The Senate Select Committee on Intelligence (SSCI) report on the CIA’s enhanced interrogation techniques (EITs) was cause for much acrimony and little informed debate about intelligence methods. But as we argued in Overt Action, the report, despite its flaws, can be used to inform our judgments about current operations. Now that the initial debate has died down, I want to revisit the question of whether the CIA misled policymakers about program’s effectiveness.
The report does not validate the argument that CIA misled policymakers. What it does show is the presence of what I call the wishful-thinking bias—the tendency of those heavily involved in counterterrorism efforts to search for and find evidence that validates their role and the effectiveness of their operations.
As CIA Director Brennan stated in his remarks about the Senate report, to the extent that CIA still disagrees with the SSCI assessment of the effectiveness of the program, it also disagrees with accounts of how it represented the program’s effectiveness to policymakers. But the more general point that Brennan made, with which I strongly agree, is that the CIA does not and would not mislead policymakers. Such an action would be inconsistent with the strong ethos that exists within the Agency: “While the Agency has a traditional bias for action and a determined focus on achieving our mission, we take exceptional pride in providing ‘truth to power,’ whether that power likes or agrees with what we believe and say or not.”
This characterization, and in my view, is spot on.
Nonetheless, within the pages of SSCI’s report, I do think you can find evidence of a subtler problem. The third section of the report’s executive summary suggests that assessments of the EIT program made by ALEC Station—the former CIA unit within the Counterterrorism Center (CTC) dedicated to finding Bin Laden—were at odds with other assessments of that program. ALEC Station officers seemed to view it as their role to support the EIT program and to marshal evidence in its favor, rather than to do a complete, rigorous analysis of its effectiveness. It was their assessments that formed the basis of representations about the program’s effectiveness that were made to Congress and the White House.
ALEC Station officers were probably extremely well positioned to evaluate the EIT program. They were surely reading through the daily cable traffic and had a deep grasp of the relevant details, as is typical of CTC officers who are immersed in their missions. But I suspect, based on details in the Senate report and on my own time in CTC, that such a complete evaluation never really occurred.
My views here are heavily influence by my time working the Iraq conflict in the 2006 – 2008 timeframe. Joint Special Operations Command (JSOC) was at that time engaged in one of most impressive counterterrorism campaigns in the history of warfare. Seemingly on a nightly basis, teams of special forces were eliminating large swaths of al-Qaeda in Iraq’s (AQI) leadership. Yet those successful operations never had the impact we hoped until they were combined with a broader counterinsurgency strategy in early 2007 (i.e. the surge). AQI was a remarkably resilient organization that could adapt to leadership losses easily.
Despite a lack of compelling evidence, JSOC’s assessments consistently overstated the impact of its operations. Indeed, if its assessments were accurate, AQI would have been destroyed several times over. Instead, the group was more powerful than ever as of early 2007. Back at CIA, our views were often more cautious. As much as we wanted to see AQI defeated, we weren’t personally invested in JSOC’s success and could make a more objective assessment. We often couldn’t find evidence of more than ephemeral impact on AQI. We were consistently proven correct.
It was clear to me at the time that JSOC was too close to its own operations to make an assessment of its impact. Its staff was so deeply integrated into operations—hunting al-Qaeda leaders, contributing to targeting packages, tasking relevant selectors—that it would then grasp at the most anecdotal of details to demonstrate evidence of progress against AQI. This is the wishful-thinking bias in action. It is the result any time those running an intelligence program are responsible for evaluating its effectiveness, and it is a particularly acute problem within the counterterrorism community, where targeters have such an emotional attachment to their targets and where traditional distinctions between operations and analysis have fallen away.
During my time in CTC, I sometimes saw that same bias at work in cases where the CIA was evaluating its own programs. And I suspect the same issue was at play at NSA, in assessments of the value of some surveillance programs that were delivered to the FISC and to Congress. Those guilty of the wishful thinking bias aren’t deceiving or intentionally misleading policymakers. They are trying to speak truth to power, but their version of the truth is clouded by their personal roles and the equities of their agencies. Their bias is the very natural result stemming from an admirable commitment to mission.
Intelligence agencies like the CIA aren’t just biased toward action, as Director Brennan stated. They are biased towards their own action. In some respect this might not be different than many other parts the federal government, but other executive agencies operate in unclassified settings, and their claims and activities are subject to great public scrutiny. That isn’t the case with the intelligence community (IC), where only those with the necessary access can evaluate particular programs.
The bias problems evidenced in the SSCI report are still present today. Here are three steps that should be taken to address that issue.
First, policymakers in Congress and the White House should examine under a microscope any claims coming from a particular intelligence agency about the effectiveness of its operations. Perhaps this could go without saying if it were not for the fact that policymakers rightly assign a lot of credibility to the views expressed by intelligence professionals. Those policymakers are accustomed to the IC’s willingness to speak “truth to power” and to put forward analytic assessments that might be inconsistent with the policies of the Administration and Congress. But not all assessments coming from these agencies should be treated equally, and those implicating an agency’s own operations should be considered with a higher degree of skepticism.
Second, Director Brennan and the new “Directorate of Analysis” leadership should take steps to guard against bias while they aggressively pursue the CIA’s reorganization. In March, Director Brennan announced that the Agency would be establishing new Mission Centers that are essentially modeled off CIA’s Counterterrorism Center and that will combine operational, analytic, and technical capabilities. I support this reform; analysts can become force multipliers for their operational colleagues, steering operations in the direction they can be most impactful, and operational insights can be critical for analysts trying to compete with 24hr news coverage. But this reorganization isn’t without pitfalls, particularly if a close relationship between the analytic and operational sides of CIA undermines its ability to provide “truth to power.”
And third, the Office of the Director of National Intelligence (ODNI) should invest greater efforts to examine the efficacy and programs of the different members of the IC. Examining intelligence value of specific programs is an appropriate role for the ODNI, rather than for the agencies that actually run those programs. The Privacy and Civil Liberties Oversight Board has called for just such as effort (see recommendation #10 here) by the ODNI as a means to achieve a better balancing between security and civil liberties. But work like this from the ODNI would also help hone the IC’s own intelligence gathering activities to maximum effect.