Martinez-Viguerias: The BIA Must Give Cogent, Record-Engaging Reasons to Reject IJ CAT Factfinding Under Clear-Error Review

1. Introduction

Case: Martinez-Viguerias v. Blanche, No. 25-9556 (10th Cir. Aug. 25, 2026).
Parties: German Martinez-Viguerias (lawful permanent resident; petitioner) vs. Todd Blanche, U.S. Attorney General (respondent).
Core issue: Whether the Board of Immigration Appeals (BIA) properly applied clear-error review when it reversed an Immigration Judge’s (IJ’s) predictive factual findings supporting deferral of removal under the Convention Against Torture (CAT).
Holding in brief: The Tenth Circuit held the BIA did not properly apply clear-error review; it effectively reweighed evidence and failed to grapple with the IJ’s key, petitioner-specific factual findings. The court vacated the BIA decision and remanded for the BIA to apply clear-error review correctly.

Procedural posture highlights: After multiple IJ decisions and BIA remands, IJ McGrail granted CAT deferral based on individualized risk factors (tattoos, former gang membership, mental illness manifestations, inability to obtain Mexican ID, and country conditions of corruption/complicity). The BIA vacated, labeling several IJ findings “clearly erroneous” for being too “generalized.” The Tenth Circuit granted the petition for review and remanded.

2. Summary of the Opinion

The court reiterated that, by regulation, the BIA must review IJ factual findings “for clear error, and only clear error.” Although the BIA recited the correct standard, the court found it misapplied that standard when it rejected three IJ findings:

  • Likelihood of torture: The IJ’s over-50% predictive finding was grounded in petitioner-specific risk factors and record evidence; the BIA rejected it with generalized critiques that did not meaningfully address those specifics.
  • Ability to relocate within Mexico: The IJ concluded relocation would not mitigate risk given petitioner’s visible markers and cartel reach; the BIA again dismissed the finding as “generalized” without engaging the key individualized evidence.
  • Government acquiescence: The IJ found torture would occur by or with acquiescence of officials (including willful blindness). The BIA’s rejection was inadequately explained and, to the extent it demanded proof of acquiescence to the petitioner’s specific torture, it applied an incorrect legal requirement.

As to remedy, the court declined to order reinstatement of CAT deferral (as it had in Ramos v. Bondi), emphasizing the general rule favoring remand to the agency. It remanded for the BIA to either accept the IJ’s findings or provide a supportable clear-error basis for rejecting them.

3. Analysis

A. Precedents Cited

1) Kabba v. Mukasey

Kabba v. Mukasey, 530 F.3d 1239 (10th Cir. 2008), supplies the decision’s doctrinal spine:

  • Standard of review architecture: Courts review the BIA’s legal determinations de novo, while BIA fact findings receive substantial-evidence review.
  • Regulatory constraint on the BIA: Under 8 C.F.R. § 1003.1(d)(3)(i), the BIA must review IJ factual findings only for clear error.
  • Appellate check: Whether the BIA actually honored clear-error review (as opposed to reweighing evidence) is reviewed de novo by the court of appeals.

The Tenth Circuit used Kabba to frame the central inquiry: not whether the court agrees with the IJ, but whether the BIA had a legally valid clear-error basis to reverse the IJ’s factfinding.

2) Ramos v. Bondi

Ramos v. Bondi, 155 F.4th 1154 (10th Cir. 2025), provided both the clear-error definition and the “warning signs” that the BIA has overstepped:

  • Clear error is a “high bar”: reversal is improper merely because the reviewer would have decided differently.
  • Indicators of misuse: A BIA decision is suspect when it fails to address “key factual findings,” re-weights certain facts, or does not explain why the IJ’s view lacked “logic, plausibility or support in the record.”

Ramos also influenced remedy analysis. The court distinguished its earlier decision to remand with instructions to reinstate the IJ order (due to extreme delay and futility) and chose a more traditional agency remand here.

3) Villalta Martinez v. Bondi

The court borrowed heavily from Villalta Martinez v. Bondi, 157 F.4th 108 (2d Cir. 2025), for the discipline required under clear-error review:

  • What counts as clear error: examples include no evidentiary support, contradiction by indisputable evidence, or obvious misunderstanding of testimony.
  • “Cogent reasons” requirement: the BIA must supply cogent reasons when declaring an IJ finding clearly erroneous.
  • Proper remand instruction: on remand, the BIA should “either accept the IJ’s findings or, if it can, provide a supportable basis for rejecting them.”

In Martinez-Viguerias, the “cogent reasons” concept is operationalized: generalized statements about “pervasive violence” being insufficient do not qualify as cogent if they fail to engage the IJ’s petitioner-specific reasoning.

4) Uc Encarnacion v. Bondi

The court cited Uc Encarnacion v. Bondi, 156 F.4th 927 (9th Cir. 2025), to rebut a common agency move in CAT cases: dismissing evidence as “general possibility” when the record actually ties risk to the applicant’s characteristics.

  • The Ninth Circuit’s critique—that a generic “general possibility” statement may be true but irrelevant to individualized evidence—was deployed to show the BIA’s reasoning here had “little relationship” to what the IJ actually relied upon.
  • Uc Encarnacion also supported the proposition that evidence of targeting of a group can be sufficiently particularized for CAT purposes.

5) Karki v. Holder

Karki v. Holder, 715 F.3d 792 (10th Cir. 2013), controlled the acquiescence analysis:

  • Definition: a public official must have awareness of torturous activity and then breach the legal duty to prevent it.
  • Mens rea: “willful blindness” is enough; actual knowledge or willful acceptance is not required.
  • No “specific threat notice” requirement: the applicant need not prove officials knew of a specific threat to the applicant personally; evidence of regular failure to prevent/punish torture can satisfy acquiescence.

The panel used Karki to identify legal error in any BIA suggestion that the IJ had to show the Mexican government would acquiesce “specifically to Petitioner’s future torture.”

B. Legal Reasoning

1) The court’s central move: separating “insufficient evidence” from “clear error”

The opinion draws a sharp line between (a) an appellate body concluding it would weigh evidence differently and (b) the regulatory authority to reverse only for clear error. The BIA’s stated rationale—evidence was “too generalized” and did not show “frequency” for “similarly situated” persons—was treated as an improper re-assessment of probative weight rather than a demonstration that the IJ’s findings were illogical, implausible, unsupported, or contradicted by indisputable evidence.

2) Predictive findings are factual findings

The case reinforces that CAT determinations often depend on predictive factual findings about what is likely to happen on return. The court treated the IJ’s findings (likelihood of detection; likely harm; inability to relocate; acquiescence) as factual determinations protected by clear-error deference at the BIA level.

3) What the BIA failed to do (and must do on remand)

The court repeatedly faulted the BIA for not engaging the IJ’s “key factual findings,” including:

  • Petitioner’s inability to obtain Mexican identification and resulting vulnerability to police stops, bribery, detention.
  • Petitioner’s visible tattoos (including gang and religious tattoos) and evidence tying tattoos to victimization risk.
  • Petitioner’s former gang status, potential recruitment/punishment dynamics, and transnational gang/cartel attention.
  • Petitioner’s noticeable mental health manifestations and heightened visibility/vulnerability.
  • Country-conditions evidence about corruption, collusion, torture to extract confessions, and low accountability.

Under the court’s approach, a proper clear-error reversal would need to explain why these petitioner-specific inferences are not just debatable, but clearly mistaken on the whole record.

4) Remedy: why the court remanded rather than ordering CAT relief

While acknowledging detention length, the court adhered to the general administrative-law principle (also recognized in Ramos v. Bondi) that appellate courts typically remand to agencies for matters primarily committed to agency hands. Unlike Ramos, the court did not find the case-specific equities and futility concerns strong enough to justify a directive to reinstate CAT deferral.

C. Impact

1) Constraining BIA “clear error” reversals in CAT cases

The decision strengthens a practical rule: if an IJ ties CAT risk to individualized markers (appearance, documentation barriers, mental illness, gang history) and anchors those markers in record evidence, the BIA cannot label the finding “generalized” without directly addressing the IJ’s chain of reasoning and explaining why it collapses under the clear-error standard.

2) Reinforcing that acquiescence does not require “personalized government intent”

By invoking Karki v. Holder, the opinion underscores that applicants can prove acquiescence through systemic patterns—corruption, collusion, willful blindness, failure to prevent/punish—without proving officials had prior notice of the applicant’s exact future torture scenario.

3) Appellate litigation strategy and agency drafting practices

  • For practitioners: build a record that explicitly connects country conditions to applicant-specific traits; highlight how the IJ made “predictive factual findings” and demand clear-error discipline on appeal.
  • For the BIA: if reversing, it must do more than assert “generalized evidence”; it must confront the IJ’s specific factual predicates and explain—under clear-error logic—why they cannot support the prediction.
  • For IJs: the decision rewards structured CAT analyses that articulate (i) what will likely happen, (ii) why, (iii) the torture nexus, (iv) acquiescence, and (v) relocation, with explicit citations and individualized reasoning.

4. Complex Concepts Simplified

  • Clear-error review: A very deferential standard. The reviewer may not reverse just because it disagrees. Reversal is appropriate only when the finding is plainly mistaken on the whole record (e.g., no evidence supports it, it contradicts indisputable evidence, or it rests on a clear misunderstanding).
  • Predictive factual findings: Fact determinations about future likelihood—what will probably happen if the person is removed (e.g., likely to be stopped, detained, targeted, tortured). These are treated as factual findings for BIA review.
  • CAT “more likely than not”: The applicant must show a greater-than-50% chance of torture if removed.
  • Torture under CAT (as used here): Severe pain or suffering inflicted by, at the instigation of, or with the consent or acquiescence of a public official (or someone acting in an official capacity).
  • Acquiescence / willful blindness: Officials need not expressly approve the torture. It can be enough that they are aware (or deliberately avoid knowing) and breach their legal duty to prevent it.
  • Internal relocation: Whether the person could avoid torture by moving to a different region of the country. If the risk is effectively nationwide (or the applicant’s identifying traits travel with them), relocation may not be safe or reasonable.

5. Conclusion

Martinez-Viguerias v. Blanche is a standard-of-review enforcement decision: the Tenth Circuit vacated the BIA because its “clear error” reversals read like reweighing and generalized dismissal rather than a cogent demonstration that the IJ’s predictive factfinding was plainly mistaken. The opinion also reaffirms that CAT acquiescence does not require proof the government would knowingly acquiesce in this applicant’s torture specifically; systemic willful blindness and failure to prevent/punish can suffice. On remand, the BIA must either accept the IJ’s findings or supply a genuinely supportable clear-error basis for rejecting them.