Constitutional Anomalies or As-Applied Challenges? A Defense of Religious Exemptions
Constitutional Anomalies or As-Applied Challenges? A Defense of Religious Exemptions
Stephanie H. Barclay 0 1
Becket Fund for Religious Liberty 0 1
Mark L. Rienzi 0 1
Amendment Commons 0 1
the Religion Law Commons 0 1
0 Stephanie H. Barclay & Mark L. Rienzi, Constitutional Anomalies or As-Applied Challenges? A Defense of Religious Exemptions , 59 B.C.L. Rev. 1595, 2018
1 Th e Catholic University of America, Columbus School of Law
Part of the Civil Rights and Discrimination Commons; Constitutional Law Commons; First
I. THE CRITICISM OF RELIGIOUS EXEMPTIONS AS ANOMALOUS AND DANGEROUS ................. 1600
III. EMPIRICAL ANALYSIS OF RELIGIOUS VERSUS SPEECH CLAIMS.......................................... 1631
A. Methodology .................................................................................................................... 1633
B. Findings ........................................................................................................................... 1639
A. Majoritarian Reasoning of Gobitis Overruled by Barnette in the Speech Context .......... 1647
CONCLUSION ............................................................................................................................. 1653
STEPHANIE H. BARCLAY*
MARK L. RIENZI**
Abstract: In the wake of Burwell v. Hobby Lobby and now in anticipation of
Craig v. Masterpiece Cakeshop, Inc., the notion that religious exemptions are
dangerously out of step with norms ofConstitutional jurisprudence has taken
on a renewed popularity. Critics increasingly claim that religious exemptions,
such as those available prior to Employment Division v. Smith and now
available under the federal Religious Freedom RestorationAct (RFRA), are a threat
to basic fairness, equality, and the rule of lawU.nder this view, exemptions
create an anomalous private right to ignore laws that everyone else must obey,
and such a scheme will result in a tidal wave of religious claimants striking
down government action. Our Article presents an observation that undermines
these central criticisms. Far from being “anomalous” or “out of step” with our
constitutional traditions, religious exemptions are just a form of “a-aspplied”
challenges offered as a default remedy elsewhere in constitutional adjudai-c
tion. Courts regularly provide exemptions from generally applicable laws for
other First Amendment protected activity like expressive conduct that mirror
the exemptions critics fear in the context of religious exercise. TheArticle
also presents original empirical analysis, including a national survey of all
fdeeral RFRA cases since Hobby Lobby, indicating that concerns of critics about
religious exemptions have not been borne out as an empirical matter. Our
findings suggest that even after Hobby Lobby, cases dealing with religious
exemption requests remain much less common than cases dealing with other
expressive claims, and are less likely to result in invalidation of government
cations. Thus, far from creating anomalous preferential treatment that threatens
the rule of law, a religious exemption framework simply offers a similar level
of protection courts have long provided for dissenting minority rights housed
elsewhere in the First Amendment.
© 2018 Stephanie H. Barclay & Mark L. Rienzi. All rights reserved.
* Stephanie H. Barclay is Legal Counsel at the Becket Fund for Religious Liberty, and she
will be joining BYU Law School as an Associate Professor of Law, effective Fall 2018. The
authors thank Eric Baxter, Jud Campbell, Justin Collings, John Corvino, Richard Fallon, Philip
Hamburger, Ira Lupu, Jim Oleske, James Phillips, Eric Rassbach, Paul Stancil, and Robert Tuttle
for very helpful comments. The authors also thank Rachel Busick, Shaina Huleatt, and Chelise
Fox for helpful research and assistance with the empirical analysis.
** Mark L. Rienzi is Professorof Law at The Catholic University of America, Columbus
School of Law, and Senior Counsel at the Becket Fund for Religious Liberty. The views expressed
in this Article do not necessarily reflect the views of Becket or its clients.
Religious exemptions create an “anomaly” within our legaslystem—
an unfair special privilege to ignore the laws everyone else must o1bey.
Worse still, protecting the rights of diverse religious claimants in our nation
will “be courting anarchy” by turning our law into “swiss cheese” and
inviting a tidal wave of litigation.2
So goes one of the most common refrains raised by critics of religious
exemptions. Some prominent free exercise cases have traded on theses- a
sumptions, most notably the famous and controversial case ofEmployment
Division v. Smith.3 Many of the recent criticisms of religious exemptions
rely on these assumptions, both in the context of exemptions offered
constitutionally or statutorily through laws such as the Religious Freedom
Resotration Act (“RFRA”).4 And these arguments are being made with increasing
frequency and volume in the wake of Burwell v. Hobby Lobby Stores, Inc.,5
and now in anticipation ofMasterpiece Cakeshop, Ltd. v. Colorado Civil
Rights Commission.6 But are these arguments really accurate?
A closer look at religious exemptions csaes—particularly in
comparison with other types of First Amendment cas—esshows that the claim of
unfair favoritism is not correct. This Article presents the claim that religious
exemption requests are just a version of what is generally thought of as one
of the most common, modest, and preferred modes of constitutional
adjudication: the as-applied challenge. This is true regardless of whether the rie-l
1 Emp’t Div., Dep’t of Human Res. v. Smith, 494 U.S. 872, 886 (1990);see also Mary Anne
Case, Why “Live-and-Let-Live” Is Not a Viable Solution to the Difficult Problems of Religious
Accommodation in the Age of Sexual Civil Rights, 88 S. CAL. L. REV. 463, 469, 471 (2015)
(arguing that Employment Division v. Smithwas correctly decided, and that the Religious Freedom
Restoration Act (“RFRA”) both causes problems of administrability for religious exemptions and
results in problematic preferential treatment of some types of religious claims).
2 Smith, 494 U.S. at 888;JOHN CORVINO ET AL., DEBATING RELIGIOUS LIBERTY AND
DISCRIMINATION 52 (2017); see also infra notes 71–92 (surveying arguments of critics); infra notes
267–303 and accompanying text (reviewing empirical analysis of religious exemption cases).
3 Smith, 494 U.S. at 886; see also Brian A. Freeman, Expiating the Sins of Yoder and Smith:
Toward a Unified Theory of First Amendment Exemptions from Neutral Laws of General
Applicability, 66 MO. L. REV. 9, 11 (2001) (“explor[ing] the extent to which the Constitution requires
exemptions from neutral laws of general applicability in order to protect the free exercise of rie-l
gion”); Kenneth Marin, Employment Division v. Smith: The Supreme Court Alters the State of
Free Exercise Doctrine, 40 AM. U. L. REV. 1431, 1433–75 (1991) (surveying the historical
development of free exercise jurisprudence). See generally Reynolds v. United States, 98 U.S. 145, 167
(1878) (concluding that allowing religious exemptions would result in a “[g]overnment . . . only in
4 42 U.S.C. § 2000bb (2012).
5 134 S. Ct. 2751 (
6 Craig v. Masterpiece Cakeshop, Inc., 370 P.3d 272, 280–81 (Colo. App. 2015),cert. denied
sub nom. Masterpiece Cakeshop, Inc. v. Colo. Civil Rights Comm’n, No. 2015SC738, 2016 WL
1645027 (Colo. Apr. 25, 2016), cert. granted sub nom. Masterpiece Cakeshop, Ltd. v. Colo. Civil
Rights Comm’n, 137 S. Ct. 2290 (2017) (mem.).
gious exemptions are offered constitutionally or through statutes such as
RFRA. Furthermore, under this form of as-applied adjudication, courts
regularly provide identical exemptions in the context of expressive conduct
that critics fear in the context of religious exercise protections.
For example, in religious exemption cases, as in other expressive
cnoduct as-applied challenges, the decision-maker is asked to find that a
constitutional right would be infringed by a particular application of an otherwise
valid law not specifically aimed at protected activityT. he remedy in both
contexts is a court order protecitng the exercise of the constitutional right,
but otherwise leaving the law in place to apply to other circumstances that
may arise. In fact, the aspect of religious exemptions that generates most of
the criticism—the limited carve-out from a law that othrewise remains in
place to apply to others—has been widely praised elsewhere as making
asapplied challenges preferable to more aggressive constitutional remedies,
such as facial invalidation.7
Furthermore, there are deep structural similarities between atsh-e
applied challenges in the expressive realm and the religious exercise realm.
Thus, far from being problematic anomalies “in tension with other
constitutional principles,” the Supreme Court hasdescribed limited carve-outs as
“the basic building blocks of constitutional adjudication.8” When religious
exemption requests are properly understood as a-sapplied challenges, they
actually look quite pedestrian, particularly in comparison to how consuti-t
tional challenges work to protect other First Amendment interests.9
But what about the concern that providing religious exemptions will
result in our society “courting anarchy?” Is there something uniquely
pervasive and dangerous aboutreligious exemption requests? Is it true that the
diverse religious views in our country mean we will face an “endless chain
of exemption demands” that are much more expansive than other types of
First Amendment activity?10 And particularly in the wake ofHobby Lobby,
will we face a tidal wave of litigation by anendless line of religious
objectors who then become a law unto themselves and strike down government
action at every turn?
7 Of course, not all religious liberty claims are exemption requests. Some religious liberty
claims seek to strike down laws on their face, including under the Establishment Clause or if the
law engages in facial targeting under the Free Exercise Clause. Church of the Lukumi Babalu
Aye, Inc. v. Hialeah, 508 U.S. 520, 526, 531–32 (1993) (holding invalid a law restricting the
Santeria religious ritual of animal sacrifice); Edwards v. Aguillard, 482 U.S. 578,–85180(1987)
(holding facially invalid a law requiring teaching “creation sciences” because it lacked a “clear
8 Gonzales v. Carhart, 550 U.S. 124, 167–68 (2007); William P. Marshall,Bad Statutes Make
Bad Law: Burwell v. Hobby Lobby, 2014 SUP. CT. REV. 71, 74.
9 See infra notes 267–303 and accompanying text.
10 See infra notes 152–178 and accompanying text.
Our original empirical analysis suggests otherwise.11 The data suggests
that expressive claims are much more pervasive than religiousclaims, both
in absolute terms and as a percentage of all reported cases. We also provide
a new survey of all federal RFRA decisions since Hobby Lobby, which
analyzes how the Supreme Court’s decision Hinobby Lobby impacted win
rates of reported religiousexercise cases. The data does not demonstrate a
dramatic increase in the win rate of religious exercise litigants under RFRA.
This may be explained, in part, because there are important legal limitations
on successful religious claims, like the requirement of sincerity.
So what explains the treatment we give to religious exemptions
mcopared to other First Amendment exemptions? One clue likely comes from the
divergent 1940s cases of Minersville School District v.Gobitis and West
Virginia State Board of Education vB. arnette.12 It may be that the Supreme
Court’s embrace of a majoritarian approach to religious exercise Ginobitis
(as later affirmed by Smith) leads critics of religious exemptions to view
reilgious exercise rights more skeptically, even though neither the jurisprudential
comparison to similar rights nor the empirical data justify such differential
criticism and alarm. In fact, the Court firmly rejected theGobitis approach
under Barnette, and instead opted for a strong countemr-ajoritarian
framework for expressive rights1.3 Viewed in this context,a religious exemption
scheme such as RFRA is simply a restoration of a pluralistic protection of
dissenting rights through as-applied challenges.
Part I of this Articlesurveys scholarly criticisms of religious
exemptions as a threat to equality and the rule of law1.4 Part II sets forth an
alternate view of religious exemptions as narrow, as-applied challenges that are
elsewhere viewed as the preferred mode of constitutional adjudication. This
Part explores how in the particularly relevant comparator context of
mcopelled speech, courts regularly provide exemptions from generally
appliacble laws that mirror the exemptions critics fear in the context of religious
exercise.15 Part III discusses the authors’ original empirical analysis of
religious versus speech claims to illustrate that, contrary to scholarly
apprehension, Hobby Lobby has not had a dramatic effect on government win rates in
religious exemption challenges, nor have religious claims undergone a
darmatic expansion in volume following Hobby Lobby. If anything, the volume
11 See infra notes 267–289 and accompanying text.
12 W. Va. State Bd. of Educ. v.Barnette, 319 U.S. 624, 640–42 (1943) (holding that
compelling students to salute the flag and say the pledge of allegiance, despite their religious objections,
violated their First Amendment rights); Minersville Sch. Dist. Bd. of Educ. vG.obitis, 310 U.S.
586, 598–600 (1940) (holding constitutional a compulsorfylag salute in school),overruled by
Barnette, 319 U.S. 624.
13 Barnette, 319 U.S. at 641.
14 See infra notes 17–67 and accompanying text.
15 See infra notes 68–232 and accompanying text.
of these cases appears to be slightly decreasing as a percentage of ael-l r
I. THE CRITICISM OF RELIGIOUS EXEMPTIONS AS ANOMALOUS AND DANGEROUS
A. Smith and Initial Backlash
Critics of religious exemptions frequently rely on Justice Antonin
Scalia’s majority opinion in Smith.17 There, the Supreme Court rejected the
notion that the Free Exercise Clause requires religious exemptions from
generally applicable and neutral law1s8. To overrule prior precedent thata-f
vored such exemptions, Justice Scalia relied on two justifications thaet- r
main influential among modern critics of religious exemptions: (
exemptions allow objectors to unfairly avoid compliance with an ohterwise
valid law, and (
) allowing exemptions in our radically diverse society
would court anarchy.19
First, Justice Scalia argued that religious exemptions were tantamount
to “a private right to ignore generally applicable laws,” which would result
in “a constitutional anomaly.”20 A person’s religious views,Justice Scalia
explained, do not “excuse him from compliance with an otherwise valid law
prohibiting conduct that the State is free to regulate.”21 Viewing exemptions
through this lens, Justice Scaliaframed the issue in the case as “decid[ing]
whether the Free Exercise Clause of the First Amendment permits the State
of Oregon to include religiously inspired peyote use within the reach of its
general criminal prohibition on use of that drug . . . .”22 This issue arose in
the context of a state’s decision to deny unemployment benefits to a Native
American person fired for violating a state prohibition on the use of peyote,
even though the use of the drug was part of a religious ritual.Justice Scalia
argued that the “government’s ability to enforce generally applicable
prohibitions of socially harmful conduct, like its ability to carry out other aspects
of public policy, ‘cannot depend on measuring the effects of a governmental
action on a religious objector’s spiritual development.’”23
16 See infra notes 233–341 and accompanying text.
17 Smith, 494 U.S. at 874, 878–89.
18 Id. at 882. The Supreme Court has since whittled away at this principle fromSmith. See,
e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 188–90 (2012)
(recognizing an exemption from the generally applicable Americans with Disabilities Act).
19 Smith, 494 U.S. at 879, 888.
20 Id. at 886.
21 Id. at 878–79.
22 Id. at 874.
23 Id. at 885 (quoting Lyng v. Nw. Indian Cemetery Protective Ass’n, 485 U.S. 439, 451
Justice Scalia even went so far as to argue that valid, generally apip-l
cable laws that did not target religion could not really burden
religiousxeercise. He noted that the Free Exercise Clause would certainly prohibit a
law that specifically targeted a religious group or practice, by doing things
like banning statues used for worship purposes2.4 But it would be quite
another thing, he argued, for the Free Exercise Clause to create an exemption
from a law that was not “specifically directed at their religious practice” and
when the law is otherwise constitutional when applied to others who engage
in the practice for non-religious reasons.25 A generally applicable and
neutral law, according to Justice Scalia, could no more burderneligious
exercise than it could “abridg[e]the freedom .. . of the press” if the law does
not target such constitutional activity; instead, the burden on them is an
“incidental effect of a generally applicable and otherwise valid provisio2n6.”
Justice Scalia thus explained that heightened scrutiny was “inapplicable” to
a challenge to “an across-the-board . . . prohibition on a particular form of
In a portion of the opinion addressing the tension this approach created
with existing law, Justice Scalia acknowledged several court decisions in
other constitutional contexts. This included speech, press, and association
cases where the “First Amendment bars application of a neutral, generally
applicable law to religiously motivated action”, or else “compelled
expression” that implicates religious freedom.28 In an attempt to distinguish these
cases, Justice Scalia developed his hybrid rights theory and observed that
these cases involved free exercise claims alongsidoether constitutional
rights, or else werebased solely upon freedom of speech2.9 Justice Scalia
also argued that employing heightened scrutiny “before the government
may regulate the content of speech .. . is not remotely comparable to using
it for” religious exemptions3.0 He argued that heightende scrutiny in the
speech context, which allows the “unrestricted flow of contending speech,”
is merely a constitutional norm, whereas a religious exemption would result
in a constitutionally anomalous “private right to ignore generally applicable
laws. . . .”31
Second, Justice Scalia argued that religious exemptions are particularly
problematic in “a cosmopolitan nation made up of people of almost every
24 Id. at 877–78.
25 Id. at 878.
27 Id. at 884–85.
28 Id. at 881–82.
30 Id. at 886.
conceivable religious preference,” and this “danger increases in direct
porportion to the society’s diversity of religious beliefs.”32 “[W]e cannot afford
the luxury of deeming presumptively invalid, as applied to the religious
objector, every regulation of conduct that does not protect an interest of the
highest order,” he explained.33 Applying heightened scrutiny for such
“religious divergence” would “open the prospect of constitutionally required
religious exemptions from civic obligations of almost every conceivable
kind.”34 Some of the contexts Justice Scalia used as examples where
prbolematic exemptions could be requested included drug laws, traffic laws, or
animal cruelty laws.35 Thus, in the view of Justice Scalia, “adopting such a
system would be courting anarchy.”36
Justice Scalia acknowledged thatother First Amendment rights, like
free speech and press,sometimes bar “application of a neutral, generally
applicable law to religiously motivated action.”37 But although he explained
descriptively this difference between religious and other First Amendment
rights under his regime, he never justifiednormatively why this distinction
between religious exercise and other First Amendment rights should exist.
When the Supreme Court’s decision inSmith was handed down, it
received widespread criticism. One scholar noted that in academia generally,
“criticism of Smith . . . has become commonplace.”38 The criticism was not
32 Id. at 888 (internal citation and quotation marks omitted).
35 Id. at 889.
36 Id. at 888.
37 Id. at 881.
38 See Ira C. Lupu, Employment Division v. Smiathnd the Decline of Supreme Cou-rt
Centrism, 1993 BYU L. REV. 259, 260 n.9 (collecting sources that discuss potential implications
of Smith); see also Church of the Lukumi Babalu Aye, Inc. v. Hialeah, 508 U.S. 520, 559 (1993)
(Souter, J., concurring in part, concurring in the judgment) (noting that there are doubts as to
whether “the Smith rule merits adherence”); Richard F. Duncan, Free Exercise Is Dead, Long Live
Free Exercise: Smith, Lukumi and the General Applicability Requirement, 3 U. PA. J. CONST. L.
850, 851–56 (2001) (arguing that despite the conclusion and subsequent criticism Somfith free
exercise is still “alive and well”); James D. Gordon IIFI,ree Exercise on the Mountaintop, 79
CALIF. L. REV. 91, 114 (1991) (“What is there to admire [aboSumtith]? The Court wanted to
reach its result in the worst way, and it succeeded.”); Douglas LaycocTkh,e Remnants of Free
Exercise, 1990 SUP. CT. REV. 1, 2–3 (arguing that Smith was incorrectly decided based on
precedent and original intent); Marin,supra note 3, at 1475–76 (1991) (arguing that “theSmith Court
has rendered the free exercise clause impotent”); Michael W. McConnell, Free Exercise
Revisionism and the Smith Decision, 57 U. CHI. L. REV. 1109, 1111 (1990) (“There are many ways in
which to criticize the Smith decision. . . . Smith is contrary to the deep logic of the First
Amendment.”); Harry F. Tepker, Jr.,Hallucinations of Neutrality in the Oregon Peyote Case, 1A6M.
INDIAN L. REV. 1, 11–26 (1991) (critiquing Justice Antonin Scalia’s use of history and precedent
in the Smith decision); Carol M. Kaplan, Note,The Devil Is in the Details: Neutral, Generally
Applicable Laws and Exceptions from Smith, 75 N.Y.U. L. REV. 1045, 1046 (2000) (arguing that
“[Smith] was not well crafted and was based on a mischaracterization of precedent”); Mark J.
Rosen, Comment, Native Americans May Be Denied State Unemployment Benefits for Ceremonial
limited to academia—religious, political, and civil rights leaders also joined
in from across the political spectrum, with Ted Kennedy, Bill Clinton, and
the ACLU joining forces with Orrin Hatch antdhe United States
Conference of Catholic Bishops in efforts to repair what they saw as damage done
This backlash resulted in the nearly unanimous passage of RFRA
toerinstate a religious exemption framework.40 When RFRA was passed in 1993,
the bill was supported by one of the broadest bipartisan coalitions in recent
political history, with sixty-six religious and civil liberties groups, including
Muslims, Sikhs, Humanists, and secular civil liberties organizaotins such as
the ACLU and Americans United for Separation of Church and State.41
B. Smith’s Academic Resurgence
In light of recent h-obtutton religious exemption cases likHeobby
Lobby,42 and now Masterpiece Cakeshop,43 legal academic support for
Ingestion of Peyote Without Violating First Amendment’s Free Exercise Clause,23 RUTGERS L.J.
209, 230 (1991) (arguing that “[t]he Court’s cursory disregard for precedent in arriving at its
holding suggests the Court’s willingness to forego longs-tanding doctrinal jurisprudence, and
subsittute its new, stultified vision of the Free Exercise liberties ‘protected’ by the Constitution”);
Whitney Travis, Note, The Religious Freedom Restoration Act and Smith: Dueling Levels of
Constitutional Scrutiny, 64 WASH. & LEE L. REV. 1701, 1707 (2007) (noting that “Smithwas met with
enormous criticism”). To be sureS,mith did have some contemporaneous academic defenders.
See, e.g., Philip A. Hamburger,A Constitutional Right of Religious Exemption: An Historical
Perspective, 60 GEO. WASH. L. REV. 915, 916 (1992) (questioning the originalist historical
evidence in favor of religious exemptions); William P. Marshall, In Defense of Smith and Free
Exercise Revisionism, 58 U. CHI. L. REV. 308, 309 (1991)(defending “Smith’s rejection of
constiuttionally compelled free exercise exemptions without defending Smith itself”).
39 See Brett H. McDonnell, The Liberal Case forHobby Lobby, 57 ARIZ. L. REV. 777, 784
(2015) (“Given the politics currently surrounding RFRA, it should come as little surprise that
many religious organizations objectedto the decision inSmith. It is more surprising that many
liberal civil rights organizations objected as wel—lthe ACLU, Americans United for the
Sepaartion of Church and State, People for the American Way, and Americans for Democratic Action
came together in a powerful coalition that proposed a statutory overturning ofSmith. . . . Thus, at
the Court, in Congress, and in the White House, a large number of liberals supported the principle
of religious liberty embodied in RFRA.”).
40 42 U.S.C. § 2000bb (2012); Douglas Laycock & Oliver S. Thomas,Interpreting the
Religious Freedom Restoration Act, 73 TEX. L. REV. 209, 210, 243–44 (1994); Ira C. Lupu, The
Failure of RFRA, 20 U. ARK. LITTLE ROCK L.J. 575, 588 (1998) (noting that “RFRA is federal law,
supported by a near unanimous House and Senate and an enthusiastic President”); Travisu,pra
note 38, at 1707 (noting that “[i]n direct response toSmith, Congress passed the RFRA in 1993
with nearly unanimous support in both the House and the Senate”).
41 Laycock & Thomas, supra note 40, at 210 n.9; Travis Gasper, Comment, A Religious Right
to Discriminate: Hobby Lobby and “Religious Freedom” as a Threat to the LGBT Community, 3
TEX. A&M L. REV. 395, 416 (2015) (noting that “the groups most active in pushing for passage of
the 1993 RFRA were ideologically left of center”).
42 134 S. Ct. 2751 (
43 Craig v. Masterpiece Cakeshop, Inc., 370 P.3d 272, 280–81 (Colo. App. 2015), cert. denied
sub nom. Masterpiece Cakeshop, Inc. v. Colo. Civil Rights Comm’n, No. 2015SC738, 2016 WL
RFRA has declined, while the once-maligned reasoning of Smith has
recently resurfaced. As many commentators have observed, “the space for
accommodating religious objections to general legal obligations is
increasingly contested in contemporary American legal, political, and ethical
discourse.”44 In particular, the question of whether “demands for exemptions
from generally applicable laws” are justified is “an issue that has recently
assumed increased significance . . . .”45
For example, in the wake of recent RFRA caseso,ne scholar recently
advanced “the normative view that Smith was correctly decided and that . . .
[RFRA] was a mistake.”46 Another argued that “Smith was decided the way
it was for a reason.”47 Another defended the principal holding of Smith that
when you have “a neutral state law that applies to ever”yonthei,s law
should apply to all without religious exceptions.48 And still another argued
that Justice Scalia correctly decided Smith in holding that “[r]eligion is not a
get-out-of-the-law-free-card.”49 Notably, most of these critics do not object
in principle to protecting religious liberty; rather, the objections of critics
generally boil down to the same two primary arguments Justice Scalia relied
on in Smith.
First, critics argue that religious exemptions from otherwise valid laws
provide an anomalous remedy—essentially an excuse to avoid obeying the
1645027 (Colo. Apr. 25, 2016), cert. granted sub nom. Masterpiece Cakeshop, Ltd. v. Colo. Civil
Rights Comm’n, 137 S. Ct. 2290 (2017) (mem.).
44 Hillel Y. Levin et al.,To Accommodate or Not to Accommodate: (When) Should the State
Regulate Religion to Protect the Rights of Children and Third Parties?,73 WASH. & LEE L. REV.
915, 918 (
); see also Frederick Mark Gedicks, One Cheer forHobby Lobby: Improbable
Alternatives, Truly Strict Scrutiny, and Third-Party Employee Burdens, 38 HARV. J.L. & GENDER
153, 176 (2015) (expressing concern aboutBurwell v. Hobby Lobby’s “unprecedented expansion
of permissive accommodation” as an “‘o-oputt’ from generally applicable legislation”); Paul
Horwitz, The Hobby Lobby Moment, 128 HARV. L. REV. 154, 170 (
) (arguing that“[a]
substantial body of opinion on this issue has moved from the view thSamtith erred grievously by
rejecting the prior regime of free exercise exemptions from generally applicable la.w., t.o a
broader questioning of religious accommodations altogether”); Maimon SchwarzschildD,o
Religious Exemptions Save?, 53 SAN DIEGO L. REV. 185, 187 (
) (“[S]upport for religious
exemptions may now be breaking down along ideological-political lines. . . . [O]utright opposition to the
idea of religious exemption was uncommonly met with until very recently, either in politics or in
the legal literature.”); Elizabeth Sepper,Reports of Accommodation’s Death Have Been Greatly
Exaggerated, 128 HARV. L. REV. F. 24, 24 (
http://harvardlawreview.org/wpcontent/uploads/2014/11/vol_128_Sepper.pdf [https://perma.cc/4W9U-DKWT] (claiming a
growing consensus “against granting religious exemptions from generally applicable laws”).
45 Laura M. Weinrib, Freedom of Conscience in War Time: World War I and the Limits of
Civil Liberties, 65 EMORY L.J. 1051, 1054 (
46 Case, supra note 1, at 469.
47 Marshall, supra note 8, at 74.
48 Haley Gray, What You Need to Know About theMasterpiece Cakeshop Case, 5280: DENVER
MAG. (June 28, 2017),
49 CORVINO ET AL., supra note 2, at 31.
laws that apply to everyone else. They have been describedas “a troubling
form of relief—special exemptions from neutral laws for a limited class of
beneficiaries—that is in tension with other constitutional principle,”s or as
not “in the tradition of American liberty.”50 Under this view, exemptions are
problematic because they allow religious objectors to avoid “play[ing] by
the same rules as everyone else.5”1 Thus, if a given law is justifiedat all,
then the law ought to be appliedconsistently.52 Offering exemptions, one
scholar argues, would “result in a kind of ‘Swiss cheese’ law.5”3 Professor
Frederick Gedicks asserts that RFRA defies common sense and the
constitution by giving religious believers “a free pass to ignore laws that bind
everyone else.”54 Professors Ira Lupu and Robert Tulett similarly argue that
extending “strict, right-sprotective review to laws that imposed an
incidental burden on religious experience thus elevated religious freedom to a
preferred position among First Amendment rights, rather than assimilating
the Free Exercise Clause with its counterpart rights of speech and press.”55
The view that religious exemptions are anomalous quickly leads to the
conclusion that recipients of such exemptions are getting preferential
treatment under the law, or as one scholar asserted,“disconcerting favoritism”
for religious objectors5.6 Some raise concerns that religious
accommoadtions would result in particularly problematic special treatment in the
ncotext of anti-discrimination laws, including the public accommodation laws
at issue in Masterpiece Cakeshop.57 Other scholars echo these allegations of
special privilege for religion under a religious accommodation scheme.58
Second, critics argue that religious accommodations are uniquely
prevasive, such that they deteriorate the rule of law and risk
anarch,yparticularly post Hobby Lobby. In this vein,Professor Case argues thatHobby
Lobby has “open[ed] up the floodgates to a host of new potential claims for
religious exemption by a host of different kinds of service providers.”59
Professor Leslie Griffin argues that because ofHobby Lobby, the “broad
reading of RFRA . . . will encourage many future lawsuits and undermine more
civil liberties.”60 She also asserted that “[a]lmost anything can btuerned
into a claim of ‘cooperation with evil’” and “all federal laws are now bs-u
ject to challenge, with the possibility of every citizen becoming ‘a law unto
himself’ until the rule of law is undermined.6”1 Professor Elizabeth Sepper
tions for religious freedom, starting with RFRA and its offshoots); Samuel R. BagensTtohse,
Unrelenting Libertarian Challenge to PublicAccommodations Law, 66 STAN. L. REV. 1205,
); Case, supra note 1, at 485 (noting that “[a]dvocates of exemptions from public
accommodation laws for service providers who refuse to provide flowers or cake for same-sex
wedding celebrations have yet to explain whether and why the claims of these Christian bakers
and florists are more worthy of accommodation” than other groups); LucienDJh. ooge, Public
Accommodation Statutes, Sexual Orientation and Religious Liberty: Free Access or Free
Exercise?, 27 U. FLA. J.L. & PUB. POL’Y 1, 58 (
) (arguing that “any conflicts between religious
liberty asserted by secular businesses and access to goods and services must be resolved in favor
of the government’s compelling interest in guaranteeing full and non-discriminatory access for all
persons,” and that this “result does not denigrate religion”) (internal footnote omitted); Martin S.
Lederman, Reconstructing RFRA: The Contested Legacy of Religious Freedom Restoration, 125
YALE L.J. F. 416, 419 (
[https://perma.cc/R72J-26CV] (arguing that “there is widespread fear in some quarter—sand
presumably hope in others—that such claims might become a template for similar claims, pursuant to
federal or state RFRAs or analogous state constitutional provisions, for religious exemptions from
laws that prohibit discrimination in employment, or in the provision of public accommodations, on
the basis of sexual orientation”); Douglas NeJaime & Reva B. Siegel,Conscience Wars:
ComplicityBased Conscience Claims in Religion and Politi,cs124 YALE L.J. 2516, 2561–64 (2015) (noting
concern that religious exemptions could be granted from public accommodations laws through
“complicity-based conscience claims”); Leslie C. Griffin, If Conestoga Wins, Watch Out Civil Rights,
HAMILTON & GRIFFIN ON RIGHTS (Mar. 24,
2014),http://hamilton-griffin.com/if-conestoga-winswatch-out-civil-rights/ [https://perma.cc/JBR8-GFVA]. But see Sepper, supra note 44, at 26 (noting
that “[p]ublic accommodations laws generally apply with full force to all businesses serving the
public, religiously affiliated or not”).
58 See Michele Goodwin & Allison M. Whelan,Constitutional Exceptionalism, 2016 U. ILL.
L. REV. 1287, 1329 (arguing that religious accommodations result in a religious “exceptionalism”
where “‘religiously based rights’ trump all other constitutionally derived interests” and result in
“problematic norms”); Schwarzschild,supra note 44, at 199 (noting that “exemptions and
accommodations puts religious people in the invidious position of demanding special privileges”
that is particularly troubling “in an egalitarian societyw, here a core idea is rejection of special
59 Case, supra note 1, at 469, 487.
60 Leslie C. Griffin, Hobby Lobby: The Crafty Case ThatThreatens Women’s Rights and
Religious Freedom, 42 HASTINGS CONST. L.Q. 641, 673 (2015).
61 Id. at 687–88; see also Erwin Chemerinsky & Michele Goodwin, Religion Is Not a Basis
for Harming Others: Review Essay of Paul A. Offit’s Bad Faith: When ReligioBuselief
Undermines Modern Medicine, 104 GEO. L.J. 1111, 1133–34 (
) (“This decision will lead to much
broader challenges. Christian Scientists, for example, will claim that they do not have to provide
argues that “[t]he Hobby Lobby decision throws open the courtroom door to
corporations and hands them the no w-powerful weapon of corporate
cnoscience to fight off regulation that protects the full and equal citizenship of
the people.”62 Professor Corvino raises concerns that the religious activities
that receive protection for “exemption and accommodation purposes are
expansive and expanding” as a result ofHobby Lobby.63 Corvino explains
that the “pervasiveness” and “endless variety of religious scruples,” provide
a strong motive not to have an extensive“exemption regime.”64 Thus,
Corvino concludes that it is “precisely for that reason that Justice Scalia opined
[in Smith] that, in a religiously diverse nation, any system requiring strict
scrutiny for laws burdening religious beielfs is ‘courting anarchy.’”65
Professor Lupu has argued that “a general regime of judicial exemptions is a
lawless, sometimes unconstitutional, and pervasively unprincipleda- ch
rade.”66 Numerous others have expressed similar administration concerns.67
II. RELIGIOUS EXEMPTIONS UNDERSTOOD AS AS-APPLIED CHALLENGES
The question of whether to provide as-applied exemptions from
generally applicable laws can be seen as a broader political question about how
our pluralistic society should treat dissenting views porractices. Should
society generally demand conformity to general policies preferred by the
majority, or should it take a “live and let live” approach by allowing
minority and nonconformist groups and individuals to live their lives and order
their communities as they see fit where possible68? That longstanding
debate continues to rage in academia, and will for decades to come6.9 But this
Part illustrates that under Employment Division v. Smith’s framework, courts
are likely to answer the question in favor of accommodating divergent
imnority positions when it comes to speech-based rights, but not religion.70
68 Levin et al., supra note 44, at 925.
69 Some scholars note, “for those Progressives who had confronted the costs of
countermajoritarian constitutionalism head on, there was ample reason to interrogate an extension of individual
rights.” Weinrib, supra note 45, at 1136; see also Charles M. Freeland, The Political Process as
Final Solution, 68 IND. L.J. 525, 526 n.11 (1993) (collecting sources and describing the school of
thought viewing reliance on political process instead of robust bill of rights protections as the long
awaited solution to the “countermajoritarian problem” that has plagued the “democracy as
mateiral equality”). Still other scholars recognize that robust individual freedoms are critical, even for a
healthy democracy, and cannot merely be what is “left over” after the political process is
complete. KENNETH L. KARST, BELONGING TO AMERICA: EQUAL CITIZENSHIP AND THE
CONSTITUTION 196–201 (1989) (arguing that denying constitutional rights excludes certain groups from
fully belonging to the American people); Charles A. Reich, The Individual Sector, 100 YALE L.J.
1409, 1412 (1991); see also Owen Fiss, A Life Lived Twice, 100 YALE L.J. 1117, 1118 (1991)
(praising judicial protection of individual liberty against majoritarian will); Michael KJ.larman,
Rethinking the Civil Rights and Civil Liberties Revolutions, 82 VA. L. REV. 1, 19 (1996) (pointing
out that many scholars thinkBrown proves that courts are “countermajoritarian heroics” who
protect minority rights); David Luban,The Warren Court and the Concept of a Righ,t 34 HARV.
C.R.-C.L. L. REV. 7, 8 (1999) (same); Michael J. Perry,Protecting Human Rights in a
Democracy: What Role for the Courts?, 38WAKE FOREST L. REV. 635, 637 (2003) (questioning whether
the role of the judiciary in protecting entrenched human rights is appropriate); Jeremy Waldron,A
Rights-Based Critique of Constitutional Right,s 13 OXFORD J. LEGAL STUD. 18, 19–20 (1993)
(questioning our deference to some rights in comparison to others); Rebecca E. Zietlow,The
Judicial Restraint of the Warren Court (and Why It Matters,)69 OHIO ST. L.J. 255, 259 n.13 (2008)
(observing over 500 law review articles “written in the past twenty years advocating the propoi-s
tion that courts should protect minorities against the will of the majority”).
70 See also Corinna Barrett Lain, Three Supreme Court “Failures” and a Story of Supreme
Court Success, 69 VAND. L. REV. 1019, 1072 (
) (“The Court’s self-conception of its role as a
countermajoritarian protector has helped it stretch to its counetrmajoritarian limits, at least in
certain contexts. Here, several of the Supreme Court’s First Amendment cases come to mind; its
protection of flag burning, cross burning, and Ku Klux Klan rallies as freedom of expression are
prime examples. . . . [M]uch work still needs to be done on why the Justices embrace their
countermajoritarian role in some contexts and not others.”).
A. As-Applied Challenges Such as Religious Exemptions Are the
Preferred Mode of Constitutional Adjudication
Relying on Smith, Professor Corvino argues that if a given law is
justified at all, then the law ought to be applied “consistentl7y1.”Offering
exemptions, he argues, “result[s] in a kind of ‘Swiss cheese’ law.”72 Thus, if a
law is facially valid, it ought to apply universally. As a result, Professor
Corvino and others argue that a judgment that invalidates a law only in one
circumstance, but leaves the law otherwise intact, creates an anomaly
resulting in “special rights” for objectors “in tension with othecronstitutional
But such a judgment can be described in much more positive term—s
as a “modest,”74 “normal,”75 “surgical,”76 “narrow,”77 and “logically
primary”78 method that comprise “the basic building blocks of constitutional
adjudication.”79 These latter terms are precisely how both theSupreme Court
and leading commentators describe as-applied adjudication.
The Supreme Court has explained that “the normal rule is that partial,
rather than facial, invalidation is the required course, such that a statute may
be declared invalid to the extnet that it reaches too far, but otherwise left
intact.”80 Thus, when examining a statute’s constitutionality the
Couratttempts to narrow its holding to address the specific problem, or “to sever its
problematic portions while leaving the remainder intact.”81 Such a principle
flows from the “axiom that a ‘statute may be invalid as applied to one
state of facts and yet valid as applied to another8.2” This approach also
allows the Court to address more concrete facts and to do the least damage to
the rule of law as envisioned by the original drafters8.3 As Justice Stevens
put it, when the Court strikes down statutes facially, rather than as-applied,
“[t]he Court operates with a sledge hammer rather than a scalpel.”84
Many scholars also note the Roberts Court’s preference for as-applied
challenges instead of facial ones.85 This preference has manifested itself in
many different legal contexts, including “abortion rights, Congress’ns- e
forcement power under Section 5 of the Fourteenth Amendment, and
cmapaign finance.”86 Professor Richard Fallon has explained that this
jurispurdential preference for limited invalidations that leave a statute otherwise
intact generally relies on a three-part rationale: (
) the constitutional
principle of “avoiding unnecessary or prematuredecisions of constitutional
sisues” where possible; (
) the fact that meanings of statutes are often best
specified “through a series of fact-specific, case-by-case decisions”; and (
the reality that constitutionally invalid applications of statutes often “could
be severed or separated from valid ones.”87 Scholars disagree about the
frequency with which the Supreme Court actually employs as-applied versus
facial analysis to strike down statutes, but there is little debate that
applied invalidation of laws generally involves a “surgical severing” of
constitutionally infirm aspects of the rule.88
What has generally gone unnoticed is that judicia-lclyreated religious
exemptions are functionally a species of a-sapplied adjudication. This is true
regardless of whether the exemption results from litignatiobrought under
constitutional free exercise grounds or statutory grounds (such asRFRA). In
both instances, the decision-maker must determine whether a constitutional
right would be infringed by a particular application of an otherwise valid
law. And in both instances, the court will order a remedy that protects the
exercise of the constitutional right, but otherwise leaves the law in place to
apply to other circumstances that may arise. AfteSrmith, it is muchmore
difficult to obtain successful religious exemptions as a constitutional matter
in many contexts. Viewed in this light, RFRA is essentially restoring a
standard that again allows for as-applied challenges to otherwise valid laws.
For example, the Supreme Court’s decision in Hobby Lobby held that
the Department of Health and Human Service“sHH( S”) contraception
mandate unjustifiably burdened a family-held business’s religious exercise
because the government had many other alternatives to accomplish its
interest of making contraception more accessible to women8.9 The Court
therefore held that RFRA required “an exemption from the rule.”90 But the Court
did not strike down the HHS mandate wholesale. Thus, this law continues
to apply to all other covered employers, but with surgical exemptions for a
limited group of religious objectors.
Courts have engaged in this same type of adjudication in other
successful challenges under RFRA and the Religious Land Use and
Institutionalized Persons Act (“RLUIPA”). For example, i2n006, in Gonzales v. O
Centro Espirita Beneficente Uniao do Vegetal, the Court held that the Con
trolled Substances Act, although generally constitutional, could not be
paplied to prohibit the sacramental use of hoasca tea for a religious grou91p.
And in 2015, in Holt v. Hobbs, the Court concluded that the prison’s ban on
beards, although generally valid, could not be applied to prohibit certain
religiously-motivated beards.92 In each case, the Court required an
exception to an otherwise valid law to protect a religious exercise right.
88 See, e.g., Fallon, supra note 76, at 956 (explaining that a-sapplied challenges involve the
surgical severing of problematic aspects of a statute from acceptable applications).
89 134 S. Ct. 2751, 2760, 2780–82 (
90 Id. at 2761.
91 546 U.S. 418, 438–39 (2006).
92 135 S. Ct. 853, 859 (2015).
B. Survey of First Amendment As-Applied Challenges
Of course, the fact that religious exemptions are functionally
applied challenges does not answer whether those sorts of a-aspplied
challenges still result in unfair preferential treatment for religious liberty claims
if they are offered to facially valid and generally applicable law- s. As
applied challenges in other First Amendment contexts provide a particularly
relevant comparator to assess that question.
The Supreme Court has recognized that “[t]he Free Exercise Clause
embraces a freedom of conscience and worship that hacslose parallels in
the speech provisions of the First Amendment . . . .”93 Free exercise
protections and free speech protections theoretically serve many similar roles in
our constitutional democracy: they both operate as important safeguards
against government overreach, implicate matters of personal choice and
identity, allow for robust pluralism in our diverse society, help curb
dissnesion and social conflict, and protect minority rights that will not necessarily
be addressed through the political process9.4 As one notable academic has
observed regarding free exercise compared to other constitutional rights, “it
seems intuitively correct that similar rights should be enforced to a similar
extent with similar doctrine.”95
Prior to Smith, in a number of constitutional cases the Supreme Court
did enforce the First Amendment rights of speech and religious exercise
quite similarly. In Murdock v. Pennsylvania, for example, a city had a
generally applicable ordinance that required “all persons canvassing for or
soliciting . . . goods, paintings, pictures, wares, or merchandise of any kind” to
pay a fee to the city to obtain a license to solici9t6. Plaintiffs who had been
arrested under this ordinance were Jehovah’s Witnesses going door to door
distributing religious literature and soliciting donations without ever
obtaining such a license.97 The Plaintiffs argued this government action “deprived
them of the freedom of speech, press, and religion guaranteed by the First
In addressing these claims, the Supreme Court first made clear that the
ordinance at issue in this case was facially vali9d9.Further, the Court
observed that the regulation did not discriminate.100 Thus, the limited question
before the Court was simply whether the ordinance “acsonstrued and
applied require[d] religious colporteurs to pay a license tax as a condition to
the pursuit of their activities.”101
The Court rejected the ogvernment’s argument that providing an
xeemption would put Jehovah’s Witnesses “above the law.1”02 Instead, it
explained, “[a] license tax certainly does not acquire constitutional validity
because it classifies the privileges protected by the First Amendmetnalong
with the wares and merchandise of hucksters and peddlers and treats them
all alike. Such equality in treatment does not save the ordinanc1e03.”Just
because a law is facially valid does not mean it can be validly enforced
when it butts up against fundamental constitutional rights. At that point, an
ordinance “is not directed to the problems with which the police power of
the state is free to deal.”104 This is because, as the Court noted, “[f]reedom
of press, freedom of speech,[and] freedom of religion are in a preferred
position.”105 The Murdock Court thus held that the ordinance’s application
here both curtailed the free press and impinged freedom of religion, and that
these rights “stand or fall together.”106
As-applied challenges brought under other spcehe-based claims still
receive fairly similar treatment to that provided underMurdock.107 But after
Smith, and under the scheme advocated by critics of religious exemptions,
there are at least two contexts where as-applied speech challenges receive
significantly different treatment than similar religious challengesm:-co
pelled action and discretionary enforcement or application of a law.
99 Id. at 110 (holding that there was no “question as to the validity of a registration system for
colporteurs and other solicitors”).
100 Id. at 115.
101 Id. at 110.
102 Id. at 116 (internal quotation marks omitted).
103 Id. at 115.
104 Id. at 116.
105 Id. at 115.
106 Id. at 117 (emphasis added). The Court arrived at a similar conclusion in several other
cases dealing with religious solicitation. See, e.g., Follett v. McCormick, 321 U.S. 573, 577 (1944)
(holding that the application of a flat license tax was a violation of the free exercise clause);
Jamison v. Texas, 318 U.S. 413, 414, 417 (1943) (holding that the law could not prohibit
dissemination of religious handbills in the street); Cantwell v. Connecticut, 310 U.S. 296, 304–07 (1940)
(concluding that the application of a regulation that required a certificate in order to solicit support
for a religion was a violation of the Constitution).
107 Murdock v. Pennsylvania, 319 U.S. 105, 117 (1943).
endless chain of exemption demands,” is a distinct threat to the rule of law
because it will resultin a tidal wave of religious claimants striking down
government action at every turn.242
To examine this claim, we conducted a modest empirical analysis to
) the likelihood that religious exemptions result in government
cation being struck down under RFRA, and how that rate has changed since
Hobby Lobby; and (
) the volume of religious objection cases being brought
compared to speec-hbased cases, and whether the volume of religious
claims seem to have increased dramatically postHobby Lobby. Notably, we
provide the first nation-wide RFRA survey of its kind since Hobby Lobby.
Our findings contradict the notion that religious objections are much
more likely to prompt a court to strike down government action under
RFRA after Hobby Lobby.243 Compared to previous scholarship assessing
government win rates in this areaH,obby Lobby does not appear to have
significantly changed the government’s win rate in the last three years. Our
findings also indicate that cases dealing with religious objections to laws
are less pervasive than cases dealing with other expressive First Amd-en
ment claims.244 These findings apply to all federal cases, as well as the
cases specifically at the Supreme Court level. The data also does not indicate a
trend of dramatic growth in the voulme of religious cases post Hobby
Lobby. More time and data will be necessary to confirm these
results.Additionally, no statistical regression analysis has been performed to isolate the
effect of variables, and the findings in this Article are thus at best suggestive.
We utilized three different methods to analyze religious objection
claims compared to other speech and association claims, both in terms of
volume and likelihood of striking down government action. Each of these
methods is discussed in turn below.
We also compared our findings to the helpful empirical research of
Professor Adam Winkler, who reviewed all federal cases from 1990 to 2003
dealing with strict scrutiny, including cases dealing with suspect
classifications, speech, religious liberty, fundamental rights, and freedom of
assocaition.245 Though somewhat dated (and inapplicable to questions about how
Hobby Lobby changed the pervasiveness of religious objections), Professor
Winkler’s research serves as a useful benchmark.
1. New RFRA Survey Methodology
To assess how Hobby Lobby has impacted the government’s win rate
in RFRA cases, we surveyed all federal RFRA court cases available in
Westlaw that have been brought in the thre-eyear period since Hobby
Lobby.246 We analyzed the court rulings to determine how frequently the court
ruled for the plaintiff on a RFRA claim versus how frequently the court
ruled for the government. The universe of RFRA cases was created using
two types of searches. The first was a general Westlaw searfcohr RFRA
terms in the federal cases database:
For our second method, we used the Westlaw citing references feature
to identify any cases citing to the five statutory sections of RFRA for the
same three-year time period. We then cros-sreferenced both search results,
removing any duplicate decisions. Only two new cases resulted from this
second citing reference search method, bringing the total universe of cases
to 480 cases.247
es, 48 SETON HALL L. REV. 353 (2018) (surveying Tenth Circuit religious liberty cases from
246 The relevant time period was June 30, 201–4July 17, 2017. Some applications of RFRA
necessarily evaded the data set if they unreported cases, or not part of Westlaw’s database. We
assume that such cases are more likely to be government wins, and that a court decision
ovetrurning government action would result in a reported case. This may mean that the government win
rate is in actuality even higher. We excluded the Religious Land Use and Institutionalized Persons
Act (“RLUIPA”) cases from the dataset because the standard for RLUIPA was not clearly
established as being as protective as RFRA until the Supreme Court’s decision a year afterHobby
Lobby in Holt v. Hobbs. To avoid potential data skewing that would result partially through the
threeyear period based on that evolving standard, and where the purpose of this analysis is to assess the
impact of Hobby Lobby, those RLUIPA cases were excluded.
247 The search terms and results for this method included the following:
• Search: Federal cases under citing references for 42 U.S.C. § 2000bb from July 1, 2014
– June 30, 2017.
o Results: 190 cases, but only 1 additional new case.
• Search: Federal cases under citing references for 42 U.S.C. § 2000b1b-from July 1,
2014 – June 30, 2017.
o Results: 180 cases, but only 1 additional new case.
• Search: Federal cases under citing references for 42 U.S.C. § 2000b2b-from July 1,
2014 – June 30, 2017.
We analyzed a judicial decision as a single application of RFRAe,- r
gardless of the number of judges on the panel. Preliminary rulings and
decisions that were subsequently reversed or affirmed on appeal were collected
but, unless otherwise specified, were excluded from the reported results to
avoid double counting.248 Additionally, only decisions in the relevant
threeyear period were counted, and it is possible that some of these decisions
could be subsequently reversed. With the help of excellent research asss-i
tance, we manually reviewed all 480 cases to ensure that any decisions not
addressing RFRA on the merits were also exclud2e4d9.This brought the
universe of unique cases addressing RFRA on the merits to a total of 101
cases. Many of theother cases were disposed of on procedural grounds or
other legal claims.
Unsurprisingly, there were a number of cases (thirty-one to be precise)
dealing with essentially the same challenge to the “contraception mandate”
of the Affordable Care Act. These cases arose in different jurisdictions but
addressed the same issue that was ultimately addressed by the Supreme
Court either in Hobby Lobby (for the for-profit organizations) or in the
consolidated appeal in Zubik v. Burwell (for the nonprofit organizations).250 In
light of the government’s announcement to resolve these casesthrough
regulatory action after the Supreme Court’s decision in Zubik, we decided that
the most conservative course was to treat these cases as separate gonv-er
ment losses.251 We made this decision given that many religious exemption
o Results: 30 cases, but 0 additional new cases.
• Search: Federal cases under citing references for 42 U.S.C. § 2000b3b-from July 1,
2014 – June 30, 2017.
o Results: 16 cases, but 0 additional new cases.
• Search: Federal cases under citing references for 42 U.S.C. § 2000b4b-from July 1,
2014 – June 30, 2017.
o Results: 30 cases, but 0 additional new cases.
248 The empirical analysis by Adam Winkler on strict scrutiny applications similarly analyzed
court rulings, but excluded cases that were preliminary or that were addressed by another court on
appeal. Winkler, supra note 245, at 844–45.
249 A decision was considered “on the merits” if the court made a determination regarding
whether there was a sincere religious belief, a substantial burden on religious exercise, the
government had a compelling interest, or the regulation was the least restrictive means. This included
when the court discussed whether plaintiff alleged enough facts to support his RFRA claim (i.e. at
the motion to dismiss stage). Not considered “on the merits” included cases where the RFRA
claim was dismissed because the defendant was a state or private actor, or the plaintiff did not
exhaust administrative remedies.
250 See HHS Mandate Information Centr,alBECKET,
http://www.becketlaw.org/researchcentral/hhs-info-central/ [https://perma.cc/D3A9-HZJ5] (compiling a database of cases
challenging the HHS mandate);see, e.g., Zubik v. Burwell, 136 S. Ct. 1557, 1560–61 (
Third, Fifth, Tenth, and D.C. Circuit Courts decisions that held HHS constitutional, and
remanding to address the issue of employers providing notice of their religious objections).
251 In May 2016, the Supreme Court unanimously overturned lower court rulings against
religious objectors, ordered the government not to fine the objectors, and said the lower courts
objectors are likely particularly concerned about these casAersg.uably,
these cases could be treated as one case, or excluded altogether. Thus, we
presented alternative findings treating these cases as two government losses
(one loss for the government against fo-rprofit challenges in Hobby Lobby,
and one loss for the government against nonprofits inZubik), or excluding
these cases completely.
A final point of clarification is in order. The purpose of this
methodology is to examine how courts have applied strict scrutiny under RFRA since
Hobby Lobby; it is not to determine how strict scrutiny might affect
litigants, legislators, government officials, or others. Our goal was not toe-d
termine how strict scrutiny impacts the decision of lawmakers to adopt
laws, government officials to enforce laws,or litigants to bring, settle,or
appeal lawsuits. That is not something that could be measured by thisa-d
ta.252 Thus, this case survey does not capture other possible
significantmipacts Hobby Lobby may have had on the willingness of government oi-ff
cials to enforce certain laws or the willingness of litigants to bring lawsuits.
2. Targeted Comparative Searches Methodology
For our second method, we wanted to find a way to compare the
volume of speech cases to religious exercise cases. Religious exercise cases
include constitutional cases and thus present a broader universe of claims
than just RFRA cases. That is why an additional search method wea-s r
quired to assess case volume beyond our new RFRA survey2.53 Initially, we
assessed a snapshot of all cases that have been brought since Hthoebby
Lobby decision in June 2014. To do this we ran three targeted searches in
Westlaw to identify the approximate number of religious cases compared to
other expressive First Amendment cases. We first ran searches using
Westlaw’s key number system to assess how many cases were assigned
Westlaw’s “speech and expression” key number versus its “religious
exershould provide the government with an opportunity “to arrive at an approach going forward that
accommodates the petitioners’ religious beliefs.”Zubik, 136 S. Ct. at 1560. In May 2017, the
President issued an executive order directing the Department Hoefalth and Human Services
(“HHS”) and other agencies to protect religious ministries from the HHS mandate. Exec. Order
No. 13,798, 82 Fed. Reg. 21,675 (May 4, 2017). Further government action to finalize policy
changes to the contraception mandate is anticipated.
252 The data’s failure to account for primary behavior is an admitted weakness, but the nature
of the relevant question makes it almost impossible to collect the type of data that would be
necessary to analyze the effects of Hobby Lobby on primary behavior.
253 If we had limited our dataset to just RFRA cases, this would have skewed the religious
claims and indicated they were even smaller compared to other First Amendment expressive cases
than they already are.
cise” key number.254 Second, we ran searches with relevant speech or rei-l
gious-exercise search terms appearing at least four times in the body of a
case.255 Third, we ran searches with the same speech or religio-uesxercise
search terms appearing in the Westlaw summary of the case2.56 Each search
was limited to the time period of June 30, 2014t,o June 30, 2017.257 Each
of these searches has its own shortcomings in perfectly capturing the iu-n
verse of speech or free exercise cases in the last three years, but together
these searches triangulate to provide laikely relevant data point for the
comparative volume of cases brought over the last three years.258
It is possible that this data, as a snapshot in time, would not reveal a
sharp upward trend in the growth of religious cases. Thus, to analyze trends
over time in volume, we used the targeted search language for the Westlaw
key number system and ran that search limited by year for each year dating
back to 1946.259 We chose this year because (as discussed below) the modern
Spaeth Database begins in 1946 as well2.60 We compared speech to religious
cases over time in absolute terms, and also as a percentage of all reported
cases by year. We used the number of all reported cases by year that was
provdied in the Lexis database. Westlaw does not report more thtean-thousand
cases in any search category, but Lexis does not cap the number of cases
porvided in search results.261 Thus, for the numerator we used the number of
erligious or speech Westlaw key cases in a given year, and for the denominator
we used the number of all reported cases in that same year. We then displayed
the findings from this methodology in two different graphs.
3. Spaeth Database Methodology
Finally, although the findings of the first two methods provide
information about volume and win rates forall federal cases, some may wonder
whether a comparatively larger proportion of religious cases percolate
through the court system to the Supreme Court, and what type of win rates
those religious cases enjoy. We thus compared religious exercise versus
other expressive Supreme Court cases using Harold J. Spaeth’s U.Su.- S
preme Court Database.262 This database codes all Supreme Court decisions
from 1946 to 2016 based on numerous factors, including the particular legal
issue, as well as the party that was successful. This coding “makes amassing
data on the Court’s First Amendment decisions a relatively straightforward
task.”263 We first looked at all of the speech and association Supreme Court
cases in which the government was clearly successful, either as the
petitioner or the respondent.264 We then did the same thing for cases coded as r ei-l
gious exercise cases.265 Notably, this Spaeth coding of a winning party does
not ensure that a religious or speech claim, respectively, is the winning issue
on the merits. Thus, the value of these win rates is limited.
In terms of providing a fully representative dataset, there are some
obvious limitations with the Spaeth database. Most notably, the database is
limited to Supreme Court decisions, which are not representative of other
federal cases.266 Still, this database does provide a useful data point for the
volume of meritorious cases that the Supreme Court is interested in
addressing for speech compared to religious exercise issues. The modern database
time-period ends in 2017 (the 2016 term).
Our findings do not indicate that government win rates have undergone
a dramatic change since Hobby Lobby. Though the data does not assess the
impact of Hobby Lobby on primary behavior, it does contradict the fear of
some critics that Hobby Lobby will “encourage many future lawsuits,”267
“throw open the courtroom door to corporation,s”268 and “invite flimsy
but readily sustainable RFRA claims . . with a high likelihood of
Additionally, our findings indicate that reported cases dealing with
speech claims are much more voluminous than reported cases dealing with
religious claims, both in absolute terms and as a percentage of all reported
cases. The trend over time indicates that religious claims are decreasing as a
percentage of all reported cases, not increasing as some critics fear.
1. New RFRA Survey Findings
In our new survey of all federal cases involving a RFRA claim in the
three years since Hobby Lobby, the government won in fifty out of 101
cas266 According to the Supreme Court’s own website, its current caseload is over 10,000 cases,
but plenary review is only granted “in about 100 cases per Term,” so less than 1% of all appeals.
The Justices’ Caseload,SUPREME COURT OF THEUNITED STATES, https://www.supremecourt.
gov/about/justicecaseload.pdf [https://perma.cc/DSD3-CQJE]; see also David R. Stras,The
Supreme Court’s Gatekeepers: The Role of Law Clerks in the Certiorari Proce,ss85 TEX. L. REV.
947, 967, 987 (2007) (book review) (reporting the decrease in cases that the Supreme Court
decides on the merits, and noting that the Court granted review over 3% of ttihmee in the early
1980s but has done so less than 1% of the time since the October 1999 Term, although the number
of certiorari petitions has steadily increased); The Supreme Court, 2009 Term—The Statistics, 124
HARV. L. REV. 411, 418 tbl.II(B) (
) (observing that the Court granted only 0.9% of 8,131
petitions for review in its 2009 Term).
267 Griffin, supra note 60, at 673.
268 Sepper, supra note 62, at 233.
269 Marshall, supra note 8, at 120. Our findings are consistent with analysis by journalists and
HHS that there have been very few corporations that have taken advantage of a contraception
mandate exemption. See, e.g., Moral Exemptions and Accommodations for Coverage of Certain
Preventive Services Under the Affordable Care Act, 82 Fed. Reg. 47,838–857 (interim final rules
with request for comment Oct. 13, 2017) (to be codified at 26 C.F.R. pt. 54; 29 C.F.R. pt. 2590; 45
C.F.R. pt. 147); Jennifer Haberkorn,Two Years Later, Few HobbyLobby Copycats Emerge,
POLITICO (Oct. 11, 2016, 5:19 PM),
http://www.politico.com/story/2016/10/obamacare-birth-controlmandate-employers-229627 [https://perma.cc/DT69-6SCH] (“Politico obtained the
accommodation notices filed by employers between [Januaryand March] 2014 .. . . Thirty of the entities are
nonprofits and 22 are for-profits.”).
es, which is a government win rate of fifty percent.270 If contraception
mandate cases are treated as just two consolidated government losses addressed
by the Supreme court in Hobby Lobby and Zubik (one for the for-profit
cases, and one for the nonprofit cases), then the government enjoys a higher
win rate of sixty-nine percent. If the thirty-one contraception mandate cases
are excluded altgoether, the government’s win rate siesventy-one
Our findings on win rates are similar to those from the Winklerr-su
vey.272 Specifically, of all of the rights Professor Winkler surveyed,
government action was most likely to succeed in the context of strict scrutiny
applied to religious claims, with afifty-nine percent success rate—“more
273 This number
inthan double the mean of the other doctrinal categories.”
creased even further to a seventy-four percent government win rate for
religious claims challenging a generally applicable law27.4 In Professor
Winkler’s findings, speech claims were the most likely to result in striking
government action, with the government action at issue surviving a speech
challenge only twenty-two percent of the time2.75 Notably, the government
action was even less likely to survive a speech challenge than an equal
New RFRA survey findings (counting all
contraception mandate cases as separate
New RFRA survey findings (counting all
contraception mandate cases as two
New RFRA survey findings (excluding all
contraception mandate cases)
Government Win Rate
270 The underlying data set for this Table is available at https://docs.google.com
(request permission to view from author).
271 See infra note 277 and accompanying text (Table 1).
272 Winkler, supra note 245, at 844–45.
273 Id. at 857–58.
274 Id. at 861. This finding by Winkler is not a perfect comparison, as it was not limited to
RFRA. Id. at 857–58. Additionally, “generally applicable law” is an undefined legal term subject
to debate, and thus difficult to classify.See Gedicks, supra note 95, at 113 (noting that
Employment Division v. Smith did not define “general applicability”).
275 Winkler, supra note 245, at 844.
277 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
Winkler win rate for all religious claims Winkler win rate for religious challenges to generally applicable laws (not targeting religion)
Winkler win rate for speech claims
Our findings do not demonstrate a dramatic drop in government win
rates post Hobby Lobby. The data does not address relevant primary
behaviors, such as the choice of government officials not to adopt or enforce
cretain laws.278 But our findings do shed light on how judges continue ton-e
force strict scrutiny under RFRA post Hobby Lobby. More time and data are
necessary to draw more concrete conclusions.
2. Targeted Comparative Searches Findings
In our targeted searches to assess volume of religious versus speech
cases, we began by looking at volume in the three years sincHeobby
Lobby.279 In our findings, thenumber of speech and expressive casesthat the
searches returned generally dwarfed the number of free exercise cases in
similar search results. Under each search result, speech cases outnumbered
religious claims at a ratio of anywhere from 3:1 to 6:1.
These findings are consistent with Professor Winkler’s findings, where
the volume of speech-based claims was notably greater than the volume of
religious claims. Specifically, Winkler found that speech claims constituted
by far the largest category of strict scrutiny ca—ses222 of 4592.80 There
were thirty-three additional association cases, totaling 255 expressive
cases.281 In contrast, religious claims in Professor Winkler’s database
accounted for merely seventy-three of the 459 claims.282
278 Notably, under the new Trump administration, the Attorney General recently issued
gudiance that is much more protective of religious exercise under RFRA, and will likely result in much
less agency action that is hostile towards religious exercise under this administration.See
generally Memorandum from Jeffrey Sessions, Attorney Gen., U.S. Dep’t of Justice, for All Executive
Departments and Agencies (Oct. 6, 2017). That sort of primary behavior is not measured by our
RFRA case survey.
279 The date range filter used for all of these searches was July 1, 2014 to June 30, 2017.
280 Winkler, supra note 245, at 844–45.
281 Id. at 815.
283 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
Search of rele
vant term at
least four times
summary of case
When analyzing thetrends of the Westlaw key number cases over
time, our findings are not consistent with the allegation that religious claims
are undergoing a dramatic expansion, particularly as compared to speech
cases. Table 3 illustrates both types of Westlaw key number cases tracked in
absolute terms over time.284
Table 4 illustrates both types of Westlaw key number cases tracked as
a percentage of all reported cases over time. The percentage of religious
cases appears to have stayed fairly constant over the years.
284 See infra note 285 and accompanying graphic.
285 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
The underlying data set is
LLOkiKGgpe_8/edit#gid=878012463 (request permission to view from author).
46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 06 09 12 15
19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 20 20
Perhaps most interesting is that if we hone in on the years immediately
preceding Hobby Lobby and then immediately following that court decision,
a fitted line graph in Table 5 illustrates that the slope of religion cases as a
percentage of the reported caseload appears to be slightly decreasing.
Table 4 286
Expression vs Religion as a Percentage of
All Reported Cases
3. Spaeth Database Findings
According to the Spaeth database from 1946 to 2016, there were a total
of 461 First Amendment cases, and of these 378 dealt with speech or
asso286 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
287 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
ciation issues. The government was coded as a party in 344 of these cases.
In contrast, during the same period of time therewere only thirty-two free
exercise cases t(hirty-five with threeRFRA cases included)2.88 The
government was coded as a party intwenty-nine of these cases2.89 Thus, the
speech-based claims outnumbered religious claims by a ratio of more than
10:1. Overall, across the last seven decades, the court ruled in favor of the
government in forty-one percent of the free speech and association cases in
which it was a party. In contrast, thecourt ruled in favor of the government
at the higher rate of forty-five percent of the twenty-nine free exercise cases
in which the government was a party. In other words, the government is less
likely to win in the context of speech and association cases than in religious
exercise cases. But this difference is not substantial, and likely not
statistically significant at the Supreme Court level. These findings are interesting
in that they indicate that a far greater number of spee-bchased cases than
religious cases are meritorious enough to percolate to the Supreme Court.
Supreme Court Cases
Number brought between 1946 and 2016
Government win rate
Speech and Association Cases
Religious Exercise Cases 344 41%
In sum, our findings are consistent with the conclusion that speech
cases are much more pervasive than religious cases.Additionally, our
findings are not consistent with the notion that religious objections
aredramatically increasing in volume, or are muchmore likely to prompt a court to
strike down government action under RFRA afterHobby Lobby. Compared
to the work by Professor WinklerH,obby Lobby does not appear to have
significantly changed the government’s win rate.
C. Jurisprudential Explanation for Empirical Findings
The legal constraints on religious exercise claims may help exnplai
why religious exemption requests are not as voluminous or as successful as
critics fear. Although some scholars critique religious claims because truth
288 Modern Database: 2017 Release 01, supra note 262.
289 Of these cases, twenty-six were constitutional free exercise cases and three were RFRA
290 This Table is permanently available at http://www.bc.edu/content/dam/bc1/schools/law/
claims about such beliefs are“insulated from ordinary standards of
evidence,”291 the same could be saidof truth claims implicated by other First
Amendment rights. What standard of evidence could be said to apply, for
instance, to the truth claims of pornography, nearly nude dancing, videos of
animals being crushed, flag burning, or swastikas? Yet all of those are
examples of speech objections that requirethe most heightened scrutiny our
constitutional law offers.292 We often protect speech based on a speaker’s
subjective belief that a law impacts their expression, even though most
others may not view the lawas touching on expression at all2.93 For example,
most drivers who had to attach a license plate with a state motto likely did
not feel like they were being compelled to “say” anything, but thaeppellees
in Wooley v. Maynard subjectively felt otherwise, which was why they could
raise a successful a-sapplied challenge.294 As the Supreme Court has noted
elsewhere, “[a] person gets from a symbol the meaning he puts into it, and
what is one man’s comfort and inspiration is another’s jest and scorn.”295
We don’t even require speech to be sincere. We would give heightened
protection to speech even if the speaker didn’t believe what he or she was
saying, but still wanted to say it. Indeed,in New York Times v. Sullivan the
Court protected speech that was not true.296 In contrast, a religious objector
must prove her beliefs are sincere to receive protection. And sincerity acts
as a significant gatekeeper to religious objectors receiving protection.
For example, in a criminal drug trafficking case,an Arizona couple
attempted to raise drug money through its operation called the “Church of
Cognizance,” founded on the teaching that marijuana is both a deity and
sacrament.297 After the Border Patrol busted their “backpack runners” from
Mexico, the couple argued that their drug-running was part of their church’s
religious activities and thus legally protected by RFRA. In an opinion wrt-i
ten by then-Judge Gorsuch, the Tenth Circuit held that the couple’s religious
beliefs were not sincere—a threshold determination in every religious
liberty case—and that the “church” was a mere front for a drug operation. The
court explained that religious liberty laws do not “offer refuge to canny
op291 LEITER, supra note 64, at 34; see also LUPU & TUTTLE, supra note 55, at 27 (arguing that
“[i]f the government cannot evaluate the significance of a particular religious practice within a
believer’s faith, it will effectively lack the ability to identify—in a principled waym—eritorious
claims for accommodation”).
292 See, e.g., Nat’l Socialist Party of Am. v. Vill. of Skokie, 432 U.S. 43, 43–44 (1977)
(holding that because there were First Amendment implications, they must “provide strict procedural
293 See, e.g., Jacobs, supra note 132, at 160 (critiquing the way speech doctrine depends on
the “subjective intensity of the speaker’s reaction” to the law’s requirement).
294 430 U.S. 705, 707–08, 717 (1972).
295 W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 632–33 (1943).
296 376 U.S. 254, 257–59, 264 (1964).
297 United States v. Quaintance, 608 F.3d 717, 718 (10th Cir. 2010).
erators who seek through subterfuge to avoid laws they’d prefer to ignore,”
such as “those who set pu ‘churches’ as cover for illegal drug distribution
Further, the belief at issue must be “genuinely ‘religious’” to receive
protection.299 “[P]hilosophical and personal rather than religious” beliefs
are not enough.300 In Cavanaugh v. Bartelt, for example, a district court
determined that the satirical adherence to the “doctrine of the Flying Spaghetti
Monster” did not constitute a religiousbelief qualifying for First
Amendment or RLUIPA protection3.01 Instead, the court explained that this creed
was “a parody, intended to advance an argument about science, the evuo-l
tion of life, and the place of religion in public educat3i0o2n.T”hus, the
plaintiff was unable to prove that his religious exercise was burdened.303
The very existence of these threshold requirements in the religious
exercise realm also surely shapes the type of cases that litigants are willing to
bring, as they assess whether their practice is really religious and based on a
belief that they can prove is both sincerely held and actually bduerned by
the government action. In this manner, religious exercise claims aren-co
strained by doctrine in ways that other constitutional claims are not. These
constraints provide at least one explanation for why the volume of religious
exercise cases is so much lower than speech and association cases.
IV. THE ANOMALOUS MAJORITARIAN JURISPRUDENCE OF SMITH
The case studies and empirical analysis above indicate that a-sapplied
challenges are praised elsewhere in constitutional jurisprudence
(particularly the speech context), but uniquely maligned in the free exercise context
without justification. What, then, accounts for this disparate treatment of
two very similar304 types of constitutional challenges? One clue may come
from analyzing the majoritarian jurisprduential foundation on
whichEmployment Division v. Smith relies, and that has been soundly rejected in the
298 Yellowbear v. Lampert, 741 F.3d 48, 54 (10th Cir. 2014) (citingQuaintance, 608 F.3d at
299 Cavanaugh v. Bartelt, 178 F. Supp. 3d 819, 829 (D. Neb. 2016),aff’d, No. 16-2105 (8th
Cir. Sept. 7, 2016).
300 Wisconsin v. Yoder, 406 U.S. 205, 216 (1972).
301 178 F. Supp. 3d at 825, 828–29.
302 Id. at 824.
303 Id. at 834.
304 The Supreme Court has observed that “[t]he Free Exercise Clause embraces a freedom of
conscience and worship that hasclose parallels in the speech provisions of the First Amendment
. . . .” Lee v. Weisman, 505 U.S. 577, 591 (1992) (emphasis added).
A. Majoritarian Reasoning of Gobitis Overruled by
Barnette in the Speech Context
In 1935, during an elementary school’s daily pledgeof allegiance
ceremony, a ten-year-old fifth grader named William Gobitis refused to salute
the flag. This daily patriotic ritual at the time involved a “sti-fafrm” salute
that some complained looked very similar to the one Hitler required in Nazi
Germany.305 Some religious leaders had even given speeches denouncing
participation in the “Heil Hitler” salute.306 When William declined to
participate, his teacher tried to force his arm up, but William held it in his pocket
and successfully resisted. The next dayW,illiam’s eleven-year-old sister,
Lillian, did the same thing. She told her teacher, “I can’t salute the flag
naymore. The Bible says at Exodus chapter 20 that we can’t have any other
gods before Jehovah God.” The teacher hugged Lillian and called her a
The classmates of William and Lillian Gobitis were first astonished
and then disgusted with what they viewed as an unforgivable lack
opfatriotism. They would chant “[h]ere comes Jehovah” at the children and
shower them with pebbles on their way to school every day. The Gobitis parents
supported the consciences of their children, and in fact this Jehovah’s Wt-i
ness family was part of a national religious movement objecting to the flag
salute. But they too were shunned by their community: theirafmily-owned
grocery store was threatened with a mob attack and they were subjected to a
boycott. After the children were expelled from school, the family turned to
This legal dispute led toMinersville School District v. Gobitis, one of
the primary cases on which theSmith decision relied.309 In this case, the
Gobitis family challenged the generally applicable pub-lsichool
requirement that students either perform a salute to the national flag as part of a
daily ceremony or face expulsion3.10 In this case, no one disputed the
nsicere religious objections the Gobitis children had to participating in this
There was no question in theGobitis Court’s eyes that the generally
applicable flag salute requirement was csotintutional. The Court stated:
“[t]hat the flag-salute is an allowable portion of a school program for those
who do not invoke conscientious scruples is surely not debatable31.”2
Rather, the issue was [“w]hen does the constitutional guarantee [of religious
liberty] compel exemption from doing what society thinks necessary for the
promotion of some great common end, or from a penalty for conduct which
appears dangerous to the general good?3”13 Another way to ask the same
question is whether an as-applied challenge to the generally applicable flag
salute requirement should be upheld for the Jehovah’s Witness children,
providing a “religious exemption from a law that bound everybody.”314
The Court rejected this as-applied challenge, arguing that “[t]he religious
liberty that the Constitution protects has never excluded legislation of general
scope not directed against doctrinal loyalties of particular sects.”315 The Court
also cited toReynolds v. United State,s observing that “[c]onscientious
scruples have not, in the course of the long struggle for religious toleration,
relieved the individual from obedience to a general law not aimed at the
promotion or restriction of religious beliefs.”316 Thus, the Court refused to
provide what it characterized as “exceptional immunity ... to dissidents,”
and the Jehovah’s Witness children were forced to either salute the flag or
be expelled from school3.17 The Court in Gobitis also made clear that its
reasoning was not limited to free exercise, but extended to a constitutional
challenge based on speech rights as well.318
The Court primarily relied on the theory of judicial restraint developed
by Justice Felix Frankfurter, arguing that courts must defer to the will of the
majority, and that it would be an “arbitrary” exercise of power undermining
the strength of the government to set such legislative determinations
aside.319 In the theory set forth by Frankfurter, “the judiciary was supposed
to defer to reasonable judgments made by legislators, not overturn them
because it disagreed with their substance.”320
311 Id. at 592–93.
312 Id. at 599.
313 Id. at 593.
314 FELDMAN, supra note 306, at 182.
315 Gobitis, 310 U.S. at 594.
316 Id. at 594–95.
317 Id. at 591, 599–600.
318 Id. at 595 (holding that “[n]or does the freedom of speech assured by Due Process move in
a more absolute circle of immunity than that enjoyed by religious freedom”).
319 Id. at 596; FELDMAN, supra note 306, at 181–82.
320 FELDMAN, supra note 306, at 182.
Justice Harlan Stone authored a vigorous dissent, arguing that simply
deferring to the general rules passed by a legislature amounted to “nloess
than the surrender of the constitutional protection of the liberty of small
minorities to the popular will.”321 Stone relied on his famous footnote four
analysis in the previous case of United States v. Carolene Products Co., and
“pointed to the importance of a searching judicial inquiry into the legai-sl
tive judgment in situations where prejudice against discrete andinsular
minorities may tend to curtail the operation of those political processes
ordinarily to be relied on to protect minorities.”322 Legislation that operated “to
repress the religious freedom of small minorities, . . m.ust at least be
subject to the same judicial scrutiny as legislation which we have recently held
to infringe the constitutional liberty of . . .racial minorities,” Stone r-a
gued.323 This was because, in his view, the “Constitution expresses more
than the conviction of the people that democtrica processes must be
perserved at all costs.”324
Just three years later, in thBearnette decision (announced on Flag
Day), Justice Stone’s reasoning won the day and the Court
overruledGobitis—though only speaking clearly in terms of First Amendment expressive
rights.325 Specifically, the Court upheld an injunction “restrain[ingn]- e
forcement as to the plaintiffs and those of that class” based on the “limiting
principles of the First Amendment.3”26 In some of the most famous lines
from First Amendment jurisprudence, the Supreme Court waxed eloquent
about the grave risk of “coerc[ing] uniformity” in support of majoritarian
sentiment.327 “Those who begin coercive elimination of dissent soon find
themselves exterminating dissenters. Compulsory unification of opinion
achieves only the unanimity of the graveyard.”328 The Court thus concluded
that “[i]f there is any fixed star in our constitutional constellation, it is that
no official, high or petty, can prescribe what shall be orthodox in politics,
nationalism, religion, or other matters of opinion or force citizens to confess
by word or act their faith therein.”329
321 Gobitis, 310 U.S. at 606 (Stone, J., dissenting).
322 Id. (citing United States v. Carolene Products Co., 304 U.S. 144, 152 n.4 (1938)).
323 Id. at 607.
324 Id. at 606.
325 W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 642 (1943).
326 Id. at 630, 639.
327 Id. at 640.
328 Id. at 641.
329 Id. at 642 (emphasis added). The foundational principle has spawned much of our free
speech jurisprudence. See United States v. Playboy Entm’t Grp., Inc., 529 U.S. 803, 826 (2000)
(holding that “[t]he history of the law of free expression is one of vindicatioinn cases involving
speech that many citizens may find shabby, offensive, or even ugly”); Texas v. Johnson, 491 U.S.
397, 414 (1989) (holding that “[i]f there is a bedrock principle underlying the First Amendment, it
is that the government may not prohibit the expression of an idea simply because society finds the
Some scholars have argued thatBarnette was not a “religious
exemption case,” and was instead a case about “bar[ring] enforcement of the
mandatory salute statute . . . .”330 To be sure, the Court inBarnette avoided the
use of typical religious exemption language. But this was not a case about
whether the pledge of allegiance was being stricken (or even partially
stricken) for facial invalidity, or whether schools were prohibited from
holding flag salute ceremonies (or aspects of the ceremonies) as a general
matter. In fact, those sorts of lawsuits would come, but much late3r3.1 Barnette
was a case in which the Court “restrained enforcement” of an otherwise
valid policy “as to the plaintiffs and those of that class33.”2 To qualify for
this exemption, Plaintiffs (and others in the future who were similarly
situated), were being protected based on their First Amendment objection to an
otherwise valid exercise of government authority. That targeted invalidation
of government action is similar to what we think of as an -aaspplied
challenge in other contexts.
Regardless of the nuances of the type of remedy offered in this case,
what is clear is that Barnette instituted a rule of law that was distinctly
protective of minority rightsagainst majoritarian rules, and whichremains
foundational law in speech jurisprudence.333
B. Gobitis Resurrected by Smith in the Free Exercise Context
Though Gobitis remains bad law in the realm of free speech law, it was
resurrected as one of the primary jurisprudential pillars of reasoning in
Smith. The Smith Court quoted the following passage from Gobitis:
[c]onscientious scruples have not, in the course of the long
struggle for religious toleration, relieved the individual from obedience
to a general law not aimed at the promotion or restriction of rei-l
gious beliefs. The mere possession of religious convictions which
idea itself offensive or disagreeable”); see also Stewart Jay, The Creation of the First Amendment
Right to Free Expression: From the Eighteenth Century to the MiTdw- entieth Century, 34 WM.
MITCHELL L. REV. 773, 774 (2008) (noting that “[r]epudiation of governmentallym-andated
orthodoxy and tolerance for unpopular speech are two sides of the guiding principle in modern free
330 LUPU & TUTTLE, supra note 55, at 187.
331 See, e.g., Elk Grove Unified Sch. Dist. v. Newdow, 542 U.S. 1, 8, 10 (2004).
332 319 U.S. at 630.
333 See, e.g., Heffernan v. Paterson, N.J., 136 S. Ct. 1412, 1417 (
) (citiBngarnette as
setting forth the First Amendment’s “basic constitutional requirement”); Obergefell v. Hodges,
135 S. Ct. 2584, 2605–06 (2015) (quoting Barnette for the principle that the purpose of the
constitution “was to withdraw certain subjects from the vicissitudes of political controversy, to place
them beyond the reach of majorities and officials and to establish them as legal principles to be
applied by the courts”) (internal quotation marks omitted).
contradict the relevant concerns of a political society does
notrelieve the citizen from the discharge of political responsibilities.334
The Court relied on this reasoning to conclude that “an individual’si-rel
gious beliefs [do not] excuse him from compliance with an otherwise valid
law prohibiting conduct that theState is free to regulate.”335 The
precedential value of Smith’s reliance on Gobitis is dubious. As Professor Michael
McConnell put it, “[r]elying on Gobitis without mentioning Barnette is like
relying on Plessy v. Ferguson without mentioning Brown v. Board of
Four justices recognized the problem with relying onGobitis. Justice
Sandra Day O’Connor, for example, relied in her concurrence instead on the
reasoning of Barnette, quoting the following passage:
[t]he very purpose of a Bill of Rights was to withdraw certain
subjects from the vicissitudes of political controversy, to place
them beyond the reach of majorities and officials and to establish
them as legal principles to be applied by the courts. One’s right to
life, liberty, and property, to free speech, a free press, freedom of
worship and assembly, and other fundamental rights may not be
submitted to vote; they depend on the outcome of no elections.337
334 Smith, 494 U.S. at 879 (quontig Gobitis, 310 U.S. at 594–95) (internal quotation marks
335 Id. at 878–79.
336 McConnell, supra note 38, at 1124. Employment Division v. Smith also relied on “a
Mormon polygamy case from 1879, [which] was decided on the theory that the Free Exercise Clause
protects only beliefs and not conduct—a premise that the Court repudiated in 1940.”Id. (footnote
omitted) (citing Cantwell v. Connectiuct, 310 U.S. 296, 303–04 (1940));see also Bruce
Ackerman, Levels of Generality in Constitutional Interpretation: Liberating Abstraction, 59U. CHI. L.
REV. 317, 326 (1992) (arguing that “Gobitisshould be read as an especially pure example of the
New Deal approach to the Bill of Rights”).
337 Smith, 494 U.S. at 902–03 (O’Connor, J., concurring) (quotingBarnette, 319 U.S. at 638)
(internal quotation marks omitted);see also GORDON S. WOOD, CREATION OF THEAMERICAN
REPUBLIC: 1776–1787, at 593–96 (1998) (noting that our Constitution did not create pure
democracy; it created many checks and balances on the excesses of majoritarian power); Jack L. Landau,
Some Thoughts About State Constitutional Interpretation, 115PENN ST. L. REV. 837, 849 (2011)
(“The Bill of Rights itself is essentially a series of limitations on the exercise of majoritarian
authority. What the framers of the federal [C]onstitution created was not a popular democracy, but a
republic of fairly elaborate checks and balances.”). See generally WILLI PAUL ADAMS, THE FIRST
AMERICAN CONSTITUTIONS: REPUBLICAN IDEOLOGY AND THE MAKING OF THE STATE
CONSTITUTIONS IN THE REVOLUTIONARY ERA 110–14 (Rita Kimber & Robert Kimber trans., 1980)
(detailing the evolution of how the words “democracy” and “republic” were used)A;LEXANDER M.
BICKEL, THE LEAST DANGEROUS BRANCH: THE SUPREME COURT AT THE BAR OF POLITICS 16–
23 (1986) (discussing the “count-emrajoritarian difficulty” of meaningful judicial review in a
democracy). See generally ROBERT A. DAHL, HOW DEMOCRATIC IS THE AMERICAN
CONSTITUTION? (2001) (arguing that the Constitution ultimately upholds antidemocratic ideals)S;ANFORD
LEVINSON, OUR UNDEMOCRATIC CONSTITUTION: WHERE THE CONSTITUTION GOES WRONG
In a nutshell, one of the principles upon whichSmith relies is that “the
true course of judicial duty” is to “keep their hands off” the majoritarian
decisions of “legislative power,” as an exercise ojfudicial self-restraint.338
In contrast, the jurisprudence Smith overruled stands for the proposition that
religious exemptions “protect the rights of those whose religious practices
are not shared by the majority and may be viewed with hostility” because
“[t]he history of . . free exercise doctrine amply demonstrates the harsh
impact majoritarian rule has had on unpopular or emerging religious
As the Supreme Court previously noted, the diversity of religious
beliefs in our nation should weigh in favor of us providing more protection—
In the realm of religious faith, and in that of political belief,
sharp differences arise. In bothfields the tenets of one man may
seem the rankest error to his neighbor .. . . [I]n spite of the
probability of excesses and abuses, these liberties are, in the long
view, essential to enlightened opinion and right conduct on the
part of the citizens of a democracy.
The essential characteristic of these liberties is, that under their
shield many types of life, character, opinion and belief can
develop unmolested and unobstructed.Nowhere is this shield more
necessary than in our own country for a people composed of
many races and of many creeds.340
As-applied challenges to generally applicable laws allow a society as
diverse as ours to accommodate the “sharp differences [that] arisei”n
beliefs or preferences while still allowing the democratically-enacted rules of
law to continue to operate for the public good. The other alternative is to
disregard the minority views that are not protected by our political process.
Regardless of one’s ultimate views on the merits of a
countermajoritarian approach, the divergent precedents ofGobitis and Barnette provide one
(AND HOW WE THE PEOPLE CAN CORRECT IT) (2006) (same); Barry Friedman, The Birth of an
Academic Obsession: The History of the Countermajoritarian Difficulty, Part Five, 112 YALE L.J.
153, 155–62 (2002) (analyzing the history of judicial review to move beyond the countermajio-r
tarian difficulty as a constitutional theory).
338 Barnette, 319 U.S. at 648, 670 (Frankfurter, J., dissenting).
339 Smith, 494 U.S. at 902 (O’Connor, J., concurring). James Madison also famously wrote to
Thomas Jefferson that that Bill of Rights was but “parchment barriers,” and that the “invasion of
private rights is chiefly to be apprehended, not from acts of Government contrary to the sense of
its constituents, but from acts in which the Government is the mere instrument of the major
number of the constituents.” JAMES MADISON, Letter from James Madison to Thomas Jefferson (Oct.
17, 1788), in 11 THE PAPERS OF JAMES MADISON, 7 MARCH 1788–1 MARCH 1789, at 295–300
(Robert A. Rutland et al. eds., 1977).
340 Cantwell, 310 U.S. at 310 (emphasis added).
explanation for the disfavored treatment of religious exemptions. In Gobitis,
the Court determined that judicial restraint required it to simply defer to the
will of the majority, which required school children to perform a st-iaffrm
flag salute against their conscience because the majority thought they
should. Just three years later, in theBarnette decision, the Court overruled
its Gobitis holding in the contextof expressive rights, explaining that the
“very purpose of a Bill of Rights was to withdraw certain subjects from the
vicissitudes of political controversy, to place them beyond the reach of
majorities.”341 Barnette thus instituted a rule of law that was distinctly
protcetive of minority rights, and that remains foundational jurisprudence in the
context of speech rights. In contrast, thisArticle demonstrates how modern
religious exercise law has (mistakenly) resurrected the deferential and
amjoritarian reasoning of Gobitis, while asserting that requests to treat
religious exercise like other First Amendment rights are somehow o-uotf-step
with our constitutional traditions.
For critics trying to evaluate which approach is better, the following
question is instructive: Do we want to live in a society where the govne-r
ment can force school childrento salute flags, simply because the majority
likes that idea at the time? If the answer is no, that worldview hearkens to
the counter-majoritarian reasoning ofBarnette. That same principlealso
underlies statutes like RFRA. To support instead a double standard that
treats religious exercise as less deserving than any other First Amendment
right, then, would be the true anomaly.
Viewing religious exemptions through the lens of as-applied
challenges makes clear that such exemptions are not anomalous at all. Whee-n r
quests for individual religious exemptions are compared to a-sapplied
challenges in other constitutional contexts, it turns out that providing religious
exemptions from otherwise valid laws is both the most modest andthe
preferred method of adjudicating conflicts between individual rights and laws
passed for the public good.
The First Amendment cases surveyed in thiAsrticle indicate that
religious as-applied challenges do not result in preferential treatment fori-rel
gious objectors as a constitutional matter.Our new RFRA survey contradicts
the notion that religious objections are much more likely to prompt a court to
strike down government action under RFRA post-Hobby Lobby. Our findings
also indicate that cases dealing with religious objections to laws are
lesvsoluminous than other cases dealing with other expressive First Amendment
claims, and the religious cases do not appear to be undergoinga trend of
dramatic growth. In fact, the trend appears to be a slight decrease in volume.
Thus, through allowing -aapsplied challenges, religious exemption
schemes like RFRA simply restore religious exercise rights to a similar level
of protection already ofered to other rights housed in the First Amendment
and necessary for the protection of minority views in our pluralistic society.
II. RELIGIOUS EXEMPTIONS UNDERSTOOD AS AS-APPLIED CHALLENGES ............................... 1608
Constitutional Adjudication.................................................................................................. 1609
B. Survey of First Amendment As-Applied Challenges ......................................................... 1612
1. Compelled Action........................................................................................................ 1614
2. Discretionary Enforcement or Application of a Law ................................................... 1618
3. What About Intermediate Scrutiny? ............................................................................ 1619
4. What About Antidiscrimination Laws and Dignitary Harms? ..................................... 1623
1. New RFRA Survey Methodology................................................................................ 1634
2. Targeted Comparative Searches Methodology ............................................................ 1636
3. Spaeth Database Methodology .................................................................................... 1638
1. New RFRA Survey Findings ....................................................................................... 1639
2. Targeted Comparative Searches Findings.................................................................... 1641
3. Spaeth Database Findings............................................................................................ 1643
C. Jurisprudential Explanation for Empirical Findings ....................................................... 1644
B. Gobitis Resurrected by Smith in the Free Exercise Context ............................................ 1650 50 Case, supra note 1 , at 469-70 ( noting that “the sorts of religious exemptions from generally
omitted)); Marshall, supra note 8 , at 74;see also Dan T. Coenen, Free Speech and Generally
Applicable Laws : A New Doctrinal Synthesis, 103 IOWA L. REV. 435 , 466 ( 2018 ) (“[S]ingling out
general and some religions over others.” (internal quotation marks omitted )). 51 CORVINO ET AL., supra note 2, at 22. 52 Id. at 22 , 31 . 53 Id. at 52 . 54 Frederick Mark Gedicks , Is Religion an Excuse for Breaking the Law? , NEWSWEEK (Mar . 12,
2016 , 10 :51 AM), http://www.newsweek. comr/ealrieg-ious-beliefs-excuse-breaking-law-435664
[https://perma.cc/R54Y-XEM4]. 55 IRA C. LUPU & ROBERT W. TUTTLE , SECULAR GOVERNMENT : RELIGIOUS PEOPLE 11-12
( 2014 ). 56 Marshall, supra note 8, at 74; see also Case, supra note 1 , at 486-87 (raising concerns
dations”) . 57 See generally MARCI A. HAMILTON, GOD VS . THE GAVEL: THE PERILS OF EXTREME RELI-
GIOUS LIBERTY (2d rev . ed. 2014 ) (discussing the problems inherent to “extreme” legal protec71 CORVINO ET AL., supra note 2, at 31. 72 Id. at 52 . 73 Id.; Marshall, supra note 8, at 74; Schwarzschild, supra note 44, at 199; see also infra notes
152- 178 and accompany text. 74 Ayotte v. Planned Parenthood of N. New Eng ., 546 U.S. 320 , 331 ( 2006 ). 75 Id. at 329 (holding that “the 'normal rule' is that 'partial, rather than facial, invalidation is
too far, but otherwise left intact'”) (quoting Brockett v . Spokane Arcades , Inc., 472 U.S. 491 , 504
( 1985 )) (emendations in original ). 76 See Richard H. Fallon , Jr., Fact and Fiction About Facial Challenge,s 99 CALIF. L. REV.
915 , 956 ( 2011 ) (observing that as-applied challenges involve surgical severing);see also Wash .
State Grange v . Wash. State Republican Party , 552 U.S. 442 , 451 ( 2008 ) (asserting that “facial
terations and quotation marks omitted)) . 77 Snyder v . Phelps , 562 U.S. 443 , 460 ( 2011 ). 78 Richard H. Fallon, Jr., As-Applied and Facial Challenges and ThirdP- arty Standing , 113
HARV. L. REV . 1321 , 1329 , 1368 ( 2000 ) (citing United States v . Raines , 362 U.S. 17 , 5 ) ( 1960 ));
see also Yazoo & Miss. Valley R.R . v. Jackson Vinegar Co. , 226 U.S. 217 , 219 - 20 ( 1912 ) (hodl-
facts before the court); Fallon, supra , at 1368 ( “[A]s-applied challenges reflect entrenched though
can ordinarily be separated from valid subrules.” (emphasis omitted)). 79 Gonzales v . Carhart , 550 U.S. 124 , 167 - 68 ( 2007 ) (quotation marks omitted ). 80 Ayotte , 546 U.S. at 329 ( internal quotation marks omitted) (emphasis added) . 81 Id. at 328-29 (citations and internal quotation marks omitted) . 82 Id. at 329 . 83 Sabri v. United States , 541 U.S. 600 , 608 - 09 ( 2004 ) (holding that “facial challenges are
terpretation of statutes' on the basis of factually barebones records”) (quotingRaines, 362 U .S. at
22) (emendations omitted); see also Wash . State Grange , 552 U.S. at 450-51 (holding that “facial
challenges threaten to short circuit the democratic process”). 84 Citizens United v . Fed. Election Comm'n , 558 U.S. 310 , 399 ( 2010 ) (Stevens , J., concur-
ring in part and dissenting in part) . 85 Luke Meier, Facial Challenges and Separation of Powesr , 85 IND. L.J. 1557 , 1557 n.3
( 2010 ) (citing David L. Franklin, Looking Through Both Ends of the Telescope: Facial Challenges
and the Roberts Cour ,t 36 HASTINGS CONST. L .Q. 689 , 697 ( 2009 ) (arguing that the Supreme
ger, Facial and As-Applied Challenges Under the Roberts Court, 36 FORDHAM URB . L.J. 773 , 773
( 2009 ) (arguing that “[o]ne recurring theme of the Roberts Court's jurisprudence to date is its
Quinn Rosenkranz , The Subjects of the Constitution, 62STAN . L. REV. 1209 , 1233 , 1239 ( 2010 )
form of constitutional challenge”) . 86 Metzger, supra note 85, at 776 . 87 Fallon, supra note 78, at 1330- 31 . 93 Lee v. Weisman , 505 U.S. 577 , 591 ( 1992 ) (emphasis added) . 94 See , e.g., RANDY E. BARNETT, OUR REPUBLICAN CONSTITUTION 167-68 ( 2016 ) (arguing
Smith, The Rise and Fall of Religious Freedom in Constitutional Discourse, 140U . PA. L. REV.
149 , 196 - 98 ( 1991 ) (discussing historic rationales for religious freedom ). 95 Frederick Mark Gedicks , The Normalized Free Exercise Clause: Three Abnormalitie,s 75
IND. L .J. 77 , 120 - 22 ( 2000 ). But see Coenen, supra note 50, at 467 (arguing that different treat-
cise rights “serve different purposes within our constitutional system” ). 96 319 U.S. 105 , 106 ( 1943 ). 97 Id. at 106-07. 98 Id. at 107 . 242 Ira C. Lupu, Where Rights Begin: The Problems of Burdens on the Free Exercise of Rei-l
gion , 102 HARV. L. REV. 933 , 947 ( 1989 ); see also supra notes 152-178 and accompanying text. 243 See infra notes 267-292 and accompanying text. 244 See infra notes 267-292 and accompanying text. 245 See generally Adam Winkler , Fatal in Theory and Strict in Fact: An Empirical Analysis of
Strict Scrutiny in the Federal Courts , 59 VAND. L. REV. 793 ( 2006 ). Other authors have also sur-
“42 U.S.C. s 2000bb”) & DA(aft 06-30-2014 & bef 07-01-
o Results: 478 cases. 254 For speech cases, the following search was used in all state and federal cases: “adv:
TO ( 92xviii ) & DA(aft 06-30-2014 & bef 07-01 - 2017 ). ” This search references the Westlaw key
number 92 , Section XVII -IFreedom of Speech, Expression, and Press, k1490k - 2309 . For reli-
gious exercise cases, the following search was used in all state and federal cases: “adv: TO(92xiii)
& DA(aft 06-30-2014 & bef 07-01 - 2017 ). ” This search references the Westlaw key number 92,
Section XIII -Freedom of Religion and Conscience , k1290 - k1429 . 255 For speech cases, the following search was used in all state or federal cases: “adv : AT-
LEAST4(“freedom #of speech”) ATLEAST4(“freedom #of press”) ATLEAST4(“freedom #of
association”) ATLEAST4(“freedom #to associate”) & DA(aft 0360-- 2014 & bef 07-01- 2017 ).”
LEAST4(“freedom #of relig!”) ATLEAST4(rfra) & DA(aft 06-30-2014 & bef 07-01 - 2017 ). ” 256 For speech cases, the following search was used in all state or federal cases: “adv:
SY((free! /5 speech! press associat!)) & DA(aft 06-3200-14 & bef 07-012-017) .” For religious
exercise cases, the following search was in all state or federal cases: “adv: SY((relig! /5 exerc!)
rfra (free! /5 relig!)) & DA(aft 06-30-2014 & bef 07-01-2017).” 257 All relevant searches were run between October 19 , 2017 , and October 24, 2017 . We un-
given that Westlaw is continually adding new cases to its library. 258 As a starting matter, Westlaw does not claim to include all unpublished cases in its search-
able databases. 259 For counting purposes, our year began on July 1, the day afterHobby Lobby was decided,
and it ended on June 30. So, a search for one year of Westlaw key number cases would look like
this: “adv: (TO(92xviii)) & DA(aft 06-30-1945 & bef 07-01 - 1946 ). ” 260 We used the Westlaw key number for our search over time because this category was
terms. 261 To obtain the number of all reported cases from Lexis, we used the term “cite(lexis)”
“cite(lexis)” search year by year, and used this number for the denominator. 262 Modern Database: 2017 Release 01,WASH . UNIV. LAW: THE SUPREME COURT DATABASE
(Aug. 14 , 2017 ), http://supremecourtdatabase.org/data.php [https://perma.cc/XVZ9-QW6Q]. 263 Id.; see, e.g., Lee Epstein & Jeffrey A. Segal,Trumping the First Amendment? , 21 WASH.
U. J.L. & POL 'Y 81 , 92 ( 2006 ) (discussing the use of the database to analyzeother aspects of First
Amendment cases) . 264 For this coding under the “Legal Provisions” section of the database, we selected the
based on when the government was the winning party. 265 For this coding under the “Legal Provisions” section of the database, we selected the
the winning party. We also added the three RFRA cases to this category . 305 Barnette , 319 U.S. at 627 . 306 NOAH FELDMAN, SCORPIONS: THE BATTLES AND TRIUMPHS OF FDR'S GREAT SUPREME
COURT JUSTICES 179 ( 2010 ). 307 Id. at 179-80; see also SHAWN FRANCIS PETERS , JUDGING JEHOVAH'S WITNESSES: RELI-
GIOUS PERSECUTION AND THEDAWN OF THERIGHTS REVOLUTION 33-40 , 53 - 54 ( 2000 ) (ex-
Logic and Psychologic in Constitutional Decisionmaking, 36 STAN. L. REV. 675 , 677 - 78 ( 1984 )
ical understanding); Lillian Gobitis, The Courage to Put God Firs ,tAWAKE!, July 22 , 1993 , at
12- 15 ( recalling what led up to her Supreme Court case ). 308 FELDMAN, supra note 306, at 179-80 . 309 Emp't Div., Dep't of Human Res. v. Smith , 494 U.S. 872 , 879 ( 1990 ) (quoting Minersville
Sch. Dist. v. Gobitis, 310 U.S. 586 , 594 - 95 ( 1940 ), overruled by Barnette , 319 U.S. at 624). 310 Gobitis, 310 U.S. at 591.