In answers to written questions preceding his confirmation hearing, now-CIA Director Mike Pompeo caused concern by indicating a willingness to revisit rules governing military and intelligence interrogations. Specifically, he said he would “consult with experts at the Agency and at other organizations in the U.S. government” on whether having a single set of rules for both types of interrogations—which is currently the case (embodied in Army Field Manual 2-22.3 (AFM))—is “an impediment to gathering vital intelligence to protect the country.” Some took his response as leaving the door open to bringing back torture.
When Senator Dianne Feinstein sought clarification, Pompeo explained that what he meant was that he “would listen to any items raised by the High-Value Detainee Interrogation Group [(HIG)] … or other career intelligence professionals that any improvements were needed to the Army Field Manual based on their professional experience.”
The HIG has since weighed in, making clear that it will not be recommending a return to coercive interrogations. In fact, just the opposite.
As many Lawfare readers will recall, in 2015 Senators McCain and Feinstein spearheaded a legislative effort to safeguard against a return to torture, and in particular to block any attempt to resurrect the CIA’s former detention and interrogation program. The result was section 1045 of the FY2016 defense authorization bill, which restricts all government agencies—with certain exceptions for law enforcement—to the list of 19 interrogation methods set out in the AFM.
Section 1045 also mandated a “thorough review” of the AFM to ensure that it complies with U.S. legal obligations, and that none of the interrogation practices the AFM authorizes involves the use or threat of force. As part of the review, the HIG—an inter-agency body comprised of the FBI, CIA and Defense Department—was charged with preparing for the Secretary of Defense, the Director of National Intelligence, and the Attorney General a report on “best practices for interrogation.”
The HIG satisfied that statutory requirement in August of last year, but the report was scant. It noted that a more comprehensive report detailing the relevant scientific research was forthcoming. The HIG released this second report in late March, after review by the FBI, the CIA, the Undersecretary of Defense for Intelligence, the Defense Intelligence Agency, Army Intelligence, and the Marine Corps.
I draw several conclusions from the two HIG reports, discussed below, which have important implications for U.S. interrogation policy. (Full disclosure: I sit on a volunteer committee that informally advises the HIG’s research arm, but I was not involved in the development of either of the HIG’s recent reports).
Humane treatment and effective interrogation are mutually reinforcing goals, a reality that U.S. policy and practice should fully reflect.
The idea that rights-respecting interrogation methods are also the most effective is not new, but the HIG powerfully reinforces it:
Based on the comprehensive research and field validation studies detailed in this report, it is concluded that the most effective practices for eliciting accurate information and actionable intelligence are non-coercive, rapport-based, information-gathering interviewing and interrogation methods.
Building rapport, the HIG says, is the key. Doing so involves “allowing the detainee to develop a sense of autonomy, making him or her less likely to feel coerced and more likely to cooperate.” The interrogator forgoes judgment of the detainee, “accepting that whatever he or she has done is only part of who the detainee is.” She demonstrates empathy. She asks open-ended questions to encourage conversation and employs strategies to enhance a detainee’s memory.
The HIG belies a common instinct, including among some policymakers, that a different (read: more coercive) approach is necessary when dealing with high-level terrorism suspects: “Across intelligence and law enforcement communities, rapport is consistently recognized as the most important aspect of a successful interview.” That is because the “underlying processes of communication, decision-making, memory, cognition, and social dynamics are fundamentally the same in the law enforcement and intelligence gathering settings.” The goal in both is also the same: fostering cooperation (so that a detainee is willing to share valuable information), rather than compliance (which risks a detainee simply telling the interrogator what she wants to hear). “This is a key distinction,” the HIG explains, “because the level of obedience experienced within compliance can lead a subject to provide information he doesn’t actually possess and/or confirm the interrogator’s premises even if he possesses contradictory information.”
Additionally, recording interrogations, which the HIG also endorses, provides a measure of oversight and accountability from both an efficacy perspective and a rights perspective.
The AFM should embody evidenced-based, best practices for interrogation.
Neither HIG report measures the AFM’s interrogation techniques against the relevant science, nor does either offer specific recommendations for revising the AFM (despite explicit statutory authorization to do so). This is unfortunate, because a side-by-side read of the AFM and the HIG reports raises questions about the extent to which some currently approved techniques—particularly one that has raised humane treatment concerns in the past—square with what science demonstrates is most effective. To take two examples:
First, Appendix M to the AFM outlines a technique called “separation,” whereby a detainee is isolated from other detainees for interrogation purposes. In circumstances where physical separation is not feasible, and as a last resort, Appendix M authorizes the use of “goggles or blindfolds and earmuffs…as a field expedient method to generate a perception of separation.” In both cases the objectives are to prevent the detainee from communicating with other detainees and to “foster a feeling of futility.” Field expedient separation has a third objective: to “[p]rolong the shock of capture.” The AFM recommends combining separation with other approaches, including “Futility” (designed to “engender a feeling of hopelessness and helplessness,” where cooperation is the detainee’s “way out”), and “Fear Up” (where the interrogator creates a fear, or identifies an existing fear, which she agrees to mitigate if the detainee cooperates).
But if a detainee feels hopeless and helpless, won’t he be more likely to comply than to cooperate? Wouldn’t stimulating fear undermine steps in the rapport-building process, like demonstrating acceptance and empathy? Does the stress associated with fear not adversely affect memory?
With respect to field expedient separation in particular, why wouldn’t the use of goggles and earmuffs to “prolong the shock of capture” risk producing effects similar to sensory deprivation, which the AFM itself warns can result “in extreme anxiety, hallucinations, bizarre thoughts, depression, and anti-social behavior”?
Second, one of the AFM’s “emotional approaches” discusses strategies for “hasten[ing] the source’s reaching the breaking point.” Is there science behind the concept of a “breaking point”? Isn’t the idea of “breaking” a detainee fundamentally inconsistent with encouraging his autonomy?
To be clear, the AFM warns repeatedly that all of the interrogation approaches and techniques it authorizes must be used consistent with domestic and international law, and in particular that no person in the government’s custody or control, “regardless of nationality or physical location, shall be subject to torture or cruel, inhuman, or degrading treatment or punishment.” But it recognizes that “certain applications of approaches and techniques may approach the line between permissible actions and prohibited actions.” In those cases, the AFM says that interrogators should apply the golden rule: “If the proposed approach [or] technique were used by the enemy against one of your fellow soldiers, would you believe the soldier had been abused?”
The golden rule is an important last line of defense against mistreatment, but it probably shouldn’t be necessary. If there is no verifiable evidence to support the efficacy of techniques or methods that “may approach the line,” and in fact such techniques or methods might be counterproductive, the AFM shouldn’t include them.
Fortunately, the HIG can still do a science-based evaluation of each of the AFM’s 19 techniques before the statutorily mandated review concludes (which—per section 1045(a)(6)(A)(i)—will not be before November 25, 2018). The AFM should then be revised accordingly.
The HIG, and its research component in particular, should have a seat at the table throughout the AFM review, and in other interrogation policy discussions going forward.
The HIG was chartered in 2010 “to deploy the nation’s best available interrogation resources against terrorism detainees identified as having access to information with the greatest potential to prevent terrorist attacks against the United States and its allies…[and to] serve as the locus for interrogation best practices, lessons learned, and research for the federal government.” Since then, the HIG has commissioned a significant body of interrogation research that has led to over 100 publications in peer-reviewed journals. “Equally important, this research has been incorporated into HIG best practices via a continuous cycle of research advising training, training informing operations, and operational experience identifying research gaps and updating training models.”
Intelligence interrogations are too important to U.S. national security to be conducted using anything other than evidence-based, rights-respecting best practices. The HIG is the federal government’s hub for such expertise, which should weigh heavily in any discussions of interrogation rules going forward.