Published by The Lawfare Institute
in Cooperation With
In Section 540F of the National Defense Authorization Act of 2020, Congress directed the Department of Defense to conduct a feasibility study, and to report the findings of that study, on a potential new “alternative military justice system.”
Given the dramatic year that military justice has had, this direction might seem bold but not surprising. In the past year, President Trump granted controversial pardons and clemency to service members for conduct in combat that could have been charged as war crimes under international law, and for which the wheels of military justice had already come to rest or were already in motion. He also expressed interest in having the military investigate and potentially charge an officer after he testified in the impeachment hearing (in the end, he was only “fired”). And Congress has expressed continued consternation about the seemingly intractable problem of sexual assault in the military and skepticism about the efficacy of command efforts to “deal with it.”
A more broadly construed National Defense Authorization Act might have asked the Department of Defense to undertake some introspection of more foundational problem areas: the independence of commanders with respect to their investigative, prosecutorial, and judicial-like power (including insulation from political considerations and political pressure); and the kinds of misconduct that those commanders can and should address wielding those independent powers.
Unfortunately, Section 540F was much more reserved in its scope. The main thrust of the provision asked whether senior commanders (the commanders of organizations like combat divisions and corps, combatant commands, fleets, special operations forces, and entire installations) should be able to decide, like a district attorney, whether alleged violations of the Uniform Code of Military Justice (UCMJ) that constitute “felonies” (those offenses carrying maximum punishment of greater than a year in prison) shall be “referred” to courts-martial. The “alternative” model that the Defense Department must address, per Section 540F is simply whether that decision-making authority should lie solely with an independent senior judge advocate outside the chain of command. Congress wants the Defense Department to study not so much an “alternative military justice system” as it wants a study of an “alternative” to one component of an unwieldy system that military law scholar Eugene Fidell recently likened to a “Rube Goldberg Machine.”
As Michel Paradis wrote recently, this would be a big change. But not really a new system.
The change would certainly be important. As it stands, a commanding (lay) officer who has this “general court-martial convening authority” is also responsible for a great many decisions in the military justice process that do not—on their face—seem within the wheelhouse of a lay officer yet have a long historical pedigree. These include selecting the panel members (roughly equal to a civilian jury), from within the chain of command (all of whom would be junior in rank to this convening authority), as well as an authority to dismiss or withdrawal charges (Rules for Courts-Martial [R.C.M.] 401(c)(1) and 604(a)) before or during trial, to approve or reject offers to plead guilty (R.C.M. 705), and to approve expert witnesses requested by the defense and fund witness travel (R.C.M. 703). This wide delegation of power naturally creates concerns about fairness. Article 37 of the UCMJ prohibits such commanders from using their rank and position to unduly influence the course of the trial, the witnesses or the accused, the judge or panel members, and the prosecutor and defense counsel. But the appearance of impartiality and independence—and therefore the fundamental fairness and justice—of this system is always under scrutiny.
To be sure, changing to a system where the prosecutorial decision is vested in a licensed attorney with deep prosecutorial experience would be a dramatic escalation from simply calling for the removal of commanders from sexual assault prosecutions (which has been the focus of most congressional concern and oversight over the past decade). Though this would mark an enormous sea change in historical practice, it gets the U.S. better aligned with most of its allies’ military systems. The respective military justice systems in the U.K. and Canada made this shift in the 1990s following civilian court decisions denouncing their respective military codes for their lack of structural impartiality and independence.
“The Shadow Knows …”
The Department of Defense has not yet submitted its report of its feasibility study. But in the interim, a group of mostly law professors, most of whom are former U.S. military judge advocates, assembled as a “Shadow Advisory Report Group of Experts” (yes, that’s “SARGE” for short) to produce a report on potential recommendations for the military justice system. The group also includes the current judge advocate general of the U.K. Armed Forces and practitioner-scholars from Canada, New Zealand and the Netherlands. On April 20, they submitted a relatively short preliminary report to the Senate and House Armed Services committees, with their recommendations for a pilot program in what appears to be an effort to get ahead of the Defense Department report, or at least influence its drafting.
There are many strong, agreeable points raised in the “shadow report.”
To name a few: The report recommends ensuring the independence and professional protection of the “540F judge advocate” while sustaining the ability of the current convening authority to submit a statement outlining his or her considered opinion about how the crime and its prosecution affects discipline, efficiency, and morale; the report highlights the need for an independent “court administrative office,” like the U.K. uses, to manage arranging for the venue, timing of the court-martial, and the availability of witnesses and parties; and the report includes the persuasive data point that the United States’s closest allies somehow manage to survive and remain combat effective without having senior lay commanders making referral decisions and panel member selections. The last point reveals that neither their national security nor their criminal justice systems is at risk of collapsing due to the increasing “civilianization” of military justice.
Section 540F only tasked the Defense Department to study if and how an alternative to the current convening authority role in referral of charges would work. Thus, the shadow report’s authors reasonably confined themselves to this incremental—though significant—process change. But there were some areas in the shadow report that risk being repudiated or contradicted by the future Defense Department report, or not persuasive to the Armed Services committees..
It is possible that the shadow group will expand its preliminary report and offer its experienced insight to the Defense Department or the committees again. My commentary is offered to raise additional questions.
The Kind of “Felony” Should Matter
Any changes made to the commander’s convening authority role will be less justifiable or even plain wrongheaded if they do not acknowledge the distinctions between military-nexus offenses with no civilian analogue (what I’ll call “martial wrongdoing”)—like AWOL, malingering, trainee abuse, disobedience, conduct unbecoming an officer, and various others that may be “prejudicial to good order and discipline”—and what the British refer to in their code as “criminal conduct”—those acts or omissions already prohibited and punished in every civilian jurisdiction (murder, rape, etc.).
From 1969 to 1987, American military commanders could prosecute UCMJ offenses only when the circumstances of the case demonstrated some nexus to military affairs. In 1987, the Supreme Court reinterpreted the logic and history of military justice and discarded that short-lived and unworkable “service-connected” test. Commanders may now, as they could before 1969, prosecute all offenses, “martial” or otherwise (see Article 2, UCMJ, for the expansive list of those over whom the military has “personal jurisdiction”).
As a consequence, in the years since the 1987 decision, the military’s introspective analysis, criticism and defense of the UCMJ system has rested largely on only half of what makes the UCMJ (or any military justice code) unique and relevant: the kinds of conduct punishable in court and the singularly decisive role played by commanders. We tend to focus on the latter half, on defending or attacking the role of the commander as a convening authority, while neglecting to consider possible reforms in the range of offenses over which military courts have jurisdiction.
This is largely the tack taken by Paradis, in his May 4 review of the shadow report. While I agree with some of his larger points, he draws some surprising conclusions. He writes that “ordinarily, and historically, commanders have had discretion to enforce or disregard the verdict and sentence rendered by the tribunal based on their judgment as to what will best serve their broader mission.” This overlooks two important factors.
First, Paradis does not note the massive changes to the convening authority’s posttrial review power now encoded in Article 60a of the UCMJ. In a general court-martial, the convening authority has zero legal power to change a verdict and has no legal power to reduce, commute or suspend an adjudged sentence that included more than six months of confinement or a punitive discharge from the service—the very kinds of sentences most likely to qualify as the serious offenses for which Congress is interested in reforming the prosecutorial protocol.
Second, at least according to the commander-in-chief in promulgating the Manual for Courts-Martial, military justice serves three purposes: promoting justice, maintaining good order and discipline, and promoting efficiency and effectiveness—not just the commander’s “ broader mission.” Paradis reaches all the way back to Blackstone’s Commentaries to say that the modern military thinks the court-martial was “never about justice,” and to Toth v. Quarles (1955) to suggest that the modern military thinks justice and due process are “incidental” to enforcing obedience. But a 2018 case he cites is actually far more apt in capturing the prerogatives that govern modern military justice: the 2018 decision in Ortiz, where the court concluded that the court-martial “stands on equal footing” with state courts as a “bona fide judicial system.” Indeed, Supreme Court cases since the 1970s have recognized that the military justice system “will vindicate servicemen’s constitutional rights.”
The most common concerns about the military justice system come under the headings of unlawful command influence, panel selection irregularities, reluctance to prosecute some cases, or unwarranted overzealous decisions to prosecute in other cases. But in all of these instances, critical attention, as Paradis’s thought-provoking piece does in spades, always comes back to this one primary issue of commander involvement.
But it’s difficult to address the scope of a commander’s convening authority without discussing the difference in the kinds of crimes for which a commander has a stronger interest than, say, a lawyer would. Section 540F and the report’s distinction only between “felonies” and lesser offenses is insufficiently nuanced. An important difference exists between “service-connected” crimes and those that overlap with civilian offenses.
By removing the “service-connected” requirement in 1987, the Supreme Court also removed the principal justification for keeping commanders in the prosecutorial decision-making loop in the first place. Experienced commanders, arguably, know far better than any other actor whether a “martial wrongdoing” offense has enough of a negative effect on that command’s discipline, efficiency, operations or morale to justify “criminalizing” it. As the Army’s judge advocate general testified before Congress in 2019, commanders are the “fulcrum to any solution;” in the same hearing, his Air Force counterpart said, “[D]iscipline is a commander’s business.” If the chain of command, and no one else, is responsible and held accountable for completing its mission, it stands to reason that the chain of command will know when its ability is compromised because of the martial wrongdoing of a subordinate.
But the argument in favor of maintaining the commander’s role is less compelling when it comes to dealing with crimes that are not “service-connected.” Imagine that Private First Class Smith, a 21-year-old supply specialist, is caught attempting to shoplift jewelry and video game systems from the Target in town just outside Fort Anywhere, USA. It seems strange to suggest that Smith’s chain of command, ultimately led by a two- or three-star general, can persuasively comment or opine on the effect this crime had on the command’s discipline, efficiency or morale. This “civilian” larceny, one that is committed off-post and “off-duty” against a civilian victim (person or entity), seems outside of both the competency and concern of military authorities.
Let’s grant, for the sake of the hypothetical, though, that it is within the concern of military authorities. There are only a few reasonable data points relevant to that commander, who has court-martial convening authority, under these circumstances. First, it would be concerning to the commander that she may lose that suspected soldier (or sailor, airman, Marine … not sure what to call members of the Space Force) from the unit during periods of the investigation, pretrial defense, trial itself, and possible incarceration and discharge. This may affect the unit’s training readiness or the completion of some other mission for which that private first class is critical. But this concern is speculative and less plausible the lower rank the suspect has.
Second, it would be concerning to the commander that there may be the appearance of having a thieving pennyweighter on the unit roster; it is “not a good look.” But for brief periods in which recruiting standards were dropped to meet operational needs, the belief that military members are worthy of respect above and beyond members of other political, social and cultural institutions is a strong one. The military, understandably, would like to sustain a high-degree of public confidence in its service members. Many nonmartial offenses are unlikely to give a convening authority heartburn about “appearances.” Situations that would worry a commander about outside perceptions would be if the offense itself is grave, the number of suspects in the unit is high, or an alleged behavior reflects a negative trend in the unit’s behavior (or all three).
Third, if this was not a private first class but instead a lieutenant colonel or a sergeant major, one additional reason for which a commander might be directly interested in such a civilian crime would be the suspect’s rank: If the alleged criminal is a senior officer (or really any leader), it might be argued that the very act of the crime being perpetrated by a person of such a rank is harmful in some meaningful way to good order and discipline or discredits the service. But the UCMJ already contemplates this possibility, and the Supreme Court has long sustained the ability of a commander to prosecute conduct that is “unbecoming an officer,” conduct that directly prejudices good order and discipline, breaches heavily valued customs of the service, or conduct that would discredit the service, even if that conduct is constitutionally protected for civilians.
Seven Deadly Sins of Martial Wrongdoing
I argue, however, that removing the lay commander-as-convening-authority from the disposition loop for all felony offenses is, perhaps, going too far. For that matter, removing them from panel selection, or plea negotiations, for the martial offenses mentioned below might also be an unwarranted step (the shadow report suggests this on page 6, saying it is a “dangerous fiction” to believe that convening authorities are able to properly determine who is “best qualified” to serve on a jury-like fact-finding panel). There is a host of bad conduct, considered “felonious” (in terms of maximum punishment), for which a commander may in fact be better suited—both more concerned and more competent. I would suggest that there are seven categories of “martial wrongdoing,” all of which the commander-as-convening-authority can and should speak about intelligently:
(1) Acts or omissions that render individual service members less ready to do their duty or perform the mission (for example, absence without leave, unfitness because of excessive alcohol consumption or drugs, and certain types of self-injury and recklessness).
(3) Acts or omissions that interfere with command’s self-policing law enforcement authorities (for example, resisting arrest, obstructing the police, interfering with [command’s] investigation, false official statements).
(4) Acts or omissions that interfere with or degrade the command’s ability to execute its mission (for example, AWOL, desertion, contravention or violation of standing orders, disobeying lawful commands, disrespecting noncommissioned officers).
(6) Acts or omissions that depict, for no other reason than their inherent scandalous, shocking or immoral nature, a service member as something other than a morally upstanding service member, or that embarrass the service itself (for example, the U.K.’s disgraceful conduct of a cruel or indecent kind, or the United States’s conduct unbecoming an officer or conduct of a nature to bring discredit upon the armed forces).
(7) Acts or omissions that prejudice good order and discipline for some other case-specific, fact-dependent reason.
Looking at the Manual for Courts-Martial, in Appendix 12, there are few martial offenses that are not felonies—very short AWOLs, minor forms of disrespect, some dereliction of duty, for instance. In fact, nearly 60 percent of those acts that are criminalized under Article 134 (because they might “prejudice good order and discipline” or be “of a nature to bring discredit” upon the armed forces) are felonies.This leaves a lot of martial wrongdoing offenses that have no civilian analogue and for which the commander, not a judge advocate lawyer, is likely the best positioned to judge whether the alleged violation is severe enough to warrant prosecution under the circumstances.
It is not clear why even a senior judge advocate (who would be exceptionally qualified, if the shadow group’s recommended qualifications on page 9 are adopted), is in a better—more “just”—position to render a disposition decision on these offenses if the very reason they are crimes is because they do one or more of the seven things listed above, and undermine command and control or degrade the unit’s functionality. By focusing only on the commander-as-convening-authority and by considering the widest aperture of any felony, instead of distinguishing between martial and civilian offenses, Congress is not addressing the full issue. It may make sense to acknowledge that for some felonies (namely, ones intimately tied to military functioning), lay officers as convening authorities are qualified enough to sustain their current authorities, provided the current protections (like guaranteed legal advice and prohibitions on “unlawful command influence”) remain in place.
Where Is the Data?
There are also other ways in which the shadow group and the Defense Department might better advise Congress so that fully informed reform is possible.
For example, the shadow report lists several ways in which Congress has “cabined” convening authority power over the past century with “no adverse effect on the good order and discipline or effectiveness of America’s armed forces” (p. 4). First, a committee member might reasonably ask, “On what factual or empirical basis can you claim this?” To my knowledge, the shadow group did not engage anything close to a relevant cohort of leaders across the services on this question. Yet, the military has embarked on similar empirical studies before, and Congress has benefited from studies conducted by outside, private groups, like the “Cox Commission” (twice) (and the chair of that commission is a member of this shadow group). Future reports would do well to mirror those efforts.
Second, the report highlights examples of what most would consider positive developments in military justice, changes that place speed bumps on unilateral and decisive convening authority discretion: preliminary hearings, panel selection based on certain statutory qualifications, and need for pretrial legal advice, to name a few. However, these processes still ultimately rest with the commander-as-convening-authority making a binding decision at the top of these actions. The report’s gestalt message is that diminishing the lay commander’s convening authority power is not something to worry about, and it’s unclear to me how these examples support that proposition.
The commission had far better options available to it to support its argument about the lack of adverse effects of diminishing the commander’s convening authority power. It could have, for example, highlighted the changes to Article 60, UCMJ, which severely cabined postconviction clemency options for these flag officers. Another compelling data point that the committee omitted is that due process rights are already protected from further unjust convening authority interventions by an appellate judiciary sitting outside the chain of command. The rulings and analysis of this judicial body are answerable ultimately only to the civilian (and Senate-confirmed) members of the Court of Appeals for the Armed Forces and, of course, the Supreme Court. Neither of these reductions of command prosecutorial and judicial authority has been shown to undermine the appearance of command disciplinary authority. This supports the shadow report’s thesis: Reducing the authority of the commanding officer’s prosecutorial authority does not crater good order and discipline.
Why Just the Commander’s Convening Authority?
The report proposes a limit to a commanding officer’s prosecutorial discretion, but in reality, there are many other aspects of the commanding officer’s power that are ripe for reform. This is another aspect of the problem that Section 540F does not address and so the Defense Department report is also likely to leave untouched. The shadow report acknowledges that the alternative system further limits the cases in which a commanding officer retains convening authority power but “leaves in commanders’ hands the things that belong there” (p. 5). It is not clear what the report authors mean by “things,” but they do mention later that nonjudicial punishment powers and administrative corrective measures—considered to be analogous to standard employment discipline—are unchanged by the proposed alternative system (p. 8, note 35). That may be the case, but there are hosts of other commander authorities other than the power to refer a case to court-martial and to select the panel members, or to grant limited forms of clemency. To name a few provided in the Rules for Courts-Martial: the power to initiate quasi-criminal investigations (R.C.M. 303); the power to authorize a search or seizure (Military Rules of Evidence 315 and 316); or the power to issue an order in, say, a company, battalion, or brigade policy letter that is “punitive,” such that its violation becomes a violation of the UCMJ. These are significant prosecutorial and judicial-like (even legislative-like in the example of the punitive policy letter) powers unaffected by this congressional interest and Defense Department report. If we feel less confident or comfortable about the generic lay officer’s decision to refer cases, then our confidence and comfort level with these other authorities ought to be acknowledged and discussed too.
Is Timing Everything?
These additional powers are wielded by junior commanders from the first notice of alleged misconduct to the trial itself. Involving the chain of command often adds time, if not more questions. The report says “the military justice moves too slowly” (p. 6). Compared to what? State justice systems? Federal prosecutions? Or only in the eyes of interested parties, like suspects and accused? It is not clear what the right metric for time is—what would be “fast enough?” Military justice already protects service members’ right to a speedy trial by imposing time hacks on the process and forcing trial within 120 days of the charging (subject to extensions, usually requested by defense counsel) (R.C.M. 707). But, speaking anecdotally and based on this author’s personal observations, there are many instances in which a case that could have been prosecuted in civilian court was instead offered by the local district attorney (who generally has “right of first refusal” over nonmartial offenses committed in the civilian community) to the convening authority expressly because the military system acted more “expeditiously.” But that is admittedly just my experience. If my anecdotal experience seems insufficient to prove that the system moves fast enough (against some inconsistent or unnamed metric), it means the report should not use anecdotal experience and unnamed metrics to claim it is too slow either.
The report also claims that the alternative system will be swifter because lawyers are “fully and intimately aware of the system and the standards that govern charging decisions” (p. 6). This is a not-so-subtle dig at the competence of senior commanders.
In reality, however, the standard that governs charging decisions is based on probable cause, as well as the myriad factors listed in Appendix 2.1 of the Manual for Courts-Martial (what used to be in the discussion to R.C.M. 306). And this appendix was mandated by Congress in Article 33 of the UCMJ, and it reflects the president’s guidance addressed directly to commanders, not just to judge advocates. This guidance is modeled explicitly after the prosecution factors published by the American Bar Association, the Department of Justice and the National District Attorneys Association. A “probable cause” standard is also not a foreign concept to lay officers. Junior commanding officers, like a company commander with six years of commissioned service, may make investigative and law enforcement-like (and judicial-like) decisions by authorizing searches and seizures, based on a probable cause standard. We allow these same junior officers to charge service members with crimes, again based on a probable cause standard.
The military also sends future commanding generals to a sort of legal “finishing school” at the Senior Officer Legal Orientation Course and General Officer Legal Orientation Course at the Army’s Judge Advocate General’s Legal Center and School. And the military still provides legal counsel at every echelon of command. Based on all that experience, trust, and access to counsel, it is probably inaccurate and unfair to imply that general court-martial convening authorities do not understand the “standards that govern charging decisions” as well as lawyers. Of course, however, understanding the legal standard is just one part of the equation. Actually applying them is another—and , as indicated above, it is an open question how fairly that standard is applied and one for which no empirical data exists.
When Consistency Matters, and When It Does Not
The report also defends the desirability of more uniform referral decisions from command to command, and how the current system allows for too much deviation (p. 7, para. F). This implies deviation is a bad thing. Is it always? Not all commands have the same disciplinary challenges or command climate issues, nor the same need for certain kinds of more stringently disciplined behavior. The 82nd Airborne Division commander, for example, has a different set of disciplinary concerns than the commander of a Corps of Engineers district, or of a basic training battalion, or of an aircraft carrier. Furthermore, a command that is deployed in combat definitely has a different calculus when it comes to what should be, or even can be, dealt with via the military justice system at that time and place.
At bottom, we want each commander to isolate and consider each case on its own merits; that’s why Congress proscribed “unlawful influence,” why judge advocates suggest that there are “10 commandments” to be followed in order to avoid even the appearance of unlawfully influencing the process or the parties, and why the president promulgated a set of guidelines in Appendix 2.1 of the Manual for Courts-Martial. If we believe that all of these policies have merit, then the argument for dramatically reducing variation from one command to the next is counterproductive—ensuring more uniformity and “consistency” in disposition decision-making works against the goal of improving public (and service member) confidence. What ought to be “consistent” in the disposition decision-making is fairness, due process, transparency and the public’s confidence in the same.
Again, Where’s the Data?
There are other claims made in the report that lack the requisite empirical data to support them. The report supposes that the alternative system “may increase conviction rates because disposition decisions by an attorney are more likely to weed out cases that are marginal in terms of successful outcomes at trial or on appellate review” (p. 8). This would suggest that the committee assumed that such “marginal” cases are currently not being weeded out. That seems to be too bold. To support this assumption, we would need actual data from the field that shows statistically meaningful evidence of convening authorities disregarding their staff judge advocate’s legal advice in order to refer cases with one of two results: either an acquittal or dismissal of charges at trial, or an appellate reversal on factual or legal insufficiency grounds. Is there such data? This strikes me as the type of information that Congress would want to know before stripping convening authorities of power.
Not All Professions Self-Regulate With Professional Responsibility Codes and Licenses
Finally, the report supports its recommendation by saying that “[j]udge advocates are bound by service and civilian professional responsibility, unlike lay commanders” (p. 13, para. B). If by this the report means “published binding professional responsibility codes,” then this is true but also a bit unfair. It implies that lay commanders are not at all bound by other nontechnical officership norms, standards and expectations. That of course is not true.
There are real professional consequences if a commander “screws up” military justice, either in a particular case for objectively wrong actions (like tampering with witnesses, harassing the victim, failing to get to trial within the 120-day speedy-trial clock, or imposing unlawful pretrial punishment on suspects) or by systematically failing across cases. In the worst-case scenario, a commander could even be prosecuted for “knowingly and intentionally failing to enforce or comply with provisions of the UCMJ,” like Article 37’s prohibition on using rank or position to exert influence over the case or the court (per Article 131f, UCMJ), or for “dereliction of duty” (per Article 92(3), UCMJ).
Far more likely, however, a commander or a convening authority’s lack of discretion, judgment or judicial temperament is subject to reprimand if noticed by superiors who have the power of the evaluation report, or duty assignment, to wield. Relief for cause is an option; administrative discipline is an option. If lawyers (both civilian and those serving in the military) screw up in certain ways, they may lose their license to practice temporarily or permanently. Likewise, commanders may lose their “license to practice” in command, and that loss is usually permanent. In both cases—that of the lawyer and that of the commander—work is peer reviewed and policed. It is not accurate, I think, to suggest that commanders are unbound and, by implication, loose cannons who can escape professional censure, sanction or removal.
Don’t Discount the Value of Complexity
For these reasons, the shadow report on the Section 540F feasibility study is welcome and necessary. But it’s just the start of what can be deeper and more precise dialogue. The conversation should be about convening authorities more generally, commander authorities other than referral power, and which class of crimes (martial wrongdoing or “civilian-type” offenses) may reasonably benefit from status quo authorities.
It is worth discovering through rigorous empirical study whether commanders and convening authorities desire the powers the law currently affords them and whether they think it enables them in ways relevant to the protection of national security. If the answers to those two questions are yes, then it follows that we should actively discover whether commanders are, in fact, using the standards directed by Congress and the president to guide their prosecutorial decision-making. More work is to be done, and as Gen. Gordon Sullivan said, “Hope is not a method.”