“Decades of overstatement of latent print examiners” has led to unrealistically high expectations of judges and juries when it comes to the forensic science of fingerprinting, according to a new report by the American Association for the Advancement of Science.

Experts should no longer be allowed to use terms like “match,” “identification,” “individualization” or other synonyms that imply certainty, the report concludes.

“These claims were clearly overstated and are now widely recognized as indefensible,” the AAAS report states. “While latent print examiners may well be able to exclude the preponderance of the human population as possible sources of a latent print, there is not scientific basis for estimating the number of people who could not be excluded and, consequently, no scientific basis for determining when the pool of possible sources is limited to a single person.”

Instead of calling something a “match,” the report states, latent examiners should instead provide a statement like:

“The latent print on Exhibit ## and the record fingerprint bearing the name XXXX have a great deal of corresponding ridge detail with no differences that would indicate they were made by different fingers. There is no way to determine how many other people might have a finger with a corresponding set of ridge features, but this degree of similarity is far greater than I have ever seen in non-matched comparisons.”

The four-person working group who authored today’s AAAS report includes one practicing forensic scientist (John Black, formerly of the South Carolina Law Enforcement Division and the Drug Enforcement Administration). The other three are a biometric and computer engineer, a statistician, and a professor of criminology.

(The statistician, Jay Kadane of Carnegie Mellon University, has previously told Forensic Magazine that latent fingerprinting ultimately is circumstantial—but not identification—evidence.)

The science of fingerprinting is a good identifier in matching a clear set of 10 prints to another set of 10 prints, the report concedes.

But the limitations arise when it comes to latents left at a crime scene, according to the AAAS working group.

Such comparisons continue to rely on the expertise and subjective opinion of the examiner, the report adds. Technology needs to advance to the point of being able to point out quantitative measures of how likely the fingerprint is to be someone other than the accused, they add.

Other points from the report:

  • The risk of false exclusions exists with the degree of intra-finger variability, it states.
  • AFIS systems need to be refined for better automated matching. The National Institute of Standards and Technology (NIST) should oversee evaluations of the commercial AFIS systems available to agencies, they add.
  • Research samples should be inserted into latent print examiners’ regular flow of casework—although “errors are to be expected and should be treated as opportunities for learning and improvement,” the report adds.
  • Although “likelihood ratios” like those at use in Europe are beneficial in testimony, the report claims, the current best quantitative characterization models are presented by the Defense Forensic Science Center of the Department of the Army (DFSC), and not the Scientific Working Group on Friction Ridge Analysis, Study and Technology (SWGFAST).

“We agree that there is a need to quantify the strength of latent print evidence so that it is not left as a matter of subjective judgment, however, the existing scientific literature does not, at present, prove an adequate basis for quantitative estimates of the accuracy of the various determinations latent print examiners make on the basis of latent print evidence.”

Forensic Magazine has conducted searches of case histories to find convictions based on erroneous fingerprint matching, but has not yet found an instance of such a problem. Some critics point to the two-week arrest of U.S. citizen Brandon Mayfield for the 2004 Madrid terrorist bombings as an example of the limitations of fingerprint matching (Mayfield was discovered through an error by a computer system, as well as mistakes by FBI examiners. But he was cleared long before prosecutors could build a case against him). 

Fingerprint and biometric experts may take exception to the charaterization by the report. Steve Johnson, a former president of International Association for Identification, said while he and others respected the work of trying to improve forensic science, rigor and high standards already make latent print examinations very reliable.

"We've been doing this since the early 1900s... I think there's merit in what we do," said Johnson, based in West Virginia. "We in the forensic community have a vested interest in getting the right answers in the pursuit of justice."

The AAAS reevaluation of American forensic science comes in response to a scathing federal report issued in 2009 by the National Academy of Sciences. Ten disciplines were schdeduled for subject reports. Arson and now fingerprints have been published by the AAAS - but funding has since made the project's future tenuous, according to a spokeswoman.

Correction: An AAAS spokeswoman told Forensic Magazine after the publication of this story that the agency's forensic science review was on hold due to lack of funding.