Hostname: page-component-78c5997874-t5tsf Total loading time: 0 Render date: 2024-11-20T03:43:55.230Z Has data issue: false hasContentIssue false

Protecting People from Their Own Religious Communities: Jane Doe in Church and State

Published online by Cambridge University Press:  20 November 2023

Eugene Volokh*
Affiliation:
Professor of Law, UCLA School of Law, USA; Senior Fellow, Hoover Institution, Stanford (Summer 2024)
Rights & Permissions [Opens in a new window]

Abstract

Suppose that people seek confidentiality in what would otherwise be a public process—such as litigating or applying for a firearms license—because they are afraid that publicly identifying them will stigmatize them in their (or their families’) religious communities. Should the law allow them to proceed anonymously to better protect their interests and to avoid discouraging their lawsuits or applications? Or would that unduly stigmatize the religious community by branding it as improperly censorious or judgmental—or interfere with religious community members’ ability to evaluate for themselves how their coreligionists are using the courts and other government processes?

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of the Center for the Study of Law and Religion at Emory University

Introduction

Debates about religious exemptions often involve a religious community seeking protection against secular law. But what should be done when religious community members seek protection against their own community, or at least against parts of that community? In particular, when should the legal system take steps to help such members conceal actions—actions that for the rest of us would have to be public—precisely to avoid the religious community learning about those actions?

Many legal rules require that people be identified in public documents. Litigants must generally litigate under their own names, not pseudonymously (or anonymously, two terms that are generally used interchangeably in these contexts).Footnote 1 Firearms licenses and license applications are public records in many states; so are liquor license applications. Public records laws sometimes require disclosing the names of people who have been involved in government actions.

Some of these laws provide for exceptions, for instance when requiring a litigant’s “disclosure of his identity in the public record would reveal highly sensitive and personal information that would result in a social stigma.”Footnote 2 And some courts have read this as authorizing confidentiality for people who might otherwise face special stigma in their religious community.Footnote 3

Considering the possible reactions of a litigant’s or applicant’s religious community does make some sense, because doing so helps accurately estimate the social stigma this person is likely to face. And the interest in shielding people from such stigma is not just individual but also social: For instance, we want to encourage victims of tortious misconduct to come forward, so that the civil liability system can better deter such misconduct. If we know some victims will not sue if they must be named because they fear being ostracized by friends and family, we might want to allow them to sue pseudonymously.

At the same time, note the premise of the analysis: The claimants are entitled to protection not just from the operation of the secular legal disclosure rule, but from what is seen as the oppressive or backward view of their religious community. The community is viewed as unfairly judgmental about (for instance) sexual assault victims, people who engage in premarital sex, people who sue fellow community members, or perhaps drinkers and gamblers and gun owners. After all, this perceived unfairness is what gives the claimant’s confidentiality argument a special edge in the request for pseudonymity—an edge lacked by claimants who belong to other religious communities (or to no religious community).

And of course, many members of the religious group might disagree with the claimant’s characterization of the group. They might, for instance, argue that their group members are more loving and forgiving than most people and thus less likely to stigmatize (for instance) the sexual assault victim or erotic dancer than the public at large would be. Such questions are of course hard to decide objectively. But a court decision allowing pseudonymity on these grounds sends a message: The legal system does not approve of the community’s attitudes.

Relatedly, the purpose of confidentiality in such cases—to allow a person to sue or get a permit or conceal certain records without fear of ostracism by coreligionistsFootnote 4—means that the legal system is deliberately denying the coreligionists information that they allegedly think is important to their judgments about fellow group members. If group members, for instance, think that erotic dancing or contraceptive use or premarital sex or extramarital sex is sinful, they have the constitutional right to think less of those who engage in such behavior or even to shun or excommunicate them.

Of course, group members who nonetheless want to engage in such behavior also have the right to try to hide it from others’ censorious eyes. But should the legal system deliberately favor one group’s interests over the other’s, by giving those group members an extra edge in the confidentiality analysis that ordinary litigants do not get? (In all the cases I describe, group members are claiming such an extra edge precisely because of their religious group membership, not simply seeking religion-neutral treatment.Footnote 5) And should the inquiry be different when the legal system is keeping confidential the dissenting group members’ voluntary behavior, such as consensual premarital sex or alcohol use or gambling, as opposed to dissenting group members’ having been involuntarily victimized (for instance, by having been raped)?

In what follows, I analyze these matters, chiefly for the benefit of judges, lawyers, litigants, and academics who are interested in the law of pseudonymous litigation and of public records. (Most of the focus is on pseudonymous litigation because that is where the cases have been so far.) But the discussion also informs the broader questions: How should the law reconcile the competing claims of religious community members?Footnote 6 And, in particular, when should the law give some people legal exemptions precisely so they can conceal their actions from religious group members who might want to react to those actions?

I suggest that the law should not give special exemptions from the norm of public identification in such situations. Protecting religious community group members from stigma may itself stigmatize the religious community, and it may involve courts and other government entities taking sides between the religious community’s mainstream and its dissenters.

To be sure, such pseudonymity is increasingly being allowed in one particularly common and appealing situation—when rape victims are facing the risk of stigma within their religious community—on the theory that such stigma is especially improper and especially socially harmful because it is likely to lead to underenforcement of rape laws. (Indeed, that is a theory that many, though not all, courts apply to allow pseudonymity to rape victims even apart from religious factors.Footnote 7) But even if that approach is accepted as to claimants who say they have been raped, it should not be extended to other situations that involve community disapproval of voluntary behavior rather than of involuntary victimization. And in any event, I hope that my analysis will prove useful regardless of whether readers agree with this bottom line.

Disclosure Rules and Stigma-Based Exemptions

I begin by considering the areas where this issue can arise: requiring plaintiffs to name themselves, allowing subpoenas used to identify defendants, disclosing information about political contributions and political petition signatures, and disclosing public records.

Requiring Plaintiffs to Name Themselves

Generally speaking, US law requires all parties to a lawsuit to be named so that the public can better monitor public courts deciding cases using public funds, in the public’s name, and relying on government’s coercive power.Footnote 8 Indeed, some courts see this as a facet of the First Amendment rights of members of the public to access court records.Footnote 9 This rule is not absolute: Sometimes parties are allowed to appear pseudonymously, and one factor that courts consider is whether publicly identifying a party would cause “social stigma” beyond mere “personal embarrassment” or mere damage to reputation.Footnote 10 But courts are sharply split on what sorts of social stigma qualify. For instance, some courts have let plaintiffs claiming to have been sexually assaulted proceed pseudonymously, but others have not.Footnote 11 Likewise, some courts have let plaintiffs proceed pseudonymously to conceal their sexual orientation, but others have not.Footnote 12

Yet when plaintiffs argue that publicly identifying them would cause special stigma because of the likely reactions of their religious communities, courts often cite that as a special reason for pseudonymity, for instance:

The Court recognizes that victims of sexual assault often wish to keep their identities secret out of fear of embarrassment or social stigmatization. Those concerns alone, however, are insufficient to permit a plaintiff to proceed under a pseudonym. Doe v. Princeton Univ., 2019 WL 5587327, at *4 (D.N.J. Oct. 30, 2019). However, if a movant shows that her specific circumstances demonstrate a risk of serious social stigmatization surpassing a general fear of embarrassment, courts may consider those circumstances in favor of granting the motion. Doe v. Neverson, 820 F. App’x 984, 988 (11th Cir. 2020) (reversing the denial of a motion to proceed under a pseudonym because the district court failed to consider the potential significant social stigmatization on account of the movant’s membership in “a strict Muslim household where under their cultural beliefs and traditions such a sexual assault would have the tendency to bring shame and humiliation upon [the movant’s] family.”).Footnote 13

The same issue has arisen regarding sexual orientation. One court allowed a plaintiff to pseudonymously litigate his sexual orientation employment discrimination and harassment claim, partly on the grounds that “his family, friends, and religious community are not accepting of the LGBTQ community.”Footnote 14 Another rejected a pseudonymity request by a defendant who had made a similar allegation.Footnote 15

And the question has arisen as to potentially controversial voluntary sex-related behavior. One court allowed an erotic dancer to litigate her wages-and-hours claim pseudonymously in part because “her parents are devoutly religious members of a Christian church.”Footnote 16 The Seventh Circuit granted, without discussion, a motion that similarly justified pseudonymity for Notre Dame students who were admitting premarital sexual activity and contraceptive use (or at least use of contraceptives that some view as abortifacients).Footnote 17

Still other cases granted motions for pseudonymity on the grounds that the plaintiffs were suing religious leaders (Orthodox rabbis), and their religious community was alleged to be hostile to those who air such accusations before outsiders.Footnote 18 Those cases also involved alleged sexual victimization, but their logic would apply to other intra-community disputes as well. One case, for instance, relied on an article that “describes at length the cultural factors within the Orthodox Jewish community inhibiting dissent among its members, including: ‘the overwhelming concern with shame (a child who makes an abuse claim can be thought to bring shame on his whole family);’ ‘the thinking that virtually any public complaint about another person amounts to slander;’ and the notion that ‘say[ing] anything bad about the community’ would be ‘desecrating God’s name.’”Footnote 19

Two other cases allowed pseudonymity for people who had been accused of sexual misconduct, and who were suing their universities on the grounds that they had been wrongly disciplined based on such allegations. Their theory was that litigating would make clear that they had engaged in premarital sex (though, they argued, consensual premarital sex), and that revealing this would stigmatize them in their community.

In one of the cases, the court accepted the argument that pseudonymity was proper because the alleged abuser was a citizen of Kuwait, “where ‘sexual activity outside of marriage goes against religious and cultural values’ and ‘sexual relations outside of marriage are illegal,’” which creates a “heightened risk of stigma and retaliation the plaintiff alleges that [he] faces in his home country.”Footnote 20 In the other, the court accepted the argument that “this case involves students who attend a strictly religious school that expressly prohibits pre-marital sex”; that “[m]any of these students, including Plaintiff, seek entry into the clergy or religious-affiliated groups after graduation”; and that “disclosure of their identities in connection with their extra-marital sexual activities risks exposing Plaintiff, as well as the unnamed students, to ridicule and even ostracization from their own religious community.”Footnote 21

And in principle, the same argument could arise with regard to lawsuits that stem from, say, altercations at bars or casinos, filed by plaintiffs whose religious communities frown on alcohol or gambling;Footnote 22 incidents at places of worship that involve litigants who may have hidden their new religious practices from their old religious communities;Footnote 23 lawsuits over interest-bearing loans filed by plaintiffs whose religious communities condemn such loans; or divorce suits—or for that matter any claims that would require mentioning a litigant’s divorce—when the litigant’s religious community condemns divorce.Footnote 24

As it happens, all the decided cases involve situations where some courts would allow pseudonymity to some litigants even absent concerns about opprobrium in a religious community. Many cases, for instance, do allow pseudonymity for plaintiffs alleging sexual assault, and for plaintiffs alleging unsound university accusations of assault.Footnote 25 Some cases have done the same for gay or lesbian litigants, for defendants who are sued for allegedly copying pornography, and for erotic dancers who are suing for labor law violations.Footnote 26 And while I know of no cases that have allowed pseudonymity related merely to claims of contraceptive use, some cases have allowed pseudonymity as to sexual matters more broadly.Footnote 27 Thus, there is no crisp scenario in which litigants would definitely be denied pseudonymity under normal circumstances, but can get it if they belong to a particular religious community.

But, in all these scenarios, some courts do deny pseudonymity to ordinary litigants.Footnote 28 And the cases cited above show that the claimed reactions of the litigant’s religious community are being treated as one factor cutting in favor of pseudonymity.

Moreover, the cases are focusing on the litigant’s religious community. For instance, the fact that a litigant’s actions—or even just what the litigant is accused of—would lead to opprobrium within the litigant’s professional community, to the point of potential economic ruin, is generally rejected as a basis for pseudonymity.Footnote 29 Likewise, if an Alcoholics Anonymous leader seeks pseudonymity in a lawsuit stemming from a drunk driving arrest or a bar fight, on the grounds that identifying him would reveal that he had been drinking and might diminish his standing among AA members, it seems unlikely that he would get pseudonymity.Footnote 30 It is the religious basis for the potential opprobrium that weighs in favor of the litigant.

To be sure, considering religion in such situations might be sound. Recognizing that some people might be more vulnerable to community stigma because of their religious community membership could well be praised as the governmental “neutrality in the face of religious differences” that Sherbert v. Verner Footnote 31 said was at least constitutionally permissible (even though it would not be constitutionally mandatory here, for reasons discussed in the next main section). My point is simply that the law here, like in some other accommodations of religion, is indeed treating religion specially.

For a helpful contrast, consider concerns about actual physical violence rather than social or professional stigma. Courts do indeed generally allow pseudonymity, entirely apart from whether the violence stems from religious views, when there is evidence of real risk of such violence—for instance, possible physical retaliation against people who cooperated with the governmentFootnote 32 or risk of violence against an asylum seeker in his home country.Footnote 33 The same would apply to people who fear religion-related violence, as in Doe v. Dordoni, which allowed pseudonymity based on a reasonable fear of violent reprisal in Saudi Arabia stemming from a Saudi citizen’s conversion from Islam to Christianity.Footnote 34

And if a woman suing for sexual assault can credibly show that, if she is publicly identified as a rape victim, she faces a serious risk of physical harm from family members,Footnote 35 that would suffice to justify pseudonymity under normal religion-neutral pseudonymity precedents.Footnote 36 The rule allowing pseudonymity aimed at diminishing a risk of physical violence would thus be religion-neutral in such cases. Not so with the courts’ allowing pseudonymity to prevent social retaliation by a religious community.

Allowing Subpoenas Used to Identify Defendants

So far, I have discussed people who want to call on the coercive power of the court system without having to name themselves as plaintiffs. But people may also want to stop coercive subpoenas aimed at uncovering their identities as potential defendants. Those people’s concerns are often just about being fired or professionally blacklisted if they are identified as having publicly criticized their employer, or about being retaliated against by the government if they are identified as having publicly criticized government officials.Footnote 37 But sometimes the defendants also argue that they would be ostracized by their religious communities.Footnote 38

Some defendants in lawsuits claiming copyright infringement by viewers and sharers of pornographic films, for instance, have sought pseudonymity based in part on the argument that “having my name or identifying or personal information further associated with the [porn film] is embarrassing, damaging to my reputation in the community at large and in my religious community.”Footnote 39 In two cases, the courts cited this in allowing the case to proceed pseudonymously, albeit temporarily.Footnote 40

Likewise, a potential defendant in a copyright case brought by the Jehovah’s Witnesses sought anonymity in part because “if Watch Tower discovers his identity, the revelation of his identity would damage or destroy his relationships with friends and family who are active members of the Jehovah’s Witness community”—“he has been part of the Jehovah’s Witness community his whole life, and so the pain of social exclusion would be overwhelming.”Footnote 41 A similar argument was made by an Orthodox Jewish blogger (“Orthomom”) whose identity was being sought, as a potential libel defendant, via a subpoena directed to her blog hosting company.Footnote 42

Courts will sometimes allow defendants to resist a subpoena even apart from the effects on the defendants in their religious communities, especially if the court concludes that the lawsuit is likely to be legally unfoundedFootnote 43 and the lawsuit is over “political, religious, or literary speech.”Footnote 44 But the question remains: Should the analysis also be influenced by evidence of a threat of stigma specifically within a religious community, as the defendants in the just-cited cases argued?Footnote 45

Note that, unlike plaintiffs seeking anonymity, defendants seeking anonymity often do aim to hide their identities even from their litigation adversaries, and not just from the public. And if the court concludes that plaintiff has a viable legal claim, then the court would presumably deny the defendant’s request for total anonymity so that the plaintiff can identify the defendant to gather information needed to allow the plaintiff’s case to go forward—for instance, information needed to establish the defendant’s state of mind, or to eventually satisfy a judgment against the defendant. Nonetheless, the court could still order that the defendant’s identity be revealed only subject to a protective order that bars the plaintiff from revealing the information to othersFootnote 46 (or perhaps even makes the information available on an attorneys’-eyes-only basisFootnote 47).

Disclosing Information about Political Contributions and Political Petition Signatures

Political contributions—either to candidates or to independent advocacy groups that seek to influence elections—often have to be disclosed under campaign finance laws and are then made available to the public. The same is true in many states for petition signatures (whether for initiative, referendum, recall, or candidate qualification).Footnote 48

The Supreme Court has held that donor or signer information could be treated as confidential if there is sufficient evidence of likely “harassment” or “reprisals” against such donors or signers,Footnote 49 including firing by employers.Footnote 50 It’s not clear just what might qualify as harassment or reprisals, but some donors or signers might argue that they face a risk of ostracism by their religious community or even excommunication if their identities become known. (Imagine, for instance, people who would like to donate to a pro-abortion-rights initiative, or to sign such a petition, but are afraid of being shunned by their or their family’s religious group, which believes abortion is murder.)

Disclosing Public Records

The possible reactions of a person’s religious community can likewise potentially affect decisions about anonymity in public records.Footnote 51 This is especially so for license applications. For instance, New York law requires a license to possess a firearm, and the licenses are public records unless (among other things) the licensing officer finds that “the applicant has reason to believe he or she may be subject to unwarranted harassment upon disclosure of such information.”Footnote 52 Applicants could presumably claim, by analogy to the pseudonymity cases, that they belong to a pacifist religious community that frowns on firearms (or at least on handguns kept for self-defense against people),Footnote 53 and that disclosing their applications might prompt “unwarranted harassment” from coreligionists.Footnote 54

Likewise, some states treat liquor licenses as public records.Footnote 55 These applications may have to include the names of individual corporate officers and shareholders,Footnote 56 and even if they just have the names of the corporate or LLC owner, public record documents for those entities will generally include the name of corporate officers. Some people might be reluctant to have their connection to alcohol businesses publicized because they might be afraid the publicity will lead them to be condemned by coreligionists who condemn alcohol. Public records laws may leave room for government agencies to accommodate such desire for privacy, if the laws have exceptions for when disclosure would produce “unwarranted harassment”Footnote 57 or would be “a clearly unwarranted invasion of privacy,” with what constitutes private information judged by “the customs, mores, or ordinary views of the community.”Footnote 58

Marriage licenses are also public records in many states.Footnote 59 Some applicants might want the records concealed on the grounds that many in their religious community would condemn their particular marriage (for example, because community members condemn interfaith marriages or reject divorces and view remarriage as bigamous).

Finally, more generally, public records laws can be used to disclose a wide range of other contacts between people and the government (such as arrest reports). In some situations, these disclosures could similarly jeopardize people’s standing in their religious communities, and public agencies might argue that they should redact those people’s names when releasing information in response to public records requests.Footnote 60

Distinguishing Debates about More Familiar Religious Exemptions

These debates are not like those about exemptions from laws that require violation of religious beliefs, categorical exemptions that create “unyielding” restraints on secular interests, or requests for religion-neutral application of generally applicable rules.

Not Like Exemptions from Laws That Require Violation of Religious Beliefs

The religious exemptions I describe are different from most traditional religious exemptions. When the law exempts religious observers from a legal requirement, it is usually trying to protect them from having to violate their felt religious obligations. Normally, you might have to be clean-shaven to be a police officer, for instance, but if you are (say) a Muslim, Sikh, or Orthodox Jew, you could be exempted from that requirement.Footnote 61 Normally, you might have to be bareheaded in court, but if you are (say) a Muslim woman or an Orthodox Jew, you could be exempted.Footnote 62 And what is good for the religious observers is generally also good for the rest of their religious community.

But the normal debates about such religious exemptions do not carry over well to the disclosure exemptions I discuss here. First, the underlying disclosure rules do not “substantially burden” religious practice in the traditional sense of forbidding people from engaging in religiously motivated behavior,Footnote 63 compelling them to “violate[] their religious beliefs,”Footnote 64 or otherwise punishing them for their religious practices.Footnote 65 None of the people discussed in this article have claimed that God requires them to be anonymous or characterized anonymity as a religious practice.Footnote 66 For the same reason, exemptions under those disclosure rules are not required by the Free Exercise Clause, a Religious Freedom Restoration Act, or a similar general religious accommodation statute.

And for the same reason, Cutter v. Wilkinson has little to say about whether the disclosure exemptions are consistent with the Establishment Clause. Cutter held that a religion-preferential rule is “compatible with the Establishment Clause” when it (1) “alleviates exceptional government-created burdens on private religious exercise,” (2) “take[s] adequate account of the burdens a requested accommodation may impose on nonbeneficiaries,” and (3) is “administered neutrally among different faiths.”Footnote 67 But element (1) is absent here: the consideration of religion is aimed at alleviating not an exceptional government-created burden, as the caselaw has defined the term, but rather the social consequences of a particular legal rule.

Second, the tension here is not simply between the claimant’s felt religious obligations and secular government interests—it is between the claimant and the allegedly wrongly judgmental members of the claimant’s own religious community. As I discuss below, that calls for considering a different set of concerns than for a typical religious exemption claim.

Not Like Categorical Exemptions That Create “Unyielding” Restraints on Secular Interests

These exemptions also likely are not rendered unconstitutional by Estate of Thornton v. Caldor. Footnote 68 Thornton held that the Establishment Cause prohibited a law that categorically required all employers to give their employees the employees’ Sabbath day off. That law, in some respects like the disclosure exemptions, did protect against private action.

But the Thornton Court condemned the law’s creating “an absolute and unqualified right not to work on whatever day they designate as their Sabbath,” which involved “unyielding weighting in favor of Sabbath observers over all other interests.”Footnote 69 The pseudonymity cases I describe treat religion just as a factor to be considered in the analysis, which could be overcome by other factors. In this respect, they are more like Title VII’s more modest, balancing-based duty of reasonable accommodation for religious objectors to private workplace rules, which Thornton did not invalidate.Footnote 70 The same would likely be true as to exemptions from the other disclosure rules.

Of course, one can still object as a policy matter to the burden that pseudonymity imposes on third parties—especially on pseudonymous litigants’ adversaries—even if that burden is constitutionally permissible. Courts do consider such burdens, and often reject pseudonymity on those grounds (as I discuss in detail elsewhereFootnote 71). Nonetheless, courts do sometimes allow pseudonymity, when they view the burden of litigating in public on the requester to be great, and the burden of pseudonymity on the opponent and the public to be small. There does not seem to be a strong reason why it would be unfair to the opponent to consider the requester’s religious community as part of the analysis.Footnote 72

Not Like Requests for Religion-Neutral Application of Generally Applicable Rules

Finally, I note again what was discussed above:Footnote 73 All these cases involve requests that courts consider religious group membership as part of the pseudonymity analysis, not that courts treat people without regard to their religiosity. Precedents such as Trinity Lutheran v. Comer, Espinoza v. Montana Department of Revenue, and Carson v. Makin thus do not apply here; their premise is that the government may not “expressly discriminate[] against otherwise eligible recipients by disqualifying them from a public benefit solely because of their religious character.”Footnote 74 Declining to consider the stigma that a person might face in a particular religious community—just as the law generally declines to consider the stigma that a person might face in a particular social or professional communityFootnote 75—would not constitute such express discrimination.Footnote 76

Judicial Evaluation of the Qualities of a Religious Community

Weighing a person’s religious community membership in deciding whether to let the person remain pseudonymous might thus not be unduly burdensome or unfair to litigation adversaries. But might it be unfair to the religious community?

Consider, for instance, two of the cases described above, plus a third:

  • A woman whose family and friends are Trinidadian Muslims seeks pseudonymity in suing over an alleged rape.

  • A woman whose family and friends are Southern Baptists seeks pseudonymity in an employment lawsuit stemming from her work as a stripper.

  • A man whose family and friends are Orthodox Jews seeks pseudonymity in a libel lawsuit stemming from statements made by an ex-lover, including allegations of “date rap[e].”Footnote 77

To begin with, a judge would have to determine not just whether the plaintiffs would be stigmatized within that community, but whether they would be unusually stigmatized compared to ordinary litigants. The risk of some such stigma, after all, is not by itself generally enough to justify pseudonymity in litigation or as to public licenses or records.Footnote 78 And even the particular attributes in these three examples—having been sexually assaulted, being a stripper, or having being accused (even without a conviction or civil judgment) of being a rapist—are often stigmatized even outside particular religious communities.

Many within the religious communities might think the stigma is not materially greater in those communities than elsewhere, and might themselves feel condemned by the implication that it is. Our religion calls us to be loving and forgiving, they might say. It does not condemn women who were attacked. It might condemn stripping and nonmarital sex, but it acknowledges that everyone is a sinner and that all we can do is repent and strive to change and to encourage our friends and families to do the same. And it does condemn date rape, but so do many other groups.

True, there might be some unduly judgmental people who will take an unkind view as to rape victims, strippers, or people who engage in sex outside marriage; but all communities have people such as that. Why are you making us out to be particularly harsh? In the course of claiming that we unfairly stigmatize certain people—and do so more than society generally does—might you be unfairly stigmatizing us?

Moreover, the magnitude of religious communities’ condemnation of these litigants is hard to measure; decisions are likely to be guesswork, based mostly on the judge’s perception of the group’s reputation. There may well be an affidavit from the litigant,Footnote 79 and perhaps from some others, making claims about such condemnation; but such self-serving claims from a litigant—or claims from friends or family members in support of the litigant’s position—are not likely to be terribly reliable. There may be media accounts,Footnote 80 but those may well be one-sided, or based on the media outlet’s own biases. Community members might thus plausibly believe that they are being incorrectly tarred as especially judgmental, retrogressive, and intolerant based simply on outsiders’ stereotypes of, say, Muslims, conservative Christians, or Orthodox Jews.Footnote 81

To be sure, in traditional religious exemption cases, courts are supposed to accept claimants’ assertions that the law substantially burdens their religious practices, at least so long as the courts conclude the claimants are sincere.Footnote 82 But that makes sense because the burden relevant to those cases turns on the claimant’s own subjective beliefs. Here, the claimants are making assertions about the likely actions of coreligionists—assertions that, if believed, reflect badly on the character of those coreligionists.

One possible solution, of course, would be to pseudonymize the religious group, by saying that the defendant belongs to a group that condemns certain behavior without naming the group. But that would deny the public (and future litigants and their lawyers) important information about the basis for a judge’s decision. How, after all, can the public effectively “oversee and monitor the workings of the Judicial Branch,”Footnote 83 if it is not told the true basis for a judge’s decision?Footnote 84

Another solution might be for the judges to take pains to note that they are just speaking of the views of some religious community members, and not talking about the religious group as a whole. But such decision making nonetheless risks the sort of government disapproval of religion that some of the Court’s Establishment Clause decisions have condemned. Consider, for instance, a litigant’s claim that, say, she “comes from a strict Muslim household where under their cultural beliefs and traditions such a sexual assault would have the tendency to bring shame and humiliation upon her family,”Footnote 85 and that she is therefore subject to “social stigma”Footnote 86 beyond that faced by a typical litigant. A judicial determination endorsing this claim—even without claiming that the view is held by all conservative Muslims—may well be seen as critical of conservative Islam, even if the judge does not expressly condemn the group for such views. After all, many of us would disapprove of a group that blames the victim this way.

This is most true when a court concludes that a group is shaming those whom the rest of us see as clearly sinned against rather than sinning, such as rape victims. But it is also true when a group is alleged to stigmatize those who engage in voluntary actions, such as premarital sex, commercial stripping, or the like. The foundation for the anonymity claim, after all, is that the claimant is entitled to special protection—a unusual exception from the norm, and one that stems from an intolerable risk that the claimant will suffer unjustified emotional and social harm. A judgment that a group inflicts such harm and that its members deserve protection against the group is therefore a condemnation of that group.

To be sure, the Court’s recent American Legion decision repudiated the endorsement test as a formal Establishment Clause doctrine,Footnote 87 and the prohibition on disapproval of religion has generally been closely linked to the prohibition on endorsement.Footnote 88 Still, even American Legion condemned government speech that “‘deliberately disrespect[s]’ members of minority faiths.”Footnote 89

Of course, one might argue that an impartial determination of the facts about a religious group is as a matter of law not disrespectful: Find the facts and let the chips fall where they may. But a determination based on little more than an outsider judge’s perception of the group, coupled with a litigant’s own affidavit (or even the affidavits of some of the litigant’s supporters), may often carry the risk of stemming from—and reinforcing—disrespectful stereotypes and not just objective reality.

And in any event, even if such determinations are not unconstitutional, they seem to me best avoided, for the reasons given above. Certainly, the American law of religious exemptions generally avoids having to decide what Southern Baptist or Muslim or Jewish communities are like, focusing instead on the beliefs of the individual claimant and not generalizations about a group.Footnote 90

The notable exception there is Wisconsin v. Yoder, where the Court’s exemption of Amish objectors from the requirement that parents must send all children to school until age sixteen stemmed in part from “evidence … show[ing] that the Amish have an excellent record as law-abiding and generally self-sufficient members of society,” and that “the Amish community has been a highly successful social unit within our society.”Footnote 91 But this feature of Yoder has been criticized,Footnote 92 and I think rightly so.

Favoring Those Who Want Anonymity at the Expense of Coreligionists Who Want to Know

Allowing pseudonymity because of fear of coreligionists’ reactions may also be unfair to the coreligionists. First, it may take sides in an internal debate within the religious group about which behavior should be condemned and which should not be. Second, it may undermine coreligionists’ ability to monitor what is being done by secular institutions—such as courts—that operate within their communities.

Other Community Members’ Religious Interests

Suppose that some members of a religious community acknowledge that community members who act in particular ways are shamed by the community. Indeed, those members think that such shaming is an important and valuable feature of their religious life, and indeed of their exercising their First Amendment rights.Footnote 93

We believe that stripping, premarital sex, viewing pornography, taking disputes to outsiders (often included by Orthodox Jews within the rubric of “lashon hara”Footnote 94), drinking, gambling, or lending or borrowing money with interest, they might say, is contrary to God’s will. One way we deter breaches of these norms is through the threat of social shaming—much as many secular institutions threaten social shaming for what they view as immoral behavior, such as racism or sexism or hostility to homosexuality. This threat helps encourage members to stay on the right path, and helps protect people from the harms that straying can cause. And instances of such shaming also serve as teaching moments for reminding community members about these norms.Footnote 95

Someone suing as a John Doe (to give a pseudonymous litigation example) now claims that he is a member of our community, which condemns, for instance, interest-bearing lending; and because of that he wants to sue pseudonymously over such a loan, so that we, his fellow community members, do not learn about his conduct. Normally, he would not be entitled to sue pseudonymously in such a situation,Footnote 96 but here he seeks pseudonymity precisely because he fears the stigma of being labeled as a sinner by our community. But by shielding his identity, you are deliberately denying us information because you think we will use the information illegitimately, by shaming him for his religious transgression, and perhaps being less likely to trust him.Footnote 97 You are thus favoring his preferences (and yours) over ours.

Now of course these hypothetical religious community members would not be claiming some legal entitlement to surveil all their group members’ sexual behavior. They are not, for instance, trying to subpoena the bank records of all their coreligionists so that they can identify usurers or pornography buyers. When ordinary legal rules, applied entirely without regard to people’s religious communities, provide community members with privacy, the community’s mechanisms for enforcing its norms are stymied, but unavoidably so.

But with the pseudonymity rules I am describing, courts are making a deliberate choice, at least in cases involving the litigant’s voluntary behavior (as opposed to just the litigant’s being a sexual assault victim): precisely because a litigant is violating the norms of the litigant’s religious community, they are giving the litigant extra access to pseudonymity that most litigants do not have, and denying the religious community access to information about what is happening in court—access that the general community usually has (and may well valueFootnote 98) with regard to most lawsuits. And the same would also apply to decisions to treat other kinds of records as confidential, when the purpose is to allow people to conceal information from their coreligionists.

Courts are thus observing something of a schism within a religious community—between the orthodox enforcers of norms and dissenters who reject the norms. And they are choosing to support the dissenters over the orthodox, by giving the dissenters special legal treatment, precisely because the courts condemn the beliefs that the orthodox hold (or at least because the courts think those beliefs are too militantly held).

I am not arguing here that such determinations of reactions within a community are foreclosed by the First Amendment’s prohibition on courts’ resolving “ecclesiastical questions,” such as the proper interpretation of “church doctrine.”Footnote 99 Deciding whether members of a religious group harshly condemn other members who act in particular ways (or who have been victimized in particular ways) would not generally involve “the interpretation of particular church doctrines and the importance of those doctrines to the religion.”Footnote 100 Rather, it would involve estimation of how often members of a particular religious community hold particular views, not whether those views are consistent with religious doctrine or theologically important. Likewise, siding with the dissenters here rather than with the orthodox because one thinks the orthodox are being unduly judgmental is not necessarily a theological judgment as such.

Nonetheless, such a decision does involve “tak[ing] sides in a religious matter,” by deliberately favoring one religious subcommunity’s approach at the expense of another’sFootnote 101—not by applying “neutral principles of law” (in the sense of religion-neutral principles)Footnote 102 but by evaluating the perceived practices of a particular religious community. The First Amendment presumptively forbids the government from discriminating among religions, even when the discrimination targets groups based on behavior and not belief, for instance religious groups “that solicit more than fifty per cent of their funds from nonmembers.”Footnote 103 The First Amendment also bars the government from discriminating against religious groups more broadly.Footnote 104 Likewise, it should generally bar the government from favoring religious dissenters over the more religiously orthodox, or treating religious communities differently based on their more judgmental belief systems or based on their use of shame as a sanction. And even if such treatment is constitutionally permissible, it seems to me to be something that the secular legal system should generally avoid engaging in.

Other Community Members’ Political Interests

Providing pseudonymity or anonymity here also affects not just the other community members’ religious interests, but also their political interests: their rights to monitor what is happening in their communities, to better understand not just their coreligionists’ actions but also government processes.

Consider, for instance, pseudonymity in litigation. The public’s right of access to information in government records is generally framed as a right to “oversee and monitor the workings of the Judicial Branch.”Footnote 105 Indeed, as noted above, the right of access to court recordsFootnote 106—and, in the view of many courts, the right of access to parties’ names—is a “clear and strong First Amendment interest.”Footnote 107

Indeed, if a community member is suing a community leader or a community institution,Footnote 108 other members might especially want to monitor the judicial system to make sure the defendant is being treated fairly. “Public confidence” in the judiciary, courts say, “cannot long be maintained where important judicial decisions are made behind closed doors and then announced in conclusive terms to the public, with the record supporting the court’s decision sealed from public view.”Footnote 109 That may be especially true for minority religious communities’ confidence in the secular judiciary, since many minority religions and denominations might have special historical reasons to distrust the majority’s legal system.Footnote 110

To be sure, the plaintiff might understandably worry that those coreligionists will resent him for suing, and may shun him or refuse to do business with him. But that is a commonplace concern for many plaintiffs, even outside religious communities, and is generally seen as insufficient to justify pseudonymity and the concomitant interference with the First Amendment interest in public supervision of the courts.Footnote 111

Indeed, when plaintiffs are allowed to sue fellow community members pseudonymously (a nontrivial subset of the litigation I describe), the pseudonymity can sharply damage the community, even beyond the damage caused by the underlying allegations of misconduct.

A trusted institution or individual is sued by name. Community members hear about the allegations, and are naturally troubled. But because the accuser is unnamed, community members have an especially hard time figuring out how credible the allegations are.

Perhaps eventually the civil justice system will reach a result, but that will likely be years in the future. The result may not fully dispose of the allegations, for instance if the claim is dismissed on procedural grounds, or if it settles. Plus, the community might not trust the civil justice system’s results in any event.

And this effect on the religious community would be especially serious if the court accompanies its pseudonymity order with an express or implied gag order on the opposing party, for instance requiring that a religious institution or a religious leader being sued by the plaintiff “shall not publicly identify Plaintiff,” not just “in court filings” but also “otherwise.”Footnote 112 The institution or leader would be barred from communicating with fellow members of the religious community, and responding to the allegations that had publicly been made.

Even in a case that does not involve an intracommunity dispute, the right of access to court records includes people’s right to know what is going on in various government facilities, such as the courts or the public licensing systems, and the who is a big part of the what: “lawsuits are public events and the public has a legitimate interest in knowing the facts involved in them. Among the facts is the identity of the parties.”Footnote 113 “The people have a right to know who is using their courts.”Footnote 114 “[A]nonymous litigation runs contrary to the rights of the public to have open judicial proceedings and to know who is using court facilities and procedures funded by public taxes.”Footnote 115 “The Court is a public institution and the public has a right to look over our shoulders and see who is seeking relief in public court.”Footnote 116 That is why everyone has standing to intervene to assert a right of access to court records by opposing sealing or pseudonymity,Footnote 117 though in some situations courts do recognize exceptions to this right of access.

Thus, for instance, if a civil lawsuit is filed by someone who claims he was wrongly or pretextually fired or arrested for drunkenness, members of the public can generally get access to the filings and to the plaintiff’s name. A local reporter can write a story in the local newspaper, which would be especially likely if the plaintiff is someone of some standing in the community—a doctor, a lawyer, a teacher, or the like. Fellow community members could decide whether this information should lead them to trust the plaintiff less. Likewise, if a litigant sues a respected community institution, community members can take that into account in their personal judgment of the litigant’s character.

Say, though, that a court allows the plaintiff to sue pseudonymously because he belongs to a religious community that condemns alcohol consumption, or because he is suing a respected religious institution and the religious community might condemn such airing of its dirty laundry before outsiders. A reporter for a newspaper within that community then would not be able to write the same sort of story because the plaintiff will not be identified. (Reporters for other newspapers would not be able to write such a story, either, but they may well be less interested in cases within that community.) Community members would not be able to use the information about the lawsuit to update their judgments about the plaintiff. And the legal system would be opaque to them, in a way that it is not when people from other religious communities are the litigants.

Moreover, depriving the religious community members of that information would be a big part of the purpose of maintaining pseudonymity: the theory, after all, is that the plaintiff should be able to litigate without facing stigma within that community, presumably because the legal system thinks that stigma would be excessive or otherwise unfair. But, again, by doing that the secular legal system would be siding with the plaintiff and against his coreligionists.

And similar effects will be present in the other anonymity scenarios I mention. Where a reporter for a community newspaper might normally be able to write a fairly informative story about a new liquor license application, or about the donations to a local political campaign, allowing pseudonymity to people who fear religious community opprobrium would block that. A reporter for a religious community newspaper might be denied access to the details of who owns the new bar or liquor store, or who has been financing a local campaign that is of importance to the religious community.

All this suggests that providing pseudonymity to members of particular religious groups might violate the principle of the Texas Monthly v. Bullock lead opinion, which struck down a sales tax exemption for religious works on Establishment Clause grounds. That three-Justice opinion (written by Justice Brennan and joined by Justices Marshall and Stevens) stressed that the tax exemption was not a permissible accommodation of religion because it “burden[ed] nonbeneficiaries markedly”Footnote 118 “by increasing their tax bills by whatever amount is needed to offset the benefit bestowed on subscribers to religious publications.”Footnote 119

Here, too, the exemption burdens third parties, in the ways described above: it denies third parties, especially the exempted person’s fellow community members, information that they would otherwise possess, and that can be important to their community’s religious and political life. Indeed, given that some courts view the public’s right to access the information about litigant identities as a First Amendment right,Footnote 120 the exemption thus reduces third parties’ constitutional rights.

To be sure, the analogy to Texas Monthly is imperfect; and the Texas Monthly concurrences are even more distant because they stressed that tax exemption’s “preferential support for the communication of religious messages,”Footnote 121 an element that is missing here. Still, the core point remains, either as an establishment clause argument or at least as a policy argument: the benefits to some religious observers (or at least to some people who have family members within religious communities) come at the expense of the information access rights of other religious observers.

Conclusion

What then is the answer? When someone seeks to remain anonymous, whether in litigation, license applications, or public records requests, should courts specifically focus on the possible stigma this would create within the person’s religious community, as a factor in favor of anonymity?

My inclination is to say no, for some of the reasons given above. Protecting the person from stigma within the religious community may itself unfairly stigmatize the community. And it may sometimes involve secular institutions taking sides between dissenters and the orthodox within the community.

Even if a community’s stigmatizing rape victims simply because they were victimized is seen as so contemptible—and so detrimental to the enforcement of rape laws—that courts do end up considering such stigma, I doubt that courts should extend that judgment to situations where the community is disapproving not of a person’s involuntary victimization but of a person’s voluntary actions (premarital sex, extramarital sex, contraceptive use, suing a fellow community member, using alcohol, gambling, and the like). That the secular legal system does not disapprove of the actions, or disapproves of them only mildly, should not justify its taking steps to deny religious communities information that would normally be made public.

But whether or not readers agree with me on this, I hope that the analysis above can help courts and scholars analyze these questions more fully, and recognize the interests at stake here, not just for the litigants but for the religious community. And I hope this sheds light on broader discussions of whether and when the secular legal system should protect religious group members from the reactions of their fellow group members.

Acknowledgments

Thanks to Jessie Hill, Leslie Jacobs, Mark Kagan, Andy Koppelman, Jim Oleske, and Nelson Tebbe for their advice.

References

1 Eugene Volokh, The Law of Pseudonymous Litigation, 73 Hastings Law Journal 1353 (2022).

2 Raiser v. Brigham Young Univ., 127 F. App’x 409, 411 (10th Cir. 2005).

3 See infra the subheading “Requiring Plaintiffs to Name Themselves.”

4 I use coreligionists and religious group members to mean members of a religious community whose opinions are especially important to a person. The use is imprecise because people who no longer believe in a religion might still be connected with the religious community, either directly or through family, and might therefore care a great deal about what that community thinks of them. But I trade off precision here for the simplicity of just being able to say coreligionist.

5 See infra notes 2729 and accompanying text.

6 This question of course also prominently arises with regard to get statutes, which are aimed at pressuring husbands (generally Orthodox Jews) to give their wives a religious divorce (called a get in Hebrew) once a secular divorce has been entered: The reason the law intervenes is precisely that, among many Orthodox Jews, wives who are not given such religious divorces are viewed as still married, and thus any later remarriages are seen as void and the children of the remarriages are viewed as illegitimate. Those laws pose their own constitutional problems, especially to the extent they are seen as coercing the husbands into engaging in religious actions. See, e.g., Megibow v. Megibow, 612 N.Y.S.2d 758 (1994); Aflalo v. Aflalo, 295 N.J. Super. 527 (1996). See generally Lisa Zornberg, Beyond the Constitution: Is the New York Get Legislation Good Law?, 15 Pace Law Review 703 (1995). But while such laws burden one of the divorcing spouses, they do not aim at constraining the religious community’s actions: Once the religious divorce is given, including under compulsion of the law, the religious community generally has no further objection to the ex-wife’s later remarriage.

The question has also arisen with regard to attempts to limit religious communities from excommunicating or shunning members, but there the law refuses to interfere with the communities’ and community leaders’ decisions, treating group membership as a voluntary matter that either the individual or the group may terminate without legal constraint. See, e.g., Paul v. Watchtower Bible & Tract Soc’y of N.Y., Inc., 819 F.2d 875 (9th Cir. 1987) (shunning); Hubbard v. J Message Grp. Corp., 325 F. Supp. 3d 1198, 1216 (D.N.M. 2018) (shunning); Thomas v. Fuerst, 345 Ill. App. 3d 929 (2004) (excommunication); Decker ex rel. Decker v. Tschetter Hutterian Brethren, Inc., 1999 S.D. 62 (excommunication). And while of course this voluntariness is protected by generally applicable laws, such as laws preventing battery, false imprisonment, and the like, those laws are indeed generally applicable: They do not specially exempt religious community members precisely because they are religious community members.

7 See Volokh, supra note 1, at 1430–37.

8 See id. at 1366–68.

9 See, e.g., DePuy Synthes Prod., Inc. v. Veterinary Orthopedic Implants, Inc., 990 F.3d 1364, 1370 (Fed. Cir. 2021); In re Sealed Case, 931 F.3d 92, 96 (D.C. Cir. 2019); United States v. Microsoft Corp., 56 F.3d 1448, 1464 (D.C. Cir. 1995); Doe v. Stegall, 653 F.2d 180, 185 (5th Cir. 1981); Ramsbottom v. Ashton, No. 3:21-cv-00272, 2021 WL 2651188, at *2 (M.D. Tenn. June 28, 2021); Doe v. Paychex, Inc., No. 3:17-cv-2031, 2020 WL 219377, at *10 (D. Conn. Jan. 15, 2020); Doe v. Del Rio, 241 F.R.D. 154, 156 (S.D.N.Y. 2006); Dep’t of Fair Emp. & Housing v. Superior Court, 82 Cal. App. 5th 105, 110 (2022); Doe v. Kidd, 19 Misc. 3d 782, 788 (N.Y. Sup. Ct. 2008).

10 See Volokh, supra note 1, at 1405–16. This has to do with public identification; the defendant would generally need to know the plaintiff’s identity. Id. at 1362 n.25. Cf. United States v. Nordlicht, No. 1:16-cr-00640-BMC, 2019 WL 235640, at *1, *4 (E.D.N.Y. Jan. 16, 2019) (noting that the court had allowed the government to delay disclosing to defendants certain materials about the witnesses against them because “the Government claimed that these cooperating witnesses expressed fear that they would be subject to ostracism and harassment if their cooperation against fellow members of their religious community was revealed,” but noting that this was just a delay rather than a categorical denial, and “defendants received the deferred … production sufficiently in advance of trial to obviate any prejudice resulting from the delayed disclosure”).

11 See Volokh, supra note 1, at 1430–37.

12 See id. at 1406 & nn.254 & 257.

13 Doe v. Cook Cnty., 542 F. Supp. 3d 779, 784, 787 (N.D. Ill. 2021); see also Doe v. Neverson, 820 F. App’x 984 (11th Cir. 2020); Doe v. Barr, No. 1:20-cv-03553, 2020 WL 12674163 (D.D.C. Dec. 4, 2020); Doe v. City of Dalton, No. 4:21-cv-00128-LMM, 2021 WL 4618600 (N.D. Ga. July 12, 2021); Doe v. Amal, No. 1:12-cv-1359 (E.D. Va. Nov. 29, 2012), granting Motion, id. (Nov. 27, 2012); Doe v. Roe, No. 1:22-cv-08779-PGG (S.D.N.Y. Oct. 28, 2022), granting Motion, id. (Oct. 17, 2022); Roe v. Patterson, No. 4:19-cv00179-ALM-KPJ, 2019 WL 2407380 (E.D. Tex. June 3, 2019) (noting, but not heavily relying on, such an argument); Kashef v. BNP Paribas S.A., No. 1:16-cv-03228-AJN (S.D.N.Y. June 22, 2021), granting Motion [with supporting Memorandum], id. (June 7, 2021); Letter Motion, id. at 2 (Feb. 6, 2017) (making a similar request but just referring to plaintiffs’ “close knit community of Sudanese-Americans”), granted, Order, id. (Feb. 17, 2017) (again without discussion of community views).

14 Doe v. Heubach Ltd., No. 2:23-cv-01347-GAM, 2023 WL 3295528, at *1 n.1 (E.D. Pa. Apr. 10, 2023).

15 Homesite Ins. Co. of the Midwest v. Ewideh, No. 1:22-CV-1664, 2023 WL 426923, at *1, *3 (M.D. Pa. Jan. 26, 2023). The court’s rationale for rejecting pseudonymity focused on the peculiar way in which the defendant’s sexual orientation had been brought up, id. at *3: Plaintiff insurance company hadn’t said anything about sexual orientation in its filings—which focused on an insurance claim related to “water leak damage,” Homesite Ins. Co. of the Midwest v. Ewideh, No. 1:22-CV-1664, 2023 WL 3035313, *1 (M.D. Pa. Mar. 7, 2023)—but the defendants had argued in their motion for pseudonymity that defendant “is of a certain sexuality and on occasion was called homophobic slurs by the Large Loss Property Manager” of the insurance company, Motion, id., at 1 (Jan. 23, 2023).

16 Doe #1 v. Deja Vu Consulting Inc., No. 3:17-cv-00040, 2017 WL 3837730, *4–*5 (M.D. Tenn. Sept. 1, 2017).

17 Order Granting Intervenors’ Request to Litigate Anonymously, Univ. of Notre Dame v. Sebelius, 743 F.3d 547 (7th Cir. Jan. 14, 2014) (No. 13-3853), granting Motion, id. (Jan. 8, 2014) (discussion of religious community’s potential reaction is at id. at 16–18).

18 Doe No. 2 v. Kolko, 242 F.R.D. 193, 197 (E.D.N.Y. 2006); Doe v. Georgetown Synagogue—Kesher Israel Congregation, No. 1:16-cv-01845-ABJ (D.D.C. Sept. 15, 2016), granting Motion for Leave to Proceed Under Pseudonyms, id. (Sept. 15, 2016); Doe v. Georgetown Univ., No. 14-0007644 (D.C. Super. Ct. Dec. 1, 2014), granting Motion, id. at 6–7 (Dec. 8, 2014), available in Superior Court Documents, Doe v. Georgetown Univ., No. 1:15-cv-00026 (D.D.C. Jan. 8, 2015) (ECF No. 1-4). Indeed, some Jews disapprove of Jews suing other Jews—even ones who are not religious leaders—in secular courts. Michael J. Broyde, The Pursuit of Justice and Jewish Law: Halakhic Perspectives on the Legal Profession 62–64 (1996); Rabbi Yaacov Feit, The Prohibition Against Going to Secular Courts, 1 Journal of the Beth Din of America 30, 30–31 (2012) (“One who goes to secular court is considered ‘an evildoer, as if he has blasphemed, and as if he has raised a hand against the Torah of Moses.’”).

19 Kolko, 242 F.R.D. at 197.

20 Doe v. Am. Univ., No. 1:19-cv-03097, at 4–6 (D.D.C. Oct. 10, 2019), granting Motion, id. at 1, 6 (Oct. 10, 2019); see Doe v. Am. Univ., No. 19-CV-03097 (APM), 2020 WL 5593909, *1 (D.D.C. Sept. 18, 2020).

21 Memorandum, Doe v. Dordt Univ., 5:19-cv-04082-CJW-KEM, at 15 (N.D. Iowa Dec. 5, 2019) (seeking pseudonymity because “this case involves intimate details of sexual contact between two college students” and “naming Plaintiff would result in the type of harm to reputation he seeks to avoid by bringing this action,” though not specifically mentioning the harm within the religious community), granted, id. (Mar. 3, 2020). Dordt University, an evangelical Christian school affiliated with the Christian Reformed Church in North America, indeed stressed that it “firmly holds to the biblical teaching that premarital intercourse is forbidden. Further, behavior (e.g. nudity, lying in bed together) that encourages such intimacy will not be tolerated by the university. Students involved in such behavior will face disciplinary action.” Dordt University, Student Life, https://web.archive.org/web/20220808171648; https://www.dordt.edu/student-life/student-handbook/student-life.

22 Compare, in a different sort of privacy context, Oil, Chem. & Atomic Workers Int’l Union, AFL-CIO, Loc. 2-286 v. Amoco Oil Co. (Salt Lake City Refinery), 885 F.2d 697, 707 (10th Cir. 1989), where the court blocked a unionized employer’s unilateral adoption of a drug and alcohol testing policy, partly because of “the invasion of privacy threatened by Amoco’s testing program, and the potential for stigmatization and humiliation of its employees,” which “would potentially be all the more severe because of the close-knit character of the employees and the fact that the predominant religion in the community [presumably Mormonism] proscribes the drinking of alcoholic beverages,” so that “[t]he consequences of revelations about drug or alcohol use could have long term consequences for a member of such a community.”

23 See cases cited infra notes 4041 (discussing pseudonymity claims brought by people concerned that identifying them as the authors of criticisms of their old religious groups would cause them to be ostracized by friends and family members who continue to belong to those groups).

24 In one case, the court refused to vacate a final divorce when the parties had reconciled, despite the parties’ desire to avoid condemnation by their religious community:

[P]laintiff’s counsel[] urged … that the parties sought to avoid religious stigma in their communities that allegedly attaches to divorce. If that contention is the case, they should have considered alleged religious, community, and cultural opprobrium before they both consented to an uncontested civil divorce. At any rate, the Court does not bow to alleged religious sentiments or convictions that may attach to divorce. Civic marriage and divorce should not be entangled with religious marriage and divorce. Preventing embarrassment to former litigants, moreover, is not a worthy allocation of judicial resources.

Doe v. Doe, 29 Misc. 3d 483, 486–87 (2010). Yet despite that, the court took an unusual step: “In order to avoid unnecessary embarrassment to the parties, the Court has concealed their names in this version of the opinion submitted for publication, referring to the husband and wife as John Doe and Jane Doe and hiding the correct index number.” Id. at 484. (In the absence of any religious community concerns, references to divorce are not generally viewed by courts as justifying pseudonymity, see Doe v. Bd. of Regents of Univ. of N.M., No. CIV 20-1207 JB/JHR, 2021 WL 4034136, *1 (D.N.M. Sept. 4, 2021).)

25 See Volokh, supra note 1, at 1430–37.

26 See id. at 1406–09.

27 See id.

28 See id.; Doe v. Wyndham Vacation Ownership, Inc., No. 6:23-cv-01104-RBD-DCI (M.D. Fla. Aug. 22, 2023) (denying pseudonymity to a plaintiff who had alleged that she was raped, and distinguishing an earlier Eleventh Circuit case in which pseudonymity was granted on the grounds that the earlier case “differentiat[ed] general allegations of potential personal embarrassment from the situation there where plaintiff made specific allegations of being from a ‘devout Muslim family’ who would experience shame and harm to her family and reputation and submitted examples of specific harassing and threatening comments posted online”).

29 See Volokh, supra note 1, at 1420–23.

30 This would likely fall within the familiar principle that mere risk of harm to professional reputation and of social stigmatization does not justify pseudonymity. See id. at 1416–23; see, e.g., Doe v. Rackliffe, 173 Conn. App. 389, 397 (2017) (“A [party’s] desire to avoid economic and social harm as well as embarrassment and humiliation in his professional and social community is normally insufficient to permit him to appear without disclosing his identity.”) (quotation marks omitted); Doe v. Apstra, Inc., No. C 18-04190 WHA, 2018 WL 4028679, *1 (N.D. Cal. Aug. 23, 2018) (likewise rejecting a claim of pseudonymity that was based in part on the risk of “professional stigmatization” within the litigant’s “professional community”); Patton v. Entercom Kansas City, LLC, No. CIV.A. 13-2186-KHV, 2013 WL 3524157, *3 (D. Kan. July 11, 2013) (“although the Court acknowledges that Patton greatly values her reputation as a community member and future lawyer, plaintiffs alleging damage to their personal and professional reputations are generally not allowed to proceed anonymously”).

31 374 U.S. 398, 409 (1963).

32 See United States v. Doe, 655 F.2d 920, 922 n.1 (9th Cir. 1980) (pseudonymizing a litigant’s name because of the “risk of serious bodily harm if [prison inmate’s] role on behalf of the Government were disclosed to other inmates”); Doe No. 1 v. United States, 143 Fed. Cl. 238, 241 (2019) (“[D]isclosing the names of BATF employees could endanger them.”).

33 See Volokh, supra note 1, at 1397–99.

34 No. 1:16-CV-00074-JHM, 2016 WL 4522672, *3 (W.D. Ky. Aug. 29, 2016).

35 In Doe v. Barr, No. 1:20-cv-03553, 2020 WL 12674163 (D.D.C. Dec. 4, 2020), plaintiff argued that there was such a risk, id. at *3, but the judge allowed her to litigate under a pseudonym based solely on the possible danger of “reputational harm” within her community, id., without any reference to risk of physical harm.

36 The same might apply with regard to other serious harms that go beyond stigma or social or professional retaliation, even if they do not rise to the level of violence. Thus, for instance, Wolfchild v. United States, 62 Fed. Cl. 521, 553 (2004), rev’d on other grounds, 559 F.3d 1228 (Fed. Cir. 2009), allowed certain Sioux plaintiffs to proceed pseudonymously because of a concern that the tribe would disapprove of their position in the lawsuit; the court stressed the risk not just of social opprobrium but also of the tangible legal consequence of lost tribal membership (since “the community governments possess nearly a plenary power over community membership,” 62 Fed. Cl. at 553).

37 See, e.g., In re Anonymous Online Speakers, 661 F.3d 1168, 1173 (9th Cir. 2011) (discussing the “fear of economic or official retaliation”).

38 For an example of the threat of such ostracism, levied against the writer of an anonymous letter, see Brief of Appellee Rabbi Jack Bieler, Hager-Katz v. Mevlin J. Berman Hebrew Academy, 2010 WL 4890009, *7 (Md. Ct. Spec. App. Oct. 8, 2010) (quoting the rabbi’s letter): “This past Shabbat I suggested that this incident should inspire all of us to be extremely careful about engaging in Lashon HaRa [i.e., derogatory speech]. But in the event that the author’s identity can be incontrovertibly established, we think it additionally appropriate that this individual be welcomed neither into our synagogue nor our homes until such a time that she can demonstrate to the community’s satisfaction that full repentance has been achieved.”

39 See, e.g., In re BitTorrent Adult Film Copyright Infringement Cases, 296 F.R.D. 80, 90 (E.D.N.Y. 2012), report & recommendation adopted sub nom. Patrick Collins, Inc. v. Doe 1, 288 F.R.D. 233 (E.D.N.Y. 2012).

40 Strike 3 Holdings, LLC v. Doe, No. 3:19-CV-508-J-34JRK, 2019 WL 5722173 (M.D. Fla. Nov. 5, 2019), allowed at least temporary pseudonymity, citing, among other things, defendant’s argument that he is a “religious man.” A case decided by a different judge and involving a different defendant, Strike 3 Holdings, LLC v. Doe, No. 5:22-CV-398-JA-PRL, 2022 WL 16695170, *1–*2 (M.D. Fla. Nov. 3, 2022), stated that, “Although Doe alleges embarrassment in his religious community and marriage if he cannot proceed anonymously, his allegations (as currently alleged) fall short of those justifying a complete grant of his motion,” but added that, “considering different judicial approaches allowing a party to proceed anonymously, and weighing the reputational harm risks to Doe against the presumption of openness in judicial proceedings, I find this is an exceptional case warranting Doe to proceed anonymously (at least initially) until 90 days after service of the Complaint.” The latter case was voluntarily dismissed, presumably pursuant to a settlement, before the 90 days expired. Order, id. (M.D. Fla. Nov. 17, 2022).

41 In re DMCA Subpoena to Reddit, Inc., No. 3:19-mc-80005-SK, at 4, 12 (N.D. Cal. May 17, 2019).

42 See Memorandum of Law of Proposed Intervenor “Orthomom” in Opposition to Petitioner’s Application for Pre-Commencement Disclosure, Greenbaum v. Google, Inc., No. 102063/07, 2007 WL 4162535, at *27, *28 (N.Y. Sup. Ct. N.Y. Cty. Mar. 13, 2007), granted, 18 Misc. 3d 185 (N.Y. Sup. Ct. 2007) (granting motion on the grounds that Orthomom’s posts were not libelous as a matter of law, and not discussing the religious ostracism concerns):

[S]ignificant community norms in the Orthodox Jewish community disapprove of criticizing leaders, and particularly of making those criticisms in ways that bring Jews or Judaism into disrepute outside the community. Critics and their families can be shunned, even deprived of their livelihoods because many Orthodox Jews work for businesses that are run by fellow Orthodox Jews, or that depend on Orthodox customers. … Some of Orthomom’s readers have specifically taken her to task for spreading “lashon hara,” or evil talk. Thus, Orthomom faces a serious risk within her community if, as a result of Greenbaum’s petition for discovery, she is identified as the author of these criticisms of wrongdoing within the community.

43 Doe v. Cahill, 884 A.2d 451, 460 (Del. 2005); Dendrite Int’l, Inc. v. Doe No. 3, 775 A.2d 756, 760–61 (N.J. Super. Ct. App. Div. 2001).

44 Compare Obi Pharma, Inc. v. Does 1-20, No. 16CV2218 H (BGS), 2017 WL 1520085, *3 (S.D. Cal. Apr. 27, 2017) (“Cahill itself involved political speech and appears to be reserved for ‘political, religious, or literary speech.’”), with Ciabattoni v. Teamsters Loc. 326, No. N15C-04-059 VLM, 2018 WL 2418388, *3 (Del. Super. Ct. May 29, 2018) (“Plaintiff’s argument that Cahill only applies to political speech is without merit.”).

45 See, e.g., Mobilisa, Inc. v. Doe, 170 P. 3d 712, 720 (Ariz. Ct. App. 2007) (concluding that, in considering whether to enforce a subpoena in such a case, courts should look beyond just the legal validity of the plaintiff’s claims and also engaging in a “a balancing step” in which they could consider, among other things, “the potential consequence of a discovery order to the speaker”); In re Indiana Newspapers Inc., 963 N.E.2d 534, 552 (Ind. Ct. App. 2012) (same).

46 See, e.g., In re BitTorrent Adult Film Copyright Infringement Cases, 296 F.R.D. 80, 90 (E.D.N.Y. 2012), report and recommendation adopted sub nom. Patrick Collins, Inc. v. Doe 1, 288 F.R.D. 233 (E.D.N.Y. 2012).

47 See, e.g., Strike 3 Holdings, LLC v. Doe, No. 20CIV4501WFKVMS, 2021 WL 535218, *7 (E.D.N.Y. Feb. 12, 2021).

48 See, e.g., Doe v. Reed, 561 U.S. 186 (2010).

49 Id. at 200; Brown v. Socialist Workers ’74 Campaign Comm., 459 U.S. 87, 100 (1982).

50 Brown, 459 U.S. at 99.

51 I use anonymity and pseudonymity largely interchangeably here, as do the cases dealing with pseudonymity in litigation; whether it is sealing a license application entirely (which would in effect provide for anonymous licensing), or replacing a litigant’s or applicant’s name with “Jane Doe” or initials or the like, the point is that the person’s name will be concealed from the public (though, in litigation, generally not from the adversary, see supra note 8).

52 N.Y. Pen. L. 400.00(5)(b)(iii).

53 See, e.g., Amish America, Do Amish Use Guns?, https://amishamerica.com/do-amish-use-guns/ (“Amish will not bear arms against others, but they do use firearms for hunting and other purposes”).

54 I set aside here the special case of when parental notification requirements for abortion can be overridden because a court is persuaded that such parental notification would be against the child’s best interests. Compare Planned Parenthood, Sioux Falls Clinic v. Miller, 63 F.3d 1452, 1460 (8th Cir. 1995) (holding that such exemptions from parental notification are constitutionally required), with Planned Parenthood of Blue Ridge v. Camblos, 155 F.3d 352, 367 (4th Cir. 1998) (holding the contrary). Presumably when there is such a best-interests override, the anticipated reactions of parents—and whether judges would see those reactions as excessive and therefore harmful to the child—would be considered, and that would include religiously motivated reactions. But this would presumably be limited to the attitudes of the parents, whether religious or not, and would not focus on the reactions of their religious community more broadly. Even now that Roe v. Wade and Planned Parenthood v. Casey have been overruled, such questions would likely still arise under many states’ abortion laws. See Dobbs v. Jackson Women’s Health Org., 142 S. Ct. 2228 (2022), overruling Roe v. Wade, 410 U.S. 113 (1973), and Planned Parenthood of S.E. Pennsylvania v. Casey, 505 U.S. 833 (1992).

55 See, e.g., Wash. State Liquor & Cannabis Bd., On Premises, Licensees, https://lcb.wa.gov/records/frequently-requested-lists.

56 See, e.g., Cal. Dep’t of Alcoholic Beverage Control, Individual Personal Affidavit, https://www.abc.ca.gov/wp-content/uploads/forms/ABC-208-A.pdf.

57 See, e.g., Freedom Watch, Inc. v. Mueller, 453 F. Supp. 3d 139, 157 (D.D.C. 2020) (applying “unwarranted harassment” test under FOIA), appeal dismissed, No. 20-5071, 2020 WL 4931696 (D.C. Cir. July 30, 2020).

58 Michigan Fed’n of Teachers & Sch. Related Pers., AFT, AFL-CIO v. Univ. of Michigan, 481 Mich. 657, 669–75 (Mich. 2008) (quotation marks omitted).

59 Compare, e.g., Vt. Stat. Ann. § 5132 (treating marriage licenses as public records), with Cal. Family Code §§ 500–511 (generally allowing for confidential marriages, so long as the parties have lived together before marriage).

60 Cf. Holy Spirit Ass’n for Unification of World Christianity, Inc. v. U.S. Dep’t of State, 526 F. Supp. 1022, 1034 (S.D.N.Y. 1981) (withholding the names of people who had alleged misconduct on the religious group’s part because “[t]o disclose the names could subject these individuals to the fear of harassment and needless humiliation”).

61 See, e.g., Fraternal Order of Police Newark Lodge No. 12 v. City of Newark, 170 F.3d 359 (3d Cir. 1999).

62 See, e.g., United States v. James, 328 F.3d 953, 958 (7th Cir. 2003).

63 E.g., Holt v. Hobbs, 574 U.S. 352, 360–61 (2015).

64 E.g., Burwell v. Hobby Lobby Stores, Inc., 573 U.S. 682, 720 (2014).

65 E.g., Sherbert v. Verner, 374 U.S. 398, 404 n.5 (1963).

66 One can imagine such claims in some situations, for instance if someone feels a religious obligation to donate anonymously to charity, but some law requires publicizing the names of donors. See, e.g., Amicus Curiae Brief of the Philanthropy Roundtable in Support of Petitioners, Americans for Prosperity Found. v. Becerra, 2021 WL 862273, at *9 (U.S. Mar. 1, 2021) (“Many donors who desire to remain anonymous are motivated by deeply held religious or moral beliefs that have made anonymous philanthropic giving the norm when it comes to charity over the past two millennia.”); Mishneh Torah, Matnot Aniyim 10:8; Matthew 6:2–4; Quran 2:271; Bailie Mittman, First Amendment Freedoms Diluted: The Impact of Disclosure Requirements on Nonprofit Charities, 96 Indiana Law Journal Supplement 102, 120 n.158 (2021). But the cases I describe do not involve such preference for anonymity as a religious command (or a religious recommendation)—the anonymity is aimed at preventing social and professional retaliation by coreligionists.

67 544 U.S. 709, 720 (2005).

68 472 U.S. 703 (1985).

69 Id. at 709–10.

70 See id. at 711–12 (O’Connor, J., concurring) (distinguishing the Title VII duty “to reasonably accommodate the religious practices of employees unless to do so would cause undue hardship to the employer’s business” from the “absolute protection” given Sabbatarians under the law in Thornton).

71 See Volokh, supra note 1.

72 This is especially so as to the argument that pseudonymity unfairly increases the likely settlement value of the case, id. at 1381–82, by decreasing the reputational costs of litigation to the plaintiff. The settlement value of a case generally turns in large part on the ongoing costs of the lawsuit to the two parties—litigation costs, emotional costs, or reputational costs. All else being equal, if the plaintiff’s costs go down, the plaintiff will be emboldened, and the settlement value of the case will likely increase. Likewise, if the defendant’s costs go down, the settlement value of the case will likely decrease. It follows that, in cases where both sides have reputational or privacy costs stemming from the litigation, giving pseudonymity to one party but not the other would decrease the pseudonymous party’s costs and would change the likely settlement value. All else being equal, a Doe v. Smith will tend to yield a larger settlement than Jones v. Smith, which may be seen as unfair to Smith.

But by hypothesis, in the cases I am describing, any threatened Jones v. Smith litigation will have an unfairly deflated settlement value (compared to a typical case of that sort, where religious community effects are absent): Smith will know that Jones is likely to be afraid of retaliation by Jones’s religious community and that Jones will therefore likely settle the case cheaply to avoid having to file the case in the first place. Allowing Jones to sue pseudonymously may thus tend to bring the case closer to a fair settlement value that is not unduly affected by such publicity effects.

73 See supra note 2729 and accompanying text.

74 Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 462 (2017); Carson v. Makin, 142 S. Ct. 1987, 1996 (2022); Espinoza v. Montana Dep’t of Rev., 140 S. Ct. 2246, 2256 (2020).

75 See Volokh, supra note 1, at 1420–23.

76 Pseudonymity law does sometimes consider the stigma a person may face in society at large, without considering specific subcommunities. But applying such a rule, with no consideration of a religious subcommunity’s reaction, would not be expressly discriminatory and thus prohibited by cases such as Carson v. Makin—just as, for instance, a rule that the government may fund public schools without funding any private schools (religious or otherwise) is not prohibited by those cases. See, e.g., Carson, 142 S. Ct. at 2261.

77 See Complaint, Doe v. Sebrow, No. 2:21-cv-20706, ¶¶ 1, 17 (D.N.J. filed Dec. 23, 2021). The plaintiff argued, in support of pseudonymity, “Plaintiff is a member of a small insular community, the Orthodox Jewish community, and these allegations can destroy him,” Memo. in Support of Motion to Proceed Under Pseudonym, id. at 9 (Feb. 8, 2023), granted, id. (Feb. 10, 2023). The motion was granted without opinion, so it is not clear how much the court relied on a concern about stigma within a religious group.

78 See Volokh, supra note 1, at 1416–23. Many plaintiffs and even more defendants risk some degree of stigma if their identities are revealed. Usually, though, that is not enough to overcome the strong presumption in favor of public litigation. If I am sued for sexual harassment, fraud, or even malpractice, that would surely expose me to “shame and humiliation,” even if I claim that the shame and humiliation are unmerited because I am actually innocent. The same is true if I sue for wrongful firing, and my employer’s defense is that I was really fired for sexual harassment, fraud, or malpractice. Nonetheless, I generally cannot litigate such cases pseudonymously. And while plaintiffs alleging sexual assault often will be allowed to litigate pseudonymously, not all courts take that view. See id. at 1430–37.

79 See, e.g., Declaration of Jane Doe, Doe v. Neverson, No. 1:20-cv-20016-UU, ¶¶ 7–8 (S.D. Fla. Jan. 10, 2020) (ECF. No 7-1 app. A).

80 See, e.g., Motion for Leave to Proceed Under Pseudonyms, Doe v. Georgetown Synagogue—Kesher Israel Congregation, No. 1:16-cv-01845-ABJ, at 7 (D.D.C. Sept. 15, 2016).

81 In asylum cases in which an applicant raises the risk of religious persecution, immigration courts and Article III courts may have to consider some religious groups’ mistreatment of other groups. See, e.g., Sihotang v. Sessions, 900 F.3d 46, 51 (1st Cir. 2018) (noting evidence that “Islamic fundamentalist fervor seems to have intensified such that evangelical Christians may now be at special risk in Indonesia,” including both risk of discrimination by government and of private violence). But that at least involves courts reporting on conditions in foreign countries, usually bolstered by authoritative “State Department country conditions reports,” id. at 52. The cases described in the text involve courts passing judgment on communities within the United States, usually based on affidavits by litigants coupled with conventional perceptions of those communities. Such judgments about domestic religious communities are especially likely to cause religious tensions within the United States and undermine the community members’ sense of being treated equally and respectfully by the American legal system.

82 Thomas v. Review Bd., 450 U.S. 707, 715 (1981).

83 Doe v. Public Citizen, 749 F.3d 246, 263 (4th Cir. 2014).

84 To be sure, this is a decision about pseudonymity, not a decision about the bottom-line result in a case. But pseudonymity decisions are indeed significant because they affect public rights—indeed, in the view of some courts, the public’s First Amendment rights. See supra note 7.

85 Doe v. Neverson, 820 F. App’x 984, 988 (11th Cir. 2020) (quotation marks and brackets omitted).

86 Id.

87 American Legion v. American Humanist Ass’n, 139 S. Ct. 2067 (2019).

88 See, e.g., County of Allegheny v. ACLU, 492 U.S. 573, 620 (1989).

89 139 S. Ct. at 2089.

90 Thomas v. Review Bd., 450 U.S. 707, 715 (1981).

91 406 U.S. 205, 212–13 (1972).

92 See, e.g., Peter J. Riga, Yoder and Free Exercise, 6 Journal of Law and Education 449, 466 (1977) (“What the Court has done in Yoder comes dangerously close to that examination of beliefs which, in itself, is a violation of free exercise.”); Mark Tushnet, Of Church and State and the Supreme Court: Kurland Revisited, 1989 Supreme Court Review 373, 382 (“It is not unfair to read [Yoder] as saying that the claims of the Amish prevailed because they were a ‘good’ religion.”); Lisa Biedrzycki, “Conformed to This World”: A Challenge to the Continued Justification of the Wisconsin v. Yoder Education Exception in a Changed Old Order Amish Society, 79 Temple Law Review 249, 267–68 (2006) (faulting Wisconsin v. Yoder for relying on “beatific stereotypes” of the Amish); Nicholas J. Nelson, A Textual Approach to Harmonizing Sherbert and Smith on Free Exercise Accommodations, 83 Notre Dame Law Review 801, 811–12 (2008) (“The Yoder Court was even rather explicit about its function as a stamp of government approval or disapproval of specific religious beliefs. … The Court even hinted that it would not be so kind to religious views it found less appealing.”); James M. Oleske, Jr., Free Exercise (Dis)honesty, 2019 Wisconsin Law Review 689, 717–18 (2019).

93 Cf. Paul v. Watchtower Bible & Tract Soc’y of N.Y., Inc., 819 F.2d 875, 879 (9th Cir. 1987) (rejecting, on Free Exercise Clause grounds, a claim that the organized “shunning” of a dissenting religious group member constitutes tortious infliction of emotional distress); Hubbard v. J Message Grp. Corp., 325 F. Supp. 3d 1198, 1216, 1217-18, 1221 (D.N.M. 2018) (likewise).

94 See supra note 38, noting some such conduct that was labeled “lashon hara” by some (though the term is broader than just taking disputes to outsiders).

95 Such overt embrace of shaming, I expect, would be more likely for voluntary behavior. Presumably the groups would be less likely to expressly defend shaming of the involuntarily victimized, such as rape victims, and even when such shaming of victims does happen, I expect that it would be more likely to be denied by community leaders rather than overtly endorsed.

96 That is why I used this hypothetical: to avoid the complications that arise in, say, sexual assault cases, where some courts do allow pseudonymity even without reference to religious community norms and others do not.

97 To be sure, many of these groups would not view such behavior as grounds for excommunication; they may be open to people who sin but repent, especially when they do so in front of the community. One point of pseudonymity, though, is to avoid the need for public repentance.

98 See Volokh, supra note 1, at 1369–70 (citing the many cases that stress the public’s presumptive right to access information about who is using the courts).

99 See, e.g., Presbyterian Church in U.S. v. Mary Elizabeth Blue Hull Memorial Presbyterian Church, 393 U.S. 440, 445–51 (1969).

100 Id. at 450.

101 See Commack Self-Serv. Kosher Meats, Inc. v. Weiss, 294 F.3d 415, 425 (2d Cir. 2002) (concluding that laws that aimed at preventing mislabeling of food as kosher improperly favored the Orthodox Jewish understanding of what is kosher, and thus “t[ook] sides in a religious matter, effectively discriminating in favor of the Orthodox Hebrew view of dietary requirements”); id. at 426 (“As a result, because the challenged laws interpret ‘kosher’ as synonymous with the views of one branch, those of Orthodox Judaism, the State has effectively aligned itself with one side of an internal debate within Judaism. This it may not do.”); Pleasant Glade Assembly of God v. Schubert, 264 S.W.3d 1, 2, 13 (Tex. 2008) (refusing to allow liability for emotional distress—as opposed to physical injury—stemming from a religious ritual in which church members were “laying hands” on plaintiff as a means of exorcism, and concluding that, “[b]ecause providing a remedy for the very real, but religiously motivated emotional distress in this case would require us to take sides in what is essentially a religious controversy, we cannot resolve that dispute”).

102 See, e.g., Jones v. Wolf, 443 U.S. 595, 599–602 (1979) (endorsing a “neutral principles” approach for dealing with church property disputes among different members of a religious community).

103 Larson v. Valente, 456 U.S. 228, 230, 244, 246–47 (1982) (concluding that a rule that draws such a line “clearly grants denominational preferences of the sort consistently and firmly deprecated in our precedents”).

104 Carson v. Makin, 142 S. Ct. 1987, 1997 (2022) (holding that exclusion of religious institutions from generally available funding programs is generally unconstitutional).

105 Doe v. Public Citizen, 749 F.3d 246, 263 (4th Cir. 2014).

106 See, e.g., Maloney v. Murphy, 984 F.3d 50, 64 (D.C. Cir. 2020).

107 Doe v. Stegall, 653 F.2d 180, 185 (5th Cir. 1981); see also Public Citizen, 749 F.3d at 273.

108 See, e.g., supra note 13 and accompanying text.

109 Public Citizen, 749 F.3d at 263 (quoting United States v. Cianfrani, 573 F.2d 835, 851 (3d Cir. 1978)); see also Doe v. Del Rio, 241 F.R.D. 154, 156–57 (S.D.N.Y. 2006) (endorsing this view specifically as to pseudonymity); Boggs v. United States, 143 Fed. Cl. 508, 518 (2019) (likewise); T.S.R. v. J.C., 288 N.J. Super. 48, 60 (App. Div. 1996) (likewise).

110 I say and denominations to make clear that this applies to minority Christian subgroups, and not just to Jews, Muslims, and other outright minority religions.

111 See, e.g., Volokh, supra note 1, at 1420–23.

112 Doe v. Dordt Univ., No. 5:19-cv-04082-CJW-KEM, at 2 (N.D. Iowa Mar. 3, 2020).

113 Doe v. U.S. Dep’t of Just., 93 F.R.D. 483, 484 (D. Colo. 1982) (quoting Doe v. Deschamps, 64 F.R.D. 652, 653 (D. Mont. 1974)); see also A.B.C. v. XYZ Corp., 282 N.J. Super. 494, 502 (App. Div. 1995); Doe v. Frank, 951 F.2d 320, 322 (11th Cir. 1992); United States v. Microsoft Corp., 56 F.3d 1448, 1463 (D.C. Cir. 1995); In re Sealed Case, 971 F.3d 324, 326 (D.C. Cir. 2020).

114 Doe v. Blue Cross & Blue Shield United of Wis., 112 F.3d 869, 872 (7th Cir. 1997); see also United States v. Pilcher, 950 F.3d 39 (2d Cir. 2020) (quoting Blue Cross favorably); Doe v. Megless, 654 F.3d 404, 408 (3d Cir. 2011) (same); United States v. Stoterau, 524 F.3d 988, 1013 (9th Cir. 2008) (same); In re Sealed Case, 971 F.3d 324 (D.C. Cir. 2020) (same).

115 Doe v. Village of Deerfield, 819 F.3d 372, 377 (7th Cir. 2016).

116 Gibson v. Pfizer, Inc., No. 3:20-cv-03870, at 2 (N.D. Cal. Oct. 28, 2020).

117 See, e.g., Doe v. Public Citizen, 749 F.3d 246, 259, 261 (4th Cir. 2014).

118 489 U.S. 1, 15 (1989) (lead opin.).

119 Id. at 18 n.8.

120 See supra note 7.

121 Id. at 28 (Blackmun, J., joined by O’Connor, J., concurring in the judgment); see also id. at 25–26 (White, J., concurring in the judgment).