Philip Zelikow Oral History Part II

This is the second Philip Zelikow interview as a part of the George W. Bush Oral History Project. To address your concerns, I thought the first interview went wonderfully, so I hope you won’t alter your approach to answering questions. We’ll feel free to intervene if we want to sharpen items or push you ahead.

We got up through positioning you at the 9/11 Commission before, and there were a couple of staff hirings that you addressed. Your own, you mentioned Dan Marcus being brought in as the general counsel. That’s where we are in the timeline. Before I move beyond that, though, I want to turn to my colleagues and see if you’ve got any questions from the prior period that you want to go back to. Mel, was there anything that we didn’t touch on before that you’d like to deal with before we get into the 9/11 Commission fully?

No.

Were there any other key personnel decisions that you had to make at the outset in terms of setting up the Commission, or personnel decisions that became important as the work of the Commission unfolded?

The major decisions at the very outset were the basic structure and nature of the report itself. Then mapping out a research design, an investigation design, and then developing a staff plan and hiring in a way that worked for that design.

OK.

So I mapped out—even before we had offices I was basically working out of my home or out of a hotel room in Washington, and no staff, no offices, nothing. This is in now late January, February of 2003. So then you say, “I need this many teams.” Ultimately I think we originally created eight teams that turned into nine and a half as this developed. I added another team and toward the end created a one-person team.

I had to hire the team leaders as well as make decisions about the rest of the staff to recommend to Tom [Kean] and Lee [Hamilton]. Hiring the team leaders was very important, for obvious reasons. I spent a lot of time thinking about those and recruiting and identifying the people who would be the team leaders and feel very good in general about those decisions. I didn’t make all of them, or come up with all of the ideas. To give one example, one of the most important teams turned out to be headed by John Farmer, who is now the dean of the law school at Rutgers and served in the attorney general’s office in New Jersey. John was recommended by Tom Kean, I think, to me. He ended up handling the emergency response team that looked at the emergency response in both New York and Washington and in effect ran the team we ended up posting to New York City. He did, by the way, a very good job on that.

Are you working completely alone in terms of devising this work plan? Are you getting input from anybody else at this stage?

I’m substantially alone. I’d gathered a lot of paper. There was a gigantic number of résumés that had come in and other input to Tom and Lee that they then forwarded to me. I quickly roped in Stephanie Kaplan, a young woman I’d worked with at the Aspen Institute when I used to direct the Aspen Strategy Group and who is now completing her doctorate at MIT [Massachusetts Institute of Technology]. She is just a superb and very smart staffer. I had some help from her and then increasingly began to reach out and would run everything by my new deputy, Chris Kojm, who was Lee’s suggestion early on. He wanted me to hire a deputy who was his longtime former staffer. I was happy to do that. That turned out to be a fortunate choice from every perspective.

Chris was just a quiet, steady person, knowledgeable. In general it’s good for a manager to try to consciously hire to offset their own weaknesses. I actually have weaknesses, and Chris did help offset some of them. We had large arguments in the early days of the Commission staff about the hiring of the general counsel, but that was ultimately resolved in a way that I found quite satisfactory.

What would you define as your own key weaknesses?

I think probably driving too hard. I wasn’t as mellow then as I am now [laughter], so I might have been a bit brusque or impatient. I felt I needed to really give a lot of hands-on time to monitoring the substance of the investigation, and that left me less time to work on the day-to-day administrative management of the Commission, which was, after all, a small federal agency. Though small, it had three different sets of offices and a multi-million-dollar budget created literally overnight. It was an administrative challenge in a variety of ways.

All the time I spent having to worry about that side of it, which was a lot of time in the early days, was time I couldn’t spend worrying about the substantive quality of the investigation. Those were all things that Chris could help with.

What ended up happening, too, is while there were a lot of staffers who worked great with me, staffers who were annoyed with me about something—and a lot of people would be annoyed with me about something at some point—would then go to Chris, and felt like Chris would listen to their complaints. Chris and my general counsel, Dan Marcus, as well as Stephanie Kaplan, formed our front office. My habit was that they were in the loop about everything.

All of us in the front office knew what everybody else was doing. There had to be a real intimate relation of trust there. That worked, in fact really replicated the kind of relationship Tom and Lee had established with each other at the commissioner level, where you really felt that you couldn’t see any daylight between Kean and Hamilton. We created a situation in the front office where though we came from different political backgrounds—Chris, a longtime Democratic staffer on Capitol Hill; Dan Marcus, a longtime Democratic political appointee, number three in Janet Reno’s Justice Department in the [William J.] Clinton administration—we all, I think, worked together extremely well, again with different strengths and weaknesses.

Were you operating under any budget constraints or any overall framework constraints when you were laying out your original plan?

Yes, the original budget for the Commission—Congress had no idea how much something like this would cost, so they with no particular methodology guesstimated a number like $3 million. When we did the math on how much we would need, we thought we’d need at least four or five times that much money. We ultimately also realized that we would need more time than they had originally allocated.

What was the final cost?

I think we asked for and got, after some difficulty, about $14 million and actually came in under budget.

How much did you wind up spending?

I think somewhere in the $12 to 13 million range, something like that.

But your sense was they would produce what you felt you needed in order to do a thorough investigation.

Yes, we just ran the numbers on the time we needed, the number of staff we needed, because Congress had no precedent for something like this in anyone’s recent memory. So they just made it up. When we came back to them and said we need four or five times this much money, there was some sticker shock, especially from the Republicans. Remember too, a lot of these Republicans had not wanted this Commission to be created. There were a lot of people in the administration who were opposed to the Commission in the first place, so the notion that this thing is already running out of control—

I managed that problem in part because in addition to the people who were opposed, there were people in the administration, including people working in the White House, who wanted to make this work and were talking professionally with us. They could see the way we had worked through this and knew that there had been no science in the original numbers.

There were several early issues like this where folks really had not understood what would be involved in setting up an agency of this kind. We developed very early a very clear conception of what would be involved and then had to socialize all the different stakeholders to that. That turned out to be difficult.

Philip, you mentioned when you were first approached about the possibility of taking the job that you had reflected on previous historical examples that led you to think that maybe this could be an interesting thing to do and a useful public service. Did you delve into any of those historical precedents as you were trying to structure your own thinking about how this would or ought to unfold?

Yes. I looked hard at the Pearl Harbor investigation, which was the most obvious analogy. I looked hard at them right away even before I took the job. I wanted to understand the strengths and weaknesses of the different investigations. I was already pretty familiar with the [Earl] Warren Commission story. In the first weeks I looked into a few others, for instance the [Otto] Kerner Commission, the [William] Rogers Commission that investigated the Challenger launch disaster. There are even a couple of books that have been written about commissions. I checked the academic literature just a little bit to see if there was anything there that would be helpful. I derived a number of valuable design principles from that.

I may have mentioned last time such things as a unitary, not a partisan, staff; clear managerial direction for the Commission, rather than 10 commissioners running in 10 different directions; develop a chain of command. The Commission would deliberate, but then they would give authoritative guidance through Tom and Lee to me and then I in turn could give authoritative guidance to run the staff. But one of the most important and more substantive points early on, which was a strong impression from the Warren Commission, is we have to obtain and be seen to have obtained all relevant information held in the government on this. There can’t be a situation where years later, as in the Warren Commission case, did they know about Operation Mongoose and all the things associated with that. Well, it’s not clear that they did. Then although you can argue that that had no material effect on the Commission’s conclusions, it plainly damaged their public credibility for a generation.

Another more subtle lesson about the Warren Commission, when you get into the details about that story, is the way they politicized the presentation of factual findings. Let me give you an example; the famous example of this is the single-bullet theory. Some of the scholarship on the Warren Commission that I respect, from people like Max Holland, persuades me that the single-bullet theory was fudged in the Warren Commission report because of the political pressure of John Connally, so the Commission fudged and hedged. When you get into it the staff was not at all unclear about what had happened; the staff was very clear about the forensics of this, but the Commission report fudged the forensics in order to give, as someone put it, “Connally his own bullet.”

It turns out that that course was a terrible mistake. I was determined that as the report was developed we needed to be sure that we didn’t commit mistakes of that kind. There are ways—so even though we had lots of arguments later on in the development of the report about what to put in, the arguments tended to affect the interpretation of facts, but I don’t think we had arguments that obliged us to fudge the presentation of the underlying factual record itself.

Generally what happened when somebody would say, “Well, it could be interpreted another way,” or “There is another relevant fact,” we would just tend to dispassionately say, “Well, here’s the evidence for this point, here’s the evidence for another point,” and then just lay that out transparently. In situations where we weren’t that sure ourselves what the right of the matter was we would just be transparent and say, “We’re not sure. Here’s the evidence on this situation.”

In organizing the teams, did you give them a standard charge, “Here’s what I expect you to do; here’s what I need you to do”? How big was each team? How did you relate to them while they were doing their work?

I did give each of the teams a charge. Each team was mapped out with a particular portfolio that was part of the whole overall design. People were hired with that in mind. In general the teams were probably about four to seven people apiece. Some were larger. Team one—that was the al-Qaeda team—ended up splitting into two teams, what we later called team one and team one-A. We split into a general al-Qaeda team to understand the evolution of the organization over time and then a specific team just on this plot itself, the al-Qaeda development of the 9/11 plot per se.

What was the relationship of the teams to the commissioners? Did everything go through you? To what extent did people try to circumvent you and go directly to commissioners and commissioners go directly to teams and vice versa?

Some commissioners early on wanted to have their own direct relation with the team. Indeed a couple of commissioners—well, one commissioner in particular—with a little support, wanted to be able to essentially have each commissioner have their own portfolio and run their own parts of the investigation. Then they would have teams that would effectively report to them, in which case I would become somewhat superfluous.

I thought that down that road would lie madness. I could see where that would go as different commissioners would take things off to follow the particular hobbyhorses to run their part of the investigation. I reacted really strongly to that. I tried to build as strong a wall as possible, which was that all guidance to staff had to go through me and all guidance to me had to come from the Commission as a whole, through Tom and Lee. The commissioners bristled at that assertion on my part. “We need to be able to call team members ourselves and ask them questions.” This is a bargaining process, in effect. I had taken a very strong position that then got whittled around. In effect what happened is then commissioners felt that they could call team members, ask them particular questions. But then I went to the staff and said, “If commissioners contact you and ask you about stuff, you’ve got to tell us about it.”

Let me give you an example of why that is valuable. Let’s say a commissioner calls someone on the intelligence team because they want to know more details about the Kuala Lumpur lost trail in early 2000. Well, that individual—the commissioner and that team member on team two may not even know that people on two other teams are also working on that same issue. To answer the commissioner’s question adequately requires compiling some information from three different teams to give that commissioner an accurate answer to their question.

Another reason to get the staffers to report commissioner contacts is to protect the staffers. See, when a commissioner calls an individual staffer, the staffer has to be really deferential and wants to be really deferential. They may already be thinking, This is a person who may give me a reference after I leave the Commission. You don’t want to create a situation where—to be more precise, you want to minimize the inevitable situation, which is that some staffers become the pet staffers of individual commissioners. As more of that gets seen by other staffers, that staffer is in effect working for that commissioner with that commissioner’s personal agenda.

You begin to get, in effect, partisan divisions inside the staff, and it creates a lot of tension and friction in staff. We had some of this. But we discouraged it and encouraged staff to report their contacts with commissioners so that we could protect them. So if they weren’t giving the answer the commissioner wanted to hear or they didn’t want to answer the commissioner’s question because frankly they didn’t really know the answer yet or their answer was half-baked, they were only partially done, but they needed some top cover to shield them from giving an unsatisfactory answer, we could provide that for them.

There were a lot of different dynamics that we were trying to manage so that the staff could basically concentrate on their work. The commissioners were going to see the results of all this work eventually, but first the work had to be done thoroughly and professionally and vetted in the peer review process we created within the staff. Then by the time the staff would present work to commissioners it had already been pretty thoroughly scrubbed internally in a sometimes quite difficult and acrimonious internal process. But the reward of that is that in general when the staff presented stuff to commissioners, all the staffers were prepared to stand behind that because they all felt that they had been heard and participated in the development of that material and understood why it was worded the way it was worded. So it became very difficult for commissioners to find a staffer who was a dissenter and say, “Pick it apart.”

Was each team’s responsibility, in effect, to draft a chapter? In other words, did you conceive of these teams reporting to a kind of implicit table of contents?

No, I did not, because the organization of the report wouldn’t follow the actual structure of the teams. There would be issues and ideas that would run as a chronological narrative. The teams were not organized chronologically, for example. What the teams would end up doing then, as I began focusing their work after the initial phase of the investigation on the nominal objective of producing a specialized monograph—four of these were ultimately produced and are publicly available. But that would be a nominal target that they would write up. A lot of their work would in the short run go into the preparation of staff statements, which were in effect statements from the teams that were presented at the hearings.

Later in 2003 I developed a table of contents for the report, and by the end of 2003 and moving into early ’04, once I felt like I understood the narrative much better than I had at the beginning, and also had calibrated the staffers much better, I farmed out initial drafting responsibilities more to individuals than to teams.

Before we get to that stage, which we want to delve into, tell us about the data-gathering process and how the public hearings fit into the business of collecting information, how the teams were going about the work that they were doing, and in particular whether there were instances where you had to help litigate access.

Two brief preludes. First, as I said, right at the very beginning I thought that we would need access to everything. I saw that as a prerequisite for the Commission’s success, indeed to the point that I didn’t want to take the job if that wasn’t going to work. Before I took the job, after I had been approached to take it by Kean and Hamilton, I was still deliberating on whether to take it. The White House was concerned and reached out to me and invited me to a meeting with Condi [Condoleezza] Rice and Andy Card joining the meeting.

So here I am with the National Security Advisor and the Chief of Staff, both of whom are asking me if I will do this. I didn’t know at the time, I learned later, that someone in the White House had had another preferred candidate whom Kean and Hamilton had already turned down, but I didn’t know any of that.

Who was that?

I don’t know. Kean has mentioned this to me and to others. Either he doesn’t remember—and I’ve heard this too from other White House sources, but I haven’t made a determined effort to find out who it was. I still don’t know, but I know I was not their first choice. The people in the White House who were following this most closely were probably in the Vice President’s office, although I did not understand that either at the time. I came to understand it over the fullness of time.

Why is that? Why were the people in the Vice President’s office—what is implied in that?

At the time I slightly understood, but only dimly understood, the relationship of the Vice President’s office a) to all issues having to do with the War on Terror, and b) all issues having to do with protecting executive branch prerogatives in any relationship with an entity outside the executive branch, or White House prerogatives in any relationship with anybody. What is now very familiar to readers of the various books—what I really didn’t understand is the way they had organized the bureaucratic process so that—and for example the key role on these particular issues that [David] Addington played as, in effect, a deputy White House counsel, who in some respects was more important than the White House counsel. This is again back to your basic question about the investigation.

The first prelude is when I meet with Rice and Card. In that conversation I just looked at Condi and I said, “One problem with this might be that people will say I used to work with you and therefore I can’t play a sufficiently neutral role.” To which they both said, “Yes, but it wasn’t our idea to hire you. The basic impetus for this came from Lee Hamilton.” I said that that was true. They said and I later learned that Hamilton had quite thoroughly vetted me with a number of other Democrats before he came to me. They thought that that was pretty much the end of that story.

Then I pushed back hard at them saying, “You understand that if we do this we’re going to need access to everything. We’re going to need access to any intelligence materials—” I specifically mentioned intelligence materials and White House records.

Did the daily brief get mentioned?

I don’t know if I mentioned the PDB [President’s daily brief] per se, because from a purely investigative point of view, the significance of the PDB as a particular form—intelligence information is communicated to the President in many forms. The PDB is only one of those. To outsiders who don’t really see the whole river, this particular tributary to the river seems extremely important to them. It has sort of a mystical quality in part because frankly the way CIA [Central Intelligence Agency] has behaved in trying to protect its contents has actually made it seem more important relative to the other streams of information than it really is. I speak as someone who has seen all these streams, including the PDB, in wearing one hat or another on one occasion or another.

So what you’re saying is it wouldn’t be implied—

To outsiders, right. In this conversation I didn’t spotlight the PDB because frankly some of the most important information might not have been in the PDB. For instance, the covert action stuff isn’t in the PDB at all. Nothing about covert action is in the PDB, not generally. The PDB is prepared by analytic folks.

Covert actions aren’t supposed to be in the PDB.

Correct, and that’s actually more sensitive than some of the analytical stuff.

OK.

And on the White House records point, their reaction was, “We’re prepared to cooperate with this. The President has made the decision that we’re going to cooperate with this investigation.” I was very blunt with them. I said, “If you folks don’t want this to succeed just tell me now, because I won’t do it. If we’re going to be put in the position where I’m going to have to be the battering ram, then some of the politics surrounding me personally become more difficult. It would become extremely difficult and the prospects for the whole investigation would be more bleak. So just tell me right now if you don’t really want this to succeed and I won’t do this, someone else will do it.”

They said, “The President is 100 percent behind this now. We’re going to support this.” Then I raised these information issues and they said, “You need to get together with [Alberto] Al Gonzales.”

Information issue in terms of specific—

Right. I asked them up front, even before I had agreed to take the job, for an assurance that they were going to be compliant in providing all the information the Commission would need, because if they weren’t then this job looked less interesting to me. As I said, I didn’t want to be in the position of battering ram. I used that phrase.

I think you told us last time, did you not, that when you met with Gonzales you confronted major obstacles.

I did. Then I came back to the White House staffer who was handling us, principally Jay Lefkowitz, who was a deputy to Card and Josh Bolten. Lefkowitz was the staffer handling this. I said, “This is a real problem, and I’m not sure I can do this.” I reported on this to Tom Kean right away. I said, “Tom, we have a big problem right here at the start.” Then there was a flurry. Kean may have called the White House. Then I got a call from Gonzales’s deputy, whose name eludes me at the moment, but I’d remember it if I saw it.

You can add it. [David Leitch]

And Gonzales—I went over this and essentially it was a reassuring call. “We’re prepared to be cooperative.” Maybe I just wanted to be candid; maybe I wanted to hear the reassurance. He told me what I wanted to hear; I couldn’t really believe that they were going to take the attitude Gonzales had described, because Gonzales was going to treat us the way they had treated the congressional joint inquiry and that would be a failure. That was unacceptable.

So who was reassuring you? Lefkowitz?

Lefkowitz heard the blowback from me and maybe from Kean, who may have either called Lefkowitz or even Andy Card directly.

But you said you were reassured. Who is providing the reassurance?

Gonzales’s deputy, not Lefkowitz.

Oh OK.

It was a deputy White House counsel, David Leitch.

If I understood correctly your impression is that Gonzales was essentially channeling Addington?

At the time I’m not sure I understood that. Later I came to believe that. Because I came to judge that Gonzales did not have very deeply considered convictions on a variety of these issues, and a lot of these issues were frankly still pretty new to him. That, plus again, in the post-9/11 atmosphere, especially among people who hadn’t been through these sorts of crises before, there was a lot of emotional neuralgia and difficulty of stepping above all of that and just seeing things in a little larger sense and with more historical perspective on what things are really important and what things aren’t. That’s the first prelude.

The second prelude was the battle within the Commission over the hiring of the general counsel. A few commissioners were promoting a candidate for the general counsel with a concept that the general counsel will be the chief counsel of the investigation and the chief counsel will really run the investigation. Then Zelikow, this nice historian, he’ll just kind of write it up. That model was unacceptable to me. It turned out that that model was unacceptable to Kean and Hamilton as well.

I interviewed all the candidates for the general counsel’s job and we worked through a couple of candidates. I was OK with some of them. I think Hamilton, not Kean and Hamilton, interviewed multiple candidates. We all agreed on Dan Marcus. I was still running the investigation, but Dan would be the general counsel of the Commission, which meant he and his deputy, Steve Dunne, would write a lot of the information requests, would do the day-to-day work in negotiating the issues with the agencies. Although at the level of the major arguments with the agencies and their leadership I was always involved in all of those. On the day-to-day stuff our lawyer and their lawyer, Dan and Steve, would end up handling most of those. In the big battles I always got pretty involved in trying to work out the solutions.

So you have that initial encounter with the White House, then an initial encounter in the Commission. We launched our investigative efforts on a very ambitious scale, but got off to a sluggish start. To his credit, Andy Card intervened and issued an order under his signature to all the agencies reaffirming their duty to cooperate fully with the Commission, and that helped. Plus the people we were dealing with in the Justice Department were just inadequate in managing all the different agency responses, and again to the White House’s credit, they essentially got rid of the people we were dealing with at Justice and put a particular person in place named Dan Levin to manage the legal issues on behalf of the administration and really take this seriously, someone who was devoting practically full-time. This was a serious person they got to do this job who had already been at FBI [Federal Bureau of Investigation] and main Justice, a real professional, and that turned out, also, to be salutary.

Where was the original point of contact in the Justice Department? Was it in the Attorney General’s office?

It was in the Attorney General’s office, and the particular point of contact was a staffer for John Ashcroft named Adam Ciongoli. That wasn’t bad, but it was just sluggish and somewhat ineffectual and we needed to work at a pace and on a scale that was very different. We began criticizing the administration, including publicly, in the early summer of ’03 and got things into a better place.

By the summer and fall of ’03 people responded. The rivers of information were flowing in to us, and we were scheduling our interviews on a very large scale.

Was there any OLC [Office of Legal Counsel] engagement in Justice with what you were doing?

No, not that was visible to us.

OK. So you’ve got eight or nine teams at that point who got the barriers, basically, broken down.

Yes. This is an incredibly wide-ranging investigation. We’re involved in the business of a number of Cabinet departments plus the city of New York and various agencies in the city of New York plus various jurisdictions in northern Virginia. The emergency response stuff, it’s—

We had a fairly sharp little dispute with New York City and access to information there that escalated rapidly but ended up working out.

What was that about?

They wanted to blow us off, to be very blunt about it. I went up to New York and met with the people in the mayor’s office and it was almost like, “Who are you?” Here’s their response—you can talk to other people about this. I think John Farmer went with me to this meeting and another staffer. I think John would agree that I wouldn’t exaggerate when I said their view was, “Your commission is supposed to investigate the federal government. This is all a screwup of the federal government. You don’t really have a right to investigate New York City.”

I said, “Well, sorry, we have a statute from the Congress and it does explicitly call out the emergency response issues. That very much includes New York City, and we are going to do this.” We had to make it clear to them that this was going to be hardball and that if they didn’t cooperate, the news of their noncooperation was going to be on the front pages of all the New York papers very soon. Then they changed their mind. But it was like that.

Your mention of New York brings to mind the 9/11 victims’ families and their role, I guess, in wanting this Commission formed and taking an ongoing interest, if I can put it that way. How are you relating to them? What are you feeling from them? What kind of pressure? How is it getting channeled? Is it getting channeled through Commission members?

Yes, all of the above. You’re probably beginning to realize what kind of environment I’m now working in, and we’re just touching some of the things. In addition to all of the substantive issues, right? It gets worse, much worse.

The family-group issue is very interesting and not well understood by the public at all. The first thing to understand about the families is that there is not one family group, there are a number of family groups. By 2004 by far the most well-known and vocal of the family groups are what was colloquially referred to as the “Jersey Moms,” four women and a group of people associated with them. They represented a relatively small fraction of the families. The most important family group was Families of 9/11, which was headed by a man named Stephen Push. There were a couple of other family groups that were also very large, and we were relating to all of them.

Some lobbied and agitated more publicly and loudly than others, and because they had a lot of sharp things to say, they got 95 percent of the media attention. At the early stage, Families of 9/11 and Push himself had been quite outspoken and they were getting a lot of media attention because they had been so furious about the failure to create the Commission. Once the Commission was created and they saw what we were doing, our relationship with them was very positive.

Indeed, they were more than positive. It was really constructive. They basically would come to us and say, “What can we do to help?” They had ideas that didn’t even occur to us. Like, “Why don’t we go to Germany and get you all the transcripts of all the legal investigations that are being done in Germany?” That was awkward and difficult for us to do, but because they were the victims, under German law they had particular rights of access to some of the judicial processes in Germany. We didn’t have that kind of access. So they got us a ton of stuff out of Germany that we couldn’t have gotten as easily ourselves. That’s what I mean, enormously constructive help. I emphasize that because they get no attention and no credit. Our relationships with most of them were very good.

There were other family—even within the Jersey Moms, you have to think about them as different individuals, each of whom had different points of view and agendas. In general, among many of them, they had become really quite deeply embittered by the experience of trying to get the Commission set up and had become convinced that the Bush administration had something to hide because of the way they had acted. From their point of view it’s perfectly understandable why they should have come to feel this way, especially given the trauma that they were still dealing with. Some family members were already entertaining and were very interested in a variety of conspiracy theories.

There were a lot who were just very bitter about the Bush administration and I understood that. And my dealings with the family members—I was basically very straightforward with them and had a number of meetings with them early on. What ended up happening out of that, though, is that some of them came to dislike me and mistrust me personally. But some of them liked me and liked working with me, so I tended to deal with some people and not others. Chris Kojm would deal with the ones who didn’t like me, and we appointed staffers whose job full time was to work with the family groups.

Ultimately a few of the family members called for my resignation in early 2004, which Tom and Lee brushed off. But at that point you could see it was awkward.

So were they contacting you or—

By the way, one of the Jersey Moms became a very strong friend and admirer. She was killed in this plane crash in Buffalo not too long ago, an absolutely wonderful woman. It really just varied enormously among the different people and it’s very hard to generalize about the views of the family members, but they were very important to us.

Are you hearing from them almost every day? Not you personally, but—

The Commission staff was hearing from them almost every day, and one of the constructive things we did early on—they would keep saying, “We have a lot of questions,” because they were all amateur researchers into this. “We have a lot of questions and no one will answer our questions.”

We said, “Write down your questions and we’ll try to get answers. We’ll try to be sure that they are all answered.” And indeed, at the end of the day when we did our report we went back to their list of questions and gave them back an annotation, “Here’s where you should look for the answer to this question.”

We also arranged a variety of private briefings for them. Quite a few of their questions and suspicions were occasioned by not only anomalies but, quite frankly, what turned out to be falsehoods that had been told by the government in explaining what actually happened on the day of 9/11.

If you read the report, you’ll see in Chapter 1 we had to unpack the facts. The Air Force and the FAA [Federal Aviation Administration], and NORAD [North American Aerospace Defense Command] in particular, gave an account of what happened on 9/11, “OK, here is when we detected these planes. Here’s the way we responded. Here’s why our response was inadequate.”

Their whole chronological narrative, minute by minute, of what had happened on that morning turned out to be fundamentally false. This was the account that had basically been put out to the whole country, and indeed was given to us in one of the first hearings we had when they testified to what they thought had happened. As we kept working on this, it began to dawn on us that that account was just materially untrue.

Then of course as people pulled on that account all kinds of anomalies showed up because it wasn’t true. The families had noticed all these anomalies, and you can see all the suspicion and mistrust that that would inspire, understandably. One of the most important services we were able to do is, we were able to piece together what really had happened on the morning and what NORAD really had known and what the FAA really had known and then work through why did you lose these planes, right down to the second.

Once we worked it through we came back to the Air Force and said, “Is this right?” and they were forced to concede that our account was 100 percent correct. We said, “Fine, then you’re going to get up in public and the Chief of Staff of the Air Force is going to get up in public and admit that their earlier testimony was mistaken and that the narrative that we’ve now constructed was actually the right one, which the Air Force has to agree is right,” which they did. Ultimately the issue was why did they come up with this bad narrative?

And how?

And how? The explanations varied between gross negligence and deliberate cover-up. The deliberate cover-up was not really—it sounds like a very political statement—we never saw any evidence that this was done at levels above Air Force officers, who were trying to cover up the fact that they had been manifestly unready.

Which frankly, though, when you get into the real story of what happened, it’s a very human and understandable story. But they basically had put together this story where we responded coherently and we almost caught up with the planes and probably would have stopped United 93. That was deeply misleading. In fact, you had no particular idea what was going on and there is very little likelihood you would have responded meaningfully to United 93. They made it sound like there was a coherent response that barely fell short when in fact, it was a completely incoherent response that had barely understood what was going on until it was all over.

The Air Force, at the highest levels, even up to General [Richard] Myers, even to this day, continues to put forward parts of this misleading explanation. People have published books and it was unfortunate.

That circles back to some of the things that we were having to unwind and unpack and why the family members were upset and distrustful early on. They had a lot of questions. Some turn out not to be as interesting or fruitful, but some of them were understandable and natural, given the fact that the information out there did not correlate well with some knowable facts.

In general I was pleased with the fact that at the end of the day all of the families endorsed the report. Since then, a few family members have raised questions about one issue or another, but I think in general they were satisfied that the report had addressed a lot of their concerns.

Some things, for instance, we addressed that aren’t addressed in detail in the report. Let me give you an example. One of the questions was, “Was there insider trading in which someone knowledgeable connected to Wall Street had shorted stock in key airlines?” Leading to the inference that someone with access to money knew the attacks were going to happen and therefore could short these stocks just before the attacks. All right? That was an interesting question. We made sure it was investigated. It was investigated by the SEC [Securities and Exchange Commission]. We looked at the SEC investigation and did a due diligence, and the SEC did scrub this and concluded with good evidence that, no, that turns out not to be the case.

We don’t go into any detail about that in the report, but we made sure that people who had asked that question were aware that the SEC had done this work—I think the SEC report may be public—and we made sure they had access to that information. There are a dozen issues like that, because while we were working there were all kinds of rumors and allegations that were flying around.

You mention at one point that they had requested your resignation. Was there a precipitating event for that, or was there an accumulation of specific kinds of evidence about your behavior that led them to call for this at that time?

I don’t know what the precipitating event was. They or someone else at the Commission might be able to answer that better. I think, though, that the context, which was fairly logical, is that the clashes with the Bush administration over access to evidence in the PDB issue, for example, had reached a peak. We were at that point in a real confrontational mode with the Bush administration over access to information. There was that issue. Either then or shortly thereafter, Dick Clarke had probably already privately told them or published the fact that I had conversed with him, that I had been involved with these issues for a few weeks during the transition, which I discussed in the previous oral history session, which the Commission staff already knew about and I had already given my evidence to them under oath about that and turned over my papers on that.

I should add as a footnote that we had a number of staffers who had been involved in these issues before 9/11. Dan Marcus, Janet Reno’s Justice Department, Chris Kojm had been on the intelligence committees in the ’90s. Mike Hurley, who was running our policy team, had worked for Dick Clarke on the NSC [National Security Council] staff on counterterrorism issues in the Clinton period and so on. There were a number of things like that. If we had pulled everyone off the staff who had worked on counterterrorism issues in one capacity or another in the previous five or six years we would have had trouble finding much expertise on the subject.

Sure. Were there specific instances of recusals?

Yes. There was my recusal on that issue, and then I did not participate in any interviews of Condi Rice or Steve Hadley either. I attended those interviews but did not participate in them.

The others, did they—

I also recused myself from writing on that transition period in which I was involved, which is about a page in the Commission Report, I think for which team three and Dan Marcus were the principal drafters.

Then, for instance, Marcus did not conduct the interview with Janet Reno. Things like that. Hurley did not conduct the interview with Dick Clarke. I conducted the interviews with Dick Clarke, but there were a number of other people participating in those interviews who also asked questions. We conducted three different interviews with Clarke.

Can you walk us through the investigative process itself? You’re doing mostly interviews? You’re gathering documents? You’re holding public hearings?

Yes. All of that.

I can refine the question.

For anyone who is reading this doing research, they should endeavor to find—it will be in my papers but it was made public at the time, in 2007—the report I prepared for Kean and Hamilton about our effort to get access to the detainees and our effort to find out how they were being interrogated.

In the fall of 2007, when it was disclosed that the CIA had videotaped some of these interrogations and had destroyed those videotapes, it immediately raised the issue of did the 9/11 Commission attempt to get access to this. I wrote a report detailing exactly what we had done with respect to this issue. I went back to the Commission’s archives and I have and retain access to those archives to address questions that have come up since we finished our work, like these.

I wrote that report. I got help from a couple of other former staffers, especially Steve Dunne, I think, who worked with me on this. If you read that report that is a really nice glimpse. It’s an interesting illustration, it turns out a fairly important illustration, of how we went about our investigative business. It’s just one very narrow slice. You can see the layers of information requests that we made, then the follow-ups on those with the Agency lawyers. How we addressed these concerns, which ultimately, in this case, went to [George] Tenet himself, and eventually Tenet and [Donald] Rumsfeld and Gonzales met directly with Kean and Hamilton, not with me, to try to get them to fall off part of this. They agreed. The Commission debated that and then stood by Tom and Lee on a piece of this.

By the way, I dissented from that. That’s just a sidelight, but if folks will find that document, that is just one example, and if you think about that, that’s one case where we’re trying to pursue things. That’s an important one, but there are many others like that. The PDB stuff is much more involved than that, but in general there were literally dozens of information requests that we would put out. We would assemble large numbers of documents. Also, some documents would come directly to us. Some they would simply make available. Databases or large bodies of documents held at the agencies for us to peruse or mine.

We would create schedules of interviews, which we would arrange, and then figure out which staffers would do which interviews and which ones would be under oath, which ones wouldn’t be under oath formally, although, it turns out, the significance of putting people under oath is mainly ceremonial. The legal penalties involved, whether or not you’re sworn, are the same. It’s just not well understood by the nonlawyers. But they were all in the position of being under the relevant part of Title 18 because we were a government investigation of the Congress. If they committed material and knowing concealments of facts then they had felony exposure, whether or not they were put under oath.

An issue that I think remained ambiguous in the final report was what did Bush order the morning of 9/11, or did [Richard] Cheney issue the order? I think it involved authority to shoot down certain planes. How did you deal with that issue? How did it come up? How did you decide what to do in the report?

A number of us spent a lot of time on this. This was a very hard problem.

If I can pause, we want to consider this in the context of a larger question of your overall relationship with the White House throughout, but I’m happy to go ahead and proceed with this and then we can deal with the bigger question.

Yes, let’s do this and then come back to that. My relationship with the White House became increasingly adversarial. It’s one of the ironies of the outcry about my being the White House shill on the inside. The White House at the time did not view me in this way. [laughter] They viewed me as especially obnoxious and confrontational and tried as best they could to get around me, either going directly to Hamilton or finding some other way.

When you say they, the White House—

I mean Gonzales and the White House lawyers who were our key interface. I was surprised at one level, but not at another level. For instance, when Phil Shenon wrote his silly book, he got Andy Card to talk to him and Andy Card complains bitterly about me. Calls me “a bully.” You may remember back in the old days when I was working with Card to set up the [George H. W.] Bush 41 oral history project 10 years ago there was a very amicable relationship there.

You have to ask why would the White House Chief of Staff, one of the most powerful people in the country, regard this little piss-ant investigator at the 9/11 Commission as bullying him? It gives you a little bit of a glimpse into their world and into their mindset.

We’ll talk about Shenon a little more later.

Back to the question about the morning of 9/11 and these orders. First off, the team did the main homework in investigating this. Key people working on this included Dana Hyde, who now works in the [Barack] Obama administration for [Jacob] Jack Lew and who’s very good, by the way. I have a very high opinion of her ability. Dana Hyde, Miles [Kara], John Farmer got involved in this, and there were a couple of other staffers on that team who were working this very hard. Dana and I in particular had gotten involved in working on some of the extremely secret stuff that’s not in the report because we couldn’t publish in the report all the tales of what we dug into about White House emergency procedures. The White House asked us not to publicize their procedures for continuity of operations and continuity of government in any detail and we did not do that. Though privately, after the report was done, we briefed some of them on some of the lessons we had learned from that, which they took to heart and made a variety of changes in their continuity of government and continuity of operations contingency planning.

But the team, in effect, constructed this timeline and then they assembled the various logs. I was involved in questioning some of the people about this. For instance, I think I participated in the interview with Josh Bolten. I myself asked Bolten the question where he gave the answer that we reported in the Commission Report, things like that.

Several people had done this work. We had all the information. When you look at it I think some of the staff leaned toward the view, though they couldn’t prove it, that is more like what’s in Barton Gellman’s book. That is, basically, Cheney really ordered it and Bush ratified it. But they couldn’t prove that and there is countervailing evidence. It turns out that the logs don’t have some of the calls that some of the administration people told us about, but it wasn’t just Condi who told us of a call, there was another officer, a lower-level officer in the room, who also had an account that was consistent with Rice’s account and with Cheney’s account. So we basically had divided evidence. We already knew from other things that the logs were incomplete and the logs didn’t cover all the calls. You couldn’t rely only on the phone logs.

We had this problem. I think their working supposition was that Cheney had given the order and Bush had ratified it, but there was some evidence that wasn’t just from the participants that said, no, actually, there had been this call around ten o’clock in which Bush had given the green light to Cheney. Other people in the room didn’t remember that call, so you could run this argument either way.

We couldn’t just say that Cheney and Rice’s account of this is for sure right. We debated about different ways to write this up, “We think here’s the way it happened, but we can’t prove it.” What we ended up doing is I and then other staffers wrote alternative versions for the staff statement. We really litigated most of this inside the staff and came up with a version of this that we all thought was fair. The basic version just acknowledged that we were uncertain. We confessed that openly and then just said, “Here’s what we know and here’s why we’re uncertain,” and then just left it there.

I respect Gellman’s account of this, but I think we looked harder at all the available evidence on this than Gellman did. We had access to a more thorough range of the relevant documents, and to this day I’m not sure what actually happened.

Did you consult with Kean and Hamilton on how to handle this?

Oh yes. When we had difficult issues like this we would keep Tom and Lee apprised. “This is a hard problem that we’re running into for the staff statement.” Then we would explain the way we worked through it. We worked through it collegially, not just me and Chris and Dan, but also some of the other staffers. It’s a difficult process. We assembled on a piece of paper every piece of evidence we had about this and would argue back and forth about it.

The bottom line was, when we looked at it at the end of the day, we said, “This is an accurate summary of what we think we can say and all of us on the staff are prepared to stand by this.” Indeed there was a Commission meeting where this was discussed, and all the staffers who had worked on this were prepared to stand by the way we had characterized it. I think if you talk to Dana Hyde today, for example, she would say that what we put in the report was reasonably fair. Actually, I think she would be stronger than that, that it was fair.

To this day I’m not 100 percent sure what happened. I’ve read—[Karl] Rove in his memoir goes back over this again. I’m sure either Bush or Cheney will address it, and they actually might be right. The issue is interesting. It wouldn’t have made any difference that morning. It would’ve been highly damaging politically to them if we had found that they had fudged this just to avoid the appearance that Cheney was being presumptuous, although, if in fact he had been, probably a lot of people would have been somewhat understanding about it. But I’m not sure today that he was presumptuous. It may be that their account of it is right.

One of the critical points in your relationship with the White House is the interview with the President himself and the negotiations over how that was going to take place. Can you give us your account of how this was broached, what the negotiations were, and ultimately how the interview happened?

That was not as hard as some other things. We knew we had to interview all the former Presidents and the current President and the Vice President. We went out to all of them. Clinton and [Albert, Jr.] Gore pretty readily agreed, and then there were relatively easy negotiations with them as to how that would work. Bush was harder.

It’s highly unusual. Again, if you put yourself in their shoes, and to be fair to them as a historian, in almost everything that has been written on this no one has actually talked to anyone in the Bush administration about how they felt about all of this. Even though personally at the time I didn’t feel incredibly sympathetic to their point of view, as a historian, their point of view does have to fairly be taken into account. From their point of view, remember this is a group of people who are actually trying to restore Presidential prerogatives that they think have been whittled back in the previous generation, and here we’re coming in with a bulldozer trying to get them to do stuff that actually no President had done before in quite this way. That was very hard for them.

But especially once Clinton and Gore were doing it, and the issues that were raised were very clear and their stubbornness became public, the situation rapidly became indefensible for them. We had already had a harder time with Rice testifying. We’d already broken through that. Then they came in with, “Well, we’ll have Bush and Cheney together,” because I think they were afraid. I don’t know, you can ask them—I think they were a little bit afraid that if Bush and Cheney testified separately, that people—because, you see, they believed, actually with some cause, that some of the commissioners had highly partisan agendas. This is not a fanciful supposition if you understand the way the commissioners were appointed.

OK.

They thought that someone would look for microscopic distinctions between their two testimonies and then greatly exaggerate the significance of those distinctions.

I don’t know for sure what their concern was, but they came back in with this joint proposal, and of course the concern was that—most of the commissioners, especially the Democrats, had this image of Cheney as the Prime Minister. That was the very common supposition back then, especially among Democrats who didn’t know them well, and that Bush was this figurehead. They expected Cheney was going to do a lot of the talking while Bush would offer some vague generalities. Then, of course, the interview completely surprised all of them because Bush utterly dominated the conversation, did 95 percent of the talking without notes, and was clearly in command of the subject. For people who had not known Bush, this took them aback because they had a very different image of him.

The disadvantage of it all turned out not to be that we didn’t get enough from Bush, it turned out to be we didn’t get enough from Cheney. [laughter]

Where did the interview take place?

It took place in the Oval Office. I was the notetaker for the Commission. I didn’t ask any questions. I actually was the lead interviewer in the Clinton interview, but in the Bush interview I said nothing.

Did you get into highly classified materials as you interviewed them?

Some. But actually a little more with Clinton than with Bush because the covert action issues under Clinton turned out to be harder than the covert action issues pre-9/11 for Bush.

Well, tell us about that interview then.

You mean the Bush interview?

No, tell us about the Clinton interview. We’ll come back to Bush.

No, just on the procedures for Bush. I and other commissioners took handwritten notes. They all gave me their handwritten notes to use to check against my notes. I prepared a— we didn’t tape that interview. We taped Clinton. You can eventually listen to the recording and the transcript, and I think Gore, too.

When will these became available?

As soon as the government declassifies them. In theory, all of our records now are available once the agencies who hold the—

I went to the website last week and there is reasonably little available. One of the things that one sees on it is that you can get, theoretically, to Dan Marcus’s papers and some of the other people, but yours are not available at all. They’re not even listed as on the website. The website said something like 31 percent of the material is now available, but it looked like it was the type of thing that most historians wouldn’t be very interested in.

There is some interesting material in what’s been made available but it’s not the Commission—we released our hold on the material. I think we had something like five years, but it’s gone. Now it’s all agency equities, and the agencies are now controlling the declassification of further materials. It’s not an issue with the former Commission anymore.

Who would control your papers?

Whichever government agency—what has happened is that in practice it’s CIA. CIA has become the executive agent for managing the interagency declassification of Commission records, I think.

OK, Clinton.

It’s not me. They are not closed at my direction, but the point is that my written record for Bush is pretty darn close to verbatim. It’s as close to verbatim as I could make it. It’s like 30 single-spaced pages.

How many people are in the room for that interview?

That’s 10 commissioners and me, I think.

OK, so the commissioners are there.

The White House note-taker for it, which I think was [Bryan] Cunningham, told me that his record for it was almost the exact same number of pages as mine. So there is actually both a White House record of that interview and a Commission record.

So as you recollect, what were the issues that the Commission was most interested in when they interviewed President Bush? The top two or three issues.

Gosh, it’s a little hard for me to remember the main thrust now other than, What did you know? And when did you know it? And what did you do about it? In general: What did you understand about this problem, what did you try to do about, it and when?

All the commissioners went through rounds of asking questions. One of the more interesting things that came out of that, which confirms some other evidence that we had, was that Bush actually was somewhat impatient and aggressive about wanting to do something about [Osama] Bin Laden. Probably both Ernie and I thought that Bush was more aggressive about this than anyone else in the administration. The way Bush put it vividly at one point in his interview is he said, “There are a lot of people who were interested in trying to do a throat slit,” or looking for some sort of covert action or panacea that he didn’t think was going to get at it.

His sense—he didn’t use the phrase again, but it’s very reminiscent of what he was reported more contemporaneously as having said, probably in March or May of ’01, “We’re just swatting flies,” that we actually need to do something really decisive about this. The impression then and now is, my guess is, that people heard him say that and then just went on. Because he was probably the only person who was willing to think about some fairly massive operation in Afghanistan before 9/11. He was probably just ruminating about it, but to everyone else in the administration an idea like that was literally inconceivable.

On what evidence do you say that he was willing to think about invading Afghanistan prior to 9/11? He said that?

He said words very close to that in his interview with us. We had some contemporaneous evidence, actually from something that Dick Clarke had said in ’02, remembering something Bush had said in the spring of ’01, which Condi Rice had also remembered him saying in the spring of ’01. There was a little discrepancy in whether he said it in March or May, but it was consistent with what Bush was saying, which was that he was impatient with these efforts that he thought seemed relatively ineffectual and that you needed to get to the heart of this problem of al-Qaeda and Afghanistan.

He actually went into more detail about it in the interviews. He didn’t use the words “pre-emptive attack,” but in effect he was articulating a concept that these folks are at war with us and they’re going to hit us, or something like that.

I don’t want to place too great weight on it, because how much of this is post facto embellishment is hard to say. I think the way we wrote it up—we alluded briefly to this I think in Chapter 10 or 11 of the report and at the time we wrote that up we put about as much weight on it as we thought we could.

I owe you a break. Let’s park right here. Hold on to your questions.

[BREAK]

One thing I want to get us to after we talk about the work of the Commission is how the recommendations came about.

Sure. But I think that will come at the point where we’re drafting the report.

You had asked me about the Clinton interview, which was very interesting. The good news is that eventually scholars will be able to see this whole interview for themselves. I think they will find that it is a very revealing illustration of Clinton’s strengths and weaknesses. The Bush interview and the Clinton interview in their way are exemplary demonstrations of the strengths and weaknesses of both men.

How do you mean that? What are the strengths and weakness?

In the Bush interview, and in meetings, Bush is a person who is always trying to focus on the bottom line and is trying to push the decision and the action. So in that sense, he is very strategic. His whole orientation in the interview with us is honing in on the core issue. What was the core issue and bottom line? He’s one of the least digressive persons in an interview like that that you will encounter.

This has obvious strengths. My view is it’s paradigmatic of the business-school-trained mentality. I have a problem set. I want to cut to the core of the problem set. Then there is a school answer to the problem set that’s my bottom line. What’s the decision I need to make, what’s the core issue I need to understand about this that then gives me the answer or the next action? That’s a powerful trait.

The downside of that trait is that in the rush to close on the bottom line you’re not as curious about side information, about other issues or questions that need to be considered. It’s not an extremely reflective style.

Clinton is exactly the opposite. They’re polar opposites of each other in so many ways, personally and professionally. In my view, the antithesis. Clinton in his interview is an absolutely fascinating conversationalist. You ask him a question and he just launches off on these coruscating flights of excursions and analysis and this thing leads to that thing and a lot of the audience and other commissioners are bedazzled and sit listening in bewildered fascination to this roller-coaster ride. Five minutes later you forget what the original question was to begin with.

I don’t think Clinton’s doing this as a device to deflect people. He’s doing this because it’s just the way his mind works. He is instinctively discursive. Again, it’s a little bit of the lawyer-law school paradigm. He’s constantly looking to discover there are other issues hidden in this fact situation that you haven’t unearthed yet. He’s constantly unearthing more questions, more fascinating rabbit trails to pursue, but it also, of course, can be very frustrating for the staff, because remember, we actually have to make this decision.

He’s just the kind of person who’s reflective and discursive but has trouble closing, and the interview captures this very well. I opened up interviewing Clinton and I actually had some fairly specific questions to ask him, including about the intelligence budget, the covert action issues, because the intelligence budget actually doesn’t do well in the Clinton administration. There are some arguments about that, but you’ll see I’ll ask him these fairly specific questions and he’ll brush that off and then go off. I’ll work hard to try to bring him back to my question because I was once a trial lawyer, and I tried to remember what my question was.

But you’ll see, it’ll look like there is this boring, rather dull fellow Zelikow, who keeps trying to work on these niggling little points while the President is offering this magnificent tour d’horizon after Zelikow finishes asking his stupid question. That’s very much the tone.

I did that for an hour and a half or so and then I gave up and let the other commissioners throw softballs at him. He hit fungoes for the rest of the interview.

Was it productive in any material way for your investigation?

It wasn’t productive for our investigation at all.

So you didn’t find out anything that you didn’t already know?

We did not. I could get a couple of things on the record that we already knew.

How about the Bush interview?

I’ll give you an example, a very concrete example in the Clinton case, which I think I asked him about.

You have this rather amazing episode with this covert action memorandum and notification—by this time we had finally discovered that Clinton had done a kill order for Bin Laden. Discovering this was itself one of the more difficult detective inquiries of our investigative work and I was fairly proud that we cracked this puzzle, because there had been sworn testimony on both sides of this issue. Did President Clinton order that Bin Laden be killed or not? There were people who had given sworn testimony on both sides of that question, CIA and White House people. Privately, [Samuel] Sandy Berger and Bruce Lindsey, who I know as friends, were telling me, “He did. If you’ll just keep looking, you’ll find the documentary record. It’s there somewhere,” because we had not found the documentary record. Finally, I did the equivalent of turning every trunk upside down to try to find the documentary record and ultimately did.

Isn’t there a Presidential directive?

We did find the piece of paper. The thing that had confused us all along was that we had gotten copies of the relevant MONs [Memoranda of Notification], in part from the records that had been given to the overseers on the Hill, and this wasn’t there. The Commission left records behind on this issue. I wrote a memo to commissioners that explained the evidentiary problem on this issue and what our investigation had to do. So rather than my recapitulating this now, if someone will find that memo I wrote they can reconstruct why this was so hard and how, with Dan Marcus’s and Steve Dunne’s help, we solved this problem.

Can you tell us—

Later I did another memo like that on some PDB issues, went through for the commissioners, “Here is what the evidentiary issue was and here’s how we tried to work through this problem from an investigative point of view.”

So on this issue, you have this kill order for Bin Laden that President Clinton signed on, literally, Christmas Eve of 1998, at Tenet and Berger’s behest after some extraordinary stuff had gone on, which, by the way, Tenet claimed he had no memory of whatsoever. The only kill order signed by President of the United States during his tenure as DCI [Director of Central Intelligence] before 9/11, and the DCI claimed that he had no memory of the whole episode at all.

Let me just finish this anecdote on Clinton. That kill order had been written with respect to a particular proxy group that would carry out the killing. We described this in the report as the very peculiar story of what happened after he did this, how CIA tried to translate that into a guidance cable, and what happened when that guidance cable was actually received by the handlers of the tribal group that was supposed to do this. It’s almost a comical episode if it didn’t have some tragic significance.

All right, but then they wanted to use another tribal group to do the same thing. That’s in February of 1999. They needed another MON, another Memorandum of Notification, because they were using another tribal group. So they go back to the President saying, “Would you please sign this out again?” It’s the same thing he had signed in 1998. He crosses out the key kill language and writes in in his own hand; he basically turns it into capture. If you try to capture him and can’t, then you can kill him, which, by the way, completely changes the way you would run the operation and makes it infeasible.

I put this in front of Clinton and said, “Why did you do this?” He said, “I have no memory of it. I don’t know.”

That illustrates that Clinton was interviewed, we did succeed in getting him on the record saying that he had no memory of this and had no idea why he had done this.

Let me ask you a question about that. I was going to interrupt about Tenet, but this actually is a good occasion to ask you the broader question. Do you find that surprising? Do you find it credible? You’ve been on the inside of these things. Is it possible that people are so busy and that their attention is so diverted and they’re so overworked that they—

No, it is not. No, it is not credible. In my view, and by the way, the view of the whole staff, it is not credible at all. In later years I was impolitic enough to actually say so to Elsa Walsh when she was writing a piece in the New Yorker. Tenet was just livid about it and called various people to let them know. He was livid that I had dared to impugn his credibility in print, on the record, not anonymously. But of course it’s not credible.

You say, “Of course.” I just don’t know.

Think about it. You’ve instructed the CIA to kill someone and you don’t remember that?

I understand the argument that I deal with a lot of important things all day long and a many great secrets and so on and in that world. OK. This was extraordinary even in that world, and he knew it. Even the process by which they got this done was a very unusual process, and he was doing it all personally. For him to say that he had no memory—at this time in the Clinton administration this was regarded as a really unusual and extraordinary thing. When you talk to the other Clinton people, they had had run-ins with Janet Reno and issues with Jamie Baker, the NSC staff lawyer. No. As commonplace as big decisions are, this is the kind of stuff that senior officials remember.

I talked about this once with Ron Suskind, who spent a lot of time with Tenet and worked on The One Percent Solution book. Suskind has a theory in which he actually thinks that Tenet may just honestly have memory problems. Suskind thinks it may have something to do with drinking or—I’m not sure.

Would you say the same thing about Clinton’s memory on this matter? That it’s implausible that he wouldn’t remember, certainly writing in his own hand to change an order that he’s originally issued?

Yes, it’s implausible.

OK.

Well, Clinton’s memory I know more about because I have more evidence about it and these others I don’t. Not having been on the inside, I don’t know what people are inclined to remember and forget and whether it’s convenient for their memories to fail them on occasion.

Another reason why it’s implausible is that Clinton prepared for this interview. He prepared by sitting down with people like Sandy Berger. They knew this issue would come up. Of course Sandy remembers what happened.

The recommendations matter. Is this something that you are working on in parallel with the investigative part of the report? Is this something commissioners are having more of an active role in than in the investigation of what happened?

Various teams are coming up with different ideas and ideas are occurring to us as we’re working on the report. At some point—I’m having trouble remembering this—I drafted these chapters and I canvassed ideas in some fashion and then just sat down and tried to write this up. The one recommendation that several commissioners had had in their head from the start that was never far from their thoughts, as in Tim Roemer’s case—at the time some people called an NDI a National Director of Intelligence; we ended up calling it a DNI [Director of National Intelligence]—the intelligence consolidation idea. Because as Amy Zegart and others have pointed out, that’s actually a very old idea. I personally have been involved in at least one prior task force that had made this recommendation.

I didn’t go into it sure that we would recommend it or sure that that was highly relevant to this story. It turned out that it’s not a central issue in this story. This story simply offered one more illustration of the managerial weaknesses of the DCI model. It’s not so much an issue in the connecting-the-dots realm, to use that terrible metaphor, but more in examples where the DCI was trying to create a strategy of getting the intelligence community to devote a lot of new resources to this new problem of counterterrorism, and his manifest incapacity to make the intelligence community reallocate resources on a large scale.

There is a famous example where he writes this memo saying, “We’re at war. No effort should be spared.” We recount the story of how this memo had almost no visible impact whatever. It was pointed out that if you had actually wanted to put together a serious investment strategy, not just supplemental budget requests, that would have extended to language training and a lot of other things, it wouldn’t have just been more money for the counterterrorism center this year. I mean, a serious strategy that really reoriented you to a different part of the world and different problems.

We made a notional sketch. If you thought about this for 30 minutes, if you had a seminar on this for an hour on what an investment strategy for this new priority would have looked like, it would have covered a lot of agencies and involved a lot of things. No one in the DCI’s office and Tenet himself even proposed or thought hard about such a thing. In our view, if he had he would have had trouble doing it because of this lack of powers. That idea was in the back of people’s minds all along and we concluded that this case offered yet more evidence that the managerial weakness of the DCI was a serious problem.

But that did not turn out to be our lead recommendation, even on the process side. In many of our most important recommendations—the more important chapter was the one that comes before the process one, which is the one on strategy and substance of the policy, where the main thrust was to say that in general, our strategy is too militarized and needs to be more comprehensive, and that we regard “War on Terror” as a somewhat mischievous nomenclature, because you really needed to focus on Islamist extremism fairly specifically as the phenomenon and have a strategy that was more about that than about some sort of generic terror enemy. So we devoted some attention to that.

In both of those respects, by the way, I think we were pushing on an open door. I think the Bush administration was internally coming to similar conclusions. Even Rumsfeld was.

Philip, back during the entire course of the investigation, we can’t cover all the specifics, but were there other key cases or controversies that are worth your commenting on for a few minutes before we move into the drafting?

Well, there’s a lot of material on the 9/11 Commission that’s worth noting, but a lot of it can be unearthed as folks get into our records and also get into the administration’s records on their dealings with us, which no one has actually looked seriously at or even talked to the people who were involved, to the extent that they’re interested in that story.

One useful point underscores that the PDB issue actually came really close to being a monumental confrontation. The initial agreement with the administration on how to handle this broke down. We had trouble putting something back together again. Dan and I—I actually believed we would have to go to court and this would be a Pentagon Papers-scale confrontation with the administration because we were actually going to have to file a subpoena to force the administration to reveal this. Then the federal courts would litigate this and it might go very rapidly to the Supreme Court. We had already retained a law firm to draw up the papers to file that lawsuit. This is not, I think, well known. We were actually, I think at one point, within days of filing. Fortunately, Fred Fielding, who was one of our commissioners and is himself a former White House counsel with some serious experience in dealing with scandal and problems, knew this. I would talk to Fielding because I thought he was sensible, and Gonzales would talk to Fielding.

But he wouldn’t talk to you?

No, but he would talk to Fielding and either Fielding alone or Fielding with Hamilton would become key intermediaries in trying to work this. They understood that we were totally serious about going to court on this and they might not like what would happen then.

By the way, on this issue as on so many issues, in their zeal to protect Presidential prerogatives, they actually hurt the powers of the Presidency. In their zeal to protect the President, they hurt the President. One of the strengths of Gellman’s book on Cheney is he has a story about how Cheney is fighting efforts to get access to the records of his energy task force in 2001. Basically, the story is that this is a totally Pyrrhic victory. Cheney won some of the battle but totally lost the war. And this point was demonstrated exponentially in our case, even more seriously in other cases, that in their fight to protect Presidential prerogatives they actually hurt the power of the Presidency because they overreached. Had they handled this in a more sensible way they actually would have better safeguarded the very interests they were trying so zealously to protect. With which, actually, I had some modicum of sympathy.

Was their zeal—

We came very close to a large lawsuit and it was finally defused at the last minute because we came up with another PDB arrangement that, frankly, required me and Jamie Gorelick to do some pretty heroic exertions. I think at one point I was in the reading room for 30 hours straight. But we then wrote up, literally, a 7,000-word summary of all the leading intelligence documents that had come in, including all of the PDBs, that Jamie signed off on word for word. Scholars will get access to that 7,000-word summary, and I think they will find it pretty thorough.

Kean and Hamilton, by the way, in addition to us later, looked at all the PDB material also. I wanted Kean and Hamilton to be in the position to say, truthfully, I, Tom Kean, or I, Lee Hamilton, have looked at all the information I wanted to look at concerning this matter.

So ultimately it wasn’t just Jamie and I. Tom and Lee looked at all the material, too. Other people looked at some of other pieces of this, too, because, it turns out, almost all of the material in the PDBs ends up getting reflected in other intelligence documents. Because as I said earlier, there is almost a daily flow of material.

But the more serious irony is this: the CIA director can’t tell the President anything he doesn’t know. We had access to all the stuff coming to the CIA director from inside CIA. We had all the daily stuff that Tenet’s counterterrorism people were giving to Tenet every day. So we had a pretty good idea of what the government knew about al-Qaeda.

I was going to ask you about the August PDB.

You said, “He can’t tell the President what he doesn’t know.” Who is the second he?

Tenet.

Yes. Tenet can’t tell the President what Tenet, himself, does not know, and we had a pretty good bead on what intelligence was going to Tenet because we had the daily flow going into his office.

But you still don’t know whether Tenet is reading all that stuff, much less—

Right. But since you could see all the different links in the chain and then people in their interviews would tell you a lot of the things they remembered or that stood out, there are some ways of getting at this.

Go ahead, Mel.

Briefly stated, what do you think was known about al-Qaeda?

We have a nice discussion of this in the report where we call attention to the absence of the NIE [National Intel Estimate] and so on. Here’s what the CIA’s point of view on this is. You’ll read it in their hearing. It’s, “We understood this. We saw it.”

What they understood was, they tended to identify this organization as Osama Bin Laden and friends. When you look in the documents of the late ’90s they tend to be about UBL [Usama Bin Laden, also rendered as Osama bin Laden], which was the acronym for that. Some people in the organization are understanding, even by the mid-1990s, after the [Jamal al-] Fadl defection, we had his testimony from federal court in the late ’90s that we looked at—Doug MacEachin did terrific work on this—is that actually this is more than just a personalized Bin Laden network. It’s actually more of an organization and it’s more substantial. The UBL network made it sound more like the Abu Nidal Organization of the late ’80s. It seemed more like that. It made it seem somewhat more idiosyncratic.

By 2000–2001, you’ll see documents that refer to UBL and documents that refer to al- Qaeda. The reason we placed a little bit of emphasis on the absence of a National Intelligence Estimate that updated the stuff they had done in the mid-1990s is not because the National Intelligence Estimate is a magic document. It’s because it’s a document that does two things. It forces people to sit down and simply compile and examine what they already know in a somewhat systematic way, which in this case would have been salutary because they knew quite a lot about al-Qaeda by then, but they had not really assimilated it in a formal way where they put it down and said, “Here is what we know,” where they forced themselves to summarize it and reckon with it.

For instance, there has now been some very good work done by scholars on al-Qaeda using captured records and so on that many scholars are now beginning to examine that are publicly available. Thomas Hegghammer’s book, for example. Al-Qaeda’s golden years are the two years before 9/11, at which point it has thousands of people in its organization, including at least a thousand people who are fighting for the Taliban in the civil war going on in Afghanistan. It’s a substantial operation with various affiliates worldwide.

A lot of people knew that and a lot of particular intelligence items mentioned that, but in this way that is still inchoate, where it hadn’t really been fully assimilated and where folks had to come to grips with what they already knew. Although the term al-Qaeda is increasingly being used in the briefings of top leaders.

The second value of something like an NIE is the Congress knew almost none of this. The NIE goes to Congress, a formal way of announcing and consolidating known information in a way that Congress would notice. That has some important political effects. This is one of the things that drove Bob Kerrey, who was one of our commissioners, crazy. He says that he had been on the relevant committees at the time and they would have had their hair on fire if they had known some of this stuff.

Maybe Kerrey is wrong about that.

With all due respect, Philip, I just read literally three days ago the testimony of George Tenet to the Armed Services Committee in February of 2001 in which he says all these things with unmistakable clarity. He says, “There will be an attack and it could be in the United States.” He says it to the Armed Services Committee with absolute, transparent clarity.

Yes, that’s fair. He makes that point. I don’t think we quarrel with that. What we were trying to think of—here’s the intellectual problem you have, and it’s very interesting and important. It’s clear that the U.S. government as a whole knows at some intellectual level that this enemy exists. Yet there is some visceral level at which they somehow don’t think it’s a problem for which they need to think really dramatically, for instance, about even conceiving a large military operation in Afghanistan. I’m not saying ordering it; they don’t even order it planned. They don’t even order it thought about. It’s not even put on the table in an options paper. A major military operation in Afghanistan to clean this out is just not even conceivable.

If they know this intellectually, why aren’t they seeing how serious this can be? This puzzled me. I had written this essay in Foreign Affairs in 1998 on catastrophic terrorism where I even used the World Trade Center example. This is not 20-20 hindsight. I’m trying to grapple with this problem.

Even Dick Clarke, who has his hair on fire more than almost anyone about this, is writing on September 4 that hundreds may die. In other words, he’s saying we might have another attack on the scale of the embassies attack in 1998. He isn’t right. Thousands may die. There is just some level at which people, though they understood a lot of this intellectually, hadn’t quite gotten their heads around the fact that this was a potential catastrophic threat. In other words, a nonlinear jump for which they actually needed to put some nonlinear ideas on the table now if they wanted to address this before it happened.

So I’m trying to ask myself, how do you solve that problem? How do you get people to internalize that this is a truly catastrophic danger before the catastrophic danger materializes on your doorstep? This is actually a very interesting and challenging problem.

Extraordinarily interesting.

Right.

Because you can see parallel discussions going on at the exact same time about Iraq and what to do with Saddam [Hussein] in which prospective scenarios are thought through much more carefully.

The government had been playing around with war plans on Iraq for sure since late 1997, and had looked at them fairly hard in ’98, ’99. Then it put them on ice for a while, but in a way they were reviving a running discussion where this had already become conceivable, had already been conceived in fact. Plans had already been developed and they were dusting them off and trying to reconceive them. OK.

This al-Qaeda problem was a new thing. There was some sense that they felt they were working the problem sincerely, and yet clearly at some level they do not truly grasp how dangerous this problem is. Even though at another level, intellectually, you can find evidence—

Part of the problem in the report is that we weren’t trying to play gotcha with these people saying, “Well, you should have known because—” We put the evidence in the report. You can see what they should have known.

The harder problem is why, if this data is out there, did they still not quite internalize that this was a truly catastrophic danger. By the way, I think this is true for both administrations. Ernie May had a wonderful line in an essay he later wrote in The New Republic where he borrowed Clarke’s—Clarke threw a grenade at the Bush administration, saying for the Bush administration this problem was important but not urgent. Ernie thought about that and said, “Clarke was right, but it was true for both administrations.” I agree with that. I think that’s fair. I think it becomes more urgent for the Bush people after about May of 2001.

In the last few months, I see evidence, by bureaucratic standards, of increasingly frantic concern. Stuff about really beginning to lash the Predator thing out the door, some of the meetings in July of ’01, and the worries about these overseas alarms and the things they were doing about that. After about May.

Not, by the way, in the Pentagon, but from White House and CIA. So there is still this problem, and that’s what I was trying to get at with the idea of the NIE. If you’re trying to figure out a way to call attention to a giant problem that you think folks haven’t really internalized as catastrophic, the NIE is one of the classic devices you can use to do that, as a bureaucratic way of assimilating all your information and obliging people to pay attention to its dramatic conclusions. Partly because they are politically significant, because you’re giving it to the Hill.

If others can come up with a better argument for how to address this conundrum, I’d be glad to hear it.

Was your interview with Al Gore productive at all?

Modestly productive. We may cite it a few times in the report, but not very important.

It didn’t lead you against your Ernie May conclusion?

No.

In other words, that Gore was no more on to this than anybody else was?

As I said, they thought it was important. I think that that’s true, in different ways.

I think the Clinton people, intellectually—and this is one of their gifts—quickly perceived that this was a novel phenomenon and that it was significant. I think it was extremely important—probably the person who I think was more seized with this than anyone else outside of the office that was working on it was Sandy Berger, personally. In general I give the administration credit for recognizing the importance of this. They had difficulty getting the rest of the government to share this sense of importance.

In the Bush administration I think it takes them a little while to get their heads fully around it. They come in knowing what anyone who reads the newspapers knows, except for Bush and Cheney and Rice, because they are getting Tenet’s and Ben Bonk’s briefings and other briefings from the fall. Tenet’s testimony, including his private testimony and the documentary record, including from others who briefed Bush, is that Bush and Cheney get that this is significant fairly early on, personally, and care about this. They get a serious set of alarms in the spring that occasion fairly interesting meetings between Rice and Tenet, like in May of 2001, and I sense that this ratchets up several notches in the ensuing three months.

Were you surprised to find the August 2001 daily brief? Do you remember when you first encountered it, thinking, This is something that we’re going to have to deal with?

I’m trying to remember when I first encountered it. Yes, I thought it was very interesting, of course, because of the eye-catching headline. As Mel pointed out a minute ago, to anyone who is reading the river of information that’s coming into the White House, or, frankly, that Tenet was testifying to on the Hill, the statement, “Bin Laden is determined to attack the United States,” barely even raises your eyebrow.

But in the domestic political context of the Commission’s work and the way people were arguing in 2004, this was a good club with which to beat people over the head. When you actually read the document, it’s a good example of the Bush people making things so much harder for themselves. As soon as I saw this, or saw the whole text of the document, I knew, of course, that we would have to bring it out, that we would have to, and I told the Bush people, “You ought to declassify and release this immediately.”

The more you study it the more you realize what the document is and what it isn’t. Especially when we actually dug into the underlying factual information underneath almost every sentence in that document. In our footnotes you can see a lot of that. Of course what they then did instead is fought it tooth and nail, so the world understandably thinks, Man, this must be an incredibly damaging revelation for the Bush administration because they are fighting so hard to keep it from coming out. It was a tremendous gift to the commissioners who wanted to use this to bash the administration.

Do you think that particular brief was behind the vigor with which the White House fought making the entire pool available? Or was it just a more principled opposition on executive privilege grounds to having the brief—

More the latter. From previous work in declassification issues, including on the State Department’s historical advisory committee, which I served on in the 1990s when [Madeleine] Albright was the Secretary, I knew about CIA’s particular neuralgia surrounding anything having to do with declassification of PDB material. Given that longstanding view, then interacting with the particular constellation of personalities in the executive branch at this time, it was not a surprise. It was a mistake on their part, but I understood it. It was just a very serious mistake and misjudgment on their part. And indeed it creates and redoubles a sense of mystery around the PDB that I think is unwarranted.

You had mentioned earlier that joint interview. The problem ultimately was that you didn’t get enough Cheney there. Would you care to elaborate on that?

Yes. We didn’t get to have a really good one-on-one direct interview with Vice President Cheney, which would have been more helpful to us, and that was unfortunate. If I think back on what issue before 9/11 could he have offered us critical revelations that weren’t available from the documentary record? Hard to say. But it certainly would have been better.

Oh, incidentally, when I told you the story earlier about how we handled the morning of 9/11, we came up with the staff statement. There were at least two occasions, I think this was one of them, where the Vice President was so furious about what we planned to say as a staff, this was when they were first seeing what we were planning to say in the report, that the Vice President personally intervened to try to change what we were planning to say. I think this was one—

Meaning that he called Lee Hamilton?

Yes. He called Hamilton—Kean and Hamilton or just Hamilton, to try to intervene. I think this was one of the occasions.

And this was on the call about shooting down the aircraft?

Right, about how we planned to do this. The other occasion was the staff statement on how we planned to handle the Iraq issue and the issue of whether or not Iraq had been involved in the 9/11 stuff. I led the staff statement in the middle of May, I think of 2004, on this.

This is when it’s pretty good, since Bill Safire is very connected with a particular group of people in the White House, not everybody, and this is when Bill Safire wrote the column attacking me entitled, “The Zelikow Report,” where he said that I had hijacked these gullible commissioners to pursue my personal agenda, etcetera.

Which allegedly was what? I don’t remember that article.

Safire had been one of the people who had really painted himself into a corner in supporting the theory about Iraq’s involvement in 9/11 and this Czech story. So when we came out and basically said that the Iraqi government flirted a little bit with al-Qaeda, especially in the Sudanese days, but it didn’t come to anything. In effect, it never became—the way we put it was, “no collaborative operational relationship.” But this particular story, this Czech story, doesn’t appear to be true. Then Safire wrote this column blaming me and blasting me personally to delegitimize the Commission by saying it had been hijacked by a rogue staff led by Chief Rogue Zelikow.

Chief Rogue. That’s a nice tag. [laughter]

He may not have used that phrase, but I think if you go back to the column from May of ’04, you’ll see he uses phrases that are probably even better than that.

Put that on your business card.

So Philip, what issues, findings, recommendations, were most contentious among the commissioners?

The recommendation to create the National Counterterrorism Center, which was perhaps our most important process recommendation, was controversial, but mainly just with Lee. Because it was a new idea and folks had trouble getting their heads—he in particular had trouble getting his head around it and getting comfortable with it. But once I worked it through with him he was satisfied. It was not broadly controversial within the Commission.

One issue that was highly controversial among the commissioners was what to say about the FBI and what to recommend about the FBI’s future involvement in domestic intelligence. On the staff we had a fairly strong position that we did not favor the creation of a new domestic intelligence agency for the United States. We actually thought that it was better to allow the FBI to continue to try to improve rather than try to build a new domestic intelligence agency to take its place in this matter.

There was a lot of argument about that and indeed commissioners even got other staffers into the room to try to see if they could find divisions among the staff that they could then probe to help them in this argument. It was a tough issue and a tough call, but it came out more or less as you see in the report. There may have been a couple of other things, but the recommendations were not enmeshed in a huge amount of controversy. They went through them. There were some particular suggestions, maybe on the privacy and civil liberties board for example, other things, suggestions, about ways to change some of the language on the DNI recommendation. A few things here and there, but I don’t remember a great deal of controversy in the Commission about the recommendations.

The harder things, in general, were in the debate over the substance of the report itself.

Meaning what? What were the elements in the substance that aroused controversy?

In general, only a few commissioners read the drafts carefully. They tended to be the Democratic commissioners. That’s just my—

We don’t need to probe that. [laughter] But it’s surprising, because you would think Republican commissioners would be interested in knowing precisely what was said because of the current administration.

Probably the person who read the draft report more carefully than any other was Gorelick, but I think that was partly her lawyerly instincts. She read each draft with a fair degree of care.

The way we had it set up is we had distributed drafts to the commissioners. If they had issues with the draft, they would give their markups to me. I would try to incorporate their changes and put out new drafts they would then check. It was difficult for them. It was difficult for me. But in general the thrust on both sides a little bit was don’t lean forward in interpretations in ways that will be seen as bashing the Clinton administration or bashing the Bush administration.

Often there was a little bit of, “Well, if you’re going to take out that language about the Clinton administration,” John Lehman would say, “you need to take out that language about the Bush administration.” Stuff like that.

So then it reinforced the tendency to take a lot of overtly interpretative material out of the report. But where we drew the line was the presentation of the factual record, and there I really got my back up and I just fought. I won I think pretty much all those battles.

So if one read—I assume in the Commission’s records there are the previous drafts—if one read them would one see a lot more interpretation?

Yes, you would.

And would there also be a record of who argued for the deletion of X or Y interpretation?

Probably not a very good record of that. There might be a few things here or there that would shed some light, but it would be hard to trace. In some cases, too, the discussion would improve the text because the discussion would force us to say more about, “What is your source for that?” And then we would actually discuss, “What does the document say?”

It would sometimes have the constructive effect of being a little bit more like our own staff meetings, which tended to constantly, in hours and hours of internal argument, force people to go back to the earlier evidence. Actually, “What did the interview say? Well then, let’s say exactly that and no more,” just be even more rigorous. Though very stressful, that was helpful. Some of the Commission arguments had a little bit of this effect.

I’ll give you a concrete example of one of these arguments. There is some very interesting material after the attack on the USS Cole in October of 2000. There are some arguments that—we leaned forward in the interpretation of that as basically saying, in effect, that the Clinton administration waffled on this. They in effect say, “You didn’t have enough that showed it was really al-Qaeda.” That argument turns out not to be a very good argument.

We found some pretty good smoking gun material from inside the White House, where the White House seemed to pretty clearly acknowledge that they knew who was responsible for this. The notion is that we don’t know it’s al-Qaeda so we can’t strike. Sandy Berger is, meanwhile, beating on the Pentagon to give him a better set of military options than they had given him, to which he got a very poor response.

There are bitter complaints from people in the administration, say at the State Department, saying that this is really a terrible example of inaction, it’s just the exact wrong message to Bin Laden and al-Qaeda. We had some stronger language about that. All the interpretative language about that was cut out. But then some folks wanted to trim back the discussion you’ll find in the report of what we did find and what people inside the administration were saying about this. There we just drew the line, and if it came to a fight and we had evidence for what we were saying about the factual record, Tom and Lee and the Commission as a whole would back us up.

So if the report had included the interpretations that you and the staff thought ought to be in there, how much better of a report would it have been? How much was lost by taking out that stuff?

In retrospect, I actually think not very much. At the time I might have thought a little more because I, frankly like many others—different staffers, of course, get wedded to a different interpretive idea, and their idea that they’re wedded to might have been different from my idea. But I actually look back on it and find that the report has been very useful in performing its core duty, providing a foundation of factual knowledge that helps other people write interpretive works. Whether it’s Steve Coll or Lawrence Wright or others.

You must have been enormously gratified by the response this report got. You and the commissioners as well.

I was gratified by the response. Again, in the real world if you try to postulate what’s the ideal outcome of something like this, realistically, from a point of view of public reception how well read it is, what impact it has on Congress or in relationship to legislation, I have to say if you’d asked us to postulate what would have been the ideal outcome on any of those indexes before the report came out, I really feel like we were completely successful on all of those, as successful as we could have been under the circumstances. I can’t say I was surprised, because we had a design that in every way was supposed to produce these effects. But it’s extremely satisfying that the design—no one had done really in quite this way before—was maybe like the A-bomb designers when that flash occurred in the New Mexico desert for the first time. They may have been gratified, but they were not surprised.

How did you go about assuring a consistent editorial voice in something like this? Editorial voice may not be the right term, but a consistent voice in a report that presumably has an awful lot of authors, but maybe you could correct us. Did you draft the whole thing?

Oh no. Goodness, no. I couldn’t have in addition to doing everything else I was doing. I solicited draft material from people all over the Commission and I edited almost everything. I think everything in the report at some point or another went through my machine, in some cases many times. There is some material I more or less drafted or comprehensively redrafted.

There are two things that create the consistent voice. One is that process. And by the way, the others in the front office were also participating in some of this. I was more active on some chapters than on others, in part based on the quality of the work that was coming in to me and how much more I needed to do.

To give you an example, the plot team that was headed by [Dietrich] Dieter Snell turned out to be very good drafters, so I found very early on that their stuff didn’t need much work. I did a little bit. They just tended to write in the right style.

Other things, like some of the policy chapters, I got a lot of really good input. Ernie May did a lot of initial drafting, I would redraft Ernie’s, and then Ernie would redraft mine and so on. There is one chapter that’s the reflective chapter on hindsight and foresight that Ernie and I drafted together. So different stories—

That’s one secret. The second was the staff statement process, where we first arrived at our voice through this collegial process. I took the lead in drafting one of the first staff statements, the one on the Kuala Lumpur affair, because it involved the work of several teams, but then all that stuff, including the stuff that I would work on, would go through this incredibly stressful and difficult process of everyone sitting down and reviewing it line by line, often in meetings that would go through the night.

We would project paragraphs on the board one at a time and argue. It can be pretty wearisome after a while. As hard as that was, though, one of the payoffs was a wide collegial buy-in to what was done. A sense among everyone that they had gotten to be heard, that there was real evidence behind what was said. This is a real credit to Chris Kojm, who was channeling Lee Hamilton. Lee Hamilton did very little on the drafting of the report, but there is a way in which Lee Hamilton’s austere, terse, concise style channeled through Chris Kojm that had a significant stylistic effect on the report.

I think every story you’ve told about the working of this Commission has concluded with the staff working through whatever disagreements there may have been and arriving at something that you all could agree on and then presenting a solid front to the commissioners.

There were a few exceptions to that.

But in general. That is so difficult from the account in the Shenon book. How do you explain why he got it, in your view, so wrong? His is a tale of constant turmoil and disagreement with Zelikow big-footing his way in and stomping out dissent—

The biggest antidote to that is simply to talk to my staffers and especially to the team leaders and ask them about it. I’ll ride with that.

I had a staff of 85 people. About 90 to 95 percent of the staff is pretty happy and on board, 5 to 10 percent of the staff is not. Some of them, we had judged, were not doing very well and their work doesn’t end up getting reflected in the report. There was a particular person on Team 3 for example, a particular person on Team 6. The other staffers know who these people are, so part of the problem is the five or six staffers who were really disgruntled were happy to talk to Shenon. I predicted while we were still working that something like this was going to happen. The Commission is riding too high. You can’t write a book about a Commission like this that’s just a celebratory account and expect it to sell. The only way you write a book that actually is an interesting book to write has to knock us off our pedestal in some way. There are really only a couple of angles of attack. One is to go after Tom and Lee and the commissioners and the other one is to go after me. I’m a much easier target.

That’s one factor. Another factor is there really were lots of arguments and debate and stress as I was describing, but it was designed to get to the point where the staff came together and felt good about what had been done. But it was a very stressful experience. As you can tell just from our describing it here, you can begin to imagine what a pressure cooker this was. It’s the hardest thing I’ve ever done in my life. Up until fairly close to the end I wasn’t sure we would pull it off. It was a close-run thing.

Without even getting into the details of the final production process and all the things we had to do—the people working unbelievably hard, very long hours in this pressure-cooker environment with all kinds of stresses, some of which, frankly, I hadn’t anticipated when I took the job. I didn’t anticipate I’d be having all these public fights. I actually believed the administration when they said they were going to cooperate. I really did. If I had known it was going to be as confrontational as it turned to be I’m not sure I would have taken the job.

But the thing I’m extremely proud of is the degree of collegiality among the leading staff, among almost all of the staff. I think you can validate it; just talk to anyone who talks to the former staffers and finds people whom they know and trust and whose opinion they respect and then just ask them what they think. I’ll roll the dice with that.

The day that the final report was due—

By the way, just one more point about that. One of my team leaders, one of the staffers who was more disgruntled than some has gone out of his way fairly recently to let me know that, “I didn’t understand a lot of what you were doing at the time, but I understand it a lot better now,” and has said some very generous things.

It was a very tough process and some people were really unhappy, but then when you dig into the particular allegations they are problematical. There is a particular issue, too, with Shenon. I had a very big run-in with Shenon while he was covering the story for the Times. I had very good relationships in general with the journalists who were covering this, as best I could, but Shenon wrote a story that was based on a leak he had gotten from I believe one of the commissioners. Well, I believe one of our staffers was hired at the specific request of one of our commissioners, Richard Ben-Veniste. The staffer didn’t work out but I think became a source for several stories. We could triangulate where things could come from. At one point I even confronted Richard personally about some of this.

You did?

I did. Because this is one of the things that could really hurt us. It turned out fortunately the staffer was not positioned to know some of the most important secrets that we were working on at the time. So Shenon had gotten some leaks that said, in effect, that the Air Force could have stopped the plane hitting the Pentagon and that all those lives could have been saved. And that was going to be the front page of the Sunday New York Times. Now, when you get into the details of that—he had gotten this halfway in our investigation and it’s actually not true. There was never a chance of stopping the plane from hitting the Pentagon when you get into this.

I saw this story appear online in the middle of the day on Saturday and thought this would have a terrible effect because it would make the Commission look like it was about to conclude this, which we weren’t about to conclude. It would have an enormously pernicious effect in a variety of ways, including for the families. I called Shenon’s editors and actually went all the way up to New York and they ended up changing the story between the time that it first appeared online in the New York Times front page Sunday edition and the time it actually began hitting the streets after midnight. You can imagine the impact that that had.

On your relationship with Shenon?

Yes.

When you say—

By the way, the New York Times bought him out and let him go after all he finished his Commission coverage.

When you say, “all the way to New York,” you didn’t go to New York, you phoned editors all the way up to New York.

Right. First the Washington bureau chief and then it went all the way up to the editorial chiefs in New York because they’re actually going to have to rewrite the lead story in the paper they had already published. I had to convince them that these errors were so egregious that they had to do that. They were partly convinced that this was so.

You’ve earned lunch, but let me ask one question. This is about the release date of the report. This was by statute? Or had the Commission decided it wanted the report before the election?

Actually, we sought an extension of time and there was an argument. This was a significant argument among the commissioners about how long an extension to request. The debate by early ’04 came down to this: Do we go for release in July or do we actually kick it past the November election and put it out at the end of ’04 or early ’05 and get a lot more time?

The majority of the staff wanted to finish on time. A minority wanted more time. It was not a close call. Most of the staff wanted to go ahead and finish, but a nontrivial minority wanted more time.

It was going to be very hard for us to meet the summer deadline. I supported getting it out in the summer of ’04 and not asking for more time. But the commissioners had a real argument about this. It was a good, healthy argument and the bottom line came down. The Commission decided no, let’s just go with the July ’04 date.

My point was twofold to them. First of all, on the point of, can we find more work to do? Of course. Because Parkinson’s Law will operate and the work will expand to fill the space we allocate for it. But I thought we could get our job done by July and so did Chris and Dan in my front office.

Second, this was a 9/11 election in a way; 9/11 had been the dominant occurrence in the President’s first term. This was the Commission that was supposed to tell the American people about that. It seemed that it was obligatory for us, whatever we found, to make that available to the American people before they voted on Bush’s reelection. The facts that we could find about this needed to be out there before the election as a matter of civic obligation, rather than sheltering it from politics.

I also made this point. I said, “The way we’re going to write this report, the report will be examined by both sides and each of them will quote selectively from it. We will actually put the report in the hands of all the American people. The report will itself become a battlefield in the election campaign.” That’s exactly as it should be, I thought. While we’re bound to the benefit of the report, if you put the report directly into people’s hands rather than force them to rely on newspaper accounts of what it says—we had a publishing strategy that would do that. Yes, they should argue about it, yes, they would quote facts selectively from it, and it would be kind of a Rorschach blot for every one of the factions. That was the way democracy should work.

Sounds like the choice was not whether to do it before the election or after the election but before the conventions or after the elections. We’re talking about July or December, you’re saying, not during the general election campaign. Is that right?

If you kicked it past July you were looking at another six months. There was not really an intermediate argument in which folks thought we should put it out in August or September.

OK. And when the Commission was debating whether to ask for this delay—

At one time we were looking at a deadline in May and then I think we asked for an extra two months to put it out in July, but then we decided let’s just draw it in July.

OK. When the commissioners were arguing about this business of a delay was there any partisan character to those who were for it and those against it? Were the Republicans wanting to put it after the election and the Democrats wanting it out sooner?

I’m not sure. My memory of this is that it didn’t divide neatly on partisan lines but that there were people on both—I think in general the Republicans wanted to get it over with.

Oh really. OK.

But I think the Democrats may have been more split on this rather than all wanting to get it out later. I think Tom and Lee tended to defer to what most of the staff wanted to do.

There were other problems, too. If we had actually delayed six months longer we would have had some problems with retaining some of our staffers who already had on the previous dates made job commitments.

Why don’t we stop and have some lunch. I think they’re going to bring it in here and I think we’re in a good place.

OK.

Thanks, Philip. We appreciate it.

P. Zelikow, 10/4/2010 40

[BREAK]

We’re recording here. Anything else we want to get on the 9/11 Commission? Is there something we’re missing?

Is that when your role with the 9/11 Commission ended, when the report was published? Or did the next phase begin?

Much of it ended. I was very active in the remainder of 2004 in working on the legislation. I testified at the hearings that Congress immediately began holding in August of 2004 and helped place some of my staffers on the congressional committees that were writing the legislation. Then I gave testimony that helped provide some template for the design of the NCTC [National Counterterrorism Center] in the legislation itself and was involved in some of the negotiations with a little spot of controversy during the fall of 2004.

They were, of course, very difficult negotiations, as they had been in 2002 when I had been working on the Help America Vote Act coming out of the [Jimmy] Carter-[Gerald] Ford Commission exercise. The standard accounts of all this, to the extent that they’re beginning to surface, are all written from the point of view of one side, which are the protagonists. They are frustrated that the legislation is not as strong as they wish it had been and there were some tough negotiations on how to handle this.

In general I was considered a key ally of the good guys, of the people who were supporting the strong version of the legislation. Jane Harman on the House side and Senator [Joseph] Lieberman on the Senate side. [John] McCain, [Susan] Collins, but along with Senator Collins’s staff, the real trick here was actually to carry the House Republicans. That’s where the problem was. And at the committee level among the House Republicans the problem was very difficult to solve. The key interlocutor to get this bill through the House—because the Pentagon I’ve come to believe may have played a bit of a double game in the way they tried to manage the Commission and then beat it. On the Hill working with Duncan Hunter, who was the House Republican Congressman chairing the House Armed Services Committee and who is, of course, a very influential figure in this legislation.

The key factor, the reason we got this bill, is because the House Republicans were bludgeoned into going with it. The House Republicans generally didn’t like the bill. The reason they were persuaded to go along is two factors on the Republican side. One was actually that the President and the White House supported the bill, so White House input on the bill was valuable. Though other figures in the executive branch were giving different signals, the President himself and the White House staff supported the bill. The other force was the House speaker, Denny Hastert and Hastert’s chief of staff, a man named Scott Palmer. Hastert and Palmer I think are a little bit the unsung heroes in getting this bill passed. Since they are not one of the so-called “protagonists,” they don’t get any credit. I’m not sure if the bill would have passed if it hadn’t been for Hastert and Palmer figuring out a way to bridge this so they could get enough of the House Republicans on board.

Now, partly this is a problem of their own making. You may find this interesting because I know you study American politics from time to time. Hastert had an internal rule that, basically, if you could tell Hastert, “Hey, why don’t you just shave off 20 or 30 House Republicans to join with the Democrats and we can get this,” Hastert didn’t do business that way with his caucus. Hastert had an internal guideline that he would not support anything unless he could actually get a majority of his caucus supporting it. And this is the way he helped maintain his leadership within the caucus.

You can debate the merits of this approach, but that was something that they cared deeply about. Therefore it wasn’t just a matter of peeling off some loose House Republicans, it was a matter of carrying a majority of the House Republican caucus. Unless the President cared so much about this that he leaned on Hastert so hard that Hastert would break this rule, but doubtful.

I got involved in some negotiations then on how to find the right bridge language, and basically we found some bridge language that was upsetting to the protagonists because it diluted the bill a little bit. I became convinced that this was enough language to get things going and that if you had the right people it could work for this stage.

In doing that I had privately canvassed on this with people like Jim Clapper and Mike Hayden and others who could not talk to me openly but would talk to me in back channels about what they wanted and what they thought they needed and who privately were supportive of what the bill was trying to do, even though publicly they could not say so because of where Don Rumsfeld was on it.

So I then, at a key moment, not only supported the bridge language but wrote a letter that Hastert could circulate around saying, “Look, Zelikow supports this bridge language.” My allies on the protagonist side were furious with me for having written this letter. Harman called me directly and read me the riot act, but I believe—I may be mistaken and Hastert and Palmer would be good sources for this—that helped grease the wheels for the adoption of the bridge language and getting the bill passed.

The protagonists would have the argument, and they may be right, I don’t think they are, but they might be right, that, no, if you had not given them that bridge language we could have bludgeoned them into voting for what we wanted to do anyway. The political pressure was in such a place that if we had held out just that much longer we would have won. Maybe that’s right. Others are not sure, but I basically ran with Hastert and Palmer’s judgment of this and that’s what they thought they needed to carry the day, so I helped them on that. But I alienated some erstwhile friends by doing so. Other than that, most of the help I was giving during this phase in the second half of 2004 was more ordinary business.

Did Condi Rice contact you at any point when she was considering going to State?

Yes. I had been very careful after I took the 9/11 Commission job; I had really virtually no contacts with Condi. I’d met with her early on as I have described to you, and then I had another set of conversations with her because I needed to lead an overseas delegation to interview Saudis in Saudi Arabia working with the Saudi authorities and go to the war zone in Afghanistan and go to Yemen and go to Pakistan and interview leading Pakistani authorities. I needed the U.S. government to help me do that and Condi could make that happen. I needed military cooperation and so on and she agreed to that and facilitated that.

That was in late summer, early fall of ’03. Except for that famous fax—we were leaning on her to testify, and I faxed her a photograph of a prior—Admiral [William] Leahy, a predecessor in her job, testifying at the Pearl Harbor hearings—and had a little note on it, “Condi, a picture is worth a thousand words. Somebody is going to publish this picture soon,” to show that there is obviously a precedent. I don’t know whether that helped, but it may have contributed to the decision they soon made that she needed to testify, which, of course, she did.

Saudis were cooperative? I don’t want to go too far back into this.

No, actually, the Saudis were cooperative in their way. That’s not the most efficient and energetic government in the whole world, but yes, they were cooperative. That’s partly because of where the Saudi government was at that point. There are some good books like Hegghammer’s book on jihad in Saudi Arabia, which can give you a good sense of how this evolved in Saudi Arabia. By this time al-Qaeda was going to war against the Saudi government and the Saudis were getting scores of their people killed fighting these folks. They were happy to help.

Also, of course, the American media was in full cry with allegations that the Saudi royal family was somehow involved in 9/11 or in covering up 9/11. They were happy to facilitate investigation where we could talk to people for ourselves. The congressional joint inquiry, which had a lot to say about “the Saudi role,” had, as best I could tell—and I went through all of their investigative records—never talked to a single Saudi, official or unofficial of any kind. It actually helps to talk to Saudis and spend time in Saudi Arabia. It’s a complicated country and it helps to understand it.

Mel, are there any other questions about these two other consultations—

Yemen, too, by the way.

When Philip’s abroad for 9/11?

Just one more thing before I leave the 9/11 Commission, because it’s actually resonant today. Yemen was very important and it actually connects very well to the Saudi point because the issue back then was, did the 9/11 attackers have a support network in the United States of secret al-Qaeda agents? The allegation was, yes, they did, and they were all connected somehow with Saudi oligarchs of one flavor or another. We looked at that pretty hard and the evidence for that wasn’t very persuasive, but there was some pretty interesting and disturbing evidence that they did have a support network in the United States. It turns out that the people who seem to be in that support network were a couple of Yemenis. The distinction between a Saudi and a Yemeni can depend on what year you are looking at.

Bin Laden himself was once a Yemeni. His family comes from a tribe that’s based in Yemen. So their world is a little more complicated than our neat territorial divisions. But the couple of key Yemeni figures were very suspicious. One of them, by the way, was a man named Anwar al-Awlaki, whom we spotlighted as a person of great concern to us in our report in 2004 and would later reemerge very much in the news in America about five years later in connection with various things.

One of the things that struck us at the time is, in our report, in our kind of terse, matter-of-fact way, we said, “We have great suspicions that there was a support network. Here are some of the people we’re most suspicious about.” We said that and there was no particular media notice or reaction to that at all about people like Awlaki or Mohdar Abdullah, who’s another figure we singled out for particular interest.

Another thing that was in our report that we hid in plain sight was a very disturbing couple of pages on Iran. For instance, Iraq flirted with al-Qaeda, but we concluded never developed—the phrase Doug McEachin and I worked out was “collaborative operational relationship.” The Iraqi government and al-Qaeda never developed a collaborative operational relationship, though they were intrigued by each other and took some notice.

What was more disturbing and interesting was that there were much more substantial contacts between al-Qaeda and Iran. A lot of people who had studied al-Qaeda were predisposed to think that that’s impossible because the Sunni and Shia could never get together. We had found evidence even in the mid-’90s that this Sunni group was getting material assistance from Hezbollah. They both had common enemies in the Saudi government during the 1990s and there may have even been some connection between al-Qaeda and the Saudi-Hezbollah bombing of the Khobar Towers barracks in 1996, in which the Iranian government was involved. It’s even possible that some al-Qaeda got some training in Lebanon. Then we had a couple of pages in which we said there are some very peculiar things we have unearthed about the movements of the 9/11 hijackers and other al-Qaeda people through Iran in the months before 9/11. We listed the information we had about it.

None of that proved that the Iranians were witting of the 9/11 plot itself, and I tend to doubt that they were. But there is some evidence that the Iranians were supportive of what al-Qaeda was trying to do and that they tended to facilitate al-Qaeda’s movements in the pre-9/11 period. We flagged that as something we thought deserved a lot of additional attention from the U.S. government, and then those allegations got very little notice and I never saw much evidence that our government followed up much on that lead. So we didn’t actually tie off every loose thread in the report.

OK. The original question was about contacts with Condi.

In the fall of 2004, after it became clear that Condi was going to be put forward as the next Secretary of State—I’m trying to remember who contacted me first. Either Condi Rice or Bob Zoellick, whom she had recruited to be her deputy. My recollection of this is that although I talked to both Rice and Zoellick, that Zoellick played a more active role in assembling this team than Rice did.

Rice put together—what role, if any, I would have remained murky for a while but they were clearly trying to put together the all-star squad from the Bush 41 years. Zoellick, me, Nick Burns, others who hadn’t ruined our reputations yet. So eventually I think Zoellick may have played a role in this.

It worked out that I would be the counselor of the department. The counselor of the department had been a job that Colin Powell simply had left unused and had almost disappeared from the department. It’s actually a very old job in the State Department. It was the title that was given to the No. 2 man at State in the [Woodrow] Wilson administration. Then it subsided. The No. 2 man became what was called the Under Secretary of State. But then the counselor’s job was revived again in 1937 for the ironic reason that FDR [Franklin Delano Roosevelt] wanted Cordell Hull to have one guy as the No. 2 and Cordell Hull wanted to have his friend as No. 2. So FDR was able to take his guy, Sumner Welles, as the formal No. 2, and then Cordell Hull got his buddy, a Virginia Congressman, to be his No. 2, reviving the job as the counselor.

That was in ’37, and more or less since then the counselor has been a position where the Secretary parks a sort of Minister Without Portfolio. Through much of the Cold War this was a place where you parked your broad-gauged U.S.-Soviet person, though there are exceptions to that and there have been some good people in those jobs. [George] Kennan, [Charles] Bohlen, and others, though, interestingly, when Kennan was put in that job he was being pushed upstairs.

He was on his way out.

Right. When Kennan had been at Policy Planning, Chip Bohlen had been the counselor, and one of Bohlen’s jobs was to help look after Kennan and the Policy Planning staff. Occasionally, the counselor supervises the Policy Planning staff. Sometimes the job has worked and it’s been very important; Max Kampelman for George Shultz was an important figure. But sometimes not.

When you took the job, what was your understanding of what you would do?

Well, it was really murky, so I actually invented what it is I would do. I brought that to Condi and asked her to sign off on it. The first thing I was asked to do, even before I was in the job, was to go to Iraq.

But you just said a second ago you first defined what you wanted to do and then you went to Condi. So what did you want to do?

As best I could tell neither she nor Zoellick had had the time to develop a very clear idea of what it is I would do. So in effect I said, “Why don’t you make me your deputy for all intelligence issues, including all the covert action intelligence policy issues? Make me your deputy for all terrorism and homeland security issues. I’ll go to the Deputies Committee on all those issues, and use me then as a rover to help work on some difficult problems for which I will not be a line manager.” Of which Iraq was first and foremost. Iraq, India, and North Korea were the leading entries on that list.

In the first two instances, are those likely to set off turf wars with anybody else?

Yes. One thing about being the counselor, as it is in being in any of these staff jobs, everything you do imposes on someone else’s turf.

OK.

I also got involved with Iran and the Arab-Israeli-Palestinian peace process. So basically, even in the terrorism stuff there was a little bit, but people tended to defer to me a little more on the terrorism-intelligence issues. But on all the major foreign policy issues that I got involved with, everything I did was in someone else’s lane.

One thing, though, I worked hard—I keep trying to learn my lessons and get better at this stuff as I get older. I tried to find ways of being collegial and really working with people, and one of the things I’m proudest of about that experience is that with one exception I had good personal relationships with the line managers on all these issues, even though I was spending a lot of time working in their issues.

Tell us—

The exception was North Korea and Chris Hill who had the East Asia bureau. I didn’t get along very well with Chris, but nobody did. So I wasn’t exceptional.

For the record, why don’t you tell us who were your key allies on particular issues, from whom you got your most important information and what were your links across departments and with whom?

The constellation varies on every issue.

Sure.

I ended up spending quite a lot of time among all the basket of intelligence, homeland security, terrorism. By far the most contentious of those was that I waded right into the detainee interrogation arguments, which were then coming to a climax. I was involved centrally in those battles, and that was pretty tense. There, my indispensable partner in the department, and actually my only ally, was John Bellinger, the State Department’s lawyer, who had been Condi’s lawyer at the NSC staff.

I was the policy deputy at those meetings. At the first one of those meetings I tried to go to they pushed me out, claiming I didn’t have the relevant clearances. That infuriated Condi because I did have the relevant clearances. But this is one illustration of how they tried to keep possible opponents out. You understand—

Who is the they? [laughter]

Well, J.D. Crouch chaired the meeting and Crouch said, “We’re not sure you’re cleared for this.” It got straightened out, but as I say it really set Condi off. I think she called Steve Hadley very hot about it.

Crouch was Hadley’s chief assistant?

Crouch was Hadley’s deputy and he chaired the Deputies Committee meetings for most of this period.

Lara Crouch is Crouch’s daughter, right?

Yes, that’s right. J.D. is someone who I had known from the Bush 41 days when he had been a Deputy Assistant Secretary of Defense in Cheney’s Pentagon.

Was the Vice President represented at that meeting?

Yes. The Vice President was always represented at these meetings. By this time I had heard a lot about the Vice President and his staff being so large and forceful in these matters, but for whatever reason, by this time this was no longer so true. They picked their spots. But one subject that they were always deeply involved in was terrorism and CIA and these detention issues.

And who are the Vice President’s folks?

David Addington.

Addington.

Not [Irve Lewis] Scooter [Libby]?

No. Scooter didn’t really go to these meetings. Scooter stayed behind the scenes. Then by the beginning of ’06, Scooter was already in trouble and was pushed out or had to leave.

So Addington was the key figure, but on those issues the environment was extremely difficult because at first we had no allies. At first we thought we had no allies outside of the State Department, none. At these meetings everyone would be against us. Then I got one ally and then lost him. We actually developed an ally in the Deputy Secretary of Defense.

Who’s that?

Gordon England, who was Don Rumsfeld’s deputy, and to whom Rumsfeld had said, “You look after these Guantanamo issues for the department.” England had working for him people like Matt Waxman, who had also worked for Condi. England is a pretty sensible fellow, and he wanted to get on top of these issues. Important note: There is a very important recommendation in the 9/11 Commission Report that I didn’t discuss earlier that said that this war needs to be fought as a coalition war. In order to do that you need to have coalition rules of engagement, and that means adopting Common Article 3 of the four Geneva Conventions on the Law of Armed Conflict as your base standard for the humane treatment of detainees.

As a sop to the theologians that I knew were in the Bush administration legal world, I said do this as a matter of policy. You don’t have to do this conceding that you’re required to formally by international law. But do it.

The Bush administration enthusiastically accepted practically all of the Commission’s recommendations, but not that one. I was very proud of having authored that recommendation and sold it to the Commission, and proud of it too, because it helped put the Common Article 3 point on the public agenda as an important—I wasn’t the first person to notice it, but I think it did a lot to help put it squarely into the foreground of the public debate and also influence the development of the later McCain amendment that became another really important vehicle in this.

So huge battles on that, and the story is recounted in two or three books now. England and I developed a paper to try to break this issue wide open and a key dimension—Frankly, the folks on the other side were expert bureaucratic warriors and did everything that they could to circumvent the normal process when they wanted to.

Who was on the other side? Addington and who else?

Well, everyone. Let me tick through them. Addington would occasionally lose his temper and flame out, but in a way he wasn’t the most serious opponent because you had the White House counsel, main Justice, the intelligence community, the CIA.

Why don’t you tell us the names so that the record has the names.

Oh sure. White House counsel, that was Al Gonzales. The main Justice, John Ashcroft, soon to be replaced by Gonzales. But the people, now you get into the OLC world and Dan Levin and Jack Goldsmith had just been pushed out of OLC. They would have been potential allies because of the things they were trying to do during ’04 but they had now been pushed out.

Gonzales goes in and they basically clean house at OLC. They have [Steven] Bradbury as their guy in OLC, who they will not confirm in his position, so he is uncertain about his future. Gonzales is now getting the party line back in place, and then they write this remarkable set of opinions in the spring of 2005 that have now been declassified.

Reversing the earlier.

Not so much reversing—they did not—I started to get into the details of this, but the ’04 opinions had taken on one of the most egregious of the post-9/11 opinions and thrown it out. They had not yet gotten to the hardest issues. That was the next step. They were going to get to the hardest issues, really sizing up the CIA detention program, not just throwing out this terrible opinion on the definition of torture. Then they were going to have to really come to grips with the whole CIA program. They were stepping up to that when they all left, and Dan Levin, who had worked with us as the liaison to the 9/11 Commission and then had gotten transferred into other work, had now pretty well been pushed out.

Gonzales is now Attorney General. Bradbury is at the OLC. The new White House counsel is Harriet Miers, and all these people are supporting the status quo. Of course the Vice President’s office. Also, at CIA, even though Tenet has left, his replacement, Porter Goss, is a complete supporter of the status quo as well.

So Philip, tell us—

And then the Defense Department—we thought we had this ally in England, who then—I drafted this paper in the spring of ’05. The paper was leaked, not by us but by the Pentagon to the New York Times. But I drafted the paper and I think the paper itself is unclassified and is now somewhere in the public domain. Yes, it’s in the public domain.

I went over to England’s office at the Pentagon and on their word processors we finalized this paper. This was a paper that essentially was to the President. We were just going to do an end run straight to the President. One reason we thought we could do this is because Condi increasingly, and also Karen Hughes, believes that the President’s position on this point is untenable.

What the President needed was for someone to lay out to him, “Here’s how you can make a big move on this. Here’s how it can be done.” So we wrote kind of a Big Bang. Comprehensively, instead of waiting for this to eat you up, here’s how you can get ahead of this whole issue and proactively address it, including bringing the 9/11 suspects to trial at last, which absolutely ought to happen. Instead of putting them in a hole without knowing what you’ll be able to do with them.

So we write this paper. We circulate this paper at one of J.D. Crouch’s meetings and meanwhile through other channels we tried to get it to the President directly. Because the paper was written as building blocks for a Presidential speech. So if the President liked this, it could be turned into a speech literally in a week. And just jump. What happened was, first Cheney and Rumsfeld reacted very strongly. Rumsfeld told Gordon England, “You no longer work on these issues and you no longer go to these meetings. Steve Cambone will now go to these meetings instead.”

The deputy says, “This paper is inadmissible. It’s not on the agenda.” We can’t discuss it, and I have reason to believe the President may have actually seen the paper, but it could not be taken up. The bottom line was not so much that we were beaten but that we were then forced to go back, and instead of trying to win this with a coup de main, we had to go back to trench warfare. So the [Alfred von] Schlieffen Plan failed. We had to fall back to our trenches and then had trench warfare for the rest of the year, literally fighting inch by inch on every single one of the connected issues.

What was your anticipated channel into the Oval Office? If you’re not going to take this through normal channels, if the Deputies Committee is saying this is out of bounds?

Karen and Condi had their contacts with the White House.

What motivated this?

Possibly Andy or Josh, too.

Is this an illustration of human agency? i.e. Zelikow gets into this position and really is attached to this issue, partly because of his work on the 9/11 Commission, so he’s familiar with it? Or is this a more organizational motivation? That the State Department has found this issue to complicate its relationships with critical allies and therefore the Secretary of State wants to get involved in this in a constructive way? What’s the underlying motivation here?

Yes. Both of those things are true, and on the second point I would add that both Condi and I came to the conclusion—I don’t know when she came to it—that this issue, which I’ll call the whole Guantanamo cluster of issues, Guantanamo is just a euphemism you can throw on top of them, but this cluster of issues was hurting us more in our foreign policy than any other issue in the world. It was hurting us worse than Iraq.

And the Secretary of State believed that?

Yes. She did.

How early? Had she believed that when she—

Now I want to come back, because here is what I think is the major part of the story.

OK.

Unfortunately, I’m not a critical agent in it. Partly I don’t have hard evidence on how this happened, but here’s what I gather.

I believe that sometime during 2004, heading into 2005, the President himself becomes uneasy about these policies. I don’t think it’s fair to say that he thought he had made a mistake. I think it is fairer to say that the President became conflicted and ambivalent about them and was uneasy himself and that Condi was in the same place. Why?

Again, part of this is supposition. I think Abu Ghraib had an impact. I think the CIA IG [inspector general] report they got in 2004 that John Helgerson had prepared had an impact. In general they were now reflecting more and more on this and were just uneasy about it.

I think the President was genuinely conflicted. There is a side of him that was very uncomfortable with it, yet he also believed, as he was always told to believe, that these techniques were indispensable to protect the country. This is what he was told by his intelligence experts and this is what he was told by the Vice President.

Why was he uneasy with it? What’s your supposition? Because of the morality involved in it?

Yes, I think the moral issues bothered him. He is a person who does moral reasoning for better or worse. He cares about moral issues and I think that had an effect. Of course you can solve the moral equations through various devices. You can dehumanize the people who are being tormented as being some sort of despicable nonhuman category.

As I put it in the paper that I drafted to England and I sent around, “This policy is not so much about what we say about these people. It’s what we say about who we are, not who they are.” I think I even put that in all caps.

I think that actually bothered the President. It bothered Condi. Then there were prudential issues, too. It bothered some people on the Hill. It bothered some of our allies. It was creating problems for us that became increasingly evident, even in our intelligence operations. It was not just a matter of public opinion.

The point of all of that is to say that’s why, clearly, the President decided to allow people to begin dissenting.

You didn’t dissent because the President said you could dissent.

I did not. I only get to dissent because Condi lets me. I work for Condi, and I get to play this role that I’m playing because Condi lets me play it. Condi wouldn’t have let me play it if this issue was not open for discussion and debate. The President, I believe, signaled to various people that he was ready to allow his officials to discuss and debate this program and these questions.

That’s why these internal debates begin occurring in 2005 and gain increasing momentum all through that year. If Condi thought that the President thought that questioning this policy was out of bounds, she probably would not have done it. And if she did not think it was appropriate to question these policies, she would not have authorized me to work on the problem.

But again, just to focus on the point, your sense is that the environment changed and brought Condi around in 2004-05, about the same time that she’s moving—

I can’t trace the exact development of her views on this subject.

But the natural question is, if she’s been in the White House and presumably responsible for orchestrating foreign policy in a stronger or weaker fashion, why does she wait until she gets to the State Department before she starts pushing?

I suspect but cannot prove that she was uneasy about it before she left the White House. The circumstances under which these policies were originally adopted partly went around her and partly through her. I am not her defense lawyer.

Sure.

I’m not trying to exculpate her of responsibility for the adoption of these policies in ’01 and ’02 or the quality of the analysis or lack thereof that was associated with those policy decisions. But I also believe that at this period Tenet was the Combatant Commander of the War on Terror and Dick Cheney was his Secretary of Defense and Condi was not firmly inside that chain of command, though she was knowledgeable about what was going on and saw a lot of things.

As some of the press accounts point out, things like the Geneva Convention issue went around her, because once Cheney had the President’s support for something, that was the end of the story as far as he was concerned. Cheney didn’t submit his proposals to interagency discussions because he was above the interagency process; he wasn’t part of it. That was one of the defects of putting the Vice President in this kind of line management role in a major policy of the government.

Condi had been fully aware of what they were doing and had been part of it, but somewhere along the way, I don’t know when, had become uneasy.

What’s the date of the Abu Ghraib revelations?

Early ’04. Maybe January.

That helps to illustrate the timing of the two.

I just know that by ’05 she believed that the President wanted to hear alternatives, so the paper I wrote with England she had authorized me to write. She knew I was doing that. Karen Hughes, by the way, was very much in Condi’s camp on this.

Who was?

Karen Hughes.

Karen Hughes. Just in the terms of the way we understand bureaucracy working here, were there people in the European office of the State Department coming to you and saying, “It would be a good idea to address this issue”?

No.

How about political-military?

No one from inside the State Department asked me to address this.

So it’s your personal judgment, Condi Rice’s personal judgment, that this is really having a huge impact on our standing with our allies.

Yes.

But the people who were in the department dealing with our allies weren’t telling you that this is a big deal.

No. Of course if you ask them, is it a big deal, they would say, “Oh yes.”

But the pressure wasn’t coming from them.

No. And it’s actually not their nature or their habit in the EUR [State Department’s European desk] bureau to write memos saying please do something about X. Maybe it should be but it is not. That’s just not the way they—

No, but that’s important to say.

It is. Or in the Pol-Mil Bureau. And actually, Bellinger’s role in all this is even a more complicated story perhaps than Condi’s because Bellinger is present at the creation, and in all of this stuff. I think Bellinger will become, eventually, a notable dissenter who joins me in the trenches, then after I leave government he continues to wage these battles for the rest of the second term with some courage.

I believe that Bellinger’s views actually evolved over the course of the first term in ways that he can better recount than I can. Even by the spring of ’05, when these OLC opinions went around and I got to look at them, they had come by the State Department, by John’s office, and John, frankly, felt that he wasn’t in a good position to comment. In effect, they had gone through, he had signed off, though he didn’t feel like he was entitled to have much voice on them.

This is important, because later on I went squarely after these opinions and said that one of these opinions was bad law and wrote a legal memo making the argument.

This comes later.

This comes later. I needed to do it later on because of the legal position that we finally forced them to fall back to, which was an untenable legal position. It needed to be attacked squarely, even though bureaucratically, formally, I wasn’t in the position to tell the Justice Department how to interpret American constitutional law. Their interpretation was the whole way they managed the treaty interpretation. Since I had worked in American constitutional law, I thought their interpretation was so radically indefensible that someone needed to tell the President that his lawyers were adopting a position that other reasonable lawyers in America would find laughable.

That was the position on which their whole legal status was now resting after the passage of the McCain amendment at the end of ’05. That’s why I wrote a memo in February of ’06 that went squarely after them. This is the one that they got so annoyed about that they tried to collect all the copies so that there would be no record or traces of it.

You said that you had gotten the OLC memos, but that’s not in ’05, that’s in ’06?

No, I got the OLC memos that were written in May of ’05 by early June of ’05, but at this time this stuff is really closely held. At this time, there are only three human beings in the State Department who have access to these papers. That’s Condi Rice, John Bellinger, and me.

By virtue of your clearances or by virtue of who—

By virtue of clearances and how closely held information about the CIA program was.

Those memos were more recently drafted. Those were not the John Yoo—

Right. These memos were more recently drafted that were finally the new opinions on the legality of the CIA program, which they had ducked in those earlier opinions and which Goldsmith and Levin were circling around to tackling when they left office.

When you’re getting your views to the President, how are you doing that?

Well, we tried in this one case through back channels that I didn’t see. We generated a piece of paper that in May of ’05 may have gone to the President. I heard it might have gone to him. I don’t know how.

After that, though, we were just fighting it through the line arguments and that just rested on I would do battle in the Deputies Committee. Then occasionally you’d get principals meetings and occasionally you’d get full meetings chaired by the President, especially by 2006. By 2006, incidentally, you had meetings where—and this was very unusual—Condi and the Vice President argued vehemently head to head in front of the President at a couple of these meetings.

This is all, again, under the rubric of the Guantanamo stuff?

Well, there are all kinds of issues now that you’re having to litigate internally that are then being affected simultaneously by the moves that are being undertaken in the Congress and eventually by the courts.

The final coup de grace—we were making headway internally, but then the Supreme Court really helped us in the summer of ’06 and broke the dam open. We were able to get the President to make the September ’06 move that’s really the turning point for this issue.

But there are a whole lot of issues. There is first an issue, which was not so hard for us, on rendition, which was more manageable in my view. Then, what do you do and what are the standards of treatment for confining these people? What are the standards of treatment for interrogating them? If you bring them back to the United States, do you bring them to trial? If you bring them to trial, do you use military commissions? If you keep military commissions, what are the procedures for those commissions? What should be the future of the facility at Guantanamo Bay? Should that be a permanent facility or should we try to close it down? If you try to close it down what would take its place?

That’s a sampling. Underneath all those issues there are other issues.

Cheney and his subordinates were your leading antagonists on this issue—

Yes, but at the meetings themselves they didn’t have to carry most of the water. Others would do most of the work. Crouch, at this time at least, was not terribly sympathetic. Well, no one was sympathetic.

Crouch.

Harriet was not sympathetic. Main Justice. Each relevant line agency, depending upon what the particular issue was, would defend the status quo, and if necessary the Vice President’s office would intervene. Occasionally Addington would say something hot and I would say something hot right back to him, but by that time in the interagency process, Addington himself was seen as—the point I want to make is that I did not regard Addington as my principal problem. Addington would speak out and other folks would roll their eyes a little bit. My guess is when I would speak out they would roll their eyes a little bit, too. These were the neutral, gray men in the middle, and the neutral, gray men in the middle were behind the status quo. Actually, they were tougher opponents than Addington.

The nice thing in a way about Addington, although he’s in some ways not a very nice person, is that he’s very straight and open about what he thinks. He comes right at you, usually, so you can go right back at him.

What were the main motives of your opponents? Was it vested interest? They already made decisions and therefore were sticking to them? Did they really argue that they were getting important information through existing processes? Were they worried that if you changed policy the President or the country would look weak?

They made all of those arguments. Yes. First they argued that these interrogation methods are indispensable to collecting intelligence that’s needed to protect the country. I would challenge them on this. I’ve written a little bit about the argument, so I won’t recapitulate now what I’ve written about this argument.

But in challenging them would you specifically say, “Tell us what important information you received?”

Yes.

Were they able to provide illustrations?

Yes. Ultimately, J.D. would say, “The President needs some talking points or a sheet of paper that recounts success due to these interrogation methods.” But this is not so much used to answer the State Department. This is so that they can go up to the Hill and try to push John McCain aside.

They would develop those talking points, which would have their list of successes. If you want to know what those are like, listen to the speeches Dick Cheney and Lynne Cheney and others gave after the OLC opinions were declassified, because they are essentially the same talking points that were given to the President in 2005 and 2006. There are a couple of problems with that. One is that there was actually no independent review of the assertions made in those papers.

They were incredibly vague, as I recall.

Some were. Some were not. But I knew a little bit about some of those cases and I had other friends in the intelligence community who would tell me more about some of those cases. What happens, of course, is a lot of times you get an intelligence break of a mosaic of information that is made up of many pebbles. One of those pebbles might have come from one of these interrogations, but that maybe wasn’t the most important source of pebbles. Right. And then they would say, “We stopped the Heathrow attacks because we got some—” That’s a success. If you dig into where did we get the intelligence for that particular prevention, what’s the backstory behind it, you get into the intelligence world where a lot of people who contribute to a success claim that they contributed to a success. You want to encourage them in that way, but it can be misleading.

Then there’s the problem—A, a lot of examples they’re giving to the President have not been scrubbed dispassionately so that you can get a proper sense of perspective. B, there is no analysis of: Yes, but do you need these particular methods to have gotten that information from the interrogation? And, of course, there is no real cost-benefit analysis at all.

In other words, there is no serious analytical process that is applied to producing talking points that are almost designed more for a political purpose than for serious, rigorous analysis. No one had ever really done a rigorous analysis about whether these methods were uniquely necessary.

The Vice President is often said to be a person who is pretty rigorously analytic. Was he not rigorously analytic on this point? What do you sense motivated him?

The honest answer is I don’t know. I have tried to read all the available information on this and occasionally ask questions. My hypothesis for reconstructing this points to a set of key decisions that were made in late 2001 and early 2002 in which this stuff was sold largely on anecdotal evidence coming from CIA, reinforced by the claims of the people who ran the SERE [Survival Evasion Resistance Escape] Program about their ability to reverse-engineer the program and what that could yield. This was a program that was familiar to people in the CIA and some parts of the military because they had gone through it in their training. Then you talked to the odd Israeli or Jordanian or Egyptian who’d say, “Yes, stuff like that can work,” not having carefully analyzed current Israeli practice, much less recent British or German practice.

I have never been able to find evidence of a really rigorous analytic process, even in that period. Maybe it occurred, but I don’t think so, partly because I think that there wasn’t a real process of any kind. It was done really in the CIA channel, and the only interagency process was a lawyers’ committee process where the lawyers had a relatively narrow remit for what they could review.

This is one of the process deficiencies. I think when we worked on this in ’05 it was the first time this program had been thoroughly debated by a policy interagency committee.

As opposed to—

As opposed to a lawyers’ committee where they ask lawyers is it legal or not, which had become kind of a moral—in our current civic life, where lawyers assume the role of secular priests, it was tantamount to getting the question answered: Is this OK? The lawyers say it’s OK, then you have received absolution.

Who participated in this?

—and the lawyers, of course, are giving you this answer in a situation where the client wants a certain answer and the law itself is incredibly murky.

Who participated in these meetings on behalf of the CIA?

It varied. By 2005, of course, Goss, if the director got involved, or [John] Negroponte if the DNI got involved. Negroponte was somewhat uneasy about all this but tended to defer. His deputy on this, David Shedd, was more sympathetic to my camp, but laid low.

Incidentally, Juan Zarate, the counterterrorism guy on the NSC staff, and that whole counterterrorism office of the NSC staff, didn’t even play in these issues.

Bob Grenier—that name is a public name, so I’m not disclosing anything—was involved in some of these meetings but clearly had his hands tied in what he could say and answered questions very narrowly. The CIA’s lawyers would regularly be at these meetings. Above all, John Rizzo, who now speaks a lot on the circuit.

OK.

That gives you a little bit of an impression.

What percentage of your time would you say, in 2005-06, was spent on this issue? The Guantanamo cluster, is that 20 percent of your time?

No. It’s probably no more than 10 percent. Of course it spikes at various times, depending on opportunities and occasions for a major move, but no more than 10 percent. Probably the issue I spent the most time on was Iraq. I spent a lot of time on India in the early months, and then much less later after the initiative was launched. Nick Burns ran that for Condi and I only got involved with it when it looked like it might go off the tracks.

And you spent your time on this detainee issue because you felt it was a really important issue?

Yes, and because Condi felt it was important. I was operating at the very edge of what Condi wanted. At one point Condi had gotten some complaints about how—

You were bullying people again? [laughter]

Or I was being obstreperous and was not displaying enough understanding for all the backstory. Clearly some people had come to her concerned about this, and these people were still her friends, too. Harriet Miers and others. Harriet was, and I presume still is, a very good friend of Condi. To Condi’s credit, although a couple of times she was a little concerned about me, at no point did she really try to yank in the reins, which she could certainly have done, or just pulled me off of it.

I think Bellinger was a determined advocate and a little more mild-mannered, but she gave me a lot of rope to work with on this.

Could you have prevailed on this issue if it had remained strictly within the administration, or was it the action by Congress and the Court that—

I think that we would not have prevailed internally alone. On the other hand, I think that it would have been hard to do this all externally. I think it was the combination that was valuable. Having these issues being pushed on the inside, presenting increasingly plausible alternatives in front of the President on the inside, combined with the outside pressure. The McCain amendment we were already moving for. We made a big move on this in December of ’05 that essentially adopted the cruel, inhumane standard for all of our treatment everywhere, which we did as a policy matter before the McCain amendment was passed and then announced that in advance of Condi’s trip to Europe in December of ’05.

In anticipation of that amendment?

Partly, but it was actually a battle we had been fighting all year long.

That’s the trench warfare that had been going on for over a year.

That’s one aspect of the trench warfare. Then with both that decision and the McCain amendment having passed, it turned out that Justice was forced into the legal opinion that said, “The CIA program and everything it does, does not constitute cruel, inhuman, or degrading treatment.” That forced them to the final fallback position, which they had tried to avoid in the fourth of their four legal opinions in May of 2005. That was the point at which they had fallen back to a quite obnoxious legal stance that I thought was indefensible. No one in the government would raise their hand and say, “This is a terrible legal argument.”

That’s when, in February of ’06, I wrote down that this was a terrible legal argument. Not saying that my interpretation of the law was entitled to deference, but frankly, there needed to be a debate. I needed to do what I could to provoke a debate on that. Four months later the Supreme Court simply settled the issue.

Typically, because they had overreached, they lost even more. They’re obliged to accept Common Article 3 not only as a matter of policy but as matter of law. Of course one of the arguments that Mel introduced in his question earlier, when he asked was this one of their motives, was concern to protect their past decisions, including the legal liability people potentially faced for their past decisions. Yes, of course, that was a motive.

At the point that the Supreme Court made its decision in the middle of ’06, I got the news of this decision while I was with Condi in Moscow. I printed the decision out and read it. Reporters were looking over at me because they saw me at the dinner reading this Supreme Court decision page by page. Then I figured out—I knew how to read Supreme Court decisions and I knew what this meant. I went to Condi on the plane and said, “There is some felony exposure here, for a lot of people potentially, even including you. You and others need to—” because it was my job to help her do the right thing. They needed to be aware of the significance of this exposure.

To explain this to you and to others who don’t follow this very carefully, the law has changed on this point right now. If you look up the current lawbooks, it won’t show you this. It’s a felony punishable by up to life imprisonment under federal criminal law to commit a war crime. What is a war crime? A-B-C, and one of those things that is a war crime is a violation of Common Article 3 of the four Geneva Conventions. At the point that Common Article 3 is applied to al-Qaeda, if you violate Common Article 3, you have committed a felony punishable by up to life imprisonment. Common Article 3 standard is the cruel, inhuman, and degrading treatment standard, which we had just finished arguing was the standard the CIA program had never met. All right.

I’m surprised they didn’t run you off, Philip.

Frankly, from another perspective, you need people to help get you out of the hole.

Oh, I don’t disagree with that. But you don’t want people to tell you you’re in a hole.

No. We then did a lot to prepare the President’s move in September of ’06 that began climbing out of all this, that went to Congress for new legislation, which also, by the way, amended this particular war crimes law and other things. It didn’t end the problem, but it fundamentally moved this into a different place. It changed the whole character of the public debate. Once you came out into the open with the program, once you brought these people out of the black sites and back into Guantanamo, then the democracy could actually work and you could have a genuine and more healthy public debate over how to strike the various balances and trade-offs that you need to strike on a program of this kind.

Ever since then, I think that debate has oscillated back and forth but has been moving in a healthier direction that allows us to arrive at policies that better reflect the values of the democracy.

Philip, I want to go back and ask you again about the circumstances that bring the first set of Justice legal opinions to your desk, what were the circumstances behind that again, and your own reaction?

Condi gave me the responsibility and I went to Bellinger and said, “I need to see these opinions.” He had the only copy in the building.

This was of the 2001.

No. These are actually the latest 2005 opinions that had just been adopted with that, without my knowing about it until after it had already happened. I got to sit down and read the opinions. That was my way of getting briefed on all the interrogation techniques in the CIA program, described in extremely dry clinical language. You have to take a little time to understand what that would really mean for human beings and how this would really work.

OK.

It was actually very difficult for anyone to get information about this program and how it worked. Eventually I tried to inquire a little bit into meetings at CIA as to which of these techniques do you really think is valuable to figure out some way of trying to draw the lines.

One important way the Iraq war helped me with all of this is at the same time as I was working on these issues I was going to Iraq and I knew what we were doing against al-Qaeda in Iraq. I knew what we were doing in our intelligence work in Iraq and what we were doing to interrogate al-Qaeda in Iraq.

Mel may have heard this before, but one of the fascinating aspects of this issue is we conducted, without meaning to do it, a natural double-blind experiment in which we had one interrogation program for al-Qaeda worldwide operating under one set of rules and one institution, and then we had another interrogation program for al-Qaeda in Iraq—who, by the way, are probably more hardened and dangerous and seasoned killers than the other guys we were picking up—run under different rules with a different institutional setup. In real time you could compare the results you were getting in these two programs and you could talk to the intelligence people running the two programs.

By the middle of ’05 we already knew that what we were doing with al-Qaeda in Iraq, which had been cleaned up quite a bit after Abu Ghraib exposés in ’04, what we were doing against al-Qaeda in Iraq was highly successful and was being run under manageable international rules. Indeed, I was even present at an NSC meeting where the JSOC chief, the Joint Special Operations Command chief, who was running our program in Iraq out of Balad, told President Bush in my presence that the interrogation program against al-Qaeda in Iraq was entirely successful and got them the information they needed very quickly. Incidentally, the commander who gave President Bush that information was one Stan McChrystal.

And they were following the Army’s code of conduct?

Yes, they were in effect following the Army Field Manual. There is an Atlantic magazine article and some other things that have tried to depict the way—this was in a magazine article I think [Mark] Bowden did on how we got [Abu Musab] al-Zarqawi, and he talks some about the interrogation program and the way it worked. It used FBI guys who are some of the most knowledgeable interrogators in U.S. government institutions. But CIA, military people—see, the FBI, after very early on, refused to have anything to do with the CIA program because they didn’t want to be within a mile of it. I think Bob Mueller partly was worried about the legal issues, too, quite sensibly.

That was a very revealing illustration of the argument we were making at the time about whether this was uniquely necessary. We could appeal to various historical illustrations, including in American history, but here we were running a natural experiment right at that moment on al-Qaeda. Interestingly, it took a while for this to filter in and register.

We’ll want to get into the Iraq piece, which is a big one, but you mentioned earlier that at the early phases the India piece had been an important part of your portfolio. You want to give us a brief explanation of what you were doing there?

Yes. Let me give you a short capsule of how the India civil nuclear deal came to be, which I think will historically look like one of the most important foreign policy initiatives of the second term of the Bush administration and maybe one of the ones with more lasting importance. The issue at the beginning, in early 2005, actually was a Pakistani issue. The President had made a decision in 2004 that he was going to go ahead and finally deliver the F-16s to Pakistan that Pakistan bought all those years before, for reasons in Pakistani policy. Then they had a problem. If we’re going to go ahead with the F-16s for Pakistan, which is a long-simmering issue, how are we going to manage this with India?

That issue was bouncing around the State Department when Condi came into office. It was on Condi’s desk and also on Bob Zoellick’s desk. Zoellick roped me into this first. Zoellick had mixed views about the person who then ran our South Asian bureau, a woman named Christina Rocca. He asked me if I would take an independent look at how to manage this Pakistani F-16 problem. Both Zoellick and I had the instinct that the way to manage this problem, rather than just figure out some reassurances for India, was to enlarge the whole way in which we were thinking about this issue beyond just Pakistani F-16s.

That opened up a door into how should we think about our strategic relationship with India and what could we do with India that made it clear that this hyphenated relationship, where everything had to be balanced between India and Pakistan, was no longer a useful way of even thinking about our relationship with India.

I did a lot of homework on where things stood with India. To make a long story short, ever since the late Clinton administration, but certainly since the beginning of the Bush administration, everybody had wanted to greatly strengthen our relations with India. I even put some language in the National Security Strategy, which was warmly endorsed by Bush and Rice, about the importance of the relationship with India for the future.

But the relationship with India had a structural ambivalence right at its heart connected to these nonproliferation issues. The nuclear nonproliferation issues sound narrow on the surface, but in fact the way the nuclear nonproliferation problem interacts with many other nonproliferation and arms transfer strictures and technology transfer strictures in U.S. law and regulation is such that with India being outside the NPT [Nuclear Proliferation Treaty] world, it complicated your ability to move forward with India on all sorts of things. India was not an NPT outlaw. India had not violated the treaty; it had just never signed the treaty. And the reasons it hadn’t signed the treaty were—here a historian’s perspective helps—they were peculiar and somewhat anomalous.

The irony was that China and India both looked at getting nuclear weapons around the same time. The Chinese just pursued it with more determined zeal and vigor than the Indians did. The Indians, characteristically, dithered about it for years and were unsure of whether they wanted to go forward, so the Chinese program got finished under Mao [Zedong] before the NPT curtain rang down. The Indian program was already well advanced, but didn’t get done before the NPT rang down, but the Indians couldn’t very well say, “Wait, wait, stop, we’re still building our nuclear weapon. You can’t adopt this treaty yet.”

But the Indians neither are in the position to just shut down the whole program. So they go ahead and test their device in 1974. We’re in the anomalous position now decades later of where we can sell nuclear reactors to China but can’t sell a nuclear reactor to India because India is caught now.

India hasn’t been a notorious proliferator. China has helped smuggle nuclear technology that allowed Pakistan to build its nuclear weapons. India wasn’t doing stuff like that. But nonetheless we have a much better nuclear trade relationship with China than we do with India. This is a very strange situation. And the structural ambivalence this puts into the relationship handicaps your ability to move forward on many fronts. The Bush administration in ’04, led by Steve Hadley at the White House as Deputy National Security Advisor and Ken Juster at the Commerce Department, had developed an initiative called Next Steps in Strategic Partnership that was supposed to try to advance technology trade with India and other things.

The bottom line about the Next Steps in Strategic Partnership, it had gotten completely tangled up in this thicket of laws and regulations and it was a Gordian knot that they could not get through. My conclusion in early ’05 was that the only way to get through this Gordian knot was to cut it. There was simply no way to really advance relations with India on a broad front unless you somehow came to grip with the structural ambivalence and somehow dealt with the India nonproliferation anomaly. The only way to do that, I thought, was to essentially grandfather India into the nonproliferation regime.

There are disadvantages and advantages to doing this. The disadvantage, of course, is that it seems to set a bad precedent. In fact, I don’t think North Koreans and Iranians make their decisions about nuclear weapons for reasons like these when you actually get into the detail of it. The upside was that India, psychologically as well as politically, is in a kind of halfway house, and you need to bring them out of the halfway house and into the big house where the master lives. See, psychologically, they feel like they are in a halfway house. It encourages a sullen insularity in the way they view the world that is not where India needs to be in the 21st century to play the kind of role we want it to play in south and central Asia, in my judgment.

I don’t know how that will work, but I just know that it needs to be more constructive, given my bets about the way I hope that continent will evolve. The argument then was that if you bring them into the club they can become a constructive partner with nonproliferation, and also, by the way, will be less reliant on fossil fuels, which will then help you a lot in all your energy and environmental arguments.

I mention a little bit of this because it’s common in the literature to find an argument that this was all done as a balance of power move against China, which it was not. Kind of an ironic argument, because the same people who did the responsible stakeholder move that ended the containment ambivalence about China, Bob Zoellick and me, were some of the same gang involved in the India move earlier in 2005.

I then develop this argument—I cut the Gordian knot as part of this broader conception of the role India needs to play in the 21st century and what America can do in connection to this. This was an occasion where reflecting on history was very valuable. Condi and I discussed this and some of this is in writing at the time. We thought hard about the role the United States had played in the development of both western Europe and Japan in the 1940s and 1950s, and the broad concept was that—Mel will remember how some Americans like to play with phrases like “a third force” in the 1950s.

A broader conceptual notion was that if you built places on the periphery of these continents that are fundamentally healthy, the basic health of those places may eventually have a positive, rippling effect outward. So that if Japan and South Korea and the East Asian periphery is fundamentally healthy, eventually that might, in some way you can’t foresee, have positive effects on the Asian land mass. If western Europe is fundamentally democratic and prosperous and healthy, that might, again in ways you might not be able to exactly foresee, ripple inward deeper into Europe.

That’s the moment India is at historically in the broader outlook for the future of south and central Asia, in which India is such a critical variable. Again in ways that you can’t precisely foresee, if India develops in a really healthy way this could reverberate into Asia in a fundamentally positive and fruitful direction, even though we cannot foresee how that will happen. The corollary issue that falls out of that is, if that’s right, what role can the United States play now to facilitate that at a point when India actually still needs the help? Because it actually hasn’t joined the ranks of the great powers and is really wondering what it should become.

That’s the broader strategic context for this move, which I talked about with Condi. Condi began talking about this in a very preliminary way on her first trip to India. She was quite seized with this on the way back. And, by the way, she totally understood this argument.

Could I interject and ask a quick question? The strategic thinking that you just outlined seems to remind me of the strategic reasoning that neocons used in the late ’90s, 2000, 2001 to promote quick democracy in Iraq. That, too, would resonate out and impact and reshape an entire region. Was this a shared assumption? Did you realize when you talked this way that other people talked the same way about Iraq?

I didn’t much. I had never been quite as sanguine about Iraq, perhaps. I just hadn’t thought about it in that way.

But you understand the point I’m making here?

Right. But India was a democracy. We didn’t make India a democracy. India is an organic democracy. So what you’re doing is you’re seizing on an organic condition and you’re trying to figure out how that condition can grow and develop and the role you, as an outsider, can play. What is the role of the outsider in influencing the direction of this development at the margin?

I think the fundamental gut objection a lot of people have to the Iraq analogy is they sense in some profound way that it feels inorganic to the place. That’s one reason why the western European and Japanese episodes are so interesting. Of those episodes, the Japanese and the South Korean ones are the most intriguing, since you can make a pretty good argument there are a lot of things about democratic institutions that are organic in Japan’s and South Korea’s political development, but a lot that are not, and in some ways, their whole modern history had been tipping back and forth between these different models of political development with deep tension and uncertainty about where they should go. We were trying to lean in that way.

Iraq would be a much more forbidding case to analyze even in those terms. But the direct answer to your question is the analogy to reasoning about Iraq in this way had not then occurred to me and didn’t occur to me even in the earlier period. In fact, if you thought about how to change the political development models of the Arab or Muslim world, Iraq would be one of the places you would be least likely to pick as the base from which to create—

Maybe that would be true for you, but not for the people—

Yes, for some of those advocates. I can’t answer for what they thought.

Iraq was a secular place.

Yes, but by saying that I have some healthy regard—

We don’t need to get slowed by this.

But we want to go ahead and wrap up India because we’ve got the Iraq piece that we haven’t gotten to yet.

But India is really important. I appreciate what you’re saying.

It is. On the trip to India, which may have been in March of ’05 or early April, I was in the cabin in the plane with Condi, who fully got and internalized the kind of ideas I’m talking about, and said, “Philip, I want you to write a memo from me to the President and here’s what I want that memo to say.” I just took notes and she basically made the argument. In a way this is not a case where I’d come up with the idea and I had socialized it with Zoellick and with her. Zoellick was quite taken with it. She was very interested in it. I have a feeling that this is the kind of argument that they got because we were all really thinking the same way.

There are times when, on a good day, Condi and I can complete each other’s sentences. Sort of the way we can sometimes sense some things. The same is true even for Zoellick, though Condi and Zoellick is a much more complex relationship, I’m afraid, than my relationship is with either of them separately.

This was a case where she dictated the memo to me. You’ll find the memo in the files because this became—and she didn’t do this very often—this is a full-bore policy memo to the President from Secretary Rice. This is one of the more important state papers of her period as Secretary of State. There is another memo like this on China that’s prepared toward the end of 2005, but these were rare.

This memo really lays out the whole case. Not only that, but also reconfiguring the South Asia bureau into a bureau for south and central Asia. The whole grand strategy analysis is laid out for the President by her in that memo, which I drafted up and then circulated to a few people on the seventh floor, because this was being very closely held, then got some input to improve the memo. Condi got it where she wanted it, signed it out to the President, and then discussed it with the President, who I believe—I wasn’t there for it—enthusiastically endorsed it.

Then the issue, as it was getting developed for the launch, got fought interagency. On this issue Department of Defense were our allies and our opponents were the nonproliferation folks, especially in the NSC staff. Bob Joseph, who is now at State but worked with his friends in the NSC staff, and Steve Hadley representing those views. They came within an ace of effectively derailing it. The way they were going to derail it, and I think this would have been in May or June, is by studying it.

The announcement in July would have been, “Here’s an idea we plan to consider and study and then we’ll return to this issue in December and decide whether to do it.” I believe if you had done it that way it would have been killed. Partly because of what would have happened with the Hill and so on.

So the choreography of how the initiative was announced was the key to the fate of the initiative. We were able to win the battle on this choreography and get it announced the right way, then Condi and Nick actually negotiated the details, which was a cliffhanger story on the very morning of the announcement itself. The night before, Nick Burns went to sleep thinking it’s not going to happen the next day, and then Condi got up in the morning at 5:30 and decided to take one more shot at it and she pulled it off. Which, incidentally, foreshadows the way the diplomacy on this thing would work for the next three years. It’s like a Perils of Pauline serial, just one cliffhanger after another. The thing was pronounced dead multiple times.

It doesn’t sound like your opponents were so formidable. You just said the opponents were mainly the nonproliferation folks in the NSC. They must have had other allies.

In State, yes. First there were the problems in our Congress, which were very difficult. Then there were problems that the Indian government had in their parliament. Ironically, the opposition in their parliament thought this was much too pro-American.

Oh yes.

This almost brought down the Indian government. The Indian government had to decide—this is a huge issue in India, much larger than it was in our politics. It was a defining issue for them in their politics. Prime Minister [Manmohan] Singh ended up deciding to stake his whole government on it and he won, but it was a close-run thing.

There were further negotiations on the implementation, which again repeatedly came down to the wire partly because of Indian politics. We decided playing this on the road that we would seek the approval of the Nuclear Suppliers Group for this with India. We would actually bring all of the nuclear suppliers—the international nonproliferation community—and ask them to formally bless what we were going to do. That also turned out to be a hard battle. The Chinese were actually fighting it behind the scenes, using other countries, because the Chinese didn’t want to be seen in public opposing India on this.

But in 2008 that battle was also won, a really nice and complicated piece of diplomatic work in which Bill Burns plays an important role along with Evan Feigenbaum. But the lion’s share of the credit for following this through goes number one to Rice, with President Bush and Steve Hadley playing regular important roles after that.

Follow-up, Mel?

No.

Iraq? You make a fact-finding visit in February of ’05 before you move into State, if my timeline is correct.

Yes, that’s right.

Tell us how you get drafted into making this trip and what happens.

Well, out of the blue Condi called me in and said, “You need to come to a meeting I’m holding tomorrow,” or the day after tomorrow. She was convening a study group to go out to Iraq and work on what the State Department should be doing there. The Pentagon had just done a study group on Iraq at the end of ’04 called the Luck Group, named for General [Gary] Luck. This was going to be the State Department equivalent to evaluate our programs and the situation and what we should be doing. The nominal head of this was a veteran career diplomat who had served in Iraq named [Richard] Jones. Ambassador Jones.

Not as propitious as Luck.

The Jones Group. Ray Odierno was on this team, who would later become much more well-known but who had also been in the Luck Group, then a senior AID [Agency for International Development] official and one or two others and me. We went out to Iraq. This is my first visit to Iraq, in February 2005. I would end up making probably a total of 11 or 12 trips to Iraq over the next two years, but this was the first, and spent several days there looking around. I had some good deliberations and conferences and then I ended up drafting the report for Condi of this group’s work, and I was still learning.

The report gets most frequently quoted now in books like [Bob] Woodward’s for having the line in it that “Iraq is a failed state.” At the time I wrote that I didn’t think I was writing anything remarkable or controversial. Condi in one of the books actually says, “Yes, Philip sometimes is given to hyperbole.” Well, maybe so, but using that language in this memo was not one of those cases. It was a failed state undergoing a civil war. It’s an incredibly difficult situation, and what we were doing on the governance side wasn’t very good. We needed to do a lot more, so the report had a number of recommendations.

Then she asked me to continue to be involved as sort of an unofficial deputy on Iraq. Occasionally I would go to the Deputies Committee meetings on Iraq as her deputy, usually not. Usually the line manager for Iraq would be one, or another ambassador—Jim Jeffrey or David Satterfield would be the person who would go. My working relationships with both those men was good, especially with Jeffrey, who was then the DCM [Deputy Chief of Mission] in Baghdad but would later come back to run the Iraq portfolio at the State Department before going back overseas again. He would eventually replace J.D. Crouch as the Deputy National Security Advisor during the very end of the Bush administration after I had already left the government.

Jeffrey and I wrote an important set of ideas about Iraq in the spring of ’06 that foreshadowed a little bit the surge ideas again, but that’s later on in the story. So I got involved in a variety of issues that—

Philip, did you have anything in your past that prepared you for going into a war zone in a failed state? Any standard of judgment or points of comparison and contrast that help you interpret properly what you’re seeing?

No, not really, I suppose. I had done quite a lot of work on civil conflicts of one kind or another as a historian or in policy work, whether it was the Arab revolt in Palestine during the 1930s or I had worked on Northern Ireland, doing field work in Northern Ireland during the 1990s and wrote a couple of case studies for Harvard on that. I felt by that time that I had a pretty fair understanding of military and intelligence issues and knew counterinsurgency writings and counterinsurgency doctrine pretty well.

But as an eyewitness, the Irish case would be—

No, when I was on the 9/11 Commission I had gone out and done essentially field diagnostic work, including in Afghanistan back in the fall of 2003. I looked back later on how well did I read that situation, what should I have seen that I didn’t see, that kind of thing, to try to improve my own judgment. I concluded that I had done OK, that our conclusions were reasonably sound, not great. I know a lot more about Afghanistan now than I knew then, and I think there are several things about Afghanistan I never understood until fairly recently, including when I was working on Afghanistan when I was the counselor of the department. I wish I had understood Afghanistan as well then as I think I do now.

What did you not understand?

It’s a useful aside. Back when I worked on Afghanistan in ’05 and ’06, and even when I looked at it in ’03, I tended to look at the problem in the conventional way that is very common now, which is this is basically a narrative about what we are doing in Afghanistan and whether what we are doing is working. A better, and I think more useful, starting point for that analysis is to reconstruct the narrative in entirely Afghan terms and then relate us to that narrative. You construct a narrative: Here is what Afghan history is in this period and here is what is going on in the Afghan-centered narrative of the different contenders inside Afghanistan and the character of Afghan governance national and subnational. Work out that narrative, then bring the United States into that story as an outside actor.

Usually the starting place is the one I initially described. You conventionally slide into appraising your policy in that place, and semiconsciously we’re always the subject in that story and the Afghans are the object. Without meaning to be patronizing, it’s just the natural paradigm through which you approach the study of your own country’s policy.

Reconstructing the Afghan narrative, by the way, even if you try to do it, turns out to be somewhat difficult to do unless you can find really good sources of information or people who can really help clue you in much better to what’s going on. Recently I’ve had the opportunity to do that much better, even for the earlier periods. I feel like I now understand what happened. I worked pretty hard to try to understand what happened in the earlier years in Afghanistan, and I feel like I’m now getting a pretty decent bead on that. Understanding that better now is humbling to me about the inadequacies of how well I understood it even in 2005 and 2006.

On Iraq I think I did somewhat better, but all this is still somewhat humbling.

So you come back from Iraq. You’ve made this visit and are invited into the fold of the Secretary of State’s operation, and Iraq continues to be a major piece of your portfolio from then on?

Right. Really from then on. From then on my whole time in this phase of the government until the beginning of ’07 I’m constantly involved in Iraq issues. I was her deputy in the Iraq strategy review at the end of 2006.

Can you relate to us a narrative, how you see this unfolding during the course of the second term? What your engagement is?

I’ll offer a general narrative. We could go on and on on Iraq for days.

Are you OK to continue?

Yes. I’ve cleared my schedule for the rest of the day, so if you guys are interested I’m happy to go.

We’re fine. I may step out for a second and invite anybody if you want to just so we can keep the clock going.

Just to preview what you’re going to say, probably. I read your Frontline interview. That seemed to be a very detailed account.

Yes. I haven’t reread it myself anytime recently, so I forget what I said.

OK. My basic query to you is, is that a good summation of it? Because it goes through it in detail and in a very luminous way, I would say. I was just curious. You haven’t reread it.

I vaguely remember it and I think that that is still an accurate reflection of my views. Since I gave that interview I’ve done more work and I think I understand what was going on in my place in it a little better now than I did then. But that’s also useful as a historian because the evidence may be more valuable for being somewhat more contemporaneous and less reflective.

In general, in the early days, there were two things I worked on pretty hard right away. One is policing and intelligence issues in Iraq. If you work on counterinsurgency problems of any kind, you will very quickly understand that counterinsurgency problems are driven by intelligence. As Creighton Abrams once put it, but it’s been repeated several times, it’s 90 percent intelligence and 10 percent kinetic action to the extent that you want to get at the enemy. Then there is a lot of stuff about how you protect the population and improve governance and all of that, but the part about getting at the enemy is 90 percent intelligence.

One of the things that struck me right away is that no one seemed to have done any serious work on the quality of our intelligence work in Iraq. By the way, there is a very intimate connection between the quality of your policing and the quality of your intelligence. Just leave that as a statement without elaborating too much on it. I think if you reflect on it, it becomes apparent.

The other thing I worked on a lot from very early on was the creation of what now are called PRTs, Provincial Reconstruction Teams. We actually put this into our February ’05 report to Condi, and even then it was something that Odierno had suggested to me a little bit. We called them back then PSTs, Provincial Support Teams. Jones had liked this idea too, because he was dismayed at the fact that the CPA [Coalition Provisional Authority] regime had actually had a better provincial presence than we did now under the Embassy regime. We’d withdrawn from the provinces even more.

Are you going to come back and talk about the quality of intelligence work?

Yes. When I went back and worked on that, I was there formally as Condi’s counselor, but unofficially I got a hunting license from John Negroponte and his deputy, Mike Hayden—Director of National Intelligence and the Deputy Director of National Intelligence—to look at the Iraqi intelligence issues on their behalf as well.

I got very good access to the intelligence situation in Iraq. This material is not really written up in a way that’s going to show up well in the records, nor are my early findings on policing written up very well. A lot of this material was just briefed orally, including at one point I was asked to come brief what was then and might be still called the President’s Foreign Intelligence Advisory Board or PFIAB. They’ve dropped the foreign part, quite rightly. But I was asked to brief the PFIAB and one of the things I hit really hard was deficiencies in our intelligence work in Iraq.

So tell us what you found about the quality of intelligence work in Iraq. What were the deficiencies? And what were your prescriptions and to whom did you try to “sell” your prescriptions?

My basic findings were that the technical intelligence side was very good and that there was really remarkable interagency cooperation in the field on the technical intelligence side.

What does that mean, technical intelligence?

That includes things like signals intelligence. I can’t elaborate very much more beyond that except to just say that technical intelligence collection, the use of machines to collect information, including signals intelligence, was one of the strengths. Frankly, as one person once put it to me, SIGINT [signals intelligence] saved our butt because our strength in technical intelligence collection and the quality of interagency work in the field down at places like Mosul and so on—there are turf battles in Washington, but in the field it was much better, and that was proving very useful.

The human intelligence collection was not very good. It wasn’t working well at a couple of levels. We didn’t have effective policing and we didn’t have effective police training. This is actually a State Department responsibility. I was in the process of negotiating to turn that over to the military in the cases of Afghanistan after Condi had had to fire the person who ran that bureau in the State Department. That was a big problem.

The State Department—this is just one of many examples—basically did not have the institutional ability to do many things it was being asked to do, and this was one of them. To this day it still doesn’t have this institutional capacity. It takes time to build this capacity up, and we’re even today not making the investments we need to make to do it in a proper way.

One of the core problems in policing is the way we tend to do police training. We tend to get a bunch of people and bring them to a police academy in a country like Jordan, give them a course on how to be a policeman, and then send them back and expect that that will work. The way you need to make that work is you need to field mentor your people on the ground, but that means you actually have to have people on the ground to do that. The State Department had no people to do that, so it contracted people to do all of this work, including the academies. The contractors don’t do stuff in the field, except visit a police station from time to time to take, back then, an inventory of how many guns they had left.

I talked to a lot of these trainers. Some of them are great and some of them are terrible. They are usually like a cop from Memphis who wants to do something exciting or wants to earn a lot of extra money this way or wants to serve his country. Some of these people are heroic and wonderful and role models and some of them are not. In general they were pretty ineffective as field police trainers. The only people who could do that in the field, it turned out, were our soldiers, because they could live out there, and in general that meant you had to have people who had military police specialties. Almost all those people are in the Reserves and we had burned through almost all of those reservists by this time.

Just to clarify, it was a State Department responsibility. The State Department carries out its responsibility through private contractors.

That’s correct, on this and many other things. Almost all development and governance work in the State Department is carried out through private contractors. This is another profound pathology. I’ve been helping the Obama administration on this problem a little bit recently.

And you were the very occasional appointee to actually assess the efficacy of the private contracting work. That’s essentially what you were doing, because you were essentially assessing how the police work was?

In this case, yes, that’s right. And indeed, the people who supervised this for the State Department often were more general contractors than experts on policing. Because then you actually have to have substantive notions as to how you go about policing the community and what the job is, and it’s not like Memphis. It’s different, and it has different challenges.

Well, it’s good if you are judging something to have some notion of what other criteria to use to judge it.

Correct. So this a really profound problem. One problem in the human intelligence side is very weak policing. A second problem was that our soldiers were picking up a lot of people and were out there in a lot of these cities. What I would repeatedly tell people is the most valuable thing you could have in Iraq is a trained, tactical HUMINT [human intelligence] team. That’s a jargon phrase. Or THT [Tactical HUMINT Team]. A tactical HUMINT team ideally consists of an American military intelligence professional working in tandem with an Iraqi who knows the terrain. That is your tactical team so that out there with a company doing a sweep through a street or neighborhood we assign to every company, or every battalion at a minimum, a tactical HUMINT team that fuses in partnership an intelligence professional who knows how to collect, sift, and report information and an Iraqi who knows the people and the language.

If you have a good tactical HUMINT team, that’s a fantastically valuable asset to have at the patrol level. The military intelligence guys, a lot of them were also substantially in the reservists, and again, we had burned through almost all these Reserve units at this point in the Iraq story. This can sometimes be hidden from you because we’ll put people into a military intelligence billet but they might not have military intelligence training. The really good reservists had already been used up and were now in their stateside rotations. We had very few tactical HUMINT teams that were effective.

What happens is you corral a bunch of people who are suspects. You don’t have very good tactical HUMINT capability to figure out what it is you’re learning and what you’re doing. Someone needs to question those suspects. Preferably that’s not done back in Baghdad. Preferably that’s done with the capability that you have out in the field at the battalion or brigade level.

Now a sidenote that is very important to understand, I’ll just make this point here. One of the many misunderstandings about the war in Iraq is to think about it as one war.

To think what?

Think about it as one war. The war in Iraq is very usefully thought of as at least five or six different wars going on inside Iraq. The wars actually have different combatants and actually quite different dynamics. It’s very much a field officer’s war—it’s a war fought at the local level. It’s very much a war fought at the company and battalion level and maybe at the highest at the brigade level or brigade/regimental level. You really have to get into the local dynamics of who you are fighting here. Who are your local partners? Who are the authority figures that you’re empowering and fighting with? And that turns out to be the key for how you’re going to be doing in that particular war. But the war you’re fighting in Mosul is so different from the war you are fighting in Anbar or in Diyala or in Baghdad or the Baghdad belts or Basra or Tikrit. Therefore your intelligence work, the sifting of the information from these captives, the development of really good information, has to be done closer to the field level. You need to have that capability, and we didn’t have it.

What was remarkable, and you’ll see this in the documents if you just look at the documents from, say, up through almost all of ’05, and just even the ways they depict the insurgency to the extent they will even acknowledge they’re facing an insurgency, the quality of the analysis on the insurgency is just so poor and superficial and banal. You’ll see again and again the same sorts of phrases of some combination of Ba’athist dead-enders, Sunni zealots. Basically, there is some amalgamated mix, and we don’t really know how many there are; we don’t really know why they’re fighting. We had a lot of difficulty saying very intelligent things about the enemy.

Incidentally, very recently, the military intelligence commander in Afghanistan, a general named Michael Flynn, wrote a piece for the Center for New American Security, critiquing military intelligence work in Afghanistan. That’s a brilliant piece. What he was writing about Afghanistan in 2009 applied in triplicate in Iraq in 2005 and before. The example of Flynn and a few officers like him shows these are the cadre of people now rising to senior ranks who have realized and internalized the lessons from these very painful experiences I’m describing. He’s not the only one.

So who should have been responsible for making the judgment that you came to make in the middle of 2005? Who was responsible for making such an assessment for the previous 18 to 24 months? That is, you came to the conclusion that the quality of local intelligence was banal; how come you’re the first person to do that?

I don’t know that I’m the first person to make it. My impression—I tried hard to talk to a lot of field-grade officers—is that usually any consultant ends up telling the company what it actually already knows but just doesn’t know it knows it. So what’s happening here is, of course, there are lots of guys in the field who already knew how deficient they were. But command hadn’t really internalized this in a way that stepped up to that recognition.

Partly the command didn’t at this point believe it was doing a counterinsurgency fight, didn’t really want to recognize that it had to write a counterinsurgency strategy, and even had trouble using the word “insurgency.” But there were some deficiencies in higher-level command, and I think that probably includes General [George] Casey in not really understanding enough about the enemy they were facing.

Below the level of General Casey, it really varies from case to case. One of the things you’ll find in Iraq all through ’05 and most of ’06 is that there is no military strategy for Iraq. That sounds like a controversial and provocative statement, but in effect their strategy for Iraq was train Iraqis and then turn it over to them to figure it out. So we didn’t really have to have a strategy for how to win the war or defeat the enemy. Our strategy was to train Iraqis so they would defeat the enemy.

At first I didn’t believe this was true. I thought that there was a strategy and that somehow someone just hadn’t shown me the document and I just hadn’t figured it out yet because I was too stupid. It wasn’t until later in 2005 that it finally began to dawn on me that the emperor really might be that threadbare after all. I just couldn’t believe it.

Philip, you started a statement and then went off a little bit and maybe I didn’t understand the completion of the statement. You said our soldiers were picking up people all the time and then you never completed that. Do you mean to say people were picking up people all the time and not getting appropriate evidence? Soldiers were picking up people and mistreating them and making more enemies?

Maybe some of that, too, but in general a lot of interesting people were passing through our hands and we didn’t have a good system for collecting information from those people and exploiting it. The ideal way that you run this sort of thing is first of all—

May I ask, were private contractors—our soldiers picked up people. Who did the interrogation?

In the military case the military people would do it. The CIA sometimes uses contractors in a variety of ways, but the Army and the Marines tend to do things for themselves, especially this kind of work. Some supply work and other things, but intelligence work they tend to do themselves. And they just didn’t have very good systems in general for systematically collecting adequate human intelligence and sifting and correlating this human intelligence at the field level.

They still had a lot of smart, well-trained people. Not enough linguists, of course, and not enough capable Iraqis. They didn’t have enough at the field level, but at the field level is where you actually get to know the Iraqis best and get to know their neighborhoods the best.

One of the lessons that we learn in Iraq over the next two years is how to get that neighborhood familiarity, how to respect the need for that knowledge, and how to acquire it. One thing that means that you have to do is you have to stay around a while. You can’t just clear in and clear out. You have to clear and hold. If you stay there you can actually turn people into informants. You can run networks of informants. You and your Iraqi counterparts—

Wasn’t one of the key problems going on in 2004 and ’05 and ’06 that there were fewer and fewer and fewer American soldiers actually on the streets?

There were a pretty static number spread very thinly and they would move from place to place as they would run to whatever the new trouble spot was and be concentrated there. Because in general, we were still tending to run more sweeps where we would run through an area and try to clear out the bad guys, but then we would leave.

Most of our soldiers then would be stationed—

—in these larger installations and would do a lot of aggressive patrolling. Already by the middle of ’05 the pattern emerging that was striking to me was that in the absence of any kind of strong central strategic guidance as to what you’re substantively trying to do to win the war, in effect local commanders would improvise their own strategies for better or worse. You then begin to see a lot of local variation in the strategies. For instance, one of things that struck me is by the summer of 2005 the counterinsurgency dissenters are already making their arguments.

I have become one of them, but I am not the only one. There is a “red team” that was working and had been commissioned by the new Ambassador [Zalmay] Khalilzad, though he didn’t do very much with their report. He and Casey mandated them and set them in motion and it was a group of—this is before everyone started using the fashionable acronym COIN [counterinsurgency], but these were COIN guys. I got access to this red team report I think in August of ’05, when it was still only a handful of people who had seen it. Prime Minister [Anthony] Blair saw it long before President Bush did, interestingly.

But the example I wanted to give is, you have, for instance, the famous story of Tal Afar and Colonel [Herbert] McMaster and his very good operations in this particular town, which is near Mosul in northwestern Iraq. This is partly about McMaster improvising with his regiment, but actually McMaster was part of a two-star command led by a major general named David Rodriguez, who actually is the current ground commander in Afghanistan. Rodriguez was actually innovating across his entire command, not just McMaster’s regiment. He was doing very good things in Mosul and he was doing very good things all the way down to the Euphrates valley in western Iraq and into Anbar province, where he was beginning to both clear and hold. He was throwing in his engineers and he was building a combat outpost in the middle of these bad towns and just planting his feet and staying there.

And so the famous McMaster story is just one part of a larger effort. But then you would say, “Well, is Rodriguez doing what is the general strategy all over Iraq?” No. He’s just doing what Rodriguez is doing while Rodriguez happens to be in command of this particular divisional sector. That’s a very typical story. Some sectors do one thing or another.

There is a very nice essay that George Packer, the New Yorker correspondent, wrote I think in late 2005, where he goes out and helps promote McMaster in Tal Afar. He wrote about all the neat things that are going on there. Then he goes over to some place in the Tigris valley, up north of Baghdad, where there is an infantry division based there and everybody is just hanging around on the base. Packer can’t quite figure it out. How come we’re doing one thing over in this part of Iraq and at the same time this other division seems to have a completely different strategy?

That’s because they made up a different strategy. But there was no general sense of guidance or purpose. Partly that’s because there was a central purpose, but the central purpose was to leave. The great irony, the Democrats were beating up on the Bush administration although they don’t realize that Carl Levin and Don Rumsfeld have about the same view as to what should happen in Iraq. Now the two of them detest each other, so they would never acknowledge this, but Rumsfeld, I’ve come to believe, really wanted to get out of Iraq. As soon as Saddam was down, he wanted to get out.

He was vested in the idea that you needed a limited force?

Yes. I don’t know if I made this point last time, but just to say it very briefly, in 90 percent of the literature the way this is written up is, “What an idiot Rumsfeld is. He wants this limited force and you obviously need a much, much larger force in order to properly occupy the country.” So it looks like Rumsfeld is this incredible fool for being so grossly negligent.

Well, maybe he’s negligent in some other ways, but of course he had just the force he wanted to do the mission he thought should be carried out. The mission he thought should be carried out is you knock the guy over and then you leave. You do not occupy the country. So of course he’s not going to put the force in to occupy the country because he doesn’t think we should occupy the country. He’s scared to death that if he acts like he might want to occupy the country, someone will tell him to do it. It turns out that even though he didn’t prepare to occupy the country, the President ended up telling him to do it anyway.

This opens up a different set of pathologies; it’s just not quite the same pathologies that you see in all the literature.

When did people first start talking about committing more troops to Iraq? In this 2005-06 window, or was it not until ’07?

At the top of the government I saw no point in 2005 where there was any groundswell for committing more troops to Iraq in 2005. There was a significant push that began building in the summer of 2005 to get some kind of—here I would insert some strong curse words—some kind of strategy. The President had no strategy. The President kept giving speech after speech as to why Iraq was important. The American people actually agreed that Iraq was somewhat important. What they wanted to hear was how the President planned to succeed there, at least a somewhat plausible story, and the President couldn’t even explain such a story as to how we planned to succeed there. Increasingly, people were beginning to conclude the President didn’t have a story.

And this I think formed part of a break point, which combined with Katrina finally broke the camel’s back in Bush’s remarkably durable popularity. It began sliding in the second half of 2005 and it was a slide from which he never really recovered. Part of this is because people began to conclude, as Woodward later put it, that Bush was in a state of denial, or was somehow clueless, and simply he had no story to tell about what he was going to do to fix Iraq.

This is the context for this big push to have a strategy. There are these COIN arguments going on. One of my allies on this in the State Department was Condi’s communications guru, Jim Wilkinson, who later became Hank Paulson’s chief of staff in Treasury. He went over there in 2006. But Wilkinson was a tough little communications operator and he was just livid and much coarser and more abrasive about it than I could ever be and couldn’t believe that we didn’t have something to say and that we weren’t putting someone in charge to do this.

Condi kept hoping that someone else would devise and explain the strategy. She really didn’t think that she should be taking the lead to do that because it was primarily a military task. Ultimately, she concluded, by around September, October 2005—I came in with another really major report in September, which was in writing. Little pieces of it have been quoted in some of the books.

You made a second trip at least?

Oh yes. I’ve now been out there several times, and this is a major report. I had just gone around the country, myself and a staffer and a couple of soldiers to watch out for us and that’s it. Traveling very light, and I went around the country for more than a week, talking to people all over the place, and came back and wrote a long paper for Condi with about 14 different issues that we raised. Now at this point—

When was this?

This was September of ’05. There is a paper version of this, and then I did a bowdlerized version in PowerPoint, because after Condi got this paper she asked me if I would give this briefing all over the government. She arranged for me to brief right below Rumsfeld in the Pentagon, I think Pete Pace and the Deputy Secretary of Defense. Steve Hadley at some length. John Negroponte. I had slides to use that were even simpler. The point I did make in these slides, in the written portion it’s written very carefully, but the point that’s in the slides and the point I said in person is, “We need to make another huge push in Iraq.” That was the bottom line of the message.

Actually, the message had two big headlines. The first message was that we need a strategy and ultimately this would become “clear, hold and build,” which is a phrase I drafted for Condi’s testimony that she gave in October of ’05, when she finally chose to lean forward and explain the strategy we already had as if we already had it, even though we didn’t.

That aggravated some people who didn’t—

It sure did. It caused quite a little explosion. One headline was first we need a strategy that we needed to explain coherently, and the other headline was we need to make a big new effort. The analogy I would draw is in the second half of ’03, when we had the “Oh my God, we actually need to fix this place” moment. We went to the Congress and got $20 billion for reconstruction money for Iraq. In a frenzy of congressional requests that occurs in this unbelievable period of about two months in the late summer of 2003, and I said we need another really big moment like that.

Because the line that’s quoted in Woodward’s book, I guess, is the line that’s from this paper that’s something like this—I honestly said this—“If I were asked what are the odds that we will succeed in Iraq, I’m not sure I could tell you the answer. I can argue this round and I can argue this flat. I have enough information to spin it any way you want, but the honest answer is, I just don’t know whether we’re going to succeed or not.”

What did “succeed” mean?

I spelled this out. I don’t remember the exact words I used, but in the memo I even defined my terms. I said, “Here’s what success means. Here’s what failure means. And here’s what catastrophic failure means.” I said, with these definitions, the success was defined in relative—it was not, to borrow President Obama’s recent phrase, a Jeffersonian democracy in Iraq.

No, but what was it?

It was a reasonable level of public order in Iraq so that the state was viable, and we could get out leaving behind a viable state. That the insurgency was substantially defeated. That the state was more or less able stand on its own and we could get out. I even had something in there about if we are successful this will mean we have no more than this number of troops there within X years. I don’t remember anymore what the exact numbers were, but I tried to define success by my own standards from earlier writings of mine. I believe when you define objectives, you should define them concretely enough so that you know if you’ve done it. I tried to define this concretely.

The same for failure and the same for catastrophic failure. Then I said that maybe on a good day, if I’m feeling really optimistic, I might say I think the chance that we will succeed is 70 percent if I woke up feeling great. But that’s not nearly good enough. That would mean we have a 30 percent chance of failure, and who knows, some chance of catastrophic failure. For the President of the United States and for our country at war, a 30 percent chance of failure is unacceptable, and that’s the best case on an optimistic day. I would say, “Is there anybody here in this room who would say that 70 percent is too low? Would anyone here—” And there is a sense of we’re doing just enough to get by, and maybe if things go well we’ll win by a field goal.

The way America should plan in its wars, we should plan to win by three touchdowns. So just doing enough to get by is not nearly enough. We have to redouble and make another really big effort in Iraq. The time to make this argument is the fall of ’05, because if you actually need some additional dollop of troops, if you work the timeline of when that stuff needs to start getting put in motion so the troops could actually arrive in the spring of ’06, that clock needs to start in the fall of ’05.

That’s the argument we’re making in this paper. That’s what I briefed around. I actually went out of my way to find Scooter Libby because I thought the Vice President was a potential ally in this. I didn’t think the Vice President wanted to accept a risk that we were going to lose this war.

We almost lost the war in 2004, by the way. We had had a pretty scary couple of weeks there, which we just bailed out with some really hard fighting. Killed a lot of people. Now we were moving to a phase where I thought this could all blow up again.

By the way, the argument that Iraq descending into civil war and sectional violence all started with that attack on that shrine in February of 2006 is not a very good argument. I wrote in this paper, and I remember briefing Hadley, there are Shiite death squads in Ministry of Interior vehicles roaming the streets of Baghdad right now, and it’s notorious in Baghdad that this is going on. In fact, the CIA station chief had been telling me about it and I had heard a lot about this while I was there.

The sectional violence was already flickering and simmering all over the place. What happened is that a lot of these little isolated brush fires converged into a giant conflagration in early 2006, but the signs were already ominous—so the argument was we need to get on top of this. This would all be in conjunction with the election that’s going to be held and the installation of a new government that’s going to occur in the spring of 2006.

That was the precursor to Condi’s testimony laying out a notional new strategy that would—the catchphrase we developed to help people crystallize that message was “clear, hold and build,” but unfortunately it became no more than lines in speeches and White House press releases. We didn’t make it real.

Who opposed it?

It wasn’t actually so much that anyone opposed it. People heard what I had to say. They reacted sympathetically or noncommittally and then nothing happened.

Now when Condi gave her testimony, Rumsfeld took off. Rumsfeld blew up. There was a private argument in which he said he hadn’t cleared her testimony and that turned out to be kind of a dodge. We had sent it to him. He just hadn’t responded.

So he was very angry mostly because he thought she was interloping, or because he disagreed with the substance?

Both. The interloping part he kept private. There is an incredibly revealing press conference that he gave that really is worth reading in detail in which he is asked about the clear, hold, and build stuff and he just launched off. When he launches off his answer is so interesting. He says, “Clear, hold, and build, why do they think this is all about us and our strategy? It’s the Iraqis who should clear. It’s the Iraqis who should hold. It’s the Iraqis who should build.” I think Rumsfeld does this in November of 2005 if you find this.

Notice, though, the mental image that’s underneath that comment. He clearly believes that the substantive strategy for how you’re going to fight in Iraq is being developed or should be developed by Iraqis. Now, this is a very strange statement, and it’s strange at two levels. At one level you could say, “Well, gee, we need to have strategy, too.”

But the second level that’s more revealing and more surreal is, does he actually believe the Iraqis formulate strategy in Iraq? The Iraqis have no general staff at this time. At this time in Iraq we are the general staff of the entire effort. All higher-level strategic thinking in the Iraqi Armed Forces is being done by the MNF [Multinational Force], by the United States.

The Iraqis don’t develop grand strategy at all. They have no capacity to do it. We’re trying to stand up some units to go into the field, and with some success, but the Iraqis don’t formulate the strategy for the country, and when he makes a comment like that it implies that at some intellectual level he does not understand that.

How do you explain that? He’s not a stupid man.

No, he’s not a stupid man. I’ve thought about this, how to explain that. That’s why I find it so interesting and revealing. My best hypothesis, Mel, and I really welcome others who can come up with a better one, is that there is some level of cognitive dissonance going on in which he needs the Iraqis to be making those decisions, not the Americans, because the Americans can’t be making those decisions or else we’d be taking responsibility for fighting and winning this war.

There is some level at which, after the President has flipped on what to do about Iraq in the spring of 2003 and has basically forced Rumsfeld into the mission he didn’t want, there is some level at which I think Rumsfeld never changes his mind about the mission we should be performing and at one level never really quite accepts this as what the mission should be. And of course there is an ambivalence and dysfunction in the leadership of the United States that lets this happen, in which Rumsfeld is consistently—you’ll see this pattern again and again—people will complain that Rumsfeld keeps wanting to withdraw more troops. The pattern occurs time after time.

We try to pull out more. The White House waves its arms and says, “No, no, you can’t pull that stuff out yet. We’re not ready.” Then maybe the White House is able to stop the withdrawal. A month later there is another attempt at withdrawal. This pattern occurs over and over again and meanwhile, the Pentagon keeps saying, “Look at all the new Iraqis we trained. We’re cranking them out like fresh hot rolls and putting them in the field.”

It’s like he keeps wanting to consistently sail to a point on the horizon and the wind is actually pushing him on to this lee shore but he’s still steadily sailing to the point of sail that he originally picked. There’s just some psychological issue going on here. At one point Condi privately told me—I’ll say this for the oral history—that in the summer of 2003, Don put the football down on the 50-yard line and walked off the field, which was a remarkable comment for her to make. By the way, she is not a person who says derogatory things about her colleagues in private conversations. She is a really good soldier and she has a good personal relationship with both Rumsfeld and his wife, I think. But there is something deeply strange about this situation.

But he remains the Secretary of Defense for the next three years.

Indeed at some point, as this got more and more tense, into the winter of ’05 and ’06, because we are getting more and more frantic, this becomes a constant chronic thing. We need a big new push with these political and military and economic elements. There is a whole effort in the spring of ’06 to devise a 24-week plan, mapping out by each week what we’re going to do, that I’m leading with Wilkinson and others. I even briefed this to Josh Bolten, who is the White House Chief of Staff. We get the White House communications people behind us; they even managed to shoehorn Rumsfeld into making a joint trip with Condi to Iraq.

Who set this up, Philip?

Wilkinson and I did a lot to set this up this 24-week plan and then we tried to have a big war council that will pull the strategy up by its roots. That turns into the Camp David meeting of July of ’06, which then turned into a piece of press memory, and was it serious. This is when Jim Jeffrey and I sat down and wrote a draft of military strategy that would have been a counterinsurgency strategy in the spring of ’06 that called for substantial additional troops.

None of these things do we actually spell out, “Here’s how many more troops we need,” because at the State Department you really should not be writing documents that specify the force requirements. What you should be doing and what you’re trying to do is say, “We need to have another big push. Here are the policy requirements from that.” The national policy makers should agree to this and then go to the Pentagon and say, “Now do a troop-to-task analysis for this.” One of the great flaws in all of our strategic documents is I could never find troop-to-task analysis, which is the most fundamental thing in how you generate force requirements. They wouldn’t write up troop-to-task analysis because the atmosphere in their higher analytics had become so politicized and stultifying. It was a very bad situation.

At one point a few minutes ago you said you tried to get the help of Scooter Libby and the Vice President’s office. You never finished that point.

Yes, because I thought, and this was in late—when I was briefing this around, these PowerPoints, in September, I thought that Scooter and the Vice President might become essential allies.

And they rebuffed you? Or they were indifferent?

Typical Scooter took it on board sagely, noncommittally, no real reaction. Inscrutable. And that was that.

Condi told me later in the spring of ’06 that the Vice President intervened a little bit to help and Rumsfeld got a significant course correction in the spring of ’06 or spring and early summer of ’06. It seems relatively modest, and in the larger sense it was relatively modest, but it was something then that was—basically, this is by May of ’06, the President and I think the Vice President and Condi helped with this, in effect told Rumsfeld, “We don’t know if we’re going to keep going down on troops. This is going to be conditions-based and it’s possible we might need to go up. You need to change the way you talk about this so that you offer the possibility that actually troop strength might need to go up, not just down, because we are going to do what we need to do.”

The President begins to slowly, slowly squeeze on this stuff through the summer of 2006 and there is a fairly good account in Woodward’s State of Denial about a meeting I attended in August of ’06, where he gets the happy talk from Khalilzad and Casey and I try to dispel it a little bit. Josh Bolten asks some nice, probing questions. But the most interesting aspect of it is the President is really blunt and forceful, saying to Casey, “You should ask me for what you need. Am I making myself clear? We’re not going to quit this,” and is trying to convey some guidance. What he needs is the military to step up, and by this time I’m becoming convinced that this is hopeless. That is, the military is simply not going to develop a strategy along the lines that we would prefer that they develop.

If the military won’t do it and the Pentagon won’t do it, it just can’t happen. You can’t do this. You have to develop a strategy that adjusts for what the military is willing to do.

And that’s why you’re said subsequently to be an opponent of the surge.

Right. Although that story gets a little bit more complicated, too. By the time you get to the debates in late ’06, here’s the situation. Among your different wars, you’re making some significant headway against al-Qaeda in Iraq because the Special Operations folks are now doing a very good job. It’s actually an extraordinary success story and they continued to do a very good job on through ’07 and ’08.

That’s still McChrystal?

Yes, it is, and it’s a very impressive piece of work that he did.

You’re finding this out by being on the ground there and picking it up?

I had been to Balad and had become acquainted with what they were doing and then stayed in touch with it after that. I think it’s actually one of the great success stories in Iraq. In addition to that, the battle up in Mosul you’ve fought to a pretty good draw, and your battle against al-Qaeda in Iraq is also very important up in Mosul and to some extent in Baghdad. I understood the Anbar issue in ’06 imperfectly because I didn’t spend as much time out there as I did in some other places. I’d been to Fallujah in the fall of ’05 but I hadn’t been back to Anbar, I think, in ’06.

The Anbar situation is already beginning to turn for the better by the middle of ’06, but that’s because of the local dynamic. In effect, the local Iraqis decided to turn against al-Qaeda. They began turning against al-Qaeda in early ’06, and the Iraqis who were leading that were murdered. Then another group of Iraqis come to the fore in the middle of ’06 to lead that movement and they stay alive long enough. One of them was killed in ’07, but they begin leading a movement of local Iraqis. It turns out they’ve got Marines who are ready to be terrific partners for them. Plus a very good Army officer, [Sean] MacFarland, in Ramadi. So they have excellent American partners, but it really is well understood, an Iraqi narrative in which the Americans are able to play a potent supporting role for local Iraqis winning their civil war.

When you begin to understand that dynamic, the whole way in which you interpret COIN strategies changes. This has very large implications for Afghanistan, but I’m not going to go into that today. So that’s already happening positively in Anbar. The problem, though, is that you have raging sectarian violence, an effective civil war in Baghdad, the communities surrounding Baghdad, and in some other places. In effect, the issue there is what do you do about that?

There are a couple of things that are happening in the fall of ’06. The first is a concerted effort, almost a conspiracy, in which Condi plays a very active part, to convince President Bush that his strategy is profoundly broken and that he needs to make a change. Incidentally, to go back, at one point in late ’05 and early ’06, several of us who worked for Condi privately believed that we could never achieve the kind of strategy we wanted in Iraq while Rumsfeld remained the Secretary of Defense. I believe that Condi agreed with that conclusion, though she would not say so. There are some other books that recount various arguments among the inner circle as to whether to get rid of Don, I think even in early ’06. But the bottom line is the President won’t do it, and I think the conclusion that none of these moves will succeed with Rumsfeld as the SecDef [Secretary of Defense] is true.

Your interpretation is the President wouldn’t do it because he fundamentally supported Rumsfeld? Or because he didn’t have the wherewithal to get rid of him?

I think the President was never one to worry about his wherewithal. I think the President wasn’t sure that he should get rid of him. I think the Vice President may still have had some influence on this. But the real answer is you can read the same accounts I can and I don’t know any more than you do. Maybe you’ll learn more as you work on this project.

The first thing that you have to do, while Don is still the SecDef, is the President has to see that this is broken and really face that. Condi discusses this and I’m involved in this too, with the Joint Chiefs of Staff and with Mike Hayden at CIA.

Philip, can I get you to refine that for just a second? Because surely he knows something is broken already, right?

This is really you have to face the fact that not only are things going badly, but your strategy is broken. You need a different way to approach this.

OK. Not only is it broken, but there is a strategy that’s responsible for the brokenness.

Or the absence of a strategy.

OK.

Because we had made these various efforts, beginning with my paper in September of ’05, Condi’s testimony in October of ’05, the big 24-week push that we developed in the spring of ’06 that had sputtered. Jim Jeffrey and I later in the spring of ’06 wrote a military strategy memo in which we laid out a selective counterinsurgency approach that would call for some additional amount of troops. We laid out in some detail how this would work and thought that was very daring. We circulated that around. No effect.

By now we’re in September of ’06. We need to do something differently.

You’ve been pushing on this and pushing on this again. Don’t lose the train of thought, but I want to interject this. You’ve already indicated that this is a President who, when you were comparing him with Clinton, thinks in the business-school model. He identifies the core problem and is looking for the proper solution to it. This seems to me to be an example of a case where his method is not serving him very well.

I agree with you, and it puzzles me, too. I should confess that there are still many aspects of this story that I feel like I don’t understand well enough yet for my satisfaction. I’m still trying to understand better. So I read and try to piece through some things to try to figure this out.

OK. Forgive me.

But here’s what makes this interesting. This is one of those cases where the President has developed a high level of curiosity and is an avid consumer of information about Iraq, I think to the point where he’s had trouble seeing the forest because he was practically counting the trees. He’s not inattentive. Indeed, Meghan O’Sullivan is writing him memos updating what’s going on in Iraq practically every single day. He’s holding an NSC meeting practically every week on Iraq, he’s doing video conferences and so on, but what happens in those meetings is you get caught up in a huge amount of minute detail. There is a pattern in which you are constantly—“Here’s the briefing, Mr. President, on what’s been going on lately.”

There are always some things in the foreground, and you can get caught up in the things, but it’s very hard to see the broader shape of what’s going on and then try to actually steer and move that. You keep seeing the capillaries because your microscope is tuned up so high and it’s very hard to see the broader way that the blood circulation is working.

So there is this effort to—and we had supported the [James] Baker-[Lee] Hamilton inquiry, the Iraq Study Group. I had worked a little bit with Frank Wolf on this going back to the fall of ’05 because we thought that this would be a good bucket of cold water and might be a way to get a new strategy adopted, circumventing our inside bureaucracy. We cooperated with the Baker-Hamilton group and Condi laid out a little bit of our strategic thinking to the Baker-Hamilton group. Mike Hayden gave them a really strong downbeat picture that was, in a way, also designed to get to the President. Hayden, to his credit, was giving people pretty bracing information about Iraq. He was not participating anymore in the happy talk. Good on him. He became Director of the CIA in about the middle of ’06.

The Joint Staff also was convinced that things were broken, and the chairman, Pete Pace, was fretting about what his role was and what he should do, and I think perhaps never really resolved this even to his own satisfaction.

But we had some success with this. There’s another parallel story in which Steve Hadley is writing questions for George Casey that is almost more of an example of questioning the capillaries and trying to label each one of them and defeating the effort to see the big picture, which was a bit of the flaw that all the people over there, especially Meghan, tended to reinforce without, of course, meaning to do anything wrong.

The first achievement, though, in the fall of ’06 was the President was convinced that things were failing and that he needed to make a change. The moves to do that were underway before the election and then he changed out after the election.

By this time the strategy reviews around the government had already begun. By the time the NSC runs its strategy review in late November of ’06, we had already done our own at State. JCS [Joint Chiefs of Staff] had already done a strategy review internally of theirs. The NSC staff was working on one of theirs.

But one of the key insights that the Joint Staff had told me, and I was now able to go visit very privately with the J3 and the J5—the job [Lieutenant] General [John F.] Sattler had, the Marine general. The J3 who was the Joint Staff’s Director of Operations was General Doug Lute, whom I had known since graduate school, off and on, when he was a captain. Basically, Doug said, “We’re out of Schlitz. If you want to do Baghdad, that’s going to take six to eight additional brigades. We just don’t have six to eight additional brigades.”

Therefore the strategy we feel we need to adopt in the strategy review is as follows. One, treat this as a series of local conflicts, basically, the continued unfolding of Iraq’s civil war. The civil war that probably was going to happen after Saddam’s fall anyway, in which Iraq completes the political and social revolution that it’s undergoing as the people who have been in charge of the place for the last few centuries are no longer in charge and don’t like that. That has to play out. The mantra in the strategy for that piece of it was “diversify and decentralize.” You need an approach where you are supporting a lot of different people around the country, and you decentralize and don’t concentrate on the green zone as where you do things in Iraq. Because the green zone doesn’t really matter for a lot of what happens in Iraq.

The second part of it is that you reinforce success with our local partners around the country. But our basic choice with Baghdad and what are called the Baghdad belts, the surrounding areas near Baghdad, which is where nearly a quarter of the country’s population lives, is you either go into Baghdad and try to clear it out or you don’t.

If the Army says you can’t have the extra six to eight brigades, you can’t do Baghdad. Therefore, the approach was, essentially, circle Baghdad. Try to control the exit and entry points and it’s going to burn out. What you try to do is contain the sectarian violence from turning into a full interethnic genocide and manage it and hope it turns. Then you try to contain it.

Meanwhile you’re stabilizing these other different parts of the country. So again, part of the problem of seeing the Iraq war as a monolithic entity with the surge as a monolithic strategy is that it misses a lot of these other different stories in which the surge doesn’t matter very much.

One of the great ironies is we’re writing this strategy memo. There is really no strong, coherent alternative that surfaces from anywhere else. We had a wonderful and important meeting on November 11 with NSC staffers talking about this.

This is ’06?

—November 11, 2006, on a Saturday morning, where Condi and Steve and several of us sat for hours talking about this. There is a very good memorandum from this meeting that was prepared by the NSC staff, I think Brett McGurk. I scrubbed it to make sure it was right. It’s a pretty good record of this meeting. Some of the NSC staffers like McGurk, to his credit, were basically arguing, “Well, why not increase forces, why not try to do more?” I think by this time this was tempting to several people on the NSC staff, including Peter Feaver. J.D. Crouch, I think, went out there to look for himself for the first time.

The argument I made back to that is, without naming Lute and putting him in jeopardy, I said, “We are told—” because the State Department doesn’t do troop-to-task analysis— “that this will require six to eight brigades and that the military just doesn’t have it. So we have to adopt a strategy that doesn’t assume we can get it. Besides, you have a military that doesn’t want to do this strategy and doesn’t want to do it anywhere in the chain of command.”

That’s where that was left. We have a fairly fruitful strategic review, in many respects, that clears the underbrush on a lot of other issues and the basic depiction and diagnosis of the problem, especially by the middle of December. The first slides were still a little too happy talk, then we got some really good slides for the President by about the middle of December of ’06.

Let me interrupt here, because November 11 is just a few days after the midterm elections. Did the results of that election figure into your discussions at all?

No. The results were pretty much expected. Things had been going downhill for a long time. No, the fact that we had a new SecDef raised my hopes. I knew Bob Gates from the old days and thought that that might be a breath of fresh air. By this time, too, I’ve already given my notice that I’m going back to Virginia in the beginning of 2007.

By the way, I’m writing reports from these meetings for Condi almost every time. Those are all on the record somewhere and can capture a contemporaneous sense. Condi’s guidance to me at all these meetings is, “We’re not against the surge in principle, but someone needs to tell us where they’re going to get the troops and what they’re going to do with them. Because if the President goes all-out on one more big venture and no one has a plan as to how it would work, that doesn’t help the President.”

He basically puts all his chips on the table for a Hail Mary pass. This is such an amazing story I need to spend a few minutes on it. I didn’t understand a lot of this story at the time. Meanwhile, there’s the cabal that’s being organized by Jack Keane. [Frederick] Kagan is helping with that, but Keane is, I think, the key figure.

Which Kagan?

Fred Kagan, I think, but I get them mixed up sometimes. But I think this is Jack Keane, and Keane has formed a network that includes [David] Petraeus, who has been busy writing a counterinsurgency manual. I met Petraeus when he was out in Baghdad doing the training command. He was not an apostle of these beliefs back then, but he had become one, having gone back and thought about this stuff. He had joined the COIN dissenters, in other words. So he’s talking to Petraeus, he’s got some kind of contact out to Odierno, who’s going out to take the Corps command in Iraq, and he himself is hoping to coalesce a military alternative that says the military can do more.

They have their sources in the Pentagon. It’s important to understand just how deeply strange this is. The President is basically getting a military option for how to continue in the war that is being midwifed for him by Steve Hadley and is being developed for him entirely outside the established military chain of command. It will develop recommendations. We at the State Department—I know nothing of this. Boy, would I have loved to talk to those people, because of course, if they had sat down with me, “Oh, you’re willing to put all that on the table. Those are interesting ideas. Let me show you some memos that I wrote a few months ago.”

But that conversation never happens. So we’re all in our little bubble and then in a separate compartment that we don’t see but that I think the Vice President is interacting with. The Vice President’s staff at these meetings is presenting ideas about how we ought to completely side with the Shiites and let the Sunni get screwed, which are ideas that we don’t sympathize with very much. But the Vice President, I think, is indulging and working with Keane. They’re developing these really interesting ideas completely outside the military chain of command, then those ideas essentially get dropped into the mix in December of ’06. Even then the surge has different variants, including small, medium, and large surge.

Just to redouble how unusual this is, at the time these ideas are put on the table, Condi hadn’t come in with that and had been really wary of these ideas because they seemed kind of cockamamie. No one would explain exactly what you would do, and no one in the military wanted to do it, which didn’t look good for the President. So we felt we were still writing a strategy for what we knew you could do, understanding that the big difference that this makes is above all what you do in Baghdad per se, not in the whole country, though Baghdad is very important.

The President now is getting a set of ideas in addition to the strategic review, much of which he likes. The diversify and decentralize—a lot of that general diagnosis and prescription he likes and buys off on. That will include a doubling of the PRTs, a redoing of the whole way you do the PRTs, which I helped write up and defended at deputies and principals discussions. But in addition to that he’s getting an idea that is now opposed by his Secretary of Defense (Rumsfeld), by the Chairman of the Joint Chiefs of Staff, by his Combatant Commander in the field, and by the Prime Minister of Iraq, all of whom he will overrule.

One of the most interesting aspects of this is that the Iraqi Prime Minister was opposed to the surge. You have to ask yourself, “Why? Wasn’t he worried about his country falling apart?” Yet, if you think about that a little while, the answers are very revealing. Why would [Nouri al-] Maliki oppose the surge? He would oppose the surge because he didn’t think what was happening was so terrible. At least it may be not as bad as bringing in a lot of additional American troops and further Americanizing the war for a while, which tells you his relative judgment of values and what he thought about the ethnic cleansing that was going on inside of Baghdad.

On the other hand, if Malaki thinks that, here is my choice. My choice is not surge or status quo, my choice is surge or the American presence in Iraq collapses entirely and they withdraw precipitously. He’ll vote for the surge because they vitally depend on us to win the various civil wars that they’re in.

You mentioned the Iraq Study Group earlier, which you said—

Yes, and the Iraq Study Group—

That’s really over this—

Well, it turned out to be a disappointment in this sense. Several of my friends were involved with it, including my former 9/11 Commission staffers and Chris Kojm and so on, but they didn’t have much of a staff working for them. So the report itself is partly terrific, had this wonderfully trenchant stuff in the diagnosis of the situation. But the prescriptive part of the report is terrible. There is some stuff in it having to do with the Kurds that actually had people laughing out loud, it was just so ill-informed.

Oddly, the main military strategy they endorse is what Casey was doing. They endorsed the status quo strategy and didn’t seem to fully understand that that was what they were endorsing. There is a line in there that a couple of people in that group insisted be put in that says, “And we should hold out the idea of increasing,” which has allowed them to say, “Hey, we advocated the surge.” But this is not quite right.

On the plus side, the report redoubled the general sense that your current strategy is a complete failure, and it created additional public momentum for that. That was helpful, and it was a good thing they did. But the prescriptive part of the report was not useful, and in fact was somewhat harmful, unfortunately, partly because they just didn’t have the staff and the horsepower to produce something that was really formidable on all those detailed issues.

Philip, your account seems to track fairly closely with what I remember in Woodward’s account. Is his take on this in that book a reasonably good description of what was going on?

It’s pretty good for what he has. Based on the literature today, in October of 2010, the other thing to look at, it’s very short—at the time Woodward’s book came out, Michael Gordon wrote a long article in the New York Times. It was partly a spoiler piece, with Michael Gordon’s take on what happened. That article is excellent and is pretty dead-on. Gordon is writing a book on this, which, if he will ever finish it, will probably supersede Woodward’s.

OK. But your sense is that Hadley was indeed the linchpin in this account?

I think Hadley, Keane. Hadley to some degree, for which he deserves credit after a long period of difficult performance. But I give him a lot of credit for managing this choreography at the end of ’06. Above all, I give a lot of credit to the President, who really stepped up to the role of Commander in Chief in a way that was not only novel for him but is nearly unprecedented in modern American history. The extent to which he disregarded almost his entire formal chain of command and all of his formal military advice to take this other set of advice—because now to make this a little more interesting yet, when Bush orders the surge, two different people in the White House asked me to do a draft of the speech the President would give announcing the surge. Who was the communications guy back then?

[Daniel] Bartlett?

Yes. Dan Bartlett asked me to do it and Steve Hadley asked me to do it. To this day I don’t know why. But I did. They didn’t use much of it, maybe one or two words. I wanted to give the President a speech where he frankly showed the American people, “I get the problem, we’re in a tough place now, but it’s going to get better and here’s how,” and so on. Then they wrote a speech that was a little more the routine way they wrote speeches at this point in the administration. Not a Mike Gerson product. I thought that was a little bit unfortunate, and it hurt them politically in the early months doing this.

In effect what the President did, to his credit, is he said that among small, medium, and large surge options I’m going to go for the large one. I want the maximum option. But as for Condi’s challenge, what are the troops going to do, that challenge was never adequately answered. In effect, what the President did was he simply chose the maximum option, pushed all the chips he could push on the table, and threw the Hail Mary pass.

But he had changed the Secretary of Defense. He had changed his Combatant Commander. He had changed his Ambassador, another one of our recommendations. We thought Zal [Zalmay Khalilzad] also needed to be replaced, and I liked Ryan Crocker. Ryan Crocker was brought in. You need a completely new team in Baghdad. So he sent his very best receivers flying downfield and threw the Hail Mary.

I think when you get into this you will find that at the beginning of 2007 if you were to ask Petraeus and Odierno, “Do you have your strategy ready?” they would say no. They start writing the strategy on the fly, essentially making up their pass pattern as they are running downfield to catch this. Just to finish the point, and with a very quick summary, what was the major impact of the surge that they did?

It did maybe three big things. One, it provided a general strategic guidance for what we’re trying to do around the whole country, including the PRTs and the diversify and decentralize effort. It gave a strategic concept for all of it instead of every little segment making up its own. It gave a coherent theme to the whole, with Petraeus and Odierno each making major contributions for which there is some rivalry now as to who should get credit for what.

Second, it eased the zero-sum resource strain all around. So instead of choking off Anbar while it’s working, it’s not that they really created the dynamic in Anbar, Anbar was already going the right way, but it made sure that Anbar would have enough stuff to succeed, because you weren’t going to take anything away from it. Instead of having to turn Mosul into an “economy of force” operation so that you could do something in Baghdad, you could do Mosul, too. It gave you enough resources to continue the good momentum where things were already going well.

Three, they doubled down on clearing Baghdad. Now in the double-down clearing Baghdad, I think the reality of this is that a lot of the ethnic cleansing had already occurred, and a lot of the neighborhoods were already burned out in effect.

But the surge did two other things in Baghdad, which were very important, and in some of the surrounding communities. First, it stabilized the current ethnic situation and allowed them to secure and calm the community situation as they found it in 2007, which was very difficult. OK, let’s just hold it here and make this work and then provide security in every one of these communities. The creation of the 77 joint security stations.

Petraeus invited me to come spend some time in Iraq in August of ’08. I got a chance to look around at some of this and it’s very impressive. The first thing they did was basically stabilize the situation and slow and calm it. Then they became partners in the next civil war in Baghdad that, unlike the civil war in Anbar that I described to you, which was Sunni on Sunni, the civil war in Baghdad was Shia on Shia. Now the Shia government had allies. Some Kurdish folks, some Sunni troops, but the primary focus was the Shia government against its Shia enemies who were being backed by Iran.

The Iranian problem is one that runs all the way back to 2005, and I had already spent a lot of time on it. Earlier on there had been some dispute as to how involved Iran was. We made some statements and people said, “Oh no, you’re trying to invent another casus belli against Iran.” All that’s gone now, of course, because the intelligence we have on what Iran has been doing is enormous. Iran had been killing our people in large numbers. They had been masterminding operations. They had been providing tactical training, tactical supervision, providing the key armaments, these explosively formed projectiles. They had been very active against us in Iraq. So we began, finally, as I had been pressuring for a long time, reacting to that.

But that’s a really intense fight in Baghdad. So partly you’re stabilizing the situation. You are still also working the al-Qaeda part of the problem, but a lot of the really heavy fighting, the really intense engagements, are the engagements in this Shia-on-Shia civil war. In that civil war we have a local partner. We have a lot of Shia who are on our side and who are authentically local and organic to the place who are fighting with us.

Philip, would you say that the surge, the additional troops—

Incidentally, the later Basra story in 2008 is the same sort of sequel where we become the enablers for the Shia winning their civil war against other Shia who are backed by Iran in Basra in 2008.

Would you say that the surge, by which I mean the additional troops and the new uses to which they were put, was required to execute the clear, hold, and build strategy that Secretary Rice enunciated originally?

In the conditions of 2006 and 2007, yes. If you decide you’re going to try to clear and hold and build in Baghdad, instead of simply containing Baghdad and accepting a bloodbath in Baghdad, if you want to clear Baghdad, then you need additional brigades of troops to do that without having to shortchange your troop commitment elsewhere in the country. That would be my view.

But a question: Did you need a large additional increment of troops in ’05 and ’06? Had we implemented this strategy a year earlier would you still have needed a number of additional brigades? One of the ways they got the additional brigades is by extending tours of their combat troops across the Army, and my honest answer to that is I’m not sure.

If we’d had a more purposeful strategic direction across the whole country, could we have achieved much better results across the country without a large additional dollop of troops in the conditions of ’05 and ’06? The answer is perhaps. You’d want to do the troop-to-task analysis. At the time, I said and I wrote that our troops were spread too thin and we didn’t have a strategic reserve. We had just committed our last strategic reserve in combat in the fall of ’05.

We also were very short on the number of specialist troops. What I was sure about as a State Department guy looking at this in ’05 was that in ’05 we were spread too thin and we were short in certain kinds of specialist needs. So you would need some additional increment. But whether or not you would have needed a massive additional increment of troops back then, I’m not sure. You’d need a major plus-up of resources, a different way of fighting, and a major plus-up in the civilian commitment, including the PRTs.

What role does Secretary Gates play coming in, I guess mid-November of ’06, in these decisions you’ve been describing?

I think Gates’s role is very important. I only saw a glimpse of this, but then heard a lot more about it later, and actually felt a little wistful sometimes that I couldn’t be in the government in the last two years when it would have been a much better government to work in than the one I had worked in. Because of course it immediately cleared the air hugely when he came in. You could have much more collegial work done between State and Defense. The whole way the interagency process worked was just qualitatively better.

Of course their political capital was almost all gone by then, but the atmosphere was much improved and people felt freer to do straight analysis and to say that things weren’t working. Although Rumsfeld has this image of being this terrible bully, Gates actually has fired a lot more senior people than Rumsfeld ever did. But somehow, coming from Gates, the Armed Forces accepted it in a totally different way.

Gates is very much the master of his brief. He really sweats the details and is on top of things, so you can have a different quality of conversation with him. There is a delicate point, but one that has to be acknowledged, because some of the people in the higher levels of the Bush administration were aware of it, which is some uncertainty about Don Rumsfeld’s physical and mental condition, especially in the later years of his service as Secretary of Defense. In the literature I think you’ll occasionally encounter a brief reference to this. But I saw this myself in some meetings where Rumsfeld would just display symptoms of what you would characterize as old age or short-term memory or attention issues, but they were of the kind that would be jarring. People noticed it and they wondered about it.

[Leffler reenters]

Mel, the point I was just adding, which is a sensitive one, is I was alluding to the fact that especially by these later years there were some concerns about Don Rumsfeld’s physical and mental condition. Because he occasionally, even in meetings sometimes that I would witness, displayed symptoms of serious old age, where he would appear not to have heard things or would repeat things in odd ways, short-term memory issues or other things. People who were around him in these meetings more saw this more, but it was the kind of thing people would maybe talk quietly about or whisper about, but people were afraid—how do you talk about that? But senior officials in the administration could see this.

Philip, let me ask a question about the embassy in Baghdad, sort of a structural question. The ambassador, as I understand it, had a direct reporting line to the White House during a good part of the Bush administration?

No.

That’s not true.

Under Khalilzad, he would want that. But first you start with the [L. Paul] Bremer era. [laughter] Bremer, of course, reported to the Secretary of Defense, who wished to disown him, so Bremer, in fact, reported to the White House and in effect worked for Condi Rice to the extent that Condi Rice brought in Bob Blackwill to help her manage Iraq and Bremer and do some other things. That’s important in the second half of ’03 and ’04.

This is a peculiar and dysfunctional situation. But then the CPA leaves. Ambassador comes in. Negroponte comes in, a good Foreign Service officer, and reports to the Secretary of State. Has on occasion VTCs [video teleconferences] with the President in one or another NSC meetings. There is a sense that you have a direct relationship there. But Negroponte works through the chain of command.

Negroponte is replaced by Khalilzad. He knows he reports to the Secretary of State and he talks to her, but Khalilzad is managing lots of things, this is his way. He’s running back channels to the White House, to the Vice President’s office, to the Secretary of Defense. He’s on these VTCs with the President. He’s frequently on the phone with Steve Hadley and Hadley in turn with him. His relationship and Condi’s trust in him erodes. How much it erodes is hard to say, but I think I can say erodes. He was regarded as gifted for certain things, but eventually even confidence in that lessens. He’s widely disliked by his peers and everybody who is working with him in the State Department, for lots of reasons, maybe some good and maybe some not so good, but some of them are good reasons. Some of this too goes back to Afghanistan. A lot of people knew the people who had worked for him in Afghanistan, and some of the people who worked for him most closely in Afghanistan came to have very negative views about his work there.

So were these State people the groundswell of support for him to run for President of Afghanistan?

They were not. I am not aware—that groundswell is mighty hard to find. [laughter] Even in Afghanistan. Even among Afghans.

Philip, if I might—

So Khalilzad was a difficult management problem.

And Crocker.

Crocker came in just as I left, and from all I can tell was just splendid. Worked right through the chain of command, professional with everybody. We had seen Crocker in Pakistan, which was not an easy post, either. He had done a very good job there.

But there are these direct communications back and forth between the Ambassador and the State Department where the Secretary is not a direct party to those conversations?

Oh no. Khalilzad hated to talk to anyone at the State Department except for the Secretary. In general Zal wouldn’t talk to anybody except Rice. When we needed to go over Iraq policy reviews with him, we would have to schedule teleconferences with his embassy team. This is a story in its own right and it’s one of the things that began to get Condi’s patience with him to wear a little bit.

Gotcha. Do you have any idea whether those video conferences were recorded?

I think they were not recorded. These were in the secure video teleconference system run from the State Department’s operations center, not from the White House Sit Room. These are usually done—The White House Sit Room runs the White House meeting using what people refer to as SVTS, Secure Video Teleconferencing System, which was just beginning to be introduced as a brand-new innovation when I worked at the State Department Operations Center as a watch officer in the 1980s.

OK. But your sense is that there will be no historical record of these conversations.

No. Except some notes. I may have mentioned last time that the documentary record for the administration is in general pretty spotty. People tended not to take reliable or extensive notes of meetings. You’ll find meetings have a summary of conclusions attached and so forth, which are singularly uninformative. There are not good memcons [memoranda of conversations] that were done in meetings in general, and even where you find memcons, depending on the circumstance of how they were prepared, one needs to be careful how much to rely on their verisimilitude. One of the most valuable documentary resources from the Bush administration that historians should look for are the records kept of the Bush-Blair teleconferences. These were frequent and candid and extremely well documented. It just happened that the people who were taking notes for those were good professionals. At least the ones that I saw were done extremely well and are highly useful documents for historians.

But your general sense is that the same kinds of anxieties about close record keeping that prevailed in the Clinton White House and so forth were carried over into the Bush 43 White House?

Yes and no. Actually, the record keeping in the Bush administration, since I looked at both of their records, at least for the early Bush period—in general, records in the Bush administration are much better than records in the Clinton administration, at least for the issues that I followed.

So they are not great for historians but they are a lot better than the Clinton records are. The Clinton people would have meetings for which there were virtually no records at all. You actually had trouble even finding that the meeting had happened. The Bush people tended to be very bureaucratic about this: meetings are scheduled, they are held on time, there is a memo for the meeting, so finding out about the meeting, finding out about its nominal agenda, maybe who was there, was all pretty well papered up. Finding briefing memos that were written to prepare principals for the meetings, those are all there. What will be weak is if you actually want to find out what was said at the meeting. That’s hard.

Right.

The Clinton administration meeting records are much worse than that. For the issues that I know about.

You’ve probably read Peter Feaver’s draft manuscript of the change in policy in 2006 and 2007 regarding the surge.

I did. I sent him a set of written comments about that and had a series of exchanges with him in response to my comments, which seemed constructive. I’ve kept my copy of the exchange.

In the draft of this article that I’ve read that explains Presidential decision making during the surge and portrays the President very favorably, I would say extremely favorably. One of the things that struck me in reading this article was how different Bush appears in 2006-07 in terms of structuring decision making in contrast to 2002–03. I’m wondering if you agree with that observation, and if you do agree with it, how would you account for these changes? The most striking difference being the systematic decision making, setting up of meetings, the careful orchestration of a policy that could be used collectively mandated, even though clearly, decisively orchestrated by the President.

Although a fair part of the process and maybe even some of the real process, which is the Keane-related process, is actually offline, unsystematic. Had it produced results that were catastrophic, we would regard it as an example of terrible process. Because the Keane stuff wasn’t actually mainstreamed into his formal decision-making process until very late in the game, at which point it was difficult to do a lot of analysis with it. One consequence of that is at the time when the President threw his Hail Mary pass the strategy wasn’t particularly well developed, but fortunately he had sent some great receivers downfield and was throwing in a generally good direction. It turned out all right.

So far.

Right, so far. This is a longer discussion that I won’t get into, but I actually don’t believe the quality of policy making should be judged by how it turns out five years later.

I know this may seem remarkable, but in general I don’t think you should judge the quality of thought by judging what happened, if it has to do with variables people didn’t or couldn’t know about. I don’t think that sheds much insight on the quality of their thought. You can learn a lot about the quality of thought by how people appraise the things that they knew about or should reasonably have known about at the time. You can actually have very high-quality thought that produces bad results and very poor-quality thought that produces great results. I’m afraid historians constantly go along with the foolish temptation to say that should determine the awarding of grades, which historians love to do.

The odd thing about historians making this mistake so often is that the habit is profoundly ahistorical. But I agree nonetheless with the premise of your question, which is that whatever its flaws, even in ’06–’07, it’s a way better process than the process of ’02. What accounts for that? I don’t know. My hypothesis would be the banal hypothesis that they probably had learned something from experience and they knew they had had very painful experiences. Neither Condi nor Steve Hadley looked back on Iraq and patted themselves on the back for the past four years and thought, Boy, haven’t we handled this perfectly? That doesn’t mean they’re going to go public and air this dirty linen, but privately these are not stupid people. They know everything did not go all that great.

But my reading of it—and feel free to disagree, in fact, I’m sure you will disagree—is as I read the literature and as I read that article, for a variety of reasons I walk away from it thinking that Steve Hadley did a far better job in 2006-07 than Condi Rice did in 2002-03.

I agree with you. I would have added that to complete my earlier answer, which is that Hadley is a deeply methodical person, right down to his bone marrow. This can sometimes be a strength and sometimes be a weakness. In this case it was a strength.

You could even argue that earlier in 2006 it was a weakness, but by this point in 2006, because there’s a way in which he’s doing this methodical stuff in ’05 and ’06, it’s like the house is burning down and as the flames are moving through the bedroom you’re cataloging which items you wish to take out when you flee. By late 2006 a lot of those qualities are real strengths.

On some of the detention issues—at the point Steve steps in and begins managing that process in a very hands-on way in later 2005-06, the quality also goes up because he has a systematic, orderly mind and he’s very smart. He can do a very good job sometimes in this way and he is more methodical than Condi is. But I don’t acquit either of them for the way it was handled in 2002 and 2003. The problems back then are so serious that I think that there is a pretty profound indictment there. Also, already by 2002, the pattern that is constantly complained of, of all the tough issues are being bumped to principals, which is a frequent chorus, that already tells you the deputies’ process has collapsed. That’s Steve’s process.

Now why has the deputies’ process collapsed? It may not be Steve’s fault that it’s collapsed. It may be that it’s collapsed because basically Steve has no authority to make it work because they won’t pass any authority downward in a way that will empower Steve to make it work. That may be fair, and Steve would be a good person to comment on how that process evolved. But I just simply noticed that the deputies’ process is a key point of failure.

You mean in 2002-03?

Yes. When you see really deep problems in policy development, the problems usually mean that there has been a profound failure at the sub-Cabinet level, because it’s usually very difficult to do policy analysis of first impression at the level of the Cabinet principals. Cabinet principals don’t usually have time to write analytic papers and really take the time to slowly pull out all the different options that need to be put on the table. So if you think that needs to happen, that needs to happen at the sub-Cabinet level.

But the principals could subvert the deputies’ process by refusing either to acknowledge the legitimacy of it or by refusing to pass down to it—

That’s right, and if that was true, that would be an exculpatory argument saying it’s not Steve Hadley’s fault that the deputies’ process failed. The fault goes higher up in emasculating Steve Hadley and his Deputies Committee so that it couldn’t do its job.

As well as Condi Rice’s role in the first term of ’02 and ’03.

The answer is I don’t really know enough, not having participated in that administration to judge why the process failed. I’ve read the same books that others can read, but I have no special insight into that question. I simply observed that it failed, and for some of these things I think it’s very hard to excuse these failures. I think somebody’s got to tell the President, “Mr. President, you have not been adequately served on this issue.”

Why do you think it failed in 2002-03? What’s your definition of failure there?

It depends on the issue. The most high-profile story is the Iraq story. On the Iraq story it’s already failing in the spring of 2002 in not really integrating the military plan into a wider set of policy considerations about the “whether.” They back into that after they have their midsummer crisis that I think I described last time and then go with the UN [United Nations] option. They actually do have what look like proper interagency meetings about the “whether” as they deliberate on the Presidential directive that they adopt in late August of 2002 that really codifies at last the Presidential decision at last to go forward if they don’t succeed in the UN.

It looks like they’re cobbling together a decent policy process, but actually all that’s really happening in a few weeks of August of 2002, and they back into beginning to do a proper policy process after having drifted way too long on too many issues earlier in 2002. So then what happens is they find themselves constantly backed up in the fall of 2002.

Let me give a concrete example of what I mean by that. Let’s suppose you wanted to have the option of occupying Iraq and developing the institutions that would have run a proper occupation of Iraq, if you actually think you needed to invade Iraq. I’m being agnostic about that issue if I am being a good staffer. If you work back from the institutions that you will need to occupy Iraq and then work back to, “If I needed to build a proper institution to do this, when would I have needed to start putting the planning in place to gather the people and so on?” Like standing up at corps headquarters dedicated to that task and training to it and recruiting the civilian people and so on. And being able to work your way very quickly back at least to the spring of 2002. You need at least a year lead time to start getting all that ready just so you have the organization capability to consider that option should the President choose it.

You still haven’t made a decision as to whether, in what circumstances, you’re going to war, but at the point that you want this option, you can’t invent that option in a month. At the point you’re doing very serious war plans, you have to have a process that at least opens the door to the possibility that we may need these institutions. If you’re saying no, you need to understand early on that I don’t have the institutions to occupy Iraq because I made a decision that I didn’t want them. I ruled that option out. But if you say to yourself in the spring of ’02, I might need that. I’m not sure I’ll need that. There is a 20 percent chance I need it, then you need to start doing things to prepare the way for that. That’s just one illustration of it.

Another case could be, No, we’re not going to build the institutions to occupy Iraq because we’re going to make a very clear decision that we’re not going to do that. The President will understand that he has made this decision and he will understand what’s involved in making that decision. Iraq might turn into violent anarchy and we’re not going to be able to arrest that because we’re not staying. He made that decision and assumed those risks.

But that didn’t happen either. Then you have a dysfunctional policy period, especially in April and May of 2003, in which the President does the 180 degree turn on occupying Iraq and on Bremer’s mission. The interagency process surrounding that is almost invisible, even to Bremer, who is at the center of it.

Hadley’s role in the surge is a very rare role for a National Security Advisor. As a policy advocate rather than a—

No, he really is—I think he would regard himself as a facilitator and midwife. He studiously sought not to become an advocate on this issue. Incidentally, I’m not sure that he should not become an advocate. There are different models of how a National Security Advisor plays the role, and I’m not sure that the traffic-cop model is actually always the right one to protect the President. Sandy Berger, for instance, did not just think of himself as a neutral traffic cop, and in some ways I thought he was a good National Security Advisor for that particular President.

Nor Henry Kissinger. Nor [Zbigniew] Brzezinski. Nor [Brent] Scowcroft.

Yes. Scowcroft is one of the funniest examples because on some days Scowcroft will just say he was an honest broker.

On some days.

On some days. All right.

One interpretation of what I’ve heard this afternoon is that the President was like a guy who doesn’t pay his bills and doesn’t pay his bills and doesn’t go to work and realizes he’s in real trouble and goes to Vegas and bets the remaining money he has on a number on the roulette wheel and wins. And somehow that constitutes a happy ending, or an improvement in that person’s decision-making process. That’s not necessarily what I embrace but—

No, there’s a famous story of St. Elizabeth of Hungary, who is bringing food to the oppressed. The bad soldiers intercept her and they can see she’s got this basket and it’s got a big cloth over it. They’ll hang her if she’s carrying food to these people, and the soldiers say, “What have you got there?” and she says, “Just roses.” They whip the cloth off the basket and there is nothing but roses. That’s one reason why she becomes a saint, because the Lord had intervened to protect this good woman. [laughter] Maybe the Lord looks out for the righteous. I’m not sure.

In this case I think Bush deserves more credit than that. My own opinion is that, although the White House bitterly, bitterly resented Woodward’s book title, State of Denial, I think the title is nonetheless reasonably accurate. Not because Bush was incurious or inattentive, but there was a dimension that can be called a state of denial. But once somehow he breaks through that, and I think much too late—I think thousands of people died because that happened much too late—so I don’t have a relaxed view about these timing issues. But once he breaks through on that issue, he did more than simply put a chip on the number. He really sought out some additional ideas, including creative, out-of-the-box ideas. He displayed real resolve in the purpose he was trying to achieve. It wasn’t a wild bet. Serious people told him they could make it work.

In a way, it’s like the receiver who is in the huddle and says, “You throw that to me, somehow I’ll get it,” but more seriously than that. Then he went with the biggest possible option against a very difficult headwind internally and in external politics, then stuck to his guns during some really difficult months in 2007. There were, by the way, some very significant opportunity costs for this. I give Bush a lot of credit for that. I would give him actually a lot of credit for that even if it had turned out badly, because it required serious thought, a deep consideration of where he was that was pretty straight on and honest and a hard decision about what he was going to try to do under very adverse circumstances, and it might not have succeeded. But it did succeed. I think he deserves some credit for that.

That’s why I think your analogy of a Hail Mary pass really understates what it took to do that and the process that went into it.

It may. And it may be a little too callow to use that metaphor for that. I hope that you or someone else will come up with a better way to depict it that’s not quite so casual. I guess I’m trying to clarify that. I respected that at the time and I respected it now. But it’s a useful reminder about the importance of timing in world affairs.

Two other little corollaries. One is, for instance, the opportunity costs of this decision. One very important factor that influenced me personally, and maybe Condi, but influenced the department’s recommendations in late 2006 on doubling down, is before you tell me what you want to do in Iraq, tell me what you want to do about Iran. People who usually analyze these issues take them one at a time, but you actually have to see these issues all working simultaneously. We had just made a big move on Iran in the spring of 2006 to try to make a significant diplomatic opening and see if we could break the logjam there. I thought that was a really good move and I still think so. I think it actually broke the logjam and got the first UN Security Council resolutions done, which people didn’t think we could do. All that to the good.

That was going to tap out sometime in 2007, and then, based on your intelligence estimate, you needed to make a decision as to what you were going to do about Iran. Understand that if you throw everything on Iraq, you’ve now made the Iran decision. In effect, the President’s Iraq decision was also an Iran decision. I didn’t know the President had made an Iran decision. So it’s my job as a strategist to work on a strategy that doesn’t foreclose his Iran decision that I don’t think he’s made yet. I haven’t seen anybody who’s written on this Iraq story realize that this story had very significant implications for his Iran options.

Today, we’re much more sanguine about the Iran situation, since our intelligence estimates have kept pushing the red line further to the right. But, again, based on what people reasonably knew or foresaw at the end of 2006 and 2007, as to what people thought the red lines were, there was a live Iran issue that the President had to consider as to what he was going to do about that.

I was not then and am not now an advocate of launching a huge bombing strike out of the blue on Iran, but I do think there are some other very significant military options that involve very serious risks of escalation for which you need to have combat forces available to support the move.

I’ll give you another example. The optimal timing for the big move on the Israeli-Palestinian peace process was in late ’06, not the end of ’07. The Bush administration made a huge, intense move to try to get that done, in my view, a year too late, and that’s unfortunate. If you examine what they were able to accomplish in that last year, it was not trivial. They’re able to do some really interesting things to get that going in a year and then they ran out of time.

The real window for that was right after the Lebanon war for a lot of different reasons, and we knew it at the time. It’s not hindsight. Condi knew it. We took a run at it. It didn’t work. I even got my name in the papers for five minutes and then just kicked it down the road and then they thought again, Yes, we need to do it. Finally the President decided to commit to it. They held the Annapolis Conference in December of ’07. But December of ’07 is not the same as December of ’06. It was too late. Because of the timing and because Iraq is in flames at the end of ’06, now making a really big, high-profile Israeli-Palestinian move seems harder and your attention is utterly preoccupied with Iraq in flames.

My view and Condi’s view at the time is we could have done them both and they might have reinforced each other, but you can see the way these things interact sometimes, the choices you make about timing and when you decide to act.

I do give President Bush a lot of credit for what he did in ’06 and ’07 and the perseverance on this in ’07, which was a pretty close-run thing and relied a lot on General Petraeus’s political capital to be able to carry it off. Even so, I think that they could have made some of the same moves a year earlier and maybe even if they had done so the Republicans might have kept control of the Congress in ’06. But there are thoughts of counterfactuals. If he had replaced Rumsfeld at the beginning of the second term as so many people suggested he should—frankly, if he had replaced Rumsfeld before that—there were some very serious problems that had shown up with Rumsfeld fairly early on, but the President didn’t see them. The Vice President had been at the very core of the reason Rumsfeld was put in that job in the first place and the Vice President, especially in the first term, was a core figure in the administration.

It’s hard to rerun the counterfactuals too far, but I’m sure that a lot of senior officials in the Bush administration will have a lot of time to run counterfactuals about things that might have been.

Well, anything else?

There are a lot of questions, but it’s five o’clock, so—

It is five o’clock and we don’t want to abuse our privileges with you. You have been enormously cooperative and I’m very appreciative for your giving us two full days for this. I always say at this point that we never exhaust all the topics we can talk about but we do a pretty good job of exhausting the person in the hot seat.

We’ll get you a transcript to look over, and of course we’re all abiding by the ground rules of confidence until this comes out in a few years. If you look over it and you decide that there are some things that we didn’t cover that deserve consideration at some point when Mel’s back down here and doesn’t have to make a special trip, then maybe we can convene everybody for another couple of hours or I can sit down with you, since you’ll have an idea of what you want to talk about, and we can do a supplement. You’ve already done us an enormous service with this. I should say on the record, I don’t think we’d be doing this in the first place if it weren’t for you. We recognize where our patrimony is, and it runs through the Zelikow years at the Miller Center, and so we proudly acknowledge your sponsorship and your vision in seeing this through and we’re pleased to continue to be in business because of it. So thanks, Philip.

Well, all you can do is choose some good people and then they actually might make something out of it, and Russell, you were one of the good ones.

Well, I appreciate that, Philip, and thanks to both of you for your indispensable help. Thanks a lot.