DOE v. Pennsylvania Department of Corrections

IN THE UNITED STATES DISTRICT COURT

FOR THE MIDDLE DISTRICT OF PENNSYLVANIA

JOHN DOE, No. 4:19-CV-01584

Plaintiff, (Judge Brann)

v.

PENNSYLVANIA DEPARTMENT

OF CORRECTIONS, et al.,

Defendants.

MEMORANDUM OPINION

NOVEMBER 1, 2019

Plaintiff is a transgender man bringing claims of discrimination on the basis

of his gender identity against his former employer, the Pennsylvania Department of

Corrections, as well as former supervisors and coworkers in their individual

capacities. Plaintiff moves to proceed in this lawsuit under the pseudonym “John

Doe.”

Federal Rule of Civil Procedure 10(a) typically requires complaints to

“include the names of all the parties,” but courts permit plaintiffs to proceed

pseudonymously in limited circumstances. The plaintiff must show “(1) a fear of

severe harm, and (2) that the fear of severe harm is reasonable.”1 “Examples of areas

where courts have allowed pseudonyms include cases involving ‘abortion, birth

control, transexuality[sic], mental illness, welfare rights of illegitimate children,

AIDS, and homosexuality.’”2 Courts of Appeals have “held that a district court must

balance the need for anonymity against the general presumption that parties’

identities are public information and the risk of unfairness to the opposing party.”3

The United States Court of Appeals for the Third Circuit’s balancing test,

originally laid out by the Eastern District of Pennsylvania in Doe v Provident Life

and Accident Insurance Co4 and adopted by the Court of Appeals in Doe v Megless,5

is unwieldy. It first instructs the trial court to evaluate six factors in favor of

anonymity:

(1) the extent to which the identity of the litigant has been kept

confidential; (2) the bases upon which disclosure is feared or sought to

be avoided, and the substantiality of these bases; (3) the magnitude of

the public interest in maintaining the confidentiality of the litigant’s

identity; (4) whether, because of the purely legal nature of the issues

presented or otherwise, there is an atypically weak public interest in

knowing the litigant’s identities; (5) the undesirability of an outcome

adverse to the pseudonymous party and attributable to his refusal to

pursue the case at the price of being publicly identified; and (6) whether

the party seeking to sue pseudonymously has illegitimate ulterior

motives.6

Next, the court is to weigh three additional factors disfavoring anonymity:

(1) the universal level of public interest in access to the identities of

litigants; (2) whether, because of the subject matter of this litigation,

the status of the litigant as a public figure, or otherwise, there is a

2 Megless, 654 F3d at 408, quoting Doe v Borough of Morrisville, 130 FRD 612, 614 (ED Pa

1990).

3 Wagner v Holtzapple, Case No 13-cv-3051, at 4 (MD Pa July 10, 2014), quoting Does I thru

XXIII v Advanced Textile Corp, 214 F3d 1058, 1067 (9th Cir 2000).

4 176 FRD 464 (ED Pa 1997).

5 654 F3d 404 (3d Cir 2011).

particularly strong interest in knowing the litigant’s identities, beyond

the public’s interest which is normally obtained; and (3) whether the

opposition to pseudonym by counsel, the public, or the press is

illegitimately motivated.7

Despite their length, we are told these factors are nonexhaustive.8

While appropriate in some circumstances, the flaws of multifactor balancing

tests are well documented—even well-crafted multifactor tests can be difficult to

apply, difficult to predict, and invite needless litigation.9 And the Megless factors

are not the crown jewels of multifactor tests. To start, they are hopelessly imprecise

and redundant. For instance, the “universal level of public interest in access to the

identities of litigants,” whether “there is a particularly strong interest in knowing the

litigant’s identities,” “the magnitude of the public interest in maintaining the

confidentiality of the litigant’s identity,” and whether “there is an atypically weak

public interest in knowing the litigant’s identities” are befuddlingly four separate

factors. Surely, if there is a “particularly strong interest” in knowing the litigant’s

identities, then there is not an “atypically weak” public interest, which itself tells us

something about the “magnitude” and “universal level” of the public interest in

7 Id.

8 Id.

9 See, for example, Jerome B. Grubart, Inc v Great Lakes Dredge & Dock Co, 513 US 527, 547

(1995) (rejecting a proposed multifactor test because it “would be hard to apply, jettisoning

relative predictability for the open-ended rough-and-tumble of factors, inviting complex

argument in a trial court and a virtually inevitable appeal”); Antonin Scalia, The Rule of Law as

a Law of Rules, 56 U Chi L Rev 1175, 1179–80 (1989) (criticizing balancing tests’ lack of

maintaining the litigant’s confidentiality. These inquiries, and others, meander and

criss-cross into each other’s paths, to the extent they differ at all.

What’s more, the test does not provide what weight each enumerated factor

should be given, let alone how unenumerated factors should tip the balance. One

could hardly be blamed for thinking they carry equal gravity. But some are clearly

central—Is there a substantial basis for fear?—while others are secondary or

infrequently implicated—Do the parties have illegitimate motives in seeking or

opposing pseudonymity? The inclusion of so many factors flattens the seeming

importance of each.10

This morass not only adds pages to briefs, it obscures the heart of the inquiry:

Does the Plaintiff risk severe harm by proceeding under his or her real name? And,

if so, is this risk outweighed by a particularly strong public interest in knowing the

Plaintiff’s identity?11 The Megless factors may assist by providing a compendium of

possible considerations, but this Court will focus on these two questions.

10 Indeed, while I will not criticize my colleagues for correctly applying the law given to them, I

simply note that opinions applying Megless and similar tests from other circuits frequently read

as a rote recitation of factors with a conclusion tacked on the end. This style is not conducive

to the reader scrying which factors were determinative in the court’s decision. Or, perhaps more

troublingly, the court may in fact have treated all the factors as coequal.

11 See Megless, 654 F3d at 409 (“[E]ach court agrees that the purpose of the balancing test is to

allow a district court to determine whether a litigant has a reasonable fear of severe harm that

outweighs the public’s interest in open litigation.”). That this is the heart of the inquiry can also

be seen from the Megless factors themselves, if we do a little sifting. Factor 2 asks the threshold

question of what harm is feared and whether that fear is reasonable. Factors 1 and 6, his prior

identity disclosures and his motives, examine the plaintiff’s credibility for claiming fear, and

Disfavoring Factor 3 examines the defendant’s credibility in opposing the motion. The

Courts have long recognized that the harms arising from disclosing a person’s

transgender status are among those that make protection by pseudonym

appropriate.12 As litigation around transgender issues has increased, so too have

more courts permitted transgender plaintiffs to proceed pseudonymously.13 The

Third Circuit in Megless specifically listed it as an area where courts have allowed

pseudonyms.14 As the United States Court of Appeals for the Second Circuit has

succinctly summarized, “[T]he excruciatingly private and intimate nature of

transsexualism, for persons who wish to preserve privacy in the matter, is really

beyond debate.”15

The Plaintiff in the case at bar faces these harms. Litigating under his real

name would out his transgender status to the world. He avers that up to this day he

the nature of the legal issues at stake (Factor 4) or whether the plaintiff is a public figure

(Disfavoring Factor 2).

12 See, for example, Borough of Morrisville, 130 FRD at 614; Doe v McConn, 489 F Supp 76, 76

(SD Tex 1980) (permitting transgender plaintiffs to proceed pseudonymously); Doe v Blue

Cross & Blue Shield of Rhode Island, 794 F Supp 72, 74 (DRI 1992) (same).

13 See, for example, Doe v Gardens for Memory Care at Easton, Case No 18-4027, Doc 3 (ED Pa

Sept 21, 2018); Doe v Parx Casino, Civ A No 18-5289, Doc 4 (ED Pa Jan 2, 2019) (permitting

lesbian plaintiff with a masculine gender expression to proceed under a pseudonym); Foster v

Andersen, 2019 WL 329548, at *2 (D Kan Jan 25, 2019); Doe v City of Detroit, 2018 WL

3434345, at *3 (ED Mich July 17, 2018).

14 See Megless, 654 F3d at 408 (“Examples of areas where courts have allowed pseudonyms

include cases involving ‘abortion, birth control, transexuality[sic], mental illness, welfare rights

of illegitimate children, AIDS, and homosexuality.’”). The Third Circuit has also recognized

the significance of a person’s privacy interest in the closely related context of sexual orientation,

which the court stated was “an intimate aspect of his personality entitled to privacy protection”

and found that forced disclosure of it violated the Fourth Amendment. See Sterling v Borough

of Minersville, 232 F3d 190, 196 (3d Cir 2000).

15 Powell v Schriver, 175 F3d 107, 111 (2d Cir 1999). See also Whitaker v Kenosha Unified School

District No 1 Board of Education, 858 F3d 1034, 1051 (7th Cir 2017) (“There is no denying

that transgender individuals face discrimination, harassment, and violence because of their

has kept his transgender status a closely guarded secret, disclosing it only as

necessary to comply with legal requirements. Divulging it in connection with this

complaint, which alleges harassment arising from intensely personal subject matter,

poses a particularly high risk of severely compromising the privacy the Plaintiff has

labored to preserve.

The public interest in knowing the Plaintiff’s identity is relatively low. He is

not a public figure. I further note that he makes a limited request, moving only to

have his name and address concealed. This means that, unlike a sealed proceeding,

the public will have access to everything else in the proceedings as they would in a

typical case. Given that, the public interest in monitoring the courts will not be not

greatly harmed.16 There is no risk of prejudice to the opposing parties because the

Plaintiff has agreed to fully participate in discovery and depositions. Neither the

public nor the parties have a strong interest in publicly revealing the Plaintiff’s name.

Because the Plaintiff faces a significant risk of severe harm to his privacy, and

the public interest in knowing his identity does not outweigh that risk, I conditionally

grant the Plaintiff’s motion to proceed under the pseudonym “John Doe.” Should

information come to light that would justify this Court revisiting this ruling,

Defendants may raise it at a later date.

16 See Lior Jacob Strahilevitz, Pseudonymous Litigation, 77 U Chi L Rev 1239, 1246–47 (2010)

(explaining why pseudonymous litigation does not meaningfully harm the public’s ability to

An appropriate Order follows.

BY THE COURT:

s/ Matthew W. Brann

Matthew W. Brann

United States District Judge