• HIGHLIGHTS
• A strategy lacking military-political synergy
• Insurgency versus civil war in Iraq: why the distinction matters
• The advice he delivered to the Bush administration
• Is the surge Kagan's plan in action?
• Next steps in Iraq if the surge works

Let me ask you: the broad state of play around the time of the Camp David meeting [with President Bush about Iraq war strategy in June 2006] -- what was your sense of where matters stood vis-à-vis Iraq. What was the situation on the ground there from what you could tell?

We were continuing to have the Sunni armed insurgency that was sort of raging, not properly controlled, primarily in Anbar and up the Euphrates and the Tigris and the Diyala River valleys. ... And we had begun to see the shift toward a real sectarian conflict after the Samarra [Golden] Mosque bombing, but it still looked very much as though the Sunni Arab insurgency was a major driver of that sectarian violence. ... And the immediate reaction in Iraq to the Samarra mosque bombing was actually a little heartening, because the Iraqi security forces managed to shut the country down, and you did not have an immediate explosion. But it looked as though we were starting to move in that direction.

Because of the complexity of trying to deal with the Shi'a population … to me it seemed as though the best thing to do, still at that point, was to try to get the Sunni Arab insurgency under control so you could then go back to the Shi'a and say: "OK, listen. We've got that under control. You're not going to be attacked anymore. Put down the weapons, and let's move forward with the political process." So that was my analysis of the situation in Iraq.

I think in general there was a recognition that things were not moving in the right direction, especially after the mosque bombing and violence was picking up. … But there was also a lot of enthusiasm right then, because [Al Qaeda in Iraq leader Abu Musab al-] Zarqawi had just been killed in Baquba, and we had had the elections, and we had the government finally seated. So the mosque bombing aside, it looked as though there was being some progress in some important areas.

One of the things that concerned me was that I knew we did not have large-scale operations going on in Baquba, and I knew that we'd gotten a tip on taking out Zarqawi from some of the locals. I was very concerned about the possibility they would be targeted in revenge and that one of the things that we should have done as a priority was to go in and clear and hold Baquba as a way of rewarding the locals for turning over Zarqawi. Of course that wasn't done, and Baquba subsequently collapsed into an extremely chaotic situation. I don't know what happened to the people who actually informed on Zarqawi, but Baquba is now a very bad place. So that was, in my view, a real missed opportunity.

... What was your assessment of Gen. [George] Casey and the MNF [Multi-National Task] Force at that time?

I think you really have to go back to late 2003 and put yourself in the position that Gen. [John] Abizaid, [CENTCOM commander, 2003-2007], was in when he took over. He had inherited a situation where we hadn't laid the preconditions for political success because the invasion had not been conducted properly, and [Gen. Jay] Garner and [Ambassador L. Paul] Bremer, [head of the Coalition Provisional Authority, 2003-2004], hadn't done things necessarily perfectly, and we didn't have enough forces in the country to control the country adequately.

Gen. Abizaid came to the job very clearly with an attitude that focused heavily on two problems of counterinsurgency; namely, the fact that the presence of foreign forces in any society is an alien element which tends to generate a certain amount of hostility, ... and the fear that if we undertook to do a lot of things for the Iraqis that they needed, ... they would become dependent on us and we would never be able to leave. He focused very heavily on those two problems when he developed his strategy.

Now, the situation then deteriorated very rapidly, I would argue in part because he focused so heavily on those two things that he did not, in fact, focus on maintaining civil order and security, which was actually the obligation under international law of any occupying force, but also a basic principle of counterinsurgency. ... He sort of ignored that, or downplayed it.

So by the time you get to Gen. Casey's arrival in Iraq, the situation had been very bad. Actually, the insurgents had taken control over cities like Fallujah and really seized it and ran it, and you had the Sadrists beginning to rise and becoming very powerful in Najaf and Karbala. So it really looked like a very, very bad situation ... where the insurgents actually control territory. ... When he got there, he found himself in the midst of this incredibly chaotic situation. ...

By the end of 2004 things were looking up in the sense that the insurgents no longer controlled large cities. They were coming back into Tal Afar, but they didn't control Fallujah anymore; they didn't control Najaf; they didn't control Karbala; Sadr City wasn't exploding.

So Gen. Casey could be forgiven for feeling like he'd accomplished something very significant, because he had, coming from a position where things had been. But he remained very heavily focused on the two principles that Gen. Abizaid had been working on; namely, getting the Iraqis to do it for themselves and minimizing the American presence because it was thought to be an irritant. So he still did not take on responsibility for establishing and maintaining security for Iraq, which is one of the things that allowed the insurgency to continue to bubble and continue to grow.

So what some people in the press have said, and others have talked about -- that if [Casey] went with a strategy at all, what he went with or was instructed to have was a kind of exit strategy -- you don't perceive that?

I do perceive that. I think that from the outset our emphasis was on getting out of Iraq as quickly as possible, and that had always been the president's strategy; it had always been Secretary [of Defense, 2001-2006, Donald] Rumsfeld's approach, and it was the approach that Gen. Abizaid and Casey had. ... The objective is to get Iraq under control at a basic level, train up Iraqi security forces, turn over responsibility to the government and leave.

These guys tell me that they came in and they were told, basically, [hold an] election [in] January, hit that target, clean this place up, get some security going, preferably using Iraqi forces. It was all about that election, all about getting it there. ...

I think the notion that was widespread in the administration, and I think also in CENTCOM, was that the political process would end up absorbing the insurgency back into peaceful life and that we would use the political process as the lever of doing this.

It's very ironic to me, the criticism that the Bush administration has taken for preferring the use of force and being very heavy-handed and liking the military instrument. The thing is, the military instrument, as the Bush administration has applied it in Iraq, has constantly tried to solve its problems through political means and has resorted to actual kinetic operations, as they say, only as a last resort everywhere, and I think, frankly, too little. I think they frankly put too much weight on the political process, because I think there was an oversimplification of the problem and a belief that it's true to say that any insurgency ultimately has a political solution. ...

It's also true that in the modern world, the way we behave, there are no purely military solutions for any insurgencies either. ... You can't simply go in and smash people and establish order that way.

What was missing in the CENTCOM strategy was an understanding of the synergy that has to exist between a military kinetic strategy to defeat the insurgency as a prerequisite for convincing the insurgents that they have to play in the political process, because what you had going on was Sunni Arabs were unwilling to accept the fact that they would not control the new Iraq, and a lot of them supported the few of them who turned to violence as a way of gaining leverage in the political process. …

If you don't take that tool away from them, if you don't make it clear to them that violence is not an option and is not going to provide them with leverage, then you're encouraging them to continue to be violent. And this was the problem that I saw all along with the CENTCOM strategy as it was playing out.

... What happened in the summer of '06 that has [led people to think], "Ooh, this isn't going the right way." Is it just the accumulation of all of [the struggles with the insurgency leading up to that time], or is there something post-Samarra that we know?

Both. I mean, there's a lot of things that are going on here. This is a very complex situation. To begin with, there's the disappointment that the election and seating of an Iraqi government is not drying up the violence, and there's a recognition that that's probably not going to happen in the way that we had hoped that it would.

In addition to that, you continue to have very virulent Al Qaeda in Iraq and associated movements attacking the Shi'a relentlessly in the belief that if they get the Shi'a to retaliate against the Sunni community, they can mobilize the Sunni community. It's a very weird strategy, ... a very cynical one, a very horrible one. But it's unfortunately successful. And they're continuing with attempts at spectacular attacks before and after the Samarra mosque bombing. So that's going on.

The Sunni Arab insurgency is not going away, and the elections don't make it go away. In my view, the reason is because we haven't persuaded the Sunni Arabs that violence is not going to provide them with additional leverage. After the Samarra mosque bombing, you start to see the very beginnings of the mobilization of the Shi'a population to defend itself against these relentless Sunni Arab attacks which have been going on now for over two years. Now you have the real danger that this is going to morph from an insurgency into an insurgency plus outright sectarian violence that is not connected with insurgent aims.

I think it's worthwhile talking a little bit about the differences between these terms: "insurgency," "civil war," "sectarian conflict" and all like that. Any insurgency is a civil war by definition, because it occurs within a single polity, and any conflict within a single polity is a civil war. So this had been a civil war from the outset, and to hold debate about when it became a civil war is a non-debate. It was a civil war from April 2003.

The issue is that an insurgency has a fundamentally political purpose, and that political purpose is usually to depose a sitting government and then ideally to replace it with another one. [It's] possible, by the way, to do one without doing the other, which is a frequent occurrence, which is how you get cycles of chaos.

Sectarian violence can just be sectarian violence. It doesn't have to be in the service of any particular political agenda. That is to say, you can simply get a degree of hatred between communities where people go out and try to kill members of the other community just because they think it's a good thing to do, or because they think otherwise they will be killed, which is the more common element of it. And that is a kind of war of all against all, which is what people have in mind when they talk about civil war, and which is what we began to see in the very early stages after Samarra.

It should have led to a re-evaluation of our approach. It should have led to a fundamental change in strategy by the summer of 2006, because you can't approach that kind of sectarian conflict in the same way that you would approach a straightforward insurgency. A straightforward insurgency, you could conceivably imagine if you had enough time that you would build up indigenous forces that would be able to suppress the insurgents. And that is a model that has worked in a variety of places. Now, it usually takes many, many years, and we were always wildly optimistic about how rapidly we were going to be able to make that happen. ...

But when you actually are dealing with a sectarian conflict, you cannot rely on indigenous forces to put down the violence, because every single person in those indigenous forces belongs to one sect or another. And when that is the issue on the table, indigenous forces cannot be impartial imposers of peace. That's why you need to have external forces play that role. This is something that we've seen in Bosnia and we've seen in Kosovo, and we've seen it fail where we didn't really do it at Somalia. We've seen the catastrophe that resulted in Afghanistan where we didn't try at all. And we've seen the explosion of all of sub-Saharan Africa when we didn't do it in Rwanda.

[What did you hope to propose when you were invited to meet with the president and his advisers at Camp David?]

... I had published an article [in The Weekly Standard] advocating conducting major clear and hold operations, ... basically coming down the river valleys toward Baghdad, modeled on the Tal Afar operation that had been so successful the previous fall. ... It was a multiphase proposal saying, "Don't try to do it all at once." That's a militarily lousy plan. ... You need to break these things into phases; otherwise they just become even intellectually unmanageable, let alone from the standpoint of resources.

So I said: "OK, look, we know how to do this. We have the model of Tal Afar; we know how to clear and hold these cities. Let's start clearing and holding cities. We know where the problem is." I tried to deconstruct the problem in Iraq and say: "There's a lot of people we're talking about; there's 25 million Iraqis. ... Most of them are not the problem." The Sunni Arabs comprise a population of about 20 percent of that, so about 5 or 6 million people that you're talking about. If you throw in the whole population of Baghdad, maybe you're talking about 10 million people which you need to worry about.

That's not an unmanageable figure. You can do operations that work on populations of that size with reasonable force levels, especially considering the insurgency at the time was heavily based in individual cities and towns up the upper Tigris, upper Euphrates and Diyala rivers. The nice thing about that is you can do what they did in Tal Afar and places like that. You can seal the town off and clear through the whole thing and then restore it, and it's isolated from other areas. When you get into Baghdad and you start trying to do that with neighborhoods, it's more complicated.

So I proposed this plan that was a very detailed -- this is how many brigades you would need to do this, ... the phases -- that I published in The Weekly Standard. I suspect that is the reason why I was brought in at that time.

And what did you go there hoping would happen?

Honestly, I went there just eager to have the opportunity to lay a plan before the president to say, "Let us seize hold of this opportunity." I knew that Zarqawi had been killed. I thought that it was an opportunity to have a major turning point if it was seized upon and to try to encourage the administration, which had really been slipping into a lot of long-war language. It had really been starting to say, "This is going to take a long time. and there's no quick fixes," and like that. I came in to say, "Look, I actually think you can get this problem much more under control in a relatively short period of time, you know, 12, 18, 24 months." ...

So I was hoping to change the nature of the discourse that was going on and have people think a little bit more about a concrete plan for us to take more positive control of what was going on in Iraq.

... [What was] your reaction to the president picking up his tent and heading off to Baghdad [before the Camp David meeting had wrapped up]?

It didn't affect me. We'd spoken with him on the previous day, and they didn't give me what the plans were for the rest of the conference anyway. So it was news that what he had done was a surprise; it was a dramatic gesture. Since I don't know what the deputies had been trying to accomplish, I can't really speak to how much of a problem it was that he did that.

Was it the first time you'd been to Camp David?

Oh, yeah, sure. ... It was a very cool experience. They flew [us] up on a Chinook from Fort McNair, which is also a cool experience, to fly along the Potomac like that. It's a beautiful place, and it was quite a good setting, I think. I thought it was very positive that they had decided to do a two-day offsite really to focus on this issue, understanding what presidential administrations are like and how many crises they have to deal with on a minute-by-minute basis and how hard it is to get them to focus on any particular issue for a long period of time.

Were you nervous presenting to the president of the United States?

Sure, sure. The thing is, I mean, yes, it's the president of the United States, and there's a lot at stake here, and I take very seriously the burden on anyone who's going to make a recommendation about this war, because you do have soldiers' lives on the line. ... Any minute that I'm speaking about this, I have a number of my closest friends and former students who are in Iraq being shot at, driving around, worrying about IEDs [improvised explosive devices]. And I feel a very serious burden to make sure that the advice that I'm giving is the best possible advice. So when you have an opportunity to do it in a way that may actually have a real impact, then yeah, of course you feel nervous about it.

And after it was over, you felt you had had a fair opportunity to present it, and you felt good about it?

I felt that I had accomplished what I wanted to accomplish and that he was going to make the decision he was going to make. …

I was disappointed with how the strategy developed after that point. I mean, I'd been disappointed with the strategy all along, but I think that we continued to drift; we continued to focus too much on training Iraqis with the belief that they would solve the problem. We did not seize the capture and killing of Zarqawi to clear Baquba. We let Baquba go to heck; we let other areas go to heck. We continued to engage in this sort of raid-patrol kind of operation, which is really antithetical to good counterinsurgency. And all the while, you have the continuing mobilization of sectarian strife. …

I did not understand, and I still don't understand, the commitment to this same strategy and the apparent unwillingness to reconsider it fundamentally in light of two facts: one, that it clearly wasn't working; and two, that the situation had changed with the Samarra mosque bombing, which had become apparent by the summer and especially by the late summer and the fall. …

At the same time -- and I think it's important to mention this -- we have really suffered from the fact that the opposition to the war has not been constructive and that there have been many, many opportunities for critics of the war to challenge the administration on the way that it was fighting the war that had been missed as critics through the end of the 2004 presidential campaign focused on whether we should have gone to war or not at all. ...

There's a name we haven't mentioned yet in our conversation. How much do you figure Abizaid's theory when he came in, and Casey's responsibilities when he came in, are an outgrowth of Secretary Rumsfeld's [policies and views]?

We're not going to know the real answer to that until 25 years from now and all the records are declassified, in my view. I've heard a lot of conflicting stories about what Rumsfeld's role in all of this was.

Look, I was at West Point when Abizaid was commandant, and I didn't meet him, but he had a phenomenal reputation, and he has always had a phenomenal reputation as a thoughtful officer who did what he thought was right on every occasion. I don't have any reason to believe that he was executing a strategy other than the one that he thought would succeed. I would be very surprised if we found that he had allowed himself to be driven into some sort of strategy that he thought would fail without fighting tooth and nail in a way that would have become apparent.

So I don't know what Rumsfeld's influence was, but I am pretty confident that the strategy you saw executed in Iraq was the strategy that Gen. Abizaid was comfortable with and was a strategy that Gen. Casey was comfortable with.

We talked to [State Department Counselor Philip] Zelikow, who says he went over [to Iraq] on behalf of Secretary [Condoleezza] Rice, looked around in '05. Keeps coming back, keeps writing her memos. ... First one says "tipping point." Second one says, "Uh-oh, uh-oh, watch out." Third one says: "We've got to clear, hold and build. You've got to seize at least the semantic lead. The White House isn't saying the right things; you've got to say the right things. The Army clearly doesn't have a strategy. I don't know what the hell they're doing over there." What's going on with that?

I don't know. First of all, it's certainly not fair to say the Army didn't have a strategy. The Army had a strategy. We may not have liked the strategy -- I didn't like it -- but they knew what they thought they were doing. And so that rap is not a fair rap.

Secretary Rice clearly got the president sold on clear, hold and build. She put it out in a speech, and then the next thing you know, he's holding up [Col. H.R.] McMaster and Tal Afar as the example of what's going to be done.

And the weird thing is that the president, having endorsed that particular strategy, seems to have no effect on the ground, because the president has then announced this strategy which the military then doesn't execute. And I think Rumsfeld at one point even says: "Oh, we're not doing clear, hold, build. We're not doing that." I don't know the insider politics of the administration to tell you what was going on.

What I can tell you from the outside is that's a dysfunctional administration. Something's not working there if you've got the secretary of state enunciating a policy like that, the president adhering to it and endorsing it, and then the military commanders not executing. Something is broken. I don't know where exactly it was.

Zelikow said a lot of the problem from his perspective was it was a Baghdad-driven policy; that the White House and even the civilians at the Pentagon were listening to Baghdad all the time and then reacting, and that it should have been the other way around. ...

Look, the president can always control the agenda. Those guys work for him. And any moment that he chooses to say, "I don't like what you're telling me; I want you to tell me something else; I want you to find a new plan; I want you to give me a new approach -- this doesn't seem to me to be working," they have a legal requirement to do that. That's the funny thing about how chains of command work.

So if [what you describe is] going on, it's because the president is choosing to allow that to go on, which he said he was. I mean, it's not as though Bush has been in any way opaque about this. He said all along that he was listening to his generals, and he was following the advice of the experts on the ground, and he was explicitly not going to repeat the mistakes of Vietnam, where there was micromanagement, and the mistakes of the Kosovo war, where there was micromanagement, and he wasn't going to tell the experts how to do their jobs.

I have always felt that that was a problem, frankly, because first of all, at the end of the day the strategy is the president's strategy. The military commanders execute it, they can develop it, but the president is the commander in chief, and at the end of the day the responsibility is his, and therefore he should be involved enough to be sure that he's really comfortable with what's going on. And this managerial style, I'm not comfortable with.

But it's part of a larger problem that we have in American society with our notion of what civil and military relations are supposed to be, because we have this idea that civil and military relations are good when they're peaceful, when everyone's on the same sheet of music, when everyone's working together. I don't think that's true. I think civil and military relations are good when they're tense, because the truth of the matter is that you've got military experts who are expert in certain things; you've got experts at the State Department who are expert in other things. The president is a guy who's got his own degree of expertise; it's certainly political expertise and in managing a variety of things. You've got experts on the National Security Council. Unless you have phenomenal "group think" going on, these people are not going to agree. That's great. That's an opportunity for real dialogue and discourse and for problems to be surfaced and solutions to be sought. …

So I think we've had a problem with the way that this has been run; that the military's been given -- the uniformed military, in this regard -- was given too much leeway to develop and execute the strategy in Iraq as it chose. Even apart from the question of whether it was a good strategy or not, just as a managerial technique, it's not a good way to run a war, in my view.

And yet even though they've been given responsibility, and maybe even authority, because of the lack of unified command, ... does Casey have freedom to [say], "I'm going to go do this thing, and I think that's right, and I want to do it,"? ...

There have been a lot of discussions about, of course, the problem of lack of unity of command and so forth. I have never heard anyone propose a particularly good solution to that problem. I mean, do you want to subordinate Casey to the ambassador? That would be a real revolution in the way we do military operations, if you actually made the military commander the subordinate of an ambassador who doesn't have military experience.

I think hypothetically that's what Bremer expected, right, when he went there?

That may be, but that would have been a remarkable thing to do. On the other hand, there's tremendous resistance to the idea of subordinating an ambassador, who is, after all, the personal representative of the president, to the military commander on the ground as well. This is where personalities play a big role. This is where there are staff or fixes that could have been looked into probably more than they were, in terms of making sure that their staff [is coordinated]. ...

The failure really to mobilize the government, to say nothing of the American people, behind this war allows these kinds of things to persist, because when you mobilize the government you say: "Winning this war is the most important thing. I'm going to subordinate objectives to that. I'm going to subordinate your complaints to that. Now, everybody, work this out." Then you put the burden on the individuals to make sure that things get coordinated. If you don't do that, then people have different agendas, and they want to approach them in different ways.

... So we're now in the summer of '06. Things aren't going well. [Operation] Together Forward II is a failure. ... [What happens?]

What happens is that my boss, Dani Pletka, [vice president for foreign and defense policy studies], has a conversation with a retired officer about, what do we do? At AEI what can we do to be helpful in this effort? And they have an idea. She runs it by me, and I say, "OK, look, the thing that is missing from this discussion is a serious sort of proposal for what a military operation to secure Baghdad at this point would look like," because Baghdad had become the center of gravity of all of this by this point. … The insurgents had made it clear that they thought that Baghdad was the key, and they were going to win or die in Baghdad. And the president had declared that we were going to succeed in Baghdad and really put a marker down on the capital, and Together Forward was an extension of that marker. …

So I threw away my notion of coming down the river valleys and said, "OK, we'll have to do it in reverse order then, have to start in Baghdad." The problem was that the idea of clearing and holding Baghdad was being dismissed out of hand, and people were saying it would take a million soldiers; it would take a million and a half soldiers; it would take 500,000 soldiers. …

So when Dani came to me with the idea of doing something, I said, "Look, why don't we assemble a team of expert military planners with recent experience in Iraq and our own regional team and bring in some other advisers with regional expertise, and let's sit down and figure out what this would be and what would actually be required to clear the city, to bring this under control?"

One of the guys that we brought in was Jack Keane. Tom Donnelly, who was my colleague at AEI, had been working with Jack, knew Jack. I'd had a couple of conversations with Jack. It seemed appropriate to invite him in. As we explained what we were doing, he was very enthusiastic about it. So he was one of the military experts who came in and spent a lot of time with us over the course of a three-day exercise when we sat down with satellite pictures of Baghdad and [completed a] detailed study of what the problem was, who the enemy groups were, what the neighborhoods were like, what was going on. [We] had our planners who'd been there, who'd been elsewhere, who'd been involved in Tal Afar, and said, "OK, tell me what it would take."

They sat down and started putting units against neighborhoods and came up with a recommendation. And Gen. Keane was there for part of that, and then we briefed him and Gen. [David] Barno, [former senior U.S. commander in Afghanistan], who was also there. "This is sort of what we think it would take." And he said, "OK, that looks pretty good." That's how he became involved in the project. Then he became very enthusiastic about it. We ended up sort of working as a team, selling it after we had developed it.

... [Is it unusual for retired officers to get involved in creating a war strategy, such as they did with your plan?]

There have been a lot of officers at a lot of ranks who have been frustrated with what's been going on there, and especially people who'd been associated with or seen some of the more successful operations have been very frustrated with why they weren't replicated, why efforts weren't made to follow up on the successes in a variety of ways. This is normal; it happens in war. And it doesn't necessarily mean anything, because there can be reasons that are beyond what some of these officers see for not doing stuff.

In my view, in this case they were right, and that was one of the reasons why I tapped into a number of veterans of the Tal Afar operation, because I think that they did have it right, and I think it was a model that mutatis mutandis could be applied to Baghdad, especially since we'd already had operations in Baghdad where we'd seen that we could clear neighborhoods, and we hadn't held them properly. …

None of us expected the report to have the result that it did. When I called [the retired officers who participated] up, I just said: "Listen, I'm frustrated that this debate has been so vacuous. I'm frustrated that no one has really laid the detailed military requirements of any of this on the table. How about we sit down and do that, and then we can raise the level of the debate?" And they said, "Boy, that would be great." So they came in, and that's all we were trying to do. …

I don't even know how much of an influence it had on policy, and there are important differences in the way that the president actually has been announcing and briefing the plan from what we actually proposed. It will be interesting to see how things actually develop. I'm not yet prepared to say that they're executing our plan, so I don't even know what the actual impact of the plan was.

I think in general terms what had happened was the midterm elections had obviously demonstrated the American people thought we were losing, and they thought that was unacceptable. The Baker group [the Baker-Hamilton commission, also known as the Iraq Study Group (ISG)] had presented its report, and I think the effect of that report was actually a little bit odd, because that group had tried very hard to find a middle ground between really sort of seizing control of the situation, as we recommended, and withdrawing, as other people were recommending. They tried to define this military approach that was [to] dramatically increase training but reduce forces and so on. The way that report was received, I think, made it very clear that their attempt to find a middle approach had failed, and in fact there was not a persuasive argument for any sort of middle ground and that the choices really were sort of double down or leave.

And in that context, the president's decision, knowing the president at all, is simple. The president is not just going to leave. If the president believes that there's any prospect of success, then he's going to pursue the prospect of success, which in my view is right, because I think the cost of failure is so high. ... And so I think the Baker group actually had the effect of making it easier to persuade the administration that it really did need to go bigger, that it really did need to increase forces, and it did need to take a more dramatic approach, more direct approach, to solving the problem. ...

... How did it become clear to you that whatever it was you were doing and saying or writing was having an effect on [the president]? Did you ever know that?

Look, I haven't spoken with the president since the Camp David meeting, I'll tell you that. And whatever I've heard, or whoever I've spoken to in the administration, I didn't know where the president was going until I heard the speech. ... I have no idea what kind of influence I was having on the president based on what I was writing all along. As I said, 25 years from now I'll be really interested to go back and look at the documents and see what was actually going on.

If you have a nutshell version of the plan, what is the plan that you figured out with Keane and others? ...

The plan, in a nutshell, looks at the problem of Baghdad and it says, "OK, the problem with Baghdad right now is the sectarian violence, which we need to get under control." You have the Shi'a groups that are based in and around Sadr City that are attacking into the mixed Sunni-Shi'a and Sunni neighborhoods that are between the Green Zone and Baghdad International Airport and that are around the Green Zone, which is problematic in all kinds of ways, including the fact that you've got so [many] reporters based in the Green Zone, that if the area around the Green Zone is exploding, then everyone thinks, including Iraqis, that all of Iraq is exploding. It's a huge perception problem. …

Now, we decided that going after Sadr City in the first instance was not a good idea, because the Shi'a part of this equation is a lot more complicated than most people think. There's a lot of different groups competing for legitimacy within the Shi'a community. Jaish al-Mahdi [the Mahdi Army] and [radical Shi'ite cleric Moqtada al-]Sadr are only one of them. If you assault Sadr City at the outset, then you're very likely to unite the Shi'a against you, which is not currently the situation. So we said, "We don't want to do that." ...

And then we said, "OK, now let's look at these Sunni-Shi'a neighborhoods." ... There were about 23 neighborhoods that we came up with. Then we said, "How much force do you need to put down in one of these neighborhoods to clear and hold it based on Tal Afar, based on operations we've seen, based on traditional military planning?" Do you need to put a battalion down in every one of these? ... A battalion is 600 to 1,000 [troops]. …

Support troops are really important here. … We need to have engineers, because engineers will build you the structures that you need, the defenses and so forth. They can do reconstruction activities; they can do a lot of things. So lots of things that are identified as support elements really are sort of combat elements in reality, in counterinsurgency efforts.

But to put a battalion down on each one of these things, that was more than we thought was necessary, frankly. We were trying to be on the safe side. We tried to pick high estimates of what we thought would be required. That gives you a requirement of 23 battalions; you've got about three battalions in a brigade. That gives you a requirement for eight brigades. There were already five operating in Baghdad. You need to have a reserve because the enemy's going to do things that you don't expect. And remember, this isn't a plan to go after all of Baghdad, so they're going to attack a neighborhood where you need to be able to respond to that.

So we ended up calling for the deployment of four to five combat brigades in addition to those that were already in Baghdad, giving all of the units in Baghdad the mission of securing a population, which they had not had up to that point, and using what were developed as traditional approaches here, clearing and holding neighborhoods, and we knew how to do that. So we laid all of this out.

Then we looked at the problem a little bit larger, and we said: "Now, the insurgents are going to break and run. Some of them are going to try to go to Anbar; they're going to try to destabilize Anbar. They may go to Diyala." So we recommended increasing force strength in Anbar by two Marine regimental combat teams [RCTs], the equivalent of brigades, a total of about maybe 8,000, 9,000 troops, recognizing that those units would be designated [for] Al Anbar [but] might end up needing to be sent to Diyala or Salah al-Din or somewhere else depending on where the rebels go. That's another form of a reserve.

So we tried to create a surge structure that gave the commander a lot of flexibility, that had built-in reserve. It was opposite of the approach that Casey laid out in his testimony, where he said, "I don't want to send a single American soldier to Iraq more than what I think I need." We said, "I've never heard of a war being won that way." You send more force than you need so that you have it available. We're dealing with an adaptive enemy that's going to try to make us fail. That's their job; that's what the enemy tries to do. ...

The net surge was going to be a total of five Army brigade combat teams [BCTs] and two Marine regimental combat teams -- total combat force, maybe 35,000 troops when you add the additional logistics support that we figured would be necessary, maybe up to 45,000. That was the military component of the plan. We said the Americans should be in the lead. We should have unified command structure that the Americans are at the head of in a military operation here, because the Iraqis are not capable of doing this. If they'd been capable of it, they'd have done it already.

But we also identified the need to have reconstruction efforts coming in, and we talked about that some, and we noted that this was Phase I of a report. I'm currently engaged in writing Phase II, which looks at reconstruction and training and a variety of the other things that are engaged in this. And so we said unlike the Iraq Study Group, which sort of put its report on the table and walked away, we were continuing to study and work this problem.

What happened to Phase I? It's requested by the White House?

Gen. Keane went into a publicly advertised meeting on Monday after the weekend when we did this, and he can tell you about that meeting; Eliot Cohen, [Rice adviser and director of the Center for Strategic Studies at Johns Hopkins School of Advanced International Studies], and Gen. [Wayne] Downing, [retired commander of the U.S. Special Operations Command (SOCOM)], and a couple of other people were there. And he had been here -- we briefed him on the plan. I presume that he discussed it with the president. And then there were subsequent meetings that he and I had with various officials.

So when ... you watched the president's speech [announcing the surge], ... I know there was a lot of press about, well, it's going to be the surge as invented by you guys. How did that word get out, that it might be something Keane and Kagan had figured out?

... When we did our public release of this in January [2007], you know, we had Sen. [John] McCain [R-Ariz.] and Sen. [Joe] Lieberman [ID-Conn.] there as well speaking. We'd already done a lot of selling of this, and it had already gotten a lot of attention, so it was natural for people to say that.

The big divergence between what the president described and what we had advocated was his emphasis on the Iraqis in the lead and the announcement of this dual command structure that was created. [In the president's plan], we have an Iraqi command structure and we have an American command structure, and they don't really hook up at any point. I've been very uncomfortable about that. … The other weird thing that was going on is that we've got a new strategy; we've got a new plan; we know that it's going to be executed by a new commander. But the administration starts briefing subtactical details of the plan before they'd even nominated the commander who's going to execute the plan, which is weird. ... Administration officials should not be talking about what battalions are going to be doing. That's not something that's appropriate for White House officials to be doing on the whole. They can and should be talking with generals; they should be being briefed on that. But their concern is with overall strategy, and the details of that variety should be coming out of MNF-I [Multi-National Task Force-Iraq] headquarters. ...

I'm really sort of suspending judgment [on] the degree to which this is our plan and the degree to which I think it will succeed until I see what [Petraeus does].

Why [did the president choose] Petraeus? ...

I'm a big fan of Dave Petraeus, and I think they got this one exactly right. You've got a guy who was commander of the 101st [Airborne Division] when it went in, saw the initial operation, saw the transition to the post-kinetic phase, ran Mosul, ran the area in the north. Did a pretty good job, by all accounts, but certainly had the experience of having to deal with that from the division commander's viewpoint. Great thing.

Then he came back. And then he went back over to command MNSTC-I [Multi-National Security Transition Command-Iraq], which was the organization that was designed to train up the Iraqi force. So here you've got a guy who's fought with an American division in a counterinsurgency fight in Iraq, in a tough part of Iraq, and you've got a guy who's also built up the Iraqi security forces and knows what they can do, knows what their issues are and all of that sort of stuff.

Then after that, he goes to Fort Leavenworth, [Kan.],where he oversees the production of a terrific manual on counterinsurgency, and he really gets to think through all of his practical experience and all of the theoretical basis for all of this stuff. And then you send him back.

It's as though they had been designing him to be the guy to take charge. And I think it's a great pick. It's good to remember that this is the first time you're going to have a guy in a position like that who'd actually been in Iraq before. Casey was vice chief of staff of the Army when he went over, which was an interesting pick, to choose him at that moment. ... [Casey's] a very senior general, which I suspect is why he was selected, because they wanted a four-star general in that position to make it a real position that really can coordinate, and I can understand that. ...

Presumably he's extracted a couple of promises from the president that he can operate with some freedom and get what he wants?

The president has said all along that he's listening to his commanders in Iraq and will give them whatever they say. I don't know what more promises Petraeus needs than the one the president gave to Casey and held to Casey, you know? When Casey or Abizaid said they wanted to do things in a certain way, that's pretty much the way they were done. I don't think there's any reason to imagine it's going to be different with Petraeus. ...

When will we know whether what Petraeus is involved with is the endgame?

This operation is going to have to develop over a considerable period of time. The additional forces cannot arrive in theater before the end of March and April, so the real major operation to secure Baghdad can't really begin until April or May. Depending on how the enemy reacts, ... by mid-June, if they've given it everything we've got, which in many cases is the best case for us, then you could see us moving toward a much more peaceful situation.

[The enemy] think[s] we're not serious. The enemy thinks that we're going to leave soon. So what they're more likely to do is go to ground and hide in the neighborhoods and expect to pop up again when we leave. I would bet you that you will see a surge in violence, however successful we've been, in Ramadan of 2007, which is in October, as you've seen a surge in violence in every previous Ramadan in this war.

And the question is going to be not does October 2007 look better or worse than August 2007, but does Ramadan of 2007 look better or worse than Ramadan of 2006 and by how much? That will give us a pretty good measure of where we're at, because they will have every incentive by that point to throw everything that they have left at us. If we've done our job, we should have been able to mitigate that significantly. If we can't, that hasn't succeeded.

In the future, once the surge establishes security, what's the next step? ...

You have a lot of political challenges in Iraq -- that's not surprising. You have a government that's been in power for less than a year at this point. It's getting its feet on the ground. They're very inexperienced politicians. You don't have professional civil service, and you have the sectarian conflict. We've been pressing them to move forward with political reconciliation and a variety of other things that [are] impossible to do when the violence is going on at this level. So the challenge is going to be as we bring the violence under control, we're also going to have to work with the Iraqis to develop a more inclusive political system, to bring the Sunnis in to the process while we have been showing them that the violence is not going to work. ...

Now, part of the process of all this is going to be disarming the Shi'a militias. ... The Shi'a militias are recruiting very heavily based on the sectarian violence -- that's one of their primary recruiting tools. It's "defend your people against these guys who are trying to kill you." When you take that recruiting tool away, I think we're going to find that a lot of Shi'a militiamen start to go home. But you're still going to have a hard core of folks who joined not because they wanted to defend the Shi'a community, but because they wanted to do other things. And those guys are going to have to be hunted down and taken out.

Ideally, by the end of 2007/2008, you'll have Iraqi security forces sufficiently professional and self-confident enough to hunt those guys out by themselves. ... So we've got to be prepared for subsequent operations after you've secured Baghdad into Sadr City. And then you've got to get Anbar completely under control; you've got to get Diyala completely under control.

My hope and expectation would be that if we actually break the back of this ... insurgency in Baghdad, as I think we can, the population will start to turn against the insurgents, who are maintaining violence in support of what looks like a hopeless cause, and these challenges will become smaller over time. … But we have to be prepared for more serious struggle, because this enemy has been very adaptive and determined. That's why we have argued that the surge has to be prolonged, 18 to 24 months' worth of increased presence to make sure that if things don't go the way we expect and want them to go, then we're prepared to resolve them in other ways. …

Is right now in effect the moment where we're fighting the Iraq war really for the first time? …

Well, let me put it a slightly different way. We've fought three campaigns in a row so far. The first one was the kinetic operation where we took the regime down by the end of April. The second one was the counterinsurgency operation that we followed from April 2003 through the Samarra mosque bombing. The third was the sort of countersectarian conflict operation that we have been waging in fits and starts since February 2006. We won the first one; we lost the next two.

We're now embarking on the fourth campaign. History should tell you that we don't need to be dismayed by that. How many campaigns did Lincoln win before getting through? Not many. Lost a lot. After Gettysburg, lost some. In fact, after Gettysburg the situation looked so dire that Lincoln thought he was going to lose the 1864 election until [Gen. Sherman took] Atlanta, which turned it around. Is this Gettysburg? Is it Atlanta? Is it something else entirely? We'll see. But the fact that we've had mixed success in the campaigns that we've waged to date doesn't tell us anything about whether we're going to succeed or fail in this next campaign.

And if we succeed -- you know, the great thing about wars, usually if you win the last battle, that's the one that matters. The battles and campaigns that you've lost before become a lot less important as long as you win the last one. So the trick is to make sure that this is the beginning of the end, and it's the beginning of an end leading to a victory. ...

What of the fact that this policy, this latest strategy, depends upon Iraqi forces to a large extent? ...

The administration has been briefing that the Iraqi forces are going to be in the lead and that the Iraqi forces are reporting their own commander. I think both of those are mistakes. Are they fatal mistakes? Not necessarily. The fact is, you're going to have a surge; you're going to have 50,000 American troops run around Baghdad, and their mission is going to be partnering with Iraqis to help them secure the capital.

Now, one of the things we've seen over and over in this war is that when American troops partner with Iraqi troops, the Iraqi troops generally behave well and generally are seen in a positive light by the population. It's interesting that I've seen a number of blogs from Sunni Arabs in Sunni neighborhoods in Baghdad … where they send messages out saying, "Now remember, if the Iraqi National Police show up by themselves, bolt all the doors, hide your women," and [things] like that. "If they show up with the Americans, it's OK. Let them in. It's fine; don't worry about it." That tells you a lot. ...

I've always been critical of this strategy that Casey and Abizaid have been pursuing that says you have to let the Iraqis do everything for themselves. I don't think the Iraqi military needs to have the capability to do a large-scale clear and hold operation of insurgent-controlled territory by itself. … We need to do that stuff for them. Once we've done that, we need to stay with them, and part of staying with them is making sure that they continue to behave and that we help the population come to trust them and we help the Iraqi security forces come to trust the population. We brokered that deal.

We've done this successfully in the past. This is what's gone on in Bosnia; it's what's gone on in Kosovo. And we have helped to bring the level of violence down. It doesn't automatically lead to a political solution. But the thing that we need to understand is the first requirement of American national security strategy in the Middle East is to end the violence in Iraq, because the violence itself is eroding the region. If we do nothing beyond getting the violence under control, we will have accomplished a great deal. Even if we then have to maintain significant American forces in a benign environment to help secure it, as we did in Bosnia for 10 years, we have still accomplished a great deal in place of simply letting it explode.