Steve Coll is a New Yorker writer and the author of Ghost Wars: The Secret History of the CIA, Afghanistan, and Bin Laden, from the Soviet Invasion to September 10, 2001. His interview here offers an overview of George Tenet, his relationship with President Bush, his leadership of the CIA, and, in particular, his management of the intelligence community's assessment on Iraq's WMD programs -- an assessment that contributed to the case for war, but soon after was proven wrong. This is the edited transcript of an interview conducted on Jan. 12, 2006.

• Some highlights from this interview
• Assessing how we did in the 2002 Afghanistan war
• Tenet's role and judgment in the flawed WMD intelligence on Iraq
• Tenet's resignation
• The Porter Goss appointment

When George W. Bush is first elected president, George Tenet gets to stay as head of the CIA in the new administration. Can you give me some background on him?

George Tenet is a product of official Washington. He started as an aide to a lobbying group for solar power. He then worked for a long time on Capitol Hill, mainly for moderate Democrats. The most important mentor in his career was David Boren, a conservative Democrat from Oklahoma who chaired the Senate Intelligence Committee during a period where Tenet was his chief staffer. It was through Boren's mentorship that Tenet moved from the Senate to the Clinton National Security Council [NSC], where he was in charge of intelligence, budgeting and decision making at a level that put him right at the heart of the bureaucratic process that is the intelligence community in D.C. For several years, completely out of the limelight, he worked at the NSC at the heart of this culture.

In the mid-90s, at the beginning of Clinton's second term, the CIA leadership imploded, and Clinton sent Tenet out there as deputy director, because he was the only person who could be confirmed. George Tenet is a man that just about everybody in Washington likes. He's a product of Georgetown University; he's a gregarious, outgoing, warm, blunt character who knows how to work the aisle.

And so with that experience, he became an indispensable figure for Clinton, who wasn't interested in transforming the world through the CIA or running risky covert action programs all around the world. He wanted the CIA to be responsive to his relatively modest foreign policy agenda. He wanted smooth political sailing. George Tenet delivered all of that to him.

At the CIA, first as deputy director and then as director, Tenet was a popular figure. The CIA had had an unusually rapid amount of turnover among the directors, and they didn't for a good while have a director who really wanted to be there and who seemed to love the place and its mystique and its culture and its self-mythologizing. Tenet loved the place, the people. He liked walking around; he liked being of the place. And that showed; that made people feel good about his leadership.

That he was underqualified in areas traditionally associated with CIA leadership mattered less given the agency's need for this morale boost. He had almost no experience overseas. He had traveled to a limited extent. Most of his knowledge of foreign policy had been developed in the Senate, reviewing other people's decisions rather than making them. And he hadn't operated in any sense. He hadn't been a spy; he hadn't overseen spies. Yet his natural instincts for people and for process served him well during the Clinton term.

By the time Clinton left office and Bush came in, Tenet had built supporters on both sides of the political aisle. He had built support inside the intelligence community, crucially and particularly in the Directorate of Operations [DO] at the CIA, which is the secret-stealing, covert action-running, self-mythologized heart of the agency. He had won support. From all of those quarters, the new President Bush, George W. Bush, heard that he ought to keep Tenet around, that it would provide a source of continuity, that it would provide a signal of bipartisanship that his Cabinet needed as it was formed.

You talked about his connection to the CIA Directorate of Operations side. What does it mean that they liked him?

The culture of the DO from the '50s right through the '90s, it's a secret brotherhood in which the members believe that they bring to their work a rare blend of intellect and a bias to action. ... Tenet's personality fit perfectly with that culture. He was a man of short, clear sentences; energy; warmth; gregariousness. He wanted to get things done. He also understood, I think, in a tactical sense, having analyzed the failed directorships that preceded him, that history seemed to suggest to new directors of the CIA that you could not succeed unless you had the Directorate of Operations with you. Even if you wanted to reform the Directorate of Operations, even if you wanted to change the way it operated, even if you wanted to point it in an entirely new direction, you had to do so by co-opting it, by bringing it into your orbit. And Tenet was enough of a politician to understand how to go about that.

It wasn't just that he went in there and said, "Do what you please, men." It was that he went in there and said: "I am of you; I believe in you; I believe in your work; I'm going to defend you in the crunch. But I also need you to respond to my agenda for change."

What kind of change did he have in mind?

Well, at the beginning, in the Clinton period, the change he had in mind was simply reviving the morale and basic functioning of the Directorate of Operations, which had been pinched by declining budgets to the point where I think it's been acknowledged that in 1997, the incoming class of new case officers, which is the profession of spy in the Directorate of Operations, was about a dozen people. This was an organization that at the height of the Cold War had been inducting hundreds and hundreds of new operatives each year. The forced early retirements of senior officers in the Directorate of Operations had created a culture of disappointment, if not bitterness, in many corners of the Directorate.

Tenet's initial goal was to revive a sense of possibility and a sense of belonging in a Directorate that had really hit rock bottom because of these forced retirements, layoffs and declining budgets.

George W. Bush gets elected. How does the Tenet personality capture, if that's what happened, the president of the United States?

At the time, the principal privilege of the director of the CIA was that if he wished, and if he was in town, he had a daily appointment scheduled with the president, one on one, to provide him with a direct briefing of overnight developments in the world that the president of the United States ought to know about. Tenet used the opportunity in the morning briefing to build a personal relationship with the president.

... It's very easy to see how Tenet and George W. Bush would have gotten along on a personal level at these morning briefings. They're both men who aren't going to spend a lot of time pulling apart nuances of international relations. They're interested in short, sharp facts -- a clear sense of direction. Tenet is a great briefer. He has a relaxed personality. He, like the president, is casual in his demeanor, casual in his speech, funny, quick to banter. It's easy to imagine at these briefings that two men who didn't know each other at all discovered that they were similar personalities.

The president would have seen in Tenet a fellow traveler. His White House, though, would have had questions about Tenet's ultimate loyalty to the Bush administration. Tenet was, although this was not widely known, a registered Democrat. He had worked on the Democratic side of the aisle in his earlier career in the Senate. And though he had carefully developed a reputation as a nonpartisan Washington professional, even to the extent that he had Democratic connections, they were all on the conservative side of the party. Still, he had not come to this administration from the trenches. He had not established a record of contacts and support within the Republican Party.

So before Sept. 11, Tenet had a paradoxical relationship with the White House. His relationship with the president was very good and very personal, but his relationship with the Bush administration, particularly the true believers in the administration, was underdeveloped, and there were questions among some of the conservative and most loyal travelers in the Bush administration about whether Tenet really belonged among them.

Is the CIA on the verge of irrelevance when the George W. Bush administration takes over?

Well, in this period, the CIA's relevance is a function to an unusual degree in Washington of the relationship between the director and the president. The CIA has a unique status in Washington in the sense that its operations, particularly its covert operations, can only be authorized by the president, and the law that surrounds covert action contemplates a direct and confidential relationship between the president and the DCI [director of central intelligence]. It's different, for instance, than the relationship between the president and the director of the FBI, where there's an arm's-length relationship about such matters. It's different than the relationship between the president and the director of the National Security Agency [NSA], for instance, who reports to the president through the director of central intelligence.

So the irrelevance of the CIA in Washington at the end of the Clinton administration was a function in part of its ragged track record, but also of the fact that Clinton had no energetic interest in the agency and its operations. In fact, he had come to distrust it. And even though he trusted and liked Tenet, he had felt burned by a number of mistakes that, in his judgment, had come out of the CIA's bureaucracy, most notably the targeting error that led to the destruction of the Chinese Embassy in Belgrade during the Kosovo war, which was just an egregious goof that led Clinton to believe, I think, that his earlier skepticism about the CIA was well justified.

But when George W. Bush came into office, there was every reason for the CIA to believe that the estrangement from the White House that they had experienced during the Clinton administration was about to end, because in recent history, no president had been more attentive to the CIA than George W. Bush's father. George Herbert Walker Bush was a former director of the CIA; he had long-standing personal, social contacts among senior spies at the agency. He would invite them to the White House Christmas party; he would reach out to them separately from official channels. When his son took power, there was a sense of restoration at the CIA.

So now we are at 9/11, and Tenet is CIA chief. Can you talk about that?

Well, the advantage that Tenet had on Sept. 11 was [that] among the president's Cabinet members, he was the only one who really had a grip on where this attack had come from, because for the previous two or three years, his agency had been put in the lead in trying to chase Al Qaeda, and he had been a principal source of warning that Al Qaeda was planning a massive attack of some sort, though he couldn't pin it down, and he couldn't break up the plot that ultimately produced Sept. 11.

So in those early, fraught, intense meetings of President Bush's National Security Cabinet to assess what Sept. 11 meant, Tenet came to those meetings with briefing books that were fuller, more detailed, and that even contained a series of plans that had previously been written that contemplated military action in Afghanistan with a strong paramilitary special forces and intelligence component. [Secretary of Defense Donald] Rumsfeld at the Pentagon had less well-developed plans, and his aides had a less well-developed understanding of the Afghan nexus from which Sept. 11 had arisen.

So almost from the beginning, there's that level of at least bureaucratic competition, if not interpersonal competition?

It seems from everything we've now come to understand about the fall of 2001, Rumsfeld and others at the Pentagon were frustrated that their own preparedness for the war in Afghanistan was less that the CIA's preparedness, and there clearly was a tension and a competition to perform at this moment of national need. Because the Pentagon's planning for Afghanistan was underdeveloped, it mostly involved air targeting and bombing coordinates. It certainly didn't involve the kind of nimble, relatively innovative blend of collaboration with militias, special forces operations, air operations and integrated targeting that Rumsfeld saw as the new warfare.

When Rumsfeld turned to his colleagues after Sept. 11 and said, "Let's wage the new innovative war that I have in mind; show me the plan for such a campaign," they didn't have it. But the CIA had a plan that looked a lot like a Rumsfeldian war, in a sense, so it was Tenet in a way who was carrying forward the principles of new warfare that Rumsfeld himself advocated.

In those immediate days right after Sept. 11, people like John Yoo [then-deputy assistant attorney general for the U.S. Justice Department] and Judge [Jay] Bybee and others were rewriting the war powers authorization, thinking seriously about broadening powers. What is the impact of those things on George Tenet, Cofer Black, the CTC [Counterterrorist Center], the CIA?

Well, you have to understand that in the two years before Sept. 11, CIA officers like [former CTC Director] Cofer Black felt frustrated by what they regarded as the legalisms and the constraints imposed upon their attempts to go after [Osama] bin Laden and Al Qaeda in Afghanistan. They felt that both the policy that they were given to fight Al Qaeda and the rules that they had to operate by were unnecessarily constraining, and in fact had inhibited their ability to get to bin Laden before Sept. 11.

So there was this enormous sense among the officers that had lived in this campaign before Sept. 11 that now the gloves could come off. Finally, these lawyers and these cautious decision makers who had gotten in our way before can be overcome, and we can be given the license that we deserve to have had previously.

They also had a specific agenda, and they had a specific plan to attack the Taliban in Afghanistan.

And of course because of Tenet, they could?

Tenet had the relationship with the president to make this plan plausible, to communicate it, to talk about it, to answer questions, to defend it, to personify the agency and the people who stood behind him. People like Cofer Black also played a role in coming over to the White House and looking the president in the eye and saying: "We can do this; give us the running room. Here's exactly what we're going to do. We're not making this up; we've got a plan."

The Camp David meeting on the 15th [of September] was, I guess, the formal moment where the CIA plan slid across the table and the president took it on board. At least they considered it over Sunday and talked about it on Monday. Is that the way you heard the story?

... One thing to keep in mind about that Camp David meeting and the discussions within the National Security Cabinet in September of 2001 is that everyone understood broadly what Al Qaeda was, and they understood broadly that bin Laden was in Afghanistan. ... But neither the president nor most of his Cabinet members had a very well-developed understanding of what Al Qaeda actually was at that stage. ...

It was the CIA who understood what Al Qaeda was better than anybody else, so they were able to control to some extent the discussion of what should be done in September and October and November of 2001, because they knew the answers to the most urgent questions. Now, that dynamic changed over time as the Pentagon and other agencies of the government became better educated about their enemy, but it also explains, I think, some of the disconnect that people like [counterterrorism expert] Richard Clarke have described about September 2001, where you have some members of the president's Cabinet ask questions about Iraq and Saddam Hussein that strike the Al Qaeda experts like Richard Clarke and George Tenet as goofy.

Also, a lot of the Republicans in Bush's administration hadn't been in government effectively since the Cold War period, ... and so in the Cabinet, there was a natural inclination to think that an attack of this scale, this power, this sophistication, had to have roots in state sponsorship, not only because there was a natural willingness to believe that Iraq would be interested in carrying out such an attack, but also because this was a group of people whose experience of terrorism led them to believe that it was almost always state-sponsored in some sense.

And so it justifies why, for example, as it's been reported, [then-National Security Adviser] Condi Rice says: "You know, Mr. President, fighting Al Qaeda is fighting shadows. It's easier for a nation to fight a nation, to take on a state."

Yes. I think that if [then-Deputy Secretary of Defense Paul] Wolfowitz or Condi Rice were here, what they'd say was: "Well, we wanted to do both things. The CIA told us we needed to attack the network as a network, fight it in its statelessness. But at the same time, in order to diminish Al Qaeda's potential, we had to strip it of access to the resources that hostile states could provide. That's why we went in and destroyed the government of the Taliban, and that's why we began immediately to worry about the potential connections with Iraq."

That Sunday morning following 9/11, the vice president [Dick Cheney] is on Meet the Press and says to Tim Russert: "We're going to go to the dark side. The American people need to understand that." What did he mean by that, do you think?

Well, we now understand that in the first 10 days or so after Sept. 11, there was a very intensive discussion within the administration about which rules governing the covert operations of the CIA and the Pentagon needed to be rewritten. ... Cheney would have understood by the time he went on Meet the Press that a new architecture was being contemplated.

What specifically are we talking about? First of all, to carry out covert action in this country, you need the authority of the president, and the president must provide that authority in writing. When you provide written authority to the CIA to carry out covert action, you may provide a certain level of specificity, but you are mostly just giving umbrella authority to the agency to operate. Once the covert action begins, lawyers and policy-makers at the CIA, at the White House, begin to confront how much they're going to write down about the rules of engagement out there in the field.

I think Cheney was signaling that, already in the first 10 days after Sept. 11, there was a consensus building in the administration that practices that had been ruled out during the 1990s were going to be ruled in. This certainly would have included practices associated with detention, interrogation of prisoners and the use of force in targeted killings against individuals known to be planning imminent attacks against the United States or its interests.

Because of the revelations that followed Abu Ghraib in 2003-2004, there's a tendency to believe that these rules for interrogation and detention were invented in the immediate aftermath of Sept. 11. The truth is there was a long-running discussion and a set of precedents about extrajudicial detention, about capturing terrorist suspects in one country and shipping them clandestinely to another country, and even about targeted killings. There was a rich series of classified precedents -- Justice Department memoranda that predated the Bush administration -- that created the framework in which the Bush administration began to work on these issues.

Certainly, for instance, let's take renditions, capturing a terrorist suspect in, let's say [in] Italy, and putting them on a private plane with a hood over his head, flying him to Egypt for interrogation. Now, the legality of that practice under American law was well established by Sept. 11. It had been sanctioned as far back as the Reagan administration, and the legal rationales for such activity had been very well developed in Justice Department memoranda produced during the Clinton administration. Indeed, there had been a series of court cases that had seemingly ratified the legality of a lot of this activity.

So these questions had antecedents by the time the Bush administration started wrestling with them in September of 2001. Rather than inventing a whole new set of rules in which they said, "Oh, let's go kill anyone on a list of terrorists," they were taking a debate that already existed, albeit in classified channels, and just moving the lines. Then they were creating new language to justify those changes.

I don't know if you know this, but in these early days, where is the vice president on all of this? Do we have any fix on him as a character, the positions he has? ...

I don't have anything that you haven't read elsewhere. Completely elusive to me -- and maybe intentionally it was elusive; I don't know. One thing we can say is that if you just read the vice president's own words, he makes plain that for him, Sept. 11 is evidence that human affairs exist in a state of near-permanent warfare, and that the defense of the United States requires a clear-eyed recognition that war is almost a natural human condition, and that it is the failure to recognize this semi-permanent state of war, the vice president seems to believe, that made the United States vulnerable to Sept. 11 attacks in the first place.

If you were inclined to see Al Qaeda and bin Laden as criminals who can be addressed through the mechanisms of constitutional law enforcement, in the vice president's view, you are mistaken. These are warriors, marauders who can only be defeated through acts of war. It's that mind-set, I think, that the vice president, by his own public statements, has made clear he brought to each of these deliberations.

I've heard many stories about Tenet in those months right after 9/11, traveling Europe, opening up avenues that would not have otherwise been available, perhaps, to the Bush administration.

Tenet after Sept. 11 was given authority to ramp up the CIA's global campaign against Al Qaeda. Tenet recommended that this campaign be carried out primarily through the development of liaison relationships with intelligence services in the Muslim world in Europe and wherever Al Qaeda operated in Southeast Asia. To some extent, the CIA already had been campaigning against Al Qaeda through these liaison relationships. But after Sept. 11, Tenet had almost unlimited resources in terms of cash, technical assistance, personnel, to offer these intelligence services around the world, whether they were in Europe or in Jordan or in Morocco or in Egypt or in Pakistan. These were intelligence services that themselves had been confronting Al Qaeda for years, that understood who they were up against, and that also had the kinds of officers that the CIA didn't have -- officers who could blend on the street; officers who spoke Arabic; officers who could carry out successful interrogations by threatening a suspect's sense of well-being, his family, his sense of place in his society.

So Tenet went to allied intelligence agencies with satchels full of resources and essentially a promise to these agencies: "Work with us, we'll make you richer; we'll make you stronger; we'll make you better at everything you do, not just our counterterrorism agenda. Yes, we have common cause." Particularly among the most trusted allies, which I would include Jordan, Morocco and Egypt on that list, there was a sense of mutual opportunity that was almost palpable in the fall of 2001.

There were other liaison relationships where there was more mutual suspicion present between the CIA and the national intelligence service: Pakistan, to some extent, because Pakistan had previously supported the Taliban and had contact with Al Qaeda through its intelligence service; Saudi Arabia, where during the late 1990s the intelligence liaison had been a mixed picture of some mutual suspicion. But Tenet was in his element. What does the director do that nobody else can do at the CIA? He's the face of the agency.

And the downside?

I think the CIA recognized all along ... that these allied intelligence services, first of all, they have their own national agendas that are not always the same as the agenda of the United States.

Secondly, their methods of operation are perhaps not as subject to the review of constitutional legal procedures as those of the United States. This was not a new issue. During the '90s, there was a great deal of attention paid within the CIA to the problem of intelligence liaison relationships with national intelligence services that had records of abuse and torture, for instance, in the interrogation of their own nationals. There were times when the CIA felt it had to suspend aid to certain intelligence services, including Egypt's for a brief period, because of concerns, sometimes emanating from the United States Congress, sometimes generated within the CIA itself, about whether or not the Egyptians or another service was abusing the aid that it was receiving from the CIA.

So this was not a new issue. But as with retentions, as with detentions, there was a sense in the fall of 2001 that while this wasn't a new issue, the rules could change, and the balance of interests had changed. The United States had just endured the worst surprise attack on its soil since Pearl Harbor. There was an uncertainty about whether more attacks were imminent; there was a sense that the only way to prevent more attacks was to identify and disrupt follow-on cells, and that the only way to do this was to get out in the parts of the world where those cells were incubating and to work with friendly national governments to identify, disrupt, detain suspects.

So what is the Afghanistan plan?

The Afghanistan plan is to immediately collaborate with the only armed group in the country, the Northern Alliance, which is itself a coalition of militias that has been at war with the Taliban steadily since about 1996. The plan is that the CIA in the lead, but the Pentagon as well, will go into northern Afghanistan, connect with the leaders of the Northern Alliance, offer them money, equipment, political support, and with them drive against the Taliban in Kabul and in some other cities that they control in the north.

And who do they bring to the table to execute the plan?

Well, the man that they chose to lead the first trip into northern Afghanistan is a guy named Gary Schroen who was a career CIA officer in the Directorate of Operations who had developed an expertise on Afghanistan and in particular had gotten to know the Northern Alliance leaders quite well from previous trips into Afghanistan and from his work during the anti-Soviet war in the 1980s.

How big a risk was what they were trying to do, and how great a success was it really?

... There were considerable risks. ... They were cut off, isolated. They had no reliable transportation once they got into the country. They were basically going to be on their own until they either fought their way into Kabul or somebody came in and extracted them. There was some uncertainty also about the reception they were going to receive, because the Northern Alliance had just had its own leadership decapitated two days before Sept. 11, when Ahmed Shah Massoud was assassinated by Al Qaeda. So they were coming in to meet a group that was itself in some disarray and that, as it turned out, was going to require a little bit of persuasion to operate the way the CIA and the Bush administration wanted them to operate.

Was the outcome ever in doubt?

Of course it was. ... The CIA and the Pentagon didn't have any reason to be confident that the Taliban would melt away in the face of American military action. To the contrary, the Taliban represented tribes and militias in Afghanistan that historically had sliced and diced every foreign invading army that had ever come their way; the British twice, the Russians. So the United States went into this campaign concerned that they could get bogged down. ... The whole idea of working with the Northern Alliance was rooted in a desire to avoid putting American combat troops into a position analogous to that which had destroyed two British armies and decimated the 40th [Army] of the Soviet Union.

What's the assessment at the time, and even now, about how successful we really were in Afghanistan?

Well, there are two objectives in the war in Afghanistan in the fall of 2001. One is to defeat the Taliban, which is the government of Afghanistan, and to take power in Afghanistan, to control the cities, the mechanisms of government. That campaign was extraordinarily successful. It happened very quickly. It happened with very few casualties among the Americans; it happened with relatively few casualties even among our Afghan allies. The Taliban melted away and retreated to Kandahar, which was their heartland, and then were fairly quickly rooted out of Kandahar as well and forced to flee across to Pakistan and into the hills of south central Afghanistan.

So two objectives: First, defeat the Taliban -- that's enormously successful; second, to defeat Al Qaeda and to deprive it of a physical sanctuary that it had been using prior to Sept. 11 to carry out attacks. The campaign against Al Qaeda has a more mixed success. The movement does lose access to Afghanistan as a source of training camps, as a source of meetings, as a source of planning, as a source of recruits, but its leadership escapes more or less intact. Only one significant leader of Al Qaeda we now understand was killed in the war of late 2001, and Osama bin Laden and his number two, Ayman al-Zawahiri, as well as dozens of senior volunteers in the organization, managed to get away.

Why?

The full history of how Al Qaeda's leaders got away in the fall of 2001 isn't available to us yet because none of them have talked about it, but from what we know, it seems that in the late stages of the war, bin Laden and a lot of his senior Arab and Chechen and Uzbek and Southeast Asian volunteers had gathered in a mountainous area of eastern Afghanistan well known to them from their involvement in the anti-Soviet war of the 1980s. They were under attack from the air by American forces, and they were under attack on the ground from Afghan allies of the United States. They decided to retreat into Pakistan, and they succeeded into escaping, essentially, over the hills into Pakistan.

Why? There was no blocking force on the Pak-Afghan border large and effective enough to stop them as they retreated, and it may have been impossible to establish a large and effective enough blocking of a force to stop such a retreat. These are remote mountains, very underpopulated. To the extent there is a local population, it's all hostile to the United States. The Pakistan paramilitary forces that are the main guards of the borders are themselves tribal levies from the same groups that are sympathetic to Al Qaeda. The Pakistan regular army, which might have been more reliable than some of these tribal levies, was in December of 2001 being drawn away from that border to face India because of a war scare with India at that time.

So the conditions for Al Qaeda's retreat were quite favorable, and the United States did not do the one thing that the Pentagon had within its power to do, which was to move regular U.S. troops into a blocking position behind these mountains. And of course, the commander of this operation, [Gen.] Tommy Franks, was later criticized for not ordering, in particular, the 10th Mountain Division, which was then largely at a base in Uzbekistan and which was trained to fight in conditions such as it would have encountered on these hills, down into a position to block the Al Qaeda retreat.

It's interesting: In his memoir and in interviews, Franks has been asked, "Why didn't you put American troops into position to block this Al Qaeda retreat?" What he said is that he was afraid of the very thing we talked about a few minutes ago: He was afraid that he would inflame local Afghan opinion by putting a big, heavy, occupying American footprint in the heartland of Taliban country and that he was afraid he was just going to make everything worse at that stage.

So in the aftermath of that, many things happened. How does Tenet play the escape?

The escape of Al Qaeda's leadership I don't think was regarded as the CIA's responsibility, because the CIA didn't have the divisions to block the back side of those mountains. To the extent that the CIA was judged to have miscalculated -- and this was a criticism that came later with hindsight -- wasn't voiced much at the time: It was because it had chosen, particularly in the south and east of Afghanistan, local allies who turned out to be less reliable than the Northern Alliance had been in the north. This was, in fact, a problem. The Northern Alliance was motivated to collaborate with the CIA because they were already at war with the Taliban and Al Qaeda, but the Northern Alliance really didn't have strength in eastern and southern Afghanistan, which is the area to which bin Laden and his allies retreated during the war. In those areas, the CIA had much less attractive choices as it looked for local militias to work with, so they started to try to choose anti-Taliban leaders from these areas. But these anti-Taliban leaders hadn't been previously engaged in warfare against the Taliban, were not always unsympathetic to Al Qaeda. They were in too many cases mercenaries who were happy to take the Land Cruisers and the bags of cash on offer to join as the local forces against the Taliban. To be fair, some of them did reject the Taliban's ideology and were prepared to fight against them, but they had no history in this war. They were newcomers, and some of the mercenaries who allied with the CIA and the special forces units that operated in Afghanistan during this period just turned out to be unmotivated, except by greed.

Some I've spoken with say that the CIA didn't send in the right guys with the expertise.

The truth is there wasn't a lot of Afghan expertise in the CIA or in any other quarter of the government. When the Counterterrorist Center put together its teams to go in and work with the Northern Alliance, it rather quickly ran through the Afghan expertise that it had. But the war demanded many more CIA officers and special forces commanders on the ground than the CIA had Afghan experts, so rather quickly, they were drafting volunteers who had very little prior experience with the terrain, with the tribes, with the language. Undoubtedly that interfered with the operations, especially where the local Afghan allies turned out to be less reliable than they were in the north.

You know, it was really the first time I remember CIA guys being on TV and interrogating people and arresting people. You really saw on television a kind of peek into the culture.

I think in some sense, both Bush and Tenet made a decision to put the CIA in the lead in the campaign against Al Qaeda, at least initially, partly because the CIA had a plan and nobody else did, and partly because a terrorist adversary is a natural target for an intelligence service.

And for Tenet -- I don't mean that he was operating cynically, but this was an opportunity to complete the revitalization and rebuilding of the agency that he had set out to accomplish in 1997. Now he had the money that he previously had been begging for and often failed to receive. Now he had the role in American national security that hearkened back to the glory days of the CIA in the '50s and the '60s and even during the Second World War.

It was now reasonable for him to believe for the first time that he could recruit some of the best and brightest young people in the United States, that he could revive a sense of purpose at the agency. So Tenet being a leader with a political touch, and with also a touch for media and public relations, wanted to take advantage of that opportunity and did.

And then what happens? ... Was the turn to Iraq just natural and inevitable, and that they [CIA] would find themselves rolling downhill from that high point?

Well, if the CIA had not contributed to or overseen a failed assessment of Iraq's weapons of mass destruction, if it had not collaborated so enthusiastically with the war planning that turned out to be inadequate, then the sense of triumphalism that I think many at the agency felt after the Afghan war might have survived, at least to some extent.

But the successes that the agency had built up politically in its relationship with the White House, in its reputation in the country, quickly ran away as the Iraq war unfolded in the way it did, and as the intelligence assessments that provided the primary rationale for the invasion turned out to be wrong.

And what is the feeling at the agency when they first get a whiff that Iraq's back in the air? Do they feel the war on terror slipping away from them?

There were two kinds of reactions that counterterrorist types in the CIA and elsewhere had as it became clear that the Bush administration was really going to invade Iraq. Among many Al Qaeda specialists, [it] was a sense of dismay: "Wait a second. This isn't the enemy. The enemy is over here. Saddam may be a bad guy, but that's not who attacked us on Sept. 11, and we're not done getting after these guys." So one source of dissent was essentially an intellectual dissent: You are attacking the wrong enemy.

But even among those who supported the invasion of Iraq, who agreed with the White House's sense that Saddam had to be stripped of the weapons of mass destruction it was suspected he possessed, they recognized and feared what was coming, because they knew that it would drain resources away from the campaign against Al Qaeda. For all of the increased budgeting, for all of the recruitment of young people that had been unleashed after Sept. 11, the amount of competent, experienced resources that the CIA possessed in its Directorate of Operations in, say, late 2002, early 2003 was finite, and everybody on the inside understood they were finite, and that what was going to happen when you started a big old war in the heart of the Middle East was that those finite resources were going to be shifted over to the Iraq campaign.

As the Iraq occupation went sour, more and more resources were poured into reconstruction and counterinsurgency, and all kinds of officers in the Directorate of Operations who had previously been focused on Afghanistan were rotated into Iraq to contribute to that effort because it was the number one priority once the invasion was completed.

There was certainly a sense among the CIA Afghan types that they had been left with diminished resources and that this had hurt their ability to work in Afghanistan in the period of 2003 and 2004.

Even then, as the Bush Doctrine is beginning to emerge and we seem to be being prepared for a war with Iraq, does Tenet know that whatever the capital he has built, it is slipping? What's happening to him through the spring and summer of 2002 that you know of?

Tenet had an opportunity after the Afghan war to declare victory and go home. He had succeeded in the objectives he'd set for himself when he took over in the sense that he'd rebuilt the agency's budgets, its staff, and he rebuilt its morale. But if he thought about stepping aside in 2002, he didn't. One can imagine that he was caught up in the momentum of this historical campaign, that there were a lot of CIA officers out in the field every day cabling back to the director's office, "I'm after this cell tomorrow; I broke into that bank account yesterday." There was a great deal of activity under the director's supervision that was rote. It was a sense of momentum and achievement that I suspect was very difficult to walk away from.

And these stories of the morning briefing and Tenet bringing in, you know, "We got that bad guy; we broke into that bank account," and Bush checking off the list, that must have been --

Must have been heady stuff. Yes, I mean a lot of what the president was interested in, at least in the first six or 12 months of the campaign against Al Qaeda, were operations that the CIA was carrying out. The CIA was in the lead on a lot of this. After the invasion of Iraq, that began to change, and the Pentagon became much more central to the president's agenda.

Was the CIA prepared to answer the call in terms of providing intelligence to the president of the United States about the weapons of mass destruction and the connection of Al Qaeda to Saddam Hussein?

Clearly not. I think it was reasonably well known among the Bush Cabinet, and even in the late Clinton administration, that American intelligence collection about several important hard targets, including Iraq, but also Iran and North Korea, was inadequate. This was one reason why the debate over the quality of particular bits of intelligence was so lively, because there was a great sense of shared understanding that we didn't have much, and so we're really kind of pulling fragments and trying to connect things together.

And of course, the director of central intelligence wears two hats. He runs the CIA, but he's also responsible for the whole intelligence community and its analytical assessments about matters such as whether or not Iraq possesses an active weapons-of-mass-destruction program. So the fragments that were being evaluated in the fall of 2002 and in early 2003 came not just from the CIA's collection and that of its liaison services, but also from the National Security Agency, Defense Intelligence Agency [DIA], the intelligence services of the individual military services, the Army, the Navy and the Air Force.

You put all of this together in a pot, and the CIA director is the one who decides how much of this ingredient and how much of that. This put Tenet in a particularly visible position within the intelligence community and at the White House. It was his responsibility.

And he's got to be feeling the pressure to discover. I mean, in the Defense Department, they create the PCTEG [Policy Counterterrorism Evaluation Group], [Undersecretary of Defense for Policy Douglas] Feith['s] operations that gathered its own intelligence and was moving it up to the vice president's office. He's got to be sensing that he's in some competition with others.

I don't know about that. At the heart of it I think is that there was a conviction among nearly all of the most expert analysts in the intelligence community, and not only at the CIA, that Saddam really did have a hidden weapons-of-mass-destruction program. This was a conviction that was conditioned by experience, particularly the experience in Iraq after the first Gulf War, when they discovered that they had vastly underestimated Saddam's WMD program, and particularly his nuclear program. So there was a conviction that he really had this stuff. Once you more or less make up your mind that something is so, then you're going to channel your actions in service of that conviction, and I think that's what happened here.

Tenet's role was to be the communicator of the intelligence community's view, the communicator directly to the White House. That was his responsibility in the way the intelligence community was then organized. So he was speaking not only for himself in his sort of charismatic, casual style that he'd developed with the president, but he was also speaking for the broader community of analysts. It equally was his responsibility to wrestle to the ground the controversies that developed within that community about particular questions. In that role, he was in the run-up to the invasion of Iraq, overseeing disputes between particularly the Pentagon and the CIA, but sometimes the Pentagon and the Department of Energy or the State Department's Bureau of Intelligence and Research [INR] over specific fact questions.

What were those disputes about? They were purely disputes about fact questions? They were not anything more than that?

In that system, it was the responsibility of the director of central intelligence to present a unified assessment, with dissents, about crucial questions that the president asked the intelligence community to address. In the case of Iraq, the White House had asked the intelligence community to assess the strength and current activities of several key weapons programs in Iraq -- nuclear weapons, biological weapons, chemical weapons and delivery systems, meaning primarily missiles, but also other covert delivery systems like unmanned aircraft.

Each of these questions had a rich history in the intelligence community dating back through the period of U.N. inspections, wrestling with Saddam over inventories, interviews with scientists involved in these programs. But the information available about each of these crucial questions had more or less been cut off when the inspection regime came to an end, so you no longer had current reports via the U.N. and associated inspectors about questions that had been reasonably well documented through the mid- to late '90s. For the analysts, I think, it was the lack of current information that was the most daunting.

And the extent to which we can look back on it now, and we say the whole process was thoroughly politicized, was it actually? To what extent was that determinative in terms of the way nuance was handled?

Well, there are two different questions. One was how much politicization was there in the process by which professional, nonpartisan, bureaucratic analysts reached their judgments about fact questions, and then how much politicization was there in the process by which the Bush Cabinet sold those judgments to the public and the international community as a sound rationale for the invasion of Iraq?

My own sense is that the politicization was much greater in the second endeavor than in the first; that in the first there was pressure that some analysts felt, but the intelligence analysts that I've known are like university professors. They are innately stubborn, and they have a sense of always being besieged by their supervisors and always being picked on by their bureaucratic rivals. If they're experienced and they're really experts, after a time they just become ornery and they stick to their guns, even when people are after them.

When the full record of the debates that produced this judgment is available, if it's ever available, we'll see a lot of that going on. Now, at the very top, you have Tenet and his group making the final judgments about how to present the consensus, how to manage the dissents, how to resolve debates that really no reasonable person could resolve with 100 percent confidence. They are judgments, and Tenet, while not by any means a partisan figure in the Bush administration -- he's a political actor at this stage, and he has a very close relationship with the president, and he serves at the president's pleasure.

So does that mean that colors his analysis?

I myself don't think you can make that leap -- and I say that because if you're talking about the judgment about weapons of mass destruction in Iraq, it was a conviction that was held among the heads of intelligence services who had nothing but contempt for the Bush administration, including the French, and Tenet knew that. So Tenet's judgment that this evidence was reliable was reinforced not only by his direct contact with the president, but by his relationships with other intelligence services that he would have reason to think weren't spinning.

Then, having reached that conviction, he did what he does well, which is that he communicated more forcefully, more clearly, more publicly -- at some instances such as when he sat behind [then-Secretary of State] Colin Powell at the U.N. presentation -- than anybody else. He made himself the public face, in effect, of the consensus that was very complex and involved quite a lot of different actors.

But in reaching that conviction, my own estimate would be that he was influenced less by the Republican true believers at the White House and more by the agreement that he found in his judgments in Europe and among allied intelligence services.

So when the president looks at the NIE [National Intelligence Estimate] and says, "This is the best you've got?," and Tenet says, "Slam dunk," what's at play? What's happening? What does that reveal about him?

Well, Tenet himself has said that he wishes he'd never used the words "slam dunk." But those words reflect his style of speech and to some extent his effectiveness with the president and with the Cabinet. He was able to distill facts and ideas into salty, everyday language that they found attractive. Particularly he had a yen for sports metaphors.

But I think it says something larger than that, which is by the time the president was wrestling with the NIE and with the campaign of diplomacy and public communication that was going to be required in the run-up to the invasion of Iraq, George Tenet was one of the people around him who could offer the kind of common-sense language that the president would innately trust, that they were both men who valued plain speech and who wanted a yes or no answer to many questions.

And one could argue it was also his flaw?

I saw it in a different way. The conviction he held at the time that he said "Slam dunk" was wrong. Was it wrong because of the character flaw? I don't think so. I think it was wrong because the whole analytical process that led him, alone among people, to use the words "slam dunk" was itself flawed. He has a role of responsibility in that flawed analytical process. You might say that his is the most important role, but it was not the only important role, and the conviction that he developed was not a product of blind loyalty to the president; it wasn't a product of arrogance. It was a product of a very flawed, complicated process that he had bought into to the extent that "slam dunk" -- I mean, it was a natural expression of his character. But it wasn't the decisive decision; it wasn't the turning point for him. It was just another moment along the road.

With the basis of all that we know now, it's possible to imagine a director of central intelligence who, presented with the responsibility to oversee this analytical process and presented with the frustrating fragments of information and advocacy on behalf of untrustworthy Iraqi exiles that Tenet faced, would have said: "You know what? I don't trust this judgment. I don't think that it's well enough grounded. I don't think that we have examined alternative scenarios carefully enough. What if Saddam is bluffing?"

That's the question that nobody asked. Tenet didn't ask it; nobody else did, either. And a brilliant director of central intelligence who history today would remember fondly would have stroked his chin at one of these crucial meetings and said, "What if it's all a bluff?" Nobody did.

What is the moment at which Director Tenet, if there is a moment, as [Washington Post Pentagon correspondent] Tom Ricks would say, "drinks the Kool-Aid" on Iraq? Is there ever a moment?

I think the moment where he had an opportunity to step aside was after the Afghan campaign. Once he committed to continue in office as it became plain that an invasion of Iraq was next, then throughout the entire narrative by which that invasion was rationalized with intelligence assessments and then advocated before the American public and the international community on the basis of those intelligence assessments, Tenet became a full and articulate partner of that decision. I presume that, like many supporters of the invasion of Iraq, whether they're Democratic senators like Hillary Clinton or the vice president of the United States, that since then, Tenet has had occasion to reflect on the convictions that led him to support the invasion and to endorse the intelligence.

He was responsible more than anyone else for one thing, which was the quality and reliability of the NIE about the threat Iraq posed to the United States and the Western world, and for that matter the Middle East. And the truth is that the NIE that he oversaw was deeply flawed. That has to be at some very important level his responsibility. It is also the responsibility of the intelligence community and the allied intelligence services and the political ideologues that also participated in that assessment.

Imagine what it must have been like for him as the intelligence is being proven to be wrong. Can you imagine what that was like?

The thing is, Tenet had a support network within the Bush administration as he was going through this experience. He was close to [then-Deputy Secretary Richard] Armitage and Powell at the State Department; he had friends and allies on Capitol Hill, including important allies in the Republican Party. So he wasn't fighting a lonely campaign, but I think as things went bad in Iraq, as the estimates of Iraq's WMD program were proven to be so flawed as the invasion yielded to insurgency and a troubled occupation, and as the 2004 election approached, the administration began to divide within its own ranks, with Iraq as the galvanizing issue. As that division occurred, I think Tenet naturally would have seen himself as in the Powell-Armitage camp, standing across from what they might have called the neoconservative camp.

And the meaning of that?

Well, some of it was about the blame game, and some of it was about policy. The blame game became particularly acute as the 2004 election unfolded, and you had a nominee in the opposition party, John Kerry, daily criticizing the conduct of the war in Iraq and to some extent the assumptions that produced the war. So the Bush administration and the president, as he battled for his re-election, had to decide how to reply to this criticism. Would he take full responsibility for everything that had appeared to go wrong in Iraq, or would he attempt to answer some of the charges by blaming flawed advisers or flawed advice? I think this was a question that not only the president himself confronted every day, but so did every operator in his election campaign in the Republican Party.

As 2004 went along, there were certainly some in the Republican Party who thought the right answer for the president was to blame the CIA for giving him a flawed estimate in the first place. Why should it be the president's fault that the CIA told him that Saddam had weapons of mass destruction? There was certainly a school within the Republican Party and within Bush's campaign who believed that this was just wise politics, regardless of the substance. If George Tenet had to fall on his shield so that the president could be elected, so be it. That's what directors of CIA do, isn't it? And after all, he's a registered Democrat, isn't he?

And that's in fact what happened.

That's in fact what happened. Then Tenet would have seen this developing, as well he did, as particular controversies became public during 2004; for instance, the controversy over [Ambassador] Joe Wilson's trip to Niger and the assessment that Niger might have been supplying uranium to Iraq. As those controversies unfolded in the summer of 2004, anyone reading what the White House was saying, both on the record and on the background, would have understood that there was a substantial group now around the president that had decided to manage the Iraq issue in the campaign at least in part by blaming the CIA for getting it all wrong.

I don't know what George Tenet felt as he saw that unfold, but I can imagine that he was dismayed and increasingly resentful that he was being singled out for blame. At the same time, he's such an operator and such a student of Washington that surely he understood what was happening; that he was being asked, in effect, to fall on his shield so that the president could be re-elected.

In the end, was he pushed, or did he jump?

I think that he understood that his position was untenable and that it was only a matter of time and that he would serve both himself and the president best if he resigned. I suspect that he felt a great sense of personal loyalty to the president himself, with whom he had had a very successful one-on-one relationship. To the extent that he had questions about whether or not the president's advisers were attempting to undermine him in some way that was nefarious, I doubt that he would have blamed the president personally for that activity. I think he would have recognized by the summer of 2004 that there was a coincidence of interests; it was in his interest to get out, and it was also in the interest of the Bush administration that he leave.

And the Medal of Freedom was a consolation prize?

I don't know anything about that. I presume that for President Bush, it was a signal that he wasn't making Tenet a scapegoat. It would be the natural thing to do, right? I mean, you've seen this episode of "I, Claudius." You know? You put the knife in one side and the medal on the other side and that's politics.

And then Porter Goss was appointed.

I think the CIA lost a substantial piece of its history and its power after Porter Goss came over and the intelligence reorganization was inactive. What made the CIA different from any other agency in the intelligence community was the direct contact between the Director of the CIA and the President of the United States and the way that the legal framework governing intelligence activity in this country contemplated a close, direct relationship between the CIA and the President. That's now gone. The national intelligence director [John Negroponte] meets the president every day. The national intelligence director decides what overnight developments the president needs to hear about, what intelligence is reliable. The CIA is now more analogous to the FBI in the sense that it is one step removed from direct contact with the president and is more likely to receive instructions from an intermediary than it is to directly influence the president's thinking about matters of foreign policy and national security.

And the policy impact of that?

It's hard to judge so early. But I think there's one other thing that worth trying to articulate, given your interest in this narrative. What was going on in the summer of 2004 when Tenet was pressured to resign? There was a substantial group of people in the White House who believed that elements of the Central Intelligence Agency were opposed to the president's re-election, and might even be running a quasi-covert action to prevent the President's re-election. One reason why some Republicans contemplated such a historically sort of stunning idea that in the United States the CIA might be operating against its own president was because of the permission that the CIA granted to Michael Scheuer to publish his book, Imperial Hubris, in the summer of 2004. Permission that the CIA described as the fairly routine and well precedented [sic] use of its authority to allow officers in good standing to publish non-classified works, but which the White House interpreted as part of a subtle campaign to undermine the president's standing. After all, in this book Michael Scheuer was critical of the invasion of Iraq and of the Bush Administration's conduct of the war on terrorism. And while his views are complex, the headline of his book, as it was received in the heat of this political campaign, was that the Bush Administration had not done well in responding to the terrorist threat.

So I think that Porter Goss was sent out there not just to reform the mechanisms of the intelligence community that had failed in the assessment of the Iraqi weapons of mass destruction, but also to assert political control on behalf of the Republican Party eight weeks before an election in case there was any reason to be concerned that the CIA was somehow opposed to the president's re-election as an institution.