The concept of religion

Yale Law Journal

Volumn 107, Issue 3, 1997, Pages 791-821

  Peñalver, Eduardo ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33750244406     PISSN: 00440094     EISSN: None     Source Type: Journal    
DOI: 10.2307/797254     Document Type: Short Survey

References (191)

1. U.S. CONST. amend. I.
2. See, e.g., Lee v. Weisman, 505 U.S. 577 (1992) (holding that a rabbi's prayer at a public high school graduation ceremony constituted an establishment of religion); Employment Div. v. Smith, 494 U.S. 872 (1990) (holding that Oregon's dismissal of two Native Americans for their religious use of peyote did not violate the employees' free exercise rights); Edwards v. Aguillard, 482 U.S. 578 (1987) (holding that a Louisiana law requiring public schools to teach creation science whenever evolution was taught was an unconstitutional establishment of religion); Wallace v. Jaffree, 472 U.S. 38 (1985) (holding that a daily period of silence in Alabama public schools amounted to an establishment of religion); Wisconsin v. Yoder, 406 U.S. 205 (1972) (holding that mandatory high school attendance laws violated the free exercise rights of Amish parents).
3. See, e.g., Yoder, 406 U.S. at 215 (limiting the exemption of children from mandatory school attendance laws to religious groups without supplying a definition of religion to help separate such groups from secular ones).
4. The Framers probably shared these assumptions. See ARLIN M. ADAMS & CHARLES J. EMMERICH, A NATION DEDICATED TO RELIGIOUS LIBERTY 90 (1990). As a judge on the Third Circuit, Adams wrote two highly influential opinions on the issue of defining religion. See Africa v. Pennsylvania, 662 F.2d 1025 (3d Cir. 1981); Malnak v. Yogi, 592 F.2d 197, 200-15 (3d Cir. 1979) (Adams, J., concurring).
5. See generally LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW §§ 12-1 to -39 (2d ed. 1988) (describing Supreme Court jurisprudence on "speech").
6. See Texas v. Johnson, 491 U.S. 397, 406 (1989).
7. See Tinker v. Des Moines Indep. Community Sch. Dist., 393 U.S. 503, 505 (1969).
8. See Police Dep't v. Mosley, 408 U.S. 92, 95 (1972).
9. E.g., U.S. CONST. art. II, § 1, cl. 3.
10. Some commentators have taken the view, for example, that "religion" in the First Amendment is more like "speech," arguing that "religion" should be construed much more broadly for constitutional purposes than it would be in ordinary language. See, e.g., Note, Towards a Constitutional Definition of Religion, 91 HARV. L. REV. 1056, 1056 (1978) (advocating the inclusion of whatever constitutes a person's "ultimate concern" within "religion" for First Amendment purposes).
11. 592 F.2d 197 (3d Cir. 1979) (per curiam).
12. 662 F.2d 1025 (3d Cir. 1981).
13. The definition of religion in the First Amendment has, at least in part, played a role in many other cases. See, e.g., United States v. Seeger, 380 U.S. 163, 166 (1965) (holding that a man's opposition to war based upon his "belief in and devotion to goodness and virtue for their own sakes" is "religious"), discussed infra notes 38-51 and accompanying text; United States v. Meyers, 95 F.3d 1475 (10th Cir. 1996) (holding that the Church of Marijuana is not a religion), discussed infra note 78; Alvarado v. City of San Jose, 94 F.3d 1223 (9th Cir. 1996) (holding that commissioning a statue of Quetzalcoatl, an ancient Aztec god, is not a religious act for Establishment Clause purposes); Peloza v. Capistrano Unified Sch. Dist., Nos. 92-55228 & 92-55644, 1994 WL 382635 (9th Cir. July 25, 1994) (holding that "evolutionism" is not a religion); Johnson v. Pennsylvania Bureau of Corrections, 661 F. Supp. 425 (W.D. Pa. 1987) (holding that the "Spiritual Order of Universal Beings," a group founded by the plaintiff, is not a religion).
14. See, e.g., Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993) (holding that a law against killing animals, which was passed in order to prevent the exercise of the Santeria religion, was a violation of the group's First Amendment rights).
15. See, e.g., Malnak, 592 F.2d at 200 (per curiam).
16. FREDERICK MARK GEDICKS, THE RHETORIC OF CHURCH AND STATE 116 (1995).
17. Although plaintiffs from minority religious groups have won free exercise claims in lower courts, such victories are relatively rare. A computer search reaffirmed Gedicks's assertion. Search of WESTLAW, Allfeds Database (May 26, 1997) (terms and connectors search for cases containing "free exercise"). Over the past two years, I could find only one case in which a non-Christian plaintiff prevailed on a free exercise claim in the lower federal courts. See Cheema v. Thompson, 67 F.3d 883 (9th Cir. 1995) (upholding a free exercise claim by Sikh students seeking accommodation of their possession of ceremonial daggers on school property, despite a generally applicable no weapons policy). The same search, however, yielded a variety of cases in which the free exercise claims of non-Christians were denied, see, e.g., May v. Baldwin, 109 F.3d 557 (9th Cir.) (rejecting a free exercise claim by a Rastafarian inmate who did not want to unbraid his dreadlocks), cert. denied, 66 U.S.L.W. 3282 (1997); Brock v. Carroll, 107 F.3d 241 (4th Cir. 1997) (rejecting a free exercise claim by a prisoner deprived of his "prayer pipe"); Fawaad v. Jones, 81 F.3d 1084 (11th Cir.) (rejecting a free exercise claim by a Muslim prisoner who was forced to use his given name in addition to the Muslim name he had chosen upon conversion to that faith), cert. denied, 66 U.S.L.W. 3282 (1997); Hamilton v. Schriro, 74 F.3d 1545 (8th Cir.) (rejecting a free exercise claim by a Native-American prisoner seeking access to a sweat lodge and permission to grow long hair), cert. denied, 117 S. Ct. 193 (1996); Hicks v. Gamer, 69 F.3d 22 (5th Cir. 1995) (rejecting a free exercise claim by a Rastafarian seeking to grow long hair), and a number of cases in which the free exercise claims of Christian plaintiffs were upheld, see, e.g., Mockaitis v. Harcleroad, 104 F.3d 1522 (9th Cir. 1997) (upholding a free exercise claim by a Catholic priest and bishop concerning a prosecutor's secret taping of a confession by a suspect to a Catholic priest); Tucker v. California Dep't of Educ., 97 F.3d 1204 (9th Cir. 1996) (holding that a rule barring employees from posting religious messages at a Christian plaintiff's workplace violated his free exercise rights); Sasnett v. Sullivan, 91 F.3d 1018 (7th Cir. 1996) (holding that the refusal of prisons to make exceptions to a no jewelry policy to accommodate prisoners seeking to wear crucifixes violated the prisoners' free exercise rights), vacated on other grounds, 117 S. Ct. 2502 (1997). I am not making a strong statistical claim here, but merely suggesting the possibility of judicial bias against non-Christian plaintiffs. In a country in which most people consider themselves to be Christian, it might be argued that the majority of free exercise claims would be made by Christians and that therefore it is not surprising to find a numerical disparity between free exercise victories by Christians and non-Christians. On the other hand, it might be the case that, in a country where most people are Christian, the laws are written in such a way that they do not interfere with the free exercise of the majority religion. If this is the case, we would expect more free exercise claims to be made by non-Christians and hence a statistical disparity in favor of non-Christian plaintiffs. There is no a priori way to determine which possibility actually obtains. Clearly, further empirical research (more rigorous than my simple WESTLAW search) needs to be done on this issue.
18. Cf. ADAMS & EMMERICH, supra note 4, at 92 (admitting that the definition of religion presents real dangers of western bias).
19. Note, supra note 10, at 1056.
20. See Jesse H. Choper, Defining "Religion" in the First Amendment, 1982 U. ILL. L. REV. 579, 593-97.
21. See id. at 597-604. This definition fails, however, because belief in extra-temporal consequences to human action (that is, reward or punishment) is not a necessary feature of belief systems that almost all would concede to be religious. Ancient Judaism, for example, professed no such belief. See Wayne T. Pitard, Afterlife and Immortality, in THE OXFORD COMPANION TO THE BIBLE 15, 15-16 (Bruce M. Metzger & Michael D. Coogan eds., 1993).
22. See Andrew W. Austin, Faith and the Constitutional Definition of Religion, 22 CUMB. L. REV. 1, 33-43 (1991-1992). As Austin himself notes, however, faith by itself is not sufficient for something to be called a religion. See id. at 40-43. Austin therefore adds the requirement of faith in a "greater power than man." Id. at 42. This addition makes Austin's definition unworkably vague and seems to convert it into a theistic definition (at the expense of nontheistic religions).
23. See George C. Freeman, III, The Misguided Search for the Constitutional Definition of "Religion," 71 GEO. L.J. 1519, 1534-48 (1983).
24. See Kent Greenawalt, Religion as a Concept In Constitutional Law, 72 CAL. L. REV. 753 (1984).
25. See Freeman, supra note 22, at 1553; Greenawalt, supra note 23, at 762, 767-78.
26. The Supreme Court and appellate court cases discussed in this part are widely regarded by commentators as representing the most significant judicial discussions on the topic of the definition of religion. See, e.g., Choper, supra note 19, at 587-91 (summarizing the Supreme Court jurisprudence on the issue of defining religion).
27. 133 U.S. 333 (1890).
28. Id. at 342.
29. Id. at 341.
30. Id. at 341-42 ("To call their advocacy [of polygamy] a tenet of religion is to offend the common sense of all mankind.")
31. Although by the late 19th century the United States contained a diversity of sects, almost all religious people were Christians and therefore unlikely to object to a definition of religion that excluded nontheistic religions. See J. GORDON MELTON, The Development of American Religion: An Interpretive View, in ENCYCLOPEDIA OF AMERICAN RELIGIONS 1, 14-15 (5th ed. 1996).
32. 283 U.S. 605 (1931).
33. Id. at 625 (citing Holy Trinity Church v. United States, 143 U.S. 457, 470-71 (1892)). The theistic definition from Davis is quoted directly in a dissent by Justice Hughes, see id. at 634 (Hughes, J., dissenting), but the majority clearly applied the same theistic definition.
34. See WINTHROP S. HUDSON, RELIGION IN AMERICA 244-46 (4th ed. 1987). The religious homogeneity of the United States prior to the 20th century should not be exaggerated, however, because some variation did exist. See id. at 179-87. Prior to the immigration of large numbers of Asians in the late 19th century and the annexation of Hawaii in 1898, however, very few nontheists practiced their religions in the United States. See id. at 3-4, 108, 244 (describing religious pluralism in the United States before the beginning of the 20th century as pluralism within a Christian consensus); Melton, supra note 30, at 14-15 (describing the changing religious demography of the United States during the 20th century). There was little pressure on a theistic definition of religion. The number of religious bodies in the United States has exploded over the course of this century.
35. See MELTON, supra note 30, at 15; J. GORDON MELTON, Selections from the Introduction to the First Edition, in ENCYCLOPEDIA OF AMERICAN RELIGIONS, supra note 30, at xvii, xvii.
36. See TRIBE, supra note 5, § 14-6.
37. 367 U.S. 488 (1961).
38. See id. at 489-90, 495.
39. Id. at 495 n. 11. The breadth of Black's notion of religion is often overstated. "Ethical Culture" and "Secular Humanism" were not references to broad social movements, but to specific organizations, as is made clear by the two cases cited by Black in support of their inclusion. See id. (citing Washington Ethical Soc'y v. District of Columbia, 249 F.2d 127 (1957); Fellowship of Humanity v. County of Alameda, 315 P.2d 394 (Cal. Dist. Ct. App. 1957)).
40. 380 U.S. 163 (1965).
41. Pub. L. No. 80-759, § 6(j), 62 Stat. 612, 613 (1948) (codified at 50 U.S.C. app. § 456(j) (1958)).
42. Id.
43. Id.
44. To some extent, the historical context may be relevant here. Perhaps in part because of the escalating war in Vietnam, the Supreme Court shied away from the option of holding the statute unconstitutional (and thus leaving inductees without the possibility of conscientious objection), and instead sought to broaden the meaning of "religion" within the statute itself. Cf. Robert L. Rabin, When Is a Religious Belief Religious: United States v. Seeger and the Scope of Free Exercise, 51 CORNELL L.Q. 231, 238-39 (1966) (speculating about the Court's reluctance to leave people without the means of claiming conscientious objector status). That this concern is not the entire story, however, is indicated by the fidelity of the Seeger definition to the Court's broad language in Torcaso in 1961.
45. Seeger, 380 U.S. at 166 (quoting Seeger's letter to the draft board).
46. Id. at 176; see also id. at 180 (citing 2 PAUL TILLICH, SYSTEMATIC THEOLOGY 12 (1957)).
47. Evidence for this view can be found in the Court's statement that, in interpreting "Supreme Being" as used in the statute, it was faced with a choice of interpreting the term to require belief in the "orthodox God" or in a "broader concept of a power or being, or a faith, 'to which all else is subordinate.'" Id. at 174 (quoting WEBSTER'S NEW INTERNATIONAL DICTIONARY OF THE ENGLISH LANGUAGE 2105 (2d ed. 1943)).
48. Id. at 176.
49. See Torcaso v. Walkins, 367 U.S. 488, 493 (1961).
50. See Seeger, 380 U.S. at 188 (Douglas, J., concurring). This constitutional reading of the Seeger holding has been endorsed by several lower courts and most scholars. See, e.g., Malnak v. Yogi, 592 F.2d 197, 204-05 (3d Cir. 1979) (per curiam); Choper, supra note 19, at 588; Donald A. Giannella, Religious Liberty. Nonestablishment, and Doctrinal Development: Part I. The Religious Liberty Guarantee, 80 HARV. L. REV. 1381, 1425 (1967); Rabin, supra note 42, at 238.
51. See Seeger, 380 U.S. at 180 ("[W]e believe this construction embraces the ever-broadening understanding of the modern religious community.").
52. See Austin, supra note 21, at 15. But see A. Stephen Boyan, Jr., Defining Religion in Operational and Institutional Terms, 116 U. PA. L. REV. 479, 497 (1968) (praising the Seeger definition as a step in the right direction).
53. I understand conscience to be a broader concept than religion, as it encompasses both religiously and nonreligiously motivated beliefs. See infra Section III.A (discussing the reasons for rejecting a broad understanding of "religion" in the First Amendment).
54. 398 U.S. 333, 341 (1970) (including within the "parallel position" definition someone who crossed the word "religion" off of his application for conscientious objector status and who described his beliefs as based upon his readings in "history and sociology").
55. 406 U.S. 205 (1972). For discussions of the apparent narrowing of the Court's definition in Yoder, see Austin, supra note 21, at 17; and Choper, supra note 19, at 585.
56. Yoder, 406 U.S. at 216.
57. Id. at 215.
58. See id. at 216. Thoreau's motivation, the Court said, "was philosophical and personal rather than religious." Id.
59. Thoreau's beliefs were far more religiously based than, for example, those of the defendant in Welsh. Compare Welsh, 398 U.S. at 340-41 (describing Welsh's beliefs), with Freeman, supra note 22, at 1559-60 (describing the religious roots of much of Thoreau's belief system).
60. See, e.g., Thomas v. Review Bd. of the Ind. Indep. Employment Sec. Div., 450 U.S. 707, 714-16 (1981) (expressing hesitancy about the enterprise of determining the religious nature of beliefs, but stating that certain beliefs could be too bizarre to qualify as "religious").
61. 592 F.2d 197, 207-10 (3d Cir. 1979) (Adams, J., concurring).
62. See id. at 197-98 (per curiam).
63. Puja is Hindi for "homage, reverence, worship." R.C. TIWARI ET AL., HINDI-ENGLISH ENGLISH-HINDI DICTIONARY 146 (1993). Guru Dev is translated, literally, as "teacher-god." Id. at 56, 113.
64. See Malnak, 592 F.2d at 199 (per curiam).
65. Id. at 207 (Adams, J., concurring).
66. Id. at 207.
67. Id. at 207-08.
68. Id. at 208.
69. See id. at 209.
70. Id. at 210.
71. Id.
72. 662 F.2d 1025 (3d Cir. 1981).
73. Id. at 1026 (quoting Africa).
74. Id. at 1027.
75. Id. (quoting Africa).
76. Id. at 1034. The suggestion that religions must be concerned primarily with the "spiritual" or "other-worldly," and not with reforming society, assumes a strong distinction between the public, secular sphere and the private, devotional sphere that is characteristic of western, particularly Protestant, faiths.
77. Id. at 1035 (citing Malnak, 592 F.2d at 209 (Adams, J., concurring)).
78. See id. The same bias that Judge Adams exhibited in focusing on MOVE'S apparent lack of concern with the "spiritual" or "other-worldly" can also be seen in this list of structural features typical of religions. It is clear that his baseline of comparison was a Christian church community.
79. Id. at 1035. Judge Adams limited his holding to MOVE as presented in the evidence before him. Perhaps as a result of his own uncertainty regarding the correct classification of the organization, he left open the possibility that a more thorough presentation of evidence could make a difference'as to the result of his analysis in the future. See id. at 1036 n.22.
80. See, e.g., United States v. Meyers, 95 F.3d 1475 (10th Cir. 1996). In Meyers, the Tenth Circuit adopted a definition of religion following the same comparative approach as Adams in Malnak and Africa, but using a larger number of indicia. The court provided five indicia: ultimate ideas, metaphysical beliefs, moral or ethical system, comprehensiveness of beliefs, and accoutrements of religion (which in turn included 10 sub-indicia). See id. at 1483-84. The case involved a defendant who tried to avoid conviction for drug violations by arguing that he was the "founder and Reverend of the Church of Marijuana" and that it was "his sincere belief that his religion commands him to use, possess, grow and distribute marijuana for the good of mankind and the planet earth." Id. at 1479. The court only considered the definition of religion with regard to the Religious Freedom Restoration Act (RFRA), Pub. L. No. 103-141, 107 Stat. 1488 (1993) (codified as amended at 42 U.S.C. §§ 2000bb-1 to -4 (1994)), but the close connection between RFRA and prior Religion Clause jurisprudence would seem to imply that a definition of religion for the purposes of RFRA adjudication would be relevant to the question of a constitutional definition. In any event, the Supreme Court has since held the RFRA unconstitutional. See City of Boerne v. Flores, 117 S. Ct. 2157 (1997).
81. 718 F.2d 1210 (2d Cir. 1983).
82. WILLIAM JAMES, VARIETIES OF RELIGIOUS EXPERIENCE (Harvard Univ. Press 1985) (1910).
83. Moon, 718 F.2d at 1227 (quoting JAMES, supra note 80, at 31). Interestingly, James referred to his own proposed definition as "arbitrary" and limited its application to the purpose of the lectures from which the definition was drawn. JAMES, supra note 80, at 32, 34.
84. As with the discussion in the courts, scholarly debate over the correct definition of religion for First Amendment purposes has been inconclusive. Constraints on space prevent me from discussing in detail the various definitions that have been proposed and their many weaknesses. For a more detailed critique of the various definitions, see Freeman, supra note 22, at 1534-48.
85. By this I mean a relatively narrow understanding of the word, one which excludes such systems of thought as philosophy or mere ideology.
86. See Michael W. McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 HARV. L. REV. 1410, 1481 (1990).
87. Michael W. McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 HARV. L. REV. 1410, (1990). Id.
88. (quoting 1 ANNALS OF CONG. 757-59, 796 (Joseph Gales ed., 1789)).
89. See id.
90. See id. at 1495.
91. See id.
92. See id.; see also Freeman, supra note 22, at 1521-22 (arguing that the history of the congressional debates should be interpreted as endorsing an intention by the Framers to protect "freedom of conscience only in matters of religion" and not "freedom of conscience per se"); Stanley Ingber, Religion or Ideology: A Needed Clarification of the Religion Clauses, 41 STAN. L. REV. 233, 277-78 (1989) (endorsing the narrow interpretation of the Framers' choice to omit the word "conscience" from the First Amendment).
93. See Christopher L. Eisgruber & Lawrence G. Sager, The Vulnerability of Conscience: The Constitutional Basis for Protecting Religious Conduct, 61 U. CHI. L. REV. 1245, 1270 (1994).
94. Madison, for example, defined religion as "the duty which we owe to our Creator and the Manner of discharging it." JAMES MADISON, To the Honorable the General Assembly of the Commonwealth of Virginia. A Memorial and Remonstrance., in THE MIND OF THE FOUNDER: SOURCES OF THE POLITICAL THOUGHT OF JAMES MADISON 6, 7 (Marvin Meyers ed., 1981).
95. See infra notes 129-134 and accompanying text.
96. Cf. ADAMS & EMMERICH, supra note 4, at 90 (observing that the Framers probably thought that the meaning of religion was self-evident).
97. See, e.g., Eisgruber & Sager, supra note 90, at 1282; Douglas Laycock, Religious Liberty as Liberty, 7 J. CONTEMP. LEGAL ISSUES 313, 317 (1996).
98. See, e.g., HANS KUNG, GLOBAL RESPONSIBILITY 75-76 (1991).
99. See infra note 101. I would not deny that at times persons have been ruthlessly persecuted on account of their nonreligious conscience, as in the communist red scares of the first half of the 20th century. But this phenomenon has not been nearly as pervasive as religious persecution and is adequately covered by other provisions of the Constitution.
100. See, e.g., Malnak v. Yogi, 592 F.2d 197, 200 (3d Cir. 1979) (per curiam).
101. Further, both the Establishment and Free Exercise Clauses limit the freedom of the state. Thus, an overly broad definition of religion could severely hamper the ability of the state to function effectively. See Choper, supra note 19, at 592. Choper makes the plausible argument that because free exercise benefits granted to religions constitute burdens on the state, the wider the definition of religion adopted, the less robust the protections the Religion Clauses are likely to provide. See id. This provides another reason for tailoring the Religion Clauses narrowly to those belief systems that need protection.
102. This historical justification is itself reinforced by original intent, because the bloody history of religious conflict and intolerance in post-Reformation Europe was one of the major concerns that generated support for the Religion Clauses among rationalist humanists like Madison. Cf. MADISON, supra note 91, at 11 ("Torrents of blood have been spilt in the old world, by vain attempts of the secular arm to extinguish Religious discord . . . .").
103. Norman Malcolm, drawing on Wittgenstein's concept of the "language game," has compared the discourse of an individual religion to a unique language game with its own internal logic, basic assumptions ("framework propositions"), and game-specific criteria for what counts as evidence. See Norman Malcolm, The Groundlessness of Belief, in REASON AND RELIGION 143 (Stuart C. Brown ed., 1977).
104. This view of religion as somehow less than "rational" by "objective" scientific standards has been challenged by more orthodox philosophers of religion such as Colin Lyas. See, e.g., Colin Lyas, The Groundlessness of Religious Belief, in REASON AND RELIGION, supra, at 158. Malcolm does not seek to denigrate religious belief as somehow less than reasonable. Indeed, following Wittgenstein, he rejects any game-independent notion of reasonableness. See Malcolm, supra, at 146. Scientific belief, he argues, is no more "objectively" reasonable than religious belief, but from the perspective of one language game, behavior within another language game seems "unreasonable." Id. at 152. It is thus impossible for someone operating from within one language game to justify her beliefs to someone operating within another language game. See id. at 151-52. Nevertheless, the presence of individuals who, like Lyas, are willing to subject their "religion" to the scrutiny of scientific and philosophical standards of "reasonability" indicates the impossibility of using the rejection of some notion of public reason within religion as a necessary condition for calling something a religion. Such exceptions, however, should not prevent us from making a broad argument about the value of protecting "religion" as such based on the role of reason within religions in general.
105. See, e.g., Thomas Nagel, Moral Conflict and Political Legitimacy, 16 PHIL. & PUB. AFF. 215, 232 (1987) (describing debates between different religious believers as normally consisting of bare assertions of each side's truth). Philosophical ethicists have argued in a similar vein about the apparent irresolvability of many moral and political judgments.
106. See ALFRED JULES AVER, LANGUAGE, TRUTH AND LOGIC 21-22 (1956) (arguing that moral disagreements not based on disagreements over the "facts" of a situation are irresolvable because moral judgments are fundamentally subjective);
107. ALASDAIR MACINTYRE, AFTER VIRTUE 6-8 (1981) (observing that modern moral and political disagreements are uniquely irresolvable through rational discussion). But unlike philosophically derived moral or political judgments, which are often experienced as dictated by the individual conscience, the religiously based belief is experienced by the believer as imposed from outside by a reality that transcends the individual. See Ingber, supra note 89, at 282.
108. Philosophical moral objectivists also think of ethical judgments as somehow transcending the individual, see, e.g., IMMANUEL KANT, FUNDAMENTAL PRINCIPLES OF THE METAPHYSICS OF MORALS (Thomas K. Abbott trans., The Bobbs-Merrill Co. 1981) (1785) (setting forth a philosophical moral system, the content of which transcends the individual moral subject), but as with other philosophical ethicists, they participate in a shared philosophical mode of discourse that provides an alternative to violent confrontation. I understand that not all philosophers agree as to what constitutes necessary and sufficient conditions for "truth," but I believe that philosophers do share a more or less common mode of discourse, or methodology, for evaluating and discussing such issues. The feeling that the religious belief is externally imposed gives religiously motivated disagreements a sharper, more violent edge than disagreements based on differences between secular ethical systems (for example, between Kantianism and utilitarianism). To the religious believer, the one who does not believe is often seen as one who refuses to acknowledge a self-evident truth. The potential for violence resulting from the irresolvability of moral and political disputes, however, is mitigated by their more transparent and self-conscious subjectivity.
109. See Eisgruber & Sager, supra note 90, at 1282-83.
110. See Nagel, supra note 101, at 232.
111. See Leo Pfeffer, Equal Protection for Unpopular Sects, 9 N.Y.U. REV. L. & SOC. CHANCE 9, 9-10 (1979-1980).
112. For arguments that empirically based, scientific belief systems constitute the acceptable norm for public discourse, see, for example, Nagel, supra note 101, at 232; and John Rawls, The Idea of an Overlapping Consensus, 7 OXFORD J. LEGAL STUD. 1, 8 (1987).
113. The experience of the Mormons in the 19th-century United States exemplifies this pattern. Mormons were lampooned by their critics as '"men of perverted intellect.'" KENNETH H. WINN, EXILES IN A LAND OF LIBERTY: MORMONS IN AMERICA, 1830-1846, at 72 (1989)
114. (quoting Letter to the Editor of the OHIO STAR, reprinted in PAINESVILLE TELEGRAPH, May 20, 1836). As would be expected from the discussion of the nature of religious belief systems as language games, see supra note 100, the Mormons also accused their critics of irrationality, see WINN, supra, at 73. The Mormons were tarred and feathered, run out of town, and sometimes killed. See id. at 96-97, 139-42. Because of their minority status, the Mormons' appeals to majoritarian political institutions were fruitless. See id. at 142-47. In a statement that succinctly illustrates the inability of majoritarian political systems to protect religious minorities, the Governor of Missouri explained his decision not to protect Mormon settlers from anti-Mormon violence: "[T]he quarrel was between the Mormons and the mob," he said, explaining that the best solution was to let the two sides "fight it out." Id. at 139 (internal quotation marks omitted). The treatment of adherents of the Santeria religion by some local communities provides a more recent example of the vulnerability of minority religious groups to persecution through the majoritarian political process. In Hialeah, Florida, for example, the city government sought to prevent the practice of the Santeria religion by banning the killing of animals within city limits. See Church of Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 526-28 (1993). Animal sacrifice plays an integral role in Santeria rituals. See id. at 525. Santeria rituals have sparked resentment in other cities as well.
115. See, e.g., Marty Sabota, North Side Residents Upset About Animal Sacrifices, SAN ANTONIO EXPRESS-NEWS, Sept. 26, 1997, at 10D, available in 1997 WL 13206688.
116. The State Department has been accused of anti-Santeria bias for its refusal to allow 31 Cuban babalawos (Santeria high priests) into the United States in August 1997 for a religious conference in San Francisco. See Deborah Ramirez, U.S. Bans 31 Cubans from Religious Meeting, FT. LAUDERDALE SUN-SENTINEL, Aug. 4, 1997, at 1A, available in 1997 WL 11395827.
117. See, e.g., HERBERT MCCABE, GOD MATTERS 2-9 (1987) (arguing that science can answer "how" questions but only religion can answer "why" questions).
118. See MORTON KLASS, ORDERED UNIVERSES: APPROACHES TO THE ANTHROPOLOGY OF RELIGION 38 (1995). Klass argues that religion, as a universal institution, provides answers to fundamental "why" questions. See id. at 56. He points out, however, that the general answers to these fundamental "why" questions are not interesting to adherents of a particular belief system, because these answers are part of the framework of the adherents' worldview, the most basic of their taken-for-granted assumptions. See id. Instead of answering the question "Why is there death?" the believer is interested in explaining ihis particular untimely death. See id. at 57. Although I agree with Klass's basic point that religions are unique in their ability to answer these "why" questions, I do not think this is adequate as a definition of religion.
119. See STEPHEN L. CARTER, THE CULTURE OF DISBELIEF 51-56 (1993) (criticizing a "public square" that excludes religious thinking or at least forces it to be "bracketed" or translated); GEDICKS, supra note 16, at 32-37 (discussing the hostility of modern, secular culture toward religious knowledge).
120. The recent mass suicide of the Heaven's Gate group became the occasion for widespread reflection in the media on the unhealthy effects of the oversecularization of American culture. See, e.g., Marty Kaplan, Maybe Reason Isn't Enough, N.Y. TIMES, Mar. 31, 1997, at A15 (discussing the inability of secular culture to satisfy human needs for meaning).
121. See, e.g., id. Kaplan describes this need: This is the sadness at the heart of our secular lives. No one wants to live in a pointless, chaotic cosmos, but that is the one that science has given us, and that our culture has largely championed. We may yearn for the divine, but our feet are stuck in the moral relativism (or even nihilism) that such a culture breeds. The post-modern Dadaism that's hip today is the best we can do; everything's a joke. But inside it feels awful. Id.
122. See, e.g., Ingber, supra note 89, at 240; Val D. Ricks, To God God's, to Caesar Caesar's, and to Both the Defining of Religion, 26 CREIGHTON L. REV. 1053, 1061-64 (1993);
123. Jonathan Weiss, Privilege, Posture and Protection: "Religion" in the Law, 73 YALE L.J. 593, 622 (1964);
124. Sharon L. Worthing, "Religion" and "Religious Institutions" Under the First Amendment, 7 PEPP. L. REV. 313, 345 (1980).
125. See Austin, supra note 21, at 6.
126. Establishment concerns would also seem to bar the adoption of a dual definition of religion such as the one proposed by Professor Tribe in the first edition of his textbook on constitutional law. See LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW § 14-6, at 831 (1978). Professor Tribe defined religion for establishment purposes as that which is not arguably nonreligious and for free exercise purposes as that which is arguably religious. See id. As Judge Adams noted in Malnak v. Yogi, however, such a definition creates three categories of belief systems under the Religion Clauses: first, those covered neither by the Establishment Clause nor by the Free Exercise Clause; second, those protected by the Free Exercise Clause but not subject to the Establishment Clause; and third, those protected by the Free Exercise Clause and subject to the Establishment Clause. See Malnak v. Yogi, 592 F.2d 197, 212 (3d Cir. 1979) (Adams, J., concurring). The second group, likely to be made up of new religious groups, would be uniquely favored in that it could (as both arguably religious and arguably nonreligious) invoke free exercise benefits without suffering the burdens associated with the Establishment Clause. See id at 213. Perhaps for reasons such as these, Professor Tribe has abandoned this proposal in the most recent edition of his textbook. See TRIBE, supra note 5, § 14-6, at 1186.
127. This seems to have been the approach taken by John Locke in his discussion of natural kind terms (terms that refer to natural categories, like species) and nominal essences. See JOHN LOCKE, AN ESSAY CONCERNING HUMAN UNDERSTANDING 260, 270 (A.D. Woozley ed., Meridian 1964) (1689). Nominal essences, Locke explained, are the summation of all the ideas essential to the "sort."
128. JOHN LOCKE, AN ESSAY CONCERNING HUMAN UNDERSTANDING 260, (1964) (1689). Id. The membership of a particular in the sort is determined by its possession of the qualities that form the nominal essence of that sort. See id.
129. See John McDowell, Wittgenstein on Following a Rule, in MEANING AND REFERENCE 257, 257, 288 (A.W. Moore ed., 1993).
130. Several philosophers of language have adopted this approach to word meaning. See, e.g., id. at 276; Gilbert Ryle, The Theory of Meaning, in THE IMPORTANCE OF LANGUAGE 147, 162 (Max Black ed., 1962).
131. See McDowell, supra note 116, at 275.
132. See LUDWIG WITTGENSTEIN, PHILOSOPHICAL INVESTIGATIONS § 198, at 80 (G.E.M. Anscombe trans., MacMillan Co. 1953) ("[W]hat has the expression of a rule - say a sign-post - got to do with my actions? What sort of connexion is there here? - Well, perhaps this one: I have been trained to react to this sign in a particular way, and now I do so react to it.").
133. According to William Alston (a philosopher of language heavily influenced by Wittgenstein), however, definitions are attempts to teach the meaning (that is, use) of a word by substituting for it a word or phrase whose use we already know. See WILLIAM P. ALSTON, PHILOSOPHY OF LANGUAGE 22 (1964).
134. See WITTGENSTEIN, supra note 119, § 197, at 80; McDowell, supra note 116, at 288.
135. 662 F.2d 1025 (3d Cir. 1981). For a discussion of Africa, see supra notes 70-77 and accompanying text.
136. See 18 THE OXFORD ENGLISH DICTIONARY 264 (2d ed. 1989).
137. See id.
138. See WEBSTER'S NEW WORLD DICTIONARY OF THE AMERICAN LANGUAGE 1500 (2d College ed. 1984).
139. 133 U.S. 333 (1890); see supra text accompanying notes 26-29.
140. Davis, 133 U.S. at 342.
141. Edward Conze, a scholar of Buddhist thought quoted by Justice Douglas in United States v. Seeger, points out that the exact opposite could conceivably have occurred. Instead of extending our notion of religion to include nontheistic groups, we could have declined to apply the word "religion" to those groups at all. See United States v. Seeger, 380 U.S. 163, 191 (1965) (Douglas, J., concurring) (citing EDWARD CONZE, BUDDHISM: ITS ESSENCE AND DEVELOPMENT 38-39 (1959)). Undoubtedly, the many similarities between Christianity and nontheistic systems of thought like Buddhism made the conclusion that the concept of God was not essential to religion the more likely result.
142. Weiss, supra note 112.
143. See id. at 622.
144. 367 U.S. 488, 495 & n.11 (1961).
145. The opposite (though perhaps less likely) possibility, that the everyday definition of religion might become more narrow in the future, does not pose the same establishment threat. If the everyday definition of religion becomes more narrow in the future than it is today, a constitutional definition of religion that adhered to today's standards would simply protect more types of behavior than it had to. This would not create an establishment problem, however. Only a constitutional definition that is more narrow than the everyday use of the word poses the risk of establishing a form of religion.
146. See Freeman, supra note 22, at 1534-48. There is little consensus as to the "essence" of religion. There is, of course, a dizzying array of definitions that have been unsuccessfully proposed in the legal literature alone over the past three decades. See supra notes 18-24 and accompanying text. But the confusion is not limited to the law. Anthropologists have also proved unable to agree upon a single definition that successfully distinguishes all religions from all nonreligions. See JACOB PANDIAN, CULTURE, RELIGION AND THE SACRED SELF 11-13 (1991) (listing and critiquing a range of definitions proposed within the anthropological literature). The most popular anthropological definition, based upon the presence of the "supernatural," has been forcefully criticized by anthropologist Morton Klass as rooted in an atheistic and scientific worldview and therefore as "too ethnocentric to be of use for cross-cultural study." KLASS, supra note 108, at 28. Philosophers have likewise failed to locate any one essence or combination of essences that would allow them to formulate a persuasive definition of religion.
147. See, e.g., MICHAEL PETERSON ET AL., REASON AND RELIGIOUS BELIEF 3-6 (1991) (discussing the many definitions of religion that have been proposed by philosophers and proposing another);
148. NINIAN SMART, THE PHILOSOPHY OF RELIGION 26 (1979) (discussing the difficulties of formulating a definition of religion). William Alston, a philosopher of language, argues that religion is an inherently vague concept, unamenable to specific definition. Given the same borderline case, he argues, "mature native speakers" will disagree as to whether something is or is not a religion. ALSTON, supra note 120, at 87-90.
149. In two articles that appeared almost simultaneously, George Freeman and Kent Greenawalt both argued that the search for a dictionary-style constitutional definition of religion was doomed to end in failure. See Freeman, supra note 22, at 1548; Greenawalt, supra note 23, at 763. What Greenawalt calls the "dictionary approach" to defining religion involves a quixotic search for an essence of religion, the presence of which would create the infallible conclusion that the entity in question is religious and the absence of which would mean that it is nonreligious. Both Greenawalt and Freeman deny that any such essence exists, arguing instead for a determination of what is or is not a religion based on the process of analogy. See infra notes 148-154 and accompanying text.
150. Both English and non-English speakers have argued that the language is itself imbued with Protestant notions of religion. For example, Josiah Strong, a 19th-century American proponent of "manifest destiny," claimed that the English language, saturated with Christian ideas, was the agent of Christian civilization throughout the world. See JOSIAH STRONG, America the Embodiment of Christian Anglo-Saxon Civilization, in OUR COUNTRY: ITS POSSIBLE FUTURE AND ITS PRESENT CRISES (New York, 1885),
151. reprinted in CHURCH AND STATE IN AMERICAN HISTORY 136, 138 (John F. Wilson & Donald L. Drakeman eds., 1987). Similarly, Anton Walburg, a German Catholic priest in Cincinnati, argued that English was so imbedded with Protestant ideas that English-speaking Catholicism could never prosper. See HUDSON, supra note 33, at 242 (discussing Walburg's views).
152. See Benson Saler, Religio and the Definition of Religion, 2 CULTURAL ANTHROPOLOGY 395, 395 (1987) (arguing that the term "religion," in its current usage, reflects primarily western, particularly Protestant, cultural traditions and experiences).
153. See 13 THE OXFORD ENGLISH DICTIONARY, supra note 123, at 568.
154. See id. We can see a humorous example of such a use of the word in the statement of Parson Thwackum in Henry Fielding's History of Tom Jones: "When I mention religion I mean the Christian religion; and not only the Christian religion, but the Protestant religion; and not only the Protestant religion, but the Church of England." HENRY FIELDING, THE HISTORY OF TOM JONES 82 (Random House 1943) (1749), quoted in KLASS, supra note 108, at 17.
155. See KLASS, supra note 108, at 22 (describing the sacred-profane dichotomy associated with the term "religion" in European-derived societies); Saler, supra note 136, at 395 (discussing the common understanding of religion as a finite set of beliefs and practices).
156. See KLASS, supra note 108, at 22.
157. See Bimal K. Matilal, Towards Defining Religion in the Indian Context, in MEETING OF RELIGIONS 31, 37 (Thomas A. Aykara ed., 1978).
158. See Saler, supra note 136, at 395 (attributing the emergence of a discrete sphere of religion, and by extension nonreligion, to the Protestant Reformation).
159. Catholic Liberation Theology, for example, establishes transcendence of the religious-secular axis as a normative goal. See GUSTAVO GUTIERREZ, A THEOLOGY OF LIBERATION 43-46, 86 (Sister Caridad Inda & John Eagleson trans., Orbis Books 1988) (1971).
160. See Freeman, supra note 22.
161. See supra Sections III.A-B.
162. See supra Section III.C.
163. See supra Section III.D.
164. Greenawalt's and Freeman's claims about the impossibility of dictionary-style definitions of religion are based upon the difficulty of discerning any "essence" of religion. Their arguments are particularly convincing when considered against the background of a general lack of consensus regarding the correct descriptive definition of religion in a whole range of disciplines. See supra notes 18-24 and accompanying text; see also note 133.
165. The term "family resemblance concepts" actually appears in section 67. WITTGENSTEIN, supra note 119, § 67, at 32; see also JOHN HICK, AN INTERPRETATION OF RELIGION 4 (1989) (discussing family resemblance concepts).
166. WITTGENSTEIN, supra note 119, § 66, at 31-32.
167. Freeman, supra note 22, at 1553. This is the way that Alston thinks we actually do go about applying the word "religion" to new cases. See ALSTON, supra note 120, at 88-89.
168. Freeman, supra note 22, at 1553.
169. See id.
170. Greenawalt, supra note 23, at 767-78.
171. This is more true of Greenawalt than of Freeman, who does recognize that the selection of the baseline is important to the outcome. See Freeman, supra note 22, at 1562 (recognizing that Judge Adams's selection of Thoreau as the paradigm of nonreligion biased his conclusion about Frank Africa's beliefs in Africa v. Pennsylvania, 662 F.2d 1025 (3d Cir. 1981)). Nevertheless, Freeman's depiction of the paradigm religious belief system remains somewhat biased in favor of "traditional" western religions. Freeman provides a list of the characteristics of the "paradigm" religious belief system. See id. at 1553. Freeman then says that "a belief system will be more or less religious depending on how closely it resembles this paradigm." Id. But it is clear that more of the characteristics he lists (for example, belief in God) are present within western religions than within their nonwestern counterparts. This leads to the conclusion that nonwestern (for example, nontheistic) religious belief systems (that meet the rest of the criteria) are inherently "less religious" than western (for example, theistic) ones.
172. To the extent that they do discuss the baseline for the comparative process, both Freeman and Greenawalt display some of the same biases that motivate the search for a definition of religion in the first place. See supra notes 18-24 and accompanying text. Greenawalt exhibits a clear western bias in his examples of "indubitably religious" organizations that may serve as baselines for comparison: Roman Catholicism, Greek Orthodoxy, Lutheranism, Methodism, and Orthodox Judaism. See Greenawalt, supra note 23, at 767. Freeman's approach, based upon characteristics drawn from a number of "Eastern and Western religions," see Freeman, supra note 22, at 1553, is an improvement over Greenawalt's, but it still favors western religions, see supra note 155.
173. Both Freeman and Greenawalt admit that their definitions present problems for judging borderline cases. See Freeman, supra note 22, at 1565; Greenawalt, supra note 23, at 816.
174. Which characteristics are relevant may vary from case to case, but clearly not all will be relevant all the time. Some may never be relevant.
175. Freeman, supra note 22, at 1565. He suggests Frank Africa's belief system as a "borderline-case paradigm." Id. "Under these circumstances, other borderline cases could qualify for protection only by having more in common with the religious paradigm than Africa had." Id.
176. Because western religion focuses so intensely on the deity, the greatest risk of judicial bias arises with respect to religions that deviate from the theistic model. Thus it is important to require judges to consider particular religions that lack the concept altogether. The requirement that the judge include nontheistic and pantheistic religions essentially stands in as a requirement that the judge consider nonwestern models of religion in making her comparison. A baseline could also be established by requiring geographical or cultural diversity in the baseline. A judge might, for example, be required to include one African, one Native-American, one Asian, and one European religion. The problem with such a geographical approach (as opposed to a conceptual approach like the one I have proposed) is that religions generally do not confine themselves within fixed boundaries. Christianity could be described as a European or Asian or African religion, depending upon one's perspective; Islam could be considered both Asian and African; and Santeria could be considered both African and American. A conceptual categorization of religion allows for a reasonably diverse baseline without the confusions that would be created by a geographical categorization.
177. A detailed comparison of this sort would undoubtedly be a time-consuming exercise. For this reason, I agree with Judge Adams and Emmerich's suggestion that the definitional question should only be addressed when it would make a difference to the outcome of the case. See, e.g., ADAMS & EMMERICH, supra note 4, at 91. Thus, for example, if a plaintiff would lose a free exercise claim even if his belief system were classified as a religion, then the court should avoid even addressing the definitional issue.
178. See supra Section III.D.
179. 662 F.2d 1025 (3d Cir. 1981). For a discussion of Africa, see supra notes 70-77 and accompanying text.
180. Africa, 662 F.2d at 1035-36.
181. Id. at 1034.
182. See id. at 1027. Africa called society "impure" and "blemished" while asserting that his raw food diet was "pure" and "innocent." Id.
183. See id. ("Africa testified that MOVE members participate in no distinct 'ceremonies' or 'rituals'; instead, every act of life itself is invested with religious meaning and significance.").
184. See id. Africa said that by living according to the teachings of MOVE, a person is put "in touch with life's vibration," a concept resembling communion with God or Nirvana. Id. Judge Adams failed to compare MOVE with a particular religion from the pantheistic tradition. Instead, Adams relied on abstract definitions of pantheism provided in The Oxford English Dictionary and in the Encyclopedia of Philosophy to deny that MOVE was pantheistic. See id. at 1033 n.16.
185. Although no one of the following three suggested bases for comparison is essential to religion, the resemblance of the belief system in question to religious belief systems in any two of these three features would strongly indicate that a belief system was religious. The resemblance in only one or in none of the suggested criteria would suggest that the belief system was not a religion. There seems to be no way to formulate a hard and fast rule here. Much depends on such nonquantifiable factors as the degree of resemblance.
186. Africa, 662 F.2d at 1027.
187. See supra note 100.
188. This is especially true when one considers that it is the failure to subject beliefs to generally accepted criteria of reasonability that makes religious minorities especially vulnerable. See supra Subsection III.A.2.
189. See Richard J. Norman, The History of Moral Philosophy, in OXFORD COMPANION TO PHILOSOPHY 586, 587 (Ted Honderich ed., 1995). A MOVE supporter at Frank Africa's trial said that "the MOVE 'religion is total'; it encompasses every aspect of MOVE members' lives." Africa, 662 F.2d at 1028 (quoting Ramona Johnson); see id. at 1027 (describing Africa's beliefs about purity achieved through the MOVE lifestyle and impurity that results from failing to do so).
190. The mandated baseline of comparison and the three negative guidelines also provide the basis for systematized appellate review of a judicial determination that a belief system is or is not a religion. Application of the analogical methodology constitutes a question of law, and is therefore reviewable by higher courts de novo. See, e.g., Professional Real Estate Investors, Inc. v. Columbia Pictures Indus., Inc., 508 U.S. 49, 67 (1993) (Souter, J., concurring) ("[T]he question . . . is purely one of law, which we are obliged to consider de novo"); Peel v. Attorney Registration & Disciplinary Comm'n, 496 U.S. 91, 108 (1990) (describing the appropriate standard of review for questions of law as "de novo"). Failure to compare the belief system in question with particular religions from the three enumerated categories of religions would constitute reversible error, as would erroneous reliance on the absence of one (or a combination) of the traits outlined in the negative guidelines. An appellate court faced with such error could either apply the methodology properly or, if the facts were insufficient for correct application, remand for further factfinding and correct application of the methodology. Further, an appellate court that disagreed with the lower court's conclusions (even if the lower court applied the test properly) should be allowed to reapply the test using the facts found by the lower court. As observed above, see supra Section III.B, the process of applying a word to a new situation is not completely determinate. That is to say, the decision to apply or not to apply the word "religion" to a new belief system is not completely determined by the concept itself. Hence, the issue of whether a judge is "right" or "wrong" in her application of the word is really a question of whether most people, when faced with the same set of facts, would apply the word in the same way. The best way to avoid "error" in the judicial application of "religion" to new belief systems is for as many different people as possible to consider the question independently. De novo review allows for a wider consideration of the proper classification for the belief system than would otherwise occur.
191. I recognize that this methodology does not completely eliminate the risk of such bias. Nevertheless, it does seem to represent an improvement over an unfettered definition by analogy. Further, the risk of bias under the proposed definition is preferable to the establishment problems presented by dictionary-style definitions and to the considerable risk of bias in the current, unregulated system of adjudication on this issue.