Past Torture and CAT: How Do You Connect Earlier Harm to the Risk of Return?

Earlier torture can be powerful evidence of future danger, but it does not create an automatic regulatory presumption that a Convention Against Torture claim succeeds. The useful task is to explain what the earlier harm shows about the risk on return: Who inflicted it, why, whether those actors can still reach the person, and what has changed since then.

That is different from minimizing past abuse or merely repeating that the applicant bears the burden. 8 C.F.R. section 1208.16(c)(2)–(3) requires a more-likely-than-not showing of future torture and consideration of all relevant evidence, expressly including past torture, relocation evidence and country conditions.

Identify what the earlier decision actually found

Read the immigration judge’s findings rather than substitute a broad summary such as “the judge believed me.” A credibility finding, a past-persecution finding and a finding that conduct met the legal definition of torture address different questions. The same incident can be relevant across them without making the findings interchangeable.

The CAT definition addresses severe intentionally inflicted physical or mental suffering and the required official involvement, instigation, consent or acquiescence, among other requirements. Persecution and torture are not simply different labels for any serious harm. See section 1208.18(a). A focused review identifies which findings were made, which were assumed for analysis and which elements remained disputed.

The statutory-withholding regulation expressly provides a future-threat presumption following qualifying past persecution. The CAT provisions use a different structure. Importing the former presumption into CAT skips the separate forward-looking inquiry. Our comparison of protection standards explains the broader distinctions; this guide focuses on the evidentiary connection between past harm and present risk.

Build a “then and now” comparison that answers the future-risk question

Organize the comparison around the feared actors and the mechanism of risk, not just the passage of time.

These are evidence questions, not a requirement that every applicant produce a new personal threat or one particular document. Test the explanation against the complete record. If an official position ended, explain why the former role still matters—or acknowledge the evidence suggesting reduced interest. If relatives have been questioned, identify what was said about the applicant rather than assume every family incident has the same cause.

Explain changes instead of treating them as automatic answers

Release from detention may bear on future risk, but its meaning depends on the circumstances. An unconditional release, an escape, intervention by others and a release accompanied by continuing threats are different evidence. Likewise, a period without further harm needs context: Where was the person, what did the feared actors know and what ability did they have to act?

Do not hide facts that appear unfavorable. A reliable account explains travel, continuing residence, changed employment, protective measures and later contact alongside the original harm. The objective is a coherent account of why risk continues or has changed, not a narrative that treats every event as favorable by definition.

Use country evidence to explain the connection

Select passages addressing the relevant actors, methods, groups, locality or treatment of returnees. Identify the report’s date and source and explain how its information bears on the applicant’s history and proposed return. General evidence may be important; the explanation should show the proposition it supports rather than dismiss it as merely general or expect the decision-maker to infer every connection.

Preserve conflicting information too. A report describing national improvements may not establish safety for a particular person; a report documenting widespread abuse may not resolve the precise probability disputed in a particular record. The regulation requires consideration of all relevant evidence, not a fixed incident count or a mechanical probability calculation.

Zahid illustrates the disputed inference—not a California outcome rule

In Zahid v. Blanche, Nos. 25-60296 and 25-60524 (5th Cir. Oct. 20, 2026), the majority rejected an automatic CAT presumption from a past-persecution finding and concluded that the existing record did not compel a finding of the necessary future risk. It assessed the prior harm alongside later circumstances. The dissent disagreed with the separate treatment of new evidence that could show continued targeting based on a perceived protest role. Official opinion, pages 5–14 and 17–21.

The Fifth Circuit decision is not controlling Ninth Circuit precedent. Its practical value here is identifying the exact inference being disputed and the evidence relevant to it. Our case analysis explains the majority, dissent and different review standards without merging them.

Keep the evidence’s procedural home visible

Mark whether each item was before the immigration judge, supplied to the Board or obtained later. A strong factual point can still require the appropriate procedural vehicle. The new-evidence guide addresses that separate stage-dependent question.

For review, bring the decisions, relevant testimony, prior and current reports, later documents and a short comparison connecting them. Identify the actual adverse finding and both the strongest support and the hardest contrary fact. That gives counsel a basis to evaluate a present-risk argument under the governing law without treating past harm as either an automatic win or legally unimportant.