E-Mail This Article to a Friend Bad ballistics, continued

In fact, in two cases the Phoenix knows of, homicide detectives asked the ballistics lab to examine evidence for a match even after their digital-imaging identification system — used to narrow down likely matches — had rejected them. And in both cases, the BPD ballisticians declared that the evidence did match.

If the ballisticians are susceptible to confirmation bias, their new commander appears not even to understand it. In the same murder-trial hearing two weeks ago, Doherty said that the prosecutor approached her on Saturday night, less than 48 hours before trial, to ask her to review the cartridge casings that Camper had previously examined, and be prepared to testify about her findings. Asked by the defense attorney whether this was an instance where confirmation bias might be a concern, Doherty repeatedly insisted that she had no idea that finding a match might be helpful to the prosecutor’s case. Her answer makes one wonder if she knew what she was talking about.

CROSSING THE LINE

A Phoenix review of six trial transcripts and testimony at an ongoing trial suggests that BPD ballisticians often seem eager to help the prosecution and stymie the defense. At times they even seem to declare a match on evidence that they know does not justify the claim. That is, when only the "class characteristics" match, but not the "individual characteristics."

To better understand the distinction, envision an assembly machine making thousands of 9mm Ruger handguns. Each gun’s barrel is drilled out by the same steel machine, which adds the same spiral grooves on the inside of each barrel, which spin the bullet similar to the way a quarterback throws a spiral pass.

Every one of those 9mm Rugers coming off that assembly line will have exactly the same grooves, which will mark every bullet shot through them the same way. Similarly, parts of the gun that strike the cartridge casing can leave tell-tale markings.

Those are class characteristics, in ballistics parlance, and they can help determine that two bullets, or two casings, could or could not have been fired by guns of the same make — that is, guns whose grooves were made by the same machines.

But inevitably, each barrel coming off that assembly line will contain random, minute imperfections, such as a microscopic bump of steel. Those imperfections will be quite different from those left on another gun coming through the same assembly line, and will leave a unique microscopic mark — "individual characteristics" — on each bullet that passes through. Individual characteristics indicate that the bullet could only have passed through that particular Ruger; again, similar markings can be found on the cartridge casing that gets ejected from the gun.

To declare a "match" based on class characteristics alone would be wildly irresponsible — like declaring a vehicle the getaway car because it is a Ford Taurus.

Yet in a federal trial this March, Camper testified that there was no doubt in his mind that four recovered cartridge casings were fired from the specific Astra pistol in evidence, "because they all shared the same similarities, class characteristics." And Carl Washington, in a hearing before the Prochilo retrial in 2000, declared that he would, as a rule, declare a match if all he had was a particular class characteristic which he knew not to be sufficient for an individual match. "If that was all that I had, yes, I would call that a match," he said. That transcript has recently surfaced as part of an attempt, in federal court, to get ballistics evidence in one case ruled inadmissible.

Washington’s remarks and Camper’s testimony about the Astra baffle other ballistics experts. "If true, that would break all the rules of being an honest examiner," says Gannalo. "If he tries to make a result sound like more than what it is, that’s an injustice."

"I like most of the people I work with in the Boston Police Department, and they mean very well," says Lamagna. "But if you don’t understand the underlying science behind forensics, it’s much easier to make statements overstating the evidence."

Defense attorneys deserve equal blame, says Lamagna, for not learning enough, or hiring experts, to help them catch and challenge these and other questionable claims. Timothy Perkins, an attorney in Ipswich who is also officially recognized by the state’s public-counsel office as a firearms expert, agrees. "Are they lying? I don’t know. But I know there are some inadvertent errors" in BPD ballistics experts’ testimony, Perkins says. "A lot of it has to do with DAs and judges and defense attorneys not knowing very much about ballistics, and glossing over what seems to be a point that really ought to be more strenuously pushed."

That appears to be changing. Local defense attorneys are planning to challenge ballistics experts in several other upcoming cases, both in state and federal court.

Time will tell whether juries will grow skeptical, now that defense attorneys have begun exposing the ballisticians’ lack of expertise and credentials. The testimony Camper gave this week is critical to whether a man accused of murder goes to prison for life or goes free. Camper says that a cartridge casing found at the murder scene was fired by a gun found in the suspect’s hand two weeks later; it is the only physical evidence linking the defendant to the scene. The jury will decide next week whether they believe him.

David S. Bernstein can be reached at dbernstein[a]phx.com.