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

by Michael V. Hayden

  I could. “I can assure you, by the physics of the intercept, by how we actually conduct our activities, that one end of these communications is always outside the United States of America.”

  It was a long thirty-two minutes. We didn’t have much success in advancing either human knowledge or national understanding of what we were trying to do.

  We kept at it. We spent Christmastime 2005 answering the phones and initiating calls to folks in the press we had talked to in the past. It was a hard sell. Some reacted viscerally. As good a reporter as John Diamond of USA Today was furious at us for what he viewed as our betrayal of his trust in earlier stories. We didn’t see it that way, but it hardly mattered.

  We also spent Christmas week trying to get our hands on Risen’s book, State of War. A junior ODNI officer finally tracked down a copy before they went up on the shelves of a local bookstore, brought it back to the office, and proceeded to xerox multiple copies for us. We pretty much knew what the Stellarwind content would be, but were willing to risk a minor copyright violation to see what other state secrets were going to be revealed. I must admit that it also felt good not to pay for it.

  After the holidays, we were tapped to brief the entire FISA Court on the Stellarwind program. Since I had set the program up, I was to be the lead briefer. I spent two long days at Fort Meade preparing.

  On January 9 we briefed the chief FISA judge, Colleen Kollar-Kotelly, and eight of her associate judges. It was a long session: four hours without a break. Attorney General Gonzales was there at the beginning to thank the court for the opportunity to brief and to introduce me—as Colonel Hayden, a misstep that provoked some laughter and actually seemed to break the ice a bit. We laid it all out. Lots of questions, many mixed with skepticism. One FISA judge, James Robertson, had already resigned in protest, stepping down a few days after the original Times story, although it appears he acted based on the news accounts rather than any fuller description. At this session, though, after about two hours the center of gravity of the discussion seemed to shift in the direction of how the FISA regime could be made more relevant to twenty-first-century threats and technology.

  When I got back to my security detail after this session, I had a note to call my sister Debby in Steubenville. My arrival at the FISA Court had apparently been reported in the media, and in our old neighborhood in Pittsburgh showing up in front of a judge was hardly ever good news. I assured her that I wasn’t appearing before that kind of court. Not yet, anyway.

  In late January, I had another shot at a public presentation at the National Press Club. I reminded folks that this was targeted and focused, “the hot pursuit of communications entering or leaving America involving someone we believe is associated with al-Qaeda. . . . It is not a drift net over Dearborn or Lackawanna or Fremont grabbing conversations that we then sort out by these alleged key-word searches or data-mining tools or other devices that so-called experts keep talking about.

  “This is not about intercepting conversations between people in the United States. When you’re talking to your daughter at state college, this program cannot intercept your conversations. And when she takes a semester abroad to complete her Arabic studies, this program will not intercept your communications.”

  Since we were at the press club, I decided a modest lecture to the press itself would be in order. “I know how hard it is to write a headline that’s accurate and short and grabbing,” I conceded. “But we really should shoot for all three—accurate, short, and grabbing. I don’t think domestic spying makes it. One end of any call targeted under this program is always outside the United States. . . . I’ve taken literally hundreds of domestic flights. I have never boarded a domestic flight in the United States of America and landed in Waziristan. In the same way . . . if NSA had intercepted al-Qaeda ops chief Khalid Sheikh Mohammed in Karachi talking to Mohamed Atta in Laurel, Maryland, in say, July of 2001—if NSA had done that, and the results had been made public, I’m convinced that the crawler on all the 7 by 24 news networks would not have been ‘NSA domestic spying.’”

  That was the only example that I used that day, but there were more available. Accusations are always simple; truth is often complicated. And as the quotation etched into the wall of the CIA lobby suggests, it is truth—not simplicity—that will make you free.

  Then at the press club I said something I believed to be true then and I continue to believe to be true now. “Had this program been in effect prior to 9/11, it is my professional judgment that we would have detected some of the 9/11 al-Qaeda operatives in the United States, and we would have identified them as such.”

  It was a long speech, but it was the clearest and most comprehensive public exposition of what NSA was doing and why that we had ever made. Of course, it was limited to that part of the Stellarwind program that the president had confirmed, intercepting the content of one-end US international phone calls. I didn’t announce that there were other activities that I couldn’t talk about, but I did caution that “this is a little hard to do while protecting our country’s intelligence sources and methods. And, frankly, people in my line of work generally don’t like to talk about what they’ve done until it becomes a subject on the History Channel.”

  Nonetheless, I continued: “I much prefer being here with you today telling you about the things we have done when there hasn’t been an attack on the homeland. This is a far easier presentation to make than the ones I had to give four years ago telling audiences like you what we hadn’t done in the days and months leading up to the tragic events of September 11.”

  I took questions, a measure of modest courage, since the press club definition of “press” is rather broad and C-SPAN was streaming all this live. It wasn’t quite the bar scene on Tatooine in the first Star Wars movie, but it had its moments. Some questions were based on genuine efforts to understand a complex topic (like the details of an emergency FISA warrant). Others were unarguably combative (“Are you asserting inherent so-called constitutional powers . . . to violate the law when [the president] deems fit?”). Still others were thinly veiled infomercials for specific groups (like something called The World Can’t Wait!: The Call to Drive Out the Bush Regime).

  As near as we could tell, the session did not create any change in the tone (or from our view, the accuracy) of the press coverage.

  Two weeks later Jeanine and I were in Detroit to see our Steelers play the Seahawks in Super Bowl XL as guests of my Catholic junior high football coach, Dan Rooney, the owner of the team. I think Dan was taking a little pity on me, what with all the publicity. In any event it was great to see a win from the owner’s box even if I had to do two Sunday talk shows that morning. Near the end of the game, with the Steelers up 21–10, Dan apologized that he had to leave the box early to go down to the field to accept the Lombardi Trophy.

  “Dan,” I assured him, “this is a good thing.”

  There had been so many Steelers fans at the game that there was a long backup at the tollgate between the Ohio and Pennsylvania turnpikes the next morning. No one seemed to complain, though.

  I flew back to Washington that Monday morning, too, and basically reattached myself to my deputy DNI job. We had made our case as best we could. No need to keep feeding the media beast.

  We just held the line like the president said we were going to. We kept on doing what we were doing on the grounds that it was proportional, lawful, effective, targeted, had not been abused, and was consistent with the US Constitution.

  When the public debate subsided (a little), the White House and Justice continued to press for legislation to backstop what the president had authorized to create a political and legal safe harbor for the executive. It took time, but they got their way. In a series of steps from 2006 through 2008—renewal of the Patriot Act, the Protect America Act, the FISA Amendment Act—almost all that President Bush had authorized via Stellarwind was codified in law. And—despite the earlier journ
alistic outrage—a lot more to boot. Go figure.

  SEVEN

  THE PUBLIC’S RIGHT TO KNOW . . . AND BE SAFE

  FORT MEADE, MD, AND LANGLEY, VA, 1999–2009

  The Stellarwind story was symptomatic. It was but one example of growing administration-press tensions, or more accurately, it was an expression of the continuous tension between transparency and secrecy in a free society.

  I was accustomed to my public affairs officer running into my office at either CIA or NSA and suggesting or requesting or demanding that I call an editor or publisher to scotch a particularly egregious story. When I agreed to do that, I would invariably begin the conversation with, “Thanks for taking my call. I know that we both have a job to do protecting American security and American liberty. But I’m afraid that how you are about to do yours is going to make it more difficult for me to do mine.”

  To be fair, those calls I made to slow, scotch, or amend a pending story were worth making. Many on the other end of the line were open to reasonable arguments. In one case a writer willingly changed a reference that had read “based on intercepts” to “based on intelligence reports,” somewhat amazed that this change made much of a difference. (It did.)

  Another reporter, quite experienced, casually and quite responsibly volunteered to my press office, “I know this is SIGINT, but I don’t need that for my story.”

  The Washington Post’s Dana Priest, not one with high favorables at Langley, did not publish the locations of alleged CIA detention sites in her Pulitzer-winning series in late 2005. The agency argued that such a revelation would put the citizens of these countries at increased risk from terrorists. She agreed.

  But sensitivity to the national welfare was not a universal condition, and I wasn’t alone in this assessment. In January 2008 twelve senior leaders of the intelligence community (including me) signed a letter to the leadership of the Senate opposing a pending journalist shield law, which would have put even more procedural steps in the process we had to follow to track down and prosecute leakers. We said that it would “undermine our ability to protect intelligence sources and methods.”

  None of us were insensitive to the principles of the First Amendment, to the role of the press in our democracy, or to the delicate balance and inherent tension between security and openness. We just thought that some things were legitimate secrets. I laid out that case at an informal, off-the-record gathering of journalists in October 2006. I was admittedly a little pissy after several sensitive NSA and CIA programs had become press fodder.

  I began with the premise that there exist in the world things that are legitimate secrets: this is true for the family, the PTA, the Lions Club, and the press. Society understands and recognizes this need and has established rules for when secrets will be recognized and afforded a privilege and when they will not. We recognize the marital, the attorney-client, and the priest-penitent privilege, and rules have been developed for when these privileges are overcome because of the greater community need.

  Like other communal entities, governments, too, need secrets. Within the federal government, all three branches claim this right. The deliberations of the courts are secret. Congressional markup sessions are done in private. The Federal Reserve keeps secret its discussions about the overnight discount rate so as not to affect markets. Crop support levels are kept secret until a certain date and time set by the Department of Agriculture.

  The position is actually unassailable. In November 2008 New York Times correspondent David Rohde was kidnapped by the Taliban in Afghanistan and taken by the Haqqani network to Pakistan. Everybody knew. But none of this appeared in the press until Rohde escaped. None of this: not the fact that he had been kidnapped, not what the Times was doing, not what I was doing diverting CIA-controlled resources to look for him (a fact about which I quietly informed the Times). The Times thought that Rohde’s plight was a legitimate secret. So they exist! Even the press seems to concede that there are necessary secrets (including especially its own sources, of course).

  But that wasn’t always the rhetoric, so I was challenging a public talking point of the press people gathered with me in 2006 when I said, “So let me set up a contrary position to what the press speaks of as the ‘public’s right to know.’”

  The “right to know” is far from an absolute. In fact, in some ways the public has already decided what it does and does not want to know. The president and other members of the executive branch have been authorized by legislation to classify information whose release, in their collective judgment, would damage national security. The Supreme Court has repeatedly recognized this authority.

  The press kind of recognizes this right. According to David Broder of the Washington Post, “It’s the government’s responsibility to keep its secrets.” But then he goes on to say, “And it’s our responsibility to ferret out information so the public is aware of the actions being taken in its name.” In other words, the government, under the authority of the public’s lawfully elected representatives, has the right to keep secrets, but they should remain secret only unless and until someone leaks them to the press. You don’t have to drive the premise to the extreme before this begins to border on the incoherent.

  Of course, the counterpoint is that the press would never divulge legitimate secrets, and (an important corollary) the press has to be the final arbiter of what is and is not a legitimate secret. Even assuming goodwill and setting aside the question of legal authority, there remains the issue of competency. David Ignatius, no apologist for intelligence community secrecy, summarizes it nicely: “We journalists usually try to argue that we have carefully weighed the pros and cons and believe that the public benefit of disclosure outweighs any potential harm. The problem is that we aren’t fully qualified to make those judgments.”

  I laid into the journalists at my 2006 off-the-record gathering with a catalogue of specific harm done by leaked secrets. I had leaned hard on my own ops folks to get my talking points cleared for the discussion.

  I told the journalists that one recent spate of stories cost us five promising counterterrorism and counterproliferation assets, who feared we couldn’t guarantee their security, and most of them weren’t even reporting on the same subject as the stories. In August 2002 a sensitive clandestine source actually saw his reporting disclosed on TV during a meeting with his case officer. When a covert CIA presence in a denied area was revealed in the media, two assets in the area were detained and executed. Our officers there wrote: “Regret that we cannot address this loss of life with the person who decided to leak our mission to the newspapers.”

  Another leak, on weapons of mass destruction, led to one of our assets being arrested by a foreign government on suspicion of espionage; details in a follow-up story convinced the government that they had the right man and he was convicted and sentenced; a third story caused his family’s property to be seized, and we had to move them out of the country to protect them.

  I concluded by pointing out that several years prior to the 9/11 attacks, one chief of station reported that a press leak of liaison intelligence had “put us out of the bin Laden reporting business” locally. The service ceased counterterrorism cooperation for two years.

  I wasn’t done. Beyond possible (I was being kind) errors in judgment, there was a question of journalistic slant. Everyone seemed anxious to double down on the prevailing story line. Brian Ross (whom I have since gotten to know and respect) led off the ABC nightly news during the Stellarwind surveillance kerfuffle, waving an NSA memo at the camera and claiming that “whistle-blower” Russ Tice had been warned not to talk to the House Intelligence Committee about some activities because the committee was not cleared for the material. Having so much trouble containing his outrage, Ross was apparently unable to discover that Tice—who was cleared for and talking about DOD rather than NSA special access programs—needed to talk to the Armed Services Committees, rather than the intelligence
committees, about his allegations.

  I then asked the assembled journalists some rhetorical questions. Did any of you report that Tice did indeed finally appear before congressional staff? How did that go? Did you report the results of those sessions? Any congressional outrage to tell us about? (None that I know of.) Maybe a good question to ask would have been how long Tice had been at NSA (not long; DIA kind of dumped him on us) and then follow up with how long he was at NSA and cleared (even shorter; he spent most of his time on the loading dock). Actually kind of important stuff, but what the country got was Ross and Tice ominously walking out of the shadows (literally) on prime-time news.

  The New York Times reprised this genre of drama-above-all reportage in an only slightly different format in June 2006, when Eric Lichtblau and James Risen (coauthors of the NSA stories) reported that the US government had access to the financial transaction database of SWIFT (Society for Worldwide Interbank Financial Telecommunication). The government and the intelligence community fought against the disclosure, since we knew it would harm our counterterrorism activities. Following the money was an incredibly useful tool.

  The Times later argued that its story was protected by the First Amendment (unarguably true) and provided “information the public needs to make things right again” (a very arguable point). The Times trotted out the old saw that terrorists obviously knew that this was going on, and then vouched for their own patriotism by claiming that this bore “no resemblance to security breaches, like disclosure of locations, that would clearly compromise the immediate safety of specific individuals.”

  Lichtblau’s personal defense was that this was “above all else an interesting yarn about the administration’s extraordinary efforts since 9/11 to stop another attack.” Interesting yarn?

  In October 2006—months after the original article—the public editor of the Times reversed his earlier position. He said that it still was a close call but now declared that, in his view, the story should not have been published. Wow! That was a relief.