1. These phenomena—the interaction of nonlegal and legal controls—are of recurring interest to legal academics; most recently, see, e.g., Special Issue, Mediting Institutions: Beyond the Public/Private Distinction, 61 U. Chi. L. Rev. 1213 (1994)Google Scholar; Symposium, , Law, Economics, and Norms, 144 U. Pa. I. Rev. 16–13 (1996).Google Scholar

2. On the importance of religious behavior in American life, see the statistics and case studies in Gallup, G. Jr., and Jones, S., 100 Questions and Answers: Religion in America (1989)Google Scholar; Roof, W. C. and McKinney, W., American Mainline Religion (1987)Google Scholar; Caplow, T. et al. , All Faithful People: Change and Continuityin Middletown's Religion (1983)Google Scholar; Bellah, R. N. et al. , Habits of the Heart: Individualism and Commitment in American Life (1985)Google Scholar; Biddle, J. E., Religious OrganizationsGoogle Scholar, in Who Benefits from the Nonprofit Sector? (Clotfelter, C.T. ed. 1992).Google Scholar

3. The liierature is too voluminous to cite, but a recent collection of articles can be found in Symposium, , Religion in Public Life: Access, Accommodation, anil Accountability, 60 Geo. Wash. L, Rev. 599 (1992).Google Scholar I discuss some particular contributions, infra.

4. Their rare claims about the behavioral consequences of legal action are rarely persuasive, indeed, are impossible to evaluate. “In the aggregate … it seems likely thai adoption of [the ‘no endorsement’ principle] would create incentives that would intensify religious conflicts.” Smith, S. D., Symbols, Perception, and Doctrinal Illusions: Establishment Neutrality and the “No Endorse-ment” Test, 86 Mich. L. Rev. 266, 305 (1987).CrossRefGoogle Scholar “[R]eligion-specific policies promote the concept of religion as an interest, looking after the tangible concerns of its members, rather than acting as a force for moral good.” Lupu, I. C., Reconstructing the Establishment Clause: The Case Against Discretionary Accommodation of Religion, 140 U. Pa. L. Rev. 555, 597 (1991).CrossRefGoogle Scholar “[C]onfrontations between government and religious institutions may force both institutions to think in more sustained anil focused ways about their own commitments.” Tushnet, M., The Rhetoric of the Free Exercise Doctrine, 1993 Byu L. Rev. 117, 130.Google Scholar If the Supreme Court adopted the doctrine of formal neutrality and legislatures did not produce accommodations, “religious practice would be pressured toward homogeneity by the sheer pervasiveness of modern regulation.” Laycock, D.. The Remnants of Free Exercise, 1990 Sup. Ct. Rev. 1, 56.Google Scholar State designation of an official religious faith would produce “caste-based factions.” Eisgruber, C L., Political Unity and the Power of Government, 41 UCLA L, Rev. 130–1, 1312 (1994).Google Scholar

5. See Lemon, v. Kurtzman, , 403 LUS. 602 (1971).Google Scholar

6. 494 U.S. 872 (1990).

7. see Sherbert, v. Verner, , 374 U.S. 398 (1963).Google Scholar

8. The paper takes a rational choice approach. See, e.g., Iannaccone, L. R., A Formal Model of Church and Sect, 94 Amer.J. Soc. 241 (1988).CrossRefGoogle Scholar Economic discussions of religion go back to Adam Smith. See Smith, A., The Wealth of Nations 793ff. (1976 ed.)Google Scholar; see also Anderson, G. M.. Mr. Smith and the Preachers: The Economics of Religion in the Wealth of Nations, 96 J. Pol. Econ. 1066 (1988).CrossRefGoogle Scholar The only paper on religious groups in the law and economics literature is McConncll, M. W. and Posner, R. A., An Economic Approach to Issues of Religious Freedom, 56 U. Chi. L. Rev. 1 (1989).CrossRefGoogle Scholar The rational choice approach can be contrasted with various, more complex sociological theories, which generally emphasize social structure and social conditions over individual choice; see Wuthnow, R. J., Sociology of ReligionGoogle Scholar, in Handbook of Sociology (Smelser, N.J. ed. 1988Google Scholar). For a critique of the rational choice approach, see Bruce, S., Religion and Rational Choice: A Critique of Economic Explanations of Religious behavior, 54 Soc. Rev 193 (1993).Google Scholar

The contribution of this article to the economic literature on religions groups consists of its attempt to incorporate legal questions into the analysis in a systematic way. Most of the existing literature focuses on such questions as whether religious groups are efficient, what private or social purposes they might serve, and how they overcome problems of cooperation and coordination. Aside from some articles on establishment and aside from a few passing references to legal issues, which will be discussed infra, the articles do not say much about legal regulation.

9. The importance of recognizing the problem of collective action in analyzing religious groups is noted by Wallis, J. L., Modelling Churches as Collective Action Groups, 17 Intert. J. Soc. Econ. 59 (1990)CrossRefGoogle Scholar; see also Finke, R. and Stark, R., The Churching, of America, 1776–1900: Winners and Losers in Our Religious Economy 252 (1992).Google Scholar

10. Compare the definition in Hargrove, B., The Sociology of Religion 29 (2d ed. 1989)Google Scholar; for a legal and philosophical discussion, see Grcenawalt, K., Religion as a Concept in Constitutional Law, 72 Calif. L. Rev. 753 (1984).CrossRefGoogle Scholar

11. See Posner, E. A., The Legal Regulation of Groups: The Influence of Legal and Nonlegal Sanctions on Collective Action, 63 U. Chi. I. Rev. 133 (1996).CrossRefGoogle Scholar

12. Of course, the spiritual benefits can be characterized in other ways. The classic work is James, W., The Variety of Religious Experience (1958 ed.).Google Scholar For a useful discussion, see Stark, R. and Glock, C. Y., American Piety: The Nature of Religious Commitment 125–28 (1970)Google Scholar (“all religious experiences, from the vaguest to the most frenzied, constitute occasions defined by those undergoing them as an encounter—some sense of contact—between themselves and some supernatural consciousness” (p. 126, emphasis omitted)); and for examples see Bellah, et al. , supra note 1, at 227–43.Google Scholar A survey of psychological theories can be found in Argyle, M. and Beit-Hallahmi, B., The Social Psychology of Religion 178ff. (1975 ed.)Google Scholar; see also Roberts, K. A., Religion in Sociological Perspective 90–90 (1984).Google Scholar The thrust of these theories is that religion responds to a cognitive need for a coherent explanation of phenomena, or to an emotional need for relief from anxiety or guilt caused by failures to meet social standards or personal ambitions.

13. See Iannaccone, L. R., Religious Markets and the Economics of Religion, 39 Social Compass 123, 125–26 (1992)CrossRefGoogle Scholar (analyzing religious goods as “experience goods”); Hull, B. B. and Bold, F., Towards an Economic Theory of the Church, 16 Inter'l J. Soc. Econ. 5, 10–12 (1989)CrossRefGoogle Scholar (analyzing religious goods as “credence goods”). Both articles make the point that the value of spiritual benefits depends to an unusual extent on their credibility.

14. For discussions, see Elster, J., Ulysses and the Sirens 65ff. (1979)Google Scholar (discussing inconsistent time preferences and weakness of will);Elster, J., Sour Grapes 43ff. (1983)Google Scholar (discussing behavior that cannot be directly willed).

15. “The law has grasped the paradox that freedom should include the freedom to enter into enforceable contracts; it seems to overlook the need that people often have, and perhaps the right they should have, to constrain their own behavior for their own good.” Schelling, T. C., Choice and Consequence 103 (1984)Google Scholar; see generally id. at 57–112. The best explanation, as Schelling acknowledges, is that courts seem unsuited for the difficult, information-sensitive task of forcing people to adhere to personal commitments. But that docs not bar the state from playing a useful role. If self-commitment is a good, a utilitarian state should promote it, and it can by subsidizing institutions, such as religious groups, that enforce self-commitment more sensitively and accurately than courts can.

16. Religious conversion is more effective at curing alcoholism than medical therapies. See Reynolds, V. and Tanner, R., The Social Ecology of Religion 299 (1995).Google Scholar

17. This is a theme in Garet, R. R., Dancing to Music: An Interpretation of Mutuality, 80 Ky. L.J. 893 (1991–1992)Google Scholar; Garet, R. R., Communality and Existence: The Rights of Groups, 56 S. Cal L. Rev. 1001, 1005–75 (1983).Google ScholarSee also Varenne, H., Americans Together 103–06 (1977)Google Scholar: id. at 104 (respondents to a Methodist survey said that they join a church because “it has a ‘good’ minister, a ‘nice’ church building, a ‘congenial’ congregation” (emphasis added)). See also id. at 115–21 (describing the community-generating aspects of a Protestant religious service).

18. For a discussion of the literature on the relationship between religious affiliation and deviant behavior. See Cochran, J. K. and Akers, R. L., Beyond Hellfire: An Exploration of the Variable Effects of Religiosity on Adolescent Marijuana and Alcohol Use, 26 J. Res. Crime and Delinquency 198 (1989).CrossRefGoogle Scholar This study finds statistical support for the hypothesis that religious youth use marijuana and alcohol less than irreligious youth; more generally, the authors conclude that “religiosity is inversely related to delinquent behavior.” Id. at 221. See also Welch, M. R., Tittle, C. R., and Petee, T., Religion and Deviance among Adult Catholics: A Test of the “Moral Communities” Hypothesis, 30 J. Scientific Study Religion 159 (1991)CrossRefGoogle Scholar (religiosity of adult Catholics is inversely related to propensity to engage in drinking, tax evasion, and personal use of an employer's property); Francis, L.J., The Influence of Religion, Gender, and Social Class on Attitudes Toward School Among 11-Year-Olds in England, 60 J. Experimental Edu. 339 (1992)CrossRefGoogle Scholar (positive relation between religiosity and positive altitude about school): Ross, L.E., Religion and Deviance: Exploring the in Imtact of Social Control Elements, 14 Soc. Spectrum 65 (1994)CrossRefGoogle Scholar; Lipford, J., McCormick, R. E., and Tollison, R. D., Preaching Matters, 21 J. Econ. Beh. and Org. 235 (1993)CrossRefGoogle Scholar; but see Cocbran, J. K., Wood, P. B. and Arneklev, B. J., Is the Religiosity-Delinquency Relationship Spurious?, 31 J. Res. Crime & Delinquency 92 (1994).CrossRefGoogle Scholar

19. On the moral effects of religious groups, see Morgan, S. P., A Research Note on Religion and Morality: Are Religious People Nice People?, 61 Soc. Forces 683 (1983)CrossRefGoogle Scholar (claiming positive relation between religiosity and pro-social morality).

20. A related idea is that religious groups permit intergenerational contracts. Gary Becker notes that the child and the parent would be better off if the parent could agree to invest in the child's human capital in return for a promise from the child to pay for the parent's old-age care. Naturally, no such contract could be made, let alone enforced by courts; and Becker suggests that public education, on the one hand, and Social Security, on the other, provide a rough substitute for the contract See Becker, G., A Treatise on the Family 367 (enlarged ed. 1991).Google Scholar But even without the involvement of the state, a religious group could serve a similar function. In effect, the parent invests in the child and, by entering the child into the religion, extracts a fair guarantee of return payment in the form of old-age care, resulting from the child's religiously-induced loyalty. More directly, religious groups often provide services to the elderly as well as education to children. Judaism emphasizes an obligation to respect and aid the old. Christian groups support and operate old-age institutions. However, other religious groups, such as Islamic and Buddhist groups, do not put special emphasis on the needs of the elderly. See Reynolds, and Tanner, , supra note 16, at 200–03.Google Scholar

21. Religious groups have played leading roles in every great social movement, including abolition of slavery, temperance, and civil rights. Fora discussion of the moral role of religious groups in secular politics, see Reichley, A. J., Religion in American Public Life 350–59 (1985).Google Scholar On the importance of social action to many members of religious groups, and their reliance on religious groups as means, see Bellah, et al. , supra note 2 at 228–32, 239–40.Google Scholar

22. Biddle estimates that religious groups devote 29 percent of their expenditures on philanthropic activities, including activities that benefit both the non-poor and the poor. He estimates that religious groups spend about $1.43 billion on welfare-type assistance to the poor in the United Suites. They spend about $1.26 billion on aid to people overseas (but not all of this is relief). Religious groups spend about $4 billion on social services, although some portion of this benefits members. They donate $2.6 billion to other organizations, some of which must be aid agencies. Finally, religious groups are frequent and important providers of disaster relief. See Biddle, , supra note 2, at 115–19.Google Scholar

Congregations report supporting the following secular activities, for example: day care (30.8%); housing/shelter for homeless (32.5%); relief abroad (71.4%); institutional care (hospitals, clinics, nursing homes, hospices) (56.2%); elementary education (37.6%); improvement of environmental quality (26.9%). Independent Sector, From Belief to Commitment: The Activities and Finances of Religious Congregations in the United States 18 (1988).Google Scholar

23. Compare the models in Greeley, A. M., Religious Change in America 123 (1989)Google Scholar; Hull, and Bold, , supra note 13, at 8ff.Google Scholar

24. Small sects often ensure mutual aid through communal living. See Knnter, R. M., COmmttment and Community (1972).Google Scholar Other groups, such as the Amish, allow private property, but have rigorous mutual insurance systems. See Ferrara, P. J., Social Security and Taxes, in The Amish and the State 127–28Google Scholar (Kraybill, D. B. ed. 1993Google Scholar). More mainstream religious groups often provide aid to poorer members (i.e., wealthier members appear to subsidize poorer members) and provide aid to members who suffer financial misfortune. See Riddle, , supra note 2, at 114.Google Scholar

25. A description of Amish self-support can be found in Ingersoll, B., GOP's Plans to Curtail Government Benefits Bring No Pain to Amish, Wali. St. J., 12 22, 1995, at A1.Google ScholarSee generally The Amish and the State, supra note 24. Another example is the Peace Mission Movement, which, during the Great Depression, prohibited its members (all of whom were black) from receiving public welfare and provided for them in various ways itself. See Light, I. H., Ethnic Enterprise in America: Business and Welfare Among Chinese, Japanese, and Blacks 141–51 (1972).Google Scholar

26. The Githolic Church runs the most extensive system of religious schools, but other religious groups operate schools as well. Catholic school educations are superior to public educations, even controlling for the backgrounds of students. See Coleman, J. S., Hoffer, T., and Kilgore, S., High School Achievement: public, Catholic, and Private Schools Compared (1982).Google Scholar

27. See Anderson, , supra note 8 at 1070–72Google Scholar (discussing Smith's theory that religious groups produce and distribute information about their members, which allows members to assess risks in transactions); Iannaccone, L R., Religious Practice: A Human Capital Approach, 29 J. Scientific Study Religion 297, 298–300 (1990).CrossRefGoogle Scholar

28. For examples of the commercial adivines of religious groups, see Bernstein, L., Opting out of the Legal System: Extralegal Contractual Relations in lite Diamond Industry, 21 J. Legal Stud. 115 (1992)CrossRefGoogle Scholar; Carr, J. L. and Landa, J. T., The Economics of Symbols, Clan Names, and Religion, 12 J. Legal Stud. 135, 152–55 (1983)CrossRefGoogle Scholar; Robbins, T., Cults, Converts and Charisma: The Sociology of New Religious Movements 127–33 (1988)Google Scholar (discussing the commercial orientation of many modern “cults,” including the Unification movement, Scientology, the People's Temple, est. The Process, Shiloh, the Happy-Healthy-Holy Sikh movement, the Vajradhatu Tibetan Buddhist church, and Hare Krishna); Light, , supra note 25, at 141–51Google Scholar (Peace Mission Movement). Evidence of the commercial benefits of more mainstream religious organization includes the commercial success of Jewish groups, the Quakers, and the Mormons. Commercial opportunity is clearly only a part of the motivation for joining or not exiting a religious group, but it is sometimes a significant part. See Roof, and McKinney, , supra note 2, at 162Google Scholar (“the young Baptist salesman who becomes an Episcopalian sales executive,” quoting Berger, P. L., The Precarious Vision 74 (1961))Google Scholar; id. at 175–76.

29. See Wallis, , supra note 9, at 63ff.Google Scholar One might argue that attending services, for example, or even paying a tithe is not really a “cost” because the member derives pleasure from performing these duties. But so long as every demand on the resources of a member docs not align with the member's private desires, one can point to “costs.”

30. Religious groups sometimes charge, in effect, membership fees. Synagogues usually charge a fee, sometimes amounting to hundreds of dollars per year. Many religious groups, such as the Mormons and many fundamentalist Protestant sects, charge the Biblical tithe. More mainline Christian groups have in the past raised money by renting pew to congregation members, but more recently they have relied on voluntary donations. Some sects require large lump-sum payments, including donation of the member's personal property. And almost all members of religious groups must suffer the opportunity costs of spending time at services, meetings, social activities, pilgrimages (such as the Muslim once-per-lifetime pilgrimage to Mecca), and other events if and when they have other things that they would rather do. See Caplow, , supra note 2, at 84–86, 309Google Scholar (working-class families in sample contributed 3.3% of their income to churches).

31. On the closely analogous problem of opportunism in long-term contracts, see, e.g., Williamson, O. E.. The Economic Institutions of Capitalism (1985)Google Scholar; Goetz, C. and Scott, R., Principles of Relational Contracts, 67 Va. L. Rev. 1089 (1981).CrossRefGoogle Scholar

32. See Wallis, , supra note 9.Google Scholar

33. See Olson, M., The Logic of Collective Action (1965)Google Scholar; Hardin, R., Collective Action (1982).Google Scholar For an example, see Bellnh, et al. , supra note 2, at 240–11.Google Scholar

34. See Mueller, D. C., Public Choice II, 385–90 (1989)Google Scholar (discussing Arrow's theorem).

35. In the sociological literature this problem is called the “routinization of charisma.” See Roberts, , supra note 12, at 180–93.Google Scholar

36. See, e.g., id. at 198–99.

37. Among the Amish, although the leadership declares that a person is ostracized, all members must participate for the sanction to be effective. See the essays collected in The Amish and the State, supra note 24.

38. See Hull, B. B. and Bold, F., Hell, Religion, and Cultural Change, 150 J. Inst'l & Theoretical Econ. 447 (1994)Google Scholar; Hull, and Bold, , supra note 13Google Scholar; Telser, L.G.. A Theory of Self-Enforcing Agreements, 53 J. Bus. 27, 29–30 (1980).CrossRefGoogle Scholar Even religions that do not have well-developed conceptions of an afterlife have developed approaches to this problem. For example, Judaism cultivates a deep sense of obligation toward one's children.

39. See Cialdini, R. B., Influence: The New Psychology of Modern Persuasion 115–62 (1984)Google Scholar (discussing “social proof”). For an economic discussion, see Shiller, R. J., Conversation, Information, and Herd behavior, 85 Amer. Econ. Assoc. Papers & Proceedings 181 (1995).Google Scholar

40. See Iannaccone, , supra note 8, at 260.Google Scholar

41. See Wuthnow, , supra note 8, at 495–96Google Scholar; Iannaccone, L. R., Sacrifice and Stigma: Reducing Free Riding in Cults, Communes, and Other Collectives, 100 J. Pol. Econ. 271, 283–89 (1992)CrossRefGoogle Scholar; Iannaccone, , supra note 8, at 242–32Google Scholar (discussing Clock, C.Y. and Stark, R., Christian Beliefs and Anti-Semitism (1966)Google Scholar, a study of Northern California religious organizations). Sects in the study had a mean of 173 members, who had an average income of $6,944 per year and 12.5 years of education; churches in the study had a mean of 939 members, who had an average income of $10,140 per year and 14.5 years of education. Sect members attended on average 49.1 Sunday services per year and 96.8 evening meetings per year, and contributed 8.43% of their incomes. Church members attended on average 33.8 Sunday services per year and 34.5 evening meetings, and contributed 2.64% of their incomes. Sect members tended to have more friends in the congregation. For further discussion and evidence, see Iannaccone, L. R., Why Strict Churches Are Strong, 99 Amer. J. Soc. 1180, 1189ff. (1994).CrossRefGoogle Scholar For further discussion of sects, see Robbins, , supra note 28.Google Scholar

42. This discussion is highly simplified; see generally Hargrove, , supra note 10. at 175Google Scholar; Corbett, J. M., Religion in America 52–74 (1990)Google Scholar (Protestant denominations); id. at 90–91 (Catholic Church), 112–19 (Judaism). See also Burkart, G., Patterns of Protestant OrganizationGoogle Scholar, in American Denominational Organization: A Sociological View (Scherer, R. P. ed. 1980Google Scholar) [hereinafter, American Denominational Organization] (Protestantism); Kim, G., Roman Catholie Organization Since Vatican IIGoogle Scholar, in American Denominational Organization, supra (Catholicism); Elazar, D. J., Patterns of Jewish Organization in the United States, in American Denominational Organization, supra (Judaism).Google Scholar

43. For a detailed analysis, see Allen, D. W., Order in the Church: A Property Rights Approach 27 J. Econ. Beh. & Org. 97 (1995).CrossRefGoogle Scholar

44. See, e.g., Pagels, E., The Gnostic Gospels (1979)Google Scholar (tracing theological commitments to the resurrection of Christ, monotheism, and martyrdom to the organizational needs of the early Church).

45. see Wuthnow, , supra note 8, at 494–95.Google Scholar For example, it is often pointed out that ritual is public, requiring individuals to display their commitments in front of the entire religious community. For a discussion, see Stark, and Glock, , supra note 12, at 81–86.Google Scholar

46. See the discussion in Kosmin, B. A. and Lachman, S. P., One Nation Under God: Religion in Contemporary American Society 18–18 (1993)Google Scholar; see generally Ahlstrom, S. E., A Religious History of the American People 383–84 (1972).Google Scholar

47. For a discussion of the relationship between religion and economic prosperity, see Heath, W.C., Waters, M.S., and Watson, J.K., Religion and Economic Welfare: An Empirical Analysa of State Per Capita Income, 27 J. Econ. Beil and Org. 129 (1995).CrossRefGoogle Scholar

48. See, e.g., Hall, T.L., Religan and Civic Virtue: A Justification of Free Exercise, 67 Tul. L Rev. 87 (1992)Google Scholar; and, somewhat indirectly, Cover, R.M., The Supreme Court, 1982 Term—Forward: Nomos and Narrative, 97 Harv. L. Rev. 4 (1983).CrossRefGoogle Scholar

49. See text accompanying supra note 39.

50. see Finke, and Stark, , supra note 9Google Scholar (disestablishment of state churches in the United States caused rise in religious activity); Olds, K., Privatizing the Church: Disestablishment in Connecticut and Massachusetts, 102 J. Pol. Econ. 277 (1994)CrossRefGoogle Scholar (disestablishment followed by rise in the demand for preachers); Iannaccone, L. R., The Consequences of Religious Market Structure: Adam Smith and the Economics of Religion, 3 Rationality & Society 156 (1991)CrossRefGoogle Scholar (religious participation is negatively correlated with religious market concentration in Protestant (but not Catholic) countries); Iannaccone, , supra note 13 (similar)Google Scholar; Chaves, M. and Cann, D. E., Regulation, Pluralism, and Religious Market Structure: Explaining Religion's Vitality, 4 Rationality & Society 272 (1992)CrossRefGoogle Scholar (religious participation is negatively correlated with state subsidization and regulation both in Protestant and in Catholic countries). See also McConnell, and Posner, , supra note 8, at 54–55.Google Scholar Of course, some would see this as a virtue, not a vice. See Hume, D., 3 History of England, 24–26 (1854 ed.).Google Scholar

51. This argument is a more formal version of Madison's view that close relations with the state corrupt churches. See Drakeman, D. L., Religion and the Republic: Madison and the First Amendment 238Google Scholar, in James Madison and Religious Liberty (Alley, R. S. ed. 1985Google Scholar) (quoting Madison's “Detached Memoranda”). It is illustrated by Howe's discussion of the ways in which American states in the 19th century injured religious groups that they attempted to support. The states felt compelled by logic and politics to impose (democratic) organizational requirements on the congregations that they supported, but these requirements interfered with the ability of the congregations to choose and maintain the optimal organizational structures. See Howe, M. DeW., The Garden and the Wilderness 32ff. (1965).Google Scholar

52. In post-revolutionary Massachusetts, for example, local governments provided financial support to the local church that enjoyed the adherence of a majority of the residents of the parish. See Howe, , supra note 51, at 32–11.Google Scholar

53. See Walz v. Tax Comm'n, 397 U.S. 664 (1970)Google Scholar (property tax exemption); Hernandez v. Comm'r of Internal Revenue, 490 U.S. 680 (1989)Google Scholar (discussing income lax deduction for charitable contributions).

54. See Posner, , supra note 11.Google Scholar

55. See, e.g., Tilton, v. Richardson, , 403 U.S. 672 (1971)Google Scholar (upholding grants for the benefit of secular buildings operated by religious universities); Bowen, v. Kendrick, , 487 U.S. 589 (1988) (upholding grants for counseling centers run by religious groups).Google Scholar

56. See Walz v. Tax Comm'n, 397 U.S. 664 (1970)Google Scholar (property tax exemption); Hernandez v. Comm'r of Internal Revenue, 490 U.S. 680 (1989)Google Scholar (discussing income tax deduction for charitable contributions).

57. See, e.g., Church of the Lukumi Babalu Aye, Inc. v. Hialeah, 508 U.S. 520 (1993)Google Scholar (striking down public health law that prohibited ritual sacrifice of animals).

58. see Lyng v. Northwest Indian Cemetery Protective Ass'n, 485 U.S. 439 (1988)Google Scholar (upholding law that provided for road through religious site).

59. See Wisconsin, v. Yoder, , 406 U.S. 205 (1972)Google Scholar (striking down law that provided for compulsory education up to the eighth grade, as applied to Amish).

60. see Employment Div., Dep't of Human Resources v. Smith, 494 U.S. 872 (1990)Google Scholar (upholding law that prohibited peyote use even for religious reasons).

61. See 10 U.S.C. § 774 (1988 & Supp. 1995).

62. See Sherbert, v. Verner, , 374 U.S. 398 (1963)Google Scholar (requiring exemption for Sabbatarians to unemployment laws).

63. Cf. McConnell, and Posner, , supra note 8, at 58.Google Scholar

64. This is according to McConnell, M. W., Accommodation of Religion: An Update and a Response to the Critics, 60 Geo. Wash. L. Rev. 685, 734–35 (1992).Google Scholar

65. This is a theme in the literature on accommodations. For example, compare McConnell, , supra note 64, at 736–38Google Scholar (arguing that accommodations can be administered fairly) with Gey, S. G., Why Is Religion Special?: Reconsidering the Accommodation of Religion Under the Religion Clauses of the First Amendment, 52 U. Pitt. L. Rev. 75. 185–86 (1990)Google Scholar (arguing that accommodations cannot be administered fairly). See also Smith, 494 U.S. at 888–90 and n.5 (similar).Google Scholar

66. See, e.g., Allegheny County v. ACLU, 492 U.S. 573 (1989).Google Scholar

67. For some preliminary empirical support for the view that state regulation crowds out religious groups by providing substitutes, see Wuthnow, R. and Nass, C., Government Activity and Civil Privatism: Evidence fmm Voluntary Church Membership, 27 J. Scientific Study Religion 157 (1988).CrossRefGoogle Scholar

68. For a more extensive discussion, see Posner, , supra note 11.Google Scholar

69. See text accompanying supra notes 12–17.

70. For a discussion of civil religion, see Varieties of Civil Religion (Bellah, R. N. and Hammond, P. E. eds. 1980Google Scholar); Hargrove, , supra note 10, at 105–06.Google Scholar

71. Following dc Tocqueville, Bellah argues that in the United States religious groups are essential for inculcating civic virtue (and that civic virtue is necessary to a just and stable democracy). See Bellah, , IntroductionGoogle Scholar, in Varieties of Civil Religion, supra note 70. He thus assumes that religion and civil religion are complements, rather than substitutes, as I assume.

72. On the idea or the civil religion as a substitute for Christianity, see Demerath, N.J. III and Hammond, P. E., Religion in Social Context 202–12 (1969).Google Scholar

73. See supra note 50.

74. Consider laws against prostitution, pornography, adultery, sodomy, fornication, drug use, obscenity, the purchase or consumption of alcohol, cruelly to animals, and other laws that cannot be given a plausible liberal justification (in the sense of protecting people's rights not to be coerced), but appear to prohibit conduct entirely on the basis of its supposed immorality.

75. See Glazer, N.. Fundamentatism, This World 109, Summer 1982.Google Scholar

76. Cf. McConnell, and Posner, , supra note 8, at 57–58.Google Scholar

77. See text accompanying supra note 24.

78. A recent bill proposed in Congress allows stales to transfer money from federal welfare block grants to private religious organizations for charitable purposes. The bill prohibits groups from refusing aid to recipients on religious grounds and from using the money to proselytize. See Novak, V., Fight Looms Over Welfare Provisions That Funnel Aid Through Churches, Wall. St. J., 09 7, 1955, at A16.Google Scholar

79. There have been attempts to measure this; see Biddle, , supra note 2.Google Scholar

80. See text accompanying supra notes 27–28.

81. See Hull, and Bold, , supra note 13, at 6, 10Google Scholar (as the cost of state enforcement declines, the role of the church in enforcing property rights should decline); Hull, B. B. and Bold, F., Hell, Religion, and Cultural Change, 150 J. Inst'l & Theor'l. Econ. 447 (1994)Google Scholar (religious doctrines about hell will he more important in societies where weak governments, families, and communities are unable to produce cooperation).

82. See Bernstein, , supra note 28Google Scholar; Cooter, R. and Landa, J. T., Personal Versus Impersonal Trade: The Size of Trading Groups and Contract Law, 4 Intern'l Rev. L. & Econ. 15 (1984)CrossRefGoogle Scholar; Greif, A., Contract Enforceabitily and Economic Institutions in Early Trade: The Maghribi Traders' Coalition, 83 Amer. Econ. Rev. 525 (1993)Google Scholar (deseribing Jewish group of medieval international traders).

83. See, e.g., Hadnot, v. Shaw, . 820 P.2d 978 (Ok. 1992)Google Scholar; Paul v. Watchtower Bible and Tract Soc., 819 F.2d 875 (9th Cir. 1987)Google Scholar; Grunwald, v. Bornfreund, , 696 F. Supp. 838 (E.D.N.Y. 1988)Google Scholar; Brown v. Mt. Olive Baptist Church, 124 N.W.2d 445 (la. 1963)Google Scholar; Bear v. Reformed Mennonite Church, 341 A.2d 105 (Pa. 1975).Google Scholar Adescription of nineteenth-century cases, which are similar to the 20th-century cases, can be found in Weisbrod, C., The Boundaries of Utopia 126–61 (1980).Google Scholar

84. See, e.g., Jones, v. Wolf, . 443 U.S. 595 (1979)Google Scholar (discussed infra); Serbian E. Orthodox Diocese v. Milivojevich, 426 U.S. 096 (1976)Google Scholar; Presbyterian Church v. Hull Church, 393 D.S. 440 (1969)Google Scholar; Kedroff v. St. Nicholas Cathedral, 344 U.S. 94 (1952).Google Scholar

85. For a discussion of this assumption, see Ellman, I. M., Driven from the Tribunal: Judicial Resolution of Internal Church Disputes, 69 Calif. L, Rev. 1378, 1402–05 (1981).CrossRefGoogle Scholar

86. This is not true if the group's norms are suboptimal and courts are able to choose superior norms, but this seems particularly unlikely when talking about religious groups. Cf. Posner, E. A., Law, Economics, and Inefficient Norms, 144 U. Pa. L. Rev. 1697 (1996)CrossRefGoogle Scholar (discussing conditions under which group norms can be expected to harm members).

87. The incompetency of judges in evaluating religious doctrine is an old theme in scholarship; see, e.g., Howe, , supra note 51, at 48–53, 79–80Google Scholar, and in the cases, see, e.g., Watson, v. Jones, , 80 U.S. (13 Wall.) 679, 729 (1871)Google Scholar; and the sophistication of the dispute resolution mechanisms used by religious groups has been frequently observed. Cf. Bernstein, , supra note 28Google Scholar (arbitration system of Jewish diamond club) and Harmer-Dionne, E.. Optimal Enforcement in the Disciplinary Proceedings of the Church of Jesus Christ of Latter-Day Saints (unpublished, on file with author) (Mormon system).Google Scholar

88. This is not necessarily the case. In United Kosher Butchers Ass'n v. Associated Synagogues of Greater Boston, Inc., 211 N.E.2d 332 (Ma. 1965)Google Scholar, some Jewish businesses allegedly tried to corner the market in certain kosher products, and obtained from some rabbis a condemnation of a competitor's goods as inconsistent with Jewish law. If such behavior occurred, it coultl easily have been the kind of high-value opportunism that a group's sanctions are too weak to deter. Even in this case, however, the court refused to intervene on evidentiary grounds, possibly fearful of substituting its judgment for that of the group.

89. Paul v. Watchtower Bible and Tract Soc. of New York, Inc., 819 F.2d 875 (9th Cir. 1987).Google Scholar

90. Grunwald, v. Bornfreund, , 696 F. Supp. 838 (E.D.N.Y. 1988).Google Scholar

91. Brown v. Mt Olive Baptist Church, 124 N.W.2d 445 (la. 1903).Google ScholarSee also cases involving removal of leaders and removal of teachers, e.g., Miller v. Catholic Diocese of Great Falls, Billings, 728 P.2d 794 (Mt. 1986).Google Scholar

92. See, e.g., Bear v. Reformed Mcnnonite Church, 341 A.2d 105 (Pa. 1975)Google Scholar (allowing tort arising from excommunication and shunning).

93. E.g., Paul, 819 F.2d 875.Google Scholar

94. Guinn v. Church of Christ of Collinsville, 775 P.2d 7GG (Okla 1989).Google Scholar

95. 443 U.S. 595 (1979).

96. see supra note 87.

97. See Ellman, . supra note 85, at 1411–12.Google Scholar

98. This was the approach of Georgia law approved by ihe majority in Jones. 443 U.S. at 599–601.Google Scholar

99. This was the approach of the dissent in Jones. see 443 U.S. at 618–20.Google Scholar

100. This analysis thus converges with Ellman's contractual analysis for schism cases, but argues that what he calls the hands-off approach (and which he rejects) is appropriate for ostracism cases. See Ellman, , supra note 85.Google ScholarSee also Laycock, D., Towards a General Theory of the Religion Clauses: The Case of Cliurrh labor Relations and the Right to Church Autonomy, 81 Colum. L. Rev. 1373, 1403–05 (1981)CrossRefGoogle Scholar; Mansfield, J. H., The Religion Clauses of the first Amendment and the Philosophy of Hie Constitution, 72 Calif. L Rev. 847, 863–68 (1984)CrossRefGoogle Scholar; Gedicks, F. M., Toward a Constitutional Jurisprudence of Religious Groups Rights, 1989 Wisc. L. Rev. 99 (and cites therein).Google Scholar

101. U.S. Const, amend. I.Google Scholar

102. On the historical fear of religious influence on state action, see Howe, , supra note 51, at 62ff.Google Scholar

103. Many religious groups deny that receiving wealth or power from the state is a good thing, because it limits their ability to criticize the state. See Hargrove, , supra note 10, at 259.Google Scholar But they do not seem to object to tax benefits and accommodations, so the analysis is relevant.

104. This was, in fact, the experience of some religious organizations in the 19th century. States that began by supporting them over time increasingly insisted on controls. See Howe, , supra note 51, at 32ff.Google Scholar

105. A statutory analogue of this phenomenon is the lobbying restriction on nonprofit organizations, including religious groups, that receive tax exemptions. See 26 U.S.C § 501(c)(3) (1988 & Supp. 1995)Google Scholar; Wuthnow, and Nass, , supra note 67, at 162.Google Scholar Totalitarian states often suppress religious groups, probably calculating that the risks posed by the groups' ability to obtain political power outweigh any gains.

106. Compare the secularism of Lupu, which would not allow states to compensate religious groups for the benefits they confer, and the accommodationist approach of McConncll, which would allow states to promote religious groups, regardless of benefits conferred, so long as the encouragement is nondiscriminatory. See, e.g., McConnell, , supra note 64Google Scholar; Lupu, , supra note 4.Google Scholar

107. Cf. Howe, , supra note 51Google Scholar, who argues that the religion clauses reflect Roger Williams' theological conception of separation, not Jefferson's secular conception of separation; and that the former is more supportive of general, nondiscriminatory governmental support of religious groups than the latter is.

108. U.S. CONST, amend. V.Google Scholar

109. U.S. CONST, amend. I.Google Scholar

110. For discussions of normative theories, see, e.g., Smith, S. D., The Rise and Fall of Religious Freedom in Constitutional Discourse, 140 U. Pa. L. Rev. 149, 196–223 (1991)CrossRefGoogle Scholar; Garvey, J. H., Free Exercise and the Values of Religious Liberty, 18 Conn. L. Rev. 779 (1986)Google Scholar; Smith, M. E., The Special Place of Religion in the Constitution, 1983 Sup. Ct. Rev. 83.Google Scholar