University of Notre Dame

The Limits of Church Autonomy

March 25, 2023

View PDF



Lael Weinberger*

American courts apply “church autonomy doctrine” to protect the self-governance of religious institutions, based on both of the First Amendment’s religion clauses.  Church autonomy’s defenders have sometimes described the doctrine as establishing distinct spheres of sovereignty for church and state.  But critics have argued that church autonomy puts religious institutions above the law.  They contend that church autonomy doctrine lacks limiting principles and worry that the “sphere sovereignty” theory of church and state leaves no room for accountability for wrongdoing in religious institutions.  The courts, for their part, have recognized that church autonomy must have limits but have struggled to articulate them, leaving the caselaw in a state of ferment. 

This Article makes the case that, contrary to the critics, church autonomy is limited by an accountability principle, itself resting on the same bases that have been used to defend the most robust version of church autonomy.  First, the social pluralist theory of sphere sovereignty does not just defend a place for religious institutions to exercise their own self-governance over religious matters; it also has an important place for the state to hold wrongdoers accountable for civil harms.  Second, the deep history of church-state relations that has shaped the pro-church autonomy caselaw and scholarship alike also has rich resources to defend a principle of accountability. 

After presenting the theoretical case for the coexistence of autonomy and accountability principles, this Article presents a doctrinal roadmap for how courts can locate the limits of church autonomy.  Drawing on doctrinal elements already present in the caselaw, the approach outlined here can be applied to provide accountability and limit church autonomy in key cases—and it can be done without contradicting any existing Supreme Court doctrine.


The topic of religious liberty has become increasingly divisive in recent years.1  While its advocates think it embattled, critics think that religious liberty is more expansive than ever.  This concern has even seeped into the church autonomy cases, a body of caselaw that for a long time managed to stay out of the limelight.2  The courts have held that religious institutions have a right to internal self-government in managing their own affairs,3 a doctrine most commonly known as church autonomy.4  Defenders of the doctrine have gone so far as to describe the church as possessing a “sphere of sovereignty” distinct from that of the state.5  Critics, though, argue that church autonomy threatens to put religious institutions above the law.6  And as more and more religious liberty arguments are asserted by institutions—churches, parachurch organizations, religious schools, and more—the stakes for figuring out what is protected by church autonomy grow ever higher.7

No issue raises this concern as sharply as the clergy sexual abuse cases.8  The cases cry out for legal accountability.  But the critics of church autonomy argue that it is impossible to reconcile accountability in a principled way with the strong form of church autonomy currently ascendant in the courts.  “Church governance” could cover any number of abuses.9  Even some scholars sympathetic to some form of church autonomy worry that the stronger forms of church autonomy being adopted in the courts and defended by some scholars provide overbroad immunities for religious entities.10  If religious institutions are treated as having a sphere of sovereignty of their own, surely this goes too far as it provides no means for the state to hold actual wrongdoers accountable. 

The critique is a challenge to the courts and to the scholars who support a robust theory of church autonomy.  Can church autonomy legal protections shield churches from accountability for serious wrongs?  If not, what is the outer limit on church autonomy? 

This Article explores the limits of church autonomy.  It argues that church autonomy does not, and should not, protect abuse and many other forms of wrongdoing.  And it does this even after stacking the deck in favor of a robust church autonomy.  For purposes of argument, it takes as a given the Supreme Court’s precedents creating church autonomy doctrine.  It also takes as its starting point two of the more controversial bases for expansive church autonomy: first, that autonomy is based on a theory of institutional sovereignty, and second, that the long history of church-state contestation in English law can illuminate its contours.  Even with all of this granted—weighting the scales heavily in favor of church autonomy—this Article argues that limits to autonomy can be found and a baseline of state authority to hold religious actors accountable for wrongdoing can be established. 

While courts have generally intuited the coexistence of church autonomy and accountability, they have had a harder time articulating how to put the two together.  The courts have lacked a consistent account of church autonomy’s limits.  So, finally, this Article explains how these principles could be cashed out in concrete doctrinal form, using doctrinal resources that are already in the toolkit of the courts.  The conclusion is that a robust version of church autonomy is compatible with accountability.

Parts I and II of this Article are descriptive, identifying and analyzing the challenges for church autonomy as a field; Parts III and IV are prescriptive.  Part I examines the critique of church autonomy, focusing particularly on the potential it provides to cover for abuse.  Part II examines the doctrinal tools that the courts are currently using to try to identify the outer limits of church autonomy, arguing that the caselaw is currently in chaos as the courts use doctrines in inconsistent ways.

Part III offers two arguments for what I will call an accountability principle alongside autonomy.  It looks at the rationales that have been employed by church autonomy’s advocates to justify church autonomy and argues that each of these rationales also supports a place for the state in requiring accountability for harmful wrongdoing like clergy abuse.  In other words, autonomy and accountability are not opposed principles but should be seen as complementary.  Finally, Part IV turns to the doctrine, with an argument for how to implement the accountability principle in court.  In the process, this Part proposes a path toward refining the doctrine to make the caselaw more predictable and less chaotic. 

I.     Church Autonomy and the Problem of Abuse in Religious Organizations

Courts have been struggling to find the limiting principle for church autonomy.  The Supreme Court has articulated a core of protection for religious institutions but has not said more than broad generalities about its limits.11  The field is vulnerable to critique for lacking a clear limiting principle.12

Lower courts have tried to articulate limits.  But they have come up with such a variety of limiting principles as to create a tangle of confusing statements about the law.  Courts now pull from a grab bag of doctrinal tools without much clarity as to when one approach or another will be used.  The same principles come up from one jurisdiction to another but in inconsistent ways, creating circuit splits and doctrinal tensions among the lower courts and sometimes even with the Supreme Court’s precedent.13

Section A introduces church autonomy as a legal principle.  Section B shows how church autonomy has been used to complicate accountability for churches in cases of abuse.  Section C turns to the scholarship arguing that the problem of clergy abuse (and similar issues) should lead to a rollback of church autonomy.  

A.   What Church Autonomy Protects

Religion is usually a collective activity.14  Church autonomy protects the internal self-governance of religious organizations.15  Church autonomy is usually explained as a means of protecting religious institutions from state control, ensuring that religious organizations can control their own beliefs and internal affairs, or ensuring that the state does not establish a religion.  No one wants the state telling religious bodies whom it can or can’t retain as a minister, rabbi, or imam, or endorsing the theological distinctives of one side of a church split when the competing factions disagree about what counts as the “true” form of a given faith.16

The courts have based church autonomy on both religion clauses of the First Amendment.  Interfering with the internal governance of a religious institution would violate religious liberty (free exercise) and establish a religion by allowing the state to dictate the conduct of the religious body.  Scholars disagree about whether the Establishment Clause17 or the Free Exercise Clause18 provides the better foundation,19 or whether it is best to view church autonomy as the combined effect of the religion clauses, as the Supreme Court has said.20  But in any case, church autonomy ensures the institutional separation of church and state.  The generic “church” refers to a religious institution with its own religious (ecclesiastical) government; “church autonomy” protects this governance from being controlled by the civil authority.21

The judicially recognized church autonomy doctrine makes a claim about the content of the positive law—modernly, that the Constitution protects autonomy for religious institutions.22  The positivist right has, in turn, been justified on several different normative bases:

  • Liberal: Church autonomy is simply the institutional outworking of a liberal policy of religious liberty for individuals, coupled with voluntarist social organizations.23
  • Pluralist: Society consists of multiple centers of human activity and authority.  The state is only one among several, and ensuring that it not claim ultimate authority over all of life is essential to preserve freedom.  Religious institutions serve a useful function in preserving a sphere of authority outside the state.  Some also argue that religious institutions are, in principle, essential institutions to recognize as separate from the state.24
  • Religious: Some of the arguments for why religious institutions must be institutionally separate from the state are themselves religious arguments: that God and Caesar are distinct authorities in Scripture, that a Christian (or other) tradition requires an institutional separation of religious authority from the political authority.25

The principle of church autonomy can be divided into two main applications.  First, courts should not decide matters that require a particular position on religious doctrine or belief.26  So if the court is asked to decide whether a member was wrongfully expelled on the basis of the church’s doctrinal standards, the court should decline.27  Second, courts should not decide matters that interfere with the religious institution’s internal governance.28  This component of the doctrine “protect[s] their autonomy with respect to internal management decisions that are essential to the institution’s central mission.”29  This includes selecting ministers, which is known as the “ministerial exception” from employment nondiscrimination law.30  It also has been long understood to protect churches from defamation lawsuits challenging church discipline proceedings.31 

These core issues of church autonomy are well settled and relatively straightforward in application.32  But beyond the core concerns, both principles get tricky to apply.  The principle of not deciding religious doctrine in church property cases was historically complicated.  A quick version of this history shows why.  In the early nineteenth century, American courts followed the common-law tradition that religious property was held by trustees in trust for the benefit of a congregation, defined in terms of doctrine.  If there was a church split, each side would claim to adhere with more fidelity to the founding principles of the church and thus have the better claim to the property.33  This was transparently a doctrinal determination.  When the Supreme Court discarded this approach,34 the remaining question was whether courts should strictly apply the terms of a deed, or if the courts should defer to any internal church hierarchy that resolved which faction was the rightful claimant to the property.  Both approaches live on in the cases today, and courts and scholarly commentators alike are divided as to which is the more serious interference with religious autonomy.35

Perhaps even more challenging is figuring out a limit for the self-governance principle.  If the self-governance principle only applies when there is a religious reason for the particular decision, doesn’t that collapse the self-governance principle into the doctrinal autonomy principle?  It would also be inconsistent with the way the Supreme Court has interpreted the ministerial exception from employment laws.  In that setting, the Court has said that the right to self-governance of the religious body precludes the application of employment nondiscrimination law, regardless of whether the employment decision made by the church was based on a doctrinal reason.36

The difficulty in figuring out exactly where self-government ends for matters conducted within the church, in turn, raises the prospect of a church covering up wrongful acts done within the church as matters of church governance.  How far could a church go in controlling, manipulating, or abusing a member while claiming to be engaged in internal self-governance and hence immune from legal accountability? 

B.   Church Autonomy, Accountability, and the Scale of the Religious Abuse Problem

The courts widely share a sensible intuition that crimes and some subset of torts are not generally subject to a church autonomy defense.37  As the Second Circuit put it, “[t]he minister struck on the head by a falling gargoyle as he is about to enter the church may have an actionable claim”—that is, a claim not subject to church autonomy defenses.38  But at the same time, the courts agree that there is a set of torts—notably defamation—that definitely can be defeated by a church autonomy right.39  The courts have yet to supply a clear dividing line between those tort (and tort-like) matters that are covered by church autonomy and those that are not.40

In cases that held that church autonomy principles precluded a claim based on hostile workplace under the antidiscrimination rules of Title VII,41 the courts have sometimes said that there need be no concern that this holding would prevent future tort claims.42  But it is not at all clear why not.43  The Supreme Court too said that its ministerial-exception case, Hosanna-Tabor, was not deciding whether, or to what extent, church autonomy protected religious institutions from claims based on tort or contract.44  But if church autonomy did not protect churches from any tort claims, the court would be discarding a substantial body of cases protecting churches from defamation claims arising out of church discipline proceedings.  On the other hand, if churches are protected from all tort claims, then that leads to the troubling conclusion that churches are effectively immune from civil accountability for even the most heinous cases of abuse or harassment taking place under a church’s watch. 

The clergy abuse cases make this point particularly salient.  There is tragically a long list of stories of abuse happening in religious institutions, ranging from direct abuse by a religious leader on a member, but also including many stories of leaders mishandling reports of abuse or misconduct by other leaders as well as by congregants.45  Church autonomy principles are periodically litigated in clergy-abuse lawsuits.46  Sometimes, to be sure, not always.  Religious institutions don’t always choose to raise church autonomy as a defense.47  Churches may simply recognize on principle that whatever misconduct was at issue was not in fact pursued as part of the church’s governance.48  The church could view the conduct analogously to the way the Supreme Court analyzed government misconduct in Ex parte Young: conduct beyond the rightful authority of the religious institution is not protected with the immunity given to the institution.49  But even where religious institutions do not try to use a church autonomy defense, critics still worry that church autonomy could be used to shield misconduct. 

The scale of the problem of clergy abuse is taken by some critics to be fatal to church autonomy.  Tragically, religious institutions have never been free of situations of abuse, misconduct, and harassment.50  Religious institutions are often sites where people feel comfortable and safe, where congregants repose trust.  Yet they can be all too easily perverted into opportunities for abuse; the trust can provide an opportunity for harassers to take advantage and manipulate.  The news stories are, sadly, innumerable.  The Roman Catholic Church alone has been rocked by one scandal after another in the last twenty years.  In 2002, Boston Globe journalists brought national attention to a major coverup of priests sexually abusing minors in the Boston area.51  An extensive grand jury investigation in Pennsylvania in 2018 identified more than 1000 victims of clergy sexual abuse of children.52  Individual cases appeared from across the country.  A wave of new court filings in 2021 might represent another 5000 claimants.53

A steady stream of news reports and court cases examine abuse across religious traditions in America: reports of clergy members sexually abusing children,54 women,55 and men.56  A newspaper investigation into a network of Southern Baptist churches identified 700 victims of sexual abuse in churches over a twenty year period.57  A long list of religious institutions and leaders failed to properly investigate and report assaults that took place in ministry contexts.58  Assault, abuse, and harassment are found in Jewish and Muslim religious communities too.  Whether it was a rabbi abusing a minor59 or an imam abusing a woman working at a religious school,60 the bottom line is that abuse can be found across faith traditions. 

Of course, it is not just religious bodies that have serious abuse and harassment problems.  Other institutions have their complex of factors that can create heightened risk.  Schools are perhaps the most obvious example of another institution that tends to have significant levels of trust coupled with power differentials which can be all too easily abused.61  The #MeToo moment—and movement—has also played a role in highlighting problems.62  Sadly, there’s reason to think that what we’ve seen so far is just the beginning of a major reckoning with abuse of many sorts, in many places.

But religious institutions pose some unique challenges because of the legal protections they possess.  It is possible for a religious institution to defend itself against a claim of harassment or tortious conduct by claiming that the issue was one of church governance.63  A slap could be claimed to be an exercise of discipline, or an alleged hostile workplace could be a form of pastoral correction.  More commonly, churches defend against claims for supervisory responsibility for church leadership for abuse and harassment committed by others—and church autonomy comes up in this litigation with some regularity.64  Consider the following examples:

  • Employers can be liable for harms caused by an employee that the employer hired negligently.65  But when victims of abuse sue religious institutions alleging negligent hiring, the religious institution sometimes responds that it is protected in its hiring and firing decisions regarding ministers.66  Thus, the religious institution cannot be held liable for its employment decisions, at least when a minister was allegedly hired negligently.
  • Employers can be liable for negligently supervising an employee when a reasonable employer would have supervised.67  Sometimes, superiors in a religious institution are plausible supervisors.  And sometimes the supervisor had reason to suspect that the employee was a potential abuser.  This would make out a strong case for negligent supervision in many settings.  But the religious institution can respond that the nature of supervision is a religious matter.68  A court establishing the standard of care owed by the bishop to congregants to supervise a priest, for instance, is a matter of pastoral judgment; a court would be intruding on church autonomy to attempt to adjudicate the matter.69
  • Employers generally are vicariously liable for wrongs done by employees in the course and scope of their employment.70  But when religious institutions have been sued by victims, with the victims claiming that the abuser was an employee of the church, the church can respond by arguing that the nature of the employment relationship is a matter of religious determination itself.  A court would interfere with the pastoral decisions of religious leadership, excessively entangle itself in the internal affairs of the church, if it reviewed the reasonableness of employment matters.71

This could make it sound as though church autonomy stands as a roadblock to church accountability.  In practice, though, this is usually not the case.72  Courts often find ways around church autonomy arguments in the most problematic cases.73  Courts protect churches from run-of-the-mill defamation claims arising out of normal cases of pastoral rebuke or church discipline but tend to allow accountability for flagrant cases of abuse, fraud, and harassment, denying that church autonomy insulates churches from liability.74  That is not to say that they get it right all the time.  Nonetheless, as a practical matter, there are usually ways to make it come out right.

Yet even if the courts have the right instincts, there has been confusion about how these fit with their doctrinal tools.75  The broad principles they invoke appear mutually contradictory.  Or at least, the courts haven’t done a good job explaining how they can be reconciled. 

C.   Does the Conflict Undermine Church Autonomy?

Critics of church autonomy sometimes point to immunity from church abuse accountability as the most egregious consequence of a robust church autonomy doctrine.  Caroline Mala Corbin has argued that church autonomy places churches above the law.76  Marci Hamilton calls it a manifestation of “extreme” religious liberty and argues that it treats religion as if it can do no wrong.77  Both agree that it inappropriately removes incentives for care and caution that might reduce the incidence of abuse.  It also robs victims of abuse of the chance to obtain some relief—however inadequate tort remedies are in the face of abuse. 

Defenders of church autonomy have not often delved deeply into this most troubling domain for the application of church autonomy doctrine.  Courts and scholars alike have devoted quite a bit of effort and attention to figuring out the core domain for application of church autonomy and debating the many (and fascinating and important) issues about the nature and basis of church autonomy: to what extent does the doctrine rely more on one religion clause than another;78 whether church autonomy is best thought of as based on individual rights or as distinctly institutional (and as such, more than the sum of the individual rights of the persons who make up the religious body);79 whether church autonomy could be reduced to an associational right not based on religion at all;80 and so on.

The handful of works that tackle church autonomy and clergy abuse head-on from a perspective favorable to church autonomy have provided valuable and thoughtful analyses of how to apply church autonomy principles in specific situations.  Ira Lupu and Robert Tuttle have written two careful articles explaining why clergy malpractice is problematic and teasing out precisely which aspects of supervisory (respondeat superior) authority implicate religious standards and which do not.81  (Liability for supervisory authority is a common situation for applying church autonomy principles in abuse-related litigation.)82  Robert Cochran suggested a reworking of the liability framework around strict liability as a means of getting around the evaluation of religious doctrine.83  These projects deal with important issues and proceed with concern for both accountability for wrongdoers and also protection for the religious exercise of churches and other religious institutions. 

Yet church autonomy’s critics still might see them as simply tinkering around the edges of a deeper problem.  The default is still immunity for the internal workings of religious institutions.  And that, the critics argue, is the central problem.  It might be better to have strict liability for churches regarding some set of clergy misconduct,84 but that reform to the tort law side of things doesn’t address the deeper problem, that churches are treated as immune for internal affairs.  Indeed, one might think that a strict liability reform would violate a constitutional principle that broadly immunizes internal church affairs from any liability. 

This Article takes a broader look at church autonomy’s relationship to accountability.  Recall that church autonomy protects “the right of religious institutions ‘to decide for themselves, free from state interference, matters of church government as well as those of faith and doctrine.’”85  Faith and doctrine can provide alternate starting points.  Much of church autonomy falls under the doctrine heading, and this is where most of the analysis in the literature has focused in trying to establish the limits of church autonomy.  But if church governance provides an alternative starting point, it’s essential to figure out as well what that means and where its limits lie.  If church government is taken seriously as a basis for church autonomy, separate or apart from “matters . . . of faith and doctrine,” it has the potential to swallow up much of the category.  That is, it’s uncontroversial that religious institutions should be free to make their own doctrinal decisions, that courts should not take positions on matters of theology, that theology is not the domain of the state.  What is controversial is whether religious institutions are insulated in some category of self-governance matters apart from whether the decisions are based on religious doctrine (whatever that means).  If the church has a sufficiently broad and aggressive view of self-government, this could theoretically insulate a lot.  On the other hand, if the courts have no concept of church government as government, then church autonomy loses its most distinctive characteristic—its institutional character.  Either way, this issue is essential for the future of church autonomy as a legal doctrine. 

II.     The Search for Solutions

The legal doctrine assumes both that church autonomy is important86 and that religious institutions have to be held accountable at some level.87  But there has been considerable variation in the approaches courts take to reconciling these two commitments.  Courts have seized on one or another of various analytical threads to explain what is in and what is out from the protections of church autonomy.  This section looks at some of the problematic approaches that appear in the caselaw.88  It seeks to demonstrate, first, that the doctrinal analysis in the courts is in chaos, and second, that there are a number of approaches being employed by lower courts that are in tension with the Supreme Court’s precedents or inconsistent with the purposes of church autonomy doctrine.89

A.   Neutral Principles of Law as the Starting Point

One approach is to ask whether a given dispute can be resolved with “neutral principles” of law.  The problem is that this approach is sometimes applied in ways that conflict with the Supreme Court’s approach or are simply incompatible with the continued application of the church autonomy doctrine. 

1.   Neutral Principles in the Courts

The Fifth Circuit recently held that a case could proceed where the complaint simply “asks the court to apply neutral principles of tort law to a case that, on the face of the complaint, involves a civil rather than religious dispute.”90  The director of a church’s local mission board brought a defamation claim against the church’s national mission board based on a dispute about the local director’s professional relationship with and accountability to the national board’s leadership.  The defendants moved to dismiss on the basis of a church autonomy argument.  The district court dismissed the case.91  But on appeal, the Fifth Circuit reversed, reasoning that the court could proceed toward adjudicating the case simply by applying neutral principles of law.92

The neutral principles rationale appears repeatedly in the cases, articulated in various ways:

  • “Most courts . . . have adopted a narrower view of the doctrine and hold that the rights guaranteed by the First Amendment are not violated if the tort claims can be resolved through the application of ‘neutral principles’ of tort law, particularly where there is no allegation that the conduct in question was part of a sincerely held religious belief or practice.”93
  • “A court may exercise jurisdiction over a controversy if it can apply neutral principles of law that will not require inquiry into religious doctrine, interference with the free-exercise rights of believers, or meddling in church government.”94  Dissenting, Justice Boyd of the Texas Supreme Court recently argued that “the First Amendment does not bar a defamation claim, even if it arises from a religious context, when courts can resolve the claim by applying only non-religious, neutral principles.”95
  • “One way to distinguish between ecclesiastical from non-ecclesiastical claims is whether the dispute can be resolved on neutral principles of law that will not collide with church doctrine.”96 
  • “If neutral principles of law can be applied without determining questions of religious doctrine, polity, and practice, then a court may impose liability.”97

There’s something obviously appealing about this analysis.  Neutrality has been one of the recurring touchstones of Religion Clause analysis over the last fifty years.98  The Supreme Court applied neutral principles analysis in two cases involving church property disputes.99  Where it was possible to decide the ownership of church property by simply referring to the property-law documents—deeds and trusts—then it was permissible to adjudicate the case.100  What was not permissible was to decide which of two factions claiming church property was the faithful church—which faction had the better claim to be faithful to the Presbyterian tradition, for instance.  Lower courts have often seized on the “neutral principles” concept to guide their analysis of cases arising out of church disputes beyond the church property context.  

2.   Free Exercise Neutrality

The Supreme Court brought in another neutrality test for its Free Exercise doctrine.  In Employment Division v. Smith, the Court held that neutral and generally applicable laws do not violate the Free Exercise Clause.101

Occasionally this principle has been invoked in cases involving religious institutions.  In a case involving a claim against a Roman Catholic diocese for negligent hiring of clergy, this provided the Vermont Supreme Court with its starting point: “Laws that are neutral and of general applicability do not violate the Free Exercise Clause.”102  Accordingly, neutral and generally applicable tort laws can be applied to religious institutions and entities.103  “We do not believe defendant’s generalized assertion that requiring it to hire and supervise priests in a nonnegligent manner would constitute undue interference in church governance.”104

Yet the generic Free Exercise doctrine really doesn’t fit church autonomy cases.105  The Supreme Court said as much in Hosanna-Tabor, though it didn’t really explain why.  In Hosanna-Tabor, the government invited the Supreme Court to eliminate the ministerial exception to federal antidiscrimination law, an argument premised on Smith’s principle of neutrality.  If neutral principles did not raise a Free Exercise problem, then why couldn’t courts apply religiously neutral employment law when a church fired a minister?  The Supreme Court quickly rejected this argument.106  It didn’t explain very clearly why Smith was different from the ministerial exception cases.107  But that it was different, and its rule was not to be applied to matters involving the internal governance of the church, was emphatic.108

It’s clear enough that church autonomy as a doctrine would be essentially eliminated were the Smith rule applicable to issues arising from internal church matters.  Neutral rules prohibiting employment discrimination would eliminate the ministerial exemption.  Neutral rules of defamation would require courts to adjudicate defamation claims arising out of church discipline proceedings.  And so on.  But why didn’t Smith simply require this result?109

One difference that some scholars and courts have pointed out is that the ministerial exception—and church autonomy more broadly—relies on both the Free Exercise and Establishment Clauses.110  The combination—and as some have argued, the Establishment Clause in particular111—gives church autonomy protection a structural character.112  That is, church autonomy is not just about individual rights.113  It is about the government’s inability to infringe on the domain of religion, specifically in this case on the domain of a religious institution.  Older cases called this principle jurisdictional.114  Some courts still do115(though the Supreme Court has clarified that it’s not jurisdictional in the strict sense for purposes of federal civil procedure).116

The point for present purposes is that Smith by itself doesn’t tell us how to think about religious liberty in the role of a structural constraint on the government, which is what church autonomy requires.  Its framework of neutral principles isn’t designed to handle that kind of analysis.  This is the best way to understand Hosanna-Tabor’s apparently casual dismissal of Smith.117  It said that Smith was not about the internal workings of a religious institution, which of course it wasn’t, but that by itself didn’t explain why Smith was inapplicable.  Taken alone, Hosanna-Tabor is unpersuasive on this point.  But add in the idea that church autonomy is a structural principle in a way that’s different from individual Free Exercise claims, that perhaps this structural character arises from the Establishment Clause, and there is indeed a good reason to see Smith as inapposite. 

B.   Church Property Cases and Neutral Principles

For church autonomy, the proper source of neutral principles analysis isn’t Smith.  A different line of cases about church property disputes provides the more important and relevant precedents.118  The big question here is whether the neutral principles applied in these cases can be applied beyond the property context. 

Church property cases have long been intertwined with church autonomy issues.  Often it was a dispute about church property that brought matters of internal church governance before courts.  A church splits into two factions.  In various cases, the factions are divided by a disagreement about doctrine, about the choice of a pastor, or in one famous case from the American Civil War period, about the ethics of slavery.119  Each faction in each of these cases claims the right to the church building and they file suit.120  The court then has to figure out which side of the dispute has the rightful claim to the property.  Many of the key church autonomy precedents arose from cases where the courts realized that there was something wrong about courts—instruments of the state—deciding which side of the church split was in the right.121

The doctrinal character of these adjudications was particularly clear in the early days, when the American courts widely applied the Pearson test, borrowed from eighteenth-century English law.122  What the Supreme Court found problematic in the traditional Pearson approach was that courts assumed that religious institutions held property under an implied trust to serve the original religious body.123  If the church body changed its character in some way and split, the court had to decide which faction had stayed true to the founding principles of the church and award the property to that group.  This put the courts in the position of scrutinizing church doctrine.124  In the Pearson case, a church ceased to be Trinitarian and became Unitarian.125  Could the Unitarian congregation retain control of the property held in trust?  The court said that a switch to Unitarianism could be a fundamental change to the church, in which case the Unitarian congregation wouldn’t have rights to the building, and appointed a special master to deal with that issue.126  Similar issues arose repeatedly in America.  A number of American cases went deep into historical theology to try to parse out the original commitments of a given church and whether the church deviated from them.127  It raised concerns about the courts deciding what was orthodox in religion.

One way out was to defer to any internal church decisionmaker about matters of doctrine or internal self-governance.128  This is what the Supreme Court did in Watson v. Jones,129 an 1872 case that applied general common law, but invoked religious liberty values that were easy to convert into constitutional principles years later, after the court recognized the incorporation of the First Amendment.130  In that case, a local church body split and a higher-up synod of the Presbyterian church adjudicated the dispute and declared that one side of the faction was in the right and the other in the wrong.131  The Supreme Court opted to simply defer to the hierarchical church authority.132

That still left the problem of what to do with nonhierarchical entities.  Many states retained the Pearson test.  But the Supreme Court was sensibly concerned that there were First Amendment problems with the Pearson test.133  Neutral principles of law provided a way out.  It placed the burden on the church to get its property title and trust documents in order, while getting the courts out of the business of regularly scrutinizing church doctrine.134  

There’s a reasonable argument that these cases are best confined to the church property setting.135  In that setting, the question of land ownership is distilled into a set of legal documents that are highly formalized.136  Deeds set out ownership.  Express trusts articulate the terms of the trust.  Courts can look at these documents and decide who owns a given piece of property.  The Supreme Court has never applied the neutral-principles analysis outside of the property-law context. 

1.   The Problems with Raising Neutral Principles as the Starting Point for Analysis

With the lack of guidance from the Supreme Court, lower courts borrowed the neutral-principles analysis from the property law cases to serve as the touchstone of church autonomy’s limits.137  Extending the neutral principles rationale beyond the church property context raises new questions.  While it is possible to use neutral principles in a way that is reconcilable with other aspects of what the Supreme Court has said about church autonomy, it is not always used that way.138  For now, I want to flag one way in which the neutral-principles approach, as some lower courts apply it, is inconsistent with church autonomy. 

If the court simply asks whether “neutral principles” can resolve the case,139 the answer will (almost) always be yes.140  Unless the law targets or discriminates against religion,141 the law itself will be neutral.  This formulation of the idea has more in common with Smith than with the church property cases.  And as we’ve already seen, it is working on a fundamentally different problem than that addressed by the church autonomy principle, which is the idea that the internal affairs of the church are off-limits to civil government.142  All of the classic instances of church autonomy protection can appear as exemptions from neutral laws.  Defamation law gives way when a church is exercising discipline over a member.  Antidiscrimination law gives way to a religious body’s right to choose its own clergy.  Neutral principles in this form do not provide an outer limit for church autonomy.  They eliminate church autonomy.143

This is the mistake that the Fifth Circuit made in its recent defamation case, McRaney.  It made neutral principles into the first step of the analysis.  Indeed, it treated neutral principles as a pleading test: if the claim could be plead without raising religious issues, it was presumptively not a church autonomy matter.144  But that covers virtually everything.  The same happened in a Florida appellate case.  Without any analysis of whether church allegations of homosexuality were related to discipline, the court reversed a dismissal because the allegations were simply framed as a neutral tort claims.145  On the court’s analysis, though, this would swallow any church autonomy defense to claims arising out of church discipline.146

The church property cases don’t treat neutral principles primarily as a pleading standard.  Rather, they ask substantively whether neutral principles of law can be applied without deciding matters of church doctrine or belief.  What is somewhat misleading about this is that the neutral-principles language is no longer really doing the work.  Instead, the courts are simply articulating the principle that civil courts should not decide matters of religious doctrine.  If the case can be decided only by the application of neutral principles, without deciding at the same time a matter of church doctrine, then the court can proceed. 

Confusion about how to do neutral principles analysis can also bleed into an administrability concern.  When neutral principles is the starting point for analysis, without clearly recognizing that what is at issue is whether the court will impinge on religious doctrine, it is all too easy for courts to overlook or misclassify conduct in the religious organization in order to be able to adjudicate a case.  There are instances of courts holding that internal church matters are neutral because they are not sufficiently religious.147  But de minimis religion is slippery territory indeed.148

When formulated this way, it becomes evident that neutral principles don’t help in figuring out when a matter is protected as a matter of church governance.149  They may work in the church property context, as a shortcut to just applying the deeds and trusts as enacted.  More broadly, it is best to read the “neutral principles” language as just an imprecise way of articulating the familiar principle that courts should not adjudicate matters of church doctrine and belief. 

This still does not help us figure out the meaning of church governance.  Again, the courts list doctrine and belief as one basis for church autonomy and church governance as another.  That is why, in both Hosanna-Tabor and Our Lady of Guadalupe, the Court insisted that the ministerial exception applied without the church needing to show that its employment decision implicated religious belief at all.150

In the end, then, neutral principles fail to solve the puzzle of church governance.  

C.   Religious Principles Always Required

Some courts seem to assert that religious autonomy is applied only when religious principles would be implicated by a court deciding the matter.  For instance: 

  • “[A] plaintiff alleging particular wrongs by the church that are wholly non-religious in character is surely not forbidden his day in court.”151
  • “[T]he critical issue underpinning the church autonomy doctrine is whether the dispute is secular or religious.”152
  • “The First Amendment does not prevent courts from deciding secular civil disputes involving religious institutions when and for the reason that they require reference to religious matters. . . . [N]either the district court nor we have made any decision for or against any religious doctrine or practice.”153

In the law reviews, one can find the argument that a religious reason for action should be the limiting principle for church autonomy.154

But at least in the ministerial-exception cases, courts say that the harm of adjudicating internal affairs of a religious organization does not depend on whether the particular issue was based on religious convictions.155  In the context of the ministerial exception to employment law, the Supreme Court holds that religious reasons for the employment decision are unnecessary.156  Churches are simply exempt from the employment discrimination laws (and other interferences with the church-minister relationship).  The cases are none too clear on how to sort out this tension. 

Indeed, common descriptions of church autonomy take a disjunctive character: church autonomy covers matters of “discipline, faith, internal organization, or ecclesiastical rule, custom or law.”157  “[T]he courts may not adjudicate disputes concerning religious law, principle, doctrine, custom or administration . . . .”158  Courts should not decide matters of church belief or of church governance.159

A broad statement that church autonomy doctrine applies only when the issue is religious just assumes the central issue: is church governance inherently religious?  Maybe the answer is yes, but the courts have been less than clear about it.160  And if it is so, this would solve one question but raise another: Is a religious institution automatically exempt whenever it claims that something happening within its community is a matter of internal governance?  Surely that cannot be right, especially if we entertain the possibility of unusual or extremist religions that embrace violence.

We might also be concerned that a bright-line requirement of a religious reason for action will miss another doctrinal point often made in these cases—entanglement.  Imagine that a church fires a minister because of an intangible bad fit between the minister and the congregation, but the fired minister insists that this poorly articulated fit concern is a mask for discrimination on the basis of a protected characteristic (sex, race).  The church doesn’t have chapter and verse for why fit matters, but the congregation as a whole is frustrated with what it perceives as bad shepherding by the minister.  Should the congregation not get the benefit of church autonomy?  Should it depend on how good the congregation is at connecting its intuition about bad personal fit to a theological account of what makes a good pastor?  Choosing a minister is a plausibly religious decision, to be sure, but a given church may not always be effective at explaining it.161  Requiring a religious principle as the deciding factor for whether a religious institution’s decision is protected may not be so good at capturing these kinds of issues. 

American appellate cases point in different directions as to exactly what kind of religious basis challenged church action must have in order to get church autonomy protection.  Some say the conduct in question must be “rooted in religious belief.”162  Some have said that they can at least review conduct that is “wholly non-religious.”163  Some courts have arguably extended the principle that no religious justification is needed for religious institutions to situations involving other allegations of employment violations.164

To recapitulate: a hard-and-fast rule that church autonomy only applies when the conduct at issue is specifically justified on religious grounds either solves nothing or else works a major change in the law.  If it means simply that the conduct must be justified either on the basis of religious doctrine or belief or on the basis of being part of church government, then that just dodges the question.  If it means that church autonomy is only available when the specific issue challenged is explicitly justified on the basis of religious doctrine or belief, then that requires significant change in the law—it would imply that both of the Supreme Court’s ministerial exception cases were wrongly reasoned and many lower court decisions are going astray.  There may be strong arguments that such a change would be desirable, but that is not the law now, and lower courts solve nothing by occasionally suggesting in a perfunctory manner that this is the law. 

D.   Subject-Matter-Specific Carveouts

Another way of cabining church autonomy is to draw some bright-line rules about subjects that are in or are out.  The Supreme Court at least gave lip service to this option in Hosanna-Tabor when it said that it took no position on whether the church autonomy principle reached similar tort or contract issues.  But exactly how the issues would be distinguished was not evident.  Lower courts have largely been left to figure it out on their own. 

The Ninth Circuit has held that the ministerial exception applies only to “tangible employment action” (like hiring and firing decisions).165  Employment matters that might be challenged on a hostile workplace theory can be brought because it need not require the church to justify the choice of a minister.166  But the Seventh and Tenth Circuits have rejected this as a bright-line rule,167 and it is a more plausible reading of the cases that this is really just an application of a different principle—that the court won’t require a church to explain its employment decisions, and that the courts disagreed about whether that precluded a hostile work environment claim.168

A more serious effort to draw bright-line categories around particular subject areas of church autonomy has revolved around the distinction between intentional and negligent torts.  The Missouri Supreme Court distinguished intentional from negligent torts, holding that churches could be held liable for intentional torts but that church autonomy categorically precluded any negligent torts on supervisory theories.169

There are various ways in which this rule is over- and underinclusive, as is usually the case with bright-line rules.  Whether this rule is plausible rests in part on some empirical claims by the Missouri court.  The Missouri court reasoned that church autonomy’s rule against adjudicating religious matters would come into play regularly if the court were to adjudicate the authority of the church to manage its employees.  Managing employees, on this theory, would often involve religious issues about how to pastor and discipline individuals, how to shepherd a congregation, how to inculcate religious precepts in the community.  But intentional torts are different.  First, they could often be adjudicated without the context-dependent analysis required in negligence.170  Second, intentional torts on the part of the church less often have religious motives, the court thought.171

It’s easy to imagine objections to this latter point—what if a given sect thought that corporal punishment was an appropriate form of church correction?  The best defense of the Missouri court’s position is that, while it’s logically possible that intentional torts could be religiously motivated, it’s unlikely in practice—based on the observation that, in the existing caselaw, religious motives have rarely been invoked to defend intentional torts.  

This is all useful for that part of church autonomy that is concerned primarily with avoiding adjudicating matters of religious doctrine or belief.  But it does not answer the question of what, if anything, comes under a general “church governance” heading. 

E.   Membership and Consent

Another possible way of limiting the scope of the church governance form of church autonomy is to focus on church membership.  For at least some claims, courts have said that churches could be liable (that is, not protected by church autonomy) for conduct regarding nonmembers, but not liable if the same conduct was directed at members. 

  • “[The] shield from liability evaporates for claims that arise after a member has separated from the church and is no longer a church member.”172
  • “[A]bsolute First Amendment protection for statements made by a Church member in an internal church disciplinary proceeding would not apply to statements made or repeated outside that context.”173

An unpublished pair of Sixth Circuit cases illustrates the way membership can be invoked in denying church autonomy protection.174  Both cases grew out of the same set of incidents.  Two pastors from the same denomination went on a trip together.175  Afterwards, Pastor A alleged that Pastor B sexually harassed him on the trip.176  Pastor A reported Pastor B to denominational authorities and also called out Pastor B as an untrustworthy person in front of Pastor A’s congregation.177  Pastor B then sued for defamation.  The report to denominational authorities was held to be protected within church autonomy protection.178  But Pastor A’s comments to his own congregation were not protected because the court wasn’t convinced that this was called for by their religious obligations.179  Among other things, the fact that Pastor B wasn’t a part of Pastor A’s church was important to the court because Pastor A had no formal relationship to correct Pastor B in front of Pastor A’s church.180  One can imagine similar questions arising in the not-uncommon situation where a church starts a discipline process and the person subject to discipline then leaves the church.

The rationale is that membership shows an individual’s submission to church government.  Without membership, the plaintiff did not consent to church authority.  Aspects of this, at least, seem quite right.  Membership should matter; it should be possible to know when you’re in and when you’re out.  Having an exit mechanism can relieve some of the concern about removing liability protections from church actions.181  “An important aspect of church autonomy is how every insider has the right to leave, the right to become an outsider.”182  And membership provides a touchstone for figuring out who needs to be involved in church governance in congregational churches. 

Some cases have involved individuals leaving a religious organization, and in those cases, courts have been careful to respect their decision to disaffiliate.183  But none of the recent church autonomy decisions have given a clear answer as to when and in what circumstances membership is a necessary element for the application of church autonomy.

Imagine a church that doesn’t have a formal membership process.  It will not be easy to tell who is in and who is out for purposes of church discipline.  And yet it cannot be the case that a lack of formal membership divests the church of autonomy protections.  Imagine a scenario where an individual comes to a church, does not become a member, but quickly begins spreading what the church leadership would consider false teaching throughout the church; that individual also happens to be openly engaged in conduct that the church considers sinful.  Church leadership calls out this misconduct along with the subversive false teaching in a members’ meeting.  (The New Testament provides a basis for public rebuke for public sin.)184  This is easily a matter of church governance and upholding church standards on almost any way of approaching the issue except for membership.  If this discipline is not protected by church autonomy, it would seem as though very little of the church governance rationale is in fact doing any work; instead, it is all about consent.  If we do not want to strip church autonomy down simply to consent, then the church meeting rebuke should be protected on a church autonomy theory from a potential defamation suit.

Some scholars suggest distilling church autonomy to consent, or at least constructive consent.185  Even organizations without formal membership still can be analyzed in a constructive-consent framework.  But no court has been willing make consent or association the sole basis of the doctrine.  Consent is part of the rationale in the cases, but not the whole of it.

Beyond this, the most extreme form of the consent-focused rethinking of church autonomy fails to fully account for the importance of church autonomy’s religious character.  Church autonomy is appropriately based on the religion clauses (even if it also draws on the assembly and association principles from elsewhere in the First Amendment).186  Without the church as a religious organization, the justifications for church autonomy are considerably narrowed.  And the courts are right to say that religious institutions have a distinctive sphere of authority and expertise which it is inappropriate for the civil authority to infringe.187  There are historical reasons for this: religious institutions have had a fraught relationship with the civil authority.188  The church as an alternate center of authority and allegiance was a point of jealousy for princes and potentates.189  The American approach to separating the church and state, with neither institution dictating the other’s conduct, is a (arguably the) central tenet of America’s constitutional law of religious liberty.190

If this argument is well taken, then it is a mistake to try to resolve all of church autonomy’s conundrums through consent. Yet exactly how much can be resolved through consent—looking at who is a member of the church, for instance—is still poorly explained in the cases. 

F.   Balancing Tests

The courts have rejected balancing tests in the church autonomy cases.191  This would be the easiest way to deal with the problem of clergy abuse, to be sure.  If the courts could simply say that there is a compelling government interest in protecting church autonomy, then it would be free to consider whether the government’s interest in thwarting abuse could justify liability in the extreme cases.192  Douglas Laycock, one of the leading scholarly proponents of church autonomy, has urged that the courts employ balancing tests specifically to meet the challenge of clergy abuse cases.193  But as the doctrine stands now, there is no balancing used in the church autonomy cases.194  Instead, the courts have opted for bright-line rules.195

III.     The Principle of Accountability

This Article takes as its starting point two assumptions that courts rightly profess: First, religious institutions should have a zone of protection for their self-government; call this principle “autonomy.”  Second, religious institutions must not have carte blanche to cause harm (at least physical harms); call this principle “accountability.”  This Article takes for granted that autonomy is a desirable goal, having sketched the arguments made elsewhere in the literature at considerable length for church autonomy.196  Recall that church autonomy has been justified by positivist constitutional arguments, liberal theory, pluralist theory, and explicitly religious theories.197

This Article now turns to make the case for accountability on the same bases as those which justify autonomy.  The goal is not just to show that accountability is a worthy objective.  It is also to argue that autonomy and accountability ought to go together even on the very strongest forms of church autonomy.  And it is to show that both have been deeply intertwined as a matter of theory and of history.  Certainly, not everything done within a church is protected.  This is not just a premise of liberal theory.  It’s also a reasonable deduction from history and pluralist theory—the theoretical bases for the most strongly pro-church-autonomy rationales in the literature. 

Section A presents an argument from sphere sovereignty and, more generally, social pluralism.  The social pluralist theories have been used to defend the most expansive versions of church autonomy.  We start with this, first, because it is often assumed that the “sphere sovereignty” model of church autonomy is the conceptualization least amenable to an accountability principle binding the church.198  I want to show that, even accepting a set of strong theoretical commitments to a “sphere sovereignty” theory of religious institutions, there is nonetheless an important place for accountability.

Section B turns to the history.  Following the lead of the Supreme Court, it looks deep into the history of common-law church-state relations to illuminate the development of ideas—here, an accountability principle—that can inform constitutional interpretation. 

A.   Pluralism and Accountability

A number of scholars have defended church autonomy on the basis of pluralist social theories, particularly as developed by the Dutch theologian and social theorist Abraham Kuyper.199  Several scholars have argued that social pluralism (1) descriptively fits what the courts seem to think they are doing in the church autonomy cases, and (2) normatively supports a church autonomy principle.  I turn to Kuyper now as representing an expansive theory of church and state—sphere sovereignty.200  And while Kuyper’s theory is distinctively theological, its pluralist social vision has appealed to scholars who don’t necessarily share his Christian faith.

Kuyper’s theory of sphere sovereignty starts from a theological premise: that God alone is sovereign.201  As a result, any effort by human beings to claim for themselves, or for any single human institution, absolute sovereignty is an affront to God.  It is, in effect, to try to play God.202  Yet human beings are not infinitely good, wise, knowing, loving, and just.  They are, to the contrary, sinful.  Consolidation of absolute sovereign power or authority in a single place or institution would be to invite abuse of that power.203

Human authority is derivative from divine authority.204  Among human beings, God divided authority between various spheres of human life and activity.205  Kuyper pointed to church, state, and family, among others, as “unique spheres, each with its own sovereignty.”206  They should each be free, within their rightful spheres, to live differently from each other.207  No one sphere should dictate or micromanage the affairs of the others.  The church should not run the state nor the state the church.208  This not only honored God as the ultimate sovereign, but also respected a division of authority among human institutions which safeguarded freedom, law, and order.  It is like a society-wide version of checks and balances—an idea familiar to anyone with a passing acquaintance with Madisonian political theory.209

Often, the emphasis in the church autonomy literature (and caselaw) is on the independence of the church from the state.210  Taken alone, this can give the misleading impression that it’s all about autonomy with little or no place for accountability.  But sphere sovereignty does not ineluctably result in hermetically sealed spheres.  The sovereignty of human spheres is relative.  The idea of separating the spheres is not just to shield the spheres from each other, but also to check each other.  For this to happen, the spheres must be interactive.

Kuyper himself wrote of the possibility that the church could exceed its authority.  He had in mind not the clergy-abuse scandals of our day, but historical examples of churches seeking to compel entities outside the church to live according to the church’s internal discipline (which is how he viewed the Roman Catholic Church at the time of Reformation).211  If such an abuse would occur, it would be the role of the state to protect the individual from the tyranny of the church: “[W]herever, in violation of this principle, transgression of power may occur, the government has to respect the claims on protection of every citizen.”212

Consider concretely what it might look like for one sphere to check abuses by another.  First, imagine a religious institution—a church—seeking to hold the state accountable for abuses.  Many preachers of the American Revolutionary period thought they were doing precisely that when they critiqued English policy toward the American colonies from their pulpits.213  In their sermons, they interpreted and applied biblical principles about Christian obligations to the state and made the argument that the Christian is bound to honor the lawful actions of the state but can, and sometimes must, resist actions of the state that exceed its lawful authority.214  Whether England had in fact exceeded its lawful authority turned on a secondary set of arguments about English law and government (including the structure of the English constitution, the meaning of the Glorious Revolution, and in some cases, the authority of charters issued to certain of the colonies).  Similarly, consider the suggestion that the Roman Catholic Church deny communion to politicians who publicly identify as Roman Catholic but do not affirm the church’s pro-life stance.215

Note that in these examples, the churches did not themselves take up arms, which would have been an inappropriate blurring of the lines between the authority of the state and the church.  In biblical terms, the state bears the sword, while the church the authority to exhort and to control access to the ordinances or sacraments of the faith.  In these examples, the churches operated within their sphere to assert their perspectives and to hold state actors accountable by means of exhortation or through the proposed method of denying a political figure communion.  These are appropriate ways of exercising authority of the church on the state.

The church, for its part, is susceptible to sin, as is the state.  If this sin implicates the authority of the state, then it is by no means appropriate for the state to stand back as if it had no authority to engage.  As Kuyper explained, the state has an important responsibility endowed by God Himself: “The magistrate is an instrument of ‘common grace,’ to thwart all license and outrage and to shield the good against the evil.”216  But sin is an attack on “God’s handiwork.”217  As a result, “God, ordaining the powers that be, in order that, through their instrumentality, He might maintain His justice against the strivings of sin, has given to the magistrate the terrible right of life and death.”218

Again, in the western Christian tradition, the authority of the church is limited to spiritual authority.  It is manifested in instruction, exhortation, sometimes rebuke; and in the administration of ordinances or sacraments of the faith including communion and baptism.219  Churches can wield these instruments to engage with the state.220

Meanwhile, the state can and should use its temporal sanctions against wrongdoers in its domain.  That includes those within the houses of worship in its jurisdiction.  There is no good reason to say that the sphere of the state is artificially bounded by the fact that a matter against its laws happens within a church. 

That said, there must be some ability for each institution to manage its internal affairs.  Precisely because there is overlap, each sphere must use self-restraint to avoid impinging on the other excessively.  If the state was to decide that every appointment of a pastor was the state’s business, the church would soon lose a substantial degree of independence.  If the church were to decide that every vote taken by a legislator who was also a church member was grounds to threaten excommunication, there would again be little in the way of sphere sovereignty.  The exact bounds here are tricky, and the high level of generality provided by sphere sovereignty does not solve the question of where exactly to draw those bounds.  But it does establish that the same pluralist principle which counsels a degree of autonomy for church and state must at the same time recognize a place for overlap in the fulfillment of the distinctive ends of each of the spheres.  Sphere sovereignty thus counsels not only autonomy, but also accountability.  In Part IV, we will turn to a more concrete set of issues as to what is in and what is out of the protection of church autonomy law.  In the meantime, the point is simply that a very strong theory of church autonomy based on sphere sovereignty (or something like it) need not—and does not—logically lead to the church existing above the law and outside accountability for civil wrongs.

B.   History and Legal Meaning

There are a few different ways to make the case for accountability alongside autonomy in a more positivist mode.  One is to look to history.  This is a commonplace in constitutional law, of course.  What is a bit unusual in the church autonomy cases is the tendency to look back, not just to the Founding period, but far earlier.  We can call this the “deep history” approach to church autonomy. 

1.   A Note on Historical Methodology

The Supreme Court and some lower courts have looked deep into the medieval history of church-state relations to interpret the evolution of church-state relations in the common law and then, subsequently, in the American constitutional order.221 These courts, and a line of scholarship too, suggest that the roots of church autonomy principles can be traced far back in time through many European conflicts over church and state.222  This history provides lessons about the importance of including mutual accountability along with institutional separation and independence. 

The assumption behind this move is that the Constitution’s meaning was shaped by the background history and context in which it was drafted.223  That included the history of religious establishment in England and in many of the colonies before the Revolution.224  To the extent the new American nation’s ideas of religious liberty worked to disestablish the church and increase the institutional differentiation between church and state, then elements of separation and distinction in earlier English law would have been relevant.  One could imagine their relevance in a few different ways: Maybe elements of institution separation in earlier English history seeded ideas that would later foster broader American ideas of disestablishment and religious liberty.  Maybe the American positions were attempts to go beyond.  Either way, the assumption is that pre-independence ideas about disestablishment or institutional separation in the relationship of church and state set baselines that are presumptively going to have some continuing relevance to the American regime of religious liberty after independence.  Or maybe the deep history is relevant but in an even more attenuated form of common-law or Burkean development, such that the evolution of the common law helps us to see lessons learned in the past in a way that can help inform contemporary legal decisionmaking. 

I confess some ambivalence about the deep history arguments.  On the one hand, such history plausibly does inform the development of American constitutional traditions.  On the other hand, the historical stories told in the cases are often simplistic. More work remains to be done to provide a sophisticated account of how the “deep history” connects with American law and could inform constitutional interpretation.225

For purposes of this Article, I will take the “deep history” as a given part of the argument.  The courts are using it and it is worth offering an argument in terms they will recognize.226  I also find it plausible that the history of English law on this point is in fact relevant, even if more work remains to be done to make the connection in scholarship.  (A much more granular story remains to be told about religious institutions and religious liberty in early America.)227  What follows is the sketch of an argument from English common-law history for accountability—an argument in dialogue with the “deep history” account of church autonomy that has found traction in the courts.228

2.   The Evolution of Accountability in English Law

In the medieval period, one can find a wide array of conflicts between church and state.  The two powers theory articulated by the early medieval Pope Gelasius provides a point of reference for those who see a deep-seated commitment to distinct and coequal institutional authority for church and state.  “There are two powers,” he wrote to Roman Emperor Anastasius I, “by which this world is chiefly ruled, namely, the sacred authority of the priests and the royal power.”229  Gelasius’s clear distinction between religious and political authority is often heralded as an important development in political thought.230  Episodes from the medieval period (Becket at the cathedral, Holy Roman Emperor Henry IV at Canossa) have been mined (with varying levels of historical detail) to draw lessons about the independence of the church.231  Holy Roman Emperor Henry IV tried to appoint a bishop against Pope Gregory’s wishes; in the power struggle that ensued, the emperor backed down and acknowledged the church’s power.232  Analogously, English King Henry II fought with his onetime friend Archbishop Thomas Becket over the extent of royal power over ecclesiastical appointments as well as over criminal justice in churches; Becket was murdered by friends of Henry, but Henry in the end moderated his demands.233  Magna Carta’s protection for “freedom of the church” has been cited by scholars and judges alike.234  The direction of change was toward greater delineation of the bounds between the state’s jurisdiction over such subjects as criminal law and property, and the church’s jurisdiction. 

3.   Benefit of Clergy as an Immunity Based on Status as Clergy

One of the most telling transitions for accountability purposes was the evolution of the law of benefit of clergy.  The English criminal law rule initially established which court, ecclesiastical or common law, would handle a felony.235  Clergy who committed felonies were committed to the ecclesiastical court, while everyone else would be tried in a common-law court.236 The default in any case was that it was the king’s courts that would decide the case.  But if the person charged was a cleric (a term which covered a variety of different religious positions), they could assert their status after indictment and would then be transferred to the ecclesiastical authority.237  In the mid-thirteenth century, that would be the end of the matter, as far as the common-law courts were concerned.238  In the common-law courts, felonies worked a forfeiture of “life or member”239 and were punishable by death, but the ecclesiastical courts could not “pronounce a judgment of blood.”240  To receive recognition as clergy was a “benefit” because it at least meant that the accused would be spared the death penalty. 

Moreover, the ecclesiastical courts had a reputation for leniency to defendants.241  When an individual successfully asserted benefit of clergy in front of a common-law judge, the accused then transferred from the king’s court to the ordinary, an ecclesiastical court.  The ordinary’s procedure suffered some disrepute among lay English and one can detect a note of rivalry between the king’s and the ecclesiastical courts.242  Both adjudicator and procedure alike tipped heavily in favor of the accused.  A trial was held before the bishop or his representative and a jury of (usually twelve) clerks.243  The accused could make an oath of his innocence supported by “compurgation,” a procedure in which a specified number of persons would swear a ritual oath to the accused’s trustworthiness.244  Then the jury would deliver a verdict, which was (not surprisingly, given the heavy thumb-on-the-scales in the proceedings) often acquittal.245

If defendants could get the benefit of clergy, they were unlikely to be convicted.246  And even if they were, they would receive a sentence significantly less severe for the crime.247  Some historians have suggested that benefit of clergy proved an incentive for English church recruitment—joining the lowest church order and receiving the tonsure provided this limited but valuable legal immunity.248

What made the difference was not the subject matter—sacred versus secular—but the person of the defendant.249  The traditional rationale, explained by Sir William Blackstone in his classic exposition of English law, was respect for the church as a distinct jurisdiction: “[T]he benefit of clergy[] had [its] original from the pious regard paid by [C]hristian princes to the church in [its] infant state.”250  But its growth into a general claim of the clergy’s exemption from felony law jurisdiction of the king’s courts was, Blackstone complained, an instance of the church taking advantage of the pious regard with which the state treated ecclesiastics.251

During the medieval period, the details of how benefit of clergy was administered evolved.252  For present purposes, only a few points about the complex evolution of the law need be made.  Some of the reforms were designed to cut back on invocation and abuse of the privilege.  For instance, a statute passed in 1487 established that no one could claim benefit of clergy more than once, and provided that the accused was to be burned on the hand to leave a mark after claiming clergy (making it easy to identify recidivists later).253  Initially, benefit of clergy was accorded to anyone with a tonsure and clerical attire.254  But that was not a very reliable test (too easy to imitate).255  Then the criteria for identifying a clergyman shifted to the ability to read.256  The accused would be asked to read the first verse of Psalm 51, long thereafter remembered as the “neck verse.”257  This was also obviously overinclusive, because it ended up broadly protecting educated people—or indeed anyone who had the forethought to memorize the “neck verse” before committing a felony.258  The actual effect of the benefit of clergy was shifting, from protecting clergy to simply moderating the severity of the criminal law.  That was the rationale that would ultimately sustain the benefit of clergy centuries later.259  But that severing of the benefit from religion was in the future.  Still through the medieval period the justification was that the person of the clergy, not the offense, defined the scope of authority.260

4.   Benefit of Clergy and the Problem of Accountability

While historians debate exactly how high the incidence of offenses by clerics actually was, there were certainly periodic high-profile cases that worried the public about a lack of ecclesiastical accountability.261  English laymen periodically complained that the church’s ecclesiastical courts didn’t adequately address actual wrongdoing by clergy.262  Lay people were frustrated at the wrongdoing of clerics and at a perceived lack of accountability for clerics.263  Henry II’s original conflict with Becket was fundamentally about state jurisdiction over the perceived increase in criminal conduct that was shielded by benefit of clergy: the king “was told that more than a hundred murders had been committed by clerks since the beginning of his reign.”264  A case of national notoriety involved an archdeacon who allegedly murdered his archbishop with a poisoned communion cup; the accused invoked his privilege as clergy for an ecclesiastical trial, to the outrage of some observers and the embarrassment of the king.265  Henry’s Constitutions of Clarendon were an attempt to assert royal authority and significantly cut back ecclesiastical immunities.266  After Becket’s murder, Henry repealed the most controversial clauses of the Clarendon Constitutions.267  But assorted crimes would keep the issue in the public eye for generations to come.268

It even became a trope in fiction.  In a popular set of stories from fifteenth-century England, the frustration found an outlet in a fantasy of extrajudicial revenge on rogue clerics: in one popular story, a cleric seduces the wife of an honest workman, but the story concludes with the workman killing the cleric in a fit of rage.269  The ecclesiastical courts might be lenient and unreliable, but the educated upper class English who were the audience for this book sought for it a form of accountability outside legal channels.

Even during the heyday of the benefit of clergy, the clerical class was not categorically immune from the king’s criminal jurisdiction.270  Blackstone reflects typical early-modern-English anti-Catholic attitudes in his description of the privilege’s history: “[A]though the usurpations of the pope were very many and grievous, till Henry the [E]ighth entirely exterminated his supremacy, yet a total exemption of the clergy from secular jurisdiction could never be thoroughly effected, though often endeavoured by the clergy.”271

5.   Reformation and Accountability

Major change came with the Reformation.272  Reformation theology rejected the idea that the essential jurisdictional fact was a person’s status, clergy versus laity.  Luther famously argued that all Christians have a calling to serve God in whatever (nonsinful) vocation they assume.273  Luther and the other magisterial reformers denied that clergy held a holier status than persons who worked lay jobs.  Lutherans and Calvinists affirmed that there was a calling to a vocation as a clergyman—a pastor or elder.  The magisterial Reformation (as opposed to at least some elements of the radical Reformation) didn’t reject the idea of church or of church governance.  But it changed the place of the clergy within the broader social landscape, not by de-sacralizing the clergy, but by re-sacralizing all other vocations.274

This had implications for ecclesiastical jurisdiction.  The central issue was now conduct rather than status.  What mattered was whether particular substantive issues—theft, marriage, baptism—belonged in the jurisdiction of the church or of the state. The distinctions and debates on these issues reached their most refined level in the Reformed tradition influenced by John Calvin, who outlined an institutional distinction between church and state.275

Theft, murder, and mayhem should be handled by the state, which wields the sword to restrain violence.  It was through the civil government, John Calvin explained, that “the Lord has designed . . . to provide for the tranquillity [sic] of the good, and to restrain the waywardness of the wicked.”276  According to Calvin, the civil government was to “secure[]” the “safety of mankind”—“for except the fury of the wicked be resisted, and the innocent be protected from their violence, all things would come to an entire confusion.”277  Respect for civil government “is the only remedy by which mankind can be preserved from destruction, it ought to be carefully observed by us, unless we wish to avow ourselves as the public enemies of the human race.”278  Calvin defended these views against the radical reformation, which issued more fundamental challenges to authority of all kinds, of civil government as well as church.  Calvin argued that Christians were obligated to defend the faith against both those who were overly deferential to princes as well as those who fostered anarchy by taking the authority of civil government insufficiently seriously.279

Calvin exemplified this in his own life, where he had a long-running conflict with Genevan city leaders about the ability of the church to regulate access to the Lord’s supper.280  He also defended the authority of the church.  A powerful faction in Geneva rejected the standards of holiness promoted by Calvin and his allies.  They insisted that their Christian liberty permitted them to rely on grace and not bother with repentance and holiness.281  They also apparently rejected the authority of the state.282  Calvin, for his part, believed they were engaged in sin and labeled them libertines.283  Control over baptism or eligibility to partake in “communion” or the Lord’s supper was, on the other hand, a matter for church governance.284  When members of the libertine faction insisted on their right to receive the elements of the Lord’s supper, Calvin addressed them in the church service and promised to block their way physically if need be.285

Not everything was so clear cut.  Marriage, for instance, had long been the domain of the church’s jurisdiction.286  Within the Protestant world, some thought this should remain the case.  Others thought it was a shared domain, with a role for the state in authorizing marriage as a civil institution.287  Still others (notably radical Puritans) thought that marriage was entirely the state’s domain and refused to so much as engage in religious solemnities for marriage.288

The idea that it was conduct rather than status that would be decisive for the relative jurisdictions of church and state had concrete implications.  There is a longer and larger story to tell about how this shaped ideas about an institutional separation between church and state.  For the moment, it is sufficient to turn to England to note a smaller—but important—concrete outworking as it related to the criminal jurisdiction of the state over clergy. 

With the Reformation came a rollback of benefit of clergy.  Blackstone made the link clear in his own account: the modification of benefit of clergy was a reaction to “popish” excesses.289  The ecclesiastical proceedings, which Blackstone derided as “mock trials” based on “factious and popish tenets,” had the tendency to “exempt one part of the nation from the general municipal law.”290  The Reformation provided the occasion for change: “[I]t became high time, when the reformation was thoroughly established, to abolish so vain and impious a ceremony.”291

The changes started in the Henrican reformation.  In 1532, Parliament for the first time stopped the practice of shifting venue for the recipient of benefit of clergy—in the past the beneficiary would be transferred from the king’s court to the ecclesiastical authority (the local ordinary).292  As laymen had already been recipients of the benefit of clergy by virtue of their ability to read, in 1536, Parliament abolished other distinctions based on clerical status, essentially retaining the name of “benefit of clergy” but as a way of making the common law more merciful, rather than as a special privilege to clerics.  Under the 1536 statute, what had been developing for years became much more explicit: the benefit of clergy was equally applicable to laymen and clerics.293

Under the ultrareformed Edward, in 1547, Parliament passed a statute that provided new subject-matter limitations on invocation of clergy.  No longer would benefit of clergy be available for murder, for “cases of burglary and housebreaking, in which any person was in the house at the time,” for highway robbery, horse stealing, or robbing churches.294  Blackstone highlighted yet another change, a statute enacted under Elizabeth I that made permanent the rule that no longer would recipients of benefit of clergy be handed over to the ecclesiastical jurisdiction of the ordinary.295

In sum, by roughly the start of the seventeenth century, the benefit of clergy had lost all meaningful connection to the church.  It also had many significant subject matter-specific exclusions.  By 1600, there was a distinction between felonies considered too serious to be “clergyable” and minor ones that were still granted an immunity.  The serious ones were murder, house-breaking (burglary), horse-stealing, and some kinds of robbery, arson, and religious crimes.296  Meanwhile, its connection with the church was basically at an end.  Anyone who could read could claim benefit of clergy.297  Its function was no longer to protect clergy because they were sacred or belonged in the ecclesiastical court system.  Benefit of clergy remained in the law simply as a way of moderating the harshness of a criminal law with a lot of felonies for relatively minor crimes—most important among them, “petty larceny and manslaughter, that is theft not aggravated by other threatening behavior and under a certain monetary value and unlawful homicide without the malice and forethought required in murder.”298  The person who received the benefit was spared a death sentence but was branded on the hand.  Judges also had the option, which evolved over the sixteenth century, to order forfeiture of the defendant’s goods to the crown and even to order jailing for up to a year.299

The history of benefit of clergy often focuses on the incremental widening of coverage, diluting the privilege’s original orientation toward the church and repurposing the doctrine to moderate the harshness of the criminal law.300  All of which is true.  But there is more to the story than that.  Blackstone’s account makes it clear that the Reformation marks an evolution of legal thought about the jurisdiction of the church.301

Where English law started out was a (more or less) categorical immunity for clergy based on their status as clergy.  Where English law ended up by the 1600s was that the state had jurisdiction over acts that caused direct harm to others, regardless of who did it.  Benefit of clergy was no longer a status-based immunity but now simply a means to mitigation for punishment of a subset of crimes.302  For purposes of the relationship between church and state, what distinguished the domains was their function.  A clergyman who committed a crime had no immunity from what Blackstone would call the “municipal law.”  Religious liberty did not protect libertinism.  This was a principle that would be often repeated in American cases articulating religious liberty principles.

6.   Accountability Alongside Religious Liberty in America

This idea became a commonplace in early American state constitutions articulating religious liberty protections.  Religious liberty, these constitutions repeatedly asserted, does not protect licentiousness or conduct leading to societal disorder.303 Religious liberty would exist alongside accountability. 

Fairly typical was the New York Constitution of 1777: 

[T]he free exercise and enjoyment of religious profession and worship, without discrimination or preference, shall forever hereafter be allowed, within this State, to all mankind: Provided, That the liberty of conscience, hereby granted, shall not be so construed as to excuse acts of licentiousness, or justify practices inconsistent with the peace or safety of this State.304

This was not unique to New York.  New Hampshire protected religious liberty so long as the one invoking it “doth not disturb the public peace, or disturb others, in their religious worship.”305  Maryland’s Declaration of Rights in 1776 protected “religious liberty . . . unless, under colour of religion, any man shall disturb the good order, peace or safety of the State, or shall infringe the laws of morality, or injure others, in their natural, civil, or religious rights . . . .”306  The Massachusetts Constitution of 1780 protected religious liberty so long as it does not lead its exponents to “disturb the public peace, or obstruct others in their religious worship.”307  South Carolina’s Constitution included the proviso that “the liberty of conscience” it protected “shall not be so construed as to excuse acts of licentiousness, or justify practices inconsistent with the peace or safety of this State.”308  Rhode Island retained in effect its 1663 charter, which protected “differences in opinione in matters of religion” so long as those with difference “doe not actually disturb the civill peace of our sayd colony,” and that they are “not useing this libertie to lycentiousnesse and profanenesse, nor to the civill injurye or outward disturbeance of others . . . .”309

In sum, at the time of the First Amendment’s creation, “[n]ine of the states limited the free exercise right to actions that were ‘peaceable’ or that would not disturb the ‘peace’ or ‘safety’ of the state.”310  Four of these singled out “licentiousness or immorality” for specific prohibition.311

There’s at least a plausible argument that these kinds of ideas shaped the meaning of what was covered by religious free exercise at the time of the drafting of the First Amendment.  To the extent that is the case, then religious liberty itself should be thought of as defined in relationship to an accountability principle. 

To be sure, these state constitutional provisions are not specific enough to sort out the hard issues of where exactly to draw the line between church autonomy and accountability.  But they do remind us yet again that an accountability principle exists right alongside religious liberty and autonomy principles.  As this Part has argued, both autonomy and accountability are justified side-by-side from the same theories, whether it is pluralist sphere sovereignty or more positivist historical arguments.

IV.     Framework for Analyzing a Church Autonomy Defense Based on Church Government

Both church autonomy and accountability are important principles.  The problem in the current law is that the courts are unclear on how to avoid allowing one to swallow up the other.  As we’ve seen, the most troublesome—and potentially hardest to cabin—is the idea of church government as a separate header for protection.  Beyond that, there is the question of when to turn to which doctrinal tool.  This Part proposes some analytical revisions that would help to clean up what has become a doctrinal mess.  In terms of doctrine, the shifts suggested here are modest and can be done without any change by the Supreme Court.312  Indeed, most of what I suggest can be found in the best practices of the lower courts when one looks for it.  I hope that putting the pieces together here will help clarify the doctrine and also highlight the best practices for other courts and practitioners. 

A.   The Mistaken Quest for a Single, Unifying Principle

There is not a single, bright-line test that neatly fixes everything in the church autonomy space.  Most attempts to delineate the borders of the concept have been unsatisfactory, leaving counterexamples unaddressed.  But a key part of the problem is that the attempts to delineate the borders attempt to do so with a single doctrinal tool as the cure-all: if there’s a religious reason, then it’s protected, otherwise not; if there’s a neutral way to decide, then the court can proceed, otherwise not; if the issue is one of membership, then the court can proceed, otherwise not; if the tort is intentional, the court can proceed; if negligence, then it cannot. 

Religion clause jurisprudence is not easy to fix with just one principle.  Establishment Clause jurisprudence has long been torn among multiple conflicting tests and values.  The attempt to simplify Free Exercise Clause analysis into a single, overriding neutrality principle313 has had mixed results.314  And while it may not be necessary or desirable to try to create a Religion Clause doctrine as complex as modern Free Speech Doctrine,315 there’s certainly space for a little less obsessing over cure-all doctrinal tools.316

Academics have periodically criticized the dominant tendency in the legal world to seek unified doctrinal tests that paper over complexity.  Guido Calabresi and Philip Bobbitt famously suggested that law must recognize when it is dealing with “tragic” choices, ones that pose irreconcilable values which cannot be properly satisfied.317  Marc DeGirolami adapted the category of tragedy to develop a critique of religion clause law in particular.318  It is an area of law that has tended to err by pursuing sweeping, rationalist answers that comply with some single theory or principle—in the process, failing to recognize competing considerations.  A step further in abstraction, J. Joel Alicea has argued that one of the fundamental divides in constitutional theory is between more rationalist modes of analysis, which he associates with liberal political theory, and more incremental, tradition- and practice-based modes, which he associates with Burkean conservatism.319

This is not the place to make a substantive case for an entire philosophical approach to the enterprise of judging or the development of legal doctrine.  For the moment, it is enough to note that my approach of selecting from the various doctrinal tools already extant in the cases builds on this conversation.  I am cautious about overarching claims that a particular theory or a single principle can resolve all the hard issues.  My proposed solution is incremental.  It is not to eliminate the doctrinal tools and leave courts essentially with an all-things-considered, fact-intensive approach to deciding the issues.  It is instead to draw from the existing caselaw and suggest a way of putting the doctrinal pieces together in a way that makes better sense of the multiple considerations at play in church autonomy cases.

B.   Refashioning the Doctrinal Pieces into a Framework for Analysis

We now come back to where we started, the doctrinal confusion in the courts.  We bring to it normative commitments and a methodological perspective.  The normative commitments are to church autonomy and to accountability.  The methodological perspective is the openness to the common-law method of incremental development and eclecticism.  The question for this Section is how to put the doctrinal pieces together to achieve accountability while still retaining a robust church autonomy protection for churches, one that covers not just religious doctrine and belief but also church governance.

1.   Religious Belief

The first piece of the puzzle is the most complex: does conduct have to be religious in order to receive church autonomy protection?  There has been confusion about this point.  Particularly in the employment cases, courts (including the Supreme Court) have frequently said that there need be no showing that particular conduct is religiously motivated.  But in other cases, courts have said that only when the challenged action is religious is it protected.  As we’ve seen, some courts say yes; others, including the Supreme Court, say no.  It’s really a semantic difference—but one with potentially costly consequences. 

The best way to reconcile the cases is to note, first, that church autonomy protects religious conduct.  Church autonomy principles protect religious conduct, not just anything done within a church.  It’s a plausible way to read the religion clauses to the Constitution.  If the conduct is religious, then the church autonomy doctrine can be applied; if not, don’t.  But this is not the same as saying that the church has to have a religious reason for each challenged act.  That would, as we’ve seen, contradict the settled rule that ministerial hiring-and-firing decisions do not depend on the church having a religious reason. 

In the employment cases, the religious character of the conduct is still proven, but at an earlier step: once the institution is shown to be religious, then courts will conclusively presume that employment decisions regarding its ministers are religious.320  It is still a showing that religion is implicated—a point about which the courts have long been unclear.  It’s just that religion is implicated at an earlier stage: the institution must be religious before its decisions to hire and fire get church autonomy protection.321  Its selection of ministers implicates its religious character for the same reason the selection of any expressive association’s leaders implicates its message.322  (Limiting the ministerial exception to ministers roughly aligns with the selection of a leadership or authority role in the organization.)323  Similarly, membership standard setting could be presumed religious.

There is more than one way for conduct to be religious.  One way is for it to be religious is the obvious one: have a religious reason for the challenged conduct.  The other way is less obvious: be engaged in running a religious entity.  But because this latter idea is really a presumption, it has to be limited to a subset of practices that are classic church governance lest it swallow up any possibility of exception.  This is where precedent comes into the picture.  Courts cannot pretend to be totally agnostic as to the form church governance takes.  Instead, a handful of issues that are recognizable as church governance—including hiring and firing decisions and didactic instruction and correction—can appropriately be absolutely immunized.  There are other issues that could hypothetically be matters of church governance but are so far outside the norm that there should be no presumption. Imagine a church that said it administered spankings on members for violations of membership standards.  There’s no reason that this couldn’t be part of church governance, but it is not a standard part of church governance and so it should not get immunity unless it is coupled with the element discussed in the next section: consent. 

2.   Doctrine or Religious Rationale

The most straightforward way for a court to find that the conduct at issue is religious is when it is done directly pursuant to religious doctrine or beliefs.  If the claim is simply that deciding the case would require the court to decide a matter of church doctrine or religious conviction, then the matter is relatively straightforward. 

In a variegated set of cases, courts are asked to decide issues that could impinge on matters of church doctrine.  It is in these cases that the courts should ask whether the resolution of the case necessarily impinges on matters of doctrine or belief, or if it can be avoided and the case decided simply by neutral principles of law.  If doctrine or beliefs are an unavoidable part of the case, the court should not take it.  That is why clergy malpractice suits are inherently violations of church autonomy principles—it requires the court to determine a legal standard based on distinctively religious job descriptions.  A court should not say that Pastor A was negligent because a reasonable pastor would have cared for his congregants in manner X, which is a religious determination about pastoral duties.  Clergy malpractice, in which pastoral conduct is turned into a liability standard by a court applying judicially adopted standards of what a reasonable pastor would do, is unconstitutional.324

But a court can say that Pastor A was negligent for hiring a serial sex abuser to run childcare.  This isn’t a religious standard.  One doesn’t have to consult religious doctrine because one isn’t holding that this is how a reasonable pastor would run a job search.325

On the other hand, this is why most property cases should be determined by looking to the formal legal deed and trust documents, allowing the court to avoid matters of doctrine (like deciding which of two church factions best conforms to the original church doctrine). 

3.   Classic Church Governance

Church governance is a matter of religious importance.  Churches and all other religious traditions that have institutional expressions do so because they attach religious importance to the lived-out expression of faith, in a community with some kind of structure.326  But it’s not always clear that individual decisions are matters dictated by doctrine.  That doesn’t mean that governance of a religious institution is not itself a matter of religious conduct.327  To the contrary, as the Supreme Court appropriately recognized, “internal church government” is “an issue at the core of ecclesiastical affairs.”328

Church government is susceptible of being stretched to cover anything and everything that happens inside the church; the question is how to stop short of that point.  A few cases have led to courts more or less directly holding that not everything done by a religious organization is church government.  Some have held, for example, that the details of a health-insurance contract between a religious employer and its employees was a matter of general employment law rather than church government, and so not covered by church autonomy.329  A California court noted that the employees were not church members,330 but otherwise offered no elaboration on how it reached this conclusion; it was simply presented as obvious and commonsensical.  The holding does not provide much guidance for the next case.

A set of classic, recognizable governance actions should be considered religious.331  This is the best way to rationalize the ministerial exception with the rest of church autonomy doctrine.  But the zone of protected matters of institutional governance are not, and should not be, unlimited.332  Historically, courts dealing with church autonomy matters talked about the necessity of churches controlling their own governance in order to propagate their faith and maintain their faith communities.333  They also talked about consent principles coming into play when a religious institution had membership.334

The list of subjects that should be considered religious without detailed demonstration should be narrow.  Three categories emerge from the cases:

First, the selection, supervision, and retention of ministers is a quintessential component of the religious governance of any religious institution.335  Such decisions are complex.336  They take into account standards of orthodoxy and orthopraxy.  But they also take into account considerations of fit in ways that can be often fairly subjective, hard to articulate, and can vary widely.337  It shouldn’t be necessary for the religious institution to justify its employment decisions to a court.338  It is in this way that the courts could best rationalize protecting the religious entity even when it wants to discriminate on the basis of a protected characteristic, like sex or sexual orientation—this is still part of the protected domain of church governance.339

Second, matters of membership should be considered religious.340  A church should be able to say who is in and who is out of membership without a court scrutinizing the matter to decide whether the church got the matter right.341

Finally, matters of church discipline are classic instances of church governance.342  This has the most potential for abuse.  One could imagine any number of ways in which this could be manipulated into a cover for clergy abuse.  Pastor A says that spanking of members is a form of discipline, or that his having sex with a female member was a way to teach the woman submission, or that sexual favors extorted from a minor were a way of teaching absolute obedience, and so on.  There are reasons that all of these points are implausible as theological positions within the mainstream of Christian orthodoxy and many other major religious traditions.  But that said, it is still the case that sects could claim to believe these things, and a court would have a hard time telling whether it’s true or not.  (While sincerity should, I think, be required, it’s still often hard to judge.)343

Rather than an open-ended protection of church government, the forms of church governance here need to be carefully limited.  The presumption, historically, was that two forms of church discipline were immunized from judicial scrutiny: verbal corrections and excommunication.  By verbal, I mean correction by words, whether written or spoken.344  By excommunication, I mean the removal of the person from membership in the religious institution and the exclusion of that person from ritual or sacramental observances.345

Church discipline, in the Christian tradition, is not primarily about punishment or punitive measures.346  It is about correction in order to restore an individual to fellowship.347  The courts have long taken for granted that discipline will not be unduly punitive, assuming that discipline would look more or less like the Christian position.348  Limiting discipline to verbal correction and exclusion from various rites of the religious community is a plausible way to generalize this casual assumption in the caselaw, avoiding the potential problem of shielding physical abuse under the guise of discipline without adopting an excessively Christian-specific rule.

Conduct taken as part of church governance can be religious, even if it does not always have specific religious reasons for each individual act.  But this presumption of a religious character should only be for a relatively short list of classic governance functions that have been developed in the cases. 

The upshot of this is that religious congregations will be able to hire and fire ministers without justifying their decisions to a court.  Church autonomy doctrine provides them with exemption from the usual antidiscrimination laws.  Religious institutions should not be subject to suits in equity seeking relief from a membership decision to (for instance) bar a member from participating in communion.  That would be interfering with the religious organization’s autonomy over membership and discipline.  Similarly, defamation suits based on the corrective process of church discipline are generally related to church governance.349

4.   Consent

Consent principles help define the scope of church governance issues.  Consent figured heavily in the preconstitutional common-law cases that provided the backdrop for modern constitutional church autonomy.350  It has been referenced less frequently, and with more ambivalence, in recent cases.351  Meanwhile, some scholars have advocated making consent the whole of church autonomy.352  This Article argues for a pragmatic middle ground.  Consent deserves to be part of the analysis, as evidence of what counts as protected church governance in a given situation.  But it is not the totality of church autonomy.

Where consent comes in is to solve the problem of when to treat particular conduct as religious that does not easily fit the categories that have become classic instances of church governance in the caselaw. 

Suppose, for instance, that Church A has an unusual form of discipline: a ceremonial slap on the wrist.  This doesn’t fall into the narrow and cautious category of verbal rebuke that the cases (construed conservatively) have recognized.  Arguably this is because physical slap is not a standard part of church practice.  But if it were, there would have been as many battery suits as there have been defamation suits arising out of church discipline. 

The solution on this point is consent.353  A religious institution can have the benefits of immunity when there is a clear consent to the later-challenged conduct.  This is likely to be most important with unusual forms of discipline.  A member could consent to a physical slap as a form of discipline, for instance.  If consent was not coerced and was informed, this will count as a consent defense under tort law. 

That may raise a theoretical or conceptual concern that the defense is redundant.  If it’s already a defense under tort law, what does it add to have it be a constitutional church autonomy defense under the heading of consent?  The simplest answer is that it makes consent in the context of church autonomy beyond legislative change.  Imagine a paternalistic legislature that wants to eliminate the ability to consent to torts.  It wants to end boxing, MMA, and other violent forms of conduct.  But a religious group that incorporates corporal punishment into its discipline, or that engages in the kinds of crucifixion reenactments common in the Philippines as part of its passion week observance,354 could have a constitutional claim that its members should be able to consent. 

Consider where this could come into play in a more concrete form: a mandatory reporting statute requiring church leadership to report child abuse.  Imagine that a group objects to this as an infringement on its ability to self-govern.  But there is no existing established tradition that reporting crime infringes on the prerogatives of the religious institution’s self-governance.  To the contrary, the history suggests that this is what one would have expected religious entities to do, building on the English and common-law history taken into the American context.  Meanwhile, the people being protected by the mandatory reporting laws—children, the abused—either can’t or won’t consent to not report.  This just isn’t going to be covered by church autonomy. 

Consent also helps to solve the problem of privileging a particular set of religious practices.  One possible objection to the approach outlined so far is that it privileges traditional forms of religion.  Christian churches have had by far the most litigation in America.  So of course, what is “recognizable” as church governance in the caselaw is going to be disproportionately shaped by this Christian tradition.  If we basically freeze identifiable church governance activities with the easily recognizable functions in the caselaw, that is effectively freezing a set of practices that are most identified with church practices: selection of ministers, discipline in the form of rebuke and disfellowshipping.  On what basis could an unusual practice make its way into the category of church governance?  Again, this is where consent could come into the picture. 

5.   Neutral Principles

The last step is neutral principles.  If a challenged church action is not based on a religious reason and doesn’t fall into a recognized form of church governance, the courts should consider whether there is consent, which in this context has a constitutional significance.  If none of these elements are in place, then the court can adjudicate the matter using neutral principles of law.  These are principles that do not take a position on church doctrine (which is what “neutral” means in the church property cases) and do not target or discriminate against religion (which is, more or less, what “neutral” means in the Free Exercise cases applying Smith and its progeny).355

C.   Subject Matter as a Backstop?  Or, the Impossibility of Ultimate Neutrality

The approach outlined so far will cover most cases that actually arise.  It will also take the odds and ends of doctrinal tools already in use by the courts and organize them into a logically structured analysis, rather than leaving them to be drawn upon in a grab-bag manner, with unpredictable and inconsistent results.  The analysis so far in this Part provides a way of sorting out the kinds of issues that really might be part of religious-liberty interests of religious institutions from those that could be manipulated under the confusion about what constitutes church governance.

Yet all that said, one might still worry about a more drastic reduction of church autonomy: a religious institution that actually has criminal conduct as part of its religious precepts.  Why doesn’t religious autonomy protect human sacrifice or ritual sex with underage victims?356

1.   The Worry (and a Word About Its Hypothetical Character or Lack Thereof)

Maximalist proponents of church autonomy sometimes suggest that the religiously motivated abuser is more a hypothetical than an actual threat.  And they are mostly right.  There is a disjuncture between the cases that are actually brought in court, on the one hand, and the most extreme predictions of church autonomy’s critics on the other hand.  Very rarely do the actual cases involve church autonomy being asserted to directly protect clergy abuse or clergy coverups.357  In part this is because religious bodies do not want to claim serious clergy misconduct as part of their religious tradition.358  And there simply aren’t many practitioners of (say) ancient Carthaginian child-sacrifice cults in existence, seeking institutional religious autonomy.359

But I don’t think the paucity of real-world offenders is a sufficient answer to the objection.  At least for purposes of argument, one should concede that there is a real worry here.  We may not have Carthaginian Moloch worshippers conducting ritualized child sacrifice, but we can easily hypothesize a religious sect with harsh physical discipline.  This is where it makes sense to assert a final backstop principle. 

2.   Articulating a Backstop Principle

Consider the following minimal backstop principle: 

Church autonomy does not protect historically recognized criminal conduct or direct acts of physically harmful tortious conduct (and the latter could be rendered nontortious by consent where consent would be a defense in tort). 

This is entirely compatible with the caselaw as it exists.  Something like this backstop has occasionally been hinted at in the courts, but it has been inconsistently theorized and articulated.  (Consider the assertion that church autonomy doesn’t cover issues of tort, which comes up occasionally but is flatly incompatible with an important line of church autonomy cases.)360

The assumption behind this backstop principle is simply that what the courts have recognized as church autonomy so far can be taken as a given; going beyond the categories protected so far should be met with skepticism.  The backstop principle sets a line: thus far and no farther.  It is informed by the cases, but it’s also consistent with the broader principles of accountability spelled out in Part III above.  It is thus not arbitrary but informed both by theory and by the pragmatic tradition of common-law decisionmaking.  It’s worth pointing out a few characteristics of this backstop principle.

First, this gets at the truth underlying the overbroad statements that church autonomy doesn’t protect against tort.  The courts simply have not allowed church autonomy to shield one who directly causes physical harm. 

Second, this is narrow (it is only a backstop).  Formulated as it is here, it does not challenge the courts’ long line of decisions to allow church autonomy to defend against claims arising from economic harms in the form of the ministerial exception.  Nor does it question the application of the church autonomy defense to defamation, which causes reputational harm but not physical harm.  And by requiring direct acts, it does not preclude church autonomy protection for at least some kinds of supervisory decisions (although as noted above, I am inclined to think that negligence should be available against a supervisor when it does not implicate religious doctrine). 

This all leads to the third observation, that this minimal, bright-line test is consistent with the existing church autonomy caselaw.  Fourth, this is in line with the idea that there is some area of licentious or unlawful acts that is outside the protection of religious liberty generally.361  This is the case whether one looks to positivist evidence of the common law and original public meaning, to pluralist theory, or indeed even to Christian theology.  But by keying this to historical standards of criminal law, this makes it clear that religious autonomy can’t be whittled down by creative, religion-targeting laws that try to play within this category.362

3.   The Character of the Backstop

The backstop principle I’m suggesting here operates as a subject-matter articulation of limits to church autonomy.  This points to a larger characteristic of church autonomy which is not always articulated adequately.

Church autonomy is best thought of as a discrete carve-out from laws of general applicability.  It is jurisdictional, but not in a territorial way.  It’s not as though things belong to the church because they happen in the church building, any more than it is the case that everything that happens in the state can be dictated by the state.  Rather, church autonomy represents a kind of legal pluralism.  The church is a jurisdiction separate from the state, in the literal sense of being an authority distinct from the state.  Its jurisdiction goes to subject matter, not territory and not person.363  The subject matter of the church is religious.  The subject matter of the state is civil conduct.  Render to Caesar what is Caesar’s and to God what is God’s.364

When we recognize that this is a jurisdictional principle, in the sense of articulating the subject matter that belongs to the church and the subject matter that belongs to the state, then there is no escaping the fact that, in the end, there is a substantive judgment to be made here about the rightful domains of each.  One must recognize that the state is the one to primarily address civil harms, and the church has the authority to address specific spiritual and relational aspects of this in its formative and corrective discipline.  Occasionally the state’s laws can intrude on the church’s ability to conduct its basic operations.  The church autonomy principle can be described as a way of making sure that a handful of basic religious functions can proceed without direct government regulation.  But the church autonomy principle does not divest the state of its basic role in protecting the public peace, and that is why there is a need for a backstop rule.  In an area of law full of bright-line rules and less clear on limiting principles, such a backstop could be useful in making clear that there are some issues which won’t be protected even in the extremely unlikely event that church autonomy principles align around them.

One might object that even this backstop principle is, in its own way, religious.365  Perhaps it is itself a religious establishment.  By deciding what are the borderlands of the subject-matter jurisdiction of church and of state, one is inevitably making a decision freighted with theological weight.  The church autonomy doctrine, in articulating its own limits, is not—arguably, cannot be—absolutely agnostic about the nature of the church or of the state.  Rather, it inevitably draws on some assumptions about what constitutes the rightful domain of religion.366  In America, the law was historically shaped by centuries of religious history including Christian theological baggage.367  The cases developing the doctrine deal overwhelmingly with Christian religious institutions of one sort or another.368  It builds, often unconsciously, on a culture shaped by an idea of church and state as spheres, one with the keys of the (spiritual) kingdom, the other with the sword to enforce the laws of the (temporal) kingdom.369

This pervasive influence of religion on the law in this area extends even to the backstop suggested here.  Pointing to physical harms as a clear backstop principle builds on a theological understanding about the domains of church and of state.370  It certainly can—and I hope will—appeal, at least as a starting point, to people from other backgrounds.  But it is also the case that, if the backstop principle was to be designed by a minority religious group with traditions of corporal punishment, maybe the assumptions would be different. 

If the influence of Christianity in this context is raised as an objection, though, it is an objection that proves too much, for it could be leveled against any form of religious liberty.371  No system of interaction between religious institutions and the state occurs outside a history shaped by religious traditions, conflicts, and beliefs.372  There is no neutral, nonreligious ground for evaluating the subject if one is rigorous enough about neutrality.373  Neutrality in any absolute sense is, and must always be, more myth than reality.374  The idea of the autonomy of religious institutions, of the responsibilities of the state, and more, are all shaped by the religious histories of those concepts.  They might be retheorized in any number of ways but without getting rid of the ultimate objection.  Even Rawlsian liberalism, veil of ignorance and all, is at least open to the charge (leveled from many directions) of being religious (though in what way has of course been a subject of heated debate).375  Without trying to resolve the debate about liberalism or other grand questions of political theory, the approach in this Article has been to accept the history that we’ve been given and try to come up with a workable approach. 

Rather than try to escape this set of historical legacies and commitments, I recognize them and am willing to build on their foundation.  If one is not willing to take a stand on these religiously informed grounds developed over the centuries of the common-law experience, there really is no religiously neutral vantage point to escape to.  In this field, the sand shifts underfoot ceaselessly.  Religious history and religious commitments inevitably shape the assessment.  But from a wide variety of backgrounds and with a diverse range of theoretical commitments, at least this minimal backstop limit on church autonomy should be appealing.

*     *     *

Actions taken for religious reasons are relatively straightforward to identify.  It occasionally requires some fine line drawing, as in the distinction between clergy malpractice (unconstitutional) and general negligence (constitutional).  A constrained set of the governance decisions of a religious institution are presumptively religious.  Courts can and should look at the historical precedents to decide what counts as religious governance.  Within the governance decisions of a religious body, such as hiring or firing a minister, the body need not individually justify each decision on a religious basis.  These actions can also be defined cautiously—for instance, church discipline is protected insofar as it consists in words.  But if the case is an alleged battery in the course of church discipline, the church has left the realm of standard religious discipline.  Explicit consent could still be obtained, which is what keeps this area of law from ossifying into categories almost entirely defined by Christian churches.  If courts get to this point though and none of the above principles apply to bring church autonomy doctrine into consideration, then they can adjudicate the case.  If it’s not a core internal governance function, nor a decision that requires review of an issue of belief or doctrine, then it is a matter that “neutral principles of law” can address.  The standard of “neutral principles” is the end of the analysis, not the beginning.  Finally, there’s a backstop principle, that church autonomy does not protect the direct cause of physical harm.  This catches the egregious but exceptional case that slips through the other filters: if there is a religiously motivated act that causes physical harm, the perpetrator should still be subject to accountability before the civil authorities. 


Church autonomy doctrine protects the religious conduct of a religious institution.  But autonomy should not exclude accountability for civil wrongs.  Even accepting the most robust version of church autonomy in the caselaw and the most expansive version of church autonomy’s theory, the recognition of a rightful authority for religious institutions only exists alongside a rightful authority on the part of the state to protect against harm.  Sphere sovereignty honors the duty of the state as well as the responsibility of the church.  The history of church-state development going back into English history—the very history which the Supreme Court has drawn upon to defend robust church autonomy principles—contains lessons that counsel careful protection of accountability principles.  The original meaning of the First Amendment plausibly incorporates a backstop principle that religious liberty does not protect breaches of the peace. 

All of this, in turn, can inform church autonomy doctrine in the courts.  Out of the chaotic grab bag of doctrinal tools currently to be found in the cases, this Article has suggested a way to systematize the analysis and design a backstop that will assure that church autonomy not descend into lawlessness for churches.  Church autonomy and church accountability should exist side-by-side, as both essential parts of a well-ordered constitutional system.

©2023 Lael Weinberger.  Individuals and nonprofit institutions may reproduce and distribute copies of this Article in any format at or below cost, for educational purposes, so long as each copy identifies the author, provides a citation to the Notre Dame Law Review, and includes this provision in the copyright notice.

*Olin-Searle-Smith Fellow in Law & Lecturer, Harvard Law School.  For conversations and feedback that made this Article better, I thank Evelyn Atkinson, Daniel Blomberg, Charlie Capps, Nathan Chapman, Marc DeGirolami, John Ehrett, Carl Esbeck, Noah Feldman, Owen Gallogly, Doug Laycock, Zalman Rothschild, Zahra Takhshid, Diana Thomson, Sarah Weinberger, Laura Weinrib, and participants in a faculty workshop at St. John’s University School of Law and in the Nootbaar Institute Fellows Workshop at Pepperdine Caruso Law.

1See, e.g., Paul Horwitz, Comment, The Hobby Lobby Moment, 128 Harv. L. Rev. 154 (2014); Elizabeth Sepper, Reports of Accommodation’s Death Have Been Greatly Exaggerated, 128 Harv. L. Rev. F. 24 (2014).

2Church autonomy is sometimes called “religious autonomy” because (of course) it applies to religious institutions from all traditions—synagogues, mosques, and more, not just churches.  For purposes of this Article, I will use the term “church autonomy” most often—not to exclude other faith traditions, but for two practical reasons: first, church autonomy is the most common term in the cases.  See, e.g., Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2061 (2020).  Second, it is descriptive of the historical and sociological context in which the doctrine developed, which is predominately in litigation from Christian churches. 

3See, e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 181 (2012).

4This doctrine also appears under other labels, including “religious autonomy” and “ecclesiastical abstention.”  In employment discrimination cases (and sometimes other contexts, like tort and contract), the “ministerial exception” is the term used for the application of church autonomy doctrine.

5See, e.g., Westbrook v. Penley, 231 S.W.3d 389, 395 (Tex. 2007); Paul Horwitz, Churches as First Amendment Institutions: Of Sovereignty and Spheres, 44 Harv. C.R.-C.L. L. Rev. 79, 83 (2009); Mark DeWolfe Howe, The Supreme Court, 1952 Term—Foreword: Political Theory and the Nature of Liberty, 67 Harv. L. Rev. 91, 94 (1953); see also Korte v. Sebelius, 735 F.3d 654, 678 (7th Cir. 2013).

6See, e.g., Caroline Mala Corbin, Above the Law?  The Constitutionality of the Ministerial Exemption from Antidiscrimination Law, 75 Fordham L. Rev. 1965, 1968 (2007); Jean L. Cohen, Sovereignty, the Corporate Religious, and Jurisdictional/Political Pluralism, 18 Theoretical Inquiries L. 547, 551–52 (2017).

7See, e.g., Zoë Robinson, The First Amendment Religion Clauses in the United States Supreme Courtin The Cambridge Companion to the First Amendment and Religious Liberty 219, 247 (Michael D. Breidenbach & Owen Anderson eds., 2020) (describing religion clause jurisprudence since 2010 as dominated by institutions); 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, 474 & n.42 (2015) (describing the rise of church autonomy as part of a “corporatist turn in Supreme Court jurisprudence”).

8See Marci A. Hamilton, The Waterloo for the So-Called Church Autonomy Theory: Widespread Clergy Abuse and Institutional Cover-Up, 29 Cardozo L. Rev. 225, 228 (2007); Angela C. Carmella, Catholic Institutions in Court: The Religion Clauses and Political-Legal Compromise, 120 W. Va. L. Rev. 1, 54 (2017); Christopher J. Merken, Recognizing Hosanna-Tabor’s Limited Scope and Inapplicability to Clergy Sex Abuse Litigation, 2020 Ark. L. Notes 60, 64.  Scott C. Idleman, Tort Liability, Religious Entities, and the Decline of Constitutional Protection, 75 Ind. L.J. 219 (2000), predicted—incorrectly, as it turned out—that church abuse cases would precipitate a drastic decline in protection for religious institutions.

9See Richard Schragger & Micah Schwartzman, Against Religious Institutionalism, 99 Va. L. Rev. 917, 918 (2013); Elizabeth Sepper, Zombie Religious Institutions, 112 Nw. U. L. Rev. 929, 932 (2018); Leslie C. Griffin, The Sins of Hosanna-Tabor, 88 Ind. L.J. 981, 982 (2013); Corbin, supra note 6, at 1968; Cohen, supra note 6, at 575; Marci A. Hamilton, God vs. the Gavel: The Perils of Extreme Religious Liberty (2d rev. ed. 2014) (arguing that an array of social harms are protected from legal accountability in the name of religious liberty).

10See, e.g., Ira C. Lupu & Robert W. Tuttle, The Mystery of Unanimity in Hosanna-Tabor Evangelical Lutheran Church & School v. EEOC, 20 Lewis & Clark L. Rev. 1265, 1302 (2017); Michael A. Helfand, Implied Consent to Religious Institutions: A Primer and a Defense, 50 Conn. L. Rev. 877, 893 (2018).

11See Watson v. Jones, 80 U.S. (13 Wall.) 679, 722 (1871); Gonzalez v. Roman Cath. Archbishop of Manila, 280 U.S. 1, 16 (1929); Kedroff v. Saint Nicholas Cathedral of the Russian Orthodox Church in N. Am., 344 U.S. 94, 116 (1952); Presbyterian Church in the U.S. v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440, 446 (1969) (quoting Watson, 80 U.S. (13 Wall.) at 729); Serbian E. Orthodox Diocese for the U. S. & Can. v. Milivojevich, 426 U.S. 696, 708–09 (1976); Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 188–89 (2012); Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2060 (2020).

12See, e.g., Schragger & Schwartzman, supra note 9, at 946; Griffin, supra note 9, at 998; see also Lupu & Tuttle, supra note 10, at 1298.

13See infra Part II. 

14See Zoë Robinson, What Is a “Religious Institution”?, 55 B.C. L. Rev. 181, 206 (2014).

15See, e.g., Bryce v. Episcopal Church in the Diocese of Colo., 289 F.3d 648, 655 (10th Cir. 2002); Douglas Laycock, Church Autonomy Revisited, 7 Geo. J.L. & Pub. Pol’y 253, 254 (2009).

16See generally Noah Feldman, Non-Sectarianism Reconsidered, 18 J.L. & Pol. 65 (2002) (noting the historical importance of keeping the state out of interreligious disputes).

17For arguments that the Establishment Clause provides the better justification for church autonomy, see Carl H. Esbeck, The Establishment Clause as a Structural Restraint: Validations and Ramifications, 18 J.L. & Pol. 445 (2002) [hereinafter Esbeck, Validations and Ramifications]; Ira C. Lupu & Robert W. Tuttle, Sexual Misconduct and Ecclesiastical Immunity, 2004 BYU L. Rev. 1789, 1808 n.58; Carl H. Esbeck, The Establishment Clause as a Structural Restraint on Governmental Power, 84 Iowa L. Rev. 1, 42–51 (1998) [hereinafter Esbeck, Governmental Power]; Thomas C. Berg, Vouchers and Religious Schools: The New Constitutional Questions, 72 U. Cin. L. Rev. 151, 210–11, 218–19 (2003). 

18See Douglas Laycock, Towards a General Theory of the Religion Clauses: The Case of Church Labor Relations and the Right to Church Autonomy, 81 Colum. L. Rev. 1373, 1373 (1981); Douglas Laycock, Theories of Interpretation: Free Exercise Clause and Establishment Clausein Religion and American Law: An Encyclopedia 516 (Paul Finkelman ed., 2000), reprinted in 1 Religious Liberty 103, 110–11 (Douglas Laycock ed., 2010); Kathleen A. Brady, Religious Organizations and Free Exercise: The Surprising Lessons of Smith, 2004 BYU L. Rev. 1633; Michael W. McConnell, The Problem of Singling Out Religion, 50 DePaul L. Rev 1, 19–20 (2000); Laycock, supra note 15, at 263–64; see also Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 946 (9th Cir. 1999).

19For description of the debate, see Helfand, supra note 10, at 890–91.

20Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2060 (2020).

21The definition of a religious institution is itself a contestable issue, see Shaliehsabou v. Hebrew Home of Greater Wash., Inc., 363 F.3d 299, 309–11 (4th Cir. 2004); Conlon v. InterVarsity Christian Fellowship/USA, 777 F.3d 829, 833–34 (6th Cir. 2015); Brian M. Murray, The Elephant in Hosanna-Tabor, 10 Geo. J.L. & Pub. Pol’y 493 (2012), as is the definition of minister, see Gordon Coll. v. DeWeese-Boyd, 142 S. Ct. 952 (2022) (Alito, J., respecting the denial of certiorari) (joined by Justices Thomas, Kavanaugh, and Barrett); Seattle’s Union Gospel Mission v. Woods, 142 S. Ct. 1094 (2022) (Alito, J., respecting the denial of certiorari) (joined by Justice Thomas); Helen M. Alvaré, Church Autonomy After Our Lady of Guadalupe School: Too Broad?  Or as Broad as It Needs to Be?, 25 Tex. Rev. L. & Pol. 319, 330–32 (2021).  This Article brackets those important questions to consider the subject matter protected by church autonomy doctrine once it applies.

22Early cases before the incorporation of the First Amendment against the states reached church autonomy results but as a matter of common-law reasoning.  See, e.g., Watson v. Jones, 80 U.S. (13 Wall.) 679, 722 (1872).

23See, e.g., Helfand, supra note 10, at 901.

24See, e.g., Luke C. Sheahan, Why Associations Matter: The Case for First Amendment Pluralism (2020).

25See, e.g., Gerard V. Bradley, Church Autonomy in the Constitutional Order: The End of Church and State?, 49 La. L. Rev. 1057, 1085 (1989).

26See Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2060 (2020).

27See, e.g., Bryan R. v. Watchtower Bible & Tract Soc’y of N.Y., Inc., 738 A.2d 839, 848 (Me. 1999).

28Our Lady of Guadalupe Sch., 140 S. Ct. at 2060 (quoting Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 186 (2012)).


30See id.Hosanna-Tabor, 565 U.S. at 188.

31See, e.g.In re Diocese of Lubbock, 624 S.W.3d 506, 516 (Tex. 2021), cert. denied, 142 S. Ct. 434 (2021) (mem.); Westbrook v. Penley, 231 S.W.3d 389, 396 (Tex. 2007); Hubbard v. J Message Grp. Corp., 325 F. Supp. 3d 1198, 1214 (D.N.M. 2018); Pfeil v. St. Matthews Evangelical Lutheran Church of the Unaltered Augsburg Confession of Worthington, Nobles Cnty., Minn., 877 N.W.2d 528, 541 (Minn. 2016); Hiles v. Episcopal Diocese of Mass., 773 N.E.2d 929, 936–37 (Mass. 2002); Paul v. Watchtower Bible & Tract Soc’y of N.Y., Inc., 819 F.2d 875, 879 (9th Cir. 1987).  It is sometimes casually said that church autonomy does not protect personal tort suits, for example, Demkovich v. St. Andrew the Apostle Par., Calumet City, 3 F.4th 968, 982 (7th Cir. 2021) (en banc), but defamation at least must be an exception. 

32See, e.g., J. Gregory Grisham & Daniel Blomberg, The Ministerial Exception After Hosanna-Tabor: Firmly Founded, Increasingly Refined, 20 Federalist Soc’y Rev. 80 (2019).

33See Kellen Funk, Church Corporations and the Conflict of Laws in Antebellum America, 32 J.L. & Religion 263, 273 (2017); Kurt T. Lash, The Second Adoption of the Establishment Clause: The Rise of the Nonestablishment Principle, 27 Ariz. St. L.J. 1085, 1111 (1995); Bernard Roberts, Note, The Common Law Sovereignty of Religious Lawfinders and the Free Exercise Clause, 101 Yale L.J. 211, 217 (1991).

34See Presbyterian Church in the U.S. v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440, 443–46 (1969).

35Compare Michael W. McConnell & Luke W. Goodrich, On Resolving Church Property Disputes, 58 Ariz. L. Rev. 307 (2016), with Richard W. Garnett, A Hands-Off Approach to Religious Doctrine: What Are We Talking About?, 84 Notre Dame L. Rev. 837 (2009).

36See Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2055 (2020); Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171,194–95 (2012).

37See, e.g., Malicki v. Doe, 814 So. 2d 347, 351 n.2, 358 (Fla. 2002) (summarizing over a dozen relevant state and federal cases).

38Rweyemamu v. Cote, 520 F.3d 198, 208 (2d Cir. 2008).

39See, e.g., Hubbard v. J Message Grp. Corp., 325 F. Supp. 3d 1198, 1214 (D.N.M. 2018).

40For example, compare the majority and dissent in Demkovich v. St. Andrew the Apostle Parish, 3 F.4th 968 (7th Cir. 2021) (en banc); see also Bilbrey v. Myers, 91 So. 3d 887, 891 (Fla. Dist. Ct. App. 2012).

41Demkovich, 3 F.4th at 968.

42Skrzypczak v. Roman Cath. Diocese of Tulsa, 611 F.3d 1238, 1245 (10th Cir. 2010).

43See Demkovich, 3 F.4th at 989 n.3 (Hamilton, J., dissenting).

44Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 196 (2012).

45See, e.g., Esther Yoon-Ji Kang & Susie An, ‘Fertile Soil for Abuse’: A Reckoning at Covenant Fellowship Church, WBEZ Chi. (August 2, 2021, 12:00 PM), [https://‌‌/BC2Z-FMPY].

46See, e.g., Turner v. Roman Cath. Diocese of Burlington, Vt., 987 A.2d 960, 972 (Vt. 2009); Jones v. Trane, 591 N.Y.S.2d 927, 931 (N.Y. Sup. Ct. 1992).

47See Alvaré, supra note 21, at 373; Paul Horwitz, Act III of the Ministerial Exception, 106 Nw. U. L. Rev. 973, 984 (2012).

48It also casts a church defendant in a better light if it can claim to lack knowledge of abuse (for example) rather than try to claim immunity from accountability.  See, e.g., Doe v. Or. Conf. of Seventh-Day Adventists, 111 P.3d 791, 797 (Or. Ct. App. 2005).

49Cf. Ex parte Young, 209 U.S. 123, 159–60 (1908).

50The abuse goes all the way back to ancient times.  See, e.g., 1 Samuel 2:22–23 (describing how priests sinned by having sexual relations with women at the tabernacle—likely taking advantage of women who either served at the tabernacle or came to worship at the tabernacle); see also C.F. Keil & F. Delitzsch, 9 Clark’s Foreign Theological Library: Biblical Commentary on the Books of Samuel 37 (James Martin trans., Edinburgh, T. & T. Clark 1872).

51The 2003 Pulitzer Prize Winner in Public Service, Pulitzer Prizes, [].

52See Laurie Goodstein & Sharon Otterman, Catholic Priests Abused 1,000 Children in Pennsylvania, Report Says, N.Y. Times (Aug. 14, 2018), https://‌‌/2018‌/08‌/14‌/us‌/catholic-church-sex-abuse-pennsylvania.html [].

53Bernard Condon & Jim Mustian, Surge of New Abuse Claims Threatens Catholic Church Like Never Before, PBS NewsHour (Dec. 3, 2019, 11:27 AM), https://‌‌/newshour‌/nation‌/surge-of-new-abuse-claims-threatens-catholic-church-like-never-before/ [].

54See, e.g., Tricia L. Nadolny, ‘The Tongue is a Fire’: Southern Baptist Church Fractures over Secrets and Spiritual Abuse, USA Today (Feb. 13, 2020, 11:57 AM), https://‌‌/in-depth‌/news‌/investigations‌/2020‌/02‌/13‌/southern-baptist-sex-abuse-pastors-history-divided-church‌/4586698002‌/ []; Janice Broach, Memphis-Based Church of God in Christ Facing Lawsuit Following Sexual Abuse Allegations, WMC Action News 5 (Feb. 3, 2020, 11:19 PM), https://‌‌/2020‌/02‌/04‌/memphis-based-church-god-christ-facing-lawsuit-following-sexual-abuse-allegations‌/ [].

55Jack Healy, ‘There Is No Excuse.’ Methodist Pastor, Accused of Sexual Harassment, Steps Down., N.Y. Times (Dec. 27, 2019, 6:59 PM), https://‌‌/2019‌/12‌/27‌/us‌/methodist-sexual-harassment-me-too.html [].

56Jeff Burlew, Eric Dudley, St. Peter’s Founder and Outspoken LGBT Critic, Subjected Men to Sexual Misconduct, Tallahassee Democrat (Nov. 26, 2019, 5:20 PM), [].

57Robert Downen, Lise Olsen & John Tedesco, 20 Years, 700 Victims: Southern Baptist Sexual Abuse Spreads as Leaders Resist Reforms, Hous. Chron.: Abuse of Faith (Feb. 10, 2019), https://‌‌/news‌/investigations‌/article‌/Southern-Baptist-sexual-abuse-spreads-as-leaders-13588038.php[].

58Kathryn Post & Bob Smietana, With Abuse Allegations, Conservative Anglican Diocese Faces Questions About Structure, Religion News Serv. (Aug. 19, 2021), https://‌‌/2021‌/08‌/19‌/acna-abuse-upper-midwest-grenhouse-allegations-bishop-ruch-structure‌/ []; see also Megan Fowler, Faced with Allegations, Anglicans Want to Change the Trajectory of Abuse Response, Christianity Today (July 27, 2021, 2:00 PM), https://‌‌/news‌/2021‌/july‌/anglican-acna-abuse-response-investigation-wheaton-midwest.html [].

59Rabbi Convicted of Sex Abuse Gets 12 Years in Prison, Associated Press (Dec. 2, 2019), https://‌‌/article‌/223937dcac8645d0a10c36d3e0a9a48c []; Susan Edelman, Brooklyn Man Accuses Hasidic Rabbis of Child Sex Abuse in Hebrew ‘Hell’, N.Y. Post (Nov. 9, 2019, 5:50 PM), https://‌‌/2019‌/11‌/09‌/brooklyn-man-accuses-hasidic-rabbis-of-child-sex-abuse-in-hebrew-hell‌/ [https://‌‌/4EBC-4Y9R].

60Alleged Victim in Illinois Imam Sex Abuse Case Speaks Out, Al Jazeera (Feb. 19, 2015, 9:30 PM), http://‌‌/articles‌/2015‌/2‌/19‌/alleged-victim-in-imam-sex-abuse-case-speaks-out.html [https://‌‌/X793-BKUV].

61See Charol Shakeshaft, U.S. Dep’t of Educ., No. 2004–09, Educator Sexual Misconduct: A Synthesis of Existing Literature 31 (2004); Charol Shakeshaft & Audrey Cohan, In Loco Parentis: Sexual Abuse of Students in Schools: What Administrators Should Know (1994); Marci A. Hamilton, Justice Denied: What America Must Do to Protect Its Children 101 (2008); Greg Toppo, Gaps in Checking Teaching Credentials Can Miss Predators, USA Today (Aug. 21, 2006, 9:11 PM), [https://‌‌/BU7L-THWW].

62See, e.g., Jessica Johnson, #Churchtoo: Apology of Evangelical Pastor Accused of Sexual Assault Shows Why Sorry Isn’t Enough, Religion Dispatches (Jan. 16, 2018), https://‌‌/churchtoo-apology-of-evangelical-pastor-accused-of-sexual-assault-shows-why-sorry-isnt-enough‌/ []; Casey Quackenbush, The Religious Community Is Speaking out Against Sexual Violence with #ChurchToo, Time (Nov. 22, 2017, 1:34 AM), https://‌‌/5034546‌/me-too-church-too-sexual-abuse‌/ []; Whitney Woollard, What the Church Can and Should Bring to the #MeToo Movement, 9Marks (Apr. 17, 2018), https://‌‌/article‌/what-the-church-can-and-should-bring-to-the-metoo-movement‌/ [].

63See, e.g., Jones v. Trane, 591 N.Y.S.2d 927, 931 (N.Y. Sup. Ct. 1992).

64See, e.g., Hazen v. Great Plains Ann. Conf. of the United Methodist Church, No. 21-4046, 2021 WL 5867217, at *2 (D. Kan. Dec. 10, 2021).

65See, e.g., Doe v. Cath. Bishop of Chi., 82 N.E.3d 1229, 1233 (Ill. App. Ct. 2017); Carson v. Canning, 62 N.E. 964, 964 (Mass. 1902); Miller v. Wal-Mart Stores, Inc., 580 N.W.2d 233, 238 (Wis. 1998); see also Cindy M. Haerle, Employer Liability for the Criminal Acts of Employees Under Negligent Hiring Theory: Ponticas v. K.M.S. Investments, 68 Minn. L. Rev. 1303, 1306–07 (1984); J. v. Victory Tabernacle Baptist Church, 372 S.E.2d 391, 394 (Va. 1988).

66See, e.g., Byrd v. Faber, 565 N.E.2d 584 (Ohio 1991).

67See, e.g., Dan B. Dobbs, Paul T. Hayden & Ellen M. Bublick, The Law of Torts § 423 (2d ed. 2022); Ehrens v. Lutheran Church, 385 F.3d 232, 235 (2d Cir. 2004); Green v. City of Mount Vernon, 96 F. Supp. 3d 263, 297 (S.D.N.Y. 2015).

68See, e.g., Doe #2 v. Norwich Roman Cath. Diocesan Corp., No. HHDX07CV125036425S, 2013 WL 3871430, at *2–3 (Conn. Super. Ct. July 8, 2013); L.L.N. v. Clauder, 563 N.W.2d 434, 443 (Wis. 1997).

69See, e.g., ; Swanson v. Roman Cath. Bishop of Portland, 692 A.2d 441, 445 (Me. 1997); Gibson v. Brewer, 952 S.W.2d 239, 247 (Mo. 1997) (en banc); see also Doe 122 v. Marianist Province of the U.S., 620 S.W.3d 73, 77–81 (Mo. 2021).

70Restatement (Third) of Agency § 7.03(2) (Am. L. Inst. 2006).

71See, e.g., Nutt v. Norwich Roman Cath. Diocese, 921 F. Supp. 66, 72–73 (D. Conn. 1995).  An alternative argument, not related to religious liberty, is that any sexual abuse was not done to advance the interests of the employer and therefore is outside the scope of employment.  See, e.g., Tichenor v. Roman Cath. Church of the Archdiocese of New Orleans, 32 F.3d 953, 960 (5th Cir. 1994); Nutt, 921 F. Supp. at 71; Kennedy v. Roman Cath. Diocese of Burlington, Vt., Inc., 921 F. Supp. 231, 233 (D. Vt. 1996).

72See supra notes 41–42 and accompanying text. 

73See Robert F. Cochran, Jr., Church Freedom and Accountability in Sexual Exploitation Cases: The Possibility of Both Through Limited Strict Liability, 21 J. Contemp. Legal Issues 427, 436–40 (2013).

74See, e.g., Redwing v. Cath. Bishop for the Diocese of Memphis, 363 S.W.3d 436 (Tenn. 2012); Pleasant Glade Assembly of God v. Schubert, 264 S.W.3d 1, 12 (Tex. 2008); C.B. v. Evangelical Lutheran Church in Am., 726 N.W.2d 127, 137 (Minn. Ct. App. 2007).  Sometimes the court stretches for a reason to not apply church autonomy, as did a Pennsylvania court in a clergy abuse case.  The church hierarchy sought to defend against a negligent hiring claim, but the court refused to hear the argument, treating it as forfeited because it had not been raised below.  Hutchison v. Luddy, 763 A.2d 826, 833 (Pa. Super. Ct. 2000), vacated on other grounds, 870 A.2d 766 (Pa. 2005).

75See Griffin, supra note 9, at 998.

76Corbin, supra note 6, at 1968.

77Marci A. Hamilton, God vs. the Gavel: The Perils of Extreme Religious Liberty 13 (2d rev. ed. 2014).

78See, e.g., Helfand, supra note 10, at 890–91 (describing the debate).

79See Horwitz, supra note 5, at 91–99; Nicholas Wolterstorff, Abraham Kuyper on the Limited Authority of Church and State, 7 Geo. J.L. & Pub. Pol’y 105 (2009); Paul Horwitz, First Amendment Institutions 181–82, 214 (2013); Schragger & Schwartzman, supra note 9, at 922–26.

80Lawrence Sager, Why Churches (and, Possibly, the Tarpon Bay Women’s Blue Water Fishing Club) Can Discriminatein The Rise of Corporate Religious Liberty 77, 78 (Micah Schwartzman, Chad Flanders & Zoë Robinson eds., 2016).

81Lupu & Tuttle, supra note 17; Ira C. Lupu & Robert W. Tuttle, #MeToo Meets the Ministerial Exception: Sexual Harassment Claims by Clergy and the First Amendment’s Religion Clauses, 25 Wm. & Mary J. Race, Gender & Soc. Just. 249 (2019).

82Lupu & Tuttle, supra note 17, at 1794.

83Cochran, supra note 73.

84Id. at 445.

85Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2055 (2020) (quoting Kedroff v. Saint Nicholas Cathedral of Russian Orthodox Church in N. Am., 344 U.S. 94, 116 (1952)).

86See, e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 188 (2012).

87See, e.g., Pleasant Glade Assembly of God v. Schubert, 264 S.W.3d 1, 12 (Tex. 2008).

88This is not to deny that virtually all of these analytical threads have a place in the analysis of church autonomy problems, a subject developed in Part IV.

89Multiple petitions for certiorari were filed in the last several years on one or another aspect of church autonomy doctrine.  So far, the Supreme Court has shown no interest in hearing any of these cases outside of the ministerial exception context, while several Justices have recently signaled interest in potentially hearing still more ministerial exception cases.  See Gordon Coll. v. DeWeese-Boyd, 142 S. Ct. 952 (2022) (Alito, J., respecting the denial of certiorari) (joined by Justices Thomas, Kavanaugh, and Barrett); Seattle’s Union Gospel Mission v. Woods, 142 S. Ct. 1094 (2022) (Alito, J., respecting the denial of certiorari) (joined by Justice Thomas).

90McRaney v. N. Am. Mission Bd. of the S. Baptist Convention, Inc., 966 F.3d 346, 349 (5th Cir. 2020).

91McRaney v. N. Am. Mission Bd. of the S. Baptist Convention, Inc., No. 17-CV-00080, 2019 WL 1810991, at *3 (N.D. Miss. Apr. 24, 2019), rev’d, 966 F.3d 346 (5th Cir. 2020).

92McRaney, 966 F.3d at 349.

93Bilbrey v. Myers, 91 So. 3d 887, 891 (Fla. Dist. Ct. App. 2012) (citing Malicki v. Doe, 814 So. 2d 347, 358 (Fla. 2002)). 

94In re Diocese of Lubbock, 624 S.W.3d 506, 513 (Tex. 2021) (citing Westbrook v. Penley, 231 S.W.3d 389, 398–400 (Tex. 2007), cert. denied, 142 S. Ct. 434 (2021)).

95Id. at 531 (Boyd, J., dissenting).

96In re Roman Cath. Diocese of El Paso, 626 S.W.3d 36, 43 (Tex. App. 2021); see also Bixler v. Superior Court, No. B310559, 2022 WL 167792, at *13 (Cal. Ct. App. Jan. 19, 2022); de Laire v. Voris, No. 21-cv-131, 2021 WL 5860723, at *4 (D.N.H. Dec. 9, 2021).

97Gibson v. Brewer, 952 S.W.2d 239, 246–47 (Mo. 1997) (en banc).

98See, e.g., Carson ex rel. O.C. v. Makin, 142 S. Ct. 1987, 1998 (2022); McCreary Cnty. v. ACLU of Ky., 545 U.S. 844 (2005); Epperson v. Arkansas, 393 U.S. 97, 104 (1968); John Witte, Jr., Joel A. Nichols & Richard W. Garnett, Religion and the American Constitutional Experiment 170–205 (5th ed. 2022); Thomas B. Colby, A Constitutional Hierarchy of Religions?  Justice Scalia, the Ten Commandments, and the Future of the Establishment Clause, 100 Nw. U. L. Rev. 1097, 1099–1101 (2006); Lael Weinberger, Carson v. Makin and the Relativity of Religious Neutrality, 20 Harv. J.L. & Pub. Pol’y Per Curiam (2022). 

99Jones v. Wolf, 443 U.S. 595, 603 (1979); Presbyterian Church in the U.S. v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440, 449 (1969).

100Erdman v. Chapel Hill Presbyterian Church, 286 P.3d 357, 367 (Wash. 2012).  Whether and when church bylaws should count as the kind of documents which can be interpreted by “neutral principles” has posed challenging questions for courts.  See, e.g., Taylor v. Evangelical Covenant Church, 2022 IL 1-21-210524, ¶ 22 (Ill. App. Ct. 2022); In re Thomas, No. 06-21-00106-CV, 2022 WL 126708, at *9–10 (Tex. App. Jan. 14, 2022).

101494 U.S. 872 (1990).

102Turner v. Roman Cath. Diocese of Burlington, Vt., 987 A.2d 960, 972 (Vt. 2009).

103Id. at 973.


105See Michael A. Helfand, What Is a “Church”?: Implied Consent and the Contraception Mandate, 21 J. Contemp. Legal Issues 401, 413 (2013); Marci A. Hamilton, Religious Institutions, the No-Harm Doctrine, and the Public Good, 2004 BYU L. Rev. 1099, 1194–95.

106Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 190 (2012).

107Mark Chopko, Regulating Religious Charity: Current Issues and Future Challenges, 44 Rutgers L.J. 55, 80 (2013) (“The Hosanna-Tabor court distinguished Smith in a most superficial way.”); Mike Dorf, Ministers and Peyote, Dorf on Law (Jan. 12, 2012), http://‌‌/2012‌/01‌/ministers-and-peyote.html [https://‌‌/5F9M-YA8H] (“With due respect: huh???”); Michael W. McConnell, Reflections on Hosanna-Tabor, 35 Harv. J.L. & Pub. Pol’y 821, 835 (2012) (noting that the treatment of Smith “create[s] uncertainty”); Helfand, supra note 10, at 888. 

108For a defense of Hosanna-Tabor’s analysis of Smith, see Douglas Laycock, Hosanna-Tabor and the Ministerial Exception, 35 Harv. J.L. & Pub. Pol’y 839, 854–58 (2012).

109See, e.g., Corbin, supra note 6, at 1981–2004; Griffin, supra note 9, at 992.

110Paul G. Kauper, Church Autonomy and the First Amendment: The Presbyterian Church Case, 1969 Sup. Ct. Rev. 347, 375.

111Esbeck, Validations and Ramifications, supra note 17; Lupu & Tuttle, supra note 17, at 1808 n.58; Esbeck, Governmental Power, supra note 17, at 42–51; see also Michael A. Helfand, Religion’s Footnote Four: Church Autonomy as Arbitration, 97 Minn. L. Rev. 1891, 1901 (2013) (arguing for “reorienting our religion clause jurisprudence away from the structural and jurisdictional limitations we place on courts and towards the autonomy and authority we grant religious institutions”).  But seeLaycock, supra note 15, at 263–64 (arguing that church autonomy need not rely on the Establishment Clause for its structural character).   

112See, e.g., Lee v. Sixth Mount Zion Baptist Church of Pittsburgh, 903 F.3d 113, 118 n.4 (3d Cir. 2018); Conlon v. InterVarsity Christian Fellowship/USA, 777 F.3d 829, 836 (6th Cir. 2015); Napolitano v. St. Joseph Cath. Church, 308 So. 3d 274, 275 (Fla. Dist. Ct. App. 2020); Roman Cath. Diocese of Jackson v. Morrison, 905 So. 2d 1213, 1250 (Miss. 2005) (Smith, C.J., dissenting); Carl H. Esbeck, An Extended Essay on Church Autonomy, 22 Federalist Soc’y Rev. 244 (2021).

113See, e.g., Richard W. Garnett, Do Churches Matter?  Towards an Institutional Understanding of the Religion Clauses, 53 Vill. L. Rev. 273, 274 (2008).

114See, e.g., Watson v. Jones, 80 U.S. (13 Wall.) 679, 733 (1872); Hollins v. Methodist Healthcare, Inc., 474 F.3d 223, 225 (6th Cir. 2007).

115See, e.g., Bilbrey v. Myers, 91 So. 3d 887, 890–91 (Fla. Dist. Ct. App. 2012); Church of God in Christ, Inc. v. L. M. Haley Ministries, Inc., 531 S.W.3d 146 (Tenn. 2017); In re Roman Cath. Diocese of El Paso, 626 S.W.3d 36 (Tex. App. 2021) .

116See Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 195 n.4 (2012).

117See also, to similar effect, EEOC v. Catholic University of America, 83 F.3d 455, 462 (D.C. Cir. 1996).

118Jones v. Wolf, 443 U.S. 595 (1979); Presbyterian Church in the U.S. v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440 (1969).

119See Watson, 80 U.S. (13 Wall.) at 690–91.  On the case’s history, see Eric G. Osborne & Michael D. Bush, Rethinking Deference: How the History of Church Property Disputes Calls into Question Long-Standing First Amendment Doctrine, 69 SMU L. Rev. 811, 817–821 (2016).

120See, e.g.Jones, 443 U.S. at 598–99.

121See, at 602.

122See Lash, supra note 33, at 1111–12; McConnell & Goodrich, supra note 35, at 312–13.

123See Presbyterian Church, 393 U.S. at 440, 444.

124See id. at 449–50.

125Att’y Gen. ex rel. Mander v. Pearson (1817) 36 Eng. Rep. 135, 156–57; 3 Mer. 353, 418–19.


127See, e.g., Miller v. Gable, 2 Denio 492 (N.Y. 1845); see also Funk, supra note 33, at 273–78.

128Jones v. Wolf, 443 U.S. 595, 604–05 (1979).

129Watson v. Jones, 80 U.S. (13 Wall.) 679 (1872).

130See Kedroff v. Saint Nicholas Cathedral of the Russian Orthodox Church in N. Am., 344 U.S. 94, 116 (1952).

131See Watson, 80 U.S. (13 Wall.) at 681–85.

132See id. at 727.

133See Presbyterian Church in U.S. v. Mary Elizabeth Blue Hull Mem’l Presbyterian Church, 393 U.S. 440, 449–50 (1969).

134See supra note 31.

135See, e.g., Esbeck, supra note 112, at 262.

136See Jones v. Wolf, 443 U.S. 595, 602–03 (1979).

137See Erdman v. Chapel Hill Presbyterian Church, 286 P.3d 357, 376 (Wash. 2012) (Chambers, J., dissenting in part and concurring in part).

138See subsection IV.B.3, infra.

139See, e.g., Bilbrey v. Myers, 91 So. 3d 887, 891 (Fla. Dist. Ct. App. 2012).

140See Idleman, supra note 8, at 262.

141See Tandon v. Newsom, 141 S. Ct. 1294 (2021); Fulton v. City of Philadelphia, 141 S. Ct. 1868 (2021); see also Lael Weinberger, Religious Liberty, Exceptions, and Targeting, Nat’l Rev. (July 16, 2021, 12:33 PM), https://‌‌/bench-memos‌/religious-liberty-exceptions-and-targeting‌/ [].

142See, e.g., Bryce v. Episcopal Church in the Diocese of Colo., 289 F.3d 648, 656 (10th Cir. 2002).

143See, e.g., Erdman v. Chapel Hill Presbyterian Church, 286 P.3d 357, 368 (Wash. 2012).

144See McRaney v. N. Am. Mission Bd. of the S. Baptist Convention, Inc., 966 F.3d 346, 349 (5th Cir. 2020).

145See Bilbrey v. Myers, 91 So. 3d 887, 890–91 (Fla. Dist. Ct. App. 2012).

146See Jennison v. Prasifka, 391 S.W.3d 660, 665–66 (Tex. App. 2013); Kelly v. St. Luke Cmty. United Methodist Church, No. 05-16-01171-CV, 2018 WL 654907, at *8 (Tex. App. Feb. 1, 2018).

147See, e.g., Welter v. Seton Hall Univ., 608 A.2d 206, 217 (N.J. 1992).

148See Idleman, supra note 8, at 261–66.

149See El-Farra v. Sayyed, 226 S.W.3d 792, 795 (Ark. 2006); see also Heard v. Johnson, 810 A.2d 871, 880 (D.C. 2002).

150Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 194–95 (2012); Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2069 (2020); see also Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 946 (9th Cir. 1999); Combs v. Cent. Tex. Ann. Conf. of the United Methodist Church, 173 F.3d 343, 350 (5th Cir. 1999); Corbin, supra note 6, at 1975.

151Rweyemamu v. Cote, 520 F.3d 198, 208 (2d Cir. 2008).

152Gaddy v. Corp. of the President of the Church of Jesus Christ of Latter-Day Saints, 551 F. Supp. 3d 1206, 1221 (D. Utah 2021).

153Martinelli v. Bridgeport Roman Cath. Diocesan Corp., 196 F.3d 409, 431 (2d Cir. 1999).

154See, e.g., Jeremy Weese, Comment, The (Un)Holy Shield: Rethinking the Ministerial Exception, 67 UCLA L. Rev. 1320 (2020).

155See, e.g., Rayburn v. Gen. Conf. of Seventh-Day Adventists, 772 F.2d 1164, 1169 (4th Cir. 1985); Bollard, 196 F.3d at 946.

156See Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 194–95 (2012); see also Shelly Aviv Yeini, The Ministerial Exception: Our Lady of Guadalupe School and Antidiscrimination Employment Laws, 54 Vand. J. Transnat’l L. 955, 990–91 (2021).

157Bell v. Presbyterian Church (U.S.A.), 126 F.3d 328, 331 (4th Cir. 1997) (emphasis added) (quoting Serbian E. Orthodox Diocese for the U.S. & Can. v. Milivojevich, 426 U.S. 696, 713 (1976)); Bryce v. Episcopal Church in the Diocese of Colo., 289 F.3d 648, 657 (10th Cir. 2002) (quoting Bell, 126 F.3d at 331). 

15877 C.J.S. Religious Societies § 121 (2022).

159See Combs v. Cent. Tex. Ann. Conf. of the United Methodist Church, 173 F.3d 343, 350 (5th Cir. 1999).

160See, e.g., St. Brendan High Sch., Inc. v. Neff, 283 So. 3d 399, 402 (Fla. Dist. Ct. App. 2019).

161Fratello v. Archdiocese of N.Y., 863 F.3d 190, 197 & n.15, 203 (2d Cir. 2017) (“Judges are not well positioned to determine whether ministerial employment decisions rest on practical and secular considerations or fundamentally different ones that may lead to results that, though perhaps difficult for a person not intimately familiar with the religion to understand, are perfectly sensible—and perhaps even necessary—in the eyes of the faithful.  In the Abrahamic religious traditions, for instance, a stammering Moses was chosen to lead the people, and a scrawny David to slay a giant.”).

162Bryce, 289 F.3d at 657 (quoting Wisconsin v. Yoder, 406 U.S. 205, 215 (1972)); see also Billard v. Charlotte Cath. High Sch., No. 17-cv-00011, 2021 WL 4037431, at *11 (W.D.N.C. Sept. 3, 2021) (quoting Bryce, 289 F.3d at 657), appeal filed, No. 22-1440 (4th Cir. 2022).

163Rweyemamu v. Cote, 520 F.3d 198, 208, 208–09 (2d Cir. 2008).

164See, e.g., Demkovich v. St. Andrew the Apostle Par., 3 F.4th 968, 980 (7th Cir. 2021).

165See, e.g., Elvig v. Calvin Presbyterian Church, 375 F.3d 951 (9th Cir. 2004) (quoting Nichols v. Azteca Rest. Enters., 256 F.3d 864, 877 (9th Cir. 2001)); Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 944–45 (9th Cir. 1999).  But see Werft v. Desert Sw. Ann. Conf. of the United Methodist Church, 377 F.3d 1099, 1101–02, 1104 (9th Cir. 2004) (church autonomy defense applied to an employment accommodation claim); Alcazar v. Corp. of the Cath. Archbishop of Seattle, 627 F.3d 1288, 1290–91 (9th Cir. 2010) (ministerial exception defense applied to a wage and hour claim). 

166See Elvig, 375 F.3d at 961–62. 

167See Skrzypczak v. Roman Cath. Diocese of Tulsa, 611 F.3d 1238, 1245–46 (10th Cir. 2010); Demkovich, 3 F.4th at 984–85.

168See Skrzypczak, 611 F.3d at 1245; see also Winnie Johnson, Comment, A Balancing Act: Hostile Work Environment and Harassment Claims by Ministerial Employees, 96 Tul. L. Rev. 193, 220 (2021).

169See Gibson v. Brewer, 952 S.W.2d 239, 247–48 (Mo. 1997) (en banc); accord Schultz v. Roman Cath. Archdiocese of Newark, 472 A.2d 531, 546 (N.J. 1984).

170Gibson, 952 S.W.2d at 249.

171Id. at 246; see also Jones v. Trane, 591 N.Y.S.2d 927, 932 (N.Y. Sup. Ct. 1992); Byrd v. Faber, 565 N.E.2d 584, 590 (Ohio 1991).

172Doe v. First Presbyterian Church U.S.A. of Tulsa, Okla., 421 P.3d 284, 289 (Okla. 2017); see also Guinn v. Church of Christ of Collinsville, 775 P.2d 766, 775 (Okla. 1989).

173Hiles v. Episcopal Diocese of Mass., 773 N.E.2d 929, 937 n.12 (Mass. 2002) (emphasis added).

174Ogle v. Church of God, 153 F. App’x 371 (6th Cir. 2005); Ogle v. Hocker, 279 F. App’x 391, 396 (6th Cir. 2008).

175Church of God, 153 F. App’x at 373; Hocker, 279 F. App’x at 392–93.

176Church of God, 153 F. App’x at 373; Hocker, 279 F. App’x at 393.

177Hocker, 279 F. App’x at 393.

178Church of God, 153 F. App’x at 376.

179Hocker, 279 F. App’x at 396.


181See Christopher C. Lund, Free Exercise Reconceived: The Logic and Limits of Hosanna-Tabor, 108 Nw. U. L. Rev. 1183, 1203 (2014); see also Lael Weinberger, Why American Courts Care About Church Membership—and Why You Should, Too, 9Marks (July 2, 2021), https://‌‌/article‌/why-american-courts-care-about-church-membership-and-why-you-should-too‌/ [].

182Lund, supra note 181, at 1203; see also Bixler v. Superior Ct., No. B310559, 2022 WL 167792, at *10 (Cal. Ct. App. Jan. 19, 2022); Guinn v. Church of Christ of Collinsville, 775 P.2d 766, 777 (Okla. 1989); Helfand, supra note 10, at 906–07.

183See Guinn, 775 P.2d at 786; Helfand, supra note 10, at 907.

184See 1 Corinthians 5; see also Matthew 18:15–17 (directing public rebuke in front of the church body for unrepentant sin after unsuccessful private attempts at resolution).

185See, e.g., Helfand, supra note 111, at 1901–02; Sager, supra note 80, at 78.

186See Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 199 (2012) (Alito, J., concurring) (“Throughout our Nation’s history, religious bodies have been the preeminent example of private associations that have ‘act[ed] as critical buffers between the individual and the power of the State.’” (quoting Roberts v. U.S. Jaycees, 468 U.S. 609, 619 (1984))).

187See Richard W. Garnett, Religion and Group Rights: Are Churches (Just) Like the Boy Scouts?, 22 St. John’s J. Legal Comment. 515, 523 (2007).

188See Michael W. McConnell, Religion and Constitutional Rights: Why Is Religious Liberty the “First Freedom”?, 21 Cardozo L. Rev. 1243, 1246 (2000) (stating that the notion of two kingdoms is the “most powerful possible refutation of the notion that the political sphere is omnicompetent”).

189See, e.g., Steven D. Smith, Freedom of Religion or Freedom of the Church?in Legal Responses to Religious Practices in the United States: Accommodation and its Limits 249, 266–69 (Austin Sarat ed., 2012).

190That church and state are to be preserved as separate institutions is, plausibly, the minimum necessary content for the First Amendment’s religion clauses. 

191See, e.g.Hosanna-Tabor, 565 U.S. at 196; Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 946 (9th Cir. 1999); see also Corbin, supra note 6, at 1974 n.45.

192See, e.g., Helfand, supra note 10, at 895–96.  In a case decided two years after Wisconsin v. Yoder, 406 U.S. 205 (1972), one of the Supreme Court’s classic balancing-test decisions, the Fifth Circuit suggested that a compelling government interest analysis could be employed in a church autonomy case.  See Simpson v. Wells Lamont Corp., 494 F.2d 490, 493 (5th Cir. 1974).  The court cited Yoder, and the balancing language in this case could not remain good law. 

193See Laycock, supra note 15, at 274.

194See Stephanie H. Barclay, First Amendment “Harms, 95 Ind. L.J. 331, 375–76 (2020). 

195For criticism of balancing tests, see Joel Alicea & John D. Ohlendorf, Against the Tiers of Constitutional Scrutiny, Nat’l Affairs (2019), https://‌‌/publications‌/detail‌/against-the-tiers-of-constitutional-scrutiny/ [https://‌‌/KL2S-AZH7].

196See supra Section I.A.

197See supra notes 22–25 and accompanying text.

198See, e.g., Schragger & Schwartzman, supra note 9, at 946 (“The strong form of sphere sovereignty claims that churches have a special, unique, and exclusive mission . . . . [S]tated in its baldest form, it seems to countenance very few limits on church immunity.”); Lupu & Tuttle, supra note 10, at 1302; Helfand, supra note 10, at 893 (“To the extent the sovereignty approach to religious institutions derives the authority of such institutions from an inherent sovereignty that stands beyond the authority of the state, it is difficult to see what legal principles could be marshaled to articulate limits on religious institutional authority.”).

199See Horwitz, supra note 5; Wolterstorff, supra note 79; Jonathan Chaplin, Toward a Social Pluralist Theory of Institutional Rights, 3 Ave Maria L. Rev. 147 (2005).

200See Abraham KuyperSphere Sovereigntyin Abraham Kuyper: A Centennial Reader 461, 466–67 (James D. Bratt ed., 1998).  On the history of the concept, see Peter S. Heslam, Creating a Christian Worldview: Abraham Kuyper’s Lectures on Calvinism 154–60 (1998); James D. Bratt, Abraham Kuyper: His World and Workin Abraham Kuyper: A Centennial Reader, supra, at 1, 7–16; Johan D. van der Vyver, The Jurisprudential Legacy of Abraham Kuyper and Leo XIII, 5 J. Mkts. & Morality 211, 212–13 (2002); Lael Daniel Weinberger, The Relationship Between Sphere Sovereignty and Subsidiarityin Global Perspectives on Subsidiarity 49, 50–53 (Michelle Evans & Augusto Zimmermann eds., 2014).

201See Wolterstorff, supra note 79, at 111.

202See Kuypersupra note 200, at 466.

203See G. Groen van Prinsterer, Unbelief and Revolutionin Political Order and the Plural Structure of Society 53, 65–74 (James W. Skillen & Rockne W. McCarthy eds., 1991).

204Kuypersupra note 200, at 466.

205Abraham Kuyper, The Antirevolutionary Programin Political Order and the Plural Structure of Society, supra note 203, at 235, 241; Henk E.S. Woldring, Multiform Responsibility and the Revitalization of Civil Societyin Religion, Pluralism, and Public Life: Abraham Kuyper’s Legacy for the Twenty-First Century 175, 175–88 (Luis E. Lugo ed., 2000).

206See Kuyper, supra note 205, at 259–60; Richard J. Mouw, Some Reflections on Sphere Sovereigntyin Religion, Pluralism, and Public Life: Abraham Kuyper’s Legacy for the Twenty-First Century, supra note 205, at 87, 91.  For attempts to elaborate Kuyper’s spheres in the context of a comprehensive social theory, see generally Herman Dooyeweerd, A Christian Theory of Social Institutions (John Witte, Jr. ed., Magnus Verbrugge trans., 1986); Herman Dooyeweerd, The Christian Idea of the State, Essays in Legal, Social, and Political Philosophy 121, 146–47 (John Kraay trans., 1996). 

207Kuyper, supra note 205, at 259–60.

208See Wolterstorff, supra note 79, at 113.

209See Mark J. Larson, Abraham Kuyper, Conservatism, and Church and State 48–59 (2015).

210See, e.g., critiques of “freedom of the church” rhetoric in Andrew Koppelman, “Freedom of the Church” and the Authority of the State, 21 J. Contemp. Legal Issues 145 (2013); Schragger & Schwartzman, supra note 9, at 932–39.

211See, e.g., Abraham Kuyper, Calvinism: Six Lectures Delivered in the Theological Seminary at Princeton 168, 187 (N.Y., Fleming H. Revell Co. 1899) (discussing the Roman Catholic Church’s subjugation of European universities).

212Id. at 140.

213See Gary L. Steward, Justifying Revolution: The American Clergy’s Argument for Political Resistance, 1750–1776, at 1–2 (2021) (discussing how American clergy viewed civil resistance as “a moral duty” justified by the Protestant faith).

214See id. at 10.  See generally 2 Political Sermons of the American Founding Era, 1730–1805 (Ellis Sandoz ed., 2d ed. 1998) (collecting political sermons from the founding era, frequently discussing resistance and political responsibilities).

215See Francis X. Rocca, Bishops Advance Effort That Could Restrict Communion for Biden, Wall St. J. (June 18, 2021, 7:58 PM), https://‌‌/articles‌/bishops-vote-to-prepare-document-on-communion-for-catholic-politicians-who-support-abortion-rights-11624041010/ []; Ian Lovett & Francis X. Rocca, Catholic Bishops Drop Effort to Ban Communion for Politicians Who Support Abortion Rights, Wall St. J. (Nov. 17, 2021, 5:14 PM),‌/articles‌/catholic-bishops-drop-effort-to-ban-communion-for-politicians-who-support-abortion-rights-11637169752/ [https://‌‌/3HPL-EXJJ].

216Kuypersupra note 211, at 104–05.

217Id. at 105.


219These two are recognized as sacraments or ordinances in the Reformed tradition.  See The Westminster Confession of Faith ch. XXVII (Marcus Dods & Alexander White eds., Edinburgh, T. & T. Clark 1881) (1647) (using the term “sacraments”); A Confession of Faith [Second London Baptist Confession] ch. XXVIII (1677) (using the term “ordinances”).  The Roman Catholic tradition recognizes seven sacraments.  Catechism of the Catholic Church § 1210 (2d ed. 1994).

220In extreme cases, this can be disruptive engagement, as in the case of civil disobedience.  See, e.g., Martin Luther King, Jr., Letter from Birmingham Jail, 80 Christian Century 767, 767 (1963).

221See, e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 182–84 (2012); McRaney v. N. Am. Mission Bd. of the S. Baptists Convention, Inc., 980 F.3d 1066, 1076–78 (5th Cir. 2020) (Oldham, J., dissenting from the denial of rehearing en banc); InterVarsity Christian Fellowship‌/USA v. Bd. of Governors of Wayne State Univ., 534 F. Supp. 3d 785, 802 (E.D. Mich. 2021).

222See, e.g., Richard W. Garnett, The Freedom of the Church, 4 J. Cath. Soc. Thought 59 (2007); Richard W. Garnett, “The Freedom of the Church”: (Towards) An Exposition, Translation, and Defense, 21 J. Contemp. Legal Issues 33 (2013); Steven D. Smith, The Rise and Decline of American Religious Freedom 31–34 (2014); Garnett, supra note 187, at 524; Bradley, supra note 25, at 1085–86. 

223See Hosanna-Tabor, 565 U.S. at 183.

224See generally Mark Douglas McGarvie, One Nation Under Law: America’s Early National Struggles to Separate Church and State (2004) (arguing that church disestablishment in early America was characterized by transition from public to private institutional status); Disestablishment and Religious Dissent: Church-State Relations in the New American States, 1776–1833 (Carl H. Esbeck & Jonathan J. Den Hartog eds., 2019) (surveying the process of disestablishment in early America on state-by-state basis).

225For critiques of the deep history analysis, see Paul Horwitz, Freedom of the Church Without Romance, 21 J. Contemp. Legal Issues 59 (2013); Steven K. Green, The Mixed Legacy of Magna Carta for American Religious Freedom, 32 J.L. & Religion 207 (2017); Koppelman, supra note 210, at 149–52; Schragger & Schwartzman, supra note 9, at 932–39; see also N.Y. State Rifle & Pistol Ass’n v. Bruen, 142 S. Ct. 2111, 2136 (2022) (“English common-law practices and understandings at any given time in history cannot be indiscriminately attributed to the Framers of our own Constitution.”).

226See supra note 206.

227See Sarah Barringer Gordon, The First Disestablishment: Limits on Church Power and Property Before the Civil War, 162 U. Pa. L. Rev. 307 (2014); Sarah Barringer Gordon, The African Supplement: Religion, Race, and Corporate Law in Early National America, 72 Wm. & Mary Q. 385 (2015); Funk, supra note 33.

228For a contrasting interpretation, see Hamilton, supra note 105, at 1122–35 (suggesting that the story of benefit of clergy suggests that church autonomy in general is not well grounded in legal history); Griffin, supra note 9, at 988 (same).

229Gelasius, Duo Sunt (494), reprinted in Church and Empire 107, 107–08 (Maria E. Doerfler & George Kalantzis eds., 2016).

230See 1 George Klosko, History of Political Theory: An Introduction 260 (2d ed. 2012); Phil Booth, Crisis of Empire: Doctrine and Dissent at the End of Late Antiquity 227 (2014); Lester L. Field, Jr., Liberty, Dominion, and the Two Swords: On the Origins of Western Political Theology (180–398), at 259–60 (1998).

231See, e.g., Smithsupra note 222, at 31–34.

2321 C.W. Previté-Orton, The Shorter Cambridge Medieval History 489–91 (1952).

233See Harold J. Berman, Law and Revolution: The Formation of the Western Legal Tradition 256–57 (1983); John Gillingham, The Early Middle Ages(1066–1290)in The Oxford Illustrated History of Britain 104, 125 (Kenneth O. Morgan ed., 1984); Arthur R. Hogue, Origins of the Common law 38–44 (Liberty Fund 1986) (1966).

234See, e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 182 (2012); cf. R.H. Helmholz, The Church and Magna Carta, 25 Wm. & Mary Bill Rts. J. 425 (2016).

235See J. Earll Miller, A History of Benefit of Clergy in England: With Special Reference to the Period Between 1066 and 1377, at iii (May 17, 1917) (Ph.D. dissertation, University of Illinois), [].

236See Robert E. Rodes, Jr., Lay Authority and Reformation in the English Church: Edward I to the Civil War 31 (1982); 1 Frederick Pollock & Frederic William Maitland, The History of English Law Before the Time of Edward I 441 (2d ed., Boston, Little, Brown, & Co. 1899); John G. Bellamy, Criminal Law and Society in Late Medieval and Tudor England 115 (1984).

2371 Pollock & Maitlandsupra note 236, at 441–42. 

238Id. at 442.

2392 Pollock & Maitlandsupra note 236, at 466.

2401 Pollock & Maitlandsupra note 236, at 444.

241See id. at 443; Helen M. Cam, 46 Eng. Hist. Rev. 125, 126 (1931) (book review); Newman F. Baker, Benefit of Clergy—a Legal Anomaly, 15 Ky. L.J. 85, 86 (1927); George W. Dalzell, Benefit of Clergy in America & Related Matters 11 (1955).

2421 Pollock & Maitlandsupra note 236, at 443.

2434 William Blackstone, Commentaries *365.

244Id.; see also 1 Pollock & Maitlandsupra note 236, at 443–44; Rodessupra note 236, at 31.

2454 Blackstonesupra note 243, at *365.

246See 1 Pollock & Maitlandsupra note 236, at 443.

247See Candace Gregory-Abbott, Sacred Outlaws: Outlawry and the Medieval Churchin Outlaws in Medieval and Early Modern England: Crime, Government and Society, c.1066–c.1600, at 75, 78 (John C. Appleby & Paul Dalton eds., 2009); 1 Pollock & Maitlandsupra note 236, at 444–45. 

248See Jo Ann Hoeppner Moran, Clerical Recruitment in the Diocese of York, 1340–1530: Data and Commentary, 34 J. Ecclesiastical Hist. 19, 24, 35 (1983).

249See Gregory-Abbott, supra note 247, at 78.

2504 Blackstone, supra note 243, at *365 (emphasis omitted); see also Baker, supra note 241, at 85–86.

251See 4 Blackstone, supra note 243, at *365. 

252See Bellamy, supra note 236, at 115–64.

253Rodessupra note 236, at 33.

2544 Blackstone, supra note 243, at *366–67.

255France had similar clerical immunities and there are records of painstaking inspections of a prisoner’s head to evaluate when exactly a tonsure was given, in an attempt to identify pretenders to the clerical office.  See Henry C. Lea, Studies in Church History: The Rise of the Temporal Power.—Benefit of Clergy.—Excommunication. 203 (London, Sampson Low, Son & Marston 1869). 

256Bellamy, supra note 236, at 116.

257See, e.g., Norman F. Hesseltine, Benefit of Clergy—a Historical Curiosity, 16 Mass. L.Q. 73, 73 (1931); Emily Steiner, William LanglandinThe Cambridge Companion to Medieval English Law and Literature 121, 131–32 (Candace Barrington & Sebastian Sobecki eds., 2019).

258See John Hostettler, A History of Criminal Justice in England and Wales 4445 (2009); 1 James Fitzjames Stephen, A History of the Criminal Law of England 464 (London, MacMillan & Co. 1883).

259See, e.g., Rodessupra note 236, at 32; Hostettler, supra note 258, at 45.

260See Gregory-Abbott, supra note 247, at 78.  For a more acerbic version of this point, see Jeremy Bentham, The Rationale of Punishment 376 (Richard Smith ed. & trans., London, Robert Heward 1830). 

261See Gregory-Abbott, supra note 247, at 84.

262See, e.g., John Baker, Introduction to English Legal History 140 (5th ed. 2019); Gordon K. McBride, Once Again, the Case of Richard Hunne, 1 Albion 19, 19–20 (1969).

263See, e.g., Shannon McSheffrey, Whoring Priests and Godly Citizens: Law, Morality, and Clerical Sexual Misconduct in Late Medieval Londonin Local Identities in Late Medieval and Early Modern England 50, 55–56 (Norman L. Jones & Daniel Woolf eds., 2007); Rodessupra note 236, at 32; C.B. Firth, Benefit of Clergy in the Time of Edward IV, 32 Eng. Hist. Rev. 175, 186 (1917).

264Miller, supra note 235, at 53.

265See Avrom Saltman, Theobald: Archbishop of Canterbury 124–25 (1956); Miller, supra note 235, at 54; James W. Alexander, The Becket Controversy in Recent Historiography, 9 J. Brit. Stud. 1, 3 (1970).

266See Bermansupra note 233, at 255–69; C.R. Cheney, The Punishment of Felonous Clerks, 51 Eng. Hist. Rev. 215, 222 (1936); F.W. Maitland, Henry II and the Criminous Clerks, 7 Eng. Hist. Rev. 224, 224–226 (1892).

267See Cheney, supra note 266, at 219. 

268See, e.g., McBride, supra note 262, at 19; Gregory-Abbott, supra note 247, at 75–89.

269See McSheffrey, supra note 263, at 52–53.

270See Bellamy, supra note 236, at 117.

2714 Blackstone, supra note 243, at *366.

272At least as to the benefit of clergy rules.  The Reformation’s broader impact on England’s system of ecclesiastical courts was complex and not always direct.  See R.H. Helmholz, Conflicts Between Religious and Secular Law: Common Themes in the English Experience, 1250–1640, 12 Cardozo L. Rev. 707, 710 (1991); Rodessupra note 236.

273See Mary J. Streufert, Contemporary Theology and Church Life, Influence onin 1 Encyclopedia of Martin Luther and the Reformation 161, 162 (Mark A. Lamport ed., 2017); Alister E. McGrath, Christianity’s Dangerous Idea: The Protestant Revolution—A History from the Sixteenth Century to the Twenty-First 52–53 (2007).

274See Lee Hardy, The Fabric of This World: Inquiries into Calling, Career Choice, and the Design of Human Work 45–63 (1990).

275See John Witte, Jr., The Reformation of Rights: Law, Religion, and Human Rights in Early Modern Calvinism 43–80 (2007). 

276John Calvin, Commentaries on the Epistle of Paul the Apostle to the Romans 480 (John Owen ed. & trans., Edinburgh, T. Constable 1849). 



279See 2 John Calvin, Institutes of the Christian Religion bk. 4, ch. 20 (Henry Beveridge trans., Edinburgh, The Edinburgh Printing Co. 1845) (1536).

280Jung-Sook Lee, Excommunication and Restoration in Calvin’s Geneva, 1555–1556, at 73–81 (May 1997) (Ph.D. dissertation, Princeton Theological Seminary) (ProQuest).

281See Mirjam van Veen, “Supporters of the Devil”: Calvin’s Image of the Libertines, 40 Calvin Theological J. 21, 27 (2005).


283Jean-Pierre Cavaillé, Libertine and Libertinism: Polemic Uses of the Terms in Sixteenth- and Seventeenth-Century English and Scottish Literature, 12 J. Early Mod. Cultural Stud. 12, 12 (2012).

284See Robert J. Renaud & Lael D. Weinberger, A Tale of Two Governments: Church Discipline, the Courts, and the Separation of Church and State 115 (2012).

285Thomas H. Dyer, The Life of John Calvin 370–71 (London, John Murray 1850).

286See John Witte Jr., From Sacrament to Contract: Marriage, Religion, and Law in the Western Tradition 5 (2d ed. 2012); Rodessupra note 236, at 18–19; Helmholz, supra note 272, at 711.

287Wittesupra note 286, at 6.

288See E. Brooks Holifield, Peace, Conflict, and Ritual in Puritan Congregations, 23 J. Interdisc. Hist. 551, 560 (1993); Baker, supra note 262, at 138, 517, 528.

2894 Blackstone, supra note 243, at *369.



292Lesley Skousen, Redefining Benefit of Clergy During the English Reformation: Royal Prerogative, Mercy, and the State 5–6 (2008) (M.A. Thesis, University of Wisconsin), [].

293See Bellamy, supra note 236, at 144–45; Skousen, supra note 292, at 5–6.  For context, see R.H. Helmholz, Roman Canon Law in Reformation England 159 (1990).

2941 Stephen, supra note 258, at 465.

2954 Blackstone, supra note 243, at *369.

296Jeffrey K. Sawyer, “Benefit of Clergy” in Maryland and Virginia, 34 Am. J. Legal Hist. 49, 53 (1990); see also Bellamy, supranote 236, at 129–150, 154–163.

297J.A. Sharpe, Crime in Seventeenth-Century England: A County Study 24 (1983).

298See Sawyer, supra note 296, at 53.

2994 Blackstone, supra note 243, at *370; Sawyer, supra note 296, at 54; Sharpesupra note 297, at 24; Baker, supra note 262, at 102.

300See, e.g., Dalzellsupra note 241, at 12–13.

3014 Blackstonesupra note 243, at *430.

302This was the doctrine’s form at the time it made its way to America, with minor modifications.  See generally Sawyer, supra note 296; Dalzellsupra note 241; George C. Thomas III, Colonial Criminal Law and Procedure: The Royal Colony of New Jersey 1749–57, 1 N.Y.U. J.L. & Liberty 671, 696–99 (2005).

303This is a point that both advocates of religious liberty and critics of the current scope of religious liberty can agree upon.  Compare Michael W. McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 Harv. L. Rev. 1409, 1456–57 (1990) [hereinafter McConnell, Origins], with Marci A. Hamilton, The “Licentiousness” in Religious Organizations and Why It Is not Protected Under Religious Liberty Constitutional Provisions, 18 Wm. & Mary Bill Rts. J. 953, 968–76 (2010).  The state constitutional provisions excluding licentiousness conduct are important pieces of evidence in the debate about whether the Free Exercise Clause institutes a regime of exemptions from generally applicable laws or instead articulates something more like a jurisdictional principle.  See, e.g., Vincent Phillip Muñoz, The Original Meaning of the Free Exercise Clause: The Evidence from the First Congress, 31 Harv. J.L. & Pub. Pol’y 1083, 1097–98 (2008); Michael W. McConnell, Freedom from Persecution or Protection of the Rights of Conscience?: A Critique of Justice Scalia’s Historical Arguments in City of Boerne v. Flores, 39 Wm. & Mary L. Rev. 819, 830–31 (1998).

304N.Y. Const. of 1777, art. XXXVIII, reprinted in 5 The Federal and State Constitutions Colonial Charters, and Other Organic Laws of the States, Territories, and Colonies Now or Heretofore Forming the United States of America 2623, 2637 (Francis Newton Thorpe ed., 1909) [hereinafter Federal and State Constitutions]; see also David H.E. Becker, Note, Free Exercise of Religion Under the New York Constitution, 84 Cornell L. Rev. 1088, 1105 (1999).

305N.H. Const. of 1784, pt. I, art. V, reprinted in 4 Federal and State Constitutions, supra note 304, at 2453, 2454.

306Md. Const. of 1776, Declaration of Rights, art. XXXIII, reprinted in 3 Federal and State Constitutions, supra note 304, at 1686, 1689.

307Mass. Const. of 1780, pt. 1, art. II, reprinted in 3 Federal and State Constitutions, supra note 304, at 1888, 1889.

308S.C. Const. of 1790, art. VIII, § 1, reprinted in 6 Federal and State Constitutions, supra note 304, at 3258, 3264.

309R.I. Charter of 1663, reprinted in 6 Federal and State Constitutions, supra note 304, at 3211, 3213.

310McConnell, Originssupra note 303, at 1461.  Pennsylvania is not included in this list, but it also had by this time a history of this kind of qualification.  See Gary S. Gildin, Coda to William Penn’s Overture: Safeguarding Non-Mainstream Religious Liberty Under the Pennsylvania Constitution, 4 U. Pa. J. Const. L. 81, 96, 121 (2001).

311McConnell, Originssupra note 303, at 1461.

312It is because of the self-imposed limit of working with existing doctrine that this Article does not further engage with the alternative of simply using a balancing test, as proposed by Laycock.  See supra note 15; supra note 193 and accompanying text. 

313Emp. Div. v. Smith, 494 U.S. 872 (1990).

314The briefing by petitioners and their many supporters in Fulton v. City of Philadelphia, 141 S. Ct. 1868 (2021), outlined many critiques of the existing Free Exercise doctrine based on Smith.  E.g., Reply Brief for Petitioners at 7, Fulton v. City of Philadelphia, 141 S. Ct. 1868 (2021) (No. 19-123).

315See Fulton, 141 S. Ct. at 1883 (Barrett, J., concurring).

316For an article proposing a more nuanced framework for handling scrutiny, see Charles F. Capps, Incidental Burdens on First Amendment Freedoms, 96 Notre Dame L. Rev. Reflection 136 (2021). 

317Guido Calabresi & Philip Bobbitt, Tragic Choices: The Conflicts Society Confronts in the Allocation of Tragically Scarce Resources (1978).

318Marc O. DeGirolami, The Tragedy of Religious Freedom (2013).

319J. Joel Alicea, Liberalism and Disagreement in American Constitutional Theory, 107 Va. L. Rev. 1711 (2021). 

320Several courts have adopted this as the first step of a ministerial exception analysis, a practice that should be generally followed.  See, e.g., Conlon v. InterVarsity Christian Fellowship‌/USA, 777 F.3d 829, 833 (6th Cir. 2015); Grussgott v. Milwaukee Jewish Day Sch., Inc., 882 F.3d 655, 657–58 (7th Cir. 2018). 

321The most influential case on what makes an institution religious is Shaliehsabou v. Hebrew Home of Greater Washington, Inc., 363 F.3d 299, 310 (4th Cir. 2004).  SeeGrisham & Blomberg, supra note 32, at 84.

322See, e.g., Boy Scouts of Am. v. Dale, 530 U.S. 640, 647–48, 655–56 (2000); Roberts v. U.S. Jaycees, 468 U.S. 609, 623 (1984).

323This Article brackets the question of how to decide who is a minister.  See Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2062–63 (2020); id. at 2069–71 (Thomas, J., concurring); id. at 2073–76 (Sotomayor, J., dissenting).

324See Nally v. Grace Cmty. Church of the Valley, 763 P.2d 948, 960 (Cal. 1988).

325See, e.g., Mabus v. St. James Episcopal Church, 884 So. 2d 747, 755 (Miss. 2004); 3 W. Cole Durham & Robert Smith, Religious Organizations and the Law § 22:8 (2d ed. 2021).  But see Swanson v. Roman Cath. Bishop of Portland, 692 A.2d 441, 445 (Me. 1997).

326See Hosanna-Tabor Evangelical Lutheran Church & Sch. v. EEOC, 565 U.S. 171, 199 (2012) (Alito, J., concurring).

327See Watson v. Jones, 80 U.S. (13 Wall.) 679, 728–29 (1872); see also Kedroff v. Saint Nicholas Cathedral of the Russian Orthodox Church in N. Am., 344 U.S. 94, 116 (1952).

328Serbian E. Orthodox Diocese for the U.S. & Can. v. Milivojevich, 426 U.S. 696, 721 (1976).

329Cath. Charities of Sacramento, Inc. v. Superior Ct., 85 P.3d 67, 77 (Cal. 2004); Cath. Charities of the Diocese of Albany v. Serio, 859 N.E.2d 459, 465 (N.Y. 2006).

330Cath. Charities of Sacramento, 85 P.3d at 77.

331See EEOC v. Roman Cath. Diocese of Raleigh, N.C., 213 F.3d 795, 801 (4th Cir. 2000) (examining whether church governance served a “spiritual function”).

332The Supreme Court hinted at a limiting principle like this when it referenced “matters of faith and doctrine and in closely linked matters of internal government.”  Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2061 (2020) (emphasis added).

333See, e.g., Watson v. Jones, 80 U.S. (13 Wall.) 679, 728–29 (1872). 

334See, at 729.

335See Kedroff v. Saint Nicholas Cathedral of the Russian Orthodox Church in N. Am., 344 U.S. 94, 116 (1952); Skrzypczak v. Roman Cath. Diocese of Tulsa, 611 F.3d 1238, 1243 (10th Cir. 2010); United Methodist Church, Balt. Ann. Conf. v. White, 571 A.2d 790, 794 (D.C. 1990); Rayburn v. Gen. Conf. of Seventh-Day Adventists, 772 F.2d 1164, 1169 (4th Cir. 1985).

336See Alvaré, supra note 21.

337See, e.g., Fratello v. Archdiocese of N.Y., 863 F.3d 190, 203 (2d Cir. 2017); Bollard v. Cal. Province of the Soc’y of Jesus, 196 F.3d 940, 946 (9th Cir. 1999); McClure v. Salvation Army, 460 F.2d 553, 560 (5th Cir. 1972) .

338See, e.g., Combs v. Cent. Tex. Ann. Conf. of the United Methodist Church, 173 F.3d 343, 350 (5th Cir. 1999).

339One question that remains in this field is whether to protect hostile workplace claims.  One could make arguments each way under the principles set out in this Article.  There is a split of authority on this issue in the courts.  Compare Demkovich v. St. Andrew the Apostle Par., Calumet City, 3 F.4th 968, 980 (7th Cir. 2021) (holding the ministerial exception to be a defense to hostile workplace claim), and Skrzypczak, 611 F.3d at 1245 (same), with Bollard, 196 F.3d at 944 (holding that the ministerial exception does not extend to hostile workplace claims), and Elvig v. Calvin Presbyterian Church, 375 F.3d 951, 960 (9th Cir. 2004) (holding that hostile workplace claims leading to tangible employment action—firing the claimant—are protected by the ministerial exception, but without the tangible action, are not protected).  For an argument for treating hostile workplace as tortious behavior not protected by religious autonomy principles, see Rachel Casper, When Harassment at Work Is Harassment at Church: Hostile Work Environments and the Ministerial Exception, 25 U. Pa. J.L. & Soc. Change 11 (2021). 

340This is usually easy where church membership in modern America almost never brings with it property or civil rights.  Old cases sometimes distinguished issues where church membership came with a property right or a civil right.  See, e.g., Watson v. Jones, 80 U.S. (13 Wall.) 679, 728–29 (1872).

341See, e.g., Westbrook v. Penley, 231 S.W.3d 389, 397 (Tex. 2007).

342Serbian E. Orthodox Diocese for the U.S. & Can. v. Milivojevich, 426 U.S. 696, 717 (1976) (“Nor is there any dispute that questions of church discipline . . . are at the core of ecclesiastical concern . . . .”).

343See Nathan S. Chapman, Adjudicating Religious Sincerity, 92 Wash. L. Rev. 1185, 1205–10 (2017).

344See, e.g.Westbrook, 231 S.W.3d at 392–93. 

345See, e.g., O’Connor v. Diocese of Honolulu, 885 P.2d 361, 370–01 (Haw. 1994); Paul v. Watchtower Bible & Tract Soc’y of N.Y., Inc., 819 F.2d 875 (9th Cir. 1987).

346See, e.g.Galatians 6:1.

347See, e.g., Martin Bucer, Concerning the True Care of Souls 70 (Peter Beale trans., 2009).

348See, e.g., Grimes’ Ex’rs v. Harmon, 35 Ind. 198, 254 (1871).

349See, e.g., cases cited supra note 31.

350See Watson v. Jones, 80 U.S. (13 Wall.) 679, 728–29 (1872).

351See, e.g., Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2056–58, 2066 (2020) (referencing employment agreements but without explaining the significance of consent in the decision to apply the ministerial exception).

352See Helfand, supra note 111, at 1901–02.

353See Lupu & Tuttle, supra note 17, at 1798 n.23 (describing consent in older charitable immunity cases).

354See Doug Criss, Every Year a Filipino Man Marks Good Friday with an Actual Crucifixion.  He Just Did It for the 33rd Time, CNN (Apr. 19, 2019, 11:45 AM), https://‌‌/2019‌/04‌/19‌/asia/philippines-crucifixion-practice-33rd-year-trnd‌/index.html [https://‌‌/R3WX-D2Z5].

355See Tandon v. Newsom, 141 S. Ct. 1294, 1296 (2021); Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1877 (2021).

356Hypotheticals along these lines featured in recent briefing: Brief for Petitioner at 2–3, Williams v. Kingdom Hall of Jehovah’s Witnesses, 440 P.3d 820 (Utah Ct. App. 2019) (No. 20170783-CA).

357On the border, consider Doe v. Roman Catholic Bishop of Springfield, No. 2179CV00049, 2021 WL 3493884, at *6 (Super. Ct. Mass. June 17, 2021), where the church hierarchy allegedly misrepresented whether plaintiff alleged suffering abuse.  Church autonomy is raised with some regularity as a defense to negligence claims, including when there are allegations of negligent hiring or supervision resulting in abuse.  See, e.g., Buettner-Hartsoe v. Balt. Lutheran High Sch. Ass’n, No. 20-3132, 2021 WL 2580385, at *10 (D. Md. June 23, 2021).

358See Alvaré, supra note 21, at 373; Horwitz, supra note 47, at 984; Griffin, supra note 9, at 998.

359Historians have long debated whether the Carthaginians engaged in child sacrifice.  For an argument that they did, see Patricia Smith, Lawrence E. Stager, Joseph A. Greene & Gal Avishai, Cemetery or Sacrifice?  Infant Burials at the Carthage Tophet, 87 Antiquity 1191 (2013).

360See supra Section II.D.

361See supra subsection III.B.6.

362E.g., Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993) (holding one such law unconstitutional).

363See Lael Weinberger, Is Church Autonomy Jurisdictional?, Loy. U. Chi. L.J. (forthcoming) (manuscript at 17–18), [].

364Matthew 22:21; Mark 12:17.

365See Winnifred Fallers Sullivan, Church State Corporation: Construing Religion in US Law 29–45 (2020).

366See Daniel O. Conkle, Religious Truth, Pluralism, and Secularization: The Shaking Foundations of American Religious Liberty, 32 Cardozo L. Rev. 1755, 1774 (2011).

367See Sullivansupra note 365, at 10; see also Noah Feldman, The Intellectual Origins of the Establishment Clause, 77 N.Y.U. L. Rev. 346, 354–72 (2002).

368See Daniel O. Conkle, The Path of American Religious Liberty: From the Original Theology to Formal Neutrality and an Uncertain Future, 75 Ind. L.J. 1, 4 (2000).

369See Robert Joseph Renaud & Lael Daniel Weinberger, Spheres of Sovereignty: Church Autonomy Doctrine and the Theological Heritage of the Separation of Church and State, 35 N. Ky. L. Rev. 67, 94 (2008). 

370See, e.g., Griffin, supra note 9, at 998.

371See, e.g., Conkle, supra note 368, at 4 (noting the pervasive influence of Christianity on the development of American religious liberty).

372See, e.g., Robert Louis Wilken, Liberty in the Things of God: The Christian Origins of Religious Freedom (2019) (describing the contributions of the Christian tradition to the idea of religious liberty); Witte, supra note 286 (describing the contribution of Reformation thought to religious liberty and the development of rights concepts); Eric Nelson, The Hebrew Republic: Jewish Sources and the Transformation of European Political Thought 88–137 (2010); Feldman, supra note 367.

373See Weinberger, supra note 98, at 7–8.

374See generally Roy A. Clouser, The Myth of Religious Neutrality: An Essay on the Hidden Role of Religious Belief in Theories (rev. ed. 2005).

375See, e.g., Eric Nelson, The Theology of Liberalism: Political Philosophy and the Justice of God 109–33 (2019); Adrian Vermeule, Liturgy of Liberalism, First Things (Jan. 2017), https://‌‌/article‌/2017‌/01‌/liturgy-of-liberalism/ []; see alsoJames Fitzjames Stephen, Liberty, Equality, Fraternity 3–4 (Stuart D. Warner ed., Liberty Fund 1993) (1874) (discussing Millian liberalism).