Substantial Burdens Imply Central Beliefs

Any society that is open to religious accommodation will want to know about the quality of the burden its laws impose on religious belief and exercise. This short essay reflects on the nature of that inquiry. It argues to speak of a substantial burden on religion is by implication to understand religion as constituted by a system, within which certain beliefs and exercises occupy different positions of relative importance or centrality.

Religious accommodations are exemptions from compliance with the law. Before granting a religious accommodation, it would seem necessary to inquire about precisely how the law interferes with a claimant’s system of religious belief and practice. And yet one of the most vexing issues in the law of religious accommodation concerns not merely the nature of a “substantial burden” on religious exercise, but even the propriety of any legal inquiry about religious burdens at all. Any assessment of the importance or centrality of a religious belief or practice within the claimant’s belief system is strictly forbidden: “Repeatedly and in many different contexts, we have warned that courts must not presume to determine the place of a particular belief in a religion or the plausibility of a religious claim.”1

 There are two crucial reasons for deference to the claimant as to the quality of the burden. First, courts simply are not competent institutions to evaluate religious beliefs and practices. As the Supreme Court put it in United States v. Lee and Thomas v. Review Board: “It is not within ‘the judicial function and judicial competence,’ however, to determine whether appellee or the Government has the proper interpretation of the Amish faith; ‘[c]ourts are not arbiters of scriptural interpretation.’”2 The argument from “incompetence” suggests precisely that courts are poor judges of what religion may require. Second, courts should defer not merely because they are poor judges of religion or are likely to make mistakes, but because even if they were good judges of religion they would risk excessively entangling church and state with too searching an inquiry. That is, their inquiries might trigger anti-establishment concerns. Thus, the Court has said that the First Amendment prohibits civil courts from interpreting “particular church doctrines” or opining on the “importance of those doctrines to the religion.”3 The Court’s understandable reticence to tell Hobby Lobby that it was wrong about its own beliefs, or that its beliefs were “flawed,” clearly suggests a reluctance to deal with issues that might entangle it in “religious and philosophical question[s]”4 or “draw [it] into impermissible questions of theology.”5

 The law of religious accommodation therefore puts courts in a bind. On the one hand, courts are required independently to evaluate, not only whether a burden on religion exists, but also whether that burden crosses what one might be forgiven for suspecting is an intentionally imprecisely defined threshold of substantiality. On the other hand, courts are required not to evaluate independently the burden or its substantiality for reasons of institutional incompetence and anti-entanglement, but instead to defer, and to defer completely, to the claimant. The more rigorously or abjectly courts defer, the less coherent the legal inquiry they are undertaking becomes; so much so that courts may lose sight of what they were supposed to be inquiring into in the first place.

 In this short essay, I propose neither to resolve this tension nor to confront the host of issues and controversies involving religious accommodation. Instead, I aim merely and more modestly to clarify the inquiry, and to address several understandings that differ from my own. I mean these clarifications neither as exercises in statutory interpretation nor as policy recommendations about religious accommodation more broadly. A society that rejects religious accommodation will not face any of these issues. But any society that is open to religious accommodation will have to confront the problem of the threshold showing a claimant must make about the burden imposed by the law.

 First, a burden on religious exercise is a weight on it—or, less metaphorically, an interference with religious exercise. This perfectly natural definition uses the very terms in the Religious Freedom Restoration Act in its Congressional Findings and Declaration of Purpose: “laws ‘neutral’ toward religion may burden religious exercise as surely as laws intended to interfere with religious exercise[.]”6 The implication is that laws may burden religious exercise just as much when the “interference” is unintentional as intentional. A legal interference may take many forms: compulsion through the law to do or not do certain things about which the claimant is unwilling, certainly, but also other acts that may make it more difficult for the claimant to exercise his religion. As Dean John Garvey has put it: “[b]elief or conduct may be commanded, recommended, rewarded, encouraged, desired, permitted, discouraged, forbidden, or punished within a claimant’s belief system.”7 A government act that interferes with the capacity of a claimant to believe or practice his faith burdens the claimant’s religion.

 Second, a burden on religious exercise is substantial if it interferes in a significant, important, or central way with the claimant’s religious system. Notwithstanding its prohibition on inquiries into centrality, the Religious Land Use and Institutionalized Persons Act (RLUIPA) text suggests exactly this answer: the substantiality of the burden is to be measured against the “system of religious belief” of which the religious exercise at issue forms a part. A system is a group of interdependent items—in this case religious beliefs and practices—that together constitute a unified whole that is greater than the sum of its individual parts.8

 Therefore, to claim that a law imposes a “substantial burden” on religious exercise is to claim that (1) the law interferes with religious exercise; (2) it does so significantly, importantly, or centrally; and (3) its significant, important, or central interference with religious exercise can be understood and evaluated by a legal decision maker against a background system of other, interconnected religious beliefs and practices.

 There can be no evaluation of the substantiality of a burden without some understanding of the place (“place,” of course, trades on the metaphor of “centrality” and “periphery”) or comparative importance of the exercise at issue within a religious system. It is possible, in theory, that a religious system might be constituted by an interconnected set of beliefs and practices, each of which is precisely equal in importance, so that no belief or practice would be more central or important than any other. It is more probable, however, that some beliefs and practices will be located at the core or center of the system, while others will lie at the periphery, and that the question of whether a law substantially burdens a particular belief or practice will depend to some degree on its position and relationship to other beliefs or practices within the religious system. Courts surely ought to defer to claimants’ understandings of their system of religious belief. But, as RLUIPA recognizes implicitly, claimants also ought to have a system of religious belief before alleging a substantial burden on it.

 Several scholars and courts resist the conclusions at steps (2) and (3) above, and they adopt several stratagems to avoid them. One stratagem it is to argue that a “burden” on religion is only the exertion of pressure or coercion on the claimant to do or not do something—the putting of a choice to the claimant with negative consequences for failure to comply with the law.9 As one court put it: the object of the substantial burden inquiry is to “assess[] the coercive impact of the government’s actions on the individual claimant’s ability to engage in a religious exercise, as he understands that exercise and the terms of his faith.”10 A second, related, stratagem is to segregate “secular” or “civil” burdens from “religious” burdens, and to argue that substantial burdens as used in accommodation statutes and cases exclusively concern an assessment of whether the “secular” or “civil” penalty is substantial.11 Professors Ira Lupu and Robert Tuttle, for example, have said that there are “two measures of substantiality”—religious and secular.12 Professor Frederick Gedicks argues in a similar vein that courts may use “neutral principles of secular law” to assess the burden question without making any evaluations of religion.13 Professor Michael Helfand likewise writes that courts may simply assess how burdensome the “civil penalty” might be without recurring at all to religion.14 The attraction of the stratagems is that, if successful, they negotiate the tension between the standard reasons for deference and the obligation of courts to inquire into the substantial burden on religious exercise.

 But they are not successful. The interpretation of a “burden” to mean only some type of coercion or pressure is unnatural. If I carry a burden, I carry a weight or a load. That weight or load may at some point coerce me to do or not do something. But it may not. It may simply interfere with, pose an obstacle to, or frustrate my ability to do or not do something. Sometimes a weight coerces. When Giles Corey cried out “more weight!” as he was slowly pressed to death in his trial for witchcraft, the state was coercing him to confess (indeed, it intended so to coerce him).15 But he still would have borne the burden of the stones if the state were not so coercing him—if he were simply being punished for his beliefs, for example—and the burden still would have interfered with his capacity to do many things (such as to live). The equation of a burden with coercion is too narrow and misses a good deal of the former’s meaning.

 That interference, rather than coercion, is the more natural interpretation of a burden is made manifest in Lyng v. Northwest Cemetery Association.16 The Supreme Court there held that the government’s plan to build a road directly through a burial site sacred for Native American religious ritual did not constitute a burden because the road did not compel or coerce anybody to do or not do anything.17 It reached this conclusion notwithstanding its own concession that the road might well “virtually destroy” the Native Americans’ capacity to practice their religion.18 In agreeing with the outcome in Lyng, Professor Chad Flanders writes:

[T]he Forest Service wasn’t really doing anything directly to the tribes, wasn’t making them do anything, wasn’t putting them to a choice between their faith and a penalty . . . if the government destroys your sacred site, but doesn’t force you to do anything, it isn’t much different than if an avalanche did the damage. It’s a bad thing, but it’s not a burden.19

 The key point at which this argument misfires—and at which the Court in Lyng was also mistaken—is in the elision of the government not “really doing anything directly to the tribes” with the government not “making them do anything” or not “putting them to a choice.”20 Forthe government surely was doing something directly to the NativeAmericans in Lyng. The government was destroying Chimney Rock, which had been used historically and by tradition as a place of piety, ritual, and spiritual solitude by the Native Americans, as a report by the Forest Service acknowledged.21

 The road burdened religious exercise because it interfered with the Native Americans’ religious practice. And it substantially burdened religious exercise because it interfered with a vital or central feature of the religious system within which the religious exercise at issue fit. True, the road did not compel the Native Americans to do anything; and this may be at least one reason to doubt that burdens are coterminous with compulsions. There is no reason to think that a law burdens religion any less when it makes the exercise of religion impossible than when it compels action or inaction inconsistent with religious commitment. The point about a natural disaster destroying Chimney Rock is a non sequitur: if a meteorite collapsed onto Chimney Rock, then the government would not have been responsible for the destruction of the Native American religion. As it was, however, there was no meteorite and the government was responsible. Justice Brennan’s dissent in Lyng rightly observes that a legal burden on religion is a law that “frustrates or inhibits religious practice.”22 And in contextualizing these frustrations and inhibitions, he takes pains to emphasize the function or place of Chimney Rock as central and essential to a mature and long-standing religious system.23

 In rejecting Justice Brennan’s view concerning a burden’s substantiality, some scholars segregate civil or secular from religious penalties. They assume that the two may be isolated and evaluated independently.24 This is the latest iteration of an old and persistent confusion, not only in academic scholarship about law and religion, but also in a certain prominent strand of liberal political philosophy: the conviction that religious reasons and secular reasons are altogether different in kind, and that the former have no place at all in the life of liberal democratic governance.  Professor Steven Smith has cogently criticized the clean separation of the secular and the religious in a related context: “bracketing” religious reasons for political judgments (1) is impossible for religious believers; and (2) inaccurately reflects their actual reasons for supporting or opposing any given policy.25 Breaking apart and isolating religious and secular reasons risks grossly misunderstanding a religious believer’s true reasons for action. It assumes, as Smith says, that “religion” may be set up “against something else that is not ‘religion[,]’”26 which certainly may be a plausible secular assumption but may run into problems from the perspective of a religious believer.

 Something similar may be said of the proposals to examine only the nature of the civil penalties imposed in evaluating a religious burden’s substantiality. Professor Helfand argues, for example, that courts assessing substantiality may ignore religious matters altogether and focus exclusively on the secular or civil penalties imposed.27 A penalty of $1 is objectively, or perhaps secularly, insubstantial. But a penalty of $1,000, or a period in jail, is objectively, or perhaps secularly, substantial.28

 The difficulty with this approach is that it does not, in the end, successfully avoid religious questions and inquiries. Indeed, it could not do so, since the basic premise of the substantial burden standard is that the religious claimant’s perspective is the one that counts. The only alternative is that the government’s perspective about religion is the one that counts, an alternative that is both conceptually and constitutionally problematic. Worse than this, because the bifurcated civil-secular/religious approach purports to sidestep religious questions but fails to do so, it may result in a wooden and possibly skewed evaluation of the burden. By bracketing religious reasons, it may misunderstand them.

 Burwell v. Hobby Lobby Stores, Inc. helpfully puts the bifurcated approach to the test.29 In the part of the decision involving the issue of substantial burden, the Court held that forcing Hobby Lobby and other companies to choose whether to comply with the contraceptive mandate or pay the statutorily prescribed penalty constituted a substantial burden.30If the Hahns and Greens and their companies do not yield to this demand,” the Court said, “the economic consequences will be severe.”31

 Yet consider two alternatives that were available to Hobby Lobby. First, suppose that Hobby Lobby had dropped coverage altogether. Some amici in the case and some prominent academics argued that it should do so—indeed, that it would be economically advantageous for it to do so.32 Had it done so, Hobby Lobby would have faced penalties of $2,000 per employee each year.33 Yet providing health insurance costs companies like Hobby Lobby a good deal more than $2,000 per employee.34 Dropping insurance might save a company money, even if the company consequently increased salaries to compensate employees for buying coverage on one of the new federal exchanges. Is putting Hobby Lobby to such a choice a substantial burden? The Court avoided the question by claiming that it could not entertain arguments that had not been raised below.35 Yet if it had, any distinction between secular or civil penalties and religious penalties would have collapsed, because there very well may be no purely secular/civil penalty at all. Saving money is not a civil punishment. But if it comes at the cost of failing to engage in religiously “commanded, recommended, rewarded, encouraged, [or] desired”36 behavior, then it might nevertheless constitute a substantial burden.

 Second, suppose that had Hobby Lobby dropped insurance coverage, the civil/secular penalty in Hobby Lobby would have amounted to approximately $26 million per year.37 Hobby Lobby has roughly $3.7 billion in annual revenue.38 The Green family likely has profits in the hundreds of millions of dollars each year.39 Many families could live comfortably on even a small fraction of those millions. And a certain number of millions of dollars is likely what the Greens have given up in profits by declining to open for business on Sundays.40 If, therefore, the Greens have already shown that degree of financial sacrifice for religious reasons, is the $26 million quantum a burden? A substantial burden? $26 million is quite a bit more than $1,000, which seemed to be the threshold of a burden’s civil/secular substantiality proposed by Professor Helfand.41 Yet how onerous is it from the perspective of a business that produces profits orders of magnitude greater than $26 million each year and that is already sacrificing millions of dollars in sales for religious reasons?

 These questions are complicated irrespective of which metrics are used, but they are virtually unanswerable without recurring to a baseline that incorporates the Greens’ religious convictions—that is, the system of beliefs within which their objection to the contraception mandate makes sense. The issue is not merely that in “a perfect world, only burdens on religious exercise that are truly substantial from the claimant’s internal point of view would generate such relief.”42 It is that even in this imperfect world, the substantial burden inquiry is incoherent without considering the religious perspective of the claimant. Indeed, the Court itself recognized that “the Hahns and the Greens . . . have religious reasons for providing health-insurance coverage for their employees.”43 It is these reasons—”the religious dimension of the decision to provide insurance”44—and not any other reasons divorced from their religious commitments that render their claim of a “substantial burden” on their religious exercise comprehensible. More than this, a test of substantial burden that segregates religious and secular reasons, and that fails to incorporate or account for the religious reasons at stake, is likely to misunderstand a religious claimant’s true reasons and motivations. To argue that the bare fact of assessing a company like Hobby Lobby a $1,000 fine would pose a substantial burden to its religious exercise strains credulity. It is to mistake a family’s, or a company’s, money for its principles.

 None of these clarifications to the substantial burden inquiry addresses important objections concerning the real dangers of judicial incompetence and excessive entanglement with religion. I consider these objections at length elsewhere.45 And as to this standard’s application to the pending nonprofit litigation against the Obama Administration’s contraception mandate,46 those cases would be easy ones on the issue of substantial burden. There is little doubt that the contraception mandate interferes with a crucial or central feature of the system of belief and practice espoused by the Little Sisters of the Poor and the other claimants.47 The claimants may lose or they may win, but if they lose, it should not be because a court concludes that they have not satisfied the threshold showing. That so many courts of appeals have concluded exactly that is a caution against any standard, such as this one, that would impose greater rigor on the substantial burden inquiry. The cure may be worse than the disease. For now, the Supreme Court has decided against deciding the substantial burden question.48 But it cannot avoid it forever.

Professor of Law and Associate Dean for Faculty Scholarship, St. John’s University School of Law.

1. Emp’t Div. v. Smith, 494 U.S. 872, 887 (1990).

2. 455 U.S. 252, 257 (1982) (quoting Thomas v. Review Bd., 450 U.S. 707, 716 (1981)).

3. Presbyterian Church v. Mary Elizabeth Blue Hull Memorial Presbyterian Church, 393 U.S. 440, 450 (1969).

4. Burwell v. Hobby Lobby, 134 S. Ct. 2751, 2778 (2014).

5. Michael A. Helfand, Identifying Substantial Burdens, 2016 U. Ill. L. Rev. (forthcoming Aug. 2016), available at, at *22.

6. 42 U.S.C. § 2000bb(a)(2) (2012).

7. John H. Garvey, Free Exercise and the Value of Religious Liberty, 18 Conn. L. Rev. 779, 785 (1986).

8. 42 U.S.C. § 2000cc (2000).

9. Chad Flanders, Insubstantial Burdens 3 (Working Paper), available at

10. Yellowbear v. Lampert, 741 F.3d 48, 55 (10th Cir. 2014).

11. See id. at 54.

12. Ira C. Lupu & Robert W. Tuttle, Secular Government, Religious People 241–42 (2014); Ira C. Lupu & Robert W. Tuttle, Symposium: Religious Questions and Saving Constructions, SCOTUSblog (Feb. 18, 2014, 11:12 AM),

13. Frederick Mark Gedicks, “Substantial” Burdens: How Courts May (and Why They Must) Judge Burdens on Religion Under RFRA, 84 Geo. Wash. L. Rev. (forthcoming 2017), available at, at *28–29.

14. Helfand, supra note 5, at *22.

15. See Stacy Schiff, The Witches: Salem 1692, at 315–16 (2015).

16. 485 U.S. 439 (1988).

17. Id. at 450–51.

18. Id. at 451–52.

19. Flanders, supra note 9, at 15.

20. Id.

21. Lyng, 485 U.S. at 442.

22. Id. at 459 (Brennan, J., dissenting).

23. Id. at 460–61.

24. See Lupu & Tuttle, supra note 12; Flanders, supra note 9, at 2; Helfand, supra note 5.

25. Steven D. Smith, The “Secular,” the “Religious,” and the “Moral”: What Are We Talking About?, 36 Wake Forest L. Rev. 487, 494–96 (2001).

26. Id. at 499.

27. Helfand, supra note 5, at *25.

28. See id.

29. 134 S. Ct. 2751 (2014).

30. Id. at 2775–76.

31. Id. at 2775.

32. See Marty Lederman, Hobby Lobby Part III—There is No “Employer Mandate,” Balkinization (Dec. 16, 2013), 

33. Hobby Lobby, 134 S. Ct. at 2776.

34. See Lederman, supra note 32.

35. Hobby Lobby, 134 S. Ct. at 2776.

36. Garvey, supra note 7, at 785.

37. Hobby Lobby, 134 S. Ct. at 2776.

38. See America’s Largest Private Companies: # 118 Hobby Lobby Stores, Forbes, (last visited Apr. 14, 2016).

39. See Brian Solomon, Meet David Green: Hobby Lobby’s Biblical Billionaire, Forbes (Sept. 18, 2012, 7:51 AM),

40. See Brief for Respondents at 21, Hobby Lobby, 134 S. Ct. 2751 (No. 13-354), 2014 WL 546899, at *8 (“All Hobby Lobby stores close on Sundays, at a cost of millions per year, to allow employees a day of rest.”).

41. See Helfand, supra note 5, at *24.

42. Gedicks, supra note 13, at *35.

43. See Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2776 (2014).

44. Id.

45. Marc O. DeGirolami, Religious Accommodations and Religious Systems (unpublished manuscript) (on file with author).

46. See Zubik v. Burwell, 578 U. S. ____ (2016).

47. Brief for Petitioner at 16–17, 20, Zubik v. Burwell, 578 U.S. ____ (2016) (No. 14-1418), 2016 WL 93988.

48. Zubik v. Burwell, 578 U. S. ____ (2016).

The full text of this Article is available to download as a PDF.