News | July 1, 2015

From the Chairman: An Interview with Martin E. Dempsey

By R.D. Hooker, Jr., and Joseph J. Collins Joint Force Quarterly 78

On January 7, 2015, Dr. R.D. Hooker, Jr., Director of Research and Strategic Support at the National Defense University (NDU), and Dr. Joseph J. Collins, Director of the Center for Complex Operations in the Institute for National Strategic Studies, interviewed General Dempsey at NDU. Giorgio Rajao and Joanna E. Seich transcribed the interview.

U.S. Marine Corps M-198 155mm Howitzer gun crew of 4th Battalion, 14th Marines, Mike Battery, Gun 4, at Camp Fallujah engage enemy targets November 2004 (U.S. Marine Corps/Samantha L. Jones)

U.S. Marine Corps M-198 155mm Howitzer gun crew of 4th Battalion, 14th Marines, Mike Battery, Gun 4, at Camp Fallujah engage enemy targets November 2004 (U.S. Marine Corps/Samantha L. Jones)

Joseph J. Collins: Can you tell us how your views on the wars in Iraq and Afghanistan have evolved over your assignments as division commander, Multi-National Security Transition Command–Iraq [MNSTC-I] commander, acting U.S. Central Command commander, U.S. Army Training and Doctrine commander, Chief of Staff of the Army, and Chairman of the Joint Chiefs of Staff? That is an impressive set of perspectives on these wars.

General Dempsey (DOD)

General Dempsey (DOD)

General Martin E. Dempsey: I’d like to start with a vignette. I arrived in Iraq late June 2003 and took command of the 1st Armored Division. I had watched developments from Riyadh, where I was the program manager of the Saudi Arabian National Guard. There, I was being fed a pretty steady diet from my Saudi interlocutors about what was going well and what was not. I was also getting fed a heavy diet of Sunni Islam, obviously, and so I, like Bing Crosby, went on the road to Baghdad.

When I got to Baghdad, there was a sense of constant transition almost to the point of turmoil. For instance, I arrived just after Lieutenant General Dave McKiernan pulled out the CFLCC [Coalition Forces Land Component Command]. If you remember the CFLCC story, he was told, I wasn’t in the room, but I was led to believe LTG McKiernan was told by Secretary [Donald] Rumsfeld to take as much risk going out as coming in, which sounds like something Secretary Rumsfeld would have said. So CFLCC had literally taken this command and control architecture, unplugged it, and went back to Kuwait in the process of redeploying. V Corps, as you recall, was left behind with General [Ricardo] “Rick” Sanchez. And my sense was that V Corps was little suited as a command and control headquarters, understaffed and under-resourced, fundamentally a tactical headquarters.

My sense was that we were a bit adrift frankly, at least in Baghdad. I can’t speak to what was happening in Mosul, Ramadi, or Diyala Province. But in Baghdad, there was a bit of almost discovery learning, about what it means to have gone from this exquisite maneuver across the desert from Kuwait to Baghdad, to now being fundamentally responsible for the safety of a city of 7 million people, 75 square miles with a river running through it, and with deep ethnic and religious tensions.

I was trying to learn as quickly as possible what the mission was going to be because it was, quite frankly, unclear. The Iraqi army had been disbanded and de-Ba’athification had occurred. General David Petraeus at this time famously asked, “How does this thing end?” It was a fair question.

General John Abizaid came to see me around the time I took command, and I had a candid conversation with him about my initial observations, and I asked him as CENTCOM [U.S. Central Command] commander: “What is my mission, how would you articulate the intent?” And he replied, “Look, you’re going to have to take this armored division, you’re going to have to adapt it so that it can provide stability operations capability. . . . I don’t have to tell you how to do that.” But he added, “That’s job number 1: how do you take this organization that you have and tailor it in order to provide a safe and secure environment in Baghdad?” I replied, “That’s a pretty heavy lift, a safe and secure environment in Baghdad.” He also asked, “How long do you think we have here?” I knew exactly what he [was thinking] because he’s an Arabist; he’d been a scholar and an Arabic speaker. And I knew that he was asking whether the United States had a shelf-life here, or was this something we could consider doing in perpetuity, if necessary. From my experiences in Saudi Arabia, I answered, “Three years.” He stated, “I think you’re right.”

That was in 2003. Since then, I have realized in a conflict that either creates or inherits a failed state—in a conflict where the issues are historical as opposed to topical, in a conflict where religion is a factor—you separate yourself from your adversaries by innovation, not necessarily by size and technology. The rate of innovation and adaptation is likely to be the most important quality of a military campaign, not the things we normally focus on, such as Force Management Level [FML]. It seems like a recent thing with this administration, but we have been debating FML from the very start with Secretary Rumsfeld. We debated and negotiated resources before we debated and negotiated objectives. That’s my observation of my time between 2003 and the end of the Iraq War. You might place this observation on the negative side of the ledger, debating resources rather than objectives, but when objectives change, we should simply recognize this and adapt accordingly. Sometimes changing objectives is portrayed as mission failure, when in fact in a protracted campaign the likelihood of renegotiating objectives is 100 percent.

On the positive side we were able to adapt. One could argue some were late to need, but the changes we made in intelligence gathering, assessment analysis, exploitation, and dissemination were important. When I visited a combat outpost on the Pakistani border in 2008 as the CENTCOM commander, Captain McChrystal, Stan’s [Stanley McChrystal’s] nephew, was in command. The captain had more access to national technical means and all kinds of intelligence in 2008 than I did as division commander in 2003, and that’s not hyperbole. So we did make a lot of great adaptations to all of the battlefield functions, whether fires and maneuver or command and control, and we began to describe it as mission command. We decentralized, we began to empower the edge, and we began to develop the leaders who could work, seize, and execute initiatives. We began to improve intelligence functions and logistics. We learned a lot about contractors on the battlefield, some good and some bad. But we made a lot of incredible adjustments over time.

Let me finish by going back to the somewhat negative side. Architectures in organizations begin to develop a momentum of their own, and it becomes difficult to disassemble them. The architectures themselves become self-fulfilling. I didn’t think we were ever going to get out of Baghdad with all of the architecture—intelligence, logistical, command and control—we had built there.

Moreover, we probably retained a little too much control for a little bit too long. We probably didn’t make our relationship with former Iraqi Prime Minister [Nouri] al-Maliki as transactional and conditional as it should have been. As a result, we began, toward the end of the campaign, to be talking past one another. So that’s kind of the front end and the back end.

On the MNSTC-I side, which is right in the middle for me from 2005 to 2007, I know some of your questions relate to a particular one: Can we actually build and develop indigenous forces to take control of their own country? Here is where I find myself today on this question. If we take ownership in every sense of the word, which we did in the early days both in Iraq and Afghanistan, and then try to begin to build an indigenous force in an institutional design to control it—that is to say not only tactical-level fighters but also the logistics architecture, intelligence architecture, school systems, and the ministries—that’s far more difficult than making the indigenous force own it from the beginning with our enabling it.

So you might ask, what would you do differently. First of all, I would have absolutely not disbanded the Iraqi army, and I would have absolutely not de-Ba’athified. We lost all of the bureaucrats who knew how to run the country. And I would have, in a transactional and conditional way, made it clear how we would help the Iraqis regain control of their own country, put it back on its feet. But there would have been no doubt from the start that it would be their responsibility and not ours.

The enduring lesson about MNSTC-I is this: The art of campaigning and building a foreign military is establishing ownership and managing that from the start. If you take too much ownership too soon, it is almost impossible to give it back.

Sheik Abdullah Sami Obeidi, a Sunni Arab tribal leader, signs declaration of support for Sons of Iraq program as U.S. Army Colonel David Paschel, commander of 1st Brigade Combat Team, 10th Mountain Division, looks on in Al Noor, Iraq, March 9, 2008 (DOD/Samuel Bendet)

Sheik Abdullah Sami Obeidi, a Sunni Arab tribal leader, signs declaration of support for Sons of Iraq program as U.S. Army Colonel David Paschel, commander of 1st Brigade Combat Team, 10th Mountain Division, looks on in Al Noor, Iraq, March 9, 2008 (DOD/Samuel Bendet)

R.D. Hooker, Jr.: I interviewed a Combined Security Transition Command–Afghanistan deputy commander who believes that we tried too hard in both Iraq and Afghanistan to make those militaries like our own. Do you agree or disagree?

General Dempsey: I have thought about that a lot. Early on that was indeed a valid criticism. I remember going to visit Bernard Kerik, who was the senior Coalition Provincial Authority [CPA] police trainer. Kerik was passionate about not wanting the military involved in the training and only wanted occasional support with resupply as we conducted patrols in Baghdad, thus assuring the police stations were getting what they needed. He was training them to be beat cops, traffic circle cops, training them in law enforcement techniques. Then they graduated, went out into the streets of Baghdad, and were slaughtered.

Kerik left and the next guy that came in—I can’t remember his name—I went to him and said, “Look, this can’t be a competition, but I’m telling you the police you are producing are not going to stand and fight this insurgency because they’re underarmed, they don’t have protection around their stations, [and] they’re getting slaughtered in the streets.” So we forged a partnership. [Years later] I came back as commander of MNSTC-I. The next guy who came in was actually open-minded about having the police effort subordinated to MNSTC-I. It was on my watch that we gained oversight not only of the army but also of the police, and we were able to harmonize the efforts. But to your point, there’s no question that early on we were trying to create these forces in our image. I don’t regret that actually because we probably had to see if that was possible before we adapted.

We also had coalition partners that would take sectors of Iraq. The boon and bane of a coalition, as you know, is that it is a coalition—so everyone gets a voice. The boon is they’re there, and you get 26, 28, or 45 flags. But there’s no doubt in my mind, I can give you chapter and verse, that the way the British were developing the security forces in Basra was different than the Poles were developing security forces, and it was different than the way the [U.S.] Army was developing security forces in Diyala Province, different than the way the [U.S.] Marines were developing security forces in Al Anbar. Even in our own Service we had different approaches, a different way of partnering. Now is that a strength or a weakness? Initially it was a weakness because we were a little inconsistent. I think over time, however, we were able to harmonize that.

I remember visiting a country—I won’t mention which country but it wasn’t ours—and I went to its training center for the Iraqi security forces, and the trainers had a [significant amount of] instruction on drill and ceremonies. You see, the Iraqis loved to march—I mean they loved to march. But it wasn’t doing them much good to keep them alive. But because they loved to march so much, and they were well behaved when marching, this particular partner was spending a lot of time teaching them how to march.

Hooker: Many sources, including the recent RAND study by Linda Robinson, have discussed the tension between civilian decisionmakers and their military advisors in making wartime decisions, particularly in the formation of objectives and the development of strategic options. What has been your experience, and what is your advice to pass on to successors?

General Dempsey: I think the system is actually designed to create that friction in decisionmaking. Our entire system is built on the premise that we require friction to move [forward]. Physics even says that. You have to have friction before the wheels on a car make contact with the road and propel it forward. So our system is designed to create a certain amount of friction, and it succeeds. There are always [institutional] equities, or the objectives as articulated by the Department of State and USAID [U.S. Agency for International Development]. One of the debates in Iraq early on was which comes first, economic development or security. It was a chicken-and-egg argument. Those were heated debates about whether we should lock down the country and then kind of loosen the reins on it and do economic development in a secure environment, or whether we should invest mightily into transitioning state-owned enterprises into private-owned enterprises. I can remember really serious, important, and constructive debates about that dichotomy. It was a false dichotomy, but it was presented as a dichotomy nonetheless.

First, I would advise future leaders that friction and disagreement in decisionmaking is not a negative. Frankly, you should embrace friction. What I found was, and I can’t put a percentage on it, but in general the person at the table with the most persuasive argument tends to prevail in those environments.

Let me segue to an important factor. There is an article, I don’t know who wrote it, but it was written in 2013, and it focuses on the uncanny ability of military and political leaders or elected officials to talk past each other. In the military culture, as you know, we spend decades learning how to do campaign planning, and we start with a well-stated and clear objective. Then we build a campaign to achieve that objective, with intermediate objectives and milestones along the way. Then we come up with three courses of action: high risk, medium risk, and low risk. We pick the middle-risk option and execute. If you are an elected official, the likelihood of your conceiving a well-crafted and well-defined objective at the beginning is almost zero. Rather, as an elected official, your first instinct is to seek to understand what options you have.

So militarily I know I’ve got it, I have a nuclear option, but let’s just park that for a moment. What other options do I have in this magnificent toolbox called the U.S. military? What tools do I have that I can apply pressure with, that I can manage escalation with, and that I can integrate with the other instruments of national power? Elected officials are hardwired to ask for options first and then reverse-engineer objectives. And the military is hard-wired to do exactly the opposite.

Now what do we do about that situation? Nothing frankly. But that is the environment that we live and work in. I learned that pretty early on. I learned it by reading [Bob Woodward’s] Obama’s Wars [Simon & Schuster, 2011]. I read it not to get inside information on the intrigue or the kiss-and-tell aspect, but I wanted to try and understand why Woodward was able to find the seam between the advice that was given to the President and his willingness to accept that advice. And it came down to what I just described: it wasn’t articulated that way in the book, but that’s what I drew from the book. When you read a book, the author wants you to take what you want to take from it, and not necessarily what he is trying to give you. But I’ve decided that we’re just hardwired differently. Knowing that, I think it’s incumbent on us to work inside that culture and not to rebel against it. [This is a factor] in my relationship with the President, in my relationship with the JCS [Joint Chiefs of Staff], and it informs my relationship with the COCOMs [combatant commands] as I try to manage demand and supply. It has been quite helpful to me.

Getting back to the question, my advice to my successors is get to know how our government functions. Don’t come to Washington thinking you’re going to get Washington to conform to your beliefs because that is generally never going to happen. You have to have a moral compass, but you have to understand the way people in this city make decisions. Also, you must understand that most big decisions are made in conjunction with budget cycles, not in conjunction with current events. If you want to change something in our system of government, you change it in the budget. Can you do things in between budgets cycles? Of course you can; we built in a certain amount of flux, but big changes are [usually] made in budget cycles, and that includes big changes in campaigns.

Collins: If we could just follow up on that. You talked about the surge process in Afghanistan, and of course there was a surge process in Iraq. Any reflections on things that you have in your own personal knowledge or that you’ve learned from those particular cases in terms of decisionmaking?

General Dempsey: To thread that, or to link that back to the question about the RAND study and whether friction is a negative or a positive, the way that Multi-National Force–Iraq [MNF-I] was constructed was [that it would be led by] a strategic 4-star—[General] George Casey at the time, later David Petraeus—and two 3-stars. The two 3-stars were the Multi-National Corps–Iraq [MNC-I] commander, and me as the MNSTC-I commander. Both 3-stars had equal access and equal voice to the strategic command. MNC-I measured his success on levels of violence, but the MNSTC-I commander measured his success on the development of the Iraqi security forces. When the question of the surge came up, the advice of the MNC-I commander, not surprisingly, was that in order to drive down violence, he needed five brigades. (By the way, I may be off by a couple. Initially it was only two brigades, then eventually it went to five brigades.) And my advice was that we probably should knock violence down, but let’s be careful on how we do it because we could give the Iraqis the idea that every time violence spikes, we would rush in and retake control of things. We could be actually setting back the development of the Iraqi security forces. Or, stated in another way, I said, “Look, we have two options here, General Casey. You can double down on [U.S.] activities and you will probably knock the violence down pretty quickly, or you can double down on the development of the Iraqi security forces. In other words, embed at greater numbers, enable at greater numbers, but actually make them responsible for pushing the surge and bringing the spike in violence down. And my advice is the latter: we have said that our exit strategy here runs through the Iraqi security forces. So if you want my advice as the MNSTC-I commander, I think we ought to double down on the Iraqis and not double down on ourselves.”

That is exactly how the conversation went. Somehow along the way I’ve been painted with the brush of being anti-surge. I was never anti-surge. My question was simply who was going to surge. And my advice as the MNSTC-I commander was that the surge ought to be carried out by the Iraqis. It is debatable whether they could have pulled it off, but we had two separate 3-star commands in Iraq for that purpose. The decision was taken to dial down on our efforts, and I saluted, and we executed. Did it work? It did actually; it knocked the level of violence down, and the surge gave decision space for the Iraqi government, but it failed to take advantage of that space. One might make the case that they failed to take advantage because we had sent the message that if they get into trouble, we will rescue them. And I believe that, too; if you’re trying to restore stability to a failed state, do you do it or do they do it? And the surge sent a signal that if something really went badly, we would take control of it, and then we would give them another chance. The other way to do it would have been violent; it would have taken longer. I’m not suggesting I was right and they were wrong, but I think I was there to give exactly that advice. And I gave it.

The other way of considering the surge as the right course of action is to look at the transactional and conditional nature of relationships, especially in that part of the world. What actually made the surge work? Again, this is debatable, but in my judgment, what made the surge work was less about the introduction of additional U.S. forces and more about the fact that we co-opted the Sunni tribes by paying them and arming them on the promise that the Iraqi government would absorb them into their security forces. Well, okay. It didn’t happen. And because it didn’t happen, the loyalty of the Sunni tribes went to us and not to the Iraqi government. Once we took the other decision to stop paying them and stop supporting them, and they didn’t have a safety net in the Iraqi government, I think we are where we are today somewhat as a result of that. But that’s controversial.

I do think the structure of MNF-I was designed so that the strategic command would get advice on both sides of the equation, which is how much should we do and how much should they do. It was my responsibility to argue for what they should do. I made the case, and the decision went the other way. History will decide if this was correct.

Collins: The other question was the Afghanistan surge. You touched on that with your mention of [Woodward’s] Obama’s Wars. In crafting options for those situations, should the most senior military people address all the options, or only the options they think are the ones that are going to work?

General Dempsey: The one thing that has to be clear: every option in military doctrine has to be suitable, feasible, and acceptable. I could never conceive of a circumstance where I would either recommend or, if asked, support an option that I didn’t find to be feasible, acceptable, and suitable. But with that said, in particular because of what I’ve said earlier, I want to make sure this is not lost because I have been giving a lot of thought to this. In the use of the military instrument of power against state actors, we differentiate ourselves by size and technology. We are bigger, badder, our tanks shoot further, penetrate more deeply, and can operate at night in a way that our adversary cannot. So we overmatch with size and technology in state conflict. Training, good leadership, and better logistics system—all of these are important.

When you talk about conflict against nonstate actors, and that is really what we are talking about here, we were fighting an insurgency on behalf of a government. We were fighting an insurgency on behalf of Iraq and an insurgency on behalf of Afghanistan, simultaneously trying to restore their abilities to govern. In that kind of conflict, the use of military [forces] against nonstate actors, I think size and technology matter, but what matters more is the rate at which we innovate. The rate of innovation becomes a better predictor of success than the Force Management Level, for example. Size matters, but the rate at which we can innovate, adapt, and respond to changes in the environment matters more.

In that context, this is where I answer your question. The options are far broader in conflicts with nonstate actors because decisions are temporal in a way. If I am right about the need to adapt more frequently, then the last thing we want to do is flop in there with 150,000 [personnel], 12 mega-forward operating bases, [and then] begin to funnel in TGI Friday’s and Baskin-Robbins.

When I look back, conflict against nonstate actors does not lend itself to industrial-strength solutions. And I’m not sure exactly what I would have done differently, but I would have been far more expeditionary, far more austere, and far more attuned to the need to [innovate and adapt] than negotiating the Force Management Levels. For example, in Afghanistan we did surge, and that one ultimately may have had a better effect than the one in Iraq, but even in conducting that surge, we surged traditionally with BCTs [Brigade Combat Teams]. We took BCTs and surged for 12 consecutive cycles. By so doing, the industrial machine began to crank, and we started to build big FOBs [forward operating bases]—and big FOBs increase demand, demand increases requirement for money, et cetera. There is probably a way to redefine surge, but we looked at it through the lens of Force Management Levels. I wasn’t in the system at the time. I was the TRADOC commander, but the President was told: “Look, it’s 40,000 or nothing; 40,000 or let’s get out.” That is how it was portrayed. Is that right, though, is that really true, 40,000 or let’s get out?

So we have to be a little less dogmatic in conflict against nonstate actors than we are in conflict with state actors. When we are in conflict with a state actor, it tends to be more existential, it tends to be a little clearer on how you differentiate yourself, and therefore I think the options become a little crisper. I don’t find the options to be that crisp in this kind of conflict, and therefore we have to be more thoughtful and more open to negotiating them, remembering that we have to have a moral compass.

And by the way, I have one tenet that I generally rely on in making recommendations in places like Iraq and Afghanistan, and here it is: A squad’s work for a squad. If you want me to do X, here is what I think I need to do. If you think I need to do it for less, then I am going to do less. My military advice is what you can accomplish with a squad, what can you accomplish with a battalion, what can you accomplish with a brigade, and we will not ask a brigade to do a division’s worth of work. That is it, and we have had some success in discussions that are built on that principle.

Hooker: Historians are going to wrestle with whether the outcomes of the Iraq and Afghanistan campaigns were fundamentally ascribable to the military effort or civilian effort. There is a narrative that asserts the military was asked in both conflicts, at least once we got into the counterinsurgency game, to secure the population, and it did that fairly well. The military was able to build up large numbers of host nation military units that took over the transition. But the failure of the campaign was the inability of the host-nations, both in terms of the capacity and in terms of rule of law, to carry their loads. That was the vulnerability we were never able to overcome. Do you see it that way?

General Dempsey: Remember earlier when I said that in conflict against nonstate actors in failed states or failing states, I have come to believe that support needs to be transactional and conditional. I believe that because, generally speaking, in these failing and failed states the issues are societal—they are not political issues. Sometimes they begin as political issues, or they’ll start as representational—for instance, the fruit vendor in Tunisia self-immolating because the government wanted to tax his fruit stand. It starts political, but it goes pretty quickly to sectarian issues, to religion, and ethnicity because these are historic impulses that have been suppressed for generations. In those environments, it’s absolutely predictable that the “victor and vanquished” mentality will quickly come forward. Those who have been suppressed will see themselves as victors, and they will come and vanquish those oppressing them, and I think whether we are asked to conduct military operations in Iraq, Afghanistan, Libya, Syria, [or] Nigeria, that “victor-vanquished” instinct is the dominant societal instinct. If I’m right about this, then there can be no unconditional support, in my opinion, because unconditional support will simply reinforce the “victor-vanquished” paradigm as it emerges.

So let’s fast-forward to Iraq today. Some people are saying, “Why aren’t you doing more, and sooner?” Our support needs to remain as support and not ownership. Furthermore, support needs to be conditional. If the Iraqi government does not meet its commitments to create a more inclusive political environment and to address some of the grievances of the Sunni and Kurd populations, then nothing we do will last. It will be painting over rust. We have eight lines of effort, two of which are military, and generally the military lines of effort leap out in front—and I do mean leap. That is who we are, right? If it is worth doing, it is worth overdoing. The military lines of effort will always be achieved. And that can be detrimental to the other lines of effort. I don’t know if that answers your question, but it is why I believe now that the use of the military instrument of power in issues of nonstate actors and failed states needs to be far more conditional and transactional than anything we do with state actors.

One more thing: [U.S. interests must lead.] The tragedy of human suffering and the situation in Syria is awful, but I will also tell you the use of the U.S. military instrument of power without consideration of what I just described can actually create more harm and even further suffering, I think.

Collins: We have also had some grand failures in intelligence, in particular in the war in Iraq. One criticism, that in particular of General Michael Flynn, is that we are here fighting among people and we do not know much about them, and intelligence is not focused on that problem. How do you see intelligence functioning and its level of proficiency both operationally and strategically in Iraq and Afghanistan?

General Dempsey: The Intelligence Community was slow in adapting to what really mattered in the environments we found ourselves in. Back to the difference in state actors and nonstate actors: if I’m right about the fact that you differentiate yourself in a state conflict by size and technology, then the intelligence architecture is going to build itself in such a way to determine where capabilities are placing you at a disadvantage. In an environment with nonstate actors, where it is all about innovation, then you have to understand the factors that would cause you to need to innovate, and they largely reside in societal factors. You try to drive the insurgency or the terrorist group from the population. Mike [Flynn] was right; we could list the deck of cards or the wiring diagram of any number of organizations and networks in Iraq and Afghanistan, but if we were to ask a commander on the ground in Afghanistan to tell us something about this particular tribe in this particular valley and who are its affiliates, that was often discovery learning. Every time we had to RIP [relief in place] out a unit, it was discovery learning again, so we fell into a bit of the 12 1-year campaigns instead of one 12-year campaign. Mike’s article actually helped a lot with that, and we had done some things with TRADOC, with the HTTs [Human Terrain Teams], not without controversy by the way.

My TRADOC G2, a guy named Maxie McFarland, who passed away recently, was instrumental in developing and fielding the HTTs. It was his brainchild to reach out to academia, to anthropology, to form these teams and to offer them to BCTs. We would try to keep them there so the HTT would stay in the last 6 months of a brigade, and the first 6 months of a new brigade, so there was continuity. And they paid big dividends. We got this, and it was controversial still, because of the notion that we were perverting science, using science to the detriment of culture rather than to the benefit of it. But it was addressing the question you asked: how do you learn about the environment? And that is one answer.

This is in the spirit of learning lessons and not throwing stones. It took a while for the Intelligence Community to adapt to help us—that is to say the tactical commander to understand the environments—but there was progress. Now the question is whether can we sustain it. Or is the institution likely to forget that the understanding of culture, religion, and economics of a local society is important? I hope not, and with all the chiefs we seem to be committed to making sure that we don’t forget those lessons, but often the institution will. It is like a rubber band; you stretch it and then you let it go, and it will go back to its normal form or shape. I’m afraid some of that might occur, if we are not careful.

My successor will face state and nonstate challenges in about equal measures. One thing Jim Baker [Principal Deputy Director, Strategic Plans and Policy, J5] helped me think through [is] the meaning of the reemergence of Russia. I was feeling kind of constrained by Russia, and the President asked me, “Can I meet my [North Atlantic Treaty Organization] Article 5 responsibilities?” And I replied, “Mr. President, that is a great question, so let me get back to you.” I was feeling uncomfortable about the ability of our forces to use forward basing because in the last 10 years the Russians have developed some capabilities that actually could coerce and constrain us.

Jim pointed out that the world looked similar to that of the early years of my career. And there was truth in that. During the first 15 years of my career, 1974 to 1989, Russia was a near-peer competitor, not just nuclear but also conventionally. We were constrained, and our military planning took into account the fact that we were constrained in military operations by a near-peer competitor. We didn’t like it, but we learned how to deal with it. The next 20 years, 1989 to 2009, had no constraints. So most of the officer corps today lived in a world where they were unconstrained. No one could prevent them from doing anything they wanted to do. But guess what? We are back to what is probably normal, I think, in the course of recent events—that is, to where you have near-peer competitors in certain domains, and then you have to account for this in military planning.

So my successor will have to deal with the reality of state actors who can now coerce and constrain us, as well as nonstate actors. So, to your point, I don’t think the pendulum will swing entirely back to Russia or China as peer competitors, but I think the institution will have to adapt to have aspects of both in them.

Hooker: How should future senior officers who are combatant commanders or the Chairman view their role in the highest councils of government? Are they there to provide the best military advice only, or are they there, as Clausewitz noted, to be both the statesman and general?

General Dempsey: When you become a senior military leader, you have multiple responsibilities, one of which is to give the best military advice possible, and another is to help the force. But there is a third one, too. I have the responsibility to contribute to foreign relations strategy as a statutory advisor of the National Security Council. In my early days, we would go around the room, and the staff would be talking about something I didn’t want to talk about. Pick a topic. Whatever it was, as it came around the room for me, I would say, “I am here as your military advisor; that is not a military issue.” And the President would say, “Yes, but you are here, and I want your view on this strategic issue that has national security implications.”

If you are going to understand how decisions are made in our government, you must build relationships, and if you’re going to build relationships, you have to demonstrate a certain gravitas. You’ve got to be able to have a conversation about grand strategy, not just military strategy. If I had to give advice to my successors about job number one in terms of being influential inside decisionmaking boardrooms, it would be that relationships matter most of all. If you can’t develop a relationship of trust and credibility—credibility first and trust second, because trust is earned—then you won’t be successful in contributing to our national security strategy.

F/A-18C Hornet assigned to “Blue Diamonds” of Strike Fighter Squadron VFA-146 launches from USS Ronald Reagan to conduct close-air support missions in June 2011 as part of Operation Enduring Freedom (U.S. Navy/Alexander Tidd)

F/A-18C Hornet assigned to “Blue Diamonds” of Strike Fighter Squadron VFA-146 launches from USS Ronald Reagan to conduct close-air support missions in June 2011 as part of Operation Enduring Freedom (U.S. Navy/Alexander Tidd)

Collins: You come down almost exactly at the same point JFK did after the Bay of Pigs. He wrote instructions to the Joint Chiefs that said very much what you just said in the last 2 or 3 minutes. We have had a number of issues having to do with detainees’ enhanced interrogation. Some of those shoes have not dropped yet for the Department of Defense, military commissions, and so forth. Were these problems inevitable, or did we get off on the wrong foot? If 10 years from now we have another situation akin to the situations in Afghanistan and Iraq, what would you tell your successor about the lessons and how we did it in these two cases?

General Dempsey: The detention operations have to be included in any campaign plan that includes the use of military force because we can’t ever put a young man or woman in the position where there’s no possibility of detention. The alternative is capture and release, or kill not capture.

I think this is what happened in these conflicts regarding detention operations. You know what they say about campaign assumptions: if the assumptions are flawed or invalid, the campaign has to be adapted. That’s why you make assumptions about campaigns. So one of the assumptions I think we made, again I was in Riyadh when all this was being developed, but I think one of the campaign assumptions, probably driven more by political aspirations than the reality, was that [we would] go into Iraq and we would be welcomed because we would be seen as liberators, and we could take as much risk getting out as we took getting in. One of the risks we took going in was that we went in with fewer forces than the commander thought he needed to accomplish the task. Fewer forces mean fewer capabilities. We didn’t have the number of MPs [military police] that we probably needed to account for detention operations because we didn’t think that we’d be detaining enemy personnel. Or if we were detaining, we would be turning them over to I don’t know whom, but the assumptions were flawed. So yes, we got off on the wrong foot, but we also hadn’t done detainee operations since 1991.

But if you remember in 1991, the 96 Hour War, I can remember as part of VII Corps accumulating large numbers of Iraqi soldiers surrendering, and we pulled them down into Saudi Arabia into temporary camps. But I think we repatriated them within weeks, not months. And so if you go back to the time before when we did detainee operations, you have to go all the way back to Vietnam, so there was a lot of rust on that function. If there is a lesson here, it’s the lesson that comes to us instinctively, which is to address the worst-case scenario. We always do that, but we’re talked out of it sometimes, and I think in the case of detainee operations in future conflict, we shouldn’t allow ourselves to get talked out of that function.

Collins: Are enhanced interrogation techniques a bigger issue for the Central Intelligence Agency or the Department of Defense?

General Dempsey: Our issue was an initial lack of doctrine. Then we had the terrible incident in Abu Ghraib, and then the Army republished its doctrine and from that point forward, we had no further problems.

So linking it back to your last question, the key here is to continue to refresh our doctrine in order to manage the functional area. As we shrink the force, we have to be careful not to eliminate that capability. A lesson of this conflict will be that leaders need to be involved. Again, I was in Baghdad, and the only thing I controlled in Abu Ghraib was external security, but I’m pretty sure I’m correct in saying that we just turned it over to the MPs to manage without proper oversight. So leadership matters whether you’re talking about combat operations, detention operations, or intelligence operations. Remember, I described that scenario in June 2003 when we actually didn’t settle in on a definition of the enemy, an organizational principle to design against it, and a campaign that acknowledged that this was going to take some time. Even in August 2003 we were talking about the possibility of being home by Christmas. So we didn’t grasp the fact that this was going to be a protracted campaign until October. So think about the time between March and October—that’s 6 months. So there’s a 6-month period of indecision there, and that’s where some of these bad habits, worse than bad habits, this misconduct began to manifest itself. There was a list of enhanced interrogation techniques, but Abu Ghraib was clearly not a problem of enhanced interrogation—it was misconduct.

Collins: General [Daniel P.] Bolger states clearly that the wars in Iraq and Afghanistan have been lost. In fact, that is the title of his book [Why We Lost: A General’s Inside Account of the Iraq and Afghanistan Wars (Houghton Mifflin Harcourt, 2014)]. And senior military leaders bear much of the responsibility. He suggests that we should have left Afghanistan and Iraq somewhere within the first 6 months. Would such a thing have been possible?

General Dempsey: No, that’s not who we are as a nation. I wish things were that simple. Imagine being able to just go in and crush that which you can find, declare that you’ve accomplished your task, and care nothing about what you leave in your wake and withdraw. But that’s not the American way of war, not to sound too much like Russell Weigley [Distinguished University Professor of History at Temple University and noted military historian], but that’s not the American way of war. The American way of war tends to be that—out of a sense of not only obligation and responsibility to protect, although that is not really doctrine, but also compassion—we assist those who have been defeated to reestablish themselves in a more moderate and inclusive way. As far as whether we made mistakes, I take no exception to that, but I consider it more about learning than about negligence. And I think as we learned, we changed. Now to Bolger’s point about whether we stayed there too long, 6 months certainly was not possible. If he would have said in the book that we should have had an idea as to how long we were willing to make this commitment from the start and that should have informed our thinking about how to organize the campaign, I accept that, but 6 months is absurd.

Hooker: Can you compare your experience working for three different Secretaries of Defense? Do you have any thoughts on varying styles of civil-military negotiations from one to the other?

General Dempsey: All very different. As Chief of Staff of the Army and at CENTCOM, I had the opportunity to work with Secretary [Robert] Gates. So really I’ve had some close relationships with Gates, [Leon] Panetta, [Chuck] Hagel, and [Ash] Carter. First of all, I don’t think I’ve changed who I am to adapt to them, but I have adapted the way I interact with them and that’s probably an important distinction. I’ll give you some examples.

Secretary Gates was a voracious reader and a very close reader. You could give him a read-ahead document and you could expect that when you engage with him, he would have some close and crisp questions. He let the written word inform him, and so when you engaged with him one on one or in a meeting, it tended to start at a higher level.

Secretary Panetta was a man of uncanny instinct. Even before reading something or discussing it, he had been around so long and had had so many experiences inside of government—whether in the White House as Chief of Staff, in Congress, or as the Director of the CIA—that he learned less by reading and more by interacting. He also believed deeply in relationships. So if you were able to forge a relationship with him and you gained his trust, it made the interactions extraordinarily collegial.

So going from one to the other—understanding that the written word was important—I focused a lot on, especially when I was at CENTCOM, I wrote my own weekly reports, organized them and selected [the precise] words, and I managed the length of report in a way that I knew would match Secretary Gates’s way of learning. With Secretary Panetta, I’m not sure; I may have given him maybe three documents in 18 months or in 2 years. In any case, he probably knew what I was writing before I wrote it, and what he really wanted was to engage me on it. So we had a very close relationship built mostly around the time we spent in his office.

Secretary Hagel also comes from a background of long government service, whether it was as President of the USO, Deputy Director of the VA [Veteran’s Administration], an academic at Georgetown University, the Senate, and Secretary of Defense. He has […] a greater instinct not for detail but for the theory of the case. He likes to understand not only the tactical question, but also how that question fits into a broader frame. If Secretary Panetta was the quintessential extrovert, Secretary Hagel was kind of the quintessential introvert. That doesn’t mean he’s without humor. He’s pleasant, he’s engaging, he’s compassionate, and extraordinarily connected to soldiers. Not only soldiers [but also] the lower ranking enlisted [of all the Services] expressed their deep disappointment that he’s leaving. Somehow he actually found a way to make a connection with the sergeants, petty officers, Airmen, and ensigns that was quite remarkable. He is one of them. He doesn’t care for detail, and he doesn’t care for big groups, whereas with Panetta, you couldn’t put enough people in the room for him because he could just own it, honestly. He was a remarkable facilitator of huge audiences. Secretary Hagel was much more comfortable with smaller groups. He also likes to read, and so he’s kind of a hybrid of Gates and Panetta. Secretary Hagel is a one-on-one guy. He does his best thinking, his best work, and his best interaction one on one. So back to the question. I think I’m still the same person I was 3 years ago, but I’m a little savvier. I’ve adapted the way I interact with these leaders based on the way they learn, and you have to figure that out.

B-1B Lancer aircraft drop six GBU-38 munitions onto insurgent torture house and prison in Northern Zambraniyah, Iraq, March 2008 (U.S. Air Force/Andy Dunaway)

B-1B Lancer aircraft drop six GBU-38 munitions onto insurgent torture house and prison in Northern Zambraniyah, Iraq, March 2008 (U.S. Air Force/Andy Dunaway)

Collins: In our crowdsourcing of strategic lessons of the wars, a number of folks are saying that when we look back through history on a grand scale, foreign expeditionary forces in counterinsurgency operations are successful only in rare circumstances. The British were successful in Malaya, but then again the British were the government, so there was no sanctuary. The United States was successful in the Philippines in 1902, but again it was the government, and again there was no sanctuary. But other than those two cases, many experts claim that this cannot be done. We have now been involved in Iraq, Afghanistan, and Vietnam. How should your successors think about this in the future? Is this the sort of mission we have to still prepare for, or is this something that is beyond the pale?

General Dempsey: If you mean by “beyond the pale” large-scale intervention against nonstate actors or insurgents and failing states, let me use Iraq as an example.

Today, I’m at bat for the third time in Iraq. I was literally in Iraq sitting on what we later called Route Tampa in the aftermath of the 96 Hour War, blocking that major highway that runs from Kuwait to Baghdad. I sat there for a couple of months. You remember the debate about whether we should have pursued the Republican Guard and end the event right then and there, or whether we should have settled for the narrowly defined objectives in the [United Nations] Security Council resolutions and so forth.

The 96 Hour War accomplished the objectives, and as a result, we ended up leaving for the first time in the central region a large force that was there both as a deterrent and a reassurance for our allies. Prior to 1991, we had a permanent naval base in Bahrain, but I don’t think we had a permanent land base (we may have had access to some air bases). We weren’t in Qatar and we weren’t in [the United Arab Emirates]. One might argue that although the Gulf War looks to have been a lot cleaner, it did result in a requirement to place a pretty significant footprint there that has put a strain on the force ever since.

Then we go from 2003 to 2011, and I think we’ve wrung that out as much as we can, but here’s what I think is different this time when we talk about our reentry into Iraq. I think we’ve got it about right, which is to say, we’ve made it clear that we will support and enable, that we’ll keep the eight lines of effort apace, even if some of them will get a little ahead of each other on occasion. But we’re not going to take ownership of Iraq again. And I think that we can accomplish this task with a light footprint and the use of some of our key enablers, but we’ve got to have resolve and courage. If the government of Iraq proves to be incapable or unwilling, we’ve got to be willing to dial it back. In other words, it has to be conditional; it just can’t be unconditional in this kind of environment because we have other options to deal with the terrorist threat, but Iraq doesn’t have any other options to deal with this insurgent threat.

I would like to tell you that large-scale intervention during insurgencies will be a thing of the past, but we have to retain the capability. That’s why we’ve established building partner capacity as a core competence of the entire force, not just special operators. Security force assistance as part of theater campaign plans is prominent in the Phase 0 side of operations, but I don’t think we should size the force to counterinsurgency; we should size the force for treaty obligations against state actors and then retain enough slack in the system so that we can ensure readiness.

We’re getting ready to enter a huge debate about the correct balance of forward stationing, rotational deployments, and readiness of standing forces. Right now, the model we have produces readiness and deploys contingent upon combatant commander demand signals. That’s especially true in the Air Force, the Navy somewhat, pretty much true in the Marine Corps, and except for a handful of brigades, pretty much true in the Army. So we have everything distributed globally, but if there were some major contingency or if there were something that would surprise us that would exceed the capacity of a particular COCOM to deal with it, the only place to get forces and readiness would be from some other COCOM. There’s almost nothing stationed in CONUS [the continental United States] that is unused capability or readiness, so we’ve got to go back and address that.

Prior to 1991, we were a lot bigger—781,000 in the Army alone. We would have a fraction of the force forward deployed, and we would have these big Reforger exercises, for example. The forces would be stationed mostly inside CONUS, and the idea was that these forces would be at various degrees of readiness, but more or less ready in CONUS for deployment into contingency operations and the forward presence part of it wasn’t the priority. The priority was the readiness part of it.

Since 1991, the paradigm’s reversed. The priority now is forward presence to include rotational presence, thereafter security cooperation, Phase 0, Phase 1—that’s the priority now. Phase 3, combat, we’re taking risk, frankly, because we’ve got much less than we probably should have in readiness in CONUS.

I’m not suggesting we’re going to flip it again. Some would argue that we should flip the paradigm back to where we prioritize surge capacity and readiness as the primary effort. I don’t think we’ll do that, but I think you’ll see us try to rebalance it.

The part of the force that tends to be forward is the part that is most capable of doing the kind of things you’re talking about in terms of counterinsurgency, counterterrorism, building partner capacity, and so forth. As we change the paradigm, we have to figure out a way to make sure that the training that supports the new paradigm accounts for both maneuver warfare and low-intensity conflict.

I think my successor will probably have to refresh the military lexicon a bit. You said before that Bolger said, “We lost Iraq and Afghanistan.” This statement implies we didn’t win in Iraq and Afghanistan. Yet one of my premises is that the definition [of winning] is frequently redefined in [the current] environment. Saying categorically “win” or “lose” seems to be far less applicable to those kinds of conflicts than it is in a high-intensity conflict with a peer competitor that is trying either to take your territory or deny your freedom of movement.

Hooker: Is it a mistake to think in terms of war termination, or are we in the middle of an ongoing conflict, maybe less than a war, but a conflict that is unlikely to end anytime soon and that we ought to adapt to?

General Dempsey: You remember back in the early part of the last decade the phrase “the Long War.” That phrase attracted antibodies of all kinds […] fiscal antibodies, political antibodies, and intellectual antibodies. Then George Casey kind of led the charge on use of the phrase “persistent conflict.” And his view, if you don’t want to concede that this is actually a war in the strictest or loosest definition of the word, you should at least accept the fact that we’re going to be in persistent conflict. Of course, eventually that fizzled as well. I don’t remember exactly why that one fizzled, but it hasn’t actually been replaced. If we have indeed ended the wars in Iraq and Afghanistan, what do we have? We still have Soldiers and Sailors and Marines deploying in harm’s way. If you’re a pilot flying over Iraq, the distinction between combat advising and no boots on the ground is meaningless.

We’ve had some success, but there’s work to be done on acknowledging and understanding what runs from Western Pakistan to Nigeria: a group of organizations that sometimes work with each other and sometimes operate independently, depending on their objectives, that are trying to take advantage of lack of governance almost everywhere, that are playing to this victimization psychology, and that use the tactics of terror. And there is a fine line here: are these people terrorists, or are these people using the tactics of terror? What we’ve discussed, both inside the military and with the administration and Congress, is that this threat [. . .] of violent extremist organizations—most of which also happen to be radical Islamic organizations—we as a nation just haven’t had a conversation about that. I’ve been accused of being anti-Islam by some and pro-Islam by others. I guess in that sense I’m succeeding in managing the conversation. But the point is there are violent extremist organizations that are using a religious ideology to brand themselves and to gain support from disenfranchised populations, both Sunni and Shia.

As long as this conflict persists, every 10 years or so a new generation will be sucked in. And until this cycle is broken—and that cycle is likely not to be broken exclusively and not even primarily with military force—the despair, lack of hope, lack of inclusive governance, and grotesque economic disparity will continue, and the U.S. military will be called upon to have a role in addressing it.

How we define that role is to be determined. Right now, we’re defining it one country and one group at a time. In fact, if you look at the way our country plans are written for the counterterrorism, if you look at the way the State Department organizes itself and interacts with us, interacts with the combatant commanders, it is one group, one country at a time. But it’s a common threat. We have not successfully helped our elected officials address this threat in its totality, and until we do and until we can actually find the right vocabulary, I think we will continue to be effective at containing the threat and to the greatest extent possible and keeping it from our shores, but we will not be effective at ultimately defeating the threat until we capture the right framework, which is actually transregional, and until we capture the right vocabulary. That’s not to suggest, by the way, I think that absolutely the wrong thing would be to agree that it’s transregional and find the right vocabulary, and then decide that we’re going to invest enormous military resources to stabilize all of these countries and put them back on a firm footing for their future. Because that won’t happen. They will allow us to do that—you know many of them will. We’ll be embraced initially, then disdained and attacked ultimately by the very people that we think we’re helping.

You asked a great question right at the beginning of this interview about the future of counterinsurgency. Is it possible to build an indigenous force that will actually take control of its own destiny? I don’t know. But I think that’s the path to addressing that challenge in the future. In my judgment, the wrong answer is for elected officials to ask me the question they often ask, which is, “What are we [the U.S. military] going to do about it?” I get that all the time: “What are you going to do about Syria?” Here’s my response: “I’m going to try my best to find a way to integrate the military instrument of power with the other instruments of government and look for our diplomats to form coalitions and find a political path that we can enable with the use of military power.” JFQ