Hart on Legal Powers as Legal Competences

Abstract This paper first recapitulates the objections by H.L.A. Hart to the ways in which John Austin’s command model of law obfuscated the importance and the very existence of power-conferring laws. Although those objections are familiar in the world of contemporary legal philosophy, their insightfulness is highlighted here because they contrast so sharply with Hart’s own neglect of power-conferring laws at some key junctures in his theorizing. In the second half of this paper, I ponder a few of the junctures where Hart failed to heed the admonitions which he had so deftly leveled against Austin.

In much of the opening half of The Concept of Law, H.L.A. Hart set the stage for the elaboration of his own jurisprudential theory as he first dissected the model of law that had been propounded by the nineteenth-century jurist John Austin. 1 As virtually everyone among the ranks of present-day Anglophone legal philosophers is aware, one of the chief complaints about Austin by Hart was that the former theorist had disregarded and obscured the major role of power-conferring norms in the structures and operations of legal systems. A preliminary bit of clarification is needed here, however. Hart in The Concept of Law did not specify very precisely what he took a legal power to be. According to the influential analysis presented by the American jurist Wesley Hohfeld, a legal power consists in an ability to bring about some change(s) in legal relationships through the adoption of some course of conduct. 2 Under that broad conception of a legal power, miscreants who contravene legal requirements have thereby exercised powers to alter their own legal positions and the legal positions of certain other people such as law-enforcement officers; they have made themselves liable to undergo arrest or other measures of enforcement, and they have invested certain people with legal powers and legal liberties to resort to such measures. Now, although Hart knew Hohfeld's work well and admired many aspects of it, his own conception of legal powers was narrower than the Hohfeldian conception. He did not delimit the contours of his conception with any precision in The Concept of Law, but there are at least two apt ways of understanding those contours.
First, we can take Hart as confining the category of powers to Hohfeldian powers that are normally beneficial for the people who are endowed with them. In other words, being vested with a legal power (in the relevant sense) is normally better for a holder of it than is not being vested with it. Of course, Hart did not assume that the justification for a power-conferring law will always reside in the fact that the power conferred is typically beneficial for the people who hold it. Any such justificatory assumption would be particularly outlandish in connection with laws that confer public powers. Nevertheless, whatever the justification may be for a law that confers upon anyone a Hartian legal power, the possession of a power of that kind is typically advantageous for its possessor. Like any Hohfeldian legal power, a legal power in this circumscribed sense is correlated with a liability on the part of the power-holder or of someone else. A liability is a position of susceptibility or exposure to the effects of the exercise of a power by oneself or by someone else. The existence of a legal power entails the existence of a legal liability with the same content, and vice versa. (Liabilities in this Hohfeldian sense-in contrast with liabilities in an everyday sense-are by no means always or even characteristically detrimental for the people who bear them. For example, being liable to undergo the effects of the exercise of a donative power by a munificent and wealthy cousin is typically beneficial rather than typically detrimental.) A second way of construing Hart's conception of legal powers has been delineated by Visa Kurki, who has furnished the following explication of legal competences and acts-in-the-law: Legal competence: (1) A person X holds the competence C to effect the legal consequence r if and only if X can perform an act-in-the-law to bring about r.
(2) If X holds C, any act by X that effects r is an exercise of C.
Act-in-the-law: An act a, performed by X, constitutes an act-in-the-law if and only if (1) X performs a with the intention to bring about the legal consequence r, and (2) the fact that X has performed a in order to bring about r is an element of a set of actually occurrent conditions minimally sufficient for r. 3 Though Kurki is chiefly aiming to supply an independently solid exposition of the notion of legal competences, he maintains that his exposition captures the gist of Hart's conception of legal powers. His account is obviously different intensionally from the first way of construing Hart's conception which I have broached, but the two approaches are probably equivalent extensionally. At any rate, for the purposes of this paper, there is no need for me to choose between them-since, whether or not there is extensional equivalence between the two approaches across the full range of legal powers, there certainly is extensional equivalence between them in application to all the legal powers that are discussed herein. This paper will first recapitulate Hart's objections to the ways in which Austin's command model of law obfuscated the importance and the very existence of power-conferring laws. Although those objections are familiar in the world of contemporary legal philosophy, their insightfulness should be highlighted here because they contrast so sharply with Hart's own neglect of power-conferring laws at some key junctures in his theorizing. In the second half of this paper, I shall briefly ponder a few of those junctures where Hart failed to heed the admonitions which he had so deftly leveled against Austin.

Hart's Critique of Austin on the Matter of Power-Conferring Norms
Hart allowed that there is some resemblance between the mandates of an Austinian sovereign and the statutes of criminal law or tort law that impose legal duties on people. However, he submitted that there is no such resemblance between Austinian mandates and the sundry laws of any jurisdiction that confer legal powers on people. Among those power-conferring laws are norms that enable people to carry out private transactions, such as the formation of contracts or the conveyance of real estate or the donation of funds to charities. Also among the power-conferring laws are norms that authorize public officials to perform their functions of legislation or administration or adjudication. Such laws are integral to the very existence and operativeness of any legal system, yet they find no place in Austin's jurisprudential theory. In his theory, all laws are commands that impose requirements which are supported by threats of violence for disobedience. As Hart proclaimed, "there is a radical difference between rules conferring and defining the manner of exercise of legislative powers and the rules of criminal law, which at least resemble orders backed by threats." 4 Austin, like Jeremy Bentham, did leave room for liberty-conferring laws-as expressions of the sovereign's will that retract or modify the requirements imposed by previous commands-but the role of power-conferring laws is effaced by his theoretical schema. Hart, in his criticisms of Austin, showed time and again that the Austinian model of law presupposes the operativeness of power-conferring norms even while it fails to supply any account of their operativeness. (Worth noting here is that Hart should also have frowned upon Austin's disregard of immunity-conferring laws. An immunity, as explicated by Hohfeld, is a position of insusceptibility to the bringing about of some change in legal relations. For example, the First Amendment to the U.S. Constitution confers upon everyone in the American populace an immunity against being deprived of certain legal liberties by Congressional enactments. An Austinian command theory cannot generate an adequate account of immunity-conferring laws any more than of power-conferring laws.) Hart underscored the differences between duty-imposing laws and power-conferring laws in several ways. He observed for example that, whereas duty-imposing laws establish unconditional requirements, the requirements specified by a power-conferring law are conditional on someone's wishing to exercise the power that is conferred. While a duty-imposing law provides that some specified mode of conduct is "to be avoided or done by those to whom [the law] applies, irrespective of their wishes," a power-conferring law does "not require persons to act in certain ways whether they wish to or not." 5 To be sure, especially in the public sector, people are sometimes under legal duties to exercise legal powers with which they are endowed. In such circumstances, a person who holds one of those powers is legally required to follow the specified procedure for exercising it -whether or not that person wishes to do so. However, the categorical requirement is created not by the conferral of the power but instead by the imposition of the duty that accompanies the power. Even if the power and the duty are established by the same statute, the power-conferring component and the duty-imposing component of the statute are not equivalent.
A closely related way in which Hart marked the differences between dutyimposing laws and power-conferring laws is that, whereas the former laws normatively close off opportunities by prohibiting modes of conduct, the latter laws expand opportunities by presenting individuals with "facilities for realizing their wishes." 6 When drawing the contrast in this fashion, Hart particularly had in mind the opportunities presented to individuals to form private-law arrangements such as contracts, wills, trusts, marriages, and corporations. He further noted that noncompliance with the procedures specified for the creation of some such arrangement will nullify an effort to create it but will not constitute "a 'breach' or 'violation' of any obligation or duty nor an 'offence' and it would be confusing to think of [the noncompliance] in such terms." 7 Dissimilarities between duty-imposing norms and power-conferring norms are even more salient in public-law settings than in private-law settings. While recounting a host of laws that empower officials to engage in their legislative or administrative or adjudicative activities, 8 Hart drew on a legislative example to accentuate vividly the divergences between such laws and dutyimposing mandates: If a measure before a legislative body obtains the required majority of votes and is thus duly passed, the voters in favour of the measure have not 'obeyed' the law requiring a majority decision nor have those who voted against [the measure] either obeyed or disobeyed [that law]: the same is of course true if the measure fails to obtain the required majority and so no law is passed. 9 As Hart declared: "The radical difference in function between [a law that prescribes the majoritarian procedures and a law that imposes some duty] prevents the use here of the terminology appropriate to conduct in its relation to rules of the criminal law." 10 Hart's point here is again best understood with reference to the distinction between conditional requirements and unconditional requirements. Legislators who wish to vote on a proposed enactment will comply with the procedures for voting, and motorists who wish to avoid collisions and sanctions will comply with a legal mandate that forbids them to drive in a certain direction on a one-way street. However, whereas the verb phrase 'comply with' is appropriate both in application to conformity with conditional requirements (such as the procedures for voting) and in application to conformity with unconditional requirements (such as the mandate pertaining to a one-way street), the verb 'obey' is pertinent only in application to the latter. A legislator who does not wish to vote for the proposed enactment is in no way required to act in accordance with the procedures for doing so, but a motorist who wishes to drive in the wrong direction on the one-way street is nonetheless strictly required to abide by the mandate that forbids such conduct. Hence, the concept of obedience as submission to an unconditional requirement is applicable to the conduct of a motorist who refrains from driving in the wrong direction on a one-way street, but in any ordinary circumstances it is not applicable to what a legislator does in casting a vote for a proposed law.

Should Power-Conferring Laws be Reconstrued as Duty-Imposing Laws?
A defender of Austin might retort to Hart by contending that power-conferring laws are in fact duty-imposing laws. A defender so inclined would submit that, when something which appears to be a legal arrangement of some kind is deemed to be null or invalid because it does not comply with the legally specified conditions for the effecting of such an arrangement, the nullity or invalidity is a sanction administered in response to a contravention of a legal requirement. Hence, the defender would maintain, any law that specifies the procedures for forming a 9. Ibid at 31-32. 10. Ibid at 32. contract or some other legal arrangement is a duty-imposing law that carries the threat of a sanction for nonconformity.
Hart posited and cogently countered a rejoinder along these lines. He began by observing that a law which specifies the conditions for the exercise of a power is thereby supplying the normative structure of a certain practice rather than laying down some duties with sanctions to be imposed on people who engage in activities or endeavors which differ from that practice. He compellingly substantiated this first point with reference to legislative and game-playing contexts: Even more absurd is it to regard as a sanction the fact that a legislative measure, if it does not obtain the required majority, fails to attain the status of a law. To assimilate this fact to the sanctions of the criminal law would be like thinking of the scoring rules of a game as designed to eliminate all moves except the kicking of goals or the making of runs. This, if successful, would be the end of all games; yet only if we think of power-conferring rules as designed to make people behave in certain ways and as adding 'nullity' as a motive for obedience, can we assimilate such rules to orders backed by threats. 11 Persuasive though the examples from legislative and game-playing contexts are, Hart regrettably overstated this first point against the defenders of Austin. He should have acknowledged that nullity is sometimes functionally equivalent to a sanction that is designed to steer people away from certain modes of behavior. After all, there are undoubtedly some power-conferring laws that specify conditions in order to deter people from adopting certain courses of conduct. For example, when a law provides that an agreement will be invalid as a contract unless each party to it is above a specified age and is of sound mind, those two conditions specified as necessary for the exercise of a contract-forming power are doubtless prescribed partly in order to deter people from engaging in some exploitative modes of behavior; and the nullity resulting from noncompliance with those conditions is often functionally tantamount to a sanction. Hart could and should have granted as much, because all that he needed to maintain was that many power-exercising conditions-such as those that are operative in the legislative and game-playing contexts to which he referred-are markedly different from the conditions for the formation of contracts which I have broached here. Many power-exercising conditions are specified to supply the normative frameworks of various activities and enterprises, rather than to deter undesirable conduct.
At any rate, even more powerful is a second point advanced by Hart against the defenders of Austin. He noted that a duty-imposing mandate can intelligibly exist even if no sanctions are attached to it for noncompliance. Indeed, as I have discussed elsewhere, legal mandates that are utterly unenforceable rather than merely unenforced can nonetheless perform important conduct-directing functions. 12 By contrast, a power-conferring law without any specification of necessary conditions 11. Ibid at 34. 392 Kramer for the exercise of the conferred power would be unintelligibly vacuous, and therefore a power-conferring law that involves no prospect of nullity for noncompliance with conditions prescribed by that law would be unintelligibly vacuous-since one's noncompliance with necessary conditions for the exercise of a power entails the nullity of one's attempt to exert the power in question. As Hart declared, if failure to comply with [any] essential condition [for the exercise of a power] did not entail nullity, the [power-conferring] rule itself could not be intelligibly said to exist : : : even as a non-legal rule. The provision for nullity is part of this type of rule itself in a way which punishment attached to a rule imposing duties is not. If failure to get the ball between the posts did not mean the 'nullity' of not scoring, the scoring rules [and the game for which they supply the normative structure] could not be said to exist. 13

Another Reconstrual of Power-Conferring Laws as Duty-Imposing Laws
Neil MacCormick did not seek to uphold an Austinian command theory of law, nor did he rely at all on the thesis that nullity is tantamount to a sanction. Indeed, he robustly rejected that thesis. 14 Nevertheless, through a different route, he reconstrued power-conferring norms as duty-imposing norms. Having recounted and amplified Hart's exposition of a moral norm that confers powers on people to bind themselves by satisfying the conditions for the making of promises, MacCormick asserted that that norm "belongs to Hart's class of 'obligationimposing rules.' It tells us what one 'is bound' to do." 15 MacCormick added: "The obligation in question is, however, conditional. It is a condition of my coming under that [promissory] obligation to you that I have said the appropriate words in the appropriate circumstances, both of us being appropriate persons." 16 Slightly later, MacCormick characterized the promissory power-conferring norm as "a fairly simple 'obligation-imposing rule.'" 17 He reiterated that characterization several pages further on in his discussion, when he asserted that the promissory power-conferring norm is "an 'obligation-imposing rule'; the 'powers' [are] conferred by the rule because of the conditions set for determining when a person is 'bound' in virtue of the rule." 18 MacCormick averred that "the rule in question does not solely confer power nor solely impose obligation. It does both. By imposing an obligation which is conditional on the performance of a 'rule-invoking' act, it also confers a power." 19 As might be inferred from the flurry of cautionary quotation marks in these extracts-including the threefold instances of such quotation marks on the phrase 'obligation-imposing rules' or 'obligation-imposing rule'-MacCormick went astray in his conflation of power-conferring norms and duty-imposing norms. Contrary to what he presumed, the conditional obligation to which he referred is not an obligation at all. Rather, it is a Hohfeldian liability; specifically, it is a liability to incur an obligation. Let us recall that a legal or moral liability is a position of susceptibility to the effects of someone's exercise of a legal or moral power. In other words, a liability and a power are correlative positions, as the existence of either of them with a certain content entails the existence of the other with that same content. Thus the moral power of Jane to impose a promissory obligation on herself entails, and is entailed by, her moral liability to incur such an obligation through her exercise of that moral power. Pace MacCormick, the conditional obligation to which he referred is a liability of this very kind. Hence, pace MacCormick, the promissory norm which he recounted is not an obligationimposing norm. Instead, it is a norm which confers a power (one's power to impose promissory obligations on oneself through a specified procedure) and which establishes a correlative liability (one's liability to incur promissory obligations through one's adoption of that specified procedure). It is, in short, a power-conferring norm-exactly as Hart contended.
Because the existence of any power with a certain content entails the existence of a liability with that same content and vice versa, every power-conferring norm is also a liability-establishing norm. A norm cannot confer a power without also establishing a liability. Utterly unsurprising, then, is that the promissory powerconferring norm expounded by Hart is also a norm that establishes liabilities. Despite MacCormick's suggestions to the contrary, such a norm does not impose any obligations-though of course a successful exertion of the power conferred by it will impose a promissory obligation on the person who has performed that exertion.
A further reason for rejecting MacCormick's conflation of power-conferring norms and duty-imposing norms is that not all powers are powers to impose obligations. There can also be powers to create liberties, powers, or immunities. A power to create a liberty is correlated not with a liability to incur an obligation but instead with a liability to acquire a liberty. Likewise, of course, a power to create a power is correlated with a liability to acquire a power, and a power to create an immunity is correlated with a liability to acquire an immunity. Hence, a norm that confers a power of any of these kinds will not be establishing a conditional obligation (that is, a liability to incur an obligation). For example, when a norm confers a power on a person to invest somebody else with a certain immunity through a specified procedure, the norm has thereby established a conditional immunity (that is, a liability to acquire an immunity). Consequently, even if we were to commit MacCormick's error of perceiving conditional obligations as obligations, we should deny that all power-conferring norms are properly construable as obligation-imposing norms. We have heretofore examined a couple of misconceived attempts to reduce power-conferring laws to duty-imposing laws. We should now ponder the following proposition, which I will designate as the 'Subsumability Thesis': Instead of being reducible to duty-imposing laws, power-conferring laws are subsumable into duty-imposing laws as mere components or elements thereof.
In The Concept of Law, Hart chiefly mulled over a version of this thesis that had been propounded by Hans Kelsen, 20 whereas in some later essays he principally probed a milder version that had been developed by Bentham. 21 The Subsumability Thesis does not run afoul of the errors exposed hitherto. It depends neither on the classification of nullity as a sanction nor on MacCormick's conflation of duties and conditional duties. Indeed, Hart did not object to the thesis on logical or formal grounds; instead, as we shall see, he objected to it on the ground that it obscures the distinctive functions or social roles of power-conferring laws.
Though Hart in The Concept of Law concentrated predominantly on the extreme version of the Subsumability Thesis which he associated with Kelsen, he did also recount there the moderate version developed by Bentham. According to the moderate version, duty-imposing mandates such as statutes that prohibit theft and arson are full-blown laws that are addressed to citizens, whereas ostensible power-conferring laws are not full-blown laws but are instead specifications of sufficient conditions for the applicability of certain duty-imposing laws. Proponents of the moderate version of the Subsumability Thesis thus understand power-conferring legal norms as parts or fragments of veritable laws, which impose duties. Advocates of the extreme version of the Subsumability Thesis agree that power-conferring legal norms are mere parts or fragments of complete laws, but they submit that the sole complete laws are addressed not to citizens but to officials. Those laws direct officials to apply sanctions under certain conditions. Power-conferring legal norms are themselves some of the conditions that are cumulatively sufficient to trigger sanctions, and they specify further such conditions. Duty-imposing legal norms addressed to citizens are likewise some of those conditions, according to the partisans of the extreme version of the Subsumability Thesis.
Hart convincingly argued that the extreme form of the Subsumability Thesis inverts the paramount aspects and the subordinate aspects of law. Paramount among the functions of a system of legal governance, through the promulgation of authoritative standards, are the provision of guidance to citizens and the coordination of citizens' doings and the preservation of public order and the establishment of the normative frameworks of sundry activities. 22 Crucial, but ancillary to those main functions, are mechanisms for the resolution of disputes and for the disciplining of malefactors. Hart again fruitfully analogized the norms of a legal system to the rules of a game such as baseball. He underscored the perversity of treating all such rules as instructions to umpires and scorers about the conditions under which they are to reach certain determinations and perform certain actions. Important though the rules addressed specifically to umpires and scorers are, most of the rules of a game such as baseball are addressed principally to participants. We would darken counsel if we were to construe all the rules as addressed primarily or exclusively to umpires and scorers, for we would be blinding ourselves to the ways in which the rules furnish guidance to players and enable them to interact with one another concertedly. As Hart declared, the uniformity imposed on the rules by this transformation of them conceals the ways in which the rules operate, and the manner in which the players use them in guiding purposive activities, and so obscures their function in the co-operative, though competitive, social enterprise which is the game. 23 Although the moderate version of the Subsumability Thesis does not contend that all legal norms are addressed chiefly or exclusively to officials, it too purchases uniformity at the price of distortion and obfuscation. Hart encouraged his readers to contemplate power-conferring laws "from the point of view of those who exercise [the conferred powers]." 24 When we attend to that internal point of view, we can grasp that the laws which bestow powers on private individuals are "an additional element introduced by the law into social life over and above that of coercive control." 25 Hart elaborated: This is so because 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. He is made competent to determine the course of the law within the sphere of his contracts, trusts, wills, and other structures of rights and duties which he is enabled to build.

Kramer
Hart drove home his point with a rhetorical question: "Why should rules which are used in this special way, and confer this huge and distinctive amenity, not be recognized as distinct from rules which impose duties, the incidence of which is indeed in part determined by the exercise of such powers?" 27 (Note that Hart did not claim, and did not need to claim, that private powers are abundantly present in every legal system or even in every central instance of a legal system. In some rudimentary legal systems and in Communist legal systems, the presence of private powers is quite exiguous. Hart would readily have granted as much while maintaining that the Subsumability Thesis obscures the distinctiveness of the laws which confer private powers even in such systems-and while maintaining that the Subsumability Thesis is egregiously distortive in its obscuring of the distinctiveness of the laws which confer private powers in the many legal systems where those powers abound.) Norms that confer public powers of law-ascertainment and legislation and adjudication and administration on officials are likewise misrepresented when they are treated as mere fragments of duty-imposing norms. As I have recounted elsewhere, Hart discerned that such power-conferring norms are pivotal to the very existence and operativeness of any legal system. 28 He therefore wrote: "To represent such rules as mere aspects or fragments of the rules of duty is, even more than in the private sphere, to obscure the distinctive characteristics of law and of the activities possible within its framework." 29 Here as elsewhere, Hart objected to the Subsumability Thesis not on logical or formal grounds but on the ground that any such understanding of power-conferring laws badly fails to capture the import of those laws in the structuring of social institutions and intercourse.

Hart's Neglect of Power-Conferring Norms
Given the forcefulness of Hart's animadversions on Austin's disregard of powerconferring norms, it is surprising that Hart himself omitted to take account of such norms at some key junctures in his theorizing. Quite a few examples of his neglect of power-conferring norms could be adduced here, but-to keep this paper to a manageable length-I will confine myself to three. the philosophy of law-was focused squarely on duty-imposing norms and not on power-conferring norms (nor on immunity-conferring norms). With that exposition he purported to be recounting the perspective of anyone who accepts social norms, but in fact he was recounting only the perspective of anyone who accepts social duty-imposing norms.

The Internal Point of View
In the fourth chapter of The Concept of Law, Hart delineated the critical reflective attitude that is the hallmark of the internal point of view. 30 Through that delineation, he was illuminating two distinctions that vitally informed his critique of Austin and his own alternative theory of law: the distinction between normguided behavior and merely habitual behavior, and the distinction between the characteristic perspective of a participant in a practice and the characteristic perspective of an observer of a practice. Comprised by each of those distinctions is the internal viewpoint, which involves the acceptance of some norm or practice or institution by a participant.
Though the critical reflective attitude is an affect, it manifests itself as a trio of behavioral dispositions. (An affect of a person is obviously internal to that person's psyche, but the adjective 'internal' in the phrase 'internal point of view' does not denote interiority in that sense. It instead indicates that certain normupholding patterns of behavior are internal to institutions or practices or activities in that those behavioral patterns are characteristically exhibited by the participants in institutions or practices or activities.) If Jane evinces the critical reflective attitude in relation to some norm N, she is generally disposed to comply with N's requirements insofar as they are applicable to her conduct, and she is also generally disposed to criticize any contraventions of those requirements by other people, and she is likewise generally disposed to acknowledge the appropriateness of censure directed against her on any occasions when she herself has-perhaps unwittingly-contravened N. These dispositions can be underlain by various motivations, as Hart persistently emphasized. 31 Now, in the present context, what is disconcerting about Hart's otherwise insightful analysis of the internal point of view is that the analysis applies only to duty-imposing norms and not to power-conferring norms. 32 Throughout his discussion of social norms and the internal point of view, Hart was envisaging the categorical requirements established by duty-imposing norms. As he wrote, "where there is [a social norm,] deviations are generally regarded as lapses or faults open to criticism, and threatened deviations meet with pressure for conformity." 33 In other words, the social norms contemplated at this juncture by Hart were markedly different from the power-conferring norms which he envisioned  Kramer when he contended that the role of many of those latter norms is to provide the structures or frameworks of sundry legal arrangements rather than to deter people from engaging in modes of conduct that are perceived as wrongful. He did of course advert elsewhere in The Concept of Law to power-conferring social norms such as uncodified rules of games that specify how goals or runs are to be scored, but he omitted to cover such norms in his exposition of the critical reflective attitude. Similarly, although he invoked the viewpoint of power-holders in his endeavors to highlight the functional differences between power-conferring laws and duty-imposing laws, he did not elucidate that viewpoint with an analysis comparable to his analysis of the critical reflective attitude that is displayed by anyone who accepts a duty-imposing norm. This lacuna in Hart's theory is significant not only because it aligns Hart malgré lui with the throng of other legal philosophers who have overlooked the roles of power-conferring norms in systems of law, but also because his account of the internal perspective that pertains to duty-imposing norms is not straightforwardly modifiable into an account of an internal perspective that pertains to power-conferring norms. Though there will be some clear parallels between the former account and any satisfactory account of the latter kind, there will also have to be some major dissimilarities between them. Let us, then, consider two pathways for an exposition of the internal perspective of a participant in relation to power-conferring norms. These pathways are not mutually exclusive, and will probably have to be combined in an appositely thorough approach to this matter.
First, we might hold that Jane has adopted the internal viewpoint of acceptance in relation to some power-conferring norm PN only if (1) she is generally disposed to recognize the effects produced by any acts of exercising the power(s) which PN has bestowed, and she is generally disposed to recognize the nonoccurrence of such effects when any attempts to exercise the power(s) in question are unsuccessful; (2) she is generally disposed to criticize or correct other people who fail to recognize the occurrence or non-occurrence of the aforementioned effects; and 3) she is generally disposed to acknowledge the appropriateness of criticism directed at any failures of her own to recognize the occurrence or non-occurrence of those effects. These three elements are, patently, counterparts of the three elements in Hart's explication of the critical reflective attitude. However, instead of being oriented primarily toward PN itself, each element is oriented primarily toward acts of exercising the powers that have been conferred by PN. Recognizing the effects of any acts of exercising those powers is both a cognitive matter and a behavioral matter. One recognizes the effects by apprehending them whenever one has any occasion to apprehend them, and by adjusting one's conduct and decisions in response to them. Of course, the adjustments in one's conduct and decisions might be utterly routine and unreflective in many contexts. Still, insofar as one fails to undertake those adjustments when one has any occasion to undertake them, one is pro tanto failing to adopt the internal viewpoint of acceptance in relation to PN-unless one promptly corrects one's lapses either as a result of self-criticism or as a result of remonstrations from others.
Second, we might hold that Jane has adopted the internal viewpoint of acceptance in relation to PN only if (1a) she is generally disposed to exercise some power bestowed on her by PN, in contexts where her exercising of that power will plainly be beneficial and legitimate; (2a) she is generally disposed to criticize other people who have persistently omitted to exercise some power bestowed on each of them by PN, in contexts where the exercising of that power would plainly have been beneficial and legitimate; and (3a) she is generally disposed to acknowledge the appropriateness of objections directed against her own persistent failures to exercise the aforementioned power in contexts where her exercising of it would plainly have been beneficial and legitimate. Again, of course, each element in this triadic distillation is a counterpart of an element in Hart's explication of the critical reflective attitude.
Somebody might worry that the foregoing two paragraphs have not expounded the internal point of view in relation to power-conferring norms, and that they have instead expounded the internal point of view in relation to certain duty-imposing norms. According to such a line of thought, the penultimate paragraph above has distilled the internal viewpoint of a person who accepts a norm that imposes a duty to recognize the effects of any acts of exercising some specified power, and the final paragraph above has distilled the internal viewpoint of a person who accepts a norm that imposes a duty to exercise some specified power in contexts where doing so will plainly be beneficial and legitimate. Readers inclined to raise this worry will thus presume that I have not managed to provide an account of an internal perspective that applies to power-conferring norms.
Two rejoinders to such a query are pertinent here. In the first place, even if we were to grant arguendo that each of my explications of the viewpoint internal to a power-conferring norm PN has specified the viewpoint internal to a certain dutyimposing norm, we should continue to maintain that those explications have together recounted the internal viewpoint of anyone who accepts PN. Given that the relevant duty-imposing norms pertain either to recognizing the effects of exercises of powers or to exercising those powers, the specifics of the critical reflective attitude in relation to each such norm will coincide with the specifics of the internal viewpoint in relation to PN. Someone who adopts the internal perspective that consists in accepting one of the relevant duty-imposing norms will pro tanto have adopted the internal perspective that consists in accepting PN.
Also militating against the worry outlined in the penultimate paragraph above is that my explications of the internal viewpoint of someone who accepts PN are more capacious than the worry implies. Each of those explications does recount a viewpoint internal to a certain duty-imposing norm, but each of them also ranges more widely. My first account covers any situation in which the criticism to which the account adverts is concerned not with a breach of duty but with an instance of intellectual obtuseness. Similarly, my second account covers any situation in which the criticism to which the account adverts is concerned not with a 400 Kramer breach of duty but with an instance of imprudence. In other words, people can adopt the internal perspective of acceptance in relation to PN without presuming that they are under any duties to recognize the effects of exertions of the powers conferred by PN, and without presuming that they are under any duties to exercise any of those powers in contexts where the exercising of them would plainly be beneficial. Hence, my distillation of the internal perspective of somebody who accepts PN is not reducible (in either of its two versions) to a distillation of the internal perspective of somebody who accepts a certain duty-imposing norm. Now, although I have just sought to rebut a query about my two expositions of the perspective that is internal to power-conferring norms, I have not propounded either of those expositions as a definitive formulation. Rather, each of them is meant to be suggestive and to stimulate further thinking about this matter. As I have already mused, the two expositions will probably have to be combined in any full treatment of this problem; each of them articulates a necessary tripartite condition, rather than a sufficient tripartite condition, for the existence of an internal point of view in relation to PN. Of key importance for my present purposes is simply the fact that Hart neglected power-conferring norms (and immunity-conferring norms) in his analysis of the internal point of view. Quite remarkable is such an oversight by a philosopher who did so much to draw the attention of his fellow philosophers to the import of power-conferring norms.

Self-Imposed Obligations
In the final chapter of The Concept of Law, Hart presented a largely commendable discussion of the limits on national sovereignty that are imposed by the constraints of international law. However, one brief portion of his discussion is disastrously confused-and the confusion stems from another instance of his neglect of power-conferring laws. 34 Hart there set out to refute voluntarist theories of international law, which purport to "reconcile the (absolute) sovereignty of states with the existence of binding rules of international law, by treating all international obligations as self-imposed like the obligation which arises from a promise." 35 In this characterization of the voluntarist theories, Hart correctly declared that the explanandum of any such theory is the existence of obligations incumbent on states under international law. A proponent of the voluntarist approach maintains that any such obligations have been activated through the acceptance of them by each nation-state that is now under them. Having opened his discussion of voluntarism correctly in this manner, however, Hart stumbled into confusion by resolving to demonstrate the incoherence of "the argument designed to show that states, because of their sovereignty, can only be subject to or bound by rules which they have imposed upon themselves." 36 Notwithstanding that he had begun by correctly attributing to the voluntarists a thesis about self-imposed obligations, Hart here incorrectly attributed to them a thesis about self-imposed rules. His blunder on that point permeated his whole attempt to expose the incoherence of voluntarism, as he argued that "the [voluntarist] view that a state may impose obligations on itself by promise, agreement, or treaty is not, however, consistent with the theory that states are subject only to rules which they have thus imposed on themselves." 37 He elaborated: For, in order that words, spoken or written, should in certain circumstances function as a promise, agreement, or treaty, and so give rise to obligations and confer rights which others may claim, rules must already exist providing that a state is bound to do whatever it undertakes by appropriate words to do. Such rules : : : obviously cannot derive their obligatory status from a self-imposed obligation to obey them. 38 Though Hart was correct in contending that any blanket denial of the possibility of acceptance-independent norms in the international arena would be incoherent if combined with the proposition that states can impose obligations on themselves, he was thereby attacking a straw man-because voluntarists deny not the possibility of acceptance-independent norms but instead the possibility of acceptance-independent obligations. Their denial of that latter possibility is consistent with an affirmation of the former possibility, since the norms applicable to states include power-conferring norms. Some of those power-conferring norms are acceptance-independent. Power-conferring norms are not in themselves obligatory; they do not sans plus impose any obligations on the states that are within their power-conferring sway. Hence, the acceptance-independence of some norms which confer powers in the international domain is consistent with the proposition that all obligations incumbent on states are self-imposed through acceptance.
When Hart wrote in the passage above that the norms under which a state imposes obligations on itself cannot derive their own obligatoriness from any self-imposed obligations to obey them, he was strangely failing to notice that those norms are power-conferring rather than duty-imposing. Like other power-conferring norms, they are not obligatory; and thus they do not need to derive obligatoriness from any source. Instead of imposing obligations on each state, they endow each state with powers to impose obligations on itself. What is so puzzling and dismaying is that I am here making points which Hart himself made mutatis mutandis in his own discussions of power-conferring norms in the third chapter of The Concept of Law. He there emphasized that power-conferring laws are not obligatory on anyone to whom they are addressed-"[s]uch laws do not impose duties or obligations" 39 -and he rightly held that promissory norms