J.D. v. Azar, No. 18-5093 (D.C. Cir. 2019)

Annotate this Case
Justia Opinion Summary

Plaintiffs have shown a likelihood of success in connection with their claim that ORR's restriction on abortion access infringes their protected right to choose to terminate their pregnancies. In 2017, the government instituted a policy effectively barring any unaccompanied alien child in the custody of the Office of Refugee Resettlement (ORR) from obtaining a pre-viability abortion. The district court granted a preliminary injunction and the government appealed.

Agreeing that the case was not moot, the DC Circuit affirmed and held that the district court did not abuse its discretion in certifying a class consisting of pregnant unaccompanied minors in the government's custody. On the merits, the court held that, under binding Supreme Court precedent, a person has a constitutional right to terminate her pregnancy before fetal viability, and the government cannot unduly burden her decision. Consequently, these controlling principles dictate affirming the district court's preliminary injunction against the government's blanket denial of access to abortion for unaccompanied minors.

The court vacated in part and remanded to the extent that the preliminary injunction barred disclosure to parents and others of unaccompanied minors' pregnancies and abortion decisions. The court held that this portion of the preliminary injunction warranted further explication to aid appellate review.

Download PDF
United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued September 26, 2018 Decided June 14, 2019 No. 18-5093 J.D., ON BEHALF OF HERSELF AND OTHERS SIMILARLY SITUATED, ET AL., APPELLEES v. ALEX MICHAEL AZAR, II, SECRETARY, HEALTH AND HUMAN SERVICES, ET AL., APPELLANTS Appeal from the United States District Court for the District of Columbia (No. 1:17-cv-02122) August E. Flentje, Special Counsel, U.S. Department of Justice, argued the cause for appellants. With him on the brief were Hashim M. Mooppan, Deputy Assistant Attorney General, and Michael C. Heyse, Attorney. Ken Paxton, Attorney General, Office of the Attorney General for the State of Texas, Scott A. Keller, Solicitor General, Kyle Hawkins, Assistant Solicitor General, David J. Hacker, Special Counsel for Civil Litigation, Leslie Rutledge, Attorney General, Office of the Attorney General for the State of Alabama, M. Stephen Pitt, General Counsel for the 2 Governor of Kentucky, Jeff Landry, Attorney General, Office of the Attorney General for the State of Louisiana, Eric Schmitt, Attorney General, Office of the Attorney General for the State of Missouri, Doug Peterson, Attorney General, Office of the Attorney General for the State of Nebraska, Dave Yost, Attorney General, Office of the Attorney General for the State of Ohio, Mike Hunter, Attorney General, Office of the Attorney General for the State of Oklahoma, Alan Wilson, Attorney General, Office of the Attorney General for the State of South Carolina, and Patrick Morrisey, Attorney General, Office of the Attorney General for the State of West Virginia, were on the brief as amici curiae States of Texas, et al. in support of appellants. Brigitte Amiri argued the cause for appellees. With her on the brief were Meagan Burrows, Jennifer Dalven, Arthur B. Spitzer, Scott Michelman, Daniel Mach, and Melissa Goodman. Barbara D. Underwood, Solicitor General, Office of the Attorney General for the State of New York, Anisha S. Dasgupta, Deputy Solicitor General, Ester Murdukhayeva, Assistant Solicitor General, Brian E. Frosh, Attorney General, Office of the Attorney General for the State of Maryland, Maura Healey, Attorney General, Office of the Attorney General for the Commonwealth of Massachusetts, Gurbir S. Grewal, Attorney General, Office of the Attorney General for the State of New Jersey, Hector Balderas, Attorney General, Office of the Attorney General for the State of New Mexico, Joshua H. Stein, Attorney General, Office of the Attorney General for the State of North Carolina, Ellen F. Rosenblum, Attorney General, Office of the Attorney General for the State of Oregon, Josh Shapiro, Attorney General, Office of the Attorney General for the Commonwealth of Pennsylvania, Xavier Becerra, Attorney General, Office of the Attorney 3 General for the State of California, William Tong, Attorney General, Office of the Attorney General for the State of Connecticut, Kathy Jennings, Attorney General, Office of the Attorney General for the State of Delaware, Russell A. Suzuki, Attorney General, Office of the Attorney General for the State of Hawaii, Kwame Raoul, Attorney General, Office of the Attorney General for the State of Illinois, Thomas J. Miller, Attorney General, Office of the Attorney General for the State of Iowa, Aaron Frey, Attorney General, Office of the Attorney General for the State of Maine, Robert W. Ferguson, Attorney General, Office of the Attorney General for the State of Washington, Karl A. Racine, Attorney General, Office of the Attorney General for the District of Columbia, Thomas J. Donovan, Jr., Attorney General, Office of the Attorney General for the State of Vermont, and Mark R. Herring, Attorney General, Office of the Attorney General for the Commonwealth of Virginia, were on the brief for amici curiae States of New York, et al. in support of appellees. Jennifer R. Cowan was on the brief for amici curiae The American College of Obstetricians and Gynecologists, et al. in support of plaintiffs-appellees. Joel Dodge and Jane Liu were on the brief for amici curiae Reproductive Rights, Health, and Justice Organizations and Allied Organizations in support of appellees. Roxann E. Henry was on the brief for amici curiae Immigrants Rights Advocates supporting plaintiffs-appellees. 4 Before: SRINIVASAN and WILKINS, Circuit Judges, and SILBERMAN, Senior Circuit Judge. Opinion for the Court filed PER CURIAM. Dissenting opinion filed by Senior Circuit Judge SILBERMAN. PER CURIAM: Among the scores of persons who come to the United States each year without lawful immigration status, several thousand are “unaccompanied alien children.” Unaccompanied alien children have no parent or legal guardian in the United States to care for them. They are thus committed to the custody of the federal government. At some point, an unaccompanied minor might be released to an approved sponsor (usually a relative) pending determination of her entitlement to stay in the United States. If no suitable sponsor exists, an unaccompanied minor might remain in the government’s custody for an extended period. Certain unaccompanied alien children are pregnant when they arrive in federal custody, after what is often a hazardous journey. Though many carry their pregnancies to term, some desire to terminate their pregnancies. But in 2017, the government instituted a policy effectively barring any unaccompanied alien child in its custody from obtaining a pre-viability abortion. This case concerns the constitutionality of that new policy. The policy functions as an across-the-board ban on access to abortion. It does not matter if an unaccompanied minor meets all the requirements to obtain an abortion under the law of the state where she is held—including, for instance, demonstrating she is mature enough to decide on her own whether to terminate her pregnancy. Nor does it matter if she secures her own funding and transportation for the procedure. 5 It does not even matter if her pregnancy results from rape. Regardless, the government denies her access to an abortion. And the government’s newfound ban applies only to pregnant minors: anyone aged 18 (or older) in immigration custody is allowed to terminate her pregnancy. Minors alone, that is, must carry their pregnancies to term against their wishes. The claim of one minor in this case brings the policy’s breadth and operation into stark relief. She had been raped in her country of origin. After her arrival here and her placement in government custody, she learned she was pregnant as a result of the rape. She repeatedly asked to obtain a pre-viability abortion, to no avail. She remained in government custody as an unaccompanied minor because there was no suitable sponsor to whom she could be released. Nor was there any viable prospect of her returning to her country of origin: indeed, she eventually received a grant of asylum (and lawful status here) due to her well-founded fear of persecution in her country of origin. Still, the government sought to compel this minor to carry her rape-induced pregnancy to term. She is one of the named plaintiffs who brought this challenge to the government’s policy on behalf of a class of pregnant unaccompanied minors. The district court granted a preliminary injunction in favor of the plaintiffs, and the government now appeals. We initially agree with the district court that the case is not moot, and we find no abuse of discretion in the court’s certification of a plaintiffs’ class consisting of pregnant unaccompanied minors in the government’s custody. On the merits, we sustain the district court’s preliminary injunction in principal part. Under binding Supreme Court precedent, a person has a constitutional right to terminate her pregnancy before fetal viability, and the government cannot unduly burden her 6 decision. The government accepts the applicability of that settled framework to unaccompanied alien children in its custody. Those controlling principles dictate affirming the district court’s preliminary injunction against the government’s blanket denial of access to abortion for unaccompanied minors. We are unanimous in rejecting the government’s position that its denial of abortion access can be squared with Supreme Court precedent. We vacate and remand, though, a separate aspect of the district court’s preliminary injunction, which bars disclosure to parents and others of unaccompanied minors’ pregnancies and abortion decisions. That portion of the preliminary injunction, we conclude, warrants further explication to aid appellate review. I. A. Unaccompanied alien children (UACs) are minors in the United States with no lawful immigration status and no parents or legal guardians in the country able to care for them. See 6 U.S.C. § 279(g). According to the government’s published information about UACs, “[u]naccompanied alien children have multiple inter-related reasons for undertaking the difficult journey of traveling to the United States, which may include rejoining family already in the United States, escaping violent communities or abusive family relationships in their home country, or finding work to support their families in the home country.” U.S. Dep’t of Health and Human Servs., Office of Refugee Resettlement, About Unaccompanied Alien Children’s Services (June 15, 2018), https://www.acf.hhs.gov/ orr/programs/ucs/about (“ORR, UAC Services”). The “age of these individuals, their separation from parents and relatives, and the hazardous journey they take make unaccompanied 7 alien children especially vulnerable to human trafficking, exploitation[,] and abuse.” Id. The Office of Refugee Resettlement (ORR), a program in the Department of Health and Human Services, bears responsibility for the “care and placement” of UACs. 6 U.S.C. § 279(b)(1)(A). Most UACs are referred to ORR by the Department of Homeland Security (DHS) after having been apprehended by immigration authorities at the border. See U.S. Dep’t of Health & Human Servs., Office of Refugee Resettlement, Unaccompanied Alien Children Program Fact Sheet 1–2 (March 2019), https://www.hhs.gov/sites/default/files/UnaccompaniedAlien-Children-Program-Fact-Sheet.pdf (“ORR, UAC Fact Sheet”). Some unaccompanied minors who hail from countries contiguous with the United States may be immediately repatriated to their countries of origin by DHS. See 8 U.S.C. § 1232(a)(2). But the overwhelming majority of UACs are from non-contiguous countries and are therefore transferred to ORR custody. See id. § 1232(a)(2)(A), (a)(3), (b); see also U.S. Customs & Border Patrol, U.S. Border Patrol Southwest Border Apprehensions by Sector Fiscal Year 2019 (May 8, 2019), https://www.cbp.gov/newsroom/stats/swborder-migration/usbp-sw-border-apprehensions. In fiscal year 2018, almost 50,000 unaccompanied minors were referred to ORR. ORR, UAC Fact Sheet 2. Federal law requires prompt placement of UACs “in the least restrictive setting that is in the best interest of the child.” 8 U.S.C. § 1232(c)(2)(A). Pursuant to that requirement, ORR usually places unaccompanied minors in one of roughly 100 federally funded shelters across the country. ORR, UAC Fact Sheet 2. 8 B. An unaccompanied minor ordinarily remains in ORR custody until one of five events occurs: (i) she is released to a sponsor in the United States; (ii) she turns 18, at which point she is transferred to the custody of DHS; (iii) she obtains lawful immigration status in the United States; (iv) she is permitted to voluntarily depart the country; or (v) she is removed from the country. According to recent government data, the average length of time an unaccompanied minor remains in ORR custody is approximately 90 days. Id. A minor might remain in ORR custody for substantially more (or less) time, however, depending on her individual circumstances. Most UACs are released to a sponsor at some point, and they remain with their sponsor while awaiting immigration hearings. See id. The search for a suitable sponsor begins as soon as an unaccompanied minor comes into ORR custody. See Office of Refugee Resettlement, ORR Guide: Children Entering the United States Unaccompanied § 2.2 (Jan. 30, 2015), https://www.acf.hhs.gov/orr/resource/childrenentering-the-united-states-unaccompanied (“ORR Guide”). A sponsor might be an immediate relative or legal guardian, a distant relative, or an unrelated adult with a bona fide social relationship with the minor or her family. Id. §§ 2.2.1, 2.2.4. “All potential sponsors for UAC[s] are required to undergo background checks and complete a sponsor assessment process that identifies risk factors and other potential safety concerns.” ORR, UAC Fact Sheet 2. Accordingly, the “process for the safe and timely release of an unaccompanied alien child from ORR custody” to a sponsor “involves many steps.” ORR Guide § 2.1. Those steps include: “the identification of sponsors; the submission by a sponsor of the application for release and supporting 9 documentation; the evaluation of the suitability of the sponsor, including verification of the sponsor’s identity and relationship to the child, background checks, and in some cases home studies; and planning for post-release.” Id. In some cases, ORR is never able to identify an appropriate sponsor. Whether or not released to a sponsor, a UAC may be able to attain lawful immigration status in the United States. Any unaccompanied minor who gains lawful immigration status while in ORR custody must be released into an alternative placement. Id. § 2.8.6. According to ORR, “[m]any unaccompanied alien children meet conditions that make them eligible for legal relief to remain in the United States.” ORR, UAC Services. Those forms of relief include but are “not limited to asylum; special visas for children who have been abused, neglected, or abandoned by the parents or guardian; special visas for victims of severe forms of trafficking and other types of crime; or adjustment of status for those who have a legal resident or citizen family member.” Id. The first of those forms of immigration relief, asylum, entitles a person who demonstrates a well-founded fear of persecution in her country of origin to remain in the United States and, eventually, to obtain lawful permanent residence. See 8 U.S.C. §§ 1101(a)(42)(A), 1158. The other described types of immigration relief include Special Immigrant Juvenile Status (through which juveniles subjected to abuse or neglect can attain lawful permanent residence), see id. §§ 1101(a)(27)(J), 1153(b)(4), as well as T or U nonimmigrant visas for victims of qualifying crimes or human trafficking, see id. §§ 1101(a)(15)(T)–(U), 1184(o)–(p). Barring a path to lawful status in the United States, a UAC can also apply for “voluntary departure” to her country of 10 origin. “Voluntary departure is a discretionary form of relief that allows certain favored aliens . . . to leave the country willingly” rather than undergo removal. Dada v. Mukasey, 554 U.S. 1, 8 (2008). Although a grant of voluntary departure does not entitle an alien to remain in the United States, it is a form of immigration relief because it relieves her of some of the penalties that would attach if she were removed (including, for example, the five-year bar on reentry). Id. at 11. The grant of voluntary departure is at the government’s discretion, see 8 U.S.C. § 1229c(a)(1), and is contingent on the withdrawal of claims to other forms of relief, a concession of removability, and a waiver of the right to appeal, see 8 C.F.R. § 1240.26. Finally, unaccompanied minors in ORR custody or released to sponsors are subject to removal from the United States. See 8 U.S.C. § 1229a. In that respect, though, they are entitled to greater procedural protections than either the subset of minors from contiguous countries subject to immediate repatriation or adults who can be summarily removed. See, e.g., 8 U.S.C. §§ 1182(a)(6)(C), (a)(7), 1225(b)(1)(A)(i), (iii). For example, removal cases for UACs must be adjudicated by immigration judges, see id. § 1229a, and the government must ensure that unaccompanied minors have the assistance of counsel in removal proceedings “to the greatest extent practicable,” id. § 1232(c)(5). C. Roughly thirty percent of the unaccompanied minors to arrive in the United States in recent years have been female. See U.S. Dep’t of Health & Human Servs., Office of Refugee Resettlement, Facts and Data (Feb. 13, 2019), https://www.acf.hhs.gov/orr/about/ucs/facts-and-data. The healthcare services afforded to them while in ORR custody include “family planning services, including pregnancy tests 11 and comprehensive information about and access to medical reproductive health services and emergency contraception.” ORR Guide § 3.4. Each year, ORR has several hundred pregnant unaccompanied minors in its custody. See Email from Kate Wolff to Bobbie Gregg (Feb. 24, 2016), Mot. for Class Certification Ex. B at 2 (filed Oct. 18, 2017), ECF No. 18-5 (726 pregnancies in 2014 and an estimated 450 pregnancies in 2015). At least 21 shelters, in states such as Texas, Arizona, Virginia, and Washington, have housed pregnant UACs. See Mot. for Class Cert. Ex. C (filed Oct. 18, 2017), ECF Nos. 19-1 to 19-4. In fiscal year 2017, the only year for which there is data in the record concerning abortion requests, 18 pregnant unaccompanied minors in ORR custody requested an abortion. In March 2017, ORR announced that shelters “are prohibited from taking any action that facilitates an abortion without direction and approval from the Director of ORR.” Memorandum from Kenneth Tota, Acting Dir., Office of Refugee Resettlement, to ORR Staff (Mar. 4, 2017), Mot. for Prelim. Inj. Ex. A (filed Oct. 14, 2017), ECF. No. 5-4. Previously, there had been no need for a shelter to secure the Director’s approval before assisting a minor with accessing abortion services (unless federal funds were to be used directly for the procedure). A shelter thus could assist a minor if an abortion would be consistent with the relevant state’s laws. If a shelter objected to permitting a minor abortion access on religious or other grounds, ORR would transfer her to a shelter willing to provide access. Under the new policy’s requirement to secure the ORR Director’s approval before permitting abortion access, Scott Lloyd, who became Director in March 2017, denied every abortion request presented to him during his tenure. He refused 12 every request regardless of the circumstances, including when the pregnancy resulted from rape. See Dep. of Scott Lloyd, Dir., Office of Refugee Resettlement, at 64:19–21, 153:9–14 (Dec. 18, 2017), G.C.A. 207, 229; Dep. of Jonathan White, Deputy Dir. for Children’s Programs, at 17:20–18:3 (Dec. 19, 2017), P.A. 33–34. The requirement to obtain the Director’s approval thus functions as a blanket ban. The ban, though, applies only to those unaccompanied minors who are in ORR custody (including those at ORR grantee shelters). A minor who is released to a sponsor, or who obtains lawful immigration status, thus is no longer subject to the abortion bar. The same is true of unaccompanied minors who turn 18 and are then transferred to DHS custody. DHS, unlike ORR, allows pregnant women in its custody to obtain abortions. See Immigration & Customs Enforcement Guidelines, Detention Standard 4.4, Medical Care (Women) (Dec. 2016), https://www.ice.gov/doclib/detentionstandards/2011/4-4.pdf. D. This class action was brought in the name of four plaintiffs who were unaccompanied minors in ORR custody and whose requests for an abortion were denied under the new policy. 1. Jane Doe was 17 years old when apprehended at the border and remitted to the custody of an ORR shelter in Texas. After a medical examination showed she was pregnant, Doe requested access to an abortion. Texas law requires parental consent or a judicial bypass, and Doe secured a judicial bypass in Texas court so that she could decide on her own to terminate her pregnancy. ORR notified Doe’s mother of her pregnancy 13 and her request for an abortion, despite indications that doing so could expose her to a risk of serious abuse by her family. Although Doe identified two potential sponsors to ORR, neither was determined to be suitable or willing to sponsor her. At the time, Doe was seeking a determination in state court that would have permitted her to apply for Special Immigrant Juvenile Status—which, as noted, is a form of immigration relief available to children who are victims of abuse. ORR emails stated that Doe had also applied for voluntary departure but that her application was “not likely to be far [along] at all” at the time. Email from Jonathan White, Deputy Dir. for Children’s Programs, to Scott Lloyd, Dir., Office of Refugee Resettlement (Sept. 22, 2017), P.A. 26. Doe obtained private funding for the abortion procedure and arranged her own transportation to and from the provider. But even though Doe secured her own funding and transportation, and even though she had satisfied the conditions under Texas law to obtain an abortion, ORR, per Director Lloyd’s instruction, refused to authorize her release from the shelter for the procedure. On October 14, 2017, a month after she initially requested an abortion, Doe brought the present suit challenging ORR’s abortion policies on behalf of herself and a class of similarly situated individuals. On October 18, the district court granted Doe a temporary restraining order. The order enjoined the government from preventing her transport to an abortion facility or from otherwise interfering with her decision to terminate her pregnancy. A panel of this Court vacated that decision on October 20, see Garza v. Hargan, No. 17-5236, 2017 WL 9854552 (D.C. Cir. Oct. 20, 2017), but four days later, this Court, sitting en banc, vacated the panel order and 14 reinstated the district court’s temporary restraining order, see Garza v. Hargan, 874 F.3d 735 (D.C. Cir. 2017). Doe obtained an abortion the next day, October 25. See Azar v. Garza, 138 S. Ct. 1790, 1792 (2018) (per curiam). At the time, she was estimated to be at least 14 weeks pregnant. Almost three months later, on January 15, 2018, just before Doe turned 18 years old, ORR released her to a sponsor. On June 1, 2018, the Supreme Court vacated our en banc order because Doe’s claim had become moot. Id. at 1792–93; see United States v. Munsingwear, Inc., 340 U.S. 36, 39 (1950). 2. A second named plaintiff, Jane Poe, was 17 years old and pregnant when apprehended at the border in November 2017. During her initial health screening, Poe disclosed that she had been raped by a stranger in her country of origin. A subsequent medical examination revealed that her pregnancy was the result of the rape. Poe repeatedly requested an abortion even though her mother (in her country of origin) and a potential sponsor (in the United States) threatened to beat her if she attempted to terminate her pregnancy. ORR’s Deputy Director for Children’s Programs wrote a memorandum to Director Lloyd in early December 2017, explaining the circumstances surrounding Poe’s request for an abortion. The Deputy Director reported that Poe “would like an abortion on the grounds of being raped.” Memorandum from Jonathan White, Deputy Dir. for Children’s Programs, to Scott Lloyd, Dir., Office of Refugee Resettlement (Dec. 6, 2017), P.A. 16. The Deputy Director further explained that Poe “does not have any viable sponsors” to whom she could be released, and that her pregnancy had reached 21 weeks, such that the state-law deadline for an abortion was fast approaching. Id. at 16–17. As a result, the Deputy Director 15 urged, it was “critical that a decision to approve or deny her request” be made “as soon as possible.” Id. at 17. Ten days later, Director Lloyd denied Poe’s request for permission to obtain an abortion. Id. at 18. The next day, Lloyd issued a file memorandum documenting his decision. He noted that Poe had become pregnant as the result of rape but explained that ORR provides refuge “to all the minors in our care, including their unborn children.” Note to File from Scott Lloyd, Dir., Office of Refugee Resettlement (Dec. 17, 2017), P.A. 20, 23. “In this request,” Lloyd determined, “we are being asked to participate in killing a human being in our care,” and “we ought to choose [to] protect life rather than to destroy it.” Id. at 23. One day later, the district court granted Poe’s motion for a temporary restraining order over the government’s opposition. Garza v. Hargan, No. 17-cv-02122, 2017 WL 6462270, at *1 (D.D.C. Dec. 18, 2017). Poe then obtained an abortion. As of July 30, 2018, Poe had not been released to a sponsor and remained in ORR custody. On December 13, 2018, however, counsel informed the court that Poe had been granted asylum and was no longer in ORR custody. 3. The final two named plaintiffs are Jane Roe and Jane Moe. Unlike Jane Doe and Jane Poe, each of whom received a temporary restraining order and obtained an abortion while still in ORR custody, Jane Roe and Jane Moe were released from ORR custody before terminating their pregnancies. In Roe’s case, she learned of her pregnancy in November 2017, while in ORR custody. She claims, and the government believed, she was 17 at the time. She requested an abortion 16 from her shelter but was not allowed access to an abortion provider. On December 18, the district court granted Roe a temporary restraining order. See id. at *1. The government filed an appeal but soon dismissed it upon discovering information allegedly indicating that Roe in fact was not a minor and thus not properly in ORR custody. Roe was then transferred to the custody of DHS, which, as noted, allows immigration detainees to obtain an abortion. As for Jane Moe, around late December 2017, she informed her ORR shelter that she desired to terminate her pregnancy. On January 11, she joined the suit and filed an application for a temporary restraining order, claiming that the government had already delayed her abortion access by two weeks. But three days later, Moe was released to a sponsor. E. While this class action is brought in the name of four named plaintiffs, only two of them, Doe and Roe, serve as class representatives. They moved to certify a class of “pregnant [UACs] who are or will be in the legal custody of the federal government.” Mot. for Class Certification at 1 (filed Oct. 18, 2017), ECF No. 18. They sought a preliminary injunction, claiming that ORR maintains a blanket ban on abortion access, a parental-notification-and-consent requirement, and compelled religious counseling, in violation of the Fifth and First Amendments. 1. On March 30, 2018, the district court certified a class of plaintiffs consisting of “all pregnant, unaccompanied immigrant minor children (UCs) who are or will be in the legal custody of the federal government.” Garza v. Hargan, 304 F. Supp. 3d 145, 150 (D.D.C. 2018). The court certified 17 the class under Federal Rule of Civil Procedure 23(b)(2), which applies when a defendant acts on grounds that apply generally to the class, such that an injunction (or declaratory relief) is appropriate as to the entire class. On the merits, the court granted a preliminary injunction to the class. The court explained that “the government ‘may not prohibit any woman from making the ultimate decision to terminate her pregnancy before viability.’” Id. at 162 (quoting Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 879 (1992) (plurality)). That “basic proscription,” the court determined, “controls the outcome in this case.” Id. ORR’s policies, the court observed, “apply to all pregnant [UACs] in its custody—even those whose pregnancy is the result of rape.” Id. at 161. Under those policies, the court explained, “ORR effectively retains an absolute veto over the reproductive decision of any young woman in its custody, a veto that is exercised routinely to bar [UACs] from obtaining abortions, despite the fact that no public funds are expended to procure the procedures and notwithstanding the [UAC’s] own wishes or intentions.” Id. at 162. “In other words,” the court concluded, “ORR’s absolute veto nullifies a [UAC’s] right to make her own reproductive choices.” Id. And “ORR’s policy vests the power to decide the future of a [UAC’s] pregnancy in one man: Director Lloyd,” whose “ultimate decision is substantially controlled by—if not entirely based on—his ideological opposition to abortion.” Id. at 163. 2. The district court initially entered its preliminary injunction on March 30, 2018, and then clarified it on April 16, 2018. The injunction contains two relevant provisions. 18 First, it enjoins the government from “interfering with or obstructing any class member’s access to . . . an abortion” or “other pregnancy-related care” (and also enjoins any interference with access to a judicial bypass or abortion counseling). Prelim. Inj. Order (Apr. 16, 2018), G.C.A. 275. That access mandate pertains solely to pre-viability abortions. See Order Granting in Part and Denying in Part Motion to Stay (June 4, 2018) (concurring statement of Srinivasan, J.). Second, the court enjoined the government from revealing, or forcing class members to reveal, the fact of their pregnancies or their abortion decisions to anyone. (Although the government filed a notice of appeal only as to the original March 30, 2018, order, the April 16 order merely clarified the prior order in relevant respects, such that the March 30 order as clarified on April 16 is properly before us. See Fed. R. App. P. 4(a)(4)(B)(ii); cf. Sorensen v. City of New York, 413 F.3d 292, 296 & n.2 (2d Cir. 2005).) Those two aspects of the preliminary injunction—the access mandate and the disclosure bar—have been appealed by the government. The government also appeals the district court’s grant of class certification. The government, though, does not appeal other provisions of the preliminary injunction that bar retaliation against class members or shelters for abortion-related decisions and actions. II. The government devotes a majority of its principal brief to arguing two threshold issues before addressing the merits of the district court’s preliminary injunction: (i) mootness, and (ii) class certification. We first take up the government’s mootness challenge, which we reject. The class’s claims persist here because of the “inherently transitory” exception. 19 “Mootness is a pragmatic doctrine meant to limit ‘judicial power to disputes capable of judicial resolution.’” DL v. District of Columbia, 860 F.3d 713, 722 (D.C. Cir. 2017) (quoting U.S. Parole Comm’n v. Geraghty, 445 U.S. 388, 396 (1980)). The mootness inquiry is a “claim-specific analysis.” Daingerfield Island Protective Soc’y v. Lujan, 920 F.2d 32, 37 (D.C. Cir. 1990); accord Coal. of Airline Pilots Ass’ns v. FAA, 370 F.3d 1184, 1189–90 (D.C. Cir. 2004). The party seeking jurisdictional dismissal bears the “initial ‘heavy burden’ of establishing mootness,” but the “opposing party bears the burden of proving an exception applies.” Honeywell Int’l, Inc. v. Nuclear Regulatory Comm’n, 628 F.3d 568, 576 (D.C. Cir. 2010) (quoting Motor & Equip. Mfrs. Ass’n v. Nichols, 142 F.3d 449, 459 (D.C. Cir. 1998)). The government challenges all the claims, including those on which the district court declined to issue preliminary relief. “If an intervening circumstance deprives the plaintiff of a ‘personal stake in the outcome of the lawsuit,’ at any point during litigation,” the district court must dismiss her individual claim as moot. Genesis Healthcare Corp. v. Symczyk, 569 U.S. 66, 72 (2013) (quoting Lewis v. Cont’l Bank Corp., 494 U.S. 472, 477–78 (1990)). For every claim, at least one named plaintiff must keep her individual dispute live until certification, or else the class action based on that claim generally becomes moot. United States v. Sanchez-Gomez, 138 S. Ct. 1532, 1538 (2018); see also Cruz v. Am. Airlines, Inc., 356 F.3d 320, 331 (D.C. Cir. 2004). Here, the district court selected Doe and Roe as representatives. (Although the selection was not revealed until the April 16, 2018, order, we exercise pendent jurisdiction to review the certification portion, see Wagner v. Taylor, 836 F.2d 578, 583 (D.C. Cir. 1987).) 20 The government has met its burden regarding the abortionaccess claims. The Supreme Court held that Doe’s claim “became moot after” her October 25, 2017, abortion. Garza, 138 S. Ct. at 1793. Roe’s claim became moot in late December 2017, when she left ORR custody and was no longer subject to ORR’s policies. The First Amendment claims also are moot. The minors have presented two theories: that (i) ORR compels them to speak with third parties about their abortion decisions, and that (ii) ORR commits proselytism by forcing them to meet with certain religiously affiliated counselors. But those claims extinguished when Doe and Roe left ORR custody on January 15, 2018, and on or around December 19, 2017, respectively. In the pleadings, the minors raised two types of Fifth Amendment disclosure claims: one predicated on their right to “informational privacy,” the other on their right to choose whether to terminate the pregnancy. Having obtained their abortions and exited ORR custody, respectively, Doe and Roe no longer have the latter claims. As for informational privacy, we need not decide whether the claims are moot, because they, like the others, would satisfy the “inherently transitory” mootness exception. The Supreme Court sometimes has permitted the lower courts to “relate [a] certification motion back” to a date when the individual claims were live. Genesis Healthcare, 569 U.S. at 71 & n.2. A properly certified class is deemed to have attained on that date a “legal status separate from the interest asserted” by the representatives. Sosna v. Iowa, 419 U.S. 393, 399 (1975). Because the class possesses a concrete legal interest, the mootness of individual claims does not affect the ability of representatives to litigate a controversy between the defendants and absent class members. Id. at 402. 21 The relation-back date depends on the case. For instance, “where a certification motion is denied and a named plaintiff’s claim subsequently becomes moot, an appellate reversal of the certification decision may relate back to the time of the denial.” Genesis Healthcare, 569 U.S. at 71 n.2; accord DL, 860 F.3d at 721–23. Relevant here, “[w]here a named plaintiff’s claim is ‘inherently transitory,’ and becomes moot prior to certification, a motion for certification may ‘relate back’ to the filing of the complaint.” Genesis Healthcare, 569 U.S. at 71 n.2 (quoting Cty. of Riverside v. McLaughlin, 500 U.S. 44, 51– 52 (1991)). We applied the “inherently transitory” doctrine once before, but we did not elaborate on its contours. See Basel v. Knebel, 551 F.2d 395, 397 n.1 (D.C. Cir. 1977) (per curiam). We do so now. The Supreme Court crafted the exception in injunctive class actions challenging criminal and immigration detention procedures. In Gerstein v. Pugh, 420 U.S. 103 (1975), the first case to apply it, four individuals who were arrested without warrants in Florida sued state officials and asserted a federal constitutional right to a judicial probable-cause hearing as a prerequisite to pretrial detention, id. at 105–07. The Supreme Court noted that “the record d[id] not indicate whether any of [the four plaintiffs] w[as] still in custody awaiting trial when the District Court certified the class.” Id. at 110 n.11. Because the individuals sought a hearing for pretrial detention, their claims necessarily became moot when the detention ended. See id. Nonetheless, the Court let the class action survive. It reasoned: The length of pretrial custody cannot be ascertained at the outset, and it may be ended at any time by release on recognizance, dismissal of the charges, or a guilty plea, as well as by acquittal or conviction after trial. It is by no 22 means certain that any given individual, named as plaintiff, would be in pretrial custody long enough for a district judge to certify the class. Moreover, in this case the constant existence of a class of persons suffering the deprivation is certain. The attorney representing the named respondents is a public defender, and we can safely assume that he has other clients with a continuing live interest in the case. Id. The Supreme Court applied Gerstein’s holding in three other cases. See Nielsen v. Preap, 139 S. Ct. 954, 963 (2019) (plurality); McLaughlin, 500 U.S. at 50–52; Swisher v. Brady, 438 U.S. 204, 213 n.11 (1978). In Brady, the Supreme Court considered the Double Jeopardy implications of a state regime where juveniles in criminal proceedings had been found not guilty in proposed rulings by so-called “masters” but were convicted after prosecutors filed exceptions and juvenile court judges reversed the masters’ proposals. 438 U.S. at 206–13. Nine juveniles filed the injunctive class action in November 1974, asserting that a state procedural rule creating the regime was unconstitutional. Id. at 206, 209. The Supreme Court noted that, prior to certification, the injunctive claims for the juveniles became moot because the State either had withdrawn its objections (thus removing the minor from alleged jeopardy) or secured a ruling from a juvenile court judge (thus completing the allegedly unconstitutional second prosecution). Brady, 438 U.S. at 213 n.11. Still, the Court emphasized the “rapidity of judicial review of exceptions” for all class members and allowed the class’s claims. Id. The Court also highlighted that expired individual claims need not end a class action if mootness occurs before the district judge “can 23 reasonably be expected to rule” on certification. Id. (quoting Sosna, 419 U.S. at 402 n.11). In McLaughlin, the Supreme Court applied Gerstein in a factually similar context. The Court in Gerstein recognized the constitutional requirement for a judicial hearing and noted that it must occur “promptly” after the warrantless arrest. Gerstein¸ 420 U.S. at 125. Pretrial detainees brought an injunctive class action challenging the promptness of hearings taking place in the County of Riverside, California. McLaughlin, 500 U.S. at 47–48. The Supreme Court noted that the individual claims had become moot before certification because the named plaintiffs either “received probable cause determinations or were released.” McLaughlin, 500 U.S. at 51. But like in Gerstein, some claims are “so inherently transitory that the trial court will not have even enough time to rule on a motion for class certification before the proposed representative’s individual interest expires.” Id. (quoting Geraghty, 445 U.S. at 399). The Court concluded that lower courts may invoke “the ‘relation back’ doctrine” to “preserve the merits” of such claims “for judicial resolution.” Id. Most recently, in Preap, the Supreme Court considered the scope of an immigration detention provision in the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (“IIRIRA”), Pub. L. No. 104-128, div. C, 110 Stat. 3009546 (codified as amended in scattered sections of Titles 8, 18, and 28 of the U.S. Code). With exceptions not relevant here, the IIRIRA provision mandates the detention without bail of immigrants who had been convicted of certain crimes, who were later arrested upon belief of their inadmissibility or deportability, and who are awaiting the conclusion of removal proceedings. See 8 U.S.C. § 1226(c). In two injunctive class actions, immigrant plaintiffs who were detained under the provision sought a bail hearing, which federal regulations 24 ordinarily would provide. See Preap, 139 S. Ct. at 959–60; id. at 975 (Thomas, J., concurring in part and concurring in the judgment); see also 8 C.F.R. §§ 236.1(c)(8), (d)(1), 1003.19, 1236.1(d)(1). “[B]y the time of class certification[,] the named plaintiffs had obtained either cancellation of removal or bond hearings.” Preap, 139 S. Ct. at 963 (plurality). The government thus argued that the class actions were moot. And two justices found that the “inherently transitory” exception does not apply because the immigrants “are held, on average, for one year, and sometimes longer” and the trial judges could rule on certification within such a time frame. Id. at 976 (Thomas, J., concurring in part and concurring in the judgment) (citing Jennings v. Rodriguez, 136 S. Ct. 830, 860 (2018) (Breyer, J., dissenting)). But the plurality disagreed with the government and those two justices. Id. at 963 (plurality). Unmoved by the one-year average length of time, the plurality found detention to be sufficiently “transitory” because it “ends as soon as the decision on removal is made.” Id. (plurality). As for the cancellation of removal and bond hearings, the plurality found irrelevant the fact that the “named plaintiffs obtained some relief before class certification.” Id. (plurality). Gerstein, Brady, McLaughlin and Preap confirm that the relation-back doctrine requires us to analyze the “practicalities and prudential considerations” of the class action under review. Geraghty, 445 U.S. at 404 n.11; see also Basel, 551 F.2d at 397 n.1 (“[W]hether the certification can be said to ‘relate back’ to the filing of the complaint may depend upon the circumstances of the particular case and especially the reality of the claim that otherwise the issue would evade review.”); cf. DL, 860 F.3d at 722 (noting the “pragmatic” nature of the mootness doctrine). 25 None of the cases purports to outline all factors relevant to the inquiry. Still, two requirements emerge. First, as the exception’s moniker implies, we must consider the extent to which the individual claims are “inherently transitory.” As Gerstein puts it, the district court must determine whether it is “by no means certain” that an individual claim will persist long enough for it to adjudicate class certification. 420 U.S. at 110 n.11 (“It is by no means certain that any given individual, named as plaintiff, would be in pretrial custody long enough for a district judge to certify the class.”); accord Sanchez-Gomez, 138 S. Ct. at 1538. Because the mootness inquiry depends on whether the claim is potentially fleeting, we must determine what qualifies as too brief. Once the district court deems a Rule 23 class valid, the subsequent mootness of individual claims does not terminate litigation. See Sosna, 419 U.S. at 399, 402; see also Genesis Healthcare, 569 U.S. at 75. The “inherently transitory” exception serves only to salvage claims that will, or at least might, not survive until certification. Thus, we must consider whether “mootness problems” might arise to end the claim “before the district court can reasonably be expected to rule on a certification motion.” Brady, 438 U.S. at 213 n.11 (quoting Sosna, 419 U.S. at 402 n.11). The inquiry may rely on reasoned supposition. In Gerstein, the Supreme Court expressed concerns that release, dismissal of charges, a plea, or a verdict “may” end the pretrial detention claims before a class-certification decision. Id. The Court never attempted to figure out which—or even whether— these events in fact occurred to a class member; the record did not reveal such details. See id. The Court instead hypothesized events that “may” occur, based on the “practicalities” of the litigation at issue, Geraghty, 445 U.S. at 404 n.11, and its 26 understanding of how the criminal justice system works in general. Second, the record must sufficiently assure us that some class members will retain a live claim throughout the proceedings. See Gerstein, 420 U.S. at 110 n.11 (“Moreover, in this case the constant existence of a class of persons suffering the deprivation is certain.”); see also SanchezGomez, 138 S. Ct. at 1538; Genesis Healthcare, 569 U.S. at 76. Even in the class-action context, a “live controversy” must always exist throughout the litigation. See Sosna, 419 U.S. at 402. Indeed, the Supreme Court in Gerstein noted that some class members—to wit, absent clients of the plaintiffs’ counsel—had a “continuing live interest” while the case was before it. 420 U.S. at 110 n.11. In sum, the “inherently transitory” exception to mootness requires us to determine (i) whether the individual claim might end before the district court has a reasonable amount of time to decide class certification, and (ii) whether some class members will retain a live claim at every stage of litigation. An affirmative answer to both questions ordinarily will suffice to trigger relation back. Doe and Roe have demonstrated that the exception applies in this case. The claims at issue likely will, or at least might, end quickly. The average length of custody for a minor was 41 days in fiscal year 2017, when the initial complaint was filed, and was roughly 90 days by the beginning of fiscal year 2019. Of course, that is just an average, and certain events could end the claims earlier. A minor under ORR custody may turn 18 years old or successfully seek voluntary departure to her country of origin. The government acknowledges that it may find a sponsor at any point and that it may obviate the claim by finding one swiftly. See Oral Arg. Recording 1:16:13–36. 27 Such is the case for Moe, for whom the government located a sponsor three days after she had joined the case. Thus, as the district court noted, “the length of time that pregnant [minors] will remain in ORR custody is uncertain and unpredictable.” Garza, 304 F. Supp. 3d at 159. The government responds that, for any individual claim, the district court will know in advance the viability date and the relevant abortion deadline for the state where ORR keeps the minor, and that the motion may be decided ahead of those dates. Gov’t Br. 24–26; Gov’t Reply Br. 5–6. The argument errs in ignoring sponsorship and voluntary departure as potential terminating events. And even if viability were the appropriate time frame for the mootness analysis, we would reject the government’s argument. Just as the one-year immigration detention in Preap would end too soon, so too would a full term of pregnancy, let alone the remaining weeks for obtaining a pre-viability abortion after a minor becomes aware of her pregnancy. Furthermore, the Court in Gerstein underscored that a defendant’s pretrial detention could come to an end (thus mooting the claim) “upon acquittal or conviction after trial,” which might not occur for many months. See 420 U.S. at 110 n.11. The government also stresses that Doe’s and Roe’s claims remained live long enough for the district court to decide their merits through temporary restraining order applications. Gov’t Br. 25–26. The district court “necessarily ha[d] time to take action” on the certification request because it had enough time to decide the merits. Gov’t Reply Br. 5–6. No case law supports this argument. We reject its upshot as unworkable and inequitable: that some class actions would evaporate because the irreparable harm to individual plaintiffs was clear enough to warrant immediate relief but the class definition issue was complex enough to require discovery. Relatedly, we 28 fear that accepting the argument would vitiate the mootness exception. Courts may issue temporary relief in virtually every case; a judge sometimes will sign a restraining order on the day the plaintiff files her complaint. Indeed, there would have been no need to apply the exception in Gerstein or McLaughlin, because the lower courts could have granted interim relief releasing detainees from pretrial custody. Accordingly, we find irrelevant the issuance of emergency relief in this case. See Preap, 139 S. Ct. at 963 (plurality) (“[T]he fact that the named plaintiffs obtained some relief before class certification does not moot their claims.”). As for the second question, the district court found—and the government does not dispute—that some class members will have live claims at every stage of litigation. See Garza, 304 F. Supp. 3d at 160 (noting that “the claims of numerous potential class members remain unaddressed”). ORR continues to keep pregnant minors, and the plaintiffs represent that about a dozen expressed an interest in abortion or related information during the first six months after the issuance of the injunction. See Oral Arg. Recording 1:34:38–54. Based on a faithful application of the Supreme Court’s precedents, we find that the plaintiffs have established both requirements of the “inherently transitory” exception. Accordingly, the district court’s class certification relates back to the date of the pleadings, and all the class’s claims remain live. III. We next consider whether the district court properly certified a class of plaintiffs consisting of pregnant UACs who are (or will be) in ORR custody. Federal Rule of Civil Procedure 23(a) requires putative class representatives to show that: “(1) the class is so numerous that joinder of all members 29 is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a). Here, the government contends that the class fails to meet each of those requirements. A proposed class must also satisfy one of the three requirements of Rule 23(b). The district court certified the class under Rule 23(b)(2), which applies when a defendant “has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole.” Fed. R. Civ. P. 23(b)(2). The government raises no argument of any error in certifying an injunctive class under Rule 23(b)(2), instead confining its arguments to Rule 23(a)’s requirements. We review the district court’s certification of the class only for abuse of discretion. See DL, 860 F.3d at 724. We find no abuse of discretion and thus affirm the certification of a class of pregnant UACs in ORR custody. A. The government raises arguments under each of Rule 23(a)’s four elements of numerosity, commonality, typicality, and adequacy. 1. The government’s primary contention is that the class representatives (Doe and Roe) fail to meet Rule 23(a)(4) in that they cannot “adequately protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). We are unpersuaded. 30 The adequacy requirement aims to ensure that absent class members will not be bound by the outcome of a suit in which they were not competently and fairly represented. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 625 (1997). Adequacy embraces two components: the class representative (i) “must not have antagonistic or conflicting interests with the unnamed members of the class” and (ii) “must appear able to vigorously prosecute the interests of the class through qualified counsel.” Twelve John Does v. District of Columbia, 117 F.3d 571, 575 (D.C. Cir. 1997) (quoting Nat’l Ass’n of Regional Med. Programs, Inc. v. Mathews, 551 F.2d 340, 345 (D.C. Cir. 1976)). The government advances arguments under each of those prongs. a. The government first contends that Doe and Roe are poorly positioned to vigorously prosecute the class’s interests. That is because, the government submits, Doe’s and Roe’s individual claims are moot, and they thus lack an ongoing stake in the case. Mootness alone, though, does not establish their inadequacy as representatives. Though the mootness of named plaintiffs’ claims can raise “adequacy concerns,” see DL, 860 F.3d at 726, the very existence of the inherently-transitory exception disproves any suggestion that the mootness of a plaintiff’s claims necessarily demonstrates her inadequacy as a representative. The entire object of that exception is to allow a class action to proceed even though the inherently fleeting nature of the class’s claims will predictably render a given class member’s claims moot before the class is certified. Our decisions thus find it “clear that mootness and adequacy are ‘separate issues’ and that plaintiffs with moot claims may adequately represent a class.” Id. (alteration omitted) (quoting Geraghty, 445 U.S. at 407). 31 And the Supreme Court has specifically recognized that a plaintiff with a moot claim may serve as a class representative. See Geraghty, 445 U.S. at 404; Sosna, 419 U.S. at 402–03. The government, beyond noting the mootness of Doe’s and Roe’s claims, identifies no reason to doubt their ability to vigorously press the action. The government suggests that their claims were live for a briefer period than was true in other cases in which class representatives were deemed adequate notwithstanding mootness. But the ostensibly short-lived duration of the representatives’ claims only reinforces their inherently transitory nature; it says nothing on its own about the representatives’ fitness or commitment to prosecuting the action. Indeed, the ephemerality of individual claims makes class-action treatment “particularly important” so as to “ensur[e] that a justiciable claim is before the Court.” Gratz v. Bollinger, 539 U.S. 244, 268 (2003) (citation omitted). The government also asserts that the issue of abortion is one as to which a person’s views are known to evolve over time, such that Doe and Roe, the government supposes, might become less enthusiastic proponents of the class’s claims before the litigation ends. The government’s assertion is unsupported speculation, though, and could equally be said of any class representative in any abortion case. The district court, in short, saw no reason to question the class representatives’ ability “to vigorously prosecute the interests of the class through qualified counsel.” Twelve John Does, 117 F.3d at 575 (quoting Nat’l Ass’n of Regional Med. Programs, 551 F.2d at 345). The government does not challenge the district court’s assessment that class counsel is “fully competent and qualified.” Garza, 304 F. Supp. 3d at 158 n.3. And as for the representatives, the mootness of their claims affords no ground for us to find an abuse of discretion 32 in the district court’s conclusion that they can adequately press the class’s interests. See DL, 860 F.3d at 726. b. The government next contends that the class representatives “have antagonistic or conflicting interests with the unnamed members of the class.” Twelve John Does, 117 F.3d at 575 (quoting Nat’l Ass’n of Regional Med. Programs¸ 551 F.2d at 345). The government’s argument on that score centers on one consideration: that most members of the class, which includes all pregnant unaccompanied minors in ORR custody, will ultimately choose to carry their pregnancies to term rather than obtain an abortion. As a result, the government urges, the class representatives’ interests conflict with those of the unnamed class members. i. In our view, the fact that the class representatives chose to terminate their pregnancies, whereas the lion’s share of the class might make the opposite choice, entails no “conflict[] of interest between named parties and the class they seek to represent” for purposes of Rule 23(a)(4)’s adequacy standard. Amchem, 521 U.S. at 625. The constitutional right asserted by the class is a woman’s “right to choose to terminate her pregnancy” before viability. Stenberg v. Carhart, 530 U.S. 914, 921 (2000) (quoting Casey, 505 U.S. at 870 (plurality)). The class members all assert a common entitlement to make that choice on their own, free from any veto power retained (unconstitutionally, the class says) by ORR. And on the plaintiffs’ theory, they are all denied the right to terminate their pregnancies by a veto power that effectively supersedes it. The class representatives are suited to press that interest on the class’s behalf, even if various class members might make varying ultimate decisions about how to exercise their choice. 33 The government posits that class members who decide to carry their pregnancies to term “likely want pregnancy-related, delivery, and post-partum care” rather than the “abortionfocused remedy ordered by the [district] court.” Gov’t Br. 30. But the court’s preliminary injunction bars interfering not only with “abortion counseling” and “an abortion” but also with “pregnancy-related care.” Prelim. Inj. Order (Apr. 16, 2018), G.C.A. 275. At any rate, there is no reason to think that the district court’s bar against interfering with a UAC’s choice to obtain an abortion would somehow diminish the services afforded to a UAC who instead carries her pregnancy to term. One minor’s choice has no necessary effect on the care given to another minor. The remedy sought by the class thus will have no deleterious effect on the care received by absent class members. ii. To be sure, as the government has argued, some pregnant unaccompanied minors who wish to carry to term might have “little interest in challenging” the veto power since it does not ultimately impede their ability to continue their pregnancies. Gov’t Opp’n to Class Certification 19 (filed July 30, 2018), G.C.A. 67. Those minors thus might be uninterested in pursuing the class action and in vindicating their right to terminate their pregnancies without undue government interference. But the presence of uninterested individuals in a class does not compel a finding of inadequacy. A basic object of the class action device is to permit an aggregated suit when an individual might forgo pressing a free-standing claim because she has too little at stake on her own. See Amchem, 521 U.S. at 617. A lack of interest among absent class members, then, is an everyday feature of class actions. 34 That is frequently so in the context of class actions for injunctive or declaratory relief under Rule 23(b)(2), like the action in this case. A principal purpose of Rule 23(b)(2) class actions is to enable class resolution of civil-rights claims alleging classwide deprivations of protected rights. See WalMart Stores, Inc. v. Dukes, 564 U.S. 338, 361 (2011). “The key to the (b)(2) class is the indivisible nature of the injunctive or declaratory remedy warranted—the notion that the conduct is such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them.” Id. at 360 (internal quotation marks omitted). The Rule thus authorizes class actions for injunctive or declaratory relief when “the party opposing the class has acted . . . on grounds that apply generally to the class.” Fed. R. Civ. P. 23(b)(2). When a challenged policy is generally applicable to the class for purposes of Rule 23(b)(2), the history of the Rule confirms the propriety of certifying the class even if some members may be uninterested in pressing the claims. The Advisory Committee explained that a challenged action “is directed to a class within the meaning of this subdivision even if it has taken effect or is threatened only as to one or a few members of the class.” Note of Advisory Committee on Rules—1966 Amendment, 28 U.S.C. App. at 812 (2012) (“1966 Adv. Comm. Note”); cf. Wal-Mart, 564 U.S. at 361 (relying on 1966 Adv. Comm. Note to interpret Rule 23(b)(2)). For instance, (b)(2) classes challenging voter-qualification laws often include anyone disenfranchised by the challenged laws. That is so regardless of whether class members ultimately intend (or are even registered) to vote. E.g., Frank v. Walker, 196 F. Supp. 3d 893, 901 (E.D. Wis. 2016) (class defined as “all those eligible to vote in Wisconsin who cannot with reasonable effort obtain a qualifying photo ID”), appeal filed, No. 16-3052 (7th Cir. July 28, 2016); Woodsum v. Boyd, 35 341 F. Supp. 448, 450 (M.D. Fla. 1972) (class defined as “those bona fide residents . . . who do not fulfill the durational residency requirements of [the challenged law] but who are otherwise qualified to register and vote”); Ferguson v. Williams, 330 F. Supp. 1012, 1018 (N.D. Miss. 1971) (class defined as “citizens . . . who would be qualified to vote in the 1971 state elections except for their failure to register four months prior to November 2”), vacated on other grounds, 405 U.S. 1036 (1972). Here, likewise, the class represented by Doe and Roe can encompass anyone whose right to terminate her pregnancy is allegedly infringed by the government’s policy, regardless of whether she ultimately intends to exercise her right. The ORR Director’s veto power acts on grounds that “apply generally to the class,” Fed. R. Civ. P. 23(b)(2), because it allegedly denies to any pregnant UAC the right and ability to terminate her pregnancy. Even if some class members are uninterested in the claims because they are not among the “members of the class” against whom the challenged policy “has taken effect” in a way that matters to them, 1966 Adv. Comm. Note, the class definition need not exclude them. Our dissenting colleague warns that such an approach would permit classes that are, in his view, self-evidently too broad. See Dissenting Op. 10. For example, he hypothesizes, it would allow certification of a class of all federal employees in a First Amendment challenge to a ban on federal employees’ membership in particular organizations. See id. But analogous classes are not unfamiliar. Thus, in United States Civil Service Commission v. National Ass’n of Letter Carriers, AFL-CIO, the Supreme Court addressed a First Amendment challenge involving a materially identical (b)(2) class—described as “all federal 36 employees”—to the Hatch Act’s bar against certain political activity by federal workers, even though some class members presumably had no desire to engage in the prohibited conduct. See 413 U.S. 548, 551 (1973). Similarly, in Sugarman v. Dougall, a challenge to a citizenship requirement for civil service positions, the (b)(2) class was defined to include anyone who would “otherwise be eligible to compete for employment,” not only those in fact interested in competing. 413 U.S. 634, 636 n.2 (1973) (emphasis added). Those examples exhibit the principle that “[a]ll the class members need not be aggrieved by or desire to challenge defendant’s conduct in order for some of them to seek relief under Rule 23(b)(2).” 7AA Charles Alan Wright et al., Federal Practice and Procedure § 1775 (3d ed. 2019). Rather, “whether everyone in the class is interested in challenging the policy at issue . . . is largely irrelevant.” 2 William B. Rubenstein, Newberg on Class Actions § 4:28 (5th ed. 2018). iii. Shifting course, the government submits that, among the absent class members, some pregnant minors “may strongly oppose abortions,” and they thus may “support ORR’s challenged polic[y]” and its no-exceptions mandate to carry a pregnancy to term including in cases of rape. Gov’t Br. 30. Because those minors disapprove of the class’s objectives on ideological grounds (presumably including as a matter of sincere religious and moral conviction), the government contends, they are not adequately represented by Doe and Roe. The government, though, did not argue to the district court that the potential presence in the class of minors who oppose abortion on such ideological grounds precludes class certification. If faced with an objection of that kind, a district court could consider measures such as “narrow[ing] the definition of the class, divid[ing] the proposed class into 37 subclasses[,] and permit[ting] class members to opt out of the class.” Wagner, 836 F.2d at 590 (footnotes omitted). The government did argue in the district court, as noted, that some class members wish to carry their pregnancies to term and so may be uninterested in challenging the policy (an argument we have just considered, pp. 33–36, supra). But the government at no point contended in the district court, as it does now, that some of those minors may strongly oppose abortion and that such ideological opposition affords added grounds to deny class certification. That argument was never made. In that respect, we do not understand our dissenting colleague’s misimpression that we see no difference between (i) pregnant minors who are “merely uninterested” in challenging the ORR policy because they seek to carry their pregnancy to term, and (ii) pregnant minors who are opposed to abortion on religious or moral grounds. Dissenting Op. 8. Insofar as the failure to draw any distinction between those two groups may be “quite intolerant of religious views,” id., we draw exactly that distinction. The government opposed class certification in the district court on the ground that the class includes minors uninterested in pressing the claims; but it made no argument that certification should be denied because the class includes minors opposed to abortion. We are hard-pressed in these circumstances to find an abuse of discretion in the district court’s failure to narrow the class definition on a ground that the government did not assert to the court. See United States v. Regan, 627 F.3d 1348, 1354 (10th Cir. 2010). True, the party seeking certification bears the burden of showing compliance with Rule 23. Wal-Mart, 564 U.S. at 350. But that does not require anticipating and affirmatively disproving every ostensible conflict that might later be pressed in an appeal. None of the decisions cited by the government in which a court of appeals examined an 38 asserted conflict between the representatives and the absent class members involved a ground that had not already been considered by the district court. See Spano v. Boeing Co., 633 F.3d 574, 586–88 (7th Cir. 2011) (reviewing Spano v. Boeing Co., No. 06-0743-DRH, 2008 WL 4449516, at *7 (S.D. Ill. Sept. 29, 2008)); Mayfield v. Dalton, 109 F.3d 1423, 1427 (9th Cir. 1997); Schy v. Susquehanna Corp., 419 F.2d 1112, 1117 (7th Cir. 1970). Even if the government had argued in the district court that Doe and Roe could not adequately represent the class because it includes minors who oppose abortion on ideological grounds, such a circumstance, standing alone, would not be considered a conflict of interest of a kind that precludes certifying a class. There might often be a possibility that some absent class members possess conscientious beliefs running counter to an interest in redressing an alleged infringement of their rights. Indeed, “[i]n any conceivable case, some of the members of the class will wish to assert their rights while others will not wish to do so.” Charles Alan Wright, Class Actions, 47 F.R.D. 169, 174 (1969). That is especially so in the civil rights cases that make up the heartland of actions under Rule 23(b)(2), which by nature can involve polarizing issues. In such situations, courts have been “reluctant to find the class representatives inadequate” even if “some class members have an explicit desire to maintain the status quo.” 2 Rubenstein, Newberg on Class Actions § 3:64; cf. Charise Cheney, Blacks on Brown: IntraCommunity Debates over School Desegregation in Topeka, Kansas, 1941–1955, 42 W. Hist. Q. 481 (2001) (describing opposition to school desegregation among black Topekans in the lead-up to Brown v. Board of Education, 347 U.S. 483 (1954)). As courts have long recognized, “[i]t is not ‘fatal if some members of the class might prefer not to have violations 39 of their rights remedied.’” Lanner v. Wimmer, 662 F.2d 1349, 1357 (10th Cir. 1981) (quoting U.S. Fid. & Guar. Co. v. Lord, 585 F.2d 860, 873 (8th Cir. 1978)). In that respect, the government’s reliance on decisions like Mayfield v. Dalton, 109 F.3d 1423, falls short. In that case, service members sought to challenge a DNA collection program established by the military to assist in the identification of remains. The class representatives were deemed inadequate because a significant share of the proposed class might favor retaining the challenged policy. See id. at 1427. There, though, the class members had a concrete interest in preserving a means by which their remains could be identified: as the district court had observed, it was not the military but rather “the next of kin of service members who will derive the greatest benefit, and solace, from the speedy and definite identification of the remains of their loved ones.” Mayfield v. Dalton, 901 F. Supp. 300, 304 (D. Haw. 1995). In that context, the Ninth Circuit determined that “[t]his is not [merely] a case where some class members might prefer to leave [a] violation of their rights unremedied.” Mayfield, 109 F.3d at 1427. Instead, “[w]e have here a conflict between the interests of the putative representative and members of the class.” Id.; accord Spano, 633 F.3d at 587 (finding representatives inadequate because the class was defined “so broadly that some members will actually be harmed by [the] relief” sought in the action). A similar sort of conflict might be present in hypothetical claims of the sort envisioned by our dissenting colleague: if, say, a class of employees challenging a restriction on employing neo-Nazi sympathizers were defined to include Jewish employees, the latter employees would have to work alongside the former if the class succeeded in lifting the restriction. See Dissenting Op. 10. 40 Assuming Mayfield was correctly decided, this case is different. Here, insofar as the class includes pregnant unaccompanied minors who are ideologically opposed to abortion, the relief ultimately sought in the action would not itself impose a tangible harm on them of the kind present in situations like Mayfield. The ability of minors who may sincerely oppose abortion to carry their pregnancies to term would not be compromised by the grant of relief securing another class member’s ability to make a different choice. With regard to the formal inclusion of minors who oppose abortion within the class definition during the pendency of the action—as opposed to the effect on them of the relief ultimately sought in the action—nothing is required of them while the case proceeds to a result. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 810 (1985). Indeed, because there is no notice given to absent class members in a (b)(2) action, see In re Veneman, 309 F.3d 789, 792 (D.C. Cir. 2002), putative class members in (b)(2) cases may have no particular reason to know they are nominally part of an ongoing action—they are, after all, absent class members. What about situations in which absent members do know about a pending action they ideologically oppose, and also learn about the class certification order and their formal inclusion in the certified class? The Supreme Court has held that “a desire to vindicate value interests”—i.e., ideological interests—does not “provide a judicially cognizable interest” sufficient to confer standing to bring an action. Diamond v. Charles, 476 U.S. 54, 66–67 (1986). And that rule specifically encompasses a “conscientious objection to abortion.” Id. Yet if such an interest, standing alone, does not suffice to enable bringing an action, it is unclear why it compels being excluded from one. We are aware of no decision holding otherwise. 41 That is unsurprising, as a class might often include absent members who, if alerted to their membership, would prefer to be excluded from the action. Rule 23(b)(2)’s very design— which, unlike Rule 23(b)(3), has no mandate for an opt-out opportunity, see Veneman, 309 F.3d at 792—presupposes the presence of some absent members who might exclude themselves if given the chance: again, in “any conceivable case, some of the members of the class . . . will not wish” to “assert their rights.” Wright, 47 F.R.D. at 174. How, then, would courts distinguish the circumstances in which subjective opposition gives rise to a conflict of interest precluding class certification from those in which it does not? If, say, a Second Amendment claim were brought on behalf of a class of persons subject to a firearms regulation, would there be a need to deny class certification because some absent members may strongly support gun-control measures like the challenged one? Or if a (b)(2) challenge to the consideration of race in a college-admissions process were brought on behalf of all members of the allegedly disfavored racial groups (e.g., Caucasian) denied admission—as was precisely the case in Gratz v. Bollinger, 539 U.S. at 252–53— should certification be denied because the class may include absent members who ideologically support affirmative action? That manner of logic could ultimately stymie certification in virtually any (b)(2) case. Our dissenting colleague’s responses to those situations only confirms that (b)(2) classes can include ideologically opposed members. With regard to the challenge to the firearms regulation, our colleague says that the class would be limited to those who “claim injury.” Dissenting Op. 14. But of course a class could not in fact be defined simply as persons who “claim injury”; and regardless, even in a class including only those who, say, own firearms and are therefore subject to the 42 regulation, some members may still oppose the action because they generally support gun-control measures. With regard to the college-admissions challenge, our colleague notes that the class in Gratz was limited to applicants who “suffered an injury” when denied admission. Id. at 15. Again, though, some of those rejected applicants may have opposed the suit based on their support of affirmative action, yet they were still part of the class. Nothing in Rule 23 affords a basis for distinguishing between class members ideologically opposed to the action in Gratz and class members ideologically opposed to the action in this case. Our colleague suggests that, in a case like Gratz, a class member who supports affirmative action may be “satisfied with the status quo” but still not be “opposed to the recognition of the right not to be discriminated against.” Id. at 14. But that ostensible distinction is not discernible anywhere in Rule 23. At any rate, the distinction depends on the level of generality at which one conceives of the right: a class member supportive of affirmative action might well oppose “recognition of the right not to be discriminated against” when it specifically concerns an asserted right to strict race neutrality. The suggested distinction between supporting the status quo and opposing the right is elusive in this case too. Indeed, our colleague says that there is “hardly a differen[ce]” between “people who support[] the Government’s challenged policy” and people who “have moral/religious objections to abortion.” Id. at 6. The upshot is that, in a Rule 23(b)(2) action, there is no fixed requirement to give absent class members who may be ideologically opposed to the case the effective equivalent of an opt-out from the action. After all, in a (b)(2) case, an “injunction prohibiting a defendant’s action against ‘the class as a whole’ would halt the action regardless whether some of 43 those affected might have withdrawn from the suit if given the option.” Richards v. Delta Air Lines, Inc., 453 F.3d 525, 530 (D.C. Cir. 2006) (emphasis added) (quoting Fed. R. Civ. P. 23(b)(2)); see Horton v. Goose Creek Indep. Sch. Dist., 690 F.2d 470, 487 n.32 (5th Cir. 1982). And while our dissenting colleague hypothesizes that absent class members have a First Amendment entitlement to be excluded from an action, see Dissenting Op. 11—a view that could confer notice and opt-out rights in many if not all (b)(2) actions—no one has asserted any such entitlement at any point in this case. iv. While our colleague does not indicate what would be a proper class definition in this case, he believes the class cannot include pregnant minors who oppose abortion on moral or religious grounds. But a class definition that excludes persons based on their ideological convictions would necessarily turn on a person’s subjective (and possibly unannounced) beliefs. Such a definition would be, in the parlance of class actions, “unascertainable” or “indefinite,” because membership is not readily discernable by objective criteria. See 1 Rubenstein, Newberg on Class Actions § 3:3; City Select Auto Sales Inc. v. BMW Bank of N. Am. Inc., 867 F.3d 434, 439 n.3 (3d Cir. 2017); Mullins v. Direct Dig., LLC, 795 F.3d 654, 657 (7th Cir. 2015). Our sister circuits to have considered the implications of an unascertainable (b)(2) class have fallen into two camps, neither of which supports the narrowed class envisioned by our colleague. In the first camp, an unascertainable class definition is per se impermissible, even in (b)(2) actions. See 1 Rubenstein, Newberg on Class Actions § 3:7 n.9 (collecting cases). In the second camp, an unascertainable class definition presents no particular concern in a (b)(2) case, but only because the precise membership of a (b)(2) class is considered largely inconsequential: “the focus in a[ ](b)(2) class is more heavily 44 placed on the nature of the remedy sought, and . . . a remedy obtained by one member will naturally affect the others, [so] the identities of individual class members are less critical.” Shelton v. Bledsoe, 775 F.3d 554, 561 (3d Cir. 2015); cf. Cole v. City of Memphis, 839 F.3d 530, 542 (6th Cir. 2016); Yaffe v. Powers, 454 F.2d 1362, 1366 (1st Cir. 1972). In either camp, as long as the relief sought in a (b)(2) action is uniform and indivisible and the members of the class are otherwise similarly situated, a district court need not append to the class definition a carve-out for members who subjectively oppose the action. And while our court has not addressed whether Rule 23 contains an ascertainability requirement for class certification, regardless, defining the class to exclude minors who are ideologically opposed to abortion would raise vagueness and manageability concerns because of the difficulty of identifying which individual minors subjectively have (or may develop) those beliefs. We in no way question the sincerity and intensity of the beliefs of persons who oppose abortion, including as a matter of deep religious and moral conviction. See Stenberg, 530 U.S. at 920. As the Supreme Court has recognized, many “believe that life begins at conception and consequently that an abortion is akin to causing the death of an innocent child; they recoil at the thought of a law that would permit it.” Id. Many others “fear that a law that forbids abortion would condemn many American women to lives that lack dignity, depriving them of equal liberty and leading those with least resources to undergo illegal abortions with the attendant risks of death and suffering.” Id. Notwithstanding the strength and sincerity of the competing beliefs, for the reasons explained, we do not understand the applicable precedents and governing principles to compel finding an abuse of discretion in the district court’s conclusion that Doe and Roe are adequate representatives. 45 c. The government makes one final argument on the adequacy of the class representatives, this one pertaining to Roe alone. The government contends that she is an inadequate representative because she was 19, rather than 17 as the government previously believed, when in ORR custody and thus never a member of the class she seeks to represent. At the outset, we note that the government’s objection affords no basis to set aside the class’s certification because Doe could still serve as the class representative. In any event, even with regard to Roe alone, the district court made a factual finding that she was 17 years old based on her declaration to that effect. See Garza, 304 F. Supp. 3d at 152. The government, for its part, has introduced no evidence substantiating its claim that she was 19 during the time in question. We therefore have no basis to upset the district court’s finding and no occasion to consider the legal effect of her alleged non-membership in the class. 2. The class certified by the district court also satisfies Rule 23(a)(2)’s commonality requirement and Rule 23(a)(3)’s typicality requirement. a. The commonality requirement asks whether “there are questions of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). That requirement, the Supreme Court has explained, serves as a “guidepost[] for determining whether under the particular circumstances maintenance of a class action is economical.” Wal-Mart, 564 U.S. at 349 n.5 (quoting Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 157 n.13 (1982)). If 46 the class members’ claims involve no common question of law or fact, there will be “no cause to believe that all their claims can productively be litigated at once.” Id. at 350. Conversely, the commonality requirement is met if the class’s claims “depend on a common contention . . . of such a nature that it is capable of classwide resolution—which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Id. “A common question,” in other words, “is one in which the issue is susceptible to generalized, class-wide proof.” Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1051 (2016) (Roberts, C.J., concurring) (formatting modified and citation omitted). The presence of a single such common question can suffice to satisfy Rule 23(a)(2). See Wal-Mart, 564 U.S. at 359. Here, the class presents “common contention[s] . . . capable of classwide resolution,” such that it would be “productive[]” and “economical” to consider the class’s claims together. Id. at 349 n.5, 350. Consider the class’s core claim: that the ORR Director’s veto power, exercised as a ban, violates the class members’ protected right to choose to terminate their pregnancies before viability. That policy applies on uniform grounds applicable to every member of the class, regardless of the circumstances of her pregnancy. Not only do all class members present the same challenge to the policy, but there also is no evident variation among them concerning their ultimate entitlement to relief: if any person in the class has a meritorious claim, they all do. Correspondingly, the government’s arguments in defending against that claim cut across the entire class. The government maintains that denying a pregnant minor the choice to obtain a pre-viability abortion while in ORR custody 47 works no deprivation of her due process rights because she can seek to voluntarily depart the United States or she might be released to a sponsor. Those arguments, by the government’s logic, apply to the full class of pregnant minors in ORR custody, underscoring the existence of common classwide questions and the suitability of a classwide resolution. The government notes the existence of certain factual variations among the class members—namely, their age, maturity, stage of pregnancy, mental health, length of sponsorship search, and ability to return to country of origin. See Gov’t Br. 33–34. But the class members assert an entitlement to relief that is entirely unaffected by the myriad factual differences noted by the government. The common questions therefore are “apt to drive the resolution of the litigation.” Wal-Mart, 564 U.S. at 350 (citation omitted). b. Typicality differs from commonality in that typicality concerns the relationship between the representative’s individual claims and the class’s claims rather than the relatedness of the entire class’s claims. Yet they “[b]oth serve as guideposts for determining whether under the particular circumstances maintenance of a class action is economical.” Gen. Tel., 457 U.S. at 157 n.13. To that end, the two inquiries “tend to merge.” Id. The government’s typicality argument reiterates that Doe and Roe sought to terminate their pregnancies whereas most class members will seek to carry their pregnancies to term. But to destroy typicality, a distinction must differentiate the “claims or defenses” of the representatives from those of the class. Fed. R. Civ. P. 23(a)(3) (emphasis added). As a result, typicality is ordinarily met “if the claims or defenses of the representatives and the members of the class stem from a single 48 event or a unitary course of conduct, or if they are based on the same legal or remedial theory.” 7A Wright et al., Federal Practice and Procedure § 1764 (footnote omitted). Here, Doe and Roe allege they are subject to the same policy that operates against every pregnant minor to deprive her of her constitutionally protected right. Though absent class members may elect to exercise their choice in a different manner than Doe and Roe, the claims and defenses of the representatives are substantially—arguably entirely—identical to those of the class. And for the reasons just explained, the factual variations noted by the government in arguing against commonality also do not render the representatives’ claims atypical. Rule 23(a)(3)’s typicality standard thus is satisfied. 3. Finally, the government contends that if the class were narrowed to embrace only those pregnant UACs who seek an abortion, as the government thinks necessary, that class would not be “so numerous that joinder of all members is impracticable.” Fed. R. Civ. P. 23(a)(1). The government, though, does not argue that there is any numerosity problem with the class certified by the district court—i.e., pregnant unaccompanied minors who are or will be in ORR custody. And because we sustain that class definition, there is no numerosity objection to consider. It bears noting, though, that even as to a narrowed class of the kind contemplated by the government, the government’s numerosity argument would run into significant questions about how to account for the class’s size. The government stresses that it received 18 abortion requests in fiscal year 2017 and that classes containing fewer than 20 members generally fall short of the numerosity requirement. See Gov’t Br. 36. But there is nothing talismanic about the one-year timeframe hand- 49 picked by the government. Indeed, there presumably would be a need to account for minors who will request abortions in future years. And classes including future claimants generally meet the numerosity requirement due to the “impracticality of counting such class members, much less joining them.” 1 Rubenstein, Newberg on Class Actions § 3:15; see also Jones v. Diamond, 519 F.2d 1090, 1100 (5th Cir. 1975) (“Smaller classes are less objectionable where . . . the plaintiff is seeking injunctive relief on behalf of future class members as well as past and present members.”). Additionally, the government recently advised Congress that the number of unaccompanied minors coming to the country has dramatically risen of late. See Letter from Russell T. Vought, Acting Dir., Office of Mgmt. & Budget, to U.S. Cong. Leaders at 1–3 (May 1, 2019), https://www.whitehouse.gov/wp-content/uploads/2019/05/ Pence.pdf; Letter from Kirstjen M. Nielsen, Sec’y of Homeland Sec., to U.S. House of Representatives at 3 (Mar. 28, 2019), https://www.dhs.gov/sites/default/files/ publications/19_0328_Border-Situation-Update.pdf (“Nielsen Letter”). That significant increase may in turn give rise to an increased number of UACs seeking an abortion. That population also presents non-numerical considerations that might make joinder impracticable, including the fluidity of ORR custody, the dispersion of class members across the country, and their limited resources. See Garza, 304 F. Supp. 3d at 155. Classes have been certified in like circumstances even if they have fewer than 20 members. See Jackson v. Danberg, 240 F.R.D. 145, 147–48 (D. Del. 2007) (16 members); Bublitz v. E.I. du Pont de Nemours & Co., 202 F.R.D. 251, 255–56 (S.D. Iowa 2001) (17 members); Gaspar v. Linvatec Corp., 167 F.R.D. 51, 55–57 (N.D. Ill. 1996) (18 members); Manning v. Princeton Consumer Disc. 50 Co., Inc., 390 F. Supp. 320, 324–25 (E.D. Pa. 1975) (15 members), aff’d, 533 F.2d 102 (3d Cir. 1976). In the end, though, we need not decide whether a narrowed class would satisfy the numerosity standard. We sustain the class certified by the district court, and the government has lodged no numerosity challenge to that class. * * * The district court did not abuse its discretion in certifying a class consisting of all pregnant unaccompanied minors who are or will be in ORR custody. Both sides to the dispute thus can economically deal with an issue affecting many individuals in one fell swoop, consistent with the objectives of Rule 23. B. The presence in the class of individuals who wish to carry their pregnancies to term, for the reasons explained, does not pose an issue under Rule 23(a). But could it raise an issue under Article III of the Constitution—in particular, under Article III’s basic requirement that a plaintiff bringing an action demonstrate a constitutionally cognizable injury? See, e.g., Sierra Club v. EPA, 292 F.3d 895, 898 (D.C. Cir. 2002). Although the government does not argue the point, one could assert that a UAC who wishes to carry her pregnancy to term would lack the requisite injury to sue if she attempted to bring this challenge on her own. If so, would her status as an absent class member present a concern under Article III? We take up the question in light of our obligation to assure ourselves of jurisdiction, see Exelon Corp. v. FERC, 911 F.3d 1236, 1240 (D.C. Cir. 2018), and we see no Article III problem, even assuming that absent class members who wish to carry their pregnancies to term lack an Article III injury. 51 It is settled that in a case involving joined, individual plaintiffs bringing a shared claim seeking a single remedy, Article III’s case-or-controversy requirement is satisfied if one plaintiff can establish injury and standing. See, e.g., Rumsfeld v. Forum for Acad. & Institutional Rights, Inc. (FAIR), 547 U.S. 47, 53 n.2 (2006). In that event, it is immaterial that other plaintiffs might be unable to demonstrate their own standing. See id.; Bowsher v. Synar, 478 U.S. 714, 721 (1986); New Jersey v. EPA, 703 F.3d 110, 115 (D.C. Cir. 2012); Nat’l Mining Ass’n v. U.S. Dep’t of the Interior, 70 F.3d 1345, 1349 (D.C. Cir. 1995). The “irreducible constitutional minimum” of Article III, id. at 1355, that is, is “[a]t least one plaintiff”—and only one plaintiff—with “standing to seek each form of relief requested in the complaint,” Town of Chester v. LaRoe Estates, Inc., 137 S. Ct. 1645, 1651 (2017). The same “one plaintiff” rule presumably applies with equal force to a Rule 23(b)(2) class action advancing a uniform claim and seeking uniform injunctive and declaratory relief. There is some support for the view that an even more liberal standing rule applies in the class-action context, and that the standing of absent members is always irrelevant in class actions regardless of whether the class members seek uniform relief. See Lewis v. Casey, 518 U.S. 343, 395 (1996) (Souter, J., concurring in part, dissenting in part, and concurring in the judgment). We need not reach that question here, though, because this class action satisfies Article III even under the narrower one-plaintiff rule that governs joined claims seeking a uniform remedy. To be sure, the one plaintiff with standing must be a class representative—an absent class member’s individual standing will not suffice. See Warth v. Seldin, 422 U.S. 490, 502 (1975). But it is undisputed here that at least one class representative possesses standing. 52 Our dissenting colleague acknowledges the general rule that a single plaintiff with standing satisfies Article III in a case involving joined plaintiffs. Dissenting Op. 12. In his view, though, that understanding does not control in the context of a class action because, he submits, absent class members are full-fledged parties to a case in a way that joined plaintiffs are not. That counterintuitive proposition—if anything, one would expect joined plaintiffs to be full-fledged parties more so than absent class members—is unfounded. There is no reason to think that the one-plaintiff rule has lesser application in the class-action context than in the joined-plaintiffs context. For starters, any suggestion that absent class members (unlike joined plaintiffs) must themselves demonstrate standing is belied by the accepted understanding that only one of the class representatives needs standing. Cf. Frank v. Gaos, 139 S. Ct. 1041, 1046 (2019) (per curiam) (observing that “federal courts lack jurisdiction if no named plaintiff has standing” and remanding for a determination whether “any named plaintiff” has standing (emphasis added)). If even a class representative’s individual standing is immaterial as long as one representative has standing, an absent class member’s individual standing must also be immaterial in that instance. In addition, there is no support for our colleague’s supposition that, when there is one plaintiff with standing in a multi-plaintiff case, other plaintiffs who lack individual standing are some inferior type of quasi-party rather than full-fledged parties. To the contrary, no decision suggests that joined plaintiffs in a case, even if they lack individual standing, can avoid the preclusive effects of a judgment against them. Indeed, our court has reduced an award of attorneys’ fees to exclude the hours devoted to supporting every plaintiff’s standing because only one plaintiff with standing is needed. New Jersey, 703 F.3d at 115. Those hours would have been 53 well spent if the additional plaintiffs needed to prove their own standing to avoid some lesser, quasi-party status. Accordingly, the Supreme Court has stated, with no quasiparty proviso of the kind envisioned by our colleague, that “the presence of one party with standing is sufficient to satisfy Article III’s case-or-controversy requirement.” FAIR, 547 U.S. at 52 n.2. That rule governs here. As a result, the inclusion in the class of unaccompanied minors who desire to carry their pregnancies to term no more gives rise to an Article III concern than it poses a problem under Rule 23(a). IV. Having established our jurisdiction and having sustained the class’s certification, we finally come to the merits of the preliminary injunction granted by the district court. The decision whether to enter a preliminary injunction turns on four factors: (i) whether the plaintiff is likely to succeed on the merits of the action; (ii) whether she will suffer irreparable harm absent an injunction; (iii) whether the balance of the equities tips in her favor; and (iv) whether an injunction is in the public interest. See Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008). We review the district court’s legal conclusions de novo and its balancing of those factors for abuse of discretion. See Doe v. Mattis, 889 F.3d 745, 751 (D.C. Cir. 2018). The government asks us to set aside two components of the district court’s preliminary injunction: (i) the injunction’s central provision barring interference with a class member’s access to a pre-viability abortion; and (ii) the injunction’s prohibition against disclosing, or forcing a class member to disclose, the fact of her pregnancy and her abortion decision. We sustain the first provision protecting access to a pre- 54 viability abortion, but we set aside the disclosure-related provision and remand for further explanation. A. We first consider the portion of the district court’s order enjoining the government from interfering with or obstructing class members’ access to pre-viability abortions and abortion-related care. We conclude that the plaintiffs have established each of the four preliminary-injunction factors with respect to the abortion-access claim. 1. The first preliminary-injunction factor concerns whether the plaintiff has shown a likelihood of success on the merits. Here, the parties agree on the basic legal principles governing our review. The Supreme Court’s precedents establish “that the Constitution offers basic protection to the woman’s right to choose.” Stenberg, 530 U.S. at 921 (first citing Casey, 505 U.S. 833; and then citing Roe v. Wade, 410 U.S. 113 (1973)). In particular, “before viability the woman has a right to choose to terminate her pregnancy.” Id. (formatting modified) (quoting Casey, 505 U.S. at 870 (plurality)). The government may not “impose[] an undue burden on the woman’s decision before fetal viability.” Id. (quoting Casey, 505 U.S. at 877 (plurality)). And an “undue burden is shorthand for the conclusion that a state regulation has the purpose or effect of placing a substantial obstacle in the path of a woman seeking [a pre-viability] abortion.” Id. (alteration omitted) (quoting Casey, 505 U.S. at 877 (plurality)). Importantly, the government does not dispute the applicability of those settled principles to unaccompanied 55 minors in ORR custody. The district court held that the “[p]laintiffs’ Fifth Amendment right to decide whether to continue or terminate their pregnancies is not diminished by their status as undocumented immigrants.” Garza, 304 F. Supp. 3d at 162 n.5. The government did not contend otherwise in the district court, see id., and does not appeal the district court’s conclusion to that effect, “presumably based on its reading of Supreme Court precedent.” Garza, 874 F.3d at 753 (Kavanaugh, J., dissenting). We thus take the case on a common understanding that the class members have a constitutionally protected right to terminate their pregnancies before viability and that the government cannot impose an undue burden on that choice. Because this case involves minors, the Supreme Court’s precedents concerning parental consent also bear on our analysis. The Court has held that the government may condition a minor’s access to an abortion on parental consent only if the minor is provided a bypass mechanism. See Lambert v. Wicklund, 520 U.S. 292, 294 (1997) (per curiam); Bellotti v. Baird, 443 U.S. 622, 643 (1979) (opinion of Powell, J.). That bypass must, among other things, provide the minor an opportunity to show she is sufficiently mature to make the decision on her own (or that, immaturity notwithstanding, an abortion is in her best interest). See Ohio v. Akron Ctr. for Reproductive Health, 497 U.S. 502, 511–13 (1990); Bellotti, 443 U.S. at 643–44 (opinion of Powell, J.). The government thus acknowledges that, under the undue-burden framework, a state could not simply ban minors from choosing to terminate a pre-viability pregnancy. And the government, as noted, accepts in this case that the undueburden framework fully applies to unaccompanied minors in ORR custody. Yet there is no question that, under ORR’s challenged policy, the Director bars those minors from 56 accessing a pre-viability abortion. government defend ORR’s policy? How, then, does the First, the government contends that permitting unaccompanied minors in its custody to access pre-viability abortions requires it to “facilitate” abortions, which the government says it is not obligated to do. Second, the government asserts that unaccompanied minors may voluntarily depart the country and that the ban thus does not impose any cognizable burden. Finally, the government argues that, because many unaccompanied minors are released to sponsors, banning abortions while in ORR custody does not impose an undue burden. We address those arguments in order and conclude that the plaintiffs have shown a likelihood of success on each. Notably, our dissenting colleague likewise does not accept any of the government’s arguments for denying UACs access to a pre-viability abortion. Instead, as we explain below, our colleague believes the government should be permitted to delay access to a pre-viability abortion for a limited period in which it can continue searching for a suitable sponsor. But any delay, our colleague allows, must end if the pregnancy “come[s] close to viability.” Dissenting Op. 16–17. Our colleague therefore agrees with us in rejecting the government’s central submission: that it can deny abortion access altogether for unaccompanied minors in ORR custody. a. The government first argues that ORR’s policy is not a ban on access to an abortion but rather is a refusal to subsidize abortion. The government thereby seeks to come within the fold of Supreme Court decisions holding that an individual’s right to obtain an abortion does not generally include a right to have the government fund it. See Rust v. Sullivan, 500 U.S. 57 173, 201–03 (1991); Webster v. Reproductive Health Servs., 492 U.S. 490, 509 (1989); Harris v. McRae, 448 U.S. 297, 317–18 (1980); Poelker v. Doe, 432 U.S. 519, 519–21 (1977) (per curiam); Maher v. Roe, 432 U.S. 464, 475–77 (1977). Those decisions are grounded in the notion that the “Due Process Clauses generally confer no affirmative right to governmental aid.” Webster, 492 U.S. at 507 (quoting DeShaney v. Winnebago Cty. Dep’t of Social Servs., 489 U.S. 189, 196 (1989)). Thus, the government can fund and provide medical services incident to childbirth but not abortions, see id. at 509–10; McRae, 448 U.S. at 317–18; Poelker, 432 U.S. at 519–21; Maher, 432 U.S. at 475–77, and can prohibit recipients of federal funding from using the funds for abortion-related activities, Rust, 500 U.S. at 201–03. i. The government’s effort to situate this case in that line of precedents is misconceived. This case does not involve government funding of abortions. A preexisting policy specifies that ORR funding is not to be used to pay for abortion services (except with the Director’s approval in cases of rape, incest, or danger to the pregnant minor’s life). See Memorandum from David Siegel, Acting Dir., Office of Refugee Resettlement, on Medical Services Requiring Heightened ORR Involvement (Mar. 21, 2008), https://perma.cc/LDN8-JNL5. Jane Doe, for example, secured private funding and transportation for the abortion she sought to access. Yet Director Lloyd still denied the shelter authorization to permit Doe to leave the facility for the procedure. That is not a refusal to fund an abortion; it is a refusal to allow it. The government seeks to invoke the funding decisions in service of the more abstract proposition that it need not “commit any resources to facilitating abortions.” Gov’t Br. 40 58 (quoting Webster, 492 U.S. at 511). But even if we assume that those decisions support a general “non-facilitation” principle, they are of no help to the government here. Consider the specific government conduct at issue. The government cites the need to “maintain[] appropriate custody over the child” during the procedure and “monitor the child’s health” immediately afterward as a form of “facilitation.” Gov’t Br. 40–41. But the government is obligated to “maintain appropriate custody” over a UAC and “monitor the child’s health” regardless of any decision about an abortion. See 6 U.S.C. § 279. If those costs must be incurred regardless, and if a UAC secures private funding for an abortion procedure and associated travel expenses, what “facilitation” is left? The government points to a need to “provide direction to the shelter on its role in connection with the procedure” and to “draft and execute approval documents”—which appear to be essentially ministerial functions associated with any procedure for any medical condition. Gov’t Br. 41. Such functions are a far cry from funding of abortions. And when we consider the Supreme Court’s funding decisions alongside its decisions establishing a constitutional right to terminate a pregnancy pre-viability, we do not read the former decisions to mean that the government can compel a minor to carry her pregnancy to term against her wishes so that the government can be relieved of a self-imposed administrative requirement. In that respect, the government’s effort to extract its non-facilitation principle from the Supreme Court’s funding decisions fundamentally misapprehends those precedents. Those decisions are rooted in the idea that, “although government may not place obstacles in the path of a woman’s exercise of her freedom of choice, it need not remove those obstacles not of its own creation.” McRae, 448 U.S. at 316 59 (emphasis added). Here, though, the government asserts that it need not remove obstacles that are of its own creation. The “facilitation” the government says it wants to avoid involves constraints it imposed on itself: an unaccompanied minor’s abortion hinges on ORR’s drafting and executing approval documents only because ORR itself has conditioned abortion access on its execution of approval documents. But if the government could create such a requirement for itself and then refuse to satisfy it as a form of “non-facilitation,” the right to terminate a pregnancy before viability could be entirely subsumed by the principle recognized in the funding cases. To say that the government has an interest in declining to satisfy self-imposed constraints on abortion, then, would be to say that it has a cognizable interest in blocking abortion. That is, whereas the funding cases involved an interest in not “facilitating” abortion in the sense of not affirmatively funding abortion, here, the claimed interest in not “facilitating” an abortion amounts to an interest in not allowing abortion. But as the Supreme Court’s precedents establish, a “purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion” is by definition an impermissible, “undue burden.” Stenberg, 530 U.S. at 921 (quoting Casey, 505 U.S. at 877 (plurality)). The government’s interpretation of the funding decisions would render that understanding a nullity. For essentially the same reasons, the government is not helped by tying its non-facilitation interest to its “important and legitimate interest in potential life.” Casey, 505 U.S. at 870 (plurality) (quoting Roe, 410 U.S. at 163)). That interest, according to the Supreme Court, must be pursued through means “calculated to inform the woman’s free choice, not hinder it.” Id. at 877 (plurality). Here, though, the ORR policy 60 is calculated to hinder—indeed, disallow—an unaccompanied minor’s choice to terminate her pregnancy. The class members, in short, do not assert an “affirmative right to governmental aid.” Webster, 492 U.S. at 507 (citation omitted). They instead ask for the government to step out of the way. We thus cannot accept the government’s effort to reconceive of ORR’s no-exceptions ban on access to abortion as a mere refusal to “facilitate” abortion. ii. What the government deems “facilitation” that it wants to steer clear of giving to an unaccompanied minor, moreover, is something it willingly gives to all others in federal custody. Pregnant adults in immigration custody are held by DHS, which allows access to abortion, as does the Bureau of Prisons for federal inmates. See ICE Guidelines, Detention Standard 4.4, Medical Care, https://www.ice.gov/doclib/ detention-standards/2011/4-4.pdf; 28 C.F.R. § 551.23. The government, then, is sensitive to the asserted burden of “facilitating” abortions solely when it comes to minors in immigration custody. The result is that a 17-year-old (or even someone considerably younger) is compelled to carry her pregnancy to term against her wishes, cf. Doe v. ORR, 884 F.3d 269 (5th Cir. 2018) (per curiam) (14-year-old pregnant minor in ORR custody), whereas an 18-year-old can choose otherwise. Consider the circumstances of Jane Doe’s pregnancy. Recall that Director Lloyd barred her Texas shelter from releasing her for a (privately funded) abortion even though she qualified for an abortion under Texas law. At the time, she was at least 14 weeks pregnant and 17 years old. She turned 18 several weeks later, at a time she still would have been pregnant had ORR’s refusal to allow an abortion stood. But while the federal government permits a person in immigration 61 custody at age 18 to obtain an abortion, Doe’s pregnancy presumably would have become post-viability by that time, such that her right to seek a pre-viability abortion would have expired. The government’s policy thus would have denied her access to an abortion while she had that right, only to allow her access when it was too late. Or consider the implications of the government’s on-and-off interest in “non-facilitation” if Doe’s pregnancy still would have been pre-viability when she turned 18. In that situation, ORR’s policy would have simply forced her to wait a significant number of weeks to obtain a pre-viability abortion, pending her inevitable transfer from one form of government custody (ORR) in which abortion is barred to another (DHS) in which it is allowed. Why require her to wait several weeks for a later-term abortion that is sure to occur rather than permit her to obtain it earlier in her pregnancy? Solely so that the procedure then happens on DHS’s watch rather than ORR’s. Those unjustified and anomalous disparities substantially undercut the government’s reliance on its ostensible interest in non-facilitation. And they reinforce the inapplicability of the funding decisions in this case. b. The government next contends that, even if ORR’s policy works as a ban on access to an abortion rather than as a mere withholding of funding, an unaccompanied minor can easily avoid the ban by seeking voluntary departure from the United States. And because she could readily avoid the ORR ban by making use of voluntary departure, the argument goes, the ban does not impose any cognizable burden—much less an undue burden—on her choice to terminate her pregnancy. The government’s argument is misguided. 62 To see why, think about the government’s argument in the context of the Supreme Court’s decision in Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292 (2016). There, the Court invalidated Texas laws that had the effect of significantly reducing the number of available abortion providers in the State and forcing women to travel longer distances for an abortion without an offsetting health benefit. The Court held that the laws imposed an undue burden on access to abortion. But under the logic of the voluntary-departure argument the government now advances here, the Texas laws struck down in Whole Woman’s Health in fact imposed no undue burden because women desiring an abortion could always travel to another state. That sort of argument presumably could not have carried the day in Whole Woman’s Health. And the government’s voluntary-departure argument cannot carry the day here either. A state could not ban abortions outright on the theory that pregnant women can just go elsewhere. And the federal government has no greater leeway to ban abortions on the theory that pregnant women can go elsewhere via voluntary departure. In a number of respects, in fact, the voluntarydeparture argument pressed by the government here is even less tenable than an (already-unacceptable) argument that the Texas laws invalidated in Whole Woman’s Health should have been sustained because a person could leave the State. To begin with, the government’s voluntary-departure argument is grounded in an assumption that an unaccompanied minor can easily avoid the ORR ban by simply departing the country. The ready ability to leave the jurisdiction may be true of a Texas resident: she would be free to leave the State (at least as a legal matter) and go elsewhere to access an abortion. Voluntary departure, though, does not work that way. 63 Recall that voluntary departure is a form of immigration relief granted only at the discretion of the government. See p. 10, supra; 8 U.S.C. § 1229c(a)(1) (“The Attorney General may permit an alien to voluntarily depart . . . .” (emphasis added)). That is because voluntary departure leaves an individual better off than if she were removed from the country. See Dada, 554 U.S. at 8. So whereas a Texas resident can leave the State without needing any approval, an unaccompanied minor could obtain voluntary departure only if the government grants it. Voluntary departure, then, is not a freely available escape hatch from a government veto on abortion. It is instead a second government veto. Relatedly, a Texas resident could choose how soon to leave the State to access an abortion elsewhere. That is significant because the right to a pre-viability abortion has an inherent expiration date. And even before that date, forcing a person to delay obtaining an abortion itself entails an escalating burden and risk over time. With voluntary departure, though, it is not just that the government can choose when (and whether) to grant relief. It is also that the government has given no sense of how long the process can (or usually does) take from the time a person seeks the relief to the time the arrangements can come into order for her departure to another country. The process might well take considerable time in a context in which time is of the essence. In the case of Jane Doe, for instance, government correspondence in the record observed that her voluntary departure case was “not likely to be far [along] at all given [her] recent referral date” to ORR, indicating that voluntary departure, though possible, is far from instantly available. Email from Jonathan White, Deputy Dir. for Children’s Programs, to Scott Lloyd, Dir., Office of Refugee Resettlement (Sept. 22, 2017), P.A. 26. 64 What is more, any suggestion that a ban on abortion works no impermissible burden because a person can freely go elsewhere necessarily assumes the availability of abortion in the destination. That might be true of a Texas resident traveling to another state. But that is not true in the case of a pregnant unaccompanied minor in ORR custody: even assuming the government were to grant her voluntary departure and could expeditiously arrange for her return to her country of origin, she very likely could not obtain an abortion there. That is because virtually all minors in ORR custody— more than 90%—come from Honduras, Guatemala, or El Salvador, the so-called Northern Triangle countries. See ORR Fact Sheet 2. Abortion is criminalized in all three countries, so much so that only Guatemala even provides an express exception for a threat to the life of the pregnant woman. See World Health Org., Global Abortion Policies Database, https://abortion-policies.srhr.org/ (last accessed May 14, 2019); see also Immigrant Rights Advocates Amicus Br. 4, 18. It is no surprise, then, that the government conceded in Jane Doe’s case that abortion would have been unavailable to her in her country of origin. In light of the unavailability of abortion in unaccompanied minors’ countries of origin, the supposed accessibility of voluntary departure could not be seen to alleviate the burden imposed by ORR’s ban. Finally, even if voluntary departure would in fact pave the way for a UAC to obtain an abortion in her country of origin, it comes at a significant cost. Voluntary departure requires withdrawing any claims for other forms of immigration relief. See 8 C.F.R. § 1240.26. And according to ORR, “[m]any unaccompanied alien children meet conditions that make them eligible for legal relief to remain in the United States including[] but not limited to asylum; special visas for children who have been abused, neglected[,] or abandoned by the 65 parents or guardian; special visas for victims of severe forms of trafficking and other types of crime; or adjustment of status for those who have a legal resident or citizen family member.” ORR, UAC Services; see also Immigrant Rights Advocates Amicus Br. 4, 17–23. In that context, requiring an unaccompanied minor to carry her pregnancy to term unless she waives any claim for relief to stay in the United States is itself a substantial burden on her exercise of her rights. See Garza v. Hargan, No. 17-5236, 2017 WL 9854555, at *3–4 (D.C. Cir. Oct. 20, 2017) (Millett, J., dissenting). Jane Poe’s case vividly illustrates the point. Recall that she became pregnant as the result of rape, but Director Lloyd still refused to authorize an abortion, forcing her to go to court to obtain an order (over the government’s opposition) enabling her to terminate her pregnancy. She ultimately received a grant of asylum entitling her to stay in the United States. According to the government’s voluntary-departure theory, she should have been forced to choose between (i) carrying her unwanted pregnancy to term even though it resulted from rape, and (ii) returning to her country of origin, a place where, according to the government’s own determination in granting her asylum, she faced a well-founded fear of persecution. Putting her to that choice, to say the least, amounts to a “substantial obstacle in the path of a woman seeking an abortion.” Stenberg, 530 U.S. at 921 (quoting Casey, 505 U.S. at 877 (plurality)). For those reasons, we cannot accept the government’s effort to justify ORR’s ban on access to abortions on the theory that unaccompanied minors can voluntarily depart the country. The undue-burden framework has never been thought to tolerate any burden on abortion the government imposes simply because women can leave the jurisdiction. That is especially so for voluntary departure, which: is granted only at the government’s discretion; may not come soon enough even 66 if awarded; is exceedingly unlikely to enable an unaccompanied minor to obtain an abortion in her country of origin in any event; and requires abandoning potentially viable claims of entitlement to stay in the United States. c. The government’s last defense involves the prospect that an unaccompanied minor could be released to a sponsor, at which point ORR would no longer prevent her from obtaining an abortion. Because of that possibility, the government argues, the ban on accessing an abortion while in ORR custody does not impose an undue burden. i. The government’s sponsorship argument, in our view, is ultimately no more persuasive than its voluntary-departure one. Those arguments share important parallels. In both, the central idea is that an unaccompanied minor may find herself no longer in ORR custody—either because she voluntarily departs the country or because she is released to a sponsor—in which event she would be free to access an abortion without the burden of ORR’s policy. Some of the same deficiencies that require rejecting the government’s voluntary-departure argument also undermine its sponsorship argument. First, as with voluntary departure, sponsorship is not simply there for a UAC’s taking. Rather, release to a sponsor is at the discretion of the government and is contingent on factors outside the UAC’s control—most importantly, the existence of a willing and qualified sponsor. The process of approving a suitable sponsor is understandably an involved one. As ORR explains, “the safe and timely release of an unaccompanied alien child from ORR custody involves many steps including: the identification of sponsors; the submission by a sponsor of the application for release and supporting documentation; the evaluation of the suitability of the sponsor, 67 including verification of the sponsor’s identity and relationship to the child, background checks, and in some cases home studies; and planning for post-release.” ORR Guide § 2.1. As with voluntary departure, moreover, there is no guarantee that release to a sponsor—if it occurs at all—would happen in a timely fashion, even though the right to a pre-viability abortion carries a finite endpoint and its delayed recognition entails an accumulating burden and risk over time. According to recent ORR data, the average length of time a UAC spends in ORR custody is roughly 90 days. ORR Fact Sheet 2. And because that is merely an average, a sizable share of unaccompanied minors will spend more time than that in ORR custody before any release to a sponsor, with some never released to a sponsor because a suitable one is never found. Jane Doe, for instance, was released to a sponsor after months in ORR custody, at a time when she would have been at least some 26 weeks pregnant had she not been permitted to terminate her pregnancy under court order. And Jane Poe remained in ORR custody for approximately one year without ORR ever locating a suitable sponsor. See also Santos v. Smith, 260 F. Supp. 3d 598, 603–04 (W.D. Va. 2017) (no release for 29 months). In defending its sponsorship argument, the government relies on Supreme Court decisions sustaining judicial-bypass regimes that can impose some delays on a minor’s ability to obtain an abortion absent parental involvement. The government notes that, in one decision, the Court upheld against a facial challenge a bypass procedure for minors that could have taken up to 22 days to complete (in an improbable, “worst-case analysis that may never occur”). Akron Ctr. for Reproductive Health, 497 U.S. at 514. Here, though, the average length of time an unaccompanied minor remains in 68 ORR custody before release to a sponsor is more than four times that number of days; and the worst-case scenario is no release, ever (at least until age 18, at which point a person would be transferred to DHS custody as an adult). Apart from the arithmetic, there is a more fundamental flaw in the government’s effort to analogize the time before release to a sponsor to the time to secure a judicial bypass. The government does not contend only that, for that subset of minors who can gain placement with a sponsor within a given period (e.g., in less than 22 days), it would be constitutional to delay access to an abortion to them for that time. Instead, the government’s argument extends to all minors in ORR custody, regardless of whether they are ever released to a sponsor. In the government’s apparent view, the possibility that some minor could be released to a sponsor quickly enough to obtain a pre-viability abortion justifies a ban on every minor for the entire time she is in ORR custody. Thus, Jane Poe, for whom a suitable sponsor was never found, can be compelled to carry to term a pregnancy resulting from rape, because someone else might be more fortunate in securing release to a sponsor. In other words, that lightning might strike for some, the government evidently believes, means it can deny an abortion to all. That of course cannot be so. Under that logic, a state could enforce a blanket requirement of parental consent even without affording any judicial bypass because at least some parents might change their minds. But see Bellotti, 443 U.S. at 643 (opinion of Powell, J.); id. at 654 (opinion of Stevens, J.). The abstract possibility of release to a sponsor thus affords no basis to deny access to abortion across the board. It bears emphasis, moreover, that the judicial-bypass context fundamentally differs from the sponsorship context. The delay from a judicial bypass is justified by the need to 69 ensure that a minor is sufficiently mature to decide to terminate her pregnancy without parental involvement. See, e.g., id. at 636 (opinion of Powell, J.). That interest was served in Jane Doe’s case, for instance, when she obtained a bypass from a Texas court that found her competent to make her own choice to terminate her pregnancy. By contrast, ORR’s policy here has been justified only on the basis of the government’s interest in avoiding “facilitation.” And nothing in the record suggests that ORR had in mind any potential benefits of sponsorship for unaccompanied minors when it instituted and applied the ban. Nor does the structure of ORR’s policy suggest any aim to provide pregnant UACs with any benefits of sponsorship when deciding whether to terminate a pregnancy. The search for a sponsor begins for every unaccompanied minor at the moment she (or he) enters ORR custody, and there is no intensified effort to identify a sponsor just because a minor is pregnant and considers an abortion. Plus, far from improving the conditions for a minor’s decision-making, barring abortion access unless a sponsor is found would deny some minors the ability to make a decision at all (in cases like Jane Poe’s, in which no suitable sponsor is identified before viability). When Director Lloyd explained his denial of Poe’s request for an abortion, he thus made no mention of sponsorship or its potential advantages. Our dissenting colleague suggests that, regardless of whether Director Lloyd (or anyone in the government) considered the advantages of sponsorship when establishing or implementing the ORR policy, the government in its briefing before us has now referenced the “benefits of adult guidance” afforded by sponsorship and has argued “that sponsorship does provide that benefit to a pregnant minor.” Dissenting Op. 17. That may be so, but it is beside the point. Everyone agrees that sponsorship benefits UACs: that is why the government seeks release to a sponsor for all UACs, pregnant or not. The salient 70 question, though, is whether the challenged ORR policy—i.e., the ban on abortion access—is adequately aimed to (or operates to) realize those benefits of sponsorship for unaccompanied minors, even though it denies them abortion access without regard to the prospects of their release to a suitable sponsor. For the reasons explained, we think the answer is no. In short, nothing in the ORR policy’s design or operation suggests that its purpose or effect is to confer upon UACs deciding whether to terminate a pregnancy the benefit of adult consultation and support. The government’s justification for the policy instead has consistently been the same one it invokes with respect to voluntary departure—i.e., the interest in withholding government “facilitation” of abortions while a pregnant minor is in ORR custody. In the context of that interest, the government can no more deny her abortion access based on the abstract possibility of sponsorship than it can do so based on the abstract availability of voluntary departure. ii. The government suggests a fallback variation of its sponsorship argument in its reply brief: even if the abstract possibility of release to a sponsor does not allow the government to bar abortion access outright in all instances, the district court at least should have narrowed its preliminary injunction to give the government additional time to find a sponsor “when it is expeditious.” Gov’t Reply Br. 18. Our dissenting colleague accepts a form of that argument. In his view, when a UAC “says she wants an abortion,” the government should be allowed to delay abortion access for a “limited time” while it continues searching for a sponsor; but the delay in all events would need to end before the pregnancy “come[s] close to viability,” at which point the government would have to allow an abortion. Dissenting Op. 16–17. Our colleague does not specify the length of the “limited time,” but says that it would need to be “expeditious” and that “up to a 71 three-week delay ha[s] been recognized as expeditious” in the context of a judicial bypass. Id. That species of fallback argument does not advance the government’s cause. As an initial matter, the government has not provided a sound reason for any delay in the circumstances of this case, as we have already rejected its reliance on the one interest it presses—non-facilitation. And in any event, it is far from clear that the government could justify delaying abortion access for all minors for an additional period of up to three weeks on the mere chance that placement with a sponsor could occur in that window for someone. The government would have been looking for a sponsor for every minor, all along, and would continue to do so with the same intensity regardless of an abortion request. There is then no reason to suppose that the search will happen to yield a sponsor in an additional three weeks often enough to justify imposing a blanket delay of that duration—and the government has not attempted to make any argument or showing to that effect. Our dissenting colleague nonetheless asserts that the sponsorship process is in fact affected by an abortion decision. He supposes that the government’s search for a sponsor “surely” is “accelerated” when an unaccompanied minor says she wants an abortion. Dissenting Op. 16. But there is no reason to think our colleague’s speculation is accurate, and every reason to think it is not. The government has never suggested, in any materials submitted in this case or in any of its published guidance on the sponsorship process, that its efforts to find a sponsor go from half-hearted to full-hearted if it learns that a particular UAC desires an abortion. Rather, the imperative to find a suitable sponsor exists for all unaccompanied minors from the moment they enter ORR custody until a sponsor is found because 72 release to a sponsor is desirable in all instances, regardless of pregnancy (or abortion). The government thus continued trying to find a sponsor for Jane Doe after she obtained an abortion, with no evident deceleration in its (eventually successful) efforts. Indeed, even as to a minor who is pregnant, release to a sponsor presumably would be at a premium not just if she says she wants an abortion, but also if she does not: she can then have the direct assistance and support of a family member or other responsible adult as she proceeds to delivery. Our colleague’s proposed up-to-three-week delay when a minor says she wants an abortion has no relation to anything we know about the sponsorship process. Though the Supreme Court upheld a bypass mechanism that could take up to three weeks to complete in a worst-case circumstance, Akron Ctr. for Reproductive Health, 497 U.S. at 514, the delay was justified by the particular interest served by the bypass and the time needed for that process to run its course. There is nothing sacrosanct, though, about a three-week waiting period. The Court did not hold that any three-week delay advancing any interest to any degree is invariably constitutional. Here, without knowing whether, why, or how often an extra three weeks is likely to result in placement of minors with sponsors, we cannot weigh a three-week delay’s putative benefits against its burdens. See Whole Woman’s Health, 136 S. Ct. at 2309. The three-week figure has simply been plucked from the judicial-bypass context and rotely transposed here, rather than justified based on a three-week delay’s purported benefits pertaining to identifying a sponsor. At any rate, even assuming the government could justify some additional period of delaying access to an abortion that is tied to its efforts to identify a sponsor, the district court had no obligation to fashion its interim injunction around that 73 possibility. True, “an injunction must be narrowly tailored to remedy the harm shown.” Gulf Oil Corp. v. Brock, 778 F.2d 834, 842 (D.C. Cir. 1985). That principle, however, does not require district courts enjoining unconstitutional government policies to fashion narrower, ostensibly permissible policies from whole cloth. After all, what might a narrower policy look like here? The government never said, and the district court did not need to propose a range of conceivable polices to the government until the government found one to its liking. For instance, perhaps in lieu of the blanket ban the government sought to defend, it might consider a regime of the kind hypothesized by our dissenting colleague, under which it would delay access to an abortion for a given period in which it might somehow augment its ongoing efforts to find a sponsor, with viability as a backstop. Our colleague traces that approach to then-Judge Kavanaugh’s opinion in an earlier stage of the proceedings in this case. Dissenting Op. 15. But after that opinion, the government at no point asked the district court in the ensuing proceedings to consider crafting a remedy that would give the government some extra time period to find a sponsor upon learning of a UAC’s interest in an abortion. Nor did the government suggest the contours of any such approach. The district court, then, did not abuse its discretion by leaving such a regime out of its preliminary injunction. What, for instance, would be an appropriate amount of additional time to identify a sponsor? And if the start time, per our colleague’s proposal, is when a minor requests an abortion, would the extra time to find a sponsor vary depending on the amount of time remaining until viability, which in turn might vary depending on the particularities of state law? Additionally, would there be a fixed period of delay for all unaccompanied minors regardless of the prospects of identifying a sponsor, as our dissenting colleague appears to 74 presume, or would delay be warranted only for those minors for whom the prospect of expeditiously identifying a sponsor is sufficiently promising? If the latter, by what criteria would those minors be identified? And by what proper purpose would the government justify an additional period of delay, given that, as explained, it has yet to assert such a purpose? This case, in short, is not one in which crafting a narrower remedy is a “relatively simple matter.” Ayotte v. Planned Parenthood of N. New Eng., 546 U.S. 320, 329 (2006) (quoting United States v. Treasury Emps. Union, 513 U.S. 454, 479 n.26 (1995)). To the contrary, devising a narrower remedy would have entailed “making distinctions in a murky constitutional context . . . where line-drawing is inherently complex” and is laden with the kinds of policy judgments a court typically does not make on its own. Id. For instance, if a state were to enact a blanket ban on abortions for minors without any bypass, a court of course could simply enjoin the ban without needing to fashion (and carve out from its injunction) a permissible bypass regime that the state might someday wish to construct. So too here. Confronted with a blunt policy barring access to an abortion across the board for anyone in ORR custody, the district court understandably issued a corresponding remedy. d. i. The district court’s remedy does not compel “abortion-on-demand,” contrary to the government’s (and our dissenting colleague’s) characterization. See Gov’t Reply Br. 14; Dissenting Op. 18. With or without the preliminary injunction, an unaccompanied minor can obtain an abortion only if state law permits it. That is why Jane Doe needed to obtain a judicial bypass from a Texas court. A regime in which an unaccompanied minor needs to secure a judicial bypass 75 under Texas law cannot be seen as “abortion-on-demand” unless Texas law itself is seen to allow abortion-on-demand. What is more, even if we set aside the need to satisfy state law and consider only the federal government’s efforts to add its own constraints, nothing in our analysis necessarily prevents the government from attempting to regulate UACs’ abortion access under a different policy consistent with the Supreme Court’s precedents. If the government wishes to devise a narrower policy than a blanket veto and ban (including one tethered to the sponsorship process), it can do so and then test the policy in the courts. But until then, the district court is not obligated to undertake the task of chiseling from the government’s across-the-board ban a different policy the government never identified, endorsed, or defended. ii. As something of a last resort, the government asserts that upholding the district court’s decision “would constitutionally mandate what would amount to abortion tourism, where minors who cannot obtain abortions lawfully in their country of nationality demand abortion services at our border or upon illegal entry to our country.” Gov’t Br. 45. The government’s supposition that the possibility of a lawful abortion would alone cause unaccompanied minors to attempt the journey here is unsupported. As ORR itself explains, “[u]naccompanied alien children have multiple inter-related reasons for undertaking the difficult journey of traveling to the United States, which may include rejoining family already in the United States, escaping violent communities or abusive family relationships in their home country, or finding work to support their families in the home country.” ORR, UAC Services. And the “age of these individuals, their separation from parents and relatives, and the hazardous journey they take make unaccompanied alien 76 children especially vulnerable exploitation[,] and abuse.” Id. to human trafficking, Those perils have become only more pronounced. A recent letter from the DHS Secretary to Congress advises that “[r]eports of violence and sexual assault along the route are now pervasive, meaning that many arriving migrants require especially focused care. In some cases, girls as young as 10 years old in DHS custody require pregnancy tests so we can make sure they get essential medical support.” Nielsen Letter 2. That hazardous journey for minors is not “tourism,” much less “tourism” to “demand abortion.” Gov’t Br. 45. More fundamentally, even if the availability of constitutional rights in this country affords an inducement to attempt that journey, we are unable to accept the government’s argument. It is difficult to imagine the government arguing, say, that unaccompanied minors should be denied the right to freely worship while in ORR custody because those denied religious liberty in their native countries might otherwise be enticed to come to the United States. And correspondingly, we cannot accept the suggestion that minors in ORR custody should be compelled to carry pregnancies to term against their wishes—even in cases of rape—so that others will be deterred from desiring to come here. To be sure, the “right to an abortion” is viewed to have a “controversial nature,” as to which people “sincerely hold directly opposing views.” Stenberg, 530 U.S. at 920–21. But the Supreme Court “has determined and then redetermined that the Constitution offers basic protection to the woman’s right to choose.” Id. at 921. And we are not free to dilute a constitutional right recognized by controlling Supreme Court precedent—a right the government affirmatively assumes 77 unaccompanied minors here have—so that others will be dissuaded from seeking a better life in this country. * * * We therefore agree with the district court that the plaintiffs have shown a likelihood of success in connection with their claim that ORR’s restriction on abortion access infringes their protected right to choose to terminate their pregnancies. 2. The district court, having found a likelihood of success on the merits of plaintiffs’ abortion-access claim, made relatively quick work of the remaining preliminary-injunction factors. The court held that: (i) the class would be subject to irreparable harm in the absence of an injunction, including, “at a minimum, increased health risks, and perhaps the permanent inability to obtain the abortion to which they are legally entitled,” Garza, 304 F. Supp. 3d at 165; (ii) the balance of the equities favors the plaintiffs because the government lacks “any legitimate interest that will be harmed by the issuance of a preliminary injunction,” id.; and (iii) the public interest favors a preliminary injunction because the case “involves the protection of constitutional rights” and “the public has an interest in the government maintaining procedures that comply with constitutional requirements,” id. (citation omitted). We see no basis to set aside the district court’s assessment of those remaining preliminary-injunction factors. The government argues that the plaintiffs’ class will not suffer irreparable harm absent a preliminary injunction. Noting the district court’s grant of temporary restraining orders to Jane Doe, Jane Poe, and Jane Moe, the government contends that the court could similarly deal with abortion requests on a rolling, emergency basis as they arise in the course of the 78 litigation. But it is precisely because of a risk of irreparable harm that an individual litigant would be entitled to emergency relief in a given case. And the government provides no reason why the district court was required to privilege one form of interim relief that individual class members could seek in the future (one-by-one temporary restraining orders) over another form of interim relief that the class seeks now (a classwide preliminary injunction). Efficiency considerations, if nothing else, plainly favor the latter. Next, the government contends that the “government (and public) have a legitimate and significant interest in protecting potential life and therefore refusing to affirmatively facilitate abortions that the Constitution does not require.” Gov’t Br. 49. And the public interest, the government continues, “weighs against incentivizing illegal immigration.” Id. The first argument assumes the correctness of the government’s position on the merits (which we have found is unlikely to succeed). And as to the second argument, there is no evidence supporting the idea that pregnant unaccompanied minors make the journey to the country and attempt to (unlawfully) enter solely for the sake of obtaining a (lawful) abortion. Finally, the government submits that “the public interest weighs heavily in favor of allowing ORR to fulfill its statutory obligations,” and notes that “ORR must assume a de facto parent role for [pregnant UACs], who often lack the maturity to independently make such crucial life decisions.” Id. at 49– 50. Consequently, the government contends, it “is in the public interest to allow ORR to fulfill that obligation, including by assisting these minors with navigating complex moral, mental, and physical issues they face.” Id. at 50. But ORR’s policies are not calculated to assist minors with navigating the decision whether to continue (or terminate) a pregnancy. Instead, ORR’s policies aim to block a decision in one direction alone 79 until ORR can absolve itself of its de-facto parental role via either voluntary departure or release to a sponsor (or release to DHS when a minor turns 18). There was no reversible error in the district court’s consideration of the public interest. * * * Because the plaintiffs have made a sufficient showing that the various preliminary-injunction factors weigh in their favor, we affirm the district court’s order preliminarily enjoining the government from interfering with unaccompanied minors’ access to a pre-viability abortion. B. We consider next that portion of the district court’s order enjoining the government from revealing (or from forcing the plaintiffs to reveal) “the fact of their pregnancies and/or abortion decisions to anyone.” Prelim Inj. Order (Apr. 16, 2018), G.C.A. 275. The plaintiffs allege that ORR maintained a practice of disclosing, or of coercing class members to disclose, information about their abortion decisions and pregnancies to their parents (and, in some cases, to potential sponsors). They argued that the disclosure practices violate their rights under the First and Fifth Amendments. The district court evidently agreed, and the government now appeals that aspect of the preliminary injunction. We decline to reach the merits of that challenge, however, because we lack an adequate understanding of the content of the disclosure policies we would be charged with reviewing. The district court, in the portion of its decision granting certification of the class, noted that the “[p]laintiffs have offered evidence to support their contention that” ORR maintains “a series of restrictions that apply to all pregnant UCs in its custody,” including “a parental notification 80 requirement that cannot be bypassed.” Garza, 304 F. Supp. 3d at 156. But while the court observed that the plaintiffs had “offered evidence” supporting a parental notification requirement, the court did not make an ultimate finding on whether such a requirement in fact exists. Nor did the court make findings about the parameters of the alleged notification requirement. For instance, there is no finding or statement speaking to whether parental notification pertains to (or is triggered by) an abortion request, or whether notification may also encompass the fact of a minor’s pregnancy regardless of an abortion request. There is also no finding addressing whether (and to what extent) ORR undertakes a case-by-case determination concerning a minor’s best interests before notifying parents, as the government claims, or whether any notification instead occurs on a blanket basis. Nor is there a finding concerning whether and to what extent ORR is in the practice of disclosing abortion requests (or the fact of pregnancy) to prospective sponsors as well as parents, and whether the ORR disclosure practice is the same both pre- and post-viability. The answers to those sorts of questions would inform the constitutional analysis and bear on its contours. And the answers would enable us to ensure that we avoid issuing an advisory opinion about aspects of an alleged disclosure policy that do not in fact exist. In addition, a remand would allow the district court to identify which of the plaintiffs’ three constitutional theories about the alleged disclosure policies (a Fifth Amendment claim under Planned Parenthood v. Casey, a Fifth Amendment “informational privacy” claim, or a First Amendment claim) promises a likelihood of success on the merits and why. As it stands, the court’s legal analysis speaks to ORR’s challenged interference with abortion access without separately discussing the alleged disclosure polices. Nor did 81 the decision address the other preliminary-injunction factors as they may specifically pertain to the alleged disclosure policies. We appreciate the pressure on the district court to move with haste, particularly given the nature of the claims and the multiple prior rounds of emergency litigation. We also acknowledge the breadth of the legal issues raised—as the length of this opinion manifests—and the understandable desire to streamline the order. And the parties did not always draw clear distinctions between the access restriction and alleged disclosure mandate in their arguments. Nonetheless, we are unable to meaningfully review the portion of the district court’s preliminary injunction addressed to the alleged disclosure policies without a more developed understanding of the content of those polices. We thus vacate the portion of the district court’s order preliminarily enjoining the government from disclosing (or forcing class members to disclose) information related to their pregnancies and abortion decisions, so that the district court on remand can give a more fulsome account of its findings and conclusions in that regard. * * * * * For the foregoing reasons, we affirm the class certification order and the portions of the preliminary injunction enjoining obstructions to abortion access. We vacate those portions of the order relating to ORR’s alleged disclosure policies and remand for further explanation. So ordered. SILBERMAN, Senior Circuit Judge, dissenting: I should note at the outset my disagreement with the majority regarding the propriety of the district judge’s certification of the class. Three of the four factors to be considered in Rule 23(a) governing class actions – commonality, typicality, and adequacy – “tend to merge,” as Justice Scalia once observed. Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 349 n.5 (2011) (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 157-58 n.13 (1982)). The crux of my dispute on this issue is that I believe that the class is much too broad; it should not include pregnant minors who do not wish an abortion, whether you refer to that as a violation of commonality, typicality, or adequacy. Indeed, I believe that the class not only violates Rule 23(a), it also violates the Constitution since it includes many class members who do not claim injury-in-fact. But before I reach that matter, I also disagree with the majority regarding another constitutional question, as to whether this case is moot, because the class representatives’ claims are moot, or whether mootness is avoided because of the exception for claims that are “inherently transitory.” Finally, since the majority decides these procedural questions contrary to my views and reaches the merits, so do I; for I disagree on the merits as well. I. It is common ground in this case that the only relevant exception to the mootness doctrine is based on the inherently transitory exception. That is to say the claims must be inherently transitory. Generally, the cases in which Supreme Court majorities have recognized this exception have been cases in which the class members’ alleged legal violations have, from the outset of the claim, been expected to last only days or a very brief period of time – too short a period for a conscientious 2 district judge to be expected to rule on a motion for certification. See County of Riverside v. McLaughlin, 500 U.S. 44, 47, 50-52 (1991). Therefore, even if the class representatives’ claims are mooted, the class of plaintiffs remains so long as members still have live claims. Which presents the question: What period is too short? My colleagues put it in terms of whatever amount of time would be “reasonable,” keeping in mind the demands placed on a busy judge. I am afraid that is a way of turning any injunctive claim that is indefinite into an inherently transitory claim. The exception would swallow the rule. Determining whether a claim is inherently transitory requires us to judge whether the claim on its face (no matter when the certification motion is actually heard) was one that a hypothetical judge would not have had adequate time to hear and decide the certification issue. In other words, the claim must be evaluated ex ante – when made – not in the context of events that transpire after the claim. The words “inherently transitory,” therefore, necessarily suggest apparent fleeting claims. The Supreme Court recently seemed to indicate my analysis is correct – it said “this doctrine has invariably focused on the fleeting nature of the challenged conduct giving rise to the claim.” Genesis Healthcare Corp. v. Symczyk, 569 U.S. 66, 76-77 (2013). My difficulty with the majority’s formulation is that it seems to suggest that a judge may certify a class despite the mootness of the representatives’ claims if the judge has acted within a reasonable time given his or her schedule – an ex post approach which would be virtually impossible to review. Moreover, the number of variables implicit in that approach is inconsistent with the term “inherently transitory.” See Contractors’ Labor Pool, Inc. v. NLRB, 323 F.3d 1051, 1057 (D.C. Cir. 2003) (concluding that the word “inherently” precludes reliance on “independent variables”); see also Wilson 3 v. Gordon, 822 F.3d 934, 959 (6th Cir. 2016) (Sutton, J., dissenting) (“The question is whether the claim has a built-in expiration date.”). Turning to the claims at issue, they were certainly not fleeting. As the majority points out, according to recent data, the average stay in custody for minors was roughly ninety days. Op. at 67. This is hardly too brief for judicial action on the certification motion. The majority points to age, sponsorship, and voluntary departure as events that might end the claims of class representatives. Op. at 26. Yet the Appellees dispute quite vigorously the availability and operation of sponsorship and voluntary departure. Indeed, the district court’s findings certainly suggest that these factors do not support the application of the exception. The court found sponsorship to be “typically a lengthy, complex process involving multiple stages.” Garza v. Hargan, 304 F. Supp. 3d 145, 164 (D.D.C. 2018). Nor do Appellees argue that sponsorship happens quickly, see Appellees’ Br. 45, or suggest that voluntary departure is a significant option for pregnant minors, see id. at 44. Of course, there might be some movement out of the class, but that does not ex ante make the claims of putative class representatives fleeting for purposes of the inherently transitory exception. It should be recalled that the inherently transitory exception has a second requirement that there be a constant class of individuals with live claims (this is, of course, related to the Rule 23(a) numerosity requirement, see infra note 2). I have rather serious doubts that this prong is met either. The majority emphasizes that the Government does not contest this point, but since it goes to Article III standing, we have an obligation to examine the matter ourselves. The district court merely determined that “the claims of numerous potential class 4 members remain unaddressed.” Garza, 304 F. Supp. 3d at 160. Similarly, Appellees rely on the large size of the class, but if I am correct that the class is much too large, this requirement also becomes problematic. The majority opinion, on the other hand, relies on a statement at oral argument that only “about a dozen [minors] expressed an interest in abortion or related information” in the six-month period after the issuance of the injunction. Op. at 28. I do not think that such a small number is sufficient to assure the court of a continuing existence of a class of individuals with live claims for purposes of the inherently transitory exception, and we are obligated to be “certain” of the existence of a live class. See Gerstein v. Pugh, 420 U.S. 103, 110 n.11 (1975). Candidly, I am troubled by a recent Supreme Court decision, Nielsen v. Preap, 139 S. Ct. 954 (2019), on which my colleagues rely. There, a three-Justice plurality, after determining that at least one class representative’s claim was not moot, went on to state that all the claims of the class were “transitory.”1 Id. at 963 (plurality). Contrary to the majority’s suggestion that the plurality concluded that a claim lasting one year was too brief, Op. at 27, the plurality never mentioned the amount of time at stake. The transitory exception apparently applied because the class representatives’ claim directed to the Government’s failure to give a bond hearing to detained aliens could be obviated at any time if they received a decision on removal. Preap, 139 S. Ct. at 963 (plurality). Rather surprisingly, the word “inherently” is missing from the plurality’s terse statement. (I can’t believe that the Court without any explanation meant to drop the 1 The Ninth Circuit’s statutory analysis was a terribly tempting target. Sometimes the Court is more driven by issues than procedural limitations. 5 inherently transitory test.) The facts, however, are quite similar to the original case, Gerstein, but since the average length of time of detention in Preap was one year, it can hardly be called fleeting, which is what Justice Thomas pointed out in his concurrence. Id. at 976 (Thomas, J., concurring in part and concurring in the judgment). That leads me to believe that although the plurality provided us with no analysis, its justification for ignoring the lengthy average time of detention was the Government’s complete control of the timing of any removal – which, in some cases, could be fleeting. That isn’t true in our case. Two factors affect the timing of the Appellees’ claims – voluntary departure and sponsorship. But as I noted, obviously the Government cannot unilaterally effect a departure, and finding an appropriate sponsor is hardly in the Government’s sole control. In sum, the majority’s imaginative employment of the limited exception to mootness – inherently transitory claims – blows a hole as wide as the Haleakala Crater in a constitutional restriction on judicial power. II. The Government asserted that the class representatives have antagonistic or conflicting interests with members of the certified class, i.e., those pregnant minors who do not wish an abortion. As I observed, that argument implicates three requirements of class certification under Rule 23(a): commonality of the claims, typicality of the claims, and adequacy of the class representatives, which overlap.2 The 2 Resolution of this argument implicates the fourth requirement, numerosity. I am doubtful that a properly defined class could meet the numerosity requirement, which is likely why the Appellees defined the class in such broad terms. I think the majority’s suggestion that even 6 majority recognizes the Government argued in the district court that many of the putative class members wished to carry their pregnancies to birth, but contends that the Government’s elaboration of that point on appeal – that many of the pregnant minors could be expected to have moral/religious objections to abortion – is somehow a different argument. But the Government in the district court relied on Mayfield v. Dalton, 109 F.3d 1423 (9th Cir. 1997), holding that a class could not include people who supported the Government’s challenged policy. Id. at 1427. That many of the class members in our case have moral/religious objections to abortion is hardly a different point; it is not only implicit, it is obvious in today’s world. In that regard, the majority ignores the independent obligation of the district court to assure itself of the propriety of class certification. Although, in my view, both the pleadings and common sense presented this issue, the Supreme Court has cautioned that “sometimes it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question.” Falcon, 457 U.S. at 160. Any court faithfully performing the “rigorous analysis” mandated by the Supreme Court should have anticipated and engaged with the obvious problems posed by this class. Id. at 161. So, too, any district judge in the United States should have recognized that many pregnant minors who favored the Government’s policy would have moral/religious reasons for doing so. In any event, the majority confronts the argument on its merits and dismisses it as mere value or ideological interests. Op. at 40. The Government makes a related powerful argument with respect to adequacy: the class representatives, both of whom sought an abortion, cannot adequately represent the interests of a smaller class might be appropriate is quite speculative. See Op. at 48-50. 7 pregnant minors who wish to carry their babies to birth. The majority’s response is that the district judge’s order ostensibly is also directed to the benefit of the minors who don’t want an abortion. It states the Government is banned from interfering with pregnancy-related care. But that is obviously window dressing; no one has been arguing that the Government has been derelict in providing care to those minors who wish to give birth. Of course, the adequacy of the class representatives is directly related to the question whether the class has a common legal interest. The Appellees contend that all pregnant minors – whether or not they want an abortion – are really aligned with the class representatives because the relevant constitutional right in their view is the right to choose whether or not to have an abortion. I think that confuses a political slogan3 with a constitutional right. No one questions any woman’s legal right to carry a fetus to birth; the only constitutional issue is whether there is a constitutional right to have an abortion under some circumstances. So it is misleading for Appellees to speak of a constitutional right to choose whether to carry a pregnancy to birth. No one, least of all the Government, threatens such a “choice.” The Appellees’ articulation of constitutional rights carried over, for example, to the Second Amendment would be phrased as a choice whether to keep arms or not, implying a constitutional right not to keep arms – which would be ridiculous. The majority adopts the Appellees’ choice theory: “class members all assert a common entitlement to make that choice on their own, free from any veto power retained (unconstitutionally, the class says) by ORR.”4 Op. at 32. 3 4 Like “right to life.” To be sure, the “choice” wording comes from the plurality opinion in Casey, but it seems to me that that strange wording itself had a political objective. The more natural phraseology would have 8 As a consequence of its thinking, the majority treats pregnant minors who wish to deliver a baby – the much larger group – as merely indifferent to or uninterested in the class action.5 It is claimed there is no conflict between those pregnant minors who wish to give birth and those who wish an abortion. Op. at 32-43. In my view, given the fierce differences, based in part on religious beliefs, that divide people in the United States and all over the world on the morality of abortion, to blithely suggest that those pregnant minors who do not want an abortion are merely uninterested in the Government’s policy seems quite intolerant of religious views. Alternatively, the majority determines, even if some members of the class, hypothetically, are strongly opposed to abortion, there are two reasons to ignore their concerns. First, identification of those minors who have such convictions would not be easily ascertainable for the purpose of carving them out of the class action. That is so. But the inappropriateness of the class is not to be cured by carving out those with a hypothetical level of moral or religious opposition to abortion. It is sufficient to recognize, as apparently the majority does, that many persons, including presumably many of the pregnant minors, have these convictions. (Can one imagine how terrified pregnant alien minors would be if questioned about their views on the Government’s policy?) After all, it is the plaintiff who bears the burden in a class certification proceeding to meet the requirements of Rule 23(a). Richards v. Delta Air Lines, Inc., been the right to have an abortion. See Whole Woman’s Health v. Hellerstedt, 136 S. Ct. 2292, 2321-22 (2016) (Thomas, J., dissenting). 5 Of course, if the majority were only indifferent to the claims of the class representatives, that itself would undermine the adequacy of the class representatives. 9 453 F.3d 525, 529 (D.C. Cir. 2006). Thus, even if – which I do not advocate – minors with moral/religious convictions in favor of the policy should be excised from the class, it would be the Plaintiffs’ burden to identify them, and since the majority acknowledges that would be virtually impossible, it is the Plaintiffs who do not satisfy their burden. It would be willful blindness not to recognize that any randomly selected group – particularly one drawn largely from countries with substantial Catholic populations – would include women who fiercely oppose abortion as murder. In other words, once it is realized – and I think we can take judicial notice of it – that a number of the pregnant minors are likely to have moral/religious convictions that abortion is murder, that is a powerful reason to conclude the class is improperly certified. But – and this is important – I do not argue that the class is improperly certified only because it includes minors who have moral/religious objections to abortion. That a number are likely to have religious objections just illustrates the lack of commonality. The second reason the majority brandishes is that mere ideological opposition – which apparently includes religious views – is not the kind of conflict of interest that threatens the class certification. Op. at 38. Tell that to the bakers who risked sanctions for religious reasons, rather than bake a cake for a gay wedding. The Supreme Court was more sympathetic. See Masterpiece Cakeshop, Ltd. v. Colo. Civil Rights Comm’n, 138 S. Ct. 1719 (2018). Indeed, the whole body of jurisprudence dealing with abortion is infused with religious views and secular values. The majority argues that because a (b)(2) class action has no “opt-out” requirement, we need not worry about the inclusion of members who do not belong and likely do not wish to be 10 included in the class. But the justification for the lack of an optout right in a (b)(2) case is that “injunctive relief benefits the entire class.” Richards, 453 F.3d at 530. That only holds true when the class is properly defined. Indeed, since (b)(2) class members lack the ability to opt out, the district court must be particularly vigilant in defining the class. To be sure, plain ideological considerations have been repeatedly held as insufficient to constitute injury-in-fact for standing purposes. Sierra Club v. Morton, 405 U.S. 727, 738-40 (1972); Diamond v. Charles, 476 U.S. 54, 66-67 (1986). But when one is placed in a class represented by political antagonists, one’s objection is not merely an abstract ideological concern.6 For instance, the majority’s concept would permit a class of all federal government workers (approximately 3 million) if, let us say, Congress passed a law banning continuing government employment of individuals belonging to an organization that secretly supports ISIS or similar groups, and a class action based on the First Amendment was brought. All government employees, including persons employed in the CIA and the Defense Department, would have a “choice” whether or not to belong to such an organization, so their ideological opposition to the target organization supposedly would be irrelevant. According to the majority’s reasoning, they would all be merely indifferent to the claims of the hypothetical class representatives. By the same token, a neo-Nazi group that brought that sort of class action could include all Jewish government employees, and a white supremacist class action could include African-Americans. 6 After all, class members are compelled to accept representation from a quasi-political organization, the ACLU. 11 In response to my hypothetical relating to a neo-Nazi First Amendment claim in the form of a class action that would include all federal employees, the majority points out that Jewish employees might have grounds to claim a conflict with the class because they would have to work alongside the neoNazis. (Why only Jewish employees? And would half-Jews qualify? And what about those employees who are working thousands of miles away from where the neo-Nazis wish to work?) I believe being associated in an action on behalf of the neo-Nazis is a good deal more troublesome than working in the same location. I can just imagine how the dissident groups in my examples would react when told they are part of the class. The majority suggests “not to worry” the dissident groups might never know they were included in the class action. Op. at 40. I wouldn’t bet on that. Indeed, the resulting publicity would horrify most of the unnamed members of the class. I think they would probably have a First Amendment freedom of association claim themselves, i.e., an objection to a governmentally imposed membership in a politically offensive class. Cf. Maximilian A. Grant, Comment, The Right Not to Sue: A First Amendment Rationale for Opting Out of Mandatory Class Actions, 63 U. Chi. L. Rev. 239 (1996). In other words, such putative members of a class might actually have an injury-in-fact when placed in such a class. That is quite different from the Supreme Court’s recognition in Diamond v. Charles that someone supporting legislation for ideological reasons, without injury, lacks standing. See Op. at 40. In sum, the implications of the majority’s interpretation of Rule 23(a) make obvious its flawed reasoning, but in my view, its concept of a proper class, like its mootness analysis, also runs counter to constitutional standing doctrine. The Supreme Court has stated that “Rule 23’s requirements must be interpreted in 12 keeping with Article III constraints.” Amchem Products, Inc. v. Windsor, 521 U.S. 591, 613 (1997). Certification of a class consisting mostly of class members who plan to carry their pregnancies to term and who do not claim to be injured by the Government’s policy violates Article III’s requirement that a plaintiff suffer injury-in-fact.7 The majority concludes that it is irrelevant whether all members of the class claim an injury, relying primarily, by analogy, to the way the Supreme Court treats a case of joint plaintiffs. It is black letter law that once a federal court determines one of joint plaintiffs has standing, it is unnecessary to determine whether the others have standing. See, e.g., Bowsher v. Synar, 478 U.S. 714, 721 (1986). But the analogy, in my view, doesn’t hold because in the joint plaintiff case, it is irrelevant whether the “others” have standing – they are only interested in the resolution of the legal issue, like everyone else. In other words, they are not parties. See Planned Parenthood of Idaho, Inc. v. Wasden, 376 F.3d 908, 918 n.6 (9th Cir. 2004). A class action presents an entirely different legal picture. All members are parties to the case bound by the judgment with the right to appeal, inter alia, a challenge to a settlement. Devlin v. Scardelletti, 536 U.S. 1, 9-11 (2002). Similarly, the filing of a class action tolls the running of a statute of limitations for unnamed plaintiffs. Am. Pipe & Constr. Co. v. Utah, 414 U.S. 538, 551-54 (1974). The majority complains that there is no decision supporting my analysis, passing by the acknowledgment that there is no Supreme Court decision supporting its view either. I am prepared to rest on logic. 7 As the Supreme Court has recognized, a party that has not yet suffered damage can have injury-in-fact if there is “substantial risk” of future injury. Susan B. Anthony List v. Driehaus, 573 U.S. 149, 158 (2014) (quoting Clapper v. Amnesty Int’l USA, 568 U.S. 398, 414 n.5 (2013)). 13 To be sure, there is a conflict in the circuits as to whether all members of a class must have a claimed injury, compare, e.g., Denney v. Deutsche Bank AG, 443 F.3d 253, 264 (2d Cir. 2006), with Neale v. Volvo Cars of N. Am., LLC, 794 F.3d 353, 362 (3d Cir. 2015) – indeed, “the same injury,” Wal-Mart, 564 U.S. at 350 (quoting Falcon, 457 U.S. at 157) – and although the Supreme Court has not settled the issue, in discussing calculating damages in class actions seeking monetary relief, the Chief Justice has said, “Article III does not give federal courts the power to order relief to any uninjured plaintiff, class action or not.” Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1053 (2016) (Roberts, C.J., concurring). In a typical individual case, a plaintiff who claims an injuryin-fact can be challenged and put to proof as to his or her injury. I understand the Chief Justice, in addressing his argument to damages rather than class certification, to be implicitly recognizing that in a class action for damages it would be duplicative and chaotic to force each unnamed member to actually prove his or her injury twice. By analogy, the Court has recognized that unnamed members could not be challenged factually to defeat diversity – that could destroy the class action as a procedural tool by causing enormous delay. Devlin, 536 U.S. at 10. But at the certification stage it seems to me all members of the class must at least claim the same injury. This is particularly true in a (b)(2) class action that lacks a damages stage. I concede that in a class action seeking an injunction to allow a defendant to challenge factually every unnamed member’s injury similarly would be chaotic, but at a minimum, all class members should be claiming the same injury. The Supreme Court has recently recognized at least one class representative is in the same position as an individual plaintiff. He or she must demonstrate – not just claim – 14 standing and, therefore, must be prepared to prove injury-in-fact. Frank v. Gaos, 139 S. Ct. 1041, 1046 (2019) (per curiam). The majority appears to ignore the difference between a claim of injury and demonstrating (proving) the injury. The Court has never suggested that a class can include members who not only do not claim an injury, but who are opposed to even recognizing the constitutional right that the class representative asserts. In that regard, the majority’s suggestion that civil rights cases “can involve polarizing issues” does little to advance its argument. Op. at 38. For the typical civil rights case, the issue would not be divisive with respect to the class (assuming the class is properly defined), despite the fact that the general population might have strong feelings about the topic. For example, in a discrimination case, a court generally need not worry about class members being opposed to the recognition of the right not to be discriminated against, even if they are satisfied with the status quo. On the other hand, Lanner v. Wimmer, 662 F.2d 1349 (10th Cir. 1981), in which a class challenging a school released-time program for religious education purposes on First Amendment grounds included parents who approved of the program, id. at 1357, is rather similar to our case, and I think was wrongly decided. The majority’s hypothetical class challenging a regulation alleged to violate the Second Amendment apparently has no limit; it theoretically could include all Americans who live in or might drive to the affected location. The proper approach is to limit the class to those residents who claim injury because the regulation does them harm.8 8 In Parker v. District of Columbia, 478 F.3d 370 (D.C. Cir. 2007), based on prior precedent, we held that the only one of the plaintiffs who had standing was Heller because he had been denied a license. Id. at 374-78. 15 Similarly, in Gratz v. Bollinger, 539 U.S. 244 (2003), the class was limited to members of racial and ethnic groups who were treated less favorably on the basis of race in considering their applications to University of Michigan. Id. at 252-53. In other words, only those who suffered an injury by being denied admission. III. Since the majority deems the class certification appropriate, I turn to the merits. I will incorporate then-Judge Kavanaugh’s persuasive prior opinion. Garza v. Hargan, 874 F.3d 735, 75256 (D.C. Cir. 2017) (Kavanaugh, J., dissenting) (explaining why expeditious transfer to a sponsor is not an undue burden given the Government’s “permissible interests in favoring fetal life, protecting the best interests of a minor, and refraining from facilitating abortion,” id. at 752). I note that Judge Henderson has a powerful alternative argument, id. at 746-52 (Henderson, J., dissenting) (concluding that an unaccompanied alien minor in ORR custody lacked a constitutional right to an abortion), but that argument was not made by the Government, so in accordance with my views of judicial restraint, I will not consider it. To be sure, the Supreme Court once – because, as so often happens, it was anxious to reach the merits of an argument – sanctioned deciding an issue that had been waived, U.S. Nat’l Bank of Or. v. Indep. Ins. Agents of Am., Inc., 508 U.S. 439, 446-48 (1993), (which is why I once referred to the Court as a “non-court court,” United States v. Moore, 110 F.3d 99, 102 (D.C. Cir. 1997) (Silberman, J., dissenting from denial of rehearing en banc)). But since the Court set forth no standard – other than it be an anterior legal question – to guide the lower courts as to when they should follow that practice, most courts ignore it. I try to – as did Judge Kavanaugh. 16 The Government did make a somewhat different argument in our case not presented in the previous individual claims. It asserted, alternatively, that the district court’s preliminary injunction, even assuming an injunction was proper, was much too broad. The majority responds by asserting that a narrower injunction would oblige the district court to make policy judgments so a blunderbuss rather than a rifle was necessary. The majority’s apprehension seems rather ironic. After all, the whole body of constitutional jurisprudence relating to abortion is more driven by policy concerns than law. Be that as it may, the majority’s concerns seem overwrought. It discounts narrowing concepts such as giving limited time to the Government to find a sponsor because it would require the district judge to make “policy” decisions as to when the duty would start and when it would end. But it is our equitable obligation to fashion narrow relief. The Government’s obligation to search for a sponsor should start the moment the pregnant minor says she wants an abortion. Although the Government may well have been seeking a sponsor before that time, and, as I noted earlier, that process can be lengthy, surely its efforts would be accelerated in light of the new information – or would be if my opinion were adopted. The Government’s interest is to find a sponsor as quickly as possible in that situation. (We must assume the U.S. Government would be acting in good faith.) Of course, even if the Government is moving expeditiously, it could be unsuccessful for a period so there would have to be an end point. Using the analogy of the parental consent and judicial bypass cases, procedures that could cause up to a three-week delay have been recognized as expeditious. See Ohio v. Akron Ctr. for Reprod. Health, 497 U.S. 502, 514 (1990). Of course, any injunction would have to include a recognition that the mother’s health must be paramount and in no event could the process come close to 17 viability. I would leave it to the district court to arrive at a figure that meets the Supreme Court’s expeditious requirement. The majority seems to find it irrational that the Government might treat minors differently than adults. Yet “[t]he law does not always treat minors in the same way as adults, as the Supreme Court has repeatedly emphasized in the abortion context.” Garza, 874 F.3d at 755 (Kavanaugh, J., dissenting). I agree with the Government that, even if the class were properly certified, the district court’s injunction is far too broad because it ignores the Government’s legitimate interest in the minor’s welfare. The majority mistakenly analogizes the Government’s policy articulated by counsel before us to a piece of legislation which would have to be accepted or rejected. Since we are presented with the Government’s alternative argument regarding its policy that the injunction the district court adopted was too broad, as I noted, we have an equitable obligation to fashion the narrowest feasible injunction. See State of Neb. Dep’t of Health & Human Servs. v. U.S. Dep’t of Health & Human Servs., 435 F.3d 326, 330 (D.C. Cir. 2006). After all, “[t]he need for narrow tailoring . . . is particularly important in the context of a preliminary injunction or temporary restraining order, where the court has yet finally to resolve the merits of the dispute.” U.S. Ass’n of Reptile Keepers, Inc. v. Jewell, 106 F. Supp. 3d 125, 126 (D.D.C. 2015). The majority suggests that the Government’s real interest in sponsorship is nothing more than avoiding facilitation of abortion. This is unfair. It seems to me irrelevant as to what the “record” reveals of ORR’s “thinking” – whatever that means – about the benefits of adult guidance for a pregnant minor. Contrary to the assertions of the majority, in this litigation, the Government has argued that sponsorship does provide that benefit to a pregnant minor. See Appellants’ Br. 44 (“Particularly in light of the benefits sponsorship provides, such 18 a modest waiting period does not impose an undue burden on the minor’s ability to pursue an abortion.”). The majority seems to regard this case as if it were an APA challenge to agency action in which the express reasoning of the agency must be defended. But we are not reviewing administrative action, so we can rely upon the legal arguments made before us. And it is clear that the Government has invoked the benefits of sponsorship for the pregnant minors who are “considering and pursuing such a personal and sensitive decision as abortion.” Id. I am afraid the majority’s refusal to consider narrowing the scope of the district court’s order justifies Judge Kavanaugh’s accusation that the court is endorsing abortion on demand – at least as far as the federal Government is concerned.
Primary Holding
Plaintiffs have shown a likelihood of success in connection with their claim that ORR's restriction on abortion access infringes their protected right to choose to terminate their pregnancies.

Disclaimer: Justia Annotations is a forum for attorneys to summarize, comment on, and analyze case law published on our site. Justia makes no guarantees or warranties that the annotations are accurate or reflect the current state of law, and no annotation is intended to be, nor should it be construed as, legal advice. Contacting Justia or any attorney through this site, via web form, email, or otherwise, does not create an attorney-client relationship.