Psychological research continues to demonstrate that evidence such as eyewitness identifications or confessions do not necessarily prove guilt; eyewitnesses can misremember events, lineup processes can encourage witnesses to misidentify suspects, and confessions can be coerced. The unreliability of these types of evidence stems from psychological and cognitive processes such as tunnel vision, (in which a criminal justice official becomes so entrenched in the idea that the suspect is guilty that he or she starts to reinterpret information to support this idea,) and the malleability of human memory, (which causes witnesses to remember a crime differently from how it really occurred, and allows suspects to be convinced that they are guilty of a crime that they did not commit cite). Psychology has proved time and time again that not only do humans often incorrectly remember and interpret events, but also that these psychological flaws often contributes to wrongful convictions.
However, we also have to accept the fact that hard scientific evidence is also not always reliable. Over the years, countless types of forensic evidence have been found to be less dependable than they were once thought to be. For example, in 2002, the National Research Council conducted a study the FBI’s use of bullet lead analysis. Bullet lead analysis was performed to find an association between a bullet used for a particular crime, and bullets associated with suspects for that crime. The NRC found that despite the FBI’s equipment and procedures being up to snuff, their interpretations of analysis results were questionable; the FBI overstated the significance of their results to jurors, making the analysis results seem incriminating to a misleading degree. Following the report, the FBI chose to discontinue these tests, but did not thoroughly notify all the defendants who were affected by this misleading testimony, thus limiting their chances to appeal their convictions.
An even more contemporary example from the FBI is the April 2015 announcement that a large majority of FBI cases that included microscopic hair analysis also overstated the results as more incriminating than they actually are. This is especially relevant here at the Innocence Project of Florida due to the high percentage of Floridian defendants who had been convicted with the help of these analyses. Just because these analyses were misrepresented in court does not necessarily prove the defendants’ innocence, but it does speak to the fragile nature of forensic evidence.
Besides flaws in the significance assigned to results of forensic testing, there are also some fields of forensic science that were previously so plagued with errors that they have had to have been almost completely rewritten. The most noteworthy example is that of arson science. Cases from Cameron Todd Willingham, to David Lee Gavitt, to the very current case of Letitia Smallwood have all been questioned due to the flawed arson science used in their convictions. Arson science from previous decades was actually considered more of an “art”. Rather than experimenting and training, older arson science was based on certain types of burn marks that were thought to indicate arson. However, more recent science, in which researchers set actual fires and observe them, reveals that these same marks appear in accidental fires as well. Arson science reminds us that just because a certain forensic test is considered “science” doesn’t necessarily make it truly “scientific”.
This brief overview of different types of evidence is not meant to demonstrate the hopelessness of preventing wrongful convictions or of using evidence to find the real perpetrators of crimes. Instead, it is a reminder that the resources the criminal justice system has at its disposal to prove someone guilty, from burn marks to confessions, need to be handled with care.
For more information on the fallibility of forensic science, take a look at this article.