Article contents
Correlativity and its Logic: Asymmetry not Equality in the Law
Published online by Cambridge University Press: 08 February 2019
Abstract
This article provides a scheme of intelligibility for correlativity, recognising its importance for analytical and normative aspects of legal relations. It considers a variety of types of normative correlativity, investigates the logic of correlativity, and distinguishes three forms of correlation involving legal rights. It undertakes careful re-examination of Aristotelian texts to reveal neglected or misrepresented insights, restores certain Hohfeldian distinctions, and argues for a more complicated relationship between correlativity and reciprocity than previously acknowledged. Specific sections employ the scheme to provide critiques of Weinrib’s use of correlativity in his understanding of private law as corrective justice, and Zylberman’s amalgam of reciprocal correlativity in his non-instrumental view of human rights. A brief concluding section notes the deep asymmetry of law and suggests an understanding of corrective justice based on asymmetry rather than equality. More speculatively, it raises doubts about the core conviction of Kantian thinking on legal and social relationships.
- Type
- Research Article
- Information
- Copyright
- Copyright © Canadian Journal of Law and Jurisprudence 2019
Footnotes
I am grateful to Rowan Cruft for a particularly stimulating comment at an early stage in the development of this project, and for subsequent helpful comments from David Frydrych, Nicole Roughan, Andrew Simester, Alec Stone Sweet, Patrick Taylor Smith and Andrew Botterell.
References
1. Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning (Yale University Press, 1919) [FLC].Google Scholar
2. Weinrib, Ernest, The Idea of Private Law, 2nd ed (Oxford University Press, 2012) [IPL] at 142.CrossRefGoogle Scholar
3. Kant, Immanuel, Metaphysics of Morals [1797], translated by Mary Gregor (Cambridge University Press, 1991) at 56;Google Scholar cited in Weinrib, ibid at 95.
4. Cane refers to Weinrib’s “fundamental idea of correlativity (or reciprocity)”—“Corrective Justice and Correlativity in Private Law” (1996) 16 Oxford J Legal Stud 471 at 471. Steiner endorses Cane when arguing for a distributive implication of Kant’s principle of right, “construable in terms of bilateral reciprocity or correlativity.”—“Corrective Rights” in Mark McBride , ed, New Essays on the Nature of Rights (Hart, 2017) at 217.
5. Zylberman, Ariel, “Why Human Rights? Because of You ” (2016) 24 J Pol Phil 32.CrossRefGoogle Scholar
6. FLC, supra note 1 at 36, 38.
7. IPL, supra note 2 at 80-83.
8. Kramer, Matthew, “Rights without Trimmings” in Kramer, Matthew H, Simmonds, NE & Steiner, Hillel, A Debate Over Rights: Philosophical Enquiries (Oxford University Press, 1998) at 24-49;Google Scholar Nigel Simmonds, “Rights at the Cutting Edge” in Kramer, Simmonds & Steiner at 222-23. The dispute between Simmonds and Kramer involved analytical disagreement over how Hohfeldian correlativity could be understood but was at least partly motivated by Simmonds’ adherence to a will theory of rights in contrast to Kramer’s support for an interest theory of rights, as his discussion on these pages reveals. Some of the controversies over correlativity are raised in a brief but valuable paper by Stepanians, Markus, “Classical and Anti-classical Views on the Relationship between Rights and Duties” in Bluhm, Roland & Nimtz, Christian, eds, Selected Papers Contributed to the Sections of Gap.5 (mentis, 2004), available at http://www.gap5.de/proceedings/html/inhalt_au.htm.Google Scholar
9. Frydrych, David, “Rights Correlativity” in Balganesh, Shyam, Sichelman, Ted & Smith, Henry, eds, The Legacy of Wesley Hohfeld: Edited Major Works, Select Personal Papers, and Original Commentaries (Cambridge University Press) [forthcoming] available at https://papers.ssrn.com/abstract=3023807.Google Scholar
10. Zylberman, Why Human Rights?, supra note 5 at 330.
11. Ignoring complications from laboratory assisted human reproduction.
12. To be rigorous, we need to restrict the phrase “child of” to refer to “child born of” rather than “child fathered by”.
13. Similarly, when the active-passive structure is applied to normative correlativity.
14. The contract will also include another arrangement, between B and A whereby B provides what amounts to the consideration for the $200. The importance of treating the correlativity within each of these arrangements separately is noted below.
15. Whatever benefit B envisaged when contracting with A to obtain payment to C.
16. The point that normatively regulated conduct covers both acts and omissions also arises below (text following note 36) in discussion of the appropriate terminology for protected liberties, often misleadingly referred to as active rights despite a number of them involving omissions.
17. Weinrib, IPL, supra note 2 at 78.
18. Ibid.
19. In the light of Jamie Dow’s assessment of Aristotle’s approach to proof in the Rhetoric being that “a pistis [proof] is successful to the extent that the conclusion is demonstrated” and “that what it is to be a good proof is (at least in part) a matter of its credentials as a demonstration”, Weinrib’s suggestion becomes even more bewildering— Dow, Jamie, Passions and Persuasion in Aristotle’s Rhetoric (Oxford University Press, 2015) at 55, 56 [emphasis in original].CrossRefGoogle Scholar
20. The consequences for Weinrib’s own position will be picked up in section III.
21. This word is used in the translation given by Weinrib, and also in the translation found in Kennedy, George, On Rhetoric: A Theory of Civic Discourse, 2nd ed (Oxford University Press, 2007).Google Scholar
22. Aristotle, Rhetoric, II 23.3, my translation—quoted in his own translation by Weinrib, IPL, supra note 2 at 78 n48.
23. The Greek here is more cumbersome as noted in the commentary in Edward Meredith Cope & John Edwin Sandys, eds, 2 Aristotle, Rhetoric (Cambridge University Press, 1877—online 2010) at 242,and as reflected in the above translation. Cope and Sandys suggest that the argument here relies on an assumption that carrying out a command is regarded as a probable consequence of being commanded. An alternative way of unpacking the argument would be: giving the command was just, so being given the command was just, so carrying out the command one had been justly given was just. The argument from correlatives strictly only applies to the first two stages, but the third stage is then treated as a necessary implication of the second, and thus relies on the proof from correlatives.
24. That is, being sold—compare (2A) above.
25. Cope and Sandys, in their commentary, supra note 23 at 242, are puzzled by the inclusion of (v) in the passage after (ii) has already been stated, but this is because they see (ii) as already providing “the general expression of the relation between agent and patient” or, as they previously put it, between what one “has done” and “the other has suffered”. They accordingly take (v) to be referring to “a particular exemplification of it, in the justification of what would otherwise be a crime”, but there is no textual justification for reading these particular details into (v). I suggest that a more probable account is as provided here, that (ii) indicates a way the proof operates but the full explanation in terms of the mutual relation between party doing and party to whom it is done is not provided until (v). In any event, Cope and Sandys do see Aristotle as providing a general explanation for how the proof from correlatives works in this passage in terms of the connection between agent and patient (or active and passive).
The lines immediately following this excerpt provide further evidence that the final sentence is giving a general account of how this proof operates, in that they then note an exception or fallacy, that might occur if the conduct connecting the two parties is not specified according to the actual parties involved. Aristotle points out that although it may be just for B to be killed by A (in which case B’s suffering the death and A’s inflicting the death are both just), it does not follow that it would be just for B to be killed by C (Rhetoric, II 23.3—prefiguring Hohfeld’s concern with the analysis of bipartite relations).
26. In the case of the selling and buying of the right to collect taxes being pronounced shameful, it may be that both parties are under a (moral) duty not to respectively sell or buy, but neither party has a claim-right that the other complies with the duty.
27. In the example found in the continuation of the excerpted passage, where it is unjust for B to be killed by C, we can identify a right in B and a duty in C.
28. As legal and moral normative schemes are commonly accepted as doing.
29. Some conduct may be evaluated within a normative scheme as being good or just without being required. Conduct falling under supernumerary duties is an obvious case.
30. The one perspective entails the other perspective over the same conduct. As Zylberman, Why Human Rights?, supra note 5 at 324, puts it, they are part of “an interdependent inferential network”.
31. Another such issue is the matter of conceptual priority between right and duty. For discussion of both issues, see Frydrych, Rights Correlativity, supra note 9, and Stepanians, Classical and Anti-classical Views, supra note 8.
32. This is evident in contemporary debates over rights-based duties and duty-based rights, where the basic correlation between right and duty is taken to depend on a justificatory priority found in the right—or, in the duty. Examples and further discussion are provided by Frydrych and Stepanians (previous note).
33. This simple distinction between justification and normative requirement is obscured by employing the same word for both as they affect a particular party. So, right and right instead of justificatory interest and right; duty and duty instead of responsibility and duty.
34. FLC, supra note 1 at 36. The remaining two, power-liability and immunity-disability covering changes in legal relations, are briefly referred to in note 85 below.
35. See Finnis, John, “Some Professorial Fallacies about Rights” (1972) 4 Adel L Rev 377;Google Scholar Matthew Kramer, “Appendix: Getting Hohfeld Right” in Kramer, Simmonds & Steiner, supra note 8.
36. FLC, supra note 1 at 36. The complete correlation of liberty-(no-right) is regarded by Hohfeld as the negation of a (claim-right)-duty correlation, with the liberty holder now being permitted to do what as a former duty holder he was prohibited from doing, and the former claim-right holder who was previously entitled to the performance of the duty now having no-right in that respect.
37. Ted Sichelman, “Very Tight ‘Bundles of Sticks’: Hohfeld’s Complex Jural Relations” in Balganesh, Sichelman & Smith, supra note 9.
38. A protected liberty is a viable alternative term (not suffering from the limitation of “active right” noted above) but this rather downplays the importance of the protection, the constraints on others, in securing the enjoyment of this type of right. “Positive” here refers to the law’s positive securing of the right and is not to be confused with its use in positive human rights, which impose a positive burden to provide welfare.
39. Although it is ultimately a sound negation.
40. If a party is under no duty to do something then it follows that the party is free (has a liberty) not to do it. This is a standard implication within systems of deontic logic, and is naturally drawn since the positive opportunity provided by the liberty is of more interest than the mere absence of a duty.
41. Hurd, Heidi, Combat, Moral: The Dilemma of Legal Perspectivalism (Cambridge University Press, 1999).CrossRefGoogle Scholar
42. Ibid at 33, 281. This move lacks any justification from Hurd, and is unjustifiable. Hohfeld himself did not use liberty in normatively inert situations—see FLC, supra note 1 at 39—and Hurd herself acknowledges the normative significance of a Hohfeldian liberty in analyzing one component of a full positive right, styled by her as a permission, at 280 n7.
43. Ibid at 32, 280.
44. To assist with the flow of the discussion, we shall maintain a consistent use of liberty, where Hurd herself opts for permission.
45. Heidi Hurd & Michael Moore, “The Hohfeldian Analysis of Rights” (2018) 63 Am J Juris 295.
46. Hurd, Moral Combat, supra note 41 at 3.
47. Ibid at 3-4.
48. Ibid at 32.
49. Although immediately before the excerpt it is abundantly clear, when Hurd speaks of codependent actions.
50. Representing respectively type-(b) correlation and type-(a) correlation, which we have seen above both adhere to basic normative correlativity and the associated logic of correlativity, dependent upon finding an active-passive structure tied to a single occurrence of conduct.
51. Hurd, Moral Combat, supra note 41 at 271-72.
52. Including its negation, type-(b) liberty-(no-right) correlation.
53. IPL, supra note 2 at 57-58.
54. Nicomachean Ethics, V, 5, translation by H Rackham in the Loeb edition. This immediately follows a passage Weinrib himself cites elsewhere as demonstrating, “Correlativity … [being] first highlighted in Aristotle’s account of corrective justice” (Weinrib, Corrective Justice (Oxford University Press, 2012) at 15, referring to Nicomachean Ethics, V, 4).
55. IPL, supra note 2 at 114.
56. Identifying corrective justice with correlativity becomes even more problematic if there are two types of correlation to contend with; the logic of correlativity, as we saw above, applies to one type of rights correlation but not the other; and relatedly, justificatory considerations operate differently within each of them; which, in turn, spills over into any inferences of equality.
57. Nicomachean Ethics, V, 4, referring to the doing and suffering of harm.
58. Aristotle’s actual discussion of correlativity occurs in the Rhetoric, in the place cited by Weinrib to introduce a logic of correlativity, which we examined in section II.C. In that passage, Aristotle provides examples of correlativity, but none drawn from corrective justice.
59. We saw in section II.C how this was achieved by falsely attributing no explanation to Aristotle for how normative implications arise out of the recognition of correlativity.
60. It is crucial to the success of Weinrib’s idea of private law. The presence of (claim-right)-duty correlation within private law is uncontroversial. If that manifestation of correlativity can be regarded logically as an operation of corrective justice, then treating private law as corrective justice follows close behind. The emphasis for Weinrib is on treating private law as corrective justice, rather than recognizing that private law can involve corrective justice.
61. Nicomachean Ethics, V, 4, Rackham (Loeb) translation; IPL, supra note 2 at 77. This occurs in Aristotle’s discussion of the difference between distributive justice which pays attention to the status or merit of the parties and corrective justice which ignores it.
62. IPL, supra note 2 at 77.
63. The Greek ἴσος can be translated as “equal to” or “the same as”.
64. Nicomachean Ethics, V, 4; with the discussion of voluntary transactions continuing into chapter 5.
65. IPL, supra note 2 at 78, 142.
66. Ibid at 78.
67. Ibid at 80-83.
68. Ibid at 78 (in his comment on the key passage from the Rhetoric).
69. The same error can be found in Martin Stone, “The Significance of Doing and Suffering” in Postema, Gerald, ed, Philosophy and the Law of Torts (Cambridge University Press, 2001) at 157, 159, 160.CrossRefGoogle Scholar The treatment of correlative parties as equals is assisted by Weinrib’s choice of the mirror image metaphor for correlativity (IPL, supra note 2 at pp xi, 144), but sore knuckles are not the mirror image of a bloody nose (text at note 13 above), nor is receiving the benefit of a claim-right the mirror image of discharging the burden of a duty.
70. Why Human Rights?, supra note 5 at 328.
71. Ibid at 329.
72. Ibid.
73. Ibid.
74. Moreover, the specific variations are not entailed by the initial gift: different possibilities are open in fulfilling the requirement of reciprocity. Even if, from some warped sense of humour, you give me the very thing I gave you as a present, it cannot be the identical present in that it is now older, pre-owned, and altogether amounts to a different experience from the one you enjoyed some time before.
75. Why Human Rights?, supra note 5 at 328.
76. Ibid at 338. Zylberman includes these as human rights.
77. Even where A and B are connected as contracting parties by the same contract, their respective duties are not coexisting or mutually inferable from each other. These aspects of correlativity hold twice in a contract, between the claim-right and duty to perform a contractual obligation in each direction, but not between the two duties. Of course, both duties are inferable from the contract, but that is a different matter. As for coexistence, the duties will coexist in the contractual document, but need not coexist thereafter: one may be (and usually is) discharged before the other.
78. Why Human Rights?, supra note 5 at 330.
79. Ibid. In his earlier discussion (at 324), Zylberman appears unequivocal in referring to an example of basic correlativity: “For me to have a claim right to something entails and is entailed by the correlative directed duty of another.” However, a switch occurs by the time Reciprocity is being articulated.
80. B’s duty to respect A’s independence amounts to B’s duty not to interfere with A’s independent conduct, and A’s right to independence amounts to A’s liberty to engage in independent conduct.
81. The example is given by Zylberman, Why Human Rights?, supra note 5 at 336, alongside other illustrations involving the conduct of B, such as killing and use of force.
82. See section II.E-F above.
83. Zylberman, Why Human Rights?, supra note 5 at 343, does not aim to provide a fully fledged scheme of human rights but merely to set the direction for a viable project, so the possibilities for disputes and conflicts over (and between) human rights, that increase as the detail is filled in (as borne out by other such projects that do grapple with increasing levels of detail) are being kept out of view.
84. As examples, see Ripstein, Arthur, Force and Freedom: Kant’s Legal and Political Philosophy (Harvard University Press, 2009);CrossRefGoogle Scholar Sweet, Alec Stone & Ryan, Clare, A Cosmopolitan Legal Order: Kant, Constitutional Justice, and the ECHR (Oxford University Press, 2018).Google Scholar
85. The precise nature of the correlativity here across the range of disputes reaching judgment could be explored further but need not detain us for present purposes. Briefly, it can be mentioned that the basic type of correlativity applies to (claim-right)-duty and liberty-(no-right) cases; that at the point of judgment in a type-(c) protected liberty dispute what will standardly be at issue is the availability of a specific instance of protection thus reverting to type-(a) (claim-right)-duty correlativity; and, that for power-liability and immunity-disability disputes the central issue can be expressed in terms of an active-passive structure correlativity over the changing and being changed of legal relations.
86. Reinforcing the distinction emphasized at the conclusion of the critique of Weinrib: a normative evaluation equally applied to both differs from an evaluation of both as equals.
87. For Aristotle, distributive justice does require an express standard by which the just distribution is to be effected, for which equality is only one candidate (Nicomachean Ethics, V, 3—discussed in section III above). Even in a case of redistribution guided by equality there is an obvious asymmetry in the initial unjust holdings and also an asymmetry in the redistributed holdings between those who attain an equal holding through grant and those who attain it through deprivation. A valuable reflection on asymmetry in discrimination law is provided by Colin Campbell & Dale Smith, “Deliberative Freedoms and the Asymmetric Features of Anti-Discrimination Law” (2017) 67 UTLJ 247.
88. IPL, supra note 2 at xvi, 84.
89. Ibid at 142, 74.
90. No identification of corrective justice with correlativity; no outstanding puzzle over a logic of correlativity; no justificatory considerations intrinsic to correlativity due to its identification with corrective justice; and, no issue of equality between the parties arising for corrective justice and correlativity.
91. Interestingly, Steiner, supra note 4 at 215, 217, suggests that a blind spot to distributive implications is attributable to “a key misstep in Kant’s own reasoning about rights”. This suggests that the opening up of correlativity to distributive considerations is, as Steiner himself argues, more broadly important for an effective understanding of the Kantian principle of right.
92. See text at note 3 above.
93. IPL, supra note 2 at 82.
- 6
- Cited by