The final two months of 2018 have been a remarkably eventful period for observers of American civil-military relations—even for the Trump administration. In just the final two months of 2018, there was the pre-midterm election deployment of troops to the southwest border in response to the supposed “invasion” of the migrant caravan. There was the abrupt announcement of “mission accomplished” against the Islamic State in Syria and the decision to withdrawal substantial forces from Afghanistan, contrary to advice from senior military commanders and to the surprise of most of Congress. There was Secretary of Defense James Mattis’s unusual resignation in protest, giving two months’ notice, followed by President Trump’s decision to relieve him effective immediately. There was Trump’s first trip to a combat zone after two years in office, during which he defied expectations and broke security protocols by filming and naming special operators, and during which he later appears to have enjoyed a “MAGA” hat-autographing political rally with uniformed troops.
Somewhat lost amidst the clutter of the headlines is still another instance of political encroachment into military affairs: Trump’s recent interest in the case of former Army Maj. Matt Golsteyn, who is accused of killing a suspected combatant in a manner that would have violated the law of war and the Uniform Code of Military Justice. The president tweeted that he was “reviewing [Golsteyn’s] case”—prompting speculation over whether Trump might derail the case by issuing a pardon. Though the Golsteyn matter may be less high-profile as a question of civil-military relations than some of Trump’s other activities, it raises a novel concern. Is the prudence of a president’s involvement in an ongoing military prosecution a question of judicial independence, as such interjections are usually framed, or is it is instead a question of civil-military relations as an issue of military justice?
The answer has become less straightforward thanks to a recent Supreme Court opinion. But whether such presidential involvement is defined as a CMR problem or a justice problem affects public expectations on at least three fronts: how a president should engage with institutions of government; the legal posture and professional interests of the military’s chain-of-command; and the individual accused servicemember. To be clear, choosing one frame over the other does not indicate a preference for presidential interest over presidential abstention. The choice is over which frame to use to interpret and judge the presidential action, whichever it might be.
The Tweet Heard ‘Round the Courtroom
On the afternoon of Dec. 16, 2018, Trump injected himself into an ongoing criminal prosecution via Twitter.
At the request of many, I will be reviewing the case of a “U.S. Military hero,” Major Matt Golsteyn, who is charged with murder. He could face the death penalty from our own government after he admitted to killing a Terrorist bomb maker while overseas. @PeteHegseth @FoxNews
— Donald J. Trump (@realDonaldTrump) December 16, 2018
The president’s interest was apparently sparked that morning when Fox & Friends aired a segment about Golsteyn, a former Special Forces officer who the Army has now investigated twice for premeditated murder. Golsteyn has been accused of unlawfully killing a suspected, but unarmed, Taliban bomb-maker during an intense series of engagements in the Battle of Marja, Afghanistan, in 2010.
This is not Golsteyn’s first brush with public notoriety. The Army closed the first investigation into him in 2014, which led only to a non-judicial Board of Inquiry—essentially, a neutral panel of officers determining whether to separate Golsteyn for cause, and whether his serve would characterized as “honorable.” Golsteyn was then stripped of his Special Forces qualification and the Silver Star for valor he earned for actions in the battle, and was eventually discharged from service. Later, in an interview with Fox News in 2016, he defended his actions as simply performing his “end of the bargain”—relying on his judgment, as well as the responsibility entrusted to him by his superiors, to execute his mission “in good faith” given the rules and the “context” in which he was serving. The Army reopened the investigation, recalled him to active duty and found enough evidence to convince his chain-of-command to charge him.
This case, and the president’s reaction to this case, raises novel questions at the intersection of military criminal justice and the president’s role atop that system; the nature of war; principles of international humanitarian law and what behavior the military chooses to punish; and the president’s affirmative duty to “take care that the laws be faithfully executed.” In short, Golsteyn’s case reveals what happens when the immovable tension between the Commander-in-Chief Clause and the Take Care Clause of Article II of the Constitution meets the unrelenting force of a vocally opinionated Commander-in-Chief with little history of concern for traditional judicial norms of independence.
Commander-in-Chief Prerogative or Unlawful Command Influence?
In response to the president’s interest in the case, some voices have argued for a presidential pardon of Golsteyn. While the former star officer is not above the law, a representative argument reads, “it is understandable for Golsteyn to conclude that he had to kill the bomb maker” to quash the seemingly grave and immediate threat the suspect posed to Golsteyn’s troops and his mission. In this view, Golsteyn’s is the “unfortunate guilt” described by Alexander Hamilton in Federalist No. 74, which the pardon power was intended to remedy. Steve Vladeck, on the other hand, argues that Trump risks thwarting justice by exerting this unlawful command influence (UCI) over a case before due process systematically resolves it. Others disagree, pointing out that Trump merely said he was “reviewing” the matter, without signaling his preference or directing the chain-of-command to take a specific action, like dismissing the charges. Still others suggest that, even if the presidential tweet was not an explicit act of unlawful influence, it still creates problems by undermining public confidence in a fair judicial system and possibly by increasing pressure on military commanders to act in a manner they think the commander-in-chief prefers rather than what their sense of justice, the evidence and conscience demands.
The Court of Appeals for the Armed Forces opines on UCI regularly, but it has never squarely decided the legal question of whether this taint might come from a president along with a commanding general or a brigade commander. Article 37 of the Uniform Code of Military Justice (UCMJ) expressly prohibits such influence from any “person subject to this chapter,” and neither courts nor Congress have determined that a president is a “member of the regular component of the armed forces” and thus subject to the UCMJ. This cuts against arguments favoring an expansive view of UCI. Nevertheless, in 2013, a court-martial’s military judge did hold that President Obama’s public statements about his desire to see sexual assault offenders dishonorably discharged constituted such UCI, and imposed a curative remedy in that case that barred a punitive discharge if the accused had been found guilty. Moreover, in 2013, the Court of Appeals for the Armed Forces addressed how the civilian secretary of the Air Force could have committed actual or apparent UCI, even though that civilian, like the president, is not subject to the military jurisdiction Article 2 of the UCMJ.
Those opinions suggest that comments by a senior civilian in the military command hierarchy—like the commander-in-chief—have UCI-like effects, and thus raise a fundamental constitutional quality implicating a citizen’s Fifth Amendment due process rights. Vladeck, by analogy, believes the purpose of the UCI prohibition is a due process protection naturally binding the commander-in-chief, himself a de facto “military commander,” and therefore should chill presidential statements or guard against direct influence over these military justice cases. At the very least, whether presidential conduct or statements constitute UCI is still an open question—though strong arguments have, of late, directly reasoned that it must certainly apply.
And here is where the matter of judging the president’s action in the Golsteyn case gets squirrely. It is easy to say, as some have, that this is another example of Trump’s “extraordinary intervention into the American judicial system” in violation of separation of powers, similar to his comments before and after the court-martial of Sgt. Bowe Bergdahl or his other attacks on the FBI, the Department of Justice, federal courts or individual judges (and justices). But what has not been addressed in any depth is how the president’s role as commander-in-chief, a commander of the military if not a per se military commander (perhaps an illusory distinction) and titular head of the military’s criminal justice system affects these arguments.
There is a place where a president’s responsibility for and authority over the Armed Forces bumps up against a president’s law enforcement duties. Sometimes, such duties complement one another. But sometimes—like in the Golsteyn matter—they stubbornly meet head-to-head. In this case, the matter might be viewed from either perspective: that is to say, understood as civil-military relations problem, or understood as a judicial independence problem.Which framework is preferable is crucial to the question of understanding Trump’s actions.
Military Justice or Military Justice?
Current military law practice and procedure, and the traditionally integrated and capacious role of the commander in that process—all the way up to the president as commander-in-chief—would suggest that the Golsteyn case should be considered a civil-military relationship matter. Take a moment to consider the distinctiveness of the military justice system. It is the commander, not a prosecutor, who determines if a soldier will be charged under the UCMJ. It is a commander who determines whether the case will go from investigation to court-martial. It is a commander who selects the members sitting on the panel (something like a jury), and it is a commander who can dismiss charges. It is a commander who drives the post-trial process and it is a commander sitting at the end of the lengthy appellate process who takes final action to close the case. The military prosecutor, along with the military judge and defense counsel (and special victims’ counsel in sexual assault cases), and independent law enforcement authorities, manage the system itself acting as agents of the commander, the accused and the system itself. Commanders are vested, in degrees depending on rank and role, in tactics, operations and strategy against criminal activity undermining the command’s “good order and discipline.” Therefore, when a president interjects as the ultimate commander, he creates a series of challenges: He is a political actor invested with legitimate “amateur” authority, yet is involving himself in subject matter usually managed by his subordinates who possess highly-technical military “professional expertise.” In other words, it is a matter of civil-military relations. In this argument, Trump’s interjection into a far-from-resolved military justice prosecution is similar in nature to, say, President Lincoln’s direct interest in Gen. George McClellan’s campaign tactics, or President Lyndon Johnson’s similar conduct during the Vietnam War.
There is a ground for this view. Golsteyn’s alleged offenses, for example, occurred during, and as a result of, direct combat operations. It is worth emphasizing what this implies. Golsteyn was a legitimate and lawful combatant. He was engaged in pursuit of military objectives intended to accomplish an assigned mission on the ground in an active war zone. He used military means and resources to commit and attempt to hide his actions during that pursuit. This may be called a “combat-incidental” crime. Whether it should also be labeled a “war crime” is academic. The United States does not expose its servicemembers to investigation and trial by the International Criminal Court for such grave offenses. Rather, the accused faces trial by court-martial in the same manner as any other accused servicemember subject to the UCMJ, is exposed to the same punishments and enjoys the same due process protections. The circumstances of the alleged offense may affect the court-martial’s initial disposition of whether to charge, any potential mitigation, an affirmative defense or a victim-impact statement.
Golsteyn’s case illustrates this point: He explained that the extant rules of engagement forced him, begrudgingly, to release the suspect. He justified his later decision to target the Afghan following the man’s release from military custody by assessing him as an imminent danger—not just as a bomb-maker threatening troops, but to fellow Afghans who aided the United States in his original capture. To an objective observer, it would have been a cold-blooded assassination; but to Golsteyn, it represented mitigation of a tactical risk and a solution to a problem the rules of engagement created. This explains why the offense’s combat-incidental nature is relevant. Both Golsteyn’s mission and the rules of engagement came from a higher military headquarters—a command authority. And the plan of that higher headquarters was nested in the objectives and strategies of still higher headquarters and even more elevated command authority. At the top of that pyramid of rank and responsibility, regardless of how far removed from the battlefield he is, sits the president as the commander-in-chief.
Of course, none of this legally excuses the deed; whether Golsteyn’s conduct is described as a war crime or combat-incidental is relevant only so far as it contextualizes the offense and the offender. The issue is not whether Golsteyn should be judged. The issue is by what standard or gauge a commander-in-chief’s involvement in the military justice process should be judged, after someone like Golsteyn is accused of a crime.
A Case of Civil-Military Relations
If Golsteyn’s case is to be understood through the frame of civil-military relations, then the president’s involvement can be evaluated under traditional, well-traveled CMR routes—the kind that tend to define and defend the American experience of ensuring civilian supremacy over its armed defenders. Samuel Huntington’s “Objective Control” is the modern U.S. military’s lodestone for CMR. This theory would counsel the president to step away from the bully pulpit. No matter how angry he gets, how many Twitter followers tweet their support, or how loudly Fox News opines, he ought to allow the highly-specialized niche field of military justice practice to operate as far from rhetoric and sounds bites as possible. The military justice practitioners are autonomous and undisturbed to the same extent and for the same reasons that platoon leaders, company and battalion commanders, and pilots and ship captains (to name a few) have wide discretion in tactical combat planning and execution. Provided their work stays within their technical lanes of expertise and does not encroach, unsolicited, into policy advocacy or policy pressure, these cases remain far from of civilian strategist policy-makers’ direct intervention, under the conventional Huntingtonian view.
Peter Feaver’s argument that the civilian political leader always retains the “right to be wrong,” however, complicates the discussion of amateur authority over professional specialization because it anticipates and accounts for scenarios in which the political leader countermands or ignores the advice of his experts. Those experts, frustrated and uncomfortable though they may be, must accept and adapt to the civilian’s authoritative determination—or register their disagreement and maybe resign. On the other hand, the various types, values, and principles of “agency theory” that Feaver and others believe to be inherent structural attributes of the civil-military relationship suggest more of a case-by-case, personality- and interest-based range of acceptability. The “meeting of the minds” in which a civilian principal sketches out the scope of authority under which his military agent may freely act on his behalf imposes “fiduciary-like” duties of care, candor, confidentiality and fidelity in using that technical expertise to do what the principal cannot or will not do himself. That means, necessarily, that a principal can contract, expand or even puncture his agent’s sphere of responsibility, like granting a military field commander complete discretion, or “total authorization,” to plan and execute strikes in Afghanistan that, in previous administrations, may have required presidential approval. When applied to a CMR context, it might just mean that the president’s involvement is not per se good or bad, helpful or damaging; that involvement must be taken in its messy and uncertain context with an appreciation of each party’s expectations and their desired end states.
Eliot Cohen’s notion of the “unequal dialogue” paints such a picture. Recognizing that the scales are never balanced and that the weight of the civilian authority will always trump its military counterpart, the only constant in a healthy relationship between strategic-level civilian politicians and military leaders is that a dialogue in fact occurs—that is, that one party provides a venue and consideration for the considered views offered by the other. Sometimes, as with Lincoln and McClellan, the dialogue ends and the politician simply has to assert direct influence and ultimately dictate operational planning or even tactical choices if the military cannot or will not perform in a manner consistent with the civilian’s intent. Such is the nature of civilian supremacy over a military; recent history again seems to support the theory.
The Ortiz Case
A recent opinion out the Supreme Court has now made the task of interpreting and judging a president’s involvement in a case like Golsteyn’s is even more challenging. Ortiz v. United States, about which I wrote for Lawfare in July 2018, would seem to suggest that the UCMJ system is far more akin to civilian practice and purpose. If so, a president’s involvement should be discouraged notwithstanding his commander-in-chief duties. Under the Take Care Clause, presidents should be restrained in the realm of military justice in the same way and for the same reasons that such principles of non-interference are applied to the rest of the criminal justice system.
In Ortiz, the Supreme Court finally confirmed that it has review authority over the military’s criminal justice system, including its highest tribunal, the Court of Appeals for the Armed Forces. This, however, is not what makes the case an important statement about military justice. Justice Elena Kagan’s majority opinion took pains to discuss and refute a novel argument put forward in an amicus brief. In short, the amicus argued military justice was solely and purely an executive branch creation and responsibility, and was therefore subject only to executive branch processes. This would deny the Supreme Court any ability to review military justice cases because they are not really “cases” in a judicial sense at all. Because this take on the nature of the military’s criminal justice system—including its critical appellate components and their proper jurisdiction—was both “new” and “serious,” Kagan wrote, the Supreme Court felt compelled to explain why that theory was misplaced. Kagan’s rebuttal emphasized the myriad ways in which the military’s “integrated court-martial system” of justice paralleled both federal criminal practice and state criminal justice “structures.” For example, Kagan pointed to various due process protections, a hierarchy of independent appellate review, a long-established and stable body of governing case and statutory law, the res judicata effect of its decisions and statutory offenses that are indistinguishable from civilian jurisdictions. In effect, any disciplinary effect of a prosecution, court-martial and sentence that aides the commander in his duty to maintain “good order and discipline” is attendant, or supplementary, to the process’s “fundamentally judicial nature,” Kagan wrote.
This argument is not new. The U.S. military justice system, as currently designed, evolved from the traditional Articles of War. The UCMJ, enacted by Congress during the Korean War in part to prevent apparent biases and abuses of an unchecked executive branch practice, established the system as more than a creature of a president’s commander-in-chief responsibility. Subsequent amendments and appellate decisions have slowly, if not inevitably, created a “civilianization” of its processes, procedures and substantive law.
What is new, though, is that the Supreme Court described the military’s system without mentioning the role of the commander. In the seminal cases of military justice decided by the Supreme Court, like whether service members are shielded by First Amendment protections in 1974’s Parker v. Levy, the court has emphasized the fundamental differences—including the critical role of the commander—distinguishing military justice from civilian justice. “[T]he different character of the military community and of the military mission,” existing solely to strengthen national security and operating worldwide under dangerous and harsh conditions, the Parker court opined, deserves a system protecting and enabling constitutional due process somewhat differently than in say, Ohio, or in the U.S. Court of Appeals for the Third Circuit. By ignoring the chief arguments of past precedent, the Ortiz Court distanced itself (by implication) from all the reasons that elevate the generic commander to primary managerial roles over military justice.
The Ortiz view of military justice has been a long time coming. Consider the past decade of increasing congressional interest in how sex crimes are handled in the military, with serious suggestions for extracting the commander from the decision-making process altogether. The “civilianization” of military justice in the form of recent changes to the Rules for Courts-Martial, Military Rules of Evidence and even the form of some crimes—which really began in earnest with the UCMJ’s enactment in 1951—continues in force. Over this time, the commander’s central role has waned. Indeed, if the Supreme Court’s explanation of military justice in Ortiz is taken to its logical conclusion, the role of the commander may be at risk of disappearing entirely. After Ortiz, military justice exists primarily to promote justice for a certain class of people, rather than, as the Preamble to the Manual for Courts-Martial currently says, “to assist in maintaining good order and discipline in the armed forces, to promote efficiency and effectiveness in the military establishment, and thereby to strengthen the national security of the United States.” If the generic commander is no longer of central importance to the fundamental fairness and process of military justice, then by extension neither is the commander-in-chief.
There are two ways to understand this separate criminal justice system under military authority, which is ultimately subject to the control and direction of a civilian commander-in-chief. The choice between a CMR frame and a judicial independence frame depends on which word—which idea—is emphasized in the phrase “military justice.”
One solution, of course, is to say a president’s prerogative over military affairs inherently makes “good order and discipline” of service-members—necessary for their combat readiness—a proper subject for his concern and review. That is to say, this issue is always a CMR issue and ought to be discussed in those terms alone. On the other hand, since military criminal law is arguably (at least to the Supreme Court) a matter of fundamental justice with only subsidiary consequences for a particular command or commander, presidents—like his subordinate commanders—must keep their distance. That is to say, the issue is always a judicial independence issue and ought to be discussed in those terms alone. (And if Vladeck is correct, and the courts continue reasoning that senior civilians can commit actual or apparent UCI, the accused’s Fifth Amendment rights will be protected along with notions of judicial independence.)
Alternately, perhaps the issue should be considered as a sliding scale. The degree of presidential involvement can be thought of as being dependent on nature of the offense. When the nature of the alleged offense is clearly comparable to a civilian common law offense and has no nexus to a military context (other than the accused being a servicemember), then the president should deliberately abstain from direct or indirect influence before or during trial, notwithstanding his statutory ability to do so. For example, imagine a decorated combat veteran accused of killing his spouse at Fort Bragg, North Carolina, six months after returning from his fourth tour overseas in seven years. At this end of the scale, there is no reason—other than possible arguments for mental responsibility or sentence mitigation—to invade the province of the military’s court system and risk tainting the public’s perception of independent, unbiased and self-regulating military justice. The weight here falls heavily in the direction of judicial independence. But when the nature of the offense is more clearly associated with combat (“combat-incidental” as I define it here), as it does in Golsteyn’s case, only then should the president’s intentional influence be judged by whether it reflects, produces, or endangers a healthy CMR.
Trump’s involvement in Golsteyn’s case is not only a question of possible UCI. The case has too many subtle and nuanced implications. The wisdom of using commander-in-chief authority when prudence might dictate restraint; the proper role of military expertise in military courts; and the independent integrity of a criminal justice system ostensibly under presidential command all show up as a result of a single presidential tweet. The complexity of the issues at hand underlines that agreement or disagreement with the president’s intervention should not be left to instinctual resistance or support.
Thanks to Steve Vladeck for his helpful comments.