridical limit” is itself the will of a political authority (albeit quite often, though not always, a different political authority, such as “the people”). If by “political authority” one only means the legislator (and maybe the executive), then one is just stipulating the problem away, if not straight-out begging the question. No law is beyond the reach of the “political authority” in unentrenched and entrenched constitutional systems alike; nor does the Constitution need to be beyond the reach of the “political authority” to act as a “juridical limit” (as opposed to a mere political one) on ordinary law-making. Therefore, I agree with everything Paolo has to say about legal constitutionalism: but he does not need the ius/lex distinction to articulate it, nor does he need to de-politicize the constitution-making process.

34Third, not only the ius/lex distinction is at best useless and at worst a distortion of the way constitutions operate, but it undermines Paolo’s whole point about the creation/application distinction being central to constitutional democracy. Since democracy is premised on the idea of self-government by the people, the laws enacted by the people’s representatives should be applied in a way that distinguishes them from mere law (re)creation (Sandro 2022: 76-78). Therefore, as Paolo argues contra the so-called political constitutionalists, “there is no difference in kind between the language used in legislative and constitutional texts” (Sandro 2022: 78): judicial review of statutes applies the Constitution the same way ordinary judiciary activity applies legislation. Constitution-making is as much an act democratic of self-rule as legislation. Of course, Paolo agrees that laws can be relatively indeterminate and that the law-applying process is not always, if ever, straightforward (Sandro 2022: 214). As Paolo insists, discretionary application (as opposed to bound application) is still application (see Sandro 2022: 116-168). He also agrees that the legitimacy of judicial review is, in fact, context-dependent (Sandro 2022: 74). It is clear, however, that according to Paolo, the legitimacy of constitutional adjudication rests on the creation/application distinction, and the idea that applying the Constitution is no different than applying a statute.

35But then, Paolo writes:

This is where, arguably, the duality between lex and ius that we have discussed in chapter two comes back to the fore: the more the decision of a court is not based on the application of lex but on the exercise of ius , the more the decision by the court will have to be justified on the grounds of its merits and other systemic considerations, rather than on being (linguistically) warranted by the general norm at its basis. This might be, in the end, the most important upshot of the distinction between creation and application put forward in this work: not the (descriptively untenable) rejection of the law-creation power of courts, but the clarification that in such cases the decision cannot be fully justified by the idea of application of law. (Sando, 2022: 245).

36Indeed, I for one agree that constitutional (and much of administrative) adjudication is sui generis to some extent; whenever a judge is empowered to review the constitutionality of a statute (or the “lawfulness of a law” more generally) the distinction between creation and application is muddied at best. But from Paolo’s own point of view, this cannot be so. If ius is at stake when a judge reviews the constitutionality of a statute, then applying ius is quite different from applying lex. If all constitutional law is ius and is therefore “constructed” as much as “construed”, then constitutional judges are bound to be at least in part law-creators. Even if you agree that constitutional principles are not bluntly created by judges, or even if you claim that they are not ‘legislated from the bench’ so to say, but they are somewhat discovered via a Dworkinian law-as-integrity interpretation (which I think Paolo would reject), even then it is still very difficult to describe such an operation as mere application of pre-existing law (if only a discretionary one). And, Paolo is right that constitutional adjudication involves value-judgements typical of the law-creating process.

  • 13 I am not here directly concerned with the question of the democratic legitimacy of judicial review, (...)

37The upshot is that Paolo cannot have it both ways. He cannot on the one hand insist that applying the constitution and applying the statute — the two types of norms being both acts of democratic self-rule — is roughly the same process, with no real difference in terms of legal indeterminacy, then on the other hand acknowledge that since the constitution belongs to ius it is bound to be applied in a creative way. If the constitution is lex, then constitutional review can be democratic, but then the ius/lex distinction fails; if the constitution is ius, then the application/creation distinction collapses insofar as constitutional adjudication is involved, and judicial review of statutes cannot be described as democratic.13

3 Lex superior derogat inferiori

38Let us take stock. The ius/lex distinction Paolo draws is 1) inaccurate insofar as Roman law is involved; 2) irrelevant as far as general jurisprudential inquiries are concerned; and 3) useless for the explanation of legal constitutionalism. It even backfires up to the point that it threatens the very consistency of Paolo’s defence of legal constitutionalism.

39Given that I agree with a lot of what Paolo has to say on the nature of law and adjudication, as well as with his defence of legal constitutionalism, the solution I propose is simple: Paolo should drop the ius/lex distinction. In fact, he doesn’t need it since there is a much simpler explanation of the way modern constitutionalism frames the “unlawful law” issue: lex superior derogat (legi) inferiori.

  • 14 Montesquieu 1989: 158-166.

40Legal constitutionalism is roughly the acknowledgement that the political and mechanical conception of the Constitution, which is roughly Montesquieu’s, has to be translated into legal terms, into a “normative” conception of the constitution. Montesquieu’s idea14 is that the Constitution is a set of powers allocated among the various organs of the State with the hope that the interplay of the various powers at hand will yield an equilibrium and prevent abuses of power. This is roughly (mutatis mutandis) the model of checks and balances. Legal constitutionalism, drawing on this mechanical conception of the Constitution as a fine-tuned mechanism, goes further: a constitution is a set of laws, i.e., of legal norms, which set limits to what the organs of the State are empowered to do. The equilibrium is not obtained from within (that is by the interplay of the various powers), but from the application of external limits set by legal norms. Like all legal norms, constitutional norms are meant to be enforced by a certain type of organ, namely, the judicial organ.

41The crucial point here is that in translating the power-allocating mechanism into a set of legal norms, legal constitutionalism did not aim, pace Paolo, to create a “legal otherness” (Sandro 2022: 64 ff). On the contrary, legal constitutionalism aimed to assimilate, to homogenize constitutional rules with statutes. The reason is simple: when a conflict between the constitution and a statutory provision occurs, the legal way to solve the conflict is to treat them as belonging to the same genus, so that one norm can be construed as derogating the other. In a nutshell, to ensure political actors are indeed constrained by the constitution, the best way is to ensure that whenever their lex clashes with the constitutional lex, the latter prevails. Thus, the best way to empower judges to enforce constitutional norms is to treat the conflict between a statute and the constitution as an ordinary conflict of norms. In other words, the adages Lex posterior and Lex specialis must be supplemented with Lex superior.

  • 15 See, notably, Qu’est-ce que le Tiers-Etat? Sieyès 2003 and his defense of a judicial review mechani (...)

42This idea, which is present in a somewhat rudimentary form in Sieyes’s writings throughout the French Revolution,15 is closely associated with the birth of judicial review in the United States. Both Hamilton in the Federalist 78 and Chief Justice Marshall in Marbury v. Madison take up a similar line of reasoning.

This exercise of judicial discretion, in determining between two contradictory laws, is exemplified in a familiar instance. It not uncommonly happens, that there are two statutes existing at one time, clashing in whole or in part with each other, and neither of them containing any repealing clause or expression. In such a case, it is the province of the courts to liquidate and fix their meaning and operation (Hamilton, 2008: 382).

  • 16 Hamilton seems to think that the Constitution is always prior to statutory rules. But when a subseq (...)

43It is the province of the courts to solve normative conflicts, and judicial review instantiates a species of ordinary normative conflicts. Hence, courts ought to review the constitutionality of statutes, QED. Hamilton then explicitly (albeit a bit confusedly16) contrasts Lex posterior and Lex superior:

It is a rule not enjoined upon the courts by legislative provision, but adopted by themselves, as consonant to truth and propriety, for the direction of their conduct as interpreters of the law. They thought it reasonable, that between the interfering acts of an equal authority, that which was the last indication of its will should have the preference. (…) But in regard to the interfering acts of a superior and subordinate authority, of an original and derivative power, the nature and reason of the thing indicate the converse of that rule as proper to be followed. They teach us that the prior act of a superior ought to be preferred to the subsequent act of an inferior and subordinate authority; and that accordingly, whenever a particular statute contravenes the Constitution, it will be the duty of the judicial tribunals to adhere to the latter and disregard the former (Ibid.).

  • 17 Marbury v. Madison 1803 5 U.S. 137.

44This line of reasoning is explicitly endorsed in very similar terms by Marshall CJ in Marbury v. Madison17:

It is emphatically the province and duty of the judicial department to say what the law is. Those who apply the rule to particular cases, must of necessity expound and interpret that rule. If two laws conflict with each other, the courts must decide on the operation of each.

So if a law be in opposition to the constitution: if both the law and the constitution apply to a particular case, so that the court must either decide that case conformably to the law, disregarding the constitution; or conformably to the constitution, disregarding the law: the court must determine which of these conflicting rules governs the case. This is of the very essence of judicial duty.

If then the courts are to regard the constitution; and the constitution is superior to any ordinary act of the legislature; the constitution, and not such ordinary act, must govern the case to which they both apply.

  • 18 Of course, in many instances Lex superior, Lex posterior, and Lex specialis clash with one another.

45This assimilation of judicial review to the resolution of a normative conflict, itself a routine part of judicial activity, presupposes that constitutional norms are of the same kind as the statutes they occasionally clash with, since the conflict between constitution and statute is not fundamentally different from a conflict between two statutes. The resolution of both conflicts follows a similar logic: lex superior when the norms in conflict rank differently within the normative hierarchy; lex posterior when they are “acts of an equal authority”.18

  • 19 See Troper 2005.
  • 20 As Kelsen argued, constitutional review is more akin to an act of legislation (albeit a “negative” (...)
  • 21 See on this Carpentier 2020: 146-147.
  • 22 Halperin 2012: 354.

46I am by no means claiming that Hamilton’s and Marshall’s reasoning is sound. Indeed, there is good reason to think that it rests in part on a fallacy.19 There is also much reason to doubt that judicial review is a mere routine judicial act,20 and that it always, if ever, involves the resolution of a genuine normative conflict.21 But that is beyond the point for the present purposes. My only point is that the standard line of reasoning involved in what can be described as the acme of legal constitutionalism (Hamilton/Marshall) does not rest on the ius/lex distinction, but rather on the very idea that judicial review involves a conflict between two leges. What makes “legal constitutionalism” possible, then, is not the distinction between ius and lex, but the idea of a normative hierarchy between various kinds of leges. To some extent, this notion of a normative hierarchy among leges was implicit in Roman law and throughout its reception in modern jurisprudence;22 but its explicit recognition, now pervasive and almost intuitive to most lawyers, is indeed the intellectual legacy of the legal constitutionalist movement.

Top of page

Bibliography

Carpentier, M. (2020). Kelsen on Derogation and Normative Conflicts. Archiv für Rechts und Sozialphilosophie (Beiheft), 163, 125–148.

Chiassoni, P. (2011). Constitutionalism Out of a Positivist Mind Cast: The Garantismo Way. Res Publica, 17 (4), 327–342.

Ducos, M. (1984). Les Romains et la loi. Recherches sur les rapports de la philosophie grecque et de la tradition romaine à la fin de la République. Les Belles Lettres.

Ferrajoli, L. (1989). Diritto e Ragione. Teoria del garantismo penale. Laterza.

Ferrajoli, L. (2007). Principia Iuris. Teoria del diritto e della democrazia (Vol. 1). Laterza.

Finnis, J. (2011). Natural Law and Natural Rights (2nd ed). Oxford University Press.

Halperin, J.L. (2012). Lex posterior derogat priori, lex specialis derogat generali. Jalons pour une histoire des conflits de normes centrée sur ces deux solutions concurrentes. Tijdschrift voor Rechtsgeschiedenis / Revue d'histoire du droit / The Legal History Review, 80(3–4), 353–397.

Hamilton, A. (2008). The Federalist No. 78. In A. Hamilton, J. Madison & J. Jay, The Federalist Papers (L. Goldman, Ed.). Oxford University Press. (First published 1787-1788).

Hart, H.L.A. (2012). The Concept of Law (3rd ed). Oxford University Press.

Kelsen. H. (1929). Wesen und Entwicklung der Staatsgerichtsbarkeit. 2. Mitbericht von Professor Dr. Hans Kelsen in Wien. Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer, 5, 30-84.

Magdelain, A. (1978). La Loi à Rome. Les Belles Lettres.

Magdelain, A. (1986). Le ius archaïque. Mélanges de l’école française de Rome, 98(1), 265–358.

Montesquieu, C. (1989). The Spirit of the Laws (A. Cohler, B.C. Miller & H.S. Stone, Trans.). Cambridge University Press. (Orginial work published 1748).

Moreso, J.J. (2021). An Italian Path to Legal Positivism: Ferrajoli’s Garantismo. In T. Spaak & P. Mindus (Eds.), The Cambridge Companion to Legal Positivism (pp. 606–623). Cambridge University Press.

Peczenik, A. (2005). Scientia Juris: Legal Doctrine as Knowledge of Law and as Source of Law. Springer.

Roznai, Y. (2015). Unconstitutional Constitutional Amendments. The Limits of Amendment Powers. Oxford University Press.

Sandro, P. (2018). Unlocking Legal Validity: Some Remarks on the Artificial Ontology of Law. In P. Westerman, J. Hage, S. Kirste & A.R. Mackor (Eds.), Legal Validity and Soft Law (pp. 99–123). Springer.

Sandro, P. (2022). The Making of Constitutional Democracy. Hart Publishing.

Schiavone, A. (2012). The Invention of Law in the West (J. Carden & A. Shugaar, Trans.). Harvard University Press. (Original work published 2005).

Sedley, S. (2015). Lions under the Throne Lions Under the Throne. Essays on the History of English Public Law. Cambridge University Press.

Shecaira, F. (2013). Legal Scholarship as a Source of Law. Springer.

Sieyes, E.J. (2003). What is the third Estate? In E.J. Sieyes, Political Writings (M. Sonescher, Trans. & Ed., pp. 92–161). Hackett Publishing Company (Original work published 1789).

Troper, M. (2005). Marshall, Kelsen, Barak and the constitutionalist fallacy. International Journal of Constitutional Law, 3(1), 24–38.

Vermeule, A. (2022). Common Good Constitutionalism. Polity.

Villey, M. (2006). La Formation de la pensée juridique moderne (2nd ed). Presses universitaires de France.

Villey, M. (2009). Critique de la pensée juridique moderne. Dalloz.

Top of page

Notes

1 These cursory remarks cannot do justice to Vermeule’s common good constitutionalism. Suffice it to say that framing the ius/lex distinction both as a primarily interpretive matter and as a battleground between positivism and non-positivism seems problematic to me.

2 As we shall see shortly, the English language doesn’t have a distinction between Gesetz/loi/legge/lex and Recht/droit/diritto/ius.

3 On Ferrajoli, see notably, Moreso 2021 and Chiassoni 2011.

4 Thomas Bonham v College of Physicians [1610] 77 Eng. Rep. 638.

5 R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22.

6 R (Miller) v The Prime Minister and Cherry v Advocate General for Scotland [2019] UKSC 41.

7 The existence, as well as the nature and content, of leges regiae, which predated the Republic, is disputed among specialists (Magdelain 1978: 24; Schiavone 2012: 91) and I will not delve into this debate here.

8 Besides, as Magdelain notes, this (allegedly) non-derogable ius was also comprised of… previous leges! Leges were incorporated into ius, just as pretorian edicts were Magdelain 1978: 61-64.

9 This is the formula that Cicero mentions in the Pro Caecina. The exact formula may be different, and it most certainly varied from statute to statute.

10 For a defense of the idea of legal scholarship as a source of law, see, notably, Peczenik 2005: 17 and Shecaira 2013.

11 Military and Paramilitary Activities in and Against Nicaragua, Nicaragua v United States, Merits, Judgment, 1986 ICJ Rep 14, at 137.

12 Could ius describe only those parts of the constitution that are not amendable, such as supra-constitutional norms? Paolo alludes to it when he mentions eternity clauses Sandro 2022: 58. I for one am not quite convinced by the literature on unconstitutional constitutional amendments notably Roznai 2015, even though they find solid ground in eternity clauses and basic structure doctrines developed by some supreme or constitutional courts around the world.

13 I am not here directly concerned with the question of the democratic legitimacy of judicial review, and I emit no opinion on the matter. I am merely pointing out the way the ius/lex distinction threatens Paolo’s own defense of judicial review as the hallmark of democratic constitutionalism.

14 Montesquieu 1989: 158-166.

15 See, notably, Qu’est-ce que le Tiers-Etat? Sieyès 2003 and his defense of a judicial review mechanism during the constitutional debates in 1795.

16 Hamilton seems to think that the Constitution is always prior to statutory rules. But when a subsequent constitutional amendment clashes with an earlier statute, Lex superior still applies (rather than just Lex posterior).

17 Marbury v. Madison 1803 5 U.S. 137.

18 Of course, in many instances Lex superior, Lex posterior, and Lex specialis clash with one another.

19 See Troper 2005.

20 As Kelsen argued, constitutional review is more akin to an act of legislation (albeit a “negative” one) than to a strictly judicial one (Kelsen 1929: 56).

21 See on this Carpentier 2020: 146-147.

22 Halperin 2012: 354.

Top of page

References

Electronic reference

Mathieu Carpentier, Legal constitutionalism and the Ius/Lex distinctionRevus [Online], 52 | 2024, Online since 25 March 2024, connection on 19 March 2025. URL: http://journals.openedition.org/revus/10130; DOI: https://doi.org/10.4000/revus.10130

Top of page

About the author

Mathieu Carpentier

Profesor of Public Law’Université Toulouse 1 Capitole and member of l’Institut Maurice Hauriou
E-mail: mathieu.carpentier(at)ut-capitole.fr

Top of page

Copyright

CC-BY-SA-4.0

The text only may be used under licence CC BY-SA 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search