Widely used methods to trace complex DNA samples, bullets, tread and bite marks to criminal defendants fall short of scientific standards, limitations that federal prosecutors and judges should seriously consider before entering forensic evidence in trials, reports the Washington Post. The unanimous report by the 20-member President’s Council of Advisors on Science and Technology does not advocate banning testimony or putting limits on tools used for investigations. But when it comes to evidence at trial, the report says, the Justice Department and federal judges should give greater weight to scientists’ view of forensic evidence, particularly in light of a landmark 1993 Supreme Court decision that ruled courts should act as gatekeepers and admit only “scientifically valid” expert testimony.
“For a forensic science to be scientifically valid, you need actual, empirical evidence of its reliability and accuracy, period,” said Eric S. Lander, founder of the Broad Institute of Harvard and MIT. “Historically that hasn’t been the case.” Lander and White House science adviser John P. Holdren co-chaired the council. The report found that techniques that link a person to a crime scene through bite marks or treads from shoes or tires fell “far short” of scientific standards and lacked “meaningful evidence” of their accuracy. The report said evidence supporting firearms tracing was promising but needed more corroboration.