PUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
No. 21-2074
O’NEIL LEWIS KERR, Petitioner,
v.
MERRICK B. GARLAND, Attorney General, Respondent.
On Petition for Review of an Order of the Board of Immigration Appeals.
Argued: December 6, 2022 Decided: April 24, 2023
Before HARRIS and QUATTLEBAUM, Circuit Judges, and KEENAN, Senior Circuit Judge.
Petition for review denied by published opinion. Judge Harris wrote the opinion, in which Judge Quattlebaum and Senior Judge Keenan joined.
ARGUED: Peter Cameron Alfredson, CAPITAL AREA IMMIGRANTS’ RIGHTS (CAIR) COALITION, Washington, D.C., for Petitioner. Liza Murcia, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Respondent. ON BRIEF: Melody L. Vidmar, Alison Steffel, CAPITAL AREA IMMIGRANTS’ RIGHTS (CAIR) COALITION, Washington, D.C., for Petitioner. Brian M. Boynton, Principal Deputy Assistant Attorney General, Anthony C. Payne, Assistant Director, Office of Immigration Litigation, Civil Division, UNITED STATES DEPARTMENT OF JUSTICE, Washington, D.C., for Respondent.
2
PAMELA HARRIS, Circuit Judge:
O’Neil Lewis Kerr petitions for review of the denial of his claim to protection under the Convention Against Torture. The immigration judge found that Kerr, a bisexual man and former gang member, had not shown the requisite likelihood that he would be tortured if returned to his home country of Jamaica. Kerr now challenges that finding on appeal, arguing that it does not properly account for his aggregate risk of torture as required by our decision in Rodriguez-Arias v. Whitaker, 915 F.3d 968 (4th Cir. 2019). We disagree and deny the petition for review.
I.
A.
O’Neil Lewis Kerr is a native and citizen of Jamaica. In 1990, at the age of ten, he entered the United States as a lawful permanent resident to live with his father in Maryland. Kerr’s father, unfortunately, was a high-ranking member of a notorious transnational Jamaican gang known as the Shower Posse, and Kerr developed a close relationship with many of his father’s fellow gang members. Kerr joined the Shower Posse when he was in his late teens, and by the early 2000s, he was trafficking cocaine on behalf of the gang.
According to Kerr, he was soon targeted for harm by Shower Posse members for two reasons. First, in 2003 or 2004, he began an intimate relationship with a man named Kenny, who worked as a driver for the Shower Posse and accompanied Kerr on drug transports. The Shower Posse was aggressively homophobic, viewing the prospect of a gay gang member as a “stain on their reputation,” A.R. 459, and when they discovered
3
Kerr’s relationship with Kenny, Shower Posse members verbally denigrated and physically threatened him. Second, Kerr, who believed he was not being fairly compensated for his work, began to steal drugs from the Shower Posse and sell them for his own profit. This too elicited a hostile response from suspicious gang members, and over a period of several years, Kerr was shot at and threatened with death on multiple occasions.
In approximately 2009, Kerr, fearful for his safety, ended his relationship with Kenny and left the Shower Posse, moving away and changing his name and appearance to distance himself from the gang. By then, however, he had already been convicted several times in Maryland court of drug-related offenses, all stemming from his cocaine trafficking as a Shower Posse member.
B.
In 2020, the Department of Homeland Security served Kerr with a notice to appear, charging him with removability based on his criminal convictions. See 8 U.S.C. § 1227(a)(2)(A)(iii) (removal based on aggravated felony); 8 U.S.C. § 1227(a)(2)(B)(i) (removal based on controlled substance offense). Kerr, through counsel, conceded his removability as charged.
Kerr then applied for deferral of removal under the Convention Against Torture (“CAT”), which provides for relief if an applicant establishes that it is “more likely than not” that he would be tortured if removed to his home country. See 8 C.F.R. § 1208.16(c)(2). Torture, a “term of art” under the CAT, see Turkson v. Holder, 667 F.3d 523, 526 (4th Cir. 2012), is an “extreme form of cruel and inhuman treatment” rising to the level of “severe pain or suffering,” intentionally inflicted by or with the “consent or
4
acquiescence” of a government, 8 C.F.R. § 1208.18(a); see Turkson, 667 F.3d at 526. In his application, Kerr claimed that he would be tortured and killed if returned to Jamaica, primarily because of his affiliation and disputes with the Shower Posse and his sexual orientation. In support of his application, Kerr submitted, among other evidence, an expert declaration of Dr. Damion Blake, Ph.D. and general country-condition reports and articles focused on gang activity and treatment of the LGBT community in Jamaica.
An immigration judge (“IJ”) heard Kerr’s case in March of 2021. As Kerr presented his case, he feared torture at the hands of four actors if returned to Jamaica: the Shower Posse gang, other gangs in Jamaica, Jamaican authorities, and Jamaican civilians. Each actor, he contended, was more likely than not to target him for torture, for sometimes overlapping reasons including his prior affiliation with the Shower Posse, his theft from the gang, his sexual orientation, and his status as a deportee. And once the risks from those four potential sources were aggregated, as required by our decision in Rodriguez-Arias, the likelihood of torture became something well over 50 percent. See A.R. 1463 (“The independent likelihood of torture from each of these four actors surpasses 50%, meaning that Mr. Kerr has more than met his burden under CAT when aggregating the risks.”).
In a 16-page single-spaced opinion, the IJ began by analyzing the likelihood that Kerr would be tortured in Jamaica by each of the four actors he had identified, concluding that the risk as to each was minimal or not significant. The IJ started with Kerr’s lead claim, based on the Shower Posse’s prior threats against him: that the Shower Posse would torture or kill him if he were returned to Jamaica. Given Kerr’s history with the Shower Posse, the IJ concluded, it did appear that gang members, at least at one time, were primed
5
to do him harm, either because he stole from them or because of his sexual relationship with Kenny. Where Kerr’s claim faltered, the IJ determined, was in connecting that feared harm to Jamaica: All of Kerr’s problems with the Shower Posse originated and occurred in the United States, and while Kerr still feared gang reprisals in this country, there was “little to indicate” that he would face such risks were he removed to Jamaica. A.R. 83.
The IJ thoroughly analyzed the record evidence on this point, starting with the fact that Kerr had not been in contact with Shower Posse members in Jamaica. Nor, contrary to Kerr’s suggestion, was there evidence that Shower Posse members in the United States who knew of Kerr’s problems with the gang had been deported to Jamaica, where they could lie in wait for him. And in any event, the IJ determined, it was unlikely that Shower Posse members were now intent on finding and torturing Kerr in Jamaica, given the many years since Kerr’s offenses against the gang and his last contact with its members. 1 Finally, though country-condition reports confirmed the prevalence of gang violence in Jamaica, there was limited information about the Shower Posse specifically, and nothing to demonstrate that Kerr himself would be targeted. All told, the IJ concluded, while the fact that Kerr “defied and deserted the gang a number of years ago” put him at some risk of torture by the Shower Posse in Jamaica, that risk was “well below 50%.” A.R. 84.
1
Kerr’s expert, Dr. Blake, suggested that the Shower Posse also might target Kerr in Jamaica because they assumed he was an informant, cooperating with the police. The IJ considered that testimony but was unpersuaded, explaining that it remained “unclear [] exactly why Jamaican Shower Posse members would make that connection, given that [Kerr] never worked with the U.S. or Jamaican authorities and does not intend to.” A.R. 84.
6
“For similar reasons,” the IJ found only a “minimal” risk of torture at the hands of other Jamaican gangs. Id. Again, while country-condition reports showed generalized concern about rival gangs engaging in reprisal killings, there was little to suggest that Kerr personally would be targeted: He was no longer a member of the Shower Posse; he had no contact with gang members in Jamaica, Shower Posse or otherwise; and it did not appear that he was known to Jamaican gangs.
With respect to the last two sources of torture identified by Kerr, Jamaican civilians and Jamaican authorities, the IJ credited reports showing widespread discrimination against both deportees – “shunned as undesirable criminals” – and the LGBT community. A.R. 85. Deportees like Kerr could face stigma and harassment from the population and heightened scrutiny and monitoring from the police. And the country-condition evidence with respect to the LGBT community was still “more troubling,” showing “alarmingly high” levels of homophobia and discrimination by civilians and police alike. Id. “Such conditions are truly disheartening and surely indicate [Kerr] will likely face such discrimination in Jamaica as a bi-curious man.” Id. 2 But that was a different question, the IJ emphasized, than whether Kerr would likely suffer the kind of extreme harm that rises to the level of torture under the CAT. The IJ carefully examined the country-condition evidence and data regarding physical attacks and
2
Kerr self-identified as “bicurious” before the IJ. In his brief on appeal, Kerr identifies himself as bisexual, and we follow suit in this opinion. Both the agency and Kerr’s appellate brief use “LGBT” to refer to lesbian, gay, bisexual, and transgender people, and we likewise adopt that terminology here.
7
threats by Jamaican civilians against LGBT people – some but not all of which would qualify as torture for CAT purposes – and extrajudicial violence by Jamaican authorities against deportees or members of the LGBT community. Such cases, the IJ determined, though “certainly alarming,” were not so prevalent that they established any “significant” risk of torture to Kerr. A.R. 86.
Having evaluated separately the risk of torture from each actor identified by Kerr, the IJ moved on to the inquiry at the heart of this appeal: whether the aggregated risk of torture from all four sources exceeded 50 percent. As the IJ explained, “[u]nder Fourth Circuit precedent, the risks of torture from all sources must be aggregated when determining whether an individual is more likely than not to be tortured in a particular country.” A.R. 81 (citing Rodriguez-Arias v. Whitaker, 915 F.3d 968 (4th Cir. 2019)). But here, the IJ concluded, even when the risks of torture were “considered cumulatively,” Kerr had “not shown that his aggregate likelihood of torture-level harm, as opposed to other, lesser harm or increased scrutiny, exceeds 50%.” A.R. 87. 3 Kerr appealed to the Board of Immigration Appeals (“BIA” or “Board”), arguing, as relevant here, that the IJ had erred in its aggregation analysis, failing to properly consider all potential sources and motives for torture. The Board disagreed and dismissed Kerr’s
3
The IJ went on to find that Kerr in any event was ineligible for CAT relief because he had not shown a likelihood that any severe harm he might face would be with the consent or acquiescence of the Jamaican government. See 8 C.F.R. § 1208.18(a). The Board of Immigration Appeals did not reach this issue – because, as detailed below, it agreed that Kerr had not made a threshold showing that he was likely to face harm rising to the level of torture – and we likewise have no occasion to address it here.
8
appeal. In the Board’s view, the IJ had properly aggregated the risk from all sources, consistent with Rodriguez-Arias, in determining whether Kerr was more likely than not to be tortured in Jamaica. The BIA was satisfied that the IJ considered each facet of Kerr’s identity and reason he might be tortured in reaching its conclusion, and it discerned no “clear error” in the IJ’s ultimate “predictive finding” that even taken in the aggregate, the risk that Kerr would experience harm rising to the level of torture was less than 50 percent. A.R. 7; A.R. 3 (setting out “clear error” standard of review for IJ factual findings).
Kerr timely filed this petition for review.
II.
A.
When, as here, the BIA adopts and affirms an IJ’s decision and supplements it with an opinion of its own, we review both decisions. Ibarra Chevez v. Garland, 31 F.4th 279, 289 (4th Cir. 2022). We review factual findings for substantial evidence, “treating them as conclusive unless any reasonable adjudicator would be compelled to conclude to the contrary.” Id. at 288–89 (internal quotation marks omitted). Legal conclusions are reviewed de novo. Id. at 288.
As noted above, an applicant for deferral of removal under the CAT bears the burden of showing that it is “more likely than not” that he will suffer harm rising to the level of torture in the country of removal. See Rodriguez-Arias, 915 F.3d at 971; 8 C.F.R. § 1208.16(c)(2). As the IJ here explained, only “severe pain or suffering” qualifies as “torture” under the CAT. See Rodriguez-Arias, 915 F.3d at 971; 8 C.F.R. § 1208.18(a)(1).
9
And as both the IJ and the BIA recognized, we have held that in making this “more likely than not” assessment, the agency must consider the “cumulative probability of torture,” aggregating the risk of torture from all feared sources. See Rodriguez-Arias, 915 F.3d at 972; see also Ibarra Chevez, 31 F.4th at 289.
Kerr’s central argument on appeal is that the agency violated both the letter and the spirit of Rodriguez-Arias, failing to properly aggregate all potential sources of torture and reasons for torture. Because we find no defect in the agency’s aggregation analysis or overall assessment of Kerr’s CAT claim, we deny Kerr’s petition for review.
B.
As we have explained, in Rodriguez-Arias, “we joined our sister circuits in holding ‘that the risks of torture from all sources should be combined when determining whether a CAT applicant is more likely than not to be tortured in a particular country.’” Ibarra Chevez, 31 F.4th at 289 (quoting Rodriguez-Arias, 915 F.3d at 973). Under this “aggregation rule,” an applicant need not establish that the probability of torture from any one source is over 50 percent. Id. Instead, the question for the IJ is “whether the cumulative probability of torture by all the entities, or for all reasons, exceeds 50%.” Id. (quoting Marqus v. Barr, 968 F.3d 583, 589 (6th Cir. 2020)).
We also have clarified that the IJ need not engage in “statistical or quantitative analysis” – or indeed, use any “specific methodology” – when considering and aggregating multiple risks. Id. at 290. In Ibarra Chevez, the IJ, after considering the likelihood of torture from three separate sources, simply stated that “the threat to [Ibarra] from [all] sources is less than a fifty percent chance of torture both separately and in the aggregate,”
10
and the BIA affirmed those findings as supported by substantial evidence. Id. at 289, 291. That explanation and analysis was sufficient, we held, because we could “tell that the IJ and BIA” had followed Rodriguez-Arias by “combin[ing] the threats” in deciding that the petitioner had not shown the requisite likelihood of torture. Id. at 290.
Here, too, it is abundantly clear that both the IJ and the BIA applied the aggregation rule of Rodriguez-Arias, considering not only the individual risk of torture from each actor identified by Kerr but also the cumulative probability of torture. The IJ recognized from the start, in laying out the applicable law, that “the risks of torture from all sources must be aggregated when determining whether an individual is more likely than not to be tortured in a particular country.” A.R. 81 (citing Rodriguez-Arias, 915 F.3d 968). And at the end of her lengthy opinion, the IJ did just that, identifying “some risk of harm from each possible source” but finding that “even when considered cumulatively,” Kerr had not shown an “aggregate likelihood of torture-level harm” that exceeded 50 percent. A.R. 87. Likewise, the BIA expressly acknowledged and applied Rodriguez-Arias, holding that the IJ properly considered and aggregated all relevant risks and that her ultimate factual finding – that Kerr had not established the requisite aggregate risk – was supported by the record and not clearly erroneous. At this basic level, in other words, there is no question but that the agency complied with our aggregation rule from Rodriguez-Arias. Cf. Quintero v. Garland, 998 F.3d 612, 645 (4th Cir. 2021) (remanding where Board “inexplicably failed to even mention or discuss, let alone aggregate, the risk of torture from the different sources alleged”).
11
1.
On appeal, Kerr argues primarily that the IJ nevertheless erred in her aggregation analysis by failing to consider certain reasons why Kerr might be tortured by certain actors. To be clear, there is no dispute that the IJ methodically identified and aggregated the four potential sources of torture alleged by Kerr and considered multiple potential reasons for such torture. Kerr argues, however, that this painstaking analysis overlooked the risks associated with three specific pairings of source and reason: the risk that the Shower Posse would torture Kerr based on his sexual orientation, the risk that other gangs in Jamaica would do the same, and the risk that Jamaican authorities would torture Kerr because of his former gang membership. Like the BIA, we disagree, and are satisfied that the agency’s aggregation analysis properly accounted for all relevant risks.
First, the IJ did consider the risk that the Shower Posse would target Kerr because of his sexual orientation. As the BIA explained, the IJ appears to have discounted that risk somewhat, noting that the gang continued to work with Kerr long after discovering his relationship with Kenny. Nevertheless, the IJ was persuaded that Shower Posse members in the United States might target Kerr for harm based on his sexual orientation. A.R. 82 (“[I]t appears that some Shower Posse members in the U.S. may have a problem with him or see him as having defied the gang in some way, whether it is because he stole from the gang, his relationship with Kenny, or other unknown reasons.” (emphasis added)). It is true that the IJ did not expand on this point – but that is because she also found that whatever the nature of the risks from Shower Posse members in the United States, Kerr had failed to establish that those risks would carry over to Jamaica were he removed.
12
Under those circumstances, there was no reason for the IJ to further dissect the fault lines between Kerr and his fellow gang members in the United States.
With respect to the risk from other gangs in Jamaica, we agree with Kerr that the IJ’s analysis focused on the prospect that Jamaican gangs might target him for “reprisal” killings, rather than for harm based on sexual orientation. That is likely because Kerr’s presentation before the IJ also focused on reprisal killings by rival gangs, 4 and nowhere identified Kerr’s sexual orientation as a potential ground for abuse by other Jamaican gangs. Indeed, as the BIA pointed out, the entire record contains only a single reference to Jamaican gangs as a threat to the LGBT community: one sentence from Kerr’s expert witness reporting – as part of his discussion of the Shower Posse itself – that “Jamaican gangs have a zero-tolerance policy for homosexual and bisexual men.” A.R. 4. There is nothing to identify any non-Shower-Posse Jamaican gang as a threat, or to tie any such threat to Kerr. Kerr did argue, at length, that Jamaican society more broadly is hostile and violent toward LGBT people – and the IJ carefully parsed that evidence when she analyzed Kerr’s fear of torture from the Jamaican population as a whole. On this record, we are persuaded that the agency appropriately accounted for the risk of sexual-orientation-based torture from both the Shower Posse and other Jamaican gangs.
4
To the extent Kerr also suggested that Jamaican gangs might target him on behalf of the Shower Posse, that concern was addressed by the IJ when she analyzed the Shower Posse itself as a potential source of torture and found it unlikely that the gang would attempt to reach out to Jamaica to harm Kerr so many years after their last United States interaction.
13
Finally, the IJ carefully considered the risk that Jamaican authorities would torture Kerr not only because of his sexual orientation but also because of his status as a deportee, removed from the United States after criminal convictions. Kerr now contends that the IJ missed a separate potential reason for torture by Jamaican authorities, in the form of his status as a former gang member. But as Kerr understood when he was before the IJ, “former gang member” and “criminal deportee” amount to much the same thing for these purposes: Because both are associated with a criminal element, both are subject to special scrutiny by Jamaican authorities. See A.R. 1467 (arguing that “the Jamaican Government is more likely than not to torture Mr. Kerr because he is a former gang member and criminal deportee”); A.R. 244 (arguing that the Jamaican police kill and detain “those that are believed are gang members and/or are responsible for crime” or “criminal deportees”). In any event, as the Board explained, Kerr’s evidence on this point goes only to the targeting of suspected current gang members, and Kerr had not shown “why Jamaican authorities would suspect him of participating in gang activity” at this late date. A.R. 5.
2.
On appeal, Kerr also suggests that the IJ did not sufficiently explain her aggregation analysis, failing to show her work and giving only a “rote” recitation of her conclusion. See Reply Br. of Petr. at 3 (describing analysis as “aggregation in name only”). That argument is largely foreclosed by Ibarra Chevez – decided, we note, only after briefing in this case. In Ibarra Chevez, we sustained an aggregation finding presented in terms very similar to those used by the IJ here, see 31 F.4th at 289, holding expressly that no “specific methodology” is required and that an IJ need only say enough that we can “tell that [she]
14
combined the threats from each source” in determining the likelihood that an applicant would be tortured if removed, id. at 290 (cleaned up).
That aggregated-risk finding, of course, remains subject to review under the substantial-evidence standard. See id. at 291 (holding that substantial evidence supported IJ finding that aggregate risk of torture was less than 50 percent). And there may be cases in which that review will require more by way of explanation. See Zelaya v. Holder, 668 F.3d 159, 168 (4th Cir. 2012) (remanding where lack of “reasoned explanation” impaired “meaningful[] review”). If, for instance, an IJ has intimated, or it otherwise appears from the record, that the likelihood of torture from each of several individual sources is close to 50 percent, but the IJ nevertheless puts the combined risk at less than 50 percent, then some further explanation may be necessary. In this case, however, the IJ made clear that there was no substantial risk of torture-level harm from any individual source: from the Shower Posse, “well below 50%,” A.R. 84; from other gangs, only “minimal,” id.; and from Jamaican civilians and authorities, something short of “significant,” A.R. 86. Like the Board, we think those findings are substantially supported by the record, 5 and from them,
5
To the extent Kerr argues on appeal not only that the agency overlooked certain risks in its aggregation analysis but also that its findings are not supported by substantial evidence, we disagree. Contrary to Kerr’s suggestion, there is no reason to think that the agency ignored any portion of his country-condition or expert reports. On the contrary, the IJ engaged thoroughly with that evidence, see A.R. 83–86; as the BIA explained, though the IJ did not adopt certain of Dr. Blake’s expert testimony as her findings, that does not mean she did not consider that testimony, and she fully explained why she was not persuaded by it, see A.R. 5. Nor was the agency “required to discuss every piece of evidence in the record.” Ai Hua Chen v. Holder, 742 F.3d 171, 179 (4th Cir. 2014). The IJ and BIA said more than enough that we can “perceive that they have heard and thought”
15
we “may reasonably [] discern” the thinking behind the IJ’s ultimate and aggregated risk determination. See Bowman Transp. Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974). Under Ibarra Chevez, no more is required.
3.
Kerr’s last contention on appeal is that in considering and aggregating the risks of torture from various sources, the IJ failed to recognize that different facets of his identity might combine with each other to put him at an especially high risk. For instance, though the IJ considered both his risk of torture as a former gang member and his risk of torture as a bisexual man, she did so only separately, without evaluating his vulnerability as someone who is both a former gang member and also bisexual – a joint identity that is “more than the sum of its parts” when it comes to the risk of torture. Br. of Petr. at 11. The BIA rejected this argument, finding that the IJ appropriately considered the “demonstrated” risks associated with the facets of Kerr’s identity. A.R. 4. We agree.
The problem with Kerr’s argument is not its premise. Though the question is not squarely before us, we have no reason to doubt that if there were record evidence of the “compounding effect” hypothesized by Kerr – that one or more of the actors he identified were particularly likely to target persons with his combined set of characteristics – the agency would be obliged to consider that evidence. See 8 C.F.R. § 1208.16(c)(3) (providing that in “assessing whether it is more likely than not that an applicant would be
about the evidence, and we are satisfied that the agency sufficiently accounted for the record in front of it. Id. (internal quotation marks omitted).
16
tortured in the proposed country of removal, all evidence relevant to the possibility of future torture shall be considered”) (emphasis added); Rodriguez-Arias, 915 F.3d at 972– 73 (applying § 1208.16(c)(3) in formulating aggregation rule). Indeed, we do not understand the government to argue to the contrary. See Br. of Resp. at 31 (discussing Kerr’s claim regarding the “interconnected facets” of his identity and explaining that § 1208.16(c)(3) requires consideration of all relevant evidence).
The problem, rather, is that Kerr presented no such evidence before the agency and points to none on appeal. Before the IJ and BIA, Kerr identified no record evidence suggesting that he would be singled out for torture as, say, a “bisexual former gang member” and thus subjected to a risk of torture greater than the “sum of its parts” – greater, that is, than the risk captured by aggregating the likelihood of torture based on sexual orientation with the risk of torture based on former gang membership. Instead, Kerr proffered his evidence to the agency in precisely the form the agency analyzed it: Proceeding through his list of four potential sources of torture, he discussed separately each reason – sexual orientation, gang affiliation, deportee status, and the like – he might be harmed, and then called for aggregation of those separate risks under Rodriguez-Arias. See A.R. 1465–72 (brief before IJ). 6 The agency addressed Kerr’s case as it was presented,
6
The only record support Kerr can cite for his claim is a stray reference in the report of his expert, Dr. Blake, to Kerr’s sexual orientation “compound[ing]” the risk that Kerr would be tortured by the Shower Posse for stealing. See A.R. 610. We do not think this single use of the word “compound” can carry the weight Kerr assigns it. To our mind, Dr. Blake’s statement suggests only that Kerr faced torture at the hands of the Shower Posse for two different reasons that should be considered cumulatively – as the agency did here, under Rodriguez-Arias.
17
and we have no grounds for disturbing its application of Rodriguez-Arias’s aggregation rule to the evidence before it.
III.
For the foregoing reasons, we deny the petition for review.
PETITION FOR REVIEW DENIED