espite the hysterical shuffling of chairs formally known as instituting Homeland Security, the sprawling intelligence-gathering apparatus that existed pre-9/11 likely will survive the restructuring largely unscathed. This does not bode well for averting future attacks. With the existence of terrorist "sleeper cells" in the American heartland apparently confirmed and the color-coded alert system ratcheted up recently to the second-highest level, the need for accurate, well-analyzed information has never been higher, but confidence in the quality of intelligence work never lower. By all appearances, the spy bureaucracy, now comprising fourteen agencies and about 100,000 employees--much of which will remain intact even with a new Department of Homeland Security--practically begs for the same kind of housecleaning under discussion at the moment in other establishment pillars, from the FBI to the INS.

In late September a joint Congressional committee was generating daily headlines as it revealed a pattern of pre-9/11 intelligence failures and as sentiment belatedly coalesced around the need for an independent commission of inquiry. But it is one thing to dance vigorously in the glare of a harsh spotlight. It is another thing entirely to do the hard, unheralded and ultimately more meaningful work of policing and directing the intelligence community over the long haul. Even in recent days, random anecdotal evidence mounted that Congressional intelligence oversight bears too many markings of a rubber-stamping politburo. Two examples: the panel's inability to refer by name to a key 9/11 planner, though his identity could be found on the front page of the New York Times, and the Administration's determination to keep Colin Powell and Donald Rumsfeld from even having to comment on the general quality of intelligence they receive.

Congressional impotence was seldom clearer than early in the summer, when the White House accused the intelligence committees of leaking sensitive information. Panicky members rushed to invite the FBI to investigate and even polygraph them, raising the question: Who's investigating whom here? In the months prior to the recent dramatics, the prevailing winds were not those of reform but of fear--and less fear of terrorists than of the forces that wish to suppress information. Even most committee members known as comparatively reform-minded on other topics declined to respond to interview requests for this article.

One who did agree was Richard Shelby, vice chair of the Senate Intelligence Committee, a moderately conservative Republican from Alabama who is one of the panel's few open critics of the intelligence community. In a long chat in his suite at the Hart Office Building during the summer before the public hearings had been scheduled, Shelby discussed the challenges of dealing with a closed subject in an open society. At one point, he mentioned having just received a handwritten letter from Zacarias Moussaoui. The Moroccan-born Frenchman, who is awaiting trial in connection with the 9/11 attacks, was requesting permission to address Congress, saying he wanted to talk about an FBI cover-up. Whether Moussaoui is crazy--or crazy like a fox--what he has to say seems worth hearing, so I asked Shelby if I could get a copy of the letter. Shelby consulted an aide, and concluded that he was able to accommodate me because the document had not been classified--yet. Absent any evidence that the letter was a coded instruction to sleeper cells from a man who was already in jail on September 11, 2001, it seemed an unlikely threat to national security. But in today's climate, everything is sensitive, everything quickly gets locked away from public view.

Apart from the intelligence agencies, only Shelby and his colleagues get to see at least some of these documents, and to ask tough questions. And Shelby himself admitted they're doing a lousy job of it. Decrying excessive "coziness" between committee members and those on whom they're supposed to ride herd, he had lamented, "Some people on the committee don't want us to have any public meetings."

With the convening of the open hearings, these antidemocratic tendencies have been momentarily neutralized. But history provides little assurance that Congress is about to transform itself into a body determined to force substantive changes on the intelligence community. The reality is that the Hill's would-be watchdogs face crushing liabilities, from poor information to information overload, from compromised members to compromised staff, from ignorance to inefficiency, from being bamboozled by the spy agencies to being intimidated by the White House, to--worst thing of all--willful myopia.

Since Congress first instituted oversight in the 1970s in response to egregious intelligence abuses (including a role in the violent overthrow of Chile's left-wing President Salvador Allende), vigilance has steadily declined--periodic fits of enthusiasm notwithstanding. Today there are fewer open meetings than a decade ago, and fewer independent witnesses are brought in to testify. The committees, with their aura of power and exclusivity, have become so popular that senators and representatives fight for appointments. But with seniority or pull rather than skill determining membership, the committees are full of people who either don't inspire confidence or who have reasons to fear the investigative capabilities of the agencies they're supposed to scrutinize. Thus, the Senate Intelligence Committee's Ron Wyden, who couldn't locate Bosnia on a globe during his first Senate race in 1996--after serving in the House for fifteen years--or name Canada's leader. And lame duck Gary Condit, with Chandra Levy's disappearance hanging over his head, still sits on the House Intelligence Subcommittee on Terrorism and Homeland Security.

ven the savviest members need years before they've mastered the skills to challenge reticent intelligence officials. "First you've got to know what to ask for," says Shelby, "and second, to ask the right thing at the right time of the right person." Eight-year committee term limits, which sounded good when originally proposed as a way of preventing members from building power bases, have ended up weakening the effectiveness and independence of the committees. "About the time you are really into these programs, you're off the committee," says Shelby, who is leaving soon, at the peak of his knowledge and willingness to criticize the agencies.

"The intelligence officers' testimony is highly sophisticated," notes a former panel chairman. "You have to be alert and suspicious to penetrate. They snow you with fancy jargon. They jack you around. It is not unlike pre-emption with any department, except that these guys are very good at it." They have also already checked with the White House to be certain that nothing they say will cause political headaches. "Intelligence should drive policy," notes Lee Hamilton, a former House Intelligence Committee chairman, "but often it is the other way around. Policy drives intelligence." Instead of independent analysis, oversight committees get information that Hamilton says has been "distilled" for Congressional consumption. Agency officials are also good at dangling carrots before committee members, he says. "They are awful nice to them, invite them to the CIA, give them a nice dinner, court them, seduce them." Meanwhile, spy agency personnel view committee members as political animals eager to drag their analysts into a debate, hoping to use "pure" intelligence to beat up or to support the White House. As one intelligence officer told the author of a 1997 report on the relationship between the agencies and the lawmakers, "It's bad enough that policy-makers get this stuff and run with it. Can you imagine what would happen if we gave it to Congress?"

Very likely, nothing at all. The same 1997 report concluded that most intelligence provided to lawmakers is read neither by staff nor by members, who apparently have more important demands on their time. One Congressional staff member told the author of the report, "I cannot, in good conscience, recommend to my Member that it is worth his time to come in here and read this stuff. Frankly, it is not even worth my time."

It turns out that the whole vast enterprise of intelligence oversight is balanced on the head of a pin. According to a 1992 report from a senior Senate Intelligence Committee staffer, the multibillion-dollar budget of the nation's intelligence agencies is reviewed by a Hill staff of about a dozen people. "They are dependent on the agencies for information and do not have the means for independent confirmation of that information," says Steve Aftergood, director of the Federation of American Scientists' Project on Government Secrecy.

Applying the principle of "it takes one to know one," members of oversight committees turn almost exclusively to insiders to fill staff posts. It's such a revolving door that there's almost no incentive for anyone to risk alienating anyone. When I tried to track the whereabouts of a former staff director whom one Congressman recalled as being of refreshingly independent spirit, I found him working at the CIA (he quickly hung up on me). Just how convoluted does this game of musical chairs get? The House intelligence panel is chaired by Porter Goss, a former CIA officer, who is charged with "overseeing" a CIA director, George Tenet, who is himself a former Senate Intelligence Committee staff director. While it might be advisable to have an independent-minded committee member who knows the CIA inside and out, Goss is an unrelenting cheerleader. One insider neatly sums it up: "Porter Goss goes beyond support to protection."

Rob Simmons is the intelligence community's trifecta. A former CIA officer and a former staff director for the Senate Intelligence Committee, he now represents Connecticut's second district in Congress. Though not yet on any of the intelligence committees, the first-termer has been permitted by Goss to read classified materials and obtain briefings. Some who have testified say they're less worried that what they say will be leaked to the public than that it will get back to the agencies themselves, which are known for their zealousness in tracking down whistleblowers.

Even party labels mean next to nothing. Two of the more comparatively tough-minded members, Senator Shelby and Representative Saxby Chambliss, are from the GOP, while Democrats Bob Graham and Representative Jane Harman are considered stalwarts for the status quo. When the Democrats took back the Senate in May 2001, many in the CIA could barely conceal their glee at the prospect of Floridian Graham as committee chairman--a choice they found infinitely preferable to the more critical Shelby. Graham sometimes issues tough statements, but he is considered totally "on board." His intriguing past includes the controversial sale of his home, through a middleman, to Carlos Cardoen, a Chilean arms dealer who exported cluster-bomb technology to Iraq during its war with Iran (supposedly with an official US OK). Cardoen, who was accused of using real estate to launder money, raised funds for Graham's first Senate race. Graham was also close to David Paul, who went to jail for his role in the failure of an S&L with links to the intelligence-connected and now-defunct Bank of Credit and Commerce International. Graham gets along famously with Republican Goss. Said a well-respected former chairman about the duo, "I find them inoffensive at a time when they should be aggressive."

Vincent Cannistraro, former CIA chief of counterterrorism operations and analysis, says that the chairmanship usually goes to someone who has been a big supporter of the intelligence community. "When they are not, the community leaks adverse information," he says. "People with presidential ambitions won't fuck with the intelligence community--especially if there is weird stuff in their backgrounds."

hen secrecy comes up against accountability, the former almost always wins. Yet the fear of harmful leaks has little basis in fact. As a CIA consultant noted in a 1997 report, "Apart from a handful of widely reported and somewhat dated examples, no intelligence agency personnel interviewed for this study could point to instances of compromise by Members or their staffs." In fact, even the now-famous memo by FBI Minneapolis bureau agent Coleen Rowley about deficiencies at the bureau has never been leaked in its entirety.

The June "leak" regarding the National Security Agency's Arabic intercept the day before September 11 (in which someone declared "the match begins tomorrow" and "tomorrow is zero hour") gave the Administration an excuse to rough up the committees. But Loch Johnson, a respected former intelligence committee staffer, says bluntly, "Chances are, the leak came from the executive branch anyway. And if not, the Congress can read the riot act to members and staff, and they will almost always shape up." As in fact they did, scrambling to demand an inquiry into themselves. Nancy Pelosi of San Francisco, the liberal minority whip and ranking Democrat on the House Intelligence Committee, declared, "It is appropriate that we ask for this investigation and let the chips fall where they may." Several committee members balked when the probers suggested lie-detector tests all around (Shelby said he wouldn't comply, and Pelosi objected as well), but others, including Republican Senator Jon Kyl, indicated an enthusiasm for getting hooked up to a polygraph. According to the Associated Press, the offices of thirteen of seventeen committee members confirmed that they were turning over telephone logs, memos, visitor sign-in sheets and other documents--and not a single member indicated he or she would resist. In the panic and rush to establish bona fides, no one even dared to suggest that the public might be better off knowing about such matters.

John McCain underlined this when he recently stormed out of a so-called top-secret briefing, complaining that senators were learning less than they might from reading the paper. Although the committees looking into 9/11 and more generalized intelligence failings typically meet in closed session in order to protect "sources and methods," former House committee chairman Dan Glickman remembers, "They didn't tell us anything about that anyway." Glickman recalls the constant frustration of facing intelligence personnel who refused to declassify, or even talk about, information that was already out in the press. Their refusal means that Congress members are checkmated from mentioning such matters in public. It's even worse on the rare occasion when members have some worthwhile knowledge as a result of their briefings. "Most excruciating for me," recalls Lee Hamilton, "was when the President would say something at a news conference and he was flat wrong. On the basis of available intelligence, you know information that makes it wrong, but you can't say anything about it." Shelby left the impression that little has changed in this regard. Even more fundamental, Congress members have almost no way to assess the quality of raw intelligence or how it is analyzed.

Institutional cocooning is best illustrated by the continued classification not only of line items in the intelligence budget but even of the single figure that represents total intelligence spending. Periodic efforts by committee members to eliminate this ridiculous tiptoeing have largely gone silent since 9/11. The 1997 and 1998 budget totals were declassified only after the Federation of American Scientists sued for the information. Another current suit demands the release of the 2002 budget (widely believed to exceed $30 billion) and, incredibly, the 1947 and 1948 budgets, which the CIA still deems too sensitive to make public. One might well conclude that this attitude is less about protecting legitimate secrets than hiding the faults of the system itself. "There is probably a lot of classified information that might be embarrassing to people but not undermine the Republic," says Shelby.

Committee members routinely complain that agency personnel try to keep them in the dark on the high-tech equipment that consumes so much of the spying budget. Where such nondisclosure lurks, can pork be far behind? Recruiting Farsi or Arabic speakers may result in better intelligence for a relatively small outlay, but it's the hugely expensive technology contracts (and the jobs they provide) that allow Congress members to reward contributors and constituents, often without an accurate sense of what largesse they are dispensing. It's further downhill from there. "Would-be critics do not even know what is at issue," laments Aftergood. And, of course, there's that old revolving door: Typical is former Senate committee staff director Taylor Lawrence, who went over to defense contractor Northrop Grumman.

The oversight process certainly has its "moments" (one of which is right now), but its overall performance seems designed to protect a dinosaur. Times have changed, the world has changed, but we don't really know what--except for buying increasingly fancy hardware--the spy agencies have done to change their strategies and tactics since the cold war. As the long-delayed "open" hearings unfold, the confluence of events and the public mood may have finally created a rare window for real reform. Even Shelby agrees. "We think it will take public pressure to modernize these agencies," he says. Fair enough. But the pressure will have to start with someone who knows something, such as committee members themselves. What if one or two of them decide to go public with a more detailed critique of the current system? What if they declare openly that "sources and methods" is a red herring, and that it's time for the public to be told some of the hard truths about the intelligence community's quotidian operations?

They might start by challenging the idea that government works best by keeping "sensitive" information secret. There is evidence that much successful investigative work and pre-emptive action depends on a collaboration with the public. Given what we now know about tips and leads that were ignored by the bureaucracy before 9/11, it can be argued that bringing more openness to the process would improve accountability and increase public safety. Opening up the system, within prudent limits, will require not just more "open" hearings but changes in the intelligence community to insure that criticisms and suggestions are heard and, where appropriate, acted on. The oversight committees can take steps to put together well-informed and more vigorous staffs, whose mission includes not just asking agency officials better questions but cultivating alternative sources of information. Where can the committees find such staffers? Some knowledgeable insiders suggest that the best place to look might be among disaffected former intelligence officers who got into trouble for speaking out or who retired in disgust.

Right now, we need some big, concrete moves. For example, if homeland security is going to be "centralized," then perhaps so should the intelligence oversight and budgeting process, which is now shared among committees, including Armed Services and Judiciary as well as Intelligence. But the most important first step is to insure that the impending independent commission on 9/11 has the clout and know-how to dig deep and effect change. As important, after it looks into 9/11, it should be converted into a non-Congressional variant of the original Church committee of the 1970s, which created oversight. It needs a firm mandate to reassess the entire intelligence mission and infrastructure, and the ways in which a democratic society can have a say over its spies. Archimedes said, "Give me a place to stand, and I can move the Earth." Moving the intelligence community toward greater efficacy and accountability will take the concerted efforts of tough people who combine law enforcement and intelligence experience with a willingness to seek answers in the face of opposition from some of the most formidable powers in and out of government.