Betschart v. Washington County Circuit Court Judges

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT WALTER BETSCHART, on their No. 23-2270 behalf, and on behalf of all others similarly situated; JOSHUA SHANE D.C. No. BARTLETT, on their behalf, and on 3:23-cv-01097-CL behalf of all others similarly situated; CALEB AIONA, on their behalf, and on behalf of all others OPINION similarly situated; TYRIK DAWKINS, on their behalf, and on behalf of all others similarly situated; JOSHUA JAMES- RICHARDS, on their behalf, and on behalf of all others similarly situated; TANIELA KINI KIN LATU, on their behalf, and on behalf of all others similarly situated; RICHARD OWENS, on their behalf, and on behalf of all others similarly situated; LEON MICHAEL POLASKI, on their behalf, and on behalf of all others similarly situated; ALEX SARAT XOTOY, on their behalf, and on behalf of all others similarly situated; TIMOTHY WILSON, on their behalf, and on behalf of all others similarly situated; JEFFERY 2 BETSCHART V. STATE OF OREGON DAVIS; RICHARD AARON CARROLL Sr.; JENNIFER LYN BRUNETTE; NICHOLAS WALDBILLIG; DEREK PIMENO ZAVALA; CURTIS RAY ANTHONY REMINGTON; CRISTA JEAN DAVIS; NICHOLE LYNN WHALEN; JACOB ISSAC NATHANIEL COLE, Petitioners - Appellees, v. STATE OF OREGON, Respondent - Appellant, WASHINGTON COUNTY CIRCUIT COURT JUDGES, in their official capacities; PATRICK GARRETT, Sheriff, Washington County Sheriff, in his official capacity, Respondents. WALTER BETSCHART; JOSHUA No. 23-3560 SHANE BARTLETT; CALEB AIONA; TYRIK D.C. No. DAWKINS; JOSHUA JAMES- 3:23-cv-01097-CL RICHARDS; TANIELA KINI KIN LATU; RICHARD OWENS; LEON  BETSCHART V. STATE OF OREGON 3 MICHAEL POLASKI; ALEX SARAT XOTOY; TIMOTHY WILSON, Petitioners - Appellants, v. STATE OF OREGON; WASHINGTON COUNTY CIRCUIT COURT JUDGES, Respondents - Appellees. Appeal from the United States District Court for the District of Oregon Michael J. McShane, District Judge, Presiding Argued and Submitted February 6, 2024 Pasadena, California Filed May 31, 2024 Before: John B. Owens, Patrick J. Bumatay, and Salvador Mendoza, Jr., Circuit Judges. Opinion by Judge Owens; Dissent by Judge Bumatay 4 BETSCHART V. STATE OF OREGON SUMMARY * Habeas Corpus In a case in which a class of incarcerated indigent criminal defendants awaiting trial in Oregon (Petitioners) filed a federal habeas corpus petition under 28 U.S.C. § 2241, the panel affirmed the district court’s preliminary injunction requiring that counsel be provided within seven days of the initial appearance, and failing this, Petitioners must be released from custody subject to reasonable conditions imposed by Oregon Circuit Court judges. Addressing whether a federal court should wade into these state court criminal proceedings, the panel wrote that it could not abstain, even assuming all four factors set forth in Younger v. Harris, 401 U.S. 37 (1971), are met, because the unthinkable situation for Oregon’s defendants—those who are incarcerated, awaiting trial, and without counsel in direct violation of the watershed command of Gideon v. Wainwright, 372 U.S. 335 (1963)—is an extraordinary circumstance that requires federal action. The panel held that the district court did not abuse its discretion in concluding that Petitioners were likely to succeed on the merits of their Sixth Amendment claim because, without counsel, Petitioners could not understand, prepare for, or progress to critical stages. Although it did not need to definitively resolve the question, the panel wrote that it was not an abuse of the discretion for the district court to conclude, alternatively, that bail hearings are critical stages * This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.  BETSCHART V. STATE OF OREGON 5 that trigger the Sixth Amendment’s counsel requirement. The panel held that the district court did not abuse its discretion in concluding that Petitioners are suffering and will continue to suffer irreparable harm, and that the district court was within its discretion to find that the public has an interest in a functioning criminal justice system and the protection of fundamental rights. In a concurrently filed memorandum disposition in No. 23-3560, the panel rejected Petitioners’ cross-appeal from the denial of a preliminary injunction as to a proposed class encompassing indigent criminal defendants not incarcerated but subject to liberty constraints as a condition of their supervised release. In a concurrently filed order in No. 23- 3573, the panel denied permission to appeal the denial of class certification of that class. Dissenting, Judge Bumatay wrote that the jailbreaking solution crafted by the district court and endorsed by the majority is not a legally permissible response. He focused on five errors: (1) this court lacks subject-matter jurisdiction; (2) the district court order violates the Younger abstention doctrine; (3) on the merits, the district court and majority’s Sixth Amendment analysis is disconnected from precedent; (4) the Fourteenth Amendment’s Due Process Clause doesn’t justify the injunction; and (5) the district court failed to properly balance the interests of the public and the parties in crafting the injunction. 6 BETSCHART V. STATE OF OREGON COUNSEL Julie P. Vandiver (argued), Jessica G. Synder, Peyton E. Lee, and Robert B. Hamilton, Assistant Federal Public Defenders; Stephen R. Sady, Chief Deputy Federal Public Defender; Fidel Cassino-DuCloux, Federal Public Defender; Federal Public Defender’s Office, Portland, Oregon; David F. Sugerman and Nadia H. Dahab, Sugerman Dahab, Portland, Oregon; for Petitioners-Appellees/ Appellants. Michael A. Casper (argued), Senior Assistant Attorney General; Benjamin Gutman, Solicitor General; Ellen F. Rosenblum, Oregon Attorney General; Office of the Oregon Attorney General, Salem, Oregon; James Aaron, Assistant Attorney General, Oregon Department of Justice, Salem, Oregon; for Respondent-Appellant/ Appellee. Kenneth A. Kreuscher and Kassidy N. Hetland, Oregon Innocence Project, Portland, Oregon, for Amicus Curiae Oregon Innocence Project. Trisha Trigilio and Emma Andersson, American Civil Liberties Union Foundation, Criminal Law Reform Project, New York, New York; Jason D. Williamson, Center on Race, Inequality, and the Law, New York University School of Law, New York, New York; Rosalind M. Lee, OCDLA Amicus Committee, Rosalind Manson Lee LLC, Eugene, Oregon; Kristin Asai, Holland & Knight LLP, Portland, Oregon; Kelly Simon, American Civil Liberties Union of Oregon, Portland, Oregon; Gia L. Cincone, NACDL Amicus Committee, San Francisco, California; Athul K. Acharya, Public Accountability, Portland, Oregon; for Amici Curiae Civil Rights Litigators.  BETSCHART V. STATE OF OREGON 7 OPINION OWENS, Circuit Judge: The state arrests a citizen and incarcerates him pending trial. Days, weeks, and months pass without any legal representation. He seeks relief from the authorities—surely a lawyer should help him? In response, he gets a shoulder shrug, a promise that they are “working on it,” and nothing more. He remains in jail, without legal counsel or any relief in sight. You might think this passage comes from a 1970s State Department Report on some autocratic regime in the Soviet Bloc. Unfortunately, we do not need to go back in time or across an ocean to witness this Kafkaesque scene. This is the State of Oregon in 2024. The Supreme Court outlawed this practice more than sixty years ago, in Gideon v. Wainwright, 372 U.S. 335 (1963), which held that the Sixth and Fourteenth Amendments guaranteed trial counsel for indigent criminal defendants. The Court explained: “lawyers in criminal courts are necessities, not luxuries. The right of one charged with [a] crime to counsel may not be deemed fundamental and essential to fair trials in some countries, but it is in ours.” Id. at 344. The Sixth Amendment right to counsel, as outlined in Gideon, is the only “watershed” right that the Supreme Court has recognized in the habeas context. See, e.g., Edwards v. Vannoy, 593 U.S. 255, 267 (2021). Yet, due to an “ongoing public defense crisis” of its own creation, Oregon does not provide indigent criminal defendants their fundamental right to counsel despite Gideon’s clear command. For several reasons, there are not 8 BETSCHART V. STATE OF OREGON enough qualified attorneys in Oregon to represent criminal defendants, some of whom remain detained without counsel. Even worse, Oregon cannot proceed in prosecuting these defendants “unless and until an attorney is appointed to represent” them. Accordingly, an innocent person may languish in jail for months awaiting trial, simply because no lawyer has been provided to review or investigate his case. Those that manage to appear before a judge can count on little help and scant information. When one Petitioner asked the judge at a pretrial hearing when he would be appointed counsel, the judge simply responded, “I don’t know.” When the Petitioner said that continuing without a lawyer was unconstitutional, the judge responded that the Petitioner “won’t get a disagreement from me or from the prosecutor that you should have a lawyer. It is an unfortunate circumstance that we are in with the state.” The hearing then proceeded, with the Petitioner left without counsel. This is no anomaly—the record contains many similar stories, including a Petitioner who remained in jail without counsel for nearly a year. A class of incarcerated indigent criminal defendants awaiting trial in Oregon challenged this untenable situation via habeas corpus in federal court. Rather than avoid a “judicial jailbreak” by making counsel available to defendants as the Constitution requires, Oregon insisted on fighting the solution. After extensive litigation, the district court issued a preliminary injunction requiring, among other things, that “counsel . . . be provided within seven days of the initial appearance,” and “[f]ailing this, defendants must be released from custody, subject to reasonable conditions imposed by [Oregon] Circuit Court judges.” Considering the extraordinary (and extremely prejudicial) circumstances facing criminal defendants in Oregon in direct violation of  BETSCHART V. STATE OF OREGON 9 Gideon, we cannot say that the district court abused its discretion in issuing this preliminary injunction, and so we affirm. I. FACTUAL AND PROCEDURAL BACKGROUND A. Oregon’s Public Defense Crisis To understand how this Sixth Amendment nightmare became a reality, we review how Oregon provides counsel to indigent defendants awaiting trial. Rather than employ state or county public defenders, Oregon contracts with individual private attorneys for these services. Until January 2024, the Public Defense Services Commission (“PDSC”) oversaw Oregon’s public defense system. The PDSC made a bad situation worse when, in 2021 and 2022, it altered the rules governing compensation and caseloads for these private attorneys. These changes rendered public defense work financially untenable, and many private attorneys stopped taking criminal defense cases. While individuals continued to be arrested and charged with crimes, there were no longer enough lawyers to represent them. 1 Between March and June 2023 alone, the number of unrepresented criminal defendants increased by 198 percent. By September 2023, that number had grown another 48 percent, with almost 3,000 people awaiting their Gideon-guaranteed 1 This explanation reflects Oregon’s public defense system at the outset of this case. Oregon is reforming this system through state legislation. For example, effective January 2024, the PDSC was abolished and replaced with a new agency in the state government. Despite these early reforms, the crisis persists. Many of Oregon’s planned reforms will not become effective until the late 2020s and into the 2030s. 10 BETSCHART V. STATE OF OREGON counsel. More than 100 of these defendants remain incarcerated pretrial. 2 B. The Litigation In July 2023, ten indigent defendants in custody awaiting trial without representation in Washington County filed a joint habeas corpus petition in federal district court, seeking class status and alleging violations under the Sixth, Eighth, and Fourteenth Amendments. They moved for a Temporary Restraining Order (“TRO”) requiring release if they were not appointed counsel within seven days. The district court provisionally certified “individuals in physical custody in Washington County Detention Center” as the “Custody Class” and entered a TRO. Under the TRO, if class members were not provided representation within ten days of their initial appearance, or within ten days of their previous counsel’s withdrawal, Oregon would have to release them. The State of Oregon subsequently intervened as Respondent. Petitioners filed their Second Amended Petition for habeas corpus, adding unrepresented defendants across the state, both in jail and out of jail on restrictive 2 The State of Washington is facing similar problems and consequences. See Daniel Beekman, WA’s public defender system is breaking down, communities reeling, Seattle Times (Feb. 25, 2024, 7:00 AM), https://www.seattletimes.com/seattle-news/politics/was-public- defender-system-is-breaking-down-communities-reeling/ (“Staffing shortages and burnout-inducing caseloads are squeezing urban . . . [and] rural areas . . . . In some instances, people presumed innocent are languishing in jail without counsel.”); see also Colin Rigley, Confronting a Crisis, Washington State Bar News (Feb. 8, 2024), https://wabarnews.org/2024/02/08/confronting-a-crisis/.  BETSCHART V. STATE OF OREGON 11 conditions. 3 Petitioners requested that the district court convert the TRO into a preliminary injunction for the Custody Class and reduce the time in which counsel must be appointed from ten days to forty-eight hours. C. The Injunction After briefing and extensive argument, the district court granted the motion for a preliminary injunction. It declined to abstain under Younger v. Harris, 401 U.S. 37 (1971), holding that “this remains a case of ‘extraordinary circumstances’ that demands federal intervention.” The court rejected Oregon’s argument that “Petitioners can challenge their right to counsel after trial without risk[ing] irreparable harm,” reasoning that “the Sixth Amendment entitles the accused to adequate representation at all critical stages of trial.” The court also applied the logic of Page v. King, 932 F.3d 898 (9th Cir. 2019), which held that a “complete loss of liberty for the time of pretrial detention is ‘irretrievable’ regardless of the outcome at trial.” Id. at 904. The court reaffirmed its provisional class certification of the Custody Class and expanded it statewide. The court then found that Petitioners were likely to succeed on the merits of 3 Petitioners then motioned for certification of and a preliminary injunction for a “Restrictive Conditions Class,” a second proposed class encompassing indigent criminal defendants not incarcerated but subject to liberty constraints as a condition of their release. The district court denied certification and a preliminary injunction as to this proposed Restrictive Conditions Class, abstaining under Younger v. Harris, 401 U.S. 37 (1971), and rejecting Petitioners’ claims on the merits. Petitioners cross-appealed the denial of the preliminary injunction and, in the related case Betschart v. Garrett, No. 23-3573, appealed the denial of class certification. A concurrently filed memorandum disposition rejects the cross-appeal in this case, in which we abstain under Younger. A concurrently filed order in No. 23-3573 denies permission to appeal the denial of class certification of the Restrictive Conditions Class. 12 BETSCHART V. STATE OF OREGON their Sixth Amendment and due process claims and subsequently “order[ed] that counsel must be provided within seven days of the initial appearance, or within seven days of the withdrawal [of] previously appointed counsel,” and “[f]ailing this, defendants must be released from custody, subject to reasonable conditions imposed by [Oregon] Circuit Court judges.” 4 Oregon timely appealed from the preliminary injunction. A motions panel stayed the preliminary injunction pending appeal and expedited the appeal and cross-appeal. II. JURISDICTION We have jurisdiction under 28 U.S.C. § 1292(a)(1). The district court had jurisdiction under 28 U.S.C. § 2241. See Rodriguez v. Hayes, 591 F.3d 1105, 1117 (9th Cir. 2010) (recognizing that class actions may be brought “pursuant to habeas corpus”), abrogated on other grounds by Rodriguez Diaz v. Garland, 53 F.4th 1189, 1199-1201 (9th Cir. 2022); Cox v. McCarthy, 829 F.2d 800, 804 (9th Cir. 1987) (“[A] class action may lie in habeas corpus.”); Mead v. Parker, 464 F.2d 1108, 1112-13 (9th Cir. 1972) (same). With great bluster but without any legal citations, the dissent contends that we lack jurisdiction because Sixth Amendment violations supposedly do not merit release from custody. Not even the State of Oregon made this argument at the district court or on appeal. And that is because that argument ignores the basic history of Gideon (and many 4 Relying in part on Article I, Section 43 of the Oregon Constitution, the district court later amended the preliminary injunction to exclude “class members who fire their attorneys,” those charged with “murder and aggravated murder,” and those who are released under the order but have their release revoked.  BETSCHART V. STATE OF OREGON 13 other cases), where Sixth Amendment violations have led directly to defendants being released from prison. For example, the State of Florida released, or released and retried, over 4,000 prisoners after the Supreme Court issued its decision in Gideon. 5 The dissent concedes that the public defense crisis in Oregon has resulted in “a delay, and sometimes a lengthy delay,” in proceedings after a defendant is detained. This delay is enough to bring Petitioners’ suits within the “core of habeas corpus” as required by Nettles v. Grounds, 830 F.3d 922, 934 (9th Cir. 2016) (citing Preiser v. Rodriguez, 411 U.S. 475, 487 (1973)). See Preiser, 411 U.S. at 487-88 (holding that even if the state prisoners’ requested relief had “merely shortened the length of their confinement . . . their suits would still have been within the core of habeas corpus in attacking the very duration of their physical confinement itself”). The dissent next claims that the district court’s jurisdiction is “irregular[]” because it did not order release from custody for Petitioners charged with murder and aggravated murder. That the district court would have habeas jurisdiction to hear a case only if it ultimately ordered release from custody is an odd argument. District courts routinely deny release from custody under habeas jurisdiction. The district court’s decision, which was consistent with state law as to those Petitioners and mitigates 5 See Bruce R. Jacob, Memories of and Reflections about Gideon v. Wainwright, 33 Stetson L. Rev. 181, 222 (2003); M. Alex Johnson and Vidya Rao, A ‘nobody’s’ legacy: How a semi-literate ex-con changed the legal system, NBC News (Mar. 18, 2013, 2:40 AM), https://www.nbcnews.com/news/us-news/nobodys-legacy-how-semi- literate-ex-con-changed-legal-system-flna1C8914521. 14 BETSCHART V. STATE OF OREGON the “jailbreak” the dissent so ardently fears, was within its discretion. The dissent also challenges class certification and whether class actions can lie in habeas, citing dicta from a footnote in a concurrence to a case on a completely different question. See Jennings v. Rodriguez, 583 U.S. 281, 324 n.7 (2018) (Thomas, J., concurring). Oregon does not challenge class certification on appeal, and we decline to do so for them. See United States v. Sineneng-Smith, 490 U.S. 371, 375-76 (2020) (“[A]s a general rule, our system ‘is designed around the premise that . . . [the parties] are responsible for advancing the facts and argument entitling them to relief.’” (quoting Castro v. United States, 540 U.S. 375, 386 (2003) (Scalia, J., concurring in part and concurring in the judgment))). And in any case, the dissent acknowledges but completely disregards our binding precedent, which establishes that a class action can lie in habeas. See Hayes, 591 F.3d at 1117 (“[C]lass actions may be brought pursuant to habeas corpus.”). It instead suggests that because the Supreme Court has reversed our immigration-detention class-action cases on different grounds, our precedent “may be an outlier.” This assertion does not reflect an understanding of precedent. The Supreme Court overturned our immigration-detention class-action cases because of the special discretion the Immigration and Nationality Act gives the government, see Jennings, 583 U.S. at 303, which is irrelevant in this context, see Marin, 90 F.4th at 1240. 6 6 The dissent, relying on Jennings, also says that “the Supreme Court has instructed our court to ‘consider’ whether a Rule 23 class action is an  BETSCHART V. STATE OF OREGON 15 Thus, we are satisfied that our long-standing law remains valid, and that we have jurisdiction. See Miller v. Gammie, 335 F.3d 889, 899 (9th Cir. 2003) (en banc) (“[A] three- judge panel may not overrule a prior decision of the court.”). 7 III. DISCUSSION A. Standards of Review We review de novo the district court’s application of the Younger abstention doctrine and must “conduct the Younger “appropriate vehicle” for providing habeas relief in light of Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011).” That is simply not accurate. The Court in Jennings said that this court should consider on remand whether a Rule 23(b)(3) class action, specifically, “continues to be the appropriate vehicle for respondents’ claims” after Dukes, because there, the Court held “that ‘Rule 23(b)(2) applies only when a single injunction or declaratory judgment would provide relief to each member of the class’” and “[t]hat holding may be relevant on remand because the Court of Appeals has already acknowledged that some members of the certified class may not be entitled to bond hearings as a constitutional matter.” Jennings, 583 U.S. at 313 (quoting Dukes, 564 U.S. at 360). The Dukes holding is not relevant here, and we fail to see how that quote from Jennings could logically be construed as an instruction to this court to consider our precedent on an entirely different question. Additionally, the dissent brings up the Solicitor General’s recent comments before the Supreme Court on class actions in the Eighth Amendment context. As with the dissent’s cite to a footnote in a concurrence in an unrelated case, we strain to see how this is relevant. 7 The dissent also calls into question whether “habeas relief can be granted prospectively to individuals who are not yet even in custody.” But “[t]he inclusion of future class members in a class is not itself unusual or objectionable,” because “[w]hen the future persons referenced become members of the class, their claims will necessarily be ripe.” A.B. v. Hawaii State Dep’t of Educ., 30 F.4th 828, 838 (9th Cir. 2022) (quoting Hayes, 591 F.3d at 1118). 16 BETSCHART V. STATE OF OREGON analysis ‘in light of the facts and circumstances existing at the time the federal action was filed.’” Duke v. Gastelo, 64 F.4th 1088, 1093 (9th Cir. 2023) (citation omitted). We “review the district court’s decision to grant or deny a preliminary injunction for abuse of discretion.” Hernandez v. Sessions, 872 F.3d 976, 987 (9th Cir. 2017) (citation omitted). “The district court abuses its discretion when it makes an error of law.” Id. The district court’s legal conclusions are reviewed de novo, and its factual findings for clear error. Id. The abuse of discretion standard is “highly deferential to the district court.” Microsoft Corp. v. Motorola, Inc., 696 F.3d 872, 881 (9th Cir. 2012). B. Younger Abstention We first address whether a federal court should wade into these state court criminal proceedings. “[A] federal court’s ‘obligation’ to hear and decide a case is ‘virtually unflagging.’” Sprint Commc’ns, Inc. v. Jacobs, 571 U.S. 69, 77 (2013) (quoting Colo. River Water Conservation Dist. v. United States, 424 U.S. 800, 817 (1976)). However, federal courts must exercise caution when the relief sought impacts state court criminal proceedings. Younger abstention, “an extraordinary and narrow exception to [this] general rule” of hearing cases, reflects this concern. Cook v. Harding, 879 F.3d 1035, 1038 (9th Cir. 2018) (citation omitted). Under Younger, federal abstention is warranted when “(1) there is an ongoing state judicial proceeding; (2) the proceeding implicates important state interests; (3) there is an adequate opportunity in the state proceedings to raise constitutional challenges; and (4) the requested relief seeks to enjoin or has the practical effect of enjoining the ongoing state judicial proceeding.” Page, 932 F.3d at 901-02 (quoting Arevalo v. Hennessy, 882 F.3d 763, 765 (9th Cir.  BETSCHART V. STATE OF OREGON 17 2018)). Even when all four Younger factors are met, abstention is nevertheless inappropriate “where there exist other ‘extraordinary circumstances in which the necessary irreparable injury can be shown even in the absence of the usual prerequisites of bad faith and harassment.’” Kugler v. Helfant, 421 U.S. 117, 124 (1975) (quoting Younger, 401 U.S. at 53). “It is well established that the deprivation of constitutional rights ‘unquestionably constitutes irreparable injury.’” Arevalo, 882 F.3d at 766 (quoting Hernandez, 872 F.3d at 994). Even assuming all four Younger factors are met here, we cannot abstain. As the district court explained, the unthinkable situation for Oregon’s defendants—those who are incarcerated, awaiting trial, and without counsel in direct violation of Gideon’s watershed command—is an extraordinary circumstance that requires federal action. The situation here mirrors that in Page, in which we ruled that Page, who had been wrongfully civilly committed, had suffered a “complete loss of liberty” during his pretrial detention that was “‘irretrievable’ regardless of the outcome at trial.” 932 F.3d at 904. We reasoned that “a post-trial adjudication of his claim [would] not fully vindicate his right to a current and proper pretrial probable cause determination.” Id. As a result, we concluded that Page’s claim “fit[] squarely within the irreparable harm exception.” Id. (quoting Arevalo, 882 F.3d at 766). This case is also like Arevalo, in which the defendant was arrested following a domestic dispute. 882 F.3d at 764. A few days after his arrest, the state trial court set his bail at $1.5 million. Id. He filed a motion for a bail hearing, contending that the bail amount was excessive. Id. The trial court, without discussing Arevalo’s “ability to pay or what government interests the bail amount would serve,” lowered 18 BETSCHART V. STATE OF OREGON bail to $1 million, an amount Arevalo still could not afford. Id. at 764-65. Arevalo filed a habeas petition, and the district court abstained under Younger. Id. at 765. We reversed, holding that Younger abstention was not appropriate in part because Arevalo was irreparably harmed when he was incarcerated “without a constitutionally adequate bail hearing.” Id. at 767. Here, Petitioners suffer irreparable injury for the duration of their unlawful pretrial detention. See id. (“Deprivation of physical liberty by detention constitutes irreparable harm.”). Oregon does not dispute that its failure to provide counsel lengthens Petitioners’ pretrial detention. Its violation of Petitioners’ core Sixth Amendment rights undoubtedly impacts their ability to mount a vigorous defense. See Gideon, 372 U.S. at 343 (“The Sixth Amendment stands as a constant admonition that if the constitutional safeguards it provides be lost, justice will not ‘still be done.’” (citation omitted)). Oregon contends that the district court’s reliance on Page is misplaced, because in Page “the challenged procedure . . . was ‘distinct from the underlying criminal prosecution,’ and because the relief . . . could be granted without an ongoing intrusion into the state court proceedings.” Oregon conflates the fourth Younger factor and the extraordinary circumstances exception. See Bean v. Matteucci, 986 F.3d 1128, 1133 (9th Cir. 2021) (“[E]ven where the Younger factors are satisfied, ‘federal courts do not invoke it if there is . . . “some [] extraordinary circumstance that would make abstention inappropriate.”’” (quoting Arevalo, 882 F.3d at 765-66)). In both Page and Arevalo, the extraordinary circumstances exception constituted an independent basis for federal intervention, regardless of whether the Younger factors were met. See  BETSCHART V. STATE OF OREGON 19 Page, 932 F.3d at 904 (citing Arevalo, 882 F.3d at 767 n.3) (rejecting the state’s argument that Page’s failure to meet the third Younger factor categorically barred the irreparable harm exception and required abstention); Arevalo, 882 F.3d at 766 (“Younger abstention doctrine also does not apply because this case fits squarely within the irreparable harm exception.” (emphasis added)). Indeed, if meeting the preliminary four-factor test precluded application of the exception, there would be no exception at all. Citing no authority, the dissent contends that the extraordinary circumstances exception does not apply here because “the right’s vindication can come after trial through vacatur of the conviction.” The dissent does not explain how indefinite pretrial detention while a defendant waits for counsel can be repaired after trial. See Bean, 986 F.3d at 1134 (“[P]retrial rights, like those protecting unlawful pretrial detention, ‘cannot be vindicated post-trial.’” (quoting Page, 932 F.3d at 905)); Lopez-Valenzuela v. Arpaio, 770 F.3d 772, 781 (9th Cir. 2014) (“[T]he costs to the arrestee of pretrial detention are profound. ‘Pretrial confinement may imperil the suspect’s job, interrupt his source of income, and impair his family relationships.’” (quoting Gerstein v. Pugh, 420 U.S. 103, 114 (1975))). Again citing no authority, the dissent claims that “the harm the Sixth Amendment protects against is a conviction obtained through uncounseled critical stages” and that “[t]here’s no independent Sixth Amendment protection against being held in pretrial custody without counsel.” In other words, the dissent apparently believes there is no Sixth Amendment protection for those jailed by the state before conviction, when they are presumed innocent, and that Sixth Amendment protection only kicks in after they have been proven guilty beyond a reasonable doubt. This cannot be 20 BETSCHART V. STATE OF OREGON correct. The Sixth Amendment’s protection applies to “all criminal prosecutions.” U.S. Const. Amend. VI. The dissent would edit the Sixth Amendment from “prosecutions” to “prosecutions that result in convictions.” This view also ignores all of the caselaw, discussed infra pp. 23-27, holding that the Sixth Amendment provides essential protection for defendants awaiting trial. Because we conclude that Petitioners suffer irreparable injury and thus extraordinary circumstances exist here, we do not abstain under Younger. C. Preliminary Injunction “[T]o obtain a preliminary injunction a plaintiff must establish (1) ‘that he is likely to succeed on the merits,’ (2) ‘that he is likely to suffer irreparable harm in the absence of preliminary relief,’ (3) ‘that the balance of equities tips in his favor,’ and (4) ‘that an injunction is in the public interest.’” Hernandez, 872 F.3d at 989-90 (quoting Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008)). “Under our ‘sliding scale’ approach, ‘the elements of the preliminary injunction test are balanced, so that a stronger showing of one element may offset a weaker showing of another.’” Id. at 990 (quoting Pimentel v. Dreyfus, 670 F.3d 1096, 1105 (9th Cir. 2012) (per curiam)). The district court determined that the relief Petitioners sought was a mandatory injunction, because it “order[ed] a responsible party to take action” going “well beyond simply maintaining the status quo.” Marlyn Nutraceuticals, Inc. v. Mucos Pharma GmbH & Co., 571 F.3d 873, 878-79 (9th Cir. 2009). To obtain a mandatory injunction, a plaintiff must “establish that the law and facts clearly favor [their] position, not simply that [they are] likely to succeed.” Garcia v. Google, Inc., 786 F.3d 733, 740 (9th Cir. 2015) (en banc).  BETSCHART V. STATE OF OREGON 21 But see Hernandez, 872 F.3d at 998 (questioning whether the distinction between mandatory and prohibitory injunctions is meaningful). “Mandatory injunctions are most likely to be appropriate when ‘the status quo . . . is exactly what will inflict the irreparable injury upon [the] complainant.’” Hernandez, 872 F.3d at 999. “Because our review is deferential, ‘[w]e will not reverse the district court where it “got the law right,” even if we “would have arrived at a different result,” so long as the district court did not clearly err in its factual determinations.’” Garcia, 786 F.3d at 739 (quoting All. for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1131 (9th Cir. 2011)) (alteration in original). 1. Likelihood of Success on the Merits The district court concluded that Petitioners were likely to succeed on the merits of their Sixth Amendment claim, and that the law and facts clearly favored their position, because (a) the lack of counsel prevented them from preparing for or progressing to critical stages and (b) bail hearings, to which Oregon custodial defendants are entitled within a certain time frame, are critical stages. We hold that the district court did not abuse its discretion in reaching either conclusion. 8 We address each in turn. a) Preparation for and Progression to Critical Stages The district court concluded that Petitioners were likely to succeed on the merits of their Sixth Amendment claim 8 The district court also concluded that Petitioners were likely to succeed on the merits of their Fourteenth Amendment due process claim. Because we decide that Petitioners are likely to succeed on the merits of their Sixth Amendment claim, we do not reach their Fourteenth Amendment claim. 22 BETSCHART V. STATE OF OREGON because, without counsel, Petitioners could not understand, prepare for, or progress to critical stages. Indigent defendants have a fundamental right, guaranteed by the Sixth and Fourteenth Amendments, to “the aid of counsel in a criminal prosecution.” Gideon, 372 U.S. at 343 (citation omitted). The right attaches “at or after the time that judicial proceedings have been initiated against [the defendant] ‘whether by way of formal charge, preliminary hearing, indictment, information, or arraignment.’” Brewer v. Williams, 430 U.S. 387, 398 (1977) (quoting Kirby v. Illinois, 406 U.S. 682, 689 (1972)). Once the right attaches, the defendant is guaranteed counsel “during any ‘critical stage’ of the postattachment proceedings.” Rothgery v. Gillespie County, 554 U.S. 191, 212 (2008). “The Constitution’s guarantee of assistance of counsel cannot be satisfied by mere formal appointment.” United States v. Cronic, 466 U.S. 648, 654-55 (1984) (citation omitted). “[C]ounsel must be appointed within a reasonable time after attachment to allow for adequate representation at any critical stage before trial, as well as at trial itself.” Rothgery, 554 U.S. at 212. The district court’s conclusion that Petitioners were likely to succeed on the merits was not an abuse of discretion. There is a high likelihood that the failure to appoint counsel in Petitioners’ cases impairs their Sixth Amendment right to counsel. Lack of counsel not only interferes with indigent criminal defendants’ progression to critical stages by delaying those stages but also prevents any meaningful advocacy. The Sixth Amendment requires not just that counsel show up on the day of a critical stage but prepare for it too. See id.; McMann v. Richardson, 397 U.S. 759, 771 n.14 (1970) (“It has long been recognized that the  BETSCHART V. STATE OF OREGON 23 right to counsel is the right to the effective assistance of counsel.”). The right to counsel encompasses myriad attorney duties beyond mere presence at certain pretrial hearings. It is a continuous right to competent and zealous advocacy outside of the courtroom. It includes: • counsel’s investigation of lines of defense; 9 • counsel’s “available advice about an issue like deportation;” 10 • counsel’s ensuring that the defendant is competent to stand trial; 11 9 Strickland v. Washington, 466 U.S. 668, 690-91 (1984) (defining counsel’s general duty to investigate lines of defense); see also Hart v. Gomez, 174 F.3d 1067, 1071 (9th Cir. 1999) (“Hart’s counsel ‘failed to fulfill his duty to investigate [Hart’s] most important defense.’” (alteration in original) (quoting Sanders v. Ratelle, 21 F.3d 1446, 1457 (9th Cir. 1994))); Thomas v. Lockhart, 738 F.2d 304, 308 (8th Cir. 1984) (determining counsel had a duty to interview key witnesses and investigate defendant’s mental problems). 10 Padilla v. Kentucky, 559 U.S. 356, 371 (2010) (“It is quintessentially the duty of counsel to provide her client with available advice about an issue like deportation and the failure to do so ‘clearly satisfies the first prong of the Strickland analysis.’” (quoting Hill v. Lockhart, 474 U.S. 52, 62 (1985))). 11 See Douglas v. Woodford, 316 F.3d 1079, 1085 (9th Cir. 2003) (“[C]ounsel has a duty to investigate a defendant’s mental state if there is evidence to suggest that the defendant is impaired.”). 24 BETSCHART V. STATE OF OREGON • confidentiality in communication with counsel; 12 • counsel’s communication of formal plea offers; 13 • counsel’s warning of possible risks in sentencing; 14 • counsel’s assistance with a defendant’s attempt to cooperate; 15 • guidance through the plea-bargaining process, including counsel’s competent 12 See Nordstrom v. Ryan, 762 F.3d 903, 906 (9th Cir. 2014) (holding that a prison guard reading a prisoner’s letter to his lawyer violates the prisoner’s Sixth Amendment right to confide in his lawyer); Mangiaracina v. Penzone, 849 F.3d 1191, 1198 (9th Cir. 2017) (holding that a prison guard opening a prisoner’s letter from his lawyer outside the prisoner’s presence is “sufficient to state a claim for violation of [the prisoner’s] Sixth Amendment right to counsel”). 13 Missouri v. Frye, 566 U.S. 134, 145 (2012). 14 Risher v. United States, 992 F.2d 982, 984 (9th Cir. 1993) (holding that counsel “cannot be said to have been functioning as counsel within the meaning of the Sixth Amendment” when they did not warn defendant of “significant risk” he would be sentenced as a career offender). 15 United States v. Leonti, 326 F.3d 1111, 1122 (9th Cir. 2003) (“The Sixth Amendment guarantee of competent counsel applies to the process of cooperation with the government . . . .”).  BETSCHART V. STATE OF OREGON 25 advice on how to plead 16 and the right to appeal; 17 • other rights, including counsel “keep[ing] abreast of Supreme Court decisions affecting their clients’ interests.” 18 The dissent ignores these bedrock Sixth Amendment cases and dismisses every Sixth Amendment violation that occurs prior to a jury verdict as “collateral.” That is incorrect. The Sixth Amendment is not a haphazard jack-in- the-box that occasionally appears when cranked. As the Supreme Court made clear when rejecting a similar argument, it is an ongoing right that persists throughout trial court proceedings. See Lafler, 566 U.S. at 164-65 (“[T]he Solicitor General claim[s] that the sole purpose of the Sixth Amendment is to protect the right to a fair trial. Errors before trial, they argue, are not cognizable under the Sixth Amendment unless they affect the fairness of the trial itself. The Sixth Amendment . . . is not so narrow in its reach.” (citations omitted)); see also Padilla, 559 U.S. at 373 (detailing counsel’s duty to provide competent immigration advice to defendants during plea bargaining). Leonti, a case the dissent itself cites, also undercuts its argument. Leonti held that the right to counsel extended to the entire period of time in which a defendant could “attempt[] to render substantial assistance to the government” to lower his 16 Lafler v. Cooper, 566 U.S. 156, 162-63 (2012). 17 See Marrow v. United States, 772 F.2d 525, 529 (9th Cir. 1985) (“[C]ounsel has a duty to advise his client of the right to appeal the conviction.”). 18 See United States v. Loughery, 908 F.2d 1014, 1018 (D.C. Cir. 1990). 26 BETSCHART V. STATE OF OREGON sentence—after pleading guilty and thus, after the merits of his case were already decided. 326 F.3d at 1122. 19 Oregon and the dissent suggest that counsel’s pretrial duty to appear at critical stages encompasses only presence and not preparation. This view is fundamentally incompatible with the decades of precedent defining what counsel must do to provide criminal defendants their Sixth Amendment right to counsel. See, e.g., De Roche v. United States, 337 F.2d 606, 607 (9th Cir. 1964) (“It is of course true that the right conferred by the Sixth Amendment to effective assistance of counsel implicitly embraces adequate opportunity for the accused and his counsel to consult, advise and make such preparation for arraignment and trial as the facts of the case fairly demand.”). Even assuming that a bail hearing is not a critical stage, Oregon and the dissent fail to explain how an indigent criminal defendant could progress to critical stages without counsel or without being pressured into giving up their rights altogether. How, for example, would an indigent criminal defendant investigate their case from a prison cell? 20 Or 19 In any event, confinement pretrial does affect trial outcomes. See, e.g., Lopez-Valenzuela, 770 F.3d at 781 (“Pretrial confinement . . . may affect the defendant’s ability to assist in preparation of his defense” and “considerable evidence [shows] that pretrial custody status is associated with the ultimate outcomes of cases, with released defendants consistently faring better than defendants in detention.” (citations omitted)); Faheem-El v. Kilncar, 841 F.2d 712, 719 (7th Cir. 1988) (“[P]retrial detention lessens the defendant’s ability to assist in preparing his or her defense for trial.”). 20 See, e.g., Weeden v. Johnson, 854 F.3d 1063, 1070 (9th Cir. 2017) (holding that counsel’s failure to conduct psychological evaluation constituted ineffective assistance because counsel had a duty to investigate).  BETSCHART V. STATE OF OREGON 27 establish alibis? 21 Or interview witnesses? 22 Or review electronic discovery? How, without a formal legal education, would they know what rights they possess? Oregon and the dissent provide no answers because there are none. Lawyers are uniquely situated to carry out these tasks. See Gideon, 372 U.S. at 345 (“Even the intelligent and educated layman has small and sometimes no skill in the science of law . . . . He requires the guiding hand of counsel at every step in the proceedings against him.”). The bottom line is that the dissent would allow indefinite detention without counsel, as long as the accused has not yet been tried. Not even Oregon goes that far. The Sixth Amendment imposes responsibilities on counsel to ensure that indigent criminal defendants’ cases are not neglected, and defense strategy is formulated before counsel shows up in court, before making tactical decisions that could make all the difference. See Reynoso v. Giurbino, 462 F.3d 1099, 1112 (9th Cir. 2006) (“[C]ounsel cannot be said to have made a tactical decision without first procuring the information necessary to make such a decision.”). The discussions and interactions between a defendant and his attorney are integral to his defense. The “necessarily close working relationship between lawyer and client, the need for confidence, and the critical importance of trust” are all unfulfilled when a defendant has no lawyer at all. Luis v. United States, 578 U.S. 5, 11 (2016) (describing the 21 See, e.g., Bemore v. Chappell, 788 F.3d 1151, 1164 (9th Cir. 2015) (“[C]ounsel cannot neglect to investigate both the possible alibi and alternative defenses.”). 22 See, e.g., Baumann v. United States, 692 F.2d 565, 580 (9th Cir. 1982) (“We have clearly held that defense counsel’s failure to interview witnesses that the prosecution intends to call during trial may constitute ineffective assistance of counsel.”). 28 BETSCHART V. STATE OF OREGON attorney-client relationship in the context of the Sixth Amendment right to counsel of choice). The dissent contends that we are in “uncharted constitutional territory.” It is true that there is not a case that says, “an indefinite delay in counsel probably does not stand under the Sixth Amendment.” There is good reason for that: Our law assumes that the system is working the way that it should. Our law assumes that our state governments would pay to provide counsel to indigent defendants. Our law assumes that state governments would want to swiftly bring those proven guilty to justice, and to promptly release those who do not merit prosecution. It is Oregon’s uncharted refusal to adequately pay lawyers, not some new-fangled right, that forced the district court to make a tough call. Oregon and the dissent’s myopic view that the Sixth Amendment is a scattershot right—and not a consistent and ongoing one—ignores decades of controlling precedent and effectively erases the Sixth Amendment from the Constitution. The district court did not abuse its discretion in rejecting this radical take. 23 b) Bail Hearings The district court alternatively held that bail hearings are critical stages that also trigger the Sixth Amendment’s counsel requirement. The court reasoned that because Oregon law required bail hearings for all criminal defendants 23 The dissent tries to dismiss generational precedent by labeling it “New Deal.” There are at least two problems with this argument. First, the Court decided the landmark Powell v. Alabama, 287 U.S. 45 (1932), on President Herbert Hoover’s watch. President Franklin Delano Roosevelt took office in 1933. Second, and more importantly, Supreme Court precedent is precedent, even if it dates back to the 1930s and remains unpopular in certain quarters.  BETSCHART V. STATE OF OREGON 29 within a certain period, Petitioners were likely without counsel for such critical stages. While we need not definitively resolve this question here, it was not an abuse of discretion for the district court to have reached this conclusion. In determining what constitutes a critical stage, we consider three factors, any one of which “may be sufficient to render a stage of the proceedings ‘critical’”: whether “(1) ‘failure to pursue strategies or remedies results in a loss of significant rights,’ (2) ‘skilled counsel would be useful in helping the accused understand the legal confrontation,’ and (3) ‘the proceeding tests the merits of the accused’s case.’” Hovey v. Ayers, 458 F.3d 892, 901 (9th Cir. 2006) (citations omitted). The district court, quoting Coleman v. Alabama, 399 U.S. 1, 9 (1970), reasoned that the Supreme Court has specifically stated bail is one matter where “counsel can . . . be influential . . . in making effective arguments for the accused.” The district court also noted that the Second Circuit has drawn on that language from Coleman to hold that “[t]here is no question” that a bail hearing is a critical stage. Higazy v. Templeton, 505 F.3d 161, 172 (2d Cir. 2007). The dissent attempts to distinguish Higazy (ignoring the resulting circuit split) by arguing that the hearing there was an adversarial preliminary hearing, not just a hearing to set bail. 24 But Oregon bail hearings are also adversarial. The district court applied the Ninth Circuit test, quoted supra p. 29, and concluded that all three criteria were met, because in a bail hearing, “witnesses are called, evidence is 24 The dissent also discounts Higazy as “out-of-circuit,” while citing state court cases from Maryland and Alaska to cobble together its argument. 30 BETSCHART V. STATE OF OREGON presented, facts are mitigated, alternatives to incarceration are proposed, and the defendant can address the court.” The dissent, quoting United States v. Ash, 413 U.S. 300, 310 (1973), contends that “[b]ail hearings are not a critical stage because they are not ‘pretrial events that might appropriately be considered to be parts of the trial itself.’” The dissent’s reliance on Ash is misplaced. Ash did not stand for the proposition that there needs to be a merits determination at a pretrial proceeding to make it a critical stage; rather, the Supreme Court in that case was concerned with “whether the accused required aid in coping with legal problems or assistance in meeting his adversary.” Id. at 313. As to the first critical stage factor, the dissent, relying on Hovey, contends that the Ninth Circuit has “said that a proceeding is not a critical stage if there’s no ‘risk of permanent deprivation of any significant rights during the hearing.’” But we have said no such thing. See Hovey, 458 F.3d at 901-02 (holding that, because the defendant had not met any of the other factors—any one of which would have been sufficient—and did not meet the first factor because the defendant could raise questions of his attorney’s competency in the future, an attorney competency hearing was not a critical stage). The dissent next claims that Oregon law “doesn’t forbid a new bail determination once counsel is appointed.” This claim highlights the circularity of the dissent’s logic: Petitioners do not have counsel, they are bringing this suit because Oregon refuses to provide them counsel, yet the dissent crows that if they had counsel, there would not be a problem. Additionally, the statute the dissent is referencing—Or. Rev. Stat. § 123.245—only provides for modification of release agreements “[i]f circumstances concerning the defendant’s release change.” There is no indication that modification requires a hearing at all. So,  BETSCHART V. STATE OF OREGON 31 even if a defendant were to convince the court to consider a modification request, the defendant would not have the same opportunity to argue his case. As to the second factor, the dissent relies on Gerstein to argue that bail hearings do not “present . . . complex legal issues.” But Gerstein concerned a “nonadversar[ial] proceeding” of “limited function” to determine probable cause that could be decided “on hearsay and written testimony.” 420 U.S. at 120, 122. The dissent’s contention that Oregon bail hearings are only probable cause hearings is simply wrong. Under Oregon law, the magistrate considers all of the “primary release criteria,” which includes evidence of a defendant’s propensity for law- breaking and flight. Or. Rev. Stat. § 135.230(7). The defendant has the right to appear and present evidence, as do the district attorney and the victim. Id. § 135.245(5). These competing presentations of evidence make up the very “critical confrontation” to a defendant’s interests that Hovey requires to satisfy the second factor. 458 F.3d at 902. As to the third factor, the dissent quotes McNeal v. Adams, 623 F.3d 1283, 1288 (9th Cir. 2010), for the proposition that “[c]ritical stages [must] involve ‘significant consequences’ to the defendant’s case.” We note that the dissent misquotes McNeal by injecting “must” to artificially prop up its point. The actual quote is: “Critical stages involve ‘significant consequences to the defendant’s case.’” 623 F.3d at 1288. In any case, McNeal held that a motion to compel a defendant’s DNA did not have significant consequences for the defendant because his counsel had time to object, and the taking of physical evidence is otherwise “subject to meaningful challenge through the adversar[ial] process.” Id. In contrast here, the bail hearing is the 32 BETSCHART V. STATE OF OREGON adversarial process through which a defendant may meaningfully challenge his pretrial detention. Our standard of review—which the dissent appears continually to forget—is clear: “A [district] court abuses its discretion when it fails to apply the correct legal standard or bases its decision on unreasonable findings of fact.” Briseño v. Henderson, 998 F.3d 1014, 1022 (9th Cir. 2021) (alteration in original) (citation omitted). The district court appropriately applied the Ninth Circuit test for critical stages, and it does not appear, nor is it argued, that the district court’s factual findings regarding bail hearings were clearly erroneous. Id. * * * The dissent’s insistence that today we establish a “brightline rule that the Sixth Amendment right to counsel is violated by a seven-day gap” is a gross mischaracterization that demonstrates the dissent’s confusion over our standard of review. We merely hold that it was not an abuse of discretion for the district court to conclude, when faced with a complete collapse of Oregon’s indigent defense attorney network, that Gideon guarantees pretrial counsel to those incarcerated and awaiting trial. 25 25 While we agree that “[c]riminal prosecutions do not proceed in a one- size-fits-all fashion,” the district court is best positioned to make fact- specific judgments. For instance, the dissent takes issue with a part of the amended injunction concerning attorney withdrawal. That amendment was made to accommodate concerns that the parties had raised. Allowing the district court to fashion an equitable remedy based on the facts it is uniquely situated to address is the very purpose of abuse of discretion review.  BETSCHART V. STATE OF OREGON 33 2. Irreparable Harm, Balance of Equities, and the Public Interest The district court’s conclusion that Petitioners are suffering and will continue to suffer irreparable harm was not an abuse of discretion. See Hernandez, 872 F.3d at 994 (“It is well established that the deprivation of constitutional rights ‘unquestionably constitutes irreparable injury.’” (quoting Melendres v. Arpaio, 695 F.3d 990, 1002 (9th Cir. 2012))). “When the government is a party, [the third and fourth preliminary injunction factors] merge.” E. Bay Sanctuary Covenant v. Garland, 994 F.3d 962, 975 (9th Cir. 2020) (citation omitted). The district court concluded that the balance of equities tips in Petitioners’ favor because providing counsel “will guarantee efficiency, make criminal proceedings less burdensome on all involved, and will prevent cases from being needlessly delayed,” without raising administrative costs. The court also concluded that the preliminary injunction is in the public interest because “all citizens have a stake in upholding the Constitution.” Hernandez, 872 F.3d at 996 (citation omitted). The district court was within its discretion to find that the public has an interest in a functioning criminal justice system and the protection of fundamental rights. See Baird v. Bonta, 81 F.4th 1036, 1042 (9th Cir. 2023) (“A plaintiff’s likelihood of success on the merits of a constitutional claim also tips the merged third and fourth factors decisively in his favor. Because ‘public interest concerns are implicated when a constitutional right has been violated, . . . all citizens have a stake in upholding the Constitution.’” (quoting Preminger v. Principi, 422 F.3d 815, 826 (9th Cir. 2005))). 34 BETSCHART V. STATE OF OREGON Oregon contends that the preliminary injunction “will impair the State’s ability to protect victims, witnesses, and the public because it requires the State to release defendants, including potentially dangerous defendants, who are lawfully detained.” But the preliminary injunction does not unconditionally release defendants; it recognizes “the [Oregon] Circuit Court’s independent authority to set reasonable pre-trial conditions for release.” See Roman v. Wolf, 977 F.3d 935, 944 (9th Cir. 2020) (per curiam) (upholding issuance of a preliminary injunction requiring population reduction in immigration detention facilities “particularly in light of . . . the alternative means available to prevent [detainees] from absconding if they were released, such as electronic monitoring”). In Brown v. Plata, 563 U.S. 493 (2011), the Supreme Court affirmed a remedial order that effectively did the same, but with convicted prisoners. Id. at 500-01 (“[A]bsent compliance through . . . other means[,] . . . the State will be required to release some number of prisoners before their full sentences have been served.”). In Plata, the Court considered overcrowding in California prisons. After extensive litigation, a three-judge district court panel had ordered the state to “reduce its prison population to 137.5% of design capacity.” Id. The population reduction, by the panel’s estimate, “could [have been] as high as 46,000 persons.” Id. In upholding the order, the Supreme Court considered its impact on public safety. It reasoned that considering the public interest “necessarily involves difficult predictive judgments regarding the likely effects of court orders” and that “[t]hese questions are difficult and sensitive, but they are factual questions and should be treated as such.” Id. at 535. The Court then held that the district court had properly  BETSCHART V. STATE OF OREGON 35 “credited substantial evidence that prison populations can be reduced in a manner that does not increase crime to a significant degree,” id., and that “any negative impact on public safety would be ‘substantially offset, and perhaps entirely eliminated, by the public safety benefits’ of a reduction in overcrowding,” id. at 536-37 (citation omitted). In this case, the district court found that Petitioners’ requested relief did not pose a “fiscal or administrative burden on the government” and that Oregon’s fear of the threat to community safety was “theoretical.” Indeed, Oregon neither disputes that the relief imposes little or no fiscal or administrative burden nor provides any evidence that releasing non-convicted defendants, whom Oregon could monitor by any other appropriate means, would threaten community safety so drastically as to justify continuing to deny Petitioners their constitutional rights. Here, as in Plata, the relief could be characterized as “of unprecedented sweep and extent.” Id. at 501. But “so too is the continuing injury and harm resulting from these serious constitutional violations.” Id. And here, as in Plata, “[t]he State’s desire to avoid [Petitioners’ requested relief] . . . creates a certain and unacceptable risk of continuing violations of the rights of [Petitioners], with the result that many more will . . . needlessly suffer.” Id. at 533-34. “Whenever a court issues an order requiring the State to adjust its incarceration and criminal justice policy, there is a risk that the order will have some adverse impact on public safety . . . .” Id. at 534. The dissent, without any elaboration, cites the dissent in Plata to argue that the risks to the public outweigh the Petitioners’ constitutional rights. Relying on a dissent is not the best argument. A better one is that “‘[e]ven in times of 36 BETSCHART V. STATE OF OREGON crisis,’ judges must ‘not shrink from our duty to safeguard th[e] rights’ guaranteed by the Constitution.” United States v. Olsen, 21 F.4th 1036, 1057 (9th Cir. 2022) (Bumatay, J., concurring in the denial of rehearing en banc) (alterations in original) (quoting Tandon v. Newsom, 992 F.3d 916, 939 (9th Cir. 2021) (Bumatay, J., dissenting in part and concurring in part)). Denying a watershed right to criminal defendants, presumed to be innocent, is a textbook example of shrinking from this duty. The dissent cites nothing in the record to support the fear-mongering parade of horribles it claims will result if Petitioners are released. Instead, it details the crimes that Petitioners are accused of. First, we remind the dissent that criminal law features people accused of horrible things—it is criminal law after all. If the dissent were to go into the record of all the convicted prisoners in Plata, it would undoubtedly find conduct similar to, or even much worse than, what Petitioners are accused of here. The simple reality is that our Constitution protects people regardless of the accusations against them. Second, the dissent ignores a crucial part of the preliminary injunction—Petitioners are not going to be given free rein in the community. Instead, they “are subject to the conditions of release set forth in [Or. Rev. Stat.] § 135.250 and any other conditions that the Circuit Court may impose that are related to assuring the appearance of the class member and the safety of the community.” No-contact orders, GPS monitoring, and check-ins with Probation are available. The dissent does not explain why any of these standard measures would fail. The injunction further provides that if a Petitioner violates these conditions, their release can be revoked, and they are not entitled to a new seven-day period.  BETSCHART V. STATE OF OREGON 37 The dissent also asserts that the district court “failed to consider alternatives” and suggests that the court could have compelled members of the bar to represent indigent criminal defendants. But first, Oregon, despite multiple hearings and hundreds of pages of briefing, has never proposed a single alternative remedy to the district court (or our court); making up such alternatives on the fly would hardly have been an appropriate exercise of discretion. Second, the district court did, in fact, consider compelling members of the bar to represent indigent criminal defendants, and concluded that doing so in the past had not worked and repeating that mistake would be ill-advised: THE COURT: . . . The idea that judges can just grab somebody out of the hallway or grab – I mean, there was a great idea. Let’s take associate attorneys from law firms who never spent a day in a courtroom, and we’ll have them represent people. It’s kind of insulting to people who practice criminal law, first of all, and second, I – it just seems like we’re setting things up for malpractice. Indeed, the record supports this concern. One named Petitioner, who was not sure whether she had been arraigned, was appointed an attorney that had been forced out of retirement, refused to look at her case, and promptly withdrew. The dissent says that this “anecdote” does not justify the injunction. Meanwhile, the dissent—again— points to nothing in the record that supports its contrary 38 BETSCHART V. STATE OF OREGON position. 26 In any event, this practice likely also would violate the Sixth Amendment. See Barber v. Nelson, 451 F.2d 1017, 1019 (9th Cir. 1971) (“[I]f no time to prepare is available to counsel, his assistance is ineffective as a matter of law.”). The dissent even suggests that this court could order Oregon to pay their defense bar more money, but cites no authority for the extraordinary idea that we could set state wage rates under habeas. Oregon has that power yet has chosen not to wield it. The preliminary injunction respects the Oregon Constitution and state law by excepting from release those charged with murder and aggravated murder. See E. Bay Sanctuary Covenant, 994 F.3d at 985 (noting public interest in “ensuring that ‘statutes enacted by [their] representatives’ are not imperiled” (alteration in original) (citation omitted)). The district court was well within its discretion to follow Oregon law with respect to these defendants. See Melendres, 695 F.3d at 1002 (holding there was “no abuse of discretion in the district court’s determination that the equities favor issuance of a narrow, limited preliminary injunction” that does not enjoin the enforcement of valid state laws). IV. CONCLUSION Despite nearly fifty pages, the dissent never focuses on the standard of review or the Winter factors. It repeatedly disregards controlling precedent, raises new issues and 26 The dissent also states that the district court “rejected this option because it feared that some lawyers might find it ‘kind of insulting.’” That misreads the transcript. The district court was commenting that it was insulting to the criminal defense bar to suggest that their essential work could be replicated by lawyers who lack criminal defense and/or trial experience.  BETSCHART V. STATE OF OREGON 39 arguments, and either ignores authority or misreads it to prop up its personal opinions of our jurisdiction and the limits of the Constitution. The dissent’s unbounded approach is an ode to classic judicial overreach. It remains unclear why the dissent blames the district court for a “judicial jailbreak.” Consistent with the Sixth Amendment, Oregon could solve this problem overnight simply by paying appointed counsel a better wage. It is Oregon, and not the district court, that created this crisis. At the end of the day, our question is a narrow one: did the district court abuse its considerable discretion in issuing a preliminary injunction to address an unprecedented situation where, in direct violation of Gideon, unrepresented and indigent defendants wait in cells for months, helpless and powerless, while favorable evidence goes cold or disappears altogether? With that question in mind, we cannot say the district court abused its discretion. AFFIRMED. 40 BETSCHART V. STATE OF OREGON BUMATAY, Circuit Judge, dissenting: I do not say this lightly—the injunction the majority affirms here is both reckless and extreme. It orders the State of Oregon to release from jail all criminal defendants not appointed state-funded counsel within seven days of their initial appearance. Given the complexities of the situation and the shortage of public defense counsel, the result of this order is that more than a hundred criminal defendants will be immediately released from jail. And those being released are not sitting there for some petty offense. Just look at the charges of the named Petitioners here—they are accused of rape, kidnapping, strangulation, assaulting a police officer, public indecency, and burglary. All will now be released into Oregon’s communities. But this is not the end of it. Countless others will be released on an ongoing basis because the injunction applies prospectively. To avoid the inevitable chaos, our court wisely paused the district court’s extraordinary order pending appeal. But that wisdom has run out. The majority now endorses the release scheme, lifts the stay of the injunction, and lets it take immediate effect. By doing so, the Ninth Circuit is now complicit in a judicial jailbreak. I fear the coming disorder. *** For the first time in our Nation’s history, we order the release of pretrial criminal defendants from jail based solely on a delay in appointing state-funded counsel. While the Sixth Amendment grants indigent defendants the right to government-funded counsel, Gideon v. Wainwright, 372 U.S. 335 (1963), this right applies only at “critical stages” of the criminal process, Iowa v. Tovar, 541 U.S. 77, 80–81 (2004). But the district court and the majority make up a new rule: defendants must receive appointed counsel within  BETSCHART V. STATE OF OREGON 41 seven days or be released from jail. That’s an incredibly short deadline cut from whole cloth. Rather than analyzing the nuances of each defendant’s case, the district court and majority establish a categorical, one-size-fits-all rule mandating appointed counsel within a brief period. And it does so not by applying the traditional remedies of suppression or vacatur of conviction, but with blanket release from detention. If that relief were not extraordinary enough, the district court’s injunction applies on a class-wide basis, meaning that this order will lead to the immediate release of more than a hundred defendants from jail. So defendants who were denied bail—those considered too dangerous to release— will immediately be let loose into the community. See Or. Rev. Stat. § 135.240. This jailbreak applies regardless of the posture of a particular case or the individualized assessment of the defendant’s dangerousness. And if all this were not damaging enough, the district court extended this remedy to all future criminal defendants in the State. So it will lead to the ongoing release of an unknown number of defendants from Oregon jails—even those not even arrested yet. In approving this order, we have effectively commandeered the state courts and indefinitely dictate to Oregon judges when a defendant must be released from a state jail. Never mind that habeas corpus is a remedy that may be invoked only by those currently in “custody” based on the illegality of that custody. Even on its own terms, the injunction here makes little sense. It is purportedly based on the Sixth Amendment’s fundamental right to counsel—a right which attaches to all criminal defendants charged with felonies. Yet the order picks and chooses which defendants are entitled to 42 BETSCHART V. STATE OF OREGON immediate release. While freeing dozens of defendants, the injunction decrees that some defendants must remain in jail without appointed counsel—defendants accused of murder and aggravated murder. While keeping those defendants in jail makes practical sense, it doesn’t make constitutional sense. Those defendants possess the same constitutional right to counsel as everyone else. So it is baffling that the district court and now the majority somehow conclude that the Sixth Amendment doesn’t apply equally to those charged with murder. This sort of interest-balancing reeks of policymaking, not dispassionate application of the rule of law. Tellingly, the majority doesn’t even try to defend this. And most ironically of all, the order doesn’t even cure the alleged Sixth Amendment violation. Petitioners complain that Oregon has failed to appoint them their state- funded counsel. But under the order, not one defendant will receive appointed counsel. Whether in jail or on bond, Petitioners will still be left unrepresented. Sure, the injunction may inflict so much harm on Oregon that it may push the State to work harder to fix the problem, but it doesn’t directly remedy the supposed Sixth Amendment injury for any defendant. In fairness, the district court faced challenging circumstances. Oregon suffers from a critical shortage of public defense attorneys. In 2021, state officials exacerbated the problem by limiting the number of cases public defense attorneys may take. Following this rule change, the number of indigent defendants without state-funded counsel skyrocketed. The State responded by seeking to overhaul the public-defense system and by allocating $100 million in new funds to it. Still, as of the district court’s hearing on the matter, roughly 106 criminal defendants remained in jail without state-appointed counsel.  BETSCHART V. STATE OF OREGON 43 While I share the concerns for the challenges facing Oregon, this injunction is not the solution. The delay in the appointment of counsel is troubling. The Sixth Amendment is a fundamental right and must be adhered to in all applicable criminal proceedings. And the State must fix this problem. But the jailbreaking solution crafted by the district court and now endorsed by the majority is not a legally permissible response. *** Several reasons show that the majority was wrong to affirm the district court injunction here. This dissent focuses on five errors: First, we simply lack subject-matter jurisdiction. Petitioners seek habeas corpus relief under 28 U.S.C. § 2241. But by its plain language, habeas relief requires a person “in custody in violation of the Constitution.” 28 U.S.C. § 2241(c)(3). Neither the district court nor the majority explains why failing to appoint state-funded counsel makes pretrial custody unconstitutional. Often, the remedy for a Sixth Amendment violation is suppression of evidence or vacatur of the defendant’s conviction. The right to counsel is, after all, about defending against the merits of a prosecution—protecting against “results [that] might well settle the accused’s fate and reduce the trial itself to a mere formality.” Maine v. Moulton, 474 U.S. 159, 170 (1985) (simplified). But no court has ever ruled that a failure to appoint state-funded counsel makes pretrial custody by itself unconstitutional. The majority ignores these limits on our authority by bounding over this important issue. And even putting aside the fundamental incompatibility between the right, the remedy, and habeas corpus, the application of class-wide relief using this writ is itself dubious. See 44 BETSCHART V. STATE OF OREGON Jennings v. Rodriguez, 583 U.S. 281, 324 n.7 (2018) (Thomas, J., concurring) (questioning whether habeas relief may be granted on a class-wide basis). To top it all off, it is impossible to see how the district court had authority to issue prospective relief for those who have yet to be accused of a crime and have yet to be placed in “custody”—the core requirement for habeas relief. See 28 U.S.C. § 2241(c)(3). Second, the district court order violates the Younger abstention doctrine. See Younger v. Harris, 401 U.S. 37 (1971). It is not seriously contested that this case meets the four Younger factors demanding abstention. That alone should have ended the matter. But we contort the doctrine by expanding the extraordinary-circumstances exception. That exception is a limited one; it does not swallow the rule and anoint federal judges as superintendents of a state’s criminal-justice system. Third, turning to the merits, the district court and majority’s Sixth Amendment analysis is disconnected from precedent. Under the Supreme Court’s longstanding framework, the right to counsel is violated only when the defendant lacks an adequately prepared attorney at a “critical stage” of the criminal process—one that determines the prosecution’s merits. To justify the blanket release, we disregard a half-century of caselaw and hold that a seven- day stretch without appointed counsel violates the Sixth Amendment. So we no longer need to analyze the posture of the criminal case to determine whether a critical stage has occurred; we just need to count the days from initial appearance. That conclusion transforms the well-developed right-to-counsel doctrine into a novel one—one unsupported by text, history, or precedent.  BETSCHART V. STATE OF OREGON 45 Fourth, the Fourteenth Amendment’s Due Process Clause doesn’t justify the injunction either. While the district court viewed the delays in the appointment of state- funded counsel as a substantive due process violation, no court has extended substantive due process as far as this. Due process provides certain procedural protections and ensures access to courts, but it has little to say about the timing of the appointment of state-funded counsel. Fifth, even while recognizing its discretion in the matter, the district court failed to properly balance the interests of the public and the parties in crafting the injunction. No one can seriously question the obvious risks to the public by the immediate release of dozens of prisoners. Yet the district court didn’t even seriously consider more narrowly tailored alternatives before choosing its jailbreak solution. But Petitioners concede that other remedies are available, like ordering new bail hearings with counsel or directing the State to revisit its public-defender policies. Given that this order will not even remedy the lack of appointed counsel, the balance of interests cannot favor this injunction. Despite all these issues, the majority rushes to lift the stay of the injunction and endorses the prisoner-release scheme. Even worse, the majority suggests expanding this jailbreak solution to other states in the circuit. Thus, the majority’s endorsement of a seven-day rule now becomes the law in every State and federal district in the Ninth Circuit. We are now embarking into uncharted constitutional territory. This case is not only a radical reinterpretation of the Sixth Amendment and due process, but a radical reinterpretation of federalism and the separation of powers, a radical reinterpretation of the scope of habeas corpus, and a radical reinterpretation of class actions. It doesn’t push our precedent—it sets it ablaze. 46 BETSCHART V. STATE OF OREGON *** Because our court has ill-considered this radical decision, I respectfully dissent. I. Background Oregon suffers from a significant shortage of public defense attorneys. Several factors contribute to this problem, such as the backlog of cases from the COVID-19 pandemic and increased remands for new trials following the end of nonunanimous jury verdicts in Oregon, see Ramos v. Louisiana, 140 S. Ct. 1390, 1407–08 (2020). But the most acute cause is the State’s attempt to improve the quality of representation by limiting the number of cases public defenders can take. After that policy was enacted, the gap between the number of indigent defendants who require counsel and the number of defenders available to represent them increased exponentially. The result is a delay, and sometimes a lengthy delay, in the State providing government-funded counsel to criminal defendants. Under Oregon law, the initial release decision must be made at arraignment unless “good cause” is shown, in which case the hearing can be delayed up to five days. Or. Rev. Stat. § 135.245(2)(a), (7)(a). A defense attorney generally is present and available at arraignment. See id. § 135.040. A judge shall deny release if (1) the defendant is charged with murder, aggravated murder, or treason, and the proof is evident or the presumption is strong that the defendant is guilty; or (2) the defendant is charged with a violent felony and there is probable cause to believe that the defendant committed the crime and clear and convincing evidence of a danger of physical injury or sexual victimization to the  BETSCHART V. STATE OF OREGON 47 victim or public by the defendant while on release. Id. § 135.240(2), (4). Otherwise, a judge may grant release subject to conditions and bail. Id. § 135.245. Petitioners filed a joint petition for the writ of habeas corpus in the District of Oregon under 28 U.S.C. § 2241, alleging violation of their right to state-funded counsel. Petitioners sought to certify a class to include all indigent criminal defendants held in jail without counsel, as well as another class to include all indigent criminal defendants placed under restrictive release conditions without counsel. The district court provisionally certified the class of jailed defendants and entered a temporary restraining order freeing any indigent defendants in the Washington County jail who had not been appointed counsel within ten days of either their arraignment or the withdrawal of their previously appointed counsel. The State of Oregon intervened and Petitioners subsequently moved for a preliminary injunction. The district court granted a preliminary injunction and sua sponte applied it statewide. The preliminary injunction said the following: • If counsel is not secured within seven days of initial appearance for any class member currently in physical custody, or if counsel is not appointed within seven days of the withdrawal of previously appointed counsel, the sheriff of that county is ordered to release the class member. • Any future class member who has not secured counsel within seven days of 48 BETSCHART V. STATE OF OREGON their initial appearance must be released from physical custody. Less than two weeks later, the district court amended its order, materially changing the injunction’s terms without any accompanying explanation. First, the amended order redefines the scope of the class to “individuals who are or will be” physically housed in a jail in Oregon. The district court no longer uses the “future class member” language. It is unclear if there is any substantive difference between the terms. Second, the district court also clarifies that the preliminary injunction “does not apply to crimes of murder and aggravated murder.” It notes that the injunction “does not impact the provisions of Article I, Section 43 of the Oregon Constitution.” That section, like Oregon Revised Statute § 135.240(2) and (4), sets forth that defendants charged with murder, aggravated murder, treason, or a violent felony are not bailable if the court makes certain findings. Or. Const., Art. I, § 43(1)(b). Third, the amended order says that the injunction does not apply to “class members who fire their attorney.” Fourth, the amended order limits the class members eligible for release after the withdrawal of a prior counsel. Reappointment within seven days must only occur “[i]f counsel is secured within the seven-day period but subsequently withdraws due to a conflict within that period.” Thus, a class member is only entitled to reappointment within seven days of the withdrawal of a previously appointed attorney if the withdrawal was due to a conflict and was within seven days of the initial appearance.  BETSCHART V. STATE OF OREGON 49 Fifth, under the amended order, Oregon courts may set conditions of release to ensure the appearance of class members and the safety of the community. Oregon courts may also require class members to execute a “release agreement” before release. The failure to execute such an agreement “will result in the continued detention of the class member.” The State of Oregon sought an emergency stay of the district court order, which we granted. We then expedited this appeal. The majority votes to lift the stay and affirm the district court’s order. So the district court’s preliminary injunction now goes into effect. II. Lack of Jurisdiction To begin, the district court simply lacked authority to issue this injunction. Under 28 U.S.C. § 2241, federal courts have no jurisdiction over state defendants unless the defendant’s custody itself is illegal. Here, neither the district court nor the majority demonstrate how the alleged violation of the right to state-funded counsel alone renders pretrial custody unconstitutional. The district court also didn’t examine whether we have authority to grant a class-wide remedy under habeas. And there’s reason to question whether we do. Finally, the district court didn’t consider whether habeas—which requires “custody”—can be prescribed prospectively to “future class members” who may one day be detained without appointed counsel. Such relief seems at odds with the plain text of § 2241. All these reasons counsel against permitting the injunction to take effect. 50 BETSCHART V. STATE OF OREGON A. Federal habeas statutes are recognized as a grant of “jurisdiction” for courts “to inquire into violations of the United States Constitution.” See Carafas v. LaVallee, 391 U.S. 234, 238 n.11 (1968); see also Maleng v. Cook, 490 U.S. 488, 494 (1989) (per curiam) (explaining that the issue of custody goes to the “subject-matter jurisdiction of the habeas court”); Hensley v. Mun. Ct., San Jose Milpitas Jud. Dist., 411 U.S. 345, 352 n.10 (1973) (determining that habeas jurisdiction “would not merely have [been] postponed . . . but would have [been] barred . . . altogether” without a finding of custody). Because § 2241 is a jurisdictional statute, this threshold question cannot be waived. See Arbaugh v. Y&H Corp., 546 U.S. 500, 514 (2006) (noting that “subject-matter jurisdiction, because it involves a court’s power to hear a case, can never be forfeited or waived” (simplified)). To that end, it is our duty to ensure that we possess the authority to render a judgment under § 2241. We lack that authority here. Section 2241 commands that “[t]he writ of habeas corpus shall not extend to a prisoner unless . . . [h]e is in custody in violation of the Constitution or laws or treaties of the United States.” 28 U.S.C. § 2241(c)(3). Put simply, § 2241(c)(3) provides a mechanism to challenge the unlawfulness of one’s “custody.” Naturally, then, “an action sounds in habeas [under § 2241(c)(3)] . . . if success in that action would necessarily demonstrate the invalidity of confinement or its duration.” Pinson v. Carvajal, 69 F.4th 1059, 1071 (9th Cir. 2023) (dismissing a habeas petition under § 2241 for lack of subject-matter jurisdiction) (simplified). Claims that, if successful, don’t demand “the invalidity of the confinement” fall outside “the core of habeas corpus.” Id. In other words, “the relevant question is whether, based on the allegations in  BETSCHART V. STATE OF OREGON 51 the petition, release is legally required.” Id. at 1072. It is one’s custody that must be unlawful—not any other violation. Whether we possess habeas jurisdiction here boils down to a simple question—if Petitioners’ claims succeed, does it make their custody illegal so that their release is mandatory? If custody itself is not unconstitutional, the claim cannot be vindicated under § 2241 and we lack jurisdiction. See id. (“[T]he proper analytical tack when determining whether actions . . . are at the core of habeas is to consider why release from confinement is necessary to remedy the underlying alleged violation.”). A bit of background on the Sixth Amendment illustrates how release from jail isn’t a proper remedy here. As a general principle, “remedies should be tailored to the injury suffered from the constitutional violation” but “should not unnecessarily infringe on competing interests.” United States v. Morrison, 449 U.S. 361, 364 (1981). Practically, that approach “tailor[s] relief . . . to assure the defendant the effective assistance of counsel and a fair trial.” Id. at 365; see also Rushen v. Spain, 464 U.S. 114, 119–20 (1983) (“The adequacy of any remedy is determined solely by its ability to mitigate constitutional error, if any, that has occurred”). In plain terms, the general rule is—cure the constitutional defect and inflict no further harm. Before today, our practice generally followed that guidance from the Supreme Court. We constructed remedies for a violation of the right to counsel to include either suppressing evidence obtained from the violation or, in extreme cases, vacating one’s conviction. See, e.g., Cahill v. Rushen, 678 F.2d 791, 795–96 (9th Cir. 1982) (suppressing evidence obtained in violation of the right to counsel); 52 BETSCHART V. STATE OF OREGON United States v. Kimball, 884 F.2d 1274, 1280 (9th Cir. 1989) (same); United States v. Forrester, 512 F.3d 500, 509 (9th Cir. 2008) (vacating a conviction following a violation of the right to counsel). Immediate release from jail, to my knowledge, has never been a remedy for a violation of the right to appointed counsel. And there’s good reason for that. According to the Supreme Court, the right of state-funded counsel is to ensure that “the accused . . . need not stand alone against the State at any stage of the prosecution, formal or informal, in court or out, where counsel’s absence might derogate from the accused’s right to a fair trial.” United States v. Wade, 388 U.S. 218, 226 (1967). So the right’s purpose is to guarantee a meaningful defense to prosecution—not merely to assist with any independent interest of the defendant’s, such as avoiding pretrial detention. See Rothgery v. Gillespie Cnty., Tex., 554 U.S. 191, 216 (2008) (Alito, J., concurring) (“[D]efense at trial, not defense in relation to other objectives” is protected by the right.). That’s why the Court has limited the right to “critical” pretrial stages that “preserve the defendant’s basic right to a fair trial.” Wade, 388 U.S. at 227. Without more, a violation of the right to appointed counsel does not render pretrial detention illegal. In other words, the right is concerned with the ultimate merits of the criminal prosecution—not securing every possible advantage for a defendant. In attempting to refute this point, the majority unwittingly proves it. The majority cites articles discussing the release of prisoners from the State of Florida post- Gideon. But the majority misses the most basic fact of those articles—each of those prisoners had already been convicted. See, e.g., Bruce R. Jacob, Memories of and Reflections about Gideon v. Wainwright, 33 Stetson L. Rev. 181, 222 (2003)  BETSCHART V. STATE OF OREGON 53 (“[M]ore than 4,500 of the 8,000 inmates in Florida could be released and retried, or released without retrial. Of these, 4,065 had been convicted after pleading guilty while 477 had been convicted after going to trial.”) (emphases added). In this case, habeas is even more inapplicable because the alleged Sixth Amendment violation didn’t cause Petitioners’ detention. The pretrial detention determination is generally made at arraignment while Petitioners were represented by counsel. See Or. Rev. Stat. § 135.245. So nothing about delaying the appointment of state-funded counsel made the pretrial detention unconstitutional. Given that the purported constitutional violation didn’t lead to pretrial detention, the remedy is not release from “custody”—taking this case out of the scope of § 2241. Indeed, releasing a defendant from custody here would have no effect at all on the lack of appointed counsel. Complicating things even more, the district court didn’t order release from custody, in the habeas sense, it only ordered release from jail. The district court still orders that Petitioners must submit to conditions of release that ensure their appearance in court and the safety of the community. And the district court directed that the “[f]ailure of the class member to execute [a] release agreement will result in the continued detention of the class member.” But these conditions often amount to “custody” for purposes of habeas. See, e.g., Jones v. Cunningham, 371 U.S. 236, 243 (1963) (explaining that a prisoner’s conditions of release were “enough to keep him in the ‘custody’” of a parole board for habeas purposes because they “significantly restrain [his] liberty to do those things which in this country free men are entitled to do”); Hensley, 411 U.S. at 349 (“[A] substantial number of courts, perhaps a majority, have concluded that a person released on bail or on his own recognizance may be 54 BETSCHART V. STATE OF OREGON ‘in custody’ within the meaning of the statute. . . . [W]e conclude that this . . . reflects the sounder view.”). Thus, even the district court seemingly did not think that the Sixth Amendment violation here results in “custody in violation of the Constitution”—the requirement for habeas relief under § 2241. If it did, it should have ordered release from all custody, not just release from jail. The confusion over habeas’s requirement of illegal custody is also apparent from the district court’s refusal to order any relief for Petitioners charged with murder or aggravated murder, even though they suffer the same alleged violation of the right to counsel. The Sixth Amendment right to counsel does not depend on the type of felony charged. But the district court and the majority mysteriously exempt those charged with murder and aggravated murder from the district court’s seven-day rule. This is not how the Sixth Amendment works. True, Oregon may law forbid their pretrial release under some conditions, see Or. Const., Art. I, § 43(1)(b); Or. Rev. Stat. § 135.240(2), (4), but that feature has nothing to do with the Sixth Amendment analysis. Indeed, the same Oregon law provides that those charged with violent felonies are not bailable under certain conditions. See Or. Const., Art. I, § 43(1)(b); Or. Rev. Stat. § 135.240(2), (4). So it makes no sense to rely on state law here. Either the Sixth Amendment violation doesn’t render custody illegal, meaning that no defendants should be released. Or it does, meaning that all defendants should be released. There’s no room for picking and choosing who deserves a constitutional right. The bottom line—if the district court and majority were correct that Petitioners’ Sixth Amendment rights had been violated and that release from custody were the mandatory remedy, then there would be no valid basis to deprive some  BETSCHART V. STATE OF OREGON 55 defendants of their rights. The carve-out of one particularly eye-catching group of defendants only cements the irregularity of the injunction. B. We also should have questioned whether this habeas petition can be pursued in a class action. Whether habeas relief is available through class action remains an open question at the Supreme Court. See Jennings, 583 U.S. at 324 n.7 (Thomas, J., concurring) (“This Court has never addressed whether habeas relief can be pursued in a class action. I take no position on that issue here.” (simplified)). At least in the immigration-detention context, the Supreme Court has instructed our court to “consider” whether a Rule 23 class action is an “appropriate vehicle” for providing habeas relief in light of Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011). See id. at 313 (majority opinion). Wal- Mart Stores tells us that some class actions may only apply “when a single injunction or declaratory judgment would provide relief to each member of the class.” 564 U.S. at 360. True, some of our older decisions have suggested a habeas petition can be treated as a class action. See Mead v. Parker, 464 F.2d 1108, 1112 (9th Cir. 1972); Cox v. McCarthy, 829 F.2d 800, 804 (9th Cir.1987); Rodriguez v. Hayes, 591 F.3d 1105, 1117 (9th Cir. 2010), abrogation recognized by Rodriguez Diaz v. Garland, 53 F.4th 1189, 1199–1201 (9th Cir. 2022). But given more recent changes in the legal landscape regarding class actions, I question whether these cases remain good law. Further, the Supreme Court’s long history of reversing our immigration-detention class-action cases suggests that our prior views may be an outlier. See, e.g., Garland v. Aleman Gonzalez, 596 U.S. 543 (2022) (reversing Aleman Gonzalez v. Barr, 955 F.3d 762 56 BETSCHART V. STATE OF OREGON (9th Cir. 2020)); Jennings, 583 U.S. at 281 (reversing Rodriguez v. Robbins, 804 F.3d 1060 (9th Cir. 2015)). But even if a habeas petition could be grounded in a class action, other questions remain—like, what standards must we use? Whether the class-action standards of Rule 23 of the Federal Rules of Civil Procedure apply in habeas corpus proceedings “has engendered considerable debate.” Harris v. Nelson, 394 U.S. 286, 294 n.5 (1969) (simplified). Our own court has questioned whether “Rule 23 might be technically inapplicable to habeas corpus proceedings.” Ali v. Ashcroft, 346 F.3d 873, 891 (9th Cir. 2003) (simplified), opinion withdrawn on other grounds by Ali v. Gonzales, 421 F.3d 795 (9th Cir. 2005). While non-precedential, Ali also explained that an “analogous procedure by reference to Rule 23” could be applied. Id. (simplified). Two circuits have adopted the view that federal courts must create new standards for class actions brought under habeas. See United States ex rel. Sero v. Preiser, 506 F.2d 1115, 1125 (2d Cir. 1974); United States ex rel. Morgan v. Sielaff, 546 F.2d 218, 220–21 (7th Cir. 1976). Regardless of whether Rule 23 directly applies, any habeas class action will need to rely on similar considerations. And those factors are challenging when applied to habeas corpus proceedings. For example, Rule 23(a) delineates four prerequisites for class certification: (1) numerosity, (2) commonality, (3) typicality, and (4) adequacy of representation. See Fed. R. Civ. P. 23(a). While numerosity might be easily met, commonality and typicality requirements are a significant wrinkle when applied to habeas proceedings. Consider a due process challenge. “[D]ue process is flexible” and so a Due Process Clause claim “calls for such  BETSCHART V. STATE OF OREGON 57 procedural protections as the particular situation demands.” Jennings, 583 U.S. at 314 (quoting Morrissey v. Brewer, 408 U.S. 471, 481 (1972)). Because of this, a “class action litigated on common facts” might not be “an appropriate way to resolve . . . Due Process Clause claims.” Id. After all, resolving Due Process claims for hundreds of individuals is unlikely to be resolved “in one stroke.” Wal-Mart Stores, 564 U.S. at 351. The Solicitor General recently raised related issues with class actions in the Eighth Amendment context. See City of Grants Pass, Ore. v. Johnson, No. 23-175, Brief for the United States as Amicus Curiae 31 (Mar. 4, 2024). The Solicitor General cited approvingly the concern that “the need for particularized inquiries should have precluded the certification of a class because respondents cannot satisfy Rule 23(a)’s commonality requirement or Rule 23(b)(2)’s requirement that the challenged conduct must be ‘such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them[.]” Id. (quoting Wal- Mart Stores, 564 U.S. at 360). The Solicitor General then argued that the case should be vacated and remanded to “reconsider all the relevant issues in the case—including class certification.” Id. at 32. Similar concerns abound here. Criminal proceedings all have different moving pieces; they proceed at different paces for different reasons. For some defendants, delay may be the goal; for others, speed is a litigation advantage. Thus, to provide a one-size-fits-all remedy to a problem with individualized effects makes little sense. Neither the district court nor the majority grapple with these concerns. Ultimately, then, it is unclear whether and how a class action would even apply in a habeas corpus proceeding. 58 BETSCHART V. STATE OF OREGON C. There’s an even bigger problem here—it’s unlikely that habeas relief can be granted prospectively to individuals who are not yet even in custody. As stated earlier, the habeas provision invoked by Petitioners “shall not extend” except to “a prisoner” who “is in custody.” 28 U.S.C. § 2241(c)(3). But the district court’s injunction applies to “any future class member” or any individual who “will be . . . physically housed in a jail in Oregon.” So it seems that the district court has afforded habeas relief to parties who are neither “prisoner[s]” nor “in custody” now, which conflicts with the plain language of § 2241. While the Supreme Court has instructed that prisoners may apply for federal habeas relief for a sentence they have not yet served, see Peyton v. Rowe, 391 U.S. 54 (1968), or for a sentence they previously served if they are in custody for a consecutive sentence, see Garlotte v. Fordice, 515 U.S. 39 (1995), the key factor is that there must be some sort of current custody that the prisoner will experience. By opening the injunction to “any future class member,” we expand habeas relief even to individuals who have yet to be arrested. Thus, if § 2241 applies, the district court’s sweeping injunction may have exceeded its statutory scope. See Nielsen v. Preap, 139 S. Ct. 954, 962 (2019) (“Did the Preap court overstep this limit by granting injunctive relief for a class of aliens that includes some who have not yet faced—but merely ‘will face’—mandatory detention? The District Court said no, but we need not decide.”).  BETSCHART V. STATE OF OREGON 59 III. Younger Abstention That leads to Younger abstention. This petition shouldn’t have progressed this far because we should have ordered abstention from the start. The Younger doctrine is an exception to the general rule that federal courts have a duty to “hear and decide” cases falling within their jurisdiction. Sprint Commc’ns, Inc. v. Jacobs, 571 U.S. 69, 77 (2013). Younger reflects the importance of federalism in our constitutional system. It requires us to recognize “the fact that the entire country is made up of a Union of separate state governments” and to respect the “belief that the National Government will fare best if the States and their institutions are left free to perform their separate functions in their separate ways.” Younger, 401 U.S. at 44. In other words, Younger commands us to exercise some humility and acknowledge that, despite our jurisdiction, considerations more important than our desire to correct perceived wrongs require us to abstain and allow the state courts to manage their own proceedings. See Pennzoil Co. v. Texaco, Inc., 481 U.S. 1, 11 (1987) (noting that Younger helps protect “the State’s interests in the proceeding” and “avoid[s] unwarranted determination of federal constitutional questions” (simplified)). Younger applies in various circumstances where there are pending state proceedings parallel to an action for federal equitable relief. See Sprint at 77–78 (collecting cases). The quintessential application of the doctrine, however, is to avoid intervening in pending state criminal proceedings. As Younger itself said, courts “should not act to restrain a criminal prosecution, when the moving party has an 60 BETSCHART V. STATE OF OREGON adequate remedy at law and will not suffer irreparable injury if denied equitable relief.” Younger, 401 U.S. at 43–44. Unfortunately, we fail to follow Younger here. A. We apply a four-factor test to determine whether abstention under Younger is appropriate. Younger applies “when: (1) there is an ongoing state judicial proceeding; (2) the proceeding implicates important state interests; (3) there is an adequate opportunity in the state proceedings to raise constitutional challenges; and (4) the requested relief seeks to enjoin or has the practical effect of enjoining the ongoing state judicial proceeding.” Arevalo v. Hennessy, 882 F.3d 763, 765 (9th Cir. 2018) (simplified). This habeas petition meets each factor— The first two factors are easily satisfied. Indeed, no one contests them. Petitioners here seek interference with dozens, if not hundreds, of ongoing criminal prosecutions. And the prosecution of criminal law is the quintessential state interest. See Judice v. Vail, 430 U.S. 327, 335 (1977) (recognizing the importance of “the State’s interest in the enforcement of its criminal laws”). The third factor is also met. For this factor, we look to whether a procedural bar to the presentation of a federal constitutional claim exists. Commc’ns Telesystems Int’l v. Cal. Pub. Util. Comm’n, 196 F.3d 1011, 1020 (9th Cir. 1999). We “assume that state procedures will afford an adequate remedy, in the absence of unambiguous authority to the contrary.” Pennzoil Co., 481 U.S. at 15. Here, nothing disturbs this presumption. No procedural bar prevents Petitioners from asserting their federal right-to- counsel claim in state court. Indeed, we have an example in  BETSCHART V. STATE OF OREGON 61 the record. Oregon asked us to take judicial notice of an order from the Circuit Court of Multnomah County granting relief for a Sixth Amendment violation nearly identical to the ones asserted in this petition. See State v. Cutting, No. 21CR06122, slip op. 1–3 (Or. Cir. Ct. Mar. 7, 2022). The state court concluded that the State had violated the defendant’s Sixth Amendment rights by failing to provide counsel for any proceedings after his arraignment and gave the State 25 days to appoint counsel. Id. If the State failed to do so, the court made clear that it would dismiss the defendant’s charges without prejudice. Id. So it’s clear that Oregon courts not only take the constitutional rights of criminal defendants seriously, but are able and willing to grant appropriate relief. The fourth factor also favors abstention. First, in their habeas petition, Petitioners expressly asked the district court to dismiss their charges. Such action would, of course, enjoin state proceedings. But even more, the district court’s sweeping injunction constitutes a pervasive and continuous intrusion into ongoing state prosecutions. See O’Shea v. Littleton, 414 U.S. 488, 502 (1974) (observing that an injunction which requires “a major continuing intrusion of the equitable power of the federal courts into the daily conduct of state criminal proceedings is in sharp conflict with the principles of equitable restraint” recognized by Younger). In O’Shea, the Court said that Younger applies to “an injunction aimed at controlling or preventing the occurrence of specific events that might take place in the course of future state criminal trials,” even if it didn’t enjoin any criminal prosecution. Id. at 500. The purpose of abstention, the Court said, is to avoid “interference in the state criminal process by means of continuous or piecemeal interruptions of the state 62 BETSCHART V. STATE OF OREGON proceedings by litigation in the federal courts.” Id. So the Court reversed an injunction on abstention grounds because it “would disrupt the normal course of proceedings in the state courts” and “it would require for its enforcement the continuous supervision by the federal court over the conduct of the [state courts] in the course of future criminal trial proceedings involving any of the members of the [defendants’] broadly defined class.” Id. at 501. The Court also worried how an injunction, which “impose[s] continuing obligations of compliance,” would be enforced against state courts and decried a regime of constant “monitoring of the operation of state court functions.” Id. So abstention may be required even when a challenged injunction doesn’t directly enjoin the ongoing prosecution of criminal defendants. O’Shea should control here. While Petitioners’ prosecutions may proceed in some form, the district court injunction represents a “continuous . . . interruption[]” of those proceedings in general and of state pretrial-detention decisions in particular. Id. at 500. First, it would require constant monitoring of Oregon courts for compliance. Second, it would need ongoing intrusion and refinement by the district court. For example, it doesn’t apply when a defendant “fire[s]” his state-appointed counsel, but further proceedings would be necessary to determine when a “firing” has occurred. The simple reality is that the injunction calls for the ongoing federal management of state criminal prosecutions—the quintessential Younger problem. Given the satisfaction of these four factors, there’s no question that Younger abstention applies here.  BETSCHART V. STATE OF OREGON 63 B. Despite this, the majority presses ahead with our interference with Oregon’s courts based on a dubious expansion of the extraordinary-circumstances exception to Younger. The extraordinary-circumstances exception asks whether a case is so extreme as to justify displacement of our foundational principles of federalism. The Supreme Court has made clear that the exception requires “an extraordinarily pressing need for immediate federal equitable relief, not merely in the sense of presenting a highly unusual factual situation.” Kugler v. Helfant, 421 U.S. 117, 125 (1975). Outside of cases of proven harassment or prosecutions undertaken in bad faith, the exception may only apply in “‘extraordinary circumstances’ that might constitute great, immediate, and irreparable harm.” Moore v. Sims, 442 U.S. 415, 433 (1979). With no allegation of bad faith or harassment, only “irreparable harm” comes into play here. But that doesn’t fit either. The district court concluded that Petitioners have established an irreparable injury because they are “being held in custody without counsel” in violation of the Sixth and Fourteenth Amendments. But the existence of a constitutional violation isn’t enough to override Younger. After all, constitutional violations are at issue in many abstention cases. “[W]ithout some claim that a prosecution affects federally protected rights, there would be no basis for federal jurisdiction in the first place, and thus nothing from which to abstain.” Bristol-Myers Squibb Co. v. Connors, 979 F.3d 732, 738 (9th Cir. 2020). Rather, the question is whether the constitutional injury could be remedied outside of this habeas proceeding. Here, the district court seemed to conflate the injury from the 64 BETSCHART V. STATE OF OREGON alleged violation of the Sixth and Fourteenth Amendments with the injury of being detained pretrial. But that doesn’t make sense because neither alleged constitutional deprivation caused the pretrial custody here. Recall that the initial detention decision is usually made at arraignment when Petitioners are represented by counsel. So the injury of pretrial detention doesn’t invariably flow from the lack of appointed counsel. Even assuming a constitutional violation occurs later, it doesn’t necessarily cause Petitioners’ pretrial detention as a class. Thus, the injunction releasing defendants with a seven-day gap in court-appointed representation is a remedy unconnected from the claimed irreparable harm. As discussed above, the Sixth Amendment isn’t a protection against pretrial detention without counsel. Instead, the Amendment entitles a defendant to state-funded representation at critical stages of the trial—that is, stages that determine the merits of the criminal prosecution. So the harm the Sixth Amendment protects against is a conviction obtained through uncounseled critical stages. Pretrial custody is separate. There’s no independent Sixth Amendment protection against being held in pretrial custody without counsel. Indeed, the Supreme Court has never required appointed counsel at a pretrial detention proceeding. See Gerstein v. Pugh, 420 U.S. 103, 123 (1975) (“To be sure, pretrial custody may affect to some extent the defendant’s ability to assist in preparation of his defense, but this does not present [a] high probability of substantial harm[.]”). Thus, if Petitioners are left without appointed counsel at a critical stage, the right’s vindication can come after trial through vacatur of the conviction because any violation “bears directly on the framework within which the trial  BETSCHART V. STATE OF OREGON 65 proceeds, . . .—or indeed on whether it proceeds at all.” United States v. Gonzalez-Lopez, 548 U.S. 140, 150 (2006) (simplified). Think of it another way. If a defendant is made to go to trial in violation of the right to counsel, it is not the trial itself, but the lack of counsel during critical stages that constitutes the injury. So any alleged Sixth Amendment injury isn’t irreparable because redoing the relevant stage (or perhaps the whole criminal proceeding) repairs the harm. See, e.g., Forrester, 512 F.3d at 509 (reversing a conviction because of right-to-counsel violation). Thus, the claimed Sixth Amendment violation here can be remedied in later state proceedings or in a post-conviction federal habeas proceeding. The same goes for any claimed substantive due process violation. And Page v. King, 932 F.3d 898 (9th Cir. 2019), which the district court relied on, shows why its Younger analysis was off. In Page, a defendant accused of rape alleged that his due process rights were violated because the State detained him based on a “stale and scientifically invalid probable cause determination.” 932 F.3d at 904. We concluded that Younger abstention was inappropriate because the claimed due process violation directly led to his “complete loss of liberty” pretrial which is “irretrievable” regardless of the outcome of trial. Id. So, in that case, the defendant alleged that the constitutional violation directly caused the pretrial detention. Here, we have no such scenario. In fact, each Petitioner was represented at arraignment when the initial detention decision was made. No direct link can be drawn between the constitutional violation and the detention and so Page doesn’t support the Younger exception here. Likewise, Arevalo, which the majority focuses on, doesn’t show “extraordinary circumstances” either. In that 66 BETSCHART V. STATE OF OREGON case, the State conceded that Arevalo’s federal constitutional rights to equal protection and due process were violated when the state court ordered him detained on $1 million bond without considering his ability to pay and nonmonetary alternatives to bail. See Arevalo, 882 F.3d at 764–65. Despite the concession, the district court sua sponte applied Younger abstention. Id. We reversed under the “irreparable harm exception.” Id. at 766. We observed that “[d]eprivation of physical liberty by detention” could constitute an “irreparable harm.” Id. at 767. But in that case, unlike this one, the constitutional violation directly caused the pretrial detention. We noted that “the petitioner has been incarcerated for over six months without a constitutionally adequate bail hearing.” Id. But here, Petitioners—as a class—haven’t shown that any Sixth Amendment or Fourteenth Amendment violation directly caused their pretrial detention. Nor have they shown—as a class—that those rights can’t be vindicated in state proceedings or later federal proceedings. In sum, the massive federal seizure of Oregon’s criminal justice apparatus is precisely the kind of action barred by Younger. We make a mistake in shrugging off this significant federalism concern. IV. Likelihood of Success on the Merits Even setting aside the myriad of procedural hurdles barring the district court’s action here, no injunction was appropriate because Petitioners cannot show the requisite likelihood of success on the merits. In our circuit, this is not only the “most important” factor, but also a dispositive one. Garcia v. Google, Inc., 786 F.3d 733, 740 (9th Cir. 2015) (en banc) (“[W]hen a plaintiff has failed to show the  BETSCHART V. STATE OF OREGON 67 likelihood of success on the merits, we need not consider the remaining three [Winter v. Nat. Res. Def. Council, 555 U.S. 7, 20, (2008) elements for a preliminary injunction].” (simplified)). And Petitioners’ burden was even higher here because the district court imposed a “mandatory injunction” requiring Oregon state courts to affirmatively release all criminal defendants meeting the court’s seven-day test. Id. (defining a mandatory injunction as one that “orders a responsible party to take action”—not simply maintaining the status quo (simplified)). Thus, to justify this injunction, the “law and facts [must] clearly favor [Petitioners’] position”; it is “simply” not enough that they are “likely to succeed.” Id. (simplified). And we never approve mandatory injunctions in “doubtful cases.” Id. (simplified). Thus, the injunction here must meet a “doubly demanding” standard because the remedy imposed by the district court is “particularly disfavored.” See id. (simplified). The majority fails to live up to this standard in uncritically deferring to the district court’s chosen injunction. So while we normally grant some deference in reviewing preliminary injunctions, we don’t defer when the district court gets the law wrong—and we especially don’t defer when the district court orders a mandatory injunction based on an erroneous view of the law. And here, neither the Sixth Amendment nor the Fourteenth Amendment clearly justifies the district court’s sweeping, one-size-fits-all jailbreak order. 68 BETSCHART V. STATE OF OREGON A. Sixth Amendment Right to Counsel Begin with the Sixth Amendment. To refresh, the district court injunction applies to any individual who is or will be jailed in the State of Oregon. The district court concluded that those individuals suffer or will suffer a violation of their Sixth Amendment right to counsel if not appointed government-funded counsel within seven days. It then formulated a rule—Oregon must provide state-funded counsel to every detained defendant within seven days of the initial appearance (or within seven days of the withdrawal of a previously appointed attorney if the withdrawal was due to a conflict and was within the first seven days), or else release the defendant from jail. Nothing in the text nor history of the Sixth Amendment supports the seven-day rule. And the Supreme Court has been clear that the Sixth Amendment is violated only when a defendant fails to have been appointed counsel at a “critical stage” of the criminal proceedings. Thus, mandating appointment of state-funded counsel within seven days disregards the established framework for resolving Sixth Amendment questions. i. Let’s start with the basics. “In all criminal prosecutions, the accused shall enjoy the right to . . . have the Assistance of Counsel for his defence.” U.S. Const., amend. VI. Thus, by its plain text, the Sixth Amendment is concerned with counsel’s assistance for “defence” against “criminal prosecutions.” As an original matter, the Sixth Amendment right was largely understood to encompass a right to employ counsel,  BETSCHART V. STATE OF OREGON 69 not a guarantee of counsel at government expense. See Padilla v. Kentucky, 559 U.S. 356, 389 (2010) (Scalia, J., dissenting); see also Garza v. Idaho, 139 S. Ct. 738, 756 (2019) (Thomas, J., dissenting). Indeed, the text of the Amendment says nothing about government-funded counsel. Instead, there was a long history of defendant self- representation or, when necessary, ad hoc court appointment of counsel in difficult cases. W. Beaney, The Right to Counsel in American Courts 8–31, 226 (1955). It wasn’t until the 1930s that the Court suggested that a right to government-appointed counsel (rooted in the Fourteenth Amendment’s Due Process Clause, not the Sixth Amendment) might apply in capital cases, and even then, only when the defendant was unable to “mak[e] his own defense because of ignorance, feeble-mindedness, illiteracy, or the like.” Powell v. Alabama, 287 U.S. 45, 71 (1932). The Court refurbished this into a Sixth Amendment right to government-appointed and government-funded counsel in all federal criminal cases during the New Deal. Johnson v. Zerbst, 304 U.S. 458, 462–63 (1938). In 1963, the Court then incorporated the right against the States via the Fourteenth Amendment in Gideon. There, the Court reasoned that the average individual “[l]eft without the aid of counsel . . . may be put on trial without a proper charge, and convicted upon incompetent evidence, or evidence irrelevant to the issue or otherwise inadmissible.” Gideon, 372 U.S. at 345 (simplified). Thus, the right to government- funded counsel, now re-anchored in the Sixth Amendment, is incorporated as a “fundamental right” in the Fourteenth Amendment—extending it to cover state criminal defendants as well. Id. at 343. This judicial innovation ultimately required a framework for evaluating when state-funded counsel must be provided. 70 BETSCHART V. STATE OF OREGON First, the Sixth Amendment right attaches with “the initiation of adversary judicial criminal proceedings— whether by way of formal charge, preliminary hearing, indictment, information, or arraignment.” Rothgery, 554 U.S. at 198 (simplified). Second, in any “postattachment proceedings” deemed a “critical stage,” the defendant is guaranteed counsel. Id. at 212. Of course, “counsel must be appointed within a reasonable time after attachment to allow for adequate representation at any critical stage before trial, as well as at trial itself.” Id. So the dispositive question under the Sixth Amendment is whether a defendant proceeds through a “critical stage” with effective counsel. We must start then with what exactly constitutes a critical stage. They are proceedings where “the presence of [defense] counsel is necessary to preserve the defendant’s basic right to a fair trial” such that the defendant is “as much entitled to such aid (of counsel) as at the trial itself.” Wade, 388 U.S. at 227, 237 (simplified). In other words, they are “pretrial events that might appropriately be considered to be parts of the trial itself.” United States v. Ash, 413 U.S. 300, 310 (1973). But just because a hearing is important in some larger sense does not render it a critical stage. Instead, “[t]he Court has identified as ‘critical stages’ those pretrial procedures that would impair defense on the merits if the accused is required to proceed without counsel.” Gerstein, 420 U.S. at 122. What matters is “defense at trial, not defense in relation to other objectives that may be important to the accused.” Rothgery, 554 U.S. at 216 (Alito, J., concurring); see also id. at 217 (explaining that critical stages consist of “certain pretrial events [that] may so prejudice the outcome of the defendant’s prosecution that, as a practical matter, the defendant must be represented at those  BETSCHART V. STATE OF OREGON 71 events in order to enjoy genuinely effective assistance at trial” (emphasis added)). Time and again, the Supreme Court’s critical-stage analysis has centered on a proceeding’s impact on the case’s resolution—conviction and sentence—not on collateral issues unrelated to the defense against the merits of the prosecution. For example, four years after Gideon, the Court held that a pretrial lineup constitutes a critical stage because the “results might well settle the accused’s fate and reduce the trial itself to a mere formality.” Wade, 388 U.S. at 224. Then, a few years after that, the Court determined that a preliminary hearing was a critical stage because “the guiding hand of counsel at the preliminary hearing is essential to protect the indigent accused against an erroneous or improper prosecution.” Coleman v. Alabama, 399 U.S. 1, 9 (1970). More recently, when the Supreme Court has identified new critical stages, its focus remains on whether the stage affects the prosecution’s merits. Take the ruling that a plea negotiation is a critical stage. Missouri v. Frye, 566 U.S. 134, 143–44 (2012). When the Court made that determination, it again focused on how the plea negotiation affects the outcome of a defendant’s case. Id. The Court reasoned that nearly all federal and state convictions “are the result of guilty pleas,” so “the negotiation of a plea bargain, rather than the unfolding of a trial, is almost always the critical point for a defendant.” Id. Because the plea bargain is “so central to the administration of the criminal justice system” and the ultimate result of the prosecution, the Court recognized it as a critical stage. Id. at 143. Even Lafler v. Cooper, which the majority cites extensively, was resolved based on the prejudice to the outcome of the defendant’s prosecution. 566 U.S. 156, 165–66 (2012) (reasoning that 72 BETSCHART V. STATE OF OREGON plea negotiations are a critical stage since “the trial [may not] cure[] the particular error at issue . . . the defendant who goes to trial instead of taking a more favorable plea may be prejudiced from either a conviction on more serious counts or the imposition of a more severe sentence”). On the other hand, the Court has considered collateral considerations—of the kind not concerned with defense on the merits—insufficient to render an event a critical stage. Look at Gerstein. There, the Court explained that a pretrial hearing “addressed only to pretrial custody” did not constitute a critical stage. Gerstein, 420 U.S. at 123. While pretrial custody may impact the defendant, it is not considered “critical” because it doesn’t “substantial[ly] harm” “the defendant’s ability to assist in preparation of his defense.” Id. at 122–23 (simplified). That an event will dramatically affect the defendant isn’t enough to make the event a critical stage. In evaluating these concerns, the Ninth Circuit considers three factors: whether “(1) failure to pursue strategies or remedies results in a loss of significant rights, (2) skilled counsel would be useful in helping the accused understand the legal confrontation, and (3) the proceeding tests the merits of the accused’s case.” Hovey v. Ayers, 458 F.3d 892, 901 (9th Cir. 2006) (simplified). In Hovey, we held that a proceeding didn’t meet these critical-stage factors because there were (1) no “risk of permanent deprivation of any significant rights,” (2) no “complex legal problems,” and (3) no “test[ing] the merits of [the defendant’s] case.” Id. at 902. Like the Supreme Court, we’ve rejected many other pretrial hearings as “critical stages” over the years. See, e.g., United States v. Benford, 574 F.3d 1228, 1232–33 (9th Cir. 2009) (pretrial status conference); Hovey, 458 F.3d at 901– 02 (attorney competency hearing); McNeal v. Adams, 623  BETSCHART V. STATE OF OREGON 73 F.3d 1283, 1288–89 (9th Cir. 2010) (hearing on motion to compel DNA sample). ii. Under this framework, the Sixth Amendment right is concerned with adequate representation at critical stages. Whether a defendant is unrepresented during periods of the pretrial process, even prolonged periods, is not the dispositive question. Instead, the right is more nuanced, focusing on “certain steps before trial” that are seen as critical. Frye, 566 U.S. at 140 (emphasis added). Such an individualized assessment is not susceptible to blanket, brightline rules. But here, the district court creates, and the majority endorses, a brightline rule that the Sixth Amendment right to counsel is violated by a seven-day gap without government- funded representation. The district court crafted this blanket rule based on the view that (1) bail hearings, which must be held within five days of the initial appearance, are a critical stage, and (2) counsel must have time to prepare for trial within 60 days. Neither ground justifies the injunction here. Bail Hearings Bail hearings are not a critical stage because they are not “pretrial events that might appropriately be considered to be parts of the trial itself.” Ash, 413 U.S. at 310. To start, the Supreme Court has never said that bail hearings are critical stages. In fact, it has suggested the opposite in Gerstein. Because there wasn’t a high probability that pretrial detention would impair a defendant’s ability to prepare his defense, the Court said a hearing “addressed only to pretrial custody” was not a critical stage. Gerstein, 420 U.S. at 122– 23. 74 BETSCHART V. STATE OF OREGON And the district court was wrong to focus on a passing line from Coleman to suggest that the Court treats bail hearings as a critical stage. In that case, the Court said that a counsel could be useful at a “preliminary hearing” to “mak[e] effective arguments for the accused on such matters as the necessity for an early psychiatric examination or bail.” Coleman, 399 U.S. at 9. But the preliminary hearing involved multiple merits-based considerations, including “whether there is sufficient evidence against the accused to warrant presenting his case to the grand jury and, if so, to fix bail if the offense is bailable.” Id. at 8. Indeed, the Court focused on how lawyers may assist on merits issues at a preliminary hearing, like (1) “expos[ing] fatal weaknesses in the State’s case,” (2) “fashion[ing] a vital impeachment tool for use in cross-examination of the State’s witnesses at the trial,” (3) “preserv[ing] testimony favorable to the accused,” and (4) “prepar[ing] a proper defense to meet that case at the trial.” Id. at 9. It was in this context that the Court mentioned psychiatric examinations and bail—almost as an afterthought. But the Court has said that a proceeding “addressed only to pretrial custody,” Gerstein, 420 U.S. at 123 (emphasis added)—like bail hearings—is not a critical stage. The Ninth Circuit’s three-factor test confirms this conclusion. For the first factor, we’ve said that a proceeding is not a critical stage if there’s no “risk of permanent deprivation of any significant rights during the hearing.” Hovey, 458 F.3d at 902. Here, “[n]othing prevents” Petitioners from revisiting their pretrial detention status “at any point after the” bail hearing. See id. Oregon law doesn’t forbid a new bail determination once counsel is appointed. Indeed, Oregon courts appear to regularly entertain renewed motions of release. See, e.g., State v. McDowell, 279 P.3d  BETSCHART V. STATE OF OREGON 75 198, 200 (Or. 2012) (ordering trial court to grant defendant’s motion for release); In re Application of Haynes, 619 P.2d 632, 635 (Or. 1980) (reviewing trial court’s multiple denials of motions of release decided on the merits). So, the first factor does not support treating the hearing as a critical stage. So too for the skilled-counsel factor. This factor fails to establish a critical stage when the “hearing d[oes] not involve a confrontation at which an attorney would be needed to help [the defendant] cope with complex legal problems,” when a defendant’s “interests [are not] subjected to a ‘critical confrontation,’” or when there’s no “power ‘imbalance’ in the face of the state’s prosecuting authority.” Hovey, 458 F.3d at 902 (simplified). Under Oregon law, the initial release decision is usually made at the initial appearance when defendants are represented by counsel— not at bail hearings. See Or. Rev. Stat. § 135.245(2)(a). Bail hearings only come into play after a magistrate first determines that “good cause” supports postponing the detention decision. Id. And then, the bail hearing must be held within five days. Id. § 135.245(7)(a). State law provides that defendants charged with only a few offenses— murder, aggravated murder, treason, or a violent felony—are allowed to be detained. Id. § 135.240. While a defendant may present evidence, the bail hearing “may not be used for purposes of discovery.” Id. § 135.240(4)(d). At the bail hearing, the magistrate considers only limited information to determine whether release is appropriate. See id. § 135.230(7). While there may be good reason to have government-funded counsel at bail hearings, they do not present the kind of complex legal issues that would implicate this factor. Cf. Gerstein, 420 U.S. at 121 (reasoning that probable-cause bail hearings “do[] not require the fine resolution of conflicting evidence that a reasonable-doubt or 76 BETSCHART V. STATE OF OREGON even a preponderance standard demands, and credibility determinations are seldom crucial”). Finally, bail hearings do not test the merits of the Petitioners’ case. To test the merits, “[c]ritical stages [must] involve ‘significant consequences’ to the defendant’s case.” See McNeal, 623 F.3d at 1288 (quoting Bell v. Cone, 535 U.S. 685, 695–96 (2002)). The events that qualify “will determine whether a criminal conviction is possible,” see United States v. Bohn, 890 F.2d 1079, 1081 (9th Cir. 1989), or the terms of the eventual sentence, see, e.g., United States v. Leonti, 326 F.3d 1111, 1117 (9th Cir. 2003) (holding that the cooperation period for a plea bargain was a critical stage because of the “profound effect a substantial assistance motion can have on a defendant’s sentence”). A bail hearing is far from this kind of merits inquiry, focusing instead on mere releasability. No motions for dismissal. No inquiry into a privilege. No suppression of evidence. No profound effect on one’s trial or sentence. All told, under our precedent, nothing supports viewing bail hearings as critical stages in this expedited litigation. Other precedent supports this view. See Fenner v. State, 381 Md. 1, 24 (2004) (bail review hearing is not a critical stage); Padgett v. State, 590 P.2d 432, 436 (Alaska 1979) (“The setting of bail is likewise not an adversary confrontation wherein potential substantial prejudice to the defendant’s basic right to a fair trial inheres, but rather is limited to the issue of interim confinement.” (simplified)). And nothing explains why the district court chose seven days from initial appearance as the trigger point when the bail hearing must be held within five days of that appearance. The only contrary evidence the majority could muster is a single line of dicta from a single out-of-circuit opinion. See  BETSCHART V. STATE OF OREGON 77 Higazy v. Templeton, 505 F.3d 161, 172 (2d Cir. 2007). In that case, the Second Circuit held that a bail hearing implicates a defendant’s Fifth Amendment right against self- incrimination. Id. at 170. As additional “support[ for] the conclusion,” the Second Circuit observed in passing that the bail hearing is “part of a criminal case” and “the Supreme Court found that a bail hearing is a ‘critical stage of the State’s criminal process.’” Id. at 172 (quoting Coleman, 399 U.S. at 10). But, as explained above, Coleman was not about bail hearings. It was about a “preliminary hearing,” which encompasses much more than determining bail. See Coleman, 399 U.S. at 8, 10 (holding that a “preliminary hearing is a ‘critical stage’ of the State’s criminal process” which includes “whether there is sufficient evidence against the accused to warrant presenting his case to the grand jury and, if so, to fix bail if the offense is bailable”). So the Second Circuit’s Fifth Amendment ruling offers little support for the majority’s Sixth Amendment conclusion. And to my knowledge, no other circuit decision supports the majority’s novel ruling that bail hearings are a critical stage under the right to counsel. Preparation for Trial and the Progression Through Critical Stages The district court also justified its rule by reasoning that state law provides an unqualified right to trial in 60 days for defendants in custody. See Or. Rev. Stat. § 136.290. Thus, the district court concluded that counsel must be immediately appointed to allow for adequate preparation before that date. The majority adopts a different theory—it states that the lack of state-appointed counsel within seven days would “interfere[]” with the “progression to critical stages by delaying those stages” and by “prevent[ing] any meaningful advocacy.” Maj. Op. 22. While preparation for, 78 BETSCHART V. STATE OF OREGON and progression through, critical stages is important, the Sixth Amendment doesn’t support a blanket seven-day rule. These rationales are wrong for several reasons. First, the district court’s state-law analysis is inaccurate. As Oregon points out, the 60-day statutory scheme applies only to select defendants and is inapplicable to many crimes. Or. Rev. Stat. § 136.295(1) (requirement does not apply to many violent felony cases). Nor is it unqualified; it may be extended for good cause, id. § 136.295(4), including when defense counsel is recently appointed or would have trouble preparing for trial within the deadline, id. § 136.295(4)(b)(C), (D). Finally, if the deadline arrives, the remedy is statutorily provided for: release from pretrial detention. Id. § 136.290(2). So the district court’s focus on 60 days to generate a seven-day deadline makes little sense. Second, both the district court and majority wrongly establish a bright-line rule that critical stages must quickly follow the attachment of the Sixth Amendment right. The Supreme Court has explained it is an “analytical mistake [to] assum[e] that attachment necessarily requires the occurrence or imminence of a critical stage.” Rothgery, 554 U.S. at 212. Instead, determining whether a critical stage is reached must be made case-by-case. See, e.g., Benford, 574 F.3d at 1233 (“We limit our holding to what happened (and what did not happen) in this case.”); Hovey, 458 F.3d at 901 (“Based on the specific facts of this case, we conclude that the [hearing was not a critical stage].”). Criminal prosecutions do not proceed in a one-size-fits- all fashion. While I agree with the majority that the Sixth Amendment is not “a haphazard jack-in-the-box,” Maj. Op. 25, neither is it a rigid cookie cutter—invoked by a mechanical calculation of dates. Some cases may proceed  BETSCHART V. STATE OF OREGON 79 slowly. In those cases—where critical stages may not occur until later in the proceedings—the seven-day rule is disconnected from a Sixth Amendment violation. Other cases proceed quite quickly. In those cases, it’s easy to see how a critical stage could occur shortly after attachment. But even in those cases, nothing in the record supports the requirement of appointed counsel within seven days. While attorney preparation for the critical stages is required, see Rothgery, 554 U.S. at 212, it’s a mistake to assume that preparation must start within seven days in every case. If a delay in appointment does result in a critical stage without effective counsel, the Sixth Amendment provides for the remedy—vacatur of the conviction, a redo of the critical stage, or suppression of any evidence obtained. For example, the majority correctly lists several important duties counsel must undertake before trial, like investigating defenses and ensuring the defendant is competent to stand trial. But if counsel does not have adequate time to complete those tasks, those interests may be vindicated either before trial, by redoing the critical stage, or after trial, through vacatur of any conviction. Third, the majority’s belief that any delay in the “progression to critical stages” violates the Sixth Amendment puts us into uncharted constitutional territory, as the majority acknowledges. The majority blames Oregon for this unprecedented situation. But while the widespread delay in appointing counsel is extremely troubling, the Sixth Amendment is an individual right. By altering the Sixth Amendment analysis because of the large number of Petitioners involved, the majority transforms the right into a collective one. 80 BETSCHART V. STATE OF OREGON Fourth, other constitutional and statutory grounds are more focused on preventing delays in prosecutions, such as the speedy-trial right. See Doggett v. United States, 505 U.S. 647, 651 (1992) (explaining the multi-factor test for evaluating whether delay between accusation and case resolution is unconstitutional). It is that Speedy Trial Clause which “[r]eflect[s] the concern that a presumptively innocent person should not languish under an unresolved charge,” Betterman v. Montana, 578 U.S. 437, 443 (2016), not the right to counsel. As I’ve said elsewhere, “the text and history of the Speedy Trial Clause establish an enduring principle”: “[a]t its core,” the right “ensures that defendants are not locked up in jail indefinitely pending trial.” United States v. Olsen, 21 F.4th 1036, 1058 (9th Cir. 2022) (Bumatay, J., concurring in the denial of rehearing en banc). Plus, the majority forgets that Oregon law expressly accounts for this concern—ordering the release of any defendant from custody if trial does not commence within 60 days after the time of arrest. Or. Rev. St. § 136.290. And so the majority raising the specter of “indefinite detention without counsel” is textbook straw-man alarmism—a position argued by no one and detached from the realities of our criminal-justice system. See Maj. Op. 26–27. Fifth, neither the Supreme Court nor the Ninth Circuit has tackled the difficult task of setting a brightline rule for when the Sixth Amendment’s right to counsel is violated under a delay theory. See Rothgery, 554 U.S. at 213 (“We do not decide whether the 6–month delay in appointment of counsel resulted in prejudice to Rothgery’s Sixth Amendment rights, and have no occasion to consider what standards should apply in deciding this.”); Farrow v. Lipetzky, 637 F. App’x 986, 988 (9th Cir. 2016) (unpublished) (remanding to resolve “how soon after the  BETSCHART V. STATE OF OREGON 81 Sixth Amendment right attaches must counsel be appointed, and at what point does delay become constitutionally significant?”). But in one fell swoop, the majority devises a seven-day rule—on the shakiest of foundations. Without developing any constitutional standards, the majority determines—for every State and federal district in the Ninth Circuit—that seven days may be set as the outer bound for the appointment of counsel. What evidence does the majority rely on to make this determination? Not much. While seven days may have Biblical significance, it doesn’t have obvious constitutional relevance. The majority doesn’t justify its holding based on constitutional text or history. It doesn’t support its holding based on any statistics or other objective measures of criminal proceedings. And the majority makes this blanket rule without considering the varied resources, caseloads, and practices of the jurisdictions within the Ninth Circuit. While the majority proclaims it is only deferring to the district court’s seven-day rule and not adopting one itself, because the seven-day deadline is justified by both the thinnest record and the broadest Sixth Amendment principles, the majority’s rationale will apply in every case. Thus, the majority can’t ignore that its seven- day rule will effectively become the law of the land in the Ninth Circuit. Sixth, the injunction is both overinclusive and underinclusive of its Sixth Amendment rationale. As stated above, the injunction inexplicably leaves out those charged with aggravated murder and murder. But that’s not all. Notice that, in the district court’s amended order, only a Petitioner whose prior counsel has withdrawn within seven days of the initial appearance is eligible for release from jail. Under those terms, if a Petitioner’s prior appointed counsel withdraws on the eighth day, then the Petitioner may not be 82 BETSCHART V. STATE OF OREGON released over the failure to re-appoint counsel. But if any delay in “progression to critical stages” is a violation of the Sixth Amendment, it makes little sense to deny relief to Petitioners whose counsel withdraws later in the criminal proceedings—when it is more likely that a critical stage occurs. Thus, the injunction draws arbitrary lines—the hallmark of an abuse of discretion. *** No one questions how problematic the situation is in Oregon. The Sixth Amendment guarantees Oregon defendants a right to appointed counsel. And the delays in appointments raised by Petitioners may very well lead to violation of the Sixth Amendment at some point. But we are not empowered to jettison Sixth Amendment precedent, dispense with the critical-stage analysis, and fashion a blanket remedy out of thin air. And there’s simply no constitutional basis for the arbitrary choice of seven days. Given the shifting rationales for the rule and its haphazard application, it’s hard to avoid the conclusion that we are just making it up as we go along. In the normal course, we would carefully consider whether a critical stage has occurred in an individual case and, if so, whether effective counsel was available. Only then would we begin to think of appropriate remedies. B. Fourteenth Amendment Due Process Nor does the Fourteenth Amendment’s Due Process Clause justify the injunction’s jailbreak solution. According to the district court, Petitioners’ substantive due process rights are violated because they are detained pretrial without the appointment of counsel. It ruled that  BETSCHART V. STATE OF OREGON 83 Oregon disregards the “reliable process” guaranteed by the Fourteenth Amendment by requiring indigent defendants to proceed “without counsel while incarcerated.” It justified its ruling based on substantive due process cases in United States v. Salerno, 481 U.S. 739, 755 (1987), and Lopez- Valenzuela v. Arpaio, 770 F.3d 772, 779 (9th Cir. 2014) (en banc). No court has extended substantive due process to the reaches that the district court would. In Salerno, the Court rejected the view that a bail law “violates substantive due process because the pretrial detention it authorizes constitutes impermissible punishment before trial.” 481 U.S. at 746. Any due process concern related to pretrial detention in that case was alleviated by the arrestee’s right to “a prompt detention hearing” and by “the maximum length of pretrial detention” under federal speedy-trial protections. Id. at 747. Here, given that Petitioners were represented by counsel at arraignment and are protected by state and constitutional speedy-trial rights, Salerno shows that substantive due process isn’t implicated. Lopez-Valenzuela is similarly divorced from this case. There, Arizona categorically banned pretrial release for undocumented immigrants arrested for a wide range of felony offenses. 770 F.3d at 775. We held that such a regime violated immigrants’ substantive due process rights because the law was not limited to only “extremely serious offenses” and arrestees were not afforded “an individualized determination of flight risk or dangerousness.” Id. at 788, 791 (simplified). Once again, a delay in the appointment of state-funded counsel is nothing like a categorical detention law. As mentioned, Petitioners each received an individualized assessment at arraignment when they were represented by counsel. 84 BETSCHART V. STATE OF OREGON Contrary to the district court’s ruling, the Supreme Court has said that due process affords lesser protections than our modern Sixth Amendment jurisprudence when it comes to the assistance of counsel. Due process only “prohibits the conviction and incarceration of one whose trial is offensive to the common and fundamental ideas of fairness and right.” Betts v. Brady, 316 U.S. 455, 473 (1942). Indeed, “while want of counsel in a particular case may result in a conviction lacking in such fundamental fairness, we cannot say that the [Fourteenth A]mendment embodies an inexorable command that no trial for any offense, or in any court, can be fairly conducted and justice accorded a defendant who is not represented by counsel.” Id. While a fundamentally unfair conviction or denial of access to court offends due process, the Fourteenth Amendment has little to say about a delay in the appointment of state-funded counsel. And nothing supports a blanket seven-day rule under the Due Process Clause. So the Fourteenth Amendment does not save the injunction. V. Balance of Interests Finally, no injunction should have been issued because the balance of interests doesn’t support the immediate release of criminal defendants when other remedies are potentially available. See Porretti v. Dzurenda, 11 F.4th 1037, 1050 (9th Cir. 2021) (when the government opposes a preliminary injunction, the courts must consider “balance of equities and public interest” together).  BETSCHART V. STATE OF OREGON 85 A. As a general principle, “courts . . . should pay particular regard for the public consequences in employing the extraordinary remedy of injunction.” Weinberger v. Romero-Barcelo, 456 U.S. 305, 312 (1982) (simplified). Indeed, “[w]e will not grant a preliminary injunction . . . unless those public interests outweigh other public interests that cut in favor of not issuing the injunction.” All. for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1138 (9th Cir. 2011). Critically, “[a]n injunction must be narrowly tailored to remedy the specific harm shown.” E. Bay Sanctuary Covenant v. Barr, 934 F.3d 1026, 1029 (9th Cir. 2019) (simplified). It is true that courts must “not shrink from [their] duty to safeguard th[e] rights” guaranteed by the Constitution, Tandon v. Newsom, 992 F.3d 916, 939 (9th Cir. 2021) (Bumatay, J., dissenting in part), and that “it is always in the public interest to prevent the violation of a party’s constitutional rights,” Melendres v. Arpaio, 695 F.3d 990, 1002 (9th Cir. 2012) (simplified). But assuming a constitutional violation, the district court’s remedy doesn’t even rectify the alleged injury. After being released into Oregon’s communities, no formerly detained criminal defendant will have been appointed counsel. So the injunction fails to vindicate the harms to Petitioners while ignoring the risks to the public. See, e.g., Brown v. Plata, 563 U.S. 493, 577 (2011) (Alito, J., dissenting) (describing the rearrest of thousands of prisoners for committing new crimes after a court ordered a cap on the number of inmates in the Philadelphia prison system). Rather than acknowledge the problems with the injunction here, the majority categorically dismisses any 86 BETSCHART V. STATE OF OREGON safety concerns as an unsupported “fear-mongering parade of horribles.” Maj. Op. 36. But it is the majority that ignores the record in this regard. First, recall that only Oregon defendants who present a “danger of physical injury or sexual victimization” by clear and convincing evidence may be detained in the first place. Or. Rev. St. § 135.240. Second, consider just the ten named Petitioners here that will be immediately released into Oregon communities: First there is Petitioner Richard Owens, Jr., who has been convicted of two felonies—one for prior assault with a firearm. Mr. Owens’s current detention stems from a June 2023 incident, when he allegedly sped his vehicle down the road and in front of an eight-year-old’s birthday party. When victims yelled at him to slow down, he got out of his car, pulled out a gun, told the victims to “[f]uck around and find out,” and fired the gun into the air after speeding off. Next is Petitioner Tyrik Dawkins, who has two prior drug-trafficking felony convictions, a contempt-of-court conviction from Pennsylvania, a prior domestic-violence arrest from Washington, and at least three restraining orders filed since 2020 by women whom he allegedly physically abused, sexually assaulted, or threatened to murder. What brings Mr. Dawkins to the Washington County jail? Four counts of rape in the first degree, four counts of sexual abuse in the first degree, and two counts of kidnapping in the first degree. This is aside from Mr. Dawkins’s open 2021 rape investigation in Multnomah County, Oregon, for allegedly locking a victim in his hotel room, anally and orally sodomizing her for several hours, and threatening her with a firearm. We have also Petitioner Leon Polaski, who is accused of strangling his girlfriend during an argument and then fleeing  BETSCHART V. STATE OF OREGON 87 Oregon to avoid prosecution; Petitioner Joshua James- Richards, who allegedly assaulted a police officer and who had already missed mandatory check-ins with Oregon’s pretrial-release services; and lead Petitioner Walter Betschart, who was arrested for violating the terms of his previous release agreement and for violating his stalking order against his neighbor. Next to these defendants, Petitioner Timothy Wilson’s two counts of public indecency seem banal. And these are just the ten Petitioners who originated this lawsuit; it says nothing of the other 100 defendants who will also be released with the majority’s order or the countless others who will be released on an ongoing basis. B. Even more serious, the district court failed to consider alternatives that were less drastic than simply letting all criminal defendants out of jail. As the district court conceded, its fashioned injunction was a “blunt instrument” and “somewhat arbitrary.” That alone is an abuse of discretion. And less-restrictive alternatives appear readily available. Take one remedy discussed at oral argument—a court order requiring each criminal defendant to have a new, counseled bail hearing. Such a remedy would have addressed the district court’s belief that the bail hearing was a critical stage requiring appointed counsel without going further than necessary to resolve the issue. When asked about the viability of this remedy, Petitioners’ counsel agreed that it was “certainly one way that the district court could have structured its injunction.” 88 BETSCHART V. STATE OF OREGON Yet another alternative was discussed at oral argument— directing the State to reconsider the limit on the number of criminal cases a public defense attorney can handle. Again, Petitioners’ counsel was admirably honest with this potential solution: “My understanding of the crisis is that it was kicked off . . . by a change in the contracting system, and so [ordering state public defenders to take more cases] may solve the problem.” According to the majority, the district court apparently considered compelling members of the bar to represent indigent criminal defendants. But the district court rejected this option, as the majority concedes, because it feared that some lawyers might find it “kind of insulting.” But see Supreme Court of N.H. v. Piper, 470 U.S. 274, 287 (1985) (noting that members of the bar “could be required to represent indigents”); Powell, 287 U.S. at 73 (“Attorneys are officers of the court, and are bound to render service when required by such an appointment [by a trial court].”). Even so, the fear of insulting lawyers pales in comparison to the burdens on the people of Oregon imposed by the immediate release of dozens of criminal defendants. And the majority’s anecdote of one bad experience with one attorney doesn’t justify acceding to this far-reaching injunction. And finally, there’s the majority’s own proposed remedy, which it claims would fix the problem “overnight.” Maj. Op. 39. To the majority, the problem is “simply” a matter of Oregon “paying appointed counsel a better wage.” Id. If so, this would actually solve the lack of counsel without resorting to a judicial jailbreak. Ironically, the majority finally understands that its own solution would be an “extraordinary idea,” id. at 38, yet it continues to call for the jailbreak solution, which pushes the envelope even more.  BETSCHART V. STATE OF OREGON 89 Unlike the district court’s chosen remedy, all these alternatives would have more effectively and less restrictively remedied the alleged right-to-appointed- counsel violation. I do not opine on whether these alternatives are properly within the district court’s authority; I raise them only to show the how ill-considered it was to blindly defer to the injunction here. *** Given these issues, the balance of interests strongly disfavors this injunction. VI. As is obvious, the problems with the preliminary injunction here are significant. The majority brushes away these concerns—not by refuting them, but by claiming that they are merely “an ode to classic judicial overreach.” Maj. Op. 39. While this dissent may raise difficult questions and it may be easier to skirt them, it is our duty to confront them. After all, we always have a duty to ensure that a district court has authority to order an injunction. We always have a duty to respect federalism and not unduly interfere with state proceedings. And we always have a duty to follow Supreme Court precedent. The majority ignores these concerns even though the injunction here doesn’t even remedy the alleged constitutional violation. Under a proper understanding of the judicial role, we should have paused and thought through these issues before unleashing a sweeping and dangerous order on the people of Oregon. The public and the rule of law deserve better.