Playing to the Edge: American Intelligence in the Age of Terror (2 page)

I
once described NSA as “a big ship with a small rudder.” I could have added, “With a temporary captain and a temporary navigator.”

I never expected to be DIRNSA, the director of the National Security Agency. I was in Korea in 1998, serving as the deputy chief of staff for US and UN forces there following command at the Air Intelligence Agency in Texas. The Korea post was not an intelligence job. As deputy chief of staff, I kept things running smoothly for the American commander in Seoul, stayed in touch with our Korean allies, and (in my UN role) negotiated with the North Koreans at the truce village of Panmunjom.

Such was my life when my wife interrupted my shower on a bright fall morning in Seoul to tell me that Mike Ryan, chief of staff of the air force, was on the phone.

“Mikey,” he began, using a form of address he had developed for me a few years earlier when I headed up US intelligence in Europe and he was running the air war over the Balkans.

“Mikey, we’re going to nominate you to be the director of NSA. You’re not going to get it, but we’ll bring you back, you can make the rounds,
and then you’ll be well positioned when the director of DIA [Defense Intelligence Agency] position comes open next summer. That’s the one we’re really shooting for.”

Not exactly a ringing vote of confidence, but at least the chief was interested in my career, and I was happy enough to get a trip home. Actually it was a circumnavigation. I flew to Hawaii for a promotion board, continued to Washington for the job interview, and then went on to Geneva as the Korea Command representative to the four-party talks between China, the United States, and the two Koreas before I returned to Seoul.

I had only one interview for the DIRNSA job. It was with George Tenet at the Wye River Plantation on the Eastern Shore of Maryland as Tenet was shuttling back and forth between Yasser Arafat and Bibi Netanyahu trying to broker a peace deal in the Middle East.

I was picked up from Bolling Air Force Base, where I was staying, and climbed into the backseat of one of those large black Suburbans that seem to populate most spy movies. We crossed the Chesapeake Bay Bridge, and the security detail deposited me in one of the Wye Plantation’s spare lodges, where I awaited the DCI and fought off jet lag.

George came in about an hour later, all energy, seemingly happy to talk about something other than Israel and Palestine. I had not met him before, but the conversation was relaxed. We talked about intelligence in general and a bit about NSA in particular; since I had never served there, I relied a lot on what I had learned at the air force component of NSA, the Air Intelligence Agency, which I had commanded in Texas in 1996–1997.

About a month later, now back in Korea, I learned that I was going to be the next director of NSA. I doubt that I had overwhelmed George, but his new deputy was air force general John Gordon, and John and I had served in the same office on Bush 41’s National Security Council staff. Stuff like that happens in Washington.

Before my appointment was publicly announced, my wife and I took in a Friday night movie at the Yongsan army garrison base theater. It was
packed with GIs for the popular thriller
Enemy of the State
, in which Will Smith battles an omniscient and malevolent National Security Agency. As the plot twists and turns, a career NSA civilian aspiring to the deputy director position actually kills a US senator. I was sinking into my seat as my wife whispered in my ear, “What did
you
have to do for this job?”

I was sworn in for the NSA job in late March 1999. The day after the ceremony, my new agency put me into a reclining chair, hooked up sensors to my body, and asked a series of questions about my trustworthiness. All of NSA’s people have to be polygraphed. It’s the cost of admission. And if Snuffy has to do it, so does the new director. Apparently, I passed.

I arrived during NATO’s bombing campaign in Kosovo. The agency was providing solid direct support for the combat operations, as it had done in the Gulf War eight years earlier. Nothing surprising there. NSA was a national treasure. Since its founding by President Truman in 1952, it had been tasked, broadly speaking, with intercepting communications that contain information that would help keep Americans free and safe and advance our country’s vital national security interests. That is mostly the communications of adversaries, but it could also include the communications of others, not protected by the Constitution’s Fourth Amendment (which is not, after all, an international treaty), who happen to be talking about things we need to know. Picture the foreign travel agent innocently passing on the travel schedule of a terrorist or an arms trafficker.

NSA’s closest counterpart is its best friend, Britain’s GCHQ, which is only about one-fifth its size.

By 1999, though, NSA had a problem, as we have seen. It had grown up in an era in which it was the lead for a host of computer and telecommunications breakthroughs. But the outside world had passed it by in many areas. John Millis, who headed up the House staff overseeing the agency, commented that “technology has been the friend of the NSA, but in the last four or five years technology has moved from being the friend to being the enemy.” The cultural habits built up in a world where the trouble with
SIGINT was that there was too little, and it was too hard to get, were now counterproductive in a world where the trouble was in the category of too much, too hard to understand. NSA was appropriately obsessive about secrecy, but the gates and high walls that provided twentieth-century security cut the agency off from innovation as it approached the millennium.

Then there was the question of the director, described as “Christmas help” by one senior
*
(and he was a friend). Even in a public setting it was not unusual for NSA leaders to refer to the occupant of the eighth-floor suite as “the
current
director.” It was a challenge for a career military officer to sit atop a pile of civil servants who knew that they were right and everyone else was wrong, and who were fully capable of and well practiced at waiting out change that they opposed. That description didn’t apply to everyone, but it applied to enough.

Former director Bobby Ray Inman is a revered figure at NSA. (I once compared his status in retirement to that of the reverence shown to the off-camera Randolph Scott in Mel Brooks’s
Blazing Saddles
.) Early on, I made a pilgrimage to Austin to seek Inman’s advice. He warned me about the isolation of the eighth-floor director’s office. “They’ll want to put you in a sedan chair, carry you around like Pharaoh, and keep you as far away from decisions as possible.”

It’s an opaque place too. Rather than calling an office, say, the Balkans collection division, that shop is given an impenetrable alphanumeric designator. NSA employees routinely answer the phone with the last four digits of the office number: “3685. Can I help you?” That’s all good operational security, but it makes the place hard to understand, even from the inside. One officer later told me that I tried to penetrate the workforce like the agency tried to penetrate a target. I spent one of my first Saturdays just walking through the empty halls trying to decipher office symbols and locations.

And the agency had a pretty fixed culture. A veteran compared the prevailing views to that of Tevye in
Fiddler on the Roof
: “Tradition!” In the face of unprecedented change and challenge, tradition was comforting, tradition was stabilizing.

But outside observers believed that NSA was a burning platform. A group called the SSCI TAG (the Senate Intelligence Committee’s technical advisory group) was composed of high-tech gurus who warned that only about ten years of global technological advance separated the agency from operational deafness, and the clock was running.

For a signals intelligence agency, we had surprisingly antiquated IT systems, both for ourselves and to target our adversaries. Shortly after I arrived, I asked, “How do I send an e-mail to everybody?”

“Oh, we can’t actually do that,” was the response.

“But didn’t we kinda invent the computer here?”

On the mission side, according to one estimate, 70 percent to 80 percent of reporting was still coming off traditional voice intercepts. But even here we weren’t keeping up. Al-Qaeda and other targets were migrating to a new form of satellite communication. Chasing the signal required an infrastructure investment that NSA refused to make, so CIA did it on its own. When that agency briefed me on its program, I asked, “What measure of NSA sin would have caused you guys to build your own SIGINT system?”

And this was in a world in which digital communications—e-mails and the like—were also exploding. NSA largely went after non-voice data only when it was encrypted (suggesting its importance) and even then largely in fax and telex rather than e-mails.

Prior to this, e-mails had been generally limited, dedicated, point-to-point communications that came to a program loaded on your machine. Without the appropriate software, you could neither read nor send a message. We were now on the cusp of Web-based e-mail systems that promised far greater convenience and thus far greater volumes. Toying with the old point-to-point e-mails had us shooting way behind the target.

Trying to escape the tyranny of expertise (aka the existing organizational structure), I sought opinions from an outside group of experts and a group of young Turks from within the agency. They gave me a variety of recommendations and observations, the most telling of which was that the agency was less than the sum of its parts (many of which, they emphasized, were really quite good).

The outside gang went out of its way to be generous, commenting on NSA’s world-class competencies, its critical role, its response to crises, and the agency’s deep commitment to the institution and its people. It then catalogued poor systems for communication, decision making, financial and personnel management, requirements development, and business processes. It predicted rapid technological obsolescence, the product of an inward-looking culture and a leadership team that many (including some of the leadership team) believed was not up to the challenge.

The inside group, the “New Enterprise Team,” was equally scathing. They began their report with a quote from Darwin on the relationship between adaptability and survival. Then they hammered the institution and its leadership, saying that NSA’s customers had already begun to separate NSA’s products and services (which they still treasured) from NSA the institution, which they now viewed as a threat to the continued availability of those products and services. The workforce, they said, had been carrying the agency on its back for a decade.

They agreed that we lacked leadership, a strategic plan, a decision-making process, and the ability to manage our resources. What we did have, they pointed out, was a lot of committees. One of their “quick hits” was to eliminate all working groups and committees where a single individual could make decisions. I eliminated all of them by fiat and then challenged chairmen to petition for reinstatement if they still thought the panel was essential. When we were done, we had cut over a third of an amazing total of more than 450
internal
panels.

Next was the inevitable reorganization; it’s what government agencies do, after all. And there’s a reason for that. Given cumbersome restrictions on budgets and personnel, there really aren’t a lot of other tools available.

I tried to simplify things on the organizational chart, identifying two large operational boxes, one for offense (SID, the SIGINT Directorate) and one for defense (IAD, the Information Assurance Directorate). We pulled enabling functions—things like research, IT, training, logistics, human resources—out of the operating units and moved them to direct reports under me. Flattening the organization that way gave me a challenging span of control, but also let me directly use these agency-wide support activities as tools for change.

One office that was critical to get to my level was the Foreign Affairs Directorate. We had lots of partners, but for all but a few of them, relationships were managed at a fairly low operational level and decisions about them were largely based on whether or not the affected NSA office felt they needed any summer help. I wanted these ties based on strategic considerations, not just on who could or could not cover discrete targets, languages, or radio frequencies.

All the reorganization helped, but the real plus was the chance to put new leaders into a bunch of the newly created posts. We reached deep within the ranks, betting that most of these promising rookies would learn to hit a major-league curve ball soon enough. Most did.

We institutionalized that approach, setting up a one-man senior personnel shop just down the hall from my office. Federal bureaucracies are dedicated to
process
in promotions and assignments, committee upon committee deciding who advances, when, and to what job. That’s designed to reduce favoritism (and lawsuits) and to enhance stability. We didn’t want stability; we wanted disruption, and I was hell-bent to favor some over others.

We eased out a whole generation of leadership and skipped over a big chunk of their obvious successors. These were good people. They didn’t deserve this. At the time, I thought of air force general Curtis LeMay’s legendary ruthlessness when it came to shortcomings. “I can’t tell the difference between unlucky and unskilled because the results are the same,” he once said. I knew that these people were quite skilled, just unlucky. It was a tribute to the agency and to these individuals that they handled
this with grace, even though their only offense was being at the wrong place at the wrong time.

My most disruptive choice was my new deputy director. Barbara McNamara was in the position when I arrived and stayed on until June 2000. Barbara was career NSA and loved the agency with every inch of her being. She had started as a linguist and worked her way up the ranks. She was tough as nails and took nothing from her male counterparts. Her middle initial was “A,” and her signing off on countless papers over the years caused everyone to address her simply as BAM.