Noting a Lack of Research to Establish the Reliability of Many Forensic Practices, President’s Council of Advisors on Science and Technology Urges Greater Funding for Federal Science-Based Agencies to Conduct Research and to Evaluate Validity of Current Methods Recommends that Attorney General and Courts Reject Forensic Disciplines that Do Not Consistently Produce Accurate Results
(Washington D.C. – September 20, 2016) The Innocence Project applauds the President’s Council of Advisors on Science and Technology (PCAST) for its landmark report released today evaluating forensic disciplines and calling for more scientific research to establish the validity of many methods. The report urges the National Institute of Standards and Technology (NIST) to evaluate the foundational validity of all current and future forensic disciplines and recommends essential research to improve forensic methods so that they consistently produce accurate results. PCAST also expressed concern about the accuracy of testimony before criminal courts and emphasized that the attorney general must be careful to limit testimony and reports to those which have been demonstrated to be scientifically valid. Finally they urged judges to exercise their gate-keeping function and consider the absence of relevant empirical studies when deciding the admissibility of expert testimony.
“The report does a masterful job of explaining the need for more scientific research for the various forensic disciplines that are routinely used in identifying suspects. With the help of the forensic science community, this report makes very specific recommendations about the research that is needed and how existing science-based agencies can take a leadership role in that research. Strengthening forensic science disciplines through research will guarantee that these practices are based on the best science and meet the constitutional and scientific demands of the criminal justice system,” said Peter Neufeld, co-director of the Innocence Project, which is affiliated with Cardozo School of Law. “Now that the nation’s leading scientists have provided a blueprint for fixing one of the most critical problems plaguing the criminal justice system, we urge Congress, the attorney general and the courts to adopt its recommendations to prevent more innocent people from being wrongly convicted.”
The report, Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods, was prepared by PCAST, which includes some of the nation’s leading scientists and engineers. The working group that prepared the report focused on forensic practices that are typically used to identify suspects based on evidence found at crime scenes, including analysis of DNA, latent fingerprints, firearms and spent ammunition, shoe prints and bite marks. The working group conducted a comprehensive review of the existing research on these disciplines and assessed whether the discipline was capable of consistently producing an accurate result, and if so, whether the use of the discipline in a particular case meets “scientific standards reliably applying the discipline’s principles and methods.”
Significantly, the report found that bite mark analysis is unable to provide consistent and accurate results and is unlikely – even with additional research – to be successfully developed into a scientifically valid method. The field of latent fingerprint analysis, the report noted, has been advanced by recent research initiated and conducted by the forensic science community, but could be strengthened by additional studies. The report singled out DNA analysis of single-source and simple-mixture samples as an excellent example of the proper establishment of scientific validity. Existing research on DNA analysis of complex-mixture samples is promising, it noted, but additional research is needed. The report also recommended additional research on the analysis of firearms, noting there has only been one relevant study on its accuracy and reliability. For each of the disciplines reviewed, the authors detailed what additional research is needed and noted what limitations should be acknowledged if a method passes the admissibility test and used in court.
“Typically, we’ve had a situation where two scientifically untrained lawyers have argued the bonafides of scientific evidence before a scientifically untrained judge so that 12 scientifically untrained jurors could decide the weight of that evidence,” added Neufeld. “This PCAST report will replace that with the tools that will allow judges to be more rigorous in the way they evaluate scientific evidence.”
The following are some of the recommendations for strengthening forensic disciplines and for addressing the mismatch between the current state of knowledge and how forensic science disciplines are used in criminal investigations and prosecutions:
- The National Institute of Standards and Technology should assume the central and critical role of assessing the validity of existing and emerging forensic methods, and develop objective methods of analysis and evaluation for disciplines that are currently based on subjective methods. (DNA analysis of complex mixtures, latent fingerprint analysis and firearm analysis were specifically noted.)
- The attorney general should ensure that expert testimony used by the Department of Justice meets standards of scientific validity and not introduce testimony where it has not been demonstrated that the methods are capable of consistently producing accurate results.
- Federal judges, in their critical roles as the “gatekeepers” charged with ensuring the reliability of expert testimony should look to the scientific evaluation of foundational validity and applied validity in deciding the admissibility of analyst testimony.
- The FBI Laboratory should expand its forensic science research program.
The report follows a 2009 report by the National Academy of Sciences (NAS) that first brought attention to the lack of scientific research in forensic disciplines. That report prompted the Department of Justice, in collaboration with the National Institute of Standards and Technology, to create the first National Commission on Forensic Science, which is made up of scientists, judges, attorneys and forensic science leaders, to make recommendations for strengthening forensic disciplines. The PCAST report expands upon the 2009 NAS report with an updated and comprehensive evaluation of the existing research for seven forensic practices and makes recommendations for the type of research needed to validate the disciplines and ensure that they are reliable in practice. The recommendations employ existing federal science-based agencies to conduct the research rather than requiring a new entity as envisioned by the NAS report. The working group that produced the report consulted with a group of senior advisors which included two of the nation’s leading statisticians with expertise in forensic science, nine current or former federal judges and a former U.S. Solicitor General. The report also recommended several practices that could improve the overall quality of laboratories and prevent the influence of various types of biases, including routine, blinded proficiency testing and limiting the examiner’s access to potentially biasing information.
The authors acknowledged the role that the nation’s DNA exonerations played in bringing to light the lack of scientific validity in all other forensic disciplines. According to Innocence Project research, the misapplication of forensic science has contributed to nearly half of the nation’s 344 DNA exonerations.
A copy of the report is available here.
Co-Founder and C0-Director Peter Neufeld’s response to the report: