|
For more than a century, fingerprint analysis has been accepted,
first in British and then in American courts, as evidence to help convict
criminal defendants. No forensic technique is better known nor — until the
advent of DNA profiling — was more widely accepted.
But fingerprint identification is not infallible. Just ask
Brandon Mayfield, the Oregon civil rights lawyer who was wrongly implicated in
the March 2004 terrorist train bombing in Madrid, Spain, on the basis of a
fingerprint misidentification by two FBI examiners.
The FBI entered the case after Spanish police asked for help
identifying a “latent print” — an accidental impression typically invisible to
the naked eye — found on a piece of evidence in the investigation. Using the
FBI's database containing millions of fingerprint records, an examiner
identified the print as Mayfield's. Informed of the identification, a second
examiner agreed.
Mayfield, a Muslim who had once represented a convicted terrorist
in a child-custody dispute, was placed under surveillance and in May arrested
as a material witness. By mid-April, however, Spanish police had concluded that
the print was not Mayfield's; on May 19 they positively identified the
fingerprint as that of an Algerian national.
Mayfield was released and given an apology and later $2 million
in compensation. Meanwhile, the Justice Department's inspector general in March
2006 concluded that the examiners had failed to follow standard rules in their
analysis. The misidentification could have been avoided, the inspector
general's report concluded, by “a more rigorous application of several
principles of latent fingerprint identification.”
At first telling, the episode may seem to contradict the accepted
wisdom that an individual's fingerprints are unique. But the case actually
teaches a different lesson, according to Simon Cole, an associate professor of
criminology, law and society at the University of California-Irvine and author
of a book on fingerprinting, Suspect
Identities.
“The uniqueness of fingerprints doesn't really tell you anything
about the accuracy of fingerprint identification,” Cole explains. “You and I
don't need to have identical fingerprints. All we need is to have fingerprint
patterns similar enough so that a fingerprint left by me might be attributed to
you.”
In short, Cole says, misidentifications are inevitable, but their
frequency is unknown. “We really know almost nothing,” he says. When mistakes
come to light, the disclosures are often fortuitous, Cole adds. In Mayfield's
case, for example, the misidentification would never have become known if the
FBI had bowed to the doubts raised by the Spanish police and refrained from
arresting him before the real suspect was found.
Despite wide agreement among experts on the risk of error, FBI
examiners have routinely testified in court that fingerprint identification is
always accurate. Cole says he appeared as a defense witness in a federal court
hearing in Pennsylvania in 1999, where two of the FBI's leading examiners
helped persuade a judge to reject a defense motion to limit fingerprint
testimony.
Three years later, a different federal judge in Pennsylvania
stunned the legal and forensic science communities in January 2002 with a
ruling in a major murder case to prevent fingerprint examiners from declaring a
latent print to be a “match” of a defendant's print. Two months later, however,
Judge Louis Pollak reversed himself. “In short, I have changed my mind,” Pollak
wrote in the 59-page opinion.
Pollak's short-lived decision stood as the only ruling to limit
fingerprint testimony in the United States until October 2007, when a Baltimore
County, Md., judge barred fingerprint testimony altogether in a state murder
case. Citing the Mayfield case, Judge Susan Souder said the prosecution had
failed to prove that fingerprinting is an infallible methodology. Under
Maryland rules, the pretrial ruling could not be appealed. Prosecutors
responded by getting the U.S. attorney's office to indict the defendant on
federal charges. The case is pending.
Judge Louis Pollak blocked fingerprint examiners from
declaring a latent print a “match” of a defendant's print in a murder case. He
later reversed himself. (Federal Judicial Center)
|
Law enforcement groups defend fingerprinting. “It's certainly a
valid process,” says Dean Gialamas, forensic services chief for the Orange
County (Calif.) Sheriff-Coroner's Office.
Scott Burns, executive director of the National District
Attorneys' Association, calls fingerprint identification “the gold standard” of
forensic techniques, but adds, “There are always going to be some
misidentifications.”
The National Research Council report on forensic sciences
published in February 2009 noted critically that courts have accepted
fingerprint evidence “without the scientific scrutiny” that was given to DNA
profiling before its acceptance. The report calls for more research but makes
no recommendation about admitting or excluding fingerprint evidence until
then.
Cole is not calling for excluding fingerprint evidence, but he
says courts have long allowed examiners to give “scientifically unsupportable” testimony that a latent print matches the defendant's. “That individualization
testimony is too strong,” he says. “That should not be permitted.”
— Kenneth Jost
|
Read the Full Report (Subscription Required)
