DOJ Rationalizes Away Polygraph’s Failure to Catch Alleged Anthrax Killer Bruce Ivins

On Friday, 19 February 2010, the U.S. Department of Justice announced the conclusion of its investigation into the 2001 anthrax attacks. The DOJ maintains that  Dr. Bruce Edwards Ivans, who in 2002 passed a polygraph test regarding the anthrax attacks, was the sole perpetrator.

In an investigative summary (640 kb PDF), the DOJ characterizes Ivins’ passing of the polygraph as part of an effort to “stay ahead of the investigation,” alleging (at p. 84, fn. 51) that he used countermeasures to fool the polygraph:

In some sense, Dr. Ivins’s efforts to stay ahead of the investigation began much earlier. When he took a polygraph in connection with the investigation in 2002, the examiner determined that he passed. However, as the investigation began to hone in on Dr. Ivins and investigators learned that he had been prescribed a number of psychotropic medications at the time of the 2002 polygraph, investigators resubmitted his results to examiners at FBI Headquarters and the Department of Defense Polygraph Institute for a reassessment of the results in light of that new information. Both examiners who independently reassessed the results determined that Dr. Ivins exhibited “classic” signs of the use of countermeasures to pass a polygraph. At the time the polygraph was initially examined in 2002, not all examiners were trained to spot countermeasures, making the first analysis both understandable under the circumstances, and irrelevant to the subsequent conclusion that he used countermeasures.

Although the summary doesn’t state what “classic” signs of countermeasures Ivins allegedly displayed, Michael Isikoff of Newsweek reported in 2008 that the FBI “concluded he’d used ‘countermeasures’ such as controlled breathing to fool the examiners.”

While FBI and DoD polygraphers claim that Ivins showed “classic” signs of countermeasure use, it should be noted that no polygraph operator has ever demonstrated the ability to detect polygraph countermeasures. There are no journal articles or book chapters on how to detect them. And retired FBI scientist and supervisory special agent Dr. Drew Richardson’s challenge to the polygraph community to prove its claimed ability to detect countermeasures has gone without takers for more than eight years.

With regard to the psychotropic medications that Ivins had been prescribed, there are no studies on the effects of such medications on polygraph results.

Also not mentioned in the DOJ summary is the fact that the FBI searched Ivins’s premises for, among other things, “materials on how to defeat a polygraph.” Evidently, no such materials were found, or they would presumably have been mentioned in the summary. It’s worth noting that if Ivins had Googled “how to beat  a polygraph” in 2002, he likely would have found AntiPolygraph.org’s on-line book, The Lie Behind the Lie Detector [1 mb PDF], which explains precisely how to do so.

The question of whether an alleged biological terrorist fooled the polygraph is a crucial one for national security. The polygraph remains the centerpiece of America’s personnel security policy–despite the conclusion of the National Academy of Sciences that it’s junk science.

While polygraphers claim that they can now detect countermeasures, they haven’t offered any evidence to support such a notion. It’s not hard to imagine how FBI and DoD polygraphers–armed with the knowledge that Ivins had become the sole suspect in the Amerithrax case–could review his charts and divine signs of countermeasures in them. In a similar manner, polygraph reviewers claimed that the charts of Aldrich Ames–the CIA turncoat who twice fooled the polygraph while spying for the Russians–showed clear signs of deception. But such signs only became clear once other evidence pointed to Ames’ guilt.

And Ames is but the most notorious of a litany of betrayers who have fooled the lie detector: others include Ignatz Theodor Griebl, Karel Frantisek Koecher, Jiri Pasovsky, Larry Wu-tai Chin, Ana Belen Montes, and Leandro Aragoncillo.

Time and again, the polygraph has failed to protect America’s security. And time and again, these failures have been rationalized away. The polygraph operators say, “trust us.” But the failure of the polygraph to detect Ivins’ alleged deception in the Amerithrax case demands a reckoning. There must be no more rationalizations, no more sweeping the dirt under the rug.

The evidence that polygraphers can detect countermeasures generally, and that Ivins used polygraph countermeasures specifically, must be publicly disclosed so that it may be independently and critically scrutinized.

Update: An interview that Ivins gave to the FBI on 16 January 2008 undermines the notion that psychotropic drugs may have influenced his polygraph results or that he employed polygraph countermeasures. The report of that interview states regarding the 2002 polygraph examination that Ivins passed (at p. 199 of Section 4 of Ivins’s FBI case file):

Years ago, IVINS submitted to a polygraph as part of the anthrax investigation. Prior to taking the polygraph, he did not research anything about the test, to include ways to defeat its accuracy. Likewise, he did not take any steps to defeat the tests [sic] accuracy or use countermeasures. In fact, IVINS stopped taking his anti-depression/anti-anxiety medication 48-72 hours before the polygraph, and he offered to provide blood and/or urine specimens at the time of the test to prove he was not medicated.

When IVINS was interviewed in March 2005, he was asked to consent to provide handwriting examplars. Shortly thereafter, he researched experts in the field of handwriting comparisons who could possibly be consulted about the technique. IVINS has no explanation why he researched the handwriting analysis but not the polygraph examination.

In addition, comments that Ivins allegedly made regarding his polygraph examination suggest that he really did not understand polygraph procedure. According to the FBI’s report of an interview of someone who knew Ivins conducted on 5 June 2008 (see p. 68 of Section 6), this person related that:

IVINS spoke about the polygraph he took [for the investigation] and that he passed some questions but failed the one about taking stuff from work.

The question about taking stuff from work is a very common probable-lie “control” question, asked for comparison purposes. Polygraphers secretly assume that everyone has taken stuff from work, and the examinee’s denial is assumed to be less than completely truthful. Polygraphers gauge reactions to relevant questions like, “Did you send those anthrax letters?” against reactions to control questions like “Did you ever take anything from work?” If reactions to the relevant question are stronger, the subject fails, and conversely, if reactions to the control question are stronger, the person passes. (For a fuller explanation of polygraph procedure, see Chapter 3 of The Lie Behind the Lie Detector.)

Ivins evidently didn’t understand the function of such control questions, something that anyone who has researched polygraphy would grasp.

Update 2: Note that in testimony before the Senate Judiciary Committee in 2001, retired FBI polygraph operator Richard W. Keifer averred that “[b]ased on the results of scientific studies, when conducting a screening polygraph, you will have high confidence (99.99 %) on decisions to clear people.” Regarding polygraph countermeasures, Keifer maintained, “The danger from countermeasures, while real, is overstated.”