Invalidating Laws: Lessons from Suárez for Contemporary Jurisprudence

Abstract

Compare:

Human will has [by nature] the power to enter contracts, make gifts and similar things, so long as [the law-maker] does not remove or impede this ability [by means of] invalidating laws.¹

Legal rules defining the ways in which valid contracts or wills or marriages are made … provide individuals with facilities for realizing their wishes, by conferring legal powers upon them.²

These statements were made three and a half centuries apart. The first is by Francisco Suárez, writing in 1612; the second is by H. L. A. Hart, writing in 1961. They are discussing broadly the same kinds of positive laws, namely those determining the requirements for making legally valid private transactions, such as contracts, marriages, or gifts, and for validly exercising public powers, as in, for example, the issuing of valid governmental regulations or judicial rulings. Hart famously dubs all these laws “power-conferring,” whereas Suárez calls them “leges irritantes,” which means “invalidating laws.” For Hart, these laws bestow on persons a power they could not otherwise wield. For Suárez, they take away a power that persons would otherwise hold. One and the same legal reality is described in apparently opposing ways by two writers who are familiar with the relevant legal provisions. They agree that these provisions are different from those that impose duties. They also agree that these provisions concern powers. But Suárez sees a withdrawal of power where Hart sees a conferral thereof.

This article argues that we can learn important lessons about legal power by appreciating the good sense in Suárez’s approach, which differs markedly from the dominant approach in contemporary scholarship, largely inspired by Hart. The lessons are important because they bear on current debates in both public and private law—involving inter alia the separation, allocation, and review of legal powers—that general jurisprudence has long neglected.

I proceed as follows. After mapping the parallel between Suárez’s and Hart’s accounts in section I, I discuss in section II what appears to be the central difference between them: namely, that Suárez speaks of a pre-legal power, whereas Hart speaks of a power that is wholly of legal making. I go on to demonstrate that the question whether the power is of legal or pre-legal origin is a distraction from the crux of Suárez’s theory, which is easy for modern readers to miss. The crucial difference between Suárez and Hart is that the former, but not the latter, puts the spotlight on the limits on legal power. In other words, Suárez focuses on how acts may lie outside a sphere of authority, whereas Hart focuses on acts that lie fully within it. As section III explains, Hart’s focus is reminiscent of the works of Bentham, Austin and Kelsen, who consistently ignore or explain away the limits on legal power, for reasons to which neither Hart nor many of us would subscribe. Section IV claims that this undue omission in our recent tradition can be remedied by going back to Suárez’s thought. I highlight three relevant lessons that we can draw from him; these lessons concern, respectively, the reasons for limiting legal power, the procedure for doing so, and the appropriate limits on those limits (and thus, among other things, limits on powers of annulment and review).

A terminological note: throughout the article, I speak of exercises of legal power as a synonym for acts-in-the-law, to refer to acts such as entering into contracts, making wills, issuing rulings, or enacting statutes. This is consistent with accepted jargon and is not meant to prejudge the question whether the origin of the power so exercised is pre-legal. That is a moot point discussed in section II—and, as it will turn out, not one that is central to our argument.

I. THE PARALLEL BETWEEN SUÁREZ AND HART

Few people writing nowadays in our subject seem aware that H. L. A. Hart’s influential distinction between duties and powers³ finds an avant-garde and in some respects more rigorous articulation in the work of Francisco Suárez, a seventeenth-century Jesuit writing in the natural law tradition. The claims that laws regulating the way to make private transactions or official rulings are distorted if cast as mere conditions of duties, and that invalidity is ill-regarded as a sanction, long predate modern Anglo-American jurisprudence. Suárez provides an extensive defense of these claims in his 1612 Tractatus de Legibus ac Deo Legislatore (Treatise on Laws and God as Legislator). My reason for unearthing Suárez’s view on legal powers is not to make a point in the history of ideas but to argue for a corrective to a widespread understanding of legal power that, as we shall see, is not altogether helpful. Because this understanding has its mediate or immediate source in Hart, this section maps out the main tenets of Suárez’s view of legal powers by comparing his account to Hart’s. Appreciating the parallels between the accounts will make it easier to pinpoint the differences.

Suárez breaks with a centuries-long tradition that deemed the possible functions of positive laws to be exhausted by the following list: to command, to prohibit, to permit, or to punish.⁴ Although Suárez does not nominally reject the received labels, in all relevant respects the rules he dubs leges irritantes are not reducible to any of these four kinds. He is perhaps the first major jurist to coin a distinct theoretical category for these rules, and systematically to reflect on the intersection between these rules and rules involving duty or prohibition. Indeed, he devotes the entirety of book 5 of his voluminous treatise to distinguishing the different types of positive laws.⁵ Out of the thirty-six chapters of this book, more than half (nineteen) concern leges irritantes. They occupy roughly the volume of a short modern-day monograph.

In the course of these pages, Suárez insists on the distinction between nullity and prohibition, and hence between nullity and sanction. “Many prohibited things are nevertheless valid if done,”⁶ he says: for example, sales for more than the item’s just price, or marriages in breach of prohibiting (but not invalidating) impediments.⁷ Such acts are wrongful and may be subject to punishment, yet they are valid in law. Conversely, he notes, acts may be invalid despite not being prohibited or wrongful. This is so, for instance, where provisions establishing the procedure for valid marriages or the form of valid wills have not been complied with. It is not a sin to fail to make a solemn will, he writes,⁸ but the act, being null, does not generate relevant legal duties and rights.⁹ Of course, as Suárez realizes, invalidity and prohibition may also overlap. A law may invalidate an act hand in hand with prohibiting it, as in the case of a consanguineous marriage.¹⁰ But even here, Suárez observes, one can in principle distinguish between the law’s repression of wrongful conduct and the nullity of the underlying transaction.

Suárez’s study of the structure and legal logic of leges irritantes is deeply interwoven with reflection on the moral point and the moral consequences of legal invalidity. The point of leges irritantes, he says, is not to punish or to curb wrongful conduct, but to remove from individuals a power they hold by nature, that is, by natural law.¹¹ It is their “power [potestas] to enter contracts, make gifts, and similar things.”¹² That power is a “moral power” (“potestas moralis”).¹³ By this, Suárez does not mean a power every exercise of which is indefeasibly justified in contemplation of right reason. What he means is, approximately, a power every exercise of which is at least prima facie morally justified, though that justification can be morally overridden. Crucially, for Suárez, the power in question is not one of positive law’s making but exists independently of positive law. It is inherent in individuals and is in this sense pre-legal: somehow in the nature of things. This moral power is the “first foundation” and the “root” of the act’s legal validity, on Suárez’s account.¹⁴ It follows that

in order for an act to be valid, it is not necessary, strictly speaking, that it be supported or positively assisted by the superior’s will or by human law; it is enough that the latter do not oppose the act by invalidating it.¹⁵

Leges irritantes involve such opposition. Thus, their function is to take away a power the existence of which they presuppose.¹⁶ Such removal of moral power is compatible with natural law, Suárez is quick to add. Indeed, it is for the sake of the common good, as he keeps repeating.¹⁷ For experience shows that this limitation of natural power is often necessary to prevent fraud and other kinds of inconvenience (“ad vitandas fraudes et alia incommoda”).¹⁸ A man may be deprived of his power to make contracts, “just as he may be deprived of his freedom.”¹⁹

Because that is the function of these laws, says Suárez, they are properly called leges irritantes: invalidating laws.²⁰ Why, how, when, and through what procedure different kinds of leges irritantes invalidate acts, and the extent to which relevant acts may remain binding in conscience, are the subjects of intricate discussion in the ensuing chapters.²¹

There is one point in this discussion worth flagging up at this stage. It is, in fact, a curious twist in Suárez’s account. As the preceding paragraphs suggest, Suárez’s explanation of leges irritantes focuses on the private law domain. His recurrent examples are exercises of private power, such as contracts, marriages, or wills. Yet, importantly, he deems leges irritantes to operate also in the domain of official acts, notably acts “such as a judgment, an election … and all public administration or administration done with a public authority.”²² He speaks relatively little to the distinction between these acts and exercises of private power,²³ though the distinction acquires relevance at one particular juncture, in the context of Suárez’s clarification of an interpretative rule.²⁴ He asks: where a law determines a form for valid acts, without specifying the consequences of failure to comply with the form, must the rule be presumed to prescribe the form on pain of invalidity? Suárez answers “No” in respect of exercises of private power, but “Yes” in respect of exercises of public power. The reason: acts of public officials “depend on a power conferred by the prince, by the state, or by public authority,”²⁵ so the entity that imposes a form is the very entity that confers the power in the first place.²⁶

Although Suárez does not pause to say so, this entails that his earlier claims about the pre-legal moral power to perform valid acts are true only of exercises of private power. For the power to perform public acts is conferred on officials, according to the above reasoning. It is conferred by the state or public authority: in a sense, by positive law. Thus, in this public law domain, as opposed to the private law domain, leges irritantes are themselves the source of the power they confer. They do not remove, but bestow, power. Suárez’s thinking seems to be that office-holders would not wield power if they had not been appointed in the first place. This is also in line with his view that the power to rule ultimately stems from God.²⁷

Writing in the mid-twentieth century, Hart’s primary target in articulating his account of legal powers is not an old four-fold Roman classification of laws but John Austin’s command theory, first published in 1832.²⁸ Central to the criticisms Hart levies at Austin in the initial chapters of The Concept of Law is the claim that laws conferring powers cannot, without distortion, be reduced to coercive orders.²⁹ In the process of showing the distortion of these laws within Austin’s model, Hart coins the category of “power-conferring rules” and frames that category in opposition to the category of “duty-imposing rules.”³⁰

Hart’s case for the distinctness of power-conferring rules has a structure similar to Suárez’s case for the distinctness of leges irritantes. Hart also distinguishes nullity from sanction.³¹ He does so as against Austin, and with regard to both private and public powers. Hart helpfully notes that, while a duty-imposing rule can exist without a sanction, a power-conferring rule cannot logically exist without nullity as the consequence of non-compliance.³² Hart goes on to challenge Kelsen’s view that requirements for the valid exercise of powers are mere fragments of sanction-imposing norms.³³ Both Austin’s and Kelsen’s theories, says Hart, purchase uniformity at the price of distortion.³⁴ What is distorted is the relevant “social function” of these rules. By this, Hart means the ways the rules are “thought of, spoken of, and used in social life.”³⁵ According to Hart, the social function of power-conferring rules is to

provide individuals with facilities for realizing their wishes, by conferring legal powers upon them to create, by certain specified procedures and subject to certain conditions, structures of rights and duties within the coercive framework of the law.³⁶

So, not unlike Suárez, Hart gives these laws a name that is closely tied to what he understands to be their point or function.³⁷ Numerous passages in Hart’s discussion explain that the point of these rules involves giving power, rather limiting it or taking it away. Hart goes so far as to suggest that these rules bestow on persons an amenity they would otherwise lack altogether—even in the private domain:

The power … conferred on individuals to mould their legal relations with others by contracts, wills, marriages, &c., is one of the great contributions of law to social life.³⁸

If such rules of this distinctive kind did not exist we should lack some of the most familiar concepts of social life … [T]here could be no buying, selling, gifts, wills, or marriages if there were no power conferring rules.³⁹

Possession of these legal powers makes of the private citizen, who, if there were no such rules, would be a mere duty-bearer, a private legislator.⁴⁰

Consistently with this, when Hart lists the kinds of social rules one finds in a primitive or pre-legal society, he does not include moral or social powers among them, but only “restrictions on the free use of violence, theft, and deception.”⁴¹ He even makes the startling claim that such a society lives exclusively by “primary rules of obligation.”⁴² It follows that there are no social or moral powers whose existence private power-conferring rules presuppose. In Hart’s picture, these legal rules are purely and radically power-conferring.

This picture is reinforced by Hart’s association of legal powers with secondary rules, and of the latter, in turn, with the domain of public law. While earlier chapters of Hart’s book discussed private powers in critiquing Austin’s doctrines, by the time Hart launches his “fresh start” in chapter 5, his main interest lies in rules that create and empower an official class. The secondary rules he presents as “remedies” to the “defects” of a primitive society chiefly concern the workings of public institutions with capacities of legislative creation and change, adjudication, and enforcement.⁴³ He even defines a primitive society as one “without a legislature, courts, or officials of any kind.”⁴⁴ Moreover, Hart introduces the distinction between primary and secondary rules, in whose union lies the “key to the science of jurisprudence,” in terms of the distinction between duty-imposing and power-conferring rules.⁴⁵ On this reading, all power-conferring rules are secondary, and all secondary rules are power-conferring.⁴⁶ Taken together, this strongly suggests that these rules, being the tools for crafting public institutions on a clean slate, are sources of a wholly newfound power.

II. INHERENT POWER

The comparison between Suárez’s and Hart’s accounts is intriguing. In some ways, the accounts are opposed. Then again, they are opposed symmetrically. There are important parallels between them; so much so that the differences lay claim to our attention.

It would be a mistake to play down the differences by circumscribing each account to the domain of law to which it pays most attention. Suárez is primarily concerned with private law; so much so that, as we have seen, his account becomes somewhat inconsistent when he projects it onto public powers. Hart at many points writes with public law in mind; his account’s description of powers in the private domain reaches conclusions that are to some extent unpalatable. It might, therefore, be tempting to see in Suárez a theory of private powers, and one of public powers in Hart. I suspect that this compartmentalized reading of each work is the most common reading. The scarce literature on Suárez’s leges irritantes of which I am aware portrays only their implications in the private law domain.⁴⁷ Although one can hardly generalize about the copious literature on Hart’s account, readers may be familiar, as I am, with a simplistic reading of Hart that associates legal powers with secondary rules, and secondary rules, in turn, with the existence of law-making and law-applying official institutions.

But such a compartmentalized reading is untrue to the purposes of both authors. Each intends his account to apply to legal powers in general. We should not underrate the theoretical intentions of authors who have so greatly influenced our legal tradition. Nor, in any case, should we be keen on accounts of legal power perched on the private–public divide, when we witness daily how that divide becomes increasingly blurred, if not overthrown, in our legal systems. If neither Suárez nor Hart wishes to put the brakes on at that line, still less should we seek to do so.

The disagreement between Suárez and Hart must be taken seriously. Their views on legal power ought to be tested against one another (Who is right?). To that end, we need to look into the discrepancies between their views in greater detail (What exactly do they disagree about?).

Plainly, one of the moot points between Suárez and Hart concerns the origin of the power that we exercise in performing acts-in-the-law: do the relevant legal rules deal with a power that is inherent in individuals, or with one that is wholly of legal making? According to Suárez, at least in private law, these legal rules remove a pre-legal power. According to Hart, these rules confer on us a power we could not otherwise wield.

Now, there is doubtless good sense in Suárez’s insistence that persons have the ability to enter commitments, form associations, bind themselves, and otherwise shape their relations with others, and that this ability is at least to a significant extent independent of legal regulation. Even Hart’s discussion of secondary rules of change notes in passing that “an elementary form of power-conferring rule … underlies the moral institution of a promise.”⁴⁸ But the moral or pre-legal power Suárez refers to plausibly extends well beyond the capacity to promise, narrowly understood, and embraces, for example, the development of partnerships, the formation of associations, and the giving of consent in many spheres. Hart’s claim that pre-legal communities lack social rules of a power-conferring sort, and his suggestion that powers enter our world freshly through law, misconstrues the make-up not only of “primitive” communities but also of his and our own world—much of which is thus pre-legal. I think it is uncontroversial that social and moral rules determine countless forms of relationships and undertakings, many of which should not—and even cannot—be the law’s business.

But Suárez’s account is misguided insofar as it claims that leges irritantes remove that moral power. His point, recall, is that acts such as contracts, marriages, or gifts are naturally valid, unless positive laws invalidate them. Positive law’s invalidation removes the acts’ natural validity, he says. But it is hard to see how this categorical claim holds true, even on Suárez’s own terms. Surely a legally defective contract or will, though possibly not recognized in law, may retain moral binding force where the ground of invalidity is not the breach of a justified prohibition but, say, a defect of procedure or form. Suárez himself notes as much, when he gives the example of a will that has not been executed following the relevant solemnities.⁴⁹ That will, he says, cannot serve as evidence at trial or ground a legal claim, but whoever is named heir under it may justifiably keep the inheritance until, if ever, he is legally challenged. At this point, remarkably, Suárez concedes that the lex irritans does not impede the will’s “natural validity.”⁵⁰ As he puts it, the lex irritans only annuls the act externally (“in foro exteriori”), but not absolutely or “in conscience” (“in conscientia”). In more modern terminology, we might say that the named heir has moral rights as against everyone, except the legally recognized heir. Only if and when the latter sues shall the faulty will be morally and legally overridden.

It follows that the potential constraint on the testator’s inherent moral power lies not in the law’s non-recognition of the faulty will. For that recognition would not exist either, if there were no laws regulating the matter. The significant constraint lies in the will’s susceptibility to be overridden, through legal enforcement of a rival, legally valid title. But that limit—susceptibility to be overridden, as opposed to non-recognition—is at best contingent and indirect. It depends on a rival legal claim’s existing and being brought successfully. Again, Suárez hints at this when he writes, elsewhere, that laws that determine the form of transactions only invalidate the relevant acts “indirectly.”⁵¹ As Suárez is aware, the law could instead choose to curb the testator’s inherent power to make the faulty will in a more direct manner: it could seek to forestall the very exercise of that power by making it a punishable offence to craft a will of a certain description.⁵² One finds legal regulations of this sort, says Suárez, when it comes to a contract involving usury,⁵³ a consanguineous marriage,⁵⁴ or a marriage “with a machinating adulterous woman.”⁵⁵ Further examples might include the sale of stolen goods (other than to good faith third parties) or, in a more modern spirit, exploitative work arrangements. The relevant limit in these cases, the limit that involves forestalling the wrongful act, does not arise from a lex irritans. It arises from a duty-imposing (sanctioning) rule, which Suárez rightly takes pains to distinguish from the former. In short: leges irritantes neither directly nor necessarily limit an inherent moral power.

What they certainly do limit in every case is a power that only law could bestow on individuals to begin with. That is the power of which Hart speaks. It is the facility to make institutionally enforceable, widely recognizable arrangements in the currency of positive law. Suárez, as we have seen, refers to such law-conferred power when he discusses acts by public officials. He says that leges irritantes regulating those acts limit a power that these laws themselves confer. In this context, he altogether omits to speak of a moral, pre-legal power. But he does not try to qualify his claims about the private law domain in light of the insights gained from the public law discussion.

We must conclude that both Suárez and Hart are partly right on the question whether laws setting criteria of validity deal with an inherent power. More importantly, though, they are also both wrong, insofar as they suggest that the question must be answered in all-or-nothing terms. I doubt that much progress can be made in grasping the workings or the point of these laws by asking whether they either remove a pre-legal power or confer a power of legal making. I am sure that even Suárez would agree that many sound private law regimes bolster an inherent moral power, rather than merely curbing it in case of non-compliance. But even in the domain of public law, it is plausible to see pre-legal or inherent powers in play. Appeals to constituent power, for example, invoke just this kind of pre-legal ability. Of course, exercising public powers in the ranks of a state’s administration requires a prior calling and appointment. Moreover, it essentially involves a promotion of interests different from the office-holder’s personal interests. In this sense, we may say that the power they exercise is not focally inherent in them. Yet some of these features are found also in the powers exercised by agents, trustees, and corporations, which are traditionally considered in private law terms.

The helpful teaching from Suárez—the important contrast with Hart, from which jurisprudence can learn—is not to be found in Suárez’s categorical claim that leges irritantes remove an inherent power. Perhaps his insistence thereon, along with his wavering on the point, goes to explain why modern scholars have ignored his account of leges irritantes. But that claim, that question, is a distraction from what is both true and important in his approach. This comes into view once we sideline the concern with the origin or source of the power.

III. LIMITS ON LEGAL POWER

Some acts involve the exercise of legal power; some do not. Some human acts are acts-in-the-law; some are not. That is a logical implication of the existence of criteria or conditions for exercising legal power. The laws in which both Suárez and Hart are interested are laws that set such criteria or conditions. Compliance with these laws, with these criteria, yields or secures legal validity; lack of compliance does not. That general truth, which is common ground between Suárez and Hart, is independent of whether one thinks invalidity takes away an inherent power, or simply fails to buttress it with the law’s authority. It is independent of whether one thinks legal power is the law’s enhancement of an existing moral capacity, or a competence freshly created by a legal regime. Legal power regimes validate certain acts hand in hand with invalidating others—they cannot but do so. One can look at these regimes from either side: positively, as enabling the exercise of legal power, or negatively, as limiting the exercise of legal power. The distinction between these two aspects might seem trivial, and the choice between them banal. But it is neither.

This section explains why. I show how Suárez’s focus on the negative aspect follows in the steps of a long-standing tradition that has forged the basic tenets of Western legal thought. I then canvass how Hart’s immediate predecessors Bentham, Austin, and Kelsen adopt a different focus, and how that change of focus is substantially inherited by Hart. Articulating the extent to which Hart’s account, as well as more recent work on this topic, eludes discussion of the limits on legal power serves to highlight lines of enquiry that have thereby been left unaddressed.

Suárez’s primary interest lies in the rules that regulate acts-in-the-law insofar as these rules invalidate non-compliant acts. That interest is manifest not only in the name he gives these laws but also in the main subject-matters of his discussion. Core parts of his explanation of these laws problematize the notion of nullity. He also draws a resourceful picture of the procedural implications of an act’s being voidable, as opposed to being void ab initio—and of the possible variables and implications of this.⁵⁶ I shall say more about this in section IV.

In so delving into the limiting, invalidating aspect of these regimes, Suárez pursues an approach that long predates him. He stands on the shoulders of the received legal tradition, which goes back to classical Roman law. As is natural for a seventeenth-century legal scholar, Suárez sees himself as contributing to the ius commune, a Europe-wide blend of law and legal doctrine developed largely through engagement with Roman legal sources.⁵⁷ That tradition’s concern with the invalidity of acts had steadily developed over the preceding millennium and a half.⁵⁸ Indeed, Roman law sows the first seeds of the idea we now call “legal validity,” in very inchoate form, by way of devising strategies to undo or stop the effects of transactions that would otherwise be valid under the ancient provisions of the Roman ius civile. Those ancient rules provide formal, ritualistic ways to enter transactions, not free from magical overtones. Once formally performed, such transactions (for example, a gift or a will) can hardly be cancelled in contemplation of the ius civile. This is why later Roman statutes, buttressed by the creative (“judicial”) work of praetors and law teachers, work out legal avenues for undoing what in practice needs reversal or quashing, but which ancient law would doom to stand. Roman statutory contributions and doctrinal reasoning in this field therefore apply themselves to the question how to limit an act’s sphere of authority, as opposed to how to confer authority on the act in the first place. Their focal concern is not with curbing a non-legal power: the baseline ius civile rules are very much legal.

So, the concern with the limits on legal power is certainly not Suárez’s innovation. Still, his work is pioneering, in that it spells out and probes the central concepts involved: for example, by analytically contrasting invalidity with prohibition. Nor does theorizing about invalidity end with Suárez. Over two centuries after him, Friedrich Carl von Savigny lays the foundations of contemporary private law by theorizing, inter alia, invalidity (Ungültigkeit) and its causes, types, remedies, and consequences in countless kinds of legal transactions (Rechtsgeschäfte)—itself a general notion that is his brainchild.⁵⁹

But legal culture’s spirit is changing direction even by Savigny’s time. While Savigny warns of the dangers of modern codes’ sweeping away customary law by legislative fiat, Jeremy Bentham, some 30 years older than Savigny, hails the merits of legislation as an expression of the sovereign’s power, a power he portrays as legally unlimited because grounded in a habit of obedience to credible threats.⁶⁰ Among Bentham’s many salutary contributions to legal thought is an explicit mapping of different kinds of exercises of legal power, both general and particular, in both public and private law.⁶¹ But, consistent with his view of sovereign power, the legal powers exercised by individuals and institutions are not genuinely theirs, on his account. Bentham explains the nature of these powers in terms of the sovereign’s “allowing” certain persons to participate in the task of “imperation.”⁶² Since “imperation” is a task special to the sovereign, others may discharge it only by his concession. Even adding oneself to the class of persons henceforth subject to existing “mandates,” such as contractual or marriage duties, requires sovereign permission, because it involves supplementing those “mandates.” It follows that, for Bentham, legal powers are mere permissions to participate in the exercise of a power that is really still the sovereign’s, and for that reason unlimited. Limits on legal power, allocations and restrictions of power among those exercising it, causes of invalidity, and remedies thereto—are all unknown to the core Benthamite apparatus. As is well known, his disciple John Austin follows suit, further elaborating the lack of legal limits on the sovereign and the implications of regarding all laws as orders backed by threats.⁶³

About a century after Austin, Kelsen champions a view that is, in important respects, very different. It is the view that laws and acts are not rendered valid simply by the fiat of somebody’s will or order but, instead, through compliance with a norm that is valid itself—ultimately in virtue of the system’s “basic norm,” which legal thought necessarily takes for granted.⁶⁴ Kelsen’s work, widely influential in continental and Anglo-American legal thought, articulates central elements of what we would today call constitutionalist thinking. It is no coincidence that Kelsen’s intellectual origins feature an academic environment dedicated to the then-bourgeoning theory of public law, which treats the state, perhaps for the first time, as an object of juristic analysis. In various ways, that theory grapples with the puzzle that the state creates the very law that governs the state. It sees in the state’s underived, self-governing power the culmination of the ideal of the Rechtsstaat or Rule of Law.⁶⁵ Kelsen himself portrays the state as the personification of the legal order.⁶⁶ His is one of the clearest expressions of the nineteenth- and twentieth-century scholarly zeal to explain the legal order as a comprehensively, even exhaustively, self-regulating system.⁶⁷

Despite the important differences between Kelsen’s theory and the Bentham–Austin apparatus, Kelsen, too, is circumspect about invalidity and limits on legal power. The ultimate reason for this lies in Kelsen’s postulate that there can be no normative conflicts within a legal order.⁶⁸ This postulate is associated with his peculiar understanding of law’s self-regulation. Kelsen famously struggles throughout his career to reconcile this “no conflicts” postulate with the familiar reality of legally defective transactions or official acts that have legal consequences of various sorts—until withdrawn or officially cancelled, with either retroactive or prospective effect. Apparently, at least, a statute, official resolution, or private transaction that does not adequately comply with relevant validity criteria introduces a conflict in the legal order. Apparently, at least, the defective act contradicts the empowering legal regime.

Kelsen’s considered solution for reconciling this reality with the “no conflicts” postulate is not, as it might have been, to deny that the defective act is valid in the first place. It is very much valid, he insists, precisely because it is liable to be annulled by a later official pronouncement:

There must be something legally existing to which this [annulling official] decision refers … Just as everything King Midas touched turned into gold, everything to which the law refers becomes law, i.e. something legally existing. The case of absolute nullity lies beyond the law.⁶⁹

This is a striking claim. It has a striking upshot: the empowering regime, says Kelsen, must be understood to have an implicit alternative character. It empowers persons to make either compliant or non-compliant acts.⁷⁰ The regime’s “direct prescriptions”⁷¹—what we would normally call its validity criteria—are, legally speaking, optional. They may or may not be complied with, compatibly with law. Either way, the result is valid. The law, however, has a “preference” for compliance, Kelsen adds.⁷² The preference is manifested in the fact that a special organ is empowered to annul a non-compliant act. Unless and until this occurs, however, the act is fully consistent—not in conflict—with the empowering regime. It follows that, strictly speaking, on Kelsen’s account, there are no legal limits on legal power. There are only valid acts liable to be expunged from the system by a judge’s say-so.

Incidentally, this reading is confirmed by Kelsen’s very definition of legal power (Rechtsmacht). Here, he begins by noting that, in a broad sense, “any human behavior, determined by the legal order as a condition or consequence, may be regarded as being ‘empowered’ by the legal order.”⁷³ That of course includes even the commission of a crime. Since Kelsen wishes to exclude the latter, he adds that “true legal power” “impl[ies] approval” by the legal order.⁷⁴ But this narrower definition is still overinclusive, as Kelsen realizes.⁷⁵ It catches not only acts-in-the-law, such as contracts, wills, or legislation, but also any acts that simply fulfil a legal duty and thereby extinguish it. Yet Kelsen does not further elaborate on the difference between the law’s lack of “approval” in the case of invalidity and in the case of breach of duty. The loose idea of “approval,” like the idea of “preference,” does not come anywhere close to spelling out what is distinctive about legal limits on legal power. Invalidity is glossed over in silence.

Hart’s thought is deeply influenced by Bentham, Austin, and Kelsen (among others). Hart devotes much energy to challenging some of their views, precisely because he finds worthwhile insights in them. He is particularly explicit about this regarding Austin and Bentham,⁷⁶ but Kelsen’s influence on Hart’s account of the structure of the legal system is unmistakable, notably as regards the notions of legal validity, of rules about rules, and of the system’s fundamental rule determining validity criteria.⁷⁷ To be sure, Hart makes ground-breaking critiques of his predecessors’ accounts. He soundly debunks the Bentham–Austin view of the legally unregulated sovereign,⁷⁸ and puts a subtle account of legal powers in place of their reductionist understandings.⁷⁹ He also spells out the benefits of thinking of the fundamental rule as a social rule rather than as a Kelsenian logical hypothesis.⁸⁰ He, moreover, objects that Kelsen’s account of “complete” sanction-stipulating norms unduly reduces power-conferring rules to conditions for the imposition of sanctions.⁸¹

Still, to a significant extent, Hart partakes of his predecessors’ focus on acts that are within a sphere of authority, as opposed to outside it. As already noted, one of Hart’s core concerns in chapter 5 of The Concept of Law is the description of secondary rules that confer on persons a public power that they would otherwise lack. Then, in chapter 6, he dwells on the nature of the supreme rule that sets the ultimate conditions for the exercise of those powers and enables them to be recognized.⁸² Later in that chapter, under the heading of “pathology and embryology of legal systems,”⁸³ Hart discusses large-scale changes in the “complex congruent practice”⁸⁴ that underpins a legal order. But he does not also discuss individual acts that, pathologically, lack the legal warrant they purport to enjoy from an otherwise stable system.

This is not to suggest that there are no resources in Hart’s theory to make sense of legal limits on legal power. Quite the contrary. The very fact that, on his account, unlike in Bentham’s and Austin’s, even the highest official authority is rule-governed entails that any legal power, even the sovereign’s, can be legally restrained. Hart devotes a few illuminating passages to this last possibility, when he discusses how a sovereign legislature could set limits on its own powers, and the role of courts in monitoring such limits.⁸⁵ But this discussion occurs at the very end of chapter 7, after Hart has concluded his main critique of rule-skepticism, which takes pride of place in that chapter. The relevant discussion also bears the equivocal heading of “Uncertainty in the Rule of Recognition,” even though it concerns who can authoritatively settle doubts about the latter by way of restraining a legislature’s competence. To be sure, earlier in chapter 7, Hart does note the existence of legally mistaken yet final judgments.⁸⁶ But he does not delve into legal techniques and reasons for limiting official discretion, or into the legal implications of that discretion’s being overstepped. His famous analogy with a game overseen by a single “scorer” (“scorer’s discretion”)⁸⁷ directs the reader’s attention away from the delegation and allocation of powers among different decision-making institutions in the framework of a legal system.

On the other hand, Hart, unlike Kelsen, does not embrace the chimeric postulate that there cannot be conflicts between a system’s rules. Hart’s claim that, at a legal system’s foundations, one finds a social rule, rather than a logical presupposition that precludes contradictory mandates, in principle equips him to explore the nature and implications of clashes between legal provisions. Yet Hart’s discussion of the different types of legal rules—primary and secondary—and their internal relations mostly homes in on the ways in which they complement and support each other. Rules in conflict—indeed rules invalidated and invalidating—linger in an unstated background as Hart paints the basic picture of his legal world.

The neglect of limits on legal power is a mark common to much jurisprudential work that has unfolded in the wake of Hart’s. I think it is fair to say that Hart’s sparse discussions of those limits are not among the passages of his book that have attracted extensive scholarly interest. Some helpful elaborations of Hartian power-conferring rules have, in various ways, drawn on philosophical literature concerning the social rules that make up “social reality,” as John Searle famously puts it.⁸⁸ According to Searle, underlying realities such as money, language, or football are socially practiced rules that have the logical structure “X counts as Y in C.”⁸⁹ Searle dubs these rules “constitutive” because, he says, they create new forms of behavior. His contrast between “constitutive” and “regulative” rules runs in many ways parallel to Hart’s power–duty distinction. By casting legal power regimes as “constitutive” rules, in the sense of some of this philosophical literature, progress has been made in clarifying the architecture of these legal rules and their relationships with rules of other types.⁹⁰ But the focus on their constitutive—that is, positively enabling—aspect typically displaces concern with their negative—invalidating—aspect. Furthermore, giving attention to social rules is liable to cloud the workings of institutionalized settings with various layers of authority and chains of powers.

Joseph Raz’s conception of law’s authority, in turn, is a prominent portrayal of normative power in the field of jurisprudence.⁹¹ Raz argues that every legal system makes a claim to legitimate authority, which is, in essence, a power to change the normative positions (i.e., the rights, duties, powers, etc.) of the authority’s subjects by saying so. According to Raz, law claims authority of an unlimited or at least unspecified scope,⁹² although the considerations that bear on the legitimacy or moral jurisdiction of any practical authority may justify denying that the law has all the authority it claims. Raz’s rich account encourages reflection on the moral standing of legal directives, the logical structure of legal duties, and the nature of the sources of law, among many other matters. But one does not find in the foreground of his account, or in much of the literature that builds on it, an alertness to the relations among authorities within a legal order, indeed to the sharing and circumscription of their jurisdictions, both moral and legal. Blanket references to what “the authority” commands its “subjects,” though perhaps a usefully simplifying abstraction, are at best an elliptical allusion to the network of powers wielded by all of us, official and lay people, on a daily and shifting basis. By likening the law’s authority to a national legislature’s (nearly) plenary powers, as Raz and others sometimes do,⁹³ one inevitably plays down countless sorts of morally warranted legal limits on law-making and law-changing powers, legal limits that are the bread-and-butter of major areas of legal practice.

IV. LESSONS FROM SUÁREZ

The questions our recent tradition has overlooked are particularly pressing today. This section demonstrates how we can make progress on these questions by garnering insights from Suárez’s neglected approach. It identifies three valuable lessons from Suárez and shows how they can help us overcome shortcomings of currently more popular accounts. These lessons concern, respectively, the reasons for limiting legal power, the procedure for doing so, and the limits on the limits.

The first lesson to be drawn from Suárez’s account is its sensitivity to the reasons for limiting legal power. As we have seen, Suárez repeatedly refers to a moral power that may require constraint for the sake of the common good. Since it is a moral power that leges irritantes purport to remove, there must consequently be moral reasons for that removal. Suárez’s extensive discussion contemplates many different scenarios of limitation and, in so doing, it points to a range of different considerations of justice that may support limiting a power.

Scholars working in our field today tend to appreciate that sound conferrals of legal power are responsive to reasons why someone ought, in justice, to decide about something. It is easy to forget, however, that this is often the flipside of the fact that someone else ought, in justice, to be precluded from deciding that very matter, at least at that very time. A broadly Hartian defense of the social functions of secondary rules only draws attention to the general need for someone to discharge the powers of legal creation, change, and application. It does not highlight the equally important need for it to be different people who exercise the different powers involved, in ways that do not encroach on one another’s decision-making or overstep the limits of justifiable collaboration.

But such considerations are key to some of the more intricate puzzles that current legal reasoning wrestles with. Take public law. Doctrines of subsidiarity, the federal division of powers, constitutional rigidity and entrenchment clauses, the scope of delegated legislation, the demarcation of the jurisdiction of courts and other organs, and the role of fundamental rights as constitutional constraints on democratic rule all involve a circumscription of power, for reasons that go well beyond convenience or expediency. The reasons range from relative expertise or proximity to facts, all the way to political responsibility, requirements of the rule of law, and collective self-direction and its compatibility with human rights.⁹⁴ Many problems in contemporary private law similarly involve moral limits on someone’s valuable autonomy. Consider, for instance, the scope of implied terms or statutory rights in consumer or work contracts, restrictions on the powers to marry or to disinherit, asset allocation in bankruptcy, minimum capital for company incorporation, or the preclusion of certain arrangements by antitrust law.

Sound limits on these and other legal powers are applications of the basic truth that, in order for me to lead a valuably autonomous life, I do not need unrestricted freedom to do or forbear as I happen to fancy. It is part of an autonomous life that I can restrict my own ability to choose, by entering commitments, launching enterprises, and forging durable, binding relations with others. Nor is my valuable autonomy diminished, curtailed, or taken away in any meaningful sense when I am prevented from acting in an unjust manner. Political philosophy has had much to say on these questions, for example, under the guise of the harm principle and by reference to the moral limits of law. But these debates have insufficiently trickled down into the study of the allocation of legal powers.

As I have argued in section II, Suárez’s categorical claim that all leges irritantes remove a moral power is not representative of the varied interplay of moral and legal powers actually depicted in his account. For example, Suárez carefully draws attention to scenarios where limits on the legal validity of an act leave its “natural validity” intact. He also notes the existence of legal limits that are, in fact, for the benefit of the very person whose power is thereby constrained.⁹⁵ His example of a law invalidating a married woman’s waiver of her fatherly inheritance sounds antiquated,⁹⁶ but not so many others we can identify or propose in our systems. Rules that seek to even out the bargaining power of contractual parties—say, by preventing a worker from signing up to a job arrangement without minimum pay, or a consumer from waiving the right to return faulty products—in an important sense enhance the moral power of the worker or consumer precisely by limiting what he can legally commit to. The converse possibility is not far-fetched either. It can be argued, for instance, that legally empowering sex workers to enter labor law relations by legalizing prostitution may socially normalize what in truth is a harmful objectification of their bodies and thereby impair these persons’ capacity to self-determine as moral agents.⁹⁷ In short, paying attention to the reasons for limiting legal power opens up fields of enquiry that received approaches to legal power are not geared-up to traverse.

The second lesson I propose to take from Suárez concerns the procedure for limiting legal power. By this, I mean the legal method for giving effect to the relevant legal limits, and thus to the reasons for imposing them in the first place. Suárez’s emphasis on invalidation showcases a set of fascinating problems. Any lawyer is aware of the troubling duality that may arise when apparently valid acts do not fully comply with relevant validity criteria. This can be true, for instance, of contracts, wills, and appointments, all the way to administrative resolutions, judgments, and statutes. Defective acts may nevertheless exert legal effects. This duality does not come into view when our focus is on acts that lie within a sphere of authority, and thus on the validating, power-conferring aspect of the relevant rules. Only fully compliant acts warrant all the legal effects they are legally recognized as having. The sense and the way in which non-compliant acts are invalidated has long been overlooked in general jurisprudence.

We have seen how Kelsen evades this problem by defining the duality out of existence. In a move that smacks of legal realism—a line of thought he is otherwise intent on disparaging—he claims that apparently non-compliant acts in truth do comply with the regime’s implicit alternative prescription. But criteria that allow everything are not criteria. If we press Kelsen’s move far enough, the very possibility of legal reasoning dissolves. Suárez’s premise, instead, is that laws determining validity criteria are irritantes: they invalidate non-compliant acts. How they do so, however, varies between different kinds of leges irritantes.

To this topic, Suárez devotes three full chapters (roughly 10,000 words) of his book 5.⁹⁸ His discussion articulates distinctions now well-known to civil and administrative law doctrine. He distinguishes, for example, between leges irritantes that invalidate the act “immediately by the law itself at the moment the act is realized or attempted,”⁹⁹ and leges irritantes that invalidate the act by providing for a judge to annul it.¹⁰⁰ In the former case, judicial declaration of nullity has retroactive effect; in the latter case, the act is voidable until a judge annuls or confirms it.¹⁰¹ What is distinctive of Suárez’s approach, as compared to modern doctrine in relevant legal areas, is not only his generalizing aim—he writes about leges irritantes in any legal area—but his open insistence that it is the moral purpose of a given type of invalidation that ought to determine how, when, and by whom it is to be effected. He extensively reflects, for instance, on the appropriate legal effect of conduct that either justly or unjustly impedes an invalidating pronouncement,¹⁰² on whether there is room for equity in declaring voidness ab initio,¹⁰³ and on the moral status of the void ab initio act, which he says might in certain circumstances give rise to legally condoned moral rights until the judge exercises his power to totally annul the act and thus suppress its natural obligation.¹⁰⁴

I turn to the third lesson. Sometimes, if a legal challenge is not brought, or is not brought within a certain time, the defective act can no longer be invalidated. That is a limit on invalidation. But there are others. Such limits on the limits exist where the constraints on the power to review an act do not match the constraints on the original power-holder. An obvious case is that of judgments of courts of final appeal, whose pronouncements are accordingly conclusive: res judicata. But, often, the legal power of review or appeal is limited only in part. Thus, a higher court may be able to correct mistakes of law but not mistakes of fact in the lower court’s decision. Non-justiciable constraints on a constitutional power include those bearing on the grounds for impeaching the US president.¹⁰⁵ Administrative law provides for degrees of judicial deference to executive authorities, as a requirement of comity. The European Court of Human Rights, not unlike other supranational courts, gives public authorities in contracting states a margin of appreciation when it comes to implementing certain legally binding provisions.

These predicaments raise legal and moral puzzles to which the received approach to legal powers largely closes its eyes. On Kelsen’s core view, if an error is not, or cannot be, corrected, it is not an error after all. Especially where an act is indefeasible, questions about its legal correctness cannot meaningfully arise. Kelsen seems to follow the logic that two minuses make a plus. But invalidity is not arithmetic. As Hart might rightly say, this confuses finality with infallibility.¹⁰⁶ Unfortunately, Hart himself does not endeavor to generalize about the considerations bearing on appropriate limits on powers of review, or on how best to conceive of and to handle indefeasible ultra vires decisions.

We may shed light on these questions, following Suárez’s lead, by going back to the moral reasons for allocating the original power in the first place. But the answer is not straightforward. In one sense, doing justice to those reasons requires reviewing courts to second-guess whether the power-holder truly lived up to them. We find echoes of this position in doctrines of administrative law according to which courts may rule an authority’s use of discretion unlawful on grounds that the authority did not use its power to advance the purposes for which it was bestowed.¹⁰⁷ Yet, in another sense, such reviewing power undermines the very reasons it is supposed to honor. For the point of allocating the decision-making power to certain persons and not to others may be defeated if courts are allowed to reopen the very questions of justice that original choosers were meant to settle.

The preceding sentence echoes a broadly similar claim advanced by Raz.¹⁰⁸ According to Raz, it follows from law’s claim to authority that legal requirements are identifiable “exclusively” by reference to facts describable in morally neutral terms.¹⁰⁹ Law would not be capable of having authority, Raz reasons, if discerning the content of its directives required resort to the moral considerations on which the authority is there to pronounce. Note, however, that Raz’s claim concerns the identification of law, not the proper limits of review. His claim speaks to the relationship between an authority and its subjects, rather than to the relationship among authorities, that is, among those reviewing and reviewed, all of whom purport to be exercising a normative power under law.¹¹⁰ Moreover, Raz’s claim is unduly absolutist: it fails to quantify (note the term “exclusively”). This sits uneasily with basic premises of legal reasoning, as I have argued elsewhere,¹¹¹ and blinds itself to differences between moral and non-moral validity criteria that are central to current debates on, for example, judicial review. Instead, the apparatus Suárez offers probes the limits on the limits in terms of these very differences.

V. CONCLUSION

Suárez’s account of legal power is far from systematic. None of the three lessons above corresponds to a tightly argued and clearly articulated set of propositions in his treatise. But the kinds of questions Suárez asks, and the ways he goes about answering them, are a cogent reminder that theoretical argument about limits on legal power need not content itself, as it all too often does, with a despondent pointer to the promiscuous variety of positive legal regimes.

Suárez’s approach is particularly needed in our age of multiple tiers of authorities—perhaps multiple legal orders—bearing on the status of legal acts and rules whose effects extend well beyond the sphere of the national state. It is relevant to the increased blurring of the private and public spheres. It may, indeed, be complex to theorize about the limits on legal power, to broach validity’s negative, disabling aspect. But it is worth remembering that this has been the standard approach for the larger part of our legal tradition.

Footnotes

Author notes