Monday, November 05, 2007

Problems at Defense Forensics Computer Laboratory

The Defense Forensics Computer Laboratory has discovered this past August that it has inaccurately reported information in some of its cases. In a letter to the executive director of the American Society of Crime Laboratory Directors/Laboratory Accreditation Board, the director of DFCL identified discrepancies in a previously examined case, including files being reported as from the wrong profile or username, recovered filenames being misreported as actual picture files, and several computer viruses being inaccurately reported. Examination of other cases worked by the examiner in question revealed similar problems in some of those cases. [NM: available at here]

DFCL tells the accreditation board that it has informed the relevant inspectors general and investigative agencies that may have relied on the affected examinations. I'll send the source documents to No Man for posting on

This is reminiscent of the case of United States v. Luke, in which a USACIL technician gundecked DNA examinations. It will be interesting to see how this plays out in courts-martial.


Phil Cave said...

Not so similar to USACIL. Recollect, USACIL let their rogue go, gave him more training, put him back to work, and then caught him again. Only after the second time did they tell anyone outside their building.

Here, at least DCFL told someone outside their building. What is important is that their general counsel has not alerted anyone who may have prosecuted any case? At least that's what the letter implys.

The question is when do we find out which examiner by name so we can check cases.

Anonymous said...

Surpising nothing more has been mentioned about this issue at the USACIL. They supposedly re-worked all of Mr. Philip R. Mills' cases, back to 1995, using some 'high speed' DNA anaylsts like Drs. Arty Isenberg from Texas and Bob Shaler from the NYC Medical Examiner's office (9-11/WTC fame). Although nothing has been published (or is being released by the USACIL), Mr. Mills' impact on numerous cases is still pending. He worked over 450 cases, sixty of which were re-tested. Of the others, either evidence has been destroyed or those convicted pled with the government for reduced sentences. The CID command out of Fort Belvoir, VA did an admin inquiry (their SOCO) and that report is very interesting reading for anyone seriously interested in this matter. Over 250 pages of interviews, statements and exhibits that shows Mr. Mills wasn't only screwing up numerous cases but, his co0-workers had very little confidence in his work product. I'm talking about CODIS entries that were wrong, contamination in cases (not only those mentioned in the lab's memos, as well as falsification of reports. One co-worker was told (whn he started working at the lab) to be careful when he 'tech reviewed' Mills' work because he missed alot. Cases are still pending review at the service appellate courts - Luke being one of the first - and the government (and the USACIL) don't seem to be doing anything to correct such a mistake.