Articles Posted in Experts

We previously commented on an issue of voodoo science, and more on voodoo science, and the National Research Counsel report on the future of forensics.  The promoters of this piece of voodoo  — a voice stress analyser — convinced law enforcement to fund and buy their expensive machine and methods for using voice stresses to tell if a suspect was lying.  A number of reputable people published or sought to publish peer reviewed critiques.  Rather than rebut with more peer reviewed research the owners of the company sued.  That's right, rather than conduct the discussion in the research laboratory and through peer reviewed articles and research, they sought to close down critics through lawsuits.  Also, they persuaded the supposedly neutral peer reviewed journal, the International Journal of Speech Language and the Law, to withdraw the critical commentary.  The critique is still available due to the internet — an alert reader captured the article before it was removed from the journal's website.  The company had received government funding.

Well here is the next of likely several more shoes to fall on this supposed reliable technology.  It continues to be put through its own stress tests.

James D. Harnsberger, Harry Hollien, Camilo A. Martin, and Kevin A. Hollien (in press) Stress and Deception in Speech: Evaluating Layered Voice Analysis, Journal of Forensic Sciences.

Here's a filler piece from Slate.
Nina S. Rastogi, Murder, She Wrote: How forensic handwriting identification works, Slate, 26 March 2009.
Here's an interesting standard for expert testimony admissibility:

A lot of times we get caught up in the need for an "expert" witness.  If you have an expert witness the person is testifying under Mil. R. Evid. 702.  True, the expert can "get-in" more information than other witnesses, but do you always need an expert.  Federal Evidence blog reminds us that lay witnesses can often be sufficient testifying under Mil. R. Evid. 701.  This is an excellent blog for all kinds of evidentiary issues and background information.  Here is their note about one such recent case.

Reference Manual on Scientific Evidence (2d ed. 2000), Federal Judicial Center.  A third edition is in the works.

Expert Can Testify About “Role-Playing In … Sexually Explicit Conversations On The Internet”, Federal Evidence Review, 25 September 2008.

For those of us who have done internet solicitation cases, the concept of role-playing and lying amongst participants is a known serious issue.  These cases, as have mine, usually come up because the client shows up to meet the too young for prime time player.  Low and behold blow, it's a cop.  The defense then is "I didn't believe she was under age, I thought this was all part of the game, and that I was going to meet an adult interested in some role-play sex."  Of course these police stings are the ultimate role-playing game. 

In United States v. Joseph, 542 F.3d 13 (2d Cir. 2008), the divided court reversed the conviction.  In dicta, the court said the court should reconsider the denial of expert testimony about role-playing in internet chat sessions, at retrial.

What do you do if you have government sponsored or endorsed "science" that helps catch thieves, cheats, and liars, and other scientists question the validity of your science — you threaten a lawsuit for libel.

Professor Tillers has picked up on the interesting topic of voice analysis as first raised on Deception blog.  He's noted the interesting twists on the issue, including the fact that the magazine that first published the contradictory article removed it from the web — too late.  The offending article was already snapped up and is still available.

See, Prof. Peter Tillers, Voodoo Science in Some Putative Lie Detection Techniques?  Tillers on Evidence and Inference, 16 February 2009.