Under Seal v.  Under Seal

PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
 UNDER SEAL,
Plaintiff-Appellee,
v. No. 02-1683 
UNDER SEAL,
Defendant-Appellant. 
 UNDER SEAL,
Plaintiff-Appellee,
v. No. 02-1684 
UNDER SEAL,
Defendant-Appellant. 
Appeals from the United States District Court
for the District of Maryland, at Baltimore.
William M. Nickerson, Senior District Judge.
(CA-00-2631-WMN)
Argued: January 24, 2003
Decided: April 14, 2003
Before LUTTIG, MOTZ, and TRAXLER, Circuit Judges.
Affirmed by published opinion. Judge Luttig wrote the opinion,
which Judge Traxler joined. Judge Motz wrote an opinion concurring
in the judgment.

-- 1 of 18 --

COUNSEL
ARGUED: W. Neil Eggleston, HOWREY, SIMON, ARNOLD &
WHITE, L.L.P., Washington, D.C., for Appellants. Jamie M. Bennett,
Assistant United States Attorney, Baltimore, Maryland, for Appellee.
ON BRIEF: Candace N. Beck, HOWREY, SIMON, ARNOLD &
WHITE, L.L.P., Washington, D.C.; Carl S. Rauh, Alan Kriegel, Alex-
ander W. Powell, Jr., SKADDEN, ARPS, SLATE, MEAGHER &
FLOM, L.L.P., Washington, D.C., for Appellants. Thomas M.
DiBiagio, United States Attorney, Baltimore, Maryland, for Appellee.
OPINION
LUTTIG, Circuit Judge:
Appellants, two corporations who participate in the National Flood
Insurance Program ("NFIP"), object to a district court order unsealing
a False Claims Act ("FCA") complaint brought against them by the
government as intervenor in a previously sealed complaint brought
under the FCA’s qui tam provision. Finding no error in the district
court’s exercise of discretion in unsealing the record, we affirm.
I.
In 2000, an unidentified party filed a qui tam action, as relator,
under the FCA against the appellants for allegedly wrongful insurance
payments they made under the NFIP. The action was filed under seal,
pursuant to the terms of the FCA’s qui tam provision, which reads as
follows:
(1) A person may bring a civil action for a violation of
section 3729 for the person and for the United States Gov-
ernment. . . .
(2) A copy of the complaint and written disclosure of sub-
stantially all material evidence and information the person
possesses shall be served on the Government . . . The com-
plaint shall be filed in camera, shall remain under seal for
2 UNDER SEAL v. UNDER SEAL

-- 2 of 18 --

at least 60 days, and shall not be served on the defendant
until the court so orders. . . .
(3) The Government may, for good cause shown, move
the court for extensions of the time during which the com-
plaint remains under seal under paragraph (2). . . .
(4) Before the expiration of the 60-day period or any
extensions obtained under paragraph (3), the Government
shall —
(A) proceed with the action . . . ; or
(B) notify the court that it declines to take over the action
. . . .
. . . .
31 U.S.C. § 3730(b). The relator’s action lay fallow as the govern-
ment extended its investigative period. Then, on January 8, 2002, the
government notified appellants that it would intervene in the suit and
proceed against them.
Before the government filed an intervening complaint, the appel-
lants moved to compel the government to arbitrate the controversy
with them, as required by the Subsidy Arrangement that governs par-
ticipation in the NFIP. See 42 C.F.R. Ch. 1, Pt. 62, App. A (2001).
Shortly thereafter, the government filed its intervening complaint with
the district court, along with a motion to unseal the action. On June
3, 2002, the court issued an order (1) compelling arbitration, (2) stay-
ing the government’s complaint, and (3) unsealing the action.
The court stayed the effect of its unsealing order for ten days in
order to allow appellants to take an appeal to this court. Appellants
did appeal, and a motions panel of this court stayed the unsealing
order pending our disposition.
II.
As an initial matter, there is a significant question as to whether we
have jurisdiction to hear this appeal. The government contends that
3 UNDER SEAL v. UNDER SEAL

-- 3 of 18 --

the appeal is interlocutory, and therefore foreclosed from our review
by 28 U.S.C. § 1291 (providing the court of appeals with jurisdiction
to hear appeals only "from [ ] final decisions of the district courts").
Appellants, on the other hand, argue that the district court’s order is
a collateral order, reviewable under Cohen v. Beneficial Industrial
Loan Corp., 337 U.S. 541, 546 (1949). The binding effect of our pre-
cedent compels us to conclude that the district court’s order unsealing
the record is appealable.
A.
Our analysis of the collateral order doctrine begins with the twist,
uncommon in this circuit, that our precedent sets forth two different
standards for determining whether a district court order qualifies as
a collateral order.
The parties approach the collateral order analysis principally via
our decision in James v. Jacobson, 6 F.3d 233, 239 (4th Cir. 1993).
Citing James, along with Taylor v. Nelson, 788 F.2d 220, 224 (4th
Cir. 1986), they both proceed on the assumption that the district
court’s order unsealing the record must meet four criteria to qualify
for Cohen collateral appealability. They say the order must (1) con-
clusively determine the question, (2) resolve an important question
independent of the merits, (3) be effectively unreviewable on appeal
from final judgment, and (4) present a serious and unsettled question
on appeal. Elsewhere in our precedent, however, this last factor, the
fourth "factor," is not applied, and the first three factors are alone
determinative of appealability. See, e.g., Hopkins v. Prince George’s
County, 309 F.3d 224 (4th Cir. 2002).
Cohen v. Beneficial Industrial Loan Corp. first laid the framework
for the collateral order doctrine. There, the Supreme Court held collat-
eral orders appealable because they:
finally determine claims of right separable from, and collat-
eral to, rights asserted in the action, [are] too important to
be denied review and [are] too independent of the cause
itself to require that appellate consideration be deferred until
the whole case is adjudicated.
4 UNDER SEAL v. UNDER SEAL

-- 4 of 18 --

337 U.S. at 546. Since Cohen, the Court has described and/or applied
the collateral order analysis at least twenty-nine times in majority
opinions. In every one of those instances, save one, the Court either
identified or both identified and applied the three factors recited in
Cohen.1 And, in numerous of those cases the Court explicitly denomi-
nated its test as a three-factor, three pronged, or three part test.2 The
twenty-eight uniform opinions — identifying the collateral order test
as having three factors — sandwich the lone opinion, Nixon v.
Fitzgerald, in which the Court asserted that the test includes the
fourth "factor."3 On the few other occasions on which the Supreme
1Compare Cunningham v. Hamilton County, 527 U.S. 198 (1999),
Quackenbush v. Allstate Ins. Co., 517 U.S. 706 (1996), Behrens v. Pelle-
tier, 516 U.S. 299 (1996), Johnson v. Jones, 515 U.S. 304 (1995), Swint
v. Chambers County Comm’n, 514 U.S. 35 (1995), Digital Equip. Corp.
v. Desktop Direct, Inc., 511 U.S. 863 (1994), Puerto Rico Aqueduct and
Sewer Auth. v. Metcalf & Eddy, Inc., 506 U.S. 139 (1993), Lauro Lines
S.R.L. v. Chasser, 490 U.S. 495 (1989), Midland Asphalt Corp. v. United
States, 489 U.S. 794 (1989), Van Cauwenberghe v. Biard, 486 U.S. 517
(1988), Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U.S. 271
(1988), Stringfellow v. Concerned Neighbors in Action, 480 U.S. 370
(1987), Mitchell v. Forsyth, 472 U.S. 511 (1985), Richardson-Merell,
Inc. v. Koller, 472 U.S. 424 (1985), Richardson v. United States, 468
U.S. 317 (1984), Flanagan v. United States, 465 U.S. 259 (1984), Moses
H. Cone Mem’l Hosp. v. Mercury Const. Corp., 460 U.S. 1 (1983),
United States v. Hollywood Motor Car Co., Inc., 458 U.S. 263 (1982),
Firestone Tire & Rubber Co. v. Risjord, 449 U.S. 368 (1981), Federal
Trade Comm’n v. Standard Oil Co. of Ca., 449 U.S. 232 (1980), Boeing
Co. v. Van Gemert, 444 U.S. 472 (1980), Coopers & Lybrand v. Livesay,
437 U.S. 463 (1978), United States v. MacDonald, 435 U.S. 850 (1978),
National Socialist Party of America v. Village of Skokie, 432 U.S. 43
(1977), Abney v. United States, 431 U.S. 651 (1977), Eisen v. Carlisle
and Jacquelin, 417 U.S. 156 (1974), Local No. 438 Const. and General
Laborers’ Union, AFL-CIO v. Curry, 371 U.S. 542 (1963), and Swift &
Co. Packers v. Compania Colombiana Del Caribe, S.A., 339 U.S. 684
(1950), with Nixon v. Fitzgerald, 457 U.S. 731 (1982) (asserting that the
Cohen test includes the fourth "factor" without examining the Court’s
contrary precedent).
2See, e.g., Digital Equip. Corp.; Lauro Lines S.R.L.; Van Cauwen-
berghe; Gulfstream Aerospace Corp.; Mitchell; Richardson.
3The concurring opinion notes that in Eisen and Quackenbush the
Court articulated the collateral order test as having two and four factors,
5 UNDER SEAL v. UNDER SEAL

-- 5 of 18 --

Court has noted the "serious and unsettled" factor in its opinions, the
Court has noted the "factor" as a circuit rule. That is, the Court there
has noted that a particular circuit has applied the fourth "factor," not
that the "factor" is required by Supreme Court precedent. See, e.g.,
Gulfstream Aerospace Corp., 485 U.S. at 291 (Scalia, J., concurring)
("I note that today’s result could also be reached by application of the
rule adopted by the First Circuit, that to come within the Cohen
exception the issue on appeal must involve an important and unsettled
question of controlling law . . ." (emphasis added)); McDonald v.
Smith, 472 U.S. 479, 482 n.3 (1985) (noting our circuit’s application
of the fourth "factor").4
respectively. See post at 16. A close reading of those articulations, how-
ever, reveals that both cases simply repackage Cohen’s three factors in
different dress. Thus, in Eisen, the Court bundled Cohen’s second and
third factors into a single "second" factor. And in Quackenbush, the
Court divided Cohen’s second factor into two factors, its "second" and
"fourth" factors.
4The concurring opinion suggests that Hohn v. United States, 524 U.S.
236 (1988), and Brady v. Maryland, 373 U.S. 83 (1963), ought also be
included in our tabulation of Supreme Court decisions describing and/or
applying the collateral order test, see post at 14. We do not include them,
however, for the simple reason that neither describes and/or applies the
collateral order doctrine.
As to the first, Hohn, it involved no collateral order issue at all, and
merely recited the holding of Nixon v. Fitzgerald, along with other hold-
ings on different doctrinal matters, in order to illustrate that Court prece-
dent foreclosed the conclusion that "a request to proceed before a court
of appeals should be regarded as a threshold inquiry separate from the
merits," Hohn, 524 U.S. at 546. Obviously, we agree with Hohn’s recita-
tion of Nixon’s holding. But we do not agree that the Court, in reciting
Nixon’s holding for purposes entirely unrelated to conducting a collateral
order inquiry, can be said there to have described the collateral order
doctrine in the sense that we mean it, that is, as describing the doctrine
in order to enable its direct application.
As to the second, Brady, the language at issue is found in the facts sec-
tion, in a footnote. That footnote begins: "Neither party suggests that the
decision below is not a final judgment[.]" 373 U.S. at 85 n.1 (emphasis
added). From that beginning, it proceeds to cite Cohen only for the prop-
osition that the serious and unsettled factor is relevant to the conclusion
6 UNDER SEAL v. UNDER SEAL

-- 6 of 18 --

Because the Court’s most recent pronouncement and virtually all
of its other opinions establish that the collateral order test is a three
factor test, we now understand the Supreme Court’s collateral order
test to be such, and, by implication, we conclude that the fourth "fac-
tor" is not part of the test. See Cunningham v. Hamilton County, 527
U.S. 198, 202 (1999) ("[T]he collateral order doctrine [ ] provides that
certain orders may be appealed, notwithstanding the absence of final
judgment, . . . when they are "[1] conclusive, . . . [2] resolve impor-
tant questions separate from the merits, and [3] are effectively unre-
viewable on appeal from the final judgment in the underlying action."
(citing Swint v. Chambers County Comm’n, 514 U.S. 35 (1995) (cit-
ing Cohen))).
In determining that the Supreme Court’s collateral order test is a
three factor test we are not only consistent with the Court’s most
recent pronouncement, but also with the overwhelming majority of all
its pronouncements on the subject. Furthermore, our conclusion in
this regard is consistent with the Court’s earliest pronouncement on
the matter, see Swift & Co. Packers, and with its seminal pronounce-
ment in Abney v. United States, 431 U.S. 651 (1977), in which it first
adopted the term "collateral order," see id. at 657.
In Abney, the Court made abundantly clear through its detailed
analysis that under Cohen’s holding only three factors govern the
determination of whether a court order is an appealable collateral
order:
[In Cohen], the Court identified several factors which, in its
view, rendered the District Court’s order a "final decision"
that "no attack on the reviewability of the lower court’s judgment could
be successfully maintained." It nowhere suggests that a collateral order
(i.e., an order that represents a final decision, though not a final judg-
ment) is at issue. In line with the case’s posture, all the other precedent
this footnote relies on are pre-Cohen decisions. The only post-Cohen col-
lateral order decision to which it cites, Local No. 438 Const. and General
Laborers’ Union, it cites as a "cf.", that is, as a case that provides paral-
lel, not controlling, reasoning. Clearly then, Brady, in this footnote in its
fact section, does not contain an instructive description and/or applica-
tion of the collateral order inquiry.
7 UNDER SEAL v. UNDER SEAL

-- 7 of 18 --

within the statute’s meaning. First, the District Court’s order
had fully disposed of the question of the state security stat-
ute’s applicability in federal court; in no sense, did it leave
the matter "open, unfinished or inconclusive." Second, the
decision was not simply a "step toward final disposition of
the merits of the case (which would) be merged in final
judgment"; rather, it resolved an issue completely collateral
to the cause of action asserted. Finally, the decision had
involved an important right which would be "lost, probably
irreparably," if review had to await final judgment; hence,
to be effective, appellate review in that special limited set-
ting had to be immediate.
Abney, 431 U.S. at 658 (emphasis added) (citations omitted). The
Abney Court did not so much as mention Cohen’s note that the issue
there presented a "serious and unsettled question."
Our circuit began applying the fourth "factor" in Bever v. Gilbert-
son, 724 F.2d 1083, 1085 (4th Cir. 1984) (citing Cohen) (abrogated
on different grounds by Young v. Lynch, 846 F.2d 960 (4th Cir.
1988)). There, we noted that Cohen described the order being
reviewed as "present[ing] a serious and unsettled question." Since
Bever, we have, in fourteen instances, treated that notation as a fourth
element to the collateral order analysis, see, e.g., Taylor v. Nelson,
788 F.2d 220, 224 (4th Cir. 1986) (citing Bever). Those fourteen
instances compare with one-thousand-plus cases over the same period
of time in which we did not apply that fourth "factor" when applying
Cohen’s collateral order doctrine. From In re: Investigation, 563 F.2d
652 (4th Cir. 1977), to Miller v. Simmons, 814 F.2d 962 (4th Cir.
1987), to Hopkins v. Prince George’s County, 309 F.3d 224 (4th Cir.
2002), we have at all other times hewed to the Supreme Court’s pre-
scribed analysis, applying Cohen’s three factors.
As a panel of the full court, we cannot overrule prior decisions of
the court, panel or en banc, and we are bound to apply principles
decided by prior decisions of the court to the questions we address.
But here, where we confront two different rules governing the same
question, we are left no option but to choose from between our differ-
ing precedents. We would rather not hold this judicial license. How-
ever, until such time as the en banc court addresses itself to the
8 UNDER SEAL v. UNDER SEAL

-- 8 of 18 --

jurisprudential dilemma presented by these different precedents, such
is the only course available to us.
Sitting en banc in Jenkins v. Medford, 119 F.3d 1156 (4th Cir.
1997), we applied the Cohen collateral appealability test without any
reference to the fourth "factor." And we have never otherwise applied
the fourth "factor" while sitting en banc. Though Jenkins did not
claim to overrule Bever, and though after Jenkins a panel again rein-
serted the fourth "factor" into our Cohen doctrine, see Eagle Energy,
Inc. v. Secretary of Labor, 240 F.3d 319 (4th Cir. 2001), we believe
that Jenkins, along with the numerous similar panel applications of
Cohen, is grounds for concluding that our circuit’s preference in prac-
tice is to hew strictly to Cohen’s three factor test. To the extent that
any thought has even been given to the issue, this preference may be
attributable to the twin facts that, to do otherwise, would not only
conflict necessarily with Supreme Court authority but also lead to
doctrinally indefensible results. As to the first of these putative rea-
sons, imposition of an additional element conflicts necessarily with
the Supreme Court’s standard, since the elements of its collateral
order test establish a minimum threshold for our jurisdiction. And as
to the second reason, by way of example, if one may not immediately
appeal unless the issue is unsettled, then he would be foreclosed from
interlocutory review of an order which, under established precedent,
would clearly be violative of the Double Jeopardy Clause. For these
reasons, we choose not to apply the fourth "factor" here, a decision
that, happily, keeps us as true as possible to our prior circuit practice.
B.
The government pitched its argument against appealability of the
unsealing order solely on the contention that the order did not satisfy
the fourth "factor." It conceded (insofar as one can "concede" a juris-
dictional issue) that the "other" three Cohen appealability factors were
met. Obviously, the government’s position takes on significance in
light of our conclusion that the fourth "factor" is not applicable. But,
irrespective of whether we would agree with the government’s assess-
ment of the three factors, our decision in James v. Jacobson, 6 F.3d
at 237, compels the conclusion that the order in question here is a col-
lateral order. See id. (holding that a district court order denying ano-
nymity to the parties is a collateral order).
9 UNDER SEAL v. UNDER SEAL

-- 9 of 18 --

In James, the subject of the district court’s order was a matter of
discretion, involving the court’s interest in trial management. And it
involved a determination as to whether the rights of public access to
trial proceedings were outweighed by the parties’ privacy rights.
Here, in identical fashion, the subject of the court’s order is a matter
of discretion, founded on the court’s interest in trial management.
And it equally involves a determination of whether rights of public
access are outweighed by the parties’ privacy rights. Furthermore, in
both cases the privacy concern of the objecting parties is that the
court’s order will result in their being publicly identified with the
action. On these similarities, we cannot avoid the conclusion that
James applies, nor can we but hold that the order in this case is like-
wise appealable.5
5While we faithfully apply James’ principle in light of the similarities
of the cases, we are skeptical of the government’s concession, and
equally of James’ conclusion, that the district court order satisfies the
third Cohen-Abney factor — the likely irreparable loss of an important
right. The third factor — the likely irreparable loss of important rights
if review has to await final judgment — would not appear to be impli-
cated by district court orders on motions to seal court records or proceed-
ings because, unique to this realm, denial of collateral review does not
result in the consequence that review must then await final judgment.
Rather, parties dissatisfied with sealing or unsealing orders have at their
disposal what is in fact the preferred vehicle for review: petitions for
mandamus. Petitions for mandamus have long been recognized as the
appropriate vehicle for challenges to cloture and unsealing orders. Cf.
Baltimore Sun Co. v. Goetz, 886 F.2d 60 (4th Cir. 1989) ("Mandamus,
not appeal, is the preferred method of review for orders restricting press
activity related to criminal proceedings."); In re: Washington Post Co.,
807 F.2d 383, 388 (4th Cir. 1986) (treating an appeal by a non-party to
a district court order sealing documents in a criminal case as a petition
for mandamus); Central South Carolina Chapter, Society of Professional
Journalists, Sigma Delta Chi v. Martin, 556 F.2d 706 (4th Cir. 1977)
(noting that "mandamus is the proper remedy to request the relief prayed
for here" where appellant, on an interlocutory appeal, challenges a dis-
trict court protective order); see also United States v. Gonzales, 150 F.3d
1246 (10th Cir. 1998) (treating an interlocutory appeal from a district
court’s order to seal the record as a petition for mandamus, in light of the
fact that mandamus is the proper vehicle for such review).
Mandamus relief, though limited to circumstances where the petition-
er’s "right to issuance of the writ is clear and indisputable," Kerr v. U.S.
10 UNDER SEAL v. UNDER SEAL

-- 10 of 18 --

III.
Turning to the merits, we review the district court’s order to unseal
the record for an abuse of discretion. See Nixon v. Warner Communi-
cations, 435 U.S. 589, 598-99 (1978) (concluding that decisions as to
court record access are best left to the trial court’s discretion since
that court can best evaluate whether one of the three established
exceptions to the presumption of public access — (1) where disclo-
sure may be used to gratify private spite or promote public scandal,
(2) where disclosed records may serve as reservoirs of libelous state-
ments for press consumption, or (3) where disclosure might reveal
trade secrets — creates a countervailing privacy right that supports
sealing in light of the particular facts and circumstances of the case).
Appellants argue that the combined effect of the FCA’s initial seal-
ing of qui tam actions and the Subsidy Agreement’s commitment of
government suits against NFIP insurers to private arbitration, gov-
erned by the Federal Arbitration Act ("FAA"), work to create a sub-
stantial privacy interest for them that overcomes the public’s right to
access court documents. They also argue that to rule otherwise would
be to "reward the government’s improper conduct," Appellant’s Br.
at 10, which conduct they contend entails "prematurely seek[ing] to
make its case public in contravention of the private arbitration proto-
col of the Subsidy Arrangement of the NFIP," id. at 10.
In response, the government first argues, quite correctly, that the
presumption in favor of public disclosure of court records can only be
overcome by a significant countervailing interest. See Rushford v.
Dist. Court, 426 U.S. 394, 403 (1976), suffices to protect the parties’
rights here. As James’ discussion of the two varieties of "discretionary
rulings" makes clear, see 6 F.3d at 238, a properly made discretionary
ruling does not implicate important rights of the parties. Only an improp-
erly made discretionary ruling implicates the parties’ rights, either sub-
stantive or procedural. Because this distinction, on which the availability
of collateral appeal rests, can only be made after review, mandamus is
instead appropriate. Requiring that review be had via mandamus would
ensure that trials do not grind to a halt where a discretionary sealing
judgment is properly made, while also ensuring the vindication of the
important rights implicated by an improperly made sealing judgment.
11 UNDER SEAL v. UNDER SEAL

-- 11 of 18 --

New Yorker, 846 F.2d 249, 253 (4th Cir. 1988). Secondly, it rightly
points out that there is no bar on litigants filing pleadings in cases
where arbitration is required, whether that filing is to challenge the
scope of arbitration, preserve an action before a statute of limitations
runs, or for another reason. See, e.g., Long v. Silver, 248 F.3d 309
(4th Cir. 2002) (determining applicability of arbitration provision to
non-signatory, and collecting cases of court consideration of arbitra-
tion related issues). Lastly, the government correctly notes that the
FCA’s purpose in initially sealing qui tam suits does not extend to the
sealing appellants seek. See Lujan v. Hughes Aircraft Co., 67 F.3d
242 (9th Cir. 1995) (explaining that purpose of § 3730(b)(2)’s sealing
provision is to allow the government to study and evaluate, out of
public view, the relator’s information for possible intervention or
overlapping criminal investigation). We add that neither the Subsidy
Agreement nor the FAA requires that the parties maintain secrecy
during arbitration. They only deny the public any right to access arbi-
tration records. Where a party, of its own accord, places information
in a public domain, it offends neither the regulatory nor the statutory
regime.
These points, taken together, confirm that appellants have no pri-
vacy rights in the contents of the court record. Since the purpose of
the FCA does not support continued sealing, and only justifies sealing
in order that the government may investigate, appellants’ reliance on
that Act is misplaced. And, appellants cannot rely on either the Sub-
sidy Arrangement or the FAA, as neither prohibits the government
from placing its complaint against appellants in the public domain.
Since appellants lack privacy rights to this information, and since
there is no bar to filing court pleadings in connection with an arbi-
trated matter, the government’s conduct cannot be characterized as
"improper."
Unable to demonstrate any relevant privacy right, appellants fail to
satisfy Rushford’s high standard and to overcome the public’s right
to access court documents. The district court’s discretionary order
unsealing the record is therefore not an abuse of discretion.
CONCLUSION
For the foregoing reasons, the judgment of the district court is
affirmed.
12 UNDER SEAL v. UNDER SEAL

-- 12 of 18 --

AFFIRMED
DIANA GRIBBON MOTZ, concurring:
I concur in the judgment and in much of the majority’s rationale.
I write separately in an effort to reconcile some of the collateral order
precedent on the "serious and unsettled question" factor and more
fully explain why I believe this factor is not a collateral order require-
ment.
At the outset, it seems only fair to note that the conflict within the
precedent of this circuit noted by the majority, ante at 4-6 and 7-9,
undoubtedly stems from the disarray in Supreme Court precedent.
Since establishing the collateral order doctrine more than fifty years
ago in Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 546
(1949), the high court has not hewed to an entirely consistent
approach.
In Cohen itself the Court stated the rationale for the collateral order
doctrine as follows:
[When a final judgment on the merits is issued] it will be
too late effectively to review the present order and the rights
conferred by the statute, if it is applicable, will have been
lost, probably irreparably . . . . This decision appears to fall
in that small class which finally determine claims of right
separable from, and collateral to, rights asserted in the
action, too important to be denied review and too indepen-
dent of the cause itself to require that appellate consider-
ation be deferred . . . . We hold this order appealable
because it is a final disposition of a claimed right which is
not an ingredient of the cause of action and does not require
consideration with it. But we do not mean that every order
fixing security is subject to appeal. Here it is the right to
security that presents a serious and unsettled question. If the
right were admitted or clear and the order involved only an
exercise of discretion . . . appealability would present a dif-
ferent question.
13 UNDER SEAL v. UNDER SEAL

-- 13 of 18 --

337 U.S. at 546-47 (emphasis added). Thus, in the Court’s seminal
explication of the collateral order doctrine, it clearly attached some
importance to the fact that the order before it "present[ed] a serious
and unsettled question." Id.
Additionally, in four subsequent cases, the Court has mentioned the
"serious and unsettled question" factor in discussing the collateral
order doctrine. See Brady v. Maryland, 373 U.S. 83, 85 n.1 (1963)
(stating as the very first factor, in explaining its rationale for finding
the order before it appealable, that "the right to a trial on the issue of
guilt . . . presents a serious and unsettled question" (internal quotation
marks and citation omitted)); Nixon v. Fitzgerald, 457 U.S. 731, 742
(1982) (stating that "[a]s an additional requirement, Cohen established
that a collateral appeal of an interlocutory order must ‘presen[t] a seri-
ous and unsettled question’"); McDonald v. Smith, 472 U.S. 479, 482
n.3 (1985) (noting without disapproval, in a case arising from the
denial of a motion on the pleadings, that the Fourth Circuit had "ac-
cepted jurisdiction under the ‘collateral order’ doctrine," "[b]ecause
petitioner raised a ‘serious and unsettled question’"); Hohn v. United
States, 524 U.S. 236, 246-47 (1998) (describing without disapproval,
the District of Columbia Circuit’s summary dismissal of the interlocu-
tory appeal in Nixon v. Fitzgerald for failure of the order "to satisfy
th[e] threshold requirement" of presenting a "‘serious and unsettled
question’ of law sufficient to bring the case within the collateral order
doctrine"); see also Gulfstream Aerospace Corp. v. Mayacamas, 485
U.S. 271, 292 (1988) (Scalia, J., concurring) (noting "some support"
in Supreme Court precedent for viewing "serious and unsettled ques-
tion" as a factor under collateral order doctrine).
Given these cases, it is hardly surprising that several courts of
appeals, including ours, have on occasion concluded that whether an
interlocutory order involved a "serious and unsettled question" was
relevant in determining whether it was immediately appealable as a
collateral order. See, e.g., Chaves v. M/V Medina Star, 47 F.3d 153,
155 (5th Cir. 1995); Banque Nordeurope S.A. v. Banker, 970 F.2d
1129, 1131 (2d Cir. 1992); Jim Walter Res., Inc. v. Fed. Mine Safety
& Health Review Comm’n., 920 F.2d 738, 744-45 n.3 (11th Cir.
1990); see also ante at 4-6 and 7-9.
But the Supreme Court has never held an interlocutory order not
to qualify as an immediately appealable collateral order because it did
14 UNDER SEAL v. UNDER SEAL

-- 14 of 18 --

not involve a "serious and unsettled question." Furthermore, in the
years that followed Cohen, the Court has considered the collateral
order doctrine on "many occasions," Gulfstream, 485 U.S. at 276, and
the few cases mentioning a "serious and unsettled question" factor are
mere drops in this sea of collateral order jurisprudence.
Very frequently the Court has applied the abbreviated, three-factor
test (containing no mention of the "serious and unsettled question"
factor), which it originally set forth in Coopers & Lybrand v. Livesay,
437 U.S. 463, 468 (1978):
the order must [1] conclusively determine the disputed ques-
tion, [2] resolve an important issue completely separate
from the merits of the action, and [3] be effectively unre-
viewable on appeal.
See also Johnson v. Jones, 515 U.S. 304, 310-11 (1995) (quoting test
with bracketed numbers indicating the three-factors, and noting that
the Court "restated" the Cohen test in Coopers & Lybrand); Puerto
Rico Aqueduct & Sewer Auth. v. Metcalf & Eddy, Inc., 506 U.S. 139,
144-45 (1992) (quoting test, inserting bracketed numbers, and stating
that "to come within" the Cohen rule an order must meet these fac-
tors); Midland Asphalt Corp. v. United States, 489 U.S. 794, 799
(1989) (quoting test and stating that "[t]o fall within the limited class
of final collateral orders, an order" must meet these requirements);
Lauro Lines v. Chasser, 490 U.S. 495, 498 (1988) (quoting test and
noting that the Court has "held that to fall within the Cohen exception
an order must satisfy at least [these] three conditions"); Van Cauwen-
berghe v. Biard, 486 U.S. 517, 522 (1988) (quoting test and noting
that "[t]he Court refined the ‘collateral order’ doctrine of Cohen in
Coopers & Lybrand"); Richardson-Merrell, Inc. v. Koller, 472 U.S.
424, 431 (1985) (quoting test and noting "an order must at a minimum
satisfy [these] three conditions"); Flanagan v. United States, 465 U.S.
259, 265 (1984) (same); Firestone Tire & Rubber Co. v. Risjord, 449
U.S. 368, 375 (1981) (quoting test and stating that Coopers & Lyb-
rand had "recently defined this limited class of final ‘collateral
orders’ in these terms"); Moses H. Cone Mem’l Hosp. v. Mercury
Constr. Corp., 460 U.S. 1, 11 (1983) (quoting test and stating it "sum-
marize[s]" "[t]he factors required").
15 UNDER SEAL v. UNDER SEAL

-- 15 of 18 --

In addition, as the majority demonstrates, ante at 4-6, even when
the Court has not phrased the test in precisely this way, it has almost
always applied a similar test that includes no "serious and unsettled
question" factor. See, e.g., Abney v. United States, 431 U.S. 651, 658
(1977). And often the Court has expressly denominated the governing
test as a three-factor or three-pronged test. See ante at 5. But see Eisen
v. Carlisle & Jacquelin, 417 U.S. 156, 171 (1974) (characterizing
Cohen as holding an "order appealable on two grounds": "[f]irst [the
order] settled conclusively [a] claim" and "[s]econd, . . . it concerned
a collateral matter that would not be reviewed effectively on appeal
from the final judgment" (emphasis added)); Quackenbush v. Allstate
Ins. Co., 517 U.S. 706, 712 (1996) (stating four requirements: collat-
eral orders must [1] "conclusively determine a disputed question, that
is [2] completely separate from the merits of the action, [3] effectively
unreviewable on appeal from a final judgment, and [4] too important
to be denied review." (internal quotation marks and citations omitted)).1
Moreover, mandating proof that an order involves a "serious and
unsettled question" would seem to present new complications in the
already murky and often criticized collateral order jurisprudence. See,
e.g., 15A Charles A. Wright et al., Federal Practice & Proc. § 3911.5
at 432, 438 (2d ed. 1991) (noting "[t]he ambiguity of the Supreme
Court’s approach over time," on this issue and that "[t]he status of the
serious and unsettled question aspect of collateral order doctrine thus
remains uncertain"); Lloyd C. Anderson, The Collateral Order Doc-
trine: A New ‘Serbonian Boy’ and Four Proposals, 46 Drake L. Rev.
539 (1998). For example, it is unclear whether a question would have
to be "serious" to the parties, or to society in general, or to the devel-
opment of the law, or that seriousness should be determined in some
other way. Precisely when a question no longer remains "unsettled"
is also unclear. Generally, we regard a published Court of Appeals
opinion as settling a question within the circuit. See, e.g., Scotts Co.
v. United Indus. Corp., 315 F.3d 264, 271 n.2 (4th Cir. 2002) ("[A]
panel of this court cannot overrule, explicitly or implicitly, the prece-
dent set by a prior panel of this court . . . ." (internal quotation marks
and citation omitted)). But in Nixon, the District of Columbia Circuit
1The Supreme Court has increasingly emphasized that the right
asserted must be "important." See, e.g., Digital Equip. Corp. v. Desktop
Direct, Inc., 511 U.S. 863, 878-79 (1994).
16 UNDER SEAL v. UNDER SEAL

-- 16 of 18 --

had previously issued a published opinion on point but the Supreme
Court found the question "unsettled" because the Supreme Court itself
had "never so held." Nixon, 457 U.S. at 742. Given these uncertain-
ties, I think it unlikely that the Supreme Court intended, without
greater explanation, to impose this factor as a collateral order require-
ment.
For all of these reasons, I agree that the "Supreme Court’s collat-
eral order test" does not contain a fourth factor requiring the interloc-
utory order to involve "a serious and unsettled question." See ante at
7. It seems to me that when discussing the "serious and unsettled" fac-
tor initially in Cohen itself, the Court was simply describing several
features of the interlocutory order before it, which persuaded the
Court that the order was immediately appealable. Thus, the Cohen
Court noted the "serious and unsettled question" in an effort to pro-
vide a full rationale for its holding, not as part of a multi-factor "test."
In Brady, 373 U.S. at 85 n.1, the Court similarly found that the order
before it included a "serious and unsettled question," without stating
this as a collateral order requirement.
Nor did the Court in McDonald, 472 U.S. at 482 n.3, suggest that
the "serious and unsettled question" constituted a necessary ingredient
for a collateral order. The McDonald Court did not explain its own
rationale for entertaining a case arising from a trial court’s denial of
a judgment on the pleadings, but merely noted that the court of
appeals had "accepted jurisdiction under the ‘collateral order’ doc-
trine" on the theory that the "petitioner raised a ‘serious and unsettled
question.’" Id.
In Nixon, 457 U.S. at 742, however, the Court did characterize the
"serious and unsettled" factor as "an additional requirement," which
"Cohen established." And in Hohn, 524 U.S. at 246-47, the Court
recently repeated Nixon’s characterization without disapproval. These
cases do give me pause. However, the context of the Nixon Court’s
statement and the rationale for its holding on this issue seem to me
to render it questionable to rely on this characterization as signaling
establishment of a fourth factor.
In Nixon, 457 U.S. at 741, the Supreme Court was reviewing an
unpublished appellate order, apparently without a stated rationale,
17 UNDER SEAL v. UNDER SEAL

-- 17 of 18 --

"dismiss[ing] summarily" the district court’s denial of immunity to a
former president and his aides. The Nixon Court thus surmised that
the "Court of Appeals appears to have reasoned that petitioner’s
appeal lay outside the Cohen doctrine because it raised no ‘serious
and unsettled question’" since "[t]his argument was pressed by the
respondent" on the basis of an earlier holding in a similar case by the
Court of Appeals. Id. at 742-43. If we are correct as to the limits of
the collateral order test, the Nixon Court could have summarily
rejected this view. But, facing an appellate order issued without
explanation, on a question of great public interest, it is hardly surpris-
ing that the Supreme Court instead determined to meet the issue head
on. Moreover, the ease with which the Nixon Court could dispose of
the "serious and unsettled question" argument perhaps contributed to
its decision to follow this direct approach. Obviously, the Court found
the argument meritless, quickly concluding that notwithstanding the
court of appeals’ earlier decision, the immunity question was unset-
tled because the Supreme Court itself had not resolved it. Id. at 743.
Given this context and holding, I believe the fairest way to read
Nixon is as a discussion of the "serious and unsettled question" which
assumes, but does not decide, that it is a collateral order requirement.
Moreover, in view of the extensive Supreme Court collateral order
precedent post-dating Nixon, which I discuss above, I am satisfied
that even if the Supreme Court in Nixon did intend to establish a
fourth factor, it subsequently changed course.2
2Finally, it is worth noting that even if there were a "serious and unset-
tled question" requirement, the order in this case would still be appeal-
able under the Supreme Court’s articulation of this requirement in Nixon,
457 U.S. at 743. Because that case suggests that an issue remains unset-
tled until the Supreme Court itself has settled the matter, and I am aware
of no Supreme Court case that has definitively settled the issue of when
courts may unseal the record of a previously sealed complaint brought
under the False Claims Act’s qui tam provision, the issue in this case
remains "serious and unsettled."
18 UNDER SEAL v. UNDER SEAL

-- 18 of 18 --