Transparency is critically important to the legitimacy and effectiveness of democratic governments.  Transparency facilitates both the responsiveness of government to the needs of the electorate, and accountability for wrongdoing.  But the desire for transparency has always fit imperfectly with the needs of intelligence agencies, for which secrecy is a critical ingredient in the success of their operations.  Secrecy is required to protect the identity of sources, who often put themselves at great risk to provide important information to our intelligence services.  It is also required to protect the methodologies by which intelligence agencies collect information, in order to reduce the danger that foreign intelligence targets change their behavior in ways that make it harder for the government to obtain the information it needs.

Most of the time this tension is latent, as a robust architecture of compliance including all three branches of government works to ensure that the Intelligence Community behaves lawfully, even if this architecture mostly operates in secret.  And in fact, one of the byproducts (likely unintended) of the unauthorized disclosures by Edward Snowden has been the revelation of just how vigorous the government’s internal mechanisms of compliance are.  But in the last several weeks, tensions between the importance of transparency and the Intelligence Community’s need for secrecy have come into the open, as a dispute has emerged between the Senate Select Committee on Intelligence (SSCI) and the Central Intelligence Agency (CIA) about a thorough and voluminous report the SSCI wrote about the CIA’s use of Enhanced Interrogation Techniques (EITs).

A detailed discussion of the current dispute is beyond the scope of this post.  Suffice it to say, however, that concerns have emerged about whether the SSCI staffers improperly accessed an internal CIA document, publicly described as the “Panetta Review,” and whether, in turn, the CIA improperly monitored computers used by SSCI staffers to research and write their report.  The Panetta Review, whose contents remain classified, reportedly describes a number of dimensions of the interrogation program.  Most relevant here, however, is the question of “whether the program produced significant intelligence.”  That is, whether the Intelligence Community gained important intelligence information with the use of EITs that it did not obtain through other means.

The SSCI appears poised to declassify and release parts of its report, but advocates of declassification, particularly those eager to determine whether the EITs yielded unique information, are likely to be disappointed if and when some version of the document is made public.  This is because the information that would be required to make conclusive determinations about the utility of EITs would likely be too sensitive to declassify and release.  Rather than describe in detail the nature of the dispute between the CIA and SSCI, a subject well-covered elsewhere, this post will focus instead on why any partially declassified report is unlikely to resolve conclusively the debate about whether the EITs “worked.”  Those who advocate for the release of the report, therefore, should do so with other objectives in mind.

In 2009, the SSCI commissioned a report intended to “provide a detailed, factual description of how interrogation techniques were used, the conditions under which detainees were held, and the intelligence that was – or wasn’t – gained from the program.”  Over the next three years, the Committee’s staff reviewed over six million pages of records, and produced a 6,000 page report with over 35,000 footnotes, which was approved by the Committee in December 2012.  The Committee staff apparently relied “solely” on documents in compiling the report, and did not incorporate “the views of either C.I.A. or other executive branch officials.”

Discussion of both the report, and the potential for its partial declassification and release, formed an important part of the hearings held in February 2013 to consider John Brennan’s nomination to the lead the CIA.  And indeed, a movement for the partial declassification and release of the report has gained steam in the last several weeks as more information about the dispute between the CIA and the SSCI has emerged.  While there may be many good reasons to release a redacted version of the report, there are also strong reasons to be cautious while doing so.  General concerns about the ability to declassify portions of the report without jeopardizing sources and methods of intelligence gathering have been amplified by the issue alluded to above, which has been a core focus of the recent declassification debate:  Whether or not the interrogation program yielded intelligence to which the CIA did not otherwise have access.  Attempting to publicly release a redacted version of the report that can both meaningfully address this question and protect information that the intelligence community must keep secret will be extraordinarily difficult.  Here’s why.

Most declassification decisions generally are made by weighing the potential risks to sources and methods against the benefits of transparency.  In this case, however, intense attention will be paid to whether or not the value derived from the interrogation program was able to justify the moral, legal, and reputational costs that the United States bore in its aftermath.

In order to arrive at a reasonable determination of the utility of the program, however, one needs to examine all the other information to which the Intelligence Community had access both before and during the time that the EITs were used.  What else did the IC know about the networks, plots, financial support structures, weapons pipelines, and other features of al-Qaida’s operations at the time of the interrogations at issue?  How confident was the IC in the accuracy of this information?  Where did that other information come from, and what were the motives of those who provided it?

Unless one has answers to these questions it will be difficult to determine with confidence the marginal value that the use of EITs provided.  And presumably the full 6,000-page report does include a detailed discussion of just those issues.  But releasing a sufficient amount of this information for the public to form nuanced judgments of the value added by the EITs will be extraordinarily difficult to do without putting sources and methods of intelligence gathering in jeopardy.

A further concern relates to the risk of disclosure of information that came from foreign governments, and, more generally, the risk that such cooperative intelligence-sharing relationships may be jeopardized by disclosing their existence or depth.  Indeed, one may surmise that, especially in the early post-9/11 years, the U.S. government received a good deal of valuable information about al-Qaida and related terrorist networks from the countries in the Middle East and South Asia where al-Qaida operated and from which most of its members came.  The United States might not always have friendly overt relationships with some of these countries, and the foreign governments that shared information with the United States might have done so on condition that their cooperation remain private.  If the public is to be able to determine whether the EIT program yielded valuable information, it will necessarily need to determine what the U.S. government already knew (much of which might have come from foreign governments) and with what degree of confidence.  Making these determinations in a declassified setting will be exceedingly difficult.

For some, none of these considerations will matter; for them, whether or not the EIT program yielded useful information beyond what already existed is immaterial, and the report should be released for other reasons.  Chief among which may be a general commitment to as much transparency as is reasonably possible in the national security space, and a desire to learn more about the EIT program so that the country can finally move beyond this episode in our history.  But while some degree of improved knowledge about the program is within the public’s immediate grasp if a redacted declassified version of the report is released, nuanced relative judgments about what the EITs added to the store of knowledge already possessed by the Intelligence Community may need to wait for some time to come.