America's forensics system, the part of our criminal justice system responsible for scientific examinations of crime-scene evidence like fingerprints and DNA, is rife with errors. Some mistakes, like botched tests or erroneously interpreted results, are inevitable. But current error rates are needlessly high. The most recent comprehensive study of crime lab proficiency, published by the Journal of Forensic Sciences in 1995, analyzed the tests administered by the Forensic Sciences Foundation and Collaborative Testing Services as a part of the accreditation process. For many forensic disciplines, including the analysis of fibers, paints, glass, and body fluid mixtures, the rate of incorrect matches between recovered evidence and a reference sample exceeded 10 percent.
The best-performing group of disciplines, which included "finger and palm prints, metals, firearms, and footwear," had error rates at or above 2 percent. The first item on that list is especially important: False fingerprint identification usually leads to a false conviction, because of the prestige of fingerprint evidence and its undeserved reputation for infallibility. With 238,135 requests for latent fingerprint comparisons in 2002 alone, a false positive error rate of 2 percent implies up to 4,800 false convictions or guilty pleas made in hopes of a lighter sentence each year in the U.S., 1,700 of them in felony cases. (The number of improperly matched fingerprints is not completely clear. A 2005 study of fingerprint analysis suggests that the false positive rate may now be as low as 0.8 percent. But another recent study suggests it could exceed 4 percent.)
Confronted with such statistics, policy makers usually call for greater oversight—that is, finding a governmental body to watch over forensics and make sure everyone does his or her job right. In the current climate, that certainly would help. But the core problem with modern forensics isn't an absence of oversight. It's monopoly. Once evidence goes to a given lab or facility, it is unlikely to be examined by any other lab or facility. That increases the chances that a mistake will slip through undetected.
With the right reforms, we can break down that monopoly and create a working system of checks and balances. Here are eight steps to a better system:
1. Rivalrous redundancy. A jurisdiction should contain several competing forensic labs. To the extent that it's feasible, some evidence should be chosen at random for multiple testing at other labs. The same DNA evidence, for example, might be sent to three labs for analysis. The forensic scientist would not know when a given piece of evidence was being examined by another lab.
For fingerprints, multiple examinations should be routine. If the rate of false positive fingerprint error is 2 percent, triplicate examinations would eliminate 96 percent of false felony convictions due to misidentified fingerprints.
2. Independence. Coroners and forensic scientists often have a pro-police bias, thinking of themselves as a part of the prosecution team. To establish their independence from police and prosecutors, crime labs should be organized by the courts, not the cops.
3. Statistical review. Statistical review would lead to improved quality control. For example, if a given lab produces an unusually large number of inconclusive findings, its procedures and practices should be examined by an officer of the court.
4. Masking. When conducting forensic analyses, coroners and forensic scientists should be shielded from what psychologists call "domain-irrelevant information." Knowing whether the case at hand is, for example, a murder or a burglary exposes a fingerprint examiner to a powerful unconscious bias: The emotional nature of a murder case tends to make the scientist eager to get a killer off the streets and more likely to declare a match. In a 2006 study by researchers at the University of Southampton, domain-irrelevant information doubled the error rate of experienced fingerprint examiners.
5. Forensic counsel for the indigent. Although forensic science decides many criminal cases, we do not have a right to forensic counsel similar to our right to legal counsel. Just as an indigent defendant has a right to the help of a qualified attorney, an indigent defendant should have the right to the help of a scientist qualified to interpret forensic analyses.
6. Forensic vouchers. An indigent suspect on trial should also have the right to select his own forensic counsel and use a government-issued voucher to pay for it. The forensic scientist who accepts the case would later redeem the voucher at the courthouse, receiving his paycheck from an officer of the court. Such a system would give forensic counselors to the poor an incentive to provide high-quality services.
7. Division of labor between forensic analysis and interpretation. A forensic scientist who conducts a blood test, for example, should not say whether the test excludes the suspect. The interpretation of the test should be made by other forensic scientists. When a lab report comes back, it should be transmitted to two forensic scientists—one representing the prosecution and one representing the defense—for interpretation. Combined with public funding of forensic experts for defendants who cannot afford them, this will make it less likely that errors of interpretation will go unchallenged.
8. Privatization. Private labs are subject to civil liability and administrative fines for poor performance. They therefore have stronger financial incentives than publicly owned enterprises to provide good and reliable work.
While those eight reforms would establish a system of competitive self-regulation, they are not an exhaustive description of good forensic practice. For example, labs should be accredited by groups such as the American Society of Crime Laboratory Directors' Laboratory Accreditation Board, and they should have routine procedures to measure variables related to quality and a planned system of review to allow those procedures to be updated regularly.
These eight steps would also reduce the costs of the criminal justice system. The extra cost of multiple forensic tests is dwarfed by the savings associated with reduced jail time for the wrongly convicted. For example, the $100 cost of a fingerprint examination is one one-thousandth of the cost of incarcerating a wrongly convicted felon who has been given the average sentence of almost five years.
In Federalist No. 51, James Madison endorsed a "policy of supplying, by opposite and rival interests, the defect of better motives," in order "to divide and arrange the several offices in such a manner as that each may be a check on the other; that the private interest of every individual may be a sentinel over the public rights." It is time to make the private interest of every coroner, every medical examiner, and every forensic scientist a sentinel over the public rights. r
Roger Koppl, a professor of economics and finance at Fairleigh Dickinson University, is the founder and director of the university's Institute for Forensic Science Administration (IFSA) and author of "CSI for Real: How to Improve Forensic Science," a forthcoming study from the Reason Foundation. IFSA's website, alpha.fdu.edu/~koppl/ifsa.html, contains more information on competitive self-regulation.