United States v. Apperson

FILED
United States Court of Appeals
Tenth Circuit
March 9, 2016
UNITED STATES COURT OF APPEALS
Elisabeth A. Shumaker
TENTH CIRCUIT Clerk of Court
UNITED STATES OF AMERICA,
Plaintiff-Appellee,
v. Nos. 14-3069 & 14-3070
(D.C. No. 5:00-CR-40104-JTM-2)
CLYDE APPERSON; WILLIAM (D. Kan.)
LEONARD PICKARD,
Defendants-Appellants.
ORDER AND JUDGMENT *
Before HARTZ, HOLMES, and PHILLIPS, Circuit Judges.
Defendants William Leonard Pickard and Clyde Apperson (“Defendants”)
appeal from the denial of their motion to unseal a confidential informant (“CI”)
file. Defendants assert a common-law right of access to the CI file as a judicial
record. Exercising jurisdiction pursuant to 28 U.S.C. § 1291, we conclude that
the district court failed to provide an adequate explanation of its reasoning in
light of the governing legal standards to permit our merits review. Accordingly,
*
This Order and Judgment is not binding precedent, except under the
doctrines of law of the case, res judicata, and collateral estoppel. It may be cited,
however, for its persuasive value consistent with Federal Rule of Appellate
Procedure 32.1 and Tenth Circuit Rule 32.1.
we vacate the district court’s order and remand for further proceedings.
I
In 2003, a jury convicted William Leonard Pickard and Clyde Apperson of
various drug offenses related to the manufacture and distribution of lysergic acid
diethylamide (“LSD”). 1 See United States v. Apperson, 441 F.3d 1162, 1175,
1177 (10th Cir. 2006). The government’s case relied in part on the significant
cooperation and testimony of one of Defendants’ accomplices, Gordon Todd
Skinner. Mr. Skinner had worked as a confidential informant for the U.S. Drug
Enforcement Administration (“DEA”). At trial, the district court ordered the
government to turn over Mr. Skinner’s CI file to the defense. At the same time,
however, the court sealed the file.
In 2011, Defendants moved to unseal Mr. Skinner’s CI file. See United
States v. Pickard, No. 00-40104-01/02-RDR, 2012 WL 1658899, at *1 (D. Kan.
May 9, 2012), rev’d, 733 F.3d 1297 (10th Cir. 2013). Although Defendants’
counsel already had access to an unredacted copy, Defendants sought to unseal
the file in order to use it in connection with ongoing litigation under the Freedom
of Information Act (“FOIA”), 5 U.S.C. § 552, and other proceedings. The
government opposed the motion, arguing that Defendants had failed to show a
1
Defendants were convicted of conspiracy to manufacture and
distribute LSD in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(A), and 846, and of
possession with intent to distribute LSD in violation of 21 U.S.C. §§ 841(a)(1)
and (b)(1)(A). The district court sentenced Mr. Apperson to thirty years’
imprisonment. Mr. Pickard received a life sentence.
2
legitimate need to unseal the file and had not identified any public interest that
outweighed the DEA’s interest in keeping informant files confidential.
The district court denied Defendants’ motion and concluded that the CI file
should remain under seal. The court reasoned:
[S]ome aspects of these matters have been referred to by this
court and mentioned by the government at various times during
these proceedings. Nevertheless, the court fails to find that all of
the information contained in the documents has been revealed.
The court is further persuaded that some of it should not be
revealed. The defendants have failed to indicate exactly why
these documents now need to be unsealed. They have not
specifically explained what information in any of the documents
is necessary for them to use in any appropriate legal proceeding.
The court remains convinced that information concerning
confidential informants should remain private absent a
compelling reason. The court recognizes that much of Skinner’s
life has been placed under the microscope in this case, but we see
no need for any further examination of his past. The court finds
that the defendants have not sufficiently demonstrated the need
for unsealing these documents. The defendants have also failed
to adequately show why the public has any interest in these
documents.
Pickard, 2012 WL 1658899, at *3. Accordingly, it denied Defendants’ motion to
unseal the file.
On appeal, we reversed the district court’s decision. See United States v.
Pickard, 733 F.3d 1297, 1300 (10th Cir. 2013). In relevant part, we concluded
that the court erred in the manner in which it denied Defendants’ motion in three
ways. See id. at 1300, 1303–05. First, it failed to require the government to
articulate a significant interest in continuing to keep the DEA records sealed.
3
Second, the court failed to apply the presumption of public access to judicial
records. And, third, assuming that the government could articulate a significant
interest, the court did not consider whether such an interest would be adequately
served by selectively redacting the documents and unsealing the rest of the file.
Although we left open the possibility that the government could articulate an
interest that would justify keeping at least a portion of the records sealed, we
noted that “[t]he fact that some of the sealed information has already been made
public suggests that much of the information in the DEA records could be
unsealed.” Id. at 1305. We remanded for the district court to reconsider
Defendants’ motion in light of our decision.
On remand, the government argued that unsealing the records would
undercut the need for effective law enforcement, “since investigative files
‘[o]ftentimes [contain] an informant’s personal information, personal information
of investigative subjects or uninvolved third parties, law enforcement methods or
techniques, or other sensitive information,’ the disclosure of which might
‘jeopardize ongoing or future investigations.’” United States v. Pickard, Nos. 00-
40104-01, 00-40104-02-JTM, 2014 WL 789202, at *2 (D. Kan. Feb. 26, 2014)
(alterations in original) (citation omitted). Further, it expressed concern about the
deterrent effect of the routine disclosure of criminal files, which might discourage
informants or witnesses from cooperating with the government, and “would open
witnesses and law enforcement personnel named in those records to physical
4
attacks, threats, harassment, or retaliation.” Id. (citation omitted). Finally, the
government argued that these interests strongly outweigh the public’s limited
interest in a specific criminal investigative file.
The district court, however, granted in part Defendants’ motion to unseal
the file. It found that:
[W]hile the government has shown a strong and compelling
interest in the generalized confidentiality of criminal
investigative records, it has failed to make any specific showing
in this case that these interests cannot be vindicated by selective
redaction. . . .
Here, the government has not affirmatively demonstrated the
existence of any informants or confidential witness other than
Skinner. Neither has the government affirmatively represented
that Skinner’s DEA file (now apparently a decade old) would
imperil any actual ongoing criminal investigations.
Id. (citation omitted). The court imposed the following conditions on its grant of
relief:
[T]he government is authorized to selectively redact from the
materials produced (a) the identity or personal information of
informants other than Skinner; (b) the identity or personal
information of other witnesses or law enforcement officers, other
than those who testified at trial; (c) evidence which would
specifically reveal any heretofore unknown criminal investigatory
technique; or (d) evidence relating to any ongoing criminal
investigation. To the extent the government seeks redaction, it
shall produce to the court in camera both redacted and
unredacted copies of the file, together with an itemized
justification for each redaction. The defendant’s Motion to
Unseal is hereby granted as provided herein.
Id. (citation omitted).
5
Shortly thereafter, the government filed a motion to clarify the district
court’s order. It asserted a good-faith belief that it was already in compliance
with the court’s order based on its prior submissions to the court. Specifically,
the government rested on its prior submission of (1) redacted copies of the file,
along with a Vaughn index, which had been submitted as part of the government’s
earlier memorandum regarding the sealed documents, see generally Vaughn v.
Rosen, 484 F.2d 820, 827–28 (D.C. Cir. 1973) (requiring the government to
establish “a system of itemizing and indexing that would correlate statements
made in the Government’s refusal justification with the actual portions of the
document” sought to be disclosed); 2 and (2) unredacted copies of the file that
were delivered to the court. The government further requested a clarification to
specifically bar Defendants’ counsel from disclosing the DEA records to
Defendants or anyone else.
In response, the district court entered an order granting the government’s
motion and finding that the government was in compliance with its prior order.
See Aplt. App. at 1228–29 (Mem. & Order, dated Mar. 21, 2014). In summary
2
More specifically, “[a] Vaughn index is a compilation prepared by
the government agency (or intervenor) listing each of the withheld documents and
explaining the asserted reason for its nondisclosure.” Anderson v. Dep’t of
Health & Human Servs., 907 F.2d 936, 940 n.3 (10th Cir. 1990). In this case, the
Vaughn index that the government submitted to the court consisted of the list of
documents and associated objections that it had produced in related FOIA
proceedings filed by Mr. Pickard in the United States District Court for the
Northern District of California. See Pickard v. Dep’t of Justice, No. C 06-00185
CRB, 2014 WL 1868841, at *1 (N.D. Cal. May 7, 2014).
6
fashion, the court explained:
The government’s earlier Memorandum of Law references the
California Vaughn index, but does not explicitly state that it
advances the same redactions herein. The court takes the
government’s present motion as an adoption of those redactions,
and determines that the government is in compliance with the
court’s directive.
The government also requests that the court clarify its earlier
Order to specify that the disclosures rendered pursuant to the
Order are not to be further disseminated by counsel for the
defendant, William K. Rork, to any other person without specific
leave of court. This restriction is in keeping with prior
restrictions, and the court adopts the same limitation herein.
Id. (citation omitted).
Defendants, in turn, filed their own motion to clarify the court’s order.
Defendants argued that the motion to unseal related only to the CI file that the
district court had sealed during their trial, as opposed to the more general agency
file on Mr. Skinner maintained by the DEA. 3 And they pointed out that the
Vaughn index that had been prepared in the context of the FOIA litigation in the
United States District Court for the Northern District of California did not
3
According to Defendants, the DEA’s agency file (as opposed to the
CI file sealed by the court) is the subject of the FOIA proceedings in California
and was never requested as part of this action. This distinction is relevant in part
because litigation in this case has proceeded on the assumption that the CI file
sealed by the court constitutes a judicial record. See Pickard, 733 F.3d at 1302
n.5 (“The Government assumes that the DEA file is a judicial record. And we
agree.”). Defendants contend that the common-law right of public access to
judicial records, and not FOIA, provides the controlling standards. And if there
are no judicial records at issue, Defendants concede that there is no common-law
right of access to agency files maintained elsewhere by the DEA.
7
account for all the records contained in the sealed CI file. 4 More generally,
Defendants argued that the Vaughn index was insufficient, and claimed that the
government had made no effort to segregate non-exempt material in its disclosure.
Finally, they requested a modification of the nondisclosure order to allow
Defendants themselves to access the CI file.
The district court granted in part and denied in part Defendants’ motion.
See United States v. Pickard, Nos. 00-40104-01-JTM, 00-40104-02-JTM, 2014
WL 1356053 (D. Kan. Apr. 7, 2014). It first addressed the distinction between
the Skinner-related agency records maintained by the DEA and the CI file sealed
by the court. The court noted that it had reviewed the records, and observed that,
while the CI file and the Vaughn index did not relate to identical documents, “on
the whole the two files ‘significantly overlap’ each other.” Id. at *1 (alteration
omitted) (citation omitted). It characterized the Vaughn index as “precisely the
sort of careful separation and indexing” that the D.C. Circuit’s decision in
Vaughn authorized, and found that Defendants “have shown no basis for rejecting
the Vaughn index itself as a sufficient itemization of the government’s objections
to production.” Id. at *2.
The court then turned to consider the validity of the objections set forth in
4
Because the federal courts are not agencies for purposes of FOIA,
Defendants argued that the court has an obligation to redact its own judicial
records, potentially through the appointment of a special master, and cannot rely
on the government’s redactions in the form of a Vaughn index.
8
the Vaughn index, and found “no reason to vary from its earlier determination that
the government has shown a compelling interest in the confidentiality of the DEA
file.” Id. In the district court’s view:
Even if no ongoing investigation exists, the government has a
strong interest in preventing the disclosure of the file itself,
which would have a dangerous negative effect on the ability to
secure cooperation by other confidential sources in the future.
The court has recognized this importance in sealing the document
simultaneous with its reception. The defendants filed no
challenge to this seal until long afterwards.
The court further finds that the government appropriately
articulates strong and compelling rationales for withholding the
designated information, even if the ultimate result means that
much of the file remains under seal. A careful review of the
sealed DEA file confirms that it is the sort of information which
is typically subjected to the highest level of confidentiality, and
there is no meaningful basis for alternative or additional
redactions.
Id. (citations omitted).
Next, the court specifically addressed our previous observation that “[t]he
fact that some of the sealed information has already been made public suggests
that much of the information in the DEA records could be unsealed.” Id. at *3(alteration in original) (quoting Pickard, 733 F.3d at 1305). It determined “that
the limited disclosure of some information from Skinner’s participation does not
justify the broad disclosure of the majority of the CI file.” Id. After discussing
two cases that we cited in support of this inference, Mann v. Boatright, 477 F.3d
1140 (10th Cir. 2007), and United States v. James, 663 F. Supp. 2d 1081 (W.D.
9
Wash. 2009), it explained:
Neither Mann nor James involved raw files relating to
confidential informants, and the court agrees with the
government that it possesses a strong and compelling interest in
keeping confidential the information generated by confidential
informants. Accordingly, the governmental interest is far
stronger than the privacy interest in Mann and or the interest of
the sentenced defendant in James; in both cases the courts
respectively noted that the relevant documents (whether civil
complaint or plea agreement) are filed publicly as a matter of
course. James authorized the release of “boilerplate language,”
statements made in open court, and the fact of the defendant’s
cooperation with the government. Nevertheless, it explicitly
agreed to redact “the details of Ms. James’ cooperation.” [663
F. Supp. 2d at 1021] (emphasis added).
Pickard, 2014 WL 1356053, at *3.
Finally, the court set forth its view of the governmental interest implicated
by the potential disclosure here. It concluded:
The court finds the complete disclosure of the details of
Skinner’s cooperation would work substantial damage to the
government by creating a chilling effect on future cooperation.
The government’s interest is extremely strong, and the fact that
some limited information from the CI file has been referenced
during the long post-conviction proceedings does not justify the
wholesale release of the entire file. Additional redactions beyond
those contained in the Vaughn index are unjustified in light of the
government’s interest. The court further finds that the relatively
few documents which are in the CI file but not in the Vaughn
index should remain sealed as well. In both cases, a broad
release of the information would be unjustified in light of the
governmental interest involved, and the government interest is
sufficient to overcome the presumption in favor of public access
to judicial records.
Given the highly confidential nature of the informant file, the
glancing and indirect nature of much of the “publication” cited
10
by the defendants, and the inherent chilling effect of a release of
that information on future investigations, the court finds that
specific redactions beyond those identified in the Vaughn index
are insufficient to satisfy the government interest involved.
Accordingly, the court finds the defendants’ Motion to Unseal
shall remain denied, except as provided herein.
Id. at *4 (citation omitted).
The court did, however, provide a limited exception to unseal five
documents—which had already been made public—that Defendants identified in
their motion to clarify. 5 Otherwise, the court denied Defendants’ request to allow
them individualized access to the CI file and, as noted, determined that
Defendants’ motion to unseal should remain denied. Defendants timely appealed.
II
We address the legal standards governing a trial court’s decision to seal
judicial records. Ultimately, we conclude that we cannot rule at this time on the
merits of the district court’s sealing decision; in light of the controlling legal
standards, the district court’s explanation of the reasons for its decision is
inadequate to support and facilitate meaningful appellate review. We remand for
the court to more fully explicate its reasoning, consistent with the principles
5
Specifically, these documents included (a) Mr. Skinner’s Criminal
Felony Docket for Tulsa County, Oklahoma District Court, July 31, 2006; (b) the
Pottawattamie County Kansas Court order dated August 21, 2000; (c) Mr.
Skinner’s Criminal Felony Docket for Tulsa County, Oklahoma District Court
dated March 24, 2004; (d) Mr. Skinner’s Eleven-point Risk Assessment; and (e)
Mr. Skinner’s Confidential Source Agreement Form, DEA form 473, dated
October 18, 2000. See Pickard, 2014 WL 1356053, at *4.
11
discussed in this order and judgment.
A
“We review for an abuse of discretion the district court’s decisions
regarding whether to seal or unseal documents.” Pickard, 733 F.3d at 1302; see
also Mann, 477 F.3d at 1149; Colony Ins. Co. v. Burke, 698 F.3d 1222, 1241 n.28
(10th Cir. 2012) (involving a decision to seal); Nat’l Org. for Marriage v. McKee,
649 F.3d 34, 70 (1st Cir. 2011) (addressing a decision to unseal). But we review
de novo the underlying legal principles that the district court applies in
considering a motion regarding sealing or unsealing. Pickard, 733 F.3d at 1302;
accord Ameziane v. Obama, 620 F.3d 1, 5 (D.C. Cir. 2010); United States v.
Wecht, 484 F.3d 194, 208 (3d Cir. 2007). “We have previously described abuse
of discretion as an arbitrary, capricious, whimsical, or manifestly unreasonable
judgment.” ClearOne Commc’ns, Inc. v. Bowers, 643 F.3d 735, 773 (10th Cir.
2011) (quoting Winnebago Tribe of Neb. v. Stovall, 341 F.3d 1202, 1205–06 (10th
Cir. 2003)). “Of course, applying incorrect legal principles is an abuse of
discretion.” Pickard, 733 F.3d at 1302.
1
“A court has authority to seal documents before it, based upon the court’s
inherent supervisory authority over its own files and records.” Id. at 1300.
However, “[c]ourts have long recognized a common-law right of access to
12
judicial records.” 6 Mann, 477 F.3d at 1149; see Nixon v. Warner Commc’ns, Inc.,
435 U.S. 589, 597 (1978). “The right is an important aspect of the overriding
concern with preserving the integrity of the law enforcement and judicial
processes.” United States v. Hickey, 767 F.2d 705, 708 (10th Cir. 1985).
Nevertheless, the right of access is not absolute. See Colony Ins., 698 F.3d at
1241. The court, “in its discretion, may seal documents if the public’s right of
access is outweighed by competing interests.” Helm v. Kansas, 656 F.3d 1277,
1292 (10th Cir. 2011) (quoting Hickey, 767 F.2d at 708).
“In exercising this discretion, [a court] weigh[s] the interests of the public,
which are presumptively paramount, against those advanced by the parties.” Id.(quoting Crystal Grower’s Corp. v. Dobbins, 616 F.2d 458, 461 (10th Cir. 1980)).
And, as we explained in the prior appeal, the presumption in favor of access is
particularly strong, as here, “where the district court used the sealed documents
‘to determine litigants’ substantive legal rights.’” Pickard, 733 F.3d at 1302
(quoting Colony Ins., 698 F.3d at 1242). As a result, “[t]he burden is on the party
seeking to restrict access to show ‘some significant interest that outweighs the
presumption.’” Colony Ins., 698 F.3d at 1241 (quoting Mann, 477 F.3d at 1149).
“Consistent with this presumption that judicial records should be open to
6
We previously determined that the CI file that the district court
sealed in this case constitutes a judicial record. See Pickard, 733 F.3d at 1302
n.5; see also United States v. El-Sayegh, 131 F.3d 158, 163 (D.C. Cir. 1997)
(“[W]hat makes a document a judicial record and subjects it to the common law
right of access is the role it plays in the adjudicatory process.”).
13
the public, the party seeking to keep records sealed bears the burden of justifying
that secrecy, even where, as here, the district court already previously determined
that those documents should be sealed.” Pickard, 733 F.3d at 1302. Thus, in this
case, the government bears the burden of articulating an interest sufficient to
overcome the strong presumption in favor of public access to the sealed CI file. 7
2
Just as the district court has an obligawewewetion to weigh the competing
interests of the parties and the public, we too are obliged to review the district
court’s exercise of its discretion. In a variety of different contexts, we have
found reversible error where courts have failed to provide a record of their
decisional process that was adequate to support and facilitate meaningful
appellate review. See, e.g., United States v. Chavez-Calderon, 494 F.3d 1266,
1268 (10th Cir. 2007) (“In order to facilitate appellate reasonableness review and
to ensure that the [appropriate] factors have informed a district court’s exercise of
7
The government bears the burden of proof in both the common-law-
access and the FOIA contexts, where it is the party resisting disclosure; however,
the questions that district courts resolve in the FOIA context ordinarily are
deemed issues of law—not of discretion, as in the common-law-access
setting—and therefore the court’s rulings are subject to the more searching de
novo standard of review. See, e.g., Trentadue v. Integrity Comm., 501 F.3d 1215,
1226 (10th Cir. 2007) (“The federal agency resisting disclosure bears the burden
of justifying withholding. Whether a FOIA exemption justifies withholding a
record is a question of law that we review de novo.” (citation omitted)); accord
Stewart v. U.S. Dep’t of Interior, 554 F.3d 1236, 1244 (10th Cir. 2009). We have
no need in this case to fully explore the relationship between these two bodies of
law, including the extent to which they intersect.
14
discretion, we require district courts to give reasons for their sentences.”); OCI
Wyo., L.P. v. PacifiCorp, 479 F.3d 1199, 1204 (10th Cir. 2007) (“[W]e have also
cautioned that too little detail frustrates meaningful appellate review by requiring
the parties and this court to guess at why the district court reached its
conclusion.”); United States v. Roberts, 88 F.3d 872, 882 (10th Cir. 1996) (per
curiam) (“Without any reasoned elaboration by the district court we have no way
of understanding the basis of its decision . . . . As an appellate court, we are in
no position to speculate about the possible considerations which might have
informed the district court’s judgment.”), superseded by statute on other grounds
as recognized in United States v. Franklin, 785 F.3d 1365, 1368 (10th Cir. 2015);
Olcott v. Del. Flood Co., 76 F.3d 1538, 1559 (10th Cir. 1996) (“The district
court’s summary disposition of this issue renders it impossible for us to review
the propriety of its decision. . . . We have no informed basis for evaluating the
propriety of the district court’s action because in the end we are left guessing as
to the court’s purpose.”); Roberts v. Metro. Life Ins. Co., 808 F.2d 1387, 1390–91
(10th Cir. 1987) (“Where the trial court provides only conclusory findings,
unsupported by subsidiary findings or by an explication of the court’s reasoning
with respect to the relevant facts, a reviewing court simply is unable to determine
whether or not those findings are clearly erroneous.” (quoting Lyles v. United
States, 759 F.2d 941, 944 (D.C. Cir. 1985) (per curiam))).
In this case, the district court compared the sealed CI file to the Vaughn
15
index—consisting of a list of documents and associated objections that the
government advanced—and concluded that it was satisfied that the CI information
had been properly withheld. However, the record does not adequately reflect the
court’s balancing—with respect to particular documents or categories of
documents—of the specific interests of the public and the government (the party
opposing disclosure) relative to the factual circumstances of this case. Instead,
the court relied on the government’s general interests regarding confidentiality, a
potential “chilling effect,” and the need for law enforcement to secure the
cooperation of other confidential sources in the future.
Though these matters are unquestionably, in principle, legitimate
governmental interests, they are likely to be present to some degree in virtually
every case where a member of the public seeks access to law-enforcement
informant files. Therefore, lest the common-law presumption of access be
rendered a dead letter as to this class of cases, courts cannot justify denying
disclosure by endorsing such generalized governmental interests. They must
analyze the government’s interests in the context of the specific case—with
respect to particular documents or categories of documents—and explicitly
undergird their conclusions with fact-specific analysis. Absent a particularized
analysis of this type, a district court has no sound legal basis for ruling on the
sealing question. Likewise, without a district court expressly conducting such an
analysis on the record, we cannot engage in meaningful appellate review of its
16
sealing decision. Cf. Wiener v. FBI, 943 F.2d 972, 987–88 (9th Cir. 1991)
(“‘Disclosure of the factual and legal basis for the trial court’s decision is
especially compelling in FOIA cases.’ . . . ‘The reviewing court should not be
required to speculate on the precise relationship between each exemption claim
and the contents of the specific document[s].’” (alteration in original) (quoting
Van Bourg, Allen, Weinberg & Roger v. NLRB (Van Bourg I), 656 F.2d 1356,
1357, 1358 (9th Cir. 1981) (per curiam))). And that is regrettably the state of
play here. In ruling that the government had offered enough to prevail on the
sealing issue, the court provided no particularized explanation that was rooted in
the specific documents and facts of this case.
Furthermore, we cannot cobble together the elements of such a
particularized explanation by reference to the Vaughn index that the district court
accepted as the foundation for its ruling. That is because the index does a very
poor job of revealing in a particularized fashion the government’s interests
regarding the disclosure of specific documents or categories of documents, and, in
addition, does not shed meaningful light on the question that we gave high
salience in our earlier Pickard decision—that is, it does not meaningfully convey
the government’s position regarding what documents have “already been made
public,” which would “suggest[] that much of the information in the DEA records
17
could be unsealed.” 733 F.3d at 1305. 8
Notably, in the FOIA litigation in the Northern District of California, the
district court rejected as inadequate the exact Vaughn index that the district court
relied on in this case. It characterized the index as “conclusory and circular” and
“supremely unhelpful.” Pickard, 2014 WL 1868841, at *1, *4. Continuing its
assessment of the index, the court stated, “[I]t provides neither the Court nor [Mr.
Pickard] with any useful information about the content of the documents or how
the claimed exemptions apply.” Id. at *4. As a result, “[i]f [Mr. Pickard] or the
Court wishes to do anything with such representations other than unquestioningly
accept them, there is no way to do so.” Id. In light of what the court perceived as
the Vaughn index’s gross inadequacy, it ordered the government to submit an
adequate version. We agree with the California district court’s disapproving
8
As noted, the government submitted here a Vaughn index that it had
initially prepared in connection with FOIA litigation involving Mr. Pickard in the
Northern District of California. We recognize that at issue in both the FOIA and
common-law-access contexts is the public’s right of access to materials that have
been shielded from disclosure. We do not question, as a general matter, that
under certain circumstances, the government may prevail in carrying its burden in
the common-law-access context by submitting a Vaughn index (as well as perhaps
a related declaration). However, in this case, the government’s Vaughn index was
woefully inadequate to shed light on the variables relevant to the public-access
question. Moreover, the utility and responsiveness of the Vaughn index certainly
was not bolstered by the fact that the universe of Skinner-related documents
contemplated by the index was not precisely the same as the CI file.
18
assessment of the inadequacies of the Vaughn index. Consequently, we could not
successfully use it to fill the gaps in the district court’s generalized reasoning
here.
Furthermore, though the CI file is part of the appellate record, we decline
to review the documents therein and conjure up particularized governmental
interests and concerns that might justify nondisclosure of them. Such an exercise
would be:
(1) patently speculative and conjectural, Anderson v. Dep’t of
Health & Human Servs., 907 F.2d 936, 944 (10th Cir. 1990)
(“Neither the district court’s ruling from the bench nor the
subsequent written order make clear which documents the court
believed were exempt as ‘trade secrets,’ and which documents
were exempt on other grounds. Any attempt on our part to
determine which of defendants’ seven proposed classifications of
the enormous number of documents at issue were adopted by the
district court as trade secrets would be sheer speculation.”);
(2) logically at odds with the presumption in favor of unsealing
judicial records, which requires the party opposing unsealing
(i.e., the government) “to show ‘some significant interest that
outweighs the presumption,’” Colony Ins., 698 F.3d at 1241
(emphasis added) (quoting Mann, 477 F.3d at 1149); and
19
(3) at the very least, an inapt role for an appellate tribunal, 9 see,
e.g., New England Health Care Emps. Pension Fund v. Woodruff,
512 F.3d 1283, 1290–91 (10th Cir. 2008) (“We prefer to assess
the justification [for a bar order] in the first instance on the basis
of concrete facts found by the district court, and with the
assistance of the district court’s full consideration and discussion
of all of the relevant facts of the instant case and a full discussion
of the relevant persuasive authorities and the underlying reasons
and policies justifying whatever order the district court
ultimately approves.” (alteration in original) (quoting AAL High
Yield Bond Fund v. Deloitte & Touche LLP, 361 F.3d 1305, 1312
(11th Cir. 2004))); see also Davis v. United States, 192 F.3d 951,
961 (10th Cir. 1999) (“The district court’s memorandum order
neither states a set of facts relevant to the indispensability
analysis nor weighs the factors argued by the parties. This court
9
Relatedly, it would be particularly imprudent to undertake such an
initial analysis where our standard of review is abuse of discretion because this
standard allows for a range of reasonable outcomes. Cf. United States v. Mumma,
509 F.3d 1239, 1245 (10th Cir. 2007) (“[A]s we have previously explained, a
range of reasonable sentences may exist in any given case. The reasonableness of
one sentence does not, therefore, necessarily render a different sentence
unreasonable by comparison.” (citation omitted)); United States v. Angel-
Guzman, 506 F.3d 1007, 1020 (10th Cir. 2007) (“A decision upholding a
sentencing court’s discretion to sentence below the Guidelines is not precedent
for holding that another sentencing court abuses its discretion when it declines to
do so.”).
20
should not and thus will not perform the fact-finding function
reserved for the district courts.”).
In sum, we conclude that the district court’s generalized analysis of the
sealing question does not provide us with an adequate foundation for appellate
review, and we decline to undertake in the first instance a sealing analysis to
resolve the question. Accordingly, we are constrained to vacate the district
court’s order and remand for further proceedings—notably, for a fuller exposition
of the court’s reasoning.
B
As noted, on remand, the district court must analyze the government’s
interests in the context of this specific case—with respect to particular documents
or categories of documents in the CI file—and explicitly rest its sealing
conclusions on that fact-specific analysis. Given the current state of the record,
the court will encounter—at the very least—significant hurdles in accomplishing
this task. As noted, the Vaughn index in the record—which ostensibly embodies
the government’s primary statement of interests—is woefully inadequate. On
remand, the government should be given an opportunity to supplement its
showing regarding sealing or unsealing the documents in the CI file. As we noted
in an analogous context in Anderson,
The district court is entitled to considerably more assistance
. . . than it received in this case. The burden of adequately
describing the documents and of defining any narrowly-drawn
categories that may be used to deal with similar documents as a
21
group should normally be placed in the first instance on the
agency or other party claiming protection for the documents
requested. The district court’s task is essentially one of review,
but neither it nor the appellate court can adequately fulfill this
function without more particularized descriptions than those that
were provided here.
Anderson, 907 F.2d at 944 n.10.
If the government elects to supplement its showing by providing a new
Vaughn index (with or without an explanatory sworn declaration) it would be
well-advised to heed the counsel provided by the California district court when it
rejected the Vaughn index currently before us:
In W[ie]ner, [943 F.2d] at 984, the Ninth Circuit found
insufficient the Vaughn Index the FBI prepared in response to a
FOIA request about John Lennon, and explained what should
have been done as to a particular document:
Without violating the privacy interests of the informant
or the third party, the FBI could have stated that HQ-8
recites information provided by a third party to an FBI
informant detailing the third party’s knowledge of
several activists and protest activities planned at the
1972 Republican National Convention, discussing the
possibility that John Lennon would organize a series of
concerts to raise money to finance the activity, and
describing rivalries and jealousies within activist
organizations.
Id. at 984. Boilerplate explanations for withholding are
improper, and efforts must be “made to tailor the
explanation to the specific document withheld.” Id. at
978–79.
22
Pickard, 2014 WL 1868841, at *4–5 (emphasis added). 10
In this regard, we are permitted to take judicial notice of the records of
other courts. See, e.g., United States v. Ahidley, 486 F.3d 1184, 1192 n.5 (10th
Cir. 2007) (“Although we are not obliged to do so, we may exercise our discretion
to take judicial notice of publicly-filed records in our court and certain other
courts concerning matters that bear directly upon the disposition of the case at
hand.”); see also St. Louis Baptist Temple v. Fed. Deposit Ins. Corp., 605 F.2d
1169, 1172 (10th Cir. 1979) (“[I]t has been held that federal courts, in appropriate
circumstances, may take notice of proceedings in other courts, both within and
without the federal judicial system, if those proceedings have a direct relation to
10
In this same vein, the government should be alert to some of the
Wiener court’s more pointed criticism of the Vaughn index:
The index fails to tie the FBI’s general concern about disclosure
of confidential sources to the facts of this case. The index does
not describe any particular withheld document, identify the kind
of information found in that document that would expose the
confidential sources, or describe the injury to national security
that would follow from the disclosure of the confidential source
of the particular document. The FBI must have made such an
analysis in concluding that disclosure of some informants or
classes of informants would damage national security and
disclosure of others would not. Yet none of the information and
analysis necessarily considered is made available to Wiener or
the court. The index simply relies on general assertions that
disclosure of certain categories of facts may result in disclosure
of the source and disclosure of the source may lead to a variety
of consequences detrimental to national security.
Wiener, 943 F.2d at 981 (footnotes omitted).
23
matters at issue.”). And, having elected to do so, we are aware that (after one
more failed attempt), the government was able to produce a Vaughn index in the
California FOIA litigation that satisfied the district court’s specificity concerns;
notably, the index directly addressed a question similar to the one that we gave
prominence in Pickard—viz., what documents had already been officially released
in the public domain. See Pickard v. Dep’t of Justice, No. 06-cv-00185-CRB, slip
op. at 3 (N.D. Cal. Aug. 27, 2015) (“These descriptions are sufficient to inform
Pickard of why a particular document is exempt from disclosure and afford him
an opportunity to advocate for its release. . . . Moreover, the government
responded to [the magistrate’s] request that it explain which documents have been
released to the public.” (citations omitted)); see also Pickard v. Dep’t of Justice,
No. 06-cv-00185-CRB, 2015 WL 926183, at *1 (N.D. Cal. Feb. 19, 2015)
(stating, in rejecting the second Vaughn index, “After reviewing the most recent
Vaughn index and the responsive documents, this Court finds that the
index—even when viewed in combination with the government’s supporting
declaration—fails to sufficiently describe the withheld documents in adequate
detail so as to allow Pickard to challenge the government’s claimed
exemptions.”). The government would be wise to demonstrate a similar
willingness here to offer a particularized justification for its efforts to keep the CI
file documents sealed.
At bottom, the government bears the burden of demonstrating “some
24
significant interest that outweighs the presumption” in favor of public access.
Colony Ins., 698 F.3d at 1241 (quoting Mann, 477 F.3d at 1149). On remand, the
district court must assess the record with an eye toward providing a particularized
analysis—one that is suitable for possible appellate review—of the government’s
interests in sealing the documents in the CI file. In that analysis, the court may
reference the explanations that the government provides—through a new Vaughn
index or otherwise—and also, with appropriate care and circumspection, point to
individual documents or categories of documents in the CI file itself. The court
should then rule on the sealing questions as they relate to the documents in the
file. See Wiener, 943 F.2d at 988 (“After receiving an adequate Vaughn index
and conducting any additional proceedings the district court deems necessary on
remand, the court must ‘state in reasonable detail the reasons for its decision as to
each document in dispute.’” (quoting Van Bourg I, 656 F.2d at 1358)).
Our intention is not to choreograph every step of the district court’s
decision-making process or establish unnecessary boundaries on its exercise of
discretion. However, in weighing the interests of the parties and the public, we
underscore that the asserted interests for sealing cannot be generic interests that
would apply with equal force to every case. The government must articulate
specific interests that apply in the context of this case, and the district court must
balance those interests against the public’s interest in access.
25
C
Finally, Defendants have requested that (1) a copy of the CI file that was
transmitted to our court be provided to Defendants’ counsel, subject to a non-
dissemination order, and (2) the “Risk Assessment” portion of the file, which the
district court unsealed, be filed publicly and placed on the docket for access by
the parties. In light of our decision to remand this action for further substantive
analysis, we leave the resolution of these requests in the first instance to the
discretion of the district court.
III
For the foregoing reasons, we VACATE the district court’s order and
REMAND for further proceedings.
Entered for the Court
JEROME A. HOLMES
Circuit Judge
26