Intelligence Matters, Sen. Bob Graham's venture into the flood of 9/11 books, is much better than its clichéd title and generic cover art might suggest. Briskly paced and mercifully brief at 297 pages, it accuses George W. Bush of blowing the war on terrorism, covering up a crucial Saudi connection, and obstructing a congressional inquiry at every turn. As the chairman of the Senate Select Committee on Intelligence, Graham helped undertake that inquiry—and co-wrote the resulting report, which recommended a host of structural reforms, some of which Congress is now considering.
However, the main lesson of Graham's book—though he never makes this point explicitly—is that the chief failures of 9/11 were personal rather than organizational.
In the course of his narrative, Graham identifies 12 points at which the terrorists' plot "could have been discovered and potentially thwarted." Yet a close look at those 12 points reveals that in 10 of them the main problem was not the famous "wall" dividing the FBI from the CIA, but rather the sheer incompetence of the people working on either side of that wall.
This is a disturbing conclusion. If the problem is bureaucratic barriers, then tear the barriers down; doing so might be politically difficult, but at least it's clear what must be done. However, if the problem is lazy or stupid bureaucrats, what do you do about that? How do you go about ridding an agency of incompetents?
Alas, there is no denying that the terrorists of Sept. 11 were abetted by the plain incompetence of our spy agencies.
Many of Graham's 12 failures concern the U.S. intelligence community's failure to observe two of the al-Qaida hijackers, Nawaf al-Hazmi and Khalid al-Mihdhar, as they flew in and out of the United States and, finally, settled down in San Diego under the nose of an FBI informant.
The first lapse was the CIA's failure to place the two on a State Department watch list, even after receiving a report that they had attended a terrorists' summit in Kuala Lumpur. No interagency obstacles here: It's simply a failure within the CIA.
The second missed opportunity occurred when the CIA asked Malaysian intelligence agents to monitor the Kuala Lumpur summit. It turns out the Malaysians didn't bug the meeting place with listening devices, and so didn't hear the assembled discuss the plan to bomb the U.S.S. Cole. Again, this was a straightforward CIA screw-up.
The third big mistake occurred when the CIA obtained a photo of the two hijackers. Either the CIA case officers did not pass the photo on to the FBI, or they did pass it on and the FBI lost it. Graham writes that the evidence is ambiguous on this point, though he suspects the FBI is at fault. Whichever view is true, this is one flub caused, at least in part, by interagency conflicts.
The fourth muffed chance took place when the two hijackers moved near an FBI informant in San Diego. Because the FBI didn't have them on the watch list, the bureau's case officers didn't know to ask about them.
The fifth flub came soon after, when al-Hazmi started to work for a San Diego business manager who was the subject of an ongoing FBI investigation. Again, no link was made—and, again, because of utter disarray within the FBI.
No. 6: In June 2001, al-Mihdhar, who had left the United States, applied for readmission and lied on his application form, stating he had never been here before. Since he still wasn't on a watch list, no one knew otherwise.
The seventh botched opportunity was when another FBI informant reported seeing al-Hazmi with one of the Arabs he was tracking. Nobody pursued the connection. Again, an internal FBI failure.
The eighth, and catastrophically huge, screw-up occurred when Kenneth Williams, an enterprising agent in the FBI's Phoenix branch, wrote a now-famous memo warning of a possible connection between al-Qaida and Arabs taking flight-training classes. His superiors did not send the memo up to the highest levels—another case of sheer incompetence in the FBI's mid-to-upper echelons.
No. 9, and equally disastrous, occurred when FBI bureaucrats decided not to request a warrant to search the laptop of Zacarias Moussaoui, a suspected terrorist they had captured, because he did not meet the criteria of a "FISA warrant"—a special warrant authorized by the Foreign Intelligence Surveillance Act.
Some have blamed the overly legalistic Justice Department for the Moussaoui foul-up. But Graham notes that the FBI's lawyers in this case misinterpreted the FISA statute. They somehow thought the law required the target to be a suspected member of a "recognized" foreign power. Moussaoui was suspected of assisting the rebels in Chechnya, which is part of Russia and therefore not a "recognized" foreign power. Yet, Graham writes, the FISA law requires no such thing. The FBI lawyers could have requested a warrant. More puzzling still, after deciding not to pursue a FISA warrant, they could have pushed for a straightforward criminal warrant. Yet they didn't. Again, sheer, dizzying incompetence.
In the 10th bungle, German intelligence obtained the Hamburg phone number of a suspected terrorist named Mohammed Zammer and gave it to the CIA. The CIA did nothing to follow up. Sheer CIA negligence.
The 11th foul-up occurred when the CIA found out that al-Mihdhar had obtained a U.S. visa and didn't inform the FBI. This is the second instance when interagency barriers did play a critical role.
The 12th missed chance occurred on Aug. 23, 2001, when the FBI's special Osama Bin Laden Unit finally realized that the hijackers were inside the United States and asked headquarters to investigate. Headquarters declined the request and refused even to canvass the FBI's own counterterrorism team. Once more, the FBI—and the FBI alone—was to blame.
Graham discusses, but does not itemize, other, higher-level failures, most notably President Bush's failure to alert the Federal Aviation Administration after the famous President's Daily Brief of Aug. 6, 2001, which mentioned the possibility of hijackings.
In the vast majority of these foul-ups, the problem wasn't with the agencies' organizational charts but rather with the quality of the people filling those charts. The trouble wasn't so much the flow of information but rather what people did with information when they got it. The key question is not so much how the government was structured but rather who did what, when?
Washington is now debating how to restructure the intelligence community—whether to break up the CIA, expand the CIA director's powers, or create a new national intelligence authority. This is an important debate. But the chronicle of 9/11 tells us that a much more important debate should be going on over how to reward smart people inside the intelligence community—and how to recruit more of them.
As Graham notes in his book, nobody has been fired for the screw-ups that led to 9/11. (George Tenet might be an exception, but if so, he's a poor case-in-point. Rewards and punishments must be meted out clearly if they are to set examples; yet no one even knows if Tenet was fired or just quit.)
Nor does it seem that any analysts or agents have been rewarded for doing their job particularly well. Why isn't Kenneth Williams, the keen analyst who wrote the Phoenix memo, running some branch of the FBI right now? Why wasn't Sibel Edmonds, the FBI translator who spotted malfeasance in her ranks, promoted instead of hounded out of her job and treated like a criminal? Where are the lavish incentive packages to lure smart people into intelligence work? Why aren't there special scholarships for Arabic-language students?
Washington used to know how to do this sort of thing. The FBI created a whole culture of attraction. How many kids grew up and became FBI agents because they'd watched The Untouchables or read G-Man adventure books? NASA did much the same image-crafting at the height of the space program. In the early 1960s, when the Telstar satellite spun out of control, an uncle of mine who worked at NASA devised an algorithm * to bring it back in orbit. NASA splashed him on headlines all over the country as "the man with the million-mile arms." This may seem a bit cheesy by today's PR standards, but the federal government should come up with some ideas on how to accomplish what should be seen as its main challenge—to make its intelligence agencies more intelligent.
Correction, Sept. 8, 2004: This article originally mistakenly referred to a logarithm.