Unreported / Non-Citable
Background
Richard Anthony Herod was convicted in Texas state court of aggravated sexual assault and aggravated robbery and sentenced to 99 years in prison. At trial, a state DNA analyst testified that Herod could not be excluded as a contributor to a DNA mixture on a white T-shirt used to blindfold one victim. Herod’s expert agreed that he technically could not be excluded under the method used, although she believed he realistically should have been excluded.
In 2017, the Texas Department of Public Safety issued a supplemental report applying updated interpretation protocols and excluding Herod from all DNA evidence, including the mixture on the T-shirt. After the Texas Court of Criminal Appeals dismissed his second state habeas application on procedural grounds, Herod pursued federal relief under 28 U.S.C. § 2254. The district court found violations of Brady v. Maryland and Napue v. Illinois and granted habeas relief.
On rehearing, the Fifth Circuit granted panel rehearing, denied rehearing en banc, withdrew its May 20, 2026 opinion, and substituted this opinion. Judge Richman concurred in the judgment only.
The Court’s Holding
The Fifth Circuit agreed that Herod had shown sufficient prejudice to overcome the procedural default of his Brady and Napue claims. The DNA mixture evidence was the only physical evidence placing him at the crime scene, the prosecution emphasized it at trial, and the remaining evidence was contested. The court therefore concluded that the DNA-related issues undermined confidence in the verdict enough to clear the procedural bar.
Nevertheless, the court held that habeas relief was unavailable. Teague v. Lane barred Herod’s theory that the prosecution suppressed the fact of his exclusion because that exclusion was produced years after trial under updated protocols, and existing precedent did not require prosecutors to create exculpatory evidence through different forensic testing. Herod’s remaining Brady theory failed because the scientific debate over the DNA-mixture method was publicly available and thus was not suppressed; any undisclosed knowledge by DPS employees of that debate was not material.
The Napue claim also failed because Herod did not establish that the analyst’s challenged statements were actually false and known, or reasonably knowable, by the state to be false at trial. The CPI method was not scientifically unacceptable at the time, and later expert disagreement or updated scientific practices did not show that prosecutors knowingly presented false testimony. The court reversed the grant of habeas relief and remanded for proceedings on Herod’s remaining habeas claims, without requiring an evidentiary hearing on the Brady and Napue claims.
Key Takeaways
- A habeas petitioner may establish enough prejudice to overcome procedural default yet still lose because Teague or the merits foreclose relief.
- Brady does not require prosecutors to create favorable evidence by performing different forensic tests, and publicly available scientific criticism generally is not suppressed evidence.
- Later changes in forensic protocols or expert disagreement do not, without more, establish that trial testimony was false or that prosecutors knew or should have known it was false under Napue.
Why It Matters
The decision draws a sharp distinction between evidence that later science shows to be inaccurate and a constitutional violation cognizable on federal habeas review. Even though the parties apparently agreed by the time of appeal that Herod should have been excluded from the T-shirt DNA mixture, the court held that this later conclusion could not be converted into a Brady or Napue violation on the record presented.
The opinion also illustrates that clearing a procedural-default barrier does not resolve the underlying constitutional claim. Courts must separately address nonretroactivity and each claim’s substantive elements before granting habeas relief.