Science in the courtroom is important. Many elements in many criminal cases are dependent on lab tests to confirm a white powder is really cocaine or that some other substance is an illegal Schedule 1 drug. It is necessary when testing the blood of someone accused of driving while intoxicated and to determine if a bullet slug matches a particular gun used during the commission of a crime.
But it has to be done properly and it must have actual science behind the findings. In addition, if it is necessary to have an expert testify at trial, they too must actually be qualified to provide "expert" testimony, with adequate training tied to genuine scientific standards.
One would expect that the FBI, the nation's top crime investigation organization, would have the best-of-the-best process and procedures and that their crime lab would provide unimpeachable evidence analysis and testimony when called at trial.
For decades, the FBI has provided microscopic analysis of hair samples found at crime scenes. Where there is no reliable eyewitness testimony as to who was present, hair identification could be used to tie a suspect to a crime scene.
Just one problem. There were no "real" scientific standards for this type of analysis, and the vast majority of FBI lab technicians vastly overstated the certainty of their identifications.
Doubts have been cast on this type of evidence for years, but prosecutors kept presenting it, judges continued to allow it and juries continued to believe it.
The problem is not limited to the FBI, as their lab apparently trained between 500 and 1,000 state lab technicians, which mean literally thousands of cases may have used tainted or worthless evidence to convict individuals over the last 40 years.
In addition, in many of those cases, the FBI and prosecutors failed to inform those convicted of the problem with their cases. Some of the cases are so old, those convicted have died in prison. A few were executed.
Slate.com, "Pseudoscience in the Witness Box," Dahlia Lithwick, April 22, 2015