Scholarship Repository University of Minnesota Law School Articles Faculty Scholarship 2006 RRoobbeerrtt AAlleexxyy,, RRaaddbbrruucchh''ss FFoorrmmuullaa,, aanndd tthhee NNaattuurree ooff LLeeggaall TThheeoorryy Brian H. Bix University of Minnesota Law School, [email protected] Follow this and additional works at: https://scholarship.law.umn.edu/faculty_articles Part of the Law Commons RReeccoommmmeennddeedd CCiittaattiioonn Brian H. Bix, Robert Alexy, Radbruch's Formula, and the Nature of Legal Theory, 37 RECHTSTHEORIE 139 (2006), available at https://scholarship.law.umn.edu/faculty_articles/236. This Article is brought to you for free and open access by the University of Minnesota Law School. It has been accepted for inclusion in the Faculty Scholarship collection by an authorized administrator of the Scholarship Repository. For more information, please contact [email protected]. RECHTSTHEORIE 37 (2006). S. 139-149 Duncker& Humblot. 12165 Berlin ROBERT ALEXY, RADBRUCH'S FORMULA, AND THE NATURE OF LEGAL THEORY By Brian Bix, Minneapolis* I. Introduction Gustav Radbruch is well known for a "formula" that addresses the conflict of positive law and justice, a formula discussed in the context of the consideration of Nazi laws by the courts in the post-War German Federal Republic, and East German laws in the post-unification German courts. More recently, Robert Alexy has defended a version of Rad bruch's formula, offering arguments for it that are different from and more sophisticated than those that were adduced by Radbruch himself. Alexy also placed Radbruch's formula within a larger context of concep tual analysis and theories about the nature of law. Both Radbruch and Alexy claim that their positions are incompatible with legal positivism, and therefore count as a rejection (and perhaps, refutation) of it.1 In this paper, I will look at Radbruch's formula and Alexy's version of it. I want to focus not so much on the merit of the Radbruch-Alexy for mula, as on its proper characterization, and its appropriate placement within the larger context of legal philosophy. I am especially interested in the somewhat abstract and methodological question of what Radbruch and Alexy's formulations - and their strengths and weaknesses - can show us about the nature of theorizing about law. * This paper,was first presented at the Conference, "GustavRadbruch and Con temporary Jurisprudence", held at the University of Bristol. I am grateful to the participants at that conference, and to Matthew Kramer, Stanley L. Paulson and Torben Spaak, for their comments on earlier drafts of this article. 1 Alexy subtitles one of his books, An Argument from Injustice, Oxford 2002, "A Reply to Legal Positivism." In his later work, Radbruch dismissed legal positi vism as a position that "rendered the German legal profession defenceless against statutes that are arbitrary and criminal" and one "incapable of establishing the validity of statutes," G. Radbruch, Statutory Lawlessness and Supra-Statutory Law (1946), in: The Oxford Journal of Legal Studies (OJLS) 2006, pp. 1-11, 6; and he spoke approvingly of "the struggle against positivism." Ibid. For the argument that Radbruch was always opposed to at least some forms or understandings of legal positivism, see S. L. Paulson, On the Background and Significance of Gustav Hadbruch's Post-War Papers, in: OJLS 2006, pp. 17-40, 32-38. 140 Brian Bix II. Radbruch's Formula Gustav Radbruch wrote the following in a 1946 article: "The conflict between justice and legal certainty may well be resolved in this way: The positive law, secured by legislation and power, takes precedence even when its content is unjust and fails to benefit the people, unless the conflict between statute and justice reaches such an intolerable degree that the stalute, as 'flawed law', must yield to justice.,,2 Radbruch adds: "It is impossible to draw a sharper line between cases of statutory lawlessness and statutes that are valid despite their flaws. One line of distinction. however, can be drawn with utmost clarity: Where there is not even an attempt at jus tice, where equality, the core of justice, is deliberately betrayed in the issuance of positive law, then the statute is not merely'flawed law', it lacks completely the very nature oflaw.,,3 From the text, it seems clear that Radbruch intended the second quota tion to clarify the first, but the result has in fact been two quite different formulations.'! It is the first formulation that has been used by the courts,5 in part, one assumes, because the second formulation would be hard to implement, unless read in such a way that it roughly equates to the first formulation. What would it mean for a lawmaker "not even [to] attempt ... justice" or "deliberately to betrayLi equality"? In nearly every case, lawmakers attempt what is right, under their world view and con ception of what is right, however warped that worlel view and concep tion might be.6 However, under a different, reasonable reading of the text, one could/would speak of many of the Nazi laws as "not even ... attempt[ing] justice" and "deliberately betray(ing] equality" - even though the lawmakers involved may have subjectively believed that 2 Radbruch, Statutory Lawlessness and Supra-Statutory Law (note 1), p. 7. Contrast this with Radbruch's statement in 1932: "It is the professional duty of the judge to validate the law's claim to validity, to sacrifice his own sense of the right to the authoritative command of the law, to ask only what is legal and not if itis also just." G. Radbruch, Legal Philosophy, in: E. W. Patterson (ed.), The Legal Philosophies of Lask, Radbruch, and Dabin, Harvard 1950, pp. 47-224, at § 10, p. 119. This seems a world away from Radbruch's 1946 "formula", above; however, at least one commentator has argued that the contrast between Radbruch's earlier and later work has been overstated. See Paulson, On the Background and Signifi cance of Gustav Radbruch's Post-War Papers (note 1). 3 Radbruch, Statutory Lawlessness and Supra-Statutory Law (note 1), p. 7. 4 See S. L. Paulson, Radbruch on Unjust Laws: Competing Earlier and Later Views? In: OJLS 15 (1995), pp. 489-500,491; id., On the Background and Signifi cance of Gustav Radbruch's Post-War Papers (note 1), pp. 26-27. 5 Paulson, On the Background and Significance of Gustav Radbruch's Post-War Papers (note 1), pp. 26-27. 6 One can note here the connection with Alexy's "claim to correctness", dis cussedlaterin this article. Robert Alexy, Radbruch's Formula 141 those laws followed their own Nazi ideas of equality and justice - just because the resulting laws are so clearly contrary to the requirements of justice and equality. One might also argue that the first (part of the) formula is the direc tion to courts, while the second is (merely) a statement about the nature of law - 1"vhether such statements inevitably have consequences for how courts should decide cases is controversial, and will be discussed further below. In any event, this article will, like most commentators (and courts), focus on the first (part of the) formula, rather than the second. It seems clear from the examples and descriptions in Radbruch's text. that he equates the conclusion that a norm lacks legal st.atus (due to ex treme injustice) with t.he conclusion that the norm was void ab initio, or at least that it should have no application to legal disputes before a court.7 Radbruch's formula in fact has been cited repeatedly in German court decisions that refuse to give effect to certain Nazi-era and East s German rules. In recent work, Robert Alexy has endorsed Radbruch's formula and made it a centerpiece of his theory of law,» constructing a theory of law that combines the formula with Alexy's own "correctness thesis".lo The 7 E.g., G. Radbruch, Five Minutes of Legal Philosophy (1945), OJLS 26 (2006), pp. 1:3--15, 14. ("There are principles of law, therefore, that are weightier than any legal enactment, so that a law in conflict with them is devoid of validity."); id., Statutory Lawlessness and Supra-Statutory Law (note 1), p. 9 ("Objectively speaking, perversion of the law exists where we can determine, in the light of the basic principles we have developed, that the statute applied was not law at all ..." (emphasis added)). I recognize that there is, for some purposes, a practical difference between stat ing that a norm was "void ab initio" and saying that it was invalid or voidable and subsequently invalidated by a court (through constitutional review or some other basis). For example, there have been cases where a law, invalidated by a United States Supreme Court decision, was held to be valid when that decision was subsequently overruled. See 39 U.S. Op. Atty. Gen. 22 (1937). If that same norm had been held to be "void ab initio" a subsequent court action would not have revived it. Cf. Norton v. Shelby County, 118 U.S. 425, 442 (1886) ("An uncon stitutional act is not a law; it confers no rights; it imposes no duties; it affords no protection; it creates no office; it is, in legal contemplation, as inoperative as though it had never been passed."). Radbruch himself did not always seem to dis tinguish invalidity and nullity in discussing his formula. See Paulson, On the Background and Significance of Gustav Radbruch's Post-WarPapers (note 1), p. 26 n.69. II Paulson, On the Background and Significance of Gustav Radbruch's Post-War Papers (note 1), p. 18. 9 R. Alexy, A Defence of Radbruch's Formula, in: D. Dyzenhaus (ed.), Recraft ing the Rule of Law (Hart), 1999, pp. 15-39; id., The Argument from Injustice (note 1), pp. 28-31, 40-81. 10 On the "correctness thesis", see, e.g., R. Alexy, Law and Correctness, in: Cur rent Legal Problems 51 (1998), p. 205. 142 Brian Bix following sections will focus more on Alexy's work, evaluating it in the context of general legal theory. III. Legal Theory and Legal Practice To start at the most basic point: legal theory, narrowly understood (and as the term will be used in this paper), is a theory oIiered to explain (the nature of) law. There is more that can be said, and should be said in further explication: e.g., Should we assume (and if so, why?) that there is or should be a genera] or universal theory of law?; And are theories of law theories about the concept of law, and if so, how many concepts II are there?12 However, in this article those questions will be put aside, to allow us to concentrate on different basic methodological problems. The focus of both Radbruch's formula and Alexy's use of it is judicial decision-making: primarily the resolution of disputes that turn, or might turn, on the legal validity of an evil law, but also other disputes where the application of "higher law" might affect the outcome. Under the Radbruch-Alexy approach, extremely unjust laws lose their character as laws, are not to be applied in legal disputes, and therefore do not aUect citizens' legal rights and obligations. Whatever the merits of this claim,1:l it has also been presented as a claim about the nature of law, a non 14 positivistic or anti-positivistic approach offered as an alternative to or a 11 Recently, a number of theorists have raised questions about the appropriate ness of conceptual analysis for philosophy generally, and for legal philosophy in particular. E.g., B. Leiter, Beyond the Hart/Dworkin Debate: The Methodology Problem in Jurisprudence, in: American Journal of Jurisprudence 4R (2003), pp. 17-51; J. Fodor, Water's Water Everywhere, in: London Review of Books, vol. 26, no. 20, Oct. 21, 2004. The basic critique raises questions as to whether any significant truths can be discovered about concepts (arguing that philosophy should turn instead to naturalist/scientific/empirical investigation). An alterna tive attack on conceptual analysis for legal theory argues that there is marc than one tenable concept or theory of law, and that one must use moral evaluation to choose among them. E.g., S. R. Perry, Interpretation and Methodology, in: A. 1'v1ar mol' (ed.), Law and Interpretation, Oxford 1995, pp. 97-1:35. 12 See, e.g., J. Raz, On the Nature of Law, in: Archiv fur Rechts- und Sozialphi losophie 82 (1996), p. 1; id., Can There Be a Theory of Law? In: M. P. Golding/W. A. Edmundson (eds.), The Blackwell Guide to the Philosophy of Law and Legal Theory, Blackwell 2005, pp. 324-342; B. Bix, Raz on Necessity, in: Law and Philo sophy 22 (2003), p. 537; id., Raz, Authority, and Conceptual Analysis, in: American Journal of Jurisprudence 50 (2005), pp. 311-316. 13 As Alexy points out, there can be important practical benefits for a court's being able to say that an unjust rule was void ab initio: (1) with a statute negating property claims, if the statue was never legally valid one does not need to get the legislature to pass retroactive legislation; the property simply was never lost. though one does, of course, still have to get a judicial decree invalidating the leg islation; and (2) treating the norms as void ab initio may also have an advantage in cases where a claim would otherwise have been untimely. Alexy, A Defence of Radbruch's Formula (note 9), pp. 19, 36. Robert Alexy, Radbruch's Formula 143 refutation of legal positivism. The Radbruch formula as a theory of law (or as central to a theory of law) is entirely a separate claim from the formula as instructions to a judge, and the two claims must be evalu ated, differently, on their own merits. (As will be discussed, it is open to a theorist to state that a theory of law does or should give instructions to judges, but there is nothing in the nature of the task, "offer a theory of law," that makes such a connection necessary or obvious. Any such connection must be shown by the theorist.) To what extent can (or should, or must) a theory of law have implica tions for the resolution of real-world legal disputes? Here theorists have offered a wide variety of answers, which might be considered along a spectrum. At one extreme, Ronald Dworkin seems to argue that a judge's legal theory always has an impact on the resolution of individual casesY Alexy seems to hold a middle position, such that legal theory - or at least the particular legal theory he is advocating - is decisive in a small number of cases, but otherwise has little or no effect.16 At the other ex treme is the view I have advocated elsewhere17 that one's theory of law : has (or should have) no effect on the resolution of particular cases. Let us consider the connection between legal theory and practice from a dHferent perspective. Often theorists arguing for a particular legal theory will justify their preference by reference to actual cases: that the preferred theory better fits the actual results of cases (descriptive fit), or 14 By Ale:ry, The Argument from Injustice (note 1). The attribution of this posi tion to Radbruch is less certain. CL Paulson, On the Background and Significance of Gustav Radbruch's Post-War Papers (note 1), pp. 35-38. 15 E.g., R. Dworkin, Law's Empire, Harvard 1086, p. 90; id., Legal Theory and the Problem of Sense, in: R. Gavison (ed.), Issues in Contemporary Legal Philoso phy, Oxford 1987, pp. 11-20, 14-15. 16 It is important to note that Radbruch, in both his early works and his later writings, offered comments on the nature of law that were separate from, if still huving some connection to, his formula. For example, he wrote: "law is ()nly that which at least aims at serving justice. Justice is the idea of law, determining the very nature of law." G. Radhruch, Die Problematik del' H.echtsidee, in: Die Dios kuren. Jahrbuch fUr Geisteswissenschaften 3 (1924), pp. 43-50, 45, quoted in translation in Pa1tlson, Radbruch on Unjust Laws: Competing Earlier and Later Views? (note 4), p. 496; see also Radbruch, Statutory Lawlessness and Supra-Stat utory Law (note 7), pp. 6-7. His 1932 treatise seems to sound a similar theme, though in a Continental (in this case, neo-Kantian) language that makes it harder for some readers from the Anglo-American tradition to understand. E.g., Rad bmch, Legal Philosophy (note 2), p. 52: "The concept of law cannot be otherwise defined than as that reality whose meaning is the realization of the idea of law. Law may be unjust ... but it is law simply because its meaning is to be just." On Radbruch's neo-Kantianism, see Paulson, On the Background and Significance of GUstav Radbruch's Post-War Papers (note 1), pp. 29-32. 17 B. Bix, Legal Positivism, in: M. P. Golding/W. A. Edmundson (eds.), The Blackwell Guide to the Philosophy of Law and Legal Theory, Blackwell 2005, pp. 29-49, 36-38. 144 Brian Bix that the preferred theory would lead to better outcomes in certain cases (prescriptive superiority). However, one challenge (or at least one re sponse) to this way of connecting legal theories and the resolution of particular disputes is that the same legal result can be characterized, ra tionalized, or justified in different ways. That is, it may be that legal the ories are frequently orthogonal to the results of legal disputes because the same resolution can be explained or justified under most or all alter native theories. This is not to say that the actual disputes would offer no evidence at all. Consider the debate within legal positivism, in which "exclusive legal positivism" interprets legal positivism's separation thesis as requir ing that the validity and content of legal norms be ascertainable without recourse to moral norms; while "inclusive legal positivism" allows for recourse to moral norms, but only where such recourse has been author ized within the legal system by positive sources.IS Inclusive legal positi vism's view of law may more easily and elegantly explain what is going on in certain cases of constitutional judicial review based on moral sounding constitutional norms than would an exclusive legal positivist account - but that evidence would remain far from conclusive regarding the inclusive-exclusive debate.19 The "grudge informer" case in the Hart-Fuller debate, where both H. L. A. Hart and Lon Fuller discussed the merits of Radbruch's ap proach,20 is another good example of how theory can be orthogonal to practice. As described in the debate, the case was as follows: during the Nazi regime, a woman used a Nazi statute to try to get her husband killed. Under a later regime, she was tried for endangering the husband's civil rights, and she defended that her actions were allowed, if not re quired, by the Nazi law. Fuller argued that the later court was justified in treating the Nazi rule as "not law", and therefore no possible defense to the charge the woman faced. Hart would have preferred that the same result be reached by the enactment of retroactive legislation making the woman's action subject to punishment. (As Fuller pointed out, it is not clear why, if retroactive lawmaking is to be encouraged, it would make much difference whether it was done by the legislature or the court.21 ) 18 Ibid., pp. 36-38. 19 Ibid., pp. 37-38. 20 H. L. A. Hart, Positivism and the Separation of Law and Morals, in: Harvard Law Review 71 (1958), pp. 593-629, 615-621; L. L. Fulle); Positivism and Fidelity to Law - A Reply to Professor Hart, in: Harvard Law Review 71 (1958), pp. 630 672, 648-657. 21 Fuller, Positivism and Fidelity to Law (note 20), p. 649, Robert Alexy, Radbruch's Formula 145 When a court states that a particular norm is too unjust to be consid ered legally valid,22 this action can be characterized in different ways. Judges refusing to apply a statute according to its plain meaning may be interpreting the statute in light of its purposes and or in order to make the entire area of law more coherent; or they may be treating the law as invalid (or null and void ab initio) due to its inconsistency with constitu tional rules or with "higher law" that goes beyond positive sources; or they may be using their legislative power to modify or repeal existing legal norms.23 Courts, of course, tend to offer explanations and charac terizations of their own actions, but theorists need not take these charac terizations at face value (as, for example, we frequently ignore the claims by judges in prior eras that in difficult and landmark cases they were merely "discovering" the law, not making it themselves). This is a long way to make a small point: that even if there is a connec tion between legal theory and the resolution of particular cases (which I will explore further in the next section), arguments grounded in how particular cases were resolved, or how we think such cases should be re solved, may offer only uncertain evidence for choosing among alternative theories. IV. Theories of Judicial Ueasoning and Theories of Law There is, of course, a place - and an important place - for theories that can direct judges (either as a matter of law or as a matter of morality) as to how they should interpret or apply evil laws24 evil laws of their own - regimes or the evil laws of past regimes. However, it may be quite an other thing to equate such theories with theories about the nature of law, at least without further argument. As stated earlier, theorists carry a wiele range of views regarding the connections (if any) between theories of law and the resolution of partic- 22 Other examples are given by Raclbrnch, Statutory Lawlessness and Supra Statutory Law (note 1), pp. 2-6. While I do not think it is at all conclusive on this issue, we do need to keep 23 in mind Fuller's warning regarding the treatment of evil laws from old regimes: "So far as the courts are concerned, matters certainly would not have been helped if, instead of saying, 'This is not law,' they had said, 'This is law but it is so evil we will refuse to apply it.' Sure moral confusion reaches its height when a court refuses to apply somethingit admits to be law." Fulle7; Positivism and Fidelity to Law (note 20), p. 655. And to the extent that we focus on sources and judicial reasoning, we might 24 consider the disagreement between Dworkin and Raz regarding whether every norm a judge has an obligation to apply to a legal dispute thereby is or becomes a legal norm. SeeJ. Raz, Legal Principles and the Limits of Law, in: M. Cohen (eel.), Ronald Dworkin and Contemporary Jurisprudence, Rowman & Allenheld 1983, pp. n--87; R. DwoTkin, Al1eply byRonald Dworkin, in: ibid., pp. 247-300, 260-63. 146 Brian Bix ular legal disputes. When someone like Dworkin claims that there is no sharp distinction between the two, and a person's position on one entails a view regarding the other, he offers substantive arguments for that con clusion; he does not merely assume the connection. Similarly, I think the initial burden must be on the theorist who implicitly asserts that a theory of judicial reasoning is responsive (or even relevant) to inquiries about the nature of law. Again, this is not to say that the connection cannot be shown, only that it needs to be. At the same time, it is a fair question to ask Raz (and perhaps myself): if questions regarding the nature of law and when some rule or rule system warrants the labels "law" and "legal" are not closely related to the issue of which rules should be applied by judges in deciding dis putes, then what is the point of the inquiry? Raz would respond25 that his theory simply tracks our concept of law, and if the "law"/"not-law" distinction does not always track the distinction between acceptable/un acceptable sources for judges to use in resolving disputes, that is simply a reflection of both our linguistic practices and our legal practices. In the next section, I consider more closely the connection between ju dicial decision-making and legal theory in Alexy's work. V. Conceptual Analysis 1. Claim of Correctness Alexy's argument that, to be legal, an individual norm or a system of norms, must "claim correctness,,26 bears a strong resemblance to Joseph Raz's argument that legal systems necessarily claim authoritative status 27 for their norms. However, Alexy parts company with Raz when he states not only that a legal system that does not claim authority/correct ness is not a legal system, but that a legal system (or legal norm) that did not succeed at being correct/authoritative would be, for that reason, "defective. Raz, by contrast, makes it clear that a system that pur- ,,28 25 See Raz, Can There Be a Theory of Law? (note 12). 26 Alexy, The Argument from Injustice (note 1), p. 35. For a critical view of a "claim to correctness" thesis similar to Alexy's (given by Philip Soper), see Mat thew Kramer, In Defense of Legal Positivism, Oxford 1999, pp. 101-108. 27 J. Raz, Ethics in the Public Domain, Oxford 1994, p. 199. 28 Alexy, The Argument from Injustice (note 1), p. 36. Mark Murphy speaks in similar terms: that laws that "fail to be adequate rational standards" are "defec tive." M. Murphy, Natural LawJurisprudence, in: Legal Theory 10 (2003), pp. 241 267, 254. Murphy's approach raises many of the same questions and concerns as Radbruch's: is it either sensible or tenable to have an idea of "defective law" that is not reducible either to "legally valid but immoral" or "legally invalid"? Nigel Robert Alexy, Radbruch's Formula 147 parts to be authoritative but fails is still legal; in fact, Raz believes that this is the likely characterization for most legal systems.29 Raz clearly has the better of this exchange, if viewed narrowly. It does not follow logically from the fact that an entity must claim correctness or authority that its failure to achieve correctness or authority means that it is defective. If the only standard of legality is a kind of claim, then the only way to fail to achieve legality is to fail in some way in the making of this claim. (At the same time, it is an understandable move even if not one logically entailed - to go from claims about correctness to concern about its achievement.) On a similar theme, Alexy argues that it would be "defective" and "absurd" for a constitution to announce the creation of an "unjust re public.,,:1O (Though Alexy admits that this criterion would exclude very few norm systems, as most rulers make at least a show of promoting some version of justice.:l1 I would argue that Alexy's analysis confuses a ) general point about language and advocacy for something peculiar to, or essential to, lBw. If one is trying to sell, persuade, or encourage, one uses positive language. To use pejorative terms in any context that calls for support or persuasion is, at least initially, paradoxical.32 This is a point about language and rhetoric, not about law and morality. Alexy might respond that even if this part of his variation of Rad bruch's thesis is not distinctive to law it is nonetheless essential to law. However, I am not sure that even this is supportable. What if a country were to say (in its constitution, or in some other official forum) the fol lowing: "we have no time for so-called 'justice': that is the talk of weak countries; our nation is all about commercial efficiency and doing the best we can for the citizens of our great nation." Would a country's Simmonds has recently offered a quite different approach to justifying a connec tion between law and morality, arguing that the rule of law is an ideal "arche type" to which a normative system must approximate to some degree to be "law." N. E. Simmonds, Law as a Moral Idea, in: University of Toronto Law Journal 55 (2005), pp. 61-92, 85-86. For a criticism of Simmonds' view, see Matthew Krame1; Objectivity and the Rule of Law (forthcoming, Cambridge 2007). 2H Raz, Ethics in the Public Domain (note 27), pp. 200-202. :10 Alexy, The Argument from Injustice (note 1), pp. 36, 37; id., A Defence of Radbruch's Formula (note 9), p. 27. :11 Alexy, The Argument from Injustice (note 1), p. 127. 32 Alexy considers and rejects the possible rejoinder that this is merely a con vention of constitution writing. Alexy, The Argument from Injustice (note 1), p 37. However, that response misses the generality of the criticism: that it is a conven tion, or a general shared expectation, of all promotional speech. Much closer is Alexy's concession that the paradox of the unjust constitution is like asserting that "the cat is on the mat but I do not believeit is." Ibid., p. 38 n. 66.