Tue03282017

Last updateThu, 10 Nov 2016 6pm

Back You are here: Home Featured Journalists Bill Pursell The Not-So-Reliable Forensic Science

The Not-So-Reliable Forensic Science

When they are pulling out forensic science to prove a defendant guilty on CSI, we all sit back and sigh in relief. Finally, as long as the evidence is gathered and disclosed in full accordance of the law, then the infallible crime lab evidence can put the true culprits behind bars, right? 

There are many variables that need to be controlled in the court room, but even forensic science, the courtroom staple, is in ample need of reform. On July 10th it was announced that the Justice Department and the FBI, with the help of the Innocence Project and the National Association of Criminal Defense Lawyers, will review cases reaching back to at least 1985. This investigation is focusing on cases that utilized hair and fiber follicle evidence. It is unclear as to what the determining factors for cases are in this joint review, but the occurrence of a review is a positive step. These steps however, must not stop here.

Michael R. Bromwich, a former Justice Department official said, “These recent developments remind us of the profound questions about the validity of many forensic techniques that have been used over the course of many decades and underscore the need for continuing attention at every level to ensuring the scientific validity and accuracy of the forensic science that is used every day in our criminal justice system.”

In fact studies show that forensic lab analyses of body fluid, fiber, and paint, could have error rates of at least 10 percent. Fingerprint inaccuracy rates could be up to 4 percent, and bite mark evidence is described as being “notoriously unreliable though still widely used.” Often regarded as the surest form of forensic evidence, DNA has even come under scrutiny by the Wisconsin Association of Criminal Defense Lawyers in 2010.

Beside the questions on scientific validity of these crime labs, are the questions of these lab’s allegiance. The majority of forensic labs are administered by law enforcement agencies (like police departments). Normally the lab administrator reports to the head of the respective law enforcement agency. Forensic lab budgets are, and their biases very reasonably seem to be, tied to their administrating agencies.  It is tough to separate out an agency’s lack of funding and oversight, various errors, and other factors from bias, but could you imagine the working conditions if your scientific findings continually disproved your coworkers? Now allow yourself to imagine a police chief who has to okay the budget for a crime lab along with the payroll and various benefits of police officers. It seems that both the budget and bias of crime labs would benefit greatly from a move towards independence.

In 2008, a Detroit police crime lab was shut down after an outside audit ascertained that about ten percent of the evidence they used to prosecute cases, including murder cases, was false or erroneous. This crime lab was said to have “subpar” quality control and a “shocking level of incompetence.” This Detroit case is rare. Generally there is very little done to check the certification requirements of practitioners, a failure of labs to adhere to proven performance standards, and a shortage of adequate training and continued education for practitioners. There is a lack of research done into technology and innovations within the field too.

Forensic Science groups began collaborating in the early 90’s to form Scientific Working Groups (SWG). These groups were meant to set standards and improve discipline practices. In essence, they have become the main power checking the forensic science community. These SWGs have made good strides for the community, but are nearly powerless. According to a paper entitled, Solving the Problems That Plague the Forensic Science Community, SWGs do not meet regularly, or have a regular source of funding. Recommendations of SWGs are not mandated or enforceable by any laws. Many guidelines by SWGs are too vague to be helpful, and SWGs do not attempt to measure the impact of their standards by formal study or survey. 

Another inherent problem with forensic science testimony is the terminology and lack of necessary information presented in courtrooms. To begin with, commonly used phrases such as, ‘match,’ ‘consistent with,’ ‘identical,’ ‘similar in all respects tested,’ and ‘cannot be excluded as the source of,” are not words that the forensic science community has given qualified definitions. In other words, forensic scientists may have varying degrees of confidence and reliability in their research but will use these same phrases nonetheless. It is important to know exactly what degree of confidence or how reliable a forensic scientist’s results are, because these conclusions have profound effects on the outcome of cases. The meaning each forensic scientist attaches to common courtroom forensic phrases generally has subtle but significant differences. Here is an example to show how far we are from definitive and hierarchical forensic answers: one court acknowledged that an FBI fingerprint expert “testified that the error rate for fingerprint comparison is essentially zero.” As shown before, this is entirely untrue, but this statement has been used as approved precedent in later cases.

It may be excessive to ask for a full scientific report for every trial case, but when an establishment is dealing with people’s lives it seems appropriate to get some sort of statement detailing methods, materials, procedures, results, conclusions, and the sources of uncertainty in the procedures and conclusions.

Our legal system does not scrutinize the science behind forensic discoveries or the language used to express it. Why do we not question the validity of the ever shifting and evolving field of science when it’s used in the courtroom? Why is the legal system not involved in discovering the potential for inaccuracies or mistakes in the forensic process?  If we want to rely on forensic evidence in court cases, then doesn't it makes sense to have a crime lab that carries no biases, or inaccuracies caused by lack of oversight? Wouldn’t it be nice to eventually have the same confidence we use to have in this system?