« October 1, 2006 - October 7, 2006 | Main | October 15, 2006 - October 21, 2006 »

October 14, 2006

Science in the Bush-league

The Bush administration has not demonstrated a high degree of understanding of basic scientific principles or knowledge of the bases of science in their time in office.  From global warming to intelligent design, the members of the Bush-leagues seem to be existing some time circa 1600.  Today's New York Times has an editorial criticizing the EPA's rejection of new standards for soot emissions.  According to Stephen Johnson, EPA's administrator, the evidence remains "insufficient" to support new standards, despite the overwhelming recommendations of EPA's own science advisors and scientists.  It sometimes appears with the Bush administration that the only science that would be good enough to convince them that they should do what they were disinclined to do for policy reasons would be scientific findings that rode in on a golden chariot from the heavens.  The Bush-leaguers need a lot more science, and a little less policy in their science-policy.  See Editorial Here.

October 14, 2006 | Permalink | Comments (0) | TrackBack

October 13, 2006

Forensics Fired

I watched the National Geographic special last night, "Naked Science: Forensics Under Fire."  It was, as I anticipated, well worth watching.  They concentrated on arson and firearms identification (both ballistics and bullet lead comparisons).  The show was extremely critical of forensic science as it is practiced today, which is not surprising, since there is so much to be critical about.  It was only an hour long, however, so it is understandable that the producers had to focus on a couple of subjects, though they cast their aspersions widely.  My only quibble was the show's coverage of a National Institute of Justice study that was commissioned to prove that firearms identification experts can do what they claim they do.  As an intial matter, of course, research should not be designed to prove that something is true, but should be carried out to test whether something is so or not.  More problematic, however, the research involved computer evaluations of whether bullets fired from the same gun are distinguishable from bullets fired from different guns.  The researchers found that, indeed, this was the case, and concluded that thus firearms experts could validly determine whether a bullet came from a particular gun.  There is so much wrong with using this methodology to support the asserted claim that I wouldn't know where to start.  But surely, whatever this research might demonstrate, it is not that firearms examiners can validly "match" an unknown bullet to a known gun.

In the end, however, this program offers an excellent introduction to the problems endemic in forensic science.  The program's ultimate conclusion is, without question, worthy of attention.  Forensic scientists cannot rely on years of experience to support their claims;  their hypotheses must be subjected to rigorous scientific test.
--- DLF

October 13, 2006 | Permalink | Comments (2) | TrackBack

October 12, 2006

Forensics Under Fire

I wanted to move this to the front, since the show airs this evening.

Forensic science is the subject of a national geographic production to be aired on October 12th.  The notice follows.

We wanted to let you know that "Naked Science: Forensics Under Fire" is scheduled to air on the National Geographic Channel on Thursday, October 12 at 10:00 p.m. ET/9:00 p.m. CT. Produced by Jon Siskel and Greg Jacobs for Towers Productions, the show takes a close look at recent challenges to the reliability of forensic science, and how the field is responding to those challenges.

It should be well worth watching.
--- DLF

October 12, 2006 | Permalink | Comments (0) | TrackBack