ANA RUTH HERNANDEZ-LARA, Petitioner, Appellee, v. TODD M. LYONS, Immigration and Customs Enforcement, Enforcement and Removal Operations, Acting Field Office Director, Respondent, Appellant, CHRISTOPHER BRACKETT, Superintendent, Strafford County Department of Corrections, Respondent.
No. 19-2019
United States Court of Appeals For the First Circuit
August 19, 2021
Hon. Landya B. McCafferty, U.S. District Judge
APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF NEW HAMPSHIRE
Before Lynch, Lipez, and Kayatta, Circuit Judges.
Catherine M. Reno, Trial Attorney, with whom Ethan P. Davis, Acting Assistant Attorney General, Civil Division, Scott G. Stewart, Deputy Assistant Attorney General, William C. Peachey, Director, Office of Immigration Litigation, District Court Section, Carlton F. Sheffield, Senior Litigation Counsel, and Ari Nazarov, Trial Attorney, were on brief, for appellant.
Bryanna K. Devonshire, with whom Courtney H.G. Herz, Sheehan Phinney Bass & Green, PA, Gilles Bissonnette, Henry Klementowicz, SangYeob Kim, and American Civil Liberties Union of New Hampshire, were on brief, for appellee.
Hernandez subsequently filed a petition for a writ of habeas corpus in the United States District Court for the District of New Hampshire, contending that the Due Process clause of the Fifth Amendment entitled her to a bond hearing at which the government, not Hernandez, must bear the burden of proving danger or flight risk by clear and convincing evidence. The district court agreed and ordered the IJ to conduct a second bond hearing at which the government bore the burden of proving by clear and convincing evidence that Hernandez was either a danger or a flight risk. That shift in the burden proved pivotal, as the IJ released Hernandez on bond following her second hearing, after ten months of detention. The government now asks us to reverse the judgment of the district court, arguing that the procedures employed at Hernandez‘s original bond hearing comported with due process and, consequently, that the district court‘s order
I.
The parties do not dispute the relevant background facts. Hernandez was born in Usulutan, El Salvador, in 1986. Before coming to the United States in 2013, her life was marred by abusive domestic relations and gang violence. Hernandez‘s stepfather raped her when she was twelve years old and beat her mother throughout Hernandez‘s childhood. History repeated when Hernandez‘s stepfather‘s son raped Hernandez‘s then-eight-year-old daughter. Although Hernandez escaped her stepfather by living with her brother, she was unable to escape danger. Hernandez‘s brother was a member of Mara 18 (the 18th Street Gang), and after he was imprisoned for gang-related crimes, the gang began threatening Hernandez in an effort to force her to assume her brother‘s former gang responsibilities. Hernandez resisted those threats until late August 2013, when the gang told her aunts they intended to kill her and “throw [her] head in the river.”
Hernandez immediately fled to the United States and ultimately established residency in Portland, Maine, where she worked at a recycling plant and was engaged to be married.
Hernandez was taken into custody by an immigration officer on September 20, 2018, and detained pursuant to
The government‘s response provided an apt demonstration of how the burden of proof can affect immigration bond hearings. Government counsel produced a so-called “Red Notice” published by El Salvador through the International Criminal Police Organization (“INTERPOL“). The notice identifies Hernandez, describes the activities of Street Gang 18 (much as Hernandez described them), and simply states that Hernandez is subject to an arrest warrant in El Salvador under El Salvadoran “Article 13 of the Special Law Against Acts of Terrorism.”
An INTERPOL Red Notice is “a request to law enforcement worldwide to locate and provisionally arrest a person pending extradition, surrender, or similar legal action.” Red Notices, INTERPOL, https://www.interpol.int/en/How-we-work/Notices/Red-Notices (last visited August 18, 2021). In the United States, an INTERPOL Red Notice alone is not a sufficient basis to arrest, much less detain or extradite, the “subject” of the notice “because it does not meet the requirements
Hernandez denied belonging to the organization. Her counsel explained that her brother had belonged to the gang and pointed out that the Red Notice failed to specify any criminal or dangerous act that Hernandez allegedly committed.
The IJ indicated that it was not clear whether Hernandez‘s alleged involvement in the organization was due to “an inter-rival thing or [if] she was an innocent member or somehow wrongly identified.” Nonetheless, he found that there was not “sufficient evidence explaining why these allegations are being brought against her.” Stating that “it is [Hernandez‘s] burden of proof to show by clear and convincing evidence she is not a danger,” the IJ found, “based on this Red Notice, [that] she has failed to meet that burden.” Consequently, he denied her request for bond. Hernandez remained detained as she pursued claims for asylum, withholding of removal, and relief under the Convention Against Torture (“CAT“).
On April 16, 2019, Hernandez filed a petition for a writ of habeas corpus in the United States District Court for the District of New Hampshire. In her petition, Hernandez claimed that due process required the government to bear the burden of proving, by clear and convincing evidence, that she was either dangerous or a flight risk, and therefore that her initial bond hearing was constitutionally inadequate. Hernandez also claimed that because of her “prolonged detention” of over six months, due process required an additional bond hearing at which the government would bear the burden of proof. Hernandez sought as relief either her immediate release or a new bond hearing at which the government would bear the burden of proving by clear and convincing evidence that she was dangerous or a flight risk.
On July 25, 2019, the district court granted Hernandez‘s habeas petition and ordered the IJ to conduct another bond hearing at which the government would “bear the burden of justifying Hernandez‘s detention by clear and convincing evidence.” Hernandez-Lara v. Immigr. & Customs Enf‘t, Acting Dir., No. 19-cv-394-LM, 2019 WL 3340697, at *7 (D.N.H. July 25, 2019).2 Less than a week later, the same IJ who conducted Hernandez‘s first bond hearing held a second hearing in accordance with the district court‘s order. The government relied once again on the Red Notice and additionally argued that Hernandez was a flight risk because her asylum claim had been denied by both the IJ and the Board of Immigration Appeals (BIA), though it was pending before this court at the time. Hernandez countered that the Red Notice was defective, as it contained no factual allegations that Hernandez committed any crime or was part of any gang activity, and that she has no history of criminal conviction. As to flight risk, Hernandez argued she had a meaningful chance of relief in her appeal before us and that she had family ties, employment, and a residence in Maine to which she would return.
The IJ granted Hernandez‘s request for bond, setting it at $7,500. In explaining his decision, the IJ stressed the shift of burden:
Because the burden of proof is now on the Government, I do find that to be outcome determinative in this case for the reasons I stated in [the first bond hearing]. While [Hernandez] does have accusations, absent any other details or any other evidence, I‘m able to conclude that it isn‘t clear and convincing to show that she‘s a danger, especially where she has no other criminal history here in the United States.
Given her community ties, fixed address, and work history, the IJ also found that Hernandez was not a flight risk. As a result, the IJ released Hernandez after she spent over ten months in detention.
As noted, the IJ had previously denied Hernandez‘s asylum, withholding, and CAT claims on the merits, finding her credible but also concluding that “she failed to demonstrate that her familial connection to her brother was ‘one central reason’ that the gang singled her out” and that “the police would have protected [her] from the gang if she had reported the threats because the police had protected her from her ex-partner in the past.” Hernandez Lara v. Barr, 962 F.3d 45, 52 (1st Cir. 2020). After the BIA affirmed that ruling, Hernandez appealed. Nearly a year after Hernandez was released from custody, we vacated the BIA‘s decision and remanded for further proceedings, which are ongoing. See id. In the meantime, the government filed this appeal from the district court‘s grant of Hernandez‘s habeas petition.
II.
“It is well established that ‘[o]ur review of a district court‘s grant or denial of habeas is de novo.‘” Sanchez v. Roden, 753 F.3d 279, 293 (1st Cir. 2014) (alteration in original) (quoting Healy v. Spencer, 453 F.3d 21, 25 (1st Cir. 2006)). Before undertaking that review, we first survey the statutory and regulatory framework challenged by Hernandez.
The Immigration and Nationality Act (“INA“) provides that the government must detain for the duration of removal proceedings most noncitizens who have committed certain types of criminal offenses.
An Immigration and Customs Enforcement (“ICE“) officer makes the initial detention determination for noncitizens subject to detention under
In 1996 Congress enacted the Illegal Immigration Reform and Immigrant Responsibility Act (“IIRIRA“). Omnibus Consolidated Appropriations Act, Pub. L. No.
Nevertheless, following the enactment of IIRIRA, the Immigration and Naturalization Service (INS) adopted new regulations establishing a presumption of detention in the initial custody determination by the arresting officer. See
Accordingly, under current BIA precedent, a noncitizen detained under
III.
We turn now to the merits of this appeal. In Jennings v. Rodriguez, 138 S. Ct. 830 (2018), the Supreme Court held that, as a matter of statutory interpretation,
Our inquiry is guided by the three-part balancing test articulated in Mathews v. Eldridge, 424 U.S. 319, 335 (1976); see also Velasco Lopez, 978 F.3d at 851 (analyzing procedural due process challenge to prolonged
We address each factor in turn, focusing first on the allocation of the burden of proof. We then address separately the government‘s contention that, notwithstanding any analysis of the Mathews factors, precedent calls for us to rule in the government‘s favor. Finally, we address the extent of the burden to be borne.
A.
“Freedom from imprisonment — from government custody, detention, or other forms of physical restraint — lies at the heart of the liberty that [the Due Process] Clause protects.” Zadvydas v. Davis, 533 U.S. 678, 695 (2001) (citing Foucha v. Louisiana, 504 U.S. 71, 80 (1992)). The Supreme Court has repeatedly affirmed that “[i]n our society liberty is the norm, and detention prior to trial or without trial is the carefully limited exception.” United States v. Salerno, 481 U.S. 739, 755 (1987); see also Foucha, 504 U.S. at 80 (“We have always been careful not to minimize the importance and fundamental nature of the individuals’ right to liberty.“). For this reason, “civil commitment for any purpose constitutes a significant deprivation of liberty that requires due process protections.” Addington, 441 U.S. at 425 (emphasis added).
Hernandez was incarcerated alongside criminal inmates at the Strafford County Jail for over ten months. See Velasco Lopez, 978 F.3d at 850 (“[Petitioner] was not ‘detained‘; he was, in fact, incarcerated under conditions indistinguishable from those imposed on criminal defendants sent to prison following convictions for violent felonies and other serious crimes.“) During that time, she was separated from her fiancé and unable to maintain her employment. But for the relief ordered in this action, she would still be incarcerated more than two years after the jailor first locked the door behind her. There is no question that Hernandez suffered a substantial deprivation of liberty.
In an attempt to downplay that deprivation, the government notes that Congress may make rules for noncitizens “that would be unacceptable if applied to citizens,” Demore, 538 U.S. at 522, and that “detention during deportation proceedings [is] a constitutionally valid aspect of the deportation process,” id. at 523. But the same could be said for criminal proceedings. And in either case the fact that some detention is permissible does not change the fact that a detainee suffers significant liberty deprivations. Moreover, the government‘s exercise of its power to detain immigrants pending removal “is subject to important constitutional limitations.” Zadvydas, 533 U.S. at 695. That is because due process ”
The government also argues that Hernandez‘s liberty interest should be discounted because she is “not simply asserting a right to be at liberty, but rather, a right to be at liberty in the United States, where she has never held lawful status” (emphasis in original). But as the Supreme Court explained in response to this same type of argument in Zadvydas, “the choice . . . is not between imprisonment and the alien ‘living at large‘” in this country but “between imprisonment and supervision under release conditions that may not be violated.” 533 U.S. at 696; see
The government next contends that “individuals detained under
This argument is a bit like telling detainees that they can help themselves by jumping from the frying pan into the fire.
Deportation is a “‘drastic measure, often amounting to lifelong ‘banishment or exile.‘” Sessions v. Dimaya, 138 S. Ct. 1204, 1213 (2018) (quoting Jordan v. De George, 341 U.S. 223, 231 (1951)); see id. (“[D]eportation is ‘a particularly severe penalty,’ which may be of greater concern to a convicted alien than ‘any potential jail sentence.’ (quoting Jae Lee v. United States, 137 S. Ct. 1958, 1968 (2017))); Bridges v. Wixon, 326 U.S. 135, 147 (1945) (“[D]eportation may result in the loss ‘of all that makes life worth living.‘” (quoting Ng Fung Ho v. White, 259 U.S. 276, 284 (1922))). The consequences of deportation are potentially most severe for meritorious asylum seekers, for whom one might fairly say that the escape from detention offered by the government could be death. Accordingly, like the Ninth Circuit, “[w]e are not persuaded that a lower standard of proof is justified by putting people . . . to the choice of remaining in detention, potentially for years, or leaving the country and abandoning their challenges to removability even though they may have been improperly deemed removable.” Singh v. Holder, 638 F.3d 1196, 1204 (9th Cir. 2011).
We recognize that removal proceedings have an end point and that the liberty interest of a noncitizen detained under
Accordingly, we find that the first Mathews factor (the private interest at stake) weighs heavily in Hernandez‘s favor.
B.
For several reasons, the second Mathews factor — “the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards” — likewise weighs heavily in Hernandez‘s favor. 424 U.S. at 335.
First, noncitizens have no right to be provided with counsel in immigration proceedings and very often cannot obtain counsel on their own, particularly if they are detained. See
Third, noncitizens subject to immigration detention often lack full proficiency in English. See, e.g., Hernandez Lara, 962 F.3d at 55 (noting that Hernandez “does not speak, read, or write English“).
Finally, proving a negative (especially a lack of danger) can often be more difficult than proving a cause for concern. See Elkins v. United States, 364 U.S. 206, 218 (1960) (“[A]s a practical matter it is never easy to prove a negative.“). For all of these reasons, a detainee often starts out behind the eight ball in a bond proceeding, and the opportunities for prejudicial error abound.
This very case evidences how the allocation of the burden of proof can affect the likelihood of such error. With a record of employment, family relations, a settled place in the community, and no arrests, Hernandez would seem to have been a good candidate for conditional release on bail. Indeed, no party claims that she has absconded or committed any crime during the year and a half that she has been out on bail. Yet as the IJ‘s rulings make clear, the placement of the burden of proof on Hernandez decisively exploited her inability to rebut the Red Notice, even though it did not specify a single act of criminal or dangerous conduct.
As the Supreme Court has observed, a noncitizen‘s “removable status itself . . . bears no relation to a detainee‘s dangerousness.” Zadvydas, 533 U.S. at 691-92. Thus, as a practical matter, adjudication of dangerousness will naturally tend to begin with the government offering a reason to find a particular person dangerous, with that person then addressing the proffered reason. And that reason will in most cases be based on law enforcement records to which the government will have greater access. See Velasco Lopez, 978 F.3d at 853 (explaining that the government has access to “numerous databases[,] . . . to information collected by DHS, DOJ, and the FBI, [and to] information in the hands of state and local authorities,” in addition to having “broad regulatory authority” to obtain information it does not have readily available). Here, for example, it was the government that had access to the Red Notice. For all these reasons, the government is generally far more able to meet the burden of proof on the question of danger than a detained noncitizen like Hernandez.
As the government argues, detained noncitizens may certainly have a better grasp of some information relevant to flight risk -- such as family ties, length of time in the United States, or record of employment. Nevertheless, they also face significant barriers to accessing such evidence in the wake of their seizure and initial detention. Moreover, none of this is to say that an IJ cannot draw a negative inference from the fact that a detainee offers no evidence on her behalf. Rather, it is to say that the odds of error in the weighing of such evidence (or its absence) are likely reduced by placing the burden on the government, as in virtually all other instances of proposed lengthy detention.
The government‘s response to all of this is to argue that for two reasons the existing framework provides procedural protections that “exceed the constitutional minimum.” First, the government points out that the existing procedures “permit an immigration judge to consider a wide range of factors, and the alien to present any evidence that may bear on these factors.” But as Hernandez‘s experience
C.
We turn to the final Mathews factor -- “the Government‘s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail” -- which ultimately entails an assessment of the “public interest.” 424 U.S. at 335, 347. The government‘s proffered interest is the “public interest in prompt execution of removal orders” and the “importance of immigration detention to effectuate immigration proceedings.” In support of this interest, the government points to legislative history stating that
The prompt execution of removal orders is a legitimate governmental interest, see Nken v. Holder, 556 U.S. 418, 436 (2009), which detention may facilitate, see Aguilar v. U.S. Immigr. & Customs Enf‘t, 510 F.3d 1, 22 (1st Cir. 2007) (recognizing “the government‘s legitimate interest in effectuating detentions pending the removal of persons illegally in the country“). In considering that interest, we must “weigh heavily” the fact that “control over matters of immigration is a sovereign prerogative, largely within the control of the executive and the legislature.” Landon v. Plasencia, 459 U.S. 21, 34 (1982). What is at stake, however, is not the power of the government to detain noncitizens who may cause harm or flee during removal proceedings, but rather who should bear the burden of proving noncitizens pose a danger or a flight risk.
The government fails to explain why its proffered interest in securing appearance at removal proceedings and for deportation holds sway where a noncitizen is not a flight risk. See Hernandez v. Sessions, 872 F.3d 976, 990 (9th Cir. 2017) (“The government has legitimate interests in protecting the public and in ensuring that noncitizens in removal proceedings appear for hearings, but any detention incidental to removal must ‘bear[] [a] reasonable relation to [its] purpose.‘” (quoting Zadvydas, 533 U.S. at 690)); see also
Perhaps more importantly, such unnecessary detention imposes substantial societal costs. This case illustrates those costs well: Because of her incarceration, Hernandez was separated from her fiancé and unable to maintain her employment, after living peacefully in Portland for over a year. More generally, noncitizens subject to immigration detention include spouses, children, and parents of U.S. citizens, caretakers of children and elderly relatives, and leaders in religious, cultural, and social groups. The needless detention of those individuals thus “separates families and removes from the community breadwinners, caregivers, parents, siblings and employees.” See Id. at 855. Those ruptures in the fabric of communal life impact society in intangible ways that are difficult to calculate in dollars and cents. Even so, as twenty states report in an amicus brief to this court, the financial costs imposed by such widespread communal disruption are severe: “[States‘] revenues drop because of reduced economic contributions and tax payments by detained immigrants, and their expenses rise because of increased social welfare payments in response to the harms caused by unnecessary detention.”
In short, given the risk that the current procedures lead to many instances of needless detention, entailing substantial social and financial costs, the public interest in placing the burden of proof on the detainee is uncertain at best, and may well be negative.
Pointing to
Likewise, the government makes much of the Court‘s statement in Nielsen v. Preap that
139 S. Ct. 954, 966 (2019). But in context, it is clear the Court was merely contrasting
Shifting gears, the government contends that it would be “backwards” to “put the burden on the Government to justify the alien‘s detention during the interim period when the Government is pursuing removal when the burden is on the alien [to prove that he or she was admissible or to prove a defense to removal] in the underlying removal proceedings themselves.” See
In a final salvo, the government contends that two of our sister circuits have ruled in a manner inconsistent with our holding today. See Ali v. Brott, 770 F. App‘x 298 (8th Cir. 2019); Borbot v. Warden Hudson Cnty. Corr. Facility, 906 F.3d 274 (3d Cir. 2018). We see no conflict.
To start, although Ali contains dicta that portends a different result from that reached here, the Eighth Circuit made clear that it was not reaching the constitutional question that is now before us. See 770 F. App‘x at 302. Likewise, the issue presented here was not before the court in Borbot, which was a challenge based on length of detention in which the petitioner sought “to compel a second bond hearing despite alleging no constitutional defect in the one he received.” 906 F.3d at 279 (second emphasis added). And although the court in Borbot stated that the petitioner had been granted “meaningful process” under
In sum, the balance of the Mathews factors weighs in favor of Hernandez: “[T]he private interest affected is commanding; the risk of error from [placing the burden of proof on the noncitizen] is substantial; and the countervailing governmental interest . . . is comparatively slight.” Santosky, 455 U.S. at 758.
D.
The government urges that notwithstanding the foregoing assessment of the three Mathews factors, precedent precludes us from placing any burdens of proof on the government. First, it argues that the Supreme Court has in three cases upheld detention of noncitizens pending removal proceedings “on the basis of a categorical, rather than individualized, assessment that a valid immigration purpose warranted interim custody” (emphasis in original). See Demore, 538 U.S. at 531; Carlson v. Landon, 342 U.S. 524, 538 (1952); Reno v. Flores, 507 U.S. 292, 306 (1993).
Each of these cases, however, is distinguishable from the circumstances presented here. In Demore, the Court held that
In upholding the constitutionality of
The circumstances here are quite different. Unlike
The government responds that, like
Carlson v. Landon is also distinguishable. Carlson involved a challenge by noncitizens accused of participating in Communist activities to their detention pending a determination of removability. See 342 U.S. at 528-29. Although the individuals detained in Carlson had not been determined to be dangerous or a flight risk, the Court upheld their detention “by reference to the legislative scheme to eradicate the evils of Communist activity.” Id. at 543. The purpose of that legislative scheme, the Internal Security Act, was to “deport all alien Communists as a menace to the security of the United States,” id. at 541, based on Congressional findings that the “Communist organization in the United States . . . present[s] a clear and present danger to the security of the United States,” id. at 535 n.21 (quoting 50 U.S.C. § 781(15)). The Court explained that because
all alien Communists are deportable, like Anarchists, because of Congress’ understanding of their attitude toward the use of force and violence in such a constitutional democracy as ours to accomplish their political aims, evidence of membership plus personal activity in supporting and extending the [Communist] Party‘s philosophy concerning violence gives adequate ground for detention.
Thus, much as in Demore, Congress made specific findings as to the dangerousness of a class of noncitizens, and those findings were found to have justified the detention of noncitizens even in the absence of individualized determinations as to danger and flight risk. But for the same reasons that Demore is a poor analog to this case, so too is Carlson: no similar findings regarding dangerousness or flight risk have been made as to the class of noncitizens detained under
Nor does Reno v. Flores control this case. Flores involved, among other things, a procedural due process challenge to a regulation that denied bail to noncitizen minors in removal proceedings who could not be released into the custody of a parent, legal guardian, or adult relative. See 507 U.S. at 297, 306-09. The relevance of Flores to this case is not immediately apparent, as the detained minors’ challenge was not based on the allocation or standard for the burden of proof applicable to the custody determination. Rather, the minors’ principal argument was that the immigration agency should be required to determine whether “detention . . . would better serve [their] interests than release to some other ‘responsible adult,‘” even if that adult was not a parent, guardian, or relative. Id. at 308.
Undeterred, the government points to the Court‘s statement that “due process is satisfied by giving the detained alien juveniles the right to a hearing before an immigration judge,” id. at 309 (emphasis in original), and argues that because every noncitizen detained under
In another line of attack, the government shifts its focus back to Demore, arguing that the Court in that case “rejected the applicability” of Addington and Foucha in the context of noncitizens detained during the pendency of removal proceedings. The majority opinion in Demore, however, does not mention Foucha, Addington, or similar civil detention cases, despite the fact that the dissent repeatedly cites them in support of its position. We decline to read the majority‘s silence as to Foucha and Addington as an across-the-board “rejection” of their applicability in immigration detention cases. See Shalala v. Ill. Council on Long Term Care, Inc., 529 U.S. 1, 18 (2000) (“This Court does not normally overturn . . . earlier authority sub silentio.“) Addington specifically admonished that “civil commitment for any purpose constitutes a significant deprivation of liberty that requires due process protections,” 441 U.S. at 425 (emphasis added), and as the government itself acknowledges, Zadvydas, also an immigration detention case, cites to Foucha and Salerno.8 See Zadvydas, 533 U.S. at 690; see also Demore, 538 U.S. at 553 (Souter, J., concurring in part and dissenting in part) (“Nowhere [in Zadvydas] did we suggest that the ‘constitutionally protected liberty interest’ in avoiding physical confinement, even for aliens already ordered removed, was conceptually different from the liberty
Despite Zadvydas‘s reliance on Foucha, the government next argues that Zadvydas in fact supports its position that the noncitizen seeking release, not the government, should bear the burden of proof at a
This hunt for inferential support in Zadvydas overlooks the Court‘s express criticism of the underlying statute for putting the burden of proving dangerousness on the noncitizen. See id. at 691-92 (noting that “preventive detention based on dangerousness” must be “subject to strong procedural protections” and disapproving of the fact that under the statute “the alien bears the burden of proving he is not dangerous“). Moreover, the burden placed on the noncitizen in Zadvydas -- to “provide[] good reason to believe that there is no significant likelihood of removal in the reasonably foreseeable future,” id. at 701 -- is quite different from the burden placed on a noncitizen detained under
The government also points to language in the Jennings dissent which it contends approves of the existing procedures under
The government similarly contends that two district court decisions in our circuit approved of the procedures governing
Leaving no stone unturned, the government lastly points to a district court opinion which it claims held contrary to our conclusion here. See Maldonado-Velasquez v. Moniz, 274 F. Supp. 3d 11, 14-15 (D. Mass. 2017). But, beyond venturing a “guess,” the district court did not decide the due process issue. Id. at 15. Instead, it assumed arguendo that the burden had been misallocated but concluded that the petitioner could not show any prejudice flowing from that error. Id. at 13-14. We dismissed the petitioner‘s appeal as moot. Maldonado-Velasquez v. Moniz, No. 17-1918 (1st Cir. March 22, 2018). And although we stated that the petitioner‘s “due process claim is not compelling,” it is clear, as the government itself notes, that we were referring to the petitioner‘s inability to show prejudice. Id. at n.2.
For all of the foregoing reasons, we remain unconvinced by the government‘s contention that we should not view an analysis of the Matthews factors as ultimately controlling. We therefore conclude that the government must bear the burden of proving dangerousness or flight risk in order to continue detaining a noncitizen under
E.
Having decided that the government bears the burden of proof, we now turn to the extent of that burden. “[T]he function of legal process is to minimize the risk of erroneous decisions,” Addington, 441 U.S. at 425, and the standard of proof “serves to allocate the risk of error between the litigants,” id. at 423. In detention cases, applying a heightened “standard of proof . . . reflects the value society places on individual liberty,” id. at 425 (quoting Tippett v. Maryland, 436 F.2d 1153, 1166 (4th Cir. 1971) (Sobeloff, J., concurring in part and dissenting in part),
Therefore, in several contexts, the government must justify detention by clear and convincing evidence. See, e.g., Addington, 441 U.S. at 433 (involuntary civil commitment to mental hospital); Foucha, 504 U.S. at 80 (confinement of insanity acquittees). Other significant liberty interests are similarly protected: The government must satisfy the clear and convincing standard in order to terminate parental rights, see Santosky, 455 U.S. at 748, deport a noncitizen, see Woodby v. Immigr. & Naturalization Serv., 385 U.S. 276, 277 (1966), or denaturalize an individual, see Chaunt v. United States, 364 U.S. 350, 353 (1960).
As to the government‘s burden to prove that a noncitizen presents a danger, we see no reason to vary from that approach: For the reasons described above, there is a heightened risk of prejudicial error and the government has ample and better access to evidence of dangerousness. See supra Section III.B.9
But with respect to flight risk, the second Mathews factor leads us to conclude that the government need only carry its burden by a preponderance of the evidence. Simply put, there is less risk of error from a preponderance standard on this issue because, as noted, detained citizens possess knowledge of many of the most relevant factors, such as their family and community ties, place of residence, length of time in the United States, and record of employment. And because the burden is on the government, the noncitizen need not prove a negative (by showing, for example, that he or she has not fled prosecution or failed to appear at court) but is instead faced with the more straightforward task of marshalling evidence readily available to her so as to rebut the government‘s evidence. Given these considerations, the probable value of a heightened standard of proof is thus less apparent when it comes to flight risk.
Two other considerations underlie our decision. First, a noncitizen‘s flight risk (as opposed to his or her danger) has a close nexus to the government‘s interest in ensuring the prompt execution of deportation orders. Second, although the Court has consistently required a clear and convincing standard when the government seeks to detain on the basis of danger, most of those cases do not involve risk of flight. In the analogous context of pretrial criminal detention under the Bail Reform Act, where flight risk is a factor, the government need only prove flight risk by a preponderance of the evidence in order to continue detention. See United States v. Patriarca, 948 F.2d 789, 793 (1st Cir. 1991).
In sum, we hold that, in order to continue detaining Hernandez under
IV.
We consider, next, the question of prejudice. Normally “[w]hen faced with a constitutional due process claim in the immigration context, we ask whether the procedure at issue ‘is likely to have affected the outcome of the proceedings’ as a condition of relief.” Hernandez-Lara, 962 F.3d at 57 (quoting Pulisir v. Mukasey, 524 F.3d 302, 311 (1st Cir. 2008)); see also Lopez-Reyes v. Gonzales, 496 F.3d 20, 23 (1st Cir. 2007) (“Absent cognizable prejudice, there is no due process claim.“) Although Hernandez argues that “a misallocated burden of proof is a structural error [that] constitutes a per se prejudice,” we need not reach that argument. As the IJ observed, the reallocation of the burden of proof ordered by the district court proved pivotal in changing the result from detention to release. Nor has the government challenged the district court‘s finding that Hernandez was prejudiced. Cf. Hernandez-Lara, 962 F.3d at 56-57 (noting a circuit split on “whether a petitioner who was improperly denied counsel in immigration proceedings must demonstrate that the denial resulted in prejudice” but declining to decide the question given that the petitioner was clearly prejudiced).
V.
Before concluding, we address three arguments made by the dissent in support of its claim that our decision amounts to “judicial hubris.”
A.
The dissent contends first that we should grant Hernandez relief on a statutory basis, rather than on constitutional grounds. The relief proposed by the dissent under the
Hernandez, though, asks not just that the burden of proof be allocated to the government. She claims that the constitution requires the government to carry that burden by clear and convincing evidence. The district court agreed; the IJ then applied the clear and convincing standard; Hernandez was set free; and the government now appeals, asking us to rule that Hernandez was not entitled to a clear and convincing standard as to danger or flight risk. So resolving this action by deciding the APA claim developed by the dissent in Hernandez‘s favor, as the dissent proposes, would deny by neglect a central aspect of the relief sought by Hernandez under her constitutional claim.11 Ruling as the dissent proposes would also require that we more broadly vacate the relief ordered by the district court, and allow for a new hearing not just on flight risk, but on dangerousness as well. In short, what the dissent proposes is not constitutional avoidance, which entails finding an alternative basis for providing the relief sought under the constitutional claim. See Marasco & Nesselbush, LLP v. Collins, No. 20-1397, 2021 WL 3012705, at *18 (1st Cir. July 16, 2021) (declining to address due process claim under doctrine of constitutional avoidance because “the relief available under the [statutory ground] adequately addresse[d] [the plaintiff‘s] remedial requests” and so “a non-constitutional disposition [was] possible“). Rather, the dissent proposes that we simply shirk our duty to decide a properly raised claim upon which a substantial portion of the request for relief hinges.
B.
The dissent also contends that our decision infringes on the province of the political branches. That general accusation can be made in every case involving an administrative rule or congressional statute, including every due process case. Clearly, the fact that another branch has acted in an area is an insufficient reason to refrain from exercising our “duty . . . to say what the law is,” Marbury v. Madison, 5 U.S. 137, 177 (1803), even in immigration and detention cases, and even where doing so requires setting aside Congressional enactments, executive actions, or state statutes. See, e.g., Zadvydas, 533 U.S. at 695 (explaining that, despite Congress‘s “‘plenary power’ to create immigration law, . . . Executive and Legislative Branch decisionmaking in that area . . . is subject to important constitutional limitations“; construing immigration detention statute to avoid unconstitutional detention); Hamdi v. Rumsfeld, 542 U.S. 507, 536-37 (2004) (holding that even “in the context of military action, it would turn our system of checks and balances on its head to suggest that a citizen could not make his way to court with a challenge to the factual basis for his detention by his Government, simply because the Executive opposes making available such a challenge“); Foucha, 504 U.S. at 81-82 (striking down Louisiana statute under which “the State need prove nothing to justify continued detention” of insanity acquittees).
As these and many other cases make clear, ours is a system in which even the most sensitive and critical exercises of power by the political branches can be constrained by the rights of the individual. In few instances are those constraints more necessary than when the government seeks to lock up individuals behind bars. Addington, 441 U.S. at 425 (“[C]ivil commitment for any purpose constitutes a significant deprivation of liberty that requires due process protections.“). And it is precisely the role of the judiciary to define those constraints. Far from violating the separation of powers, exercising that role is integral to fulfilling the vision of the “Framers of the Constitution that, within our political scheme, the separation of governmental powers into three coordinate Branches is essential to the preservation of liberty.” Mistretta v. United States, 488 U.S. 361, 380 (1989).
We are mindful that immigration is “interwoven with contemporaneous policies in regard to the conduct of foreign relations, the war power, and the maintenance of a republican form of government.” Demore, 538 U.S. at 522. But nothing in our opinion today prevents the political branches from detaining noncitizens where necessary, let alone from exercising the power to exclude or expel noncitizens. Moreover, even where war and foreign relations are at issue, the Constitution “most assuredly envisions a role for all three branches when individual liberties are at stake.” Hamdi, 542 U.S. at 536; see also Zadvydas, 533 U.S. at 695. And as we explained above, the Court has consistently held that due process “applies to all ‘persons’ within the United States, including aliens, whether their presence here is lawful, unlawful, temporary, or permanent.” Zadvydas, 533 U.S. at 693. Though we hope and expect that the political branches exercise their authority in harmony with the rights of noncitizens, history and common sense teach that rights are most likely to be disregarded when they belong to those who cannot vote. Cf. United States v. Carolene Prod. Co., 304 U.S. 144, 153 n.4 (1938) (noting that “prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities“).
We stress as well that nothing in our decision restricts the political branches from implementing more nuanced rules for the adjudication of requests for release under section 1226. All that is required is that those rules comport with the minimum
C.
Finally, the dissent contends that our decision is overly broad because the current bond procedures are constitutional in at least some cases, dooming a facial challenge to those procedures. To this contention we offer two responses. First, and most simply, Hernandez claims that the current BIA standard of proof as applied in her case caused her to be unconstitutionally detained. And the IJ found the standard was indeed pivotal. So whatever one might say about facial challenges generally poses no bar to granting Hernandez relief.
Second, the dissent‘s reasoning seems flawed, even circular. The logic of the dissent appears to be that if there is sufficient evidence of flight risk in a particular case (e.g., per the dissent, fleeing from a checkpoint) the government need not carry the burden of proving flight risk in that particular case. But “the right to procedural due process . . . does not depend upon the merits of a claimant‘s substantive assertions.” Carey v. Piphus, 435 U.S. 247, 266 (1978). Moreover, the dissent‘s argument begs the question: What burden and standard would apply in determining whether the merits of the request for release are sufficient to obviate the need for placing the burden on the government? The dissent does not say. If the burden is as we suggest it should be, then the dissent‘s approach simply front ends the application of that requirement. And if it is a lesser burden, then the dissent‘s approach is simply a round-about way of saying that there should be a lesser burden.
Given all of the above, it is unsurprising that the Supreme Court has consistently decided procedural due process challenges in the detention context on a categorical basis (e.g., all criminal defendants or insanity acquittees). See, e.g., In re Winship, 397 U.S. 358, 364 (1970) (holding due process requires that all criminal defendants must be convicted by proof beyond a reasonable doubt); Hamdi, 542 U.S. at 533-35 (setting forth the contours of the procedures required under due process for all “citizen-detainee[s] seeking to challenge [their] classification as an enemy combatant“); Addington, 441 U.S. at 433 (holding that “the individual‘s interest in the outcome of a civil commitment proceeding is of such weight and gravity that due process requires the state to justify confinement by proof more substantial than a mere preponderance of the evidence“); Foucha, 504 U.S. at 86 (holding broadly that insanity acquittees may not be detained unless the government can show they are dangerous by clear and convincing evidence).
In none of these cases did the Court limit its holding to the specific individual before it or indicate that the requirements of due process would fluctuate based on the strength of any particular individual‘s
Nor did such cases vary the requirements of due process for different “subcategories” of detainees, e.g., those with certain types of mental illness or those who have committed certain types of crimes. Similarly, cases outside of the detention context do not slice and dice claimants (such as welfare recipients) into some unknown number of unspecified subcategories.
We are far from alone in applying procedural due process protections to well-defined categories of noncitizens (e.g.,
Recognizing well-defined categorical rules in procedural due process cases is unsurprising from the standpoint of judicial and administrative efficiency. Otherwise, every controversy would become two cases in one: a determination of the procedures required by due process, followed by a resolution of the merits. For detention pending the completion of removal proceedings, that inefficiency would be exacerbated because each case begins in an administrative proceeding, while habeas claims are heard in the district courts.
VI.
For the foregoing reasons, we affirm in part, reverse in part, and remand to the district court with instructions to allow the government, should it wish to do so, to conduct a new hearing before the Immigration Judge at which, in order to reinstitute Hernandez‘s detention, the government will need to prove flight risk by a preponderance of the evidence.
- Dissenting Opinion Follows -
LYNCH, Circuit Judge, dissenting. With respect, I cannot join the majority opinion, which is at odds with binding Supreme Court case law and creates circuit splits. First, the majority gives a backhand to the basic principle of constitutional avoidance and violates basic separation of powers principles. Second, if that were not enough, the majority‘s due process analysis is simply wrong and contrary to controlling law.
It is a “cardinal principle of judicial restraint,” that “if it is not necessary to decide more, it is necessary not to decide more.” PDK Lab‘ys Inc. v. DEA, 362 F.3d 786, 799 (D.C. Cir. 2004) (Roberts, J., concurring in part and dissenting in part). That principle is never more important than when we can resolve a case on statutory grounds to avoid reaching a constitutional question. See, e.g., Ashwander v. TVA, 297 U.S. 288, 347 (1936) (Brandeis, J., concurring). And our obligation to avoid a constitutional judgment becomes even stronger when doing so allows us to return decisions to politically accountable actors. Cf. Quill v. Vacco, 80 F.3d 716, 738-40 (2d Cir. 1996) (Calabresi, J., concurring in the judgment) (articulating theory of “constitutional remand“).
In these related cases challenging the Board of Immigration Appeal‘s (“BIA“) allocation of burdens in discretionary immigration bond proceedings to detained noncitizens
facing removal13 (“noncitizens” or “detainees“) in its 1999 decision In re Adeniji, 22 I. & N. Dec. 1102, 1113 (B.I.A. 1999) (en banc), which still controls today, the asserted violations of theI also dissent because the majority‘s due process holding is, in my view, quite wrong on the merits. No court should needlessly constitutionalize a rule that is better left to the executive and the Congress, which are, after all, responsive to the voters.
I.
We heard argument on the same day in three cases challenging the BIA‘s Adeniji decision, allocating the burdens of production and persuasion in discretionary immigration bond proceedings: this case; Doe v. Tompkins, No. 19-1368; and Pereira-Brito v. Garland, Nos. 20-1037 and 20-1119. In both Doe and Pereira-Brito, the plaintiffs pleaded their detention under Adeniji was illegal because Adeniji was in violation of the APA.14 I would resolve these cases on
Those APA arguments are properly before us. Though the courts below did not reach those arguments and though the plaintiffs have not pressed them robustly before us, we may decide a case on any grounds supported by the record. Steinke v. Sungard Fin. Sys., Inc., 121 F.3d 763, 768 (1st Cir. 1997). When we can avoid a constitutional question, we must turn to such other grounds even when the litigants lead with their constitutional claims. See Greenless v. Almond, 277 F.3d 601, 605-07 (1st Cir. 2002). Indeed, the Supreme Court has often endorsed a more lenient approach to ordinary waiver rules when that approach allows the Court to avoid thorny constitutional questions. See, e.g., Reno v. Flores, 507 U.S. 292, 300 n.3 (1993); Alma Motor Co. v. Timken-Detroit Axle Co., 329 U.S. 129, 142 (1946).
Were the Court to resolve Pereira-Brito and Doe as I propose, it would be appropriate to vacate the injunction and judgment and remand to the district court for further proceedings.
II.
Having established that the APA questions are properly before us, I turn to the regulatory and statutory context for the APA challenge.
Congress has long authorized the Attorney General to detain noncitizens in deportation proceedings.15 For most of the twentieth century, the relevant statutes vested the Attorney General with discretion to detain, release on bond, or conditionally parole such noncitizens.
In 1996, Congress passed the Illegal Immigration Reform and Immigrant Responsibility Act (“IIRIRA“), which categorically denied bond to noncitizens in deportation proceedings who have been convicted of aggravated felonies and certain other offenses.
Any officer authorized to issue a warrant of arrest [(i.e., immigration officials but not immigration judges)] may, in the officer‘s discretion, release [a noncitizen] not described in section 236(c)(1) of the Act [(a criminal noncitizen)], under the conditions at section 236(a)(2) [(permitting bond or parole)] and (3) [(prohibiting work authorization)] of the Act; provided that the [noncitizen] must demonstrate to the satisfaction of the officer that such release would not pose a danger to property or persons, and that the [noncitizen] is likely to appear for any future proceeding.
Despite this assertion of purported continuity, the final rule changed the presumption of release before immigration officials. In adopting the rule, the INS briefly explained the change relying heavily on a report from the Inspector General of the Department of Justice:
Several commenters stated that § 236 of the proposed rule as written is a reversal of long established procedure that provides that a noncriminal [noncitizen] is presumptively eligible for release. The Service has been strongly criticized for its failure to remove [noncitizens] who are not detained. A recent report by the Department of Justice Inspector General shows that when [noncitizens] are released from custody, nearly 90 percent abscond and are not removed from the United States. The mandate of Congress, as evidenced by budget enhancements and other legislation, is increased detention to ensure removal. Accordingly, because the Service believes that the regulation as written is consistent with the intent of Congress, the interim rule has not modified the proposed rule in this regard.
The regulations also provide for IJ review of initial bond determinations:
After an initial custody determination by the district director, including the setting of a bond, the respondent may, at any time before an order [of removal] becomes final, request amelioration of the conditions under which he or she may be released. Prior to such final order, and except as otherwise provided in this chapter, the [IJ] is authorized to exercise the authority in [8 U.S.C. § 1226] to detain the [noncitizen] in custody, release the [noncitizen], and determine the amount of bond, if any, under which the respondent may be released, as provided in § 3.19 of this chapter [(procedural rules)].
Following adoption of those regulations, the BIA abrogated Patel and stated -- not in a regulation but only in a reported decision in a single case -- that “for ordinary bond determinations [before IJs] under [§ 1226(a)] . . . [a noncitizen] must demonstrate that ‘release would not pose a danger to property or persons.‘” Adeniji, 22 I. & N. Dec. at 1113.
The BIA tried to justify its departure from Patel by relying on the new regulation, which did not concern IJs, and stated that the regulation required it to shift the burden of proof in detention proceedings before IJs. Id. at 1103, 1113. After determining that the regulations applied both during and after the transition period, id. at 1107-1112, the BIA held that:
[f]rom the outset . . . the regulations under the IIRIRA have added as a requirement for ordinary bond determinations under section 236(a) of the Act that the [noncitizen] must demonstrate that “release would not pose a danger to property or persons,” even though section 236(a) does not explicitly contain such a requirement. . . . We deem the regulatory provision at 8 C.F.R. § 236.1(c)(8) to contain the appropriate test, as it is binding on us and pertains directly to removal proceedings under the IIRIRA. Consequently, to be eligible for bond, the respondent must demonstrate that his “release would not pose a danger to property or persons, and that (he) is likely to appear for any future proceeding.”
Id. at 1113 (citation omitted).
III.
An agency‘s decision is arbitrary or capricious when it overlooks relevant issues or when it fails to “articulate a satisfactory explanation for its action.” Motor Vehicle Mfrs. Ass‘n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983). When an agency changes an established policy, it must show that “the new policy is permissible under the [relevant] statute, that there are good reasons for it, and that the agency believes it to be better.” FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009).
In my view, the Adeniji decision by the BIA is arbitrary and capricious. It rests on at least two erroneous and unreasoned administrative leaps. Further, I conclude the present regime is likely contrary to Congressional intent.
A.
The only reason the BIA offered for its departure from Patel was that
The text of
Indeed, the regulation which actually governs bond proceedings before IJs is different.
The regulation also is different in its description of the discretion IJs have in setting bond conditions. It authorizes IJs “to exercise the authority in [8 U.S.C. § 1226(a)] . . . to detain the [noncitizen] in custody, release the [noncitizen], and determine the amount of bond, if any, under which the respondent may be released.”
Our review is limited to reviewing the grounds the BIA offered for departing from Patel. SEC v. Chenery Corp., 318 U.S. 80, 94 (1943). The only grounds the BIA offered in Adeniji was that
B.
Not only did the BIA misinterpret
First, in adopting the regulation, the INS “entirely failed to consider an important aspect of the problem.” State Farm, 463 U.S. at 43. A key aspect of any detention regime is the relative dangerousness and flight risk of different classes of detainees. See, e.g.,
A few examples from these cases illustrate the wide range of risk different noncitizens pose. Doe was picked up after two serious criminal charges: carrying a weapon-sized knife and assault and battery. He did not even apply for asylum until after he requested a bond hearing, though he had three years to do so before his arrest.18 And while Hernández-Lara had not committed criminal offenses in the United States, an Interpol red notice said that she had done so in El Salvador and was a member of the Pandilla 18 street gang. If the IJ erred in initially denying bail based on that information, Hernández-Lara had an administrative appeal available to her, which the majority‘s opinion has pretermitted and necessarily concluded is inadequate under the Due Process Clause. On the other side of the scale, perhaps detainees who are veterans of the U.S. armed forces, and about whom the government consequently has more information, are themselves a special class.
Nor does the record reflect that the INS considered relative risk or burden as to several distinct categories of noncitizen as for which discretionary detention is authorized. The INS should have at least considered whether it was grouping like and unlike categories of discretionary detainees together under a blanket rule. Cf. Transactive Corp. v. United States, 91 F.3d 232, 237 (D.C. Cir. 1996) (“[A]n agency action is arbitrary when the agency offered insufficient reasons for treating similar situations differently.“). Since it did not, the agency‘s adoption of the rule was arbitrary or capricious.
Second, in adopting the regulation, the INS “offered an explanation for its decision that runs counter to the evidence before the agency.” State Farm, 463 U.S. at 43. An agency must “examine the relevant data and articulate a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” Id. (citation and internal quotation marks omitted). The INS explained that it shifted the presumption in Adeniji because of the INS‘s reliance on an Inspector General report that “show[ed] that when [noncitizens] are released from custody, nearly 90 percent abscond and are not removed from the United States.”
Given the agency‘s reference to irrelevant statistics alone to support the rule, I find its “reasoning to be inscrutable at best and, given the information available to the agency, facially irrational.” Marasco & Nesselbush, LLP v. Collins, No. 20-1397, 2021 WL 3012705, at *14 (1st Cir. July 16, 2021).
The INS acted arbitrarily and capriciously when it adopted
C.
The government offers three arguments for why the BIA‘s departure from Patel was not arbitrary or capricious. None are persuasive.
The government next argues that “the [BIA‘s] holding in [Adeniji] represents a reasonable interpretation of Section 1226(a) and is entitled to deference under Chevron principles.” But the BIA did not interpret
Finally, the government argues that Adeniji does not actually depart from prior decisions because it already had the authority to determine whether and how to release noncitizens on bond. Authority to act is necessary but not sufficient for an agency to change course. See Fox Television, 556 U.S. at 515. Even when an agency has broad authority, it must justify a change in how it exercises that authority. See New England Power Generators Ass‘n, Inc. v. FERC, 881 F.3d 202, 210 (D.C. Cir. 2018).
D.
Because the BIA‘s allocation of the burden of proof rests on arbitrary or capricious foundations, enforcing it against noncitizens in discretionary bond proceedings is unlawful.
IV.
I turn next to the majority‘s constitutional holding.
A.
“[P]rior to reaching any constitutional questions, federal courts must consider nonconstitutional grounds for decision.” Buchanan v. Maine, 469 F.3d 158, 172 (1st Cir. 2006) (quoting Gulf Oil Co. v. Bernard, 452 U.S. 89, 99 (1981)) (quotation marks omitted). As we can resolve this case on APA grounds, the majority‘s constitutional analysis is “unnecessary and, indeed, inappropriate.” Marasco & Nesselbush, 2021 WL 3012705, at *19.
On top of general principles of judicial restraint and constitutional avoidance, three considerations specifically support avoiding a constitutional ruling here.
The effect of the majority‘s opinion is to arrogate to the judiciary control over immigration bond procedures. In most areas of law, we should be cautious in constitutionalizing agency procedures. But in immigration,
Deciding this case on constitutional due process grounds, as the majority does, is premature and particularly ill-advised given the subject matter. See Clinton v. Jones, 520 U.S. 681, 690 & n.11 (1997). “One of the major advantages of [judicial] minimalism is that it grants a certain latitude to other branches of government by allowing the democratic process room to adapt to future developments, to produce mutually advantageous compromises, and to add new information and perspectives to legal problems.” Cass R. Sunstein, Foreword: Leaving Things Undecided, 110 Harv. L. Rev. 4, 19 (1996). In facially holding that a noncitizen may never bear the burden of proof in an immigration bond hearing, see infra Part IV.B, the majority shuns the benefits of further democratic development. Cf. Hightower v. City of Boston, 693 F.3d 61, 76-78 (1st Cir. 2012) (disfavoring facial challenges).
Further, since this litigation began, a new presidential administration has taken office and has begun to change immigration policy. See, e.g., Memorandum from David Pekose, Acting Sec‘y, Dep‘t Homeland Sec., Review of and Interim Revision to Civil Immigration Enforcement and Removal Policies and Priorities (Jan. 20, 2021) (announcing 100-day moratorium on most removal proceedings),
https://www.dhs.gov/sites/default/files/publications/21_0120_enforcement-memo_signed.pdf; Final Inadmissibility on Public Charge Grounds; Implementation of Vacatur, 86 Fed. Reg. 14221 (Mar. 15, 2021) (rescinding public charge rule). If we sent these burden of proof issues back to the BIA and required the agency to consider a wider range of circumstances, the agency may well produce a more nuanced set of bond standards. In short, we have the chance to maximize politically accountable deliberation and policy making; instead, the majority has chosen to make policy from the bench.
Finally, the majority‘s overreach will have serious practical consequences. Our immigration system is taxed to its limits.19
and by raising the required quantum of proof to detain a noncitizen in removal proceedings, the majority imposes additional strains on overburdened immigration courts and officials.20 In my view, the majority should have avoided unleashing those serious harms on our immigration infrastructure.
The majority contends that constitutional avoidance is unavailable to us in this case because deciding the APA claim in favor of Hernández-Lara would afford her only partial relief. That contention fails because equitable relief “must be ‘no more burdensome to the defendant than necessary to provide complete relief to the plaintiffs.‘” Sindi v. El-Moslimany, 896 F.3d 1, 31 (1st Cir. 2018) (quoting Madsen v. Women‘s Health Ctr., Inc., 512 U.S. 753 (1994)). If the government has acted unlawfully under the APA, she is entitled only to the necessarily relief as to that
injury. She is not entitled to the majority‘s adoption of the broad rule she proposes. This is a basic tenant of remedial law. If Hernández-Lara is entitled to any relief, that relief must be limited only to relief not more burdensome than necessary. In going beyond that relief, the majority again overreaches.21
This case demands judicial restraint. The majority opts instead for judicial hubris.
B.
Though the majority should not have reached the constitutional question, it did. I will briefly state why I think the majority‘s due process analysis is contrary to Supreme Court precedent, contrary to precedent from other circuits, and wrong. I do not take the occasion to expound on my views at great length.
The majority derives from the Due Process Clause a categorical rule. It holds that in all discretionary immigration bond cases the government must bear the burden of proving dangerousness by clear and convincing evidence and flight risk by the preponderance of the evidence. The Due Process Clause does not support that broad conclusion.
“In the exercise of its broad power over naturalization and immigration, Congress
it follows that detention of noncitizens under the government‘s current regime -- which allows noncitizens to present individualized evidence and rebut the presumption of detention -- does not offend the Due Process Clause either.
The majority also errs in rejecting the current bail detention scheme facially. A facial challenge to detention procedures fails if the procedures are “adequate to authorize the detention of at least some [persons].” Salerno, 481 U.S. at 751 (quoting Schall v. Martin, 467 U.S. 253, 274 (1984) (alteration in original)). And the current bond procedures provide robust enough bond procedures to provide many noncitizens constitutionally sufficient notice and opportunity to be heard.
Even under Mathews v. Eldridge balancing the government may require at least some noncitizens to prove that they are neither dangerous nor flight risks.22 424 U.S. 319, 335 (1976) (looking to “the private interest that will be affected by the official action,” “the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards,” and “the Government‘s interest, including the function involved and the
fiscal and administrative burdens that the additional or substitute procedural requirement would entail” to determine “the specific dictates of due process“).
First, while in general the private interest a person has in avoiding detention is strong, a noncitizen‘s interest is considerably more limited. “Detention during removal proceedings is a constitutionally permissible part of that process.” Demore, 538 U.S. at 531. The detainee‘s liberty interest is diminished by the fact that he could voluntarily remove himself
Second, the majority‘s adding to and altering of the already robust procedures would do little to improve the accuracy of bond determinations. Under current procedures, noncitizens may
introduce evidence to show that they will likely appear at their removal proceedings and are entitled to administrative and judicial review of any adverse bond determination. Indeed, the government is ill-positioned to have information beyond the criminal record.
Third, the government has a strong interest in effectively executing immigration law. “Further, it must weigh heavily in the balance that control over matters of immigration is a sovereign prerogative, largely within the control of the executive and the legislature.” Landon v. Plasencia, 459 U.S. 21, 34 (1982).
The current procedures provide detained noncitizens constitutionally sufficient notice and opportunity to be heard. Consider a noncitizen who is removable because he fled from a law enforcement checkpoint in a car. See
“It may be, of course, that in some circumstances detention of [a noncitizen] would not pass constitutional muster.
But the validity of those detentions must be determined on a case-by-case basis.” Schall, 467 U.S. at 273. The majority‘s overreaching conflicts with controlling Supreme Court precedents.
Although the majority admits that it fashions its analysis “broadly,” it contends that “judicial and administrative efficiency” justifies its holding. Maj. Op. 59. Like so many other problems of constitutional law, however, the level of generality at which we describe the problem is crucial to determining its outcome. See, e.g., Michael H. v. Gerald D., 491 U.S. 110, 127 n.6 (1989). We need not determine the level of due process required in every case through case-by-case adjudication; however, where courts can meaningfully distinguish between relevant categories, courts should not set standards at a greater level of generality. Compare Addington v. Texas, 441 U.S. 418, 431-33 (1979) (setting across-the-board standard for civil commitments on the basis of mental health given the inherent “uncertainties of psychiatric diagnosis“),
administrative burdens in proving the dangerousness or flight risk of -- veterans of the armed forces than noncitizens who have never been lawfully admitted. It also knows far more about permanent residents than those who overstay nonimmigrant visas. And, as the government has powerfully argued, it knows little about those who have recently entered the country illegally and been detained. The government‘s relative knowledge matters because it directly affects two of the key procedural due process considerations: risk of erroneous deprivation and governmental burden. The majority‘s analysis collapses those distinctions. In so doing, the majority both fails to actually apply the Mathews framework it purports to apply and reaches an overly broad holding.
C.
“[T]his issue is one where careful judicial consideration should not end with a three-judge panel, or even an en banc sitting of a circuit court of appeals, but with the Supreme Court of the United States.” Allapattah Servs., Inc. v. Exxon Corp., 362 F.3d 739, 741 (11th Cir. 2004) (Tjoflat, J., dissenting from denial of petition for rehearing en banc).
The majority‘s constitutional holding, as I have explained, “decide[s] an important federal question in a way that conflicts with relevant decisions of [the Supreme] Court.” Sup. Ct. R. 10(c). The Supreme Court should step in to bring our court
back into compliance with the Supreme Court‘s carefully considered precedents. Such an intervention would not be mere error correction: given the majority‘s facial holding, its error is not case specific. It will reverberate in thousands of immigration bond proceedings.
Additionally, the majority‘s decision conflicts with those of our sister circuits on a question of national importance. See Borbot v. Warden Hudson Cty. Corr. Facility, 906 F.3d 274 (3d Cir. 2018) (holding that initial bond hearing in which noncitizen carried the burden of proof satisfied due process, even when noncitizen had been detained for over 14 months).
In Borbot, the Third Circuit held that the Due Process Clause does not require the government to bear the burden of proof in bond proceedings. 906 F.3d at 279. The majority argues that “the issue presented here was not before the court in Borbot.” Maj. Op. 31. Not so. Borbot directly presented the question of whether the government must bear the burden of proof. The Third Circuit expressly ruled on that point of law, and it could not have justified its decision without that ruling. Had the Borbot court not rejected the petitioner‘s burden-of-proof argument, it could not have denied him a new hearing under different procedures. 906 F.3d at 277. Borbot‘s discussion of the burden of proof thus meets the textbook definition of a holding. See Garner, et al.,
The Law of Judicial Precedent 46 (2016). And the majority‘s holding squarely conflicts with it.
The majority also points to a subsequent Third Circuit decision, German Santos v. Warden Pike County Correctional Facility, 965 F.3d 203 (3d Cir. 2020), which it says “casts doubt” on the argument that Borbot accepted the
Further review of the majority‘s holding is warranted to resolve this circuit split and to bring the First Circuit back into compliance with controlling precedent.
V.
I would vacate the injunction and judgment and remand to the district court for further proceedings consistent with this opinion. I respectfully dissent.