Virginia DFS to release post-conviction reports

Reports on 78 convicted people whose DNA was excluded in Virginia’s post-conviction testing project — and whose identities have largely been kept secret by the state — will be released under the Freedom of Information Act after July 1, according to the Richmond-Times Dispatch and the Mid-Atlantic Innocence Project.

For unknown reasons, DFS has been resisting these documents’ release. However, the state budget now features an amendment that orders it.

The fact that such measures were required to release the documents implies that they will come out smelling like fish, not to mention that the department that exists to ensure the truest justice was doing the resisting. While DNA testing has mostly been reversing misidentification cases, it is also proving errors in some past and current forensic processes.

The fear of being discredited continues to stifle the truth in this and other cases. An understanding about forensics scientists needs to exist: they can make mistakes, but even more so that older determinants that they were taught were legitimate, are not necessarily.

More on the dependability of forensics here

MSNBC’s “Rock Center” reports on misidentification issues

In a mock identification test conducted on the web page of the MSNBC show “Rock Center,” 46 percent of the test takers identified the wrong suspect. These are people choosing a face out of six total suspects after seeing a very brief surveillance video, in the comfort of their homes, stress free. Even without outside factors, almost half would have sent the wrong person to trial.

The report also shows how Dallas, Texas has formed a model which they hope to be much more effective. When someone looks at a page of multiple faces they compare the faces to each other, not to the one they have in their memory. So, Dallas utilizes a photo administer, to avoid bias, who shows the photos to the victim or witness one by one. The report shows actual footage of this process, during which a woman thoroughly observes a few pictures then, when it appears, immediately slams on the photo of what seems to be the perpetrator and breaks down crying.

While it certainly proves a better effort than the truly faulty one-page photo lineup, the identification test that 46 percent of home viewers failed was administered in the same way. Photos were shown one by one, with the ability to toggle back and forth between each. Although this was unscientifically monitored test, its results should still offer fair warning to the trustworthiness of Dallas’ model.

See the video, article, and take the test here.