Playing to the Edge: American Intelligence in the Age of Terror

by Michael V. Hayden

  He challenged the report’s findings that CIA routinely went beyond the authorized interrogation techniques and misled the Justice Department, the White House, Congress, and the American people. He pointed out that the report chose to ignore the context of the time in which the program was launched and the fact that the agency was not operating alone (he noted more than thirty briefings to Congress).

  Most important, he (like the agency and Republican rebuttals) challenged the “claim that the CIA’s interrogation program was ineffective in producing intelligence that helped us disrupt, capture, or kill terrorists,” citing multiple examples of its effectiveness, including the bin Laden takedown.

  One agency wag put it more bluntly: “The Feinstein report would have you believe that the people who got bin Laden just didn’t know how it was they got bin Laden.”

  I received special mention in the Feinstein report, being cited more than three times as often as any of my predecessors. That was not because of the number of detainees I had (two in a total of about a hundred), but rather because I was the director who briefed the entire committee on the full scope of the program.

  They particularly focused on my April 2007 testimony, dedicating all of appendix three to pointing out what they believed to be inaccuracies. A lot of the issues had to do with the still raging argument over what Abu Zubaida (the first detainee to be waterboarded) told us, when, and why. Other issues could fairly be described as my briefing the standard and their searching through millions (literally) of pages to find the deviations, most of which were early in the program.

  Then there was the issue of what constituted the program. I said that the program I was briefing was created, at least in part, because of the poor agency performance with early battlefield captures—as thoroughly documented and shared with the SSCI by CIA’s inspector general. The committee knew that Gul Rahman had died in agency custody, for example, but CIA never considered him part of this program.

  And finally, there is the very real possibility that in two hours of testimony discussing things five years distant and separated from me by the administration of two other directors, I may have just gotten some things wrong. It’s possible.

  Most important, our purpose for the 2007 session—as well as similar sessions with the HPSCI—had not been to narrate a definitive history of the RDI program, but to explain its current status as a first step in building a consensus on a way ahead. That never happened, of course.

  I got wrapped around another axle in the report concerning the number of detainees. According to the report, I “instructed a CIA officer to devise a way to keep the number of CIA detainees at the same number that the CIA had previously briefed to Congress.” The report says that was ninety-eight; I think it was actually ninety-nine.

  The alleged “incident” took place in January 2009 as I was getting ready to step out the door. One CTC officer suggested that the right number of detainees in the program could be as high as 112. There had always been questions as to who should be counted in the program, and early bookkeeping had been sloppy, but I couldn’t resist offering a half smile and saying, “You people have pushed me out there for three years with ninety-seven or ninety-eight [as we added detainees].”

  The agency rebuttal reflects the consensus from that meeting that the new CTC numbers were still “somewhat speculative and incomplete.” I said that if there really were new numbers, they better make sure and then tell the new director to pass them on to Congress.

  The Feinstein report settled on “at least 119” (not 112) as the right number of detainees to book under the program.

  That said, Attorney General Holder, at the conclusion of the Durham investigation, confidently announced that “Mr. Durham examined any possible CIA involvement with the interrogation and detention of 101 detainees who were alleged to have been in United States custody.” Looks like it was always a hard number—99, 101, 112, 119—to pin down.

  I’ve since reflected on the Feinstein report and what lessons to draw from it. One positive take-away was the clear need to brief Congress fully and contemporaneously on sensitive activity. No fair relying on a director five years later to reconstruct events.

  I’m close to drawing a second, darker conclusion too. Be careful what you tell these people. Some are less interested in honest dialogue than listening to rebut and accuse and discredit.

  Which, I suppose, again raises the question of motive. Why the report? CIA was out of the interrogation business. It wasn’t going back.*

  Here, too, my darker angels suggest an answer. I picture those who really pushed the release thinking that this earned for them a sense of expiation. They get to say, and say publicly, “See. This wasn’t about us. It was never about us. We’re not like that. Those people, those people over there. The ones who lied. It’s about them. Let me show you.”

  Senator Rockefeller was an especially strong advocate for the Feinstein report. He issued his own sharply critical statement at its release in December 2014.

  In March 2003, following the capture of Khalid Sheikh Mohammed, the senator’s tone was somewhat different.

  CNN’s Wolf Blitzer asked him what would happen to KSM.

  ROCKEFELLER: Well, happily we don’t know where he is. . . . He’ll be grilled by us. I’m sure we’ll be proper with him, but I’m sure we’ll be very, very tough with him.

  That raised the question of torture for Blitzer.

  ROCKEFELLER: That’s always a delicate question. . . . On the other hand, he does have the information. Getting that information will save American lives. We have no business not getting that information.

  Blitzer again raised the question of torture.

  ROCKEFELLER: We do not sanction torture, but there are psychological and other means that can accomplish most of what we want.

  Blitzer then asked about handing him over to a third country “where the restrictions against torture are not in existence.”

  ROCKEFELLER: I wouldn’t take anything off the table where he is concerned, because this is the man who has killed hundreds and hundreds of Americans over the last ten years.

  Hmmm. You could see how the folks at the agency might have been confused.

  They, of course, disagreed with the report, its narrative, its method, and its conclusions. And they especially disagreed that this was all just about them. If it was just about them, congressional Democrats (those who had been briefed) would have begun their protest in 2002—when the trauma was recent, the threat seemed imminent, and the future was in doubt—and not in 2014, when it was not.

  Americans agreed with the agency. Poll after poll following release of the Feinstein report sided with CIA by a wide margin. Pew put the split at 51–29 that CIA’s actions were justified. The split was even larger, 56–28, that intelligence from the program prevented attacks. For those who said they were closely following the issue, the numbers were even stronger. They supported the program 59–34. CNN had similar numbers, 66–34 overall.

  Brookings, a highly regarded Washington think tank, led its coverage with the headline “Americans Have Taken Ownership of the CIA’s Interrogation Program,” pointing out that CIA tactics were approved both retrospectively and prospectively and concluding that “future presidents who refrain from doing what the Bush-Cheney administration did may well find themselves politically exposed—especially if terrorists once again manage to kill large numbers of Americans.”

  Not quite the outcome intended by the Democrats on the intelligence committee.

  But not quite the final word, either. The RDI program cries out for an accurate, dispassionate history. The six of us said as much in that WSJ op-ed, calling the Feinstein report “a missed opportunity to deliver a serious and balanced study of an important public policy question. . . . The country and the CIA would have benefited from a more balanced study of these programs and a corresponding set of recommendations. The commit
tee’s report is not that study.”

  We still await that dispassionate history.

  TWENTY-ONE

  THE PRIVATE SECTOR

  WASHINGTON, DC, 2009–2014

  In early June 2013, I was settling into a comfortable hotel in La Jolla, California, when I got an e-mail from Mark Hosenball, intelligence beat reporter for Reuters. “Is this real or nuts?” was his subject line. The text was a few words and a hyperlink to an article by Glenn Greenwald that had just been posted on the Web site of the London Guardian.

  The Guardian headline screamed, “NSA Collecting Phone Records of Millions of Verizon Customers Daily,” and a subhead pointed to a court order authorizing the operation.

  I liked Hosenball, but I didn’t answer his e-mail. This wasn’t nuts. This was real. But I wasn’t going to talk about it. This was the follow-on program to the telephone metadata we had gathered in Stellarwind, now done via a court order, but it wasn’t anything I could confirm or even try to explain. I did wonder, though, How the hell did that get out?

  The next day Greenwald in the Guardian and Bart Gellman in the Washington Post claimed that NSA had free access to the US-based servers of Microsoft, Google, Yahoo, Skype, YouTube, Facebook, AOL, and Apple under something called PRISM.

  The unfolding press coverage was fast taking on the look of a journalistic version of Tommy Franks’s “shock and awe” campaign in the second Iraq war.

  Before long we all learned that the leaks were the product of a boyish-looking twenty-nine-year-old Booz Allen contractor, Edward Snowden, who was then on the lam in Hong Kong trying to arrange transit through Russia en route to South America.

  By the summer of 2013 I had been out of government for more than four years. I was enjoying being a principal at the Chertoff Group, a security consulting firm launched by former DHS secretary Mike Chertoff, and teaching at George Mason University’s Graduate School of Policy, Government and International Affairs in northern Virginia. The Chertoff Group kept me around like-minded and like-experienced folks. It was always good to routinely share thoughts with the likes of Charlie Allen, an agency legend. The George Mason position was the product of a casual conversation with a faculty member at former senator Chuck Robb’s 2008 Christmas party. Noting what he called my “scholarship of practice,” the faculty member shepherded my forty-year-old master’s degree through GMU’s administrative wickets to land me a distinguished visiting professorship. I enjoyed teaching the course, “Intelligence and Policy,” which routinely filled up within a minute or two of online registration, although I suspect that had something to do with the hope I was going to spill some secrets.

  I was also doing the routine postgovernment thing of signing on to some boards and consultancies. I often wondered what in particular I had to offer, but the invitations were numerous and steady. At one such encounter, Charlie Allen and I spent several hours with some corporate leaders explaining the structure and dynamics of the intelligence community. It wasn’t anything exotic, and we remarked in the elevator after we were done that neither of us could believe that they wrote all that stuff down. Maybe we were underestimating what we had picked up in the near century of IC experience between us.

  An active speakers’ bureau rounded out my postgovernment portfolio and managed to fill in most of the empty spaces on my schedule.

  I missed government work less than I thought I would. When asked, I would respond that I missed the mission and I missed the people (and would occasionally add that I missed the jet), but that was about it. Most mornings I would roll over, grab my iPad, and read the CIA press clips before getting out of bed. When I later delivered my wife her coffee, she would often ask, “Anything in the clips?” and I would just as often respond, “Yet another great day to be a former senior intelligence official.”

  That said, I had taken to writing and commenting on intelligence matters since I left government, so I was quickly pulled into the debate that was swirling around Snowden.

  An early point I made was that this debate had actually been long in coming. By the late 1990s, NSA was well aware that legitimately targetable foreign communications (like those of the Soviet Union’s strategic forces) were no longer confined to isolated adversary networks. Modern targets (like al-Qaeda’s e-mails on the World Wide Web) were coexisting with innocent and even constitutionally protected messages on a unitary, integrated global communications network.

  The dilemma only got more acute after 9/11, since the enemy had demonstrated that he was already inside the gates and—even when he physically might not be inside the United States—his communications often were, as terrorists opted to use e-mail services based in this country.

  The sins that Greenwald, Gellman, and others were attempting to portray were efforts by NSA and other intelligence services to deal with these new realities. Some of the efforts did indeed raise important questions about the right balance between security and liberty, and Snowden’s disclosures no doubt accelerated and intensified that discussion. But the disclosures, and especially how they were rolled out, badly misshaped it as complex stories were misreported or, worse still, purposely pushed to the darkest corner of the room.

  Take the PRISM program, for example, disclosed on day two of the campaign. Rushing against deadline, some outlets reported (inaccurately) that NSA had direct and free access to the servers of American communication service providers and it was a short step from there to near-libelous accusations that NSA was routinely rummaging through the e-mails of ordinary Americans. In truth, under court supervision, NSA was requesting specific e-mail accounts of designated foreign intelligence targets from the providers.

  Take this hypothetical. NSA is targeting the communications of a known terrorist in Yemen. It discovers that the Yemen-based terrorist is communicating electronically with another individual; he could be in Pakistan, elsewhere in Yemen, or even in the United States. They are using a US-hosted Internet service; they send e-mails back and forth.

  Similar contours could apply to e-mail traffic detailing the delivery of dual-use chemicals to a state suspected of developing chemical weapons. Or to other communications that provided information on potential cyber attacks.

  In each of these cases, the only thing “American” about the communication is that it is physically in the United States and being hosted by a US-based Internet company. The target of the surveillance is a foreigner outside the United States.

  Director of National Intelligence (and former director of NSA) Mike McConnell saw this and convinced Congress to amend FISA (Section 702) in 2008 to make it easier for NSA to grab these kinds of communications. Access is still overseen by the FISA Court (which also compels firms to turn over the data), but the process no longer requires time-consuming, cumbersome, individualized warrants.

  NSA treats PRISM as just another collection point, an admittedly new and particularly valuable one, but still one among many designed to acquire foreign communications of foreign intelligence value.

  PRISM didn’t light up domestic protests as much as the metadata gathering, called the 215 program, because that was the section of the Patriot Act that underpinned it. PRISM was focused on foreigners; 215 was all about Americans. NSA kept a repository of American calls—not content, but facts of calls like from whom, to whom, when, for how long. It was massive, but access was tightly controlled, not just by limiting the number of people who could touch the data, but also by limiting their purpose to only counterterrorism.

  When NSA acquired a new terrorism-related phone number overseas, say, by Yemeni forces nabbing that terrorist mentioned above, a key question was always whether or not that heretofore unknown number had ever been in contact with a US phone. It’s a simple matter to query that massive 215 database and, if such a number is identified, NSA gets to ask about the other numbers with which it is in contact. Anything after that is in the hands of the FBI.

  As reasonable and constrain
ed as that sounds to me, it sure as hell didn’t go down that way. Lots of Americans were upset, and not all of them were wearing tinfoil hats. Pockets of Congress were livid. Part of the issue was that many refused to believe that the program was constrained. They suspected that the data was being used for other purposes or that massive algorithms were being launched against it to divine personal patterns of behavior.

  Others made claims that, if true, had NSA not only violating the laws of the United States but the laws of physics as well. I even heard one commentator posit that NSA could, if it suspected something of interest, click on a number in order to go back and hear what was said during a call. Huh? Click on what in reality is a phone bill to hear what was said in the past?

  I pushed back strongly against a similar notion in a discussion on Fox News Sunday with Tea Party freshman congressman Justin Amash. I advised him that he needed to be telling his constituents the truth rather than whipping them up with scaremongering.

  AMASH: And it’s important to understand that it then goes beyond metadata. So, we start with metadata but the government is not suggesting that it can’t collect your actual communications. Under this doctrine, they certainly can collect your content just as they can collect your metadata. . . .

  HAYDEN: . . . I’ve got to add, Chris [Wallace], it doesn’t make Americans more comfortable about the program to misrepresent it. This does not authorize the collection of content, period.

  An early NSA defense of the 215 metadata program was that the executive, the Congress, and the courts were all witting and supportive. In a post–Church Committee intelligence-reform era, that should have been sufficient. It wasn’t.

  The controversy created strange alignments. Fox News abandoned its usual national security credentials for a chance to beat up the president and strengthen its populist Tea Party brand. The network couldn’t address the issue without the lede “NSA scandal,” a lede that brought with it its own conclusions.