The logic of egalitarian norms

Boston University Law Review

Volumn 80, Issue 3, 2000, Pages 693-771

  Simons, Kenneth W ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0034404852     PISSN: 00068047     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (321)

1. Christopher J. Peters, Equality Revisited, 110 HARV. L. REV. 1210 (1997). Peters specifically targets what he calls "prescriptive" equality. See discussion infra Part III.C. Kent Greenawalt has replied to certain aspects of Peters' thesis. See Kent Greenawalt, "Prescriptive Equality": Two Steps Forward, 110 HARV. L. REV. 1265 (1997). Another reply is Joshua D. Sarnoff, Equality as Uncertainty, 84 IOWA L. REV. 377 (1999). Peters responded in Slouching Towards Equality, 84 IOWA L. REV. 801 (1999), and Sarnoff replied again in I Come to Praise Morality, Not To Bury It, 84 IOWA L. REV. 819 (1999).
2. Christopher J. Peters, Equality Revisited, 110 HARV. L. REV. 1210 (1997). Peters specifically targets what he calls "prescriptive" equality. See discussion infra Part III.C. Kent Greenawalt has replied to certain aspects of Peters' thesis. See Kent Greenawalt, "Prescriptive Equality": Two Steps Forward, 110 HARV. L. REV. 1265 (1997). Another reply is Joshua D. Sarnoff, Equality as Uncertainty, 84 IOWA L. REV. 377 (1999). Peters responded in Slouching Towards Equality, 84 IOWA L. REV. 801 (1999), and Sarnoff replied again in I Come to Praise Morality, Not To Bury It, 84 IOWA L. REV. 819 (1999).
3. Christopher J. Peters, Equality Revisited, 110 HARV. L. REV. 1210 (1997). Peters specifically targets what he calls "prescriptive" equality. See discussion infra Part III.C. Kent Greenawalt has replied to certain aspects of Peters' thesis. See Kent Greenawalt, "Prescriptive Equality": Two Steps Forward, 110 HARV. L. REV. 1265 (1997). Another reply is Joshua D. Sarnoff, Equality as Uncertainty, 84 IOWA L. REV. 377 (1999). Peters responded in Slouching Towards Equality, 84 IOWA L. REV. 801 (1999), and Sarnoff replied again in I Come to Praise Morality, Not To Bury It, 84 IOWA L. REV. 819 (1999).
4. Christopher J. Peters, Equality Revisited, 110 HARV. L. REV. 1210 (1997). Peters specifically targets what he calls "prescriptive" equality. See discussion infra Part III.C. Kent Greenawalt has replied to certain aspects of Peters' thesis. See Kent Greenawalt, "Prescriptive Equality": Two Steps Forward, 110 HARV. L. REV. 1265 (1997). Another reply is Joshua D. Sarnoff, Equality as Uncertainty, 84 IOWA L. REV. 377 (1999). Peters responded in Slouching Towards Equality, 84 IOWA L. REV. 801 (1999), and Sarnoff replied again in I Come to Praise Morality, Not To Bury It, 84 IOWA L. REV. 819 (1999).
5. Christopher J. Peters, Equality Revisited, 110 HARV. L. REV. 1210 (1997). Peters specifically targets what he calls "prescriptive" equality. See discussion infra Part III.C. Kent Greenawalt has replied to certain aspects of Peters' thesis. See Kent Greenawalt, "Prescriptive Equality": Two Steps Forward, 110 HARV. L. REV. 1265 (1997). Another reply is Joshua D. Sarnoff, Equality as Uncertainty, 84 IOWA L. REV. 377 (1999). Peters responded in Slouching Towards Equality, 84 IOWA L. REV. 801 (1999), and Sarnoff replied again in I Come to Praise Morality, Not To Bury It, 84 IOWA L. REV. 819 (1999).
6. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
7. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
8. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
9. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
10. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
11. Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537 (1982). Westen's article prompted numerous replies. See, e.g., Steven J. Burton, Comment on "Empty Ideas": Logical Positivist Analyses of Equality and Rules, 91 YALE L.J. 1136 (1982); Erwin Chemerinsky, In Defense of Equality: A Reply to Professor Westen, 81 MICH. L. REV. 575 (1983); Kent Greenawalt, How Empty is the Idea of Equality?, 83 COLUM. L. REV. 1167 (1983); Kenneth L. Karst, Why Equality Matters, 17 GA. L. REV. 245 (1983). Westen's writings culminated with the publication of his book in 1990. See PETER WESTEN, SPEAKING OF EQUALITY: AN ANALYSIS OF THE RHETORICAL FORCE OF "EQUALITY" IN MORAL AND LEGAL DISCOURSE (1990) [hereinafter WESTEN, SPEAKING OF EQUALITY].
12. Westen's original article emphasizes the emptiness of equality, while his book puts much greater stress on the complexity, variety, and potential confusion of egalitarian norms. Compare Westen, The Empty Idea of Equality, supra note 2, with WESTEN, SPEAKING OF EQUALITY, supra note 2. The evolution of Westen's ideas is revealed by the fact that Westen does not cite his original essay, The Empty Idea of Equality, in the bibliography to his book, and he explicitly disowns the title in his Preface. See id. at xix-xx (asserting that equality is not meaningless but instead derivative, and acknowledging the usefulness of a certain class of equality rights and of the presumption of equality). However, Westen's "Empty Idea of Equality" essay continues to be cited with much greater frequency than his book.
13. Westen's original article emphasizes the emptiness of equality, while his book puts much greater stress on the complexity, variety, and potential confusion of egalitarian norms. Compare Westen, The Empty Idea of Equality, supra note 2, with WESTEN, SPEAKING OF EQUALITY, supra note 2. The evolution of Westen's ideas is revealed by the fact that Westen does not cite his original essay, The Empty Idea of Equality, in the bibliography to his book, and he explicitly disowns the title in his Preface. See id. at xix-xx (asserting that equality is not meaningless but instead derivative, and acknowledging the usefulness of a certain class of equality rights and of the presumption of equality). However, Westen's "Empty Idea of Equality" essay continues to be cited with much greater frequency than his book.
14. See, e.g., John Kane, Basal Inequalities: Reply to Sen, 24 POL. THEORY 401 (1996) [hereinafter Kane, Basal Inequalities]; John Kane, Justice, Impartiality, and Equality: Why the Concept of Justice Does Not Presume Equality, 24 POL. THEORY 375 (1996) [hereinafter Kane, Justice]; Madison Powers, Forget about Equality, 6 KENNEDY INST. ETHICS J. 129 (1996); Amartya Sen, On the Status of Equality, 24 POL. THEORY 394 (1996).
15. See, e.g., John Kane, Basal Inequalities: Reply to Sen, 24 POL. THEORY 401 (1996) [hereinafter Kane, Basal Inequalities]; John Kane, Justice, Impartiality, and Equality: Why the Concept of Justice Does Not Presume Equality, 24 POL. THEORY 375 (1996) [hereinafter Kane, Justice]; Madison Powers, Forget about Equality, 6 KENNEDY INST. ETHICS J. 129 (1996); Amartya Sen, On the Status of Equality, 24 POL. THEORY 394 (1996).
16. See, e.g., John Kane, Basal Inequalities: Reply to Sen, 24 POL. THEORY 401 (1996) [hereinafter Kane, Basal Inequalities]; John Kane, Justice, Impartiality, and Equality: Why the Concept of Justice Does Not Presume Equality, 24 POL. THEORY 375 (1996) [hereinafter Kane, Justice]; Madison Powers, Forget about Equality, 6 KENNEDY INST. ETHICS J. 129 (1996); Amartya Sen, On the Status of Equality, 24 POL. THEORY 394 (1996).
17. See, e.g., John Kane, Basal Inequalities: Reply to Sen, 24 POL. THEORY 401 (1996) [hereinafter Kane, Basal Inequalities]; John Kane, Justice, Impartiality, and Equality: Why the Concept of Justice Does Not Presume Equality, 24 POL. THEORY 375 (1996) [hereinafter Kane, Justice]; Madison Powers, Forget about Equality, 6 KENNEDY INST. ETHICS J. 129 (1996); Amartya Sen, On the Status of Equality, 24 POL. THEORY 394 (1996).
18. "Precedential" equality is my own nomenclature. Peters describes this principle as "prescriptive" equality. See infra text accompanying note 132.
19. Consider two recent survey articles on the topic. Richard Arneson identifies two aspects of the ideal of equality: equality of democratic citizenship, which assures equal basic rights to all members of society, and equality of conditions or life prospects. Richard J. Arneson, Equality and Equal Opportunity for Welfare, 56 PHIL. STUD. 77 (1989). Brian Barry distinguishes three roles that the concept of equality plays within political philosophy: "Equal treatment for equals" specifies that those who are equal in some relevant respect should be treated equally. "Fundamental equality" refers to the idea that all human beings are of equal worth, have equal (fundamental) rights, or should be awarded equal respect and concern. "Social equality" [includes] political equality, economic equality, and equality of status among the members of a society. Brian Barry, Equality, 1 ENCYCLOPEDIA OF ETHICS 322-23 (Lawrence C. Becker ed. 1992). Barry then suggests that, while the phrase "equal treatment for equals" captures a significant sense of equality, it is an empty formula until the relevant "equals" are specified. He concludes that the popular demands for equality must refer to one of the other senses of equality. See id. at 323. My analysis of the first formula is a bit different. See infra text accompanying notes 107-28.
20. Consider two recent survey articles on the topic. Richard Arneson identifies two aspects of the ideal of equality: equality of democratic citizenship, which assures equal basic rights to all members of society, and equality of conditions or life prospects. Richard J. Arneson, Equality and Equal Opportunity for Welfare, 56 PHIL. STUD. 77 (1989). Brian Barry distinguishes three roles that the concept of equality plays within political philosophy: "Equal treatment for equals" specifies that those who are equal in some relevant respect should be treated equally. "Fundamental equality" refers to the idea that all human beings are of equal worth, have equal (fundamental) rights, or should be awarded equal respect and concern. "Social equality" [includes] political equality, economic equality, and equality of status among the members of a society. Brian Barry, Equality, 1 ENCYCLOPEDIA OF ETHICS 322-23 (Lawrence C. Becker ed. 1992). Barry then suggests that, while the phrase "equal treatment for equals" captures a significant sense of equality, it is an empty formula until the relevant "equals" are specified. He concludes that the popular demands for equality must refer to one of the other senses of equality. See id. at 323. My analysis of the first formula is a bit different. See infra text accompanying notes 107-28.
21. Peters uses the term "nonegalitarian justice" to mean "treatment of a person in accordance with the net effect of all the relevant criteria and only the relevant criteria, provided that considerations of nontautological equality cannot be relevant criteria." Peters, Equality Revisited, supra note 1, at 1228 (emphasis omitted).
22. Peters has a similarly impoverished view of John Rawls' difference principle. See infra text accompanying notes 152-59.
23. This Article addresses equality in both law and morality, and it provides numerous examples from each domain. The conceptual and general nature of my argument justifies employing the same analysis in both domains. Of course, many moral theories will support egalitarian norms that would be infeasible or unwise to recognize as legal rights or duties. Similarly, many legal institutions might value forms of equality (such as requirements of impartiality in administering discretionary benefits or burdens) that foundational moral theories would scant.
24. I will also point out occasions when theorists use the term "equality" in a different sense.
25. See Westen, The Empty Idea of Equality, supra note 2, at 547-48 (describing equality as "an empty vessel with no substantive moral content of its own" and "a simple tautology"); Peter Westen, On "Confusing Ideas": Reply, 91 YALE L.J. 1153 (1982) (arguing that the principle of equality is both "empty and confusing"); Peter Westen, The Meaning of Equality in Law, Science, Math & Morals: A Reply, 81 MICH. L. REV. 604, 618 (1983) ("The terms 'equal' and 'unequal' in law and morals are nothing but 'rhetorical' devices for talking about legal and moral rules."); Peter Westen, To Lure the Tarantula from Its Hold: A Response, 83 COLUM. L. REV. 1186, 1188 (1983) (urging that emptiness is a "logical consequence of the way equality is lexically defined") [hereinafter Westen, A Response].
26. See Westen, The Empty Idea of Equality, supra note 2, at 547-48 (describing equality as "an empty vessel with no substantive moral content of its own" and "a simple tautology"); Peter Westen, On "Confusing Ideas": Reply, 91 YALE L.J. 1153 (1982) (arguing that the principle of equality is both "empty and confusing"); Peter Westen, The Meaning of Equality in Law, Science, Math & Morals: A Reply, 81 MICH. L. REV. 604, 618 (1983) ("The terms 'equal' and 'unequal' in law and morals are nothing but 'rhetorical' devices for talking about legal and moral rules."); Peter Westen, To Lure the Tarantula from Its Hold: A Response, 83 COLUM. L. REV. 1186, 1188 (1983) (urging that emptiness is a "logical consequence of the way equality is lexically defined") [hereinafter Westen, A Response].
27. See Westen, The Empty Idea of Equality, supra note 2, at 547-48 (describing equality as "an empty vessel with no substantive moral content of its own" and "a simple tautology"); Peter Westen, On "Confusing Ideas": Reply, 91 YALE L.J. 1153 (1982) (arguing that the principle of equality is both "empty and confusing"); Peter Westen, The Meaning of Equality in Law, Science, Math & Morals: A Reply, 81 MICH. L. REV. 604, 618 (1983) ("The terms 'equal' and 'unequal' in law and morals are nothing but 'rhetorical' devices for talking about legal and moral rules."); Peter Westen, To Lure the Tarantula from Its Hold: A Response, 83 COLUM. L. REV. 1186, 1188 (1983) (urging that emptiness is a "logical consequence of the way equality is lexically defined") [hereinafter Westen, A Response].
28. See Westen, The Empty Idea of Equality, supra note 2, at 547-48 (describing equality as "an empty vessel with no substantive moral content of its own" and "a simple tautology"); Peter Westen, On "Confusing Ideas": Reply, 91 YALE L.J. 1153 (1982) (arguing that the principle of equality is both "empty and confusing"); Peter Westen, The Meaning of Equality in Law, Science, Math & Morals: A Reply, 81 MICH. L. REV. 604, 618 (1983) ("The terms 'equal' and 'unequal' in law and morals are nothing but 'rhetorical' devices for talking about legal and moral rules."); Peter Westen, To Lure the Tarantula from Its Hold: A Response, 83 COLUM. L. REV. 1186, 1188 (1983) (urging that emptiness is a "logical consequence of the way equality is lexically defined") [hereinafter Westen, A Response].
29. Westen, The Empty Idea of Equality, supra note 2, at 539 (using "equality" to mean the proposition in law and morals that "people who are alike should be treated alike").
30. Id. at 545 ("[I]f a moral standard prescribes some uniform treatment for all members of a class (for instance, that all Rhodes Scholars shall receive fellowships to Magdalen College, Oxford), equal treatment is achieved by either uniformly granting or uniformly denying that treatment to all members.").
31. Id. at 542.
32. Id. at 547-48. As Westen argues: To say that two persons are the same in a certain respect is to presuppose a rule - a prescribed standard for treating them - that both fully satisfy. Before such a rule is established, no standard of comparison exists. After such a rule is established, equality between them is a "logical consequence" of the established rule. They are then "equal" in respect of the rule because that is what equal means: "Equally" means "'according to one and the same rule.'" They are also then entitled to equal treatment under the rule because that is what possessing a rule means: "To conform to a rule is (tautologically) to apply it to the cases to which it applies." Id. at 548 (footnotes omitted). Peters agrees that such "traditional" equality arguments state the tautology that "'People identically entitled to a particular treatment are identically entitled to that treatment.'" Peters, Equality Revisited, supra note 1, at 1217.
33. See Westen, The Empty Idea of Equality, supra note 2, at 580-81. For example, equality encourages vague equal protection standards for the evaluation of fundamental rights when substantive due process would be a clearer doctrinal approach. Id. at 581 n.154; see also Zablocki v. Redhail, 434 U.S. 374, 395-96 (1978) (arguing that the application of equal protection to substantive due process "invites mechanical or thoughtless application of misfocused doctrine").
34. Id. at 581-84. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 269-70.
35. Id. at 266-67. Westen argues that "the assertion 'all men are equal' is capable of stating both an 'is' and an 'ought': an 'is' if it means 'all men are descriptively equal in their possession of certain empirical traits'; an 'ought' if it means 'all men are prescriptively equal in certain ways in which they ought to be treated.'" Similarly, Westen asserts, prescriptive equality is often grounded on descriptive equality: the claim that people "ought" to be treated equally is grounded on the empirical claim that they "are" equal. See id. at 266.
36. Id. at 271-74 ("A person who advocates something controversial in the name of equality benefits from the other equalities that his audience accepts.").
37. Id. at 274-80.
38. According to Westen: [Douglas] could have responded, not by defending inequality, but by advocating a competing equality such as the equal rights of the citizens of each state to decide for themselves whether to abolish slavery, or the equal right of all persons (free and slave, white and black) to receive the treatments to which they were entitled under state and federal law. Id. at 282.
39. Id. at 283. Joseph Raz asserts that "intellectual confusion" is the price of employing two very different conceptions of egalitarianism - the "rhetorical" and the "strict." RAZ, THE MORALITY OF FREEDOM 228 (1986). This distinction essentially tracks my distinction between lexical equalities and comparative equality rights. See infra text accompanying notes 107-28. However, insofar as Raz finds comparative equality rights to be meaningful, I do not count his observation about confusion as a criticism of equality.
40. Peters, Equality Revisited, supra note 1, at 1213.
41. See Westen, The Empty Idea of Equality, supra note 2, at 542 ("[T]he idea of equality is logically indistinguishable from the standard formula of distributive justice"); WESTEN, SPEAKING OF EQUALITY, supra note 2, at xix ("[J]ust as all normative disputes are about equality, none is really about equality."). Peters purports to offer an argument that differs from Westen's insofar as Peters focuses on successive rather than simultaneous unequal treatments. But Westen also addressed this situation within his category of "prescriptions as baselines within descriptive standards of comparison." WESTEN, SPEAKING OF EQUALITY, supra note 2, at 87-89, 98-99, 189-92, 219-23.
42. See Westen, The Empty Idea of Equality, supra note 2, at 542 ("[T]he idea of equality is logically indistinguishable from the standard formula of distributive justice"); WESTEN, SPEAKING OF EQUALITY, supra note 2, at xix ("[J]ust as all normative disputes are about equality, none is really about equality."). Peters purports to offer an argument that differs from Westen's insofar as Peters focuses on successive rather than simultaneous unequal treatments. But Westen also addressed this situation within his category of "prescriptions as baselines within descriptive standards of comparison." WESTEN, SPEAKING OF EQUALITY, supra note 2, at 87-89, 98-99, 189-92, 219-23.
43. See Westen, The Empty Idea of Equality, supra note 2, at 542 ("[T]he idea of equality is logically indistinguishable from the standard formula of distributive justice"); WESTEN, SPEAKING OF EQUALITY, supra note 2, at xix ("[J]ust as all normative disputes are about equality, none is really about equality."). Peters purports to offer an argument that differs from Westen's insofar as Peters focuses on successive rather than simultaneous unequal treatments. But Westen also addressed this situation within his category of "prescriptions as baselines within descriptive standards of comparison." WESTEN, SPEAKING OF EQUALITY, supra note 2, at 87-89, 98-99, 189-92, 219-23.
44. See, e.g., Kane, Basal Inequalities, supra note 4, at 405 (claiming that "the establishment of a genuine principle of moral equality requires substantive argument that is logically prior to the conception of justice that the principle informs"); Kane, Justice, supra note 4, at 376 ("[T]here is nothing in the basic concept of justice that implies anything of any ethical import about the equality of human beings."); Louis Pojman, Theories of Equality: A Critical Analysis, 23 BEHAV. & PHIL. 2, 25 (1995) ("When the 'principle of equality' is analyzed carefully, it becomes either an empty formal rule or becomes the endorsement of substantive values having nothing in particular to do with equality . . . . On reflection few of us value equality as an intrinsic good."); Sen, supra note 4 (criticizing Kane's analysis but not his conclusion that justice exists independent of equality).
45. See, e.g., Kane, Basal Inequalities, supra note 4, at 405 (claiming that "the establishment of a genuine principle of moral equality requires substantive argument that is logically prior to the conception of justice that the principle informs"); Kane, Justice, supra note 4, at 376 ("[T]here is nothing in the basic concept of justice that implies anything of any ethical import about the equality of human beings."); Louis Pojman, Theories of Equality: A Critical Analysis, 23 BEHAV. & PHIL. 2, 25 (1995) ("When the 'principle of equality' is analyzed carefully, it becomes either an empty formal rule or becomes the endorsement of substantive values having nothing in particular to do with equality . . . . On reflection few of us value equality as an intrinsic good."); Sen, supra note 4 (criticizing Kane's analysis but not his conclusion that justice exists independent of equality).
46. See, e.g., Kane, Basal Inequalities, supra note 4, at 405 (claiming that "the establishment of a genuine principle of moral equality requires substantive argument that is logically prior to the conception of justice that the principle informs"); Kane, Justice, supra note 4, at 376 ("[T]here is nothing in the basic concept of justice that implies anything of any ethical import about the equality of human beings."); Louis Pojman, Theories of Equality: A Critical Analysis, 23 BEHAV. & PHIL. 2, 25 (1995) ("When the 'principle of equality' is analyzed carefully, it becomes either an empty formal rule or becomes the endorsement of substantive values having nothing in particular to do with equality . . . . On reflection few of us value equality as an intrinsic good."); Sen, supra note 4 (criticizing Kane's analysis but not his conclusion that justice exists independent of equality).
47. See Westen, The Empty Idea of Equality, supra note 2, at 558; see also WESTEN, SPEAKING OF EQUALITY, supra note 2, at xiii (arguing that omission of "equality" from the proposed Equal Rights Amendment would not change the meaning of the Amendment).
48. See Westen, The Empty Idea of Equality, supra note 2, at 558; see also WESTEN, SPEAKING OF EQUALITY, supra note 2, at xiii (arguing that omission of "equality" from the proposed Equal Rights Amendment would not change the meaning of the Amendment).
49. Peters, Equality Revisited, supra note 1, at 1258.
50. Westen, The Empty Idea of Equality, supra note 2, at 540 (footnote omitted).
51. Peters, Equality Revisited, supra note 1, at 1228 (emphasis omitted). Peters believes that his formulation differs from Westen's insofar as "nontautological equality" differs from "likes should be treated alike." Peters, however, ignores the more complex analysis in Westen's book Speaking of Equality. See supra note 24.
52. Peters, Equality Revisited, supra note 1, at 1228 (emphasis omitted). Peters believes that his formulation differs from Westen's insofar as "nontautological equality" differs from "likes should be treated alike." Peters, however, ignores the more complex analysis in Westen's book Speaking of Equality. See supra note 24.
53. 118 U.S. 356 (1886).
54. Peters, Equality Revisited, supra note 1, at 1219-20; see also id. at 1258-59. I argue below that this "single person reductio" is not a logical flaw as Peters suggests, because equality principles sometimes legitimately apply to the unjust treatment of a single person. See infra text accompanying notes 137-43.
55. Peters, supra note 1, at 1212; see also id. at 1225-27, 1250, 1252, 1257, 1263-64. The wrong, Peters explains, can take the form of undeserved beneficial treatment as well as undeserved detrimental treatment. Id. at 1212.
56. Consider the following example from William Frankena: "If a ruler were to boil his subjects in oil, jumping in afterwards himself, it would be an injustice, but it would be no inequality in treatment." William K. Frankena, The Concept of Social Justice, in SOCIAL JUSTICE 17 (Richard B. Brandt ed. 1962). Westen cites Frankena and others to illustrate how equality can be unjust. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 90-92. The objection is often asserted by death penalty proponents. One major argument against the death penalty is that it is administered in a highly arbitrary way. And one form of arbitrariness is the unjustified imposition of a death sentence on some but not all of those who deserve it. Ernest van den Haag has forcefully responded with the multiplication of wrongs objection: [Some] appear to prefer equal injustice - letting all get away with murder if some do - to unequal justice: punishing some guilty offenders according to desert, even if others get away. Equal justice is best, but unattainable. Unequal justice is our lot in this world. It is the only justice we can ever have, for not all murderers can be apprehended or convicted, or sentenced equally in different courts. We should constantly try to bring every offender to justice. But meanwhile unequal justice is the only justice we have, and certainly better than unequal injustice - giving no murderer the punishment his crime deserves. Ernest van den Haag, Refuting Reiman and Nathanson, in PUNISHMENT AND THE DEATH PENALTY 207, 214 (Robert M. Baird & Stuart E. Rosenbaum eds. 1995); see also Ernest van den Haag, The Collapse of the Case against Capital Punishment, NAT'L REV., Mar 31, 1978, at 397: [T]he abolitionist argument from capriciousness, or discretion, or discrimination, would be more persuasive if it were alleged that those selectively executed are not guilty. But the argument merely maintains that some other guilty but more favored persons, or groups, escape the death penalty. This is hardly sufficient for letting anyone else found guilty escape the penalty. On the contrary, that some guilty persons or groups elude it argues for extending the death penalty to them. Id.
57. Consider the following example from William Frankena: "If a ruler were to boil his subjects in oil, jumping in afterwards himself, it would be an injustice, but it would be no inequality in treatment." William K. Frankena, The Concept of Social Justice, in SOCIAL JUSTICE 17 (Richard B. Brandt ed. 1962). Westen cites Frankena and others to illustrate how equality can be unjust. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 90-92. The objection is often asserted by death penalty proponents. One major argument against the death penalty is that it is administered in a highly arbitrary way. And one form of arbitrariness is the unjustified imposition of a death sentence on some but not all of those who deserve it. Ernest van den Haag has forcefully responded with the multiplication of wrongs objection: [Some] appear to prefer equal injustice - letting all get away with murder if some do -to unequal justice: punishing some guilty offenders according to desert, even if others get away. Equal justice is best, but unattainable. Unequal justice is our lot in this world. It is the only justice we can ever have, for not all murderers can be apprehended or convicted, or sentenced equally in different courts. We should constantly try to bring every offender to justice. But meanwhile unequal justice is the only justice we have, and certainly better than unequal injustice - giving no murderer the punishment his crime deserves. Ernest van den Haag, Refuting Reiman and Nathanson, in PUNISHMENT AND THE DEATH PENALTY 207, 214 (Robert M. Baird & Stuart E. Rosenbaum eds. 1995); see also Ernest van den Haag, The Collapse of the Case against Capital Punishment, NAT'L REV., Mar 31, 1978, at 397: [T]he abolitionist argument from capriciousness, or discretion, or discrimination, would be more persuasive if it were alleged that those selectively executed are not guilty. But the argument merely maintains that some other guilty but more favored persons, or groups, escape the death penalty. This is hardly sufficient for letting anyone else found guilty escape the penalty. On the contrary, that some guilty persons or groups elude it argues for extending the death penalty to them. Id.
58. Consider the following example from William Frankena: "If a ruler were to boil his subjects in oil, jumping in afterwards himself, it would be an injustice, but it would be no inequality in treatment." William K. Frankena, The Concept of Social Justice, in SOCIAL JUSTICE 17 (Richard B. Brandt ed. 1962). Westen cites Frankena and others to illustrate how equality can be unjust. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 90-92. The objection is often asserted by death penalty proponents. One major argument against the death penalty is that it is administered in a highly arbitrary way. And one form of arbitrariness is the unjustified imposition of a death sentence on some but not all of those who deserve it. Ernest van den Haag has forcefully responded with the multiplication of wrongs objection: [Some] appear to prefer equal injustice - letting all get away with murder if some do - to unequal justice: punishing some guilty offenders according to desert, even if others get away. Equal justice is best, but unattainable. Unequal justice is our lot in this world. It is the only justice we can ever have, for not all murderers can be apprehended or convicted, or sentenced equally in different courts. We should constantly try to bring every offender to justice. But meanwhile unequal justice is the only justice we have, and certainly better than unequal injustice - giving no murderer the punishment his crime deserves. Ernest van den Haag, Refuting Reiman and Nathanson, in PUNISHMENT AND THE DEATH PENALTY 207, 214 (Robert M. Baird & Stuart E. Rosenbaum eds. 1995); see also Ernest van den Haag, The Collapse of the Case against Capital Punishment, NAT'L REV., Mar 31, 1978, at 397: [T]he abolitionist argument from capriciousness, or discretion, or discrimination, would be more persuasive if it were alleged that those selectively executed are not guilty. But the argument merely maintains that some other guilty but more favored persons, or groups, escape the death penalty. This is hardly sufficient for letting anyone else found guilty escape the penalty. On the contrary, that some guilty persons or groups elude it argues for extending the death penalty to them. Id.
59. Consider the following example from William Frankena: "If a ruler were to boil his subjects in oil, jumping in afterwards himself, it would be an injustice, but it would be no inequality in treatment." William K. Frankena, The Concept of Social Justice, in SOCIAL JUSTICE 17 (Richard B. Brandt ed. 1962). Westen cites Frankena and others to illustrate how equality can be unjust. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 90-92. The objection is often asserted by death penalty proponents. One major argument against the death penalty is that it is administered in a highly arbitrary way. And one form of arbitrariness is the unjustified imposition of a death sentence on some but not all of those who deserve it. Ernest van den Haag has forcefully responded with the multiplication of wrongs objection: [Some] appear to prefer equal injustice - letting all get away with murder if some do - to unequal justice: punishing some guilty offenders according to desert, even if others get away. Equal justice is best, but unattainable. Unequal justice is our lot in this world. It is the only justice we can ever have, for not all murderers can be apprehended or convicted, or sentenced equally in different courts. We should constantly try to bring every offender to justice. But meanwhile unequal justice is the only justice we have, and certainly better than unequal injustice - giving no murderer the punishment his crime deserves. Ernest van den Haag, Refuting Reiman and Nathanson, in PUNISHMENT AND THE DEATH PENALTY 207, 214 (Robert M. Baird & Stuart E. Rosenbaum eds. 1995); see also Ernest van den Haag, The Collapse of the Case against Capital Punishment, NAT'L REV., Mar 31, 1978, at 397: [T]he abolitionist argument from capriciousness, or discretion, or discrimination, would be more persuasive if it were alleged that those selectively executed are not guilty. But the argument merely maintains that some other guilty but more favored persons, or groups, escape the death penalty. This is hardly sufficient for letting anyone else found guilty escape the penalty. On the contrary, that some guilty persons or groups elude it argues for extending the death penalty to them. Id.
60. Peters' objections to equality were first stated in the context of his objection to using equality to justify precedential constraints. See Christopher J. Peters, Foolish Consistency: On Equality, Integrity, and Justice in Stare Decisis, 105 YALE L.J. 2031, 2055-73 (1996) [hereinafter Peters, Foolish Consistency]; see also Peters, Equality Revisited, supra note 1, at 1226 n.32, 1263.
61. Peters' objections to equality were first stated in the context of his objection to using equality to justify precedential constraints. See Christopher J. Peters, Foolish Consistency: On Equality, Integrity, and Justice in Stare Decisis, 105 YALE L.J. 2031, 2055-73 (1996) [hereinafter Peters, Foolish Consistency]; see also Peters, Equality Revisited, supra note 1, at 1226 n.32, 1263.
62. Larry Alexander, Constrained by Precedent, 63 S. CAL. L. REV. 1, 10 (1989); see also Larry Alexander, Bad Beginnings, 145 U. PA. L. REV. 57, 85-86 (1996) (arguing that equality is not served by repeating past moral mistakes); Larry Alexander, Striking Back at the Empire: A Brief Survey of Problems in Dworkin's Theory of Law, 6 L. & PHIL. 419, 427-30 (1987) (criticizing the overemphasis on equality as an independent moral value in determining political morality); David Lyons, Formal Justice and Judicial Precedent, 38 VAND. L. REV. 495, 505, 508 (1985) (criticizing the formal justice argument that precedent should be followed to promote consistency).
63. Larry Alexander, Constrained by Precedent, 63 S. CAL. L. REV. 1, 10 (1989); see also Larry Alexander, Bad Beginnings, 145 U. PA. L. REV. 57, 85-86 (1996) (arguing that equality is not served by repeating past moral mistakes); Larry Alexander, Striking Back at the Empire: A Brief Survey of Problems in Dworkin's Theory of Law, 6 L. & PHIL. 419, 427-30 (1987) (criticizing the overemphasis on equality as an independent moral value in determining political morality); David Lyons, Formal Justice and Judicial Precedent, 38 VAND. L. REV. 495, 505, 508 (1985) (criticizing the formal justice argument that precedent should be followed to promote consistency).
64. Larry Alexander, Constrained by Precedent, 63 S. CAL. L. REV. 1, 10 (1989); see also Larry Alexander, Bad Beginnings, 145 U. PA. L. REV. 57, 85-86 (1996) (arguing that equality is not served by repeating past moral mistakes); Larry Alexander, Striking Back at the Empire: A Brief Survey of Problems in Dworkin's Theory of Law, 6 L. & PHIL. 419, 427-30 (1987) (criticizing the overemphasis on equality as an independent moral value in determining political morality); David Lyons, Formal Justice and Judicial Precedent, 38 VAND. L. REV. 495, 505, 508 (1985) (criticizing the formal justice argument that precedent should be followed to promote consistency).
65. Larry Alexander, Constrained by Precedent, 63 S. CAL. L. REV. 1, 10 (1989); see also Larry Alexander, Bad Beginnings, 145 U. PA. L. REV. 57, 85-86 (1996) (arguing that equality is not served by repeating past moral mistakes); Larry Alexander, Striking Back at the Empire: A Brief Survey of Problems in Dworkin's Theory of Law, 6 L. & PHIL. 419, 427-30 (1987) (criticizing the overemphasis on equality as an independent moral value in determining political morality); David Lyons, Formal Justice and Judicial Precedent, 38 VAND. L. REV. 495, 505, 508 (1985) (criticizing the formal justice argument that precedent should be followed to promote consistency).
66. Alexander, Constrained by Precedent, supra note 35, at 10.
67. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
68. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
69. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
70. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
71. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
72. For discussions of the problem, see LARRY S. TEMKIN, INEQUALITY 245, 279, 282 (1993); Arneson, supra note 6, at 501-02; Dennis McKerlie, Equality and Priority, 6 UTILITAS 25, 27 (1994); Dennis McKerlie, Critical Notice, 25 CAN. J. PHIL. 623, 633-36 (1995); Dennis McKerlie, Equality, 106 ETHICS 274, 276-77 (1996) (stating the "leveling down" objection); Derek Parfit, Equality or Priority?, The Lindley Lecture at the Univ. of Kan., at 17-18, 23-24 (Nov. 21, 1991); Pojman, supra note 25, at 11; Madison Powers, Forget about Equality, 6 KENNEDY INST. OF ETHICS J. 129, 132 (1996).
73. Parfit, supra note 37, at 17. Parfit continues with another example: Egalitarians would avoid this form of the objection if what they think bad is only inequality in resources. But they must admit that, on their view, it would in one way better if, in some natural disaster those who are better off lost all of their extra resources, in a way that benefited no one. This conclusion may seem almost as implausible. Id.
74. See RAZ, supra note 22, at 227, 235; see also Greenawalt, supra note 1, at 1277 n.41 (quoting Raz).
75. See Peters, Equality Revisited, supra note 1, at 1237.
76. See id.
77. See id. at 1237-39. Peters concludes that the survivors should draw lots so that each person has an equal (10/11) chance at the scarce benefit, and he asserts that this principle is grounded in nonegalitarian justice. Id. at 1240-43. Below, I will support his conclusion but challenge the assertion that nonegalitarian justice is the operative principle here. See infra text at notes 179-86.
78. For a helpful overview, see D. Kelly Weisberg, Introduction, in FEMINIST LEGAL THEORY: FOUNDATIONS 121-27, 211-20 (Weisberg ed. 1993).
79. See id.; see also MARTHA MINOW, MAKING ALL THE DIFFERENCE: INCLUSION, EXCLUSION, AND AMERICAN LAW 272, 275 (1990) (discussing the failings of formal equality in the context of divorces and family laws); Anne Dailey, Feminism's Return to Liberalism, 102 YALE L.J. 1265, 1268-71 (1993) (touching upon important themes of women's difference in feminist theory); Cynthia Ward, On Difference and Equality, 3 LEG. THEORY 65, 70-73 (1997) ("The failure to recognize certain differences has become a failure to treat all people as equal.").
80. See id.; see also MARTHA MINOW, MAKING ALL THE DIFFERENCE: INCLUSION, EXCLUSION, AND AMERICAN LAW 272, 275 (1990) (discussing the failings of formal equality in the context of divorces and family laws); Anne Dailey, Feminism's Return to Liberalism, 102 YALE L.J. 1265, 1268-71 (1993) (touching upon important themes of women's difference in feminist theory); Cynthia Ward, On Difference and Equality, 3 LEG. THEORY 65, 70-73 (1997) ("The failure to recognize certain differences has become a failure to treat all people as equal.").
81. See id.; see also MARTHA MINOW, MAKING ALL THE DIFFERENCE: INCLUSION, EXCLUSION, AND AMERICAN LAW 272, 275 (1990) (discussing the failings of formal equality in the context of divorces and family laws); Anne Dailey, Feminism's Return to Liberalism, 102 YALE L.J. 1265, 1268-71 (1993) (touching upon important themes of women's difference in feminist theory); Cynthia Ward, On Difference and Equality, 3 LEG. THEORY 65, 70-73 (1997) ("The failure to recognize certain differences has become a failure to treat all people as equal.").
82. See WEISBERG, supra note 43, at 125 (reviewing criticism of the male norm).
83. A closely related complaint is that conventional equality analysis creates an illusion of neutrality. See Mary Becker, Prince Charming: Abstract Equality, 1987 S. CT. REV. 201.
84. See Personnel Admin. of Mass. v. Feeney, 442 U.S. 256 (1979) (upholding an absolute lifetime civil service job preference for veterans with passing scores on examination, even though 98% of the beneficiaries of this preference were male).
85. See General Elec. Co. v. Gilbert, 429 U.S. 125, 133-40 (1976) (Title VII analysis); Geduldig v. Aiello, 417 U.S. 484, 496-97 (1974) (equal protection analysis).
86. Conventional equality analysis would, for example, seem inadequate to justify a legal policy forbidding employers from firing all pregnant workers.
87. See infra Part V.
88. I employ the term "normative equality" in this section rather than "prescriptive equality" simply to avoid confusion with Peters' narrower conception of "prescriptive equality," discussed below. It is unfortunate that Peters has employed the general term in such a specialized way. "Precedential equality" might have suited Peters' purposes better while avoiding confusion, as I will explain. See infra Part III.C.
89. In an earlier article, I spell out this point in more detail. See Kenneth W. Simons, Equality as a Comparative Right, 65 B.U. L. REV. 387, 393-416 (1985); see also RAZ, supra note 22, at 225 (distinguishing between a principle that fortuitously generates equal entitlements and a principle designed to achieve equality); Parfit, supra note 37, at 8 (distinguishing between comparative and noncomparative justice).
90. This assertion must be qualified, however, in the case of what I call impure equality rights. See infra Part II.C.
91. This example does raise a doubt. If I honor my promise to one child, should I not honor my promise to another? And does that deeper obligation not express a genuine egalitarian duty? I explore this point further below. See infra Part IV.A.
92. See infra Part IV.D.
93. See JOEL FEINBERG, SOCIAL PHILOSOPHY 98 (1973); Joel Feinberg, Noncomparative Justice, 83 PHIL. REV. 297, 297-338 (1974) (the first explicit discussion of comparative rights). See generally RAZ, supra note 22, at 221; WESTEN, SPEAKING OF EQUALITY, supra note 2, at 131-45 (discussing the difference between comparative and noncomparative rights with regards to the Fourteenth Amendment; and dividing comparative rights into two subcategories, antidiscrimination rights and remaining comparative rights); Greenawalt, supra note 2, at 1178-83 (discussing how substantive norms of equality relate to other comparative and noncomparative norms); Hoffman, A New Theory of Comparative and Noncomparative Justice, 70 PHIL. STUD. 165, 165 (1993) (criticizing Feinberg's characterization of comparative and noncomparative justice and proposing an alternate theory); Stephen Nathanson, Does It Matter If the Death Penalty Is Arbitrarily Administered?, 14 PHIL. & PUB. AFF. 149, 151 (1985) (applying the distinction to the debate over the capricious implementation of the death penalty); Simons, supra note 52, at 416-52 (analyzing the distinction between comparative and noncomparative rights). This Article discusses comparative equality rights. Most of the analysis would also, however, apply to and explain a duty to treat people equally, even if no right corresponded to that duty. A teleological view, for example, might decline to recognize rights in victims of inequality, but might justify a duty to bring about the best state of affairs, which in turn is defined in part as the reduction or elimination of certain inequalities. According to Raz, to have a right is to have an interest that grounds another's duty. See RAZ, supra note 22, at 165-92. On this view, an egalitarian right does imply a duty, but an egalitarian duty does not imply that anyone has a right against the dutyholder. See id. One plausible conception of an egalitarian duty that lacks a corresponding egalitarian right is this: government has a duty to level down, but no one has a right to insist that inequality be leveled down, since leveling down makes no one better off. Whether this conception is persuasive, however, depends on whether inequality creates any special injury apart from the relative level of tangible benefits or resources. See infra note 72 and accompanying text.
94. See JOEL FEINBERG, SOCIAL PHILOSOPHY 98 (1973); Joel Feinberg, Noncomparative Justice, 83 PHIL. REV. 297, 297-338 (1974) (the first explicit discussion of comparative rights). See generally RAZ, supra note 22, at 221; WESTEN, SPEAKING OF EQUALITY, supra note 2, at 131-45 (discussing the difference between comparative and noncomparative rights with regards to the Fourteenth Amendment; and dividing comparative rights into two subcategories, antidiscrimination rights and remaining comparative rights); Greenawalt, supra note 2, at 1178-83 (discussing how substantive norms of equality relate to other comparative and noncomparative norms); Hoffman, A New Theory of Comparative and Noncomparative Justice, 70 PHIL. STUD. 165, 165 (1993) (criticizing Feinberg's characterization of comparative and noncomparative justice and proposing an alternate theory); Stephen Nathanson, Does It Matter If the Death Penalty Is Arbitrarily Administered?, 14 PHIL. & PUB. AFF. 149, 151 (1985) (applying the distinction to the debate over the capricious implementation of the death penalty); Simons, supra note 52, at 416-52 (analyzing the distinction between comparative and noncomparative rights). This Article discusses comparative equality rights. Most of the analysis would also, however, apply to and explain a duty to treat people equally, even if no right corresponded to that duty. A teleological view, for example, might decline to recognize rights in victims of inequality, but might justify a duty to bring about the best state of affairs, which in turn is defined in part as the reduction or elimination of certain inequalities. According to Raz, to have a right is to have an interest that grounds another's duty. See RAZ, supra note 22, at 165-92. On this view, an egalitarian right does imply a duty, but an egalitarian duty does not imply that anyone has a right against the dutyholder. See id. One plausible conception of an egalitarian duty that lacks a corresponding egalitarian right is this: government has a duty to level down, but no one has a right to insist that inequality be leveled down, since leveling down makes no one better off. Whether this conception is persuasive, however, depends on whether inequality creates any special injury apart from the relative level of tangible benefits or resources. See infra note 72 and accompanying text.
95. See JOEL FEINBERG, SOCIAL PHILOSOPHY 98 (1973); Joel Feinberg, Noncomparative Justice, 83 PHIL. REV. 297, 297-338 (1974) (the first explicit discussion of comparative rights). See generally RAZ, supra note 22, at 221; WESTEN, SPEAKING OF EQUALITY, supra note 2, at 131-45 (discussing the difference between comparative and noncomparative rights with regards to the Fourteenth Amendment; and dividing comparative rights into two subcategories, antidiscrimination rights and remaining comparative rights); Greenawalt, supra note 2, at 1178-83 (discussing how substantive norms of equality relate to other comparative and noncomparative norms); Hoffman, A New Theory of Comparative and Noncomparative Justice, 70 PHIL. STUD. 165, 165 (1993) (criticizing Feinberg's characterization of comparative and noncomparative justice and proposing an alternate theory); Stephen Nathanson, Does It Matter If the Death Penalty Is Arbitrarily Administered?, 14 PHIL. & PUB. AFF. 149, 151 (1985) (applying the distinction to the debate over the capricious implementation of the death penalty); Simons, supra note 52, at 416-52 (analyzing the distinction between comparative and noncomparative rights). This Article discusses comparative equality rights. Most of the analysis would also, however, apply to and explain a duty to treat people equally, even if no right corresponded to that duty. A teleological view, for example, might decline to recognize rights in victims of inequality, but might justify a duty to bring about the best state of affairs, which in turn is defined in part as the reduction or elimination of certain inequalities. According to Raz, to have a right is to have an interest that grounds another's duty. See RAZ, supra note 22, at 165-92. On this view, an egalitarian right does imply a duty, but an egalitarian duty does not imply that anyone has a right against the dutyholder. See id. One plausible conception of an egalitarian duty that lacks a corresponding egalitarian right is this: government has a duty to level down, but no one has a right to insist that inequality be leveled down, since leveling down makes no one better off. Whether this conception is persuasive, however, depends on whether inequality creates any special injury apart from the relative level of tangible benefits or resources. See infra note 72 and accompanying text.
96. See JOEL FEINBERG, SOCIAL PHILOSOPHY 98 (1973); Joel Feinberg, Noncomparative Justice, 83 PHIL. REV. 297, 297-338 (1974) (the first explicit discussion of comparative rights). See generally RAZ, supra note 22, at 221; WESTEN, SPEAKING OF EQUALITY, supra note 2, at 131-45 (discussing the difference between comparative and noncomparative rights with regards to the Fourteenth Amendment; and dividing comparative rights into two subcategories, antidiscrimination rights and remaining comparative rights); Greenawalt, supra note 2, at 1178-83 (discussing how substantive norms of equality relate to other comparative and noncomparative norms); Hoffman, A New Theory of Comparative and Noncomparative Justice, 70 PHIL. STUD. 165, 165 (1993) (criticizing Feinberg's characterization of comparative and noncomparative justice and proposing an alternate theory); Stephen Nathanson, Does It Matter If the Death Penalty Is Arbitrarily Administered?, 14 PHIL. & PUB. AFF. 149, 151 (1985) (applying the distinction to the debate over the capricious implementation of the death penalty); Simons, supra note 52, at 416-52 (analyzing the distinction between comparative and noncomparative rights). This Article discusses comparative equality rights. Most of the analysis would also, however, apply to and explain a duty to treat people equally, even if no right corresponded to that duty. A teleological view, for example, might decline to recognize rights in victims of inequality, but might justify a duty to bring about the best state of affairs, which in turn is defined in part as the reduction or elimination of certain inequalities. According to Raz, to have a right is to have an interest that grounds another's duty. See RAZ, supra note 22, at 165-92. On this view, an egalitarian right does imply a duty, but an egalitarian duty does not imply that anyone has a right against the dutyholder. See id. One plausible conception of an egalitarian duty that lacks a corresponding egalitarian right is this: government has a duty to level down, but no one has a right to insist that inequality be leveled down, since leveling down makes no one better off. Whether this conception is persuasive, however, depends on whether inequality creates any special injury apart from the relative level of tangible benefits or resources. See infra note 72 and accompanying text.
97. See JOEL FEINBERG, SOCIAL PHILOSOPHY 98 (1973); Joel Feinberg, Noncomparative Justice, 83 PHIL. REV. 297, 297-338 (1974) (the first explicit discussion of comparative rights). See generally RAZ, supra note 22, at 221;
98. "In general, comparative rights are of this form: (A) If class T obtains benefit or burden B, then class S should obtain f(B) where f(B) is some function of B. If the function is equality, i.e., f(B) = B, then the result is a comparative equality right: (B) If class T obtains B, then class S should obtain B." Simons, supra note 52, at 438 n.114; see also Raz, supra note 22, at 225 ("All F's who do not have G have a right to G if some Fs have G."). This account does not accurately describe nondiscrimination rights, however. Such rights have a logic of the following form: (N) Characteristic R shall not be used to permit or justify the unequal distribution of B. It appears that negative equality rights (such as nondiscrimination rights (N)) can be restated in the form of positive equality rights (such as (B)) as follows: (N*) If class T obtains B, and if class S would otherwise be entitled to B except for S's possession of characteristic R, then S should obtain B. See infra notes 95-106 and accompanying text for a more substantive analysis of discriminatory motives and stereotypes.
99. Rawls' difference principle, which states that "social and economic inequalities are to be arranged so that they are . . . to the greatest benefit of the least advantaged," is an example of the latter category. See JOHN RAWLS, A THEORY OF JUSTICE 83 (1971).
100. See, e.g., 42 U.S.C. §1981 (1982) ("All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and to no other . . . ."); Simons, supra note 52, at 422.
101. See, e.g., David Miller, Arguments for Equality, 7 MIDWEST STUD. PHIL. 73 (1982) (noting that egalitarian arguments for equality are all sensitive to relational factors and "concerned not merely with how well off individuals are, measured along some dimension, but with the relative standing of different people on that dimension").
102. Of course, the decisionmaker's flexibility depends on the scope of the comparative right. S might have a right to equal treatment with T with respect to some benefits but not others. (Women have a legal right not to suffer discrimination with respect to job opportunities, but not with respect to obtaining private dinner invitations.) And the right might be a right, not to equality, but to the minimization of inequality. (Rawls' difference principle exemplifies the latter category, while his principle of equal basic liberties exemplifies the former.) Further, equality as to B might be guaranteed only for a certain range of B. For example, legal or moral principles might require equality in the provision of basic health care, but permit unequal distribution of elective or cosmetic surgery.
103. This feature accounts for the otherwise inexplicable standing doctrine in equal protection cases: the complainant has standing to object to the more favorable discriminatory treatment of others even if the remedy for the discrimination might level down the benefits of others, thus leaving the complainant, and others in his class, with the same benefits as before. See Heckler v. Mathews, 465 U.S. 728, 738-40 (1984); Orr v. Orr, 440 U.S. 268 (1979). Similarly, a complainant has standing even if the remedy would only directly benefit others in his class (such as other white applicants objecting to an affirmative action program). The theory in this last category of cases is that the injury from the equality violation "is the denial of equal treatment resulting from the imposition of the [racial] barrier, not the ultimate inability to obtain the benefit." Northeastern Fla. Chapter of Assoc. Gen. Contractors v. City of Jacksonville, 508 U.S. 656, 666 (1993); see also ERWIN CHEMERINSKY, CONSTITUTIONAL LAW: PRINCIPLES AND POLICIES §2.5.3 (1997); JOHN E. NOWAK & RONALD D. ROTUNDA, CONSTITUTIONAL LAW §2.12(f)(5), at 89-90 (5th ed. 1995).
104. This feature accounts for the otherwise inexplicable standing doctrine in equal protection cases: the complainant has standing to object to the more favorable discriminatory treatment of others even if the remedy for the discrimination might level down the benefits of others, thus leaving the complainant, and others in his class, with the same benefits as before. See Heckler v. Mathews, 465 U.S. 728, 738-40 (1984); Orr v. Orr, 440 U.S. 268 (1979). Similarly, a complainant has standing even if the remedy would only directly benefit others in his class (such as other white applicants objecting to an affirmative action program). The theory in this last category of cases is that the injury from the equality violation "is the denial of equal treatment resulting from the imposition of the [racial] barrier, not the ultimate inability to obtain the benefit." Northeastern Fla. Chapter of Assoc. Gen. Contractors v. City of Jacksonville, 508 U.S. 656, 666 (1993); see also ERWIN CHEMERINSKY, CONSTITUTIONAL LAW: PRINCIPLES AND POLICIES §2.5.3 (1997); JOHN E. NOWAK & RONALD D. ROTUNDA, CONSTITUTIONAL LAW §2.12(f)(5), at 89-90 (5th ed. 1995).
105. When a legislature prospectively restricts benefits, it often "grandfathers" some beneficiaries who had received unrestricted benefits in the past, continuing at least some of their benefits for some period of time, but at the same time fully imposing the new restrictions on all new beneficiaries. A strict requirement that government uniformly level down when it lowers or restricts benefits would forbid such selective grandfather provisions. Moreover, even if a grandfather clause is the legislature's response to a judicial finding that it has violated equality principles, the clause might be upheld, if the protected beneficiaries have a strong reliance interest. See Heckler, 465 U.S. at 746 (1984) ("We have recognized, in a number of contexts, the legitimacy of protecting reasonable reliance on prior law even when that requires allowing an unconstitutional statute to remain in effect for a limited period of time."); see also Nordlinger v. Hahn, 505 U.S. 1, 13 (1992); Sklar v. Byrne, 727 F.2d 633, 639-42 (7th Cir. 1984).
106. See also WESTEN, SPEAKING OF EQUALITY, supra note 2, at 136 (responding to the flexibility argument and conceding that equality rights have this type of distinctive flexibility). Westen, however, continues to understate the distinctiveness of equality rights. Thus, with respect to the claim by Raz and me that the very point of equality rights is to achieve equality in treatment, Westen partially demurs: "[W]hile all rights aim toward equality, only antidiscrimination rights aim solely toward equality. All rights require, and (if faithfully applied) result in, relationships of equality." Id. at 140. But I disagree that noncomparative rights aim toward equality even in part. Rather, equality is their incidental effect, a mere byproduct of requiring the duty-holder to provide a specified form of noncomparative treatment to a class of persons. And even this effect occurs only when the right has the form of an entitlement under a classification or rule, rather than the form of an entitlement held by uniquely specified individuals. See infra notes 116-18 and accompanying text.
107. See RAWLS, supra note 58, at 62 (discussing the "chief" primary social goods which include "rights, liberties, powers and opportunities, income and wealth").
108. See example 2.A, supra p. 706. See generally DOUGLAS RAE, ET AL., EQUALITIES (1981); William B. Griffith, Equality and Egalitarianism: Framing the Contemporary Debate, 7 CAN. J.L. & JURIS. 5 (1994).
109. See example 2.A, supra p. 706. See generally DOUGLAS RAE, ET AL., EQUALITIES (1981); William B. Griffith, Equality and Egalitarianism: Framing the Contemporary Debate, 7 CAN. J.L. & JURIS. 5 (1994).
110. But see Greenawalt, supra note 1, at 1289 (noting that he is agnostic about whether "the principle of equality should authorize giving people worse [rather than better] treatment than they otherwise deserve").
111. Compare a situation in which an employer first hires a white person, and later refuses to hire an equally qualified black, with a situation in which she makes the hiring decisions simultaneously.
112. See Simons, supra note 52, at 423. In a legal system, egalitarian norms can be constitutional, statutory, or common law. See id.
113. See supra notes 56-60 and accompanying text.
114. See generally RAWLS, supra note 58.
115. This distinction might help explain the feminist debate over whether women should frame their claims of unjust treatment in terms of "sameness" or instead "difference." See infra Part V.G. For the argument that normative equality should focus on social relations, and only indirectly on the patterns of distribution of goods, see Elizabeth Anderson, What is the Point of Equality?, 109 ETHICS 287, 313-14 (1999).
116. For discussion of the distinction between deontological and teleological (including consequentialist) accounts of equality in the context of distributive justice, see TEMKIN, supra note 37; at 11; McKerlie, supra note 37, at 274-75; Parfit, supra note 37, at 13. According to the deontological view, inequality's importance "is not that it makes an outcome better or worse, per se, but that in certain situations we have moral reasons to reduce or remove inequality." TEMPKIN, supra note 37, at 11. Peters apparently assumes that equality rights, to be genuine, must be deontological. Peters, Equality Revisited, supra note 1, at 1238. But he does not explain his assumption. For example, in replying to Greenawalt's views, he concedes that "good consequentialist reasons" may explain "why certain kinds of decisionmakers should avoid the appearance of partiality or bias," but he concludes that these considerations (which I would certainly classify as egalitarian) are entirely a matter of "nonegalitarian" justice, not of deontological equality. Id. at 1239 n.46. And, in discussing the Lucky/Unlucky lottery hypothetical, he concedes that permitting the inequality might undermine confidence in government, or reduce the willingness of people to play the lottery. Id. at 1248. Therefore, he concludes, our real concern is nonegalitarian justice, not equality. See also id. at 1250 n.58 ("We favor equal treatment of children by their parents . . . because we want children to feel that they are equally loved and that their parents are unbiased and impartial."). But again, it is not clear why equality should not be considered an important and distinctive legal and moral concept simply because its value is viewed as instrumental rather than intrinsic.
117. See Developments in the Law-Employment Discrimination, 109 HARV. L. REV. 1568, 1580 (1996). The general statutory provision is 42 U.S.C. §2000e-2. The judicially-created "disparate impact" category was recently codified at 42 U.S.C. §2000e-2(k), in response to the Supreme Court's narrower interpretation of the category in Wards Cove Packing Co. v. Atonio, 490 U.S. 642 (1989). Consider also the Supreme Court's treatment of wealth discrimination. The Court disfavors explicit use of wealth criteria to discriminate between classes of citizens. See, e.g., Cipriano v. City of Houma, 395 U.S. 701 (1969); Edwards v. California, 314 U.S. 160 (1941). But the Court permits most classifications that disproportionately affect the poor, such as fee requirements that the poor find much more difficult to meet. See, e.g., Kadrmas v. Dickinson Pub. Sch., 487 U.S. 450 (1988). Peters asserts that justice requires not just the correct outcome, but also that the reasons actually relied upon were the morally relevant ones. Peters, Equality Revisited, supra note 1, at 1229. But this is not the only defensible view of justice, in either law or political theory. An equality right, as well as other rights and justice principles, might be concerned only with reasons, or only with outcomes, or with both. In Title VII, for example, disparate impact cases focus on discriminatory outcome, while disparate treatment cases typically involve a combination of impermissible reason and discriminatory outcome. Moreover, in his ideal theory of justice, Rawls assumes that citizens will be prompted by the usual range of human motivations, from altruism to greed, but that the social system can be so structured that the outcome of aggregate human choices will be just.
118. See Developments in the Law-Employment Discrimination, 109 HARV. L. REV. 1568, 1580 (1996). The general statutory provision is 42 U.S.C. §2000e-2. The judicially-created "disparate impact" category was recently codified at 42 U.S.C. §2000e-2(k), in response to the Supreme Court's narrower interpretation of the category in Wards Cove Packing Co. v. Atonio, 490 U.S. 642 (1989). Consider also the Supreme Court's treatment of wealth discrimination. The Court disfavors explicit use of wealth criteria to discriminate between classes of citizens. See, e.g., Cipriano v. City of Houma, 395 U.S. 701 (1969); Edwards v. California, 314 U.S. 160 (1941). But the Court permits most classifications that disproportionately affect the poor, such as fee requirements that the poor find much more difficult to meet. See, e.g., Kadrmas v. Dickinson Pub. Sch., 487 U.S. 450 (1988). Peters asserts that justice requires not just the correct outcome, but also that the reasons actually relied upon were the morally relevant ones. Peters, Equality Revisited, supra note 1, at 1229. But this is not the only defensible view of justice, in either law or political theory. An equality right, as well as other rights and justice principles, might be concerned only with reasons, or only with outcomes, or with both. In Title VII, for example, disparate impact cases focus on discriminatory outcome, while disparate treatment cases typically involve a combination of impermissible reason and discriminatory outcome. Moreover, in his ideal theory of justice, Rawls assumes that citizens will be prompted by the usual range of human motivations, from altruism to greed, but that the social system can be so structured that the outcome of aggregate human choices will be just.
119. See infra Part IV.A.3. In earlier writing, I employed the misleading term "implicit comparative rights" for this idea. Simons, supra note 52, at 421-27.
120. See Simons, supra note 52, at 456-59 (citing literature); see also WESTEN, SPEAKING OF EQUALlTY, supra note 2, at 181.
121. The parent's duty to explain or justify other distinctions in treatment, such as distinctions among friends or other family members, might be less stringent.
122. See, e.g., Lucinda Findley, Transcending Equality Theory: A Way Out of the Maternity and Workplace Debate, 86 COLUM. L. REV. 1118 (1986); Diana Majury, Strategizing in Equality, 3 WIS. WOMEN'S L.J. 169 (1987).
123. See, e.g., Lucinda Findley, Transcending Equality Theory: A Way Out of the Maternity and Workplace Debate, 86 COLUM. L. REV. 1118 (1986); Diana Majury, Strategizing in Equality, 3 WIS. WOMEN'S L.J. 169 (1987).
124. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 163-64 (arguing that equality of opportunity, perhaps the most widely accepted egalitarian principle, signifies different things to different people, "enveloping all its uses with favorable connotations").
125. I believe that this is the essence of the critique of some feminists. For example, they might object (and with good reason) that modern equal protection doctrine addressing sex discrimination improperly restricts its focus to whether a classification adequately furthers a legislative purpose and thus "treats like alike." See infra text accompanying notes 248-61.
126. The "leveling down" objection normally refers to leveling down benefits, not burdens. But insofar as relief from a burden is a kind of benefit, the leveling down objection sensibly includes an objection to "leveling down" that kind of benefit as well. That is, the objection seems to have at least as much force when the decisionmaker eliminates discriminatory distribution of a burden by denying all persons relief from the burden (i.e., by extending the burden to all).
127. There are some exceptions, as noted shortly. But we should distinguish, and not count as a genuine exception, situations in which a successful plaintiff in an individual equal protection or anti-discrimination rights case obtains a remedy that rewards her but not others who are similarly situated. Such a remedy might be awarded merely for instrumental reasons, to encourage the bringing of lawsuits. The more important question is whether the class of which she is a part is itself entitled to be leveled up. If not, then the equality right that she asserts does not demand leveling up.
128. Age Discrimination in Employment Act (ADEA), 29 U.S.C. §621 (1994) ("It shall be unlawful for an employer [to discriminate against an employee because of the employee's age] or . . . to reduce the wage rate of any employee in order to comply with this chapter.")
129. The Equal Pay Act, which redresses certain forms of sex discrimination in employment, forbids leveling down as a remedy. Equal Pay Act of 1963, 29 U.S.C. §206 (1994) ("[A]n employer who is paying a wage rate differential in violation of this subsection shall not, in order to comply with the [nondiscrimination] provisions of this subsection, reduce the wage rate of any employee."). See, e.g., E.E.O.C. v. Romeo Community Sch., 976 F.2d 985, 988 (6th Cir. 1992) (holding that a prima facie case of a violation of the Equal Pay Act was made out where the defendant "cured the disparity between male and female wage rates by lowering the male wage rate to the rate for females"). For further discussion of the Equal Pay Act, see SUSAN M. OMILIAN & JEAN P. KAMP, SEX-BASED EMPLOYMENT DISCRIMINATION (1990 & Supp. 1996). The limits of these statutory obligations to level up are unclear. The employer is not required to maintain a wage level or a set of employment positions indefinitely. Thus, under the ADEA, even if a plaintiff was illegally denied a position because of age, "if a reduction-in-force or a legitimate reorganization eliminated plaintiff's job, reinstatement would not be required." MACK A. PLAYER, EMPLOYMENT DISCRIMINATION LAW 554 (1988). But presumably the reason, to be "legitimate," must be other than a simple desire to level down to rectify the inequality. By contrast, effectuating that desire is a perfectly legitimate remedial response to the violation of pure equality rights.
130. The Equal Pay Act, which redresses certain forms of sex discrimination in employment, forbids leveling down as a remedy. Equal Pay Act of 1963, 29 U.S.C. §206 (1994) ("[A]n employer who is paying a wage rate differential in violation of this subsection shall not, in order to comply with the [nondiscrimination] provisions of this subsection, reduce the wage rate of any employee."). See, e.g., E.E.O.C. v. Romeo Community Sch., 976 F.2d 985, 988 (6th Cir. 1992) (holding that a prima facie case of a violation of the Equal Pay Act was made out where the defendant "cured the disparity between male and female wage rates by lowering the male wage rate to the rate for females"). For further discussion of the Equal Pay Act, see SUSAN M. OMILIAN & JEAN P. KAMP, SEX-BASED EMPLOYMENT DISCRIMINATION (1990 & Supp. 1996). The limits of these statutory obligations to level up are unclear. The employer is not required to maintain a wage level or a set of employment positions indefinitely. Thus, under the ADEA, even if a plaintiff was illegally denied a position because of age, "if a reduction-in-force or a legitimate reorganization eliminated plaintiff's job, reinstatement would not be required." MACK A. PLAYER, EMPLOYMENT DISCRIMINATION LAW 554 (1988). But presumably the reason, to be "legitimate," must be other than a simple desire to level down to rectify the inequality. By contrast, effectuating that desire is a perfectly legitimate remedial response to the violation of pure equality rights.
131. See Parfit, supra note 37, at 5-6.
132. Parfit is discussing the view that inequality is an intrinsically bad consequence or state of affairs, to be distinguished from the more typical teleological view that inequality is often bad because it leads to further bad consequences, such as envy, civil strife, or inefficiency. Teleologists (including consequentialists) can compare the value of outcomes by the degree of inequality, as well as by the aggregate well-being, that they produce. See Parfit, supra note 37, at 5-6.
133. If we are Deontic Egalitarians, we do not believe that inequality is bad, so we are not forced to admit that . . . it would be in one way better if inequalities were removed by leveling down. We can believe that we have a reason to remove inequality only when, and only because, our way of doing so benefits the people who are worse off. Or we might believe that, when some people are worse off than others, through no fault or choice of theirs, they have a special claim to be raised up to the level of the others, but they have no claim that others be brought down to their level. Parfit, supra note 37, at 9.
134. With respect to the wrongdoer's remedial options to level down or level up, we can imagine categories of equality rights other than pure, impure, and modified impure. Suppose S obtains 10, T obtains 20, and this inequality is unjustifiable. A stingy type of impure equality right would require leveling T down, rather than leveling S up. Moreover, a super leveling down principle would justify leveling down of both groups (e.g., if this were the only feasible way to achieve equality). It might justify reducing both S and T to 9. Conversely, a super leveling up principle would justify leveling up of both groups (e.g., increasing both S and T to 21). More unusual categories are also imaginable. A leapfrog leveling down principle would justify moving the higher group to a level lower than that of the lower group (e.g., lowering T to 9), if this move reduces inequality. And a leapfrog leveling up principle would justify moving the lower group to a level higher than that of the higher group (e.g. raising S to 21). These last two categories are not entirely fanciful; they might be the only realistic options if an equal benefit is not available. Suppose an airline mistakenly gives two coach seats belonging to X and Y to other passengers, and only one coach and one first class seat remain. The airline's egalitarian duty might require that either X or Y receive a first class seat. See also TEMKIN, supra note 37, at 118-56 (thoroughly analyzing different approaches for measuring inequality). One argument in favor of Rawls' difference principle relies on the intuition that a super leveling down principle is morally objectionable. The argument goes as follows: Rawls assumes that we start from a position of equality of primary social goods (including liberty, wealth, and income), and that social and political institutions (including a market system) are permissible notwithstanding the inequalities they produce if those inequalities benefit all citizens, including those who are worst off. See RAWLS, supra note 58, at 14-15, 78. But if we reject his view that such inequalities are permissible, then it seems that we would prefer super leveling down (relative to the state of affairs in which inequality is to the benefit of all). Yet the super leveling down state of affairs not only does not help the worst off, it actually worsens their position relative to their status under the difference principle. For careful discussion of Rawls' difference principle and whether, property understood, it is open to a leveling down objection, see TEMKIN, supra note 37, at 31-37, 91-117; Parfit, supra note 37, at 35-39.
135. Of course, whether even limited leveling down is permissible depends on the force of T's right to the existing level of benefits.
136. Raz seems to suggest a negative answer, though he does not specifically address what I call "impure" equality rights. "It is only the effect of other principles which can explain our preference for giving the benefit to those who lack it to denying it to those who have it. This preference cannot be explained on the basis of the egalitarian principles themselves." RAZ, supra note 22, at 227.
137. The enactment of the Equal Pay Act poses an interesting example. Employers wanted the freedom to reduce wages to comply with the Act. Instead, Congress postponed the effective date of the Act for a year. "This will give time for unions, employers, and employees to make adjustments or changes in the job classifications, wage rates, and so forth, if necessary." 87 Cong. Rec. 8687 (1963) (remarks of Congressman Griffin).
138. This is not to say, however, that impure equality rights are only defensible when the person under the duty to rectify the inequality is the person who created the inequality. For example, government might also have a duty to rectify inequalities that are not of its own making, such as private discrimination, or natural inequalities in talents. This duty could nevertheless be a duty to level up.
139. See infra note 98 and accompanying text.
140. On the other hand, the concern could be broader, namely, a belief that the opportunities already being offered in wealthy districts should be granted to the poor. In that case, leveling up could be required. (If the right to equal educational spending were constitutionalized, it might forbid the state from adopting a formula requiring wealthier districts to lower their spending, a formula that might save the state money compared to the state subsidizing the poorer districts.)
141. See RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 227-28 (1977). For further discussion, see Simons, supra note 52, at 480 n.229.
142. See ROBERT NOZICK, ANARCHY, STATE AND UTOPIA 223 (1974): That differences in treatment need to be justified does fit contemporary governments. Here [unlike the situation of private persons, who are entitled to transfer property as they wish,] there is a centralized process treating all, with no entitlement to bestow treatment according to whim. Id. Of course, since Nozick would radically limit the legitimate domain of government decisions, the actual scope of his principle would be much narrower than the scope of Dworkin's.
143. See DWORKIN, supra note 95, at 227 (drawing a distinction between (superficial) "equal treatment" and (deeper) "equal concern" or "treatment as an equal"). If I have two children, and one is dying from a disease that is making the other uncomfortable, I do not show equal concern if I flip a coin to decide which shall have the remaining dose of a drug. This shows that the right to treatment as an equal is fundamental, and the right to equal treatment, derivative. In some circumstances the right to treatment as an equal will entail a right to equal treatment, but not, by any means, in all circumstances. Id.
144. See Palmer v. Thompson, 403 U.S. 217 (1971), and the discussion in Simons, supra note 52, at 431-33. My observations in the text explain why Peters' criticism of Palmer, supra note 1, at 1263 n.84, does not prove his claim that racial discrimination is not essentially concerned with equal treatment.
145. See Kenneth W. Simons, Rethinking Mental States, 72 B.U. L. REV. 463, 519 (1992) (discussing selective sympathy and selective indifference).
146. Even this conception poses significant difficulties. See David Strauss: Discriminatory Intent and The Taming of Brown, 56 U. CHI. L. REV. 935 (1989).
147. Of course, intermediate possibilities are also possible. If one measures "concern" literally in terms of time and attention, then the decisionmaker has a fixed amount of concern to "distribute," and might thus need to decrease somewhat the "concern" previously showed to his favorites and to increase somewhat the "concern" previously shown to those least favored.
148. See infra note 247.
149. Moreover, when the question is what remedy a decisionmaker should provide when he has shown unequal respect to different groups, it is plausible to recognize an "impure" equality right, requiring him to level up the disfavored group. See supra Part II.C.
150. See supra note 47 and accompanying text.
151. One difficulty with this example is that jobs are a scarce resource: giving a job preference to one or two groups of citizens affects the job prospects of others who are not entitled to the preference. To avoid this difficulty, imagine that the government provides medals or symbolic awards to citizens who have made sacrifices. Then it is easier to see how giving awards to two groups rather than none shows more "respect" for citizens generally. (Even here, to be sure, the resource is not unlimited. Giving awards to all citizens might so debase the award that its value would be entirely lost.)
152. See Richard McAdams, Cooperation and Conflict: the Economics of Group Status Production and Race Discrimination, 108 HARV. L. REV. 1003, 1044-49 (1995) (explaining race discrimination as a process by which a racial group seeks to produce status for itself by lowering another group's status).
153. I will note some caveats, however, in a later section. See infra Part IV.A.
154. The relevant "universe" can be qualified or limited, however. It need not include all persons. See Simons, supra note 52, at 454-55 n.153.
155. See Peters, Equality Revisited, supra note 1, at 1246-47.
156. More precisely, we must suppose that at least one person with trait T receives B while at least one person with trait T does not.
157. Thus, in (3.A), the son will complain that he was incorrectly deprived of a dessert that the similarly-entitled daughter actually received.
158. Thus, in (3.B), Mr. Unlucky will complain that similarly-entitled Ms. Lucky incorrectly received a lottery windfall, while Unlucky only received the face amount of his ticket.
159. In (3), "B" can refer to a burden instead of a benefit. In that case, the observations in the last parenthetical should be appropriately modified by characterizing the relevant benefit as relief from a burden. Incorrect deprivation occurs when a burden is incorrectly imposed, or is excessive. Incorrect beneficence occurs when a burden is incorrectly not imposed, or is too light. Although Peters addresses both types of problems, his main focus is the second problem, incorrect beneficence. See Peters, Equality Revisited, supra note 1, at 1212. He maintains that "prescriptive" equality purports to provide a reason for treating the claimant wrongly, a reason that the nonegalitarian rule does not itself provide. By contrast, he believes that the nonegalitarian rule itself suffices to explain what is wrong in the first problem, incorrect deprivation. Id. at 1226.
160. Following Peters' original example, I put aside the complication that the amount each person receives might depend on the number of lottery winners. On the question whether a competitive good should, for that reason alone, count as a comparative right, see infra note 181 and accompanying text.
161. Peters characterizes the treatment of Lucky as not just incorrect, but "unjust." See Peters, Equality Revisited, supra note 1, at 1230 n.38, 1247-51. This conclusion reflects Peters' view that incorrect treatment under a rule is necessarily an injustice. I think it less idiomatic to restrict the term "injustice" in this context to comparative and noncomparative right violations that are to the detriment of the right holder.
162. To be sure, such a case might implicitly rest on a general principle such as: "Show equal concern to all of my own children, whether natural or adopted." But it need not. It could simply reflect a commitment to treat adopted Andrew and natural Nathan with equal concern, without displaying any similar commitment as to future children. For example, a parent might admit that she would not be able to show as much concern for additional children as she shows for the first two.
163. By contrast, lexical equality, or equality of entitlements as a mere consequence of implementing a rule, typically does follow from providing specified treatment to all persons with a specified descriptive trait, because rules often identify a trait that is a necessary and sufficient condition of entitlement (as in (3.A) and (3.B)).
164. "To say that considerations of one kind (e.g. the mental) supervene on those of another kind (e.g. the physical) is to say that there are, or can be, no differences in the first kind without there being differences in the second kind." Paul Teller, Supervenience, in A COMPANION TO METAPHYSICS 484 (Jaegwon Kim & Ernest Sosa eds. 1995).
165. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 185-225 (also citing extensive philosophical literature); John E. Coons, Consistency, 75 CAL. L. REV. 59 (1987); Greenawalt, supra note 1; Peters, Equality Revisited, supra note 1.
166. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 185-225 (also citing extensive philosophical literature); John E. Coons, Consistency, 75 CAL. L. REV. 59 (1987); Greenawalt, supra note 1; Peters, Equality Revisited, supra note 1.
167. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 185-225 (also citing extensive philosophical literature); John E. Coons, Consistency, 75 CAL. L. REV. 59 (1987); Greenawalt, supra note 1; Peters, Equality Revisited, supra note 1.
168. Peters does acknowledge that the "prescriptive equality" he discusses is only one of many kinds of arguments that might be called egalitarian. See Peters, Equality Revisited, supra note 1, at 1214. Greenawalt makes a similar acknowledgement. See Greenawalt, supra note 1, at 1274, 1278 n.42; see also id. at 1274 (suggesting that, relative to "the major issues about equality in Western societies . . . the subject of Professor Peters' inquiry is narrow and comparatively insignificant. But it is definitely worth our attention."). However, Peters also leaves the unmistakable impression that he is hostile to any genuine comparative equality rights. Thus, he employs the term "nonegalitarian justice" to encompass all justice principles other than "prescriptive equality." Peters, Equality Revisited, supra note 1, at 1228. This hardly leaves much room for egalitarian principles other than "prescriptive equality." Moreover, by treating Rawls' difference principle and norms against racial discrimination as instances of "nonegalitarian justice" in the noncomparative sense, see id. at 1254-55, 1259-62, Peters confirms the impression that he would not endorse any comparative conception of equality other than "prescriptive equality" (which he ultimately rejects); see also id. at 1262-64 n.84 (simply rejecting the comparative equality component of a state constitutional right to a "thorough and efficient" education, a component that requires some equalization between poor and wealthy districts, and claiming that only the noncomparative component of the right is important).
169. Peters does acknowledge that the "prescriptive equality" he discusses is only one of many kinds of arguments that might be called egalitarian. See Peters, Equality Revisited, supra note 1, at 1214. Greenawalt makes a similar acknowledgement. See Greenawalt, supra note 1, at 1274, 1278 n.42; see also id. at 1274 (suggesting that, relative to "the major issues about equality in Western societies . . . the subject of Professor Peters' inquiry is narrow and comparatively insignificant. But it is definitely worth our attention."). However, Peters also leaves the unmistakable impression that he is hostile to any genuine comparative equality rights. Thus, he employs the term "nonegalitarian justice" to encompass all justice principles other than "prescriptive equality." Peters, Equality Revisited, supra note 1, at 1228. This hardly leaves much room for egalitarian principles other than "prescriptive equality." Moreover, by treating Rawls' difference principle and norms against racial discrimination as instances of "nonegalitarian justice" in the noncomparative sense, see id. at 1254-55, 1259-62, Peters confirms the impression that he would not endorse any comparative conception of equality other than "prescriptive equality" (which he ultimately rejects); see also id. at 1262-64 n.84 (simply rejecting the comparative equality component of a state constitutional right to a "thorough and efficient" education, a component that requires some equalization between poor and wealthy districts, and claiming that only the noncomparative component of the right is important).
170. The idea of a "presumption of equality" requiring equal treatment unless reasons are shown for unequal treatment is not without complexity, however. For a thoughtful discussion, see WESTEN, SPEAKING OF EQUALITY, supra note 2, at 230-56.
171. See Simons, supra note 52, at 393-403; Westen, A Response, supra note 11, at 1188-90; see also David Lyons, On Formal Justice, 58 CORNELL L. REV. 833 (1973) .
172. See Simons, supra note 52, at 393-403; Westen, A Response, supra note 11, at 1188-90; see also David Lyons, On Formal Justice, 58 CORNELL L. REV. 833 (1973) .
173. Greenawalt reaches a similar conclusion. He agrees that the formal principle, "equals should be treated equally," can be empty and thinks this is a good reason for ascribing to it the content of "prescriptive equality," namely, "the idea that giving a form of treatment to one equal is a reason to give the same treatment to another." Greenawalt, supra note 1, at 1268. I disagree, however, that the appropriate content of TLA is this historical or precedential principle of "prescriptive" equality. See infra Part III.C. Westen's detailed analysis of the "formal" principle of equality partially supports my analysis of TLA, though I believe he gives insufficient attention to the distinction between comparative and noncomparative principles. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 185-229; see also supra note 64. Westen notes that even Joel Feinberg, the originator of that distinction, provides a regrettably ambiguous analysis of "likes should be treated alike." See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 135 n.11 (citing FEINBERG, supra note 56, at 99); Simons, supra note 52, at 401 n.25.
174. Greenawalt reaches a similar conclusion. He agrees that the formal principle, "equals should be treated equally," can be empty and thinks this is a good reason for ascribing to it the content of "prescriptive equality," namely, "the idea that giving a form of treatment to one equal is a reason to give the same treatment to another." Greenawalt, supra note 1, at 1268. I disagree, however, that the appropriate content of TLA is this historical or precedential principle of "prescriptive" equality. See infra Part III.C. Westen's detailed analysis of the "formal" principle of equality partially supports my analysis of TLA, though I believe he gives insufficient attention to the distinction between comparative and noncomparative principles. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 185-229; see also supra note 64. Westen notes that even Joel Feinberg, the originator of that distinction, provides a regrettably ambiguous analysis of "likes should be treated alike." See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 135 n.11 (citing FEINBERG, supra note 56, at 99); Simons, supra note 52, at 401 n.25.
175. See Westen, The Empty Idea of Equality, supra note 2, at 542.
176. See Barry, supra note 6, at 323; see also Allen Buchanon, Distributive Justice, in 1 ENCYCLOPEDIA OF ETHICS 655 (L.C. Becker & C.B. Becker eds. 1992). Westen carefully examines H.L.A. Hart's articulation of this view in WESTEN, SPEAKING OF EQUALITY, supra note 2, at 225-29. Much of Westen's analysis is persuasive, but once again, he understates the importance of the comparative/noncomparative distinction. Thus, Westen's conclusion, that the "formal" principle or TLA adds nothing to the "material" principle of justice is valid when TLA simply reflects the notion of implementing a noncomparative rule. However, it is not valid when TLA is a way of expressing an egalitarian, comparative metaprinciple, or when the material principle is itself comparative.
177. See Barry, supra note 6, at 323; see also Allen Buchanon, Distributive Justice, in 1 ENCYCLOPEDIA OF ETHICS 655 (L.C. Becker & C.B. Becker eds. 1992). Westen carefully examines H.L.A. Hart's articulation of this view in WESTEN, SPEAKING OF EQUALITY, supra note 2, at 225-29. Much of Westen's analysis is persuasive, but once again, he understates the importance of the comparative/noncomparative distinction. Thus, Westen's conclusion, that the "formal" principle or TLA adds nothing to the "material" principle of justice is valid when TLA simply reflects the notion of implementing a noncomparative rule. However, it is not valid when TLA is a way of expressing an egalitarian, comparative metaprinciple, or when the material principle is itself comparative.
178. Of course, the argument sometimes is meant to invoke an egalitarian metaprinciple, such as a requirement of impartiality or freedom from bias in administering criminal punishment. This egalitarian metaprinciple has played an important role in recent sentencing reform movements. A major impetus behind the recent movements toward fixed sentencing and the abolition of probation and parole is a concern that under indeterminate sentencing, judge A and judge B would impose dramatically different punishments on the same set of facts. More rigid sentencing policies reduce that form of inequality (though they create other problems, including harshness and a crude insensitivity to the facts of individual cases). Because, under the old system, judge A and judge B had discretion, it was often not possible to criticize either judge's decision as an erroneous application of the rule, "impose punishment in accordance with guilt." Cf. Nathanson, supra note 56, at 149-64 (arguing that the death penalty should be abolished because of its arbitrary imposition).
179. Cf. Kane, Justice, supra note 4, at 378-81 (properly concluding that the Aristotelian formulation need not express a comparative right, but improperly concluding that it cannot).
180. This statement is subject to the caveats noted infra Part IV.A. The requirement of consistency could itself express an egalitarian norm, but it would be a different norm than the duty not to discriminate against women; for example, it might be a duty to explain to the second woman why she was not protected by the anti-discrimination norm when the first woman was protected.
181. Usually, as in his Lucky/Unlucky example, Peters is concerned about incorrect beneficence, and is therefore arguing for consistent departure from a rule. But occasionally, he is concerned about incorrect deprivation. Concern about the latter form of lexical inequality presumably militates in favor of consistent implementation of a rule. In SPEAKING OF EQUALITY, Westen distinguishes two ways in which we might treat people "equally" pursuant to a relevant rule. The rule can serve directly as a prescriptive standard of comparison, so that "A and B are equal" means that they have equal entitlements under a rule. Or the rule can serve indirectly as a baseline for measuring whether actual treatments are identical or nonidentical in the extent to which they depart from what the rule says they should be. In the second case, "A and B are equal" means either that both have received what they are entitled to under a baseline rule, or that both have been denied what they are entitled to under that rule. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 86-92, 189-90. This second formulation seems to include the argument I am now discussing.
182. Indeed, Peters makes a similar reductio ad absurdum argument, asserting that (incorrectly) giving Unlucky what Lucky (incorrectly) received treats Unlucky equally to Lucky but unequally to everyone else in the world who has been or ever will be treated correctly with respect to any rule. See Peters, Equality Revisited, supra note 1, at 1249. Greenawalt notes that this comparison is very odd, insofar as Peters ignores the types of egalitarian comparisons and relationships that most people consider salient, such as relationships of parents to children. See Greenawalt, supra note 1, at 1279-81. But I believe that there is a deeper problem: Peters fails to see that claims of unjustly unequal treatment must be grounded in a particular comparative right claim, not in the diffuse and extreme claim that any descriptively unequal allocation is unjust. At the same time, Greenawalt's own critique seems inadequate. His "salience" qualification seems overly strict. For example, governments must treat all citizens with equal concern and respect, yet it is not clear that citizens' relationships to their government are especially "salient." See also infra note 241.
183. But government might need to justify its choice of administrative systems when one predictably leads to more errors than another. Cf. Logan v. Zimmerman Brush Co., 455 U.S. 422 (1982).
184. Peters, Equality Revisited, supra note 1, at 1214.
185. Peters asserts that his analysis differs from Westen's as follows. A special nontautological equality problem exists, according to Peters, when X correctly gets a benefit but Y is incorrectly denied it. This isn't just a case of Y being wrongly denied a benefit, for someone with no greater entitlement to the benefit has been awarded it. See Peters, Equality Revisited, supra note 1, at 1220-21. However, Peters misunderstands Westen here. Westen no longer views all equality arguments as meaningless tautologies. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 202-04. Moreover, Westen does recognize the subcategory of equality principles that Peters denominates "prescriptive." See supra note 24.
186. Peters asserts that his analysis differs from Westen's as follows. A special nontautological equality problem exists, according to Peters, when X correctly gets a benefit but Y is incorrectly denied it. This isn't just a case of Y being wrongly denied a benefit, for someone with no greater entitlement to the benefit has been awarded it. See Peters, Equality Revisited, supra note 1, at 1220-21. However, Peters misunderstands Westen here. Westen no longer views all equality arguments as meaningless tautologies. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 202-04. Moreover, Westen does recognize the subcategory of equality principles that Peters denominates "prescriptive." See supra note 24.
187. See Peters, Equality Revisited, supra note 1, at 1223 (emphasis omitted).
188. It also seems to derive from Peters' interest in the subject of precedential constraint. See Peters, Foolish Consistency, supra note 34, at 2057-62, 2073; Peters, Equality Revisited, supra note 1, at 1226 n.32, 1263.
189. It also seems to derive from Peters' interest in the subject of precedential constraint. See Peters, Foolish Consistency, supra note 34, at 2057-62, 2073; Peters, Equality Revisited, supra note 1, at 1226 n.32, 1263.
190. Peters explains in his conclusion: "My argument against prescriptive equality has really been an argument in favor of making every decision on its own merits, of treating people the way justice dictates that they be treated, regardless of how someone else has been treated in the past." Peters, Equality Revisited, supra note 1, at 1264.
191. See 42 U.S.C. §2000e-2(a)(1) (discussed in Simons, supra note 52, at 419 n.66); see also Simons, supra note 52, at 418-20 (noting that such discrimination is "hypothetical" only because, despite the decisionmaker's intent to discriminate, the class of persons who would be affected by this intent is currently empty).
192. See Peters, Equality Revisited, supra note 1, at 1219-20.
193. See id. at 1220.
194. Discussing Tick Wo v. Hopkins, 118 U.S. 356 (1886), Peters asserts: "The Chinese applicants were treated wrongly not because they were treated differently, or 'unequally,' from the Caucasian applicants, but rather because they were treated according to an irrelevant criterion (their race, or national origin), regardless of how Caucasians were treated." Peters, Equality Revisited, supra note 1, at 1220 (emphasis added). The emphasized phrase reveals the fallacy in this argument. If Caucasians, too, were denied laundry permits, then it is quite unlikely that the criterion of race explains why Chinese applicants were denied permits. And, even if it turns out that the Caucasian applicants were denied permits for other reasons, while racial prejudice really was the actual reason for denying permits to Chinese, then the latter still amounts to an equality problem, in this straightforward sense: if the Chinese applicants had been Caucasian, then they would have received permits. (Or even if they would not have, racial prejudice still provided the government with an additional reason for burdening the Chinese applicants, a reason that made it more likely that government would burden applicants on a racial basis.)
195. To some extent, Greenawalt agrees with Peters' "one person reductio" argument. In an attempt to show that the precedential conception of equality has independent meaning, Greenawalt states: "[I]t may be a worse wrong if a white is given a benefit and a black is denied the benefit because of race, than if the black is the only person concerned and he is denied the benefit because of race." Greenawalt, supra note 1, at 1278 n.42. Perhaps. But the wrong is not much worse, if it is worse at all. In any event, a conception of equality that ignores the wrong in the latter case is sorely deficient. (Although Peters does not believe that the latter type of wrong presents an equality issue, Greenawalt correctly disagrees.)
196. 120 S. Ct. 1073 (2000) (per curiam).
197. Id. at 1074-75.
198. See Peters, Equality Revisited, supra note 1, at 1252-53.
199. As Peters concedes, if "prescriptive" equality demands that Unlucky also obtain a windfall, this can be done by giving him a larger check initially (if his treatment was second) or by retrospectively giving him an additional check (if his treatment was first and he initially received the correct amount). See id. at 1226-27 n.33. One small difference is the administrative cost of issuing a second check. If the solution to inequality is to level "Lucky" down to the correct level, a more serious difference exists. For it is often infeasible or wrong to deprive Lucky of a windfall she has already received and spent, but this is not the case if her treatment is second and we intervene promptly. (Similar problems arise if the issue is incorrect deprivation rather than incorrect beneficence.)
200. See Greenawalt, supra note 1, at 1272, 1282. Greenawalt notes that this is more likely to occur when we have two different decisionmakers. For example, judge A predicts that judge B will make a mistake. And usually it is easier to recognize past mistakes than to predict future ones; but still, in principle, "prescriptive" equality is not tied to the time sequence. See id. at 1282.
201. See Peters, Foolish Consistency, supra note 34, at 2064.
202. Greenawalt agrees with Peters that prescriptive equality does not support the doctrine of precedent, but he agrees for two special reasons. See Greenawalt, supra note 1, at 1289. First, the people involved in the two cases are often not significantly related to one another, and Greenawalt believes that a significant relationship is a precondition to applying prescriptive equality. See id. (For my criticism of this requirement, see infra note 241.) Second, in civil cases, "[I]f one litigant [in a given case] is to receive better than otherwise deserved treatment because of precedent . . . then the competing litigant [in that case] will get worse treatment than he otherwise deserves." Thus, following the precedent in the name of equality will give one of the litigants worse treatment than she or he deserves. Id. Although Greenawalt believes that prescriptive equality has force when it results in giving people better treatment than they deserve, he is agnostic about whether it has force when it results in giving people worse treatment. See id.
203. See infra Part IV.D. Alexander makes a further argument purporting to show that the doctrine of precedent does not express a defensible egalitarian norm. See Alexander, Constrained by Precedent, supra note 35, at 10. Alexander asserts that in the context of precedent, "the comparative justice [or equality] claim is actually stronger for following a decision that is seriously wrong than for following a decision that is only minimally wrong," insofar as the winner in the present cases will be much better off than her equally deserving but losing counterpart in the earlier case. Id. Although the noncomparative justice reasons for following the earlier case are correspondingly weaker, the result is still that we have the same overall balance of reasons for following seriously wrong past decisions as for following minimally wrong past decisions. But, Alexander notes, we do not in fact believe that it is as important to follow seriously wrong past decisions. Thus, he concludes, we must not believe intertemporal equality is a weighty value. See id. Alexander's argument relies on a false premise. It is not necessarily true that the equality claim for following a seriously wrong precedent is especially strong. Equality principles can be sensitive to differentials in wellbeing, but they need not be. For example, they can depend more on the permissibility of the criteria of distinction (as in prohibitions of racial and gender discrimination). In the context of precedent, the equality norm might be this: Although a court has considerable discretion concerning which principle of law to adopt in a case of first impression, equality demands that the court have a substantial reason for not following a precedent. This norm is based on equality insofar as it requires an explanation for differences in treatment of those who claim to be similarly situated. Of course, such a norm might often be less weighty than the need to correct a past mistake, and the more serious the past mistake, the weightier the competing need. But the equality norm requiring explanation can have a weight that is insensitive to the seriousness of the past mistake.
204. For instance, if L receives a permit in January, and M is denied a permit on similar facts in February, we are normally more troubled than if L receives a permit in January and L is then denied another one on similar facts in February.
205. See Peters, Equality Revisited, supra note 1, at 1250 n.58 ("The prescriptive egalitarian holds that unequal treatment of identically situated people, measured according to any legitimate standard of identicality, is always to some degree wrong.").
206. Id. at 1259.
207. Id. at 1260.
208. Id.
209. Id.
210. Id.
211. It is plausible to question whether what Rawls is concerned about is the inequality of distribution of primary social goods among different social and economic classes, or instead the need to give priority to the needs of those who are worst off. In an important sense, the latter interpretation is not egalitarian. See Parfit, supra note 37, at 35-39; TEMKIN, supra note 37, at 245. But this is not Peters' point.
212. Peters, Equality Revisited, supra note 1, at 1262. In another argument, Peters conjures a scenario in which the difference principle might incorporate "prescriptive" equality, and in which the principle should therefore, he believes, be rejected. See id. at 1261. He supposes essentially the Lucky/Unlucky lottery ticket fact pattern, and conjectures that the difference principle might support multiplying Lucky's windfall and providing it to Unlucky as well. See id. But, as Peters might realize, Rawls never intended the difference principle to apply to distributions that depart from what his theory requires. See id. at 1261 n.80. Rather, he is concerned to explain what people deserve in the first instance, as a result of implementing basic principles that would be chosen in the original position. As Rawls explains, he mainly considers "strict compliance as opposed to partial compliance theory"; but it is the latter that examines how we should respond to actual injustice in a non-ideal world. RAWLS, supra note 58, at 8.
213. Peters, Equality Revisited, supra note 1, at 1262.
214. See id. at 1220, 1230 n.38, 1254-55, 1259-62; see also WESTEN, SPEAKING OF EQUALITY, supra note 2, at 120-21 ("The significant question for purposes of equality, therefore, is whether persons and tangible things are identical in relevant respects . . . . Equality is a relationship that obtains among persons or things by reference to such standards of comparison as have been independently established as appropriate to the states of affairs one
215. See supra text accompanying note 29.
216. As Professor Paul Brest explains: [I]f all race-dependent decisions were irrational, there would be no need for an antidiscrimination principle, for it would suffice to apply the widely held moral, constitutional, and practical principle that forbids treating persons irrationally. The antidiscrimination principle fills a special need because . . . race-dependent decisions that are rational and purport to be based solely on legitimate considerations are likely in fact to rest on assumptions of the differential worth of racial groups or on the related phenomenon of racially selective sympathy or indifference. Paul Brest, Foreword: In Defense of the Antidiscrimation Principle, 90 HARV. L. REV. 1, 7 (1976) (footnote omitted); see also Larry Alexander, What Makes Wrongful Discrimination Wrong? Biases, Preferences, Stereotypes, and Proxies, 141 U. PA. L. REV. 149, 170-72 (1992).
217. As Professor Paul Brest explains: [I]f all race-dependent decisions were irrational, there would be no need for an antidiscrimination principle, for it would suffice to apply the widely held moral, constitutional, and practical principle that forbids treating persons irrationally. The antidiscrimination principle fills a special need because . . . race-dependent decisions that are rational and purport to be based solely on legitimate considerations are likely in fact to rest on assumptions of the differential worth of racial groups or on the related phenomenon of racially selective sympathy or indifference. Paul Brest, Foreword: In Defense of the Antidiscrimation Principle, 90 HARV. L. REV. 1, 7 (1976) (footnote omitted); see also Larry Alexander, What Makes Wrongful Discrimination Wrong? Biases, Preferences, Stereotypes, and Proxies, 141 U. PA. L. REV. 149, 170-72 (1992).
218. Peters does offer an explanation, suggesting that opponents of affirmative action think race is always irrelevant, while proponents argue that a preference is "based on a relevant criterion . . . : the fact that the African-American business owner, due to historical and sociological conditions, faces artificial obstacles to obtaining government contracts that the Caucasian owner does not face." Peters, Equality Revisited, supra note 1, at 1256 n.66. But it is misleading to describe this debate as a battle over which nonegalitarian criterion of relevance to use. Rather, each criterion expresses a different egalitarian norm: the opponents argue that a relative preference unfairly distributes benefits unequally based on race; while the proponents argue that a relative preference fairly distributes benefits in order to promote equal opportunity, or diversity, and to redress historical inequality.
219. See Craig v. Boren, 429 U.S. 190, 190-210 (1976).
220. See Califano v. Goldfarb, 430 U.S. 199 (1977); Weinberger v. Wiesenfeld, 420 U.S. 636 (1975); Frontiere v. Richardson, 411 U.S. 677 (1973).
221. See, e.g., United States v. Virginia, 518 U.S. 515 (1996).
222. See Barbara Underwood, Ending Race Discrimination in Jury Selection: Whose Right Is It, Anyway?, 92 COLUM. L. REV. 725, 732 n.30 (1992) (citing studies). But there is also contrary evidence. See id. at 731-32.
223. See Batson v. Kentucky, 476 U.S. 79 (1986) (finding that the equal protection clause forbids prosecutors in criminal cases from exercising preemptory challenges of potential jurors based solely on account of their race or based on the assumption that black jurors as a group are unable to impartially consider a case against a black defendant); see also J.E.B. v. Alabama ex rel. T. B., 511 U.S. 127 (1994) (applying Batson to prohibit gender-based preemptory challenges).
224. See Peters, Equality Revisited, supra note 1, at 1256 ("The wrong of . . . decisions [based on especially odious criteria like race or gender] lies not in the difference in treatments that results from them; it lies in the application of an especially harmful irrelevant criterion to produce that difference.").
225. This analysis might help explain why an equality claim against the employer in examples (2.D) and (2.E), infra text accompanying notes 200-03, can still be viable even if he promotes neither employee, if his reason for promoting neither was based on a sexist desire to avoid promoting Wilma. (He might choose uniform non-promotion in order to avoid detection of his sexist motivation.) On first glance, forbidding the employer from "leveling down" or "multiplying the wrong" in this way seems inconsistent with my earlier analysis. But it is not, because an equality right can be designed to equalize status, not merely (and sometimes not even) to equalize tangible benefits. See supra Part II.D. Denying promotion to both employees still treats Wilma worse than she would have been treated had she been a man. And it represents a distinct insult to her, possibly undermining her status in the workplace in a way that denying promotion to Mark does not. (Finally, an equality right can be impure, thus requiring leveling up.)
226. See generally CHEMERINSKY, supra note 62 (ch. 10); Simons, supra note 52, at 467-72.
227. Compare this example with Justice Scalia's argument that affirmative action should be limited to remedying specific identified instances of past discrimination, a remedy that no longer requires any reference to race at all. See City of Richmond v. J.A. Crosson Co., 488 U.S. 469, 526-28 (1989) (Scalia, J., concurring). In effect, he argues for race being considered, not as a comparative group right (blacks should be preferred to whites in certain respects), but as a noncomparative individual right (all persons are entitled to a judicial remedy for an individual rights violation; some blacks are so entitled). See id. Of course, his argument still reflects the logic of comparative rights, for the historical individual rights violations which now justify a judicial remedy were themselves violations of the comparative right not to suffer invidious discrimination.
228. These examples involve the operation of visible and invisible legal principles. Similar problems can arise within a moral system that recognizes either lexically ordered principles or reasons with widely varying weights.
229. A possible example is the Presidential Recordings and Material Preservation Act, enacted after President Nixon's resignation, explicitly limited to Nixon and regulating the release of his Presidential papers and tape recordings. See Nixon Administrator of Gen. Servs., 433 U.S. 425 (1977).
230. Westen, in an otherwise useful analysis of the Twenty-Sixth Amendment to the Constitution, fails to identify the invisible equality component within that facially noncomparative rule and thus misstates the significance of the rule. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 136-41. Westen points out that the Amendment provides both a federal and a state right to vote to those who are eighteen years or older whenever older citizens have such a right. Id. at 138. However, because the federal Constitution elsewhere establishes when older citizens have such a right, the effect of the Amendment is to give eighteen year olds a noncomparative federal right to vote for certain federal officials. See id. at 137-38. But with respect to state elections, Westen asserts, the effect of the Amendment is very different. Since the U.S. Constitution does not create any significant requirements with respect to the state offices that must be elective, the Amendment only creates a contingent right: If a state allows persons older than eighteen to vote in an election, it shall also allow eighteen year olds to do so. See id. All of this is true enough. But Westen then proceeds to analyze the federal right of eighteen-year olds to vote as if it had nothing to do with nondiscrimination. See id. at 140. Yet that right, albeit noncomparative, has the scope that it does precisely because a comparative nondiscrimination right (contained in the Amendment) was combined with other constitutional voting rights. If the Constitution were amended to provide that members of the U.S. Senate were no longer to be elected, or that Supreme Court Justices were to be elected, then the Amendment would again be triggered and would affect the scope of the ultimate federal right to vote. Consider also Raz's discussion of Dworkin's fundamental right of equal concern and respect. See RAZ, supra note 22, at 220 (interpreting Dworkin too narrowly as endorsing only a noncomparative right, when in fact, Dworkin means at least to emphasize the impermissibility of discrimination along this dimension); see also Simons, supra note 52, at 480 n.229.
231. Could the notion of "impartiality" be understood in noncomparative terms? I think not. "To act 'impartially,'" Westen suggests, "is to act without favoring one person vis-a-vis others." WESTEN, SPEAKING OF EQUALITY, supra note 2, at 224. Surprisingly, despite this plausible (and comparative) definition, Westen disagrees with my view that impartiality is a necessarily comparative principle. He gives the example of a piano teacher who contracts separately with two students to give each sixty instructional minutes a week, but who skimps on the time of one of the students. See id. at 225. Westen correctly notes that the contractual right of each student is noncomparative, and concludes: "Yet we would commonly say . . . that by denying one student (and not the others) the sixty-minute lesson to which he is noncomparatively entitled, the teacher is not treating the two students 'impartially.'" Id. True enough. But the reason we would say this is because the teacher seems to be favoring one student over the other. Such favoritism violates a higher-level egalitarian principle, that one should not differentially honor (first level, noncomparative) entitlements out of unjustifiable partiality, or simply without good reason.
232. Westen's discussion of unequal implementation of entitlements, in his category of "prescriptions as baselines within descriptive standards of comparison," neglects to consider the possibility of egalitarian metaprinciples for judging such inequalities. See WESTEN, SPEAKING OF EQUALITY, supra note 2, at 86-92.
233. Peters offers an argument to explain why only nonegalitarian principles justify proportional reduction in this type of case. See Peters, Equality Revisited, supra note 1, at 1234-35. He states that if the original rule says that A and B are each independently entitled to a particular benefit, and if scarcity occurs, then proportional reduction is required. This is not because of the demands of equality, but because the original rule meant that there was no relevant difference between A and B with respect to obtaining the promised benefit. See id. at 1235. This is a clever but unpersuasive argument. The original entitlement rule might have meant no more or less than it said: A and B are each entitled to the full benefit. If the benefit cannot be fully distributed, other principles might have to come into play, and the original rule might have little or no force. For a helpful discussion of the possible choices in this type of situation, see Greenawalt, supra note 1, at 1276-78 (discussing Peters' example of a drug in short supply).
234. See Peters, Equality Revisited, supra note 1, at 1237, discussed supra text accompanying notes 40-42.
235. Raz discusses this issue under the heading "Principles of equal distribution in conflict." He discusses the principle: (5.1) If there are n F's each is entitled to 1/n of all the G. Raz suggests that a genuinely egalitarian principle underlies (5.1), namely: (5.2) In scarcity each who has equal entitlement is entitled to an equal share. RAZ, supra note 22, at 222-24. Raz is correct that the comparative equality principle (5.2) could explain (5.1). But two noncomparative principles (call them (5.3) and (5.4)) could also explain (5.1): (5.3) Every F is entitled to as much G as possible. (5.4) No other principle is relevant in distributing G to any F. Then, when scarcity occurs, we would violate these principles if we gave a larger share of G to some Fs than to other Fs.
236. This is not necessarily the case, however. Even with a competitive or scarce good, noncomparative rights can operate. A supermarket advertising "free beach balls while quantities last" grants each customer a (defeasible) noncomparative right to a free beach ball. When quantities run out, that defeasible condition kicks in. Although the actions of other customers affect the condition, this fact does not render the right comparative in any important sense. The boundary between comparative and noncomparative rights is sometimes difficult to draw. I do not have the space to explore the issue further here.
237. See Peters, Equality Revisited, supra note 1, at 1240.
238. Id.
239. See RAE, ET AL., supra note 66 passim.
240. See supra Part II.D.
241. An interesting analytical problem arises here. Are equality principles with respect to indivisible benefits subject to the leveling down and multiplication of wrong objections? It might seem they are not. If the "benefit" is a place on the lifeboat, then one can level up (as many as possible get a seat) or down (no one does). But if the "benefit" is defined as an equal chance for a place, how could one level up or down? In the example, everyone deserves a 10/11 chance. Consider a less exotic example: discrimination in filling the only supervisor position in a job category. If the person hired is no more qualified than the person passed over, it is infeasible to level up, yet it seems absurd to level down. The appropriate analysis is to require an equal opportunity for the job: each candidate deserves equal consideration, apart from race or other impermissible traits. But how can "equal consideration" be a matter of leveling up or down? The answer is that the decisionmaker might have discretion concerning the nature of the benefit that is offered, and this discretion might permit "leveling up" or "leveling down" in the relevant sense. In the lifeboat example, the captain might be in position to save either a rickety 10-person boat or a more sturdy 8-person boat; the survivors have a right to an equal chance at the available seats on whatever boat is saved (for example a right to a random selection method). In the employment example, the employer has some discretion to advertise more or less widely, even though the extent of advertising affects the prospects of the two candidates before her. (Or she may redefine the nature or emoluments of the job.)
242. Indeed, he might have standing to raise the equality claim, but he is unlikely to have standing to raise the noncomparative right claim, considered by itself.
243. Peters implicitly argues only for the leveling up form of "prescriptive equality:" Unlucky wants to get what Lucky received, not to have Lucky leveled down. But a pure equality right is satisfied by either remedy. Moreover, the force of the noncomparative right might result in the leveling down remedy, if Lucky has already disposed of the windfall proceeds.
244. See, e.g., RAZ, supra note 22, at 229 (arguing that equality is used contextually and argumentatively, but not normatively, when we make the claim: "'You seem to acknowledge the force of the reason in one case so why do you deny it in the other?'").
245. See Kenneth W. Simons, Overinclusion and Underinclusion: A New Model, 36 UCLA L. REV. 447, 459 (1989).
246. See, e.g., City of Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 448-50 (1985) (holding that zoning which excludes a group home for the retarded cannot be rationally justified by supposed legitimate government interests, when the interests cannot explain the beneficial treatment given to other permitted uses).
247. In constitutional jargon, this is a complaint of overbreadth. See Simons, supra note 190, at 482-88. The claim also expresses one sense of the idea that "unequals should be treated unequally" and that "unalikes should be treated unalike." See Simons, supra note 52, at 437-42.
248. See Greenawalt, supra note 1, at 1177 ("For [a] rule to survive [an equal protection rational basis challenge], the state needs not only a legitimate reason for treating the claimants the way it does - that reason would suffice to meet a straightforward substantive due process attack; the state must also have a legitimate reason for drawing the lines of inclusion and exclusion as it does."); Simons, supra note 190.
249. Even when these caveats apply, however, they will often require, not full implementation of a rule, but only selective implementation.
250. However, we can distinguish egalitarian from nonegalitarian concerns about classificatory fit. For example, the egalitarian concern is whether the asserted argument for differential treatment proves too little or too much; the nonegalitarian concern might be whether the asserted purpose is a fraud, or is insufficiently furthered to justify infringement of the plaintiff's rights or interests. See Simons, supra note 190, at 452-54.
251. By contrast, Peters' lottery example, example (3.B), supra p. 724, involves incorrect beneficence rather than incorrect deprivation. Here, "multiplication of the wrong" would mean a multiplication of incorrect beneficence (giving Unlucky a windfall because Lucky received one). On whether this should count as a "wrong" or "injustice," see supra note 115.
252. For helpful discussions of discretion, see DWORKIN, supra note 95, at 31-35, 68-71; FREDERICK SCHAUER, PLAYING BY THE RULES 222-28 (1991). Recall that a benefit is optional if it could initially have been denied; but it becomes vested if, once conferred, it cannot be freely withdrawn. See supra p. 719. Optional benefits might or might not be vested. And those that are vested can be strongly or weakly so. The strength of the entitlement can be measured by its temporal duration (e.g., benefits cannot be retroactively withdrawn, or, more strongly, must be distributed prospectively for some period of time) or by its ability to override other moral or legal considerations. All legal entitlements might be considered "optional" in the sense that the political system might ultimately change the entitlement (even such entitlements as constitutional rights). But, for purposes of equality analysis, the question is more limited: does the decisionmaker (a judge, or administrator, or legislator, or parent) have the option, at the time of the decision, to redefine or alter the recipient's entitlement? If not, then the entitlement is mandatory in the relevant sense, even though the entitlement can be altered by some other authority, or by that decisionmaker at some other time.
253. Discretion might also, or instead, be limited by the requirement that the decisionmaker act only for certain reasons. For example, a sentencing judge might be permitted to act only on the basis of retribution, deterrence, cooperation with the police, and other identified factors. The decision remains discretionary, however, if the judge is permitted to weigh and balance these factors as she likes.
254. See Alexander, Constrained by Precedent, supra note 35, at 11-12 (asserting that intertemporal equality has little weight except in two situations: where the good in question is a discretionary benefit and where it is competitive, such that inequality amounts to deprivation). More precisely, the multiplication of wrongs objection is misplaced insofar as the decisionmaker has discretion to confer B within range R (e.g. a sentence from one to ten years). Suppose one judge has sentenced X to one year and Y to five years. If the judge could have imposed any sentence between one and ten years without committing a wrong, a later judge can remedy the unequal distribution of B within R either by leveling up within R (sentencing both to one year) or by leveling down (sentencing both to five years). The leveling down option depends, of course, on whether X's entitlement to a one year sentence vests once it is announced. Peters pays virtually no attention to discretionary decisions. It is therefore not surprising that he emphasizes so heavily the "multiplication of wrongs" objection to equality.
255. Federal anti-discrimination laws support "make whole" relief, generally interpreted as requiring promotion in this situation, where feasible. Thus, Title VII does presumptively entitle the victim of discrimination to back pay and retroactive promotion or reinstatement. CHARLES R. RICKEY, MANUAL ON EMPLOYMENT DISCRIMINATION LAW AND CIVIL ACTIONS IN THE FEDERAL COURTS 1-85 (2d ed. 1996) (citing Albemarle Paper Co. v. Moody, 422 U.S. 405, 422 (1975)); see also 42 U.S.C. §2000e-5(g)(1) ("If the court finds that the respondent has intentionally engaged in or is intentionally engaging in an unlawful employment practice charged in the complaint, the court may . . . order such affirmative action as may be appropriate, which may include, but is not limited to, reinstatement or hiring of employees, with or without back pay."). However, neither reinstatement of a fired employee nor promotion of an employee who was illegally passed over is required if a reduction in force or legitimate reorganization eliminated the position. See PLAYER, supra note 84, at 554 (discussing the ADEA).
256. Once an optional benefit becomes vested, however, revoking it is indeed wrongful. See supra p. 719.
257. The close relationship of the two objections also makes it unnecessary to clarify what counts as a "wrong" for purposes of the "multiplication" objection. The most plausible meaning is the denial of a person's rights or entitlements, to their detriment, because it is the meaning that justifies the rhetorical bite of the objection. Peters, however, also counts as a "wrong" any incorrect treatment, even if the error is to the benefit of the recipient. See supra note 115. Alexander, similarly, seems to adopt a broad notion of what constitutes a "wrong." He complains that equality cannot plausibly support a doctrine of precedent because we would then be committed to replicating "morally erroneous decisions." See Alexander, Constrained by Precedent, supra note 35, at 10; supra text accompanying note 35.
258. As a practical matter, Title VII, by authorizing "make whole" relief, often provides a more generous remedy than state law would otherwise have provided if the denial of the entitlement had not been based on a discriminatory ground forbidden by federal law. See supra note 200. Moreover, the Civil Rights Act of 1991 amended Title VII to permit punitive damages in some circumstances. 42 U.S.C. §1981a(b)(1).
259. More realistically, courts interpreting a constitutional or legislative equalization requirement are likely to permit the state to adopt a formula that minimizes inequalities by restricting them to a certain range. See generally ROBERT F. WILLIAMS, STATE CONSTITUTIONAL LAW: CASES AND MATERIALS 995-1035 (3d ed. 1999).
260. See supra Part II.C for discussion of the "pure/impure" distinction.
261. See ROBERT E. GOODIN, UTILITARIANISM AS A PUBLIC PHILOSOPHY 246-48 (1995), pointing out that although it is paradoxical to meet needs by destroying needed resources, leveling down can be the preferable way to eliminate inequality when the inequality takes the form of unequal satisfaction of nonbasic needs. Achieving a higher level of satisfaction of such relative needs is essentially a status good, and it is better to eliminate such status by leveling down than by leveling up. Consider the case of housing: Beyond assuring a decent minimum, government might choose to restrict the affluent from building more expensive housing rather than promote more expensive housing for the poor. Id. at 252-54.
262. Similarly, consider a weak demand-for-reasons equality right that requires the decisionmaker to have a rational reason for differential treatment. Consistent with such a right, a government could give a small economic benefit to some businesses but not other businesses that seem to be as deserving, merely for reasons of administrative convenience (so long as the criterion of unequal distribution does not violate a stronger equality right, such as a prohibition on racial discrimination). See, e.g., FCC v. Beach Communications, Inc., 508 U.S. 307, 315-16 (1993); New Orleans v. Dukes, 427 U.S. 297, 303 (1976). Leveling up all deserving businesses is considered too costly, but leveling down to achieve a stingy type of equality is considered insufficiently important in light of the value of the selective program.
263. Greenawalt gives a similar example where waste would not be objectionable. See Greenawalt, supra note 1, at 1277 (parent refuses to buy last remaining toy of a certain type for one child in order to avoid treating the other child unequally).
264. See id. at 1267, 1271 (offering another example of such a situation); see also WESTEN, SPEAKING OF EQUALITY, supra note 2, at 195-99, 211-19 (analyzing different rationales a judge may use for differential sentencing regarding the same criminal activity).
265. See Coons, supra note 119, at 69.
266. Or perhaps not. As Greenawalt points out, "sometimes we are confident that two people count as relatively equal - for example, that they should be punished equally -before we have decided what treatment they should receive." Greenawalt, supra note 1, at 1267 (footnote omitted). Westen offers an interesting response to this argument and the argument in the text. He claims that the sentencing judge implicitly is employing a rule to decide the treatment of each defendant, and it is that rule, not the concept of equality, that prompts the judge to treat them equally. The rule is to the effect that no criminal sentence shall exceed or fall short of the sentence that most closely corresponds in gravity to the defendant's blameworthiness, or that would most effectively deter, and so forth. WESTEN, SPEAKING OF EQUALITY, supra note 2, at 195-99; see also id. at 212-13 (examining various judicial rationales behind consistent and inconsistent sentencing). The problem with this response is familiar: the response asserts a noncomparative right, ignoring the comparative character of the original argument. Thus, suppose that the judge wishes to dispense a punishment corresponding to blame, and she knows what both defendants have done, but she is uncertain about what punishment they deserve. She might feel that a sentence between five and ten years in prison would be within the range of deserved punishment. But she might also know that the two defendants are equal in blame, and therefore conclude that it would be wrong to treat them differently. It would be especially troubling if the judge were to treat the two differently, even if neither the judge nor outside observers were confident in answering the noncomparative question of how much punishment each defendant deserves (without regard to the punishment that the other deserves). Westen is indeed correct that the sentencing decision embodies an implicit rule or set of rules, and that the rules themselves might permit discretion in setting the appropriate punishment. But it is critical that a comparative background rule is in effect (providing, roughly, that a judge exercising discretion in sentencing must treat equally those defendants who are identical with respect to the relevant purposes of punishment). If instead only the noncomparative rule "provide a sentence based on blame within the bounds of your discretion" were in effect, the judge could sentence one defendant to five years and the other to ten years, and the propriety of the first sentence would have no bearing whatsoever on the propriety of the second. Sarnoff's essential argument is that the right to equal treatment follows from uncertainty about which legal and moral criteria apply, and how they apply. Sarnoff, supra note 1. However, I largely agree with Peters' response that this argument proves too much, for such uncertainty always exists. See generally Peters, Slouching Towards Equality, supra note 1.
267. Or perhaps not. As Greenawalt points out, "sometimes we are confident that two people count as relatively equal - for example, that they should be punished equally - before we have decided what treatment they should receive." Greenawalt, supra note 1, at 1267 (footnote omitted). Westen offers an interesting response to this argument and the argument in the text. He claims that the sentencing judge implicitly is employing a rule to decide the treatment of each defendant, and it is that rule, not the concept of equality, that prompts the judge to treat them equally. The rule is to the effect that no criminal sentence shall exceed or fall short of the sentence that most closely corresponds in gravity to the defendant's blameworthiness, or that would most effectively deter, and so forth. WESTEN, SPEAKING OF EQUALITY, supra note 2, at 195-99; see also id. at 212-13 (examining various judicial rationales behind consistent and inconsistent sentencing). The problem with this response is familiar: the response asserts a noncomparative right, ignoring the comparative character of the original argument. Thus, suppose that the judge wishes to dispense a punishment corresponding to blame, and she knows what both defendants have done, but she is uncertain about what punishment they deserve. She might feel that a sentence between five and ten years in prison would be within the range of deserved punishment. But she might also know that the two defendants are equal in blame, and therefore conclude that it would be wrong to treat them differently. It would be especially troubling if the judge were to treat the two differently, even if neither the judge nor outside observers were confident in answering the noncomparative question of how much punishment each defendant deserves (without regard to the punishment that the other deserves). Westen is indeed correct that the sentencing decision embodies an implicit rule or set of rules, and that the rules themselves might permit discretion in setting the appropriate punishment. But it is critical that a comparative background rule is in effect (providing, roughly, that a judge exercising discretion in sentencing must treat equally those defendants who are identical with respect to the relevant purposes of punishment). If instead only the noncomparative rule "provide a sentence based on blame within the bounds of your discretion" were in effect, the judge could sentence one defendant to five years and the other to ten years, and the propriety of the first sentence would have no bearing whatsoever on the propriety of the second. Sarnoff's essential argument is that the right to equal treatment follows from uncertainty about which legal and moral criteria apply, and how they apply. Sarnoff, supra note 1. However, I largely agree with Peters' response that this argument proves too much, for such uncertainty always exists. See generally Peters, Slouching Towards Equality, supra note 1.
268. See also WESTEN, SPEAKING OF EQUALITY, supra note 2, at 197-98.
269. However, the rationale for this view is probably not that racist motivation is actually acceptable, but instead that the cost of inquiring into its existence would be too great because such inquiries would unduly limit socially beneficial exercises of discretion.
270. See J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127 (1994) (holding that the equal protection clause prohibits preemptory strikes based on gender); Batson v. Kentucky, 476 U.S. 79 (1986) (holding that the equal protection clause prohibits preemptory strikes based solely on race).
271. "Selective prosecution" is another context in which courts defer to prosecutors' need for flexibility and discretion. Chief Judge Richard Posner recently summarized the doctrine as follows: [Selective prosecution] has two meanings in law. The first is simply failing to prosecute all known lawbreakers, whether because of ineptitude or (more commonly) because of lack of adequate resources. The resulting pattern of nonenforcement may be random, or an effort may be made to get the most bang for the prosecutorial buck by concentrating on the most newsworthy lawbreakers, but in either case the result is that people who are equally guilty of crimes or other violations receive unequal treatment, with some being punished and others getting off scot-free. That form of selective prosecution, although it involves dramatically unequal legal treatment, has no standing in equal protection law . . . . The second form of selective prosecution, and the only one that is actionable under the federal Constitution, is where the decision to prosecute is made either in retaliation for the exercise of a constitutional right, such as the right to free speech or to the free exercise of religion, or because of membership in a vulnerable group. Esmail v. Macrane, 53 F.3d 176, 178 (7th Cir. 1995). Judge Posner acknowledges, however, another category of judicially cognizable "selective prosecution" - situations where a citizen is singled out for official harassment because of malice and spite. Id.
272. Greenawalt, supra note 1.
273. See id. at 1269-71, 1273-83. Greenawalt gives the following example. A professor determines that 89 is the top of the B+ range. By mistake, when S receives an 89, the professor gives her an A-. T also received an 89. Greenawalt suggests that "prescriptive equality" provides a reason to give T an A-as well, even though this contradicts the original grading scheme. Id. at 1271. As Greenawalt points out, this is a case of equality "pulling against the balance of other reasons," because the grading scheme might be partly designed to measure achievement absolutely, or consistently over time. Id. This is a justifiable instance of multiplying the original wrong, if no one else is similarly situated to S and T, and if the harm of departing slightly from the grading scheme is not serious.
274. Id. at 1281.
275. See CHEMERINSKY, supra note 62, at 529-31.
276. For discussions of the distinction, see TEMKIN, supra note 37 (ch. 9); McKerlie, Equality, supra note 37, at 275; Parfit, supra note 37, at 3.
277. An example would be Ronald Dworkin's assertion that a fundamental right to equal concern and respect underlies all other important political rights. See DWORKIN, supra note 95.
278. Parfit, supra note 37, at 6.
279. "[A] bare . . . desire to harm a politically unpopular group" is unacceptable. See Romer v. Evans, 517 U.S. 620, 634 (1996) (quoting Department of Agriculture v. Moreno, 413 U.S. 528, 534 (1973)); see also Jeremy Waldron, The Substance of Equality, 89 MICH. L. REV. 1350, 1359 (1991) (reviewing WESTEN, SPEAKING OF EQUALITY and arguing that "in advancing social justifications, we are required to accord equal concern to the interests of everyone in society"); Richard Pildes, Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism, 27 J. LEGAL STUD. 725, 744-47 (1998) (noting that Supreme Court's voting rights cases consider which state interests in the franchise are permissible and which are constitutionally excluded).
280. "[A] bare . . . desire to harm a politically unpopular group" is unacceptable. See Romer v. Evans, 517 U.S. 620, 634 (1996) (quoting Department of Agriculture v. Moreno, 413 U.S. 528, 534 (1973)); see also Jeremy Waldron, The Substance of Equality, 89 MICH. L. REV. 1350, 1359 (1991) (reviewing WESTEN, SPEAKING OF EQUALITY and arguing that "in advancing social justifications, we are required to accord equal concern to the interests of everyone in society"); Richard Pildes, Why Rights Are Not Trumps: Social Meanings, Expressive Harms, and Constitutionalism, 27 J. LEGAL STUD. 725, 744-47 (1998) (noting that Supreme Court's voting rights cases consider which state interests in the franchise are permissible and which are constitutionally excluded).
281. See infra note 257.
282. See supra Part IV.A.1.
283. See RAWLS, supra note 58, at 199 (explaining that the lexically prior principles of justice are best embodied at the constitutional level, while the lexically subsequent difference principle is best implemented through legislation).
284. For some illuminating accounts, see, for example, Alexander, supra note 162; Brest, supra note 162; Charles R. Lawrence, III, The Id, The Ego, and Equal Protection: Reckoning with Unconscious Racism, 39 STAN. L. REV. 317 (1987).
285. See supra text at notes 152-59 (discussing Peters' analysis of Rawls' arguments).
286. See infra text accompanying notes 248-54.
287. See supra text accompanying notes 21-22.
288. Peters concedes that equality can sometimes be a useful concept, but he believes that its value is quite limited: Sometimes calling attention to inequality of treatment reveals injustice in treatment; recognizing a symptom can help us diagnose the disease. But once the disease has been identified, the important thing is to remedy the injustice, not the resulting inequality. Peters, Equality Revisited, supra note 1, at 1257. I agree that a differential in tangible treatment is not the only concern of all equality rights. But Peters does not recognize that some differential (whether in tangible treatments or in status) is a necessary and distinctive concern of such equality rights.
289. See id. at 1258.
290. Indeed, it is precisely when the interest or benefit is itself constitutionally guaranteed, and not optional, that the suitability of equal protection analysis is most controversial. Thus, it is controversial whether equal protection fundamental interest analysis represents a distinct form of analysis or should instead be understood as an aspect of substantive due process. See Ira Lupu, Untangling the Strands of the Fourteenth Amendment, 77 MICH. L. REV. 981, 985 (1979); Simons, supra note 49, at 467-72.
291. See supra Part IV.C (discussing equality in both discretionary and nondiscretionary decisions).
292. For a similar example, see Greenawalt, supra note 1, at 1265-66.
293. These examples are discussed supra in the text accompanying notes 200-03.
294. But note the qualification, discussed supra in text accompanying note 170.
295. See Peters, Equality Revisited, supra note 1, at 1223.
296. Peters asserts that the prescriptive force of nontautological equality is greatest when it counsels the multiplication of wrongs, rather than the multiplication of correct treatment; for in the latter case, the nonegalitarian treatment rule already tells us to treat a second recipient correctly after a first recipient has been treated correctly. See Peters, Equality Revisited, supra note 1, at 1226.
297. See supra Part III.B.
298. See Greenawalt, supra note 1, at 1266, 1272, 1289. Greenawalt's limitations seem too restrictive, however. Unrelated persons could also have a right to demand reasons for differential treatment (as the equal protection rational basis test provides), including mistakenly beneficial treatment received by others. (Suppose the IRS made a small error that benefited 90% of a defined group of taxpayers; equality provides a reason for extending the benefit to the other 10%.) And the person who is treated worse might be entitled to equal treatment even if she is unaware of her disadvantage. (Suppose the other 10% are unaware of the error.)
299. See supra text accompanying notes 148-49, 207.
300. See supra Part II.C.
301. See NOZICK, supra note 96, at 239 (suggesting that "the envious person, if he cannot (also) possess a thing (talent, and so on) that someone else has, prefers that the other person not have it either"); see also RAWLS, supra note 58, at 530-41 (discussing envy). For a passionate, conservative critique of egalitarianism as resting on envy, see ROBERT BORK, SLOUCHING TOWARDS GOMORRAH: MODERN LIBERALISM AND AMERICAN DECLINE 66-82 (1996).
302. See RAWLS, supra note 58, at 533 (arguing that resentment is a moral feeling, while envy is not).
303. Recall the example of the university official choosing to voice a greeting rather than also shaking hands. See supra p. 752.
304. In the famous words of Justice Jackson: [T]here is no more effective practical guaranty against arbitrary and unreasonable government than to require that the principles of law which officials would impose upon a minority must be imposed generally. Conversely, nothing opens the door to arbitrary action so effectively as to allow those officials to pick and choose only a few to whom they will apply legislation and thus to escape the political retribution that might be visited upon them if larger numbers were affected. Railway Express Agency v. New York, 336 U.S. 106, 112-13 (1949) (Jackson, J., concurring). However, this political dynamic can also interfere with egalitarian goals. See Peters, Equality Revisited, supra note 1, at 1263-64 n.84 (discussing an educational funding dispute in the New Jersey school system, in which the wealthy school districts balked at "leveling down" their spending to equalize spending with poorer districts; this public outcry led the legislature to abandon the "leveling down" plan).
305. See supra notes 43-47 and accompanying text.
306. See supra text accompanying notes 116-18 (noting that a right of equal treatment does not always depend on showing that one class is descriptively equal in any respect to another class).
307. See supra Part IV.B (noting that equality is invoked not only to challenge the administration of law or the use of discretionary power under the law, but also to challenge the basic social policies behind legislation).
308. These difficulties have apparently not deterred the Canadian Supreme Court, however. It has found neutral laws discriminatory in effect, and has held that discrimination does not depend on a comparison to another class's more favorable current treatment. See Brooks v. Canada Safeway Ltd., 1 S.C.R. 1219 (1989). For a discussion of the case, see Kathleen E. Mahoney, The Constitutional Law of Equality in Canada, 44 ME. L. REV. 229, 240 (1992).
309. One possible objection is that equality's distinctive flexibility (normally permitting either leveling up or down) is missing when the egalitarian norm takes the form of a prohibition on acting with discriminatory intent or acting based on discriminatory stereotypes. I respond to that objection in Part II.D supra.
310. See generally Laura W. Stein, Living with the Risk of Backfire: A Response to the Feminist Critiques of Privacy and Equality, 77 MINN. L. REV. 1153, 1189-91 (1993) (challenging the view that feminists should adopt a uniform conceptual or doctrinal approach in critiquing injustice). In a recent article, Jane Cohen endorses a pluralistic approach to achieving gender equality that accommodates a broad range of personal goals, whether or not some of these goals are customarily considered distinctively "male." She also offers a thoughtful and qualified critique of contemporary feminist hesitation about employing egalitarian norms. See Jane Cohen, Equality for Girls and Other Women: The Built Architecture of the Purposive Life, 9 J. CONTEMP. LEG. ISSUES 103 (1998).
311. See generally Laura W. Stein, Living with the Risk of Backfire: A Response to the Feminist Critiques of Privacy and Equality, 77 MINN. L. REV. 1153, 1189-91 (1993) (challenging the view that feminists should adopt a uniform conceptual or doctrinal approach in critiquing injustice). In a recent article, Jane Cohen endorses a pluralistic approach to achieving gender equality that accommodates a broad range of personal goals, whether or not some of these goals are customarily considered distinctively "male." She also offers a thoughtful and qualified critique of contemporary feminist hesitation about employing egalitarian norms. See Jane Cohen, Equality for Girls and Other Women: The Built Architecture of the Purposive Life, 9 J. CONTEMP. LEG. ISSUES 103 (1998).
312. See DEBORAH RHODE, THE POLITICS OF PARADIGMS: GENDER DIFFERENCE AND GENDER DISADVANTAGE, FROM BEYOND EQUALITY AND DIFFERENCE ch.8 (G. Bock & S. James eds. 1992).
313. See supra Part I.G.
314. Pregnancy Discrimination Act, 42 U.S.C. §2000e(k) (1994).
315. See, e.g., Harper v. Virginia Bd. of Elections, 383 U.S. 663, 668 (1966). In Harper, the Court invalidated a poll tax and essentially ignored state interests in obtaining revenues or in limiting the vote to those with a stronger interest in voting. Instead, the Court simply concluded that if the state chooses to make an office elective rather than appointive, it may not draw de facto wealth distinctions among voters, even if those distinctions would indeed serve these state interests. See also Simons, supra note 52, at 476-77; Pildes, supra note 223, at 744-47.
316. Cohen v. Brown Univ., 101 F.3d 155, 176-81(1st Cir. 1996), cert. denied, 520 U.S. 1186 (1997).
317. Id. at 179-80.
318. Even here, however, equality can play a part, if pluralism expresses the notion that different values or ideals are equally valuable. But insofar as pluralism expresses the idea that the different values deserving of respect are incommensurable, equality is inapposite.
319. Ward, supra note 44, at 98.
320. See generally AMARTYA SEN, INEQUALITY REEXAMINED (1992). Raz argues that "[w]e only have reason to care about inequalities in the distributions of goods and ills, that is of what is of value or disvalue for independent reasons. There is no reason to care about inequalities in the distribution of grains of sand, unless there is some other reason to wish to have or to avoid sand." RAZ, supra note 22, at 235. This argument is overstated, because for some equality rights, such as the right not to suffer racial discrimination, the subject-matter hardly matters. But it is indeed true of many equality rights, especially positive rights to the equal distribution of an opportunity or resource.
321. See generally THOMAS NAGEL, EQUALITY AND PARTIALITY (1991).