Paul Pillar served for 30 years as an analyst at the CIA, finishing his career as the national intelligence officer for the Near East and South Asia. In that role, he was involved in compiling the intelligence leading up to the Iraq war and afterwards. Since retiring, Pillar has written an article in Foreign Affairs criticizing what he saw as the politicization of intelligence in the run-up to the Iraq war. Here, Pillar expands on that point and discusses the writing of the flawed 2002 National Intelligence Estimate (NIE) on Iraq's weapons-of-mass-destruction program; the faulty intelligence that made it into President Bush's 2003 State of the Union address; the 2004 National Intelligence Estimate on the aftermath of the Iraq war, which was leaked in the middle of the presidential campaign; his opinions on controversial practices such as renditions and torture; and changes at the CIA since the resignation of George Tenet.

• Some highlights from this interview
• The problem of politicizing intelligence
• Intelligence in the run-up to the Iraq war
• The 2004 NIE and politics at the CIA
• Thoughts on renditions and torture

Give me just a sense of who you were [at the CIA].

Most of my career of 28 years was as an analyst or manager of analysts. For the most part, that's done in the Directorate of Intelligence for the CIA, although you would have analysts elsewhere in the intelligence community -- agencies like the Defense Intelligence Agency [DIA], or the Bureau of Intelligence and Research in the State Department [INR]. ... I worked on the National Intelligence Council two different times, once in a junior capacity, and then the last five years of my career, I had the privilege to be the national intelligence officer for the Near East and South Asia, which has the responsibility of coordinating intelligence of those regions across the whole [intelligence] community. Through most of the 1990s, I worked on counterterrorism as head of analysis and subsequently as deputy chief in the Counterterrorist Center [CTC]. ...

What makes a good analyst?

A commitment to not just the truth, but a willingness to go wherever the evidence and the reporting takes you, to have no preconceptions and to keep striking away at that path, to try to get you to the truth. ...

I think about the vice president [Dick Cheney] and the way he and some of these other fellows talk about the CIA. They say, "Soft, squishy, don't get an answer, 'Don't know,' 'I wish I had more direction.'" The way that I read the story, they started feeling this way going all the way back to the Cold War. Give me a response to that criticism.

The intelligence community [is] always torn in two opposite directions in this regard. One is the appetite by the consumers to have clear, definite judgments and not, as people often complain about economists, having two hands: on the one hand this; on the one hand that. People like to have one-armed intelligence analysts just like they like to have one-armed economists. But ... the other pressure is not to come to closure too soon, not to be more definite than the evidence or the reporting allows you to be. Indeed, as we've seen in the whole Iraq weapons story, there is a hazard in being too definite, or at least giving the impression of being too definite if the evidence and the underlying reporting is really more uncertain.

There's no way that any intelligence officer or any intelligence agency is ever going to resolve those conflicting pressures. They're always going to feel it, and there will always be the appetite from the consumer for a more definite judgment.

Let's talk about the analysis just [before] the Persian Gulf War, Desert Storm [1991]. What did you know? What didn't you know? ... What was the analysis?

On the Iraq weapons program, there were judgments that unconventional weapons programs existed. One of the revelations after the invasion and after the inspections began in Iraq was that some of those programs had gone farther than had been believed. The intelligence community had undershot, if you will, in its assessment of just how far along, especially on the nuclear program, the Iraqis had been. ...

And the implications of a discovery like that in political terms were what, from what you could tell?

Well, the lesson that clearly was drawn ... is "My goodness, this shows us how much we might not know." And as people like the vice president and others repeatedly reminded in the lead-up to the Operation Iraqi Freedom, "We don't know what we don't know." ...

When George W. Bush is inaugurated, what's the expectation for how these guys are going to handle the use of the intelligence services of the United States government?

Two thoughts came to mind. One was this was a president who was elected ... not primarily on foreign policy issues. For the first eight and a half months before 9/11, their priorities were not in the foreign policy area. On the other hand, ... the elder President Bush, who was a former director of central intelligence, was known to be one of the best consumers of intelligence in the sense of reading it, listening to it, paying a lot of attention to it, at least in the sense of the amount of time spent with intelligence officers. There was probably some hopefulness that the son would follow in the footsteps of the father with regard to using the product.

How about the vice president?

There wasn't much of a basis for someone like myself to have a prior expectation, except his service on the House Intelligence Committee earlier on, so some of us had some interaction with him in that regard, as someone to be briefed, seemingly a good listener -- although any time you ask an intelligence analyst or an intelligence officer what they feel like interacting with a very senior consumer, having face time with a senior consumer is always wonderful, whether they're listening or not.

Why?

That is part of how you measure success. One of the worst things for an intelligence agency or an intelligence officer is to be irrelevant. It's better even to say something and be challenged than to not have any attention paid to what you're doing at all.

And did you find him attentive in that way?

In my personal experiences early on, yes. ...

And [former CIA Director] George Tenet?

George Tenet enjoys much respect and affection on the part of the great majority of people at the CIA, and many of us throughout the community as well. I'd have to say he's a very tough act to follow. ...

When he gets the job, ... what's the buzz at the CIA about him?

Well, he had already been deputy director for a couple of years, and before that he had been the senior intelligence person at the National Security Council [NSC], and before that, we had had much interaction with him as the staff director of the Senate [Select] Intelligence Committee [SSCI]. George Tenet was very much a known quantity: ... very gregarious, very engaging, very easy to deal with in many ways. I'd say the overall response was very positive.

And how important was it to you, if at all, that his gregariousness ... actually yielded him being close to the president?

It's always going to be very important to have that kind of access, ... [but] it gets back to the question of, "Are people just giving you face time, or are they actually listening to what you have to say?" In the end, though, it's always going to be a call for the policy-maker. ... Although intelligence officers are sometimes asked, "Well, what's your view on this? Which do you recommend, this policy option or that policy option?," that's not the job for either DCI [director of central intelligence] or even a junior officer.

There have been people who have come into this room and said to me, "It's a serious problem that George Tenet got so much face time with the president of the United States, because he couldn't speak the truth to power."

That is a serious issue. For decades, intelligence officers ... have argued and debated about this, about what is the ideal degree of closeness to have. On the one hand, if you're too close, your objectivity is questioned. On the other hand, if you're not close enough, you risk being irrelevant. You have to strike a balance. ...

What do you think in this case?

I think, well, the pendulum over the last two decades has swung more in favor of closeness and away from distance. It has probably swung too far. ... The intelligence community had other things to say about Iraq which would not necessarily have had the implication that war was the appropriate response. Overall, I can't really say that the intelligence community profited from its very close access in the sense of influencing the policy. I think the policy was driven clearly by other things. ...

Let's pretend for a minute that it's the day before they're inaugurated. ... What [do] you think should have been the priority right before these guys got started, based on what the CIA was bringing forward?

I would refer you to the public statements that are made every year by the director of central intelligence to Congress. ... If you looked at the statements that George Tenet made in 1999, in 2000, in 2001, he placed terrorism and specifically Al Qaeda at the top of the list of threats that he considered important. I agreed with those judgments.

Did you guys have any sense of how welcome that information was to the incoming administration?

I don't think it's a matter of welcome or not welcome. ... It's a matter of priority; it's a matter of resources. Take the issue of aviation security, for example. ... We did a National Intelligence Estimate [NIE] in 1995 on foreign terrorist threat to the United States; it highlighted the threat to transportation and civil aviation in particular. ... Should more have been done in the area of aviation security? Of course, in retrospect. But it always gets down to dollars and a reluctance on this part by one industry not to spend a lot of money before a disaster happens, which is of course our pattern in this country time and time again. ...

[Then-Deputy Secretary of Defense Paul] Wolfowitz and most of those New American Century [neoconservative] people by '97, '98, are out there agitating for finishing [what] President George H.W. Bush couldn't [in Iraq]. ... What were you making of those allegations then in the '90s?

The basic allegation that Saddam's regime was responsible for all manner of things, including the World Trade Center in 1993 and just about any other terrorism under the sun, ... was simply not credible at all. It's not as if the accusations or the line of argument had been dismissed out of hand; they hadn't been. ... They were looked at carefully, and there simply wasn't anything there.

Now, Iraq, of course, was doing other things with regard to sponsorship of terrorism, ... and they rightly were on the state-sponsored terrorism list. But what we saw happen after 9/11, trying to put together this thesis of some sort of alliance with Al Qaeda, was a manufactured issue.

Could you feel that there was among this group, the neocons, that there was a real force out there pushing to overthrow Saddam?

It was there, but it wasn't until 9/11 afforded the opportunity by making the American public suddenly much more militant that the prospect of actually going to war became real. Before 9/11, they didn't have enough to hang on. The desire was there, but the prospect was not. ...

The very first National Security Council meeting was a lot of talk about how to go to war with Iraq, how to get rid of Saddam Hussein, and not about [Osama] bin Laden and other things. Did you know about that?

No, but it was not a surprise, the focus on Iraq. All you had to do was read writings, hear speeches by some of the people who had been influential in the administration, to know that was a very high interest.

So did you know then that there was going to come a moment where people who knew what you knew, thought what you thought about Saddam Hussein and Iraq, were going to come up against a handful of fellows in the Bush administration? ...

Not really until early 2002, when the road to war became quite clear, and the very strong, energetic campaign to try to link the Saddam Hussein regime to the whole 9/11 issue, if not the 9/11 attack itself. ... Fortunately, I was not at that time in the counterterrorism business, so there were many colleagues of mine back in my old organization, the Counterterrorist Center, who had to face [it] in a more acute form than I did. ...

How does the imperative make its way to the Central Intelligence Agency?

Many different ways: The questions every morning; the tasks; the requests to look into this angle one more time, turn over that rock again. If you didn't find anything last week, look again to see if there's something there about that connection. That's the main way. And then, of course, intelligence analysts listen to speeches and read newspapers, too, and they can see what senior policy-makers are saying publicly. They can make their own judgments about, well, does this really conform, or does it not conform with what our analytic judgments are? It was a clear tension.

Do you remember what were they talking about, more specific than just "We're looking for a link"?

This has come out of things like articles in The Weekly Standard and so on. It's basically laundry lists of raw reporting in which if you make certain inferences. There's some kind of connection. There was a phone call made, or there was a meeting held. Indeed, there were some meetings held back early in the 1990s between individuals associated with the Saddam regime [and] individuals associated with Al Qaeda. They were mainly in the nature of feeling each other out, certainly nothing that came close to an alliance or a sponsorship or cooperative relationship.

But in the world of international terrorism, ... it's possible to link just about anyone to just about anyone else. If you dig hard enough for enough circumstantial reporting, meetings, phone calls, names that come up in the same breath, travel to the same place at about the same time, you and I could be seen to have a relationship, probably, if someone worked hard enough at linking us through such things like that.

That's very different from forming an analytic judgment about what is the nature of the relationship we had. ... It becomes very easy to use raw reporting to try to make the case that there is a substantial relationship, losing sight of the fact that relationships can be ones of suspicion or even competition, as well as ones of cooperation. In the case of the one we're talking about, Al Qaeda and Saddam Hussein's Iraq, [it] was more the former: suspicion of feeling each other out.

So we're talking about the difficulty of objective analysis in the face of policy-makers who already have their minds made up? ...

The ideal model, the proper relationship, is one in which the intelligence community focuses on certain subjects, either because the policy-maker is interested in the subjects -- that is, not particular lines of argument -- or because in the judgment of the intelligence officers, these are important issues that could affect or threaten U.S. national interests. Once they choose those topics by either of those criteria, they take them wherever the evidence and the analysis takes them without regard to particular policy references.

That's the ideal textbook model. When policy-makers make a decision first, then use intelligence to support the decision, that basically stands the model upside down.

What did you think would happen with the agency after 9/11 happened?

That there would be a major increase in resources devoted specifically to counterterrorism. That's the natural reaction to any kind of disaster. But other than that, it's not a matter of something happening to the agency. The basic mission would stay the same. ... What changed drastically overnight was, of course, the policy priority and the public concern.

Did you feel the agency was at all responsible for it?

No, no. The lack of tactical information is in large part a reflection of the target itself: small groups, highly secretive, very conscious of operational security, extremely difficult to penetrate. With hindsight, there are some miscues we're all aware of in terms of tardiness, of people being placed on watch lists, that sort of thing. How much difference would that have made historically? That's a matter of hypothetical alternative scenarios. But in terms of strategic understanding and strategic warning, it's hard to get more explicit than [what] the head of the intelligence community was even speaking [about] just publicly, let alone in a classified forum to Congress and to the administration leaders.

And when the critics, the vice president included, say, "Where was the human intelligence?" ...

There is always no substitute for the well-placed human source when it comes to counterterrorism; there's also nothing I can think of that's harder to do than place a human source inside a terrorist group. I don't mean at the periphery -- there have been enough of those; I mean in the innermost circles where plans are laid. It's not that people haven't thought of that or haven't worked on that very hard for a long time. It's always been extremely difficult; it always will be, no matter how the intelligence community is organized and no matter how many resources you give it. ...

Fighting the war on terror, from a Central Intelligence Agency perspective, it's a war of information, right? And it's a manhunt. It's not an Army kind of thing?

In part. Intelligence is one of, and only one, of several tools, all of which are very important. ... Related to intelligence is covert action, diplomacy, leaning on other governments to be more cooperative, to do things like arrest or roll up terrorist cells and do all manner of other things that we can't do ourselves is very important. Military action, selectively applied, including ... the overthrow of the Taliban and the rousting of Al Qaeda from its former safe haven, was extremely important.

Financial efforts, going after the money and the very aggressive use and extensive use of our law enforcement capabilities, not just because of the investigative powers that they have, but also because it has always been, and still is, one of the tenets of U.S. counterterrorist policies to "bring terrorists to justice," and that means using our criminal law enforcement system, not as an alternative to military force, ... but as one more of an assortment of tools, all of which have to be used.

... Speaking about the war on terror, [torture and renditions], those two items, comment for me, will you, on them as outgrowths of policy. ... Were you privy to the fact ... that the CIA was involved in this business before you read about them in The Washington Post?

Renditions have always been a major part of the counterterrorist effort. When I was working on counterterrorism in the 1990s, it was a major part of the effort, and not a secret one. You can look at old State Department annual reports and get a listing in some of them on the number of terrorists who have been rendered to one place or another, even if the place isn't mentioned.

An unfortunate thing is how this has been [con]flated with the whole issue of torture and treatment of detainees. You can't separate the two issues entirely, obviously, but the process of rendition, which has been a useful and important process for the handling of some terrorists ... has gotten a bad name because of its involvement with this other issue of how detainees are treated. But we really ought to try to keep them separate. ...

Do you have an opinion about whether we should have been doing those things?

... I think there are serious questions to be raised about the effectiveness of some of the treatment that had been reported.

And the "black sites"? Where do we put these guys?

I really don't have a perspective on that. It's a fair question to ask. If someone is captured, either on a battlefield in Afghanistan or someplace else, you have to do something with them. If you have strong reason to suspect this is a terrorist, you don't just set him free. One of the difficulties involved in bringing people before our ordinary civilian criminal courts of justice is that you want to make sure you've got a strong enough prosecution case that you don't get the worst result, which is to try someone and then he gets acquitted, even if as a matter of intelligence, you still have strong suspicions that he's a terrorist. That's the worst of all worlds.

... The typical detainee, if you're talking about experienced Al Qaeda-type operatives -- not someone who's just a foot soldier who's been scarfed up on the field of battle -- has to be viewed as a formidable opponent in the interrogation room and will use whatever ... counter-interrogation techniques to disguise what he continues to want to be disguised.

Take a guy to Egypt, though, we know what's going to happen to him.

Well, sometimes we know; sometimes we don't. It's not always that clear cut.

Really? Why is he going then?

That's a different question. The why is not to be tortured. The why can include that's where he's wanted for a crime; the why can include this is where, even without coercive techniques, [the] most effective interrogation can take place. There are various reasons. ...

After 9/11 did anything change about the way you did your job?

In my job as national intelligence officer in the areas of South Asia, ... the immediate expectation was that there would be action in Afghanistan, military action. Clearly that meant more work and more assessments to be done, particularly in support of our military. And in fact, that work was done. But then later on, as we got into early 2002, the Iraq side of things shaped up. ...

When do you begin to hear that Iraq is Phase II of the war on terror?

You read the newspapers, and you listen to presidential speeches and vice presidential speeches, and the message starts coming through pretty clearly. It's much more than that; that is to say, [I was] exposed to the same things that all American citizens were exposed to in interpreting where this administration was going, as opposed to any special requests or special tasks laid on the intelligence community. Although indeed, there was a lot of that with regard to trying to support the notion of a cooperative relationship with Al Qaeda.

At least the way the story is told by [journalist] Bob Woodward, is that even then the president is asking Rumsfeld to begin war plans. [Counterterrorism expert] Richard Clarke reports that [there is] lots of conversation about "Let's get ready." That's the fall [of 2001]. You're the Middle East guy for the Central Intelligence Agency, and you're not receiving any formal requests, any requests for information, any intelligence requests? ...

I did not receive any requests from a policy-maker on Iraq until about a year into the war. Now, this is to be distinguished from our military, which was doing a lot of the planning work and making many, many requests, usually at other levels, but some of my colleagues in the National Intelligence Council were involved in supporting the military planning. But no, I did not.

A year into the war meaning when?

Meaning close to mid-2004. ... Remember, things like the infamous National Intelligence Estimate [NIE] on weapons of mass destruction, which was published in the fall of 2002, that was not requested by the administration; that was requested by Congress.

Wait a minute. The Bush administration, the vice president, president, did not ever ask the Central Intelligence Agency for an intelligence assessment?

I want to be precise. From my seat on the National Intelligence Council, ... not just the CIA but the entire intelligence community, ... no, I did not receive any request on Iraq. ... There were other requests being made, from DIA to CIA, particularly by the military as they faced their very awesome job of planning.

But not the vice president?

No, sir.

Not the president? Not the national security adviser?

The course had already been set. ...

What was your take on the weapons-of-mass-destruction program of Saddam Hussein at that time in 2002?

The take was the same as everyone else's take. When I say everyone else, I mean every intelligence service on the globe, and reflecting a general consensus that there was something there in terms of programs and, [in] the case at least of chemical weapons, probably of stockpiles. There was no strong reason from an analyst's point of view to call into basic question the idea that there was something there.

What do you mean?

There were huge collection gaps; there's no question about it. ... There was not a basis for anyone to stick his or her head up and say, "Wait a minute; analytically there might not be anything there," because the prima facie case was yes, there was something there. ...

Nuclear, too?

Nuclear, too. ... And there, the consensus judgment of the community, as is well known, is that Iraq was probably several years away from acquiring a nuclear weapon. ...

[Undersecretary of Defense for Policy Douglas Feith's intelligence group within the Pentagon, the Policy Counterterrorism Evaluation Group (PCTEG)] ... did you know of their existence? ...

I knew they were out there somewhere. I really didn't regard it as part of the intelligence business. I more appropriately thought of it as extensions of a speechwriting staff, since the mission was to come up with material for a public case.

Did you know they were "stovepiping" stuff up into the vice president's office?

After the fact we found out about that. They presented some things, as we found out after the fact, to the White House that were not presented to the counterterrorist analysts and managers at CIA, particularly things that were explicitly critical of the intelligence community for, according to their version, not being able to see this supposed relationship between Iraq and Al Qaeda.

Where did they get their information?

Most of the same places analysts all around town get it. You have to understand, lots of people are reading the same raw traffic on their screens every day. That includes intelligence analysts; it includes people at the NSA [National Security Agency] and Department of Defense and elsewhere. None of this means anything unless it's analyzed, and ... in the very shadowy world of international terrorism, analysis is everything. ... That's why I say that such a unit in the Defense Department under Undersecretary Feith is best thought of as an extension of a speechwriting staff, not as part of the intelligence community. ...

But it was being used for more than just marketing. My sense was it was being used to drive decisions, but maybe those decisions had already been made?

I think the decisions had already been made. ...

[What about the information provided by Ahmad Chalabi's Iraqi National Congress?]

The Iraqi National Congress, or INC, and the information that it provided had a role, but I would not describe it as a major role. The shortcomings, the analysis on weapons of mass destruction involved mostly other sources and the lack of credibility of other sources. Possibly the INC reporting shaped attitudes elsewhere in the government, in the administration, more than they shaped it in the intelligence community. But I would summarize the role of their information as being of fairly modest effect and modest input. ...

Tell me the story of the 2002 NIE. How does it get started? How does it get written? ... Why is it requested? ...

It was requested by members of the Senate Intelligence Committee. This was already the autumn of 2002, so there was going to be a vote in Congress, as everyone knew, on what became known as the Iraq war resolution. The senators wanted a separate input on the issue of weapons of mass destruction from the intelligence community before they reached their vote. The schedule was very compressed. There was only three weeks in which the estimate was done, although it was based, of course, on previously done work. ...

Do you think it was a good document?

Well, in retrospect, there were certainly significant flaws in it, or it reflected significant flaws in the tradecraft, which mainly had to do with insufficient checking of the credibility of sources, which later were revealed. ...

Specifically what happened?

A lot of intelligence analysts were caught up in several things: a previous consensus against which there just wasn't enough intelligence to challenge it; the consensus being that yes, there were programs. The atmosphere in which they were working, in which a policy decision clearly had already been made, in which intelligence was being looked to to support that decision rather to inform decisions yet to be made, was a very important part of the atmosphere.

Exactly how that may have affected the individual judgments of particular analysts, it's impossible to say. It probably had some effect, particularly since most of the shortcomings of the analysis we're talking about come down to matters of nuance, caveat -- whether the language is too strong, that sort of thing. There were many, many opportunities for things to be shaded in the preferred direction rather than in another direction.

Even if they weren't shaded in a sense of being directly politicized, what the analysts knew was we were going to war, so there might have been an erring in the direction of warning. ... The last thing any intelligence officer would want to have happen is American armies invade and they are caught by surprise by something like chemical weapons or biological weapons, so quite possibly, there might have been a bias for that reason as well. ...

And the pressure that we've heard about from the vice president, [Cheney's former Chief of Staff Scooter] Libby, driving up to Langley?

This is something I think has been missed by otherwise very good work by the likes of the commission led by Judge [Laurence H.] Silberman and Sen. [Charles S.] Robb, which produced a very comprehensive and useful report with excellent recommendations. But one thing that's been missed by that inquiry, and the inquiry of the Senate Intelligence Committee, really has to do with the issue of politicization. They reached the judgment that no, there wasn't any evidence of politicization, but basically what they were doing was asking analysts whether their arms had been twisted.

Politicization, real politicization, rarely works that way; that is to say a blatant, crude arm twisting. It's always far more subtle. It would take the form either of these almost subconscious or subliminal adjustments that dozens of analysts might make in the course of phrasing their judgments, making it a little less nuanced, a little less caveated, which I think is the main basis for criticizing the judgments on weapons of mass destruction in Iraq.

It can take the form of ... intelligence assessments that conform with what is known to be the policy having an easier time making it through ... than assessments that don't supply the policy. ... This wasn't an inquiry into how can Iraq threaten the United States; it wasn't an inquiry into what are Al Qaeda sources of support. It instead was basically research in support of a specific line of argument. That, I think, qualifies for the label "politicization," even if analysts are doing their best job to maintain their analytic integrity when they make their individual judgments. ...

I don't know how many people have walked in here in the last couple of days and said to me that the 2002 NIE, forgive me, was just garbage. People who know say, "Poorly thought out, poorly written, wrong; criminal; just terrible; that it caused a war; that it allowed" --

It did not cause a war, as indicated by the fact that the administration didn't even ask for it. And by the way, as has been reported, very few Congressmen even read it. ...

So you feel good about what you did?

Not everything I did. The issues of tradecraft errors, nobody feels good about that. If you're looking at things that I didn't feel good about doing, I would refer to the unclassified "[white] paper" that was laid out. In retrospect, although people who worked on it, including myself, didn't have substantive problems with it at the time, it was clearly requested and published for policy advocacy purposes. This was not informing [a] decision. What was the purpose of it? The purpose was to strengthen the case of going to war with the American public. Is it proper for the intelligence community to publish papers for that purpose? I don't think so, and I regret having had a role in that. ...

The preparation of Secretary Powell for the United Nations -- what do you know about that?

The secretary, once he found out he would have to make this presentation, put an awful lot of his own time and attention into it. I was not directly involved in the preparation, but he had his team of people from the State Department working in close cooperation with many intelligence officers to craft that statement. ...

He gives the presentation. ... What do you think?

That with regard to the weapons of mass destruction, it was probably the best possible public case that could be made, ... but also satisfied Secretary Powell's minimum standards for credibility and soundness of judgment. He, of course, as he's made clear since then, feels a bit used, and I think he deserves to feel that way, particularly with regard to not having been fully apprised of some of the credibility problems with sources that were used. I would be at least as upset as Secretary Powell was.

[Powell's Chief of Staff] Lawrence Wilkerson said to us ... that [often when] Powell would complain about something at Langley, that it was Tenet who said, in effect, "Trust me; we've got this." It's the "slam dunk" line all over again. ... How does [Tenet] make assertions like that? ...

I have absolutely no reason to doubt that the director was calling things as he saw them and was not willfully misleading anyone. The director had reason to be upset and angry, too, because I think he was [let down] by people lower down in his organization who did not do sufficient checking or cross-checking on some of these issues.

Why didn't they?

Partly the strong consensus in terms of what everyone seemed to think there was in Iraq; partly because everyone knew there wasn't a decision riding on this; ... and partly it's just some outright sloppiness, to be quite honest. ...

Were there guys who raised their hands and said: "Hell no, I won't participate in this. Hell no, I'm not going to do it"?

No, and on the weapons-of-mass-destruction issue, there wasn't a basis for saying that. ...

I thought there were serious questions raised about yellowcake and ... the aluminum tubes.

There were serious questions raised about yellowcake, and that's why the people in the intelligence community advised the White House not to use it publicly, and that's why it was not used in the intelligence community's own unclassified product, because there were doubts about the credibility of the report that turned out to be fabricated.

Right. So the president uses it in the State of the Union in January. Why? Who puts it in there? How does it get in there?

It got in there because the people on the National Security Council staff who were supporting the writing of the speech wanted it to be in there. It had rhetorical appeal. ... It made more sense to a lot of the American public than something like the outside diameter of aluminum tubes. ... And you remember how the line was, "The British have so-and-so." So technically, it wasn't the United States vouching for the report.

What do you think of that?

I personally wasn't involved in the dialogue, but in retrospect, clearly it was a mistake to say or do anything that would constitute approval of putting it in, since the clear inference by the otherwise uninformed public would be yes, the U.S. government is vouching for it, even if it was sourced to the British. That was just a technicality that I think went right over the heads of most of the listeners. ...

This thing, though, that you guys didn't want -- you knew Tenet went to war to get it out of the Cincinnati speech; you knew that he personally called to pull it out, and ... the NSC guys stick it back in the State of the Union -- it didn't strike you at all?

... I can't remember exactly what the reaction was, ... but the administration was doing everything it could to make the public case for war, and this was just one additional way to do it. It was just one line, 16 words, in a whole State of the Union address, and not necessarily the one that would even get the most attention. ...

There were, as I understand it, ... lots of other agencies complaining about things that were included in the weapons-of-mass-destruction descriptions. So there was some dissent happening out there?

The dissents were completely recorded and identified in the estimate. ... There were a couple of dissents on these technical issues, ... from the Air Force with regard to whether these unmanned aerial vehicles [UAVs] had the primary mission of being attack as opposed to reconnaissance. The other issue concerned the purpose of the aluminum tubes, in which the Department of Energy's judgment was they were probably not for enrichment cascades.

One of the interesting things is both the Department of Energy and the Air Force nonetheless signed on to the main judgments about the first instance -- Iraq is reconstituting a nuclear weapons program -- and the second instance -- that they have a biological weapons program as well. So even though these specialized agencies dissented on the technical issue, they concurred in the mainline judgments.

When do you know that the weapons-of-mass-destruction case is wrong?

There wasn't any one eureka moment. It's through the summer of 2003, as U.S. coalition troops were going in more and more places and not finding things. It was more of a gradual process. After the first few months, it got to be pretty clear that there wasn't stuff there that was postulated.

First of all, how?

What I think of first and foremost, personally, is just the wisdom, or lack of it, of the basic policy decision. Of course, there are all kinds of other issues about the intelligence profession -- how it's been used, how it's been treated -- and those are important, too, but this was one of the biggest national security, foreign policy decisions this country has ever made. ... I think it was a mistake. I always thought that, and we're paying a very heavy price for it. ...

Then what comes second is thinking more about ... the proper relationship between intelligence and policy-making. It does disturb me that that relationship was, on this particular issue, basically turned upside down, with the decision coming first and intelligence being looked to more to support a decision already made than to inform a decision yet to be made. ...

Our British allies, I think, have had a much more forthcoming debate and discussion about these things. Robin Cook, former [British] foreign secretary, [who] resigned from the government on differences over Iraq, spoke before he died in almost exactly the same terms as I just did. We've been less forthright in this country, but over the longer term, these issues will acquire some perspective, and I hope we'll learn some lessons from them.

[Was the CIA asked to provide analysis of the aftermath of invading Iraq?]

... Although the intelligence community was not asked by the administration, ... the community did on its own initiative, in an effort to be as helpful as it could to those who would have to deal with the aftermath, be it an American general or an American ambassador or an Iraqi prime minister. ... We weren't trying to make specific predictions, since so much of that would depend on what the United States would do, and decisions yet to be made, but rather just to identify what the challenges would be.

... We expected the main challenges would be in the area of political transition in an Iraq which has had a political culture that is not amenable to quick adaptation of democracy. That doesn't mean that Arabs can never be democrats; ... it does mean that the political transition we expected would be long and difficult and turbulent. We expected that the divisions that are all too plain in Iraq among various sectarian and ethnic groups -- the Sunni Arabs, the Shi'a, the Kurds -- would make for a sharply divided society once the dictatorship of Saddam had been taken off the top, and for the significant chance of violence, especially if an occupying force ... did not sit on top of it to keep a full-scale civil war from breaking out.

With regard to attitudes towards the occupying force itself, what was expected was that after the very first few weeks, those attitudes would be shaped above all by how successful or unsuccessful that force would be in providing security and stability and putting Iraq back on the road to prosperity. If it failed to do that in the first few weeks or months, it probably would become the object of hostility. Looking farther out into the region, we also expected that insofar as there would be violence in Iraq, it would serve as a magnet for extremists from elsewhere in the region; that it would give a boost to political Islam, particularly to the extreme version. I think we've pretty much seen all these things take place.

How was that perceived by the administration, which at this moment was heading for war?

There is very little reception to report on. The one senior-level reaction I got to this analysis was to question it on the grounds that it was too pessimistic, that we were not seeing the possibilities. ...

Deaf ears?

From my perspective, yes. There was no delusion on our part of changing the war decision; that had been made well before. We were just trying to be of as much help as we could to those who would have to deal with the aftermath, ... because clearly, it was going to be a big change.

What did it say to you, this response?

It said that the decision had been made well in advance, and that decision was based on a basic optimism about what this expedition would bring. There's no question that from the perspective of late 2005, approaching now three years after the initial invasion, that things haven't gone anywhere near the way the decision makers would have expected them to go. ...

The NIE of August 2004 that you were [involved in writing] created a bunch of waves and [led some to see what they] define as a civil war going on between the administration and the CIA. [What is your stance on] putting that message out, and how it was received?

Unfortunately, all this work gets done in Washington, which is a political town, and when you have an election campaign going on, everything that statement means is double or triple. One of the more unfortunate things that we saw happening in 2004 [was] leaks of intelligence assessments that were speaking the truth about what was going on in Iraq, and given the hindsight of the last year and a half or so, ... shape up pretty accurately. If it's an unwelcome message to whatever administration is in power, this is a source of consternation.

It's not surprising that ... things [were] said in the editorial pages to the effect that the intelligence community is trying to undermine the policies of the administration. Wasn't true at all. You simply had the intelligence community trying to do its job of calling the situation in Iraq as it saw it, and that's not a matter of undermining anyone's policy. The fact that it was inconvenient for the administration is pretty obvious. ...

You knew it when you wrote it?

Of course, although you hope things aren't going to leak. ... A lot of the accusations that have been made about people in the intelligence community supposedly leaking things, people forget -- well, they forget a couple things. One is, people who work at the CIA, they have things like polygraph examinations. ... If anyone thought about getting into the leak game, they'd have a lot more reason to think twice than somebody in most of the rest of the government who has access to classified information.

The second thing is, people forget that when something leaks, it greatly complicates the job of the intelligence community. ... If estimates do not leak, then we don't have to worry about this public fallout, and ... the intelligence community can do its job of providing the most objective analysis possible a lot better. ...

I don't mean to disagree with you, but when you know you're writing a really tough assessment of the post-Iraq invasion, and you know it's going to be widely disseminated, ... [don't] you know the son of a gun is going to go out?

There are many things that are done where the presumed chance of a leak is very high, yes.

And in this case?

And in this case, ... in fact in any case, it's not grounds for pulling your punches, changing your judgments. That can't be done.

It might even be an opportunity to send a message without having to --

But that would not be the proper purpose, of course.

... It seems to me [there] has been an American view ... [that] something's rotten in Denmark with the CIA and that things need to be changed. Part of that, of course, fell to Tenet ... and resulted in [CIA Director from 2005-2006] Porter Goss being brought [in]. ... Help me understand what it was like when Goss came in. ... Seem like a good choice to you, personally?

I think Mr. Goss deserves much better press than he's gotten, quite frankly. I might add there are some things, just from the standpoint of intelligence analysis, that an elected politician, or someone who's had that experience, ... can understand better than most of us who are career bureaucrat intelligence officers can understand. ...

But this is an agency we don't want to politicize?

That's a separate issue. We're not talking about pushing things in one political direction rather than another. I'm just talking about what kinds of backgrounds can make a person a suitable director. ... One of the former directors of central intelligence who is still highly regarded today out at Langley as one of the best ones was George H.W. Bush. ...

When you first started talking to me here, you said what the mission was, what the CIA mission was. What did you say?

The basic mission of intelligence is to provide the most objective analysis of and collection on realities out there overseas that could affect the U.S. national security interests.

Did that happen? Honestly, did that happen in the war on Iraq?

On intelligence production related to Iraq? For the most part it did, yes. ...We're all familiar with what was erroneous, what was wrong about the weapons of mass destruction, and there have been at least one or two instances we've talked about on the terrorism side, and ... the different, subtle ways in which politicization can work its way in. ... Looking at the intelligence product as a whole, for the most part it was, yes.

... Somebody told me: "Harebrained schemes get started all the time. They filter up, but one of the grownups usually intervenes and says, 'Hold it; let's think about this again,' or, 'Let's stop this.' Or analysts step up and say, 'Wait a minute.'" ...

In the end, there's not something preventing our elected leaders from engaging in folly. If you do have an intelligence service that is in a position to render judgments, including public judgments, more independently, that will help. It won't be a foolproof solution by any stretch of the imagination. The role of Congress obviously is important, and some members -- I would mention, for example, former Sen. [Bob] Graham, former chairman of the Senate Intelligence Committee, who I think was very forthright on these issues and was saying many of the right things. Their role is very important, but as we've seen, only so much can be done by some members.

Can't stop a juggernaut.

That's right. And this was a juggernaut.