Identifying And Explaining Change: Development Of US Interrogation Policy 1949-2006

The consensus view holds that US interrogation policy has, since September 11, 2001, undergone a fundamental transformation resulting in a “New Paradigm.” However, this view grossly oversimplifies the complex pre-9/11 history of interrogation policy, a history that is largely unwritten. My paper attempts to fill this gap through (1) unearthing the history of pre- 9/11 interrogation policy and (2) identifying points of change and the reasons for that change.

The thesis finds three overlapping phases in post World War II interrogation policy: 1949-1972, – a period characterized by a “permissive ” interrogation policy that became increasingly less so; 1972-2001, – a period of generally restrictive policy as applied to interrogation conducted by US servicemen, with an important caveat explained below, but permissive as applied to interactions with foreign intelligence services; and post-9/11 – a more permissive period, compared to 1972-2001, as applied to US servicemen, but with some interesting and important continuities with developments in the preceding period. (The paper understands interrogation policy to be those interrogation methods that have been explicitly authorized or sanctioned by official doctrine (as represented in directives or manuals), training guides, course materials, or interrogation research programs. The paper relies upon US Army and CIA interrogation manuals, Army and CIA project memoranda documenting operationally applied interrogation research, as well as Army and CIA training and instructional materials for foreign intelligence counterparts.)

This exploration of the history of interrogation policy makes three sets of contributions. First, it illuminates what seems to have driven the evolution of that policy; presidential leadership and legal changes seem only weakly connected to policy changes while the priorities of the intelligence agencies seem most important. An additional finding is that there is less evidence than generally believed to support the oft-repeated view that national security operatives questioned the value of harsh interrogations; indeed, it appears that the direct use of physical force was declared ineffective often for institutional reasons that had nothing to do with the actual efficacy of such questioning itself.
Second, it places current interrogation practice in the context of a complex and novel history that reveals how open-ended terms like ‘torture’ and ‘permissible’ have been understood over time. Pre-September 11 policies have at times been relatively permissive. There are three important points to be made here. First, almost without exception, the techniques approved at any one time post-September 11 for military interrogations of unlawful combatants would have been understood to fall within the constraints of the Geneva Protections for protected Prisoners of War at different points of time in the pre-9/11 periods. Second, modern rendition and related practices reflect longstanding American tolerance of coercive techniques when used by non-US personnel. And, third, directly relevant to the current debate on interrogation policy, pre-9/11 Army interrogation policy was already structured such that doctrine differentiated starkly between classes of combatants and the types of limits applicable to their status and interrogations. This is reflected in the omission of Common Article 3 in the 1992 Army interrogation manual.

Third, much of the debate over interrogation has been cast in terms of legality as if the applicable legal regimes were precise. They are no t, as the paper demonstrates by examining the changing scope of non-torture conduct considered permissible. The paper argues that what is “legal” under these various legal regimes, at least when they were first developed and subscribed to, gives very broad scope to interrogators. But this misses the point as to what guidance we now should give to our intelligence agencies. Rather than focusing on what is legal under these various existing regimes, it would be more helpful to focus on which of the various techniques are necessary, useful, morally appropriate, and administrable.