Cooperation and conflict: The economics of group status production and race discrimination

Harvard Law Review

Volumn 108, Issue 5, 1995, Pages 1005-1084

  McAdams, Richard H ^a  

^a CHICAGO KENT COLLEGE OF LAW   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 11944256065     PISSN: 0017811X     EISSN: None     Source Type: Journal    
DOI: 10.2307/1341869     Document Type: Article

References (391)

1. WILLIAM SHAKESPEARE, HENRY V, act 4, sc. 3, 11. 19-22 (John D. Wilson ed., Cambridge Univ. Press 1968) (1623).
2. Id. 11. 57-67.
3. See, e.g., GARY S. BECKER, THE ECONOMIC APPROACH TO HUMAN BEHAVIOR 14 (1976); RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 21 (4th ed. 1992); Fred S. McChesney, Desperately Shunning Science, 71 B.U. L. REV. 281, 281 (1991).
4. See, e.g., GARY S. BECKER, THE ECONOMIC APPROACH TO HUMAN BEHAVIOR 14 (1976); RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 21 (4th ed. 1992); Fred S. McChesney, Desperately Shunning Science, 71 B.U. L. REV. 281, 281 (1991).
5. See, e.g., GARY S. BECKER, THE ECONOMIC APPROACH TO HUMAN BEHAVIOR 14 (1976); RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 21 (4th ed. 1992); Fred S. McChesney, Desperately Shunning Science, 71 B.U. L. REV. 281, 281 (1991).
6. See MORRIS JANOWITZ, THE PROFESSIONAL SOLDIER 215, 220 (1960) ("In its contemporary form a major aspect of military honor comprises a sense of brotherhood and intense group loyalty."); S.L.A. MARSHALL, MEN AGAINST FIRE 149 (1947) ("[P]ersonal honor is the one thing valued more than life itself by the majority of men. The lips of the dying attest how strongly this force influences individual conduct in battle."); UNITED STATES DEP'T OF THE ARMY, FM 22-100, MILITARY LEADERSHIP 33, 39 (1961). Echoing King Henry's St. Crispian speech, the military spends millions of dollars annually to entice volunteers for "the few, the proud, the Marines." See Paula Span, The Marines Go Medieval, WASH. POST, Mar. 22, 1992, (Magazine) at 25 (noting that the Marine Corps' ad budget is almost $10 million).
7. See MORRIS JANOWITZ, THE PROFESSIONAL SOLDIER 215, 220 (1960) ("In its contemporary form a major aspect of military honor comprises a sense of brotherhood and intense group loyalty."); S.L.A. MARSHALL, MEN AGAINST FIRE 149 (1947) ("[P]ersonal honor is the one thing valued more than life itself by the majority of men. The lips of the dying attest how strongly this force influences individual conduct in battle."); UNITED STATES DEP'T OF THE ARMY, FM 22-100, MILITARY LEADERSHIP 33, 39 (1961). Echoing King Henry's St. Crispian speech, the military spends millions of dollars annually to entice volunteers for "the few, the proud, the Marines." See Paula Span, The Marines Go Medieval, WASH. POST, Mar. 22, 1992, (Magazine) at 25 (noting that the Marine Corps' ad budget is almost $10 million).
8. See MORRIS JANOWITZ, THE PROFESSIONAL SOLDIER 215, 220 (1960) ("In its contemporary form a major aspect of military honor comprises a sense of brotherhood and intense group loyalty."); S.L.A. MARSHALL, MEN AGAINST FIRE 149 (1947) ("[P]ersonal honor is the one thing valued more than life itself by the majority of men. The lips of the dying attest how strongly this force influences individual conduct in battle."); UNITED STATES DEP'T OF THE ARMY, FM 22-100, MILITARY LEADERSHIP 33, 39 (1961). Echoing King Henry's St. Crispian speech, the military spends millions of dollars annually to entice volunteers for "the few, the proud, the Marines." See Paula Span, The Marines Go Medieval, WASH. POST, Mar. 22, 1992, (Magazine) at 25 (noting that the Marine Corps' ad budget is almost $10 million).
9. See MORRIS JANOWITZ, THE PROFESSIONAL SOLDIER 215, 220 (1960) ("In its contemporary form a major aspect of military honor comprises a sense of brotherhood and intense group loyalty."); S.L.A. MARSHALL, MEN AGAINST FIRE 149 (1947) ("[P]ersonal honor is the one thing valued more than life itself by the majority of men. The lips of the dying attest how strongly this force influences individual conduct in battle."); UNITED STATES DEP'T OF THE ARMY, FM 22-100, MILITARY LEADERSHIP 33, 39 (1961). Echoing King Henry's St. Crispian speech, the military spends millions of dollars annually to entice volunteers for "the few, the proud, the Marines." See Paula Span, The Marines Go Medieval, WASH. POST, Mar. 22, 1992, (Magazine) at 25 (noting that the Marine Corps' ad budget is almost $10 million).
10. HENRY W. GRADY, THE NEW SOUTH AND OTHER ADDRESSES 55 (1969).
11. See RONALD H. COASE, THE FIRM, THE MARKET, AND THE LAW 33-55 (1988).
12. Thomas Wolfe, Chickamauga, in CHICKAMAUGA AND OTHER CIVIL WAR STORIES 41, 60 (Shelby Foote ed., 1993) ("[The Civil War] would give us all a chance to wear a uniform . . . , to shoot some Yankees and to run 'em north, and then to come back home and lord it over those who hadn't been and be a hero and court the gals.").
13. GRADY, supra note 5, at 55.
14. DAVID HUME, ESSAYS 54, 58 (Eugene F. Miller ed., Liberty Classics 1985) (1758).
15. See ADAM SMITH, 1 THE WEALTH OF NATIONS 477-78 (Edwin Cannan ed., Univ. of Chicago Press 1976) (1776).
16. Jon Elster concisely describes the collective action problem: Suppose that each member of a group has the choice between engaging in a certain activity and not engaging in it. The group has a collective action problem if it is better for all if some do it than if nobody does, but better for each not to do it. JON ELSTER, NUTS AND BOLTS FOR THE SOCIAL SCIENCES 126 (1989); see RUSSELL HARDIN, COLLECTIVE ACTION 8-9, 16-22 (1982); MANCUR OLSON, JR., THE LOGIC OF COLLECTIVE ACTION 5-22, 53-65 (1965).
17. Jon Elster concisely describes the collective action problem: Suppose that each member of a group has the choice between engaging in a certain activity and not engaging in it. The group has a collective action problem if it is better for all if some do it than if nobody does, but better for each not to do it. JON ELSTER, NUTS AND BOLTS FOR THE SOCIAL SCIENCES 126 (1989); see RUSSELL HARDIN, COLLECTIVE ACTION 8-9, 16-22 (1982); MANCUR OLSON, JR., THE LOGIC OF COLLECTIVE ACTION 5-22, 53-65 (1965).
18. Jon Elster concisely describes the collective action problem: Suppose that each member of a group has the choice between engaging in a certain activity and not engaging in it. The group has a collective action problem if it is better for all if some do it than if nobody does, but better for each not to do it. JON ELSTER, NUTS AND BOLTS FOR THE SOCIAL SCIENCES 126 (1989); see RUSSELL HARDIN, COLLECTIVE ACTION 8-9, 16-22 (1982); MANCUR OLSON, JR., THE LOGIC OF COLLECTIVE ACTION 5-22, 53-65 (1965).
19. In the example from which the dilemma draws its name, a prosecutor offers two prisoners the same bargain. If both prisoners remain silent, the prosecutor will use existing evidence to convict each of a minor crime for which each will serve one year in prison. If only one of the two prisoners confesses, the confessor will serve no time, but her testimony will be used to convict the non-confessor of a more serious crime for which she will serve five years. If both prisoners confess, each will serve an intermediate sentence of three years. Each prisoner must make an irrevocable decision without knowing how the other decides. Under these conditions, there is a "dominant" strategy, meaning a strategy that is best for an individual regardless of what decision the other individual makes. If prisoner B confesses, prisoner A is better off confessing and thereby serving three years instead of five. If B remains silent, A is still better off confessing and thereby serving zero years instead of three. Prisoner B faces exactly the same set of considerations. The dilemma occurs because A and B, each acting to further her own interests, will each decide to confess, yet mutual confession yields an outcome (three years each) that is for each unambiguously inferior to mutual silence (one year each). Although each would prefer mutual silence to mutual confession, the individual nature of the decision making produces the latter.
20. See, e.g., ROBERT AXELROD, THE EVOLUTION OF COOPERATION 7 (1984); ELINOR OSTROM, GOVERNING THE COMMONS: THE EVOLUTION OF INSTITUTIONS FOR COLLECTIVE ACTION 23-38 (1990); WILLIAM POUNDSTONE, PRISONER'S DILEMMA 1-9 (1992).
21. See, e.g., ROBERT AXELROD, THE EVOLUTION OF COOPERATION 7 (1984); ELINOR OSTROM, GOVERNING THE COMMONS: THE EVOLUTION OF INSTITUTIONS FOR COLLECTIVE ACTION 23-38 (1990); WILLIAM POUNDSTONE, PRISONER'S DILEMMA 1-9 (1992).
22. See, e.g., ROBERT AXELROD, THE EVOLUTION OF COOPERATION 7 (1984); ELINOR OSTROM, GOVERNING THE COMMONS: THE EVOLUTION OF INSTITUTIONS FOR COLLECTIVE ACTION 23-38 (1990); WILLIAM POUNDSTONE, PRISONER'S DILEMMA 1-9 (1992).
23. Public goods are those goods that are non-exhaustible - because consumption by one does not preclude consumption by others - and non-excludable - because it is not feasible to prevent non-contributing individuals from enjoying the good once it is produced. See ROBERT COOTER & THOMAS ULEN, LAW AND ECONOMICS 46 (1988). Since everyone benefits when one produces a public good, selfish individuals attempt to "free-ride" on the contributions of others, which leads the market to undersupply such goods. See id. at 46-48.
24. When a depletable resource is publicly owned, each individual gains the full benefit of her consumption but shares the cost with all other users. The result is inefficient depletion of the resource. See OSTROM, supra note 13, at 1-3, 103-10; Garrett Hardin, The Tragedy of the Commons, 162 SCIENCE 1243, 1244 (1968).
25. The state can solve the commons problem, for example, by defining and enforcing private property rights. When one person bears all the costs of resource depletion, she will stop consuming when the marginal cost of consumption exceeds the marginal benefit. See OSTROM, supra note 13, at 12-13. When private ownership is impractical, regulation may simply prohibit over-consumption. See id. at 8-11.
26. See, e.g., Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 VA. L. REV. 1089, 1103-05, 1109-11 (1981); Saul Levmore, Monitors and Freeriders in Commercial and Corporate Settings, 92 YALE L.J. 49, 76-85 (1982). Indeed, legal rules are sometimes justified by the need to maintain problems of collective action. Each firm in a cartel, for example, has the incentive to free-ride on price-fixing by underselling the cartel price, a strategy that prevents the cartel from existing at all. Antitrust policy is urgently concerned with the tools firms might use to overcome this collective action problem. See Ian Ayres, How Cartels Punish: A Structural Theory of Self-Enforcing Collusion, 87 COLUM. L. REV. 295, 304-12 (1987). Indeed, the argument of Part II.C.2 is that antidiscrimination laws are needed to prevent collective action by whites.
27. See, e.g., Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 VA. L. REV. 1089, 1103-05, 1109-11 (1981); Saul Levmore, Monitors and Freeriders in Commercial and Corporate Settings, 92 YALE L.J. 49, 76-85 (1982). Indeed, legal rules are sometimes justified by the need to maintain problems of collective action. Each firm in a cartel, for example, has the incentive to free-ride on price-fixing by underselling the cartel price, a strategy that prevents the cartel from existing at all. Antitrust policy is urgently concerned with the tools firms might use to overcome this collective action problem. See Ian Ayres, How Cartels Punish: A Structural Theory of Self-Enforcing Collusion, 87 COLUM. L. REV. 295, 304-12 (1987). Indeed, the argument of Part II.C.2 is that antidiscrimination laws are needed to prevent collective action by whites.
28. See, e.g., Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 VA. L. REV. 1089, 1103-05, 1109-11 (1981); Saul Levmore, Monitors and Freeriders in Commercial and Corporate Settings, 92 YALE L.J. 49, 76-85 (1982). Indeed, legal rules are sometimes justified by the need to maintain problems of collective action. Each firm in a cartel, for example, has the incentive to free-ride on price-fixing by underselling the cartel price, a strategy that prevents the cartel from existing at all. Antitrust policy is urgently concerned with the tools firms might use to overcome this collective action problem. See Ian Ayres, How Cartels Punish: A Structural Theory of Self-Enforcing Collusion, 87 COLUM. L. REV. 295, 304-12 (1987). Indeed, the argument of Part II.C.2 is that antidiscrimination laws are needed to prevent collective action by whites.
29. For example, in the summer of 1993, citizens of the communities flooded by the Mississippi River volunteered hours of hard physical labor, filling and stacking sand bags to save their towns. See Ralph Dummit, Neighbors Pitch In and Win, ST. Louis POST-DISPATCH, Aug. 6, 1993, (St. Charles) at rSC. Flood control cannot be apportioned solely to those who contribute to it, and an individual's contribution is not likely to determine the success of the group effort. One would, therefore, have expected everyone to free-ride.
30. See AMITAI ETZIONI, THE MORAL DIMENSION: TOWARD A NEW ECONOMICS 25-29 (1988); ROBERT H. FRANK, PASSIONS WITHIN REASON: THE STRATEGIC ROLE OF THE EMOTIONS 212-36 (1988); HOWARD MARGOLIS, SELFISHNESS, ALTRUISM, AND RATIONALITY: A THEORY OF SOCIAL CHOICE 11-14 (1982); Amartya K. Sen, Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, in BEYOND SELF-INTEREST 25, 34-36 (Jane J. Mansbridge ed., 1990).
31. See AMITAI ETZIONI, THE MORAL DIMENSION: TOWARD A NEW ECONOMICS 25-29 (1988); ROBERT H. FRANK, PASSIONS WITHIN REASON: THE STRATEGIC ROLE OF THE EMOTIONS 212-36 (1988); HOWARD MARGOLIS, SELFISHNESS, ALTRUISM, AND RATIONALITY: A THEORY OF SOCIAL CHOICE 11-14 (1982); Amartya K. Sen, Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, in BEYOND SELF-INTEREST 25, 34-36 (Jane J. Mansbridge ed., 1990).
32. See AMITAI ETZIONI, THE MORAL DIMENSION: TOWARD A NEW ECONOMICS 25-29 (1988); ROBERT H. FRANK, PASSIONS WITHIN REASON: THE STRATEGIC ROLE OF THE EMOTIONS 212-36 (1988); HOWARD MARGOLIS, SELFISHNESS, ALTRUISM, AND RATIONALITY: A THEORY OF SOCIAL CHOICE 11-14 (1982); Amartya K. Sen, Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, in BEYOND SELF-INTEREST 25, 34-36 (Jane J. Mansbridge ed., 1990).
33. See AMITAI ETZIONI, THE MORAL DIMENSION: TOWARD A NEW ECONOMICS 25-29 (1988); ROBERT H. FRANK, PASSIONS WITHIN REASON: THE STRATEGIC ROLE OF THE EMOTIONS 212-36 (1988); HOWARD MARGOLIS, SELFISHNESS, ALTRUISM, AND RATIONALITY: A THEORY OF SOCIAL CHOICE 11-14 (1982); Amartya K. Sen, Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, in BEYOND SELF-INTEREST 25, 34-36 (Jane J. Mansbridge ed., 1990).
34. See N.L. Kerr, Motivational Choices in Task Groups: A Paradigm for Social Dilemma Research, in EXPERIMENTAL SOCIAL DILEMMAS 1-9 (Henk A.M. Wilke, Dave M. Messick & Christel G. Rutte eds., 1986); David M. Messick & Marilynn B. Brewer, Solving Social Dilemmas: A Review, 4 REV. PERSONALITY & Soc. PSYCHOL. 11, 17-29 (1983). Until the early 19705, "experimental gaming was devoted overwhelmingly to two-person games," but has since focused on "[t]he Multi-Person . . . Prisoner's Dilemma . . . and the strategically equivalent Commons Dilemma." ANDREW M. COLMAN, GAME THEORY AND EXPERIMENTAL GAMES 168 (1982) (citations omitted).
35. See N.L. Kerr, Motivational Choices in Task Groups: A Paradigm for Social Dilemma Research, in EXPERIMENTAL SOCIAL DILEMMAS 1-9 (Henk A.M. Wilke, Dave M. Messick & Christel G. Rutte eds., 1986); David M. Messick & Marilynn B. Brewer, Solving Social Dilemmas: A Review, 4 REV. PERSONALITY & Soc. PSYCHOL. 11, 17-29 (1983). Until the early 19705, "experimental gaming was devoted overwhelmingly to two-person games," but has since focused on "[t]he Multi-Person . . . Prisoner's Dilemma . . . and the strategically equivalent Commons Dilemma." ANDREW M. COLMAN, GAME THEORY AND EXPERIMENTAL GAMES 168 (1982) (citations omitted).
36. See N.L. Kerr, Motivational Choices in Task Groups: A Paradigm for Social Dilemma Research, in EXPERIMENTAL SOCIAL DILEMMAS 1-9 (Henk A.M. Wilke, Dave M. Messick & Christel G. Rutte eds., 1986); David M. Messick & Marilynn B. Brewer, Solving Social Dilemmas: A Review, 4 REV. PERSONALITY & Soc. PSYCHOL. 11, 17-29 (1983). Until the early 19705, "experimental gaming was devoted overwhelmingly to two-person games," but has since focused on "[t]he Multi-Person . . . Prisoner's Dilemma . . . and the strategically equivalent Commons Dilemma." ANDREW M. COLMAN, GAME THEORY AND EXPERIMENTAL GAMES 168 (1982) (citations omitted).
37. See ETZIONI, supra note 19, at 60-61; Linnda R. Caporael, Robyn M. Dawes, John M. Orbell & Alphons J.C. van de Kragt, Selfishness Examined: Cooperation in the Absence of Egoistic Incentives, 12 BEHAV. & BRAIN SCI. 683, 687-92 (1989); Robyn M. Dawes, Social Dilemmas, 31 ANN. REV. PSYCHOL. 169, 185-86 (1980).
38. See ETZIONI, supra note 19, at 60-61; Linnda R. Caporael, Robyn M. Dawes, John M. Orbell & Alphons J.C. van de Kragt, Selfishness Examined: Cooperation in the Absence of Egoistic Incentives, 12 BEHAV. & BRAIN SCI. 683, 687-92 (1989); Robyn M. Dawes, Social Dilemmas, 31 ANN. REV. PSYCHOL. 169, 185-86 (1980).
39. See Robyn M. Dawes, Alphons J.C. van de Kragt & John M. Orbell, Cooperation for the Benefit of Us - Not Me, or My Conscience, in BEYOND SELF-INTEREST, supra note 19, at 97, 100.
40. John M. Orbell, Alphons J.C. van de Kragt & Robyn M. Dawes, Explaining Discussion-Induced Cooperation, 54 J. PERSONALITY & Soc. PSYCHOL. 811, 811 (1988). A recent study is illustrative. In a matrix with the highest "sucker's payoff" (the payoff for cooperating when the other party defects), there was a $7 loss if one cooperated when the other defected, and a $5 gain for defecting when the other cooperated; mutual cooperation paid $2 each; mutual defection incurred a $5 loss for each. See John M. Orbell & Robyn M. Dawes, Social Welfare, Cooperators' Advantage, and the Option of Not Playing the Game, 58 AM. SOC. REV. 787, 788, 791 (1993). The cooperation rate was 51%. In four other matrices, cooperation ranged from 35% to 54%. See id. at 794. These games, however, are merely the control for the manipulations that were actually the heart of this experiment, namely, dilemma games in which the players could "opt out" of the game entirely. See id. Thus, cooperation in such experiments is sufficiently established that researchers have moved on to more interesting questions. See also John M. Orbell, Peregrine Schwartz-Shea & Randy T. Simmons, Do Cooperators Exit More Readily Than Defectors?, 78 AM. POL. SCI. REV. 147, 147, 155-60 (1984) (proposing theories to explain why cooperators are less likely to exit than defectors).
41. John M. Orbell, Alphons J.C. van de Kragt & Robyn M. Dawes, Explaining Discussion-Induced Cooperation, 54 J. PERSONALITY & Soc. PSYCHOL. 811, 811 (1988). A recent study is illustrative. In a matrix with the highest "sucker's payoff" (the payoff for cooperating when the other party defects), there was a $7 loss if one cooperated when the other defected, and a $5 gain for defecting when the other cooperated; mutual cooperation paid $2 each; mutual defection incurred a $5 loss for each. See John M. Orbell & Robyn M. Dawes, Social Welfare, Cooperators' Advantage, and the Option of Not Playing the Game, 58 AM. SOC. REV. 787, 788, 791 (1993). The cooperation rate was 51%. In four other matrices, cooperation ranged from 35% to 54%. See id. at 794. These games, however, are merely the control for the manipulations that were actually the heart of this experiment, namely, dilemma games in which the players could "opt out" of the game entirely. See id. Thus, cooperation in such experiments is sufficiently established that researchers have moved on to more interesting questions. See also John M. Orbell, Peregrine Schwartz-Shea & Randy T. Simmons, Do Cooperators Exit More Readily Than Defectors?, 78 AM. POL. SCI. REV. 147, 147, 155-60 (1984) (proposing theories to explain why cooperators are less likely to exit than defectors).
42. John M. Orbell, Alphons J.C. van de Kragt & Robyn M. Dawes, Explaining Discussion-Induced Cooperation, 54 J. PERSONALITY & Soc. PSYCHOL. 811, 811 (1988). A recent study is illustrative. In a matrix with the highest "sucker's payoff" (the payoff for cooperating when the other party defects), there was a $7 loss if one cooperated when the other defected, and a $5 gain for defecting when the other cooperated; mutual cooperation paid $2 each; mutual defection incurred a $5 loss for each. See John M. Orbell & Robyn M. Dawes, Social Welfare, Cooperators' Advantage, and the Option of Not Playing the Game, 58 AM. SOC. REV. 787, 788, 791 (1993). The cooperation rate was 51%. In four other matrices, cooperation ranged from 35% to 54%. See id. at 794. These games, however, are merely the control for the manipulations that were actually the heart of this experiment, namely, dilemma games in which the players could "opt out" of the game entirely. See id. Thus, cooperation in such experiments is sufficiently established that researchers have moved on to more interesting questions. See also John M. Orbell, Peregrine Schwartz-Shea & Randy T. Simmons, Do Cooperators Exit More Readily Than Defectors?, 78 AM. POL. SCI. REV. 147, 147, 155-60 (1984) (proposing theories to explain why cooperators are less likely to exit than defectors).
43. One explanation is that the subjects do not understand the game, but there are several reasons to reject this possibility. First, researchers in these studies typically take great pains to ensure that the subjects understand the game. See Caporael, Dawes, Orbell & van de Kragt, supra note 21, at 688. Second, considerable evidence suggests that permitting subjects to talk with each other increases cooperation, even though discussion tends to increase the subjects' understanding of the strategic tension of the game. See infra notes 40-50 and accompanying text. In fact, some evidence suggests that, absent discussion, cooperators understand the strategic structure of the game better than defectors. See Harold H. Kelley & Janusz Grzelak, Conflict Between Individual and Common Interest in an N-Person Relationship, 21 J. PERSONALITY & SOC. PSYCHOL. 190, 195-96 (1972).
44. See ANATOL RAPOPORT & ALBERT M. CHAMMAH, PRISONER'S DILEMMA: A STUDY IN CONFLICT AND COOPERATION 207-08 (1965).
45. See COLMAN, supra note 20, at 124; see also AXELROD, supra note 13, at 40-42 (discussing computer program prisoner's dilemma tournaments won by tit-for-tat); id. at 73-87 (discussing real-world examples, including reciprocal cooperation in minimizing artillery casualties among opposing forces in World War I trench warfare). By reciprocating cooperation, a player using tit-for-tat offers other players the possibility of large, long-run gains from repeated cooperative interactions. By retaliating immediately and without exception to defection, the player also forecloses the possibility of long-run gains from defection. See id. at 53-54, 118-20.
46. See AXELROD, supra note 13, at 10. Even when the game is repeated, reciprocity works only if the players do not know when the game will end. See id. Otherwise, each player reasons that the last round of play represents a one-shot prisoner's dilemma in which the dominant strategy is to defect. For any round preceding a round in which the player expects mutual defection, the player has no reason to cooperate. This logic pushes back the decision to defect to the first round of play. See JON ELSTER, THE CEMENT OF SOCIETY: A STUDY OP SOCIAL ORDER 43 (1989) (expressing uncertainty about this reasoning but noting the "nearly unanimously held view that defection is the rational strategy" in finitely repeated games). But see HARDIN, supra note 11, at 145-50 (denying that rationality requires defection regardless of how many times the game is finitely repeated).
47. See, e.g., Robert H. Frank, Thomas Gilovich & Dennis T. Regan, Does Studying Economics Inhibit Cooperation?, 13, 14 fig.6 (1992) (unpublished manuscript, on file with the Harvard Law School Library) (finding that 39.6% of economics majors and 61.2% of non-economics majors cooperated in a prisoner's dilemma); Gerald Harwell & Ruth E. Ames, Experiments on the Provision of Public Goods. I. Resources, Interest, Group Size, and the Free-Rider Problem, 84 AM. J. Soc. 1335, 1349-50 (1979) (finding that the typical subject invested 57% of her available resources in a public good, two-thirds of subjects invested more than half, and 22% invested all); Orbell, van de Kragt & Dawes, supra note 23, at 813, 816-17 (finding that with discussion, 79% of subjects cooperated, while with no discussion and other factors, 20% cooperated); Orbell, Schwartz-Shea & Simmons, supra note 23, at 153 (finding that allowing exit and discussion led over 75% to cooperate and that allowing exit and no discussion led 18% of those who chose to stay to cooperate).
48. See, e.g., Robert H. Frank, Thomas Gilovich & Dennis T. Regan, Does Studying Economics Inhibit Cooperation?, 13, 14 fig.6 (1992) (unpublished manuscript, on file with the Harvard Law School Library) (finding that 39.6% of economics majors and 61.2% of non-economics majors cooperated in a prisoner's dilemma); Gerald Harwell & Ruth E. Ames, Experiments on the Provision of Public Goods. I. Resources, Interest, Group Size, and the Free-Rider Problem, 84 AM. J. Soc. 1335, 1349-50 (1979) (finding that the typical subject invested 57% of her available resources in a public good, two-thirds of subjects invested more than half, and 22% invested all); Orbell, van de Kragt & Dawes, supra note 23, at 813, 816-17 (finding that with discussion, 79% of subjects cooperated, while with no discussion and other factors, 20% cooperated); Orbell, Schwartz-Shea & Simmons, supra note 23, at 153 (finding that allowing exit and discussion led over 75% to cooperate and that allowing exit and no discussion led 18% of those who chose to stay to cooperate).
49. One might argue that, even in "one-shot" dilemma experiments, subjects fear being "punished" outside the experiment should they ever encounter the other players. Yet subject B will not seek to punish subject A unless doing so would create some benefit for B. One might argue that B will be less likely to cooperate with A in any future dilemma setting. However, when they meet in a future real-world dilemma, A and B will have no reason to cooperate unless they expect to interact with sufficient frequency to permit reciprocity. But given sufficient reciprocity, there is no reason for B to mistrust A. One may be tempted to say that A has shown herself to be a "defector" by nature, but A merely defected in the laboratory when it was rational to do so. A new game is now afoot in which the tit-for-tat strategy is a good bet for both A and B. Subjects should therefore not fear defecting in the one-shot experiment. That so many do not defect remains a puzzle.
50. See sources cited supra note 19.
51. I am not claiming that altruism and a concern for fairness do not exist, but only that neither fully explains the residual cooperation in these experiments.
52. One might explain the data as the product of "group altruism." This term, however, obscures the fact that the individuals not only feel more "altruistic" toward members of their groups than toward non-members, but also seem willing to sacrifice the welfare of non-members for the sake of their group. See supra notes 35-37 and accompanying text. And there is no reason to think that it is less "fair" for teams than for individuals to cooperate in a prisoner's dilemma.
53. See Gary Bornstein, Amnon Rapoport, Lucia Kerpel & Tani Katz, Within- and Between-Group Communication in Intergroup Competition for Public Goods, 25 J. EXPERIMENTAL SOC. PSYCHOL. 422, 424 (1989); Orbell, Schwartz-Shea & Simmons, supra note 23, at 154-55 (reporting that, when subjects had the third option of exit, discussion not only increased cooperation but decreased exit); Orbell, van de Kragt & Dawes, supra note 23, at 811.
54. See, e.g., FRANK, supra note 19, at 142 (reporting that 37% of subjects cooperated when permitted only 10 minutes of discussion and forbidden to make promises to cooperate, 61% cooperated when permitted 30 minutes of discussion but forbidden to promise, and 68% cooperated when permitted 30 minutes of discussion without restrictions); Robyn M. Dawes, Jeanne McTavish & Harriet Shaklee, Behavior, Communication, and Assumptions About Other People's Behavior in a Commons Dilemma Situation, 35 J. PERSONALITY & SOC. PSYCHOL. 1, 5 (1977) (finding that defection rates in a prisoner's dilemma game declined from 73% with no communication to 65% with communication not regarding the game, to 26% with communication regarding the game but no promises, and to 16% with communication that included promises).
55. Orbell, van de Kragt & Dawes, supra note 23, at 811.
56. Communication should, in theory, be relevant only if it permits someone to make a credible threat or promise. There is every reason for people to try to fool others into believing their promises to cooperate or retaliate, but in a one-shot game between strangers, there is no self-interested reason for people to follow through on those promises or threats.
57. See Dawes, van de Kragt & Orbell, supra note 22, at 103-07.
58. See id. at 103-04.
59. See id. at 106 (from 33% to 31%).
60. Id.
61. Dawes and his colleagues have suggested promise-making as an alternative explanation for the effect of discussion (subjects can promise only if discussion is permitted). See Orbell, van de Kragt & Dawes, supra note 23, at 818. But more recently they found greater support in the transcripts of the discussions for the group identity explanation. See Dawes, van de Kragt & Orbell, supra note 22, at 109. Indeed, other studies on the effects of discussion on cooperation have found that most of the increase in cooperation can be obtained even if promising is not allowed. See sources cited supra note 41; see also Orbell, Schwartz-Shea & Simmons, supra note 23, at 158-59 (summarizing the group-identity explanations subjects gave for their decision not to exit a dilemma game).
62. Dawes, van de Kragt & Orbell, supra note 22, at 109.
63. Based on 10 years of research, including "27 different no-discussion conditions with 1,188 subjects in 178 groups, and for 12 different discussion conditions with 637 subjects in 95 groups," Dawes and his colleagues conclude that "with no discussion, egoistic motives explain cooperation; with discussion, group identity - alone or in interaction with verbal promises - explains its dramatic increase." Id.
64. See sources cited infra note 120.
65. See DENNIS CHONG, COLLECTIVE ACTION AND THE CIVIL RIGHTS MOVEMENT 1-12 (1991). After reading Chong's explanation of how the movement solved its collective action problems, I came to believe that the prevailing economic criticisms of Title VII are based on assumptions that would make an explanation of the civil rights movement impossible; if selfish whites cannot coordinate discrimination in the face of market competition, neither should blacks have been able to cooperate to succeed in a civil rights movement.
66. See id. at 32-33.
67. See id. at 34-35 ("The desires to gain or sustain friendships, to maintain one's social standing, and to avoid ridicule and ostracism are all social goals that constitute selective incentives for individuals to participate in collective action."). Chong explains how this non-material concern for reputation served as a motivator in civil rights activities like the freedom rides, the Montgomery bus boycott, the Selma voting drive, and the Greensboro lunch counter sit-in. See id. at 55-65. "We refuse one-time gains through free riding in order to retain the esteem, respect, and continued goodwill of those we care about." Id. at 55.
68. See id. at 35-36. Although social pressure is most effective in small, tightly knit groups, "[l]arge groups nevertheless can take advantage of the selective incentives provided by social pressures if they in turn comprise a network of much smaller organizational units." Id. at 35.
69. Marshall claims: During combat the soldier may become so gripped by fear that most of his thought is directed toward escape. But if he is serving among men whom he has known for a long period or whose judgment of him counts for any reason, he still will strive to hide his terror from them. . . . The majority are unwilling to take extraordinary risks and do not aspire to a hero's role, but they are equally unwilling that they should be considered the least worthy among those present. MARSHALL, supra note 4, at 148-49; see JANOWITZ, supra note 4, at 220.
70. 58 Id. at 151-52.
71. Richard H. McAdams, Relative Preferences, 102 YALE L.J. 1, 9 (1992) (emphasis omitted).
72. See id. at 38-48.
73. BECKER, supra note 3, at 145 (arguing that economics should strive to explain behavior with only a few preferences that reflect basic pleasures).
74. It would be trivial to say that people care about what others think of them as a means to furthering various material ends; the evaluations others make obviously affect the resources one can gain in trade. My claim, however, is that people also care quite a bit about what others think of them as an end in itself. Esteem is objectively manifested in the ways people show each other respect: for example, who is permitted to speak first or without interruption; who is asked for advice and counsel; who sets the standard of fashion or humor. These objective prerogatives are valued, in my model, only as manifestations of the esteem individuals seek and are not separate ends in themselves.
75. See McAdams, supra note 59, at 38-44. The desire for esteem also explains why people care how their salary compares to that of their co-workers and why people save according to the size of their income relative to others. See id. at 45-46.
76. Consider, for example, being the only member of an audience to begin applauding at an inopportune moment. Embarrassment does not depend on the prospect of pecuniary consequences from one's blunder. We feel discomfort because - caring generally for esteem - we suffer when others think ill of us, even if they are strangers. Of course, people may internalize norms and feel embarrassment (or guilt) even when others are not present. But my point is that there is often an increase in the intensity of feeling if others know of our offending conduct.
77. See Bibb Latané, The Psychology of Social Impact, in CONTEMPORARY READINGS IN SOCIAL PSYCHOLOGY 288, 291 (David A. Scroeder, David E. Johnson & Thomas D. Jensen eds., 1985) ("Eighty years of experimental evidence strongly shows that individuals are influenced by the actions and expectations of others.").
78. Esteem concerns primarily reduce the benefits of defecting when the other party cooperates, possibly to the extent that a player is better off cooperating when the other party cooperates. Even with esteem, a player is still better off defecting when the other party defects, but defection is no longer the "dominant" strategy. Moreover, because mutual cooperation is better for both parties than mutual defection, communication may increase cooperation by assuring each that the other will cooperate.
79. The statistical theory of discrimination, for example, posits that people use race as a proxy to evaluate others by assuming in the absence of more specific information that the person has the average qualities of her race. The individual will benefit from the use of proxies if the savings in information costs outweigh the costs of inaccuracy. See infra note 116.
80. Once the group boundary arises, subjects within one group may also anticipate evaluating, and being evaluated by, subjects in the other group.
81. See supra notes 32-33 and accompanying text.
82. Similarly, in the commons depletion experiments, subjects cooperated more if they believed the experimenter was evaluating all of them as a single group rather than as part of one of two subgroups. See supra note 37 and accompanying text. In the former case, the subjects apparently viewed depleting the pool as a form of group failure, the cost of which is a lower evaluation of the group and all its members. In the latter case, gaining the esteem of being on a successful team arguably required depleting the pool at a higher rate so that one's team of subjects would reap more than the other team. The "minimal inter-group paradigm" studies do not present a dilemma, but note that subjects rated others so as to maximize the difference in outcome between the groups rather than to maximize the absolute outcome for their own group. See supra notes 34-36 and accompanying text. The individuals behaved as if they sought to gain individual esteem by maximizing the proxy value of membership in their group.
83. See Robert Folger, Reformulating the Preconditions of Resentment: A Referent Cognitions Model, in SOCIAL COMPARISON, SOCIAL JUSTICE, AND RELATIVE DEPRIVATION 183, 184-212 (John C. Masters & William P. Smith eds., 1987); James M. Olson & J. Douglas Hazlewood, Relative Deprivation and Social Comparison: An Integrative Comparison, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON 1, 2-5 (James W. Olson, C. Peter Herman & Mark P. Zanna eds., 1986).
84. See Robert Folger, Reformulating the Preconditions of Resentment: A Referent Cognitions Model, in SOCIAL COMPARISON, SOCIAL JUSTICE, AND RELATIVE DEPRIVATION 183, 184-212 (John C. Masters & William P. Smith eds., 1987); James M. Olson & J. Douglas Hazlewood, Relative Deprivation and Social Comparison: An Integrative Comparison, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON 1, 2-5 (James W. Olson, C. Peter Herman & Mark P. Zanna eds., 1986).
85. McAdams, supra note 59, at 34; see also Faye Crosby, A Model of Egoistical Relative Deprivation, 83 PSYCHOL. REV. 85, 90 (1976) (maintaining that the first precondition of relative deprivation is that "[t]he person who lacks X must . . . see that someone else . . . possesses X").
86. See Thomas D. Cook, Faye Crosby & Karen M. Hennigan, The Construct Validity of Relative Deprivation, in SOCIAL COMPARISON PROCESSES 307, 320-23 tbl. 4 (Jerry M. Suls & Richard L. Miller eds., 1977) (collecting studies).
87. See Lise Dubé & Serge Guimond, Relative Deprivation and Social Protest: The Personal-Group Issue, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON, supra note 71, at 201, 204-10. One researcher summarizes: [F]raternalistic [relative deprivation] is useful for predicting white racism and support of black political candidates in the United States, militancy among black Americans and their approval of violence to gain their rights, attitudes of Québec nationalism among Francophone Canadians in Montreal, subjective satisfactions among members of Toronto's Chinese community, and outgroup attitudes among Chinese in Toronto and Muslims in India, respectively. Kenneth L. Dion, Responses to Perceived Discrimination and Relative Deprivation, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON, supra note 71, at 159, 174.
88. See Lise Dubé & Serge Guimond, Relative Deprivation and Social Protest: The Personal-Group Issue, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON, supra note 71, at 201, 204-10. One researcher summarizes: [F]raternalistic [relative deprivation] is useful for predicting white racism and support of black political candidates in the United States, militancy among black Americans and their approval of violence to gain their rights, attitudes of Québec nationalism among Francophone Canadians in Montreal, subjective satisfactions among members of Toronto's Chinese community, and outgroup attitudes among Chinese in Toronto and Muslims in India, respectively. Kenneth L. Dion, Responses to Perceived Discrimination and Relative Deprivation, in 4 RELATIVE DEPRIVATION AND SOCIAL COMPARISON, supra note 71, at 159, 174.
89. Although there are exceptions, individuals tend to value more highly the positive traits that they believe they possess, See, e.g., George R. Goethals, David M. Messick & Scott T. Allison, The Uniqueness Bias: Studies of Constructive Social Comparison, in SOCIAL COMPARISON 149, 149-50 (Jerry Suls & Thomas A. Wills eds., 1991);
90. Pawel Lewicki, Self-image Bias in Person Perception, 45 J. PERSONALITY & SOC. PSYCHOL. 384, 385-87 (1983).
91. Even if an individual admires only a few leaders or "heroes" in the group, she will care about the opinion of other group members, who (especially if they know her better than the heroes do) can dramatically influence the leaders' esteem for her. Thus, a few highly admired people within a group can cause everyone in the group-to care much more about the esteem of group members than they do about that of non-group members. See CHONG, supra note 52, at 59-60; ROBERT C. ELLICKSON, ORDER WITHOUT LAW: How NEIGHBORS SETTLE DISPUTES 237-38 (1991).
92. See supra notes 51-58 and accompanying text.
93. There may be costs to thinking well of others, but probably only when one grants very high levels of esteem. First, if one has a relative preference for obtaining a particular rank (for example, being a top violinist), esteeming many others may mean conceding that one has not achieved the rank one desires. See generally McAdams, supra note 59, at 7-26 (describing analyses of relative preferences in which one derives pleasure from another's consumption level in relation to one's own). For most individuals, however, it is still costless to believe that a limited number of others are exemplary in some way, either because one can still believe oneself superior to them or because one's relative preference is merely to be in the top tier. Similarly, there may be a cost in deciding how much esteem others merit. But this "decisional cost" itself may be zero it is merely the by-product of the ego-enhancing process of valuing one's own qualities and of the information picked up about others in the course of daily life. (Of course, because this decisional cost is not always zero, people use "proxies" to allocate esteem.) Finally, a cost may be incurred in expressing esteem. (In some circumstances, esteem is inferred without expression, but only because esteem is sometimes expressed under those circumstances.) But I would make the same argument for expression that I do for esteem: people have a preference for expressing their beliefs, and thus the smile, gesture, or phrase that expresses esteem, within limits, is simply not a cost. Some commentators explain voting in this manner. See infra note 217.
94. A group may also formalize the awarding of intra-group status by giving awards, special titles, or privileges. The abundance of such external tokens of esteem testifies to their success.
95. See supra note 75.
96. See McAdams, supra note 59, at 79-80.
97. See MICHAEL HECHTER, PRINCIPLES OF GROUP SOLIDARITY 40-58 (1987); WILLIAM J. GOODE, THE CELEBRATION OF HEROES 204-34 (1978).
98. See MICHAEL HECHTER, PRINCIPLES OF GROUP SOLIDARITY 40-58 (1987); WILLIAM J. GOODE, THE CELEBRATION OF HEROES 204-34 (1978).
99. See ELLICKSON, supra note 76. Other notable exceptions are GEORGE A. AKERLOF, AN ECONOMIC THEORIST'S BOOK OF TALES 95-97 (1984); and Robert D. Cooter, Structural Adjudication and the New Law Merchant: A Model of Decentralized Law, 14 INT'L REV. L. & ECON. 215, 224-27 (1994).
100. See ELLICKSON, supra note 76. Other notable exceptions are GEORGE A. AKERLOF, AN ECONOMIC THEORIST'S BOOK OF TALES 95-97 (1984); and Robert D. Cooter, Structural Adjudication and the New Law Merchant: A Model of Decentralized Law, 14 INT'L REV. L. & ECON. 215, 224-27 (1994).
101. 85 Ellickson studied disputes over damages from cattle trespass, allocation of the costs of boundary fences, and damages from automobile accidents caused by cattle wandering onto highways. He found that the first two classes of disputes (but not the third) are resolved by social norms that arise independently of and differ from legal rules. See id. at 40-103.
102. See id. at 167. Ellickson also reviewed studies that demonstrate the effectiveness of norms both in the whaling industry in the 18th and 19th centuries and among orchard owners in Washington state. See id. at 189, 191-206. He concludes that "members of a close-knit group develop and maintain norms whose content serves to maximize the aggregate welfare that members obtain in their workaday affairs with one another." Id. at 167 (emphasis omitted).
103. The enforcement mechanism for reciprocity is personal self-help. For example, when his neighbor fails to cooperate in controlling cattle trespass or contributing to boundary fences, a rancher may withhold some favor he usually grants, detain trespassing cattle, or engage in mild physical retaliation. See ELLICKSON, supra note 76, at 57-59, 79-81. Ellickson labels the strategy these neighbors employed "even-up." It is a more complex form of reciprocity than tit-for-tat, applicable to a wider variety of circumstances. See id. at 225-29.
104. I am relying on Ellickson's definition of norms as a means of third-party, not reciprocal, control. Robert Cooter, on the other hand, implies that norms add to reciprocity only because individuals "internalize" certain obligations of behavior. See Cooter, supra note 83, at 221. Like Ellickson, I categorize internalizalion of norms as "first-party" or "self" control and conceive social norms as an independent form of "third-party" control. See ELLICKSON, supra note 76, at 126 n.8, 243-46. The difficult question is why this third-party enforcement exists. My claim is that such enforcement succeeds without internalizalion (or reciprocily) as long as individuals seek the esteem of others. Thus, while Cooter emphasizes the motivation of guilt - a negative feeling that exists even when others do not know of one's transgression, I emphasize shame - a negative feeling that exists only when others believe one has transgressed. See generally JEROME KAGAN, THE NATURE OF THE CHILD 145-49 (1984) (discussing the development of distinct emotions of guilt and shame).
105. ELLICKSON, supra note 76, at 131 n.21.
106. See id. at 131, 213-19. In Ellickson's taxonomy, social norms are enforced by "vicarious self-help." Id. at 131.
107. See ELSTER, supra note 27, at 132 n.113. The individual reasons that either enough others will enforce the norm to ensure its effectiveness without her contribution or not enough others will enforce the norm to permit it to work even with her contribution. In either case, there is no material reason for her to enforce the norm.
108. Id. at 133.
109. Id.
110. Ellickson responds that "[p]erhaps the perception, or better yet the existence, of a few virtuous leaders or other committed third-party enforcers at the highest level of social control creates incentives for cooperative activity that cascade down and ultimately produce welfare-maximizing primary behavior." ELLICKSON, supra note 76, at 237. Given his assumption, Ellickson is right. If most individuals perceive that certain members of the community are willing to inflict punishment on norm-violators notwithstanding the material irrationality of doing so, this core group can terminate the otherwise infinite regress and provide a foundation or "backstop" for the whole enforcement mechanism. Yet there is no reason for people to perceive the existence of such a group unless it actually exists. And Ellickson offers no explanation of why this core group of people ignore their own self-interest. Of course Ellickson may believe that such people are in fact pursuing a non-material self-interest in the form of status, but he does not make such an argument explicitly. See id. at 238 n.23 (noting "the elevated status of leaders and their special role in the enforcement of norms"). The desire for esteem is a more general and reliable backstop for norm enforcement than material self-interest.
111. Gossip transmits information on which people make reflexive esteem judgments. Elster has responded that expressing disapproval "requires energy and attention that might be used for other purposes" and that "[o]ne may alienate or provoke the target individual, at some cost or risk to oneself." ELSTER, supra note 27, at 133. But others need not always express disapproval; sometimes one is confident that by violating a. social norm, she has lowered herself in the eyes of others. Further, because people enjoy expression as an end in itself, expressing disapproval is not always costly. One may alienate the target individual by disapproving of her - but that is the whole point. Rather than bear the cost of retaliating against those who express disapproval of one's norm violation, the rational actor often just obeys the norm in the first place.
112. See Jonathan R. Macey, Promoting Public-Regarding Legislation Through Statutory Interpretation: An Interest Group Model, 86 COLUM. L. REV. 223, 224 n.6 (1986).
113. Absent this effect, free-riding may be so great that the group never forms at all. Thus another reason intra-group status production causes more group conflict is that it ensures there will be more groups.
114. See Richard A. Posner, The Material Basis of Jurisprudence, 69 IND. L.J. 1, 19 (1993).
115. Posner attempts to explain how lawyers cooperate in rent-seeking. See id. His most plausible claim for why individual lawyers bear the cost of organizing to produce cartel-protective legislation is that they believe such activities will make them better known, "thereby increasing the likelihood that [others] will refer cases to [them]." Id. at 20. But among the other lawyers a given lawyer knows, there are more compelling reasons for making referrals, such as the expectation of material reciprocity. Status, however, provides an explanation: when the material costs are low, lawyers "pay" organizers with status; and because organizers enjoy high status, lawyers seek to associate with them by giving them referrals.
116. By raising their group's status, individuals produce status for themselves in two ways. First, they share in the enhanced status their group enjoys. Second, for the reasons discussed above, they are rewarded with intra-group status for contributing to their group's welfare.
117. See McAdams, supra note 59, at 49-55.
118. For example, even if groups A and B seek superiority over each other, there is no inherent preference conflict if group A members wish to be superior in characteristic X and group B members wish to be superior in characteristic K. The student bodies of two rival universities can both prosper, for example, if the students at the school that excels in football value only the sports comparison, while the students at the school that excels in placing students in graduate programs care only about the academic comparison.
119. There are certain characteristics that are widely valued in American society, such as wealth, education, and occupational prestige. See ROGER S. MASON, CONSPICUOUS CONSUMPTION 108, 113 (1981); Marcus Felson, Invidious Distinctions Among Cars, Clothes and Suburbs, 42 PUB. OPINION Q. 49, 56-57 (1978). One can describe an individual's general social status by roughly summarizing these widely valued characteristics. See MIHALY CSIKSZENTMIHALYI & EUGENE ROCHBERG-HALTON, THE MEANING OF THINGS: DOMESTIC SYMBOLS AND THE SELF 31 (1981) ("[Status symbols] are in some ways a summary of all the salient characteristics of the self, a global measure of the owner's standing in the community."). Even if some people are indifferent to their relative wealth, education, or occupational prestige, they tend to be esteem-seekers and care about the social status these items earn for them.
120. There are certain characteristics that are widely valued in American society, such as wealth, education, and occupational prestige. See ROGER S. MASON, CONSPICUOUS CONSUMPTION 108, 113 (1981); Marcus Felson, Invidious Distinctions Among Cars, Clothes and Suburbs, 42 PUB. OPINION Q. 49, 56-57 (1978). One can describe an individual's general social status by roughly summarizing these widely valued characteristics. See MIHALY CSIKSZENTMIHALYI & EUGENE ROCHBERG-HALTON, THE MEANING OF THINGS: DOMESTIC SYMBOLS AND THE SELF 31 (1981) ("[Status symbols] are in some ways a summary of all the salient characteristics of the self, a global measure of the owner's standing in the community."). Even if some people are indifferent to their relative wealth, education, or occupational prestige, they tend to be esteem-seekers and care about the social status these items earn for them.
121. There are certain characteristics that are widely valued in American society, such as wealth, education, and occupational prestige. See ROGER S. MASON, CONSPICUOUS CONSUMPTION 108, 113 (1981); Marcus Felson, Invidious Distinctions Among Cars, Clothes and Suburbs, 42 PUB. OPINION Q. 49, 56-57 (1978). One can describe an individual's general social status by roughly summarizing these widely valued characteristics. See MIHALY CSIKSZENTMIHALYI & EUGENE ROCHBERG-HALTON, THE MEANING OF THINGS: DOMESTIC SYMBOLS AND THE SELF 31 (1981) ("[Status symbols] are in some ways a summary of all the salient characteristics of the self, a global measure of the owner's standing in the community."). Even if some people are indifferent to their relative wealth, education, or occupational prestige, they tend to be esteem-seekers and care about the social status these items earn for them.
122. Conflict exists in a variety of other circumstances as well. See McAdams, supra note 59, at 52-54. For example, conflict arises if one group seeks superiority and the other seeks equality.
123. See id. at 55-59. An exception is when the competition encourages investment in a public good. See id. at 59-62.
124. Cf. Steven C. Salop & David T. Scheffman, Raising Rivals' Costs, AM. ECON. REV., May 1983, at 267 (discussing strategies intended to increase costs for rival firms).
125. See supra notes 32-37, 44-49 and accompanying text.
126. ELLICKSON, supra note 76, at 117.
127. See id. at 117-18, The ranchers believe the open range provides greater immunity from liability than it actually does. Ellickson says this misunderstanding leads to inaccurate decisions regarding care and insurance. See id. I argue, however, that the belief persists because it supports the group's production of status. See infra pp. 1058-60.
128. Even when one seeks status through positive accomplishments, admitting to a status motivation undermines one's status. See JON ELSTER, SOUR-GRAPES 66-69 (1983); McAdams, supra note 59, at 41-44. The esteem of others, Elster argues, is something one cannot bring about directly, but only as the "by-product" of some other activity. See ELSTER, supra, at 66 ("[N]othing is so unimpressive as behaviour designed to impress."). People still seek status, but they require a plausible non-status explanation for their conduct.
129. Status production commonly involves such rationalizations. I explain the phenomenon more fully in the context of race discrimination. See infra Part II.B.2.C.
130. See Ken Foskett, Retreat on Issue a Sore Point for Governor, ATLANTA CONST., Jan. 29, 1994, at F1.
131. Even when individuals articulate a moral motive for their political positions, status competition may better explain the legal change sought or opposed. See, e.g., JOSEPH R. GUSFIELD, SYMBOLIC CRUSADE: STATUS POLITICS AND THE AMERICAN TEMPERANCE MOVEMENT 12-35, 44-50, 188 (1963). My argument here thus falls in line with recent work on the psychology of procedural justice. Independent of outcomes, individuals care greatly about the procedures used in adjudicating disputes or allocating resources. According to psychologists, the explanation is that individuals care about the esteem they receive within their groups and adjudicatory proce-dures reveal that esteem. See E. ALLAN LIND & TOM R. TYLER, THE SOCIAL PSYCHOLOGY OF PROCEDURAL JUSTICE 230-37 (1988); TOM R. TYLER, WHY PEOPLE OBEY THE LAW 173-76 (1990). My point is to extend this insight from procedural to substantive rules: independent of the substantive effects of legal rules, individuals care about the symbolic effects those rules have on their social status.
132. Even when individuals articulate a moral motive for their political positions, status competition may better explain the legal change sought or opposed. See, e.g., JOSEPH R. GUSFIELD, SYMBOLIC CRUSADE: STATUS POLITICS AND THE AMERICAN TEMPERANCE MOVEMENT 12-35, 44-50, 188 (1963). My argument here thus falls in line with recent work on the psychology of procedural justice. Independent of outcomes, individuals care greatly about the procedures used in adjudicating disputes or allocating resources. According to psychologists, the explanation is that individuals care about the esteem they receive within their groups and adjudicatory proce-dures reveal that esteem. See E. ALLAN LIND & TOM R. TYLER, THE SOCIAL PSYCHOLOGY OF PROCEDURAL JUSTICE 230-37 (1988); TOM R. TYLER, WHY PEOPLE OBEY THE LAW 173-76 (1990). My point is to extend this insight from procedural to substantive rules: independent of the substantive effects of legal rules, individuals care about the symbolic effects those rules have on their social status.
133. Even when individuals articulate a moral motive for their political positions, status competition may better explain the legal change sought or opposed. See, e.g., JOSEPH R. GUSFIELD, SYMBOLIC CRUSADE: STATUS POLITICS AND THE AMERICAN TEMPERANCE MOVEMENT 12-35, 44-50, 188 (1963). My argument here thus falls in line with recent work on the psychology of procedural justice. Independent of outcomes, individuals care greatly about the procedures used in adjudicating disputes or allocating resources. According to psychologists, the explanation is that individuals care about the esteem they receive within their groups and adjudicatory proce-dures reveal that esteem. See E. ALLAN LIND & TOM R. TYLER, THE SOCIAL PSYCHOLOGY OF PROCEDURAL JUSTICE 230-37 (1988); TOM R. TYLER, WHY PEOPLE OBEY THE LAW 173-76 (1990). My point is to extend this insight from procedural to substantive rules: independent of the substantive effects of legal rules, individuals care about the symbolic effects those rules have on their social status.
134. See GARY BECKER, THE ECONOMICS OF DISCRIMINATION 14 (2d ed. 1971) ("If an individual has a 'taste for discrimination,' he must act as if he were willing to pay something, either directly or in the form of a reduced income, to be associated with some persons instead of others."), see also id. at 15 ("Discrimination is commonly associated with disutility caused by contact with some individuals . . . .").
135. Becker uses a "discrimination coefficient" to describe the degree to which individual market actors discriminate: The money costs of a transaction do not always completely measure net costs, and a [discrimination coefficient] acts as a bridge between money and net costs. Suppose an employer were faced with the money wage rate π a particular factor; he is assumed to act as if π(I + di) were the net wage rate, with di as his [discrimination coefficient] against this factor. Id. at 14. Becker provides a similar description of co-worker and customer discrimination, see id. at 14-15, and uses a "market discrimination coefficient" to measure differences in market wage rates or other prices due to discrimination, id. at 17-18.
136. The model of "statistical discrimination" claims that race is only a proxy for other qualities people care about. See RICHARD A. EPSTEIN, FORBIDDEN GROUNDS: THE CASE AGAINST EMPLOYMENT DISCRIMINATION LAWS 33-35, 40-41 (1992); Edmund S. Phelps, The Statistical Theory of Racism and Sexism, 62 AM. ECON. REV. 659, 659 (1972). Without more specific and costly information about individuals, discriminators assume that each member of a category has the average qualities of others in that category. Thus, white employers who discriminate are merely drawing rational statistical inferences about individual blacks given the cost of acquiring more specific information. Information costs do not, however, explain the racial animus that Becker's theory addresses; for example, they do not explain white supremacists' violence against blacks. Further, the idea of race as a proxy is inconsistent with Jim Crow definitions of race according to factors that were not at all observable. See infra p. 1042. My claim is not that statistical discrimination does not exist but that we require a different theory for the paradigmatic case of racial animus.
137. The model of "statistical discrimination" claims that race is only a proxy for other qualities people care about. See RICHARD A. EPSTEIN, FORBIDDEN GROUNDS: THE CASE AGAINST EMPLOYMENT DISCRIMINATION LAWS 33-35, 40-41 (1992); Edmund S. Phelps, The Statistical Theory of Racism and Sexism, 62 AM. ECON. REV. 659, 659 (1972). Without more specific and costly information about individuals, discriminators assume that each member of a category has the average qualities of others in that category. Thus, white employers who discriminate are merely drawing rational statistical inferences about individual blacks given the cost of acquiring more specific information. Information costs do not, however, explain the racial animus that Becker's theory addresses; for example, they do not explain white supremacists' violence against blacks. Further, the idea of race as a proxy is inconsistent with Jim Crow definitions of race according to factors that were not at all observable. See infra p. 1042. My claim is not that statistical discrimination does not exist but that we require a different theory for the paradigmatic case of racial animus.
138. "[B]y increasing . . . contact . . . trade imposes nonpecuniary, but real, costs on those members of either race who dislike association with members of the other race." POSNER, supra note 3, at 651. Similarly, Epstein's characterization of discrimination as a "refusal to deal" reflects a model in which discriminators seek to avoid association with members of other races. EPSTEIN, supra note 116, at 29-30; see also Harold Demsetz, Minorities in the Market Place, 43 N.C. L. REV. 271, 271 (1965) (equating "discrimination against" with an "aversion to association" with). An important but under-recognized exception (to which I am indebted) is George A. Akerlof, Discriminatory, Status-Based Wages Among Tradition-Oriented, Stochastically Trading Coconut Producers, 93 J. POL. ECON. 265, 265 (1985). See infra pp. 1064-69.
139. "[B]y increasing . . . contact . . . trade imposes nonpecuniary, but real, costs on those members of either race who dislike association with members of the other race." POSNER, supra note 3, at 651. Similarly, Epstein's characterization of discrimination as a "refusal to deal" reflects a model in which discriminators seek to avoid association with members of other races. EPSTEIN, supra note 116, at 29-30; see also Harold Demsetz, Minorities in the Market Place, 43 N.C. L. REV. 271, 271 (1965) (equating "discrimination against" with an "aversion to association" with). An important but under-recognized exception (to which I am indebted) is George A. Akerlof, Discriminatory, Status-Based Wages Among Tradition-Oriented, Stochastically Trading Coconut Producers, 93 J. POL. ECON. 265, 265 (1985). See infra pp. 1064-69.
140. See BECKER, supra note 114, at 35-37; see also Jennifer Roback, Southern Labor Law in the Jim Crow Era: Exploitative or Competitive?, 51 U. CHI. L. REV. 1161, 1163 (1984) ("Becker's argument that discrimination is unprofitable to the discriminators and should eventually disappear from a competitive market setting is widely accepted." (footnote omitted)). One might charge that Becker's model is proved wrong by the continued existence of discrimination. Epstein responds that little discrimination remains and that what does remain is likely to be desirable, based on efficiency considerations omitted in Becker's analysis. See EPSTEIN, supra note 116, at 47-78, 91.
141. See EPSTEIN, supra note 116, at 9, 12; Demsetz, supra note 117, at 271, 275-93; Richard A. Posner, An Economic Analysis of Sex Discrimination Laws, 56 U. CHI. L. REV. 1311, 1312 (1989) (speculating that sex discrimination law has diminished women's welfare); Richard A. Posner, The Efficiency and the Efficacy of Title VII, 136 U. PA. L. REV. 513, 514 (1987) [hereinafter Posner, Title VII].
142. See EPSTEIN, supra note 116, at 9, 12; Demsetz, supra note 117, at 271, 275-93; Richard A. Posner, An Economic Analysis of Sex Discrimination Laws, 56 U. CHI. L. REV. 1311, 1312 (1989) (speculating that sex discrimination law has diminished women's welfare); Richard A. Posner, The Efficiency and the Efficacy of Title VII, 136 U. PA. L. REV. 513, 514 (1987) [hereinafter Posner, Title VII].
143. Thus, some theorists have argued that, even under Becker's model, competitive markets can perpetuate discrimination, see, e.g., Cass R. Sunstein, Why Markets Don't Stop Discrimination, 8 Soc. PHIL. & POL'Y 22, 22 (1991), and that the laws against discrimination are efficient, see, e.g., John J. Donohue III, Is Title VII Efficient?, 134 U. PA. L. REV. 1411, 1412 (1986) [hereinafter Donohue, Title VII]; John J. Donohue III, Prohibiting Sex Discrimination in the Workplace: An Economic Perspective, 56 U. CHI. L. REV. 1337, 1347-48, 1366-67 (1989) [hereinafter Donohue, Prohibiting Sex Discrimination].
144. Thus, some theorists have argued that, even under Becker's model, competitive markets can perpetuate discrimination, see, e.g., Cass R. Sunstein, Why Markets Don't Stop Discrimination, 8 Soc. PHIL. & POL'Y 22, 22 (1991), and that the laws against discrimination are efficient, see, e.g., John J. Donohue III, Is Title VII Efficient?, 134 U. PA. L. REV. 1411, 1412 (1986) [hereinafter Donohue, Title VII]; John J. Donohue III, Prohibiting Sex Discrimination in the Workplace: An Economic Perspective, 56 U. CHI. L. REV. 1337, 1347-48, 1366-67 (1989) [hereinafter Donohue, Prohibiting Sex Discrimination].
145. Thus, some theorists have argued that, even under Becker's model, competitive markets can perpetuate discrimination, see, e.g., Cass R. Sunstein, Why Markets Don't Stop Discrimination, 8 Soc. PHIL. & POL'Y 22, 22 (1991), and that the laws against discrimination are efficient, see, e.g., John J. Donohue III, Is Title VII Efficient?, 134 U. PA. L. REV. 1411, 1412 (1986) [hereinafter Donohue, Title VII]; John J. Donohue III, Prohibiting Sex Discrimination in the Workplace: An Economic Perspective, 56 U. CHI. L. REV. 1337, 1347-48, 1366-67 (1989) [hereinafter Donohue, Prohibiting Sex Discrimination].
146. See, e.g., Mary E. Becker, Needed in the Nineties: Improved Individual and Structural Remedies for Racial and Sexual Disadvantages in Employment, 79 GEO. L.J. 1659, 1663 (1991); Richard Delgado, Rodrigo's Second Chronicle: The Economics and Politics of Race, 91 MICH. L. REV. 1183, 1188-93, 1195-97 (1993).
147. See, e.g., Mary E. Becker, Needed in the Nineties: Improved Individual and Structural Remedies for Racial and Sexual Disadvantages in Employment, 79 GEO. L.J. 1659, 1663 (1991); Richard Delgado, Rodrigo's Second Chronicle: The Economics and Politics of Race, 91 MICH. L. REV. 1183, 1188-93, 1195-97 (1993).
148. A partial exception is John J. Donohue III & James J. Heckman, Re-evaluating Federal Civil Rights Policy, 79 GEO. L.J. 1713, 1728-29 (1991). The authors briefly propose an alternative to Becker's model in which race discrimination is supported by coercive social norms and universal belief in the inferiority of blacks. See also J. Hoult Verkerke, Free to Search, 105 HARV. L. REV. 2080, 2090-93 (1992) (discussing the economic importance of coercive social norms and reviewing RICHARD A. EPSTEIN, FORBIDDEN GROUNDS: THE CASE AGAINST EMPLOYMENT DISCRIMINATION LAWS (1992)). Neither work, however, explains the intra-group cooperation necessary for discriminatory norms to function or the existence of racist beliefs. Gary Becker, for example, makes the first point implicitly by defining discrimination solely as a means of avoiding a psychic "cost" one suffers from association. See BECKER, supra note 114, at 14-15, 153. The harm of discrimination is simply the loss of the benefit economic trades would have created, a loss shared by both discriminators and victims. See id. at 29-30; POSNER, supra note 3, at 651. Epstein exemplifies the second point. In Forbidden Grounds - a 500-page, purportedly exhaustive attack on the laws that prohibit employment discrimination - there is no mention of the rich literature on the psychological harms of racism.
149. A partial exception is John J. Donohue III & James J. Heckman, Re-evaluating Federal Civil Rights Policy, 79 GEO. L.J. 1713, 1728-29 (1991). The authors briefly propose an alternative to Becker's model in which race discrimination is supported by coercive social norms and universal belief in the inferiority of blacks. See also J. Hoult Verkerke, Free to Search, 105 HARV. L. REV. 2080, 2090-93 (1992) (discussing the economic importance of coercive social norms and reviewing RICHARD A. EPSTEIN, FORBIDDEN GROUNDS: THE CASE AGAINST EMPLOYMENT DISCRIMINATION LAWS (1992)). Neither work, however, explains the intra-group cooperation necessary for discriminatory norms to function or the existence of racist beliefs. Gary Becker, for example, makes the first point implicitly by defining discrimination solely as a means of avoiding a psychic "cost" one suffers from association. See BECKER, supra note 114, at 14-15, 153. The harm of discrimination is simply the loss of the benefit economic trades would have created, a loss shared by both discriminators and victims. See id. at 29-30; POSNER, supra note 3, at 651. Epstein exemplifies the second point. In Forbidden Grounds - a 500-page, purportedly exhaustive attack on the laws that prohibit employment discrimination - there is no mention of the rich literature on the psychological harms of racism.
150. A partial exception is John J. Donohue III & James J. Heckman, Re-evaluating Federal Civil Rights Policy, 79 GEO. L.J. 1713, 1728-29 (1991). The authors briefly propose an alternative to Becker's model in which race discrimination is supported by coercive social norms and universal belief in the inferiority of blacks. See also J. Hoult Verkerke, Free to Search, 105 HARV. L. REV. 2080, 2090-93 (1992) (discussing the economic importance of coercive social norms and reviewing RICHARD A. EPSTEIN, FORBIDDEN GROUNDS: THE CASE AGAINST EMPLOYMENT DISCRIMINATION LAWS (1992)). Neither work, however, explains the intra-group cooperation necessary for discriminatory norms to function or the existence of racist beliefs. Gary Becker, for example, makes the first point implicitly by defining discrimination solely as a means of avoiding a psychic "cost" one suffers from association. See BECKER, supra note 114, at 14-15, 153. The harm of discrimination is simply the loss of the benefit economic trades would have created, a loss shared by both discriminators and victims. See id. at 29-30; POSNER, supra note 3, at 651. Epstein exemplifies the second point. In Forbidden Grounds - a 500-page, purportedly exhaustive attack on the laws that prohibit employment discrimination - there is no mention of the rich literature on the psychological harms of racism.
151. See GUNNAR MYRDAL, AN AMERICAN DILEMMA 590 (1944) ("[T]he more general allegation that there is an inherent repulsion to personal intimacies and physical contact between the two [racial] groups is unfounded."); see also C. VANN WOODWARD, THE STRANGE CAREER OF JIM CROW 31-32, 42 (3d ed. 1974) (stating that, during the two decades after Reconstruction and before Jim Crow segregation, contact between the races was considerable).
152. See GUNNAR MYRDAL, AN AMERICAN DILEMMA 590 (1944) ("[T]he more general allegation that there is an inherent repulsion to personal intimacies and physical contact between the two [racial] groups is unfounded."); see also C. VANN WOODWARD, THE STRANGE CAREER OF JIM CROW 31-32, 42 (3d ed. 1974) (stating that, during the two decades after Reconstruction and before Jim Crow segregation, contact between the races was considerable).
153. See, e.g., JOHN DOLLARD, CASTE AND CLASS IN A SOUTHERNTOWN 96 (1937) (reporting that in Southerntown, the pseudonym Dollard gives to the town in which he lived and studied racial relations in the early 19305, "97 percent of the women listed under 'domestic and personal service' [in the census] are Negro women").
154. See MYRDAL, supra note 124, at 590, 652 (noting "the acceptability [among whites] of physical contact with favorite servants" and stating that "the Negro maid knows the life of her white employer as few white persons know it").
155. See PATRICIA H. COLLINS, BLACK FEMINIST THOUGHT 71 (1991); MYRDAL, supra note I24, at 597. Indeed, black women sometimes nursed white infants. See WOODWARD, supra note 124, at 42-43; More Slavery at the South, INDEPENDENT, Jan. 25, 1912, at 196-200, reprinted in BLACK WOMEN IN WHITE AMERICA 227, 227 (Gerda Lerner ed., 1972) (under the title I Live a Treadmill Life).
156. See PATRICIA H. COLLINS, BLACK FEMINIST THOUGHT 71 (1991); MYRDAL, supra note I24, at 597. Indeed, black women sometimes nursed white infants. See WOODWARD, supra note 124, at 42-43; More Slavery at the South, INDEPENDENT, Jan. 25, 1912, at 196-200, reprinted in BLACK WOMEN IN WHITE AMERICA 227, 227 (Gerda Lerner ed., 1972) (under the title I Live a Treadmill Life).
157. See PATRICIA H. COLLINS, BLACK FEMINIST THOUGHT 71 (1991); MYRDAL, supra note I24, at 597. Indeed, black women sometimes nursed white infants. See WOODWARD, supra note 124, at 42-43; More Slavery at the South, INDEPENDENT, Jan. 25, 1912, at 196-200, reprinted in BLACK WOMEN IN WHITE AMERICA 227, 227 (Gerda Lerner ed., 1972) (under the title I Live a Treadmill Life).
158. Such relations began during slavery. See BELL HOOKS, AIN'T I A WOMAN 33 (1981); JACQUELINE JONES, LABOR OF LOVE, LABOR OF SORROW 37 (1985). Writing during the Jim Crow era, Dollard suggested that input "many, if not most, southern boys begin their sexual experience with Negro girls." DOLLARD, supra note 125, at 139; see also W.E.B. DUBOIS, DARKWATER 163-86 (1920) (discussing sexual relations between white men and black women).
159. Such relations began during slavery. See BELL HOOKS, AIN'T I A WOMAN 33 (1981); JACQUELINE JONES, LABOR OF LOVE, LABOR OF SORROW 37 (1985). Writing during the Jim Crow era, Dollard suggested that input "many, if not most, southern boys begin their sexual experience with Negro girls." DOLLARD, supra note 125, at 139; see also W.E.B. DUBOIS, DARKWATER 163-86 (1920) (discussing sexual relations between white men and black women).
160. Such relations began during slavery. See BELL HOOKS, AIN'T I A WOMAN 33 (1981); JACQUELINE JONES, LABOR OF LOVE, LABOR OF SORROW 37 (1985). Writing during the Jim Crow era, Dollard suggested that input "many, if not most, southern boys begin their sexual experience with Negro girls." DOLLARD, supra note 125, at 139; see also W.E.B. DUBOIS, DARKWATER 163-86 (1920) (discussing sexual relations between white men and black women).
161. See DOLLARD, supra note 125, at 143 ("A Negro woman reported with indignation that a number of white men in Southerntown have Negro mistresses. . . . [S]ome of these . . . are the ones who insist on segregation."); KENNETH B. CLARK, DARK GHETTO 68-69 (1965) (noting that civil rights movement put social pressure on "any Negro woman who [was] known to be the mistress of . . . a segregationist . . . to break that relationship").
162. Following RICHARD A. POSNER, SEX AND REASON 138-41 (1993), one could say that white men incurred considerably lower costs in obtaining sexual gratification from black women than from white women. To engage in sexual relations with white women, white men in the Jim Crow South usually had to assume the costs of marriage or the risks of extramarital sexual affairs (legal sanctions for rape or adultery, or extra-legal sanctions from a woman's male relatives). Regarding African American women, however, the law permitted no marriage to white men and in practice imposed no penalties for rape, while legal institutions permitted no private retaliation by blacks. See DOLLARD, supra note 125, at 145, 151.
163. See MYRDAL, supra note 124, at 1086.
164. See id. at 284, 1083-84.
165. Indeed, Myrdal reports that black women fared better at getting domestic work in the South: in other regions, many whites "have a prejudice against using Negro women in their homes, partly because they believe them to be less dependable, partly because they shun the contact with an alien race." Id. at 1083 (emphasis added).
166. Regarding long-term affairs, see DUBOIS, cited above in note 128, at 172.
167. See SUSAN ESTRICH, REAL RAPE 72 (1987).
168. BECKER, supra note 114, at 11.
169. See MYRDAL, supra note 124, at 587 ("Every single measure [of enforced segregation] is defended as necessary to block 'social equality' which in its turn is held necessary to prevent intermarriage.'").
170. See, e.g., CLARK, supra note 129, at 69; IDA B, WELLS-BARNETT, ON LYNCHINGS 5-12 (Arno Press 1969) (1892). We can reject the response that conduct aimed at other whites served merely to minimize undesired contact with blacks. Admittedly, forbidding whites to marry (or otherwise associate with) blacks made it less likely that whites would accidentally make contact with a black person through his or her white spouse (or friend). But this fact is not an adequate explanation of white-on-white coercion. A preference for non-association cannot explain the degree to which interracial social contact, especially between black men and white women, was such a central and emotionally charged matter. If the only point were to avoid associating with the black member of an interracial couple, whites should have found interracial couples more acceptable than all-black couples - the former included one person with whom it was acceptable to associate - and should have found a passing sexual affair less troubling than a long-term serious relationship. Yet segregationist whites found the idea of interracial couples abhorrent, and no less so when the relationship was fleeting. See CLARK, supra note 129, at 69; DOLLARD, supra note 125, at 165-66; MYRDAL, supra note 124, at 588.
171. This violence in turn raises another example of the overinclusiveness of the associational preference model. Even violence requires a form of physical "contact" the discriminatory white is supposed to abhor. Lynchings were often very public affairs; sometimes hundreds or thousands of whites would flock to the scene. See, e.g., ARTHUR F. RAPER, THE TRAGEDY OF LYNCHING 12 (1933). White spectators who crowded around protracted torture-lynchings apparently wished to avoid only certain kinds of contact or "association." See id. at 143-44.
172. DOLLARD, supra note 125, at 350-51 ("Any of these acts would imply social equality instead of social inferiority for the Negro.").
173. Id. at 45-46.
174. See id. at 187 ("The white person in Southerntown has no more option about calling a Negro 'Mr.' than the Negro has to demand the title."). Dollard reports the case of a "prominent white man in the state who had to meet a talented Negro woman from Southerntown on various occasions both in town and out of it" and who addressed her with the title "'Mrs.' when they were away, but by her first name in Southerntown." Id. at 346.
175. See id. at 12.
176. Dollard observes: When ['Yankees'] come down South . . . they usually accept the social arrangements and become loyal white-caste members. A number of cases of this were pointed out, especially those of northerners who had come to do business in the South . . . . [Regarding one such person,] [i]t seemed very likely that he had to accept southern views because his social contacts and those of his wife and family were with whites; he could not stand out against the tremendous pressure of white sentiment; and further, he could advance his economic interests only by coöperating with the dominant group. Id. at 48.
177. "There are constant and potent pressures to compel every white person to act his caste rôle correctly." Id. at 49; see also id. at 46-48 (discussing the meaning and effectiveness of epithets whites use against whites who do not conform to racial norms). Dollard's black informants warned him to be careful not to treat them as he treated whites nor as he treated blacks in the North. See id. at 8.
178. See id. at 181. In one town, southern white postal workers even effaced such titles on envelopes addressed to blacks. See NEIL R. MCMILLEN, DARK JOURNEY 24 (1989). In contrast, whites referred to each other with various titles of respect, id. at 23-24, and blacks referred to whites with such deferential titles as "Mr.," "sir," or "boss." See DOLLARD, supra note 125, at 181, 346. Similarly, whites permitted blacks to come to their house only at the back door, and whites typically summoned blacks from their dwellings by honking a horn. See id. at 46, 120.
179. See DOLLARD, supra note 125, at 178-79; see also MCMILLEN, supra note 146, at 23-28 (contrasting racial "etiquette" for blacks and whites in Jim Crow Mississippi).
180. A proponent of the associational preference model might claim that insulting behavior served merely to deter blacks from future association. Yet nothing in the historic literature I reviewed suggests that whites insulted blacks to deter unwanted association. Whites refused to give blacks titles of address and insisted that blacks use the back door whether or not the white initiated and desired the association.
181. See, e.g., F. JAMES DAVIS, WHO IS BLACK? 19-23 (1991); Robert J. Cottrol, The Historical Definition of Race Law, 21 LAW & SOC'Y REV. 865, 865 (1988); Cheryl I. Harris, Whiteness as Property, 106 HARV. L. REV. 1707, 1737-41 (1993).
182. See, e.g., F. JAMES DAVIS, WHO IS BLACK? 19-23 (1991); Robert J. Cottrol, The Historical Definition of Race Law, 21 LAW & SOC'Y REV. 865, 865 (1988); Cheryl I. Harris, Whiteness as Property, 106 HARV. L. REV. 1707, 1737-41 (1993).
183. See, e.g., F. JAMES DAVIS, WHO IS BLACK? 19-23 (1991); Robert J. Cottrol, The Historical Definition of Race Law, 21 LAW & SOC'Y REV. 865, 865 (1988); Cheryl I. Harris, Whiteness as Property, 106 HARV. L. REV. 1707, 1737-41 (1993).
184. See, e.g., MYRDAL, supra note 124, at 683-84; PATRICIA J. WILLIAMS, THE ALCHEMY OF RACE AND RIGHTS 223 (1991); Harris, supra note 149, at 1710-13.
185. See Harris, supra note 149, at 1739.
186. Racial ambiguity also intensifies the overinclusiveness problem. Given the lengths that some southern states went to in excluding from the "white" race people with only distant black ancestry, see id., one would suspect that the associational preference against Africans would be of unmitigated intensity. So why would whites who wished to exclude people who had any African ancestry from the white train car ever employ people of full African ancestry in their homes?
187. What is really required to explain the data is a more complex set of third-party preferences: a desire that members of one's race not associate with other races in public transportation, another that they not associate with other races by using the same rest rooms, another desire that they avoid association by using the same motel, lunch counter, and so on.
188. Even with this strategy, one cannot explain the derogation of blacks as the consequence of an associational preference. One could, of course, posit a discrete preference for disparaging blacks.
189. See George J. Stigler & Gary S. Becker, De Gustibus Non Est Disputandum, 67 AM. ECON. REV. 76, 76 (1977).
190. Gary S. Becker, A Theory of the Allocation of Time, 75 ECON. J. 493, 495 (1965).
191. For example, if we say that people buy microwaves because they have a preference for microwaving food, we can explain the fact that people with identical wealth buy different quantities of microwave ovens only by saying that some people have stronger preferences for microwaves than others. The claim is unfalsifiable because no independent measure of consumer preferences exists except the observed behavior one seeks to explain. With the household production function, however, we can avoid such unfalsifiable explanations. We could predict, for example, that people with less leisure time for cooking (labor being one production input) will make greater use of capital intensive devices like microwave ovens.
192. Compare Stigler & Becker, supra note 155, at 89 ("[N]o significant behavior has been illuminated by assumptions of differences in tastes. Instead, they, along with assumptions of unstable tastes, have been a convenient crutch to lean on when the analysis has bogged down.") with BECKER, supra note 114, at 12: ("[T]he regional difference in market discrimination . . . must be 'explained' by a regional difference in tastes.").
193. See, e.g., MYRDAL, supra note 124, at 591 ("[W]hat white people really want is to keep the Negroes in a lower status."). "[I]n the three-cornered tension among upper class whites, lower class whites, and Negroes, the two white groups agree upon the Negroes as a scapegoat and the proper object for exploitation and hatred. White solidarity is upheld and the caste order protected." Id. at 598; see also FRANK TANNENBAUM, DARKER PHASES OF THE SOUTH 8-9 (1924) (examining the "fear of changing status").
194. These social scientists reject non-instrumental theories that locate prejudice in individual cognitive or personality processes. See DAVID T. WELLMAN, PORTRAITS OF WHITE RACISM 24-26, 48-62 (2d ed. 1993); Herbert Blumer, Race Prejudice as a Sense of Group Position, I PAC. SOC. REV. 3, 3 (1958); Lawrence Bobo, Group Conflict, Prejudice, and the Paradox of Contemporary Racial Attitudes, in ELIMINATING RACISM 85, 95-98 (Phyllis A. Katz & Dalmas A. Taylor eds., 1988); Riia Luhtanen & Jennifer Crocker, Self-Esteem and Intergroup Comparisons: Toward a Theory of Collective Self-Esteem, in SOCIAL COMPARISON, supra note 75, at 211, 225.
195. These social scientists reject non-instrumental theories that locate prejudice in individual cognitive or personality processes. See DAVID T. WELLMAN, PORTRAITS OF WHITE RACISM 24-26, 48-62 (2d ed. 1993); Herbert Blumer, Race Prejudice as a Sense of Group Position, I PAC. SOC. REV. 3, 3 (1958); Lawrence Bobo, Group Conflict, Prejudice, and the Paradox of Contemporary Racial Attitudes, in ELIMINATING RACISM 85, 95-98 (Phyllis A. Katz & Dalmas A. Taylor eds., 1988); Riia Luhtanen & Jennifer Crocker, Self-Esteem and Intergroup Comparisons: Toward a Theory of Collective Self-Esteem, in SOCIAL COMPARISON, supra note 75, at 211, 225.
196. These social scientists reject non-instrumental theories that locate prejudice in individual cognitive or personality processes. See DAVID T. WELLMAN, PORTRAITS OF WHITE RACISM 24-26, 48-62 (2d ed. 1993); Herbert Blumer, Race Prejudice as a Sense of Group Position, I PAC. SOC. REV. 3, 3 (1958); Lawrence Bobo, Group Conflict, Prejudice, and the Paradox of Contemporary Racial Attitudes, in ELIMINATING RACISM 85, 95-98 (Phyllis A. Katz & Dalmas A. Taylor eds., 1988); Riia Luhtanen & Jennifer Crocker, Self-Esteem and Intergroup Comparisons: Toward a Theory of Collective Self-Esteem, in SOCIAL COMPARISON, supra note 75, at 211, 225.
197. These social scientists reject non-instrumental theories that locate prejudice in individual cognitive or personality processes. See DAVID T. WELLMAN, PORTRAITS OF WHITE RACISM 24-26, 48-62 (2d ed. 1993); Herbert Blumer, Race Prejudice as a Sense of Group Position, I PAC. SOC. REV. 3, 3 (1958); Lawrence Bobo, Group Conflict, Prejudice, and the Paradox of Contemporary Racial Attitudes, in ELIMINATING RACISM 85, 95-98 (Phyllis A. Katz & Dalmas A. Taylor eds., 1988); Riia Luhtanen & Jennifer Crocker, Self-Esteem and Intergroup Comparisons: Toward a Theory of Collective Self-Esteem, in SOCIAL COMPARISON, supra note 75, at 211, 225.
198. DOLLARD, supra note 126, at 173-74. Professor Dollard observes: The gain here is very simple. . . . [I]t consists of an illumination of the image of the self, an expansive feeling of being something special and valuable. It might be compared to the illusion of greatness that comes with early stages of alcoholization, except that prestige is not an illusion but a steadily repeated fact. Id. at 174.
199. W.E.B. DUBOIS, BLACK RECONSTRUCTION IN AMERICA 700 (1935).
200. See, e.g., DAVID R. ROEDIGER, THE WAGES OF WHITENESS 13 (1991); Derrick Bell, White Superiority in America: Its Legal Legacy, Its Economic Costs, 33 VILL. L. REV. 767, 774 (1988); Kimberlé W. Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 HARV. L. REV. 1331, 1381 (1988); Harris, supra note 149, at 1758.
201. See, e.g., DAVID R. ROEDIGER, THE WAGES OF WHITENESS 13 (1991); Derrick Bell, White Superiority in America: Its Legal Legacy, Its Economic Costs, 33 VILL. L. REV. 767, 774 (1988); Kimberlé W. Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 HARV. L. REV. 1331, 1381 (1988); Harris, supra note 149, at 1758.
202. See, e.g., DAVID R. ROEDIGER, THE WAGES OF WHITENESS 13 (1991); Derrick Bell, White Superiority in America: Its Legal Legacy, Its Economic Costs, 33 VILL. L. REV. 767, 774 (1988); Kimberlé W. Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 HARV. L. REV. 1331, 1381 (1988); Harris, supra note 149, at 1758.
203. For example, individuals seek to acquire and display "status symbols" such as consumer goods that demonstrate economic success or fashionable taste. See McAdams, supra note 59, at 38-44. Or one might invest in acquiring physical features that will enhance status. See, e.g., Body Work, U.S. NEWS & WORLD REP., Oct. 17, 1994, at 15, 15 (observing that Americans spent over $1.7 billion in cosmetic surgery in 1993).
204. When socially connected groups share a visible trait, they can coordinate either form of status production for the shared trait.
205. Race is a category constructed along the lines of observable traits, but not determined by them. See supra p. 1042. Barbara Fields observes, therefore, that "race" cannot explain "why people of African descent have been set apart for treatment different from that accorded to others" because "race is just the name assigned to the phenomenon" of categorization. Barbara J. Fields, Slavery, Race and Ideology in the United States of America, NEW LEFT REV., May/June 1990, at 95, 100. The proxy effect of observable traits does not explain which physically observable features will come to dominate the production of status (why not left- versus right-handedness?). Which observable features determine racial distinctions is a matter of historical circumstances. For Africans in America, race arose as part of the ideology of slavery. See id. at 101; Harris, supra note 149, at 1715-21. Once the categories are in place, however, status production explains their tenacity. Thus, once the idea of "whiteness" arose, white southerners protected their status by adopting stringent standards for being white, and immigrants struggled to gain the status of being white. See Harris, supra note 149, at 1744 & n.162; ROEDIGER, supra note 163, at 133-63.
206. See infra note 182.
207. See supra pp. 1020-28.
208. See supra pp. 1018-19.
209. The theory of status allocation thus explains higher levels of discrimination in smaller towns: individuals who fail to contribute to maintenance of the external status are easily identified and punished. See DOLLARD, supra note 125, at 15.
210. See CHONG, supra note 52, at 13. Even though a person cannot prevent others from free-riding on a public good, if the private benefits she obtains from such a good exceed her private costs, she will contribute to its provision.
211. The theory thus predicts higher levels of discrimination among low-status whites. See infra p. 1055.
212. ELLICKSON, supra note 76, at 180-81.
213. To Ellickson's compilation of data on social norms, Chong adds that the civil rights movement of the 19605 succeeded because it managed to combine the efforts of various pre-existing socially connected groups, such as black churches. See CHONG, supra note 52, at 35-36.
214. Race is the obvious example: because dark-skinned Africans were brought to America as slaves, socially connected groups were highly stratified by skin color. Immigration provides other examples: a minority culture may, for a time, maintain distinctive language and clothing so that its socially connected groups also share easily discernible traits.
215. The potential exists only when one expects to prevail in the status "war" that status production will bring about. If one shares traits with the majority or the economically and politically powerful, one can expect that subordination of the remainder will succeed. Attempting to produce status on the basis of traits shared by a minority may, however, prompt effective counter-subordination.
216. There are also reasons external to the status-production model to expect continued overlap between shared-trait and socially connected groups. For example, at least in economies that lack a reliable legal infrastructure for enforcing contracts, socially connected minority ethnic groups may enjoy a comparative advantage in informal contract enforcement. See Sumner J. La Croix, Homogenous Middleman Groups: What Determines the Homogeneity?, 5 J.L. ECON. & ORG. 211, 217-20 (1989); Janet T. Landa, A Theory of the Ethnically Homogenous Middleman Group: An Institutional Alternative to Contract Law, 10 J. LEGAL STUD. 349, 359-62 (1981).
217. There are also reasons external to the status-production model to expect continued overlap between shared-trait and socially connected groups. For example, at least in economies that lack a reliable legal infrastructure for enforcing contracts, socially connected minority ethnic groups may enjoy a comparative advantage in informal contract enforcement. See Sumner J. La Croix, Homogenous Middleman Groups: What Determines the Homogeneity?, 5 J.L. ECON. & ORG. 211, 217-20 (1989); Janet T. Landa, A Theory of the Ethnically Homogenous Middleman Group: An Institutional Alternative to Contract Law, 10 J. LEGAL STUD. 349, 359-62 (1981).
218. See Peggy C. Davis, Law as Microaggression, 98 YALE L.J. 1559, 1565-68 (1989); Patricia Williams, Spirit-Murdering the Messenger: The Discourse of Fingerpointing as the Law's Response to Racism, 42 U. MIAMI L. REV. 127, 129 (1987).
219. See Peggy C. Davis, Law as Microaggression, 98 YALE L.J. 1559, 1565-68 (1989); Patricia Williams, Spirit-Murdering the Messenger: The Discourse of Fingerpointing as the Law's Response to Racism, 42 U. MIAMI L. REV. 127, 129 (1987).
220. See ARISTOTLE, RHETORIC II.2.1378b-79a (W.D. Ross ed. & W. Rhys Roberts trans., 1924). The true snub occurs only when one goes out of one's way to ignore someone. Refusing to accept a gift is, other things being equal, more insulting than refusing to make a gift. Refusing to accept an invitation to do something one enjoys is more insulting than refusing to participate in an activity one does not enjoy. In each case, acting against one's own interests is both a less equivocal and a more intense signal of one's desire to insult.
221. See McAdams, supra note 59, at 40 & n.158, 42 n.170.
222. Individuals do not, however, invariably prefer to be in the numerical majority. If a minority is high-status as a consequence of enjoying disproportionate political or economic power, or other badges of status, individuals may struggle to define themselves as belonging to that group instead of to the majority.
223. For group status production, there must be at least two groups, and thus, at least one boundary. Rigorously enforcing the boundary symbolizes the greater exclusivity of the dominant group. Thus race definitions that treat black ancestry as "tainting" an apparently white person contribute to the disparagement of blacks and the production of status for those defined as "truly" white.
224. Like Becker, I believe that the appropriate starting point for explaining discrimination is we paradigmatic case of overt racial animus. Although status production suggests that economists have misclassified some subtle racial animus as statistical discrimination, status production is only a starting point. It should be supplemented with other theories, including, where appropriate, the theory of statistical discrimination.
225. See pp. 1040-41.
226. Moreover, for whites, occupying the subordinate position in a hierarchical relationship was a fate worse than equality. Having black elected officials was, therefore, anathema to discriminatory whites because it implied such a subordinate position.
227. See DOLLARD, supra note 125, at 143-45.
228. See MYRDAL, supra note 124, at 590.
229. The punishment takes the form of lowered intra-group status. "The white people enforce caste rules with ominous unanimity and one is compelled, by one's white-caste membership, to assist to some degree in the personal derogation of the Negro and the expression of hostile pressure against him." DOLLARD, supra note 125, at 349. Social ostracism was a powerful threat: "If one lives in Southerntown, 'not to be received' is a very serious matter and would be more so if one's family were there; living would be quite intolerable without opportunity for friendly contacts within the white caste." Id. at 354.
230. Id. at 66 (emphasis added).
231. See RAPER, supra note 139, at 20. When a lynching occurred, "[t]he general public either justified or condoned the lynching, and any individual or group who disagreed was made to suffer. Merchants, bankers, lawyers, and preachers faced a public boycott - or thought they did - should they take a stand in defense of law and order." Id. Raper reported that a National Guardsman who used his bayonet to cut a white man who was attempting to remove a black rape suspect from custody "was never able after that time to keep employment in [that town]." Id. at 244.
232. See DOLLARD, supra note 125, at 127.
233. The well-documented violence directed by whites against blacks is evidence that whites overcame the incentive to free-ride. After all, one incurs some cost by participating in a lynching, given the value of one's time and the risk of criminal ramifications which, while extremely low, were higher than zero. See RAPER, supra note 139, at 2, 32. The costs could be quite high if the lynching was preceded by an "extended man-hunt[ ]" by "undeputized" men, or if law enforcement officials uncharacteristically resisted or threatened to resist a mob's attempt to remove a suspect from jail. Id, at 1, 9-10, 14-15, 32, 244.
234. See DOLLARD, supra note 125, at 176-77 ("(W]hite people become aggressive as soon as Negro submission is withheld, and many stories are told of the 'what I did with that "uppity" nigger' type."); RAPER, supra note 139, at 48 ("'The Negro must stay in his place' is a common phrase, and the white people who have denned 'his place' may be expected to react violently if they think he is either getting out of it or showing dissatisfaction with it").
235. See MYRDAL, supra note 124, at 563. Myrdal reports: Economic fear is mixed with social fear: a feeling that the Negro is "getting out of his place," and [that] the white man's social status is being threatened and is in need of defense. . . , [T]his feeling [underlies the] common saying . . . that "a lynching now and then" is expedient or necessary in keeping the Negroes from becoming "uppity.". . . . [A]fter the First World War many lynchings of Negro soldiers - sometimes in uniform - were openly motivated by the fear that they had gotten "wrong ideas" about their social status while serving in France. Id.; see also WALTER WHITE, ROPE AND FAGGOT 11 (1969) (suggesting that "lynching is much more an expression of southern fear of Negro progress than of Negro crime").
236. "While the Klan undoubtedly retained large numbers of middle-class supporters [in the 1920s] until its ultimate decline, the available evidence suggests that these were largely among the socially marginal and status-deprived elements in the stratum. . . . Its leaders spoke the language of the underdog Protestant." SEYMOUR M. LIPSET & EARL RAAB, THE POLITICS OF UNREASON 125 (1978). In the decade after Brown v. Board of Education, 347 U.S. 483 (1954), "the Klan tended to be less of a resistance than a status movement for some of those left insecure or unrecognized by society." DAVID M. CHALMERS, HOODED AMERICANISM 6 (1965). Racial violence also served to unite diverse white subgroups. See RAPER, supra note 139, at 47 ("[Llynchings often serve as socializing forces . . . [that] tend to minimize social and class distinctions between white plantation owners and white tenants, mill owners and textile workers, Methodists and Baptists . . . .").
237. "While the Klan undoubtedly retained large numbers of middle-class supporters [in the 1920s] until its ultimate decline, the available evidence suggests that these were largely among the socially marginal and status-deprived elements in the stratum. . . . Its leaders spoke the language of the underdog Protestant." SEYMOUR M. LIPSET & EARL RAAB, THE POLITICS OF UNREASON 125 (1978). In the decade after Brown v. Board of Education, 347 U.S. 483 (1954), "the Klan tended to be less of a resistance than a status movement for some of those left insecure or unrecognized by society." DAVID M. CHALMERS, HOODED AMERICANISM 6 (1965). Racial violence also served to unite diverse white subgroups. See RAPER, supra note 139, at 47 ("[Llynchings often serve as socializing forces . . . [that] tend to minimize social and class distinctions between white plantation owners and white tenants, mill owners and textile workers, Methodists and Baptists . . . .").
238. "Enticement" laws, for example, restricted the ability of southern plantation owners to hire workers away from other plantations, see MYRDAL, supra note 124, at 248-49, and thus depressed the wages of blacks. See EPSTEIN, supra note 116, at 46, 246-49; Robert Cooter, Market Affirmative Action, 31 SAN DIEGO L. REV. 133, 139-41 (1994).
239. Nor does rent-seeking explain why race, as opposed to any other means of grouping, was the axis along which individuals chose to organize. The status-production theory, however, explains why there is frequent overlap between shared-trait groups, like races, and the socially connected groups that are capable of inducing their members to contribute to lobbying activities. See supra notes 175-176 and accompanying text.
240. See BECKER, supra note 114, at 17, 21, 156.
241. See WOODWARD, supra note 124, at 13.
242. An associational preference proponent might respond that slavery was more prevalent in the South because it was a more productive means of agriculture for southern than for northern farmers. Yet southern whites seemed to permit more association than slavery required, at least for house servants and in urban areas. See, e.g., id. at 12-16. Moreover, even if agricultural economics explains why slavery predominated in the South, it cannot explain why southerners were more discriminatory than northerners.
243. See DOLLARD, supra note 125, at 50-51, 186.
244. See WHITE, supra note 194, at 11.
245. See, e.g., CHARLES A. LOFGREN, THE PLESSY CASE 25 (1987) ("[T]he most direct trigger for the initial wave of Jim Crow legislation was increasing black unwillingness to defer to whites.").
246. See Michael J. Klarman, Brown, Racial Change, and the Civil Rights Movement, 80 VA. L. REV. 7, 30 (1994) ("At the beginning of the twentieth century, over 90% of American blacks lived in the eleven former Confederate states . . . .").
247. The smaller the upwardly mobile group, the less their success will push other groups down in relative terms; indeed, effectively subordinating a larger group necessarily places one higher in social rank than subordinating a smaller one. The status-production theory thus explains why, within southern counties, severity of discrimination was positively correlated with black population. See id. at 68. A larger group, however, is more difficult to subordinate than a smaller one; absent disproportionate economic or political power by the majority, we might expect that as the size of the minority population approaches a majority, subordination will level off or decline. In the Jim Crow South, however, whites enjoyed such disproportionate power that they could effectively subordinate a black majority, and the above correlation still held in such cases. See id. Becker's theory also fails to explain why segregation appeared to be so one-sided, imposed by whites on blacks. See MYRDAL, supra note 124, at 575-77. Whites appeared to have had a stronger aversion to interracial contact than blacks. Violence may have deterred blacks from expressing their desire to avoid association with whites. But in the associational model, because the only purpose of violence is to deter association, the one-sidedness of violence suggests that blacks wanted association more than whites. The status-production model offers an explanation: given the stark status hierarchy, whites stood to lose social status by associating in equal relationships with blacks, while blacks did not. This explanation is consistent with the call by some black intellectuals at the time for separatism. See, e.g., Robert S. Browne Urges "A Formal Partitioning of the United States Into Two Totally Separate and Independent Nations," in BLACK PROTEST THOUGHT IN THE TWENTIETH CENTURY 516, 526-28 (August Meier, Elliott Rudwick & Francis L. Broderick eds., 2d ed. 1971). Segregation loses some of its stigma when the minority chooses to separate rather than having the majority choose to exclude. Moreover, if one can form sufficiently intense feelings about group members, it is possible one will no longer care about the esteem granted or withheld by non-members. Cf. Julius Lester, The Angry Children of Malcom X, in BLACK PROTEST THOUGHT IN THE TWENTIETH CENTURY, supra, at 469, 483 ("[The white man] is not to be lived with and he is not to be destroyed. He is simply to be ignored . . . .").
248. The smaller the upwardly mobile group, the less their success will push other groups down in relative terms; indeed, effectively subordinating a larger group necessarily places one higher in social rank than subordinating a smaller one. The status-production theory thus explains why, within southern counties, severity of discrimination was positively correlated with black population. See id. at 68. A larger group, however, is more difficult to subordinate than a smaller one; absent disproportionate economic or political power by the majority, we might expect that as the size of the minority population approaches a majority, subordination will level off or decline. In the Jim Crow South, however, whites enjoyed such disproportionate power that they could effectively subordinate a black majority, and the above correlation still held in such cases. See id. Becker's theory also fails to explain why segregation appeared to be so one-sided, imposed by whites on blacks. See MYRDAL, supra note 124, at 575-77. Whites appeared to have had a stronger aversion to interracial contact than blacks. Violence may have deterred blacks from expressing their desire to avoid association with whites. But in the associational model, because the only purpose of violence is to deter association, the one-sidedness of violence suggests that blacks wanted association more than whites. The status-production model offers an explanation: given the stark status hierarchy, whites stood to lose social status by associating in equal relationships with blacks, while blacks did not. This explanation is consistent with the call by some black intellectuals at the time for separatism. See, e.g., Robert S. Browne Urges "A Formal Partitioning of the United States Into Two Totally Separate and Independent Nations," in BLACK PROTEST THOUGHT IN THE TWENTIETH CENTURY 516, 526-28 (August Meier, Elliott Rudwick & Francis L. Broderick eds., 2d ed. 1971). Segregation loses some of its stigma when the minority chooses to separate rather than having the majority choose to exclude. Moreover, if one can form sufficiently intense feelings about group members, it is possible one will no longer care about the esteem granted or withheld by non-members. Cf. Julius Lester, The Angry Children of Malcom X, in BLACK PROTEST THOUGHT IN THE TWENTIETH CENTURY, supra, at 469, 483 ("[The white man] is not to be lived with and he is not to be destroyed. He is simply to be ignored . . . .").
249. See, e.g., Donald L. Noel & Alphonso Pinkney, Correlates of Prejudice: Some Racial Differences and Similarities, 69 AM. J. Soc. 609, 611-12 (1964); Thomas J. Pavlak, Social Class, Ethnicity, and Racial Prejudice, 37 PUB. OPINION Q. 225, 229-30 (1973); A. Wade Smith, Racial Tolerance as a Function of Group Position, 46 AM. Soc. REV. 558, 563-67 (1981).
250. See, e.g., Donald L. Noel & Alphonso Pinkney, Correlates of Prejudice: Some Racial Differences and Similarities, 69 AM. J. Soc. 609, 611-12 (1964); Thomas J. Pavlak, Social Class, Ethnicity, and Racial Prejudice, 37 PUB. OPINION Q. 225, 229-30 (1973); A. Wade Smith, Racial Tolerance as a Function of Group Position, 46 AM. Soc. REV. 558, 563-67 (1981).
251. See, e.g., Donald L. Noel & Alphonso Pinkney, Correlates of Prejudice: Some Racial Differences and Similarities, 69 AM. J. Soc. 609, 611-12 (1964); Thomas J. Pavlak, Social Class, Ethnicity, and Racial Prejudice, 37 PUB. OPINION Q. 225, 229-30 (1973); A. Wade Smith, Racial Tolerance as a Function of Group Position, 46 AM. Soc. REV. 558, 563-67 (1981).
252. See RAPER, supra note 139, at 6, 10-11.
253. JUDITH CADITZ, WHITE LIBERALS IN TRANSITION 91 (1976). A review of psychology literature reveals a similar finding: "the more negative the self-attitudes, the greater the number of unacceptable targets and the more negative are attitudes toward them." HOWARD J. EHRLICH, THE SOCIAL PSYCHOLOGY OF PREJUDICE 163 (1973); see also JOHN E. FARLEY, MAJORITY-MINORITY RELATIONS 26 (2d ed. 1988) ("Persons in lower [socioeconomic status] groups tend to report more negative views toward out-groups, to be more ethnocentric, and to express more stereotyped thinking . . . ."); GEORGE E. SIMPSON & J. MILTON YINGER, RACIAL AND CULTURAL MINORITIES: AN ANALYSIS OF PREJUDICE AND DISCRIMINATION 66-70 (5th ed. 1985) (finding "strong evidence[] for class differentials in prejudice").
254. JUDITH CADITZ, WHITE LIBERALS IN TRANSITION 91 (1976). A review of psychology literature reveals a similar finding: "the more negative the self-attitudes, the greater the number of unacceptable targets and the more negative are attitudes toward them." HOWARD J. EHRLICH, THE SOCIAL PSYCHOLOGY OF PREJUDICE 163 (1973); see also JOHN E. FARLEY, MAJORITY-MINORITY RELATIONS 26 (2d ed. 1988) ("Persons in lower [socioeconomic status] groups tend to report more negative views toward out-groups, to be more ethnocentric, and to express more stereotyped thinking . . . ."); GEORGE E. SIMPSON & J. MILTON YINGER, RACIAL AND CULTURAL MINORITIES: AN ANALYSIS OF PREJUDICE AND DISCRIMINATION 66-70 (5th ed. 1985) (finding "strong evidence[] for class differentials in prejudice").
255. JUDITH CADITZ, WHITE LIBERALS IN TRANSITION 91 (1976). A review of psychology literature reveals a similar finding: "the more negative the self-attitudes, the greater the number of unacceptable targets and the more negative are attitudes toward them." HOWARD J. EHRLICH, THE SOCIAL PSYCHOLOGY OF PREJUDICE 163 (1973); see also JOHN E. FARLEY, MAJORITY-MINORITY RELATIONS 26 (2d ed. 1988) ("Persons in lower [socioeconomic status] groups tend to report more negative views toward out-groups, to be more ethnocentric, and to express more stereotyped thinking . . . ."); GEORGE E. SIMPSON & J. MILTON YINGER, RACIAL AND CULTURAL MINORITIES: AN ANALYSIS OF PREJUDICE AND
256. JUDITH CADITZ, WHITE LIBERALS IN TRANSITION 91 (1976). A review of psychology literature reveals a similar finding: "the more negative the self-attitudes, the greater the number of unacceptable targets and the more negative are attitudes toward them." HOWARD J. EHRLICH, THE SOCIAL PSYCHOLOGY OF PREJUDICE 163 (1973); see also JOHN E. FARLEY, MAJORITY-MINORITY RELATIONS 26 (2d ed. 1988) ("Persons in lower [socioeconomic status] groups tend to report more negative views toward out-groups, to be more ethnocentric, and to express more stereotyped thinking . . . ."); GEORGE E. SIMPSON & J. MILTON YINGER, RACIAL AND CULTURAL MINORITIES: AN ANALYSIS OF PREJUDICE AND DISCRIMINATION 66-70 (5th ed. 1985) (finding "strong evidence[] for class differentials in prejudice").
257. See CADITZ, supra note 208, at 100-01. For example Caditz found that "[t]hose affirming the importance of their ethnic group exhibit more dilemmas toward busing, apartment rentals to blacks, and hiring blacks than the comparison group." Id. at 101.
258. Individuals can obtain status by achievement or talent - for example, displaying wealth is a common means of gaining esteem. See McAdams, supra note 59, at 38-44. Those who have neither wealth nor some other positive accomplishment as a basis for esteem are the most likely to seek esteem in other ways, such as subordination.
259. Caditz's research supports this finding. She queried white "liberals" about race: In general, the more stable and status secure, the more likely white liberals will affirm favorable attitudes toward interracial social situations. Conversely, the more status insecure, the more status threatened white liberals will feel, and consequently, the more likely they will be to reveal attitudes unfavorable toward interracial situations. CADITZ, supra note 208, at 107. Yet Caditz found "no significant differences between groups compared on the basis of their actual ranks in occupation and in education." Id. at 104. She concludes that "mobility per se, irrespective of the actual ranks, is associated with the degree of favorableness toward . . . interracial situations." Id.; see also BRUNO BETTELHEIM & MORRIS JANOWITZ, DYNAMICS OF PREJUDICE 150-51 (1950) (finding the greatest prejudice against blacks among the downwardly mobile, followed by the non-mobile, then the upwardly mobile); Joseph Greenblum & Leonard I. Pearlin, Vertical Mobility and Prejudice: A Socio-Psychological Analysis, in CLASS, STATUS AND POWER 480, 491 (Reinhard Bendix & Seymour M. Lipset eds., 1953) (finding more ethnic prejudice among upwardly and downwardly mobile than among non-mobile persons).
260. Caditz's research supports this finding. She queried white "liberals" about race: In general, the more stable and status secure, the more likely white liberals will affirm favorable attitudes toward interracial social situations. Conversely, the more status insecure, the more status threatened white liberals will feel, and consequently, the more likely they will be to reveal attitudes unfavorable toward interracial situations. CADITZ, supra note 208, at 107. Yet Caditz found "no significant differences between groups compared on the basis of their actual ranks in occupation and in education." Id. at 104. She concludes that "mobility per se, irrespective of the actual ranks, is associated with the degree of favorableness toward . . . interracial situations." Id.; see also BRUNO BETTELHEIM & MORRIS JANOWITZ, DYNAMICS OF PREJUDICE 150-51 (1950) (finding the greatest prejudice against blacks among the downwardly mobile, followed by the non-mobile, then the upwardly mobile); Joseph Greenblum & Leonard I. Pearlin, Vertical Mobility and Prejudice: A Socio-Psychological Analysis, in CLASS, STATUS AND POWER 480, 491 (Reinhard Bendix & Seymour M. Lipset eds., 1953) (finding more ethnic prejudice among upwardly and downwardly mobile than among non-mobile persons).
261. See MYRDAL, supra note 124, at 603 ("[T]he development of prejudice against Negroes [was] usually one of [the] first lessons in Americanization for [new immigrants residing in the North]. Because they are of low status, they like to have a group like the Negroes to which they can be superior."); see also ROEDIGER, supra note 163, at 133-63 (discussing the history of Irish-American racial consciousness).
262. See CADITZ, supra note 208, at 118 (discussing conflict between blacks and Mexican Americans); ROEDIGER, supra note 163, at 133-63 (discussing conflict between blacks and Irish immigrants).
263. Professor Bobo reports: In 1942, approximately 60% of whites believed that blacks were less intelligent than whites. By 1964, that figure had declined to less than 25%. A substantial majority of white Americans in 1942 approved of the blatantly discriminatory proposition that "white people should have the first chance at any kind of job," whereas in 1972 nearly 100% of whites in a national survey rejected that statement. Bobo, supra note 160, at 88 (citations omitted). See generally HOWARD SCHUMAN, CHARLOTTE STEEH & LAWRENCE BOBO, RACIAL ATTITUDES IN AMERICA 71-138 (1985) (discussing trends in white racial attitudes between 1942 and 1983). Even if survey subjects are concealing their "real" views, the survey results suggest that something has changed, if only the willingness to admit to anti-black or anti-integrationist views.
264. Whites remain reluctant to vote for black politicians. Cf. Donald R. Kinder & David O. Sears, Prejudice and Politics: Symbolic Racism Versus Racial Threats to the Good Life, 40 J. PERSONALITY & SOC. PSYCHOL. 414, 427-30 (1981) (analyzing white voters' racial attitudes and preferences); John B. McConahay & Joseph C. Hough, Jr., Symbolic Racism, 32 J. SOC. ISSUES 23, 36-44 (1976) (analyzing voter preferences in terms of "symbolic racism"). That reluctance increases when blacks are a large percentage of the population. See SCHUMAN, STEEH & BOBO, supra note 214, at 204. Regarding economic discrimination, see notes 239 and 258 below.
265. Whites remain reluctant to vote for black politicians. Cf. Donald R. Kinder & David O. Sears, Prejudice and Politics: Symbolic Racism Versus Racial Threats to the Good Life, 40 J. PERSONALITY & SOC. PSYCHOL. 414, 427-30 (1981) (analyzing white voters' racial attitudes and preferences); John B. McConahay & Joseph C. Hough, Jr., Symbolic Racism, 32 J. SOC. ISSUES 23, 36-44 (1976) (analyzing voter preferences in terms of "symbolic racism"). That reluctance increases when blacks are a large percentage of the population. See SCHUMAN, STEEH & BOBO, supra note 214, at 204. Regarding economic discrimination, see notes 239 and 258 below.
266. See SCHUMAN, STEEH & BOBO, supra note 214, at 154-62 (discussing disparity between black support for principles of integration and implementation of such principles); Bobo, supra note 160, at 88-89; Thomas F. Pettigrew, Racial Change and Social Policy, 441 ANNALS AM. ACAD. POL. & SOC. SCI. 114, 119 (1979) ("[W]hite Americans increasingly reject racial injustice in principle but are reluctant to accept the measures necessary to eliminate the injustice."). Of course, white opposition may arise out of commitment to libertarian principles or beliefs about the cost and benefits of such proposals. But the question remains why whites and blacks hold such different views on these matters. See SCHUMAN, STEEH & BOBO, supra note 214, at 139-62; Lawrence Bobo, Attitudes Toward the Black Political Movement: Trends, Meaning, and Effects on Racial Policy Preferences, 51 Soc. PSYCHOL. Q. 287, 299-300 (1988).
267. See SCHUMAN, STEEH & BOBO, supra note 214, at 154-62 (discussing disparity between black support for principles of integration and implementation of such principles); Bobo, supra note 160, at 88-89; Thomas F. Pettigrew, Racial Change and Social Policy, 441 ANNALS AM. ACAD. POL. & SOC. SCI. 114, 119 (1979) ("[W]hite Americans increasingly reject racial injustice in principle but are reluctant to accept the measures necessary to eliminate the injustice."). Of course, white opposition may arise out of commitment to libertarian principles or beliefs about the cost and benefits of such proposals. But the question remains why whites and blacks hold such different views on these matters. See SCHUMAN, STEEH & BOBO, supra note 214, at 139-62; Lawrence Bobo, Attitudes Toward the Black Political Movement: Trends, Meaning, and Effects on Racial Policy Preferences, 51 Soc. PSYCHOL. Q. 287, 299-300 (1988).
268. Many have argued that voting is not instrumentally rational. One ballot can affect the outcome of an election only if the vote would otherwise be tied. Thus, the odds against one vote's affecting the outcome of a modern American election are so great that the cost of voting (even the very low risk of dying on the way to the ballot box or the five minutes it takes to vote) apparently outweighs by a wide margin any expected instrumental benefit. See, e.g., DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE 24-27 (1991). Voting is rational, however, if understood as an expressive end in itself. The utility derived from voting "is related to the utility that people derive from speaking their mind about some subject, even though nothing will be changed thereby." Richard A. Posner, What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does), 3 SUP. CT. ECON. REV. 1, 19 (1993); see also Geoffrey Brennan & Loren Lomasky, The Impartial Spectator Goes to Washington: Toward a Smithian Theory of Electoral Behavior, I ECON. & PHIL. 189, 199 (1985) (arguing that voting is "a potentially valued avenue for the expression of moral sentiments").
269. Many have argued that voting is not instrumentally rational. One ballot can affect the outcome of an election only if the vote would otherwise be tied. Thus, the odds against one vote's affecting the outcome of a modern American election are so great that the cost of voting (even the very low risk of dying on the way to the ballot box or the five minutes it takes to vote) apparently outweighs by a wide margin any expected instrumental benefit. See, e.g., DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE 24-27 (1991). Voting is rational, however, if understood as an expressive end in itself. The utility derived from voting "is related to the utility that people derive from speaking their mind about some subject, even though nothing will be changed thereby." Richard A. Posner, What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does), 3 SUP. CT. ECON. REV. 1, 19 (1993); see also Geoffrey Brennan & Loren Lomasky, The Impartial Spectator Goes to Washington: Toward a Smithian Theory of Electoral Behavior, I ECON. & PHIL. 189, 199 (1985) (arguing that voting is "a potentially valued avenue for the expression of moral sentiments").
270. Many have argued that voting is not instrumentally rational. One ballot can affect the outcome of an election only if the vote would otherwise be tied. Thus, the odds against one vote's affecting the outcome of a modern American election are so great that the cost of voting (even the very low risk of dying on the way to the ballot box or the five minutes it takes to vote) apparently outweighs by a wide margin any expected instrumental benefit. See, e.g., DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE 24-27 (1991). Voting is rational, however, if understood as an expressive end in itself. The utility derived from voting "is related to the utility that people derive from speaking their mind about some subject, even though nothing will be changed thereby." Richard A. Posner, What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does), 3 SUP. CT. ECON. REV. 1, 19 (1993); see also Geoffrey Brennan & Loren Lomasky, The Impartial Spectator Goes to Washington: Toward a Smithian Theory of Electoral Behavior, I ECON. & PHIL. 189, 199 (1985) (arguing that voting is "a potentially valued avenue for the expression of moral sentiments").
271. Supermarket tabloids and talk radio often thrive on the expression of palpably false beliefs, apparently because they are false in an interesting way.
272. See Hazel Markus, The Self in Thought and Memory, in THE SELF IN SOCIAL PSYCHOLOGY 102, 125 (Daniel M. Wegner & Robin R. Vallacher eds., 1980); Shelley E. Taylor & Jonathon D. Brown, Illusion and Well-Being: A Social Psychological Perspective on Mental Health, 103 PSYCHOL. BULL. 193, 197 (1988) ("[A] great deal of research in social, personality, clinical, and developmental psychology documents that normal individuals possess unrealistically positive views of themselves . . . .").
273. See Hazel Markus, The Self in Thought and Memory, in THE SELF IN SOCIAL PSYCHOLOGY 102, 125 (Daniel M. Wegner & Robin R. Vallacher eds., 1980); Shelley E. Taylor & Jonathon D. Brown, Illusion and Well-Being: A Social Psychological Perspective on Mental Health, 103 PSYCHOL. BULL. 193, 197 (1988) ("[A] great deal of research in social, personality, clinical, and developmental psychology documents that normal individuals possess unrealistically positive views of themselves . . . .").
274. See Taylor & Brown, supra note 219, at 197: Howard Tennen & Sharon Herzberger, Depression, Self-Esteem, and the Absence of Self-Protective Attributional Biases, 52 J. PERSONALITY & SOC. PSYCHOL. 72, 72 (1987).
275. See Tajfel & Turner, supra note 34, at 7; cf. Taylor & Brown, supra note 219, at 195 (stating that "individuals [tend] to see their intimates as better than average"); Alvin Zander & Warwick Armstrong, Working for Group Pride in a Slipper Factory, 2 J. APPLIED SOC. PSYCHOL. 293, 304 (1972) (discussing how interest in team pride affects beliefs about the team).
276. See Richard A. Posner, The Material Basis of Jurisprudence, 69 IND. L.J. 1, 10 (1993); see also ELSTER, supra note 11, at 37-38 (describing "wishful thinking," a process by which individuals form beliefs they previously wished to be true). See generally THOMAS GILOVICH, How WE KNOW WHAT ISN'T SO: THE FALLIBILITY OF HUMAN REASON IN EVERYDAY LIFE 75-87 (1991) (discussing how desire leads to self-serving beliefs); DAVID PEARS, MOTIVATED IRRATIONALITY 41-66 (1984) (discussing explanations for self-deception).
277. See Richard A. Posner, The Material Basis of Jurisprudence, 69 IND. L.J. 1, 10 (1993); see also ELSTER, supra note 11, at 37-38 (describing "wishful thinking," a process by which individuals form beliefs they previously wished to be true). See generally THOMAS GILOVICH, How WE KNOW WHAT ISN'T SO: THE FALLIBILITY OF HUMAN REASON IN EVERYDAY LIFE 75-87 (1991) (discussing how desire leads to self-serving beliefs); DAVID PEARS, MOTIVATED IRRATIONALITY 41-66 (1984) (discussing explanations for self-deception).
278. See Richard A. Posner, The Material Basis of Jurisprudence, 69 IND. L.J. 1, 10 (1993); see also ELSTER, supra note 11, at 37-38 (describing "wishful thinking," a process by which individuals form beliefs they previously wished to be true). See generally THOMAS GILOVICH, How WE KNOW WHAT ISN'T SO: THE FALLIBILITY OF HUMAN REASON IN EVERYDAY LIFE 75-87 (1991) (discussing how desire leads to self-serving beliefs); DAVID PEARS, MOTIVATED IRRATIONALITY 41-66 (1984) (discussing explanations for self-deception).
279. See Posner, supra note 222, at 11-12.
280. See id. at 11. "[A]n aid to, and perhaps even a condition of, successful cartelization is the creation of an ideological rather than a purely contractual community." Id. at 20.
281. Id. at 11. Posner compares the characteristics of medieval guilds to the modern legal profession. He argues that lawyers have used guild-like strategies to cartelize the supply of legal services and that a certain part of what passes as "jurisprudence" is this profession's self-serving ideology. See id. at 1-3.
282. 227 Thus, if voting has only expressive benefits, whites must believe, more often than blacks, that policies that help blacks are bad ideas and that black candidates will do a poor job. Indeed, many studies have found that self-interest does not explain white opposition to black candidates or other race-related voting. Personal vulnerability on an issue does not successfully predict attitudes or voting behavior concerning those issues. See James R. Kluegel & Eliot R. Smith, Affirmative Action Attitudes: Effects of Self-interest, Racial Affect, and Stratification Beliefs on Whites' Views, 61 SOC. FORCES 797, 813 (1983); David O. Sears & Harris M. Allen, Jr., The Trajectory of Local Desegregation Controversies and Whites' Opposition to Busing, in GROUPS IN CONTACT: THE PSYCHOLOGY OF DESEGREGATION 123, 128-29 (Norman Miller & Marilynn B. Brewer eds., 1984).
283. 227 Thus, if voting has only expressive benefits, whites must believe, more often than blacks, that policies that help blacks are bad ideas and that black candidates will do a poor job. Indeed, many studies have found that self-interest does not explain white opposition to black candidates or other race-related voting. Personal vulnerability on an issue does not successfully predict attitudes or voting behavior concerning those issues. See James R. Kluegel & Eliot R. Smith, Affirmative Action Attitudes: Effects of Self-interest, Racial Affect, and Stratification Beliefs on Whites' Views, 61 SOC. FORCES 797, 813 (1983); David O. Sears & Harris M. Allen, Jr., The Trajectory of Local Desegregation Controversies and Whites' Opposition to Busing, in GROUPS IN CONTACT: THE PSYCHOLOGY OF DESEGREGATION 123, 128-29 (Norman Miller & Marilynn B. Brewer eds., 1984).
284. 230 In fact, the pretense of equality served to enhance the production of status because whites could then (choose to selfishly) believe that the failures of blacks to achieve equality were their own fault. In a sense, whites could gain the benefit of cheating in a competition without the loss of self-respect that comes from acknowledging that one is cheating.
285. 230 In fact, the pretense of equality served to enhance the production of status because whites could then (choose to selfishly) believe that the failures of blacks to achieve equality were their own fault. In a sense, whites could gain the benefit of cheating in a competition without the loss of self-respect that comes from acknowledging that one is cheating.
286. See Klarman, supra note 204, at 23-26.
287. See id. at 20-21 (describing how World War II labor shortages increased black employment in skilled positions); id. at 52-65 (reporting that urbanization and industrialization increased demand for black labor and that industrial jobs created a black middle class less vulnerable to white control); id. at 65-67 (detailing increased education levels). In other words, discriminatory norms may be stable given fixed economic conditions but may change if new conditions make transacting with blacks more materially beneficial.
288. The immediate effect was to convince many northern whites that segregation was sufficiently immoral to merit national legislation. See id. at 145-49. As to southern whites, I speculate that the same force contributed to long term attitude changes.
289. See supra notes 215-216.
290. The model also provides a useful beginning for economic analysis of sexism. Like racial features, one's sex is reasonably observable. Men, like whites, often belong to exclusive socially connected groups. Racism and sexism each allow the oppressor to "feel superior; thus, a 'poor white' in the South can console himself with the thought that he is not a 'dirty nigger' . . . . [T]he most mediocre of males feels himself a demigod as compared with women." SIMONE DE BEAUVOIR, THE SECOND SEX at xxiv (H.M. Parshley ed. & trans., Alfred A. Knopf 1953) (1949). As Beauvoir notes, however, there are important differences between sexism and racism. See id. at xviii. Within the status theory, the most important distinction is the extent to which men and women depend on each other for their social status. Men typically belong to socially connected groups, such as families, that contain women. Given such relationships, a man who disparages women would seem to threaten his own status. The male's response to this problem is twofold. First, he claims the dominant position within the family relationship. Second, he seeks a means of distinguishing the females with whom he associates from the rest of the females of the world. Of course, status is obtained from other men in this manner only if it derives from some male consensus as to what constitutes desirable female behavior or attributes. Consequently, social norms arise within socially connected groups of men that define a standard of female behavior. Men then gain status by being associated with women who follow these roles well. Thus, men subordinate a group that includes family members by employing a rich ideology of gender roles that allows them to distinguish between their female relatives and other women - between "good" and "bad" women. Another key distinction between racism and sexism is that, unlike blacks, women are not in the minority. Successful subordination often requires a majority, since an equally large class of victims may respond with equally effective counter-subordination. But even a non-majority group with disproportionate material and political power can still succeed in subordinating others. That men succeed in setting the terms for esteem seems to depend upon them having economic superiority - fact that must first exist for reasons status production does not explain. For an interesting game theory explanation, see Carol M. Rose, Women and Property: Gaining and Losing Ground, 78 VA. L. REV. 421, 429-54 (1992).
291. The model also provides a useful beginning for economic analysis of sexism. Like racial features, one's sex is reasonably observable. Men, like whites, often belong to exclusive socially connected groups. Racism and sexism each allow the oppressor to "feel superior; thus, a 'poor white' in the South can console himself with the thought that he is not a 'dirty nigger' . . . . [T]he most mediocre of males feels himself a demigod as compared with women." SIMONE DE BEAUVOIR, THE SECOND SEX at xxiv (H.M. Parshley ed. & trans., Alfred A. Knopf 1953) (1949). As Beauvoir notes, however, there are important differences between sexism and racism. See id. at xviii. Within the status theory, the most important distinction is the extent to which men and women depend on each other for their social status. Men typically belong to socially connected groups, such as families, that contain women. Given such relationships, a man who disparages women would seem to threaten his own status. The male's response to this problem is twofold. First, he claims the dominant position within the family relationship. Second, he seeks a means of distinguishing the females with whom he associates from the rest of the females of the world. Of course, status is obtained from other men in this manner only if it derives from some male consensus as to what constitutes desirable female behavior or attributes. Consequently, social norms arise within socially connected groups of men that define a standard of female behavior. Men then gain status by being associated with women who follow these roles well. Thus, men subordinate a group that includes family members by employing a rich ideology of gender roles that allows them to distinguish between their female relatives and other women - between "good" and "bad" women. Another key distinction between racism and sexism is that, unlike blacks, women are not in the minority. Successful subordination often requires a majority, since an equally large class of victims may respond with equally effective counter-subordination. But even a non-majority group with disproportionate material and political power can still succeed in subordinating others. That men succeed in setting the terms for esteem seems to depend upon them having economic superiority - fact that must first exist for reasons status production does not explain. For an interesting game theory explanation, see Carol M. Rose, Women and Property: Gaining and Losing Ground, 78 VA. L. REV. 421, 429-54 (1992).
292. But see Donohue, Title VII, supra note 120, at 1412; Donohue, Prohibiting Sex Discrimination, supra note 120, at 1347-48.
293. But see Donohue, Title VII, supra note 120, at 1412; Donohue, Prohibiting Sex Discrimination, supra note 120, at 1347-48.
294. 238 See id. at 43-45 ("[I]f the production function of each firm were linear and homogeneous, [market discrimination] would equal zero if at least one employer had . . . zero [taste for discrimination]."); POSNER, supra note 3, at 651.
295. There is significant evidence of persistent discrimination in competitive markets. See, e.g., ALICIA H. MUNNELL, LYNN E. BROWNE, JAMES MCENEANEY & GEOFFREY M.B. TOOTELL, MORTGAGE LENDING IN BOSTON: INTERPRETING HMDA DATA 2 (Federal Reserve Bank of Boston Working Paper No. 92-7, 1992) (finding that black mortgage applications in the Boston area were turned down 60% more often than applications from similarly situated whites); MARGERY A. TURNER, MICHAEL FIX & RAYMOND J. STRUYK, OPPORTUNITIES DENIED, OPPORTUNITIES DIMINISHED: RACIAL DISCRIMINATION IN HIRING 37-59 (1991) (reporting Urban Institute studies that found employment discrimination); Ian Ayres, Further Evidence of Discrimination in New Car Negotiations and Estimates of Its Cause 12-16 (1994) (unpublished manuscript, on file at the Harvard Law School Library) (finding that car dealers offered worse deals to black than to white test purchasers). Some commentators reject such studies on methodological grounds. See, e.g., EPSTEIN, supra note 116, at 51-58 (critiquing the Urban Institute and Ayres studies). I merely note that the intensity of one's methodological skepticism depends greatly on whether one believes that discrimination, in theory, can exist in competitive markets. If this section succeeds in demonstrating that it can, there will be less reason to doubt such studies. And to the extent that the methodologies are sound, the studies suggest the importance of the mechanisms I describe.
296. There is significant evidence of persistent discrimination in competitive markets. See, e.g., ALICIA H. MUNNELL, LYNN E. BROWNE, JAMES MCENEANEY & GEOFFREY M.B. TOOTELL, MORTGAGE LENDING IN BOSTON: INTERPRETING HMDA DATA 2 (Federal Reserve Bank of Boston Working Paper No. 92-7, 1992) (finding that black mortgage applications in the Boston area were turned down 60% more often than applications from similarly situated whites); MARGERY A. TURNER, MICHAEL FIX & RAYMOND J. STRUYK, OPPORTUNITIES DENIED, OPPORTUNITIES DIMINISHED: RACIAL DISCRIMINATION IN HIRING 37-59 (1991) (reporting Urban Institute studies that found employment discrimination); Ian Ayres, Further Evidence of Discrimination in New Car Negotiations and Estimates of Its Cause 12-16 (1994) (unpublished manuscript, on file at the Harvard Law School Library) (finding that car dealers offered worse deals to black than to white test purchasers). Some commentators reject such studies on methodological grounds. See, e.g., EPSTEIN, supra note 116, at 51-58 (critiquing the Urban Institute and Ayres studies). I merely note that the intensity of one's methodological skepticism depends greatly on whether one believes that discrimination, in theory, can exist in competitive markets. If this section succeeds in demonstrating that it can, there will be less reason to doubt such studies. And to the extent that the methodologies are sound, the studies suggest the importance of the mechanisms I describe.
297. See AKERLOF, supra note 83, at 35 (contrasting existing analysis, which assumes that "current transactions (so long as they are legal) do not result in changed relations with uninvolved parties in subsequent transactions"); see also id. at 71-72 (arguing for a model of custom-preserving equilibrium).
298. Id. at 35.
299. Id. at 36. Akerlof develops a formal model of caste equilibrium that applies to race discrimination. See id. at 37-41.
300. According to Richard A. Epstein, The Status-Production Sideshow: Why the Antidiscrimination Laws Are Still a Mistake, 108 HARV. L. REV. 1085, 1100-04 (1995), the history of Jim Crow demonstrates that discriminatory social norms require the support of violence. As he puts it, "[c]oercion is always the main event; status production is the side show." Id. at 1092. There are two problems with this analysis. First, I do not claim - no one in their right mind would - that social norms could produce the level of discrimination and segregation that existed in the Jim Crow South without support from private violence (and state law). Of course there is no "modern system" that shuns the use of violence but "is able to maintain anything similar to the caste-like structure of Jim Crow." Id. at 1100. The answer to Professor Epstein's challenge - "[w]hy would . . . boycotters have . . . to resort to violence if vilification" were available, id. at 1103 - is simply that violence secures a higher level of compliance with group norms than vilification alone. But that discrimination will be lower without violence does not, as Professor Epstein assumes, mean that discrimination will cease to exist. See supra pp. 1060-63; infra pp. 1068-71 (explaining how subtle forms of discrimination have replaced overt forms). Professor Epstein does not seriously challenge Part I's conclusion that withholding esteem works to induce material sacrifices for group welfare. (Although he argues that some successful civil rights boycotts used violence, see Epstein, supra, at 1103, he appears to concede that others did not, see id. at 1104-05.) Thus, social ostracism is, to use Professor Epstein's term, a "stick." Id. at 1093. Whether or not status punishments constitute what he means by "coercion," id. at 1092 n.22, the evidence shows they can deter free-riding. Second, Professor Epstein assumes without explanation that violence is no longer relevant to discriminatory social norms. He optimistically claims that "[e]ven if Title VII were repealed tomorrow, . . . [p]rivate violence by anyone against anyone else would still be sternly prohibited." Id. at 1099. But whether stern prohibitions, evenly enforced, will actually preclude violence and violent threats that support social norms is an empirical question. Not only does racial violence persist today despite legal prohibitions, see infra note 296, but, more generally, history shows that "no state has been able to prevent [the use of] violence to enforce rules of behavior." ELLICKSON, supra note 76, at 140 n.9. When violence is an option, the threat of violence also enforces norms. See id. at 213; see also id. at 58-59, 215-218 (describing violent self-help by ranchers in Shasta County, California); SALLY E. MERRY, URBAN DANGER 178-86 (1981) (listing examples of violence or threats of violence as ways of managing danger within a neighborhood); Donald Black, Crime as Social Control, 48 AM. SOC. REV. 34, 42 (1983) (developing a sociological theory of tomes of self-help).
301. According to Richard A. Epstein, The Status-Production Sideshow: Why the Antidiscrimination Laws Are Still a Mistake, 108 HARV. L. REV. 1085, 1100-04 (1995), the history of Jim Crow demonstrates that discriminatory social norms require the support of violence. As he puts it, "[c]oercion is always the main event; status production is the side show." Id. at 1092. There are two problems with this analysis. First, I do not claim - no one in their right mind would - that social norms could produce the level of discrimination and segregation that existed in the Jim Crow South without support from private violence (and state law). Of course there is no "modern system" that shuns the use of violence but "is able to maintain anything similar to the caste-like structure of Jim Crow." Id. at 1100. The answer to Professor Epstein's challenge - "[w]hy would . . . boycotters have . . . to resort to violence if vilification" were available, id. at 1103 - is simply that violence secures a higher level of compliance with group norms than vilification alone. But that discrimination will be lower without violence does not, as Professor Epstein assumes, mean that discrimination will cease to exist. See supra pp. 1060-63; infra pp. 1068-71 (explaining how subtle forms of discrimination have replaced overt forms). Professor Epstein does not seriously challenge Part I's conclusion that withholding esteem works to induce material sacrifices for group welfare. (Although he argues that some successful civil rights boycotts used violence, see Epstein, supra, at 1103, he appears to concede that others did not, see id. at 1104-05.) Thus, social ostracism is, to use Professor Epstein's term, a "stick." Id. at 1093. Whether or not status punishments constitute what he means by "coercion," id. at 1092 n.22, the evidence shows they can deter free-riding. Second, Professor Epstein assumes without explanation that violence is no longer relevant to discriminatory social norms. He optimistically claims that "[e]ven if Title VII were repealed tomorrow, . . . [p]rivate violence by anyone against anyone else would still be sternly prohibited." Id. at 1099. But whether stern prohibitions, evenly enforced, will actually preclude violence and violent threats that support social norms is an empirical question. Not only does racial violence persist today despite legal prohibitions, see infra note 296, but, more generally, history shows that "no state has been able to prevent [the use of] violence to enforce rules of behavior." ELLICKSON, supra note 76, at 140 n.9. When violence is an option, the threat of violence also enforces norms. See id. at 213; see also id. at 58-59, 215-218 (describing violent self-help by ranchers in Shasta County, California); SALLY E. MERRY, URBAN DANGER 178-86 (1981) (listing examples of violence or threats of violence as ways of managing danger within a neighborhood); Donald Black, Crime as Social Control, 48 AM. SOC. REV. 34, 42 (1983) (developing a sociological theory of tomes of self-help).
302. According to Richard A. Epstein, The Status-Production Sideshow: Why the Antidiscrimination Laws Are Still a Mistake, 108 HARV. L. REV. 1085, 1100-04 (1995), the history of Jim Crow demonstrates that discriminatory social norms require the support of violence. As he puts it, "[c]oercion is always the main event; status production is the side show." Id. at 1092. There are two problems with this analysis. First, I do not claim - no one in their right mind would - that social norms could produce the level of discrimination and segregation that existed in the Jim Crow South without support from private violence (and state law). Of course there is no "modern system" that shuns the use of violence but "is able to maintain anything similar to the caste-like structure of Jim Crow." Id. at 1100. The answer to Professor Epstein's challenge - "[w]hy would . . . boycotters have . . . to resort to violence if vilification" were available, id. at 1103 - is simply that violence secures a higher level of compliance with group norms than vilification alone. But that discrimination will be lower without violence does not, as Professor Epstein assumes, mean that discrimination will cease to exist. See supra pp. 1060-63; infra pp. 1068-71 (explaining how subtle forms of discrimination have replaced overt forms). Professor Epstein does not seriously challenge Part I's conclusion that withholding esteem works to induce material sacrifices for group welfare. (Although he argues that some successful civil rights boycotts used violence, see Epstein, supra, at 1103, he appears to concede that others did not, see id. at 1104-05.) Thus, social ostracism is, to use Professor Epstein's term, a "stick." Id. at 1093. Whether or not status punishments constitute what he means by "coercion," id. at 1092 n.22, the evidence shows they can deter free-riding. Second, Professor Epstein assumes without explanation that violence is no longer relevant to discriminatory social norms. He optimistically claims that "[e]ven if Title VII were repealed tomorrow, . . . [p]rivate violence by anyone against anyone else would still be sternly prohibited." Id. at 1099. But whether stern prohibitions, evenly enforced, will actually preclude violence and violent threats that support social norms is an empirical question. Not only does racial violence persist today despite legal prohibitions, see infra note 296, but, more generally, history shows that "no state has been able to prevent [the use of] violence to enforce rules of behavior." ELLICKSON, supra note 76, at 140 n.9. When violence is an option, the threat of violence also enforces norms. See id. at 213; see also id. at 58-59, 215-218 (describing violent self-help by ranchers in Shasta County, California); SALLY E. MERRY, URBAN DANGER 178-86 (1981) (listing examples of violence or threats of violence as ways of managing danger within a neighborhood); Donald Black, Crime as Social Control, 48 AM. SOC. REV. 34, 42 (1983) (developing a sociological theory of tomes of self-help).
303. Cf. EPSTEIN, supra note 116, at 29-31 (arguing that discrimination by a large majority need not harm a minority if sufficient opportunities among non-discriminators will remain for members of the minority).
304. See J. Mark Ramseyer & Minoru Nakazato, The Rational Litigant: Settlement Amounts and Verdict Rates in Japan, 18 J. LEGAL STUD. 263, 286-87 (1989).
305. See Akerlof, supra note 117, at 266.
306. This assumption is supported by relational contracts literature, which claims that firms in ongoing relationships acquire special knowledge about each other. See, e.g., Goetz & Scott, supra note 17, at 1100-01 ; Oliver E. Williamson, Transaction-Cost Economics: The Governance of Contractual Relations, 22 J.L. & ECON. 233, 240-42 (1979).
307. See Akerlof, supra note 117, at 266-67 (relying on the search model of Peter A. Diamond, Aggregate Demand Management in Search Equilibrium, 90 J. POL. ECON. 881, 882-86 (1982)).
308. Professor Epstein makes two arguments against this analysis. First, he says that it "assumes that boycotters keep their intentions secret." Epstein, supra note 243, at 1101. To the contrary, Akerlof assumes merely that there are search costs so that, for a given time period, the probability of finding a trading partner is less than one and varies with the number of traders in the relevant geographic area. See Akerlof, supra note 117, at 269. Consequently, it does not matter if boycotters are instantly recognized as such; when the density of traders in the area falls because some are boycotters, the time (and cost) required for searching necessarily increases. Second, Professor Epstein suggests that innovators can publicly identify themselves as willing to deal with targets of discrimination and that they can do so "without fear of coercive retaliation." Epstein, supra note 243, at 1102. Even if there is no violence, however, Akerlof demonstrates that it will not pay to innovate if sufficient numbers of traders are willing to boycott innovators. In the employment setting, for example, an innovating employer may lose valuable customers or employees. The analysis of Part I further suggests that the employer may lose the approval of fellow members of his social groups. If these costs outweigh the benefits, advertising oneself as a norm violator will merely ensure a negative return.
309. Akerlof, supra note 117, at 274.
310. See id. at 268 ("[U]nlike Becker's model, there is no excess profit to be made by entrepreneurs with no taste for discrimination. And entrepreneurs with a low taste for discrimination cannot profitably purchase the capital of those with a high taste for discrimination.").
311. See Jennifer Roback, Racism as Rent Seeking, 27 ECON. INQUIRY 661, 670 (1989) (explaining that Akerlof's model assumes that "the sanctions are costless to the individual imposing them"). Roback offers a model of racial norm enforcement: individuals "gain utility from conforming to [a] norm themselves," and "from knowing that other people are conforming to the norm. The utility gain may be a gain in pure psychic income, if, for example, employers gain satisfaction from knowing that blacks are being kept properly 'in their place.'" Id. at 665 (foot-note omitted). Her model thus describes a benefit to boycottera to offset the costs of the boycott. But Roback merely posits that such gains exist. She does not explain why people would "gain utility" from conforming to or observing other people conform to a norm.
312. See supra p. 1045.
313. See supra pp. 1046-48.
314. Professor Epstein says I neglect how the mechanism of intra-group esteem allocation makes forces opposing discrimination more tenacious. See Epstein, supra note 243, at 1106. To the contrary, I pointed out that the mechanism is necessary to explain the behavior of participants in the civil rights movement that produced Title VII, behavior that economic critics of Title VII - ironically - have never addressed. See supra p. 1018. Opposition to discrimination is more tenacious than was previously understood only because previous economic models could not explain why any opposition existed. Notwithstanding this tenacity, there is no reason to suppose that less populous and poorer groups can completely eliminate subordination by the dominant group. That opponents of discrimination sometimes succeed in using esteem allocation to engage in counter-boycotts and other defensive measures is precisely the reason I argue that race discrimination is a market failure. See infra pp. 1074-78. Unless one side has such an overwhelming advantage that the other forgoes all competition for status, the endless cycle of investing and counter-investing in zero-sum status production is inefficient.
315. See supra notes 206-210 and accompanying text.
316. See supra pp. 1032, 1060-62.
317. I do not refer to highly contested definitions of ultra-subtle discrimination. In American society, discrimination is sufficiently subtle if it refrains from overt reliance on skin color alone. See, e.g., Samuel L. Gaertner & John F. Dovidio, The Aversive Form of Racism, in PREJUDICE, DISCRIMINATION, AND RACISM 61, 77-78, 84-86 (John F. Dovidio & Samuel L. Gaertner eds., 1986); Joleen Kirschenman & Kathryn M. Neckerman, "We'd Love to Hire Them, But . . .": The Meaning of Race for Employers, in THE URBAN UNDERCLASS 203, 230-31 (Christopher Jencks & Paul E. Peterson eds., 1991). For example, a white employer may consider hiring a black worker if, but only if, she stands in stark contrast to all the stereotypes the employer holds of blacks. The employer could genuinely espouse a belief in equal opportunity, but also believe blacks are frequently (but not inherently) indolent, obtuse, or quarrelsome. Such an employer might parade her decision to hire a black worker who is manifestly none of those things to prove her belief in equality, but nonetheless think that any blacks who are not model employees fall into one of the negative categories. Thus, modern discrimination is less derogatory, and therefore less productive of status, than it used to be. But that is no reason to assume that the discriminatory norm must be unraveling. There is, after all, less material cost associated with employing such a norm, since it permits the most materially beneficial trades with blacks (like hiring the manifestly qualified black), refusal of which would be branded as racist. It is thus easier to understand how norms like "never take a chance on a black" or "all things being equal, hire the white" could survive market competition. The status benefit is less than with overt discrimination, but so is the material cost.
318. I do not refer to highly contested definitions of ultra-subtle discrimination. In American society, discrimination is sufficiently subtle if it refrains from overt reliance on skin color alone. See, e.g., Samuel L. Gaertner & John F. Dovidio, The Aversive Form of Racism, in PREJUDICE, DISCRIMINATION, AND RACISM 61, 77-78, 84-86 (John F. Dovidio & Samuel L. Gaertner eds., 1986); Joleen Kirschenman & Kathryn M. Neckerman, "We'd Love to Hire Them, But . . .": The Meaning of Race for Employers, in THE URBAN UNDERCLASS 203, 230-31 (Christopher Jencks & Paul E. Peterson eds., 1991). For example, a white employer may consider hiring a black worker if, but only if, she stands in stark contrast to all the stereotypes the employer holds of blacks. The employer could genuinely espouse a belief in equal opportunity, but also believe blacks are frequently (but not inherently) indolent, obtuse, or quarrelsome. Such an employer might parade her decision to hire a black worker who is manifestly none of those things to prove her belief in equality, but nonetheless think that any blacks who are not model employees fall into one of the negative categories. Thus, modern discrimination is less derogatory, and therefore less productive of status, than it used to be. But that is no reason to assume that the discriminatory norm must be unraveling. There is, after all, less material cost associated with employing such a norm, since it permits the most materially beneficial trades with blacks (like hiring the manifestly qualified black), refusal of which would be branded as racist. It is thus easier to understand how norms like "never take a chance on a black" or "all things being equal, hire the white" could survive market competition. The status benefit is less than with overt discrimination, but so is the material cost.
319. See ELLICKSON, supra note 76, at 167.
320. See ELSTER, supra note 11, at 111; Cooter, supra note 83, at 230 n.48. Both believe there are norms that arise among strangers, although, as Cooter notes, the process by which this occurs "requires explaining." Id. at 231 n48.
321. I make this argument below at pp. 1074-78.
322. The analogy to conspicuous consumption may again be helpful. Existing norms may condemn especially ostentatious consumption because such behavior is inefficient, yet because these norms arise essentially between strangers, they are weak and do not deter more subtle forms of consumption, which are more easily rationalized as something other than conspicuous consumption. See McAdams, supra note 59, at 42-44, 79-80.
323. See Cooter, supra note 196, at 166.
324. See id. at 153, 155-156. Thus, Cooter seems to view discrimination as a productive activrty, but one productive of material ends.
325. See id. at 154 (citing Jennifer Roback, Racism as Rent Seeking, 27 ECON. INQUIRY 661 (1989)).
326. See supra pp. 1027-29.
327. Posner, supra note 222, at 10.
328. See id. at 11. Posner also expresses doubt that ideology by itself is sufficient for guilds to survive and argues that government restrictions on competition are essential. See id. at 8 (noting that guilds typically have government-granted power to exclude competitors because "prohibiting entry is necessary if the guild is to have supracompetitive profits"). Posner, like Cooter, also considers only guilds whose ends and means are material.
329. See id. at 19.
330. An example of political organizing by whites is the Citizens' Council, which operated in various southern states in the 1950s and 1960s in opposition to integration. "By the second anniversary of the public school decisions . . . 'protective societies' accounted for a considerable portion of the energies of perhaps as many as 250,000 to 300,000 people from virtually every station of southern life. . . . [N]one was more powerful than the Citizens' Council . . . ." MCMILLEN, supra note 229, at 11.
331. Many commentators have decried the role of the "old-boy network" in perpetuating discrimination against women and minorities. See, e.g., Michael M. Burns, The Exclusion of Women From Influential Men's Clubs: The Inner Sanctum and the Myth of Pull Equality, 18 HARV. C.R.-C.L. L. REV. 321, 322-23 (1983).
332. See AXELROD, supra note 13, at 13-14, 59.
333. See id. at 126-32.
334. See Burns, supra note 271, at 325-34.
335. Epstein agrees that reciprocity prevents the market from eliminating discrimination, but argues that it is efficient that such discrimination remain because members of a particular racial or ethnic group may be able to elicit greater cooperation from members of their own group than from non-members. For single-race firms: The party who cheats at work now knows that he faces stricter sanctions, given the strong likelihood that the information will be brought home to him at play, at church, or in other business and social settings. The complex network of human interactions thus induces persons to honor their deals. EPSTEIN, supra note 116, at 70; cf. Landa, supra note 177, at 361-62 (arguing that socially and ethnically homogeneous trading groups have a comparative advantage in economies without reliable legal infrastructure to enforce contracts). Whether such hiring discrimination is actually efficient depends, however, on why the underlying segregation - "at play, at church, or in other business and social settings" - exists. Epstein is correct to note that, regardless of the reasons for the underlying segregation, this form of discrimination will persist: for any particular firm, cooperation is easier to achieve among employees who socialize together.
336. See supra note 110 and accompanying text.
337. See supra pp. 1060-62.
338. See BECKER, supra note 114, at 39-42; see also Cooter, supra note 196, at 160 ("Competition can teach a sharp lesson to businesses that rely upon false signals.").
339. Although marginal productivities may vary continuously so that actual ties are extremely rare, an employer will invest only limited resources into determining the productivity of job applicants. If employers invest only enough to place applicants in one of a few ability categories, then ties could be common.
340. By ignoring the issue, they implicitly assume that ties never occur. See, e.g., EPSTEIN, supra note 116, at 41-44.
341. See sources cited supra note 119. But see Donohue, Prohibiting Sex Discrimination, supra note 120, at 1347-48; Donohue, Title VII, supra note 120, at 1423-31.
342. See sources cited supra note 119. But see Donohue, Prohibiting Sex Discrimination, supra note 120, at 1347-48; Donohue, Title VII, supra note 120, at 1423-31.
343. See, e.g., Fred S. McChesney, Boxed In: Economists and Benefits From Crime, 13 INT'L REV. L. & ECON. 225, 227-28 (1993); Richard A. Posner, An Economic Theory of the Criminal Law, 85 COLUM. L. REV. 1193, 1196, 1198 (1985); Gordon Tullock, The Welfare Costs of Tariffs, Monopolies and Theft, 5 W. ECON. J. 224, 228-31 (1967).
344. See, e.g., Fred S. McChesney, Boxed In: Economists and Benefits From Crime, 13 INT'L REV. L. & ECON. 225, 227-28 (1993); Richard A. Posner, An Economic Theory of the Criminal Law, 85 COLUM. L. REV. 1193, 1196, 1198 (1985); Gordon Tullock, The Welfare Costs of Tariffs, Monopolies and Theft, 5 W. ECON. J. 224, 228-31 (1967).
345. See, e.g., Fred S. McChesney, Boxed In: Economists and Benefits From Crime, 13 INT'L REV. L. & ECON. 225, 227-28 (1993); Richard A. Posner, An Economic Theory of the Criminal Law, 85 COLUM. L. REV. 1193, 1196, 1198 (1985); Gordon Tullock, The Welfare Costs of Tariffs, Monopolies and Theft, 5 W. ECON. J. 224, 228-31 (1967).
346. One might argue that the transfer often increases wealth because the thief gains more than the owner loses. Yet because economics measures value by willingness to pay, the opposite is likely to be the case. When transaction costs are low, prohibiting theft still permits property transfers through voluntary market exchange. The only transfers prevented by criminalizing theft are, therefore, those in which the thief is not willing to pay the owner enough to induce a voluntary exchange - where, in other words, the thief "values" the good as much as or less than the owner. See POSNER, supra note 3, at 208 & n.3; Posner, supra note 282, at 1196. Preventing transfers when the thief values the good less than the owner does is efficient. Economists do not, however, rely on this argument as much as the one in the text, perhaps
347. Tullock, supra note 282, at 231.
348. THOMAS HOBBES, LEVIATHAN 88-89 (Richard Tuck ed., Cambridge Univ. Press 1991) (1651); see also EPSTEIN, supra note 116, at 15-19 (describing Hobbes's argument for prohibitions on force and fraud).
349. EPSTEIN, supra note 116.
350. Cf. Williams, supra note 178, at 147 (referring to acts of racism as the "appropriation of psychic property").
351. Hobbes observes: For every man looketh that his companion should value him, at the same rate he sets upon himselfe: And upon all signes of Contempt, or undervaluing, naturally endeavours, as far as he dares (which amongst them that have no common power to keep them in quiet, is far enough to make them destroy each other), to extort a greater value from his contemners, by dommage; and from others, by the example. HOBBES, supra note 285, at 88. For a more detailed criticism of Epstein's use of Hobbes in Forbidden Grounds, see Richard H. McAdams, Epstein on His Own Grounds, 31 SAN DIEGO L. REV. 241, 242-49 (1994).
352. HUME, supra note 9, at 55.
353. See supra notes 193-195 and accompanying text.
354. See, e.g., Melvin L. Oliver, James H. Johnson, Jr. & Walter C. Farrell, Jr., Anatomy of a Rebellion: A Political-Economic Analysis, in READING RODNEY KING: READING URBAN UPRIS-ING 117, 120-21 (Robert Gooding-Williams ed., 1993). For a personal narrative conveying how Paragement leads to rage, see Williams, cited above in note 178, at 127-29.
355. See supra p. 1031.
356. This argument is conservative in that it omits the claim that the loss to blacks exceeds the gain to whites. Following the alternative economic argument against theft, see supra note 283, one could observe that, like thieves, discriminators do not pay their victims for their coerced gain. Prohibiting discrimination still permits voluntary market transactions in status; the only transfers prevented are ones where whites value their status gain as much as or less than blacks value the status loss. Second, though equating value with willingness to pay is controversial in the context of theft, utilitarian concerns support this efficiency analysis in the context of race discrimination. The argument is that the psychological damage to blacks exceeds the status pleasure for whites. Cf. WILLIAM M. GRIER & PRICE M. COBBS, BLACK RAGE 30-33 (1968) (discussing the psychologically devastating legacy of black subordination); ABRAM KARDINER & LIONEL OVESEY, THE MARK OF OPPRESSION: A PSYCHOSOCIAL STUDY OF THE AMERICAN NEGRO 379 (1951) (emphasizing white psychological gains by describing "the degradation of the Negro's status" during slavery as "the extreme manifestation of the ego perversion of dominance"); Chester M. Pierce, Psychiatric Problems of the Black Minority, in 2 AMERICAN HANDBOOK OF PSYCHIATRY 512, 513-18 (Silvano Arieti ed., 2d ed. 1974) (suggesting that American society's insistence that "white skin color is superior to black skin color" comes at the expense of black mental health).
357. This argument is conservative in that it omits the claim that the loss to blacks exceeds the gain to whites. Following the alternative economic argument against theft, see supra note 283, one could observe that, like thieves, discriminators do not pay their victims for their coerced gain. Prohibiting discrimination still permits voluntary market transactions in status; the only transfers prevented are ones where whites value their status gain as much as or less than blacks value the status loss. Second, though equating value with willingness to pay is controversial in the context of theft, utilitarian concerns support this efficiency analysis in the context of race discrimination. The argument is that the psychological damage to blacks exceeds the status pleasure for whites. Cf. WILLIAM M. GRIER & PRICE M. COBBS, BLACK RAGE 30-33 (1968) (discussing the psychologically devastating legacy of black subordination); ABRAM KARDINER & LIONEL OVESEY, THE MARK OF OPPRESSION: A PSYCHOSOCIAL STUDY OF THE AMERICAN NEGRO 379 (1951) (emphasizing white psychological gains by describing "the degradation of the Negro's status" during slavery as "the extreme manifestation of the ego perversion of dominance"); Chester M. Pierce, Psychiatric Problems of the Black Minority, in 2 AMERICAN HANDBOOK OF PSYCHIATRY 512, 513-18 (Silvano Arieti ed., 2d ed. 1974) (suggesting that American society's insistence that "white skin color is superior to black skin color" comes at the expense of black mental health).
358. This argument is conservative in that it omits the claim that the loss to blacks exceeds the gain to whites. Following the alternative economic argument against theft, see supra note 283, one could observe that, like thieves, discriminators do not pay their victims for their coerced gain. Prohibiting discrimination still permits voluntary market transactions in status; the only transfers prevented are ones where whites value their status gain as much as or less than blacks value the status loss. Second, though equating value with willingness to pay is controversial in the context of theft, utilitarian concerns support this efficiency analysis in the context of race discrimination. The argument is that the psychological damage to blacks exceeds the status pleasure for whites. Cf. WILLIAM M. GRIER & PRICE M. COBBS, BLACK RAGE 30-33 (1968) (discussing the psychologically devastating legacy of black subordination); ABRAM KARDINER & LIONEL OVESEY, THE MARK OF OPPRESSION: A PSYCHOSOCIAL STUDY OF THE AMERICAN NEGRO 379 (1951) (emphasizing white psychological gains by describing "the degradation of the Negro's status" during slavery as "the extreme manifestation of the ego perversion of dominance"); Chester M. Pierce, Psychiatric Problems of the Black Minority, in 2 AMERICAN HANDBOOK OF PSYCHIATRY 512, 513-18 (Silvano Arieti ed., 2d ed. 1974) (suggesting that American society's insistence that "white skin color is superior to black skin color" comes at the expense of black mental health).
359. Black separatism necessarily has the consequence of refusing some or all economic trades with whites. But separatism may be intended not as an counter-insult, but as a defensive means of achieving psychological independence from white subordination. See supra note 205.
360. Thus, an initially non-violent subordination provokes a non-violent response, but when the subordinator's efforts are frustrated by the victim's "standing her ground," there may be an escalation to violence. A number of race riots fit this pattern. See generally 9 RACE, LAW, AND AMERICAN HISTORY 1700-1990: LYNCHING, RACIAL VIOLENCE, AND LAW passim (Paul Finkelman ed., 1992) (illustrating how violence, riots, and lynching supported segregation).
361. See JOE R. FEAGIN, RACIAL AND ETHNIC RELATIONS 376 (1978) ("Open and violent interethnic conflict has been a crucial current in American history. Earlier immigrant groups have regularly attempted to subordinate later groups."); see also JACK LEVIN & JACK MCDEVITT, HATE CRIMES 137-48 (1993) (recounting racial clashes between minority groups, especially other minorities versus blacks); ROEDIGER, supra note 163, at 133-63 (relating the history of conflict between Irish immigrants and African Americans); Sumi K. Cho, Korean Americans vs. African Americans: Conflict and Construction, in READING RODNEY KING: READING URBAN UPRISING, supra note 291, at 196, 196-204 (explaining the sources of discrimination and prejudice between Korean Americans and African Americans). The conflict between the Jewish and African American communities is a timely and obvious example of this problem.
362. See JOE R. FEAGIN, RACIAL AND ETHNIC RELATIONS 376 (1978) ("Open and violent interethnic conflict has been a crucial current in American history. Earlier immigrant groups have regularly attempted to subordinate later groups."); see also JACK LEVIN & JACK MCDEVITT, HATE CRIMES 137-48 (1993) (recounting racial clashes between minority groups, especially other minorities versus blacks); ROEDIGER, supra note 163, at 133-63 (relating the history of conflict between Irish immigrants and African Americans); Sumi K. Cho, Korean Americans vs. African Americans: Conflict and Construction, in READING RODNEY KING: READING URBAN UPRISING, supra note 291, at 196, 196-204 (explaining the sources of discrimination and prejudice between Korean Americans and African Americans). The conflict between the Jewish and African American communities is a timely and obvious example of this problem.
363. See JOE R. FEAGIN, RACIAL AND ETHNIC RELATIONS 376 (1978) ("Open and violent interethnic conflict has been a crucial current in American history. Earlier immigrant groups have regularly attempted to subordinate later groups."); see also JACK LEVIN & JACK MCDEVITT, HATE CRIMES 137-48 (1993) (recounting racial clashes between minority groups, especially other minorities versus blacks); ROEDIGER, supra note 163, at 133-63 (relating the history of conflict between Irish immigrants and African Americans); Sumi K. Cho, Korean Americans vs. African Americans: Conflict and Construction, in READING RODNEY KING: READING URBAN UPRISING, supra note 291, at 196, 196-204 (explaining the sources of discrimination and prejudice between Korean Americans and African Americans). The conflict between the Jewish and African American communities is a timely and obvious example of this problem.
364. Black feminists, for example, have described how women of color have been subordinated not only by white men, but by black men and white women. See, e.g., E. Frances White, Listening to the Voices of Black Feminism, 18 RADICAL AM. 7, 11-17 (1984).
365. See COLLINS, supra note 128, at 79-82; DOLLARD, supra note 125, at 68-71; Kristal B. Zook, Light Skinned-ded Naps, in MAKING FACE, MAKING SOUL: CREATIVE AND CRITICAL PERPECTIVES BY WOMEN OF COLOR 85, 85-96 (Gloria Anzaldúa ed., 1990).
366. See CLARK, supra note 129, at 70 ("The Negro male was, therefore, driven to seek status in ways which seemed . . . antisocial, escapist, [or] socially irresponsible.").
367. See, e.g., Janita Poe, A Black-On-Black Education Barrier, CHI. TRIB., Jan. 1, 1994, § 1, at
368. See, e.g., KENNETH B. CLARK, PREJUDICE AND YOUR CHILD 49-51 (1955).
369. In fact, the very inefficiency of this system may explain the existence of an anti-discrimination norm or counter-norm. Because the counter-norm arises between strangers, however, it is not likely to cancel the effect of stronger discriminatory norms that arise within and between socially connected groups. See supra p. 1069.
370. People can concentrate, for example, on raising their group status by focusing on the positive achievements of their group rather than by derogation of other groups or by elevating the status of groups whose welfare does not inherently conflict with other groups. Similarly, people can decide to seek increased intra-group esteem through positive achievement or through ways that do not inherently conflict with the preferences of other group members.
371. One might view hate speech in this manner, as potential discriminators attempt to subvert verbally those protected by anti-discrimination laws.
372. See George P. Fletcher, The Metamorphosis of Larceny, 89 HARV. L. REV. 469, 469 n.1, 496 (1976).
373. See supra p. 1072. If blacks are already disproportionately poor in society, partitioning of white and black neighborhoods will harm blacks more than it harms whites.
374. Liability works directly to raise the costs of subordination but also works indirectly. Those who enforce discriminatory norms by threats or the use of low-level violence now find their gets are - given civil liability - harder to intimidate. Norm enforcers could use higher levels of violence, but criminal sanctions more effectively detect and deter such immoderate violence.
375. The COst one bears for subordinating one's own group does not completely eliminate this strategy. One may, for example, be willing to raise the status of a group in which one has very high intra-group status at the expense of lowering the status of a group in which one has only average intra-group status - especially if one has few alternative options for producing esteem. Racial minorities who denigrate members of their own race and women who demean other women have made precisely such a choice. See supra notes 298-299 and accompanying text.
376. The corollary is that individuals join groups that do not act to subordinate each other, or leave one of the groups when such a status conflict arises. Yet sometimes the cost of avoiding membership in either of two such groups outweighs the benefits of avoiding the status conflict. In such cases, the individual is less likely to invest in subordination of either group than those who belong only to one of the two groups, and may invest in restraining other members from investing in the subordination strategy.
377. See, e.g., David H. Swinton, The Economic Status of African Americans: "Permanent" Poverty and Inequality, in THE STATE OF BLACK AMERICA 1991, at 25, 25-65 (Janet Dewart ed., 1991).
378. My argument here is not that "contact" between the races decreases prejudice. That claim is the subject of extended discussion and debate. See, e.g., Norman Miller & Marilynn B. Brewer, The Social Psychology of Desegregation: An Introduction, in GROUPS IN CONTACT: THE PSYCHOLOGY OF DESEGREGATION, supra note 227, at 1, 2-4. To the contrary, the status-production model contemplates that people can obtain status by subordinating those with whom they have frequent contact. My point is that cross-membership makes subordination, by either group against the other, less productive of status.
379. See, e.g., Donohue, Prohibiting Sex Discrimination, supra note 120, at 1338-39 & n.6; Posner, Title VII, supra note 119, at 520. Indeed, surveys suggest a profound decrease in the number of whites who deny the principles of equal opportunity. See supra note 214. See generally Kenneth Dau-Schmidt, Legal Prohibitions as More Than Prices: Preference-Shaping Policies Under the Law, in NEW AND CRITICAL PERSPECTIVES IN LAW AND ECONOMICS (Robin Malloy & Christopher Braun eds., forthcoming 1995) (discussing Title VII as one example of laws that attempt to change "invidious preferences").
380. See, e.g., Donohue, Prohibiting Sex Discrimination, supra note 120, at 1338-39 & n.6; Posner, Title VII, supra note 119, at 520. Indeed, surveys suggest a profound decrease in the number of whites who deny the principles of equal opportunity. See supra note 214. See generally Kenneth Dau-Schmidt, Legal Prohibitions as More Than Prices: Preference-Shaping Policies Under the Law, in NEW AND CRITICAL PERSPECTIVES IN LAW AND ECONOMICS (Robin Malloy & Christopher Braun eds., forthcoming 1995) (discussing Title VII as one example of laws that attempt to change "invidious preferences").
381. See, e.g., Donohue, Prohibiting Sex Discrimination, supra note 120, at 1338-39 & n.6; Posner, Title VII, supra note 119, at 520. Indeed, surveys suggest a profound decrease in the number of whites who deny the principles of equal opportunity. See supra note 214. See generally Kenneth Dau-Schmidt, Legal Prohibitions as More Than Prices: Preference-Shaping Policies Under the Law, in NEW AND CRITICAL PERSPECTIVES IN LAW AND ECONOMICS (Robin Malloy & Christopher Braun eds., forthcoming 1995) (discussing Title VII as one example of laws that attempt to change "invidious preferences").
382. See supra pp. 1059-60.
383. See EPSTEIN, supra note 116, at 242-66 (providing a comprehensive description of such costs). For an attempt to measure them, see John J. Donohue III, Advocacy Versus Analysis in Assessing Employment Discrimination Law, 44 STAN. L. REV. 1583, 1599-1603 (1992) (reviewing RICHARD A. EPSTEIN, FORBIDDEN GROUNDS (1992)). Donohue also attempts to calculate the benefits of Title VII, but not along the lines suggested here. See id. at 1603-07.
384. See EPSTEIN, supra note 116, at 242-66 (providing a comprehensive description of such costs). For an attempt to measure them, see John J. Donohue III, Advocacy Versus Analysis in Assessing Employment Discrimination Law, 44 STAN. L. REV. 1583, 1599-1603 (1992) (reviewing RICHARD A. EPSTEIN, FORBIDDEN GROUNDS (1992)). Donohue also attempts to calculate the benefits of Title VII, but not along the lines suggested here. See id. at 1603-07.
385. See Terry Eastland, The Case Against Affirmative Action, 34 WM. & MARY L. REV. 33, 46 (1992); William Van Alstyne, Rites of Passage: Race, the Supreme Court, and the Constitution, 46 U. CHI. L. REV. 775, 809 (1979).
386. See Terry Eastland, The Case Against Affirmative Action, 34 WM. & MARY L. REV. 33, 46 (1992); William Van Alstyne, Rites of Passage: Race, the Supreme Court, and the Constitution, 46 U. CHI. L. REV. 775, 809 (1979).
387. See supra p. 1080.
388. See HOBBES supra note 285, at 88-89.
389. See Ayres, supra note 17, at 320-22.
390. See ELLICKSON, supra note 76, at 46, 167-69.
391. See COASE, supra note 6, at 33-55. Firms lower the transaction costs of production. See id. at 6-7. Chess-clubs lower the transaction costs for people who are looking for a game of chess.