“Executive Summary,” Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program  BY THE U.S. SENATE SELECT COMMITTEE ON INTELLIGENCE. 2014, 500 pp. Free online.

“Minority Views of Vice Chairman Chambliss Joined by Senators Burr, Risch, Coats, Rubio, and Coburn,” Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program  BY THE U.S. SENATE SELECT COMMITTEE ON INTELLIGENCE. 2014, 159 pp. Free online.

“CIA's June 2013 Response to the SSCI Study on the Former Detention and Interrogation Program”   BY THE CENTRAL INTELLIGENCE AGENCY. 2014, 136 pp. Free online.

“Overview of CIA-Congress Interactions Concerning the Agency’s Rendition-Detention-Interrogation Program”   BY THE CENTRAL INTELLIGENCE AGENCY. 2014, 43 pp. Free online.

At the end of World War II, the United Kingdom built memorials to every military branch save one: the Royal Air Force’s Bomber Command. The group had skillfully carried out its mission, leveling German cities in raids that killed hundreds of thousands of civilians. But by 1945, the British public had developed second thoughts about the morality and efficacy of that mission. Years later, opinion shifted again. In 1992, the country finally erected a statue, unveiled by the queen mother, of the command’s lead officer, Sir Arthur Travers Harris. Still, the debate over the bombings never really went away; for months, police had to guard the statue from vandalism.

In recent years, the United States has also developed a belatedly guilty conscience about a wartime practice: the torture of detainees in the “war on terror.” The controversy over the program of “enhanced interrogation” carried out by the CIA during the George W. Bush administration was rekindled late last year with the release of a set of reports and documents related to the U.S. Senate Select Committee on Intelligence’s five-year, $40 million investigation of the program. That investigation resulted in a 6,700-page report that remains classified. But after a long struggle between the Senate, the White House, and the CIA, the committee’s Democratic majority released a heavily redacted summary of the report last December, along with a dissenting view from the minority Republicans on the committee. Simultaneously, the CIA issued a rebuttal of the majority report and soon afterward declassified an overview of its interactions with Congress regarding the program. (A disclosure: I have led internal studies for the CIA, including one on U.S. intelligence failures relating to Iraq’s weapons of mass destruction programs, and I currently chair its Historical Review Panel.)

Official line: CIA headquarters, July 2004
Charles Ommanney / Getty Images

The many questions surrounding the interrogation program are not inherently partisan. Yet the committee’s Republican minority withdrew its staff early on, in September 2009, after Attorney General Eric Holder launched his own investigation into detainee abuse, which prevented some CIA officers from talking to the committee. (Although many others could have still provided testimony, especially after the Justice Department closed its investigation in August 2013, the committee did not interview anyone.) So the final report was essentially a product of the Democratic majority, which chose to absolve everyone but the CIA. The Republican rebuttal, in turn, amounted to a brief for the defense, based partly on the CIA’s response to the majority’s charges, with a heavy dose of snark thrown in. Most media coverage of these documents has focused almost exclusively on the majority report, in large part because its authors provided reporters with an advance copy, something that the minority’s staff neglected to do. But any sensible analysis requires reading all the documents.

Although the committee’s prodigious research into CIA records is commendable, its sources tell only part of the story, especially when it comes to determining how the agency made decisions about interrogation techniques. To paint a fuller picture of the program, investigators would have needed the records of other agencies, especially those of the White House. But executive privilege and Republican opposition stood in their way. And even with more complete access, much would be missing thanks to the government’s inhibitions about keeping records of sensitive discussions. In the words of one CIA official quoted in the majority report, “All of the fighting and criticism is done over the phone and is not put into cables.”

Prisoners went for days without sleep, often with their arms shackled to a bar overhead. Sometimes, handlers forced them to go to the bathroom in diapers (Bush expressed unease about this) and delivered food and water rectally (for medical reasons, the CIA claims).
The various reports are deeply disturbing, owing not only to the shameful abuse and torture they describe; they also demonstrate the dogmatism and partisanship that continue to suffuse the politics of U.S. national security in the post-9/11 era. To be sure, the adversarial nature of the U.S. government can prove useful, and bipartisanship isn’t necessarily a good thing. The 9/11 Commission’s commitment to presenting a common front, for instance, prevented it from drawing an obvious conclusion from the relevant facts: that the Bush administration had not seen terrorism as a major problem before the attacks. But in the case of investigating the CIA’s interrogation program, a more collegial process would have served the country better, yielded a deeper understanding of what happened, and made clearer the legitimate disagreements that remain. Instead, the reports fail to present any evidence that could undermine their cases, express any uncertainty, or acknowledge any alternative interpretations of the facts they do include. And each sidesteps fundamental questions about the proper balance between values and security, wasting a rare opportunity to fill the vacuum that allowed torture to occur in the first place.


The report and the rebuttals agree on five central points. First, they all line up on some essential facts and share a basic timetable. Six days after the 9/11 attacks, Bush signed a classified memo authorizing the CIA to “capture and detain persons who pose a continuing, serious threat of violence or death to U.S. persons and interests or who are planning terrorist activities.” (The document made no mention of interrogations.) In the years that followed, the agency detained at least 119 individuals. The use of “enhanced interrogation techniques” began in 2002. The agency ceased using one of the most controversial techniques, waterboarding, in 2003; by 2007, it had stopped using the other ones as well. But the program didn’t officially end until January 2009, when President Barack Obama signed an executive order that required the CIA to close its detention facilities and prohibited any U.S. government employee from using interrogation techniques not permitted by the U.S. Army Field Manual.

A second basic point is also beyond dispute: the “enhanced interrogation techniques” were brutal. Prisoners went for days without sleep, often with their arms shackled to a bar overhead. Sometimes, handlers forced them to go to the bathroom in diapers (Bush expressed unease about this) and delivered food and water rectally (for medical reasons, the CIA claims). Whether these and other tactics, such as slapping or waterboarding, qualify as torture under the law remains subject to debate. But they certainly count as torture in the ordinary sense of the term.

Detainees in a holding area at Guantanamo Bay, January 2002.
Reuters / U.S. Department of Defense

That said, the majority report never uses the term “torture” to describe U.S. actions. Nor does it wade into the controversy about the quality of the legal opinions composed by Justice Department officials between 2002 and 2007 that declared the CIA’s techniques permissible, most likely because the committee was unwilling to deal with the implications of taking a stance on the question of whether the CIA broke U.S. or international law. Classifying CIA practices as crimes would have raised the uncomfortable question of who should be indicted for them. The list of candidates, after all, could have included Bush and Vice President Dick Cheney. Declaring the CIA’s techniques legal, on the other hand, would have undercut the committee’s message that they should never be used again.

Defenders of the program point to the absence of more severe techniques, and they are correct that there was a line that interrogators did not cross. No one seems to have suggested breaking a prisoner’s bones, stubbing out cigarettes on his skin, attaching electric wires to his genitals, or pulling out his fingernails, for example. But a Goldilocks view of the CIA’s interrogations—that they were not so cruel as to constitute torture but just harsh enough to compel hardened terrorists to divulge critical intelligence—seems too convenient to be true.

All sides accept that the CIA was unprepared to hold prisoners and that from the program’s inception in March 2002 to early 2003, the agency managed the interrogations poorly, doing little to prevent unauthorized excesses. (Many of the most egregious abuses, including what led to the death of one prisoner, occurred during this period.) They also agree that the agency did not systematically assess the value of the information that the torture produced. Although the CIA did conduct several reviews, none was thorough, and none involved outside experts. Finally, there is consensus that the CIA often overstated the value of the information that came from the torture. It remains unclear, however, whether the agency’s leaders knowingly misled others, were misled by subordinates, or misled themselves. In any case, a lack of objective analysis helped muddy perceptions all around.

These areas of agreement aside, the various accounts diverge, beginning with the majority report’s allegation that the CIA, in a gross betrayal of duty, failed to tell Bush what it was doing. That seems unlikely. Although the majority could not find any records showing that the CIA briefed Bush on its interrogation practices before April 2006, years after the program began, here the old maxim applies: “Absence of evidence is not evidence of absence.” For one thing, the committee could not examine White House records, and Bush himself has said that he was adequately and appropriately informed. For another, the majority report notes that the National Security Council staff requested that the secretaries of state and defense not be briefed on the program’s specifics, which suggests that members of the president’s inner circle were fully informed gatekeepers. The committee also makes little mention of Cheney and his staff, who, according to many media accounts, were also deeply involved.

All 39 of the people the CIA tortured would have had to divulge everything useful they knew before being tortured or given up nothing once they were. In theory, it’s possible that’s what happened, but it seems unlikely, given the wide range of ways that different people respond to duress and pain.
The majority report’s parallel but more self-serving claim is that the CIA “actively avoided or impeded congressional oversight of the program.” In this case, too, the available facts tell a more complicated tale: the agency acknowledges that it did not inform the full Intelligence Committee about the program until late 2006, but the committee confirms that beginning in September 2002, the CIA did brief the chairs and vice chairs of the House and Senate Intelligence Committees, along with the leadership of each house of Congress. So even if the CIA should have brought in the entire committee earlier, it did inform the key congressional leaders. There is no indication, moreover, that those legislators pressed hard for additional information until early 2005.

The central argument of the majority report—that when the CIA did brief Congress and the president, the agency inaccurately claimed that its methods were working—invites similarly close scrutiny. The reasoning here is intricate, turning on legalistic definitions, detailed timelines, and implied counterfactuals. It is especially hard to follow because the report addresses the CIA’s rebuttal in footnotes and because much of the underlying evidence is available only in the committee’s full study, which remains classified. The majority report’s first conclusion is that “the CIA’s use of its enhanced techniques was not an effective means of acquiring intelligence or gaining cooperation from detainees.” Although it never defines “effective,” the report essentially claims that torture never provided any valuable intelligence in the cases it examines.

Outside the Camp Five detention facility at Guantanamo Bay, December 2008.
Reuters / Mandel Ngan
On its face, that position is suspect. For it to be true, all 39 of the people the CIA tortured would have had to divulge everything useful they knew before being tortured or given up nothing once they were. In theory, it’s possible that’s what happened, but it seems unlikely, given the wide range of ways that different people respond to duress and pain. And if torture were uniformly ineffective, there would be no need for a treaty banning it—which the Geneva Conventions do.

The majority’s argument has problems with evidence as well. In judging the torture’s effectiveness, the majority report looks for direct connections between the intelligence derived from the torture and its benefits to national security. But the minority and CIA rebuttals are right to urge a broader view. For one thing, analysts needed a great deal of information about al Qaeda before they could make sense of any one source. By the majority report’s standard, the torture was not effective if it merely contributed to a general understanding of al Qaeda, rather than leading directly to the foiling of a terrorist plot or the capture of an 
al Qaeda member. Yet crucial insights often result from indirect links. It might have been, as the majority report argues, that breaks in many cases came from prisoner interrogations that did not involve torture. But in some cases, interrogators asked those detainees questions because of intelligence that came from others who were tortured. And although the majority report lends little weight to information that simply confirmed other intelligence, such findings can prove invaluable, since tips from individual sources are rarely sufficient to merit action on their own. In essence, the report and the rebuttals talk past each other on this point: the Democrats dismiss evidence of a type that the Republicans and the CIA (rightfully) consider central.

The majority report also relies heavily on hindsight. Although the majority report is probably correct to say that in many cases the CIA had enough information to identify a terrorist or break up a plot without torturing detainees, it does not consider how the CIA actually reached its conclusions. In some instances the report cites, the CIA could have drawn multiple judgments from the existing intelligence; in others, agency analysts overlooked valuable information. Whatever criticism is due, saying that the CIA could have reached a conclusion without torture is different from saying that it would have.

The debate over whether torture produced useful information should not be confused with the broader question of whether the interrogation program did more harm than good.
Take, for example, one piece of intelligence that proved critical to the CIA’s hunt for Osama bin Laden. The majority report claims that the CIA could have identified bin Laden’s courier, Abu Ahmed al-Kuwaiti, without information derived from torture. But merely reciting other evidence that Kuwaiti was the courier, as the report does, doesn’t prove anything. The only way to test the report’s claim would be to examine the entire CIA database, remove any intelligence gleaned from torture, and then determine if Kuwaiti should have stood out from the many other suspects that were in play. The problem, in other words, wasn’t connecting the dots; it was that too many of them looked alike. What clinched the Kuwaiti case, the CIA claims, were suspicious denials by two major al Qaeda figures, after they were tortured, that Kuwaiti was close to bin Laden—something that set analysts on his trail. The agency makes a similar argument in explaining how it found Saajid Badat, who had been trained to bring down an airliner with a shoe bomb in 2001. After the plot failed, Badat was not immediately identified as one of the conspirators. The CIA claims that evidence obtained from Khalid Sheikh Mohammed, whom the agency waterboarded 183 times, was crucial in pointing them in Badat’s direction. “We were following many disparate individuals who were allegedly threatening U.S. interests,” the CIA says in its rebuttal, “and there was nothing at the time on Badat to lead us to prioritize him over the others.”


The debate over whether torture produced useful information should not be confused with the broader question of whether the interrogation program did more harm than good. Even if torture worked in the narrow sense, its costs might have outweighed its benefits: the negative global reaction to the CIA’s brutal methods decreased support for U.S. policies and may have helped terrorist groups win more sympathizers and recruits. At the very least, the torture committed alienated U.S. allies and gave ammunition to those who opposed Washington’s policies, contributing to the belief that the United States was hypocritical in its public defense of liberty and prone to treat Muslims as less than human. But neither the Senate reports nor the CIA rebuttal seeks to calculate those costs.

A U.S. Army soldier looks through binoculars while standing on a guard tower at maximum security prison Camp Delta at Guantanamo Naval Base in Guantanamo, Cuba, August 26, 2004.
A U.S. Army soldier looks through binoculars while standing on a guard tower at maximum security prison Camp Delta at Guantanamo Naval Base in Guantanamo, Cuba, August 26, 2004. 
Mark Wilson / Reuters
The CIA’s interrogation program raised a host of moral questions as well, which the Senate reports and the CIA rebuttal ignore. Perhaps this is just as well; one has no reason to expect senators or intelligence officials to be especially qualified on the subject of morality. But both the Democratic majority report and the Republican dissent take easy ways out. By claiming that the torture was ineffective, the Democratic report encourages a sense of indignation and implies that the interrogation program was morally indefensible. The Republican dissent, for its part, contents itself with claiming that the torture did produce useful information but avoids an accounting of its moral and political costs, suggesting that such concerns have no place in counterterrorism policies.

All the documents suffer from a shortcoming that has characterized the debate over interrogation ever since press reports revealed the truth about the CIA’s “enhanced techniques”: a failure (or refusal) to acknowledge the existence of tradeoffs between competing values. Looking back, no one wants to argue that the CIA’s techniques constituted torture but were also an important tool in defending against terrorism. Likewise, no one seems willing to contend that the “enhanced interrogation techniques” were ineffective but nevertheless nothing to be ashamed of. Yet the alternative conclusions aren’t all that convincing. The Republicans describe the techniques as less brutal than torture but harsh enough to make hardened terrorists betray their cause. The Democrats argue that those detainees would have provided useful information through benign interrogation techniques but were so committed that they would not yield anything useful when waterboarded. Both these formulations are highly implausible, and both spare their beholders from confronting the possibility that the CIA tortured people, acted immorally, and also saved lives.

It’s conceivable that the magnitude of the threats the United States faced after 9/11 required some sacrifice of values. The CIA’s leadership seems to have adopted that view; consider, for example, the agency’s decision to not punish those responsible for torturing Khalid al-Masri, a German citizen the agency imprisoned after mistaking him for someone else. In such cases, the majority quotes the CIA as saying, “the Director believes the scale tips decisively in favor of accepting mistakes that over connect the dots against those that under connect them.” But one could use that logic to justify virtually any tactic—and in Masri’s case, there were no dots to connect at all. Critics have also pointed out that many of the horrific contingencies officials have worried about—for example, a terrorist plot to bring down the Brooklyn Bridge by loosening its bolts and cutting its cables—have been little more than fantasies. Either way, assessing the scale of the terrorist threat today remains difficult, and Americans have not had a substantive debate about the kind of sacrifices they should be willing to make in the service of their safety. Americans generally agree that they should not change their way of life, or cede their tolerance of satirical movies or inflammatory cartoons, in order to conciliate their enemies. A corollary might be that the United States should preserve the principle of treating prisoners humanely even if Americans die as a result. The country does not pay ransom to terrorists; perhaps it should also not surrender its values to them.

Such difficult questions require a national conversation. Unfortunately, the Senate Intelligence Committee forfeited its chance to lay the foundations for one. Indeed, the majority report suggests that little further thought is needed, clearing almost all involved. According to the Democrats on the committee, the American people, Congress, the Department of Justice, and even the president himself were either kept in the dark by the CIA or deceived by it into needlessly allowing torture to continue. The majority report’s authors seem to want Americans to accept these findings, condemn the CIA, and then simply vow to never permit torture to recur.

Should the United States suffer another massive terrorist attack, U.S. officials might very well decide to use torture again.
But preventing torture from happening again would require a much less blinkered examination of the facts, a far deeper national consensus about the possible tradeoffs between rights and security, and a clearer understanding of what U.S. and international law permits and forbids. Does the United States need new laws, an executive branch that interprets them differently, or stronger punishments for offenders? The majority report doesn’t just fail to offer clear answers; it doesn’t even help Americans think more deeply about the questions themselves.

Consider the contrast between the majority report on torture and the various reports issued in 1975 and 1976 by the Church Committee, a Senate body that investigated and revealed abuses and crimes committed by the CIA and other U.S. intelligence agencies during the Cold War, including spying on American citizens and conspiring to assassinate foreign leaders. At the start, Senator Frank Church, who led the investigation, likened the CIA to a rogue elephant; by the end, he had concluded that the agency had in fact carried out presidential policy. That willingness to minimize partisanship and evaluate history with an open mind lent credibility to the committee’s policy recommendations, many of which Congress implemented.

The Senate Intelligence Committee could have taken a similar path, by candidly considering how a policy of torture resulted from decisions made at the highest levels of the executive branch and was conducted without objection from the CIA’s congressional overseers. Bush clearly supported the use of torture while it was occurring and has continued to defend its use; Cheney has been even more aggressive in pushing back against criticism, refusing to acknowledge any flaws at all in the way the CIA used so-called enhanced interrogation techniques. And even though Obama has said that he believes the CIA “tortured some folks,” his administration has declined to prosecute anyone for doing so. What’s more, for all the outrage generated in some quarters by the majority report’s revelations about the CIA’s brutal methods, an opinion poll conducted by The Washington Post and ABC News after the report’s release found that a large majority of Americans—59 percent—nevertheless support the use of those methods. This all suggests that should the United States suffer another massive terrorist attack, U.S. officials might very well decide to use torture again. If the authors of the majority report believe their efforts have made that outcome less likely, they are mistaken. In the end, a less political report might have had more influence.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • ROBERT JERVIS is Adlai E. Stevenson Professor of International Politics at Columbia University and a member of the Saltzman Institute of War and Peace Studies.
  • More By Robert Jervis