Cross-Examining Forensics

The Supreme Court says that forensics analysts can be cross-examined...for now.


On June 25, in a late-term, 5-4 ruling marked by unconventional alliances and sharp disagreement between the justices, the U.S. Supreme Court decided in Melendez-Diaz v. Massachusetts that the Sixth Amendment's Confrontation Clause requires forensic experts whose reports are admitted into evidence to be made available for cross-examination. You might have already thought that you had the right to challenge a witness who offers powerful evidence of your guilt. But until six weeks ago, that wasn't the case in many states.
Defense attorneys and critics of the forensics system were celebratory, but they might want to check next term's docket. Just before it recessed, the Court agreed to hear a case from Virginia with very similar issues at stake. The unusual move has some Court watchers speculating that the minority in Melendez-Diaz may see former prosecutor and new Associate Justice Sonia Sotomayor as an ally. Her vote could give them a majority for a reversal or a significant limitation of last term's ruling.
Melendez-Diaz v. Massachusetts allows defense attorneys to question the authors of forensics reports about their methodology and to probe those authors' testimony for possible errors. A faceless analysis that cites a 99 percent or higher probability of a forensics match can lose some of its punch if the author can be questioned in front of a jury about the possibility of bias or human error.
The basic problem with courtroom forensics is that there's too much bias and prejudice in the analysis for it to be classified as science. Peer review, for example, forms the very foundation of scientific inquiry, but it's mostly absent from forensic analysis. Scientists go to great lengths to insulate themselves from bias, such as conducting double-blind studies. Forensic analysts, on the other hand, routinely meet with the attorneys for whom they'll be testifying before conducting their examination. One 2006 study by researchers at Britain's University of Southampton found that fingerprint analysts were twice as likely to find false matches when they were given extraneous information about the case. In a recent cover story on forensics, Popular Mechanics summed up the problem: "Forensic science…was not developed by scientists. It was created by cops—often guided by little more than common sense—looking for reliable ways to match patterns from clues with evidence tied to suspects. What research has been done understandably focuses on finding new techniques for putting criminals in jail." In other words, where science is about process, forensics tends to be more concerned with outcomes.

Yet in courtrooms forensic evidence is usually presented as hard science, giving it a false authority that can have a powerful influence on jurors (particularly those who watch the various CSI series on television, where the well-funded, high-tech labs always identify the killer by the end of the hour). A congressionally commissioned report published earlier this year by the National Academy of Sciences found that across all forensic specialities, from medical examination to fingerprinting to hair and fiber analysis, not only does the methodology often lack scientific rigor, examiners commonly exaggerate the certainty of their findings on the witness stand, or testify to findings that have no basis in science at all. The Innocence Project estimates that half of all wrongful convictions are at least partly due to faulty forensic science.

The Melendez-Diaz decision recognized these problems. "A forensic analyst responding to a request from law enforcement may feel pressure—or have an incentive—to alter evidence in a manner favorable to the prosecution," Justice Antonin Scalia wrote in the majority opinion, adding, "Confrontation is designed to weed out not only the fraudulent analyst, but the incompetent one as well."

The decision didn't mandate that states change the underlying systemic problems with forensics that give rise to bias and improper incentives in the first place. It merely gave defendants the opportunity to raise these issues in court. But even that modest reform has the old guard up in arms. Last month, the Washington Post reported that all across the country, Melendez-Diaz "has prosecutors and judges shaking their heads in disgust," issuing dire warnings that "murderers could walk free" and "drunken driving cases could be dismissed." Lab workers just don't have time to traipse off to court to defend their results, the officials complain.

That's too bad. But the Bill of Rights protects us from government overreach. To say we should suspend constitutional protections because keeping them in place would prove inconvenient to the government rather misses the point. If there aren't enough analysts to both work the lab and testify in court, states will either have to budget more money for forensic analysis, or prosecutors will have to start prioritizing cases with the budgets they have. If the criminal justice system has adapted to new technology in a way that doesn't pass constitutional muster, it simply needs to change.
For example, the Post notes that an amicus brief written for the Melendez-Diaz case by a group of state attorneys general explained that the vast majority of crime-lab work involves testing for drug cases, and warned that a decision mandating confrontation could mean that more of those cases come to trial. "Even if only 5 percent of drug cases culminate in trials, the burden on the states is oppressive," the brief warned. Perhaps, then, prosecutors should stop devoting so many resources to consensual crimes. That would free up analysts to work on crimes that have actual victims.
If the Court significantly narrows the scope of Melendez-Diaz next term, the complaining judges and prosecutors may yet get their way. That's good news if you're merely concerned with efficiency. It's less reason to celebrate if your concern lies with the Constitution, or with ensuring a fair trial.
Radley Balko is a senior editor at Reason magazine.