Last week CAAF granted review in the Air Force case of United States v. Knapp II, No. 13-0512, (CCA opinion here) that involves the awesome powers of the “human lie detector,” with the following issue:

Whether the military judge (1) plainly erred by initially allowing “human lie detector” testimony, (2) abused his discretion by allowing further admission of “human lie detector” testimony, over defense objection, and (3) erred by not providing a curative instruction on the “human lie detector” testimony.

Appellant was convicted by a general court-martial composed of officer and enlisted members of aggravated sexual assault in violation of Article 120 (2006) for engaging in sexual intercourse with a female Airman while she was substantially incapacitated by alcohol. He was sentenced to confinement for 3 years, total forfeitures, reduction to E-1, a reprimand, and a dishonorable discharge.

Before trial, Appellant was interrogated by the Air Force Office of Special Investigations (AFOSI). Appellant initially stated that he had consensual sexual intercourse with the female Airman. But as the interrogation progressed over the course of a few hours, Appellant eventually admitted that the victim was unconscious when he began having sex with her. At trial, the Defense tried to show that this later admission was false, caused by AFOSI’s interrogation techniques. But an AFOSI investigator – “Agent P” – testified at trial about many “nonverbal cues” given by Appellant during the earlier, exculpatory part of his interrogation, and explained, “That is indicating to me that there is some form of deception going on.” Knapp, ACM 37718, slip op. at 7 (A.F.Ct.Crim.App. Mar. 20, 2013) (unpublished). Eventually the Defense specifically objected to “human lie detector” testimony from the agent, but the judge overruled the objection and then gave no specific instructions about such testimony.

The AFCCA found error, noting:

What the trial counsel elicited, however, was that these actions by the appellant constitute “nonverbal discrepancies” or “cues” which AFOSI agents are “trained” to recognize. Although the agent was not specifically asked to provide his opinion about the truth or falsity of the appellant’s statements, we find that his overall testimony indicated he could tell the appellant was lying based on the appellant’s physical reactions, and we thus find error.

Knapp, ACM 37718, slip op. at 11. But the CCA affirmed anyway, finding no prejudice due to the “inconsistencies and contradictions” in Appellant’s statements during the interrogation and the general instruction given to the members that they were responsible for evaluating the credibility of the witnesses. The CCA also cited the Defense’s “‘false coerced confession’ strategy” in its conclusion about there being no prejudice. Id., slip op. at 12.

The facts of this case are remarkably similar to another Air Force case involving testimony from an AFOSI agent about indicators of truthfulness that was decided by CAAF a decade ago: United States v. Kasper, 58 M.J. 314 (C.A.A.F. 2003) (opinion). In Kasper, CAAF found prejudicial error and reversed in an opinion authored by then-Judge Effron and joined by then-Judge Baker and Judge Erdmann. As CAAF still has a vacancy, we’ll watch to see if Senior Judge Effron is asked to participate in this case, giving Chief Judge Baker, Senior Judge Effron, and Judge Erdmann another opportunity to discuss the propriety of a law enforcement officer testifying about “training” to recognize untruthfulness through “nonverbal discrepancies” or “cues.”

4 Responses to “The “human lie detector” returns to CAAF”

  1. JOPAJA says:

    Is this informed at all by Salinas v. Texas, No. 12-246, 570 U.S. ___ (June 17, 2013): “[l]ooked down at the floor, shuffled his feet, bit his bottom lip, cl[e]nched his hands in his lap, [and] began to tighten up.”  Perhaps observations would go to juror’s imaginations if prosecutors better prepped officers and asked fewer questions on direct?

  2. Matt says:

    Imagine if CID and AFOSI simply recorded all interviews.  Panels could decide for themselves how the soldier looked at the beginning of the interview, and whether they were displaying characteristics of dishonesty.  For the most part, this would not help soldiers, yet CID still insists on not recording them.

  3. Charlie Gittins says:

    No, no, no.  Defense counsel do not want CID and AFOSI to record.  You can get too much mileage out of cross when you show just how easy it is to record; how most police agencies do so and that they know how easy it is and choose not to do so. Policy?  Really?  Accuracy is less important than policy?  Let the agent explain everything that happened in a 6 hour interrogation where the accused cried, etc.  “If it is so important to know what happened during your interrogation such that you take notes, why, sir, would you choose not to use a $29.00 digital recorder to record every word of this interrogation?”   “Why sir, if your office possesses a digital video recorder, would you choose not to record the interrogation in such an important case, where every word may be important?”  Let them keep f-ing up by the numbers.