RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 24a0244p.06
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
┐ UNITED STATES OF AMERICA, │ Plaintiff-Appellee, │ │ v. > Nos. 23-5821 │ │ MARTAVIOUS KINCAIDE, │ Defendant, │ │ │ DEANDRE SWAIN, │ Proposed Intervenor-Appellant. │ ┘
Appeal from the United States District Court for the Western District of Kentucky at Louisville. No. 3:22-cr-00115-1—David J. Hale, District Judge.
Argued: April 30, 2024
Decided and Filed: October 28, 2024
Before: GIBBONS, BUSH, and LARSEN, Circuit Judges. _________________
COUNSEL
ARGUED: William E. Sharp, LOUISVILLE METRO PUBLIC DEFENDER’S OFFICE, Louisville, Kentucky, for Appellant. Amanda E. Gregory, UNITED STATES ATTORNEY’S OFFICE, Louisville, Kentucky, for Appellee. ON BRIEF: William E. Sharp, LOUISVILLE METRO PUBLIC DEFENDER’S OFFICE, Louisville, Kentucky, for Appellant. Amanda E. Gregory, UNITED STATES ATTORNEY’S OFFICE, Louisville, Kentucky, for Appellee. Nos. 23-5821 United States v. Kincaide, et al. Page 2
_________________
OPINION _________________
LARSEN, Circuit Judge. Deandre Swain is a criminal defendant in Kentucky state court. He moved to intervene in another person’s federal criminal case, claiming a First Amendment right to access a sealed document. Swain believes that the sealed document might contain a cooperation agreement that could aid his defense in state court. The district court denied his motion to intervene. For the reasons that follow, we AFFIRM.
I.
In May 2023, Martavious Kincaide pleaded guilty to three federal firearms offenses in the Western District of Kentucky. Later that month, Deandre Swain filed a motion to intervene in Kincaide’s case. Swain sought to unseal a docket entry in Kincaide’s case, entitled “Plea Agreement Supplement.” The Western District of Kentucky requires that a “plea supplement” be filed in the docket of every criminal case resolved by guilty plea. United States District Court Western District of Kentucky, General Order 2010-06, https://www.kywd.uscourts.gov/general- orders-2010 (last accessed Aug. 16, 2024). Each plea supplement “will contain either a cooperation agreement or a statement that no such agreement exists.” Id. Under the District Court policy, all plea supplements are sealed, which ensures that “each case appear[s] identical” to the public. Id. Because plea supplements are sealed, third parties cannot determine whether any particular defendant’s plea contains an agreement to cooperate with the government—thus potentially protecting cooperators from those who wish them harm.
Swain is facing Kentucky charges for murder and wanton endangerment stemming from the death of Tamal Wood in a drive-by shooting. Swain’s motion to intervene alleged that Kincaide is a witness in the Wood homicide case. Kincaide was allegedly with Wood at the time of the shooting. Swain, presumably hoping to impeach Kincaide in the event that Kincaide Nos. 23-5821 United States v. Kincaide, et al. Page 3
testifies against him, “seeks to uncover any inducements or other government-conferred benefits granted Mr. Kincaide.”1 R. 24, Motion to Intervene, PageID 71.
Swain asserts a First Amendment right of access to Kincaide’s plea supplement. In his motion to intervene, he claimed an interest in the contents of Kincaide’s plea supplement, “both as an individual who enjoys a First Amendment-protected right to access and as a criminal defendant who maintains his innocence of the offenses for which he stands accused.” Id. Swain contended that General Order 2010-06 could not justify withholding access to the plea supplement because it was a facially unconstitutional blanket order of closure. The government opposed Swain’s motion.
The district court denied the motion, holding that General Order 2010-06 is consistent with the First Amendment because it “is the narrowest method of achieving the compelling interest of protecting the safety of cooperators.” R. 32, District Court Order, PageID 171. And Swain’s specific interest in the sealed material did not supersede the policy of compliance with General Order 2010-06 because “consistency is a necessary element of the District’s General Order.” Id. at 171–72. Swain timely appealed the denial of his motion to intervene.2 II.
A.
The Supreme Court has held that “the press and general public have a constitutional right of access to criminal trials.” Globe Newspaper Co. v. Superior Ct., 457 U.S. 596, 603 (1982); see generally Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555 (1980) (plurality opinion). The Sixth Amendment by its terms protects the right of “the accused” to “a speedy and public trial.” U.S. Const. amend VI. The First Amendment, by contrast, does “not explicitly mention[]” a “right of access to criminal trials.” Globe Newspaper, 457 U.S. at 604. Yet the
1 If Kincaide does in fact testify against Swain, the state prosecutors will have to reveal any cooperation agreement of which they have been made aware. See Giglio v. United States, 405 U.S. 150 (1972). But any required disclosure would come later, as the right to receive such exculpatory material is a trial right. See United States v. Ruiz, 536 U.S. 622, 629–31 (2002). 2 When a non-party is aggrieved by a final order denying its motion before the district court, we have jurisdiction under 28 U.S.C. § 1291 to hear an appeal by the non-party. See In re Siler, 571 F.3d 604, 608–09 (6th Cir. 2009). Nos. 23-5821 United States v. Kincaide, et al. Page 4
Supreme Court has held that the First Amendment is “broad enough to encompass those rights that . . . are nonetheless necessary to the enjoyment of other First Amendment rights.” Id. The Court has deemed the right of access to criminal trials to be one such right on the ground that “a major purpose of that Amendment was to protect the free discussion of governmental affairs” and “to ensure that the individual citizen can effectively participate in and contribute to our republican system of self-government.” Id. (citations omitted).
The First Amendment right of access is not limited to the criminal trial itself. The Supreme Court has extended the right to certain other criminal proceedings, such as preliminary hearings. See Press-Enter. Co. v. Superior Ct., 478 U.S. 1, 10 (1986) (Press-Enter. II). And this court has extended the right to some civil or administrative proceedings, like deportation hearings, see Detroit Free Press v. Ashcroft, 303 F.3d 681, 700 (6th Cir. 2002), and to certain documents in criminal proceedings, such as those “pertaining to” a motion to disqualify the trial judge. Application of Nat’l Broad. Co., 828 F.2d 340, 344 (6th Cir. 1987).
To determine whether the right of access attaches, we apply the “experience and logic” test articulated by the Supreme Court in Press-Enterprise II. 478 U.S. at 9.3 We ask (1) whether the proceeding or material has “historically been open to the press and the general public,” and (2) whether “public access plays a significant positive role in the functioning of the particular process in question.” Press-Enter. II, 478 U.S. at 8. If the answer to those questions is “yes,” then a qualified right attaches, and sealing is permissible only if it “is essential to preserve higher values and is narrowly tailored to serve that interest.” Press-Enter. Co. v. Superior Ct., 464 U.S. 501, 510 (1984) (Press-Enter. I). A district court’s justification for sealing under that standard must include specific findings such “that a reviewing court can determine whether the closure order was properly entered.” Id.
3 Swain at times suggests that the test is whether the document qualifies as a “judicial record.” But that is not the test for a First Amendment right of access. See Stevens v. Mich. State Ct. Admin. Off., 2022 WL 3500193, at *5 (6th Cir. 2022). The First Amendment provides a presumptive right of access to some “documents and records that pertain to a proceeding” in court, but that is because they pass the Supreme Court’s “experience and logic” test. Nat’l Broad., 828 F.2d at 344; see Stevens, 2022 WL 3500193, at *5–7. We have no occasion to comment on other doctrines that might require access to judicial records, such as the common law right of access, because Swain raises only a First Amendment claim here. Nos. 23-5821 United States v. Kincaide, et al. Page 5
B.
The threshold question is whether Swain has a qualified First Amendment right to access Kincaide’s sealed plea supplement—in other words, whether the “experience and logic” test supports recognizing a public right of access to cooperation agreements. The answer is “no.” In short, neither experience nor logic favors Swain (and to succeed, both must). See In re Search of Fair Fin., 692 F.3d 424, 431 (6th Cir. 2012); In re Reps. Comm. for Freedom of the Press, 773 F.2d 1325, 1332 (D.C. Cir. 1985) (Scalia, J.). Because no qualified First Amendment right of access attaches to cooperation agreements, we need not reach the question whether General Order 2010-06 is “essential to preserve higher values” and “narrowly tailored.” Press-Enter. I, 464 U.S. at 510. 1.
First, experience. Under the experience prong, we ask whether the relevant proceeding or record has “historically been open to the press and general public.” Press-Enter. II, 478 U.S. at 8. We do not limit our focus to the tradition of any particular jurisdiction; rather we seek an “established and widespread tradition” of open access to the “type or kind of [proceeding or record] throughout the United States.” El Vocero de P.R. v. Puerto Rico, 508 U.S. 147, 150 (2004) (quoting Rivera-Puig v. Garcia-Rosario, 983 F.2d 311, 323 (1st Cir. 1992)).
In each of the Supreme Court’s cases finding a First Amendment right of access to a proceeding, the Court pointed to a tradition of public access dating at least to the early nineteenth century. For example, in Richmond Newspapers, the plurality opinion traced the tradition of open criminal trials from “the days before the Norman Conquest” through “the judicial systems of colonial America,” to the adoption of the Constitution. 448 U.S. at 565, 567. Both Press-Enterprise cases also recounted a long and widespread tradition of public access. In Press-Enterprise I, the Court traced access to jury-selection proceedings from the Norman Conquest to the Founding, noting that “[p]ublic jury selection . . . was the common practice in America when the Constitution was adopted.” 464 U.S. at 508. And in Press-Enterprise II, the Court described a “near[ly] uniform practice” of public access to preliminary hearings, citing the 1807 trial of Aaron Burr as a prominent example. 478 U.S. at 10. The Court’s emphasis on longstanding historical practice is also largely consistent with its recent approach to determining Nos. 23-5821 United States v. Kincaide, et al. Page 6
the scope of rights guaranteed by the First Amendment. See, e.g., Vidal v. Elster, 602 U.S. 286, 295–308 (2024); Kennedy v. Bremerton Sch. Dist., 597 U.S. 507, 535–36 (2022).
This circuit’s cases applying the experience prong have also demanded a long-established and widespread tradition of public access, though not necessarily one dating to the Founding era. In Detroit Free Press, we considered whether the public has a First Amendment right of access to deportation hearings. 303 F.3d at 700–05. Applying the experience test, we noted that “[a] historical tradition of at least some duration is obviously necessary” otherwise “nothing would separate the judicial task of constitutional interpretation from the political task of enacting laws currently deemed essential.” Id. at 701 (alteration in original) (quoting Freedom of the Press, 773 F.2d at 1332 (Scalia, J.)). But we rejected the notion that a tradition of openness could count only if it predated “the adoption of the Bill of Rights.” Id. at 700. After all, we remarked, Press-Enterprise II “relied on exclusively post-Bill of Rights history.” Id. Still, it must be noted, Press-Enterprise II invoked a long and “near uniform practice of state and federal courts” conducting “preliminary hearings in open court,” dating from the 1807 “Burr [case] until the present day.” 478 U.S. at 10. And while we surmised that “a brief historical tradition might be sufficient to establish a First Amendment right of access where the beneficial effects of access to that process are overwhelming and uncontradicted,” Detroit Free Press, 303 F.3d at 701, we had no occasion to apply that proposition. Instead, reviewing the history of deportation and exclusion proceedings dating back to the “first general immigration act . . . enacted in 1882,” we concluded that the experience test was satisfied because “deportation proceedings historically have been open.” Id.; see also Nat’l Broad., 828 F.2d at 344 (finding experience prong met where a survey of “reported Sixth Circuit cases involving the disqualification of judges from 1924 to 1984[,]” in addition to those cited in the annotation to the judicial recusal statute, revealed none “in which the proceedings were closed or the record sealed”).
On the other hand, in In re Morning Song Bird Food Litigation, we rejected a claim that the public possesses a First Amendment right to access objections to a presentence report (PSR). 831 F.3d 765, 778 (6th Cir. 2016). We concluded that no such right exists because PSR “[o]bjections have not been historically available to the public.” Id. at 777. We explained that “before 1975, most courts did not permit the defendant to access his own PSR, much less Nos. 23-5821 United States v. Kincaide, et al. Page 7
comment on or object to it.” Id. And in their “relatively short history” PSR objections had not been broadly available to the public. Id.4
Here, there is no widespread, historical tradition of public access to cooperation agreements. Cooperation agreements of one sort or another appear to have deep roots in American legal practice. For example, The Whiskey Cases describe a phenomenon in which accomplice testimony was typically rewarded with an “equitable right” to a recommendation of pardon. 99 U.S. 594, 602 (1878). Commentators have described this practice as “informal” and subject to “prosecutorial discretion.” Graham Hughes, Agreements for Cooperation in Criminal Cases, 45 Vand. L. Rev. 1, 7–8 (1992). Swain does not discuss this history, much less suggest a tradition of openness. Swain instead rests much of his argument on an analogy to the more general practice of plea bargaining. But Swain acknowledges that plea agreements historically were made “sub rosa . . . shrouded in secrecy and deliberately concealed by participating defendants, defense lawyers, prosecutors, and even judges.” Appellant Reply Br. at 8 (quoting Blackledge v. Allison, 431 U.S. 63, 76 (1977)).
Today, federal cooperation agreements are largely formalized under the U.S. Sentencing Guidelines, which became effective in 1987.5 Under the Guidelines, a sentencing judge may provide a downward departure if the prosecutor files a motion “stating that the defendant has
4 Fair Finance and United States v. Miami University also rejected claims to a qualified right of access—to search warrant documents and student disciplinary records, respectively. 692 F.3d at 430–31; 294 F.3d 797, 823 (6th Cir. 2002). But neither case had occasion to elaborate on the experience prong because the records at issue had never been publicly available. Fair Fin., 692 F.3d at 430–31; Miami Univ., 294 F.3d at 823. 5 Federal cooperation agreements may also take other forms. For example, under Federal Rule of Criminal Procedure 35(b), as revised in 1984, a court may reduce a sentence previously imposed to reflect a defendant’s post- sentencing “substantial assistance in investigating or prosecuting another.” Fed. R. Crim. P. 35(b)(1); see also Pub. L. 98-474, Title II, § 215(b), 98 Stat. 2015. Motions under § 5K1.1 are more common than those under Rule 35(b), but “the primary distinction between [them] is temporal.” U.S. Sent’g Comm’n, The Use of Federal Rule of Criminal Procedure 35(b), at 4, 7 (2016), available at https://www.ussc.gov/sites/default/files/pdf/research-and- publications/research-publications/2016/Rule35b.pdf. Federal cooperation may also be rewarded in a non- prosecution or deferred prosecution agreement. See, e.g., U.S. Dep’t of Just., Just. Manual §§ 9-27.600, 620, 630, 640; Roth et al., supra, at 1359 n.10. Such agreements are generally entered and fulfilled out of public view. Informal immunity agreements typically unfold without judicial involvement and, thus, would not become public knowledge absent disclosure by the parties or a request for judicial intervention to remedy a breach. See, e.g., United States v. Minn. Mining & Mfg. Co., 551 F.2d 1106, 1112 (8th Cir. 1977) (an agreement not to prosecute did not involve the trial judge as “a participant” but was instead “a prosecutorial agreement, the inviolability of which rested completely in the province of the government prosecutors, who have the sole power and responsibility to institute criminal proceedings”); United States v. Fitch, 964 F.2d 571, 574 (6th Cir. 1992). Nos. 23-5821 United States v. Kincaide, et al. Page 8
provided substantial assistance in the investigation or prosecution of another person who has committed an offense.” U.S.S.G. § 5K1.1. These cooperation agreements (or “substantial assistance” agreements) are products of the mid-1980s. In other words, they have a relatively brief history.
Modern substantial assistance agreements have also failed to produce any consistent national practice—much less one of uniform public access. For years, judges, scholars, and practitioners have observed that “substantial assistance” motion practice in federal courts is disuniform and frequently opaque. See, e.g., Jessica A. Roth, Anna D. Vaynman & Steven D. Penrod, Why Criminal Defendants Cooperate: The Defense Attorney’s Perspective, 117 Nw. U. L. Rev. 1351, 1365 (2023); ADMIN. OFF. OF THE U.S. CTS., FINAL REPORT OF THE TASK FORCE ON PROTECTING COOPERATORS 2 (2018) [https://perma.cc/F33E-M2FS]; Daniel C. Richman, The
Challenges of Investigating Section 5K1.1 in Practice, 11 Fed. Sent’g Rep. 75, 75–76 (1998); Judge Stanley Marcus, Substantial Assistance Motions: What Is Really Happening?, 6 Fed. Sent’g Rep. 6, 6–7 (1993). Of particular relevance here, federal district courts across the country “vary considerably in the extent to which information referencing cooperation is discussed in open court or filed on the public docket rather than kept under seal.” Roth, et al., supra, at 1365.
The Western District of Kentucky’s sealing policy is far from an outlier. As of 2016, thirty-three federal districts “ma[de] criminal cases appear identically on CM/ECF to obscure cooperation information (such as requiring filing sealed supplements with a plea agreement).” MARGARET S. WILLIAMS, DONNA STIENSTRA & MARVIN ASTRADA, FED. JUD. CTR., SURVEY OF HARM TO COOPERATORS: FINAL REPORT 26 (2016). Thirty-seven districts sealed documents containing cooperation agreements sua sponte, and sixty-six districts sealed cooperation information at the parties’ request. Id. Not exactly a national tradition of public access.6
6 Swain suggests that, pre-PACER, there was a tradition of public access. But Swain’s evidence is scant. He points to one oral remark made by an individual judge speaking at a Judicial Conference meeting and one line in a 2016 memo from the Committee on Court Administration and Case Management. Those statements suggest that, pre-PACER, it was easier to access cooperator information, but they are only generalized assertions about the impacts of technological changes. They neither provide any empirical data about pre-PACER access nor discuss district-by-district or state-by-state pre-PACER access at all. They do not come close to establishing a widespread tradition of public access, for even a brief period between the Guidelines taking effect and PACER’s advent. Nos. 23-5821 United States v. Kincaide, et al. Page 9
Nor does the record reveal a widespread history of access to cooperation agreements at the state level. See El Vocero, 508 U.S. at 150–51 (“The established and widespread tradition of open preliminary hearings among the States was canvassed in Press-Enterprise [II] and is controlling here.”). The parties have given us no information on state practice. And even scholars seem to know little. “Because most states, even those that adopted their own sentencing guidelines, do not have an analogue to Section 5K1.1, cooperation at the state level is much more difficult to measure.” Roth et al., supra, at 1361 n.14. State practice seems to involve substantial variation, though “it is widely recognized that many plea agreements with cooperating witnesses at the state level are informal and unwritten.” R. Michael Cassidy, “Soft Words of Hope:” Giglio, Accomplice Witnesses, and the Problem of Implied Inducements, 98 Nw. U. L. Rev. 1129, 1148 (2004).
In sum, Swain has cited no analogue from early American history that would establish a national tradition of public access to “substantial assistance” cooperation agreements. Nor are we aware of one. And looking to more recent history proves no better. Like the PSR objections in Morning Song, federal substantial assistance agreements in the form of § 5K1.1 motions are fairly new. And there has been no unbroken tradition of public access since their inception. See El Vocero, 508 U.S. at 150. Rather, at the federal level, different judicial districts have different policies that provide varying levels of public access to cooperation agreements. Nor do we have any evidence of a “widespread tradition of open” access to cooperation agreements “among the States.” Id. at 150–51. Experience, in short, reveals no national “tradition of accessibility.” Press-Enter. II, 478 U.S. at 10; El Vocero, 508 U.S. at 150. “That alone requires a rejection” of Swain’s assertion that a qualified First Amendment right of access attaches to cooperation agreements. Fair Fin., 692 F.3d at 431; see also Hils v. Davis, 52 F.4th 997, 1002 (6th Cir. 2022). Nonetheless, we offer a few observations about the logic prong as well.
2.
The logic prong instructs us to ask whether “public access plays a significant positive role in the functioning of the particular process in question.” Press-Enter. II, 478 U.S. at 8. But the Supreme Court has provided little guidance on how to apply this test. The Court has noted that “many governmental processes operate best under public scrutiny.” Id. At the same time, the Nos. 23-5821 United States v. Kincaide, et al. Page 10
Court has made clear that “some kinds of government operations . . . would be totally frustrated if conducted openly.” Id. at 9. For instance, the “grand jury system depends upon the secrecy of [the] proceedings.” Id. (citation omitted). Beyond this, the Court’s cases have provided little guidance on how the logic prong weighs the benefits of access against competing concerns.
Swain argues that the logic prong favors public access because it “provides needed scrutiny of prosecutorial discretion and judicial sentencing decisions which, in turn, promotes public confidence in the administration of justice.” Appellant Reply Br. at 10. There is some merit to this point: cooperation agreements are no doubt relevant to the administration of justice, and public scrutiny of some aspects of the criminal justice system is often beneficial. E.g., Globe Newspaper Co., 457 U.S. at 606; Press-Enter. I, 464 U.S. at 508–10. In conducting the logic inquiry, this court has noted that “sunshine” can benefit various proceedings. For example, in Detroit Free Press, we reasoned that public access to deportation proceedings “ensures that government does its job properly” and “acts as a check on the actions of the Executive by assuring us that proceedings are conducted fairly.” 303 F.3d at 703–04; see also United States v. DeJournett, 817 F.3d 479, 485 (6th Cir. 2016) (noting that access to plea agreements helps “in monitoring the administration of justice by plea”).
But the same could be said of shining light on nearly any government function; and the countervailing interests here bear a strong resemblance to those our court has deemed sufficient to defeat the claim that “public access plays a significant positive role in the functioning of the particular process in question.” Press-Enter. II, 478 U.S. at 8; see also Fair Fin., 692 F.3d at 433. Public access to cooperation agreements endangers government cooperators, and it may also frustrate government investigations.
“Snitching” is a dangerous business. Between approximately 2014 and 2017, nearly 600 government cooperators were either harmed or threatened, and the Wall Street Journal reported that 61 were murdered. WILLIAMS, STIENSTRA & ASTRADA, supra, at 8; Jacob Gershman, Why Life for ‘Snitches’ Has Never Been More Dangerous, WALL ST. J. (June 20, 2017), https://www.wsj.com/articles/criminals-subvert-online-court-records-to-expose-snitches- 1497960000 [https://perma.cc/P9Z3-PX2Z]. Websites like “Whosarat.com” aggregate public information to identify government cooperators. See Adam Liptak, Web Sites Listing Nos. 23-5821 United States v. Kincaide, et al. Page 11
Information Concern Justice Dept., N.Y. TIMES (May 22, 2007), https://www.nytimes.com/2007/05/22/washington/22plea.html [https://perma.cc/MXJ9-LQPN]. A 2016 survey even found that, where it was disclosed, the plea agreement or supplement was “the document most frequently used to identify a defendant/offender as a cooperator.” WILLIAMS, STIENSTRA & ASTRADA, supra, at 13. In short, public access to cooperation agreements significantly jeopardizes cooperators’ safety.
In Fair Finance and Morning Song, analogous concerns led us to conclude that logic did not support a right of access to either search-warrant documents or PSR objections. In Fair Finance, several newspapers moved to unseal search-warrant documents related to the investigation of a suspected Ponzi scheme. 692 F.3d at 427–28. We rejected the claim that the documents were subject to a First Amendment right of access. Public access to those documents, we said, might, among other things, “compromise[]” the “safety of confidential witnesses.” Id. at 432. And in Morning Song, we recognized that “public access would pose a disincentive to raising [PSR] objections, which would, in turn, undermine the goal of gathering accurate and complete information to aid the sentencing court.” 831 F.3d at 777. So too here. Public access to cooperator agreements would not only put cooperators in harm’s way, but would surely disincentivize future cooperation, in turn undermining the government’s ability to gather accurate information about criminal activity. See Roth et al., supra, at 1385–86; Ian Weinstein, Regulating the Market for Snitches, 47 Buffalo L. Rev. 563, 583 (1999) (noting that “fear of physical retaliation” is a “common disincentive” against cooperation). In light of our precedent, then, we cannot say with any confidence that public access to cooperation agreements would “play[] a significant positive role in the functioning” of government processes. Press-Enter. II, 478 U.S. at 8.
C.
Swain presents several cases that he says call for a different result. We disagree. First, he points to DeJournett, which held that there is a qualified First Amendment right of access to plea agreements. 817 F.3d at 485. Swain contends that DeJournett controls this case. But DeJournett took pains to note that the plea agreement at issue there did not contain a “cooperation or substantial assistance clause.” Id. at 481. So we had no occasion to consider Nos. 23-5821 United States v. Kincaide, et al. Page 12
whether cooperation agreements that might be related to, or incorporated into, plea agreements are also subject to a First Amendment right of access. Thus, DeJournett does not resolve the issue before us.
Second, there is good reason to avoid reading DeJournett expansively. Although we purported to apply the “experience and logic” test, DeJournett lacked any analysis of whether there exists a longstanding national tradition of public access to plea agreements (and, of course, it did not consider the history of cooperation agreements either). See id. at 484–85. That lack of historical analysis is in direct tension with both Press-Enterprise II itself and our prior cases applying Press-Enterprise II. See Fair Fin., 692 F.3d at 430–31 (asking whether there exists a historical tradition of public access); Morning Song, 831 F.3d at 777 (same); Detroit Free Press, 303 F.3d at 700–03 (same); Miami Univ., 294 F.3d at 823 (same).
Third, a plea supplement (containing a cooperation agreement or a decoy)—while closely related to the plea agreement itself—is its own unique form of document. Plea supplements (and the cooperation agreement which the sealed supplement may contain) have their own unique history. And, as we discuss above, public access to cooperation agreements presents a unique set of issues implicating cooperator safety. Those issues are simply not raised by plea agreements in general and were not at issue in DeJournett. We reject Swain’s claim that DeJournett controls.
Swain next points to out-of-circuit authority, but it fails to persuade. In United States v. Bacon, the Tenth Circuit confronted a district court sealing policy virtually identical to the one here. 950 F.3d 1286, 1290 (10th Cir. 2020). The court held that the defendant was entitled to access his own plea supplement. Id. at 1292–95. But Bacon dealt only with the common-law right of access to judicial records; indeed, the defendant “disclaimed any reliance on a potential First Amendment right,” and the court made clear that it had “no occasion to address whether such a constitutional right exists.” Id. at 1293 n.3. Swain, conversely, asserts only a First Amendment right of access. So Bacon does not answer the question presented here.
By contrast, the Ninth Circuit has held that a plea agreement’s “cooperation addendum” is subject to the First Amendment right of access. In re Copley Press, 518 F.3d 1022, 1026, 1028 (9th Cir. 2008). But we are not persuaded by that authority. The court offered no rationale Nos. 23-5821 United States v. Kincaide, et al. Page 13
for its pronouncement beyond a citation to a prior case: Oregonian Publishing Co. v. U.S. District Court, 920 F.2d 1462 (9th Cir. 1990). That prior case, in turn, held only that “plea agreements and related documents” are subject to a First Amendment right of access. Id. at 1466. The “related documents” were characterized as “memoranda” relating to sealing motions and “the district court’s findings in support of its order sealing the documents.” Id. at 1464. Although Oregonian Publishing Co. noted in passing that the plea agreement “contemplated [the defendant’s] cooperation with the government,” there is no indication that the court considered that fact in its analysis of the First Amendment right. Id. at 1467. Thus, neither Oregonian Publishing Co. nor Copley Press provides any analysis of cooperation agreements or any reasoning as to why the First Amendment might provide a right of access to these documents. Moreover, a Ninth Circuit panel subsequently questioned the reasoning of Copley Press. See United States v. Doe, 870 F.3d 991, 997 (9th Cir. 2017). We are not bound to follow Copley Press’s unreasoned rule, and we decline to do so here.
***
Swain fails to show that either experience or logic favors a qualified First Amendment right of access. There is no widespread tradition of public access to cooperation agreements. Neither can we say with any certainty that public access would “play[] a significant positive role in the functioning of the particular process in question.” Press-Enter. II, 478 U.S. at 8.
We AFFIRM the denial of Swain’s motion to intervene.