A Strategy's Cautious Evolution

Personalizing the struggle to one man, he said, was "one of the fallacies" of the Clinton team's approach.

In his first week on the job, deputy national security adviser Stephen J. Hadley instructed NSC team leaders to propose subjects for high-level review. Much of the incoming staff was still finding its way around the 553 rooms and two miles of corridors in the Eisenhower Executive Office Building, once the world's largest.

Clarke did not need a map, or a second invitation. He had a three-page proposal on Hadley's desk that day.

The Jan. 25 memorandum spoke starkly. Clarke and Cressey had just navigated through the most intensive period of counterterrorist activity in American history. The millennium year marked its start with al Qaeda plots -- stopped by improbable good fortune -- to mount synchronized strikes on airports in Boston and Los Angeles, and on American tourists in Jordan. It ended with a suicide attack that killed 17 sailors and crippled the USS Cole in Yemen three weeks before the presidential election.

More attacks had almost certainly been set in motion, Clarke and Cressey wrote. American intelligence believed there were al Qaeda "sleeper cells" in America -- not a potential problem but "a major threat in being," according to people who read their proposal.

Clarke had pressed superiors since the Cole bombing on Oct. 12, 2000, to mount a military attack on al Qaeda's Afghan training camps. Clinton left the question for his successor, and what little public record there was hinted that Bush might choose to act.

"I hope that we can gather enough intelligence to figure out who did the act and take the necessary action," candidate Bush said the morning after the explosion. "There must be a consequence."

Clarke argued that the camps were can't-miss targets, and they mattered. The facilities amounted to conveyor belts for al Qaeda's human capital, with raw recruits arriving and trained fighters departing -- either for front lines against the Northern Alliance, the Afghan rebel coalition, or against American interests somewhere else. The U.S. government had whole libraries of images filmed over Tarnak Qila and its sister camp, Garmabat Ghar, 19 miles farther west. Why watch al Qaeda train several thousand men a year and then chase them around the world when they left?

Clarke asked Rice to let him begin an interagency review. As it began, he recommended five immediate steps.

Massoud's Northern Alliance fighters, in danger of defeat by the Taliban, needed enough aid "to keep them alive until we figured out what our overall strategy would be," as a new Bush appointee put it. In neighboring Uzbekistan, President Islam Karimov needed more help for an American-trained battalion he sent against fundamentalist rebels allied with al Qaeda. Treasury had to get moving on a terrorist assets tracking center, months overdue. The CIA's Counterterrorism Center could buy a lot more cooperation from foreign intelligence services if it had more cash -- the center's whole budget, sources said, did not exceed $ 50 million. And the Voice of America had to start answering bin Laden -- in local languages -- to counter his appeal in the Islamic world.

Not much came of Clarke's immediate requests. It would be months before the new team's appointees arrived in force. But Rice and Hadley liked his zeal. The inherited strategy of battling al Qaeda cell by cell, they believed, could not work.

"The premise was, you either had to get the Taliban to give up al Qaeda, or you were going to have to go after both the Taliban and al Qaeda, together," Hadley said in an interview. "As long as al Qaeda is in Afghanistan under the protection of the Taliban . . . you're going to have to treat it as a system and either break them apart, or go after them together."