KEEPING OUR BALANCE:
W
HY THE FREE EXERCISE CLAUSE NEEDS TEXT,
HISTORY, AND TRADITION
WILLIAM J. HAUN
*
In Fiddler on the Roof, the main character—Tevye der Milkhiker—
begins the play with an ode to “Tradition.” The song recounts how
the duties of religion, family, and work ensure continuity amid
change. This enduring stability is tradition’s virtue—or as Tevye
puts it, “how we keep our balance.” Without that balance, “our
lives would be as shaky as a fiddler on the roof.”
1
Fiddler’s under-
standing of tradition—a means to ensure continuity amid change—
would be a helpful corrective to current Free Exercise doctrine.
During the past decade, Free Exercise doctrine has become some-
thing like a fiddler on a roof. More than before, religious liberty is
a prominent feature of the U.S. Supreme Court’s docket. These
cases raise many doctrinal questions: What is religious speech?
2
When and how is government “neutral” toward religion?
3
What
does it mean for religious groups to participate equally in public
*
Senior Counsel, the Becket Fund for Religious Liberty; Nonresident Fellow, Amer-
ican Enterprise Institute. The views in this article are my own, and do not necessarily
reflect the views of the Becket Fund or its clients. I am grateful for the helpful insights
I received from Judge Janice Rogers Brown, Joel Alicea, Stephanie Barclay, Marc
DeGirolami, Luke Goodrich, Yuval Levin, Christopher Palko, Eric Rassbach, Laura
Slavis, Colten Stanberry, Mark Rienzi, Sophie Rizzieri, Adam White, Lori Windham,
and book club members at the National Shrine of St. Alphonsus Liguori. Ut in omnibus
glorificetur Deus.
1. JERRY BOCK ET AL., Tradition, in FIDDLER ON THE ROOF (1964).
2. Masterpiece Cakeshop, Ltd. v. Colo. C.R. Comm’n, 138 S. Ct. 1719, 1721 (2018).
3. Id. at 1723–24; Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63, 66 (2020).
420 Harvard Journal of Law & Public Policy Vol. 46
programs?
4
What protections do the religious have against govern-
ment discretion?
5
Do those protections change based on corporate
status?
6
What if the government admits it could accommodate a re-
ligious organization, but refuses to do so?
7
Can a government re-
fuse religious accommodations based on comparisons to secular ac-
commodations, and if so, what is properly comparable?
8
What
makes a church employee a “minister”?
9
And to what extent can
civil courts intrude into a religious organization’s internal deci-
sions?
10
Although these myriad contexts call the Free Exercise “fid-
dler” to dance to many tunes, one thing is clear: the fiddler is danc-
ing on unstable doctrine.
That is because current doctrine often rests on Employment Divi-
sion v. Smith.
11
Smith refused to authorize a religious exemption
from an “across-the-board-criminal prohibition on a particular
form of conduct.”
12
The folk understanding of Smith is that the gov-
ernment never has to accommodate religious believers burdened
by “neutral” and “generally applicable” laws. This baseline treat-
ment continues even as five sitting Supreme Court justices
acknowledge “compelling” reasons to overrule Smith.
13
And, as
will be discussed, Smith’s premises are disintegrating. In short, the
Free Exercise Clause needs surer footing than Smith.
4. Carson v. Makin, 142 S. Ct. 1987, 1997 (2022) (discussing Espinoza v. Mont. Dep’t
of Revenue, 140 S. Ct. 2246, 2255, 2261 (2020) and Trinity Lutheran Church of Columbia,
Inc. v. Comer, 137 S. Ct. 2012, 2021 (2017)).
5. Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1878 (2021).
6. Burwell v. Hobby Lobby Stores, Inc., 573 U.S. 682, 713–717 (2014).
7. Zubik v. Burwell, 578 U.S. 403, 407–08 (2016).
8. Tandon v. Newsom, 141 S. Ct. 1294, 1296 (2021).
9. Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 188–
92 (2012).
10. Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2055 (2020).
11. 494 U.S. 872 (1990).
12. Id. at 872–74, 884.
13. See, e.g., Fulton, 141 S. Ct. at 1931 (Gorsuch, J., concurring) (“[N]ot a single Justice
has lifted a pen to defend” Smith.).
2023 Keeping Our Balance 421
Smith should be abandoned and “text, history, and tradition”
should be adopted in its place. This latter approach is taken by
standard originalism,
14
fully expressed in the Second Amendment
context, and—notably for the Free Exercise Clause—already ap-
plies to other Religion Clause doctrines.
15
On this approach, the
Free Exercise Clause would presumptively protect a given religious
exercise unless the opposing party can show a long, unbroken tra-
dition of restriction that is analogous to the burden at issue. Text
and history are already well-established interpretive commit-
ments.
16
But tradition’s contribution is less clear. This article ex-
plains the role tradition should play in Free Exercise doctrine.
The Free Exercise Clause “has infrequently been interpreted tra-
ditionally.”
17
The complicating factor is Justice Scalia’s opinion for
the Court in Smith.
18
There, Smith responded to the textual ambi-
guity of the Free Exercise Clause toward religious accommodations
14. See, e.g., Michael W. McConnell, Textualism and the Dead Hand of the Past, 60 GEO.
WASH. L. REV. 1127, 1136 (1998). Originalism and its statutory cousin, textualism, are
“capacious term[s] for a variety of theories that are very different in their specifics.”
Marc O. DeGirolami, Traditionalism Rising, J.
CONTEMP. LEGAL ISSUES (forthcoming
2023) (manuscript at 18), https://ssrn.com/abstract=4205351 [https://perma.cc/EK4S-
PK73]. But the “Standard Approach” to defining those theories is to contrast them with
theories that interpret a legal text using something other than the text’s original public
meaning. J. Joel Alicea, Liberalism and Disagreement in American Constitutional Theory,
107 V
A. L. REV. 1711, 1714 (2022). This article will limit its discussion of originalism to
the standard approach, exemplified by Justice Scalia. See A
NTONIN SCALIA, SCALIA
SPEAKS 184 (Christopher J. Scalia & Edward Whelan eds., 2017) (describing Scalia’s
originalism); see also infra Part I.B.
15. Infra Part II. The mantra of “text, history, and tradition” seems to have first gained
interpretive force in the Second Amendment context (though there were earlier passing
usages). See Dru Stevenson, “Text, History, and Tradition” as a Three-Part Test, D
UKE CTR.
FOR
FIREARMS L.: SECOND THOUGHTS (Mar. 11, 2020),
https://sites.law.duke.edu/secondthoughts/2020/03/11/text-history-and-tradition-as-a-
three-part-test/ [https://perma.cc/E3GE-2URE].
16. See Michael W. McConnell, Lecture, Time, Institutions, and Interpretation, 95 B.U.
L. REV. 1745, 1747–51 (2015).
17. Marc O. DeGirolami, The Traditions of American Constitutional Law, 95 N
OTRE
DAME L. REV. 1123, 1148 (2020).
18. Infra Part II.D.
422 Harvard Journal of Law & Public Policy Vol. 46
with Justice Scalia’s preference: judicial restraint.
19
Smith admits
this was a “prefer[ence],” not a constitutional mandate.
20
And this
preference overrode any regard for longstanding practices of reli-
gious accommodation—evidence that Smith (and Justice Scalia
again in City of Boerne v. Flores
21
) deemed inappropriate for courts
to consider.
22
These choices make Free Exercise jurisprudence a
doctrinal outlier.
23
Moreover, by jumping straight from the Free Ex-
ercise Clause’s textual ambiguity on accommodation to Smith’s re-
straint preference, “restraint” is enforced by two abstract standards
(“neutrality” and “general applicability”) that have no necessary
connection to the Clause’s semantic or historical meaning—to say
nothing of longstanding practices toward religious accommoda-
tion. Unsurprisingly, the result of these abstract standards is not
restraint, but the interpretive tools that Justice Scalia considered un-
restrained: legislative history, decisionmaker motive, and analysis
of a law’s disparate impacts. These tools not only license judicial
manipulation to uphold government burdens on religion,
24
they re-
move the Free Exercise Clause from its ordinary understanding as
a guarantee of religious liberty.
25
Here, because there are open methodological points related to
tradition,
26
it is important to clarify what I mean when I refer to
“text, history, and tradition.” This article advocates for the use of
19. Id. Here, I am using “judicial restraint” as Smith did: ambiguity in constitutional
text means “judges should defer to the decisions of present-day representative institu-
tions.” McConnell, supra note 14, at 1136.
20. Emp. Div. v. Smith, 494 U.S. 872, 890 (1990).
21. 521 U.S. 507 (1997).
22. Smith, 494 U.S. at 889–90; City of Boerne, 521 U.S. at 541–42 (Scalia, J., concurring).
23. Infra Part II.D.
24. See Masterpiece Cakeshop, Ltd. v. Colo. C.R. Comm’n, 138 S. Ct. 1719, 1737 (2018)
(Gorsuch, J., concurring) (“Nor can any amount of after-the-fact maneuvering by our
colleagues save the Commission.”).
25. See Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1904 (2021) (Alito, J., concur-
ring) (evaluating the text of the Free Exercise Clause in “1791 (and today)”); see also id.
at 1896 (“These words had essentially the same meaning in 1791 as they do today.”).
26. See N.Y. State Rifle & Pistol Ass’n v. Bruen, 142 S. Ct. 2111, 2162 (2022) (Barrett,
J., concurring) (highlighting open questions around “the manner and circumstances in
which postratification practice may bear on the original meaning of the Constitution”).
2023 Keeping Our Balance 423
“text, history, and tradition”—in that order—when interpreting the
Free Exercise Clause. Some have argued for a form of “tradition”
that disregards text and original public meaning.
27
Others have ar-
gued for the role of “liquidation,” whereby an ambiguity in the
Constitution’s original meaning is “settled” by a post-ratification
practice or practices, regardless of their temporal endurance before
and after ratification.
28
Neither tradition alone nor liquidation is
my claim. Rather, a political community’s longstanding practices
toward particular religious accommodations—practices that can
come both before and after the Constitution’s ratification—should
illuminate what text and history do not definitively resolve about
the Free Exercise Clause’s original meaning. Illumination would re-
sult by the judiciary answering “historical, analogical questions,”
akin to the Court’s approach in the recent Second Amendment de-
cision, New York State Rifle & Pistol Association v. Bruen.
29
As Bruen
said, this approach was adopted from a “similar” one governing
Establishment Clause doctrine.
30
The church autonomy context re-
flects this approach too. All these contexts provide strong reasons
for extending “text, history, and tradition” to the Free Exercise
Clause.
In particular, this article makes three doctrinal suggestions: (1)
moving from a grand unified theory governing all Free Exercise
cases—as Smith sought—to context-specific rules rooted in
27. See Ernest Young, Rediscovering Conservatism: Burkean Political Theory and Consti-
tutional Interpretation, 72 N.C.
L. REV. 619, 674 (1994); David Strauss, Common Law Con-
stitutional Interpretation, 63 U.
CHI. L. REV. 877 (1996).
28. DeGirolami, Traditionalism Rising, supra note 14, at 20 (citing William Baude, Con-
stitutional Liquidation, 71 S
TAN. L. REV. 1 (2019)).
29. Bruen, 142 S. Ct. at 2134.
30. Id. at 2130; see also, e.g., Firewalker-Fields v. Lee, 58 F.4th 104, 122 n.7 (4th Cir.
2023) (“So, in Establishment Clause cases, the plaintiff has the burden of proving a set
of facts that would have historically been understood as an establishment of religion.
This requires proving both a set of facts, like in all litigation, and proving that those
facts align with a historically disfavored establishmentarian practice.” (citing Bruen, 142
S. Ct. at 2130 n.6)). As Bruen shows, this is a “legal inquiry” that can be decided at the
pleading stage. See Bruen, 142 S. Ct. at 2130 n.6.
424 Harvard Journal of Law & Public Policy Vol. 46
historical analogues;
31
(2) limiting any inquiry into “compelling”
interests to those that the opposing party shows, through
longstanding practice, are well-accepted reasons to burden the re-
ligious exercise at issue;
32
and (3) crafting distinct protections for
religious institutions.
33
These changes reflect tradition’s insight:
self-government requires enduring consent, and that consent is
demonstrated by the American people’s longstanding practices to-
ward their constitutional guarantees. Free Exercise doctrine, in both
its substance and its administrability, would benefit from this prac-
tical wisdom.
I. T
RADITION AS AN INTERPRETIVE AID TO TEXT AND HISTORY
Tradition’s distinct interpretive role is often “elided” when the
Supreme Court discusses text and history.
34
It is therefore im-
portant to understand what tradition itself brings to the interpre-
tive table.
That is this section’s subject.
There are many ways to distinguish tradition from text and his-
tory. One could explain why tradition is not as widely used.
35
One
could discuss tradition’s distinct justifications in morality and pol-
itics, contrast tradition with less standard forms of originalism, or
distinguish tradition from “liquidation.”
36
These distinctions have
been drawn well by others, especially Professor Marc DeGirolami.
37
Instead of retreading those grounds, this section will explain tradi-
tion’s distinct contribution to a jurisprudence of text, history, and
tradition. The first subsection will explain how tradition’s
31. Infra Part III.B; see also Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2087
(2019) (plurality opinion) (“While the Lemon Court ambitiously attempted to find a
grand unified theory of the Establishment Clause, in later cases, we have taken a more
modest approach that focuses on the particular issue at hand and looks to history for
guidance.”).
32. Infra Part III.C.
33. Infra Part III.D.
34. See DeGirolami, supra note 14, at 17.
35. See DeGirolami, supra note 17, at 1124.
36. See, e.g., DeGirolami, supra note 14, at 43.
37. See supra notes 14, 17.
2023 Keeping Our Balance 425
supplemental contribution to text and history compensates for text
and history’s potential for overtheorizing and unworkable rules.
The second subsection will explain how originalism’s standard ap-
proach—the approach of text, history, and tradition—provides ex-
amples of how to operationalize tradition’s supplementary role.
A. Tradition compensates for the shortcomings of text and history.
“Almost all interpreters, whatever their school of thought, agree
that the constitutional text (including inferences from structure) is
the place to begin, and that when the text is clear it is binding.”
38
A
commitment to the primacy of text is rooted in certain theories
about the binding nature of a written constitution.
39
A similar point
can made about the importance of history. Among all constitutional
interpreters, “[t]he importance of the temporal dimension is well
recognized.”
40
Where interpreters differ is not so much on the im-
portance of an historical “moment,” but what that historical moment
should be. For originalists, the history of “the moment at which the
Constitution was adopted” matters.
41
For living constitutionalists,
the present moment’s—purportedly—“better informed under-
standing”
42
is what matters. But no matter the preferred “moment,”
history-based jurisprudence is accepted, and text-based jurispru-
dence is too.
However, interpreting text and history can be very mechanical
and empirical.
43
That is not necessarily a problem. Technical tools
38. Michael W. McConnell, Time, Institutions and Interpretation, 95 B.U. L. REV. 1745,
1747 (2015).
39. See McConnell, supra note 14, at 1134–39.
40. McConnell, supra note 38, at 1751.
41. Id.
42. Obergefell v. Hodges, 576 U.S. 644, 671–72 (2015).
43. See N.Y. State Rifle & Pistol Ass’n v. Bruen, 142 S. Ct. 2111, 2130 n.6 (2022) (quot-
ing William Baude & Stephen Sachs, Originalism and the Law of the Past, 37 L.
& HIST.
REV. 809, 810–811 (2019)); see also Facebook, Inc. v. Duguid, 141 S. Ct. 1163, 1174–75
(2021) (Alito, J., concurring) (discussing the interplay of canons of construction and
“empirical” attempts to determine textual and historical evidence); Bostock v. Clayton
Cnty., 140 S. Ct. 1731, 1766 (2020) (Alito, J., dissenting) (arguing that “[d]ictionary
426 Harvard Journal of Law & Public Policy Vol. 46
and rules can be quite helpful. And for both text and history inter-
pretation—where the inquiries are either semantic or consider the
meaning of a word in an isolated “moment”—technical methods
can make sense.
44
But tradition-based evidence is different, because
tradition does not “view[] authoritative history as the snapshot of
a particular moment.”
45
Rather, by analyzing longstanding prac-
tice, tradition-based interpretation is analogical—finding meaning
when “multiple institutions independently reach[] the same con-
clusion” on a practice “over a long period of time.”
46
This analogical
inquiry, while necessarily comparative, “is not a mere likeness be-
tween diverse objects, but a proportion or relation of object to ob-
ject.”
47
The interpretive insight of tradition comes not from more
historically researched “facts,” but from immersing the interpreter
in social memory.
48
That is, the interpreter ascertains how Ameri-
can culture received its past, demonstrated by longstanding prac-
tice.
49
By identifying interpretive meaning in how generations
definitions are valuable . . . . But they are not the only source of relevant evidence” in
determining an “ordinary” meaning); see also John S. Ehrett, Against Corpus Linguistics,
108 G
EO. L.J. ONLINE 50, 72–73 (2019) (“Hidden beneath the fig leaf of ‘science’ are the
same value judgments that have long bedeviled all questions of textual interpretation—
only this time, those underlying value commitments are harder to immediately ascer-
tain.”).
44. Though technical approaches still have their limits and problems. See, e.g., Ehrett,
supra note 43, at 72–73.
45. Michael W. McConnell, Tradition and Constitutionalism Before the Constitution, 1998
U.
ILL. L. REV. 173, 174 (1998).
46. McConnell, supra note 38, at 1772.
47. Analogy in Metaphysics and Scholastic Philosophy, C
ATHOLIC ANSWERS,
https://www.catholic.com/encyclopedia/analogy [https://perma.cc/3X4Q-BF83].
48. See Judge Neomi Rao, The Province of Law, 46 H
ARV. J.L. & PUB. POL’Y 87, 99 (2023)
(“In analyzing the meaning of the Constitution and understanding its legal back-
ground, we must be mindful of the animating spirit and the institutional structure
of our law. We must draw on our distinctly Anglo-American legal reasons and princi-
ples.” (emphasis in original)).
49. See Josef Piper, Tradition Concept and Claim 16–22 (2010) (contrasting “historical
knowledge” and accepting a tradition).
2023 Keeping Our Balance 427
receive an understanding, the interpreter can retain “continuity
with the past” and harmony with the Constitution as a whole.
50
Tradition’s regard for continuity can be in tension with restraint.
Again, for purposes of this essay, I am discussing judicial restraint
as it is deployed in Smith—the idea that, when faced with constitu-
tional ambiguity, “judges should defer to the decisions of present-
day representative institutions.”
51
Restraint’s emphasis on pre-
sentism is in tension with tradition’s emphasis on endurance—that
is, how “the words of the Constitution . . . have been understood by
the people over the course of our constitutional history, from enact-
ment through the present.”
52
Some tradition advocates look at this
distinction and conclude that traditionalists should be “neither
committed to nor supportive of” standard originalism and judicial
restraint.
53
But text, history, and tradition is after something differ-
ent.
For text, history, and tradition, these tensions are good. Tradi-
tion’s practical focus helps ground text and history.
54
The authority
for text and history rests on the political theory that, in short, “[i]f
the Constitution is authoritative because the people of 1787 had an
original right to establish a government for themselves and their
posterity, the words they wrote should be interpreted—to the best
of our ability—as they meant them.”
55
This piece assumes that this
theory is correct.
56
But if it is correct, then the interpretive authority
for text and history can rest in abstraction and eschew knowledge
50. McConnell, supra note 14, at at 1136–37; see also Facebook, 141 S. Ct. at 1175 (Alito,
J., concurring) (“Empirical evidence might prove me wrong, but that’s not what mat-
ters.” What matters is whether such tools “accurately describe how the English lan-
guage is generally used.”).
51. McConnell, supra note 14, at 1136.
52. Id.
53. See Young, supra note 27; Strauss, supra note 27.
54. McConnell, supra note 14, at 1128.
55. Id. at 1132.
56. See id.
428 Harvard Journal of Law & Public Policy Vol. 46
from experience
57
—a form of knowledge that “comes to man in
many more forms than” syllogistic reasoning, empirical analysis, or
filtering history by theory.
58
The knowledge of experience is some-
times called “social knowledge,”
59
and it recognizes that certain
principles only receive full elucidation through application over
time.
60
When an interpreter acquires meaning from practice, he will per-
mit longstanding practices to distill the meaning suggested by the
technical analyses of text and history.
61
Such distillation does not,
57. For examples of privileging abstract conceptions of text and history, see, e.g., Bos-
tock v. Clayton Cnty., 140 S. Ct. 1731, 1750–51 (2020) (“One could easily contend that
legislators only intended expected applications or that a statute’s purpose is limited to
achieving applications foreseen at the time of enactment. However framed, the em-
ployer’s logic impermissibly seeks to displace the plain meaning of the law in favor of
something lying beyond it.”); Troxel v. Granville, 530 U.S. 57, 91–92 (2000) (Scalia, J.,
dissenting) (“[W]hile I would think it entirely compatible with the commitment to rep-
resentative democracy set forth in the founding documents to argue, in legislative
chambers or in electoral campaigns, that the State has no power to interfere with parents’
authority over the rearing of their children, I do not believe that the power which the
Constitution confers upon me as a judge entitles me to deny legal effect to laws that (in
my view) infringe upon what is (in my view) an unenumerated right.”).
58. F.A.
HAYEK, THE CONSTITUTION OF LIBERTY 125 (Ronald Hamowy ed., Routledge
2011) (1960).
59. See R
OGER SCRUTON, THE MEANING OF CONSERVATISM 31 (3d ed., Palgrave Mac-
millan 2001) (1980).
60. See J
OHN HENRY CARDINAL NEWMAN, AN ESSAY ON THE DEVELOPMENT OF CHRIS-
TIAN
DOCTRINE 185 (Univ. Notre Dame Press 1994) (1845) (distinguishing “a true de-
velopment and a corruption”); cf. P
OPE JOHN PAUL II, THE CATECHISM OF THE CATHOLIC
CHURCH 23 (2d ed. 2019) (1992) (“Yet even if Revelation is already complete, it has not
been made completely explicit; it remains for the Christian faith gradually to grasp its
full significance over the course of the centuries.”).
61. As shown by the cases discussed infra, the longstanding practices that tradition-
based interpretation considers include those both before and after the Constitution’s
ratification. If an interpreter only considers post-ratification evidence, he overlooks the
lessons taught by a practice’s roots and soil—that is, the how and why a practice became
longstanding. Moreover, a case that purported a basis in tradition, but only considered
recent practice, is not tradition-based. And the opinion would probably reveal it. Cf.
Lawrence v. Texas, 539 U.S. 558, 571–72 (2003) (“we think that our laws and traditions
in the past half century are of most relevance here,” and then claiming that history and
tradition are not “the ending point of the substantive due process inquiry”) (internal
quotation marks and citation omitted).
2023 Keeping Our Balance 429
by definition, authorize overriding what text and history defini-
tively show. Rather, longstanding practice brings to light meaning
left ambiguous by text and history. Proper tradition-based evi-
dence, then, “illustrates, not obscures; corroborates, not corrects the
body of thought from which it proceeds.”
62
On this view, the inter-
preter is not a “technician,” willing to invalidate longstanding prac-
tices because his “archaeological excavation” cannot explain them
on “rationalistic” grounds, or from a single moment’s “history.”
63
Nor is the interpreter an antiquarian, whose “wise and laudable”
interest in returning to original practices would “reduce everything
to antiquity by every possible device.”
64
An interpreter using tradi-
tion acknowledges that meaning does not change. Yet he also
acknowledges the limits of that insight when interpretation does
not only require knowing history, but also exercising judgment in
applying historical meaning to present circumstances. The inter-
preter must ensure the historical meaning’s fitting application “to
meet the changes of circumstances and situation.”
65
Tradition re-
veals the fitting application.
Tradition, as Edmund Burke illustrated,
66
gives insight into how
original meaning should apply, because inherent to a successful
tradition—that is, a tradition handed on to a new generation—is
some proven good use.
67
By definition, then, successful traditions
are not static—they show how a people carry out an understanding
of their history. That endurance depends on “interpretation and re-
formulation in order [for the preserved practice] really to reach
62. NEWMAN, supra note 60, at 200.
63. Joseph Cardinal Ratzinger, Preface,
ALCUIN REID, THE ORGANIC DEVELOPMENT OF
THE
LITURGY 11–12 (2d ed. 2005) (discussing tradition and proper development regard-
ing the Catholic Mass).
64. Pope Pius XII, Mediator Dei, E
NCYCLICAL ON THE SACRED LITURGY § 62 (1947).
65. Id. at § 63.
66. Burke never used the term “tradition,” but instead invoked “prescription,” a term
that “originated in Roman property law, where it referred to ownership by virtue of
long-term use, rather than by formal deed.” Y
UVAL LEVIN, THE GREAT DEBATE 140
(2014).
67. Id.
430 Harvard Journal of Law & Public Policy Vol. 46
each new generation.”
68
As such, consulting a tradition helps an in-
terpreter determine the difference between (acceptable) translation
of original meaning to new contexts and (unacceptable) transfor-
mation of the original meaning to a new essence. By being immersed
in how a practice underlying a constitutional provision applies over
time, a judge therefore becomes immersed in the society’s tradition
of the underlying substance. In being so immersed, the judge ap-
proaches interpretation like a “gardener,” determining the “the in-
ner structural logic” of text and history well enough to ensure that,
even as circumstances give rise to new questions and situations,
constitutional meaning is faithfully transmitted to subsequent gen-
erations.
69
Tradition’s regard for enduring practice provides a check against
overly theoretical approaches to text and history interpretation. As
Professor Michael McConnell put it, “[t]he fundamental conceptual
error with respect to all [judicial] methodologies, but especially
originalism, is the belief that they will necessarily produce a single
right answer to the disputed legal question.”
70
Rather, text and his-
tory “more often exclude[] certain possibilities” than they “pro-
vide[] clear answers.”
71
If, in the face of that ambiguity, tradition is
ignored, then the inertia of wanting a Single Right Answer will still
insist on one—even if it means contravening longstanding practices
that support an alternative reading of the original evidence. Insist-
ing on such interpretations reflects a view of text and history that
expects them to “accomplish too much” by “wrest[ing] a greater
precision” than either warrant.
72
Moreover, this approach sacrifices
the judiciary’s distinct vantage point in the federal system: an
68. JOSEF PIEPER, TRADITION 50 (St. Augustine Press 2010).
69. See Ratzinger, supra note 63.
70. McConnell, supra note 38, at 1761.
71. Id. at 1761, 1787.
72. Id. at 1760; see also Thomas Merrill, Bork v. Burke, 19 H
ARV. J.L. & PUB. POL’Y 509,
520–21 (1996) (“[O]riginalism by its very nature requires that the interpreter compre-
hend and adopt the values, aspirations, and linguistic conventions of a society several
steps removed in time from our own. . . . One can fairly question whether the average
judge or lawyer . . . is capable of carrying off this kind of inquiry.”).
2023 Keeping Our Balance 431
institution removed from political forces such that it can apply
foundational principles to the “flesh-and-blood” of an actual case.
73
Tradition upholds the judicial role by taking the “range of plausible
interpretations” from text and history and identifying “concrete
practices”—ones of “substantial duration,” from both “the political
organs of government” and also “individual citizens or groups of
citizens”—that then become presumptively “determinative” of
constitutional meaning.
74
For similar reasons, American constitutionalism “accords the
past an authority that philosophy does not.”
75
This is evident in the
embrace of stare decisis,
76
discussions of constitutional interpreta-
tion in the Federalist Papers,
77
the widespread influence of the British
common law,
78
and the role of longstanding practice in founda-
tional Supreme Court decisions, like McCulloch v. Maryland.
79
In-
deed, James Wilson—known today for his commitment to natural
law and natural rights—called custom “the most significant, and
73. See ALEXANDER BICKEL, THE LEAST DANGEROUS BRANCH 116 (1962).
74. DeGirolami, supra note 14, at 6–7. See also McConnell, supra note 16, at 1771 &
n.106 (“I wrote of this methodology years ago under the name of ‘traditionalism,’” but
“that name did not catch on,” as the Court would prefer “longstanding practice.”);
NLRB v. Noel Canning, 573 U.S. 513, 513–14 (2014). For the presumptive role, see infra
Part II.A–B (discussing Bruen’s use of presumptions and presumptions in the Establish-
ment Clause context).
75. Anthony Kronman, Precedent and Tradition, 99 Y
ALE L.J. 1029, 1034 (1991).
76. See T
HE FEDERALIST NO. 78 (Alexander Hamilton) (Clinton Rossiter ed., 2003).
77. “Laws,” Hamilton says “are a dead letter without courts to expound and define
their true meaning and operation.” Id. N
O. 22, at 146 (Alexander Hamilton). And those
meanings will be “liquidated and ascertained by a series of particular discussions and
adjudications.” Id. N
O. 37, at 225 (James Madison). As such, “the natural and obvious
sense of [the Constitution’s] provisions, apart from any technical rules, is the true cri-
terion of construction.” Id. N
O. 83, at 496 (Alexander Hamilton). Because such “rules of
legal interpretation” are determined by “conformity to the source which they are de-
rived,” id. at 495, and American law draws on authority that is “ancient as well as nu-
merous,” id. N
O. 49, at 312 (James Madison) (emphases omitted), judges must be
formed in the “long and laborious” study in not only law’s technical maxims, but also
its origins in the people’s traditions, id. N
O. 78, at 470 (Alexander Hamilton).
78. D
ANIEL J. BOORSTIN, THE MYSTERIOUS SCIENCE OF THE LAW 73 (1941).
79. 17 U.S. (4 Wheat.) 316, 401–407 (1819); see also B
ICKEL, THE LEAST DANGEROUS
BRANCH, supra note 73, at 105 (discussing McCulloch).
432 Harvard Journal of Law & Public Policy Vol. 46
the most effectual” sort of law, because its continuance shows “in-
ternal evidence, of the strongest kind, that the law has been intro-
duced by common consent; and that this consent rests upon the
most solid basis—experience as well as opinion.”
80
Such customs
are identified and upheld by text, history, and tradition. That inter-
pretive sequence assesses constitutional ambiguities by taking the
longstanding practice of a given political institution or community
and relating it—“at least analogically”—to “the historically defined
hard core” of the guarantee at issue.
81
B. Standard originalism operationalizes text, history, and tradition.
As standard originalism’s foremost expositor,
82
it is no surprise
that Justice Scalia offered the most thorough guidance for opera-
tionalizing tradition’s supplemental role to text and history.
83
Scalia’s guidance began with a crucial point: tradition “giv[es]
content only to ambiguous constitutional text; no tradition can su-
persede the Constitution.”
84
Second, “tradition” cannot be invoked
abstractly. Rather courts should identify traditions at “the most
specific level,” regardless of whether the identified tradition is
“protecting, or denying protection to, the asserted right.”
85
That is
not to say that more “general” traditions are unhelpful. They can
be helpful.
86
But the more general the tradition, the more “impre-
cise” its “guidance,” and the more important it becomes that the
80. JAMES WILSON, ON THE GENERAL PRINCIPLE OF LAW AND OBLIGATION (1790 –
1791), reprinted in C
OLLECTED WORKS OF JAMES WILSON 470 (Kermit L. Hall & Mark
David Hall eds., 2007).
81. See A
LEXANDER BICKEL, THE MORALITY OF CONSENT 18, 29 (1975).
82. See Samuel A. Alito Jr., Remarks to the 2020 Federalist Society National Lawyers Con-
vention, 45 H
ARV. J.L. & PUB. POL’Y 83, 100 (2022).
83. McConnell, supra note 14, at 1136 (“What Scalia rejects is the idea that the nation
should be governed not by the will of the people over time, but by the opinions of
judges, or of the legal elite.”).
84. Rutan v. Republican Party of Ill., 497 U.S. 62, 96 n.1 (1990) (Scalia, J., dissenting);
see also NLRB v. Noel Canning, 573 U.S. 513, 573 (2014) (Scalia, J., concurring).
85. See Michael H. v. Gerald D., 491 U.S. 110, 127 n.6 (1989) (plurality opinion) (Scalia,
J.).
86. Id. (saying you can “consult and (if possible) reason from” them).
2023 Keeping Our Balance 433
tradition is continuing and widespread before it can be determina-
tive.
87
Third, the invoked tradition must be one of “unchallenged
validity.”
88
A tradition has unchallenged validity when it “ha[s] not
been vigorously opposed on constitutional grounds,” meaning it
hasn’t been “litigated up to th[e Supreme] Court,” or “upheld only
over [a historically vindicated] dissent.”
89
When unchallenged tra-
ditions are identified, they “are the best indication of what funda-
mental beliefs [a constitutional text] was intended to enshrine.”
90
Yet fourth, if the tradition itself is not going to resolve the case—
but instead helps direct one of the Court’s “abstract tests”—then
the Court should “craft[]” the test “so as to reflect[]those constant
and unbroken traditions.”
91
Yet Justice Scalia’s guidance for “text, history, and tradition” is in
tension with his regard for judicial restraint.
92
Scalia did not “artic-
ulate the connection between these methods, or . . . explain how to
decide cases when they are in conflict.”
93
His judicial opinions sug-
gest, however, that tradition should be subordinated to judicial re-
straint. As he said in McIntyre v. Ohio Elections Commission,
94
this is
“the most difficult” issue for originalists.
95
That is because in cases
where tradition-based evidence could illuminate ambiguities in
87. Id.
88. See Rutan, 497 U.S. at 96 n.1 (Scalia, J., dissenting) (explaining why Brown v. Board
of Education was right to overrule Plessy v. Ferguson: “a tradition of unchallenged valid-
ity did not exist with respect to the practice in Brown”); see also Noel Canning, 573 U.S.
at 573 (Scalia, J., concurring) (arguing that “a self-aggrandizing practice adopted by one
branch well after the founding, often challenged, and never blessed by this Court” can-
not contravene original understanding).
89. See Rutan, 497 U.S. at 96 n.1 (Scalia, J., dissenting).
90. McIntyre v. Ohio, 514 U.S. 334, 378 (1995) (Scalia, J., dissenting).
91. United States v. Virginia, 518 U.S. 515, 570 (1996) (Scalia, J., dissenting); Planned
Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833, 981 (1992) (Scalia, J., dissenting)
(“in defining ‘liberty,’ we may not disregard a specific, ‘relevant tradition protecting,
or denying protection to, the asserted right’” (quoting Michael H. v. Gerald D., 491 U.S.
110, 127 n.6 (1989)).
92. McConnell, supra note 14, at 1137 & n.45.
93. Id. at 1137 n.45.
94. 514 U.S. 334 (1995).
95. Id. at 375 (Scalia, J., dissenting).
434 Harvard Journal of Law & Public Policy Vol. 46
semantic or original meaning, “constitutional adjudication neces-
sarily involves not just history but judgment: judgment as to
whether the government action under challenge is consonant with
the concept of the protected freedom . . . that existed when the con-
stitutional protection was accorded.”
96
But Scalia’s preference for
general rules that “hedge” judges in
97
was designed to prevent
judges from rendering judgment on “the concept of the protected
freedom” at issue. This led Justice Scalia to condition his evaluation
of longstanding practice on what would, in his view, better limit
judges. For example, in McIntyre he wrote that if “[a] governmental
practice” restricting a Founding-era practice “has become general
throughout the United States,” then it is presumptively constitu-
tional—even if it began over a century after the Founding.
98
Simi-
larly in Brown v. Entertainment Merchants Association,
99
Justice Scalia
(for the Court) acknowledged that “long (if heretofore unrecog-
nized) traditions of proscription” could allow governments to adopt
“novel restriction[s]” on speech issues.
100
But, his reasoning dis-
couraged their development, lest the Court encourage case-by-case
adjudication.
101
Scalia’s theoretical concerns about restraint also ex-
plain why he embraced tradition in the Establishment Clause con-
text. Scalia’s concerns about limiting judicial judgment aligned
with his opposition to “formulaic abstractions” that take decisions
about permissible religious expression away from a community’s
“long-accepted constitutional traditions.”
102
But in the Free Exercise
96. Id.
97. See Antonin Scalia, The Rule of Law As A Law of Rules, 56 U.
CHI. L. REV. 1176, 1180
(1989) (“Only by announcing rules do we hedge ourselves in.”).
98. Id. at 375–76 (“The earliest statute of this sort was adopted by Massachusetts in
1890 . . . .”) (Scalia, J., dissenting).
99. 564 U.S. 786 (2011).
100. Id. at 792.
101. See id. But cf. id. at 821 (Alito, J., concurring) (“I would not squelch legislative
efforts to deal with what is perceived by some to be a significant and developing social
problem. If differently framed statutes are enacted by the States or by the Federal Gov-
ernment, we can consider the constitutionality of those laws when cases challenging
them are presented to us.”).
102. Lee v. Weisman, 505 U.S. 577, 644 (1992) (Scalia, J., dissenting).
2023 Keeping Our Balance 435
context—where “long-accepted constitutional traditions” regard-
ing religious accommodation might increase case-by-case adjudica-
tion—Scalia preferred the formulaic abstraction (“neutrality” +
“general applicability” = no relief).
While there can be a tension between tradition and restraint,
standard originalism does not require a conflict. As Professor
McConnell put it, “[t]he important point here is the sequencing.”
103
Both tradition and restraint “respect the will of the people as ex-
pressed at various points in time.”
104
Neither tradition nor restraint
seek to “upend existing social policy and to substitute its oppo-
site.”
105
But invoking restraint before tradition uproots restraint
from any grounding in text, history, and analogically demonstrated
practices. Such a jump means that deference to a present majority
is no longer a command of text, history, or analogical practice. Ra-
ther, this “restraint” is just “the judges’ own view of what should be
the constitutional constraint” that is allowed to “brush[] aside” “the
conventional legal analyses of text, history, practice, and prece-
dent.”
106
By contrast, as evidenced in recent Second Amendment
and Religion Clause decisions, text, history, and tradition could
achieve durable restraint by improving the court’s analytical preci-
sion. Judges could analogize the practice or regulation at issue to
what is known about the constitutional provision’s original mean-
ing. Over time, with the development of more specific historical
analogies, the increased analogical precision would either define or
displace the court’s resort to balancing tests.
107
103. McConnell, supra note 16, at 1788.
104. McConnell, supra note 14, at 1137 n.45.
105. McConnell, supra note 16, at 1781.
106. Id.
107. See McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 360 (1995) (Thomas, J.,
concurring); see also id. at 370–71; Brown v. Ent. Merchs. Ass’n, 564 U.S. 786, 839 (2011)
(Thomas, J., dissenting) (“Whether the statute would survive an as-applied chal-
lenge . . . is a question for another day.”); id. at 806 (Alito, J., concurring) (“In consider-
ing the application of unchanging constitutional principles to new and rapidly evolving
technology, this Court should proceed with caution. . . .”).
436 Harvard Journal of Law & Public Policy Vol. 46
II. T
EXT, HISTORY, AND TRADITION IN PRACTICE
New York State Rifle & Pistol Association v. Bruen
108
is a recent Sec-
ond Amendment decision that provides a comprehensive example
of text, history, and tradition. And Bruen’s use of tradition is
“adopt[ed from] a similar approach”
109
in recent Establishment
Clause cases. This section will consider both those cases and Bruen,
along with tradition’s use in the church autonomy context. All these
contexts contrast sharply with the Free Exercise Clause, where the
Smith approach spurns text, history, and tradition.
A. Bruen.
Building on Justice Scalia’s opinion for the Court in District of Co-
lumbia v. Heller,
110
Bruen evaluated whether the Second Amendment
allowed New York to condition a license to carry a gun on a “spe-
cial need for self-defense.”
111
The Court held that “the Constitution
presumptively protects th[e] conduct” that is “cover[ed]” by a con-
stitutional amendment’s “plain text”—unless the government can
“demonstrate that [its] regulation is consistent with this Nation’s
historical tradition of” regulating that conduct.
112
On this reading,
the Constitution’s text provides a presumption that government
cannot restrict a clearly granted freedom. In response, “the govern-
ment must affirmatively prove that” it can restrict the freedom
based on “the historical tradition that delimits the outer bounds” of
the right at issue.
113
This can be satisfied via “analogical reasoning,”
which requires “that the government identify a well-established
108. 142 S. Ct. 2111 (2022).
109. Bruen, 142 S. Ct. at 2130 (citing Am. Legion v. Am. Humanist Ass’n, 139 S. Ct.
2067, 2087 (2019) (plurality opinion)).
110. 554 U.S. 570 (2008). Heller refers to a tradition-based approach in identifying ac-
ceptable limits on the right to keep and carry arms. See id. at 627. But in defining the
right, Justice Scalia explained that the Second Amendment’s history is “unambigu-
ous[].” See id. at 580, 584.
111. Bruen, 142 S. Ct. at 2122.
112. Id. at 2126.
113. Id. at 2127.
2023 Keeping Our Balance 437
and representative historical analogue” to the law at issue, “not a
historical twin.”
114
Bruen’s analogical analysis illustrates well how tradition supple-
ments text and history. For example, Bruen says that the Court
“look[s] to history” because the Second Amendment “was not in-
tended to lay down a novel principle but rather a codified right in-
herited from our English ancestors.”
115
Consulting tradition, then,
identifies proper historical analogues and excludes “endorsing out-
liers that our ancestors would never have accepted.”
116
Bruen dis-
tilled a tradition of analogous firearm regulation from “(1) medie-
val to early modern England; (2) the American Colonies and the
early Republic; (3) antebellum America; (4) Reconstruction; and (5)
the late 19th and early 20th centuries.”
117
Yet none of this evidence
was meant to serve antiquarian ends—such that a single category
of evidence or isolated practices could displace the Second Amend-
ment’s ordinary understanding. As Bruen put it, “when it comes to
interpreting the Constitution, not all history is created equal.”
118
Ex-
cluded from the inquiry would be “an ancient practice that had be-
come obsolete . . . at the time of the adoption of the Constitution
and never was acted upon or accepted in the colonies.”
119
Similar
caution is deployed toward post-enactment history. While a court
can “liquidat[e] indeterminacies in written laws,” that is no license
to “expand[] or alter[] them.”
120
“Thus, post-ratification adoption or
acceptance of laws that are inconsistent with the original meaning
of the constitutional text obviously cannot overcome or alter that
text.”
121
Even with that caveat, Bruen acknowledged that “other
cases implicating unprecedented societal concerns or dramatic
114. Id. at 2133.
115. Id. at 2127 (internal quotation marks and citations omitted).
116. Id. at 2133 (quoting Drummond v. Robinson, 9 F.4th 217, 226 (3rd Cir. 2021).
117. Id. at 2135–36.
118. Id. at 2136.
119. Id. (internal quotation marks and citation omitted).
120. Id at 2137 (internal quotation marks and citation omitted).
121. Id. (internal quotation marks and citation omitted).
438 Harvard Journal of Law & Public Policy Vol. 46
technological changes may require a more nuanced approach.”
122
“Although [the Constitution’s] meaning is fixed according to the
understandings of those who ratified it, the Constitution can, and
must, apply to circumstances beyond those the Founders specifi-
cally anticipated.”
123
In adopting text, history, and tradition, Bruen “expressly rejected
the application of any judge-empowering interest-balancing in-
quiry.”
124
At the same time, Bruen perceives no conflict between
drawing precise historical analogies and the judicial role. Rather,
“answering these kinds of historical, analogical questions” is “an
essential component of judicial decisionmaking under our endur-
ing Constitution.”
125
Under Bruen, a “text, history, and tradition test”
126
identifies how
“earlier generations addressed the societal problem”—and those
resolutions give rise to constitutional presumptions.
127
“[W]hen a
challenged regulation addresses a general societal problem that has
persisted since the 18
th
century, the lack of a distinctly similar his-
torical regulation addressing that problem is relevant evidence that
the challenged regulation is inconsistent with [the Constitution].”
128
At the same time, “if earlier generations addressed the societal
problem, but did so through materially different means, that also
could be evidence that a modern regulation is unconstitutional.”
129
B. A “similar approach” to the Establishment Clause.
Bruen is “adopt[ed from] a similar approach”
130
in recent Estab-
lishment Clause cases. Those cases “abandoned” the “ambitious,
122. Id. at 2132; see also id. at 2162 (Barrett, J., concurring).
123. Id. at 2132 (majority opinion).
124. Id. at 2129 (internal quotation marks and citation omitted).
125. Id. at 2134 (internal quotation marks and citation omitted).
126. Id. at 2161 (Kavanaugh, J., concurring).
127. Id. at 2131 (majority opinion).
128. Id.
129. Id.
130. Id. at 2130 (citing Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2087 (2019)
(plurality opinion)).
2023 Keeping Our Balance 439
abstract, and ahistorical”
131
attempt to impose a “grand unified the-
ory” of “neutrality” on all public religious expression.
132
Instead,
starting with Marsh v. Chambers
133
and confirmed by Kennedy v.
Bremerton,
134
Establishment Clause jurisprudence is governed by
text, history, and tradition.
In Marsh, the Supreme Court upheld Nebraska’s practice of open-
ing legislative sessions with prayer.
135
It did so by referencing “his-
torical practices and understandings.”
136
“Standing alone,” Marsh
said, “historical patterns cannot justify contemporary violations of
constitutional guarantees, but there is far more here than simply
historical patterns.”
137
What confirmed the historical analysis was
enduring “practice.”
138
One simply cannot “cast aside” “two centu-
ries of national practice”—such an “unambiguous and unbroken
history” is a “part of the fabric of our society.”
139
Another legislative prayer case, Town of Greece v. Galloway,
140
built
on Marsh. That case built out tradition’s distinct contribution to text
and history. That is because, unlike Marsh, the “specific practice” of
prayer in Town of Greece “lacked the very direct connection, via the
First Congress, to the thinking of those who were responsible for
framing the First Amendment.”
141
Thus, appealing to the First
Amendment’s ratification history was insufficient. Instead, Town of
Greece explained that “[a]ny test the Court adopts must
acknowledge a practice that was accepted by the Framers and has
131. Kennedy v. Bremerton Sch. Dist., 142 S. Ct. 2407, 2427 (2022).
132. Am. Legion, 139 S. Ct. at 2087 (plurality opinion).
133. 463 U.S. 783 (1983).
134. 140 S. Ct. at 2427–2428.
135. Marsh, 463 U.S. at 786.
136. Cnty. of Allegheny v. Am. Civ. Liberties Union, 492 U.S. 573, 670 (1989) (Ken-
nedy, J., concurring in part).
137. Marsh, 463 U.S. at 790.
138. See id.
139. Id. at 792.
140. 572 U.S. 565 (2014).
141. Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2088 (2019) (plurality opin-
ion).
440 Harvard Journal of Law & Public Policy Vol. 46
withstood the critical scrutiny of time and political change.”
142
And
in Town of Greece, because the town’s practice “fi[t] within the tra-
dition” carried out by the First Congress and other state legisla-
tures, it was presumptively constitutional.
143
Further, in American Legion v. American Humanist Association—
(upholding the constitutionality of the Bladensburg peace cross, a
public religious memorial)—the Supreme Court’s analysis em-
braced tradition’s regard for social knowledge. There, the Court’s
opinion relied on a historical Establishment Clause analysis, but not
one that compared the Bladensburg cross to the Establishment
Clause’s original meaning.
144
Rather, “[t]he passage of time gives
rise to a strong presumption of constitutionality.”
145
That is because
“[w]ith sufficient time, religiously expressive monuments, sym-
bols, and practices can become embedded features of a commu-
nity’s landscape and identity. The community may come to value
them without necessarily embracing their religious roots.”
146
But, if
such a community icon was removed “or radical[ly] alter[ed] at this
date,” such an act “would be seen by many not as a neutral act but
as the manifestation of a hostility toward religion that has no place
in our Establishment Clause traditions.”
147
C. Church autonomy.
Tradition influences the use of text and history in “church auton-
omy” cases too. This autonomy has several “component[s],”
148
but,
“in short,” it is the “power” of religious organizations “to decide
142. Town of Greece, 572 U.S. at 576.
143. Am. Legion, 139 S. Ct. at 2088 – 2089 (quoting Town of Greece, 572 U.S. at 577)
(plurality opinion).
144. Id. at 2078 (majority opinion).
145. Id. at 2085.
146. Id. at 2084.
147. Id. at 2074 (internal quotation marks and citation omitted).
148. Our Lady of Guadalupe v. Morrissey-Berru, 140 S. Ct. 2049, 2060–61 (2020) (“[A]
component” of church autonomy is the “ministerial exception,” but the doctrine is a
“broad principle” covering “internal management decisions that are essential to the
institution’s central mission.”).
2023 Keeping Our Balance 441
for themselves, free from state interference, matters of church gov-
ernment as well as those of faith and doctrine.”
149
Tradition’s effect on church autonomy doctrine is obvious from
the protection’s constitutional source. Church autonomy, says the
Supreme Court, is neither the result of textualism nor purposivism,
but rather “the foundation of our political principles,”
150
a “broad
and sound view of the relations of church and state under our sys-
tem of laws,”
151
and a “sphere” of authority protected by the Free
Exercise Clause working in conjunction with the Establishment
Clause.
152
The Court’s two most recent church autonomy cases on
the merits—Hosanna-Tabor v. EEOC
153
and Our Lady of Guadalupe v.
Morrissey-Berru,
154
both involving the right of religious organiza-
tions to select their ministers without judicial interference—are
good examples.
Echoing tradition’s regard for social knowledge, both Our Lady
and Hosanna-Tabor expressly rejected the use of “rigid formula” to
identify who a “minister” is.
155
One reason why, as Our Lady ex-
plains, is that “judges cannot be expected to have a complete un-
derstanding and appreciation of the role played by every person
who performs a particular role in every religious tradition.”
156
Therefore, the religious institution’s understanding is “im-
portant.”
157
Similarly, both Hosanna-Tabor and Our Lady specify that
the goal of historical analysis isn’t to capture a given “moment’s”
understanding, but rather to determine the kind of “practices”
“that the founding generation sought to prevent a repetition
149. Kedroff v. Saint Nicholas Cathedral of Russian Orthodox Church in N. Am., 344
U.S. 94, 116 (1952).
150. Watson v. Jones, 80 U.S. 679, 728 (1871).
151. Id. at 727.
152. Our Lady, 140 S. Ct. at 2060; Hosanna-Tabor v. EEOC, 565 U.S. 171, 181, 183 – 84,
188–89 (2012).
153. 565 U.S. 171 (2012).
154. 140 S. Ct. 2049 (2020).
155. Our Lady, 140 S. Ct. at 2067 (quoting Hosanna-Tabor, 565 U.S. at 190).
156. Id. at 2066.
157. Id.
442 Harvard Journal of Law & Public Policy Vol. 46
of . . . in our country.”
158
This analysis of longstanding practice thus
included both pre- and post-ratification evidence. For example, Ho-
sanna-Tabor began with a discussion of historical British statutes
and their effect on Founding era practices.
159
Practices that occurred
post-ratification were also illustrative, including Thomas Jeffer-
son’s response to John Carroll in 1806, when Carroll sought federal
guidance on appointing a Catholic bishop for the territory acquired
via the Louisiana Purchase.
160
James Madison’s reaction to the 1811
incorporation controversies surrounding the Anglican Church in
Virginia was also considered.
161
Likewise, Our Lady surveyed the
historical importance of religious education across faiths, both at
present and “from the earliest settlements in this country.”
162
D. Smith is the outlier.
Justice Scalia did not employ text, history, and tradition in Smith.
Rather, he “filtered his originalism through the twin lenses of de-
mocracy and the need for clear rules over vague standards.”
163
Excluding evidence of longstanding practices regarding religious
accommodation, Smith attempted to develop a bright line rule: “the
right of free exercise does not relieve an individual of the obligation
to comply with a ‘valid and neutral law of general applicability on
the ground that the law proscribes (or prescribes) conduct that his
religion prescribes (or proscribes).’”
164
Smith’s new test displaced
the compelling interest test articulated in Sherbert v. Verner.
165
“Un-
der the Sherbert test, governmental actions that substantially bur-
den a religious practice must be justified by a compelling
158. Id. at 2061; see also Hosanna-Tabor, 565 U.S. at 183–184.
159. Id. at 182.
160. Id. at 184.
161. Id. at 184 – 85.
162. Our Lady, 140 S. Ct. at 2064–2066.
163. Amul R. Thapar, Smith, Scalia, and Originalism, 68 C
ATH. U. L. REV. 687, 695
(2019).
164. Emp. Div. v. Smith, 494 U.S. 872, 879 (1990) (quoting United States v. Lee, 455
U.S. 252, 263, n.3 (1982) (Stevens, J., concurring in the judgment)).
165. 374 U.S. 398 (1963); see Fulton v. City of Philadelphia, 142 S. Ct. 1868, 1893 (2021)
(Alito, J. concurring) (recognizing the displacement).
2023 Keeping Our Balance 443
governmental interest.”
166
Smith, however, claimed the compelling
interest test was “never applied” to provide free-exercise accom-
modations,
167
and that this was for good reason because it would
produce a “constitutional anomaly”
168
and “court[] anarchy.”
169
Smith displays the “sequencing” problem discussed above,
whereby the supplementary role of tradition is discarded by jump-
ing immediately from textual ambiguity to judicial restraint. “As a
textual matter,” all Smith holds is that it “do[es] not think the [Free
Exercise Clause] must” be construed to require accommodations.
170
Rather, Smith rests on a self-consciously pragmatic construction—
calling its new interpretation “permissible,” “preferred,” and
“sounder” than a pro-accommodation interpretation, but never re-
quired.
171
Seven years later, when Justice Scalia would respond to
historical evidence against Smith in City of Boerne v. Flores, he made
similar defenses.
172
These arguments led Scalia in both opinions to
eschew any reliance on evidence of longstanding religious accom-
modations. “There is no reason to think [those practices] were
meant to describe what was constitutionally required (and judi-
cially enforceable), as opposed to what was thought to be legisla-
tively or even morally desirable.”
173
That a legislature was “ex-
pected to be solicitous of [religious accommodation] in its
legislation” does not mean “the appropriate occasions for [their]
creation can be discerned by the courts.”
174
True, “[i]t may fairly be
said that leaving accommodation to the political process will place
at a relative disadvantage those religious practices that are not
166. Smith, 494 U.S. at 883 (citing Sherbert, 374 U.S. at 402 – 03.
167. Id. at 884–85.
168. Id. at 886.
169. Id. at 888.
170. Id. at 878.
171. Id.
172. See City of Boerne v. Flores, 521 U.S. 507, 539 (1997) (Scalia, J., concurring).
173. Id. at 541 (Scalia, J., concurring); see also Smith, 494 U.S. at 890.
174. Smith, 494 U.S. at 890.
444 Harvard Journal of Law & Public Policy Vol. 46
widely engaged in.”
175
But that “unavoidable consequence of dem-
ocratic government must be preferred” by judges.
176
III. F
REE EXERCISE DOCTRINE SHOULD ADOPT TEXT, HISTORY, AND
TRADITION
Free Exercise doctrine would benefit from abandoning Smith and
instead applying text, history, and tradition. That approach would,
like Bruen, presumptively protect religious exercise unless the op-
posing party shows a historically analogous tradition of restricting
it. This section will begin by defending that approach against Smith
and its failed promise of judicial restraint. This section will then of-
fer specific ways in which Free Exercise doctrine could apply text,
history, and tradition.
A. Text, history, and tradition are more conducive to judicial re-
straint than Smith.
As discussed, the engine behind Smith is a certain view of judicial
restraint. This view of restraint might prompt some, who are other-
wise supportive of text, history, and tradition interpretation, to re-
sist using it in Smith’s place. On this view, “tradition” is the prob-
lem, because it is otherwise accepted that “text” and “history”
should take priority over “restraint.” This objection argues that
“tradition” could compromise “restraint” in at least three ways: (1)
by giving judges a reason to depart from what we know of the Free
Exercise Clause’s original meaning; (2) by introducing a subjective
debate over the proper way to characterize and analogize traditions
to the practice at issue; and (3) by using political choices regarding
religious accommodation as a presumptive indication of the Free
Exercise Clause’s meaning, tradition would erode the distinction
between legislative discretion and constitutional mandate. These
concerns are serious. But none is sufficient to keep the Free Exercise
175. Id.
176. Id.
2023 Keeping Our Balance 445
Clause a doctrinal outlier from tradition’s general use in Religion
Clause jurisprudence and other areas of constitutional law.
To counter these concerns, first consider that Smith is hardly up-
holding judicial restraint. As discussed, Smith skipped from the
Free Exercise Clause’s textual ambiguity on the issue of accommo-
dations to a preference of judicial restraint.
177
The result was “re-
straint” unmoored from text, history, and tradition—and thus not
really restraint at all.
178
Instead, Smith’s preference is policed by
“neutrality” and “general applicability,” two open-ended inquiries
that analyze a law’s legislative history,
179
decisionmaker motive,
180
and its disparate impacts.
181
Whether or not Justice Scalia intended
it, Smith’s inquiries resemble the Lemon test that he rightly derided
in the Establishment Clause context: “formulaic abstractions that
are not derived from, but positively conflict with, our long-ac-
cepted constitutional traditions.”
182
And just like the Lemon test, the
Supreme Court often distinguishes Smith’s inquiries, opting for a
context-specific rule instead.
183
Smith’s context—an “across-the-
177. Smith, 494 U.S. at 890.
178. See Scalia, supra note 97, at 1184–85 (“It is, of course, possible to establish general
rules, no matter what theory of interpretation or construction one employs. As one
cynic has said, with five votes anything is possible.”).
179. Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 557–558
(1993) (Scalia, J., concurring) (explaining that “[t]he Court analyzes the ‘neutrality’ and
the ‘general applicability’” questions “in separate sections . . . and allocates various in-
validating factors to one or the other of those sections,” while rejecting the need to make
“a clear distinction between the two terms” and the “legislative motive” analysis).
180. Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1919 (2021) (Alito, J., concurring)
(identifying problems with applying the “neutrality” analysis and motivations).
181. See id. at 1921–22 (Alito, J., concurring) (“Cases involving rules designed to slow
the spread of COVID-19 have driven that point home” that “[i]dentifying appropriate
comparators” “has been hotly contested”).
182. Lee v. Weisman, 505 U.S. 577, 644 (1992) (Scalia, J., dissenting); see also Fulton,
141 S. Ct. at 1922 (Alito, J., concurring) (“Much of Smith’s initial appeal was likely its
apparent simplicity. . . . Experience has shown otherwise.”).
183. Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2080 (2019) (plurality) (dis-
cussing Lemon test); see also Hosanna-Tabor v. EEOC, 565 U.S. 171, 190 (2012) (Smith
only applies to “outward physical acts”); Espinoza v. Montana Dep’t of Revenue, 140
S. Ct. 2246 (2020) (no mention of Smith); Carson v. Makin, 142 S. Ct. 1987 (2022) (no
mention); Fulton, 141 S. Ct. at 1877 (“falls outside Smith”).
446 Harvard Journal of Law & Public Policy Vol. 46
board-criminal prohibition on a certain form of conduct”
184
—sheds
little light on the lion’s share of Free Exercise cases. Treating Smith
like the doctrinal baseline, then, does not restrain judges. Rather,
judges can—and do—freely engage in “after-the-fact maneuver-
ing”
185
to retrofit government action around Smith’s inquiries.
“[S]ubscribing to Smith, particularly if one also believes the over-
stated claims of predictability made on its behalf, may mask the
truth of what judges actually do with free exercise cases.”
186
Second, these concerns overlook the fact that tradition is a supple-
mentary tool to text and history. Tradition does not override text
and history. Rather, it allows courts to analogize from longstanding
political or cultural practices toward religious exercise to resolve
ambiguities within semantic and original meaning. Here—as virtu-
ally everyone in the Smith debate acknowledges—the text and orig-
inal meaning
187
of the Free Exercise Clause are ambiguous about the
constitutional mandate for religious accommodations.
188
Therefore,
there must be some supplementary tool. For Smith, that
184. Emp. Div. v. Smith, 494 U.S. 872, 884 (1990).
185. Masterpiece Cakeshop, Ltd. v. Colo. C.R. Comm’n, 138 S. Ct. 1719, 1737–39
(2018) (Gorsuch, J., concurring).
186. M
ARC O. DEGIROLAMI, THE TRAGEDY OF RELIGIOUS FREEDOM 165 (2013).
187. Post-Civil War history cuts against Smith, as the Fourteenth Amendment’s fram-
ers “explicitly target[ed]” religion-neutral and generally applicable laws in the South
“as examples of what would become unconstitutional” via incorporation. Kurt T. Lash,
The Second Adoption of the Free Exercise Clause: Religious Exemptions Under the Fourteenth
Amendment, 88 N
W. U. L. REV. 1106, 1149 (1994).
188. Supra note 19 and accompanying text (Smith discussion on textual ambiguity);
City of Boerne v. Flores, 521 U.S. 507, 544 (1997) (Scalia, J., concurring in part) (original
meaning “more supportive of [Smith’s] conclusion than destructive”); McConnell, supra
note 16, at 1761; Michael W. McConnell, The Origins and Historical Understanding of Free
Exercise of Religion, 103 H
ARV. L. REV. 1409, 1415 (1990) (“possible interpretation”);
Philip A. Hamburger, A Constitutional Right of Religious Exemption: An Historical Perspec-
tive, 60 G
EO. WASH. L. REV. 915, 947–948 (1992) (“[T]he claim that the Free Exercise
Clause provided a right of exemption from civil laws depends upon evidence that
may be questioned.”); Fulton, 141 S. Ct. at 1882 (Barrett, J., concurring) (“history looms
large,” but not “compelling”). Moreover, Justice Alito’s Fulton concurrence does not
purport a conclusive original meaning answer either—partly because, as he explained,
the original meaning of the Clause occurred “before the concept of judicial review took
hold.” Fulton, 141 S. Ct. at 1907 (Alito, J., concurring).
2023 Keeping Our Balance 447
supplementary tool was restraint. But its reasons for that prefer-
ence—that judicial review of governmental practices could pro-
duce “danger[ous]” results,
189
be “horrib[ly]” standardless,
190
and
be “a constitutional anomaly”
191
—have all proven hyperbolic.
192
Another proposal to keep Smith would root its inquiries in an-
other preference: “principle,” one that would eschew tradition-
based evidence by claiming the Free Exercise Clause only protects
“religious worship as such.”
193
This is the view of Professor Vincent
Phillip Muñoz. Professor Muñoz candidly admits that he does not
favor the results his approach would produce—and for good rea-
son. This proposal suffers from the Single Right Answer problem:
by leaning heavily on theory and excluding practice, this approach
attempts to wrest more from original meaning than it can pro-
vide.
194
As a result, the inability to administer Smith remains,
195
189. Smith, 494 U.S. at 888.
190. Id. at 890 n.5.
191. Id. at 886.
192. For example, there is historical evidence of judicially mandated religious accom-
modations. See Stephanie Barclay, The Historical Origins of Judicial Religious Exemptions,
95 N
OTRE DAME L. REV. 55, 69 (2020). And judicial review of religious accommodations
resembles “as-applied” relief. Stephanie Barclay & Mark Rienzi, Constitutional Anoma-
lies or As-Applied Challenges? A Defense of Religious Exemptions, 59 B.C.
L. REV. 1595, 1611
(2018). Moreover, the judicial outputs in such cases show the Supreme Court is “up to
the task” of assessing “specific claims for exemptions as they ar[i]se.” Gonzales v. O
Centro Espirita Beneficente Uniao do Vegetal, 546 U.S. 418, 436 (2006).
193. See V
INCENT PHILLIP MUÑOZ, RELIGIOUS LIBERTY AND THE AMERICAN FOUND-
ING
59 (2022) (“worship as such,” a “cumbersome modifier” intended to prohibit only
“outlawing a practice on account of its religious character.”).
194. Id. at 306 (“The most fundamental tradition of American constitutionalism,
moreover, is not reliance on tradition. . . . To follow the Founders requires that we fol-
low their philosophical thinking.”).
195. For example, if a jurisdiction bans “all wine uses,” it bans religious wine uses
too. But on Professor Muñoz’s theory, that is not a Free Exercise violation. To Muñoz,
a jurisdiction would violate the Free Exercise Clause if it banned “religious uses of
wine.” But what if the jurisdiction banned “all wine uses, including religious uses?” In
every case, the violation of religious exercise is the same—but, to Muñoz, in only one
is it clear that the Free Exercise Clause cares. See id. at 260 n.10 (calling a regulation on
“drive-in spiritual services” “[a]n example of government regulation of religious exer-
cises as such,” and noting—but not explaining whether it’s significant—that the order
448 Harvard Journal of Law & Public Policy Vol. 46
while the content of the Free Exercise guarantee gets murkier.
196
And the virtue of tradition-based analysis—how ordinary Ameri-
cans understood the real-world application of their political princi-
ples—is considered a barrier to taking a principle to its furthest ex-
tent.
197
By contrast, with text, history, and tradition, religious exercise
that has a strong analogical connection to the Founding Era
198
“pertained exclusively to religious services”). While Muñoz claims that the Founders
understood “the natural boundaries” of religious liberty “to be established by the laws
of nature,” see id. at 60, applying his theory would seem to mark the boundaries on
religious liberty by legislative pedantry.
196. For example, Professor Muñoz claims that Lamb’s Chapel v. Center Moriches School
District, 508 U.S. 384 (1993), is correct, because it invalidated a New York school dis-
trict’s “after-hours-school-use polic[y]” that “forb[ade] religious exercises as such.”
M
UÑOZ, supra note 193, at 263. Muñoz is right that Lamb’s Chapel is correct, but it’s not
clear why he would think so. In Lamb’s Chapel, the religious exercise was “a film series
dealing with family and child-rearing issues faced by parents today.” 508 U.S. at 387.
Muñoz never explains how this is encompassed by what he calls “religious exercise as
such,” a concept his book equates with “the natural right of religious worship.”
M
UÑOZ, supra note 193, at 67; see also id. at 263. Indeed, one might think a jurisdiction
following his theory could prohibit showing this film series—because a film series is not
worship, and Muñoz claims that “the state may make exclusions on the basis of religion
as long as it does not exercise jurisdiction over religious exercises as such.” Id. at 269.
In Lamb’s Chapel, the government rule prohibited using school premises after hours “for
religious purposes.” 508 U.S. at 387. And the “six-part film series” was refused by the
school district because it appeared “‘to be church related.’” Id. at 389. Muñoz never
explains why either this ban or this denial crossed the line from what he considers per-
missible (“exclusions on the basis of religion”) to what he considers impermissible (“ju-
risdiction over religious exercise as such”). Unlike Professor Muñoz’s theory, the Su-
preme Court doesn’t condition religious liberty on drawing such difficult lines. See
Good News Club v. Milford Central School District, 533 U.S. 98, 126–27 (2001) (Scalia, J.,
concurring) (“we have previously rejected the attempt to distinguish worship from
other religious speech, saying that the distinction has [no] intelligible content, and fur-
ther, no relevance to the constitutional issue.” (internal quotation marks and citation
omitted) (alterations in original).
197. M
UÑOZ, supra note 193, at 226 (“[T]he aim of the inquiry is to determine as much
as possible about the original meaning of the principle itself, not any particular expected
applications of it, since these may fall short of or even contradict the principle.”).
198. This is not to say that the religious exercise at issue need have been exactly pre-
sent at the Founding. Long practice, as in American Legion, does give rise to a presump-
tion of constitutionality. Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2082 (2019)
2023 Keeping Our Balance 449
would be presumptively protected, unless the opposing party
shows that a long history of analogous restriction can overcome
that protection.
199
This approach possesses a built-in respect for en-
during democratic judgment. It cannot contravene what is known
about text and history, but it can help illuminate what they do not
definitively resolve. That is because “[e]nduring cultural and polit-
ical practices reflect the people’s judgments about what is con-
sistent with their fundamental law.”
200
This is the logic of “implied
ratification”—whereby the Constitution “derives its continued au-
thority from the implicit consent of each subsequent generation.”
201
Implied ratification only makes sense if the people’s longstanding
practices can generally be presumed to reflect what the Constitu-
tion guarantees.
202
To quote natural lawyer James Wilson again, this
is “evidence[] of the strongest kind” of enduring, “common con-
sent.”
203
The goal of this approach, then, is not one-sided. Some-
times, text, history, and tradition will benefit religious liberty.
Other times, the best analogies might justify restriction. But in ei-
ther case, Free Exercise doctrine would be more administrable than
judicially invented inquiries into “neutrality” and “general
(plurality opinion). But more recent religious exercise that implicates the government’s
reason for regulating in an analogous way would also receive the presumption. See
Ramirez v. Collier, 142 S. Ct. 1264, 1288 (2022) (Kavanaugh, J., concurring) (suggesting
that the “history of religious advisors at executions” shapes whether a compelling in-
terest in banning audible prayer and religious touch in death chamber exists, even as
“some of the history is not precisely on point”).
199. Religious liberty claimants may also use history and tradition—from that per-
spective, to affirmatively show that religious liberty presumptively includes the reli-
gious exercise at issue. See infra Part III.C (discussing Ramirez).
200. Marc O. DeGirolami, First Amendment Traditionalism, 97 W
ASH. U. L. REV. 1653,
1656 (2020).
201. McConnell, supra note 14, at 1132.
202. See id. Professor McConnell argues that implicit consent “must rest on more than
the mere fact that the people have not often amended the Constitution through the Ar-
ticle V procedures,” as that process “is sufficiently onerous that the mere lack of amend-
ments cannot, without more, be taken as proof of continued popular satisfaction with
the Constitution.” Id. Rather, it is the continued “venerat[ion of] the Constitution” by
the American people that shows enduring consent. Id.
203. W
ILSON, supra note 80, at 470.
450 Harvard Journal of Law & Public Policy Vol. 46
applicability”—and the Clause’s substance would better reflect
“the spirit of practical accommodation that has made the United
States a Nation of unparalleled pluralism and religious toler-
ance.”
204
Below are some specific ways in which these insights could be
applied.
B. Developing context-specific rules, not a one-size-fits-all test.
One way text, history, and tradition could improve Free Exercise
doctrine is, in place of Smith, courts could determine the propriety
of burdens on religious exercise through analogical reasoning
about longstanding practices. This approach would resemble Bruen
and the Establishment Clause cases: religious exercise is presump-
tively protected by the Free Exercise Clause’s text, unless the op-
posing party shows an unbroken, analogous tradition of restriction.
Used this way, tradition can help overcome the temptation toward
a single test to rule all Free Exercise cases.
For example, by using analogical reasoning to reconcile new gov-
ernment regulations with religious exercise, the judiciary can en-
sure that “the Free Exercise Clause [does not] shrink every time the
government expands its reach and begins to regulate work that has
historically and traditionally been done by religious groups.”
205
Smith, however, devised a rule from one context—“an across-the-
board criminal prohibition” enacted by a legislature
206
—and pur-
ports to apply that rule to myriad contexts, without regard to
whether those other contexts bear any resemblance to Smith’s. This
dynamic creates many awkward fits, especially with growing reg-
ulatory power. Indeed, most religious freedom cases at the Su-
preme Court in the past decade have come from administrative
204. Salazar v. Buono, 559 U.S. 700, 723 (2010) (Alito, J., concurring in part and con-
curring in the judgement).
205. Transcript of Oral Argument at 23, Fulton v. City of Philadelphia, 141 S. Ct. 1868
(2021) (No. 19-123) (statement of Lori Windham).
206. Emp. Div. v. Smith, 494 U.S. 872, 884 (1990).
2023 Keeping Our Balance 451
actions, not general, democratically enacted criminal laws.
207
As
Justice Scalia knew well, the administrative context is distinct from
legislation. Unlike the legislature, regulatory bodies—premised on
their “expertise” in technical knowledge—are generally disinclined
to accommodate religious orthodoxy or account for social
knowledge.
208
Outsourcing decisions to that context “breaks down”
Smith’s “political logic.”
209
By contrast, a context-specific approach
to religious exercise would allow text, history, and tradition to har-
monize free exercise with modern government power.
Two recent Free Exercise cases suggest a shift like this is already
underway. Tellingly, neither case cites Smith. In the first, Espinoza
v. Montana Department of Revenue,
210
the Court invalidated a fund-
ing prohibition on religious schools in part because there was no
“historic and substantial” tradition supporting such a ban.
211
Ra-
ther, the only “tradition” of such bans that did exist were the
207. See Carson v. Makin, 142 S. Ct. 1987, 1993 (2022) (executive department funding
determination); Fulton, 141 S. Ct. at 1875–76 (administrative decision from Department
of Human Services); Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63, 65–66
(2020) (per curiam) (shutdown executive order); Masterpiece Cakeshop, Ltd. v. Colo.
C.R. Comm’n, 138 S. Ct. 1719, 1723 (2018) (ruling of Colorado Civil Rights Commis-
sion); Trinity Lutheran Church of Columbia, Inc. v. Comer, 137 S. Ct. 2012, 2017 (2017)
(agency funding rule); Holt v. Hobbs, 574 U.S. 352, 358 (2015) (department grooming
policy); Burwell v. Hobby Lobby Stores, Inc., 573 U.S. 682, 696–97 (2014) (agency-
crafted mandate); Stormans, Inc. v. Wiesman, 136 S. Ct. 2433, 2434 (2016) (Alito, J., dis-
senting from denial of certiorari) (state pharmacy board rules).
208. Antonin Scalia, Rulemaking as Politics, 34 A
DMIN. L. REV. Xxv, xxxi (1982) (“More
needs to be done to bring the political, accommodationist, value-judgment aspect of
rulemaking out of the closet.”); see also Philip P. Hamburger, Exclusion and Equality: How
Exclusion from the Political Process Renders Religious Liberty Unequal, 90 N
OTRE DAME L.
REV. 1919, 1939–40 (2015) (“[T]he administrative idealization of scientism and central-
ized rationality usually renders administrative acts—compared with acts of Con-
gress—relatively indifferent and even antagonistic to religion and religious concerns.”)
209. See Hamburger, supra note 208, at 1938; see also Brief for Dominican Sisters of
Mary, et al. as Amici Curiae Supporting Petitioners at 3, Zubik v. Burwell, 578 U.S. 403
(2016) (No. 14-1418), 2016 WL 212595, at *3 (“HHS’s decision to gerrymander the ex-
emption in this way was intentional; it knew that in significant cases, virtually identical
religious groups would be treated differently based on nothing more than their classi-
fication under tax law.”).
210. 140 S. Ct. 2246 (2020).
211. Id. at 2258–59.
452 Harvard Journal of Law & Public Policy Vol. 46
nineteenth-century Blaine Amendments, laws reflecting a “big-
otry” toward Catholic immigrants—“hardly . . . a tradition that
should inform our understanding of the Free Exercise Clause.”
212
The Free Exercise Clause instead contains a “principle” of nondis-
crimination against religious status.
213
And the post-ratification ap-
plication of that principle illuminated no tradition denying reli-
gious schools the right to participate in neutral benefit programs.
214
In the second, Carson v. Makin,
215
the Supreme Court invalidated
a Maine statute that prohibited tuition assistance payments from
going to religious schools.
216
Echoing Smith, Maine (and Justice
Breyer in dissent) attempted to distinguish Espinoza from Carson:
Religious schools are not “‘bar[red] from receiving funding simply
based on their religious identity,’ but instead ‘based on the religious
use that they would make of it in instructing children.’”
217
This
means the restriction only has “the effect of burdening a particular
religious practice,” and under Smith, that is not a cognizable Free
Exercise concern.
218
But Carson didn’t limit its analysis to “how the
benefit and restriction are described”—the Court focused instead
on how “the program operates.”
219
As in Espinoza, the Court said
there was no “historic and substantial tradition” that could credit
“promot[ing] stricter separation of church and state than the Fed-
eral Constitution requires.”
220
212. Id. at 2259.
213. Id. at 2254 (quoting Trinity Lutheran Church of Columbia, Inc. v. Comer, 137 S.
Ct. 2012, 2019 – 21 (2017)).
214. Id. at 2258 – 59.
215. 142 S. Ct. 1987 (2022).
216. Id. at 2002.
217. Id. at 2001 (quoting Carson v. Makin, 979 F.3d 21, 40 (1st Cir. 2020), rev’d, 142 S.
Ct. 1987 (2022)).
218. See Emp. Div. v. Smith, 494 U.S. 872, 886 & n.3 (1990).
219. Carson, 142 S. Ct. at 2002.
220. Id. at 1997, 2002 (quoting Espinoza v. Mont. Dep’t of Rev., 140 S. Ct. 2246, 2259
(2020)).
2023 Keeping Our Balance 453
C. Determining “compelling” interests.
Text, history, and tradition could also be used to define the search
for “compelling” interests. This way, those interests are interpreted
“so as to reflect” the practices that illuminate how Americans ap-
plied the free exercise guarantee.
221
Three justices in Fulton sug-
gested something like this, saying that the compelling interest test
could replace Smith and be “rephrased or supplemented with spe-
cific rules.”
222
And Justice Scalia’s guidance for tradition’s use, dis-
cussed above, suggests some ways this could be implemented:
Identify, at the most specific level of analogy, a tradition of burden-
ing a particular religious exercise.
223
This tradition must be one of
“unchallenged validity,” meaning that Supreme Court jurispru-
dence has neither rejected such regulation nor put it into serious
doubt.
224
New regulatory traditions can emerge, but they must have
roots in older, analogous ones.
225
The Supreme Court gestured toward this approach in Ramirez v.
Collier.
226
There, the longstanding protection for clergy prayer in the
death chamber meant that Texas lacked a compelling interest in
denying an inmate’s request to have “his long-time pastor . . . pray
with him and lay hands on him while he is being executed.”
227
Alt-
hough a statutory case, Ramirez’s compelling interest analysis mir-
rors what would occur under the Free Exercise Clause. In defining
the compelling interest, the Court illustrated the role tradition plays
221. United States v. Virginia, 518 U.S. 515, 568 (Scalia, J., dissenting).
222. Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1924 (2021) (Alito, J., concurring);
see also Ramirez v. Collier, 142 S. Ct. 1264, 1288 (2022) (Kavanaugh, J., concurring) (ar-
guing that a tradition-based analysis could ensure that “the Court does not merely
point to its own policy assessment” when determining state interests).
223. See Michael H. v. Gerald D., 491 U.S. 110, 127 n.6 (1989) (opinion of Scalia, J.).
224. See Rutan v. Republican Party of Ill., 497 U.S. 62, 95 n.1 (1990) (Scalia, J., dissent-
ing).
225. See N.Y. State Rifle & Pistol Ass’n v. Bruen, 142 S. Ct. 2111, 2133 (2022).
226. 142 S. Ct. 1264 (2022).
227. Id. at 1272.
454 Harvard Journal of Law & Public Policy Vol. 46
in focusing the inquiry.
228
Justice Kavanaugh, moreover, concurred
to explain how “the history of religious advisors at executions”
meant “the Court does not merely point to its own policy assess-
ment of how much risk the State must tolerate in the execution
room.”
229
While “the history is not precisely on point,” because the
nature of the execution here was different than in prior examples,
“[s]till, the history generally demonstrates that religious advisors
have often been present at executions.”
230
That historical analysis,
as the majority opinion says, evidenced a “tradition [that] contin-
ued throughout our Nation’s history” and “continues today”
231
—a
fact Justice Kavanaugh considered “perhaps even more relevant”
than history.
232
D. Crafting specific rules for institutional religious exercise.
Text, history, and tradition would also be helpful in building out
distinct protections for religious institutions. This build-out could
happen alongside or independent of the previous suggestions.
As explained, church autonomy cases are already incorporating
tradition’s regard for social knowledge by adopting legal standards
that are not rooted in abstractions.
233
Similarly, recent Free Exercise
decisions have relied on the distinctive knowledge and mission of
religious institutions when crafting legal rules. For example, in Ful-
ton, the Court’s ruling is colored by the “incongruity” of labeling a
religious foster care agency a public accommodation when it is
asked to evaluate marriages but disregard its religious understand-
ing of marriage.
234
Similarly in Carson, the Court rejected Maine’s
“semantic” distinctions between restricting a religious school’s use
228. See id. at 1277–79 (citing “[a] tradition of such prayer continu[ing] throughout
our Nation’s history” to undermine the need for “a categorical ban on audible prayer
in the execution chamber”).
229. Id. at 1288 (Kavanaugh, J., concurring).
230. Id.
231. Id. at 1279 (majority opinion).
232. Id. at 1289 (Kavanaugh, J., concurring).
233. Supra Part II.C.
234. Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1881 (2021).
2023 Keeping Our Balance 455
of public money and its status as a religious organization.
235
Doing
otherwise would let courts “scrutiniz[e] whether and how a reli-
gious school pursues its educational mission,” thereby “rais[ing]
serious concerns about state entanglement with religion and de-
nominational favoritism.”
236
To make this point, Carson explicitly
connects free-exercise doctrine with the church autonomy cases.
237
This parallel could be further developed should the Supreme Court
consider “whether the freedom for religious employers to hire their
co-religionists is constitutionally required,” especially as “federal
statutory exemptions” and lower court decisions have long
acknowledged it.
238
Building out these tradition-based rules would
allow courts to better distinguish “internal management decisions
that are essential to the institution’s central mission” from decisions
capable of secular regulation.
239
C
ONCLUSION: KEEPING OUR BALANCE.
If the Free Exercise Clause is going to “translat[e]” its guarantees
“into concrete restraints” over time, then it needs text, history, and
tradition.
240
Applying that approach would resolve the morass cre-
ated by Smith’s unrestrained inquiries into “neutrality” and “gen-
eral applicability.” Further, adopting text, history, and tradition
would bring the Free Exercise Clause into line with the rest of Reli-
gion Clause jurisprudence, and the growing use of text, history, and
tradition throughout constitutional law. Finally, and as important,
by accounting for the people’s longstanding practices toward reli-
gious accommodation, the Free Exercise Clause would be neither a
235. Carson v. Makin, 142 S. Ct. 1987, 1999–2001 (2022).
236. Id. at 2001 (citing, inter alia, Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140
S. Ct. 2049, 2068 – 69 (2020)).
237. Id. Four justices would reiterate this connection again in Yeshiva Univ. v. YU
Pride All., 143 S. Ct. 1, 2 (2022) (Alito, J., dissenting).
238. Seattle’s Union Gospel Mission v. Woods, 142 S. Ct. 1094, 1094 (2022) (Alito, J.,
statement respecting the denial of certiorari).
239. Our Lady, 140 S. Ct. at 2060.
240. W. Va. Bd. of Educ. v. Barnette, 319 U.S. 624, 639 (1943).
456 Harvard Journal of Law & Public Policy Vol. 46
“living” text nor a “dead” letter. Rather, it would be as it should: an
enduring guarantee.
