Managing Oversight

Congressional oversight of intelligence is broken. That was the dismal conclusion of a 2006 report by the Center for American Progress, as well as the bipartisan 9/11 commission, both of which scoured the histories of congressional watchdogging -- real and imagined -- and concluded that the system set up to guard against abuses and keep intelligence in line with U.S. policies was not working the way it was intended.

What was the problem? It boiled down to this, the center's report said: Congress had all the tools it needed to conduct real oversight, but wasn't using them. That's because the committees themselves fractured along party lines. From the mid-1970s to the late 1990s, congressional oversight was, if not entirely consistent, consistently bipartisan. There were notable exceptions, but generally, members believed they had a special role to play in foreign policy, and they made an effort to become more informed about intelligence, both in terms of specific operations and the discipline as a whole.

That all went sideways with the past decade's rise in partisan rancor. No one will be surprised to know it has infected the intelligence committees. The latest evidence was on display in May when House Speaker Nancy Pelosi accused the CIA of having lied to her about interrogation of terrorist suspects.

Democrats rallied around the speaker, Republicans rallied around the legacy of George W. Bush. And rather than argue about how well they'd actually managed their oversight responsibilities, the lawmakers questioned each other's honesty, integrity and willingness to defend the nation from evil.

As the committees have become more partisan, they've also become less informed about intelligence. The center's report cites a number of historians and former staffers who regret that lawmakers spend much of their time bickering over authorizations for individual programs, using the power of the purse for tactical political advantage. That has left less time to devote to the overall mission of the intelligence community and to strategy. Back in the 1980s, intelligence committees acted as a sounding board for intelligence operations, a kind of council of political elders who were concerned about how agencies carried out their missions. That required a level of trust among the members, and while it was never ironclad, it has dissolved.

Oversight today bears a striking resemblance to the 1950s. In the CIA's early days, there were no intelligence committees per se. A small number of senators and congressmen on the appropriations and armed services committees formed an exclusive club, and for two decades, their oversight was known more for its deference to the CIA director than its skepticism of him. Once, in 1973, Senate Armed Services Committee Chairman John Stennis told CIA Director James Schlesinger, who wanted to brief him on an upcoming operation, "Just go ahead and do it, but I don't want to know."

What most historians recognize as meaningful oversight began in the mid-1970s, after revelations of foreign policy fiascoes and domestic intelligence abuses. That seems to be the model that many would like to recapture. Their prescription is this: stronger management. The center's report urged committee staff to conduct more investigations and field studies, and to communicate more with agency personnel. In other words, they should do the job they're allowed, and required, to do by law. The 9/11 commission offered a different, hugely controversial recommendation: Form a single joint committee, and eliminate term limits on membership so Congress could develop a seasoned, knowledgeable base of overseers.

The commission was less than optimistic about the chances of that happening, noting, "Few things are more difficult to change in Washington than congressional committee jurisdiction and prerogatives." Recent events remind us how true that is.

Shane Harris, a staff correspondent for National Journal, wrote about intelligence and technology at Government Executive for five years.