1:20-cv-00731-LMB-JFA) NATIONAL ASSOCIATION OF IMMIGRATION JUDGES, affiliated with… v. Sirce E. Owen

FILED: November 20, 2025
PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
No. 23-2235
(1:20-cv-00731-LMB-JFA)
NATIONAL ASSOCIATION OF IMMIGRATION JUDGES, affiliated with the
International Federation of Professional and Technical Engineers,
Plaintiff – Appellant,
v.
SIRCE E. OWEN, in her official capacity as Acting Director of the Executive Office
for Immigration Review,
Defendant – Appellee.
O R D E R
The Court denies the petition for rehearing en banc.
A requested poll of the Court failed to produce a majority of judges in regular active
service and not disqualified who voted in favor of rehearing en banc. Judges Wilkinson,
King, Gregory, Wynn, Thacker, Harris, Heytens, Benjamin, and Berner voted to deny
rehearing en banc. Chief Judge Diaz and Judges Niemeyer, Agee, Richardson,
Quattlebaum, and Rushing voted to grant rehearing en banc.

-- 1 of 36 --

2
Judge Wilkinson wrote an opinion concurring in the denial of rehearing en banc.
Judge King wrote an opinion concurring in the denial of rehearing en banc. Judge Thacker,
with whom Judge King joined, wrote an opinion concurring in the denial of rehearing en
banc. Judge Quattlebaum, with whom Judges Agee, Richardson, and Rushing joined,
wrote an opinion dissenting from the denial of rehearing en banc.
Entered at the direction of Judge Berner.
For the Court
/s/ Nwamaka Anowi, Clerk

-- 2 of 36 --

3
WILKINSON, Circuit Judge, concurring in the denial of rehearing en banc:
Notwithstanding my reservations with the panel opinion, I vote to deny rehearing
this case en banc. Mere disagreement with the merits of a panel’s decision is seldom
sufficient grounds for deviating from our normal respect for panel adjudication.
I.
A rehearing en banc represents a departure from our standard procedures—a
departure of an “extraordinary nature” that is heavily disfavored. Fed. R. App. P. 40 &
advisory committee’s notes to the 2024 amendment. “The decision to grant en banc
consideration is unquestionably among the most serious non-merits determinations an
appellate court can make, because it may have the effect of vacating a panel opinion that
is the product of a substantial expenditure of time and effort by three judges and numerous
counsel.” Bartlett ex rel. Neuman v. Bowen, 824 F.2d 1240, 1242 (D.C. Cir. 1987)
(Edwards, J., joined by Wald, C.J., Robinson, Mikva & R.B. Ginsburg, JJ., concurring in
the denials of rehearing en banc); see also Mitt v. Bagley, 626 F.3d 366, 370 (6th Cir. 2010)
(Sutton, J., joined by Kethledge, J., concurring in the denial of rehearing en banc).
A fifteen-member en banc court like ours is a bit of an ungainly beast. It rivals even
FDR’s audacious plans for court expansion. Its proceedings too often feature a cacophony
of voices each vying for time, with counsels’ arguments pushed to the periphery. (I should
apologize here and now for being part of the bedlam.) Too frequently en banc proceedings
end with splintered opinions that have little or no educative effect. And these fractures raise
considerable doubt that the en banc opinion will be any more correct than that of the
original panel. Indeed, en banc arguments are regrettably bereft of the more personal

-- 3 of 36 --

4
interactions conducive to good listening and sound decision-making that are possible with
three-, but not fifteen-, member courts.
Rules and regulations on speaking order or time are no answer to the problems
plaguing swollen tribunals. In fact, they can make the whole situation worse. They lend to
en banc hearings a stilted and artificial quality, quite at odds with the fluency and
spontaneity that panel arguments at their best reflect. And even when en banc proceedings
run smoothly, they constitute an “enormous distraction” from our “heavy schedule of brief-
reading, oral arguments, motions work and opinion-writing in connection with cases on the
regular calendar.” Bartlett, 824 F.2d at 1243.
Judicial resources like those of many institutions are finite. Over-lavishing attention
on one case can mean shortchanging another. Redirecting inordinate time and resources to
hear a case for a second time does nothing but lengthen the line for all the other litigants
waiting to have their first hearing. Litigation itself can come to seem unending. It is for
many litigants, and even perhaps a lawyer or two, an exhausting ordeal. An en banc
proceeding can seem to extend the lifespan of an already elderly case. I do not understand
the view that every assertedly “wrong” panel decision is perforce intolerably so. If only
law was that clear-cut. In short, few should dispute Justice Frankfurter’s observation that
“[r]ehearings are not a healthy step in the judicial process,” and should not be considered
a “normal procedure.” W. Pac. R. Corp. v. W. Pac. R. Co., 345 U.S. 247, 270 (1953)
(Frankfurter, J., concurring in the judgment).
My fine colleague Judge Thacker makes the legitimate point that there were often
more en bancs in the past than there are today. Conc. Op. at 14–15. But the pertinent

-- 4 of 36 --

5
question is whether that state of affairs was desirable. The trend away from en bancs, for
which I like to think I have worked, reflects a dissatisfaction with the practice’s overuse. I
write, in part, because I worry we are reverting to our former bad habits. In the past three
months alone, we have conducted five en banc polls. The fact that these polls have not
carried the day does not eliminate the danger that resort to en banc proceedings may
become all too facile a practice.
This is not to say that rehearings en banc have no place in our judicial system.
Indeed, the Rules provide for them. Such proceedings do possess occasional value in
ironing out intra-circuit conflicts and addressing questions of exceptional importance. See
Fed. R. App. P. 40(b)(2). It is, however, a matter of degree. Judging from the gist of my
concurring colleague’s view, en banc proceedings are to be welcomed, while the Rules
provide precisely the opposite. Fed. R. App. P. 40 (indicating that “rehearing en banc is
not favored”). Given the Rules, it would often seem more suitably the province of the
Supreme Court to decide which questions are sufficiently exceptional to require additional
review; that is the purpose of certiorari, after all.
As Judge Oakes has noted, some cases are “too important to en banc.” James Oakes,
Personal Reflections on Learned Hand and the Second Circuit, 47 Stan. L. Rev. 387, 392
(1995). By subjecting litigation to the en banc detour, we shield it from Supreme Court
review while the often year-long proceeding plays itself out. Shielding significant cases
from the Supreme Court for prolonged periods can have deleterious consequences. We
would often be wiser to “speed” such cases “on [their] way to the Supreme Court as an
exercise of sound, prudent and resourceful judicial administration.” Eisen v. Carlisle &

-- 5 of 36 --

6
Jacquelin, 479 F.2d 1005, 1021 (2d Cir. 1973) (Kaufman, J., joined by Friendly, C.J.,
Feinberg, Mansfield & Mulligan, JJ., concurring in the denial of rehearing en banc). While
the argument can be fairly made that cases need to marinate or incubate in the courts of
appeal, there is only so much benefit to be derived from chewing the bone of an issue
indefinitely. The Supreme Court is well-equipped on its own to resolve the competing
arguments on a question. The pathway to the Court reinforces the wise tradition of vertical,
rather than horizontal, judicial review.
Considering a case’s import also invites subjectivity: “one judge’s case of
exceptional importance is another judge’s routine or run-of-the-mill case.” Bartlett, 824
F.2d at 1242 (internal quotation marks omitted). Ordinarily en banc requests arise from
divided panels. Concerns of subjectivity become all the more exacerbated when, as here,
we question the work of a unanimous panel. The agreement of three colleagues should be
accorded substantial weight. Given the volume and variety of cases before our court, any
one of us can find a case at any time to which we might register strong objections and
ascribe to it “exceptional importance.” If we were to request an en banc poll in such cases,
I fear it would exact a terrible drain upon our judicial resources and our “sound, collegial
attitude.” Air Line Pilots Ass’n Int’l v. E. Air Lines, Inc., 863 F.2d 891, 925 (D.C. Cir.
1988) (R.B. Ginsburg, J., concurring in the denial of rehearing en banc).
Requests for polls are always framed as exceptional, but there comes a point in
which the exceptional becomes more and more the rule, and the practice more and more to
our collective detriment. I cannot hurry to make myself party to such a trend or practice. I
recognize it is best for judges “never to say never,” lest some case from far-off lands

-- 6 of 36 --

7
appears to rebuke the principle just announced. But this scenario should be a rare occasion.
We would benefit from allegiance to sound procedure even during those times when it
seems less convenient to do so. At their best, standard procedures are neutral, thereby
keeping substantive disputes year in and year out within mutually accepted bounds.
II.
Respectfully, I do not agree with the panel opinion in this case. I do not believe that
our court enjoys the prerogative to decide whether a particular statutory program is
“functioning as Congress intended.” Nat’l Ass’n of Immigr. Judges v. Owen, 139 F.4th 293,
304 (4th Cir. 2025). This “functionality” test plants the seeds of real mischief to which I
think only the Supreme Court can bring an effective halt.
Put simply, it is not our job to amend Congress’s handiwork. If a statutory scheme
is not functioning as Congress intended, then it would seem logical that Congress be the
one to fix the problem. Such matters require legislative attention, not judicial correction.
And while the panel takes the seemingly modest step of remanding the case for judicial
fact-finding, id. at 308, it is legislative oversight and fact-finding that is in order.
Furthermore, if a statute has a functionality problem, it will often be because of
some alleged malfunction in executive enforcement. The malfunction here is assumed to
be the President’s removal of the Special Counsel and members of the Merit Systems
Protection Board. Id. at 305–07. The lawfulness of those removals has yet to be resolved
and is the subject of ongoing litigation. See id. at 307. That litigation would seem the most
suitable way of addressing the problem. At the very least, it is premature to revisit the
functionality of the statutory scheme when the statute’s removal protections have not been

-- 7 of 36 --

8
held unconstitutional. Cf. United States v. Arthrex, Inc., 594 U.S. 1, 23–25 (2021)
(considering the issue of severability only after part of the statute had been held
unconstitutional).
In sum, functionality is a vague and impenetrable standard. What is dysfunctional
should often be the subject of political debate. Functionality assessments are too frequently
untethered, as I see it, to any principle other than judicial preference. Left untamed,
functionality will vest the judiciary with a general supervisory authority over both the
legislative and executive branches. I hope this test will be consigned to strong disfavor—
an emphasis only our highest court can suitably supply.

-- 8 of 36 --

9
KING, Circuit Judge, joining in the concurrence of Judge THACKER in the denial of
rehearing en banc, and separately concurring in the denial of rehearing en banc:
I join fully in the concurrence of my good colleague Judge Thacker in the denial of
rehearing en banc in this appeal. I write separately to emphasize once more that it is the
applicable Rules of Appellate Procedure that specify the requirements for en banc
proceedings. See Fed. R. App. P. 40(c) (explaining when rehearing en banc may be
ordered). If those conditions are satisfied — no matter how difficult the dispute might be
— our Court should be willing to address and resolve that case en banc. As Judge Thacker
properly recognizes, it is part of our job description to do so! Accord Dubin v. United
States, 599 U.S. 110, 132 n.10 (2023) (Sotomayor, J.) (recognizing that “resolving hard
cases is part of the judicial job description”); Doe v. Va. Dep’t of State Police, 720 F.3d
212, 214 n.* (4th Cir. 2013) (King, J., dissenting from denial of rehearing en banc and
highlighting that federal judges must “remain faithful to [their] constitutional charge to
decide cases and controversies as they are presented”).
Simply put, rehearing en banc is not at all warranted here. But Judge Thacker’s
view of our en banc procedures is “spot on” and absolutely correct.

-- 9 of 36 --

10
THACKER, Circuit Judge, with whom Judge KING joins, concurring in the denial of
rehearing en banc:
I vote to deny rehearing en banc in this case. In that regard, I agree with my good
colleague Judge Wilkinson that denial is the appropriate vote here. We part ways, however,
when it comes to the basis for that vote. Unlike Judge Wilkinson, I am in complete
agreement with the panel opinion.
Additionally, I write separately to reject what I view as a fallacy offered by Judge
Wilkinson’s concurring opinion. What I take from the overriding tone and tenor of his
opinion is that the mere act of holding an en banc proceeding is somehow untoward and
that we run the risk of doing so too frequently. Neither is accurate. Therefore, in the
interest of institutional integrity, I am compelled to push back.
Holding an en banc hearing is not some rogue act. It is provided for by the rules.
En banc proceedings have been authorized since the Judicial Code of 1948 and have been
a part of the Federal Rules of Appellate Procedure since 1968. Rule 40(b)(2) of the Federal
Rules of Appellate Procedure provides the mechanism for an en banc proceeding. Such
proceedings may be held if a majority of the members of the court who are in regular active
service conclude that doing so is necessary to maintain uniformity across cases and/or that
the question at issue is one of exceptional importance. Fed. R. App. P. 40(b)(2) and (c).
But, in Judge Wilkinson’s view, “it would often seem more suitably the province of the
Supreme Court to decide which questions are sufficiently exceptional to require additional
review….” J. Wilkinson, Concurring Op. at 5. This approach puts the cart before the horse
and is neither logical nor practical.

-- 10 of 36 --

11
First of all, that is not the rule. To be sure, the Supreme Court has the last word on
legal jurisprudence in our judicial system and resolves conflicts among the courts on
significant issues. But that does not mean that the United States Circuit Courts play no
role at all. The Supreme Court is not the arbiter of what is of sufficient exceptional
importance in order for an appellate court to empanel an en banc court in the first instance.
By rule, that is up to the appellate courts themselves; it is not the province of the Supreme
Court. Additionally, it defies practical reality to suggest that we simply punt important
cases to the Supreme Court. For the last fiscal year, our court disposed of 3,448 cases.
Monthly Statistical Report, United States Court of Appeals for the Fourth Circuit,
https://www.ca4.uscourts.gov/docs/pdfs/publicstats.pdf?sfvrsn=5524a238_324
[https://perma.cc/VX52-2JJE] (last accessed Nov. 13, 2025). And we are but a single court
amongst all of the federal courts that funnel cases to the Supreme Court.
I expect the counter-argument would be that judicial efficiency favors the view that
only the very most significant cases would be reserved for the Supreme Court to resolve
rather than spend time on en banc review by the appellate courts. But that reads the
“questions of exceptional significance” criteria out of Rule 40. And let’s play that theory
out. Suppose a majority of our court is of the view that a panel opinion is incorrect relative
to an exceptional issue. Judge Wilkinson would have the majority of the court take a pass
on en banc review in favor of the Supreme Court deciding the issue—which may never
actually come to pass. The viewpoint of the dissenting voices on the court would then
prevail and an opinion that a majority of the court views as legally wrong would remain
the law for the entire Circuit. That cannot be right.

-- 11 of 36 --

12
Beyond that, in the case at hand, my colleagues’ approach would leave plaintiff, the
National Association of Immigration Judges, with no opportunity for meaningful review
of its claim that the constitutional rights of its members are being violated. Claims for
violation of rights arising under the United States Constitution are typically brought in
federal court. 28 U.S.C. § 1331. This is true except when Congress makes clear its intent
to strip such jurisdiction from the federal courts through legislation. See Webster v. Doe,
486 U.S. 592, 603 (1988). The Supreme Court in Thunder Basin Coal Co. v. Reich, 510
U.S. 200 (1994), set forth the test to determine whether Congress so intended. Thunder
Basin directs us to look to “the statute’s language, structure, and purpose, its legislative
history, and whether the claims can be afforded meaningful review.” Id. at 207. The panel
opinion does just that, and it remands to the district court to assess whether the CSRA’s
adjudicatory scheme is functioning as Congress intended. We cannot abdicate our duty to
hear constitutional challenges in the pyrrhic hope that Congress will recognize the
dysfunction or lack of independence of the Merit Systems Protection Board or the Office
of Special Counsel and choose to act.
Judge Wilkinson may not like the Rule. And I acknowledge that en banc
proceedings are “not favored.” Fed. R. App. P. 40(c). But it is not an act of anarchy when
we as a court work within the rules to correct what a majority of us believe to be an error.
Nor is doing so an “enormous distraction” from our workload. J. Wilkinson, Concurring
Op. at 2 (quoting Bartlett ex rel. Neumann v. Bowen, 824 F.2d 1240, 1243 (D.C. Circ.
1987)). To the contrary, in my view, it is our job.

-- 12 of 36 --

13
The undergirding of Judge Wilkinson’s point of view appears to be his professed
concern that the exceptional may become the rule. And he opines that he “cannot hurry to
make [himself] party to such a trend or practice.” J. Wilkinson, Concurring Op. at 6. But
he need not worry. There is no such “trend or practice.” Indeed, the reverse is true.
As a court, we dispose of an average of 3,500 cases per year. Monthly Statistical
Report, United States Court of Appeals for the Fourth Circuit,
https://www.ca4.uscourts.gov/docs/pdfs/publicstats.pdf?sfvrsn=5524a238_324
[https://perma.cc/VX52-2JJE] (last accessed Nov. 13, 2025) (indicating the court disposed
of 3,660 cases in the 2023–2024 fiscal year and 3,448 cases in the 2024–2025 fiscal year).
In comparison, for the 2024 calendar year1 we held three en banc hearings2 and for the
2025 calendar year thus far we have held two en banc hearings. Further, if these hearings
are viewed through the lens of a court term3 as opposed to a calendar year, the numbers are
even more lean. Viewed that way, we held one en banc hearing for the 2024–2025 court
1 The Fourth Circuit Court of Appeals website maintains these statistics by calendar
year versus by court term. En Banc Cases, United States Court of Appeals for the Fourth
Circuit, https://www.ca4.uscourts.gov/opinions/en-banc-cases [https://perma.cc/CEB7-
NLTR] (last visited Nov. 13, 2025).
2 Of note, these three cases all shared the common theme of the issue of
constitutionality of firearms regulation. See United States v. Price, 111 F.4th 392 (2024)
(argued March 20, 2024) (authored by J. Wynn); Bianchi v. Brown, 111 F.4th 438 (2024)
(argued March 20, 2024) (authored by J. Wilkinson); Maryland Shall Issue, Inc. v. Moore,
116 F.4th 211 (2024) (argued March 21, 2024) (authored by J. Keenan).
3 As a court, we generally hear oral arguments from September to May (although
we decide cases and write opinions throughout the year). Here, I am considering the 2024–
2025 court term to encompass the period of September 1, 2024 to August 31, 2025 and the
2025–2026 court term to include September 1, 2025 to August 31, 2026.

-- 13 of 36 --

14
term4 and one en banc hearing thus far for the 2025–2026 court term.5 Indeed, over the
course of the most recent five years, we have averaged only 3.4 en banc proceedings per
calendar year.6 But in diametric opposition to the trend Judge Wilkinson portends—this
relatively low average of en banc proceedings was not always the norm on this court.
My esteemed colleague was on the court when, during the 1995 calendar year,7 our
court held thirteen en banc proceedings. Thirteen.8 What is more, such proceedings
4 See United States v. Chatrie, 136 F.4th 100 (2025) (argued January 30, 2025) (per
curiam).
5 See American Federation of State, County and Municipal v. Soc. Sec. Admin., 25-
1411 (argued September 1, 2025).
6 The number of en banc proceedings we have held for each of the past five years is
as follows: 2025 – two; 2024 – three; 2023 – three; 2022 – three; 2021 – six. See En Banc
Cases, United States Court of Appeals for the Fourth Circuit,
https://www.ca4.uscourts.gov/opinions/en-banc-cases [https://perma.cc/CEB7-NLTR]
(last visited Nov. 13, 2025).
7 1995 is the first year for which the Fourth Circuit website maintains statistics for
en banc proceedings.
8 These are those thirteen en banc Fourth Circuit cases: Smith v. Virginia
Commonwealth University, 84 F.3d 672 (1996) (authored by J. Chapman) (reversal of grant
of summary judgment in a Title VII pay disparity case); United States v. Barber, 80 F.3d
964 (1996) (authored by J. Niemeyer) (affirming the constitutionality of Section 571 of the
National Defense Authorization Act (commonly known as “Don’t Ask. Don’t Tell”));
Thomasson v. Perry, 80 F.3d 915 (1996) (authored by J. Wilkinson) (affirming jury verdict
convicting Appellants of money laundering cash proceeds from the sale of marijuana);
Stiltner v. Beretta USA Corp., 74 F.3d 1473 (1996) (authored by J. Hamilton) (affirming
the grant of summary judgment to employer on ERISA and state law claims); Gilliam v.
Foster, 75 F.3d 881 (1996) (authored by J. Wilkins) (affirming the grant of habeas relief);
Cochren v. Morris, 73 F.3d 1310 (1996) (authored by J. Wilkinson) (affirming the
dismissal of in forma pauperis complaint); Spicer v. Commonwealth of Virginia, 66 F.3d
705 (1995) (authored by J. Niemeyer) (Title VII case in which the jury verdict against the
plaintiff as to her retaliation claim was affirmed but the district court’s ruling in her favor
following a bench trial as to her sexual harassment claim was reversed); United States v.
Hines, 65 F.3d 392 (1995) (per curiam) (conviction confirmed on all counts); Nasim v.
Warden, Maryland House of Corrections, 64 F.3d 951 (1995) (authored by J. Niemeyer)
(affirming the dismissal of in forma pauperis complaint as frivolous); Berkeley v. Common
(Continued)

-- 14 of 36 --

15
appear to have been held as a matter of course on some relatively run of the mill issues.
But it is now that my colleague is concerned that en banc proceedings may be overused?
For the sake of completeness, having set out above the number of en banc
proceedings for the most recent five year period (a total of 17) as well as the average for
that period (3.4 per year), I do the same for the five year period from 1995 to 1999. Over
the course of that period, we held a total of 55 en banc proceedings, for an average of eleven
per calendar year.9
Thus, with all due respect to my good colleague Judge Wilkinson, I am compelled
to complete the record on the state of en banc proceedings in our court, lest his viewpoint
have an unintended chilling effect. Judge Wilkinson is, of course, entitled to his view. But
I am entitled to mine as well.
Council City of Charleston, 63 F.3d 295 (1995) (authored by J. Luttig) (holding that a
municipality is not immune from liability under Section 1983 for the enactments and
actions of the local legislative body); United States v. Langley, 62 F.3d 602 (1995)
(authored by J. Hamilton) (affirming convictions for making a false statement to a federally
licensed firearms dealer and possession of a firearm after having been convicted of a
felony); Pinder v. Johnson, 54 F. 3d 1169 (1995) (authored by J. Wilkinson) (granting
qualified immunity to law enforcement officer holding that failure to safeguard was not a
clearly established due process right); Hardester v. Lincoln Nat. Life Ins. Co., 52 F.3d 70
(1995) (per curiam) (affirming grant of summary judgment to insureds in dispute over
health insurance coverage provided under an ERISA welfare benefit plan).
9 The number of en banc proceedings we held each year from 1995 to 1999 was:
1995 – thirteen; 1996 – fourteen; 1997 – thirteen; 1998 – seven; 1999 – eight. See En Banc
Cases, United States Court of Appeals for the Fourth Circuit,
https://www.ca4.uscourts.gov/opinions/en-banc-cases [https://perma.cc/CEB7-NLTR]
(last visited Nov. 13, 2025).

-- 15 of 36 --

16
QUATTLEBAUM, Circuit Judge, with whom Judges AGEE, RICHARDSON, and
RUSHING join, dissenting from the denial of rehearing en banc:
As framed by the panel opinion, this appeal turns on a single question—is it “fairly
discernible” from the Civil Service Reform Act of 1978 (CSRA) that Congress intended to
preclude district courts from hearing claims like those brought in this case? See Nat’l Ass’n
of Immigr. Judges v. Owen, 139 F.4th 293, 304 (4th Cir. 2025) (NAIJ II). For an inferior
court, the answer to that question is unquestionably yes. Finding this answer doesn’t take
any heavy lifting. The Supreme Court has already given us the answer. Not subtly. Not by
implication. No, the Supreme Court has told us twice—emphatically and directly—that
district courts lack jurisdiction over claims like the ones the National Association of
Immigration Judges (NAIJ) asserts here. See Elgin v. Dep’t of Treasury, 567 U.S. 1, 11–
12 (2012) (“Given the painstaking detail with which the CSRA sets out the method for
covered employees to obtain review of adverse employment actions, it is fairly discernible
that Congress intended to deny such employees an additional avenue of review in district
court.”); United States v. Fausto, 484 U.S. 439, 447 (1988) (“[T]he absence of provision
for these employees to obtain judicial review is . . . manifestation of a considered
congressional judgment that they should not have statutory entitlement to review for
adverse action of the type governed by Chapter 75.”). In fact, the answer is so clear that
the NAIJ does not even contest the issue. It concedes that Elgin resolves the question. See
Op. Br. at 17.
Even so, the panel decided on its own—without any party raising the issue and
without requesting supplemental briefing—that actions of the current administration after

-- 16 of 36 --

17
oral argument and a recent Supreme Court stay decision have made Congress’ intent less
clear. To arrive at that conclusion was no small feat—the panel not only had to disregard
party presentation principles; it also had to create a new test. Instead of applying the test
that the Supreme Court told inferior courts to apply in Thunder Basin Coal Co. v. Reich,
510 U.S. 200, 207 (1994), the panel added a new consideration—whether recent political
and judicial events have caused the CSRA to cease to function as intended. Having raised
that question on its own, the panel attempts to claim that Fausto and Elgin do not control
since those decisions did not answer that specific question. It then vacates the district
court’s decision and remands with instructions to that court to conduct factfinding on
“whether the text, structure, and purpose of the [CSRA] has been so undermined that the
jurisdiction stripping scheme no longer controls.” NAIJ II, 139 F.4th at 300.
By my count, no fewer than three errors underpin the panel’s holding. First, it fails
to adhere to Supreme Court precedent that is directly on point. Second, it usurps Congress’
role by allowing unelected judges to update the intent of unchanged congressional statutes
if the court believes recent political events—like those of the current administration it
cites—alter the operation of a statute from the way Congress intended. And third, it
disregards the principle of party presentation. The result? Now, at least in our circuit, we
are free to set aside Supreme Court precedent, reimagine congressional intent and abandon
our role as a neutral arbiter of the positions presented by the parties if we divine that events
decades after a statute’s passage suggest it is not functioning as originally intended. This
cannot be right.

-- 17 of 36 --

18
Regrettably, our court denied rehearing this appeal en banc. Because in my view the
panel decision conflicts with Supreme Court precedent and involves questions of
exceptional importance, I respectfully dissent.
I.
A. The Civil Service Reform Act
I begin with an overview of the relevant statutory scheme. The CSRA
“comprehensively overhauled the civil service system.” Lindahl v. Off. of Pers. Mgmt., 470
U.S. 768, 773 (1985). One aspect of the CSRA was the creation of a “new framework for
evaluating adverse personnel actions against [federal] ‘employees’ and ‘applicants for
employment.’” Id. at 774. This was intended to centralize and streamline “administrative
and judicial review of personnel action.” Fausto, 484 U.S. at 444.
From this, two agencies emerged—the Office of Personnel Management (OPM) and
the Merit Systems Protection Board (MSPB). 5 U.S.C. §§ 1101, 1204. The OPM is the
implementing arm, responsible for enacting rules and regulations under the CSRA to
protect federal workers. See 5 U.S.C. §§ 1101–05. The MSPB, on the other hand, is the
quasi-judicial body charged with protecting and enforcing what the OPM implements. See
5 U.S.C. § 1204. It is vested with authority to adjudicate disputes arising under the CSRA.
See id. As the panel opinion notes, this includes hearing disputes regarding “federal
employees’ allegations that their government employer discriminated against them,
retaliated against them for whistleblowing, violated protections for veterans, or otherwise
subjected them to an unlawful adverse employment action or prohibited personnel
practice.” NAIJ II, 139 F.4th at 302 (citation omitted). The MSPB is composed of three

-- 18 of 36 --

19
members appointed by the President with the advice and consent of the Senate. 5 U.S.C. §
1201. And only two may be from the same political party. Id. A member serves a term of
seven years and can only be removed for inefficiency, neglect of duty or malfeasance in
office. Id. § 1202.1
The CSRA also established the role of Special Counsel. Id. § 1211(a). One of the
Special Counsel’s jobs is to receive and investigate allegations of prohibited personnel
practices against the federal government. See id. § 1212. If the Special Counsel finds
“reasonable grounds to believe that a prohibited personnel practice has occurred, exists, or
is to be taken,” it must report such determination together with any findings or
recommendations to the MSPB, the agency involved in the employment dispute and the
OPM. Id. § 1214(b)(2)(B). This report can include recommendations for the MSPB to take
corrective action. Id. If the agency fails to address the practice, the Special Counsel can
petition the MSPB for corrective action. Id. § 1214(b)(2)(C).
1 As noted below, the constitutionality of these for-cause removal provisions and
other similar provisions has recently been called into question. See Trump v. Cook, No.
25A312, 2025 WL 2784699 (October 1, 2025) (deferring ruling on the government’s stay
application pending oral argument in January 2026 in a case considering executive for-
cause removal authority under the Federal Reserve Act); Trump v. Slaughter, No. 25A264,
2025 WL 2692050 (Sept. 22, 2025) (granting stay permitting discharge of members of the
Federal Trade Commission and directing parties to brief and argue whether “statutory
removal protections . . . violate the separation of powers and, if so, whether Humphrey’s
Executor v. United States, 295 U.S. 602 (1935), should be overruled”); Trump v. Boyle,
145 S. Ct. 2653, 2654 (2025) (granting stay permitting discharge of members of the
Consumer Product Safety Commission relying on Trump v. Wilcox, 145 S. Ct. 1415 (2025)
because the CPSC exercises executive power in a manner similar to the National Labor
Relations Board); Wilcox, 145 S. Ct. at 1415 (granting stay permitting removal of members
of the National Labor Relations Board and the MSPB because the President “may remove
without cause executive officers who exercise that power on his behalf, subject to narrow
exceptions . . .” ).

-- 19 of 36 --

20
The Special Counsel is appointed by the President with the advice and consent of
the Senate for a five-year term. Id. § 1211(b). Similar to MSPB members, the President
may remove the Special Counsel “only for inefficiency, neglect of duty, or malfeasance in
office.” Id.
“Three main sections of the CSRA govern personnel action taken against members
of the civil service.” Fausto, 484 U.S. at 445. But only two are relevant in this case—
Chapter 23 and Chapter 75. See 5 U.S.C. §§ 2301 et seq., 7501 et seq.
Chapter 23 describes the merit system of employment, forbidding an agency from
engaging in “‘prohibited personnel practices,’ including unlawful discrimination, coercion
of political activity, nepotism, and reprisal against so-called whistleblowers.” Fausto, 484
U.S. at 446 (quoting 5 U.S.C. § 2302). Employees alleging violations under this Chapter
“are given the right to file charges of ‘prohibited personnel practices’ with the Office of
Special Counsel of the MSPB . . . .” Id. (citation omitted). As outlined above, the Special
Counsel then investigates, notifies the agency, the MSPB and the OPM, and takes such
action as authorized under the statute. See 5 U.S.C. § 1214(b)(2). “Any employee or
applicant for employment adversely affected or aggrieved by a final order or decision of
the [MSPB] may obtain judicial review of the order or decision” in the United States Court
of Appeals of the Federal Circuit. Id. §§ 7703(a)–(b).
Chapter 75 covers adverse actions against employees for “efficiency of the service.”
Fausto, 484 U.S. at 446. It has two subchapters that delineate between minor and major
adverse actions. Id. at 446–47. Under the second subchapter, an employee against whom a
major adverse action is proposed must be afforded: (1) 30 days’ written notice of the action

-- 20 of 36 --

21
with the agency’s reasoning (absent limited circumstances), (2) a reasonable time to
respond, (3) the right to have an attorney represent him and (4) a written agency decision
with reasoning at the earliest practicable date. 5 U.S.C. § 7513(b). From there, an employee
is “entitled to appeal to the [MSPB] . . . ” and then the United States Court of Appeals for
the Federal Circuit. Id. §§ 7513(d), 7703(a)–(b).
So, the takeaway here is that under the CSRA, an aggrieved employee must proceed
through the administrative procedures in Chapter 23 or Chapter 75—chiefly through the
MSPB and the United States Court of Appeals for the Federal Circuit. The only listed
exception is § 7703(b)(2), which applies to district court review of adverse MSPB decisions
concerning discrimination under the Civil Rights Act, the Age Discrimination in
Employment Act and the Fair Labor Standards Act.
B. The Challenged Speech Policy, the NAIJ and this Lawsuit
With that background in mind, I turn to the facts and history of this case. The
Executive Office of Immigration Review (EOIR) is a separate agency within the
Department of Justice tasked with adjudicating immigration cases under authority
delegated from the Attorney General. To carry out its purposes, the EOIR employs
hundreds of immigration judges. In October of 2021, the EOIR established a personnel
policy requiring those immigration judges to obtain prior approval before any official
speech.2 Where the judge is “invited to participate in an event because of their official
2 There are previous iterations of the policy. See Nat’l Ass’n of Immigr. Judges v.
Neal, 693 F. Supp. 3d 549, 557–58 (E.D. Va. 2003) (NAIJ I). For the sake of brevity, I do
not discuss them here.

-- 21 of 36 --

22
position, is expected to discuss agency policies, programs, or a subject matter that directly
relates to their official duties, or otherwise appear on behalf of the agency, . . . [it] will be
considered in an official capacity.” J.A. 57. In determining whether a speech is “official,”
supervisors are to consider a host of relevant factors such as the nature and purpose of the
engagement, the host and sponsors and the appropriateness of the forum for the speech.
See J.A. 57. To the extent a speech is official, the supervisor, with input from the speaking
engagement team, ultimately “make[s] the final decision concerning approval or denial of
the request and inform[s] the employee of the supervisor’s decision,” J.A. 59.
The NAIJ is a non-profit voluntary association of immigration judges, which
includes members who are required to comply with EOIR’s speech policy. Although it did
not allege a discrimination action, which would have been excepted from the CSRA’s
administrative procedures, the NAIJ still sued the Director of the EOIR in the United States
District Court for the Eastern District of Virginia on behalf of its members contending that
the speech policy violates the First Amendment by “impos[ing] an unconstitutional prior
restraint on the speech of federal immigration judges.” J.A. 13. The NAIJ asserted “[t]he
policy bans judges from sharing their private views on immigration law or policy issues,
or about the agency that employs them. Judges who violate the policy face a range of
disciplinary sanctions, including reprimand, suspension, and even removal from the federal
service.” J.A. 13. And the NAIJ alleged the speech policy is “void for vagueness under the
First and Fifth Amendments because it invites arbitrary and discriminatory enforcement,
and . . . fails to give immigration judges fair notice of what standards will be applied in
reviewing their requests for preapproval.” J.A. 36.

-- 22 of 36 --

23
After determining that the NAIJ had standing, the district court analyzed the CSRA
under the two-part test established in Thunder Basin to determine whether “Congress
intended to divest district courts of jurisdiction.” Nat’l Ass’n of Immigr. Judges v. Neal,
693 F. Supp. 3d 549, 569 (E.D. Va. 2003) (NAIJ I). At step one, it considered whether it
was “fairly [discernible] from the CSRA’s scheme that Congress intended to preclude
district-court jurisdiction over certain covered actions brought by covered federal
employees.” Id. (citing Elgin, 567 U.S. at 11–12). The clear answer—yes. See id.
Next, the district court turned to step two to evaluate whether NAIJ’s claim is “of
the type Congress intended to be reviewed within this statutory structure.” Id. (quoting
Bennett v. U.S. Sec. & Exch. Comm’n, 844 F.3d 174, 181 (4th Cir. 2016) (internal quotation
mark omitted)). After examining the possibility of meaningful judicial review, whether
NAIJ’s claims were “wholly collateral” to the statute’s review provisions and whether
agency expertise would bear on the case, the district court determined that “Congress
intended the CSRA scheme to preclude district court jurisdiction over plaintiff’s challenge
to the 2021 policy.” Id. at 581. NAIJ timely appealed, and we have jurisdiction under 28
U.S.C. § 1291 and 28 U.S.C. § 1331.
On appeal, the NAIJ took no issue with the district court’s analysis of step one of
Thunder Basin. In fact, it conceded that Supreme Court precedent foreclosed this issue.
Op. Br. at 16–17 (“Under the first step, courts must ask whether Congress’s intent to
preclude district court jurisdiction is fairly discernible in the statutory scheme. The
Supreme Court has already held that such intent is manifest in the CSRA . . . .” (cleaned

-- 23 of 36 --

24
up)). Rather, it only argued that the district court erred at step two. Id. at 17 (“[O]nly the
second step of the Thunder Basin inquiry is at issue here.”).
Despite that concession, the panel decided on its own that the district court should
look again at step one to consider the “[s]erious questions [that] have recently arisen
regarding the functioning of both the MSPB and the Special Counsel.” NAIJ II, 139 F.4th
at 313. In other words, the panel decided that, if the district court determines that actions
by the current administration and a recent stay order from the Supreme Court prevent the
CSRA from functioning as Congress originally intended, the district court can reconsider
Congress’ intent and disregard Fausto and Elgin.
We cannot do that. We must follow Supreme Court precedent. See State Oil Co. v.
Khan, 522 U.S. 3, 20 (1997) (noting that even where a Supreme Court precedent contains
many “infirmities” and rests on “wobbly, moth-eaten foundations,” it remains the Supreme
Court’s “prerogative alone to overrule one of its precedents” (quoting Khan v. State Oil
Co., 93 F.3d 1358, 1363 (7th Cir. 1996))). Nor should courts reconsider the intent of a
statute that has not changed because they perceive that statute is not functioning as
intended. Finally, a dramatic ruling like the panel opinion would be dubious under any
circumstances, but to issue such an opinion without any party raising those issues and
without ordering any supplemental briefing magnifies the mistake.
II.
A. Supreme Court Precedent
Supreme Court precedent should have made easy work of this case. District courts
have jurisdiction “of all civil actions arising under the Constitution, laws, or treaties of the

-- 24 of 36 --

25
United States.” 28 U.S.C. § 1331. Congress, however, may “impliedly preclude jurisdiction
by creating a statutory scheme of administrative adjudication and delayed judicial review
in a particular court.” Bennett, 844 F.3d at 178 (citing Thunder Basin, 510 U.S. at 207).
So, the question is then how to determine when a district court is divested of jurisdiction
by a statutory scheme.
Thunder Basin tells us. It established a two-part test for determining when a district
court is divested of jurisdiction. At step one, a court must ask whether Congress’ intent to
preclude district court jurisdiction over claims within the statute’s scope was “fairly
discernible in the statutory scheme.” Thunder Basin, 510 U.S. at 207 (quoting Block v.
Cmty. Nutrition Inst., 467 U.S. 340, 351 (1994)). This requires looking at the “statute’s
language, structure, and purpose, [and] its legislative history . . . .” Id. (citing Block, 467
U.S. at 345).3 At step two, the court looks to whether the claim “can be afforded meaningful
3 As a matter of first principles, I have some concerns about Thunder Basin.
Examining a statute’s text seems like a better way to determine whether Congress has
precluded federal courts from hearing certain claims than using judge-made multi-factor
tests. See Axon Enter., Inc., v. Fed. Trade Comm’n, 598 U.S. 175, 209 (2023) (Gorsuch, J.,
concurring) (noting that there is a “better way” than Thunder Basin to determine whether
a district court is deprived of jurisdiction, and it “begins with the language of the relevant
statutes, and when the statutory language provides a clear answer, it ends there as well”
(quoting Hughes Aircraft Co. v. Jacobson, 525 U.S. 432, 438 (1999) (cleaned up))). But as
an inferior court, Thunder Basin binds us. That includes its reference to legislative history
as a consideration. But see A NTONIN S CALIA & B RYAN G ARNER , R EADING L AW: T HE
INTERPRETATION OF L EGAL TEXTS 376 (2012) (stating that using “legislative history to
find ‘purpose’ in a statute is a legal fiction that provides great potential for manipulation
and distortion”). In Fausto, the Court did not expressly incorporate legislative history as a
separate factor in its step one preclusion analysis. 484 U.S. at 444 (“The answer is to be
found by examining the purpose of the CSRA, the entirety of its text, and the structure of
review that it establishes.” (citations omitted)). It did, however, cite briefly to legislative
history as part of its purpose analysis. Id. (“A leading purpose of the CSRA was to replace
(Continued)

-- 25 of 36 --

26
review” and is the “type Congress intended to be reviewed within this statutory structure.”
Id. at 207, 212. This examines whether the claim at issue is “wholly collateral to a statute’s
review provisions and outside the agency’s expertise, particularly where a finding of
preclusion could foreclose all meaningful judicial review.” Id. at 212–13 (cleaned up).4
But remember, no party here took issue with the district court’s analysis as to step one of
the CSRA. See Section II.B.3. And that was for good reason—because the Supreme Court
has already decided the issue.
First, in Fausto, the Supreme Court determined the CSRA was designed to foreclose
the United States Claims Court from exercising jurisdiction over employment disputes
involving certain federal employees—nonpreference employees of the excepted service.
484 U.S. at 455.5 Fausto was an employee of the Department of the Interior Fish and
the haphazard arrangements for administrative and judicial review of personnel action, part
of the ‘outdated patchwork of statutes and rules built up over almost a century’ that was
the civil service system.” (quoting S. Rep. No. 95-969, p. 3 (1978), U.S. Code Cong. &
Admin. News 1978, p. 2723)). Likewise, in Elgin, the Court makes no mention of
legislative history as a separate factor to assess. 567 U.S. at 10 (“To determine whether it
is ‘fairly discernible’ that Congress precluded district court jurisdiction over petitioners’
claims, we examine the CSRA’s text, structure, and purpose.” (citations omitted)).
Regardless of the weight afforded legislative history, the Supreme Court has already
decided that it supports the broad preclusive sweep of the CSRA to deprive federal district
courts of jurisdiction.
4 Because I take no issue with the panel opinion’s description of Thunder Basin at
step two, I do not discuss it here.
5 And while Fausto predates the Court’s establishment of the Thunder Basin test,
the Court’s analysis as to the CSRA in Fausto relied on Thunder Basin’s same principles.
See Lindahl, 470 U.S. at 779 (“[T]he question whether a statute precludes judicial review
‘is determined not only from its express language, but also from the structure of the
statutory scheme, its objectives, its legislative history, and the nature of the administrative
action involved.” (quoting Block, 467 U.S. at 345)); Block, 467 U.S. at 345 (“Whether and
to what extent a particular statute precludes judicial review is determined not only from its
(Continued)

-- 26 of 36 --

27
Wildlife Service who was suspended from his job due to unauthorized use of a government
vehicle. Id. at 440–41. He later sought backpay in the United States Claims Court. Id. at
440–41. The Court noted the conduct underlying Fausto’s claim was covered by Chapter
75 of the CSRA because it constituted an adverse action for “efficiency of the service.” Id.
at 446–47. The real question was whether or not Fausto’s classification as a nonpreference
member of the excepted service rendered him eligible for judicial review, rather than
review under the CSRA, because Chapter 75 only addressed “preference eligibles in the
excepted service.” See id. at 446–47. Fausto argued that silence as to nonpreference
members meant the statute’s scheme foreclosing judicial review did not apply to him—and
he was free to proceed in whatever judicial forum he chose. Id. at 449–50. The Supreme
Court rejected this attempt to narrowly construe the CSRA. See id. at 449–51. It
emphasized that the “CSRA established a comprehensive system for reviewing personnel
action taken against federal employees.” Id. at 445. And it found the CSRA intentionally
excluded employees in Fausto’s “service category from the provisions establishing
administrative and judicial review,” thereby preventing Fausto from seeking review in the
Claims Court. Id.
Next, in Elgin, the Supreme Court considered whether the CSRA’s scheme
precluding judicial review applied to constitutional claims in district court. 567 U.S. at 8.
There, petitioners were former federal competitive service employees who were required
under the CSRA to comply with the Military Selective Service Act but failed to do so. Id.
express language, but also from the structure of the statutory scheme, its objectives, its
legislative history, and the nature of the administrative action involved.”).

-- 27 of 36 --

28
at 6–7. As a result, they were “discharged (or allegedly constructively discharged)” from
their positions. Id. at 7. All petitioners, except one, sued in federal district court raising
constitutional challenges to the CSRA. See id. The Supreme Court considered whether the
district court was the proper forum for their claims. See id. at 8. The answer: A resounding
no. See id.
Following Thunder Basin, the Court began with the CSRA’s text and structure.
“Nothing in the CSRA’s text suggests that its exclusive review scheme is inapplicable
simply because a covered employee challenges a covered action on the ground that the
statute authorizing that action is unconstitutional.” Elgin, 567 U.S. at 13. The Court
highlighted that the CSRA contained only one exception where claims were permissible in
district court—when a covered employee “alleges that a basis for the action was
discrimination in contravention of federal employment laws.” Id. (quoting 5 U.S.C. §
7702(a)(1)(B)). But even that is limited to after the employee obtains an unfavorable MSPB
decision. See id. The Court explained that such an exception reveals that “Congress knew
how to provide alternative forums for judicial review based on the nature of an employee’s
claim.” Id. “That Congress declined to include an exemption from Federal Circuit review
for challenges to a statute’s constitutionality indicates that Congress intended no such
exception.” Id. Thus, the Court rejected the petitioner’s request to create an exception to
“CSRA exclusivity for facial or as-applied constitutional challenges to federal statutes.”
Id. at 12–13.
Next, the Court considered the purpose of the CSRA.

-- 28 of 36 --

29
[T]he CSRA’s integrated scheme of administrative and judicial review for
aggrieved federal employees was designed to replace an outdated patchwork
of statutes and rules that afforded employees the right to challenge employing
agency actions in district courts across the country. Such widespread judicial
review, which included appeals in all of the Federal Courts of Appeals
produced wide variations in the kinds of decisions issued on the same or
similar matters and a double layer of judicial review that was wasteful and
irrational.
Id. at 13–14 (cleaned up). According to the Supreme Court, the very purpose of the CSRA
was to create a streamlined and more consistent system of review precisely by depriving
district courts of jurisdiction. See id. at 14.
Fausto and Elgin work in tandem to clarify the CSRA’s sweeping preclusive effect.
“Just as the CSRA’s elaborate framework demonstrates Congress’ intent to entirely
foreclose judicial review to employees to whom the CSRA denies statutory review, it
similarly indicates that extrastatutory review is not available to those employees to whom
the CSRA grants administrative and judicial review.” Id. at 11 (cleaned up). As the Seventh
Circuit has properly explained, district courts are “not, in the first instance, the appropriate
forum to expound on the meaning of the CSRA’s various provisions. That task was
delegated to the MSPB, with later review by the Federal Circuit.” Ayrault v. Pena, 60 F.3d
346, 350 (7th Cir. 1995) (citation omitted).
As a result, step one of Thunder Basin is settled as to the CSRA—the Supreme Court
has determined Congress’ intent for the CSRA to preclude district court review is “fairly
discernible.” Elgin, 567 U.S. at 12. Thus, we need not, and indeed we cannot, go any
further. See Payne v. Taslimi, 998 F.3d 648, 655 n.4 (4th Cir. 2021) (“[O]ur mandate as an
inferior court [is] to follow the Supreme Court’s commands (vertical stare decisis).”).

-- 29 of 36 --

30
B. The Panel’s Reasoning
Despite this on-point Supreme Court precedent, the panel doesn’t stop; it marches
on. The panel refers to the current administration’s removal of the Special Counsel and two
members of the MSPB—which resulted in the MSPB lacking a quorum—and the
administration’s position in lawsuits challenging those removals because “removal
protections enshrined in the CSRA are violations of separation of powers.” NAIJ II, 139
F.4th at 305, 307 (citations omitted). The panel also points to the recent Supreme Court
order in Trump v. Wilcox, 145 S. Ct. 1415 (2025), which stayed reinstatement of MSPB
board members who were removed by the President. NAIJ II, 139 F.4th at 307. According
to the panel, these events create tension with the CSRA’s limitations on the President’s
removal power of members of the MSPB and the Special Counsel “only for inefficiency,
neglect of duty, or malfeasance in office,” 5 U.S.C. §§ 1202(d), 1211(b); the term limits
on both MSPB members and the Special Counsel; the requirement that the Senate approve
MSPB members; and the MSPB’s structural requirements that members be from separate
political parties. NAIJ II, 139 F.4th at 305–08.
These tensions, the panel finds, cast doubt—doubt that apparently did not exist from
1978, when Congress passed the CSRA, until early 2025, when the current administration
took the actions the majority describes and the Supreme Court issued its stay order in
Wilcox—on whether Congress’ intent to preclude district court jurisdiction over claims
within the CSRA’s scope was “fairly discernible in the statutory scheme.” Thunder Basin,
510 U.S. at 207 (quoting Block, 467 U.S. at 351). To address that concern, the panel opinion
changes the Thunder Basin test by adding an inquiry into whether recent political events

-- 30 of 36 --

31
prevent the CSRA from functioning as intended when Congress passed it. NAIJ II, 139
F.4th at 305. If the district court finds the answer is no, the panel concludes the district
court must then decide whether Congress would have nevertheless intended for the CSRA
to still preclude judicial review in district court. See id. at 308. Finally, if the district court
concludes the answer to that question is no, NAIJ’s case can proceed in district court
contrary to Fausto and Elgin.
To repeat, I see three problems with this decision. First, the Supreme Court has
already spoken as to the CSRA’s scheme—judicial review in district court is not
appropriate. Second, the panel’s decision requires the district court to invade the
responsibilities the Constitution vests to Congress. Third, the panel takes these
extraordinary steps on its own without the issues being raised by any party and without the
benefit of any briefing or oral argument.
1. Fausto and Elgin Bind Us
First, the Supreme Court has already told us in Fausto and Elgin that, under Thunder
Basin’s test, the CSRA’s statutory scheme deprives district courts of jurisdiction to review
federal workers’ employment claims. See Section II.A. And nowhere in Thunder Basin did
the Supreme Court inquire into whether a statute was functioning as originally intended.
See 510 U.S. at 207. The Supreme Court can of course change the test or the answer. But
we can’t. As an inferior court, we are not free to conjure up new questions to avoid Supreme
Court precedent. See Payne, 998 F.3d at 654 (“It is beyond our power to disregard a
Supreme Court decision, even if we are sure the Supreme Court is soon to overrule it.”).

-- 31 of 36 --

32
Despite this command, the panel opinion suggests that Fausto and Elgin may not
control because they did not address whether recent political events have caused the
statutory scheme to cease to operate as originally intended. Technically, that is true—those
cases are from decades ago. But the only basis the panel gives for revisiting those cases
here is to consider the functioning-as-intended component of the Thunder Basin test that
the panel creates in its opinion. That’s not a good reason. In the absence of the Supreme
Court speaking, whether Thunder Basin should include a function component is not our
decision to make.
2. Reimagining and Updating the Intent of Unchanged Statutes Invades
Congress’ Role
Second, the panel’s inquiry into whether a statute is functioning as intended creates
enormous separation of powers concerns. Congress passed the CSRA. If the statute is not
functioning as Congress thinks it should be in light of the administration’s actions or
judicial decisions, Congress can amend it. See Rivers v. Roadway Exp., Inc., 511 U.S. 298,
313 (1994) (“Congress, of course, has the power to amend a statute that it believes we have
misconstrued.”). That has happened repeatedly through the years. See, e.g., Food & Drug
Admin. v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 126 (2000) (holding that
“Congress has clearly precluded the FDA from asserting jurisdiction to regulate tobacco
products,” after which Congress enacted the Family Smoking Prevention and Tobacco
Control Act, 15 U.S.C. § 1331 et seq., to give the FDA authority to regulate the sale,
distribution, advertising, promotion and use of tobacco products). But unelected judges in
2025 cannot cast aside the CSRA’s preclusive review scheme because they believe that the

-- 32 of 36 --

33
95th Congress in 1978 would not have intended the CSRA to be exclusive if it had known
a later administration might claim a constitutional right to terminate members of the MSPB
and the Special Counsel and act on that claim. See Bostock v. Clayton Cnty., 590 U.S. 644,
654–55 (2020) (“This Court normally interprets a statute in accord with the ordinary public
meaning of its terms at the time of its enactment. After all, only the words on the page
constitute the law adopted by Congress and approved by the President. If judges could add
to, remodel, update, or detract from old statutory terms inspired only by extratextual
sources and our own imaginations, we would risk amending statutes outside the legislative
process reserved for the people’s representatives.”).
Besides, how are judges supposed to do this? There are countless reasons members
of Congress may have voted to adopt the CSRA. I don’t see how judges today can discern
a new congressional intent for a statute enacted 47 years ago based on recent political
events. Congress expresses its intent in the words of the statutes it passes and in the
amendments to those statutes. As courts, we should stick to applying and interpreting those
words, not trying to divine what Congress would or would not have done differently had it
known about the political environment today.
What’s more, consider the instability we are sowing. If the current administration’s
actions allow us to reinterpret Congress’ intent about the CSRA’s statutory scheme, future
political events will too. The nature of the CSRA’s exclusive review scheme will be in a
constant state of flux on whether district courts have jurisdiction, rendering the scheme
captive to judges’ views on political whims of the most recent administration.

-- 33 of 36 --

34
Similarly, the panel’s reliance on Wilcox is also misplaced. First, it’s a stay order—
not a final ruling on the constitutionality of the CSRA’s removal provisions. Second, even
if the Supreme Court were to decide that those removal provisions are unconstitutional, I
don’t see how that could reveal a new congressional intent about the exclusivity of the
CSRA’s review procedures. Such a ruling might present severability issues. But without a
final ruling striking the removal provisions, this is neither the time nor place for such an
analysis. Nor is it appropriate to jump the gun on those settled principles by creating and
applying a new functioning-as-intended test.
In sum, had the Congress that passed the CSRA known of the recent events the panel
identifies, who knows whether that Congress would or would not have still deprived district
courts of jurisdiction to hear employment disputes of federal employees? I certainly don’t.
But even if inferior courts didn’t have the answer to that question from Fausto and Elgin—
and we do—I don’t see how our court or the district court is equipped to answer that
question based on present events and circumstances. Whatever one thinks of Thunder
Basin, it doesn’t permit the panel’s approach.
3. Party Presentation
Last but not least, party presentation. “In our adversarial system of adjudication, we
follow the principle of party presentation.” United States v. Sineneng-Smith, 590 U.S. 371,
375 (2020). Courts rely on the parties to “frame the issues for decision,” thereby assigning
to courts the “role of neutral arbiter of matters the parties present.” Id. (quoting Greenlaw
v. United States, 554 U.S. 237, 243 (2008)) (internal quotation mark omitted). While the
principle is “supple, not ironclad,” the Supreme Court has instructed us that deviation from

-- 34 of 36 --

35
the general principle is limited to “circumstances in which a modest initiating role for a
court is appropriate.” Id. at 376.
So just what did the parties say on step one of Thunder Basin? The NAIJ conceded
in its brief that:
Under the first step [of Thunder Basin], courts must ask “whether Congress’s
intent to preclude district court jurisdiction is ‘fairly discernible in the
statutory scheme.’” Bennett v. SEC, 844 F.3d 174, 181 (4th Cir. 2016)
(quoting Thunder Basin, 510 U.S. at 207). The Supreme Court has already
held that such intent is manifest in the CSRA, see Elgin, 657 U.S. at 12, and
so only the second step of the Thunder Basin inquiry is at issue here.
Op. Br. at 16–17. And the Director of the EOIR agreed, saying:
[T]his Court and the Supreme Court have repeatedly observed that the CSRA
provides the exclusive means by which covered federal employees may bring
claims arising out of their employment. Elgin, 569 U.S. at 13; see also
Bennett v. U.S. Sec. & Exch. Comm’n, 844 F.3d 174, 180–81 (4th Cir. 2016)
(discussing Elgin); Hall v. Clinton, 235 F.3d 202, 203 (4th Cir. 2000).
Resp. Br. at 17–18. The result was that the parties confined their argument to step two of
Thunder Basin. Without question, the panel raises an issue that the parties not only failed
to raise, they conceded it was foreclosed by Supreme Court precedent.
Beyond that, the panel opinion’s discussion of step one of Thunder Basin interjects
factual events that all occurred after oral argument. So, the panel doesn’t just make
arguments that the parties did not make; it makes arguments the parties could never have
made. And it did so without asking for supplemental briefing or oral argument on these
far-reaching issues.
True, the panel opinion claims sending this back to the district court for findings in
the first instance justifies its departure from party presentation principles. But the only

-- 35 of 36 --

36
thing it sends back is the factual findings informing the new functioning-as-intended test it
adopts. The opinion decides that if the inquiry under this new test comes out a certain way,
Supreme Court precedent doesn’t apply. We should not be confused about the scope of
what the panel opinion did and did not do.
III.
To conclude, the panel claims that it “cannot allow [its] black robes to insulate [it]
from taking notice of items in the public record.” NAIJ II, 139 F.4th at 313. That may sound
well-intentioned. But I fear that in “tak[ing] judicial notice of matters of public record,”
the opinion unnecessarily and improperly inserts our court into the political controversies
of the day. Id. at 305. And worse than that, in doing so, it undermines important principles
of our system of justice. First, the panel opinion trods upon the black robes of our Nation’s
highest court, whose precedent demands concluding that Congress’ intent to deprive
district courts of jurisdiction under the CSRA is “fairly discernible” in the statute. See
Elgin, 567 U.S. at 11–12; Fausto, 484 U.S. at 447. Second, the panel opinion uses its robes
to invade the role of the legislative branch, permitting judges to reimagine the
congressional intent behind an unchanged statutory scheme because we believe the statute
is not functioning as the 95th Congress intended. And third, the panel opinion shirks party
presentation principles—taking off its black robes to argue a case different from the one
the NAIJ advanced. But our job, and only our job, is to follow the law wherever it leads.
Here, it leads to affirming the district court. Because the panel fails to do so, and because
of the far-reaching implications of its reasoning, I respectfully dissent from the denial of
rehearing en banc.

-- 36 of 36 --