Abstract

How do electoral politics affect presidential decisionmaking in war? As both commander in chief and elected officeholder, presidents must inevitably balance competing objectives of the national interest and political survival when assessing alternative military strategies in war. Yet, how and when electoral pressures influence decisionmaking during an ongoing conflict remains unclear. Drawn from the logic of democratic accountability, two mechanisms of constraint may be inferred. First, presidents may delay making decisions that are perceived to carry excessive electoral risk. Second, electoral pressures may have a dampening effect, causing presidents to water down politically sensitive courses of action to minimize any expected backlash. Recently declassified documents and interviews with senior administration officials and military figures illustrate these mechanisms in a case study of decisionmaking during the second half of the Iraq War. Both George W. Bush's surge decision of 2007 and Barack Obama's decision to withdraw troops in 2011 are shown to have been profoundly influenced by concerns related to the domestic political calendar. These findings call for further study of the nuanced ways in which the electoral cycle shapes wartime decisionmaking.

Introduction

It is an inconvenient truth, rarely admitted, that leaders habitually take electoral considerations into account when making decisions about military strategy in war. At once commander in chief and the holder of the highest elected office, presidents must inevitably balance the competing objectives of the national interest with personal political survival when assessing alternative strategies in war.1 Although true of all regime types, this is especially the case in democracies such as the United States, with clear and predictable channels through which the voting public can give its verdict on the leader's performance and thus hold them to account every four years. As John Brennan, a former director of the Central Intelligence Agency (CIA), puts it, “If you're going to be involved in some type of foreign activity that is going to take resources and entail risks to forces, you need to do it in a manner that is both most practical and most effective, but at the same time try to keep your political support at home, in terms of the Congress, in terms of the American public, and with the election cycles, both midterm elections as well as presidential elections.”2

A rich literature has emerged in recent years concerning democratic accountability in decisionmaking, especially as applied to the United States. Scholars have examined a whole host of domestic determinants of foreign policy, including the role of public opinion, partisanship, and Congress.3 Collectively, they undergird key research programs exploring varied theoretical phenomena, including audience costs and the democratic peace. Although this article touches on many of these issues tangentially, it is distinct in its focus on electoral constraints, defined here as institutional limitations imposed on presidential decisionmaking by the U.S. electoral cycle. Opinion polls and the atmosphere in Congress, of course, act as political bellwethers for any elected official, but the relative weight accorded to them will vary depending on how close the “accountability moment” of Election Day is, to use a phrase famously coined by George W. Bush.4 As such, a major theoretical contribution of this article is in highlighting the much understudied electoral face of democratic accountability, putting the politicians back into international politics.

Specifically, this article introduces two key mechanisms that explain how electoral constraints affect presidential decisionmaking on questions of military strategy in war. First, it argues that awareness of upcoming elections typically causes presidents to delay decisions where the options available are deemed to present significant risk in terms of their electoral prospects. Where circumstances dictate that deferral is not an option, however, electoral politics will have a dampening effect on decisionmaking, whereby politically sensitive courses of action will be weakened to minimize the expected backlash they may incur.

I illustrate these mechanisms through an examination of presidential decisionmaking during the second half of the Iraq War. I argue, first, that the 2007 surge in Iraq—the biggest shift in strategy of the entire conflict—was delayed for at least six months in anticipation of the 2006 midterms, which acted as a critical turning point, releasing President Bush to very quickly opt for a bold policy that he knew would ignite fierce public opposition. Second, I show that President Barack Obama's reaction to several proposals that recommended leaving a follow-on force after the December 2011 withdrawal deadline was strongly conditioned by the dampening effect, on the eve of the 2012 election season.

In making these arguments, this article addresses a number of gaps in the small but growing literature dealing with the effect of electoral politics on decisionmaking.5 The first relates to the dependent variable of interest, which in almost all existing studies relates to decisions to enter a conflict or the initial use of force by the United States short of war. The second gap in the literature follows from its overwhelmingly quantitative and theoretical nature; although analysts have established a number of important patterns regarding elections and decisions to use force, less work has been done on understanding the full causal mechanisms underpinning the relationships identified. Finally, and relatedly, the complexity of the role of electoral constraints cannot be fully understood through an artificially imposed temporal frame, centered on a presidential election year or the three months prior to November, which characterizes much existing work.

A single case study cannot fully demonstrate any wider generalizability of the claims made in this article. What it can provide, however, is both a degree of plausibility sufficient to move scholars beyond mere theoretical supposition and a spur to further research. By focusing on Iraq, analysts can also use the electoral prism to shed new light on an issue of continuing contemporary relevance and growing historical debate.6 Drawing on a series of interviews with senior figures from both administrations and the top echelons of the military, as well as recently declassified archival material, the case brings a new perspective on the origins of the continuing struggles of U.S. policy in the Middle East.

The rest of this article proceeds as follows. First, I outline the logic of electoral accountability as the norm in terms of explaining how electoral politics affect decisionmaking. Second, I describe the mechanisms underpinning this concept when applied to wartime contexts, emphasizing the electoral pressures to delay and dampen decisionmaking behaviors and outcomes. Third,

I demonstrate how these mechanisms operated in practice during the second half of the Iraq War. The article concludes with a brief summary of its main findings and a call for further research.

The Logic of Electoral Accountability

At a basic level, arguments in favor of electoral accountability stem from the classic Kantian logic that because citizens bear the brunt of the human and financial costs of war, they will hold a natural aversion to recklessly entering conflicts. In a democracy, where elections allow voters to retrospectively punish leaders for excessive belligerence, decisionmakers are incentivized ex ante to act with a degree of caution in matters of war and peace for fear of later reprisal at the ballot box. Because leaders are at least in part motivated by the desire to remain in office, it logically follows that public attitudes are a natural source of constraint on belligerent behavior.7

Recent work on the broader phenomenon of democratic constraint has added various nuances to this assumption. For instance, Jonathan Caverley argues that the average citizen can, in fact, be shielded from the financial burdens of war—notably, in democracies with high income inequality and progressive taxation—thereby eroding the Kantian constraint and permitting aggressive militarism to take hold.8 When it comes to the cost in terms of casualties, this too has arguably decreased in salience to the average voter in the United States, who will be less likely than ever to have to fight in any given conflict, particularly since the transition to an all-volunteer force. With the battlefield increasingly occupied by drones and precision-guided munitions, rather than boots on the ground, the human cost of war appears to be shrinking further still, again legitimating the use of force.9 Even in cases where considerable numbers of troops have been deployed, citizens may be willing to stomach a high number of casualties if they believe that victory is likely, that the mission is sufficiently worthy, or if a consensus exists among elites in favor of the war.10 Others who accept the basic premise of democratic constraint nonetheless argue that its strength varies from case to case, depending in part on the access the public has to information about the costs of war.11 Some scholars working on the institutional foundations of the democratic peace, meanwhile, agree that electoral accountability does not render democratic states peculiarly pacifistic, per se, so much as more discriminatory in choosing their battles. Because elected leaders are more likely to face reprisal at the ballot box for losing wars, such states tend to enter into conflicts they are likely to win, and go on to fight harder to recover sunk costs in a bid to avoid defeat.12 Few of these scholars would outright reject the notion of democratic constraint, but would offer a more sophisticated account for when it might operate.

In addition, important studies have argued that presidents not only have the power to shape and lead public attitudes, but have increasingly sought to do so.13 Given that presidents hold varying beliefs about the proper role of public opinion in foreign policy formulation, some individuals will be more resistant to the pressures of the electorate than others, and more prone to using the bully pulpit to manipulate public opinion in support of a preferred policy.14 In their examination of public attitudes during the Iraq War, Christopher Gelpi, Peter Feaver, and Jason Reifler leave room for such an interpretation. In arguing that popularly held views concerning the public's sensitivity to casualties are exaggerated, their analysis implies some latitude for presidents to increase the public's level of casualty tolerance and thereby maintain support for costly wars through carefully planned rhetorical strategies that emphasize the likelihood of success.15

The consensus, however, is that although presidents may well try to lead public opinion, they are nonetheless bound when choosing among alternative policies. The menu of feasible options open to them will usually be limited by concerns about the expected public reaction, as perceived (accurately or otherwise) through polling data, congressional sentiment, or media sources.16 Hence, as much as some might like to, presidents cannot entirely ignore public attitudes or write themselves a blank check of support for costly military operations through a smart communications strategy. Moreover, although it might be true that conventional assumptions about reflexive casualty phobia among the U.S. public are at least in part faulty, even those scholars who argue this acknowledge that policymakers persistently succumb to this apparent misconception in reality, often implementing policies that seek to minimize casualties as a result.17 Taking no particular stance on related academic arguments about the determinants of public attitudes concerning war, then, this article focuses instead on the underlying truth that policymakers worry about those attitudes. More specifically, it explores what this concern means for wartime decisionmaking in the broader context of the U.S. electoral cycle.

A more extreme approach related to the idea that presidents can lead public attitudes is diversionary war theory, in which commanders in chief not only are seen to have the ability to manipulate public opinion to support their preferred course of action, but are incentivized to use force abroad for the purpose of influencing public sentiment. Noting the existence of a “rally” phenomenon, whereby a national crisis sparks a boost of support for the leader, this research suggests that leaders may artificially create such a rally through the cynical use of force, to deflect attention away from domestic troubles or improve their polling ahead of an election.18 What makes this type of behavior distinct for this article's purposes is that the electoral consequences of any given military or diplomatic strategy are perceived not as risks, but as incentives.

The empirical record for such cynical behavior is decidedly mixed, however, suggesting that it may be the exception to the rule of constraint. Scholars have repeatedly pointed to methodological problems with the diversionary hypothesis and suggested that leaders rarely have the opportunity or the appetite to engage in such activity.19 Because the expected electoral gain is far from certain—the size of a rally effect and its duration can be small and dependent on factors outside the leader's control—such controversial applications of force typically offer greater political risk than reward.20

Democratic constraint as the prevailing norm thus appears logically robust, and this central logic drives the small but growing literature that sees elections as a force for constraint on the foreign policy behavior of leaders. The most well-known work here is Kurt Gaubatz's study that finds that democracies tend to avoid entering into wars as an election approaches, instead engaging in conflicts more often after polling day.21 More recent work has supported this conclusion, pointing out that leaders are freer to use force after an election than before, because this is where audience costs and other political constraints are lowest in an electoral cycle.22 Other studies argue that leaders facing re-election are typically less belligerent than those not due to face voters again, giving statistical credence to Alexander Hamilton's insistence that imposing tenure restrictions on U.S. presidents would cause “a diminution of the inducements to good behavior.”23

Yet, the existing literature still contains three interrelated problems that this article seeks to resolve. First, the dependent variable in almost all studies of electoral accountability is some formulation of the “use of force,” often meaning the entry of a state into war or the initial resort to military action short of war. What happens, however, when presidents find themselves in an environment where their dual roles as commander in chief and holder of the highest elected office demand equal attention?

Second, most studies tend to treat electoral pressures in an artificially bounded way that reflects the scholar's desire for neat units of analysis rather than the messiness of real life. In many cases, electoral dynamics are accounted for by dummy variables referring to a fixed period preceding an election, typically three months or a year. Even in qualitative work, the focus is too often myopic, centering on presidential elections at the height of the campaign season. This article instead assumes that presidents may be conditioned by electoral pressures in a more nuanced way at different stages of the electoral cycle. For instance, the looming prospect of an election season may be as influential as being in the midst of the campaign itself. Moreover, it seems fair to argue that it is not just presidential elections that matter, but congressional ones, too. Although the president's personal political survival is not at stake, the midterms represent a moment when the balance of power in Congress can change, with a significant impact on the leader's ability to push through legislative priorities for the remainder of their term or pursue any proposed military strategy free of congressional interference. With the president's popularity so tied to their party's fortunes, there is a similar spotlight shone on the actions of the commander in chief during these electoral periods.

These two issues could perhaps be addressed by introducing additional variables into another large-n analysis, but herein lies the third problem with existing studies.24 Overwhelmingly quantitative in nature, the literature has yielded an excellent emerging overview of the dominant patterns and correlational relationships between the electoral cycle and foreign policy. What is missing are convincing in-depth accounts of the causal mechanisms underpinning these relationships. With the insight of the existing scholarship in hand, then, this article uses a case study approach, informed by original archival research and fresh elite interview material, to examine the effect of electoral politics on wartime decisionmaking.25

Electoral Constraints and Decisionmaking in War

Building on the existing insights of the aforementioned literature, I infer two mechanisms that underpin the role of electoral constraints: delayed decisionmaking and a dampening effect.

DELAYED DECISIONMAKING

When presidents are faced with a need to change strategy or authorize a significant military operation in the lead-up to an election, their first preference is likely to be to delay making any decision until after polling day. Even if the proposed action is deemed essential, the timing of that decision may be more flexible. Because the electoral risks of making significant changes to strategy are amplified at this time of highest political sensitivity, the president may be willing to accept some substantive cost of postponement to avoid suffering any political fallout when it matters most.

Not all military strategies carry the same level of political risk, of course. The incentive to delay making a decision will be much higher when the options under consideration include, for instance, shifts to large-scale, casualty-intensive strategies, as opposed to special forces operations that may not even become publicly known. In coming to their assessments of the relative electoral risk versus the strategic cost of delay, each president will also bring different risk-propensities and ideological beliefs to the conflict. A risk-acceptant hawk is thus more likely than a risk-averse dove to make a tough call before an election, for example. Unpopular presidents with little to lose, or very popular presidents with much political capital, may also prove more likely to accept military strategies with greater potential for negative domestic fallout.

Yet, despite these various contextual factors, electoral pressure is just that, a pressure, and not a deterministic phenomenon forcing all presidents to act alike. Just as importantly, using much of the aforementioned logic of electoral accountability, analysts might reasonably surmise that the preference to delay will be common, even among hawkish commanders in chief. Given the prevalence of anti-war voices in domestic political institutions, it does not take much to sour public perceptions of a president's war record.26 It may be true that the public generally knows little about foreign affairs, and cares less, but it has been shown time and again that the salience of military operations can soar— perhaps even out of all proportion—when the number of body bags begins to mount (consider the loss of U.S. soldiers in Somalia in 1993, for instance). Even if highly confident of strategic success, the course of action may be unlikely to yield positive results in time to reap the electoral reward, instead opening space for critics to assail ongoing stalemate.

To prove the existence of this delay effect, there are several observable implications that my case study should be able to identify through a triangulation of original archival material, firsthand accounts by elites involved in the decisionmaking processes, and the secondary record. First, there must be evidence that the president believed that a change in strategy was necessary. Next, one must be able to show that a decision was deferred, in the sense that viable opportunities to subject the issue at hand to normal review processes were not pursued. Relatedly, it is necessary to show that electoral considerations lay behind this delay. Finally, if the decision was truly delayed and not simply ignored, evidence must exist of the same problem being considered in the post-election phase, likely with a degree of urgency not hitherto seen, and a decision being made quickly.

DAMPENING EFFECT

If the president assesses the cost of delaying a decision to be more significant than the electoral risk of doing so, or if a pre-election decision is made necessary by circumstances outside of the president's control (for instance, because of fixed deadlines imposed by international organizations or immediate threats posed by other actors), political sensitivities remain a critical component of the decisionmaking process. Specifically, proposals by military figures and other members of the national security team that are perceived to carry excessive electoral risk will be scaled down, with their most politically unpopular aspects removed from consideration.

This dampening mechanism operates in a fairly straightforward manner. Proposals brought to the White House must pass a test of political acceptability, in which options carrying excessive political risk will be eliminated before any serious in-depth review.27 After this initial screening, the remaining options will be assessed by the administration across a range of metrics, of which political salience continues to be an important one. The remaining proposals will now likely be finessed by advisers to provide what the president perceives to be the most optimal plan in military terms with a sufficiently low level of political risk. If no lowest-common-denominator option can be found that sufficiently minimizes the electoral risk, the president may choose to forgo any proposed strategic payoff and remain with the status quo policy.

George W. Bush and the Iraq Surge

In a primetime televised address on January 10, 2007, President Bush announced the most significant strategic shift of the Iraq War, officially named “The New Way Forward.”28 Under the revised plan, more than 20,000 additional U.S. troops would be sent to the Middle East in an attempt to boost security amid spiraling sectarian violence. Whereas the existing strategy sought to transition security responsibility to Iraqi forces, the surge shifted the focus to a more population-centric counterinsurgency (COIN) strategy with U.S. troops in the lead.

The decision was a long time coming. The following discussion demonstrates that, through the spring and summer of 2006, key actors in the national security bureaucracy, including Bush, perceived a need to try something different in Iraq. Yet, because admitting the existing strategy was failing was politically unwise, and switching to a casualty-intensive alternative at a time of growing public disillusionment with the war was even less palatable, the administration deferred making a course correction. Only after the 2006 midterm elections was a formal review process launched and a presidential decision quickly reached, in a classic illustration of the delay effect.

“IT'S HELL, MR. PRESIDENT”

Some of the first observers to perceive the need to rethink the United States’ Iraq strategy were senior officials in the Department of State. As State Department counselor, Philip Zelikow began to worry during a fact-finding mission to Iraq in February 2005. Perceiving a huge inconsistency of military activity in Iraq, Zelikow concluded that “there really was no strategy.”29 His observations fueled great consternation in Washington: “By the summer of ‘05 and certainly into the fall of 05, the whole circle around the secretary of state were seething with frustration about the Iraq situation. We could feel it killing the administration.”30

While chargé d'affaires in Baghdad in 2004, James Jeffrey sensed Secretary of State Condoleezza Rice's growing pessimism. “By the fall of 2004,” he explains, “Rice was growing far more skeptical about the Iraq adventure … despite all of the money, and all of the committees we had in Washington and at the embassy, and the military running around, not a lot was getting done.” By the time Jeffrey returned to Washington in 2005, “she was probably the most skeptical of any of the officials.” From here, matters only deteriorated further. “There was a tremendous frustration below the scenes that things weren't working,” recounts Jeffrey, who at this point had become Rice's Iraq coordinator. “Things by the spring of 2006 were pretty grim.”31 Rice admitted as much in her memoir: “It was becoming painfully obvious that we had neither the right military strategy in Iraq nor enough forces to carry out the flawed one that we were pursuing … Maintaining the status quo was not an option.”32

Among the National Security Council staff, concerns were raised during the process of writing the National Strategy for Victory in Iraq in late 2005. “During that effort it became clear that there were large disagreements within the administration about what the strategy should be and what the strategy was,” recalls Meghan O'Sullivan, deputy national security adviser for Iraq and Afghanistan, adding, “that whole effort made me uncomfortable about exactly where we were.” After the bombing of the Samarra mosque in February 2006, those initial niggling concerns had transformed into concern about the fundamentals of the course of action the United States was pursuing in Iraq. “These disconnects were hamstringing our strategy. This became really evident in the early part of 2006,” recalls O'Sullivan. “I went out to Iraq in early spring … and it became very obvious to me that our strategy and the realities on the ground were totally out of line with one another.”33 O'Sullivan's colleague on the National Security Council, Peter Feaver, similarly recounts that “over 2006 the Iraqi situation slowly deteriorated to the point where it came to correspond to the seemingly hopeless situation that critics had long claimed it to be … Senior civilians on the National Security Council (NSC) staff (myself included) believed the situation warranted launching a top-to-bottom reassessment of the strategy by late May 2006.”34 On her return to Washington, O'Sullivan and like-minded colleagues began seeking to bring about precisely that outcome, beginning with a memorandum to her boss, Stephen Hadley, in which O'Sullivan all but begged for a strategy review, concluding in bold text, “We are executing a plan based on assumptions that are no longer valid.”35 When President Bush pulled her aside after a meeting and asked what it was like on the ground, meanwhile, O'Sullivan did not sugarcoat her assessment, replying, “It's hell, Mr. President.”36

The intelligence community, meanwhile, was soon issuing similarly dark forecasts. In his summary report offered to the president in June 2006, Director of National Intelligence John Negroponte noted the ever growing violence metrics, which had reached an all-time high. The report zeroed in on the sectarian dynamics then at play, pointing out that vast swaths of disaffected Sunnis were siding with al-Qaida in Iraq (AQI) “not because they find the message appealing, but because they feel that AQI can protect them against Shia militias and the Iraqi Security Forces.”37 By implication, the existing transition strategy was arming and equipping Iraqi forces that, in many cases, were complicit in the violence. Looking back, Negroponte confirms the gravity of the problem. “It had plunged into a very severe situation … we were losing it.”38 With many CIA analysts believing Iraq to be in a state of civil war, Bush received an intelligence update, dated July 20, that read, “The deteriorating security situation is outpacing the Iraqi government's ability to respond … Violence has acquired a momentum of its own and is now self-sustaining.”39

Among the uniformed military, there was also skepticism. “We all knew it wasn't going well,” admits Michael Mullen, chief of naval operations, speaking of the views of the Joint Chiefs of Staff (JCS). “Quite frankly, there was no strategy. That's what drove us crazy.”40 Even the commanding general of coalition forces Multi-National Force–Iraq (MNF-I), George Casey, understood that the nature and scale of the challenge before him fundamentally changed during 2006. “The sectarian violence really began in April or May of ‘05,” he later recalled, yet “after Samarra it was a fundamentally different tone.”41 Although he still ultimately believed in the fundamentals of the underlying transition concept, the threat posed to this plan by worsening sectarian violence and a lack of political reconciliation was plain to see, particularly after the “Together Forward” operation in June 2006 failed to decisively stem the tide of bloodshed in Baghdad. The evolving complexity of the threat was something he had tried to brief in his video teleconferences with superiors that summer, emphasizing how the conflict in Iraq was transitioning from an insurgency against the coalition to a struggle for the division of political and economic power among Iraqis. Notably, the reaction in Washington was muted. “I was saying all the words,” recalls Casey, “but it wasn't connecting.”42

Most important of all, President Bush was aware of a need for change, and by the spring of 2006 began to count himself among the doubters of the prevailing U.S. military strategy. Although a model of sunny optimism in public, in private “the president became very depressed about it,” according to Negroponte.43 “This is a daily conversation in the White House,” adds O'Sullivan; “it was very clear that he was concerned about the situation, worried that it was going in the wrong direction.”44 Indeed, Bush tells us as much in his own words: “In the months after the Samarra bombing, I had started to question whether our approach matched the reality on the ground.” By spring, he had come to a conclusion, telling National Security Adviser Hadley, “This is not working … We need to take another look at the whole strategy. I need to see some new options.”45

“IT WAS A BUST”

If the lack of a strategic review through most of 2006 was not for want of an awareness by the Bush administration of the need for one, neither was it down to a shortage of attempts to spark such a process by concerned officials. In fact, a series of largely independent efforts were launched in the year leading up to November 2006, each of which was left to fizzle out.

At the State Department, the articulation of a strategy of “clear, hold, and build” in the fall of 2005 was perceived by its authors as “the first big effort to tear up and change the strategy.” At a minimum, argued Zelikow, “We needed a story. And finally, by golly, at least we're going to offer a story. Since no one else went out to do it. And maybe that will spur others.”46 Jeffrey confirms that already by this time there was a sense at the State Department that “you should actually do a goddamn counterinsurgency. That is, you do something like the surge.”47 That the NSC picked the mantra up and spun it into the National Strategy for Victory in Iraq was ultimately insufficient for achieving the real change the State Department was seeking. “Except for at the level of rhetoric,” regrets Zelikow, “it's a bust.”48

This articulation of a new strategy aligned with work being done in Baghdad, too. Having himself concluded that “our strategies to date had not worked,” the new U.S. ambassador, Zalmay Khalilzad, commissioned a Red Team study in Iraq, which assessed that the plan devised by General Casey was badly off course.49 Instead, the report advocated the adoption of an “ink-spot” strategy, based on classic population-centric counterinsurgency principles. Around the same time, in November 2005, the Iraqi national security adviser wrote to the U.S. team in Baghdad calling for essentially the same type of strategy, backing the view of experts that decisionmakers should shift their focus “away from hunting down and killing insurgents, emphasizing instead an effort to better protect the population.”50

Into 2006, sensing a new appetite for a fresh look by NSC staffers such as O'Sullivan, Secretary of State Rice asked Zelikow and Jeffrey to develop a proposal for a full military campaign plan. The recommended “selective counterinsurgency” strategy outlined in a June 5 memorandum called for a focus on securing and holding several key areas in Iraq, necessitating “some additional infusion of American forces in the short term.”51 Although no immediate feedback was received, the proposal was fed into planning efforts for a “war council” summit held at Camp David, Maryland, in June, widely expected to be a warts-and-all review of strategy, where the president and his advisers were “really going to have the blunt, all-out conversation that they have long needed to have so this strategy can be pulled up and thoroughly re-examined,” as Zelikow puts it.52 Again, though, “that turned out to be a bust.”53 Indeed, President Bush left halfway through the set of scheduled meetings, using the get-together as a cover for a secret trip to Iraq to meet with Prime Minister Nouri al-Maliki. “What I thought would have started a zero-base review of the strategy miscarries back into debates about implementation on the margins,” recalled Feaver.54

Two months later, on August 17, President Bush chaired a meeting in which he began to vocalize his concerns with the existing strategy. In Bush's recollection, he tried to send a “signal” to his team that he was thinking differently. “We must succeed,” he reports having said, adding, “If [the Iraqis] can't do it, we will … We have to make damn sure we do not fail.”55 Yet, it seems that Bush's memory may have embellished the strength of the signal he sent. General Casey, for instance, still felt that he had the support of the president, which was reasonable to think given that he had been asked to stay on for another year in Iraq a few months prior, and following Camp David, where Casey's update on the existing strategy was approved.56 “And then I read in President Bush's book that August was the worst month of his presidency,” the general recalls, “and I'm going, what?”57

Between the August 17 meeting and the November 2006 elections, the number of closely held studies looking at Iraq strategy proliferated among the various skeptical groups. Yet, the establishment of a full interagency review, which was critical in bringing about real change, was again deferred. An internal NSC effort to consider alternatives was held very closely, and only gradually opened up to State Department officials on a strictly informal basis. Unsurprisingly, perhaps, the importance of having military representatives in the room quickly became apparent. “We can only go so far,” explained David Satterfield, a senior adviser to Secretary of State Rice, adding, “We can't make judgments about military force capabilities, about readiness capabilities … It's the chiefs, J-3 and J-5”—the directors of operations, and strategy, plans, and policy—who could say “what exists out there.”58

In military circles, at least three studies were under way, none of which was formally connected to the other, let alone to the efforts of the NSC and State Department. By early November 2006, the stark assessment of the “Council of Colonels” group, established in the fall at request of the chairman of the Joint Chiefs, had reached the JCS: “We are not winning, so we are losing.”59 Meantime, Gen. David Petraeus was shaping revisions to the U.S. Army's counterinsurgency field manual in ways that foreshadowed the surge strategy. Although the manual was designed for much broader application, its lead author admits that “Petraeus saw this as a tool for Iraq … It was very obvious to me by the time we got to the late summer that this was going to Iraq and he was going to use it in Iraq, and that's what he was shaping it for.”60 Outside the chain of command, meanwhile, retired Gen. Jack Keane had begun leading a “cabal” aiming to trigger a shift toward a similar more troop-intensive COIN strategy, briefing Secretary of Defense Donald Rumsfeld—and indeed anyone who would listen—on the need to send additional troops in support of a revised plan.61

Had each of these disparate studies and reviews been formally brought together earlier, a full airing of views might have led to a change in strategy much earlier than it ultimately did. In a normal interagency review process, explains Zelikow, “The first thing you do is run a subcabinet process that elicits papers and arguments about strategy from State, Defense, JCS, maybe convened under the deputies and J.D. Crouch. J.D. Crouch never ran such a meeting until like the end of November. Not one.”62 As a result, dissatisfaction with the existing strategy remained isolated among disparate silos in the bureaucracy, without constructive interagency analysis and input that would help move the concerns toward a solution. Before November 2006, adds Zelikow, “That just simply never happened. These papers don't exist. At no point before the end of November of 2006 did I ever see a paper and an interagency process in which DoD [Department of Defense] or the command are being forced to engage in the arguments about their strategy.” The net effect of the White House's failure to launch a formal strategy review, then, was the deferral of any real decision on a new strategy. “Everything that happened then could have happened more than a year earlier, or two years earlier,” regrets Zelikow.63

“I DECIDED TO WAIT”

Some of the best evidence that the delay in authorizing a formal review was caused by electoral constraints appears in Bush's memoir. In a key passage explaining his thinking in the fall of 2006, he wrote, “I decided a change in strategy was needed. To be credible to the American people, it would have to be accompanied by changes in personnel … [But] with the 2006 midterm elections approaching, the rhetoric on Iraq was hot … I decided to wait until after the elections to announce any policy or personnel changes. I didn't want the American people to think I was making national security decisions for political reasons.”64 Putting aside the somewhat troubling assertion that timing critical decisions on strategy based on the domestic political calendar was somehow an example of keeping national security policy free of political considerations, this statement alone is highly suggestive of the delay effect.

Secretary of State Rice, meanwhile, though convinced of the need to change strategy, reportedly said at the time that she was not willing “to do anything that would be above the radar screen in the heavy political breathing of the November elections,” because the last thing the administration needed was “a hothouse story” that revealed that things were so bad in Iraq that the administration was looking for alternative approaches.65 Hadley agreed, telling Rice in October 2006 that discussions between the NSC and State Department concerning the situation had to remain informal: “We've got to do it under the radar screen because the electoral season is so hot.”66 Earlier that summer, Hadley reportedly elaborated on his thinking: “I've got to help this President get through what is going to be a really rugged three years. And if the Democrats take over the House and the Senate it's going to be unbelievable after 2006.” If there was a “raging debate” over Bush's war record with people arguing “we have no choice but to throw the Republicans out and bring the troops home … this is really going to be awful.”67 Looking back, O'Sullivan acknowledges that an important factor behind her inability to spark a formal review process was that “this is the summer before the midterm elections. And I think that it would have been potentially awkward for there to be a strategic review of our Iraq strategy publicly out there in the run-up to the elections.”68 Col. Peter Mansoor, meanwhile, a member of the Council of Colonels, writes of his similar view of the reason why the several informal reviews were not linked up: “With the midterm elections less than three months away … any hint that the administration was having second doubts about Iraq was politically radioactive.” Given fears of an explosive fallout from any leaks from a full review, “serious debate on changing course would have to wait until the voters had had their say.”69

At a minimum, this evidence appears compatible with the argument of John Mearsheimer that wartime democratic leaders may be inclined to cover up failing policies to maintain public support at home. As part of a “strategic cover-up,” the Bush administration may have felt that the public opprobrium sparked by any acknowledgment of failure would undermine public confidence at home and, as such, cause more damage to the war effort than continuing with the existing approach.70 Alternatively, in an “ignoble cover-up,” Bush and his advisers may have had more self-interested reasons, seeking to avoid electoral punishment for the failure of the existing strategy.71

In reality, however, these concerns related as much to the likely public reaction to a future policy as the optics of the existing one. Given that Bush was already apparently inclined toward escalation rather than withdrawal, the political logic of delaying such a move was reasonable enough.72 After Congressman Jack Murtha's powerful demand for an immediate withdrawal in late 2005, the center of gravity in the Democratic Party was by the following summer firmly in favor of a scaling down of the U.S. military presence in Iraq, with a dozen senior figures writing a scathing open letter to Bush on July 30 calling for a shift toward a “more limited mission focused on counterterrorism, training and logistical support of Iraqi security forces.”73 Even senior Republicans, including the chairman of the Armed Services Committee in the Senate, John Warner, were writing anxious memoranda outlying their concerns with any hint in the rumor mill that additional forces were being readied for deployment.74 In September 2006, meanwhile, Senator Mitch McConnell pleaded with the president to withdraw some troops from Iraq at least as a political token.75 Among the public at large, appetite for something like the surge was vanishingly small; a poll in June 2006 showed that just 6 percent of Americans favored an increase in troops.76

The president and key advisers privately understood that any possible escalation would arouse a firestorm of public criticism. “The surge option brought risks of its own,” recalls Bush. “Increasing our troop level would be deeply unpopular at home. The fighting would be tough, and casualties could be high.”77 In response to McConnell's plea, meanwhile, he later admitted that it was apparent that opting to increase, not decrease, troop levels would be “the most unpopular decision of my presidency.”78 Meghan O'Sullivan, one of the chief advocates of the surge in Bush's inner circle, acknowledges the situation as follows: “Initial opposition to the idea of the surge was virtually uniform. With a few notable exceptions, such as [Senator] John McCain, members of Congress from both parties argued strongly against it. Public opinion was firmly in the camp of bringing troops home. Even many senior members of President Bush's own administration were, throughout much of the process, arguing against the surge.”79 And in Baghdad, the constraints the political leadership were under could be sensed, with one of General Casey's aides later recalling, “I guess the read of domestic public opinion at the time, or even looking back on it, I don't thin[k] the people would have supported an increase in forces, it had to wait until after the election.”80

At the State Department, concerns about the deteriorating state of public support for the war led key figures to retreat altogether from their earlier advocacy of a better resourced COIN plan. “My concern, which led me and others around Rice to become skeptical of our own recommendations,” recalls Jeffrey, “was based on rapidly fading domestic support … We were all growing skeptical that we would have the public support for anything.”81 Indeed, before long Rice would be defending a proposal that contemplated the withdrawal of U.S. troops to the outskirts of the cities in Iraq and intervention only in cases of uncontrollable, near-genocidal violence. Why? Because “we lost the US people when we got into a one thousand year blood feud. They understood going after al Qaeda; understood Sadaa[m]ists; understood all of that. Do not understand this blood feud. Must … realize our limitations.”82

Not only would the initial decision to escalate have been unpopular, but from a communications perspective, such a strategy would only make things worse before they got better. The population-centric COIN approach, which according to multiple sources caught Bush's eye through the summer of 2006, was almost by definition a casualty-intensive military strategy.83 “We understood the inherent violence in a counterinsurgency,” notes Conrad Crane, the lead author of the U.S. Army's doctrinal manual, adding, “we use the word kill like seventy times.”84 If the experts could effectively guarantee the president that such a strategy would incur more casualties in an election year, what they could not do was offer any real hope that conclusive results would be seen in the short term. “We were pretty honest in the manual; it was going to take a lot of killing, a lot of violence, it's going to take a lot of time,” says Crane.85

Instead of pursuing alternative approaches, the Bush administration acted as one might expect in the lead-up to an election by diverting attention toward the minimization of casualties associated with the existing plan. This, indeed, is one of the aspects General Casey sensed in his interactions with Washington. In handwritten notes of one such call, the commander listed a series of questions he had been asked: “Can we shift weight away from activities that cause casualties?” “What are the risks of going to a lower visibility posture now/6mos/12mos?” “Less troops = less casualties?”86 Secretary of Defense Rumsfeld had, of course, always believed in pushing the responsibility for security onto the Iraqis; yet in the summer of 2006, his concerns seem notably driven by worries about congressional reaction to the continued exposure of U.S. troops to violence in Iraq. Hence, when General Casey reluctantly decided to extend the tours of the 172nd Stryker Brigade to shore up security in the short term, Rumsfeld fired across a note lamenting the move as “unfortunate,” as the administration was now “facing some difficulties in Alaska and Congress because of it.”87

The impact of casualties on other senior Bush administration officials’ concerns was also significant. Looking back at 2006, Rice remembers how “each morning I opened the Washington Post to ‘Faces of the Fallen.’ The Post had begun the series in 2003 to memorialize Americans killed in the war. I made myself look at every one of the photos—a harsh reminder of the costs of the war.”88 Explaining why she began to reconsider her earlier proposals for a better-resourced COIN plan during the summer, she notes that her growing concern was that putting more troops in harm's way “would only result in more casualties.”89 Her new proposal—threatening to withdraw U.S. support—might not work, she admitted, but “at least this way, fewer Americans would die.”90 As Casey himself later lamented, speaking of the views of political leaders in Washington, “Over time, casualties and violence became the de facto measure of strategic progress in Iraq.”91 Such heightened casualty-sensitivity during an election season is exactly what theories of electoral constraint predict, as the administration seeks to minimize the prospect of public backlash on the eve of Election Day.

BUREAUCRATIC BOTTLENECKS

The two most commonly cited alternative explanations for the Bush administration's delay in changing the U.S. military strategy in Iraq are wholly interconnected with related electoral concerns. The first is nicely illustrated by Zelikow's judgment that “Don Rumsfeld did more than any other single person to destroy the Bush presidency.”92 For James Jeffrey, Rumsfeld was “the eight-hundred-pound gorilla” who blocked all efforts to change strategy, on the basis on his view that “shit happens. Chill out. We've got a plan, which is to build up the Iraqi army [and] get out of there.”93 When fellow members of the administration challenged the commanders working on Iraq at weekly videoconferences, Rumsfeld would often stand in the way of attempts to elicit a full discussion of views, on one occasion even writing a follow-up memorandum threatening that “if NSC meetings are going to have the tone of the last one, I am going to have [commander of U.S. Central Command (CENTCOM) Gen. John] Abizaid and Casey not on the meetings.”94 For good reason, then, when the State Department was developing its own “selective counterinsurgency” proposal, Zelikow recalls admitting to Rice, “The strategy we have in mind, we will not be able to execute this with Don Rumsfeld as the secretary of defense … It's just not going to happen.”95

Yet, if Rumsfeld alone was the cause of the bureaucratic bottleneck, Bush had had a number of opportunities to remove him, but waited until the day after the midterm elections to do so, which suggests that political timing was wholly interrelated in the matter. “He had been skeptical for [the] better part of a year,” notes Jeffrey, “but [he was] not willing until after [the] elections to take on Rumsfeld and [Vice President Dick] Cheney.”96 Rumsfeld had offered to resign on a number of occasions before this time, including March 2006, a month after the Samarra bombing in Iraq. With Bush senior adviser Karl Rove arguing that firing Rumsfeld would place an albatross around the necks of Republicans in the congressional election campaigns, the politically safer thing to do was to wait it out, thereby lengthening Rumsfeld's grip on Iraq strategy.97 As emphatic as Jeffrey is about the detrimental role that Rumsfeld played in this decisionmaking process, he concludes, “This all broke because of the election. The election was a seminal event.”98

Another alternative hypothesis identifies the source of the delay as Bush's fear of overruling the uniformed military, which remained fairly solidly against a troop increase, at least within the formal chain of command. Admiral Mullen understood the problem as follows: “As powerful as presidents can be, they know the importance of having the military on board. By on board, I don't mean actively espousing policy. Military leaders shouldn't do that. But it does help inoculate the president from criticism if he can say he has consulted the Pentagon and his commanders helped inform the policy and are comfortable carrying it out. I think President Bush understood this dynamic well and did not want to get himself into a position where people could accuse him of bullying the chiefs or pulling us through a knothole to get our support.”99 With his vantage point at the NSC, Feaver speculates that the delay may well have come down to a “respect or fear” of opposing General Casey or to concerns about a “revolt of the generals” among the JCS. This was why Bush turned to studies by retired senior military officers, outside the formal channels, using the ideas contained therein “as stalking horses to elicit JCS views gradually,” rather than confronting the set of four-star generals with his own views outright.100

Yet, again, this explanation cannot be disentangled from the electoral context. On the most basic level, if the risk of upsetting the generals was political in nature, as Mullen notes, then it is logical for the president to have waited until after the period of heightened political sensitivity—during election season—before challenging military opinion. As President Harry Truman found out after firing Douglas MacArthur, the widely popular general, in 1951, having a vocal critic with a litany of stars on his shoulder may prove politically disastrous. Furthermore, the narrative of a slow, gentle process of consensus building among the military by Bush does not fit with the recollection of the key generals involved. Having been excluded from all of the informal reviews, General Casey identified the November elections as a crucial turning point in Bush's appetite for something new: “It was the first videotelecon we did after the midterms. It was noticeable because he was always very ebullient. And he wasn't. He was just very direct and matter of fact. That was when I first kind of said, ‘ok, there's something up here.’ It was November, right in that early part of November … I honestly didn't perceive any change in his attitude until after that midterm election.”101

A similar point could be made about the perspective of the JCS. As stated earlier, the Joint Chiefs had not been made aware either of Bush's doubts through the spring and summer of 2006 or of the informal studies undertaken into the fall. Indeed, the Council of Colonels was an independent review running on parallel tracks to the concerns of the NSC or the State Department. Both within that review and among the JCS more generally, there was limited appetite for sending additional brigades. “We thought,” says Mullen, “just from a readiness standpoint, the shape the Army was in, that two was about max.”102 The five-brigade option that the Bush administration chose came from outside the chain of command, and was “180 degrees out of sync with the uniformed military's view of things,” recalls Lt. Gen. Douglas Lute, then director of operations of the Joint Staff. “There was no appetite to surge among the JCS or the operational commanders,” he confirms, adding, “it was imposed by the president.”103 This was a story not of a president gently bringing his chiefs on board, but rather of a president confronting them with a fait accompli after the election.

Finally, instead of a long period of gentle persuasion, it appears that the enticement that brought the chiefs on board was Bush's offer in a meeting of December 13, 2006, promising more resources for the Army and Marines. The pledge to pump resources into the depleted forces not only placated the joint chiefs, who were responsible for allocating those resources, but also served to further mollify the opposition of the commanding general. When Casey drafted a memorandum for the president, stating his objections to the proposed surge strategy, CENTCOM Commander Abizaid advised him to scrap the letter, explaining that it was a done deal. “Look, this is all tied to decisions about increasing the size of the Army and the Marine Corps,” Abizaid told Casey, “It's done, get out of the way.”104 What, one might reasonably ask, stopped the president from offering these before this point? Lt. Gen. John Sattler, at that time the J-5 responsible for proposing strategy, plans, and policy to the chairman of the JCS, admits that “had we decided to grow the Army and the Marine Corps earlier, then that may have been a more viable option earlier than it was.”105 Absent Bush's caution regarding the political fallout of countermanding the JCS, then, there was a plausible route to overcome their objections much earlier. That Bush waited until after Election Day points decisively to the existence of the delay effect in operation.

OPENING THE WINDOWS

Election Day 2006 acted as a turning point, giving way to decisive and bold decisionmaking behavior by President Bush. Although the decision to surge U.S. troops was announced in early January, the record indicates that it had effectively been made by the time Bush visited the JCS on December 13, just five weeks after the congressional elections.

On November 10, three days on from the midterms, and two days after firing Rumsfeld, Bush appointed J.D. Crouch to lead a full review of strategy in Iraq. A flurry of interagency meetings subsequently took place, bringing together several of the informal efforts and injecting them with a sense of urgency not hitherto seen—Bush wanted a report in only sixteen days. “Basically the day after the election … it's as if all of a sudden the windows had been opened and fresh air is coming in,” recalls Zelikow.106 Having been largely deferential on questions of military strategy up to this point, Bush was suddenly front and center of the deliberations. During the initial stages of the review, this new posture was enunciated by Hadley, who, at Bush's behest, intervened in one of the first meetings to instruct the participants that, whatever their own proposals were, “you have got to give the president the option of a surge in forces. He will want to see it, and he'll want to know what it means. You all can take your positions for or against or in between, but you have to present him that as an option.”107

By the time the review moved from the deputies to cabinet officials, the key follow-up meeting minutes reveal a president eager to push ahead with bold steps. On December 8, for instance, Crouch had barely begun his presentation on the propositions under discussion before Bush challenged his suggestions, asking, “What is different from now? Don't see anything different?”108 Rejecting the State Department's new plan to step back and let the violence burn itself out as an abdication of responsibility at best, Bush turned to Casey, who reiterated the intention to press on with the transition of security responsibilities to Iraqis, as agreed with the Prime Minister Maliki and foreshadowing key tenets of the “Accelerated Transition” plan that Casey was scheduled to formally brief a few days later.109 “So six months more of the same stuff?” asked a seemingly incredulous Bush, before showing his own hand by asking, “Does this argue for more US forces as a bridge?”110 Casey later conceded that “our life changed after those elections … it was clear to me that the President was out of patience with the Iraqis to do anything, and we needed to focus on securing Baghdad.”111 Although Casey briefed his subordinate commander in Baghdad to think creatively about U.S. military strategy, it soon became clear that the president “was already doing it.”112

Referencing the late-2006 report of the Iraq Study Group, which effectively supported the central tenets of the existing strategy, Bush, on December 9, spoke of the “radical action we will take [to] achieve victory,” imploring his team to “understand that Baker-Hamilton [is] not setting orders. I am running this show.”113 Then, over Casey's objections, he questioned whether the Iraqi Security Forces could assume responsibility as early as proposed. “If forces [are] unreliable to protect Baghdad,” he said, it would amount to “a lot of wasted effort.” Instead, he thought aloud, it “seems [we] should move more resources into Baghdad … Do we need more troops?” Bush made plain his state of mind: “People concerned US not engaged. Think about symbols. Show muscle … Most important is put us in position where can win. Want to win. No mistake. Victory. Surge important.”114

It was not just the speed or decisiveness with which Bush reached a decision that is notable, but also the bold and risky nature of the surge option, which the president selected over an enormous amount of opposition. Having now passed the moment of highest political sensitivity, there was no reason to resort to half-hearted measures. “It's a Hail Mary pass,” says Zelikow, “because he's throwing the ball at a time where he doesn't really have a play, where he's got the receiver in mind … it was really up to Petraeus and [Lt. Gen. Raymond] Odierno to figure out how to deal with it and make it work.”115 Yet Petraeus's executive officer, Peter Mansoor, admits, “We did not know whether the surge would work; indeed, the odds were stacked against us.”116 Even the architects of the new COIN doctrine that underpinned the surge had their doubts. “Baghdad was a mess,” admits Crane, adding, “I thought you had to consider the possibility of failure.”117

Bush's release from electoral constraints is also arguably what explains the determination to go with a five-brigade option, rather than the two that the JCS said was the most they could accommodate. Lute, who originally was at the forefront of those arguments against a large injection of troops, later discussed the logic as follows: “We're going to get one shot at this. It's going to be politically tough in any case, so we're not going to get multiple bites at this apple. Do it properly: send in five.”118 This dynamic also explains the clean sweep of personnel over the winter of 2006. By the time the surge units arrived in Iraq, the secretary of defense, CENTCOM commander, MNF-I commander, and ambassador to Iraq had all been replaced.

In choosing the all-in, full-throated gamble that the surge represented, Bush also demonstrated a willingness to overrule virtually his entire national security team, in another indication of the impact that the lifting of electoral constraints can have on decisionmaking. In addition to the aforementioned opposition from the JCS and the State Department, it was clear that Casey was briefing a plan diametrically opposed to the plan Bush had in mind. During his meeting with Bush on December 12, Casey emphasized that additional U.S. troops would at best have “a temporary, local effect” in reducing violence, would not be decisive absent progress on political reconciliation on the part of the Iraqis, and in the meantime would result in additional casualties and extend the time it would take to pass on security responsibilities.119 Furthermore, Defense Secretary Rumsfeld agreed fully with Casey at the time of the formal strategy review. In a videoconference with Casey, Rumsfeld addressed the prospect of sending more troops, simply stating: “Doesn't help. Feeding alligator … Want more Iraqi troops.”120

There is even some question as to whether Prime Minister Maliki supported the proposed surge. Bush states that he secured Maliki's blessing during a November 30 meeting in Amman, Jordan.121 Yet, the main subject of these discussions was a new Baghdad security plan, a strategic concept created by the Iraqis that would see the Iraqi security forces take over more security functions from the United States. Maliki's response to Bush's offer of additional troops was negative. “I don't need your forces,” he said, “We can do it ourselves. We should do it ourselves.”122 The record shows that the Iraqis had been eager for a U.S. withdrawal to begin as soon as possible, with rough timetables for such an eventuality being proposed by August 2006.123 In truth, Bush seemed determined to drag Maliki along kicking and screaming if necessary, as the minutes of Bush's mid-December NSC meetings show. Told of Maliki's reservations about the surge, Bush grew frustrated, saying “he needs to do this with some speed … [we] cannot let Maliki decide if we surge.”124 Bush demanded that the Iraqi prime minister publicly call for additional U.S. troops, and if he continued to stall, Bush was apparently even willing to consider his removal, at one point saying, “If he's not [the] guy, then we've got to make [a] change.”125

The 2006 elections were thus a key turning point in the decisionmaking process on Iraq. Yet, it was not the case that Bush became inclined to simply reflect the concerns of the electorate, which had signaled its dissatisfaction with the president's war record by handing Democrats sweeping gains in the midterms. Instead of embracing elements of the Baker-Hamilton report, which enjoyed solid and bipartisan public support, the president doubled down. The real significance of the midterms, then, was that only now, as a lame duck newly freed of electoral constraints, could Bush make what he understood to be “the most unpopular decision of my presidency.”126

Barack Obama's Decision to “Go to Zero”

For the next five years, U.S. military strategy in Iraq would largely continue on the path determined by President Bush, owing to a series of actions taken to lock in the surge strategy. The most notable of these was the signing of a status of forces agreement (SOFA), in December 2008, which contained a deadline for the withdrawal of all U.S. troops by the end of 2011. That President Obama would decide to “go to zero” in October 2011 was by no means foreordained, however. In fact, as John Brennan recalls, “I always thought it was going to be a question of how much or how little would remain, as opposed to leaving … I was, frankly, surprised, when it went from a discussion of how many troops to just pulling out. I didn't see that coming.”127

Through an analysis of the decisionmaking process in this period, drawing on the insights of those directly involved, a picture emerges in which the electoral cycle played a key role in conditioning the Obama administration's willingness to leave a residual force and negotiate a renewed SOFA to cover their presence in theater. Having deferred the issue until after the 2010 midterms, the White House belatedly picked up the question of a follow-on force over the winter. Then, as the 2012 campaign season drew nearer, the political optics of leaving troops in theater began to outweigh the perceived benefits to the administration. As such, in an iterated process typical of the dampening effect outlined earlier, the White House rejected a series of proposals related to the size of the follow-on force. The administration's appetite for retaining troops in Iraq diminished over time until finally. And against the preferences of a strong majority of national security officials, Obama decided the effort of pushing through even a small residual force was not worth the political backlash it might engender from his Democratic base, to whom he had promised a full withdrawal from Iraq.

KICKING THE “POLITICAL FOOTBALL” DOWN THE ROAD

To begin with, the Obama administration avoided all talk of a continued U.S. military presence in Iraq during the 2010 campaign season, in another sign of the delaying effect that electoral calendars seem to have on decisionmaking in war. Deciding on what came after the December 2011 withdrawal deadline was understood to be time sensitive. Unsurprisingly, the relevant combatant commanders brought the issue up repeatedly when Pentagon officials visited Iraq, stressing that “we're going to have to start thinking about this, we're going to have to start deciding.”128 Diplomatically speaking, given that the previous SOFA had taken almost two years to negotiate, there was also good reason to think that starting early would be essential to reaching an agreement.

Colin Kahl, a senior Pentagon official, explains why the process did not begin until the winter of 2010: “It was very clear that the system was to have no discussions of a follow-on force until the Iraqi government was formed and we got past the midterm elections, our own midterm elections, so that politics didn't enter into the equation … we didn't want this to become a political football in the midterm elections and [therefore] we weren't going to talk about it right on the eve of our midterm elections.”129 Although quiet planning was taking place in Pentagon and military circles, the White House did not seem keen to discuss the idea. As Flournoy observed, “It took a while to get folks really focused and engaged.”130 By delaying the real start of the interagency decisionmaking process until the new year, “we were already on the clock,” admits Kahl.131

WITHDRAWAL BY A THOUSAND CUTS

After the 2010 midterms, the Obama administration could begin discussions of the politically sensitive issue of whether to keep U.S. troops on in Iraq. Unlike Bush, however, who entered a lame-duck period after the 2006 midterms, Obama still had to face the voters in a re-election campaign in two years’ time, so electoral constraints did not simply dissipate. Instead, the reference point changed. Now looking ahead to the 2012 presidential race, Obama had to approach the decision at hand with a keen eye to minimizing his electoral risk.

For James Jeffrey, the U.S. ambassador to Iraq during this period, this explains Obama's initial appetite to consider a follow-on force. In essence, a residual force could work as an insurance policy against a deterioration of security in Iraq in the lead-up to the 2012 campaign: “It's late 2010, you're going to be running for re-election in 2012…. The main way you think about it is not what great things you can do for the American people … [but] what can screw this thing up, where can I trip up and fall? And that's foreign policy. That's Harry Truman in 1952. That's LBJ in 1968 … Foreign policy really can blow up your presidency. Knowing that … he was like, ‘I don't want anything to go wrong that I can be blamed for having been the cause of it. Iraq seems to be going well. Why not try to keep troops on, so that the troops might be able to fix something that starts going wrong?’ Iraq had turned out to be a ‘good news’ story, much to Obama's surprise, and he didn't want to have it blow up just before the 2012 elections and get blamed for it. So, probably against his deepest instincts, he agreed to try to keep troops on.”132

This administration's apparent openness to consideration of a follow-on force was consistent with the consensus among both the military and several civilian officials in the administration that a continued U.S. presence in Iraq was essential. For a start, during this period, both Secretary of Defense Robert Gates and his successor, Leon Panetta, were staunch advocates of a residual presence.133 Even those who were instrumental in facilitating proposals for lower troop numbers and reprioritized missions still believed in the value of retaining some kind of residual capacity on the ground. For the vice chairman of the JCS, Gen. James Cartwright, concerns about the amount of intelligence, surveillance, and reconnaissance resources available dictated a smaller force structure. But even in a downsized capacity, the role of special forces, air support, and intelligence was deemed critical as an “enabler” for local security forces. Although many took issue with Cartwright's topline figures and manner of briefing them, there was little debate about the importance of retaining at least these functions; “I got really no pushback on the tailoring side of this question,” the general recalls.134 Brennan, now Obama's chief counterterrorism adviser, similarly cautioned against a full withdrawal: “I was a supporter of maintaining a presence, mainly to enable the Iraqi forces to be able to transition to full time and complete responsibility on the counterterrorism front,” he recalls, “I was in favor of the train and advise/assist mission continuing.”135 Vice President Joe Biden reportedly saw similar value in a follow-on force, too. Looking back, his national security adviser, Antony Blinken, listed as a key priority “counterterrorism, and making sure that we continued to have some forces on the ground that could work with the Iraqis, to continue to train them, to prosecute the counterterrorism mission. We were very clear with the Iraqis that while al-Qaida in Iraq was down and out, it was not dead, and if they did not keep their foot on the throat, it could come back.”136 In short, there was very little enthusiasm for “going to zero” at the outset of this decisionmaking process.

Yet if Obama, a president whose political career rested in large part on opposition to the Iraq War, was now willing to accept some troop presence after 2011, he was not about to write a blank check to the military. Indeed, in the year that followed the November 2010 elections, proposals would be screened for their political acceptability, with troop numbers repeatedly cut in a manner that correlated well with the increasing proximity of the 2012 campaign and resurgent political sensitivity that came with it. When Gen. Lloyd Austin, commander of all U.S. forces in Iraq, first recommended a 20,000 to 24,000-strong residual force, Pentagon officials, conscious of the likely reaction of the White House, were alarmed. In Kahl's telling, as soon as Austin mentioned the headline figure, his boss intervened and said words to the following effect: “‘Lloyd, this isn't going to work. A president who campaigned on not leaving a Korea-style presence in perpetuity in Iraq is not going to leave a Korea-style presence in Iraq. That's not going to happen.’”137 Indeed, Flournoy recalls that “it was just not even in the ballpark from a White House perspective … The initial number that CENTCOM put forward was way beyond what they were expecting, so DoD was asked to go back to the drawing board and massage the tasks and the role of a follow-on force, and redo the troop to task analysis, to try to come up with something that was more modest but would still fulfil the mission objectives.”138 The military's first proposal, then, had categorically failed the screening test of political acceptability before even reaching the president's desk, as the dampening effect would suggest, and the search for ever lower numbers began in earnest.

In the following weeks, Austin presented a revised set of options, stating his preference for 19,000 troops, a minimum acceptable option of 16,000, and a bare-bones proposal of 10,000. The White House, however, recoiled at what it still perceived to be excessively high numbers. Within a few weeks, the proposals on the table at an NSC meeting on April 29, 2011, had shrunk again as low as 8,000, with a ceiling of 16,000.139 The apparent allergy to the military's recommendations prompted Admiral Mullen, who had become chairman of the JCS, to author a rare memorandum laying out his concerns. Mullen was, as he recalls, squarely behind Austin's prior assessment: “Basically the ten [thousand] was equal to zero. You can put the ten thousand there, but we can't accomplish the mission with that. That was the floor. This was an important part of the process: each time the number got smaller, I pushed hard to say ok, you've got to tell me what the mission is here, because if it is twenty-eight or twenty-five, that mission is different from fifteen or sixteen, and basically from ten on down, the mission assigned could not be achieved. It's a physics problem at that point, pure and simple. And it would have been irresponsible, if infact dangerous, for me not to say so.”140 Mullen's memorandum was as unwelcome as Austin's own recommendations, landing at the Oval Office with “a thud; they didn't want to hear it.”141 According to later accounts, the president's national security adviser, Thomas Donilon, felt “the military was boxing in the White House and creating a potentially political liability. Imagine if it leaked! … The White House did not like it when the military delivered PowerPoint briefs with ‘high risk’ stamped in red on the low troop options, and Donilon liked Mullen's letter even less.”142

Fortunately for Donilon and his colleagues, one senior military officer was prepared to speak up for the politically appealing lower figures the White House favored. General Cartwright wrote a dissent from Mullen's memorandum and, in a series of briefings with White House representatives, including the vice president, proceeded to explore several troop-level options as low as 3,000, or even 1,000 with additional forces just outside theater in Kuwait. This idea of a rapid reaction force based just across the border captivated Donilon, even though it was later denigrated by Mullen as a “half-baked idea” that was “blind to reality.”143 Indeed, according to Cartwright himself, “My position was not well received by the military … it was just heresy, and remained so.”144 Still, the headline figures were lower, and as such, Cartwright's proposals got traction among the political types. Having a four-star general endorse smaller force structures was critical in legitimating the White House's embrace of numbers lower than the rest of the military was comfortable with. In a meeting on May 19, 2011, President Obama signaled that he was happy to proceed with a proposal of up to 10,000 troops.145

With several U.S. delegations now authorized to begin discussions with the Iraqis about a renewed SOFA, important political and military figures in both the United States and Iraq remained convinced of the strategic logic of a sizable residual force. During the spring and summer of 2011, the chief of staff of the Iraqi army and commanding general of Iraqi special operations forces publicly spoke of the need for continued U.S. presence, even until as late as 2020.146 Senior civilian figures, including Iraq's foreign minister, had also relayed to the U.S. ambassador that they had been clear with Prime Minister Maliki that Iraq “needs the United States to stay,” and that, in turn, Maliki had responded, “Of course we need them to stay.”147 General Austin was similarly inclined, and so delayed the beginning of the drawdown, keeping almost 50,000 troops in theater as late as August 2011 so as to provide “flexibility” for the prospect of a bigger follow-on force.148 The need to stick at least to the latest iteration of the plan was underscored in a U.S. military report of July, which noted that the previous quarter ended with the highest number of combat deaths in three years, concluding “that the conflict in Iraq is not yet over.”149

Yet, congressional pressure to reduce government spending amid a spiraling budget deficit soon began to cause the administration to become uncomfortable with the financial and, by extension, political cost associated with keeping 10,000 U.S. troops in theater. In his memoir, Leon Panetta laments how war-weary Democrats and small-government Tea Party Republicans united on the desire to reduce the “staggering investment” of troops and materiel in Iraq as a way to make overall budgetary savings. “In a town where the two parties agreed on almost nothing,” Panetta wrote, “they suddenly settled on the consensus that we could dramatically cut defense spending during wartime.”150 It was in this context that Tony Blinken and Denis McDonough took a tour of Iraq, meeting commanders and exploring options for alternative force commitments. According to Blinken, “We came back with the conclusion that the initial footprint that we were looking at was probably larger than the Iraqis could digest, and we wanted to get to yes with them.”151 Former U.S. Ambassador Jeffrey accepts that on one level, this estimate is fair (notwithstanding the important point that privately Iraqi officials were prepared to have as many as 20,000 if they could get it through their parliament). All the same, for Jeffrey, “they were the skeptics,” who, faced with the task of justifying lower numbers to the majority of the uniformed military who were opposed to them, were only really in Iraq “looking for reasons.”152 Perhaps unsurprisingly, they found them. Blinken and McDonough argued that if the U.S. military footprint around checkpoints in the north could be scaled back, the residual troop presence could go as low as 3,500, with an additional rotating component of 1,500. Obama endorsed these proposals in a meeting on August 13, 2011.

PULLING THE PLUG

On October 21, 2011, President Obama and Prime Minister Maliki concluded that negotiations over a renewed SOFA were at a dead end. And because the Pentagon's lawyers deemed it essential to have the Iraqi parliament approve the agreement, the formal U.S. military commitment to Iraq would cease at the end of the year. For some senior administration officials, it was the political climate in Iraq that ultimately settled the question. Blinken argues that “the president really was open to this smaller footprint in terms of a follow-on force … but the caveat was that this had to be something that the Iraqis embraced politically, because we didn't want to put our forces where they really weren't welcome.”153 To Blinken, Maliki's insistence that he would be unable to get U.S. demands on legal immunities through the parliament indicated that a continued U.S. presence was unwelcome. Lute, who was instrumental in negotiating the original SOFA, paraphrased the Iraqi view in blunter terms as akin to saying, “You know, we can argue about this, but you guys need to start packing your stuff.”154

Yet, if Obama really wanted a deal that badly, there were other ways to achieve it without parliamentary approval, which was, after all, a U.S. demand. Maliki, for instance, was willing to consider signing an executive memorandum of understanding governing the U.S. troop presence, something supported by a senior member of the U.S. delegation, Brett McGurk. As Flournoy recalls, although the lawyers had valid points, “it's hard to count the number of places where we've accepted a degree of ambiguity or just a complete absence of an agreement to do what we thought was necessary for our mission.”155 One of those places, of course, is Iraq, where thousands of troops have been stationed without formal legal protections since being sent back by Obama to confront the Islamic State in 2014. “What I came to believe,” continues Flournoy, “when 25,000 was too much, 20,000 was too much, 15,000 was too much, 10,000 was too much—was to realize that the president had made up his mind and he didn't want to have any U.S. troops remain in Iraq. And though I certainly believe that it's critical for U.S. forces to have legal protections anywhere we are, we have had many, many exceptions to that rule, where we have deployed forces without a SOFA in place. So that ended up being an excuse, or a public explanation, for what I believe was a policy choice.”156

Many senior officials believe that Obama's true intentions and motivations were pretty transparent from the beginning. “Look at what he said in the campaign,” argues Lute, whom Obama retained as “war czar” on the NSC, noting that “presidents tend to do what they said.”157 For Mullen, “It was very clear; you go back to his campaign promise, and it was going to be zero, it's just this question of how we were going to get there.”158 According to Kahl, “It is reasonable to conclude that Obama never believed that an enduring presence of thousands of American forces in Iraq was sustainable or in the American interest. He was a president who had made that clear in his campaign, had taken some risk to have the drawdown be more gradual than some people expected…. So it's reasonable to conclude that he wasn't going to go the extra mile to have this happen.”159 The Democratic base had already been upset about the extended drawdown until this point. As 2012 approached, a hardline stance now would play even worse.

Yet if one rejects the notion that Obama ever wanted a sustained presence in Iraq, which thus implies that electoral pressures were pushing on an open door, what explains the efforts throughout 2011—albeit belated and lukewarm—to negotiate one? If personal preferences overdetermined the decision, why did Obama not simply signal as much early on, rather than proceed with the iterated decisionmaking process described here, which was already having negative consequences for strategic planning of a “zero option” contingency?160

The decision to “go to zero” in October 2011 seems to have been the final point of a process that sought to balance the strategic benefit of a residual presence with the electoral risk of doing so. By late 2011, things in Iraq looked more or less stable. Indeed, in assessments provided by the intelligence community at the time, Obama had been advised that Iraq was unlikely to descend back into 2006-style violence any time soon, absent some exogenous shock or significant sectarian shift in Baghdad.161 With the 2012 campaign season looming, forcing through a renewed SOFA was, for Obama, not worth the political capital at such a sensitive time in the electoral cycle. In this sense, the electoral calculus had changed. In the winter of 2010–11, the risk of a deterioration in security and distance to the 2012 elections permitted some modest appetite for a follow-on force. By October, “they were nine months closer to the election [and] they didn't need as much of an insurance policy with Iraq.”162 Instead, the political damage that leaving troops could do to Obama's re-election campaign was now the more potent risk to be mitigated, leading Obama to “go to zero.”

Perhaps most telling were Obama's public comments during the third presidential debate of the 2012 campaign, when the president denied ever having tried to attempt such a negotiation in the first place. When Obama's Republican challenger, Mitt Romney, agreed with Obama's earlier ambition of reaching an agreement on a renewed SOFA, the president quickly intervened, saying, “That is not true … what I would not have done is left 10,000 troops in Iraq that would tie us down. That certainly would not help us in the Middle East.”163 Yet, this was precisely the policy Obama pursued between May and August 2011. Walking back from this position in 2012 would seem to validate the view that as the 2012 election drew closer, the idea of remaining in Iraq became a political liability as the candidate ran on a platform that reified his extrication of U.S. forces from Iraq.

In sum, the evidence presented here underscores the reality that wartime decisionmaking reflects a continual balancing act between the president's competing priorities as elected officeholder and commander in chief. Or, as Brennan summarizes, “Throughout this process [Obama] was weighing a number of factors and considerations, not just in terms of security requirements, but also some of those political dimensions, as it got closer to the elections.”164

Conclusion

Presidential decisionmaking in war is profoundly shaped by electoral pressures. Both the timing and nature of decisions regarding military strategy may be subject to constraints imposed by the domestic political calendar in ways that often do not align with the strategically optimal course of action. Seeking to balance their responsibilities as commander in chief with their political interests as holder of the highest elected office, presidents regularly moderate the professional advice of their national security team to satisfy the competing priorities they must manage.

This study goes beyond the existing literature in a number of ways. On a theoretical level, the application of existing insights regarding electoral accountability to a wartime context is novel. The article has shown that electoral constraints do not simply cease to operate once a conflict is initiated; instead, they continue to exert pressure on presidents throughout the course of a war. This finding is surprising, given that in matters so serious as war and peace, one might reasonably expect elected officials to rise above the fray of partisan politics and fully focus on their role in protecting national security free of domestic political considerations.

In addition, this article has introduced two mechanisms that outline ways in which electoral constraints operate at multiple stages of the electoral cycle. Under the delay effect, the presence of upcoming elections typically causes the president to defer decisions that may be politically sensitive until the question may be revisited at a time of less electoral risk. In cases where the timing of a decision is inflexible, the dampening effect will likely operate, whereby proposed courses of action are screened for their level of political risk, with unpopular elements removed or watered down until an acceptable balance of military utility and electoral risk can be found. These mechanisms operate not simply on the eve of a presidential election, but in anticipation of one and in relation to congressional midterms.

The article also makes a significant empirical contribution to the literature on the Iraq War. In the case of the 2007 surge decision, President Bush's response to a declining situation in Iraq was strongly conditioned by the delay effect. Although Bush perceived a need to change strategy from early 2006, he instead chose to defer a full consideration of alternatives until after the congressional elections, whereupon he suddenly sprang into action, forcefully pushing a bold and risky counterinsurgency plan against overwhelming opposition. In “going to zero” in 2011, President Obama's iterated path to the full withdrawal of U.S. troops demonstrated all the hallmarks of the dampening effect, with the looming presence of the 2012 campaign season leading to successive cuts in proposed follow-on forces until nothing remained politically viable.

It should be noted that the two empirical examples discussed in this article do not explore identical types of decisions. President Bush's surge decision involved escalation, whereas the debate in the Obama administration was about the pace and finality of a process of de-escalation. It follows that the electoral risks that mattered were different in each case. For Bush, given all we know about public sensitivity to the costs of war—and policymakers’ attentiveness to it—the political consequences of escalating before an election at a time when public sentiment ran diametrically opposed to such a course of action are relatively straightforward to identify. In Obama's case, the calculation was more nuanced. On one hand, the White House needed to minimize the political fallout associated with keeping troops on in Iraq, given the significance of Obama's prior pledge to do precisely the opposite. On the other hand, this needed to be tempered by the electoral risk associated with the collapse of Iraq back into violence should any precipitate withdrawal backfire prior to the 2012 presidential election.

Several points draw both the surge and withdrawal cases together and make them useful objects of analysis for this article, however. First, and most generally, both are excellent examples of presidents juggling their dual roles as elected officeholder and commander in chief. In both cases, electoral constraints clearly had an impact on the nature and timing of key decisions regarding military strategy, with both Bush and Obama seeking to offset the political risk of preferred courses of action in a manner strikingly consistent with the demands of the domestic political calendar. More specifically, both episodes highlight the increasing reluctance to approve and implement plans that entail a prolonged or an increased level of military commitment to a conflict as an election approaches. Elected leaders may deem such proposals strategically beneficial, but the likelihood and timing of their adoption appears to be conditional on the stage of the electoral cycle. As such, the cases illuminated here offer support for the notion of electoral accountability outlined at the outset of this article.

These findings are also substantively important when one considers what might have happened had electoral constraints not been a factor. Were the 2006 midterms not a limiting factor, President Bush might have opted to launch a strategy review at least six months before he did, and perhaps as early as the fall of 2005. An interagency consensus could have been established between the NSC and the State Department, that may in turn have elicited alliances with the doubters in the CIA and the uniformed military. Even if the bureaucracy remained ambivalent, the decisiveness that Bush showed after the election, and indeed his willingness to overrule a firm majority of dissenters, could well have resulted in a decision to shift toward a population-centric counterinsurgency plan much earlier than was the case.

In Obama's case, absent electoral pressures, one could speculate that the administration might have been more willing to stick to the military's proposals for more significant follow-on forces to remain in Iraq. At a minimum, the 10,000-strong force initially agreed to by Obama, “a force that could do real things,” according to the Pentagon leadership, might have stuck.165 Moreover, had planning for a force of this size been able to begin in earnest before the 2010 midterms, there might well have been a better chance for the U.S. side to have reached an agreement with the Iraqis about the legal protections the troops would possess.

Beyond these suppositions, however, this article does not seek to make any strong claims on the question of whether electoral constraints are a net benefit or net cost to effective decisionmaking. Although it could be argued that there is a strategic cost associated with the delay and dampening effects described here, it might also be noted that the responsiveness of elected officials to the concerns of the electorate is a fundamental and valuable feature of democracy. More specifically, taking a stand on this debate requires making prior judgments as to the wisdom of the decisions under discussion, at a time when the efficacy of the surge and the decision to withdraw in 2011 are subjects of intense and ongoing debate. While some argue that an earlier and more enduring increased presence of U.S. troops might have alleviated the deepening security issues in Iraq during this period and perhaps forestalled the resurgence of the Islamic State thereafter, others may reasonably believe that no amount of security would have solved the essentially political problems at stake, much as someone such as General Casey had been arguing all along. Because the primary objective of this article has been to explore how electoral constraints matter in decisionmaking, it is left to the reader to draw any wider normative implications as to whether they should.

The complex effect that electoral pressures have on wartime decisionmaking calls for further study. Additional in-depth analysis of other conflicts may serve to bolster the generalizability of the claims made here. Moreover, the mechanisms offered here may not be the only ones in operation; under different circumstances, the U.S. electoral cycle may well constrain decisionmaking in other ways. Further research might also fruitfully explore in more depth the connection between different phases of the electoral cycle and the strength of these mechanisms of constraint.

Acknowledgments

The author is grateful to Louise Fawcett for her invaluable guidance and support. He also thanks Janina Dill, Dominic Johnson, Duncan Snidal, and the anonymous reviewers for helpful comments and suggestions. Financial support for this research was generously provided by Merton College and the Department of Politics and International Relations at the University of Oxford, as well as the Economic and Social Research Council.

Notes

1. 

See Peter Trubowitz, Politics and Strategy: Partisan Ambition and American Statecraft (Princeton, N.J.: Princeton University Press, 2011); and Bruce Bueno de Mesquita and James Lee Ray, “The National Interest versus Individual Political Ambition,” in Paul F. Diehl, ed., The Scourge of War: New Extensions on an Old Problem (Ann Arbor: University of Michigan Press, 2004), pp. 94–119.

2. 

Author phone interview with John Brennan, July 19, 2018.

3. 

On public opinion, see especially Matthew A. Baum and Philip B.K. Potter, War and Democratic Constraint: How the Public Influences Foreign Policy (Princeton, N.J.: Princeton University Press, 2015); and Ole R. Holsti, Public Opinion and American Foreign Policy, rev. ed. (Ann Arbor: University of Michigan Press, 2004). On partisanship, see Charles A. Kupchan and Peter L. Trubowitz, “Dead Center: The Demise of Liberal Internationalism in the United States,” International Security, Vol. 32, No. 2 (Fall 2007), pp. 7–44, doi.org/10.1162/isec.2007.32.2.7. On Congress, see William G. Howell and Jon C. Pevehouse, While Dangers Gather: Congressional Checks on Presidential War Powers (Princeton, N.J.: Princeton University Press, 2007); and Douglas L. Kriner, After the Rubicon: Congress, Presidents, and the Politics of Waging War (Chicago: University of Chicago Press, 2010).

4. 

Jim VandeHei and Michael A. Fletcher, “Bush Says Election Ratified Iraq Policy,” Washington Post, January 16, 2005, https://www.wsj.com/articles/SB110591468644127305.

5. 

For a useful review, see John H. Aldrich et al., “Foreign Policy and the Electoral Connection,” Annual Review of Political Science, Vol. 9 (2006), pp. 477–502, doi.org/10.1146/annurev.polisci.9.111605.105008.

6. 

The best work on this phase of the war is arguably Michael R. Gordon and Bernard E. Trainor, The Endgame: The Inside Story of the Struggle for Iraq, from George W. Bush to Barack Obama (New York: Pantheon, 2012). Among the better journalistic accounts are Bob Woodward, The War Within: A Secret White House History, 2006–2008 (New York: Simon and Schuster, 2008); and Thomas E. Ricks, The Gamble: General David Petraeus and the American Military Adventure in Iraq, 2006–2008 (New York: Penguin, 2009). The surge has attracted the most academic attention. See, in particular, Peter D. Feaver, “The Right to Be Right: Civil-Military Relations and the Iraq Surge Decision,” International Security, Vol. 35, No. 4 (Spring 2011), pp. 87–125, doi.org/10.1162/ISEC_a_00033; Stephen Benedict Dyson, “George W. Bush, the Surge, and Presidential Leadership,” Political Science Quarterly, Vol. 125, No. 4 (Winter 2010), pp. 557–585, doi.org/10.1002/j.1538-165X.2010.tb00685.x; Kevin P. Marsh, “The Intersection of War and Politics: The Iraq War Troop Surge and Bureaucratic Politics,” Armed Forces & Society, Vol. 38, No. 3 (July 2012), pp. 413–437, doi.org/10.1177/0095327X11415492; and Stephen Biddle, Jeffrey A. Friedman, and Jacob N. Shapiro, “Testing the Surge: Why Did Violence Decline in Iraq in 2007?” International Security, Vol. 37, No. 1 (Summer 2012), pp. 7–40, doi.org/10.1162/ISEC_a_00087. For an excellent recent addition to the literature, see Timothy Andrew Sayle et al., eds., The Last Card: Inside George W. Bush's Decision to Surge in Iraq (Ithaca, N.Y.: Cornell University Press, 2019).

7. 

Bruce Bueno de Mesquita et al., The Logic of Political Survival (Cambridge, Mass.: MIT Press, 2003).

8. 

Jonathan D. Caverley, Democratic Militarism: Voting, Wealth, and War (Cambridge: Cambridge University Press, 2015).

9. 

See John Kaag and Sarah Kreps, Drone Warfare (Cambridge: Polity, 2014), pp. 53–77.

10. 

See, respectively, Christopher Gelpi, Peter D. Feaver, and Jason Reifler, “Success Matters: Casualty Sensitivity and the War in Iraq,” International Security, Vol. 30, No. 3 (Winter 2005/06), pp. 7–46, doi.org/10.1162/isec.2005.30.3.7; Bruce W. Jentleson, “The Pretty Prudent Public: Post-Vietnam American Opinion on the Use of Military Force,” International Studies Quarterly, Vol. 36, No. 1 (March 1992), pp. 49–74, doi.org/10.2307/2600916; and Eric V. Larson, Casualties and Consensus: The Historical Role of Casualties in Domestic Support for U.S. Military Operations (Santa Monica, Calif.: RAND Corporation, 1996).

11. 

Baum and Potter, War and Democratic Constraint.

12. 

Bruce Bueno de Mesquita et al., “An Institutional Explanation of the Democratic Peace,” American Political Science Review, Vol. 93, No 4 (December 1999), pp. 791–807, doi.org/10.2307/2586113; and Dan Reiter and Allan C. Stam, Democracies at War (Princeton, N.J.: Princeton University Press, 2002).

13. 

Robert Y. Shapiro and Lawrence R. Jacobs, “Who Leads and Who Follows? U.S. Presidents, Public Opinion, and Foreign Policy,” in Brigitte L. Nacos, Robert Y. Shapiro, and Pierangelo Isernia, eds., Decisionmaking in a Glass House: Mass Media, Public Opinion, and American and European Foreign Policy in the 21st Century (Lanham, Md.: Rowman and Littlefield, 2000), pp. 223–246.

14. 

Douglas C. Foyle, Counting the Public In: Presidents, Public Opinion, and Foreign Policy (New York: Columbia University Press, 1999).

15. 

See Christopher Gelpi, Peter D. Feaver, and Jason Reifler, Paying the Human Costs of War: American Public Opinion and Casualties in Military Conflicts (Princeton, N.J.: Princeton University Press, 2009), pp. 226–233, 260–262. On elite cues and public attitudes toward war, see Adam J. Berinsky, “Assuming the Costs of War: Events, Elites, and American Public Support for Military Conflict,” Journal of Politics, Vol. 69, No. 4 (November 2007), pp. 975–997, doi.org/10.1111/j.1468-2508.2007.00602.x.

16. 

Richard Sobel, The Impact of Public Opinion on U.S. Foreign Policy since Vietnam: Constraining the Colossus (Oxford: Oxford University Press, 2001); Philip J. Powlick and Andrew Z. Katz, “Defining the American Public Opinion/Foreign Policy Nexus,” Mershon International Studies Review, Vol. 42, No. 1 (May 1998), pp. 29–61, doi.org/10.2307/254443; and Bruce Russett, Controlling the Sword: The Democratic Governance of National Security (Cambridge, Mass.: Harvard University Press, 1990).

17. 

Gelpi, Feaver, and Reifler, Paying the Human Costs of War; and Richard A. Lacquement Jr., “The Casualty-Aversion Myth,” Naval War College Review, Vol. 57, No. 1 (Winter 2004), pp. 39–57, https://digital-commons.usnwc.edu/nwc-review/vol57/iss1/6.

18. 

On rally effects, see John E. Mueller, War, Presidents, and Public Opinion (New York: John Wiley and Sons, 1973). For notable examples in the literature of studies examining the electoral incentives behind diversionary uses of force, see Richard J. Stoll, “The Guns of November: Presidential Elections and the Use of Force, 1947–1982,” Journal of Conflict Resolution, Vol. 28, No. 2 (June 1984), pp. 231–246, doi.org/10.1177/0022002784028002002; Gregory D. Hess and Athanasios Orphanides, “War Politics: An Economic, Rational-Voter Framework,” American Economic Review, Vol. 85, No. 4 (September 1995), pp. 828–846, https://www.jstor.org/stable/2118234; and Benjamin Fordham, “The Politics of Threat Perception and the Use of Force: A Political Economy Model of U.S. Uses of Force, 1949–1994,” International Studies Quarterly, Vol. 42, No. 3 (September 1998), pp. 567–590, doi.org/10.1111/0020-8833.00096.

19. 

Notable examples include Jack S. Levy, “Diversionary War Theory: A Critique,” in Manus I. Midlarsky, ed., Handbook of War Studies (Boston: Unwin Hyman, 1989), pp. 259–288; James Meernik, “Presidential Decision Making and the Political Use of Military Force,” International Studies Quarterly, Vol. 38, No. 1 (March 1994), pp. 121–138, doi.org/10.2307/2600874; and James Meernik and Peter Waterman, “The Myth of the Diversionary Use of Force by American Presidents,” Political Research Quarterly, Vol. 49, No. 3 (September 1996), pp. 573–590, doi.org/10.1177/106591299604900306.

20. 

Bradley Lian and John R. Oneal, “Presidents, the Use of Military Force, and Public Opinion,” Journal of Conflict Resolution, Vol. 37, No. 2 (June 1993), pp. 277–300, doi.org/10.1177/0022002 793037002003; and John R. Oneal and Anna Lillian Bryan, “The Rally ‘Round the Flag Effect in U.S. Foreign Policy Crises, 1950–1985,” Political Behavior, Vol. 17, No. 4 (December 1995), pp. 379–401, https://www.jstor.org/stable/586592.

21. 

Kurt T. Gaubatz, Elections and War: The Electoral Incentive in the Democratic Politics of War and Peace (Stanford, Calif.: Stanford University Press, 1999).

22. 

Giacomo Chiozza, “Presidents on the Cycle: Elections, Audience Costs, and Coercive Diplomacy,” Conflict Management and Peace Science, Vol. 34, No. 1 (January 2017), pp. 3–26, doi.org/10.1177/0738894215593677; and Philip B.K. Potter, “Electoral Margins and American Foreign Policy,” International Studies Quarterly, Vol. 57, No. 3 (September 2013), pp. 505–518, doi.org/10.1111/isqu.12040. See also Paul K. Huth and Todd L. Allee, “Domestic Political Accountability and the Escalation and Settlement of International Disputes,” Journal of Conflict Resolution, Vol. 46, No. 6 (December 2002), pp. 754–790, doi.org/10.1177/002200202237928.

23. 

Paola Conconi, Nicolas Sahuguet, and Maurizo Zanardi, “Democratic Peace and Electoral Accountability,” Journal of the European Economic Association, Vol. 12, No. 4 (August 2014), pp. 997–1028, doi.org/10.1111/jeea.12074; Sean Zeigler, Jan H. Pierskalla, and Sandeep Mazumder, “War and the Reelection Motive: Examining the Effect of Term Limits,” Journal of Conflict Resolution, Vol. 58, No. 4 (June 2014), pp. 658–684, doi.org/10.1177/0022002713478561; Philip B.K. Potter, “Lame-Duck Foreign Policy,” Presidential Studies Quarterly, Vol. 46, No. 4 (December 2016), pp. 849–867, doi.org/10.1111/psq.12322; and Alexander Hamilton, “The Federalist Papers: No. 72,” Avalon Project, Yale Law School, first published March 21, 1788, http://avalon.law.yale.edu/18th_century/fed72.asp.

24. 

Causation may also be established through alternative methods—for instance, through the use of randomized control trials. Such experimental methods are not adopted here, however, given the inherent difficulties involved in simulating presidential wartime decisionmaking environments in the laboratory.

25. 

Historians have made more effective use of case studies to examine the role of electoral politics on foreign policy, but still lack accounts of fully specified mechanisms to explain it. See, for instance, Andrew Preston, “Beyond the Water's Edge: Foreign Policy and Electoral Politics,” in Gareth Davies and Julian E. Zelizer, eds., America at the Ballot Box: Elections and Political History (Philadelphia: University of Pennsylvania Press, 2015), pp. 219–237; Michael H. Armacost, Ballots, Bullets, and Bargains: American Foreign Policy and Presidential Elections (New York: Columbia University Press, 2015); and Andrew Johnstone and Andrew Priest, eds., U.S. Presidential Elections and Foreign Policy: Candidates, Campaigns, and Global Politics from FDR to Bill Clinton (Lexington: University of Kentucky Press, 2017).

26. 

See Gaubatz, Elections and War, pp. 16–18.

27. 

See Barbara Farnham, “Impact of the Political Context on Foreign Policy Decision-Making,” Political Psychology, Vol. 25, No. 3 (June 2004), pp. 441–463, doi.org/10.1111/j.1467-9221.2004.00379.x.

28. 

George W. Bush, “President's Address to the Nation,” transcript of televised remarks (Washington, D.C.: White House, January 10, 2007), https://georgewbush-whitehouse.archives.gov/news/releases/2007/01/20070110-7.html.

29. 

Philip Zelikow, “The Surge,” transcript of oral history interview by Peter Feaver and Aaron Crawford (Dallas, Tex.: Collective Memory Project, Center for Presidential History, Southern Methodist University, March 24, 2015), p. 12. Draft manuscript obtained by author from interviewee.

30. 

Author interview with Philip D. Zelikow, Charlottesville, Virginia, March 19, 2018.

31. 

Author phone interview with James F. Jeffrey, June 28, 2018.

32. 

Condoleezza Rice, No Higher Honor: A Memoir of My Years in Washington (New York: Crown, 2011), pp. 465, 507.

33. 

Author phone interview with Meghan L. O'Sullivan, June 11, 2018.

34. 

Feaver, “The Right to Be Right,” p. 101.

35. 

Woodward, The War Within, p. 69.

36. 

George W. Bush, Decision Points (New York: Crown, 2010), p. 364.

37. 

“ODNI Camp David Scene Setter,” June 12, 2006, box 114, “Commander, MNF-I, Jul 04–Feb 07, February–June 2006,” Papers of General George W. Casey Jr., National Defense University, Washington, D.C. (henceforth Casey Papers).

38. 

Author interview with John D. Negroponte, Washington, D.C., March 27, 2018.

39. 

Woodward, The War Within, pp. 72–73.

40. 

Author interview with Michael G. Mullen, Washington, D.C., March 29, 2018.

41. 

“Transcript: 2006 Huddle (#1/2),” March 9, 2008, box 104, “Commander, MNF-I, Jul 04–Feb 07, Notes, Calendars, Huddles: 2006 Huddles, Transcript/Notes,” Casey Papers.

42. 

Author interview with George W. Casey Jr., Arlington, Virginia, March 16, 2018.

43. 

Author interview with Negroponte.

44. 

Author interview with O'Sullivan.

45. 

Bush, Decision Points, p. 363. See also Bush and Hadley's comments, quoted in Sayle et al., The Last Card, pp. 53–54.

46. 

Author interview with Zelikow.

47. 

Author interview with Jeffrey.

48. 

Author interview with Zelikow.

49. 

Zalmay Khalilzad, The Envoy: From Kabul to the White House, My Journey through a Turbulent World (New York: St. Martin's, 2016), p. 234.

50. 

Memorandum, Mowaffak Al-Rubaie to George Casey and Zalmay Khalilzad, November 6, 2005, box 113, “Commander, MNF-I, Jul 04–Feb 07, November 2005–January 2006,” Casey Papers.

51. 

Woodward, The War Within, p. 55.

52. 

Zelikow, “The Surge,” p. 28.

53. 

Author interview with Zelikow.

54. 

Peter Feaver, quoted in Peter Baker, Days of Fire: Bush and Cheney in the White House (New York: Doubleday, 2013), p. 467.

55. 

Bush, Decision Points, p. 371.

56. 

Author interview with Casey; Memorandum, George Casey, “Key Recommendations, Briefing Camp David,” June 12, 2006, box 114, “Commander, MNF-I, Jul 04–Feb 07, February–June 2006,” Casey Papers; and “MNF-I Staff Notes Excerpts, May–Sep 2006,” doc. no. 0455, CENTCOM Iraq Papers, U.S. Army Heritage and Education Center, Carlisle, Pennsylvania, https://ahec.armywarcollege.edu/CENTCOM-IRAQ-papers/index.cfm (henceforth CENTCOM Papers), p. 23.

57. 

Author interview with Casey. See also Bush, Decision Points, p. 367.

58. 

Woodward, The War Within, pp. 178–179.

59. 

Ibid., p. 201.

60. 

Author interview with Conrad C. Crane, Carlisle, Pennsylvania, March 23, 2018. See also Army Field Manual 3–24: Counterinsurgency (Washington, D.C.: United States Department of the Army, December 2006).

61. 

The role of Keane is particularly emphasized in Ricks, The Gamble, especially pp. 79–93.

62. 

Author interview with Zelikow. Crouch was the deputy national security adviser at the time.

63. 

Ibid.

64. 

Bush, Decision Points, p. 372.

65. 

Woodward, The War Within, pp. 84–85.

66. 

Ibid., p. 175.

67. 

Bob Woodward, State of Denial: Bush at War, Part III (New York: Simon and Schuster, 2006), p. 491.

68. 

O'Sullivan, quoted in Sayle et al., The Last Card, p. 91.

69. 

Peter R. Mansoor, Surge: My Journey with General David Petraeus and the Remaking of the Iraq War (New Haven, Conn.: Yale University Press), p. 46.

70. 

John J. Mearsheimer, Why Leaders Lie: The Truth about Lying in International Politics (New York: Oxford University Press, 2011), p. 63.

71. 

Ibid., p. 23.

72. 

On Bush's emergent preference for escalation, see Bush, Decision Points, pp. 361–378. See also Joshua Bolten, Bush's chief of staff, quoted in Sayle et al., The Last Card, p. 111; and Karl Rove, quoted in Sayle et al., The Last Card, p. 133.

73. 

Eric Schmitt, “Fast Withdrawal of G.I.'s Is Urged by Key Democrat,” New York Times, November 18, 2005, https://www.nytimes.com/2005/11/18/politics/fast-withdrawal-of-gis-is-urged-by-key-democrat.html; and Adam Nagourney, “Democratic Leaders Ask Bush to Redeploy Troops in Iraq,” New York Times, August 1, 2006, https://www.nytimes.com/2006/08/01/washington/01pullout.html.

74. 

John Warner to Peter Pace, John Abizaid, and George Casey, “Troop Levels in Iraq,” October 25, 2006, box 115, “Commander, MNF-I, Jul 04–Feb 07, July–November 2006,” Casey Papers.

75. 

Bush, Decision Points, p. 355.

76. 

CNN poll, conducted by Opinion Research Corporation, June 8–11, 2006, http://pollingreport.com/iraq11.htm.

77. 

Bush, Decision Points, p. 375.

78. 

Ibid., p. 355.

79. 

Author email correspondence with Meghan L. O'Sullivan, January 31, 2019.

80. 

Ed Donnelly, quoted in “Transcript: 2006 Iraq Huddle (#1/2),” March 9, 2008.

81. 

Author interview with Jeffrey.

82. 

“Notes from POTUS SVTC,” December 8, 2006, box 116, “Commander, MNF-I, Jul 04–Feb 07, December 2006–February 2007,” Casey Papers.

83. 

Bush, Decision Points, p. 367; Karl Rove, Courage and Consequence: My Life as a Conservative in the Fight (New York: Simon and Schuster, 2010), p. 476; and Dyson, “George W. Bush, the Surge, and Presidential Leadership,” p. 573.

84. 

Author interview with Crane.

85. 

Ibid.

86. 

Handwritten Notes, May 2006, box 114, “Commander, MNF-I, Jul 04-Feb 07, February-June 2006,” Casey Papers. See also “MNF-I Staff Notes Excerpts, May-Sep 2006,” p. 21; and Email, Donald Rumsfeld to Stephen J. Hadley, “U.S. Casualties in Iraq,” May 8, 2006, Rumsfeld Papers, http://papers.rumsfeld.com.

87. 

Memorandum, Donald Rumsfeld to George Casey, “Decision on Stryker Brigade,” August 2, 2006, box 115, “Commander, MNF-I, Jul 04–Feb 07, July-November 2006,” Casey Papers.

88. 

Condoleezza Rice, Democracy: Stories from the Long Road to Freedom (New York: Twelve, 2017), p. 313.

89. 

Rice, No Higher Honor, p. 507.

90. 

Ibid., p. 540.

91. 

George W. Casey Jr., Strategic Reflections: Operation Iraqi Freedom, July 2004-February 2007 (Washington, D.C.: National Defense University Press, 2012), p. 168. See also John Abizaid, quoted in Sayle et al., The Last Card, p. 128.

92. 

Author interview with Zelikow.

93. 

Author interview with Jeffrey.

94. 

Memorandum, Donald Rumsfeld to Stephen Hadley, “NSC Meetings,” May 30, 2006, Rumsfeld Papers.

95. 

Zelikow, “The Surge,” p. 32.

96. 

Author email correspondence with James F. Jeffrey, June 28, 2018.

97. 

Rove, Courage and Consequence, p. 466. See also Sayle et al., The Last Card, pp. 119–121.

98. 

Author interview with Jeffrey.

99. 

Author interview with Mullen.

100. 

Feaver, “The Right to Be Right,” p. 102.

101. 

Author interview with Casey.

102. 

Author interview with Mullen.

103. 

Author interview with Douglas E. Lute, Rosslyn, Virgina, April 3, 2018.

104. 

Author interview with Casey.

105. 

Author phone interview with John F. Sattler, July 25, 2018.

106. 

Author interview with Zelikow.

107. 

Woodward, The War Within, pp. 234–235. See also Bush, Decision Points, p. 376.

108. 

“Notes from POTUS SVTC,” December 8, 2006.

109. 

Briefing Notes, “Iraq—Security: Way Ahead,” December 12, 2006, box 117, “Commander, MNF-I, Jul 04–Feb 07, Declassified Key Documents, POTUS SVTCs,” Casey Papers.

110. 

“Notes from POTUS SVTC,” December 8, 2006.

111. 

“Transcript: 2006 Iraq Huddle (#1/2),” March 9, 2008.

112. 

Ibid.

113. 

“Notes from POTUS SVTC,” December 9, 2006, box 116, “Commander, MNF-I, Jul 04–Feb 07, December 2006–February 2007,” Casey Papers. The Iraq Study Group Report contained only one sentence referring to a “surge”-like option, buried on p. 73. See James A. Baker III and Lee H. Hamilton, The Iraq Study Group Report: The Way Forward—A New Approach (New York: Vintage, 2006).

114. 

“Notes from POTUS SVTC,” December 9, 2006.

115. 

Author interview with Zelikow. Gen. David Petraeus succeeded Casey as commanding general of MNF-I. Lt. Gen. Raymond Odierno served under Petraeus as commander of MNF-I.

116. 

Mansoor, Surge, p. 56.

117. 

Author interview with Crane.

118. 

Author interview with Lute.

119. 

Casey, Strategic Reflections, pp. 141–144.

120. 

“Notes from SecDef SVTC,” November 16, 2006, box 115, “Commander, MNF-I, Jul 04–Feb 07, July–November 2006,” Casey Papers.

121. 

Bush, Decision Points, p. 374.

122. 

Woodward, The War Within, p. 264.

123. 

Presentation Slides, Unnamed Author [probably Nouri al-Maliki or Mowaffak al-Rubaie], August 2006, box 115, “Commander, MNF-I, Jul 04–Feb 07, July–November 2006,” Casey Papers.

124. 

Briefing Notes, “Iraq—Security: Way Ahead,” December 12, 2006.

125. 

“POTUS SVTC,” December 11, 2006, box 116, “Commander, MNF-I, Jul 04–Feb 07, December 2006–February 2007,” Casey Papers.

126. 

Bush, Decision Points, p. 355.

127. 

Author interview with Brennan.

128. 

Author interview with Michèle A. Flournoy, Washington, D.C., April 3, 2018. See also Joel D. Rayburn and Frank K. Sobchak, The U.S. Army in the Iraq War, Vol. 2: Surge and Withdrawal, 2007–2011 (Carlisle Barracks, Pa.: United States Army War College Press, 2019), p. 519.

129. 

Author interview with Colin H. Kahl, Washington, D.C., March 15, 2018.

130. 

Author interview with Flournoy.

131. 

Author interview with Kahl.

132. 

Author interview with Jeffrey.

133. 

Robert M. Gates, Duty: Memoirs of a Secretary at War (New York: Alfred A. Knopf, 2014), p. 553; and Leon Panetta, Worthy Fights: A Memoir of Leadership in War and Peace (New York: Penguin, 2014), pp. 356–357, 393.

134. 

Author interview with James E. Cartwright, Washington, D.C., March 26, 2018.

135. 

Author interview with Brennan.

136. 

Author phone interview with Antony J. Blinken, June 8, 2018.

137. 

Author interview with Kahl.

138. 

Author interview with Flournoy.

139. 

Gordon and Trainor, The Endgame, p. 656.

140. 

Author interview with Mullen.

141. 

Ibid.

142. 

Gordon and Trainor, The Endgame, p. 657.

143. 

Author interview with Mullen.

144. 

Author interview with Cartwright.

145. 

Gordon and Trainor, The Endgame, p. 664.

146. 

Rayburn and Sobchak, The U.S. Army in the Iraq War, Vol. 2, p. 543.

147. 

Quotation in “United States Forces-Iraq Quarterly Command Report, 2d Quarter (1 January–31 March 2011), FY11,” May 1, 2011, doc no. 471, CENTCOM Papers.

148. 

Ibid.

149. 

“United States Forces-Iraq Quarterly Command Report, 3rd Quarter, FY11 (1 April–30 June 2011),” July 1, 2011, doc no. 1003, CENTCOM Papers.

150. 

Panetta, Worthy Fights, p. 370.

151. 

Author interview with Blinken.

152. 

Author interview with Jeffrey.

153. 

Author interview with Blinken.

154. 

Author interview with Lute.

155. 

Author interview with Flournoy.

156. 

Ibid.

157. 

Author interview with Lute.

158. 

Author interview with Mullen.

159. 

Author interview with Kahl.

160. 

Richard R. Brennan Jr. et al., Ending the U.S. War in Iraq: The Final Transition, Operational Maneuver, and Disestablishment of United States Forces–Iraq (Santa Monica, Calif.: RAND Corporation, 2013).

161. 

Author interview with Kahl. These assessments were contested by several senior political and military officials.

162. 

Author interview with Jeffrey.

164. 

Author interview with Brennan.

165. 

Author interview with Flournoy.