Diverse perspectives and the religion clauses: An examination of justifications and qualifying beliefs

Notre Dame Law Review

Volumn 74, Issue 5, 1999, Pages 1433-1472

  Greenawalt, Kent ^a  

^a Och Spine at New York Presbyterian Hospitals   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0033412604     PISSN: 07453515     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (117)

1. My present interest in both topics was triggered by an illuminating set of papers delivered at a Symposium on Religion and the Constitution. See Religion Symposium, 7 J. CONTEMP. LEGAL ISSUES 275 (1996). I concentrate my discussion on those papers.
2. Within any given category, it might be relevant what potential the particular publications have to contribute to significant truth; but part of the truth justification may be that no state official should decide whether particular publications are likely to promote truth.
3. See, e.g., C. Edwin Baker, Scope of the First Amendment Freedom of Speech, 25 UCLA L. REV. 964 (1978).
4. That is, the relation between justification and implementation is similar whether the inquiry about justification is broadly normative (what really are the best reasons?), historical (what did legislators or people then believe?), sociological (what do people now believe?), or some combination of these. I believe all these matters appropriately may count in interpretation. One can perhaps imagine a "specific practice" originalism that focuses only on practices that were once accepted and does not give any weight to perceived underlying justifications. If such a position is not incomprehensible, it is very foolish. In this Essay, I mainly discuss the connection between "best reasons" justifications and applications of norms, but I assume that the connection between historically and sociologically derived justifications of norms and the applications of those norms is similar.
5. I make this effort in KENT GREENAWALT, SPEECH, CRIME, AND THE USES OF LANGUAGE (1989) and KENT GREENAWALT, FIGHTING WORDS (1995).
6. I make this effort in KENT GREENAWALT, SPEECH, CRIME, AND THE USES OF LANGUAGE (1989) and KENT GREENAWALT, FIGHTING WORDS (1995).
7. In reality, some reasons for free religious exercise might be thought to permit an established state religion or religions.
8. This brief summary draws substantially from John H. Garvey, An Anti-Liberal Argument for Religious Freedom, 7 J. CONTEMP. LEGAL ISSUES 275, 283-86 (1996). See also JOHN H. GARVEY, WHAT ARE FREEDOMS FOR? 42-57 (1996); Larry Alexander, Good God, Garvey! The Inevitability and Impossibility of a Religious Justification of Free Exercise Exemptions, 47 DRAKE L. REV. 35 (1998) (doubting that a religious justification for liberty will support privileges for mistaken views).
9. This brief summary draws substantially from John H. Garvey, An Anti-Liberal Argument for Religious Freedom, 7 J. CONTEMP. LEGAL ISSUES 275, 283-86 (1996). See also JOHN H. GARVEY, WHAT ARE FREEDOMS FOR? 42-57 (1996); Larry Alexander, Good God, Garvey! The Inevitability and Impossibility of a Religious Justification of Free Exercise Exemptions, 47 DRAKE L. REV. 35 (1998) (doubting that a religious justification for liberty will support privileges for mistaken views).
10. This brief summary draws substantially from John H. Garvey, An Anti-Liberal Argument for Religious Freedom, 7 J. CONTEMP. LEGAL ISSUES 275, 283-86 (1996). See also JOHN H. GARVEY, WHAT ARE FREEDOMS FOR? 42-57 (1996); Larry Alexander, Good God, Garvey! The Inevitability and Impossibility of a Religious Justification of Free Exercise Exemptions, 47 DRAKE L. REV. 35 (1998) (doubting that a religious justification for liberty will support privileges for mistaken views).
11. One kind of nonreligious view is explicitly antireligious. Someone who regards religion as foolish superstition and believes that freedom is the best way for people to learn this crucial truth of human life might also support religious liberty, including a rule of nonestablishment.
12. A justification for religious liberty based on happiness or autonomy might be understood in a somewhat modest way. A proponent might say, "I am not talking about ultimate human good. I am concerned with the relations between a state and its citizens, and what the state should understand as the good of its citizens. The state should aim for happiness (or autonomy)." Such a view could plausibly arise out of religious beliefs themselves - that is, a full theory from a religious perspective might assign secular government this role. Even if secular political theory was the basis on which to conclude that the state should aim for happiness or autonomy, that view could be compatible in its practical implications with religious views that assigned the government no responsibility for religious truth.
13. Religious convictions typically make the quality of choices important - not just the freedom with which they are made. The belief that true freedom consists in living in accord with God's will is a fairly common religious one, but that freedom is not the freedom of choice I have in mind here. One could imagine a religious understanding that God wishes human beings to be as unconstrained as possible, and that such freedom is the proper object of human life; but that understanding would be unusual. If the proponent of autonomy happens to believe in what has come to be called "the unencumbered self," that view of individuality is also at odds with most religious beliefs. Ordinary happiness on earth is not the ultimate good for most religions. Religious believers often think that true happiness lies in leading a life that is good or right from the best religious perspective. On this view, a final compatibility exists between true religion and true happiness; but this happiness is not achieved by being sought for its own sake.
14. See generally Garvey, supra note 7, at 281-82; Douglas Laycock, Religious Liberty as Liberty, 7 J. CONTEMP. LEGAL ISSUES 313, 316-24 (1996); Suzanna Sherry, Enlightening the Religion Clauses, 7 J. CONTEMP. LEGAL ISSUES 473, 485 (1996).
15. See generally Garvey, supra note 7, at 281-82; Douglas Laycock, Religious Liberty as Liberty, 7 J. CONTEMP. LEGAL ISSUES 313, 316-24 (1996); Suzanna Sherry, Enlightening the Religion Clauses, 7 J. CONTEMP. LEGAL ISSUES 473, 485 (1996).
16. See Laycock, supra note 11, at 324-26; Ira C. Lupu, To Control Faction and Protect Liberty: A General Theory of the Religion Clauses, 7 J. CONTEMP. LEGAL ISSUES 357, 359-60 (1996).
17. It may be, however, that a government that tries to accommodate the desire of all individual religious persons and groups for religious liberty for themselves will end up with an approach that is identical to that it would take if it respected convictions about the appropriateness of general religious liberty.
18. One might ask just how far the "transposed" reasons will fit into other nonreligious reasons and how far they will be distinctive. Very roughly, I treat "respect for convictions" as a new nonreligious reason; I treat the strife avoidance and social stability that might be achieved by accommodation as aspects of other nonreligious reasons.
19. However, a conclusion that civil strife is not a serious modern danger would leave intact a more complex argument that prevention of strife was an important and valid objective in 1789 and 1866, and that continuity in constitutional government calls for us to continue to accept some rights whose justifications have weakened over time.
20. The emphasis might be on what the "adopters" intended or on how people more generally would have understood the provisions adopted.
21. Virtually everyone now agrees that no strategy of interpretation is self-evidently correct; any strategy must be defended in terms of the political values it reflects and realizes.
22. See, e.g., KENT GREENAWALT, PRIVATE CONSCIENCES AND PUBLIC REASONS (1995).
23. International human rights documents that protect religious freedom do not bar establishments of religion. See International Covenant on Civil and Political Rights, art. 18, U.N. GAOR, 21st Sess., Supp. No. 16, at 71, U.N. Doc. A/6316 (1966) (entered into force March 23, 1976); Univeral Declaration of Human Rights, art. 18, G.A. Res. 217(III)A, U.N. GAOR, 3d Sess., pt. 1, U.N. Doc. A/810 (1948).
24. International human rights documents that protect religious freedom do not bar establishments of religion. See International Covenant on Civil and Political Rights, art. 18, U.N. GAOR, 21st Sess., Supp. No. 16, at 71, U.N. Doc. A/6316 (1966) (entered into force March 23, 1976); Univeral Declaration of Human Rights, art. 18, G.A. Res. 217(III)A, U.N. GAOR, 3d Sess., pt. 1, U.N. Doc. A/810 (1948).
25. See Garvey, supra note 7, at 278.
26. Of course, unevenness of coverage may be a reason to suppose that the religion clauses were not adopted by people who conceived autonomy just as the argument does; if they had precisely this idea of autonomy, would they not have extended liberty more fully?
27. I am assuming that protecting religious liberty is not an unjust discrimination against those asserting sexual liberty.
28. See Laycock, supra note 11, at 323-24.
29. To quantify the point, suppose a legal norm covers from 35-65 in a spectrum of practices that runs from 1 to 100. Justification A covers 25-60; it is both underinclusive and overinclusive. Justification B covers 40-85; it is similarly underinclusive and overinclusive. Justification C covers 35-40 and 60-100 (but not 40-60); it is also underinclusive and overinclusive. If one asks what territory is covered by at least two of the three justifications, it is exactly 35-65. If the justifications complement each other and are of roughly equal power, legislators may sensibly provide coverage when, and only when, at least two of the justifications are in play.
30. My argument in this section closely resembles Ronald Dworkin's claims that the power of legal justifications depends on their fit with legal materials and their soundness and that assessment of convincing justifications affects conclusions about the content of norms. See, e.g., RONALD DWORKIN, LAW'S EMPIRE (1986).
31. That might include only a bar on a formally established church and on extreme forms of government sponsorship of religion.
32. This would be one example of how justifications often carry through to interpretations.
33. I discuss many aspects of the topic in GREENAWALT, supra note 18.
34. The idea of having "reasons" involves having bases that other people could recognize as having force; however, as I explain in PRIVATE CONSCIENCES AND PUBLIC REASONS, id., at chs. 3-4, one may believe people will recognize that force only if they have some personal experience (such as a religious conversion) that they cannot be persuaded to have on the basis of ordinary reasons.
35. However, the phrase "moral duty" is sometimes used in respect to how people ideally should regard their responsibilities. Thus, a vegetarian might say, "We all have a moral duty not to eat meat."
36. When I say "they gain strength," I mean they really are stronger, not just that they have more appeal.
37. Not everyone agrees about this. Some people, including many postmodernists, believe that all starting points for argument are nonrational assumptions; either no perspectives are especially reasonable, or ideas of what is reasonable do no more than reflect cultural assumptions.
38. An argument cannot be accessible to actual reasonable people unless it has components whose force has some cultural resonance. Thus, a defender of animal rights will rely on strands of analysis that do have cultural resonance. The components of an argument may have cultural resonance even if the argument as a whole does not. The distinction between cultural resonance and accessibility is, thus, more complex than the text suggests.
39. A practical consequence of the claim that more than effective advocacy is involved is this: If enough individuals to make up a majority happen to have a variety of idiosyncratic reasons for adopting a law, none of them having either cultural resonance or general rational force, then it would be wrong for the majority to coerce the entire population.
40. I provide a more elaborate example along these lines in GREENAWALT, supra note 18, at 12-22.
41. See JOHN RAWLS, POLITICAL LIBERALISM (1993).
42. John Rawls, Justice as Fairness: Political Not Metaphysical, 14 PHIL. & PUB. AFF. 223, 229 (1985).
43. However, when one gets into the comparative value of avoiding violence and achieving other values, controversial overarching views may come into play.
44. See GREENAWALT, supra note 18, at 106-20; see also KENT GREENAWALT, RELIGIOUS CONVICTIONS AND POLITICAL CHOICE 144-72 (1988).
45. This is roughly the import of an "original position" analysis combined with the justifications that would lead people to accept the implications of that analysis as reflective of justice. See RAWLS, supra note 36; JOHN RAWLS, A THEORY OF JUSTICE (1972). I should mention that Rawls does allow overarching views to give subsidiary support to positions arrived at without direct reliance on them.
46. I am assuming, of course, that their religious perspectives lead them to accept such an approach.
47. A group that is a small minority but believes a theocracy of its own religious leaders is the best form of government, whenever it can be achieved, may well agree to political principles of liberal democracy as the best for it in present conditions. But this "agreement" may constitute a prudential judgment from its own religious perspective, rather than genuine agreement about fundamental political principles.
48. One possibility I do not discuss is that religious ideas would be brought directly to bear on some aspects of religious liberty, while other aspects, say the issue of establishment, would be decided according to shared principles.
49. See Garvey, supra note 7, at 278-82.
50. We should definitely not suppose that all religious justifications will look essentially similar, given the tremendous diversity of religious understanding. Of course, a modern Baptist conception may not look too different from a modern Presbyterian conception.
51. People who do not accept premises advanced by others may also argue that, even if the premises are granted, they do not yield conclusions that are claimed.
52. I assume that official government organs may not sponsor explicitly atheist views. Part II of this Essay addresses that and related topics.
53. Some religious justifications, especially those that take a narrow view of impermissible establishments, may accept government sponsorship of some religious ideas.
54. Some private persons, such as presidents of major corporations and universities, have great visibility and may address a broad public about subjects such as whether the religion clauses should be amended to allow public school prayer.
55. See GREENAWALT, supra note 18.
56. See Garvey, supra note 7, at 278-82.
57. Nothing I say rebuts the logical possibility that there is some religious justification now unperceived or perceived only by a few people that is really much stronger than any other justification. What I say in the text would not cover that justification.
58. See Laycock, supra note 11, at 315-16.
59. I am assuming here that the particular version of nonsponsorship does not vary with the underlying justification.
60. On this subject, see supra Part III.
61. In the summer of 1997, I attended a conference discussing John Locke's defense of tolerance and the competing views of some of his contemporaries. A participant pointed out that different theories about why governments should tolerate diverse religious beliefs could lead to different consequences for the extent of religious exercise. He suggested that: (1) if one thought only that religious truth was none of the government's business, one might accept government interference with religious conscience whenever secular interests are involved; (2) if one thought only that the government should not coerce individuals about religious belief, one also would accept a rather limited protection of religious exercise; (3) if one thought government was incompetent in religious matters, one might support a somewhat wider scope for religious liberty; (4) if one thought religious conscience was inviolable and very important to individuals, one might want robust protection of religious conscience, including government accommodation to religious liberty even when moderate secular interests are involved. I thought these comments were instructive about how various theories underlying religious liberty could carry through to different visions of the scope of religious liberty, with variant implications for religious exemptions from secular laws. However, the differences in theory on which the participant focused are not the ones on which I have concentrated in this Essay. I have assumed that any complete theory (religious or not) about religious liberty needs to include the importance people attach to religious liberty. That should be sufficient to yield a strong argument for robust protection, including exemptions from some valid laws, whether or not one believes that religious conscience is really inviolable in some special sense. Thus, I do not believe the conference observations, broadly accurate as they were, undermine the main thrust of my discussion, although they do show that one needs to be cautious about concluding that different theories will have similar carry through.
62. A court might, however, say that wide acceptance of a religious justification by citizens affects how the issue should be viewed.
63. There are various responses to this gambit: namely that (1) since exemption compensates for state-imposed disadvantage, it does not approve religion; (2) it is enough for exemption that people care deeply about religion - exemption implies no state approval; (3) if some sacrifice is made in the principle of nonsponsorship, that may be justified to warrant other religious clause objectives. Notice that this third possibility raises a critical issue about the status of nonsponsorship: is it the overarching principle of the religion clauses or one among other values to be achieved? Briefly, the main answer to each of these responses is that the government does not sacrifice religious clause values if exemptions are extended to similarly situated nonreligious claimants. A failure of extension thereby constitutes a kind of implicit approval of religion, if there are significant numbers of similarly situated nonreligious claimants.
64. 398 U.S. 333 (1970).
65. Id. at 367-74 (White, J., dissenting).
66. Id. at 344-67 (Harlan, J., dissenting)
67. See 32 C.F.R. § 75.4 (1997).
68. See Laycock, supra note 11.
69. By referring to religion in the ordinary sense, I do not mean to imply that determining the borders of religion is easy, but I do not tackle that problem here. See Kent Greenawalt, Five Questions About Religion Judges Are Afraid to Ask, in LAW AND RELIGION: THE OBLIGATION OF CITIZENSHIP AND THE DEMANDS OF FAITH (Nancy Rosenblum ed., forthcoming 1999); Kent Greenawalt, Religion as a Constitutional Concept, 72 CAL. L. REV. 753 (1984) [hereinafter Greenawalt, Constitutional Concept].
70. By referring to religion in the ordinary sense, I do not mean to imply that determining the borders of religion is easy, but I do not tackle that problem here. See Kent Greenawalt, Five Questions About Religion Judges Are Afraid to Ask, in LAW AND RELIGION: THE OBLIGATION OF CITIZENSHIP AND THE DEMANDS OF FAITH (Nancy Rosenblum ed., forthcoming 1999); Kent Greenawalt, Religion as a Constitutional Concept, 72 CAL. L. REV. 753 (1984) [hereinafter Greenawalt, Constitutional Concept].
71. 494 U.S. 872 (1990). In City of Boerne v. Flores, 117 S. Ct. 2157 (1997), the Supreme Court held invalid, at least in state and local applications, a federal statute designed to undo the rule of Smith, 494 U.S. at 872.
72. My own objections to Employment Division v. Smith, 494 U.S. at 872, are summarized in Kent Greenawalt, Should the Religion Clauses of the Constitution Be Amended?, 32 LOY. L.A. L. REV. 9, 11-18 (1998).
73. See David Tracy, Kenesis, Sunyata, and Trinity: A Dialogue with Maseo Abe, in THE EMPTYING GOD: A BUDDHIST-JEWISH-CHRISTIAN CONVERSATION 135 (John B. Cobb, Jr. & Christopher Ives eds., 1990).
74. Douglas Laycock describes himself as "agnostic about matters of religion" and says, "None of the claims of the world's religions seem to me either plausible or falsifiable." Laycock, supra note 11, at 353. I do not doubt that Professor Laycock is an agnostic; but someone is also an agnostic who says: "Some religious claims are moderately plausible but not persuasive; some crucial religious claims are falsifiable but our evidence for and against them is radically inconclusive."
75. If atheism and agnosticism do not count as religions, then motivations that flow from atheism or agnosticism will be nonreligious, but they will differ from most nonreligious reasons in deriving from answers to religious questions.
76. The choice to marry would be similar if someone would not consider anyone outside the faith.
77. It is arguable whether the chooser's perception or a more objective appraisal should control. For many purposes, the latter might be more important; but the law should rarely go beyond a person's sincere understanding of his own motivation as religious or not. In any event, my central point here is that most people have a great many nonreligious reasons for action.
78. Laycock, supra note 11, at 329.
79. See id. at 330-31. It is logically possible for Congress to adopt, and for the courts to acknowledge its adopting, a narrower or broader concept of religion than is attributed to the constitutional clauses. I assume here that the sense of religion in most relevant legislation is the same as under the religion clauses.
80. See id. at 330.
81. Some scholars argue that the government should be neutral between ideas of the good life, but this view goes far beyond existing constitutional principles, and even it would not bar teaching about matters of justice. For suggestions that the Free Speech and Free Press Clauses may place limits on government speech, see MARK YUDOF, WHEN GOVERNMENT SPEAKS: POLITICS, LAW AND GOVERNMENT EXPRESSION IN AMERICA (1983); Robert D. Kamenshine, The First Amendment's Implied Political Establishment Clause, 67 CAL. L. REV. 1104 (1979); Steven Shiffrin, Government Speech, 27 UCLA L. REV. 565 (1980).
82. Some scholars argue that the government should be neutral between ideas of the good life, but this view goes far beyond existing constitutional principles, and even it would not bar teaching about matters of justice. For suggestions that the Free Speech and Free Press Clauses may place limits on government speech, see MARK YUDOF, WHEN GOVERNMENT SPEAKS: POLITICS, LAW AND GOVERNMENT EXPRESSION IN AMERICA (1983); Robert D. Kamenshine, The First Amendment's Implied Political Establishment Clause, 67 CAL. L. REV. 1104 (1979); Steven Shiffrin, Government Speech, 27 UCLA L. REV. 565 (1980).
83. Some scholars argue that the government should be neutral between ideas of the good life, but this view goes far beyond existing constitutional principles, and even it would not bar teaching about matters of justice. For suggestions that the Free Speech and Free Press Clauses may place limits on government speech, see MARK YUDOF, WHEN GOVERNMENT SPEAKS: POLITICS, LAW AND GOVERNMENT EXPRESSION IN AMERICA (1983); Robert D. Kamenshine, The First Amendment's Implied Political Establishment Clause, 67 CAL. L. REV. 1104 (1979); Steven Shiffrin, Government Speech, 27 UCLA L. REV. 565 (1980).
84. Laycock suggests that although the Founders may not have considered atheism a religion, they were aware that the "principal antagonists in religious conflict varied from time to time and place to place." Laycock, supra note 11, at 333. Laycock urges that the term religion should be read to cover the major modern conflict between believers and nonbelievers.
85. Laycock has an interesting discussion of the linguistic problem, including the characterization of humanism as religious in the first Humanist Manifesto (1933), a characterization dropped in Humanist Manifesto II (1973). See id. at 328-30.
86. 406 U.S. 205 (1972).
87. See, e.g., European Convention for the Protection of Human Rights and Fundamental Freedoms, Nov. 4, 1950, art. 9(1)213 U.N.T.S. 221, 230; JENNIFER FRIESEN, STATE CONSTITUTIONAL LAW: LITIGATING INDIVIDUAL RIGHTS, CLAIMS, AND DEFENSES 261 (2d ed. 1996) (nine states lack establishment clauses); Richard S. Kay, The Canadian Constitution and the Dangers of Establishment, 42 DEPAUL L. REV. 361 (1992); sources cited supra note 19.
88. See, e.g., European Convention for the Protection of Human Rights and Fundamental Freedoms, Nov. 4, 1950, art. 9(1)213 U.N.T.S. 221, 230; JENNIFER FRIESEN, STATE CONSTITUTIONAL LAW: LITIGATING INDIVIDUAL RIGHTS, CLAIMS, AND DEFENSES 261 (2d ed. 1996) (nine states lack establishment clauses); Richard S. Kay, The Canadian Constitution and the Dangers of Establishment, 42 DEPAUL L. REV. 361 (1992); sources cited supra note 19.
89. See, e.g., European Convention for the Protection of Human Rights and Fundamental Freedoms, Nov. 4, 1950, art. 9(1)213 U.N.T.S. 221, 230; JENNIFER FRIESEN, STATE CONSTITUTIONAL LAW: LITIGATING INDIVIDUAL RIGHTS, CLAIMS, AND DEFENSES 261 (2d ed. 1996) (nine states lack establishment clauses); Richard S. Kay, The Canadian Constitution and the Dangers of Establishment, 42 DEPAUL L. REV. 361 (1992); sources cited supra note 19.
90. Therefore, it cannot be argued that the atheist (lacking religious convictions) has only an Establishment Clause objection to forced participation in a Mass, whereas the religious Jew has both a free exercise and establishment objection to her participation.
91. Laycock seems to imply that forced participation in religious ceremonies would violate the Free Speech Clause, see Laycock, supra note 11, at 330, and I think that conclusion is correct. Whether forced observance, without active participation, would violate the Free Speech Clause is more doubtful. I assume that compelling ordinary citizens to observe ceremonies whose significance is dominantly religious would violate their free exercise rights.
92. 406 U.S. 205 (1972).
93. Formulations of the threefold establishment test of Lemon v. Kurtzman, 403 U.S. 602, 612-13 (1971), suggest that aiding or inhibiting religion may constitute an establishment. Teaching atheism would inhibit religion.
94. See Bolling v. Sharpe, 347 U.S. 497, 499 (1954). One might also conceive of the content of the religion clauses as being influenced by the Equal Protection Clause.
95. I assume here that countries with weak establishments could provide full free exercise rights to citizens. That proposition may be disputed - that is, someone might argue that a weak establishment itself involves a mild violation of free exercise.
96. I here pass over the problem that various practices in the United States, for example, "In God We Trust" on coins, "under God" in the Pledge of Allegiance, do involve mild endorsement of positive religious views. The modern Supreme Court has never acknowledged that something is both a serious sponsorship and constitutionally permitted.
97. In contrast, an approach under which atheism and agnosticism are religions requires equal treatment automatically.
98. Welsh v. United States, 398 U.S. 333, 342 (1970) (plurality opinion) (citations omitted).
99. This arrival at a conviction largely separate from religious belief seems much more likely for an insistence on shared family time than an insistence on pacificism. If a religious person becomes a pacifist, his religious beliefs and practices will almost certainly figure significantly in his arrival at that position.
100. See Michael McConnell, Accommodation of Religion, 1985 SUP. CT. REV. 1, 10-11 ("[U]nbelief entails no obligations and no observances. Unbelief may be coupled with various sorts of moral conviction . . . . But these convictions must necessarily be derived from some source other than unbelief itself . . . .").
101. The claim might be instead that an act is religiously favored or disfavored (rather than required or forbidden), or that a law encourages or discourages it (rather than requiring or forbidding it).
102. See Sherbert v. Verner, 374 U.S. 398, 399 (1963).
103. What I have said is not accurate about proclaiming agnostic beliefs and reading agnostic (and other) literature. Agnosticism could provide a reason for these acts; but any law forbidding those acts would directly violate the Free Speech and Free Press Clauses, and perhaps the religion clauses, for all persons. Agnostics would not need an exemption from a law that would validly apply to others.
104. I put aside here the possibility that some overall philosophy includes agnosticism along with other elements. Such a philosophy might be a cohesive whole with agnosticism as a central element. This would be an analogue for agnosticism of how Marxism relates to atheism. What I say about that would be relevant if such an agnostic philosophy existed; however, I am not aware of any such approach. One might think of some Unitarian-Universalist groups as a kind of religion for agnostics. When I speak of agnostics here, I mean individual, detached agnostics. For them, it is hard to see how agnosticism could provide a reason for an exemption from a valid secular law.
105. See supra note 93. A law directly forbidding defenses of atheism or the reading of atheist books would be unconstitutional. No exemption would be needed.
106. I have already shown why an atheist's exclusion of religious reasons does not entail that all the positive reasons on which the atheist relies are atheist reasons. See supra notes 90-94 and accompanying text.
107. One might resist this conclusion by building on suggestions made by Douglas Laycock about conscientious objectors. See Laycock, supra note 11, at 335. He posits a number of claims of conscientious objection made by believers and/or practitioners of traditional religions in which the connection between religious belief or practice and the objection to military service is not strong. He assumes that neither administrative officials nor courts will look too closely at how a religious belief relates to the objection (although the statutory language requires that the objection be based on a religious belief and practice). If officials think the claimant is sincere about his objection and his religious beliefs, he will qualify. Indeed, "[v]ariations in the content and derivation of religious beliefs cannot be the basis for differences in legal treatment." Id. Someone might contend that a similar approach is appropriate for the relation between atheism and agnosticism and objections by nonreligious people. This argument would be ingenious but faulty. (On first reading, I thought Laycock was making this argument, but on subsequent readings I have concluded that his position is more complex, drawing both from the strength of the beliefs and their sources. See id. at 335-36.) One problem concerns agnosticism. Whatever may be said about atheism and conscientious objection, it will typically be hard to find any positive connection between conscientious objection and agnostic views. Even for atheists the argument is unpersuasive. If Laycock is correct that in instances of sincere religious belief and sincere conscientious objection no one would worry too much about the strength of connection, this result is partly because the connection is strong in the vast majority of instances. Most religious conscientious objectors have objections powerfully connected to their religions. Most atheists who are conscientious objectors do not have objections so strongly connected to their atheism. A more crucial difficulty is that Laycock's examples are drawn from an area of law in which, according to Supreme Court decisions, the connection between religious belief (in the usual sense) and sincere objection is actually irrelevant. Under Welsh v. United States, 398 U.S. 333 (1970) (plurality opinion), any genuine conscientious objection counts as religious. As long as an objector is deeply opposed to fighting in any war, it does not matter whether his objection is based on standard religious belief and practice. If he is a sincere objector, he qualifies as religious under the plurality's reading of the statute. Claims about (ordinary) religious belief may still be relevant in a sense; if he lies about his religion, he is probably lying about being a conscientious objector. But the strength of connection between any sincere religious beliefs and the conscientious objection is legally insignificant. A fuller test of Laycock's assumption about connection turns on examining an area in which the connection of religion to the underlying claim remains important. We may consider the religious mother who does not want to work on Saturday because she needs or wants to spend time with her family, or religious people who want to withdraw children from school after the eighth grade. Constitutional law, as it stands, appears to require that both claims be based on religious belief and practice. In both circumstances, we would expect courts to inquire about the connection between religious belief and the claim. Does the mother's insistence that she not work on Saturday relate to her religious understanding? Does the parents' strong desire to withdraw their children from school derive from the parents' religion? A court would not in either instance simply assume that a sincere religious belief of some sort plus a sincere strong wish for an exemption necessarily amounted to a religious claim for an exemption. Laycock implies that for a court to inquire about the connection between a religious belief and claim to exemption would be unconstitutional. See Laycock, supra note 11, at 331. As long as a court limits itself to the religious understanding of claimants themselves, not asking what the correct view of a particular religion is, some inquiries of this sort are acceptable. They have been required under the Religious Freedom Restoration Act (RFRA), which accords an exemption only when a law imposes a substantial burden on the exercise of religion. Not every heavy burden felt by someone who happens to be a religious believer is a substantial burden on his religious exercise. (The Supreme Court's invalidation of most (all?) of RFRA has nothing to do with the appropriateness of this requirement.)
108. One might have a religious claim based on religious activity that did not depend on one's answers to religious questions.
109. See 398 U.S. 333 (1970) (plurality opinion).
110. 380 U.S. 163 (1965).
111. See, e.g., Laycock, supra note 11, at 336. Laycock talks of the "nonatheist's belief in transcendent moral obligations" that serve "the same functions in his life" as "moral beliefs and sources of derivation serve for atheists." See also Laura S. Underkuffler, Agentic and Conscientic Decisions in Law: Death and Other Cases, 74 NOTRE DAME L. REV. 1713 (1999); Laura Underkuffler-Freund, The Separation of the Religious and the Secular: A Foundational Challenge to First Amendment Theory, 36 WM. & MARY L. REV. 837 (1995).
112. See, e.g., Laycock, supra note 11, at 336. Laycock talks of the "nonatheist's belief in transcendent moral obligations" that serve "the same functions in his life" as "moral beliefs and sources of derivation serve for atheists." See also Laura S. Underkuffler, Agentic and Conscientic Decisions in Law: Death and Other Cases, 74 NOTRE DAME L. REV. 1713 (1999); Laura Underkuffler-Freund, The Separation of the Religious and the Secular: A Foundational Challenge to First Amendment Theory, 36 WM. & MARY L. REV. 837 (1995).
113. An interesting implication is that a religious believer (in the ordinary sense) could have ȧ claim strong enough to qualify as religious, even if the claim did not connect significantly to his religious beliefs. The "ultimate concern" argument, as typically put, usually compares the moral convictions of the nontheist to the central concerns of the religious believer. The implicit assumption is that the crucial convictions of the believer will be tied to his religion; for him there will be no ultimate concern not related to his religion. But people care deeply about many different things. Some religions may have relatively little bearing on some aspects of moral life. Some adherents of more comprehensive religions do not actually make the connections between religious views and moral choices that one might expect. (For example, many Christians in the United States may treat most family responsibilities as a separate moral domain, not related to their religious belief and practice, although ministers and others who think deeply about Christianity believe it bears significantly on how we should treat members of our families.) In a society in which so much moral discourse is not religious, we must acknowledge that some believers will have moral convictions they perceive as separate from their religion that will be as strong as the deep moral convictions of nonbelievers. The logic of the position that the deep est moral convictions of nonbelievers can qualify as religious leads to the conclusion that some moral convictions of believers that are not connected to their religious beliefs might also qualify as religious.
114. I discuss this problem briefly in Greenawalt, Constitutional Concept, supra note 64, at 754-55.
115. Even if the Sabbatarian admits that work is all right for emergencies, such as saving lives, her belief will be more nearly unconditional than that any nontheist would be likely to have. By "nontheists" here, I refer to nontheists who do not belong to traditional religions. A person who is a practicing Orthodox Jew and believes that refraining from work on Saturday is a crucial, uncompromisable element of religious practice might not believe in the existence of God.
116. See Perry, supra note 19, at 432-33.
117. See Welsh v. United States, 398 U.S. 333, 362-64 (1970) (plurality opinion).