The Danger of Military Partisanship
It is ironic that the man who looked at trends in civil-military relations in the early 1990s and penned a fictitious story about a future coup observes trends in 2018 and now believes rhetoric about American civil-military relations is too inflammatory. Yet in a recent essay critiquing Major Matt Cavanaugh’s and warning that a more partisan military could lead to “dire” consequences, Major General (Ret.) Charlie Dunlap argues that growing partisanship in the ranks isn’t something we should be worried about.
Dunlap and I have met at several conferences, and I have enormous respect for him as an observer of American civil-military relations. Unfortunately, in this case, his piece is riddled with flawed logic and factual errors. While this short essay cannot address them all, I try to correct or refute the most problematic of his arguments. In our nation’s current polarized political environment, increased partisanship among military officers is a real problem, and – if we don’t arrest current trends – the consequences could be dire indeed.
Correlation, Not Causation
The biggest problem with MG Dunlap’s essay is that he attacks a strawman by suggesting that the lack of partisanship among senior military officers in the 1970s created the problems of discrimination, racism, drug abuse, and poor performance. As Hugh Liebert and I argue in our piece, “Midlife Crisis? The All Volunteer Force at Forty,” the quality of the All Volunteer Force (AVF) has increased at the same time that the percentage of senior officers who identify with a political party has grown. In 1976, only 45 percent of senior officers identified with a political party; by 1996, nearly 90 percent did so, and it has held fairly steady ever since. At the same time, the quality, discipline, and tactical and operational performance of the AVF also improved.
While it is true that these two trends developed at a similar time, growing partisanship among senior officers did not cause problems among the force to disappear over the last forty years. Rather, the norm of non-partisanship broke down in the wake of the Vietnam War due to other factors at the same time that various reforms led to improvements in the quality of the AVF. But this correlation implies no causal relationship. It is a classic case of the post hoc, ergo proctor hoc fallacy, and there is simply no reason to think that having officers embrace a norm of non-partisanship would somehow degrade the quality of the force today. So the real question we need to ask is not whether we are “nostalgic for the supposedly ‘non-partisan’ military of 1976,” but whether we would have preferred to have the military of 1976 led by a partisan officer corps. And whether we would prefer to have the military of today and tomorrow be led by partisans or professionals.
Looking Through Party-colored Glasses
Dunlap also misrepresents the scholarship on how partisanship impacts one’s political attitudes. Specifically, he suggests that “setting aside one’s political druthers is a vastly less demanding decision … and most military professionals do so instinctively.” While Dunlap offers anecdotal evidence for this claim, the political science literature suggests that one’s partisan identification is the best predictor of someone’s views on all sort of issues, even when they don’t realize it is playing a role. In other words, even if most officers want to live up to the norm of non-partisanship and isolate their personal political views from their professional thinking and behavior, doing so simply is not as easy as Dunlap thinks it is.
In the real world, it is extremely difficult for officers to separate their personal from their professional views. Military officers often don't realize how their own partisan preferences shape their own "expert" views based on "experience," even on military issues about whether and how to use military force. In my dissertation, I demonstrated that one's partisanship is a better predictor of support for the Powell doctrine, the efficacy of different types of military options (e.g., the use of ground forces, drones, special operations forces, etc.), and whether to use force under certain circumstances than whether someone is a military officer/veteran or not. Peter Feaver, Lindsay Cohn, and I recently replicated this result. Additionally, Heidi Urben’s work shows that officers’ political preferences influence the opinions they express at work and their behavior on social media. So while few officers would admit that their political preferences shape their professional opinions and behavior, the evidence suggests they often do.
The effects of one's partisanship or one's personal views are often hard for one to discern herself because they creep in through the media we consume, the friends we have, social media filters, ad-maximizing search engines, etc. And we're not always cognizant of these biases. While my research focuses on the military, the broader political science literature confirms that one’s partisan identification is the best predictor of attitudes across a wide array of topics and that one’s party has a significant impact in shaping how we see and interpret the world around us. As a result, people often have a hard time distinguishing between their own personal biases based on party identification and their professional expertise and opinions. There is simply no reason to believe, and no research to demonstrate, that military officers are immune to these forms of bias. In fact, most research suggests they are not. The challenge for professionals is to become more aware of their own biases so they can recognize and mitigate them so that they will be able to provide more objective expertise and analysis about how to use military force most effectively. This problem becomes especially acute for officers as they rise in rank and assume more responsibility for the civil-military relationship, but Dunlap assumes the problem away.
Even Perceived Bias Has Real Consequences
Although Dunlap wrongly assumes that officers can compartmentalize their partisan preferences from their personal views, he does at least acknowledge the problem of perceived bias. Unfortunately, however, he significantly understates its impact. Citing judges as officials the public trusts to set aside their personal opinions in the courtroom, Dunlap ignores both declining public confidence in the courts as well as swings in public opinion based on whether partisans think the Supreme Court represents their own views. After Neil Gorsuch was confirmed in April 2017, for example, confidence in the Supreme Court among Republicans surged by more than ten points even before he had participated in a single decision.
Whether it is deserved or not, a significant portion of the public perceives the military to be a partisan institution. In a report we wrote for CNAS in 2012, Peter Feaver, Kyle Dropp, and I found that a majority of the public thinks the military is partisan and, among those who think the military is partisan, more than 60 percent think the military leans Republican. This perception may also have an impact on public trust in the military and on recruiting and retention. As partisanship among military officers has grown, public confidence in the military also has begun to polarize. And, as Liebert and I outline in our article on the AVF, these perceptions also may be hurting recruiting efforts among certain groups in the U.S. population.
The perception that senior military officers are becoming more partisan also creates perverse incentives for political leaders to bargain with military leaders, to use them as the public face for Administration policies, or to try to use them as a shield against criticism when policies go wrong. In a deeply polarized society, this trend contributes to the increasing use of the military as a political prop, not only on military issues but also on domestic issues such as gun control, immigration policy, and even whether restaurants can refuse service to Administration officials. Not only does this phenomenon potentially transfer some responsibility for civilian governance from elected leaders, it also undermines democratic accountability when elected leaders use generals, service members, or veterans to shield themselves from public criticism.
Not What the Framers Intended
Finally, Dunlap fundamentally misunderstands the Framers’ view of civil-military relations and the protections they developed in the Constitution. According to Dunlap, “civil-military relations issue the Framers were concerned about was the physical threat that a professional military might pose, not what seems to energize critics today: the idea that military professionals, active or retired, might engage in policy debates.” While it is always fraught to attempt to ascertain what the Framers’ intended, individually or as a group, it is clear they didn’t think of questions of civilian control of the military in the way we do today. Their fears were far more nuanced than Dunlap suggests. Additionally, although the Framers clearly had concerns about the impact that the creation of a standing army might have on civil liberties and these concerns pervaded debates at the Constitutional convention, a professional military as we know it today did not yet exist in America. But armed groups and political factions did.
Although the Framers’ views were often inchoate, they were not solely concerned about a military usurpation of power; they also were concerned that a civilian leader or group might capture the military and use it for tyrannical purposes. Of course, local insurrections such as Shay’s Rebellion were a primary driver to the development of a stronger federal government under the Constitution. And the Framers were deeply concerned about how to handle to the threat of armed militant groups to local and state governments in a way that maintained political stability. Nevertheless, at least some of the Framers appeared far more worried that the federal government would consolidate power and use the military against the people than they were that a professional military would rise up against the federal government itself. In particular, the Framers were concerned that a single king-like leader or a powerful faction would be able to co-opt all or part of the military in an attempt to impose a tyrannical government. This civil-military faction or coalition would then be able to consolidate power, challenge republican governance, undermine individual liberty, and threaten the sovereignty of state governments. So the fear that the military might pick sides in domestic debates, and that political leaders might use the military institution for partisan gain, was in fact a primary concern of the Framers. As a result, they included provisions in the Constitution to make it harder, though not impossible, for any one leader or branch of the government to do so.
Indeed, at the worst moment of our own political history, this is exactly what occurred. Political leaders from one faction convinced large portions (by one estimate 26% of West Point graduates joined the secessionist cause) of our nation’s military to ignore the will of our elected political leaders. Yet, when speaking of disdainful comments about elected leaders, Dunlap concludes, “the reality is that however regrettable such commentary may be, it’s occurred within the ranks throughout our history (and especially during the Revolution and Civil War) … yet the Republic has not suffered ‘dire’ consequences.” If the Civil War does not count as a “dire” consequence, I’m not sure what ever could. So while I am not arguing that we are headed toward a civil war today, I am suggesting that we think carefully about what should cause worry. For even if elected political leaders are able to use the military to sell policies, target their political enemies, or to protect themselves from democratic accountability by using the military as the public face of some policies, this set-up is inimical to our system of government and could have problematic effects over the medium to long-term. Left unchecked, these trends will undermine military expertise, effectiveness, public confidence, and recruiting. Moreover, they could portend much bigger problems in the midst of a true political crisis.
The concern today is not a military revolt against civilian government. Moreover, many of the protections the Framers establish remain in place and they have been buttressed by additional protections that have evolved throughout American history. But growing partisanship among the military officer corps nevertheless represents a real danger for the health of the American republic. Especially during the current period of uncertainty when the legitimacy of political decisions are increasingly questioned, the potential that political or military leaders could attempt to use the legitimacy of the military institution to advance partisan aims is not something we should assume away. Public confidence in all other national institutions has either plummeted, polarized, or both, and there is no guarantee that this won’t happen to the military like it has for the Supreme Court. In fact, as I have outlined above, there is already evidence that partisan political leaders distinguish between “my generals” and “their generals,” and that partisan differences may even be having an impact on recruiting, retention, and morale. These trends should be troubling to all Americans, regardless of party.
Concerns about growing partisanship in the military officer corps have little to do with whether members of the military can or should engage in political or policy debates. Instead, they have to do with whether elected political leaders – or military officers themselves – can use the military as an institution in ways that weaken democratic accountability, harm military strength and effectiveness, and undermine the proper functioning of our republican processes. Throughout most of human history, societies and governments have failed to maintain the balance between liberty and security. It is not something we should take for granted. If we do, we once again will have realized the worst fears of the Framers and that would be a “dire” consequence indeed.
These views are the views of the author and do not represent the official positions of the Department of Defense, the U.S. Army, or the United States Mission to NATO.