Javier Martinez v. Lowell Clark

68 F.4th 1195

• Court: Court of Appeals for the Ninth Circuit
• Decided: May 30, 2023
• Docket: 21-35023
• Status: Published

Opinion

FOR PUBLICATION

UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT

JAVIER MARTINEZ, No. 21-35023

Petitioner-Appellant, D.C. No. 2:20-cv-00780- v. TSZ

LOWELL CLARK, Warden, Northwest Detention Center; ORDER NATHALIE ASHER, Tacoma Field Office Director, United States Immigration and Customs Enforcement; ALEJANDRO MAYORKAS, Secretary, Department of Homeland Security; MERRICK B. GARLAND, Attorney General,

Respondents-Appellees.

Filed May 30, 2023

Before: Jacqueline H. Nguyen, Eric D. Miller, and Patrick J. Bumatay, Circuit Judges.

Order; Statement Respecting the Denial of Rehearing En Banc by Judge Berzon 2 MARTINEZ V. CLARK

SUMMARY*

Immigration/Habeas/Detention

The panel denied a petition for panel rehearing and denied on behalf of the court a petition for rehearing en banc in a case in which the panel held that federal courts lack jurisdiction to review the discretionary determination that a particular noncitizen in immigration detention poses a danger to the community and so is not entitled to release on bond. Respecting the denial of rehearing en banc, Judge Berzon, joined by Chief Judge Murguia and Judges Wardlaw, W. Fletcher, Paez, Christen, Hurwitz, Koh, Sung, Mendoza, and Desai, disagreed with the Court’s refusal to reconsider the panel opinion en banc. Judge Berzon wrote that the panel’s characterization of the dangerousness determination as discretionary conflicts with longstanding precedents from the criminal bail context holding that dangerousness determinations are mixed questions of law and fact, subject to independent review. Judge Berzon also wrote that the panel’s ruling is at odds with Supreme Court guidance as to the sorts of determinations that constitute mixed questions rather than discretionary ones. Noting the critical importance of judicial review when liberty is at stake, Judge Berzon wrote that the panel’s ruling grants the government unconstrained discretion to determine whether individuals in removal proceedings should be detained based on dangerousness, without judicial backstop.

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. MARTINEZ V. CLARK 3

ORDER

The panel has voted to deny the petition for panel rehearing. The full court was advised of the petition for rehearing en banc. A judge requested a vote on whether to rehear the matter en banc, and the matter failed to receive a majority of the votes of the nonrecused active judges in favor of en banc consideration. See Fed. R. App. P. 35(a). The petition for panel rehearing and rehearing en banc is DENIED. An opinion respecting the denial of rehearing en banc, prepared by Judge Berzon, is filed concurrently with this order.

BERZON, Circuit Judge, with whom MURGUIA, Chief Judge, and WARDLAW, W. FLETCHER, PAEZ, CHRISTEN, HURWITZ, KOH, SUNG, MENDOZA, and DESAI, Circuit Judges, join, respecting the denial of rehearing en banc:

I respectfully disagree with this Court’s refusal to reconsider the panel opinion en banc. “Freedom from bodily restraint has always been at the core of the liberty protected by the Due Process Clause from arbitrary governmental action.” Foucha v. Louisiana, 504 U.S. 71, 80 (1992). For that reason, the Supreme Court has required “strong procedural protections”—including judicial review—when upholding preventative detention based on dangerousness. Zadvydas v. Davis, 533 U.S. 678, 691–92 (2001). Yet the panel in this case held that federal courts lack jurisdiction to review the Board of Immigration Appeals’ (“BIA”) determination that a noncitizen poses a 4 MARTINEZ V. CLARK

danger to the community and so is not entitled to be released from immigration detention on bond. See Martinez v. Clark, 36 F.4th 1219, 1224 (9th Cir. 2022). The panel concluded that a jurisdictional limitation in 8 U.S.C. § 1226(e), which applies to “the Attorney General’s discretionary judgment regarding the application of this section,” id., precludes review of dangerousness. Martinez, 36 F.4th at 1224, 1228. The panel’s characterization of the dangerousness determination as discretionary conflicts with longstanding precedents from the criminal bail context holding that dangerousness determinations are mixed questions of law and fact, subject to independent review. See, e.g., United States v. Howard, 793 F.3d 1113, 1113 (9th Cir. 2015) (per curiam); United States v. Motamedi, 767 F.2d 1403, 1405–06 (9th Cir. 1985). And the panel’s ruling is at odds with Supreme Court guidance as to the sorts of determinations that constitute mixed questions rather than discretionary ones. See, e.g., Guerrero-Lasprilla v. Barr, 140 S. Ct. 1062, 1069–70 (2020); U.S. Bank Nat’l Ass’n v. Vill. at Lakeridge, LLC, 138 S. Ct. 960, 967–68 (2018). For these reasons, this Court should have reconsidered the panel opinion en banc. I. Under 8 U.S.C. § 1226, the government has authority to detain noncitizens present in the United States during the pendency of removal proceedings. For most noncitizens, the “default rule”—set forth in subsection (a) of 1226—is that the government has statutory authority to release them on bond. See 8 U.S.C. § 1226(a)(2); Jennings v. Rodriguez, 138 S. Ct. 830, 846 (2018). In such bond hearings, release turns on whether the noncitizen poses a danger to persons or property, a threat to national security, or a flight risk. See MARTINEZ V. CLARK 5

Matter of Guerra, 24 I. & N. Dec. 37, 38 (BIA 2006) (citing Matter of Adeniji, 22 I. & N. Dec. 1102 (BIA 1999)); 8 C.F.R. § 1236.1(c)(8); see also Singh v. Holder, 638 F.3d 1196, 1206 (9th Cir. 2011). Section 1226(c) departs from the default rule by specifying categories of noncitizens who, like Martinez, are subject to mandatory detention because of criminal offenses or terrorist activities. The government generally has no statutory authority to release noncitizens covered by section 1226(c). See Jennings, 138 S. Ct. at 846-47. But here, the district court held that because Martinez’s mandatory detention was prolonged, “due process requires the government to show by clear and convincing evidence that the detainee presents a flight risk or a danger to the community.” Martinez v. Clark, No. 18-CV-01669-RAJ, 2019 WL 5962685, at *1 (W.D. Wash. Nov. 13, 2019); see also, e.g., German Santos v. Warden Pike Cnty. Corr. Facility, 965 F.3d 203, 210–11 (3d Cir. 2020) (holding that noncitizens subject to “unreasonably long” detention under section 1226(c) have a due process right to a bond hearing); Reid v. Donelan, 17 F.4th 1, 8 (1st Cir. 2021) (rejecting across-the-board rule that all section 1226(c) detainees have a constitutional right to a bond hearing once detained for longer than six months, but recognizing “the possibility that in most individual cases, detentions of six months (or of even less time) might necessitate some type of hearing to see if continued detention is reasonably necessary to serve the statute’s purposes”). In Martinez’s bond proceedings, the IJ and BIA denied him release, concluding based on his years- old drug convictions that—notwithstanding his subsequent good conduct—he is a danger to the community. The panel in this case held that the district court lacked jurisdiction to review the dangerousness determination. 6 MARTINEZ V. CLARK

Martinez, 36 F.4th at 1228. In doing so, the panel invoked another subsection of 1226, subsection (e), which imposes limits on judicial review. Section 1226(e) provides:

The Attorney General’s discretionary judgment regarding the application of this section shall not be subject to review. No court may set aside any action or decision by the Attorney General under this section regarding the detention or release of any alien or the grant, revocation, or denial of bond or parole.

Section 1226(e) “applies only to ‘discretionary’ decisions about the ‘application’ of § 1226 to particular cases.” Nielsen v. Preap, 139 S.Ct. 954, 962 (2019) (plurality opinion). It “does not limit habeas jurisdiction over constitutional claims or questions of law.” Singh, 638 F.3d at 1202. According to the panel, the dangerousness determination is unreviewable under section 1226(e) because the inquiry lacks any ascertainable legal standards, is “fact-intensive,” “subjective[,] and value-laden,” and is therefore “purely discretionary.” Martinez, 36 F.4th at 1228–30 (internal quotations marks and citation omitted). This holding both mischaracterizes the nature of dangerousness determinations and misapplies the principles that govern which decisions involve discretionary questions as opposed to legal questions. MARTINEZ V. CLARK 7

II. The nature of the dangerousness determination here may seem like an esoteric jurisdictional question. But getting it right is of enormous practical importance to a great many individuals. The panel assumed that a bond hearing required under the Due Process Clause for noncitizens detained under section 1226(c) is subject to the dangerousness standard applicable to statutory bond hearings for noncitizens detained under section 1226(a). See Martinez, 36 F.4th at 1226, 1228–29 (citing Singh, 638 F.3d at 1206, and Guerra, 24 I. & N. Dec. at 40). The panel’s jurisdictional ruling thus precludes court review of dangerousness determinations for all noncitizens detained pending their removal proceedings under section 1226, not just noncitizens like Martinez who are subject under the statute to mandatory detention because of their criminal record. Whether the government has unreviewable discretion to determine if a noncitizen should be detained as a danger to the community is a question of considerable constitutional significance. “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, 533 U.S. at 690. Under the panel’s ruling, the government could deem anyone dangerous and detain them for years while their removal case slowly works its way through the system; the constitutional protection of liberty would be eviscerated. But the Supreme Court has “upheld preventive detention based on dangerousness only when limited to specially dangerous individuals and subject to strong procedural protections.” Id. at 691. Allowing noncitizens to be 8 MARTINEZ V. CLARK

detained for prolonged periods where “the sole procedural protections available . . . are found in administrative proceedings” would raise an “obvious” constitutional problem. Id. at 692. “[T]he Constitution may well preclude granting an administrative body the unreviewable authority to make determinations implicating fundamental rights.” Id. (internal quotation marks and citation omitted). Apart from the judicial review question, the panel’s conclusion that dangerousness, and therefore release from immigration detention, “is a ‘subjective question that depends on the identity and the value judgment of the person or entity examining the issue,’” Martinez, 36 F.4th at 1227 (citation omitted), is profoundly troubling from a constitutional perspective. In upholding the Bail Reform Act against a due process challenge, the Supreme Court emphasized that “[t]he judicial officer is not given unbridled discretion in making the detention determination.” United States v. Salerno, 481 U.S. 739, 742, 751–52 (1987). The panel here, in contrast, concluded that essentially the same determination in the immigration context is wholly subjective at the agency level, as well as dependent on the identity and values of the decisionmaker—in other words, it is subject to “unbridled discretion.” Were that true, there would almost surely be a due process violation. Conditioning release from detention entirely on the identity of the decisionmaker or the decisionmaker’s personal tastes or feelings offends the central purpose of the Due Process Clause—protecting individuals from “arbitrary detention.” Rodriguez v. Marin, 909 F.3d 252, 255, 257 (9th Cir. 2018) (citation omitted); see also Foucha, 504 U.S. at 80. MARTINEZ V. CLARK 9

III. We have recognized in the criminal bail context that the determination of dangerousness is governed by ascertainable standards, holding squarely that such a determination is a mixed question of law and fact subject to independent review. See, e.g., Howard, 793 F.3d at 1113; United States v. Hir, 517 F.3d 1081, 1086–87 (9th Cir. 2008); Motamedi, 767 F.2d at 1405–06. The dangerousness determination in the immigration context is directly analogous. Yet the panel opinion does not mention the bail cases at all. That gap is telling. Had the panel acknowledged the bail precedents, it would have had to explain why the immigration bond determination regarding dangerousness lacks judicially cognizable legal standards and is therefore unreviewable, Martinez, 36 F.4th at 1228–29, when courts in the criminal bail context routinely review directly parallel determinations independently and have done so for decades. 1. The Bail Reform Act provides for release of a criminal defendant pending trial “unless the judicial officer determines that such release will not reasonably assure the appearance of the person as required or will endanger the safety of any other person or the community.” 18 U.S.C. § 3142(b). We have long held that in considering a pre-trial release determination, the appellate court reviews “the district court’s factual findings under a deferential, clearly erroneous standard,” but

the conclusion based on those factual findings presents a mixed question of fact and law. The inquiry transcends the facts presented and requires both the consideration 10 MARTINEZ V. CLARK

of legal principles and the exercise of sound judgment about the values which underly those principles.

Motamedi, 767 F.2d at 1405, 1406; United States v. Townsend, 897 F.2d 989, 994 (9th Cir. 1990). “In light of the important constitutional dimensions involved” in applications for release from detention, the appellate court has “a nondelegable responsibility to make an independent determination of the merits of the application.” Motamedi, 767 F.2d at 1405 (citation omitted). In particular, with respect to “the danger that [the detainee] poses to the community,” the court’s factual findings are reviewed for clear error, but “[t]he conclusions based on such factual findings . . . present a mixed question of fact and law.” Howard, 793 F.3d at 1113; see also Hir, 517 F.3d at 1086. The majority of circuits likewise independently review bail release determinations while deferring to the district court’s findings of subsidiary facts. See, e.g., United States v. Portes, 786 F.2d 758, 763 (7th Cir. 1985) (holding that determining whether a defendant “pose[s] a danger to the community . . . is a judgmental function [as to which] we . . . must engage in an ‘independent review’ of the case.”); United States v. Delker, 757 F.2d 1390, 1399, 1400–01 (3d Cir. 1985) (independently reviewing dangerousness determination and stating that independent review is “appropriate in light of the nature of the question to be determined” because “[a] crucial liberty interest is at stake”); United States v. Patriarca, 948 F.2d 789, 791 (1st Cir. 1991) (independently reviewing denial of release based on dangerousness); United States v. Stone, 608 F.3d 939, 945 (6th Cir. 2010) (reviewing denial of release based on dangerousness after explaining that “[w]e review the district MARTINEZ V. CLARK 11

court’s factual findings for clear error, but we consider mixed questions of law and fact—including the ultimate question whether detention is warranted—de novo”); United States v. Cisneros, 328 F.3d 610, 613, 618–19 (10th Cir. 2003) (similar); United States v. Quartermaine, 913 F.2d 910, 915 (11th Cir. 1990) (similar); United States v. Sazenski, 806 F.2d 846, 847 (8th Cir. 1986) (similar).1 2. The determination of whether a person will “endanger the safety of any other person or the community” in the bail context, 18 U.S.C. § 3142(b), is directly analogous to the dangerousness determination in the immigration context. The Supreme Court has long analogized immigration detention to criminal detention and immigration bond to criminal bail. See, e.g., Wong Wing v. United States, 163 U.S. 228, 233, 235 (1896); Zadvydas, 533 U.S. at 690–92 (citing Salerno, 481 U.S. at 747, 750–52). The substantive standards for dangerousness in the two contexts are essentially the same, and the pertinent factual and equitable considerations are as well. In both settings, the decisionmaker considers whether the historical facts give rise to an inference that the applicant for release poses a danger to the community. See, e.g., Motamedi, 767 F.2d at 1407; Singh, 638 F.3d at 1206. Comparing the case law in both contexts demonstrates that the panel was wrong to

1 A few circuits characterize dangerousness for bail purposes as a finding of fact subject to clear error review. See United States v. Manafort, 897 F.3d 340, 346 n.2 (D.C. Cir. 2018); United States v. English, 629 F.3d 311, 319 (2d Cir. 2011); United States v. Clark, 865 F.2d 1433, 1437 (4th Cir. 1989) (en banc). The Fifth Circuit applies a deferential standard of review similar to abuse-of-discretion. See United States v. Moreno, 857 F.3d 723, 725–26 (5th Cir. 2017). 12 MARTINEZ V. CLARK

conclude that there are no applicable “standards sufficient to permit meaningful judicial review” of the dangerousness determination for purposes of an immigration bond hearing. See Martinez, 36 F.4th at 1229 (citation omitted). “The decision whether to admit a defendant to bail . . . must often turn on a judge’s prediction of the defendant’s future conduct.” Jurek v. Texas, 428 U.S. 262, 275 (1976). To assess whether it is safe to release an individual for bail purposes, a district court takes into account multiple factors, including “the nature and circumstances of the offense charged,” the “history and characteristics of the person,” including “past conduct” and “criminal history,” and “the nature and seriousness of the danger” posed by the individual. 18 U.S.C. § 3142(g). The district court typically looks to objective sources concerning the individual’s history of violence: “prior convictions, police reports, and other investigatory documents” which “are, as a matter of course, used to show past histories of violence.” Motamedi, 767 F.2d at 1407. There is no question that there are legal standards applicable to such review. The Supreme Court has specifically so recognized, rejecting the notion that a requirement that a decisionmaker assess the likelihood that an individual “would constitute a continuing threat to society” relies on a standard that is “so vague as to be meaningless.” Jurek, 428 U.S. at 272, 274. “[T]here is nothing inherently unattainable about a prediction of future criminal conduct.” Salerno, 481 U.S. at 751 (quoting Schall v. Martin, 467 U.S. 253, 278 (1984)). Dangerousness determinations in the immigration context are no less subject to meaningful legal standards sufficient to permit judicial review. Consistent with the MARTINEZ V. CLARK 13

precedents from the bail context, we have explained that to determine dangerousness in section 1226(a) bond hearings, immigration judges must consider a person’s “criminal record, including the extensiveness of criminal activity, the recency of such activity, and the seriousness of the offenses.” Singh, 638 F.3d at 1206 (quoting Guerra, 24 I. & N. Dec. at 40). Singh further explained that “criminal history alone will not always be sufficient to justify denial of bond on the basis of dangerousness,” because “the recency and severity of the offenses must [also] be considered.” Id. In other words, just as in the bail context, the court considers objective evidence concerning the immigration detainee’s past conduct and criminal history to make a prediction about likely future conduct. See Motamedi, 767 F.2d at 1407. As in bail cases, the essential question is whether the evidence “demonstrate[s] a propensity for future dangerousness.” Singh, 638 F.3d at 1205. The BIA’s analysis in Guerra reinforces that, as in the bail context, sufficient legal standards do exist to permit meaningful review here. Guerra explained that other past conduct short of a criminal conviction is relevant to determining dangerousness:

[A]lthough we recognize that the respondent has not been convicted of the offenses charged in the criminal complaint, we find that unfavorable evidence of his conduct, including evidence of criminal activity, is pertinent to the Immigration Judge’s analysis regarding . . . danger to the community.

24 I. & N. Dec. at 41. Thus, IJs “are not limited to considering only criminal convictions in assessing whether 14 MARTINEZ V. CLARK

an alien is a danger to the community. Any evidence in the record that is probative and specific can be considered.” Id. at 40–41. Applying these standards to the evidence in Guerra, the BIA affirmed the IJ’s determination. Id. at 41. In other words, the BIA itself reviewed the IJ’s dangerousness decision by applying legal standards to the objective facts. There is no reason why the same legal standards are sufficiently enunciated for BIA review of IJ decisions but not for court review of BIA decisions. Nor is the BIA’s own characterization of the bond determination as discretionary pertinent. Guerra reasoned that that provision “gives the Attorney General the authority to grant bond if he concludes, in the exercise of discretion, that the alien’s release on bond is warranted.” 24 I. & N. Dec. at 39. But the agency’s own characterization of the nature of the decision – as opposed to its description of the substance of the standard – is not controlling for purposes of deciding a federal court’s jurisdiction. Otherwise “the Executive would have a free hand to shelter its own decisions from abuse-of-discretion appellate court review simply by issuing a regulation declaring those decisions ‘discretionary.’” Kucana v. Holder, 558 U.S. 233, 252 (2010). IV. As the panel acknowledged, the jurisdictional prohibition in section “1226(e) does not limit habeas jurisdiction over ‘constitutional claims or questions of law.’” Martinez, 36 F.4th at 1227; see also Singh, 638 F.3d at 1202. In concluding that “danger to the community” is a purely discretionary determination and so not a question of law, the panel reasoned that the determination requires a “fact- intensive,” “multi-factorial analysis with no clear, uniform MARTINEZ V. CLARK 15

standard for what crosses the line into dangerousness.” Martinez, 36 F.4th at 1228–29. This test for identifying an unreviewable discretionary judgment is fundamentally flawed. Nearly every consideration the panel identified to support the conclusion that the dangerousness determination is discretionary is also applicable to legal questions involving the application of law to fact, as the Supreme Court and our court have recognized. Recent Supreme Court precedent reflects that many legal questions involving the application of law to fact, often called “mixed questions,” are fact-intensive, subject to a “broad . . . standard,” and require balancing multiple facts or considerations. See U.S. Bank Nat’l Ass’n, 138 S. Ct. at 967–68; see also Guerrero-Lasprilla, 140 S. Ct. at 1069. The Court has also recognized that the application of law to fact entails consideration of competing values. See Miller v. Fenton, 474 U.S. 104, 115–16 (1985). So, contrary to the analysis of the panel in this case, the characteristics relied upon by the panel cannot serve as litmus tests for discretionary decisions. U.S. Bank National Ass’n and Guerrero-Lasprilla refute the panel’s conclusion that an inquiry must be discretionary if it is fact-intensive. As Guerrero-Lasprilla recognized, some mixed questions of law and fact “immerse[] courts in case-specific factual issues.” 140 S. Ct. at 1069 (citation omitted). The mixed question in U.S. Bank National Ass’n was “fact-intensive” and required “[p]recious little” legal work. 138 S. Ct. at 968. Some mixed questions may “compel[] [courts] to marshal and weigh evidence, make credibility judgments, and otherwise address . . . ‘multifarious, fleeting, special, narrow facts that utterly resist generalization.’” Id. at 967 (quoting Pierce v. Underwood, 487 U.S. 552, 561-562 (1988)). And mixed 16 MARTINEZ V. CLARK

questions in the constitutional context may “primarily involve[] plunging into a factual record.” Id. at 967 n.4. Also, mixed questions of law and fact can entail balancing multiple facts or weighing competing concerns. U.S Bank National Ass’n explained that a mixed question may require a court to “weigh evidence” and “balance [the facts] one against another.” 138 S. Ct. at 967–68. Guerrero- Lasprilla held that the fundamentally equitable question whether an individual acted with due diligence for purposes of equitable tolling is a question of law involving a mixed question of law and fact. See 140 S. Ct. at 1068. Mixed questions of law and fact may also entail consideration of underlying values. For example, Miller held that “the ultimate issue of ‘voluntariness’” of a confession, for purposes of determining whether it was obtained in violation of due process, “is a legal question requiring independent federal determination,” 474 U.S. at 110, even though the voluntariness inquiry “subsum[es] . . . a ‘complex of values,’” id. at 116 (citation omitted). See also, e.g. United States v. McConney, 728 F.2d 1195, 1202, 1204–05 (9th Cir. 1984) (en banc) (recognizing that the application of law to undisputed fact can require the court “to exercise judgment about the values that animate legal principles” and “balance competing legal interests”). Nor does the absence of a legal standard that mandates a “certain” outcome, Martinez, 36 F.4th at 1229, render an issue discretionary. It is commonplace for legal standards to “be given concrete meaning through a process of case-by- case adjudication.” I.N.S. v. Cardoza-Fonseca, 480 U.S. 421, 448 (1987). Some mixed questions “require courts to expound on the law, particularly by amplifying or MARTINEZ V. CLARK 17

elaborating on a broad legal standard.” U.S. Bank Nat’l Ass’n, 138 S. Ct. at 967. Legal inquiries involving the weighing of multiple factors, without a standard that mandates a particular result, are legion. Take, for example, the familiar question whether a police officer’s use of force was excessive under the Fourth Amendment—a question that is determined under the multi- part balancing test of Graham v. Connor, 490 U.S. 386, 396– 97 (1989). See, e.g., Glenn v. Washington Cnty., 673 F.3d 864, 871 (9th Cir. 2011). There is no legal standard in excessive force cases that mandates a particular outcome in all instances. But the Supreme Court has held that once the facts have been established, whether the totality of the circumstances “warrant[s] deadly force . . . is a [] question of law.” Scott v. Harris, 550 U.S. 372, 381 n.8 (2007) (citation omitted); see also, e.g., United States v. Phelps, 955 F.2d 1258, 1265–66 (9th Cir. 1992) (explaining that the process due under the three-part balancing test of Mathews v. Eldridge, 424 U.S. 319, 335 (1976), is a “question of law”); Moser v. Las Vegas Metro. Police Dep’t, 984 F.3d 900, 905 (9th Cir. 2021) (explaining that the multi-part First Amendment “Pickering balancing test presents a question of law”); Alpha Indus., Inc. v. Alpha Steel Tube & Shapes, Inc., 616 F.2d 440, 443-44 (9th Cir. 1980) (explaining that under the multi-factor AMF Inc. v. Sleekcraft Boats, 599 F.2d 341 (9th Cir. 1979), test for trademark confusion, the “determination of likelihood of confusion based on th[e] factors is a legal conclusion”). The bottom line is that multi- factor standards that require weighing competing interests are commonly understood to constitute legal standards, not to constitute subjective, purely discretionary, unreviewable decisionmaking. 18 MARTINEZ V. CLARK

**** Judicial review is of critical importance when liberty is at stake. See Zadvydas, 533 U.S. at 692; Motamedi, 767 F.2d at 1405. And that’s precisely what’s on the line here: the dangerousness determination at issue can often make the difference between years in detention awaiting a final removal decision and liberty during that period. The “prediction of future criminal conduct is an essential element in many of the decisions rendered throughout our” system of justice, and so it is a “task performed countless times each day throughout the American system of criminal justice.” Jurek, 428 U.S. at 275-76. In our Circuit as well as others, that determination is subject to independent judicial review for criminal bail purposes. See supra, Part III.1. Yet the panel’s decision concludes that for immigration detainees, there are no cognizable legal standards that would permit judicial review of the analogous determination in bond cases. In so doing, the panel grants the government unconstrained discretion to determine whether individuals in civil removal proceedings should be detained based on dangerousness, without judicial backstop. I seriously disagree with this Court’s decision to deny rehearing en banc. Should the issue arise again once the case law on the implications of U.S. Bank National Ass’n and Guerrero-Lasprilla is better developed, I hope the issue will be revisited.