Private: Who Bites the Bullet?

by Brandon L. Garrett, Justice Thurgood Marshall Distinguished Professor of Law, University of Virginia School of Law. Last fall, Harvard University Press published his new book, Too Big to Jail: How Prosecutors Compromise with Corporations.

Yesterday, the Supreme Court issued a per curiam ruling in Kulbicki v. Maryland. The ruling was brief but unusual; the Court does not often take certiorari to review state habeas rulings. Here, the Maryland Court of Appeals had granted habeas to a prisoner whose trial lawyer had utterly failed to challenge an FBI agent’s testimony about Comparative Bullet Lead Analysis, or CBLA. The FBI agent did not find an “exact” match but sufficient similarity to conclude that the bullet that killed the murder victim came from Kulbicki’s weapon (and also matched a fragment in his truck). The problem was that this CBLA bullet analysis was flawed science. The National Academy of Sciences concluded in a 2004 report that "The available data do not support any statement that a crime bullet came from a particular box of ammunition.” Fundamental flaws in the assumptions and empirical basis for CBLA analysis led Maryland courts to reject CBLA evidence 15 years later, and the FBI to itself later disavow and discontinue use of the technique in 2005.  

Should Kulbicki’s lawyer have known back in 1995, at the time of the trial, that this was flawed science? Kulbicki argued that a report co-authored by the analyst showed how the FBI analyst had doubts even in 1991 “that bullets produced from different sources of lead would have a unique chemical composition.” The Court rejected the notion, saying that “At the time of Kulbicki’s trial in 1995, the validity of CBLA was widely accepted, and courts regularly admitted CBLA evidence until 2003.” Further, “Given the uncontroversial nature of CBLA at the time of Kulbicki’s trial,” it would be asking lawyers to "go looking for a needle in a haystack” to search for such evidence that the forensics were flawed.

Compare the Court’s ruling in Kulbicki to last year’s per curiam opinion in Hinton v. Alabama, another case examining a lawyer’s failure to adequately develop forensic evidence at trial, including firearms and tool mark analysis. The outcome was different. In Hinton, a death penalty case, the Court found the lawyer to have been constitutionally ineffective. (On remand, Hinton’s conviction was vacated). From the beginning, the tone in the two per curiam opinions could not have been more different. In Hinton, the Court correctly stated the Strickland v. Washington constitutional standard for ineffective assistance of counsel as asking “if his trial attorney’s performance falls below an objective standard of reasonableness and if there is a reasonable probability that the result of the trial would have been different absent the deficient act or omission.” In Kulbicki, the Court oddly misstated the standard as “meaning his errors are ‘so serious’ that he no longer functions as ‘counsel,’ and prejudicial, meaning his errors deprive the defendant of a fair trial.” That description of the Strickland test was, at the very least, a casual and imprecise one.

Now, in Hinton, counsel mistakenly thought that the court would not pay for a better ballistics expert, so did not ask for one. The Court there emphasized: “The trial attorney’s failure to request additional funding in order to replace an expert he knew to be inadequate because he mistakenly believed that he had received all he could get under Alabama law constituted deficient performance.” Did counsel in Kulbicki similarly fail to make reasonable investigations and inquiry? We now know that the FBI’s bullet lead analysis was unscientific and flawed – and that aggressive findings of a “match" never should have been allowed in court at all. That said, in Hinton, the trial lawyer knew he had a sub-par forensic expert on a key issue and mistakenly thought there was no way to do better for his client. In Kulbicki, perhaps the lawyer had no idea that there was any way to challenge the FBI’s scientific-sounding bullet lead evidence. But that begs the question whether a lawyer should be expected to look harder into whether the forensics are actually supported by real research. Is it a good excuse that judges routinely allow a type of forensic analysis that isn’t supported by real research?

In Hinton, the Court emphasized: “[W]e have recognized the threat to fair criminal trials posed by the potential for incompetent or fraudulent prosecution forensics experts.” Nowhere did the Court even pay lip service to that threat in Kulbicki, despite the fact that the case did indeed involve a patently unscientific technique since disavowed and discontinued by the FBI. Instead, the Court emphasized that a 1991 report co-authored by the FBI expert testifying in the case would have been difficult for any defense lawyer to even find, and that it did not clearly signal awareness at the FBI that the technique was flawed. Is the unfortunate moral of the story that if a crime lab hides the flaws in the “science" for years, it can cut off access to post-conviction relief, whose success so often hinges on ineffective assistance of counsel claims?

Should the trial lawyer be the one to “bite the bullet,” as it were? Perhaps ineffective assistance of counsel claims are not the best way to remedy unscientific evidence in the courts. It is an unfortunate failing of our habeas law that a convict’s best hope may be to try to blame the defense lawyer for the failings of prosecution experts. Should courts instead consider more carefully a Brady v. Maryland theory that the unsound underpinnings of this evidence were concealed from the defense? Or a theory that the evidence was fabricated in violation of Napue? Should state courts be empowered to grant relief because the changed understanding of the science, or lack thereof, constitutes new evidence of innocence? Perhaps more states need to follow the lead of Texas and adopt statutes that permit post-conviction relief for changed understanding of scientific evidence. False facts and unsound science may be most conveniently hung on the ready post-conviction hook of a Strickland claim. But when forensic errors come to light, accountability should be placed squarely on the prosecutors and the forensic analysts who were the purveyors of flawed science. In Hinton, the Court recognized, “[p]rosecution experts, of course, can sometimes make mistakes.” When they do, the prosecution and those very experts should face the post-conviction consequences – and bite the bullet.