National Security

To COIN or Not?

A decade later, what lessons haven't we learned from the war in Iraq that we should?

On March 19, it will have been 10 years since the United States invaded Iraq. Foreign Policy and the RAND Corporation teamed up to bring together many of the key players who launched, fought, analyzed, and executed the war, including everyone from Bush national security advisor Stephen Hadley and Gen. John Allen to Doug Feith, the controversial Bush Pentagon aide who advocated for the war in 2003, and Paul Pillar, the CIA analyst who later went public with his doubts. It quickly became clear that, even a decade later, every aspect of the war -- from its rationale through each phase of its execution -- remains hotly contested. We hope this unique conversation adds to the record of how we understand that war -- and in particular, what its consequences will be, and already are, for future American national security debates. We present here edited excerpts from the second part of the discussion, on America's belated embrace of counterinsurgency in the Iraq war -- and whether and how the legacy of that shift will live on after the conflicts of the post-9/11 decade are long behind us. The session was moderated by FP's Susan Glasser, and you can find a full list of participants and their bios here. The first part of the transcript can be read here.

Susan Glasser, editor in chief, Foreign Policy: Clearly there's a debate around this table on the question of, what was the nature of the insurgency, was it an insurgency, was it a civil war? When did we understand it to be the case? And of course, we have so many interesting perspectives on that, including the person who wrote the book on [counterinsurgency]. So I am going to turn to John Nagl.

I also wanted to dial back in time for one second and say, what if we were having this conversation at the end of the Vietnam War? I assume there were many conversations like this that were in fact held in the 1970s. So why and how did we lose that thread? And in the spirit of that, what are the threads that are most important for us to take away from this going forward?

But I do want to start with John, who did write the book on that. I think there's a little bit of a new conventional wisdom that COIN, counterinsurgency doctrine, was perhaps a tactic that we've turned into a strategy, that we've embraced it as sort of a cure-all. Now maybe there's a backlash to that in the sense that, well, America's not going to get into these wars anymore, and so therefore what do we need it for?

Lt. Col John Nagl (ret.): I tried to figure out why it was that the Army had buried the lessons of Vietnam so effectively, and wrote both the best and the worst doctoral dissertation written on counterinsurgency in the 1990s because it was the only one. And then having written the book, I went and did the research in Al Anbar in 2003/2004.

This was a time when we were forbidden to use the word "insurgency" to describe the insurgency in Iraq. I can refer you to the video of Secretary of Defense Rumsfeld correcting Pete Pace, when General Pace used the word "insurgents" to describe the people we were fighting in Iraq at a time I was fighting insurgents in Iraq -- and I was, in fact, fighting both Sunni insurgents and AQI [al Qaeda in Iraq]. And I came back from Iraq, we printed up coffee mugs that said "Iraq 2003-2004." We were winning when I left. [LAUGHTER]

I went to work at the Pentagon -- big change from Al Anbar, in Al Anbar I knew who I was fighting -- and started trying to resuscitate the idea of counterinsurgency. [I] had some success with that, with Dave Petraeus as part of the general relearning process for lessons that have long existed -- that RAND, among other organizations, discovered in the first counterinsurgency era in the '60s.

There is nothing particularly new in the Counterinsurgency Field Manual [written by Nagl and Petraeus]. Two pillars: one of them, protect the population first; the second, learn and adapt, build an adaptive learning organization. And those principles, I think, stand up pretty well.

Counterinsurgency is a set of tools used by the military, and by civilian organizations, and it is primarily civilian. And the military completely agreed that we have not properly resourced those capabilities. But it can only work in the context of a political framework that knows what we're trying to accomplish.

If there's one thing that we failed to do in Iraq and in Afghanistan as effectively as we should have, it's security forces assistance -- a long-standing principle required for success and counterinsurgency. We continued to not resource that properly. That is the raison d'être for the American Army in this century. It is refusing to accept that. We are continuing to mess that up, and will continue to mess it up until somebody grabs the Army by the shoulders and shakes it and says, security force assistance is your job. Do it.

Eliot Cohen, Bush State Department official: I want to make some somewhat nihilistic remarks. [LAUGHTER] The first thing is just to remind us all, counterinsurgency is a kind of military operation. There's an American style to counterinsurgency; there was a German style to counterinsurgency; there's a Soviet or Russian style to counterinsurgency. It's just a kind of operation that militaries do, and I think particularly in the popular discussion there's this tendency to call counterinsurgency the kind of stuff that's in the manual.

Three points. One, we didn't take much away from Vietnam because we stopped thinking about Vietnam after that conflict. And I saw that up close doing some work for the Army War College in that immediate post-Vietnam period. That doesn't necessarily mean that there were "lessons" with quotation marks around them that should have been chiseled in the forehead of every lieutenant colonel. Just we stopped thinking and reflecting about it.

Second, to my mind, the critical point, we do not know the other side's story. Vietnam, we are only now really learning their story, and it's partial. It's going to be much harder for Iraq because there's a whole bunch of other sides. We've talked al Qaeda in Iraq as if we all know what that is. I'm not sure I know what al Qaeda in Iraq is -- what connection it had with Osama bin Laden, how it was structured, who it took its orders from, where it grew from.

And the Anbari sheikhs, their story of the Awakening is different than our story of the Awakening. We just need to be very sensitive to the American tendency to be completely solipsistic and tell the story of the war as if it's us and some kind of ill-defined other.

And finally, having played a very modest role in helping get the COIN manual launched, I've got two big reservations about it. Actually three. One is a technical one, which is it underestimated the killing part of counterinsurgency and particularly what Stan McChrystal and his merry men were doing [with special operations]. I think that is a large part of our counterinsurgency success. We killed a lot of the people who needed to be killed, or captured them, and that's not something you want to talk about. You'd rather talk about building power plants and stuff, but the killing part was really important, and I think we have to wrestle with that one because it's obviously problematic.

But more problematic, I really wonder about this stuff about protecting the population as opposed to controlling the population, and I have real doubts now, which I did not have before, about a lot of the development stuff that we did. I'll just give you one vignette from a visit to Iraq, where I think it was around Tikrit, and our wonderful PRT leader was telling me what happened when a commander read all the stuff about COIN, decided, you know, these people don't have good water, we're going to buy them a water purification plant. So he buys them a water purification plant.

The only problem is nobody's told the Iraqi ministry concerned that they now have that. Nobody has been trained to maintain that thing. There's no money anywhere for spare parts, so the thing works for 6 months and dies, and then the locals say, you bastards, you did this to us deliberately, didn't you? You know, you deliberately gave us a water purification plant that you knew was going to blow up in 6 months. We hate you even more than before.

I think we need to take a really close and somewhat jaundiced look at a lot of the development spending that we're doing, as opposed to simply bribing people to stay off the battlefield, which is a different matter.

Glasser: You're in favor of that --

Cohen: What?

Unknown voice: Bribing people?

Cohen: Yes, and I mean, I have a much harder-edged view of the whole problem I think than I used to.

Col. Peter Mansoor (ret.): I have four points. One, we hate COIN, we hate counterinsurgency, and we will never do it again until we do it again. And that's, I think, the lesson of the last 30, 40, 50 years. Great little pamphlet by Conrad Crane called "Forgetting Vietnam," in which he was trying to develop a course in counterinsurgency warfare [and] low-intensity conflict at Fort Leavenworth in the '80s, and he goes to the JFK School of Special Warfare thinking these guys are the COIN experts. [He asked] can I see your files on Vietnam only to be told, they were ordered to throw them away in the 1970s because we will never fight that kind of war again.

The second point is on this idea that we can kill and capture our way to victory in counterinsurgency, and I respectfully disagree. Where these networks are so robust -- and Iraq was a very robust insurgent terrorist network -- you cannot kill or capture your way to victory. We tried that for three years. And it's only when you have the synergy of conventional and special forces working together that you can eventually collapse a network.

The conventional forces take and hold ground, and control the population -- I agree, it is control, not protect -- and that forces the terrorists and insurgents to move and communicate. And when they do that, they can be targeted, killed, and captured. And then there's another component in the detention camps, where you separate the reconcilables from the irreconcilables. It's only [when] we did that in a comprehensive way in 2007/2008 that we collapsed the network in Iraq. And it was not just Stan McChrystal's guys. It was the conventional and special forces working together.

And the fourth thing is, I don't think we're going to have time or resources to train our forces for both high-end and counterinsurgency warfare. But it is much easier, I think, to dial down their skill set from high-end to low-end with appropriate pre-deployment training, than to do the reverse: train them for COIN and then ramp it up for them to be able to take on a high-end military.

So in terms of training, I would opt for training at the high end, but we need to educate our officers and our senior non-commissioned officers for the broad swath of warfare. And this is what we failed to do after Vietnam. We educated only for the high end. At best, counterinsurgency education was an elective course and sometimes not even available.

Kenneth Pollack, Brookings Institution: In response to both Eliot and Pete, you know, I'd say, I think you guys are both right. I mean, protecting and controlling the population, it's effectively the same thing. Sometimes you're doing one, sometimes you're doing the other. Where I think the difference is important, is again, the difference between what you're trying to do with an insurgency versus a civil war. With an insurgency, you are trying to help people in a whole variety of different ways, protect them from a whole variety of bad things. In the civil wars it's about breaking the relationship between the militias and the populace. In some ways I think the best way to see the difference between the insurgency and the civil war is in the outcomes of the two of them. John is right, at the tactical military level, the tactics are, by and large, the same. But once you start getting up to that political level and what you're trying to accomplish, things are very, very different.

And I think where we really came off the rails [was] in '09 and '10 in particular. Where in [an] insurgency what you're ultimately looking to do is to legitimize the existing government, in an inter-communal civil war, it's really about power sharing, because what you have is a struggle among different identity groups, all of whom see an opportunity for power [but] also who are deeply fearful that if they are not the ones in power, they are the ones who are going to be slaughtered.

Go back to John Allen out in Anbar. One of the most important shifts that nobody really has talked about too much is the shift that occurs in that '06/'07 period where we go from being the enablers of the Shia, in particular, and to a lesser extent the Kurds, to becoming the protectors of the Sunni community. One of the most important things that happens in '07 is that we reach out to the Sunnis and we say to them, we are going to bring you into the government. We are going to make sure that the government provides for your community, gives you an equal share of political power, and protects you rather than having, basically, quasi-government death squads killing all of you in Baghdad and elsewhere. That is absolutely critical in turning the Sunnis around.

And, looking forward, you want to solve the problem of Syria? Until someone can convince the Alawites and the Druze and the Kurds and the Christians that they are not going to be slaughtered under a majority Sunni government, they will keep fighting, okay? It will be the same exact situation.

Steve, I apologize, I could not disagree more with Stan's characterization of AQ. AQ is important, it's not unimportant, but to suggest that somehow this is a war about AQ -- you know, again, Stan did a remarkable job. But you know, he had his target. And you know, I think the book does a great job of portraying his target. But AQ takes advantage of an opportunity. We've seen this all around the Middle East whenever --

Stephen Hadley, Bush national security advisor: That's -- I agree with that.

Pollack: -- you know, whenever we see an inter-communal civil war, AQ jumps in, because they realize, hey, we can be on the side of the Sunnis or we can be on the side of the Muslims fighting against the Americans or whoever the oppressor is. It gives them entree into it.

Where I think that we've really missed something is in this strategic application of what it is that you're trying to do where the political meets the military. And again, what I saw in '09 and '10, what we were doing was not simply trying to leave a legitimate government and then walk away; what we were doing was actually preserving the power-sharing arrangement that we had hammered out in '07 and '08, and which is unraveling literally as we speak.

Philip Mudd, former CIA official: There was no AQ in Iraq and there still isn't. AQ is a globalist organization that said, your target in Algeria, Egypt, Saudi Arabia, is outside your borders. It's in Washington and New York. What we had in Iraq was actually an adversary that said, we're not interested in going outside our borders. We're staying home. And furthermore telling AQ core, forget about your ideology, I'm going against the Shia. So we can use the shorthand of AQI, but I don't think they were there before, and they certainly weren't there during the war.

Hadley: I've got to say, I don't get that. When Zarqawi issues a manifesto, and hooks UBL to support him on the grounds that he's going to begin to establish the caliphate in Anbar, we see the plan. I mean, this is the part of establishing an al Qaeda control, which they do in Anbar. Sunnis decide actually they don't like it, and throw it off.

So you know, I get it. Al Qaeda always comes in in these situations. They're doing it in Syria. My only point is, they came in in a big way and a successful way, and they were to able to lead a movement that resulted in about 15,000 to 20,000 fighters and to say --

Mudd: I don't agree. People who said they were al Qaeda came in, people who, they wanted the jersey -- like me wearing a sort of, Griffin the Third jersey and say I'm on the Redskins. Well, I like the brand, but I can't play the game. [LAUGHTER]

Mansoor: One, al Qaeda in Iraq is what they call themselves. It wasn't a label applied to them. And two, they did have a plan for going outside the borders of Iraq. It was a written plan. It was, first establish a safe haven in Iraq, and then destabilize Jordan and Syria, use that as a base to conquer Saudi Arabia and Egypt, and then destroy Israel. And that was a written plan that they used.

Amb. Charlie Ries: The most important thing I learned in Iraq is that economics are not some technical problem, that it is about power, and it is political economics. And so it was really important for us to understand that.

Second thing, it seems to me, is that the biggest economic stimulus in Iraq was security. The economy in Iraq boomed when the security improved. And so, again, there's not one or the other. It's integral to the way that the society functioned.

I think that, in retrospect, the huge ERF [Emergency Response Fund] funding of 2003, the multiyear $20 billion, was as much a burden for us as it was an advantage. Because we had so much money, we had this desire to commit it and spend it and deal with all these problems. If we had less money, and we had a sense of more time, we could have done a lot better. We all know that we didn't spend anywhere near as much time getting Iraqi buy-in. The point that Eliot made about people in the water security plant -- there's a gazillion such examples throughout Iraq.

A point that I think was made earlier about whether there should be a civil military command, or that, you know, put the military in charge of all aspects of reconstruction from the rebuilding of the security forces to the rebuilding of the markets, in communities. I actually think that that probably is a bad responsibility for the military, because they are not trained for it, they don't necessarily have these kinds of things. But we are really badly placed ourselves, on the civilian side of the government, to undertake this. And we ought to develop the capabilities to be able to do this so we don't have to invent it all over again.

The problem is that, with the military, you develop military capabilities as a deterrent to security challenges you might face in the future. And so the more that you have them, you could always justify them as a deterrent to bad guys doing bad things. To develop this civilian reconstruction capability in an environment in which we really don't want to do this nation-building, it is very hard to get political support for it. I don't know that there's a permanent solution for that, because we're not building capabilities that deter something; we're building capabilities that people might feel would lead us to doing things that are otherwise not popular.

Glasser: This is such an important part of the conversation. The very first thing that I remember when I drove into Basra the day that it fell to the British was, you know, on the one hand we saw this group of people gathered on a street, and they were very excited, they had just been liberated from the prison that they had been held in. And that sort of conformed with the narrative we expected, and they took us in and they showed us, you know, here's where they plugged me into the electric socket.

But then the very next thing we saw and heard -- and for the next two weeks -- was two words, which I quickly learned, and have since forgotten in Arabic, which were "water" and "electricity." And that's all we heard from the crowds on the street corners. And it was not what I had expected to hear.

Peter Feaver, Bush NSC official: So, four points. The first is that, when we talk about COIN, we mean population-centric COIN. Because there's a non-population-centric COIN that's the Russian model, which is not what we're talking about. So just as a clarification on terms.

The second is, the part of the debate in the '04/'05/'06 timeframe that hasn't come out was not a debate over the desirability of COIN, but rather, could you do light-footprint COIN because troops create the problem of resentment and producing antibodies -- this was the Abizaid view [and] was what General Casey, I think, was sort of trying to move towards. I would say that's the dominant view in the Obama administration today, is that you can do this on the light footprint. So while I think the surge proved the value of the approach we took, reasonable people were arguing on the other side, and I would say they may have actually won in terms of policy.

Third point. The phrase "Generals prepare to fight the last war" -- sometimes but not always. Sometimes the institutional Army commits itself never to fight the last war, and then takes steps and colludes with the National Command Authority to figure out ways to make it impossible to fight it. That's what the Vietnam Army did and that was part of the movement up to the Reserves, et cetera. We all know that story. I think we're in the process of doing that again, and we're making sure that we lose the capacity to do this.

The military hasn't lost it as fast as the State Department has. I think the State capacity to do this in 2013 is less than it was in 2007/'8 for sure, probably less than it was in 2004/'5, and that's a scary thought.

The fourth and final observation is, the reason the National Command Authority colludes in this self-lobotomization, is that the goal of American grand strategy over the last 150 years or so has been to avoid fighting the last war. The central goal of containment was to avoid fighting World War II -- confront the Soviet challenge without having to do it the way we confronted fascism in imperial Japan, and the central thrust of post-Cold War grand strategy has been to avoid another Cold War, to avoid a hostile peer rival. And the central goal of post-9/11 has been to avoid another 9/11 that drags us into Afghanistan.

At the grand strategy level that's sensible. You are trying to avoid fighting the last war, and the pernicious thing is that when you go from grand strategy down to strategy in operational capacity, is that it creates this perverse incentive to eliminate your capacity to fight this war that you're hoping to avoid.

Mudd: I hate to say this, one positive point that I think we learned might be a bit tactical and that is fusion and rolling operations real-time against an adversary that's networked. When we looked at what the adversary, especially the terrorist adversary, was concerned about, they were crippled, I think, by this idea that if you put all people in one place -- that is, all agencies -- take all the data in this data-intensive environment, and start rolling [operations], it's incredibly successful.

And when I think about next wars, I could see real application of this concept if you have these kinds of adversaries -- and to me a drug trafficking organization looks a lot like a foreign fighter network. They communicate, they need money, they have leadership, and they can't afford the kind of operational tempo that, with our resources, we can impose on them. To me that's a terrific lesson.

Gen. John Allen: Indeed, Phil, they laminate themselves together. So that you find them working together. And that's why the network-targeting approach works so well. It's an art form.

Amb. James Dobbins: Yeah. I'm a COIN enthusiast, but I wonder, in fact, if it's the right term or whether we should be talking about irregular warfare because counterinsurgency is essentially defensive, rather than defensive and offensive together. And in many ways, the best way of marginalizing extremists who are attached to an insurgency is not to suppress the insurgency; it's to support the insurgency.

There isn't an insurgent in the world that wouldn't rather have American support than al Qaeda support, if they were offered the choice. And that's what we did in Bosnia. That's what we did in Kosovo. That's what we did in Afghanistan in the '80s. That's what we did in Afghanistan in 2001. That's what we did in Iraq with the Sons of Iraq. That's what we did in Libya. We supported the Muslim insurgents, and marginalized the extremists as a result. That's what we should be doing in Syria.

And that means you have to be good, not just at counterinsurgency, you have to get good at insurgency. And it's worth noting that that's really what the special forces were created for. They weren't created for counterinsurgency, they were created for insurgency. And that's a lot of what they did, actually, in Vietnam, was create insurgencies in places like Laos, and places behind the enemy lines, and in Vietnam.

So I do think that in some ways, focusing exclusively on counterinsurgency is focusing on only half of the spectrum of capabilities that we need to sustain. And perhaps also focusing too much on engaging ourselves in long, lengthy conflicts against insurgencies that have some reasonable basis that we ought to be considering supporting, rather than suppressing.

Paul Pillar, former CIA analyst: Two brief observations. One, I would disagree somewhat with Eliot's observation that we just don't understand this sort of thing until we get on the ground. I think we've been in some situations where there was a lot of understanding, inside and outside government. The problem was it just wasn't harnessed or utilized.

In the case of Iraq, you had the whole Future of Iraq Project in the State Department; you had a major Intelligence Community assessment about challenges after Saddam was gone. And on the outside you had people like Les Gelb with the Council on Foreign Relations that proposed doing a similar kind of study, but it just wasn't utilized. So it's a matter of harnessing and utilizing, not so much finding out for the first time once we get on the ground.

In the larger discourse about COIN -- and I don't mean the sophisticated group in this room, but I mean the larger, unsophisticated discussion amongst the public and in places like Capitol Hill -- one of the biggest confusions has been between, on the one hand, the value and validity of COIN per se, versus the policy question in any one country as to whether the level of effort required for successful COIN is worth it for U.S. interests.

A couple years back I was giving testimony to the House Armed Services Committee on Afghanistan. And the focus of my remarks was on that second question. And I got met by one whole line of questioning from one member that was about whether, basically, you know, did I respect the military judgment of Commander X, and what he says we need. And that wasn't the issue at all. Of course I respected his judgment. The question is whether to do it successfully that was worth it.

So we keep getting some of this discourse that makes it sound like whether the manual that John Nagl and company wrote was valid. Yes, it is valid. But that's not the policy question in terms of whether to do COIN in any one instance.

Allen: I think we're really at a critical moment right now, as the clock continues to tick in Afghanistan. We're under 22 months remaining with the ISAF mission. And so as the services begin to look into the distant future in terms of how they prepare themselves for future conflict, we're going to see them looking at school curricula, professional and military education, all of those things that have been shaped so profoundly by irregular warfare and counterinsurgency operations over the last 10 years.

And the things that we have done on the ground in Afghanistan, and the things we did on the ground in Iraq, so much of that requires a really profound understanding of the social dimension of the environment in which you're operating.

I think the question will be, and what we'll need to watch very closely is, do we profoundly change our school curricula over time to reassert ourselves in expeditionary operations, or reassert ourselves in airborne operations, reassert ourselves in high-mobility, high-intensity, firepower-dominated operations, potentially at the expense of this really powerful reservoir of intellectual prowess we've built amongst our officers.

I think it's a critical moment, and as we begin to look at ourselves for the next 10 years, how are we going to remain true to what we've learned over the last 10?

So much of what we learned has been not just in countering the insurgent, but in removing and dealing with those underlying social factors that create the insurgency to begin with. And they're related, and you may have to counter the insurgent in some areas while you are seeking in every possible way to remove, or eliminate, or to address those profound underlying circumstances that generated the insurgency.

So the challenge, I think, for the future for us, is to understand how best to prepare our forces, ultimately, for irregular warfare.

John Nagl and I were in a conference not long ago where I talked about some of the immutable principles associated with warfare in the future. One of them is that we must always have a fundamental understanding of the central fabric of the environment in which we are going to be serving. We spent a great deal of time before we went to the Anbar province studying the tribes. Tribe by tribe, from the Syrian border right down to Baghdad. And I don't think anybody spent more time the sheikhs than I did. And I could tell the sheikhs stories about their grandfathers, because we spent the time learning about the tribes. Thus we were able to operate within the tribes and not around the tribes. They were not terra incognita. They, in fact, ultimately welcomed us because we took the time to learn the social dimension of them.

When the insider threat in Afghanistan began to loom large for us, my first response to that as the commander in Afghanistan was to go back and reverse-engineer our cultural training with respect to both the NATO forces that were coming into the theater, but the U.S. forces as well, to determine whether we had missed, conceivably, some aspect of the preparation of our forces, which was playing out in a way that conceivably created the insider threat, that lesson learned from Anbar province.

With respect to governance, one of the most important things that we've relearned again in Afghanistan, and we'll learn it again later, is how do we legitimize the government? And it's critical. It's not just the central government, it will have to be the subnational governance as well. And the great challenge for us in the Anbar province was to get the sheikhs to acknowledge ultimately there would need to be a civil government. That's where we used development as the key mechanism by which we could get them to the table, because if we could agree on projects that could improve both the lives of the members of the tribe, but also appear to have been flowed by the civil government, there was the opportunity, through the use of properly spent development dollars, to amalgamate that tribal and civil governance.

But then the greater challenge for us [was] to connect the outlying governance with the central government. I remember [a] sheikh reaching over and patting my thigh and saying, here is my government. Well, I couldn't be his government; it had to be al-Maliki. And ultimately, that was where we, I think, were successful in connecting the subnational governance to the national governance, but that doesn't come naturally. I mean that kind of capacity within our military does not come naturally.

And for me, the great breakthrough was the arrival in Baghdad of a fellow by the name of Ryan Crocker, who reached out to us, as we were reaching towards Baghdad. Not that the previous State team there was not capable of doing it. The circumstances had changed so that Ryan's skills now could be brought to bear as the circumstances changed, to provide the connectivity that we needed over the long term.

And finally the security forces assistance. The foreign forces in a counterinsurgency do two profound things: You shake the insurgency by the use of conventional and special operations forces through intelligence, through counterinsurgency techniques. But you also have to undertake the security forces assistance to prepare the security forces ultimately to replace you as the defeat mechanism of the insurgency. And so, for us again in the Anbar province -- a province that was uniquely Sunni, that was uniquely out of the same tribal confederation -- an early decision by the central government of Iraq to let the Anbari tribal sons remain in the province, in the police force, and in the Army permitted us to bring two divisions online and to go from about 4,500 police to nearly 30,000 police in the better part of a year. And it was a profound decision that made all the difference in the world.

As we reset our forces for the future, we've got to maintain our faithfulness to the basic intellectual principles of irregular warfare, the components of which are such things as the proper employment of development, understanding the relationship of subnational and national governance, the social fabric in which you're going to operate. These are Ph.D.-level intellectual demands on our officers. We cannot permit that to go.

Kalev Sepp, Naval Postgraduate School: I would suggest that the American Army, in particular, is not good at training foreign armies. My feeling is that the Marines have done this much better. When I was with Third Marines just a couple of years ago, I was in the middle of a discussion among their battalion commanders where they were able to discuss the difference in burial rites between the different valleys in their sector of operations. I mean, you know, that knowledge of the culture that they were going to operate in.

But I would say that the American Army does this badly, has a history of doing it badly. The Korean Army that we trained from 1945 to 1950 collapsed under North Korean attack initially. The Vietnamese Army, very famously, after several years of U.S. Army advisors -- 17,000 of them in South Vietnam. At the Battle of Ap Bac, the better part of three South Vietnamese battalions were beaten by one Vietcong company, and this, of course, eventually provides some of the raison d'être for U.S. escalation because the army that the U.S. Army was training could not do the job.

And, then, the American 104th Division was a training division inside the U.S. Army structure. It's a Reserve division, but it's called a drill sergeant division. It's only 8,000 [troops], but for general war to rapidly expand the American Army, the idea was that this division would train several American divisions. It was sent to Iraq. It failed completely in training the Iraqis. Every unit they trained was judged not combat effective and had to be disbanded and retrained.

Doug Feith, Bush Pentagon official: I just wanted to comment on the importance of a civilian component to these military efforts. When we observed the problems of getting civilians into Iraq to help the effort, we saw very clearly the institutional deficiencies. I mean, we didn't have established institutions that could handle personnel questions, that could handle contracting questions. In other words, we were trying to create a major civilian effort from scratch.

There have been some criticisms that we should have gotten the postwar planning stuff done a little earlier. Steve was explaining why we delayed a bit or kept it low key for a while because of the concerns of how a highly visible effort of that kind could undermine diplomacy. But I think the key point is whether we started the higher profile, more coordinated, more visible postwar planning effort in October, versus in late December or January, is not -- I mean, I don't think there's great consequence there. We should have started it 50 years ago. I mean, we have been doing civilian kinds of stabilization and reconstruction stuff for very many decades, and the basic pattern remains the same, despite the very nostalgic view that people have of how brilliantly we did post-World War II.

If you go back and read Dean Acheson's memoirs, he describes the post-World War II reconstruction effort as a complete catastrophe. And it wasn't for three years before we got to the Marshall Plan. And so, the basic way it happens is you start with the Keystone Cops, always. After a while, we get smart, and you get some systems in place, you get some experience, you start to learn what the picture is on the ground. Eliot's point is very well-taken that, as much as you think you might know in advance, it's practically insignificant compared to what you learn when you're there.

And sometimes a lot of what you know in advance is not only inadequate, it's exactly wrong, as was the case in Iraq over and over again. I mean, a lot of the so-called intelligence about Iraq was precisely wrong. It wasn't simply less than you wanted.

And so you start with the Keystone Cops, you get smarter, you get better, you get skilled. You get teamwork established, and then you disband everybody, and you go to next event and you start with the Keystone Cops again.

That doesn't quite happen with the military. Military tends to get better. I mean, the military is kind of, in general, a learning institution. The civilian efforts at reconstruction and stabilization have not been institutional enough to learn anything. In other words, there is no institution.

When we saw this problem -- and I think Steve Hadley deserves credit here and President Bush deserves credit here -- we said, there are some institutional things that should be done to try to put us in a better position to handle problems like this in the future. The creation -- as it turns out, over the objections initially of the State Department -- of the Office of the Coordinator for Reconstruction and Stabilization, that was an important step. And the basic idea was to try to create an institution that would have a built-in expertise that would last, that would be able to study operations the way the military does, learn lessons -- the equivalent of professional military education.

One of the things that deserves some attention is, why has that effort gone with so much difficulty in the State Department? It never really took off. It was fought by almost all the regional bureaus -- not all, but almost all. Now it's been turned into a bureau, basically what had been that office. It's still not thriving, it's still not quite doing the concept that the NSC staff and the Pentagon had in mind when we championed this idea, we had to kind of ram it down the throats of State Department people who didn't want to take on the responsibility.

And if General Allen is right, that this is a really important thing for the U.S. government to have the capability to do, to bring civilians in to assist the military, then it's worth trying to look at the recent history, where there were some proposals to deal with it institutionally.

I think Charlie Ries put his finger on it when he said that there are people on the Hill who look at it the way some people look at nuclear modernization. They say, we don't want you to do that because we don't want to make it easier for you to use that tool. And that's been a serious argument.

Part of the counterargument is the one that Steve Hadley highlighted, which emphasized how valuable this capability is, in advance of a conflict, to maybe allow you to head off a conflict. And that's a strong argument. But why this has been -- I don't want to say a complete failure, but not the success that it should have been -- deserves attention.

David Rothkopf, Foreign Policy: If I could just make a two-finger intervention on this particular point myself. Jim and I got to know each other because we were responsible for the reconstruction of Haiti into the thriving economy it is today. [LAUGHTER] And subsequent to that I ended up writing a book for Carnegie on this issue of post-conflict reconstruction and stabilization, and looking at Haiti and at Bosnia and at the Palestinian territory and some other places we've tried to do it. The recurring theme [is] this is the thing we do the most when we intervene. And no one in the United States government wants to own it.

And the problem with having the State Department own it is, they can't do it alone. This is one of those areas where you need a whole-of-government solution. And it's an institutional void that we keep tripping over. It's among the things that has dogged us since the Second World War.

The final point I want to make is, every time this happens somebody says, we need a Marshall Plan. And you know, what they neglect to say is that the reason Marshall Plan worked is, we had achieved complete victory. We could impose our will on everybody there, there was a strategic urgency to write a check for however big we wanted to write the check, and the countries we were rebuilding had actually been thriving economies before we started to do this. And none of those conditions have ever existed in any of the other places we said we needed to have a Marshall Plan with the exception of Japan.

Feaver: Elliott Abrams told me that on my tombstone he's going to write as my epitaph, "It's worse than that." Because that was what I always said in every meeting with Steve and the others. And I think it's worse than that in the following sense: that if the trio of Rice plus Gates, and then Gates plus Clinton, plus supermajorities in the House and Senate, could not fix the resource imbalance, and they didn't, we have seen the high-water mark in terms of the balance between State and Defense.

In the next 20 years it's going to be worse than it's been in the last 10 years. And I can't foresee a combination of secretaries and appropriators on the Hill that would do more to rectify the balance than we had in the last five years, and we couldn't get it done then.

So my message to the Defense Department is, you're going to have to do it as the first go-around the next time, because you will not have the capacity in the State Department, even that you've come to count on in the last five years.

Pillar: It just needs to be noted that the most authoritative Intelligence Community assessment done pre-war on what the post-overthrow challenges would be in Iraq, a redacted version of it is on the Senate Intelligence Committee website free for all to read and form your own judgments. But unfortunately most of it turned out to be pretty prescient.

Greg Jaffe, Washington Post: It struck me when Eliot said there's a Russian way of COIN and a German way of COIN and an American way of COIN. It struck me with regard to Iraqi and Afghan security forces assistance. We continually wanted both of those armies to fight an American way of COIN. You know, I can remember many U.S. company commanders and U.S. battalion commanders talking to their Iraqi counterparts saying, you're not doing COIN right. This is how you do it.

I'm haunted a little bit by something that a brigade commander said to me a couple years ago in Afghanistan, which is that we took the best mountain fighters in the world and we've turned them into a fourth-rate NATO army.

So it's in an area that I think as we shrink the military down, you know, the Army can be inclined to do less and less and less in this. And I really think that they need to do more and more and more.

Glasser: I'm struck, as I'm sure all of you are, with the really profound ambivalence reflected in this conversation, which began with everyone's sort of collective determination not to unlearn these lessons that were purchased at such a high cost, and that this isn't going to be Vietnam, we're not going to put this away. And yet the conversation inevitably circles back around to a real uncertainty and discomfort and ambivalence about whether this is a capacity we want inside our government, never mind the question of where.

Foreign Policy

National Security

Ends and Means

A decade later, what lessons haven't we learned from the war in Iraq that we should?

On March 19, it will have been 10 years since the United States invaded Iraq. Foreign Policy and the RAND Corporation teamed up to bring together many of the key players who launched, fought, analyzed, and executed the war, including everyone from Bush national security advisor Stephen Hadley and Gen. John Allen to Doug Feith, the controversial Bush Pentagon aide who advocated for the war in 2003, and Paul Pillar, the CIA analyst who later went public with his doubts. It quickly became clear that, even a decade later, every aspect of the war -- from its rationale through each phase of its execution -- remains hotly contested. We hope this unique conversation adds to the record of how we understand that war -- and in particular, what its consequences will be, and already are, for future American national security debates. We present here edited excerpts from the first part of discussion, on "Ends and Means," moderated by RAND's James Dobbins, and you can find a full list of participants and their bios here. The second part of the transcript can be read here.

James Dobbins, RAND: There are some people who feel that we bit off more than we could chew in Iraq, that we had too ambitious objectives when we went into Iraq. And there are others, including myself, who feel that we didn't resource it adequately in the early months, and as a result had a hard time clawing back from those early days. I'd like to ask Steve Hadley to say a few words about ends and means, goals, objectives, and early commitments.

Stephen Hadley, Bush national security advisor: Did we have a match between ends and means in terms of the invasion and the toppling of Saddam? I would say yes. We had one mistake: We did not pay attention to the rear area security problem. We knew in the planning it was a problem, and we failed to get it teed up in the interagency so it would really be addressed in the systematic way it needed to be addressed, and we paid for that.

Second, was there a mismatch between ends and means in terms of the objective? You know, the lore out there was we went to war to bring democracy to the Iraqi people. That was not the case. We went to war to achieve some hard national security objectives.

Before we went to war the president had, in the situation room, a conversation about, once we topple Saddam, what is our obligation to the Iraqi people? Is it simply to substitute an authoritarian who will not move against our interests by supporting terror, invading neighbors, pursuing WMD? Or do we have an obligation because we are the United States of America, and because they've suffered under 30 years of a brutal authoritarian. Do we have an obligation to give the Iraqi people a chance, an opportunity, to build a democratic future for themselves?

The president decided on the latter, and I actually think we achieved that objective. It wasn't pretty, and Iraq today is not pretty, but it has an opportunity to build a democratic future despite the enormous pressure that Syria and other events are putting on Iraq.

Third, did we have a mismatch between objectives and resources in terms of the insurgency? And you know, Stan McChrystal's book is very interesting, because it makes crystal clear that what Iraq became was a struggle against al Qaeda in Iraq. And I remember -- and I'm not getting partisan here -- I remember in the summer of 2008 when then-candidate Obama said al Qaeda was the ball. The Bush administration took their eye off the ball, and they went into Iraq, but al Qaeda isn't in Iraq. Al Qaeda is in Afghanistan.

And I asked Mike McConnell at the time -- [director of national intelligence] at the time -- how many al Qaeda fighters are in Afghanistan today? And how many are in Iraq? And he said, in Iraq there's about 15,000 down from about 20, and in Afghanistan there's 200.

So you can say we failed to foresee that Iraq would become the front line of al Qaeda's struggle against the United States, and I think we did not have the right strategy or the right resourcing in the end of the day to deal with that problem. And it took us a long time. And part of it was relearning a lot of things about counterinsurgency we did not know. So I'm not sure it's a mismatch of objectives and resources. It may be something we should have anticipated and planned for but we did not, and it took us a long time, too long, and too high a price, to finally get our arms around that problem, and defeat al Qaeda in Iraq, and that's what we did.

Fourth, post-conflict stabilization and reconstruction: We knew we had a problem there, because we did it one way in Bosnia -- I would say, the international approach. We tried another way in Afghanistan, which was the pie approach -- give every one of our national allies a piece of the pie, at least initially. And we found that everybody had a small piece of pie, which meant it was a second- or third-order priority, and nothing got done.

So in Iraq we decided, since the military actually knows how to get things done, we'll give it to the military with a lot of State Department and other agency help. And I think what we found is the military made a valiant effort but they didn't have the full skill set to solve the problem either, which raises a broader problem. We've spent 60 years investing and learning how to recruit, train, exercise, fight, and improve our military, and we have an unbelievable military. We have not made a similar commitment to develop the kind of civilian capabilities you need to do post-conflict stabilization reconstruction, or to get in pre-conflict to help unstable and post revolutionary states from descending into conflict.

I don't think that was a reason not to remove Saddam given the national security objectives, but I think one of the conclusions we should draw is we've got to make an investment in those capabilities, because I think, to assume that we're never going to need them again post conflict, and don't need them pre-conflict in -- in failing, or post revolutionary states, I think is wrong..

Eliot Cohen, Bush State Department official: I guess I just have four thoughts of ends and means. I could never get a clear sense of what democratic governance meant for a country like Iraq, and how ragged an outcome we could comfortably live with.

Secondly, I think part of our problem is we forgot how to do military government. I think the military really didn't want to have anything to do with anything associated with the sort of stuff that they had actually done quite a bit of in other places, particularly in Europe, after World War II. I think part of the problem here is the atrophy, or the disappearance of capabilities that maybe only mass mobilization military could have.

The third question I have about means is, how much development is really called for in all this, having seen how much of that got wasted, or was actually counterproductive? I really wonder whether we went overboard in how much we've spent, and that we ended up spending money in ways that created dependence, created resentment.

The fourth thing [is that] the large part of means is institutions and institutional culture. I think there were -- and there probably remain -- huge problems because I suspect that the half-life of our memory of these conflicts is going to be really short. And that the instinct of the military is not going to be to hold up a mirror to its own performance [to say], "How do we make sure that sort of thing doesn't happen again?" It will be to say, "God, we're never going to do that again." And then we'll find ourselves in a room like this -- well, maybe my kids will -- you know, 15 or 20 years from now.

Peter Feaver, Bush NSC staffer: I've gone back through a lot of the criticisms [of the war], and I've noticed several common flaws that are worth keeping in mind. The first one is a failure to do serious counterfactual analysis. What would have been the situation if we had chosen a different course?

The second level is, what if things had gone slightly differently -- those kinds of plausible what-ifs that could have happened but didn't happen that might have changed [the outcome].

And the third grand level is the counterfactual from the point of view of al Qaeda. Iraq broke al Qaeda in a way that it did not break the United States. The U.S. paid a terrible price, but Iraq turned out to be a breaking point for al Qaeda.

Lt. Col. (ret.) John Nagl: I certainly agree with Steve Hadley that there are hard national security reasons why we invaded Iraq, and certainly one of them could not have been that there was al Qaeda in Iraq in 2003, because there was not. There were an awful lot of al Qaeda in Iraq. I fought them later, but they came after we got there. So if our strategy was to attract al Qaeda to Iraq, it worked. I don't remember that being one of the arguments.

So the casus belli was the Iraqi WMD programs. The ends/means mismatch was that if we were moving to secure Iraqi WMD, we did not have sufficient forces to do so. And I know this personally. When I arrived in Iraq in September of 2003 at Taqqadum Air Base, one of the largest weapons bases in all of Iraq, it was completely unsecure. The fence around it had been dismantled by the Iraqi people who were selling the metal for scrap iron. There were literally weapons lying in the streets. They could have been anything, and many of them were.

This failure to provide sufficient troops to secure Iraqi weapon sites which was the justification for the war, had been foreseen by Marine Lt. Gen. Greg Newbold, and he's written about his struggles inside the Pentagon to try to send sufficient troops to accomplish that mission.

I'll make just one more point. We have not talked at all about some of the critical policy decisions that were made, decisions like the complete de-Ba'athification of Iraq which contributed to the events that followed. I'll talk just briefly about what I consider to have been the most grievous of those errors, which was the disbanding of the Iraqi Army.

One of my buddies was a tank battalion commander, fought into Baghdad and was working with an Iraqi major general who had essentially demobilized his division but had his soldiers ready to secure Baghdad as a responsibility of the occupying power. My friend and he were having discussions; the order came down to disband the Iraqi Army completely. My friend had to tell his interlocutor, the Iraqi major general, "I'm sorry, we're not going to be able to work with you."

The Iraqi looked him in the eye, straight in the eye, and said, "That means I will be fighting you tomorrow." My friend said, "I know." They stood up, saluted each other, and the next day the IEDs started.

So the decisions made in Washington not to provide sufficient troops to secure the WMD sites, later to de-Ba'athify, and to shut down the Iraqi Army, were huge contributors to everything that followed.

Col. Peter Mansoor (ret.), former aide to Gen. David Petraeus: We've talked about resources and whether, had we provided more resources or better resources, this war would have gone differently. And I don't think in the first three years it would have.

General Petraeus used to say that it's important for a senior leader to get the big issues right. And early on, we did not. And the de-Ba'athification of Iraqi society and disbanding the Iraqi Army did more than just create significant problems for us. In my view, those decisions created the insurgency. There would have been insurgency anyway but it would have been much smaller; it would have been, as Secretary Rumsfeld used to say, the dead-enders. But by de-Sunnifying Iraq, in the minds of the Iraqi Sunnis, we created our own problem.

I love President Bush's memoir Decision Points, because in it, he talks about these decisions, and he goes, you know, we did not talk about these in the National Security Council, and we should have. And he takes responsibility for that. And he says, now, had we talked about these issues we might have come to the same decisions but at least we would have done it knowing the second and third order effects that would follow from them.

Walt Slocombe, senior adviser to the Coalition Provisional Authority: Well, I think Mr. Mansoor has hit on a real difference of analysis here. It wasn't de-Ba'athification, it was de-Sunnification. And it really followed what Steve Hadley has described as President Bush's very serious and important question. Do we want to install another authoritarian? Which would have meant, in that context, Sunni regime in Iraq or not. You can argue that point, but that's what it was about. [Nobody thought the Iraqi army] could be reconstituted as anything other than a Sunni militia.

I think a fundamental error was having a divided command, one -- with a military command responsible for the military operations and a civilian structure under Ambassador Bremer responsible for the civilian side. The fact of the two of them, to put it mildly, General Sanchez and Ambassador Bremer, didn't exactly get along particularly well didn't help. But I think even with very good personal relationships, it was an unnecessary burden.

I think the biggest lack of committed resources was time and patience, that the assumption was, at the beginning, that it would be very easy, and then it was a period when the assumption was, well, this was going to be hard and we're going to have to stay a long time.

I don't think there was a sufficient analysis of how exactly we were going to create [a democracy], and it was certainly not a sufficient commitment of resources. I mean, I guess this is on the record so I probably shouldn't say it: We didn't let KBR waste enough money in the early days. We should have done things like tell KBR, we don't care what it costs or how you have to do it, but get the goddamned electricity fixed. That's one of the things which my sense was, was that there was an understandable sense of disappointment that things didn't suddenly get a lot better for the ordinary Iraqi, whatever their sectarian affiliations, and things like electricity would have made a big difference. It wasn't that we didn't try; it's that we didn't succeed.

It's all very well to say, well, we did very well in Germany and Japan, and for that matter Bosnia and Kosovo or Haiti, where we didn't do so well. But in none of those situations was there an active, armed insurgency opposition trying to stop us.

Kenneth Pollack, Brookings Institution: I think one of the mistakes that we made at the time, and we continue to make, is to think of Iraq as an insurgency. Iraq was an insurgency. There was an insurgency there. But that was not the primary problem in Iraq. The primary problem of Iraq was a civil war. It's a very quintessential inter-communal civil war. It was a civil war ultimately of our making because we created the security vacuum that gave rise to it.

The thing was that the insurgency was directed at us so that's what we tend to focus on. But when you think about the dynamics of Iraq and what was important to that country, and ultimately, what allowed us to achieve what we achieved in 2008/2009, it was really because we were solving the problems of the civil war, even though we oftentimes didn't refer to it that way explicitly.

Kalev Sepp, Naval Postgraduate School: I would like to go back to what Peter Feaver had raised before about counterfactuals. Related to the weapons of mass destruction, the counterfactual question would be, what if WMD had been discovered? Would that have changed the subsequent events at all?

The once-classified slide decks that talked about one of the proofs of WMD was the deception measures that the Iraqis were discovered to be making. Both deception and denial measures. And they were all correct -- all of the camouflage, the obstruction of UN weapons inspectors, signals deception that was clearly related to the WMD sites.

But there was a single presumption about what that meant. And at a very fundamental level deception is conducted for one of two purposes. You either hide strength, or you hide weakness. It does not seem anybody explored the idea that they were hiding the fact that they had no WMD at all. Of course, the target audience for that was the Iranians, and we were the unintended audience for the deception.

Philip Mudd, former CIA official: If you think of the civil unrest piece, we have a lot of experience in looking at the Sikhs, the Kashmiris, Sri Lankans, the tribal areas of Pakistan, Afghanistan. I looked at all these other insurgencies at the agency, and it's not clear to me that provision of more assistance would have been helpful.

If you have a local population that supports the civil unrest, the ability to impose will is going to be extremely difficult, regardless of the investment. And again, looking at the Indians going against the Sikhs, the Kashmiris, the southern Sri Lankans and Sinhalese going against the Tamils -- a history of this tells me that if the locals like the message of violence, you can't stamp it out.

The second and final message is local power. If you want to stamp it out, you've got to have somebody locally -- and that is the Sri Lankan Army, the Indians in Kashmir, the northern alliance in Afghanistan -- who's going to crush heads.

And this is where I think the de-Sunnification piece comes in. One of our big assumptions was, why wouldn't these guys be with us? That was wrong. If we assume we're not going to have local support, we're going to have to crush them. And to crush them, my experience with all these other areas tells me, we need a local power who will do it, because we can't.

Doug Feith, Bush Pentagon official: There was a lot of talk in the run-up to the war about basic strategic ideas, and one of the most basic was that we would have a strategy of liberation rather than occupation.

We actually had a plan for political transition in Iraq that would have been a?variation on what we did in Afghanistan, where we would not have set up a protracted U.S. occupation regime. In Afghanistan, as everybody knows, we overthrew the Taliban but didn't set up a U.S.-led occupation government. But we wound up having an occupation government in Iraq despite the fact that there had been quite a substantial discussion and formal agreement at the NSC level that we were not going to have that.

When you have a problem with a regime, I mean, as we had in Afghanistan, as we had in Iraq, as we now have in Syria, North Korea, when you have a problem with a regime, and if you're going to take action to move that regime aside, how do you define your goals? And what are the consequences of deciding to run the place for a substantial period of time? I mean, that's quite important.

One of the consequences of what we did in Iraq by shifting, essentially, to an occupation approach, despite, as I said, the pretty widespread agreement before we went to war that that would be a problem, was that all of the problems of Iraq, which were very substantial (and more, in fact, than we knew as a government -- you know, like the horrible state of their infrastructure, which was worse than we had known). But all of those problems, like the electricity problem, became an American problem.

Is there some way to deal with horrible regimes like that, that you want to move aside, and even taking some steps, as I think Steve Hadley rightly pointed out, are kind of required of the United States, you're not going to want to put a dictator in place of another dictator. You're going to want to try to give an opportunity for the people to build better, more democratic political institutions. But is there a way to [deal with horrible regimes] where you don't wind up being the occupying power; where you don't wind up owning every problem that that country has because of its decades of misrule?

But I think that a lot of bad things flowed from the fact that we took the kind of ownership of Iraq and its governance, after Saddam's overthrow, that we did, and I don't think it was inevitable.

Paul Pillar, former CIA analyst: The textbook scenario, for whenever a nation like ours undertakes something really major, and certainly a big offensive war of choice is major, is that we have a policy process in which all of the relevant parts of the bureaucracy, as well as external sources of expertise, are engaged that examines in detail, what are the objectives that we would be trying to achieve if we undertake this? What would be the cost? What are the other side effects? And you weigh the pros and cons, and that's the way a policy process works. And as Rich Armitage once observed, we didn't have that with going to war in Iraq. And there was no process to decide whether it was something this country ought to do in the first place, as opposed to discussions about getting public support for the decision, or for implementing the decision.

And in my view, that's the most basic lesson to be learned, and to be applied in the future, is have a policy process. Of course, that gets right to the key question of doing the war or not doing the war. But it also subsumes so much else of what we've discussed already. You know, we didn't anticipate this, or we didn't apply enough resources to that. Those are exactly the sorts of questions that, in a thorough pre-war policy process, would be forced to fore.

Peter Feaver mentioned, for example, a paucity of counterfactual analysis, and I think that's a correct observation in terms of what we've seen over the last 10 years. But, of course, the right time to do the counterfactual analysis is before you do the war, not the subsequent 10 years.

You know, we had, back in the Johnson administration decisions on Vietnam in the mid '60s, a very thorough policy process, all documented in the Pentagon papers, and you can read about it in LBJ's memoirs, but that didn't guarantee success.

But I would estimate that not just 10 years from now but 50 years from now, when historians look back at this, the absence of that kind of process is going to be one of the most extraordinary things that historians will comment on.

Mansoor: I want to go back to the comment that Iraq was a civil war. Although I agree with that, I don't think it shows the complexity of the conflict. Because it was an insurgency, there was terrorism there, there was organized crime there, and it wasn't just one civil war, it was multiple civil wars. Shia on Shia, Sunni on Sunni, and Shia on Sunni. And the key, which we finally got around to in 2007, was to leverage the fault lines between them to find allies.

So I think the takeaway here for the future is, we need to understand better the local dynamics of these places if we're going to go in there with an industrial strength armed force. And I don't think we understood those dynamics in 2003 at all.

David Sanger, New York Times: Picking up on that same point, I had a chance recently to go back over notes from the interviews that we had done with members of the administration, including some in this room, in the weeks leading up to the war. And a couple of things jumped out at me. One was an assumption at the time, or at least a discussion of an assumption at the time, that it would be a war of liberation, as Doug said, rather than a war of occupation, because it would be able to cut off the very tops of the ministries, de-Ba'athify each of the ministries, and the following Monday morning everybody would come into work and the infrastructure would kind of pick up where it was, and Doug made the point, nobody understood the degree to which the infrastructure didn't work, but they also didn't understand the degree to which that bureaucratic structure didn't really exist.

When you go back and you look at the speeches that President Bush gave in the run-up to the war, there was one that he gave that dealt with the democratization issues, and there were six or seven that dealt with the WMD threat, including the big Cincinnati speech, which had the discussion of how attacks could actually strike the United States, and so forth. I suspect that, had that ratio been reversed, that it might have been a lot easier to set up the later conversation about the opportunities for democratization, and might have affected in some way the way we approached this.

But since it took, instead, five or six months to come to the determination that there wasn't an active WMD program even if there had been one before, that meant that there was a lag in coming up with what became a new justification for the war. And I think that that, in the end, ended up costing pretty vital time.

Rajiv Chandrasekaran, Washington Post: The political transformation plan was, in a sense, quite aspirational before the war. And as we look forward to sort of enduring lessons from all of this, particularly in terms of resourcing, the lack of attention to political transition strikes me as at near the top of the list. And it ties directly into the issue of de-Sunnification. Its true cost, yes, was the insurgency, but also leaves us with a legacy of the politics we have now, which are far from inclusive. And that ties directly back to de-Sunnification decisions, and had there been a more robust and meaningful transition plan, we might not even have gotten there.

Look at Afghanistan today. It's a problem of politics more than anything else. And you could argue all along, we've had counterinsurgency military strategies, we've had reconstruction strategies, but there's never been a very good diplomatic political strategy there, as we failed to have early on in Iraq.

Cohen: Yes, you need political strategy; yes, it's all about local dynamics and including personalities and so on. And my point would be, you never know that stuff until you're really there on the ground. You don't know that from the outside. My strong suspicion is, as I said, you don't understand the local dynamics until you are physically on the ground, until General Allen is out there with the Anbari sheikhs, and talking to them, all the time, and figuring out who's who and what's what.

And the implication of that is if you do do this kind of thing, you are doing occupation. Don't kid yourself. And I think we did kid ourselves, you know, just looking at those simple words, liberation and occupation, they can mean different things. Liberating Prague is one thing. Liberating Baghdad is very, very different.

Feaver: I had in mind many different counterfactuals, but the one that Paul and a couple other people had raised is, what if we had had no invasion?

Here's what I think the reality would have been. One, a much stronger al Qaeda than we ended up with 10 years later. Secondly, a conviction that Hussein had a much larger WMD arsenal than he actually had, because, of course, our knowledge about his absence of WMD involved interrogations of  Hussein, which we wouldn't have had. And third, based on that same rationale, the collapse of the sanctions regime in the UN, which would have allowed him to actually start rebuilding in a way that he would have been able to.

Is that a better world than the world we're in today? Maybe. I think people can still disagree. But it is not the idyllic world that the argument is usually made.

Last observation, Doug asked sort of the rhetorical question: Could we ever avoid responsibility? I think we're running a natural experiment on that very question on Syria. President Obama right now appears to be running the experiment that, if we don't intervene, we can avoid responsibility for the very predictable chaos that's coming. Then, when and if Assad falls, and the chaos that everyone has predicted comes to pass, we will all say, we told you so. And apparently the Obama administration's position is, but we're not responsible for it, and therefore it's not our problem. And I think when we're looking for lessons to go forward on Iraq, one of the hard questions we should ask is, what are the areas in the world where our inactivity will absolve us from responsibility from cleanup?

Feith: The strategic question is, is there a middle ground between the kind of inactivity in Syria that we now see, and occupation? Eliot's suggesting that there might not be, that occupation is inevitable in some ways. But it's an interesting question, because if the only option for dealing with a problem regime is forcing it out of the way, and then owning the country for a substantial period of time, then I think your point that the American people are going to say, we'll be damned for a very long time, is almost certainly the case. And maybe that is right.

Gen. John Allen: I was a deputy commander out in Anbar Province. We spent an extraordinary amount of time preparing ourselves intellectually for what we would face in the Anbar Province, which was, in many respects, a closed system compared to other areas within Iraq that folks had to deal with, which was, perhaps, multiethnic, multi-confessional. And the environment in which we operated in Anbar presented us with the challenge of dealing with virtually the complete absence of governance with the complete ascendancy of the tribes. And so understanding the human terrain, understanding the dynamics of the social fabric into which we were ultimately thrust gave us, I think, a number of opportunities.

One of the most important opportunities was not just our ability to deal with the tribes, to have entree to tribes at the level where we could deal with sheikhs as a matter of regular intercourse. But it also gave us the ability to recognize the potential value of the Awakening when it occurred. We really sensed that something was changing dramatically in the battle space, and we were very familiar with, and very aware of, the previous tribal attempts to organize, and how those had either failed or been stamped out by al Qaeda.

Leaving Iraq, I would become the Deputy Commander at CENTCOM where, to Dr. Hadley's point, I was very concerned that many of the lessons that we had learned in terms of the capacity for civil military execution, but even more importantly, the capacity for civil military planning, I was very concerned about losing the edge, potentially, of that. And so, for us, as we undertook at CENTCOM contingency plans for other potential regional issues, we spent a great deal of time understanding the environment in Phase Zero, which is called the shaping phase, which is peacetime, if you will, but also spent a lot of time trying to understand, in areas where there could be potential contingency operations, understanding tribal, ethnic, confessional governmental dynamics, so that should a contingency begin, we had attempted within the context of our formal planning to understand that, as you shift from peacetime to the contingency, in a very real sense, you are beginning your execution for stability operations, and ultimately the shift to civilian government. You're beginning that shift the moment that the contingency begins if you've done the planning correctly.

So an immediate lesson learned for me, a really key takeaway from my personal experience in Iraq was that, in the context of contingency planning, for contingencies which were really not even on scope at that particular moment for us, understanding what would be necessary to stabilize an environment following a contingency operation, and then shift that stability over to an indigenous civil governance was a very important takeaway for us.

And of course, the big takeaway, again, back to Dr. Hadley's point, the big takeaway was, while the military is quite good at that, frankly -- quite good at shaping in Phase Zero, what might be the outcome in Phase 4, there is still a great deal of room for the development of capacity, so that it is in fact a civil military, and ultimately a civil lead in Phase 4 and Phase 5. And that was a direct outcome for me. It is a concern of mine today, as we continue to think about the future: Are we adequately resourcing ourselves intellectually so that future contingencies, as a direct result of what we learned in Iraq, are we going to be properly postured at a civil military level ultimately to move to Phase 4 and Phase 5?

Michael Gordon, New York Times: There was a contradiction at the heart of the Bush administration's strategy in Iraq, which was a very simple one: It wanted to do regime change, and it didn't want to take on any of the burdens of nation-building. In fact, Don Rumsfeld gave a speech to that in the lead-up to the war.

And so everything I saw from the inside, in terms of the planning, the extensive planning on what they used to call Phase 3 was all about how to take apart the regime, how to defeat Saddam, how to get into Baghdad, how to destroy the Republican Guard. There was very little of what they call Phase 4, which was, what were you going to do after you arrived in Baghdad and inherited this situation? And the strategy was a hopeful strategy. It was Doug Feith's liberation strategy, which is that it was -- the war was intended to be kind of a turn-key operation, where we would rather quickly hand over to some sort of Iraqi entity -- this was codified in CENTCOM's own plans. This whole intensive, laborious, complex, difficult occupation that the United States presided over was backed into. It wasn't the original design.

But the basic point I think in these operations is it's not enough to simply destroy something. If you don't have a plan to fill the vacuum that follows, you shouldn't go in in the first place. And United States didn't have such a plan, and it later evolved such a plan. What the United States should have done and what the military should have done is, instead of spending 18 months planning Phase 3 -- how to get to Baghdad and take down the war, and sort of started -- spent 18 months on Phase 4, and worked on things once we're there when all these things happen, how are we going to handle this, what's going to be the political mechanisms and military mechanisms. Because Phase 3 was the easy part. You could have attacked Baghdad from Turkey, we could have come from Kuwait; we could have come from Jordan. The result was preordained.

It was Phase 4, the part that was given short shrift, that, really, in the planning process, on the political side and on the military side, that should have been the very first planning task. And if you don't have an answer for Phase 4, then don't do Phase 1, 2 and 3.

Allen: Let me just say this again: A Phase Zero is where you do this, which is, you know, the peacetime part of this thing. When you begin the operation, or you contemplate the operation, Phase 4 begins immediately. It's not at the end of Phase 3. You're mobilizing for Phase 4 as the operation begins. So however long it takes you to get to Phase 4, Phase 4 has really started intellectually for you in terms of mobilization of resources from the beginning. And that ought to be driving our thinking on every contingency plan henceforth.

Hadley: You know, what you see depends on where you sit. So I speak from my particular vantage point of the White House, and I recognize that everything I say can be discounted because it's so self-serving. One of the things Dr. Sepp, I think, said, which I think is exactly right in terms of failure of intelligence, it's really a failure of imagination. It never occurred to me or anyone else I was working with, and no one from the intelligence community or anyplace else ever came in and said, what if Saddam is doing all this deception because he actually got rid of the WMD and he doesn't want the Iranians to know? Now, somebody should have asked that question. I should have asked that question. Nobody did. It turns out that was the most important question in terms of the intelligence failure that never got asked.?

Secondly, Peter is right. We assumed we had about 150,000 Iraqi troops that we would vet, and then they would help us reconstruct and provide security. And they -- we can talk about disbanding the Army -- they melted away, was how it looked to us. They melted away. They were not there, including all their equipment. And I still wonder today where all that stuff went.

We did not have a plan B. We tried a plan B. Colin Powell went out and tried to get other countries -- remember, Egyptians and others -- [to] put in troops, and got to zero. And we never decided, we've got to put 100,000 more troops in there because we're 150,000 short. That was less a problem of planning or a fallback. It was more we didn't face up to the consequences of where we were.

Three, process. We had process. It certainly was not adequate, though in some sense pre-war no planning process ever is.

But I'll give you three things to think about:

One, there are constraints. If you think it's a war of choice, you have lots of time to do a planning process. I don't think Iraq was a choice -- a preemption or choice -- I think it was a war of last resort. After 12 years we ran out of options. And maybe it was a failure of vision, maybe a failure of planning, but it's either let Saddam off the hook or go to war.

One of the problems about doing this wonderful 18 months of post-war planning is when it leaks, everybody says, "See, the Bush administration doesn't want to get a diplomatic solution, they just want to go in and change the regime. We told you." And suddenly, your coercive diplomacy goes out the window.

A decision I kept struggling with was, when do you start the post-war planning and how do you keep it small so people didn't participate? It was not broadly within the government. So you don't blow your possibility of solving the problem without going to war. It was a constrained planning process.

It had three problems in my view.

One, I don't think what we did in the interagency planning process ever got translated into guidance to troops down at John Nagl's level, so it may be planning with no consequence. Secondly, I'm not sure CENTCOM's heart was really in Phase 4. And third, given 10 years of war and the experience that John Allen had, we would do a planning process a lot better this time than we did.

It's only when we got a handle on al Qaeda, and then changed strategy in January 2007, that we did all the things Peter was talking about. I think as we think about this we underestimate the extent to which al Qaeda played. Al-Zarqawi was in Iraq, of course, when we went to war, but he, as [McChrystal's] book makes clear, exploited the vacuum that was created to make it the central front in the war on terror. We're responsible for that. But it changed the situation that we planned for in profound ways.

Foreign Policy