No CrossRef data available.
Published online by Cambridge University Press: 21 December 2020
According to a recent wave of work by legal anti-positivists, legal norms are a subset of moral norms. This striking “one-system” view of law has rapidly become the dominant form of anti-positivism, but its implications have so far been little tested. This article argues that the one-system view leads systematically to untenable conclusions about what legal rights and obligations we have. For many clear legal norms, the view lacks the resources to explain the existence of corresponding moral norms. And its criteria for distinguishing legal norms within morality imply an under- or over-inclusive set of legal norms. I stress the special difficulties that apply beyond obligations, in the case of privileges and powers, and I show that the view's problems do not only—or mainly—concern egregiously unjust law, or indeed morally defective law at all. I close with reflections on legal normativity and the prospects for different forms of anti-positivism.
I am very grateful to Tom Adams, Andrew Currie, Andrea Dolcetti, James Edwards, Timothy Endicott, Cécile Fabre, John Gardner, Kate Greasley, Scott Hershovitz, Tess Little, Ezequiel Monti, Adam Perry, David Plunkett, Alex Sarch, Dan Singer, Nicos Stavropoulos, Sandy Steel, Sabine Tsuruda, Fred Wilmot-Smith, Daniel Wodak, and two anonymous referees for comments on earlier drafts of this paper, as well as to participants at the Dartmouth Workshop on Truth, Power, and the Foundations of Democracy, the New Directions in Philosophy of Law conference at Surrey, and the Seminario de Problemas Contemporáneos de Filosofía del Derecho at Universidad Torcuato Di Tella, for helpful discussion.
1. Ronald Dworkin, Justice for Hedgehogs (2011), at 405.
2. Ronald Dworkin, Law's Empire (1998).
3. Examples of this approach include Plunkett, David, A Positivist Route for Explaining How Facts Make Law, 18 Legal Theory 139 (2012)CrossRefGoogle Scholar; Berman, Mitchell, Of Law and Other Artificial Normative Systems, in Dimensions of Normativity: New Essays on Metaethics and Jurisprudence 137 (Plunkett, David, Shapiro, Scott J. & Toh, Kevin eds., 2019)Google Scholar.
4. Dworkin, supra note 1, at 405. For discussion of Dworkin's argument, see Greenberg, Mark, The Moral Impact Theory, the Dependence View, and Natural Law, in The Cambridge Companion to Natural Law Jurisprudence 275, 283–284 (Duke, George & George, Robert P. eds., 2017)Google Scholar.
6. See especially Greenberg, Mark, The Standard Picture and Its Discontents, in Oxford Studies in Philosophy of Law: Volume 1 39 (Green, Leslie & Leiter, Brian eds., 2011)Google Scholar.
7. See, e.g., Ronald Dworkin, Taking Rights Seriously (1978).
9. See, e.g., Greenberg, Mark, The Moral Impact Theory of Law, 123 Yale L.J. 1118, 1308 (2014)Google Scholar; Hershovitz, supra note 5, at 1194; Nicos Stavropoulos, Legal Interpretivism, in The Stanford Encyclopedia of Philosophy (Edward N. Zalta ed., Summer 2014), https://plato.stanford.edu/archives/sum2014/entries/law-interpretivist/.
10. See, e.g., Dworkin, supra note 7, at 87, 89. My interest here is not exegetical. Excellent accounts of Dworkin's views are given in Hershovitz, supra note 5, at 1195–1202, and Stavropoulos, Nicos, The Debate That Never Was, 130 Harvard L. Rev. 2082, 2086–2087 (2017)Google Scholar.
11. Dworkin, supra note 1, at 402.
12. Id. at 404–405.
13. Greenberg, supra note 6; Greenberg, supra note 9; Greenberg, supra note 4.
14. Hershovitz, supra note 5, at 1188.
15. Id. at 1195.
16. Id. at 1202–1203.
17. Jeremy Waldron, Jurisprudence for Hedgehogs, NYU School of Law, Public Law Research Paper No. 13-45 (2013), https://papers.ssrn.com/abstract=2290309. Waldron's aim is to “elaborate and defend [Dworkin's] basic position, not just expound it.” Id. at 8.
18. Stavropoulos, supra note 9. He also attributes this view to Dworkin.
19. Schaus, Steven, How to Think About Law as Morality: A Comment on Greenberg and Hershovitz, 124 Yale L.J. F. 224, 235–239 (2015)Google Scholar.
20. A similar example is given in Waldron, supra note 17, at 11–12.
21. I assume that moral norms, facts, etc., are “robustly normative,” in the sense that they are (or give) genuine reasons concerning what one ought to do sans phrase. A skeptic about this thesis might nevertheless accept much or all of the following argument, but that will depend on the variety of skepticism.
22. This argument would therefore apply to a view that denied the Identity Thesis but affirmed the Correspondence Thesis, such as an unusually strong view in favor of a moral obligation to obey the law, according to which morality grounds an obligation (power, privilege, etc.) with the same content as every (independent) legal norm.
23. This leaves it open whether these types of lawmaking contribute to legal content by moral or nonmoral means, so it does not beg any questions against the one-system view.
24. I do not assume that all or even most legal norms in the jurisdiction are clear, only that the ones I discuss are the kinds of norms that can be.
25. Dworkin, supra note 2, ch. 1.
26. See Greenberg, supra note 9, at 1138.
27. See, e.g., Lewis A. Kornhauser, Doing Without the Concept of Law, NYU School of Law, Public Law Research Paper No. 15-33 (2015), https://papers.ssrn.com/abstract=2640605.
28. Ronald Dworkin, Justice in Robes (2006), at 2–5.
29. Even if these facts are, as Hershovitz suggests, themselves grounded in part by pragmatic considerations. For Hershovitz would presumably accept that one can speak truly or falsely about legal obligations, at least relative to some context.
30. Different sets of obligations will be picked out as legal by the one-system anti-positivist and the “two-systems” anti-positivist who allows that legal and moral norms may diverge. But does their disagreement about legal obligations matter, if they are otherwise in agreement about what people, including judges, ought to do at the end of the day? Since law is so pervasively embedded in our social and political practices, there are in my view intrinsic theoretical reasons to want a sound account of it, quite apart from practical questions. In addition, the disagreement between the one-system anti-positivist and the two-systems anti-positivist reflects and impacts various other philosophical commitments. For example, the two-systems picture sits uncomfortably with skepticism about nonrobust normativity.
31. My critique therefore does not target the natural law view that, roughly, valid legal norms may diverge from moral norms but are thereby defective or noncentral qua law, defended in, e.g., John Finnis, Natural Law and Natural Rights (2d ed. 2011); Mark C. Murphy, Natural Law in Jurisprudence and Politics (2006).
32. Dworkin, supra note 1, at 410–412; Greenberg, supra note 9, at 1337.
33. Greenberg, supra note 9, at 1137.
34. Or so I shall assume in order to assess different versions of the one-system view. A skeptic about either pro tanto or all-things-considered norms will thereby be a skeptic of the corresponding form of the one-system view.
35. Thanks to an anonymous reviewer for pressing me to clarify the following discussion.
36. For discussion, see, e.g., Stephen Darwall, Bipolar Obligation, in Morality, Authority, and Law: Essays in Second- Personal Ethics I 20 (2013); Wolf, Susan, Moral Obligations and Social Commands, in Metaphysics and the Good: Themes from the Philosophy of Robert Merrihew Adams 343 (Newlands, Samuel & Jorgensen, Larry eds., 2009)Google Scholar; R. Jay Wallace, The Moral Nexus (2019), ch. 2. Raz famously explains obligations as complexes of first- and second-order reasons: Raz, Joseph, Promises and Obligations, in Law, Society and Morality: Essays in Honour of H.L.A. Hart 210 (Hacker, P.M.S. & Raz, Joseph eds., 1977), at 210Google Scholar.
37. See, e.g., Wolf, supra note 36; Darwall, supra note 36; Wallace, supra note 36, ch. 2.
38. “Normally” allows that, on some views, a breach of duty may not always count as a wrong—where, for example, the agent has an excuse. An analogous proviso applies to the following point about reactive attitudes and actions.
39. See, e.g., Stephen Darwall, The Second-Person Standpoint: Morality, Respect, and Accountability (2006), chs. 4–5; Judith Jarvis Thomson, The Realm of Rights (1992), chs. 3–4.
40. Additionally, it is very plausible that an all-things-considered obligation specifies an action that would involve the least wrongdoing (if any) to perform.
41. Again, the assumption is favorable to the one-system view as it keeps the pro tanto variant on the table.
42. Greenberg identifies legal obligations as certain “genuine, all-things-considered, practical obligations” (which he regards as equivalent to “all-things-considered moral obligations”). The thesis that legal obligations are all-things-considered moral norms is, however, advanced “tentatively.” See Greenberg, supra note 9, at 1306–1308.
43. Id. at 1304.
44. Greenberg, supra note 9, at 1306.
45. These are generally strict liability offenses, and though some jurisdictions contain a necessity defense, in many it would not cover such a case, in the absence of a threat of death or serious injury.
46. Similar examples are given in Schaus, supra note 19, at 232.
47. The example is from Edwards, James, Laws that Are Made to Be Broken, 12 Crim. L. & Phil. 587, 591 (2018)Google Scholar, who discusses it in a different context.
48. Id. at 591.
49. As James Edwards and Dan Singer have pointed out to me, a defender of ATC may have to take this view of privileges for independent reasons. Consider a case where there is a moral duty not to φ, and so no moral privilege to φ. If the Correspondence Thesis applies to privileges, there is then no legal privilege to φ, and hence there is a legal duty not to φ. But no proponent of the one-system view accepts the false inference from a moral duty against φing to a legal duty against φing. These points are developed in James Edwards, Law as Morality (unpublished manuscript).
50. See, e.g., Immigration Rules, pt. 9, para. 320(6); British Nationality Act 1981, §40(2).
51. Hershovitz, supra note 5, at 1189.
52. See in particular Darwall, supra note 39. See also, e.g., Robert Merrihew Adams, Finite and Infinite Goods: A Framework for Ethics (2002), ch. 10; Wolf, supra note 36. For a view that stresses both the deliberative stringency and the accountability features of obligation, see Wallace, supra note 36. Thanks to an anonymous reviewer for urging me to consider such views.
53. Proponents of the one-system view have—understandably—not advanced the weaker claim that legal obligations are just certain moral reasons. Since a mere reason to φ neither is nor implies an obligation (or privilege, power, etc.) to φ, a reason could not be such a legal incident. It is also obscure what it would amount to for moral reasons (as opposed to obligations, etc.) to be judicially enforced, or generated in the legally proper way.
54. See Darwall, supra note 39; Adams, supra note 52, ch. 10; Wolf, supra note 36. Could PT suppose that breaching a defeated obligation is, ipso facto, in no way wrong, and leaves no remainder? That is a dubious view of moral obligations, implying that if one breaches a solemn promise in order to discharge a weightier obligation, questions of apology or forgiveness should not even arise. And it would not fill the gaps in the positive arguments, discussed presently, that PT must rely on to vindicate the Correspondence Thesis.
55. See, e.g., Adams, supra note 52, at 233; Wolf, supra note 36, at 349ff.; Stephen Darwall, Moral Obligation: Form and Substance, in Morality, Authority, and Law: Essays in Second-Personal Ethics I 42ff. (2013).
56. See, e.g., Darwall, supra note 36, at 33.
57. See Darwall, supra note 39; Adams, supra note 52, ch. 10; Wolf, supra note 36.
58. Indeed, this is plausible even if there is no particular urgency, so such cases do not seem to depend on conflicts of obligation.
59. Justified blame being crucial: issuers of unjust orders frequently engage in unjustified accountability practices.
61. Viehoff, Daniel, Democratic Equality and Political Authority, 42 Phil. & Pub. Aff. 337, 367 (2014)Google Scholar.
63. Notice that the question here is not whether officials may permissibly sanction you for breach of the legal obligation: that would not imply that you already had a moral obligation corresponding to the legal one.
64. Greenberg, supra note 9, at 1316.
65. I here prescind from the separate matter of any legitimate expectations, whose limitations I address above.
66. See, e.g., Dworkin, supra note 2, at 225–238; Greenberg, supra note 9, at 1316.
67. In fact, however, as we saw above, even if it would be unfair to B for A to not be subject to the obligation, it does not follow that it would be unfair for A to breach it.
68. For discussion of the conditionality of democracy's putative authority, see, e.g., Thomas Christiano, The Constitution of Equality: Democratic Authority and Its Limits (2008), at 260–300; Niko Kolodny, Rule Over None I: What Justifies Democracy?, 42 Phil. & Pub. Aff. 195, 198 (2014); Viehoff, supra note 61, at 371–375.
69. Greenberg, supra note 9, at 1314.
70. For an overview, see Green, Leslie, Law and Obligations, in The Oxford Handbook of Jurisprudence and Philosophy of Law 514 (Coleman, Jules L., Himma, Kenneth Einar & Shapiro, Scott J. eds., 2002)Google Scholar.
71. I thank Scott Hershovitz for this suggestion.
73. Greenberg, supra note 9, at 1337.
74. Waldron, supra note 17, at 26.
75. Hershovitz, supra note 5, at 1189–1190.
76. Greenberg, supra note 4, at 287.
77. See Waldron, supra note 17, at 21–22; Greenberg, supra note 9, at 1338; Hershovitz, supra note 5, at 1194.
78. I am grateful to Scott Hershovitz for pressing this concern.
79. One reason for doubt is pervasive cross-contextual legal disagreement. Suppose a government official insists that it is lawful to do something that a court later rules was unlawful. If the official and the court had sufficiently different practical purposes—which, Hershovitz supposes, they might well have done—then, on his view, their divergent legal assertions might simply have reflected their using “legal obligation” and cognate words in different senses, corresponding to different demarcations. The implication, if so, is that the court and the official did not disagree substantively, and that each party may have been correct in their legal assertions, since each concerned their own contextually appropriate sense of “legal obligation.” This seems a remarkable conclusion. Surely the court and the official were not confusedly talking at cross-purposes or engaged in some kind of charade. Was their real disagreement about which sense of “legal obligation” was relevant? There is at least a serious explanatory burden on the contextualist about demarcation here. The challenge in explaining legal disagreement generalizes to any theory that proliferates different senses of “legal obligation” that legal participants are said to employ. For Hershovitz's discussion of this issue, see Hershovitz, supra note 5, at 1202–1203. Thanks to an anonymous reviewer for encouraging me to say more here.
80. As I indicated in Section I, I take the kinds of legal obligations I have considered to be capable of being clear across various such standard contexts.
81. Hershovitz, supra note 5, at 1188.
82. Dworkin, supra note 1, at 404–405.
83. Hershovitz, supra note 5, at 1188.
84. Schaus, supra note 19, at 235.
85. Greenberg, supra note 9, at 1288.
86. Id. at 1321–1323.
87. Id. at 1322.
88. Id. at 1322.
89. Id. at 1323.
90. Greenberg, supra note 4, at 281.
91. Greenberg, supra note 9, at 1322.
92. Id. at 1323.
93. Greenberg, supra note 4, at 289.
94. Id. at 289.
95. Greenberg, supra note 9, at 1322.
96. One might deny that there could be such discretions, but that would require the further, highly controversial, Dworkinian premise that there is a single correct legal answer to all such questions.
97. Dworkin, supra note 1, at 406.
98. Schaus, supra note 19, at 235.
99. Id. at 236.
100. See, e.g., Lawrence Sager, Material Rights, Underenforcement, and the Adjudication Thesis, 90 B.U. L. Rev. 579 (2010).
101. Dworkin, supra note 1, at 412.
102. Schaus, supra note 19, at 237.
103. See Paul F. Scott, The National Security Constitution (2018), ch. 6.
104. And rightly, we may here suppose, owing to the institutional limitations of courts.
105. In A New Philosophy for International Law, 41 Phil. & Pub. Aff. 2 (2013), Dworkin imagines a hypothetical world court as an aid to interpreting the content of international law, a proposal that suffers from similar problems.
106. I am grateful to Adam Perry for this suggestion.
107. Notice, too, that if enforcement is substituted for some weaker notion, like applying or taking into account a norm, further over-inclusiveness is likely, since there may be many moral norms relevant to deciding cases in ways short of being enforced.
108. Hershovitz, supra note 5, at 1192.
109. Id. at 1191.
110. Id. at 1192.
111. Thanks to Nicos Stavropoulos and Scott Hershovitz for pressing this concern.
113. Nor would it be an epistemologically sound principle; see Timothy Williamson, The Philosophy of Philosophy (2007), ch. 7.
114. Williamson, supra note 113, ch. 7.
115. Greenberg, supra note 9, at 1138.
117. Dworkin, supra note 2, ch. 1.
118. Greenberg, supra note 9, at 1321–1323.
119. Id. at 1293, 1329, 1341.
120. Schaus, supra note 19, at 237.
122. Dworkin, supra note 2.
123. Hershovitz's view that legal obligations should be picked out depending on our practical purposes arguably shows some sympathy to this way of thinking. See Hershovitz, supra note 5, at 1199–1204. For discussion of considerations that may bear on such reforming projects, see Plunkett, David, Negotiating the Meaning of “Law”: The Metalinguistic Dimension of the Dispute over Legal Positivism, 22 Legal Theory 205 (2016)CrossRefGoogle Scholar.
No CrossRef data available.