Richard B. Myers and Richard H. Kohn

Michael Desch's "Bush and the Generals" (May/June 2007) contains significant errors of fact and interpretation. One of us, Richard Myers, has direct knowledge and personal experience with the subject; the other, Richard Kohn, has been studying and observing American civil-military relations for 45 years.

Bush administration officials did not, as Desch charges, "overrule" the military "on the number of troops to be sent" to Iraq or "the timing of ... deployment." Both were the result of over a year of questioning and discussion back and forth, and the final plan contained contingencies for different numbers of forces depending on the course of the campaign. To be sure, the combatant commander often found the probing and questioning of plans by Secretary of Defense Donald Rumsfeld and the Joint Chiefs of Staff distasteful. But in the end, all involved supported the final plan regardless of the disagreements along the way.

Contrary to Desch's interpretation, the Kosovo intervention in 1999 was not evidence of poor civil-military relations. The Joint Chiefs, the secretary of defense, and President Bill Clinton all agreed on limiting the application of force in Kosovo -- overruling the advice of General Wesley Clark, the supreme allied commander for Europe, as was legitimate in the civil-military relationship.

There was no "truce" between the military and civilians after 9/11 because there had never been a war. There was just the friction and distrust (never open but exacerbated by Rumsfeld's approach and style) inherent in U.S. civil-military relations.

Desch charges that Deputy Secretary of Defense Paul Wolfowitz's "cavalier dismissal of troop-requirement estimates by General Eric Shinseki, the army chief of staff," was "the clearest display of civilian willingness to override the professional military on tactical and operational matters." But it is not true that Shinseki's offered advice was subsequently overruled. In his congressional testimony, Shinseki told senators that such estimates should come from the combatant commander, and he never offered these troop numbers to either the Joint Chiefs or to the president. Desch is correct, however, that criticism of Shinseki's testimony by senior civilian officials was not conducive to proper civil-military relations.

Meanwhile, Lieutenant General Gregory Newbold -- another retired military official who has recently criticized the civilian leadership -- left the Joint Chiefs of Staff in 2002, before planning for the Iraq campaign was complete or the military was informed of a decision to go to war. Newbold never made his views known to the chairman or the vice chairman, for whom he worked directly.

Desch also implies that senior military officers were intimidated into silence on the number of troops needed in Iraq during the occupation. The truth, however, is that General John Abizaid, the head of U.S. Central Command from July 2003 until March 2007, and his commanders thought more U.S. troops would be counterproductive, and the Joint Chiefs agreed. Only following the deterioration of the situation several months after the 2006 bombing of the Golden Mosque in Samarra was there support for a rise in U.S. troop levels among some of the most senior military officers.

The recommendations that Desch draws from his faulty analysis are dangerous. Certainly, Secretary of Defense Robert Gates should "encourage, rather than stifle, candid advice from the senior military leadership." But to imply that Rumsfeld stifled candid advice is misleading. Some may have been intimidated by him, but he insisted that General Myers, as chairman of the Joint Chiefs, provide advice -- and General Myers always did so, candidly. (If the chairman's advice differs from that of the service chiefs, he is obligated by law to state their advice as well.)

Desch recommends returning to "an old division of labor" in which "civilians give due deference to military professional advice in the tactical and operational realms in return for complete military subordination in the grand strategic and political realms." In fact, that "old division of labor" never disappeared, even after nuclear weapons and limited and guerrilla war blurred the distinctions and injected civilians much more heavily into operations and tactics, largely through the setting of rules of engagement. But "due deference" does not mean automatic consent, as Desch implies: that clearly would negate civilian control of the military. Meanwhile, once military advice has been offered, automatic consent by the military in strategic and political matters is necessary -- regardless of whether or not the military advice is heard, listened to, and considered.

Desch questions "salute and obey" as the norm for the U.S. military, but he seems to base this on a misinterpretation of H. R. McMaster's Dereliction of Duty: Lyndon Johnson, Robert McNamara, the Joint Chiefs of Staff, and the Lies That Led to Vietnam. (Kohn supervised McMaster's master's and Ph.D. theses, which became the book.) This misinterpretation is common in the military. In reality, the book argues and implies nothing other than this: during the Vietnam War, the Joint Chiefs should have spoken up forcefully in private to their superiors and candidly in testimony to Congress when asked specifically for their personal views, and they should have corrected misrepresentations of those views in private meetings with members of Congress.

Ultimately, there is no such thing as a "proper civil-military balance." What is necessary for effective policy, good decisions, and positive outcomes is a relationship of respect, candor, collaboration, cooperation -- and subordination. Nothing would undermine that relationship more than a resignation by a senior military officer. The role of the military is to advise and then carry out lawful policies and orders, not to make them. To threaten resignation -- taking disagreement public -- directly assaults civilian control of the military. Political and international strategic considerations are the responsibility of civilians, elected and appointed. No military officer, even at the very top, can know all that is involved in the highest levels of decision-making, which is inherently political (in the generic, not partisan, sense). In other walks of life, professionals can resign, but a military leader sworn to defend the country would be abandoning it, along with the people under his or her care or command.

There may be some extraordinary or dire situation in which an officer must for personal reasons ask to be relieved or retired: for example, when people would be slaughtered for no explicable or conceivable reason or the existence of the country jeopardized with no conceivable justification. But one individual's definition of what is moral, ethical, and even professional can differ from someone else's. There is no tradition of military resignation in the United States, no precedent -- and for good reason. Even the hint of resignation would encourage civilians to choose officers more for compliance and loyalty than for competence, experience, intelligence, candor, moral courage, professionalism, integrity, and character.

The fact is that the president and the secretary of defense have the authority and the right to reject or ignore military advice whenever they wish. That is the law, in accordance with the Constitution and consistent with U.S. historical practice. Even if Desch does not understand or accept that, the military does -- and so, too, do the American people.

RICHARD B. MYERS, Colin Powell Chair of Character, Leadership, and Ethics at the National Defense University and Foundation Professor of Military History and Leadership at Kansas State University, was first Vice Chairman and then Chairman of the Joint Chiefs of Staff between 2000 and 2005.

RICHARD H. KOHN is Professor of History and Peace, War, and Defense at the University of North Carolina at Chapel Hill.



Mackubin Thomas Owens

Desch is correct to observe that there is a troubling rift between the uniformed U.S. military and civilian leaders (although it is not as great as he suggests). But Desch errs when he blames most of the current problems on the Bush administration in general and on former Secretary of Defense Donald Rumsfeld in particular. In fact, the uniformed military deserves a significant share of the blame as well.

Desch charges the administration with willfully ignoring military advice, initiating the Iraq war with too small a force, ignoring the need for preparations for postconflict stabilization, failing to foresee the insurgency, and not adapting once things started to go wrong. This criticism is predicated on two questionable assumptions. The first is that soldiers deserve to have a voice in making policy regarding the use of the military -- indeed, that they have the right to insist that their views be adopted. The second is that the judgment of soldiers is inherently superior to that of civilians when it comes to military affairs -- and that in times of war, accordingly, civilians should defer to military expertise.

Both of these assumptions are questionable at best. They are also at odds with the principles and practice of U.S. civil-military relations, which subordinate the uniformed military to civilian authority even in what might seem to be the purely military realm. As Eliot Cohen demonstrates in Supreme Command, a book that Desch cites disapprovingly, successful wartime presidents, such as Abraham Lincoln and Franklin Roosevelt, "interfered" extensively and frequently with military operations.

Desch's first assumption rests on a misreading of McMaster's Dereliction of Duty. Many serving officers believe that Dereliction of Duty concludes that the Joint Chiefs of Staff should have more openly voiced their opposition to the Johnson administration's strategy of gradualism in Vietnam and then resigned rather than carry out the policy. But in fact, this is a serious misinterpretation that has reinforced the increasingly widespread belief that officers should be advocates of particular policies rather than simply serving in their traditional advisory roles.

Desch's second assumption -- that soldiers have better judgment than civilian policymakers on military affairs -- is called into question by a review of the historical record. Lincoln constantly prodded George McClellan to take the offensive in Virginia in 1862, while McClellan constantly whined about insufficient forces. During World War II, there were many differences between Roosevelt and his military advisers. General George Marshall, the greatest soldier-statesman since George Washington, opposed arms shipments to the United Kingdom in 1940 and argued for a cross-channel invasion before the United States was ready. History has vindicated Lincoln and Roosevelt.

Many in the military blame the U.S. defeat in Vietnam on civilians. But in fact, the operational approach in Vietnam was forged by the uniformed military. General William Westmoreland adopted the counterproductive strategy of emphasizing attrition of Peoples' Army of Vietnam forces in a "war of the big battalions" -- sweeps through remote jungle areas in an effort to destroy the enemy with superior firepower. By the time his successor could adopt a more fruitful approach, it was too late.

During the planning for Operation Desert Storm in 1990-91, General Norman Schwarzkopf, then the head of U.S. Central Command, called for a frontal assault against Iraqi positions in southern Kuwait followed by a drive toward Kuwait City. That plan would probably not have achieved the foremost military objective of the ground war: the destruction of the three divisions of Saddam Hussein's Republican Guard. Accordingly, the civilian leadership rejected it and ordered Schwarzkopf to return to the drawing board. The revised plan was far more imaginative and effective.

In the case of Rumsfeld, it is clear that he was guilty of errors of judgment regarding the conduct of the Iraq war. However, as case after case makes clear, Rumsfeld's critics were no more prescient than he. Rumsfeld failed to foresee the insurgency and the shift from conventional to guerrilla war, but so did his critics in the uniformed services. The army's official historian of the campaign has placed the blame for this failure squarely on the army. "Reluctance in even defining the situation," the historian writes, "is perhaps the most telling indicator of a collective cognitive dissonance on the part of the U.S. Army to recognize a war of rebellion, a people's war, even when they were fighting it."

Critics also charge Rumsfeld's Pentagon with shortchanging the troops in Iraq -- by, for example, failing to provide them with armored Humvees. But a review of army budget submissions even after the war had turned into an insurgency reveals that the service's priority was to acquire big-ticket items. It was only after the danger of improvised explosive devices became obvious that the army began to push for supplemental spending to "up-armor" the utility vehicles.

And although it is true that Rumsfeld downplayed postconflict stabilization operations, it is also true that he was merely ratifying the preferences of the uniformed military, which has a cultural aversion to such operations. The real villain in this case is the Weinberger-Powell Doctrine -- a set of principles long internalized by the U.S. military emphasizing the requirement for an "exit strategy." When generals are thinking about an exit strategy, they are not thinking about "war termination" -- how to convert military success into political success.

In retrospect, it is easy to criticize Rumsfeld for pushing General Tommy Franks, the combatant commander, to develop a plan for a force smaller than the one called for in earlier plans, as well as for his interference with the time-phased force and deployment list (TPFDL), which lays out the schedule for deploying forces. But that charge focuses on the consequences of the chosen path (attacking earlier with a smaller force) while ignoring potential drawbacks of the alternative -- for example, that taking the time to build up a larger force would have meant losing the chance for surprise.

The debate over the size of the invasion force must also be understood in the context of the state of civil-military relations at the end of Clinton's presidency. Rumsfeld was correct that civilian control of the military had eroded during the Clinton years. When the army did not want to do something -- as in the Balkans in the 1990s -- it would simply overstate the force requirements: "The answer is 350,000 soldiers. What's the question?" Accordingly, Rumsfeld was inclined to interpret the army's call for a larger force as another case of foot-dragging. In retrospect, Rumsfeld's decision not to deploy the First Cavalry Division was a mistake, but at the time it was not unreasonable to argue that the TPFDL (like the "two major theater war" planning metric) had become little more than a bureaucratic tool that the services used to protect their share of the defense budget.

Uniformed officers have an obligation to stand up to civilian leaders when they think a policy is flawed. They must convey their concerns to civilian policymakers forcefully and truthfully. If they believe the door is closed to them at the Pentagon or the White House, they also have access to Congress. But the U.S. tradition of civil-military relations requires that they not engage in public debate over matters of foreign policy, including the decision to go to war. And once a policy decision is made, soldiers are obligated to carry it out to the best of their abilities, whether their advice has been heeded or not.

MACKUBIN THOMAS OWENS is Associate Dean of Academics and Professor of National Security Affairs at the Naval War College in Newport, Rhode Island.



Lawrence J. Korb

Desch does an excellent job of analyzing one problem in civil-military relations during the Bush administration: the failure of civilians to give due deference to professional military advice in the tactical and operational realms. But he does not deal with a more serious problem: the way in which President George W. Bush and his appointees have used military professionals to support their political agenda. The problem is illustrated both by the conduct of General David Petraeus, the current commander of the multinational forces in Iraq, and by the White House's decision not to nominate General Peter Pace for a second two-year term as chairman of the Joint Chiefs of Staff.

On September 26, 2004 -- approximately six weeks before a presidential election in which the deteriorating situation in Iraq was an increasingly important issue -- Petraeus, then in charge of training Iraqi security forces, published an op-ed in The Washington Post. He wrote glowingly of the progress the Iraqi security forces were making under his tutelage. According to the article, training was on track and increasing in capacity, more than 200,000 Iraqis were performing a wide variety of security missions, 45 Iraqi National Guard battalions and six regular Iraqi army battalions were conducting operations on a daily basis, and six additional regular army battalions and six Iraqi Intervention Force battalions would become operational by the end of November 2004. The Bush administration's policy at that time was "we will stand down when they stand up." Petraeus' article, accordingly, had the effect of telling the electorate that there was light at the end of the tunnel.

The op-ed was patently false and misleading, but that was not the worst part. If Petraeus wrote and published the article on his own initiative, he was injecting himself improperly into a political campaign. If he was encouraged (or even authorized) to do so by his civilian superiors, they were abusing military professionalism for partisan political purposes.

In his new role, Petraeus continues to be used by the administration -- willingly or unwillingly -- for partisan political purposes. When many Democrats and some Republicans criticize the president's latest strategy in Iraq, Bush and his political allies argue that the strategy is not theirs but that of General Petraeus -- and that by confirming him, the Senate essentially endorsed the current approach, obligating it to give "Petraeus' strategy" a chance to succeed. Whether Petraeus was in fact influential in getting the president to adopt the so-called surge is beside the point. The Senate confirmed Petraeus because legislators believed that he had the background and experience necessary to carry out policies decided on by the president in consultation with Congress. Regardless of where it originated, the current strategy in Iraq is Bush's strategy.

Desch is right that Rumsfeld's meddling approach contributed to the problems in Iraq and that the new secretary of defense, Gates, should give more deference to advice from military professionals. But at the same time, Gates must not encourage -- or allow -- military professionals to inject themselves into political campaigns or to be used to legitimize political decisions. The long-term ramifications of this abuse may be much worse than Rumsfeld's meddling. When Bush says that Congress must support Petraeus' strategy, Gates needs to remind him that it is the administration's strategy, and Petraeus needs to make clear that he will support whatever strategy the American people decide on -- and that unlike in 2004, he will be honest with voters and stay out of the next presidential election.

Unfortunately, by not appointing General Pace to a second two-year term as chairman of the Joint Chiefs of Staff, Gates has continued the Bush administration's policy of exploiting military professionalism to support its political agenda. Pace was the first chairman in over 40 years to be denied the customary two-year reappointment. This had nothing to do with his qualifications or a failure to act professionally on his part. (In fact, Gates earlier expressed a desire to reappoint Pace, and President Bush called him a good general.) Pace was denied his reappointment because he would have faced a contentious confirmation hearing. And why would the hearing have been contentious? Because it would have provided the Senate with another opportunity to cause political problems for the Bush administration by, once again, examining its failed policies in Iraq.

LAWRENCE J. KORB, Senior Fellow at the Center for American Progress, was Assistant Secretary of Defense for Manpower, Reserve Affairs, Installations, and Logistics from 1981 to 1985.



Like Lawrence Korb, I deplore how the Bush administration has cynically used military professionals to advance its political agenda. The most egregious example of this is how after years of ignoring military advice it did not like, the administration now deflects calls from members of Congress to reduce troops in Iraq with the pious injunction that Washington ought not substitute "the opinions of politicians for the judgments of our military commanders," as President Bush put it after vetoing Congress' Iraq spending bill. I am not sure, however, that I share Korb's assessments of Generals Petraeus and Pace. I will withhold judgment of Petraeus' independence and candor until mid-September, when he delivers his report to Congress on the situation in Iraq. Unlike Korb, I do not think that Pace was denied a second term as chairman of the Joint Chiefs simply because his role in the Iraq war would have made his confirmation hearings contentious. I believe Pace's inappropriate public statements on nonmilitary issues -- his condemnation of homosexuality and his support for clemency for Lewis "Scooter" Libby -- also played a role in the decision.

The most important issue Mackubin Thomas Owens raises concerns the respective roles of military and civilian leaders in wartime decision-making. The best system is one that allows for substantial military autonomy in the military, technical, and tactical realms (how to fight wars) in return for complete subordination to civilian authority in the political realm (when and if to fight them). Admittedly, this approach is not perfect, but, like Samuel Huntington, I believe that it strikes the best possible balance between military effectiveness and civilian control. It is also consistent with Clausewitz's dictum that war has "a grammar of its own, but its logic is not peculiar to itself." Civilians should have the final say, in Clausewitz's view, not because they have any greater expertise than military officers in the narrow military realm (the grammar of war) but because the political rationale for war (its logic) should be paramount in guiding state policy.

This important distinction highlights just how radical a departure the Bush administration's approach to civilian control has been. That approach, which Owens endorses, is that civilians are more competent than military professionals not only in the larger political sphere (a point on which we all agree) but also in the narrower military realm. This latter argument defies common sense: professionals by definition have greater expertise in their particular fields than do those who deal with them in only part of their careers and then only episodically, as is the case for most senior civilian leaders in regard to the military.

There is not much evidence supporting the proposition that civilians make superior decisions in the narrow military realm than do military professionals. The single study that Owens cites -- Cohen's Supreme Command -- is flawed. Rather than looking at all the instances in which civilians "probed" into the military realm, Cohen chooses only a handful of cases, and only ones in which civilians did so successfully. And even among the handful of cases Cohen examines, the record of civilian strategists is mixed. Winston Churchill, whatever his heroic political leadership during World War II, pushed more than his share of harebrained military schemes that resulted in disaster (Gallipoli in World War I and Norway early in World War II) or would have had they been implemented (landings in the Balkans instead of western Europe).

General Richard Myers offers his "direct knowledge and personal experience" as reasons to accept his and Richard Kohn's remarkable claim that the intense civil-military "friction and distrust" in the run-up to the Iraq war was normal. Of course, throughout his tenure as chairman of the Joint Chiefs of Staff, Myers was a good team player for the Bush administration and put the best face on the increasingly bad situation in Iraq. If he thought that the war was going swimmingly, it is not surprising that he also thought that all was as it should be in civil-military relations.

The first evidence that civil-military relations were broken was the torrent of leaks that came out in the run-up to the war from senior officers within the planning process who were unhappy with Rumsfeld's numerous interventions into the details of the operation. At the time, Myers dismissed these stories as untrue and attacked the critics as unpatriotic. Subsequently, six retired generals who were intimately involved in either the Washington or the Iraqi side of the war's planning and execution issued calls for Rumsfeld's removal on the grounds that his meddling had crippled U.S. efforts. Perhaps Kohn can think of instances when U.S. civil-military relations have been worse over his 45-year career studying the subject, but I cannot, save perhaps for the later phases of the Vietnam War. And even if that period is comparable to the situation under Rumsfeld, it is hardly a ringing endorsement of the state of U.S. civil-military relations.

Myers and Kohn criticize my claim that professional military judgment was ignored on the matters of troop levels and deployment timing. Newbold, the former director of operations for the Joint Chiefs of Staff, admits he should have spoken out sooner. But others did speak out before the war, including Shinseki. Myers and Kohn claim that Shinseki never offered advice on the matter of troop levels to the rest of the Joint Chiefs or to the president. But in his much-discussed congressional testimony in February 2003, Shinseki suggested that "several hundreds of thousands" might be needed to occupy Iraq, even if he ultimately deferred to General Franks' judgment. And in their book Cobra II, Michael Gordon and Lieutenant General Bernard Trainor report that at a White House meeting on January 20, 2003, Shinseki raised a number of issues, including troop numbers and the timing of their deployment, as qualifications of his endorsement of the war plan.

If taken at face value, Myers and Kohn's assertion that "in the end, all involved supported the final plan" is a damning indictment of the competence of the senior military leadership, including Myers himself, who assures us he had Rumsfeld's ear. I see it, instead, as an indication that after enough time and pressure, generals will eventually give their civilian bosses the answers they want. Otherwise, as the Shinseki affair so clearly demonstrated, other generals will be found to replace them. Although Myers and Kohn concede that Wolfowitz's public humiliation of Shinseki "was not conducive to proper civil-military relations," they understate the chilling effect this incident had on other senior officers who dissented from Rumsfeld's policies. The fact that Rumsfeld and his team were, through a year and a half of "probing and questioning" (which Myers and Kohn concede the combatant commander found "distasteful"), able to whittle the final troop number to less than half of the 380,000 the original war plans called for does not alter the fact that civilian views on force levels prevailed over the military consensus -- with disastrous results after the fall of Baghdad.

Kohn has chided me for years about my reading of McMaster's influential Dereliction of Duty, a misreading that according to him and to Owens, many military officers have made as well. But in defense of his interpretation of McMaster's argument as simply a plea that senior military leaders speak candidly, Kohn cites not the book itself but rather his own authority as McMaster's thesis adviser. I read McMaster's message as being that unqualified allegiance to the commander in chief needs to be rethought, based on a number of vignettes he presents involving senior military leaders who went, or should have gone, well beyond offering candid advice. Telling is McMaster's account of Army Chief of Staff Harold Johnson's despair when he realized that he "was to preside over the disintegration of the Army" because "he did not resign, resist, or object to the president's decision." Today, the turmoil inside the U.S. Army is so great that many junior officers reportedly believe -- and one, Lieutenant Colonel Paul Yingling, has publicly charged -- that the general officer corps is guilty of a similar dereliction of duty for going along with policies foisted on the military by intrusive civilian leaders. If one reader misconstrues an author's argument, the reader is probably at fault. When numerous readers all come to the same erroneous conclusion, the author was probably just not clear.

Finally, Myers and Kohn warn that my policy recommendations are "dangerous." My suggested civil-military division of labor implies, they charge, "automatic consent" to military preferences. And they claim that my suggestion that senior generals who protested Rumsfeld's Iraq policies would have been more effective had they resigned while on active duty undermines civilian control. In fact, I do not argue that civilian leaders should rubber-stamp military policies, even in the tactical or operational realms, so this charge is a red herring. The problem with civilian meddling in the run-up to the Iraq war was not that Rumsfeld and Wolfowitz overruled the senior army leadership on the number of troops necessary for "Phase IV," or reconstruction, but that they did so claiming superior military expertise, rather than offering a compelling political reason for ignoring these military recommendations. I suspect that Rumsfeld and Wolfowitz had political reasons for reducing the size of the force -- public support for the war would likely have been less if it could not have been sold as a "cakewalk" -- but they were unwilling to subject their reasoning to critical scrutiny and wanted to confine the debate to the more esoteric realm of military strategy, on which most of the public was unlikely to challenge them.

Their second charge is based on the questionable assumption that the fear of senior officers' requesting retirement or relief of duty in protest against policies they found objectionable would lead civilians to appoint senior officers based on "compliance and loyalty." Myers and Kohn cannot seriously maintain that during the golden age when protest was rare, senior officers were never appointed for their pliability. Moreover, it does not follow logically that such behavior would undermine civilian control. Indeed, by raising the bar for military dissent, it would make it more likely that serving officers would "salute and obey" in response to orders with which they disagreed in all save the most extreme circumstances.

Myers and Kohn conclude that "what is necessary for effective policy, good decisions, and positive outcomes is a relationship of respect, candor, collaboration, cooperation -- and subordination." It is hard to characterize our Iraq policy in such positive terms. The civil-military relationship under Rumsfeld lacked respect, candor, collaboration, and cooperation. If the best we can say is that the U.S. military did not become openly insubordinate, that is damning with faint praise.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now