War Stories

Dismantle the CIA?

What’s behind Sen. Roberts’ awful intelligence-reform proposal?

You’d think there must be something good about a CIA-reform proposal that’s denounced by George Tenet and unnamed White House officials. But no, Sen. Pat Roberts’ plan to overhaul the U.S. intelligence bureaucracy is a true stinker, every bit as bad as his establishment critics contend.

Roberts, a Kansas Republican, is chairman of the Senate Intelligence Committee, so his pronouncements on such matters can’t be casually dismissed. What, then, was he up to when he unveiled his scheme on last Sunday’s Face the Nation?

Anyone with the slightest insight into the workings of Washington would know that his proposal has no chance of passage. Anyone who studies the “intelligence community” as much as Roberts does would also know—or should—that the proposal, if it were put into effect, would do more harm than good. So again, what’s going on here?

Roberts called for creating a “national intelligence director” who would have “complete budget and personnel authority” over intelligence agencies throughout the federal government. This is pretty much what the 9/11 commission called for—and it’s what President Bush has given only lip service to. (He’s agreed to appoint someone with the title of national intelligence director but with none of the authority—in short, a glorified paper-pusher.)

But then Roberts’ plan goes about six bridges too far. He would dismantle the CIA altogether. He would turn its clandestine shop (the directorate of operations) and its analytical shop (the directorate of intelligence) into separate, distinct agencies and place those agencies under the new national intelligence director. He would also strip the Pentagon of its current control over the Defense Intelligence Agency and the National Security Agency. And he would put the new national director in charge of intelligence units now working at the departments of State, Energy, Treasury, and Homeland Security.

Clearly, intelligence agencies are scattered too widely across the bureaucracy. The long list of units to be absorbed by the national director attests to this fact. However, placing them all under a single, centralized entity is not the answer; it will not make for better, or even better-coordinated, intelligence.

Let’s look at what Roberts would do to the CIA. One problem with the current arrangement is that the spymasters in the directorate of operations and the analysts in the directorate of intelligence almost never talk with each other. The director of the CIA, who lords over them, could encourage or force them to work together, but he rarely does. Why? Because the director is but a temp at Langley; the heads of DO and DI—and their upper-echelon subordinates—are insiders, veterans, pros, and they value their quasi-independent status. The director has to work with them, can’t afford to alienate them, and so steps back from confronting them. (Flynt Leverett, a former CIA senior analyst, told me that he has watched three CIA directors “flinch” from the prospect of trying to bring the two directorates together.)

Obviously, something has to be done about this situation; someone has to figure out a way to commingle these two branches without violating their morale and mores. One clear way not to do this is to chop them completely and formally apart. If the head of the CIA has a hard time coordinating the spies and the analysts when they’re all in the same agency and working in the same building, how is some überhead going to do any better after the two branches have been split into autonomous agencies and he’s sitting across town, simultaneously trying to manage a dozen other headaches?

There are other problems. It is an extremely bad idea, for instance, to lasso all intelligence analysts under one roof. Separate and competing analytical teams make for more open, creative, better analysis—and, therefore, wiser, more informed policy-making. (This assumes, of course, that the policy-makers haven’t made up their minds ahead of time and then pressured the analysts into coming up with the “right” conclusions.)

Some of the resistance to Roberts’ proposals—and to those of the 9/11 commission—stems from parochial interests. For instance, the secretary of defense currently controls three-quarters of the U.S. intelligence budget (in other words, about $30 billion, to say nothing of the enormous power that goes with it), and he doesn’t want to give it up. However, it still isn’t a good idea to separate the Pentagon from units involved in tactical military intelligence.

The various panels investigating the failures leading up to 9/11—including Roberts’ committee—reached several conclusions in common. They found too many “brick walls” between and within agencies, too many “stovepipes,” too much “groupthink.” Roberts’ plan would thicken those walls, further insulate the stovepipes, and intensify the group-think.

What’s his motivation here? There are two possibilities.

The first is that he’s advancing a deliberately extreme proposal in order to prod the stuffy, stodgy bureaucracy into moving. He’s telling the White House that if Bush doesn’t start making serious reforms, Congress will—possibly in ways that the executive branch won’t like. And he’s shifting the definition of “acceptable” reform: By proposing a plan that goes well beyond the 9/11 commission’s proposals, he is making those commission proposals seem more moderate by comparison.

Some of Roberts’ comments on Face the Nation support this theory. “My worry,” he said at one point, “is that … the administration comes out and does not go far enough in response to the 9/11 commission … And then we’ll fuss about it. And then the [election] campaign will take over and we won’t get anything done. Now I have said time and time again that we have an urgent need to move, but we have to get it right … [This proposal] is at least a marker that we can start the debate … Let’s get it out on the table and let’s talk about it.”

However, there is a second, more cynical, and, alas, more plausible theory: He’s putting out a proposal that’s deliberately out-to-lunch, in order to distract the debate from more reasonable resolutions, to deflect attacks on Bush, and to discourage the whole idea of organizational reform.

Sen. Carl Levin, a Michigan Democrat and another member of the Senate Intelligence Committee, appeared with Roberts on Face the Nation, and it was clear—both senators said so explicitly—that he knew nothing of Roberts’ proposal ahead of time. When the host, Bob Schieffer, asked Levin for his reaction, he replied, “I think it’d be better to start on a bipartisan basis with a bipartisan bill … And frankly, I hope we’ll start with that as soon as we get back and not fool around with other political gestures to spend weeks in September, when we should be focusing on reform of the intelligence committee.”

Schieffer broke in: “Well, you’re not saying Sen. Roberts has put out a political gesture.”

Levin replied, “No, what I’m saying is that it is not a bipartisan bill. …”

But actually, what Levin was saying is that it is a political gesture. And he has a point. The Senate Intelligence Committee generally operates as a bipartisan panel, at least by the standards of most congressional panels. As recently as July, Roberts had no problem sharing the podium with Sen. John D. Rockefeller IV, the ranking Democrat, when their committee released its own report on the failures leading up to 9/11.

If Roberts were serious about putting something out on the table, so we can get “real reform” before the election, he could easily have worked out a plan with his colleagues. Surely Roberts knew the administration would oppose his idea. (Dick Cheney and Donald Rumsfeld have spoken out against similar but far milder suggestions by the 9/11 commission.) By coming out solo, Roberts guaranteed that the Democrats on his committee would oppose it, too. (Rockefeller has since said that he, too, knew nothing about the proposal until he heard about it on television and that in any case, it’s unwise: “Disbanding and scattering the Central Intelligence Agency at such a crucial time,” he said, “would be a severe mistake.”)

In short, Roberts gets a twofer: He draws attention away from Bush’s refusal to enact serious reforms—and he creates a situation in which the Democrats appear to be the foot-draggers. Crafty.