The Contemporary Presidency: Civil-Military Friction and Presidential Decision-Making

The Contemporary Presidency: Civil-Military Friction and Presidential ...
Davidson, Janine
Presidential Studies Quarterly; Mar 2013; 43, 1; Research Library
pg. 129
984503137150
strategy and an uncertain timeline. The president and his civilian advisors could not understand why the military seemed incapable of providing scalable options for various goals and outcomes to inform his decision-making. Meanwhile the military was frus- trated that their expert advice regarding levels of force required for victory were not being respected (Woodward 2010).
Such mutual frustration between civilian leadership and the military is not unique to the Obama administration. In the run-up to the Iraq War in 2002, Secretary of Defense Donald Rumsfeld famously chastised the military for its resistance to altering the invasion plan for Iraq. The military criticized him for tampering with the logistical details and concepts of operations, which they claimed led to the myriad operational failures on the ground (Gordon and Trainor 2006; Ricks 2007; Woodward 2004). Later, faced with spiraling ethnic violence and rising U.S. casualties across Iraq, George W. Bush took the advice of retired four-star General Jack Keane and his think tank col- leagues over the formal advice of the Pentagon in his decision to launch the so-called surge in 2007 (Davidson 2010; Feaver 2011; Woodward 2010).
A similar dynamic is reflected in previous eras, from John F. Kennedy's famous debates during the Cuban Missile Crisis (Allison and Zelikow 1999) to Lyndon Johnson's quest for options to turn the tide in Vietnam (Berman 198 5; Burke and Greenstein 1991), and Bill Clinton's lesser-known frustration with the military over its unwilling- ness to develop options to counter the growing global influence of al-Qaeda.2 In each case, exasperated presidents either sought alternatives to their formal military advisors or simply gave up and chose other political battles. Even Abraham Lincoln resorted to simply firing generals until he got one who would fight his way (Cohen 2002).
What accounts for this perennial friction between presidents and the military in planning and executing military operations? Theories about civilian control of the military along with theories about presidential decision making provide a useful starting point for this question. While civilian control literature sheds light on the propensit y for friction between presidents and the military and how presidents should cope, it does not adequately address the institutional drivers of this friction. Decision-making theories, such as those focused on bureaucratic politics and institutional design (Allison 1969; Halperin 1974; Zegart 2000) motivate us to look inside the relevant black boxes more closely. What unfolds are two very different sets of drivers informing the expectations and perspectives that civilian and military actors each bring to the advising and decision- making table.
This article suggests that the mutual frustration between civilian leaders and the military begins with cultural factors, which are actually embedded into the uniformed military's planning system. The military's doctrine and education reinforce a culture of “military professionalism,” that outlines a set of expectations about the civil-military decision-making process and that defines “best military advice” in very specific ways. Moreover, the institutionalized military planning system is designed to produce detailed
2. Telephone interview with former Deputy Assistant Secretary of State James Steinberg, September 6, 2012. Dr. James Steinberg served in the Clinton administration as the Deputy National Security Advisor and in the Obama administration as the Deputy Secretary of State.
and realistic military plans for execution—and that will ensure “victory”—and is thus ill suited to the rapid production of multiple options desired by presidents. The output of this system, framed on specific concepts and definitions about “ends,” “ways,” “means,” and expectations about who provides what type of planning “guidance,” is out of synch with the expectations of presidents and their civilian advisors, which in turn have been formed from another set of cultural and institutional drivers.
Most civilian leaders recognize that there is a principal-agent issue at work, requir- ing them to rely on military expertise to provide them realistic options during the decision-making process. But, their definition of “options” is framed by a broader set of political objectives and a desire to winnow decisions based, in part, on advice about what various objectives are militarily feasible and at what cost. In short, civilians’ diverse political responsibilities combined with various assumptions about military capabilities and processes, create a set of expectations about how advice should be presented (and how quickly), how options might be defined, and how military force might or might not be employed. These expectations are often considered inappropriate, unrealistic, or irrel- evant by the military. Moreover, as discussed below, when civilians do not subscribe to the same “hands off’ philosophy regarding civilian control of the military favored by the vast majority of military professionals, the table is set for what the military considers “meddling” and even more friction in the broken dialogue that is the president's decision-making process.
This article identifies three drivers of friction in the civil-military decision-making dialogue and unpacks them from top to bottom as follows: The first, civil-military, is not so much informed by theories of civilian control of the military as it is driven by disagreement among policy makers and military professionals over which model works best. The second set of drivers is institutional, and reflects Graham Allison's organiza- tional process lens (“model II”). In this case, the “outputs” of the military's detailed and slow planning process fail to produce the type of options and advice civilians are hoping for. Finally, the third source of friction is cultural, and is in various ways embedded into the first two. Powerful cultural factors lead to certain predispositions by military planners regarding the appropriate use of military force, the best way to employ force to ensure “victory,” and even what constitutes “victory” in the American way of war. These cultural factors have been designed into the planning process in ways that drive certain types of outcomes. That civilians have another set of cultural predispositions about what is appropriate and what “success” means, only adds more fuel to the flame.
First Order (Civil-Military) Friction: Differing Expectations for Civilian Control
In the classic The Soldier and the State (required reading by all military officers), Sam Huntington (1981) argued for what he labeled “objective” control of the military. In this model, which Eliot Cohen (2002) labeled the “normal” theory of civil-military relations because of its widespread acceptance across the military profession, civilian leadership should provide the military with broad objectives and then stand aside while the military
professionals plan and execute the mission in the way they see fit. Just as one would not dictate to one's doctor how to perform surgery, presidents or civilian secretaries are considered unqualified to scrutinize details of military operations.
In the minds of many professional military officers, the Huntington (1981) model was validated by the negative example of President Johnson's hyperscrutiny of bombing targets during the Vietnam War, and by the positive example of George H. W. Bush’s deference to the military in the execution of the first Gulf War in 1990 (Kitfield 1997). This narrative permeated the education system and the culture of the U.S. military for decades. Indeed, as an Air Force officer in the 1990s, I was taught that our failures in Vietnam were due to the fact that the military was forced to fight “with one hand tied behind its back.” When President Bush claimed after Desert Storm, “By God—we've kicked the Vietnam syndrome once and for all,” military professionals understood that they had a commander in chief who would, as Ronald Reagan (1980) had promised, “never again ask young men to fight and possibly die in a war our government is afraid to let them win.”
In contrast, Eliot Cohen (2002) demonstrates in his book, Supreme € ommand, that a more “hands on” approach to civilian control is likely to yield better outcomes. In his model, civilians respect the military's operational expertise, but the commander in chief is understood to have broader responsibilities and insight in his role as president, thus requiring him—indeed legally authorizing him—to determine whether or not various military options are sound. What the military considers the appropriate plan to achieve an operational “victory,” a president might view as more strategically or politically risky given other macro objectives. President Kennedy colorfully described this mis- match when he remarked that the Joint Chiefs “advise you the way a man advises another one about whether he should marry a girl. He doesn't have to live with her” (Zegart 2000, 45).
The president's responsibility to see broader strategic issues and goals often leads him to disregard or override military advice. In Cohen's study (2002), for example, Winston Churchill insisted the military continue advancing beyond the point where the generals had declared the enemy defeated because he understood that where forces physically stood and held ground when the bullets stopped flying would dictate the terms of peace, especially in critical places like Berlin. Of course, the right to disregard operational expertise also grants presidents the right to make bold strategic errors as well. Consider that George W. Bush’s decision to divert operational resources from an ongoing fight in Afghanistan in order to invade Iraq led to failure in the battle of Tora Bora, missed opportunity to kill or capture Osama bin Laden, and tragic and unnecessary U.S. casualties in the mountains of Afghanistan (Berntsen and Pezzullo 2006; Fury 2008).
For best results, Cohen (2002) proposes the need for a respectful, but “unequal” dialogue between the military and the president. In this model, the military provides its best advice, and the president can and should question it until there is ideally a mutual understanding. Regardless of reaching consensus, the president has the final say in how or if to execute. In this unequal dialogue, the roles are clear as each side has distinct responsibilities based on position and expertise. Because the military are considered the experts in the art of war, their advice should inform presidential decision making by
offering operational military options as well as details about resources required and timelines. A president must know how a potential conflict will unfold, how many forces it might require, and how long it might take to achieve various objectives (or alternative “end states” in military language). The military is not necessarily expected (nor invited) to offer strategic and political advice, such as whether or not the mission is in the “national interest” or if the American people will or will not support it. Presidents would consider those issues theirs to determine after weighing the myriad factors and options.
Cohen's model is useful in that it outlines the perspective and responsibilities of the commander in -chief, helps one understands his constraints and expectations, and draws clear lines around the roles each side should play in an “unequal” civil-military dialogue. Cohen's book, which was published in 2002, is a clear challenge to the “normal” Huntington model, which has been the prevailing model for military profes- sionals for decades. Based on the historical record of frustrated presidents, it appears that Cohen's model comes closer to how civilians assume the process should work. It easily follows that where presidents and their civilian advisors subscribe to Cohen and military officers to Huntington (1981), friction is bound to ensue. This mismatch in expectations creates not an “unequal” dialogue, but a broken one.
Graham Allison's account of the decision to execute a naval blockade during the Cuban missile crisis is a classic example of the dilemma that emerges given the mismatch in philosophy over civilian control, coupled with the principal-agent problem of military expertise. In this account, after a great deal of wrangling and maneuvering among advisors and between civilians and the military, the president decided that a blockade was a better option than airstrikes. No sooner had the Navy gotten to work executing its mission, than President Kennedy and Secretary of Defense Robert McNamara began to wonder just what—exactly—they had ordered the Navy to do. They sought answers:
McNamara put his questions harshly: Who would make the first interception? Were Russian-speaking officers on board? How would submarines be dealt with? ... Picking up the Manual of Navy Regulations (Chief of Naval Operations, Admiral Anderson) waved it in McNamara's face and shouted, “It's all in there.” To which McNamara replied, “I don't give a damn what John Paul Jones would have done, I want to know what you are going to do now!” The encounter ended on Anderson's remark, “Now Mr. Secretary, if you and your Deputy will go back to your office the Navy will run the blockade.” (Allison 1969, 707).
This exchange demonstrates how a president, subscribing to the Cohen (2002) model, can fall into a trap by not asking the “right” questions up front of his Huntingtonian- minded military advisors. During the decision-making deliberations, President Kennedy chose a naval blockade over what he perceived was the more aggressive and operationally and politically risky option of airstrikes. Details about rules of engagement would have been important for the president to know in order for him to understand how various options might unfold and to determine the level and nature of risk he may be undertak- ing in each option. Thus, the questions McNamara was asking were critical to this understanding—and arguably should have been asked before the decision was made. Still, civilian decision makers might also have assumed that such an assessment of risks would have been offered by the Navy as part of its presentation of options and in its role
as advisor. In short, civilians may not know what they do not know. But, from the classic Huntingtonian perspective, which clearly the Chief of Naval Operations subscribed to, such details would have been considered too “in the weeds” for civilians to concern themselves with and such questions perceived as “meddling” in military affairs.
During the Clinton administration, Jim Steinberg, who was the deputy national security advisor, was told by the military leadership that he was not “qualified” to see a particular set of detailed, highly classified, war plans. Steinberg concluded, “the military thinks that this military stuff is so complex and complicated that civilians don't have the expertise to understand what they are seeing. So they need to ‘protect’ them from making poor decisions.” He says that for high-level civilians who need to advise the president, it is important to “understand the game.” But, he says, he refuses to play it.’
When Steinberg insisted on being briefed, he was alarmed to note that there were large movements of forces written into the early, precombat, phases of the plan. He commented that this might be perceived as provocative to the enemy and thus risk unnecessary escalation, something civilian leadership desperately wanted to avoid. But from the military's perspective, not having those capabilities in the region when the time came introduced a level of operational risk that made the military very uncomfortable. Steinberg's instincts to query the military and to delve deeper into the details—against the preferences of the military—paid off in creating the opportunity for each side to better understand the tradeoffs between the political and operational objectives. Thus, the president could be presented with a more holistic understanding of the risks when and if the time came.
In some ways, both sets of expectations revealed in these examples are under- standable. Military professionals expect to be given clear objectives for their military planning and then be free to execute toward those objectives based on their military expertise and their judgment regarding what constitutes operational success (a.k.a. “victory”). Presidents and secretaries should know what they want done and issue explicit orders accordingly. But in order to decide what they want done—in fact in order to provide the type of up-front guidance the military desires—they need to weigh risks associated with how various options might unfold. And for that they need a level of military expertise they likely do not possess, and a degree of familiarity with plans and rules of engagement that the military considers beyond their need to know. Presi- dents do not normally study military doctrine and operational manuals in preparation for their role as commander in chief. They expect their military advisors to present such advice and expertise to them.
A more recent example, recounted by former Pentagon official Rosa Brooks, further illuminates this “chicken and egg” dilemma (Brooks 2012).4 As early as 2009, White
J. Telephone interview with former Deputy Assistant Secretary of State James Steinberg, September
6, 2012.
1 The author also participated in this planning debate, as a colleague of Dr. Brooks in the Pentagon, 2010-11. Dr. Brooks was a senior advisor to the undersecretary of defense for policy at the time, working on rule of law and humanitarian issues; the author was the deputy assistant secretary of defense for plans. In this broken dialogue, our role as Pentagon civilians was to translate what the White House wanted into concrete planning “guidance.”
House officials in the Obama administration began to worry about potential genocide- level violence in the wake of the pending referendum on South Sudan. They wanted to know what the military could do about it—what military options could they recommend to the president in the event such mass killing commenced or appeared imminent? But before offering options, the military said they needed more guidance. Their queries led to exasperation on both sides in a classic frustrated dialogue that went something like this:
Civilians: What are our military options to prevent or stop mass atrocities? Military: That depends. How much money is the president willing to spend? Civ: What? How much does this kind of thing normally cost?
Mil: What kind of thing? What do you want to do, exactly?
Civ: We don't want genocide. We want to protect innocent people from getting killed.
Mil: Ok, then we need some proper planning guidance. How many troops is the president willing to commit? For how long? What are his end states? Is it purely humanitarian or do you want plans to fight the Sudanese in north Sudan? Are we talking about targeting the bad guys or just evacuating civilians? Oh, and it will take us about six months to gin up a plan.
Civ: Six months? The referendum is in 4 months! This could happen any day! And what do you mean by “end state”?
Rosa Brooks characterized the broken dialogue as follows:
The ensuing back and forth was tense and occasionally broke out into open expressions of anger and mistrust. At best White House staff members considered their military counterparts rigid, reductionist, and unimaginative. At worst, they were convinced that the Pentagon was just being difficult—that the military “didn't care” about Sudan or about atrocity prevention and was determined to flout the president's wishes by stonewalling and foot-dragging at every turn instead of getting down to work.
The military representatives involved in the discussions were equally exasperated. What was wrong with these civilians? Didn't they know what they wanted? Were they too naive—or uncaring—to understand that the potential mobilization of thousands of people and millions of dollars of equipment required greater specificity in terms of assumptions, constraints, and desired end-states? (Brooks 2012)
Clearly, a mismatch in perspective over how civilians can and should oversee the military and provide guidance and how the military should present plans and advice contributed to this very frustrated dialogue. But a simple Cohen-Huntington debate does not fully illuminate why, once the military has come to the table, it seemed so difficult for them to understand what civilians wanted to accomplish or to generate options in what civilians might consider a reasonable amount of time. Civil-military scholar and former white house official, Peter Feaver might explain this behavior as classic military “shirking,” which according to his principal-agent model is likely to occur when “the gap between what the military would prefer to do and what civilians have asked it to do is great ... and/or when the expectation of punishment is small” (Feaver 1998, 415).
The Sudan episode may seem to fit Feaver's model; but military professionals who concede the importance and legitimacy of civilian control adamantly reject the shirking thesis. As a former chief of the planning office on the Joint Staff, Brigadier General Hix, put it, “don't ascribe malice of intent to either side of the equation; there are simple laws
of physics at play in military planning”’. Hix's “laws of physics” include realities of time and space associated with logistics as well as the sheer work load faced by the limited numbers of trained planners using insufficient computer modeling technology to produce viable military plans. A detailed examination of the military's planning process and the doctrine that informs that process shows how these elements affect the military advice offered to the president. Decision-making theories, such as the bureaucratic politics (Allison 1969; Halperin 1974) or the organizational process models (Allison 1969), further illuminate the dynamics at play.
Second Order (Institutional) Friction: Military Planning Systems Meet Presidential Decision Making
The classic bureaucratic politics model applied to presidential decision making presents a vision of men and women in a room, each with a different desired course of action, bargaining and debating for the final outcome. Their positions are determined by the interests and culture of their organizations, demands placed on the individuals within the organization, and personal goals of decision makers and advisors that may not be directly related to the issue at hand. That is, “where you stand is where you sit” (Allison 1969, 711).
In this model, the president and his military advisors each come to the table with a very different set of factors influencing “where they sit.” As General Michael Hayden, former Director of the CIA, described the environment, “each person has a seat at the table; but each has come to that seat through a very different door” (Hayden 2012). The president is charged as commander in chief to be a responsible steward of military resources; while also ensuring he is safeguarding the national security of the country. He may also have any number of domestic agenda items for which he will need to husband his political capital. As Hayden also points out, the president of the United States, who is elected based on his vision for the country, is therefore “vision-based” and inherently optimistic about the United States’ ability to solve problems. This is in contrast to his intelligence advisors, who are “facts based,...