Junk Science and the American Criminal Justice System [Paperback]

No one in America will ever know the number of innocent people convicted, sent to prison, and even executed because of the flood of rotten forensics and bogus scientific opinions presented to juries. In this intriguing and beautifully crafted book, Innocence Project lawyer M. Chris Fabricant illustrates how wrongful convictions occur, and he makes it obvious how they could be prevented. An attorney uses the cases of clients who were wrongly convicted and imprisoned to question the status of forensics as reliable evidence, arguing that it is subjective speculation masquerading as science, typically tilted in the governments favor against an indigent person of color. Readersespecially those fond of TV detectives and their infallible crime labswill be flabbergasted by [Fabricant's] list of forensic techniques long used by labs, including the FBI's, and proclaimed by highly paid 'expert witnesses' that, when investigated by competent researchers, turn out to be unreliable or worthless . . . A brilliant rebuttal of junk science in the courtroom. This? M. Chris Fabricant's ground-breaking new work is here to explain (and explain well) what happened, where we went wrong, and how many have paid the price of junk science convictions. Fabricant's long involv[ment] with the Innocence Project lends his book credence and weight, as he details overturned convictions, shoddy research, and the vast egos that underpinned some of the worst work in the history of criminal (in)justice. Spellbinding . . . I believed in the polygraph test, in an unthinking way, right up until last week when I read a new book by M. Chris Fabricant,? Few people are more qualified to write about wrongful convictions in the US than M. Chris Fabricant . . . Fabricant's book shows how faulty ideas from blood spatter analysis to shaken baby syndrome were developed, infected court systems, and ruined a still-untold number of lives. Fabricant expertly delves into now-discredited forensic tactics, includl#-