Scholarship Saturday: The ability of appellate courts to reverse convictions that are factually insufficient keeps the military justice system respectable
In July 2019, the U.S. Navy-Marine Corps Court of Criminal Appeals (NMCCA) used its power to reverse convictions for factual insufficiency to overturn the trial court’s determination that Marine Colonel Daniel Wilson raped a subordinate’s six-year-old daughter. The NMCAA reasoned that inconsistencies in the child’s testimony rendered the evidence insufficient “to exclude every fair and rational hypothesis except that of guilt.” United States v. Wilson, No. 201800022 (N.M.C.C.A, July 1, 2019) at 53 (quoting the language used to describe the beyond a reasonable doubt standard in United States v. Loving, 41 M.J. 213, 281 (C.A.A.F. 1994)).
The decision in Wilson was immediately condemned by mainstream news outlets like ABC News and Newsweek, with the latter publishing a critique from Don Christensen, who was credited as being a “retired U.S. Air Force colonel and president of Protect Our Defenders”:
The decision is extremely disappointing and highlights the absurdity of the military justice system. Few if any civilian appellate courts have the ability to substitute their judgment for that of the trial court. The civilian justice systems understand that the persons who actually heard the witnesses testify are the only ones who should judge the credibility of witnesses, not appellate judges.
That assessment in the lay press eventually made it to a policy wonk publication that focuses on informing lawmakers and their staff: Congressional Quarterly’s Roll Call. In May 2020, Roll Call published an article commenting on the Wilson case that questioned whether the military Courts of Criminal Appeals ought to retain their factual sufficiency powers. The article was published under the headline: Victims’ lives turn on coming military sexual assault debate, and featured this assessment from former Deputy Assistant Secretary of State, and present-day pro-bono victim’s counsel, Mary Rose Hughes (who is also Of Counsel at the well-connected law firm, Perkins Coie):
The military system is remarkably different, but not in a way that advances the cause of justice.
A few months later, in early-December 2020, as the Fiscal Year 2021 National Defense Authorization Act (NDAA) was in its final stages of negotiation, Roll Call again published an article concerning the factual sufficiency power of the military Courts of Criminal Appeals: NDAA envisions new military policies on sexual assaults. This time, Roll Call featured commentary from Eugene R. Fidell, credited as “a senior research scholar at Yale Law School and professor at New York University who specializes in military law.” Roll Call summarized Professor Fidell’s comments as:
The current standard [allowing for factual sufficiency review] is a throwback to a time before there were judges or even many lawyers in courts martial, and so it was important for military appellate judges to almost completely reassess every conviction being appealed. Convictions are rarely overturned . . . but when that has happened in several rape and assault cases in recent years, it has led to uproars. . . . ‘In a certain category of high-profile cases, this will rebalance the scales.’
Professor Fidell’s assertion that factual sufficiency review is a relic of the past, and that its weight is no longer required to balance the scales of military justice bears some examination. A study of the legislative history discloses that, indeed, as Professor Fidell asserts, the lack of lawyers in the process and the lack of judges in the courtroom were part of the reason Congress gave the Service Courts of Criminal Appeal the power of factual sufficiency review. But, those were not the primary concerns that animated that decision. Rather, the Courts of Criminal Appeals were given factual sufficiency review powers out of concern that court-martial members make for poor finders of fact. Specifically, when it comes to court-martial panels, Congress was worried about 3 things:
(1) Congress gave the Courts of Criminal Appeals the power to review cases for factual sufficiency because the members of court-martial panels have more important things to do than reaching reliable verdicts.
This concern by Congress was well-demonstrated during the hearings wherein factual sufficiency review was created. That concern – that court-martial members have more important duties to attend to – is voiced directly in this exchange between the Chairman of the House Armed Services Legal Subcommittee, Paul Kilday (D-TX) and the Chairman of the Veterans of Foreign Wars (VFW) Special Committee on Military Service, Judge Donald E. Long (of Portland, Oregon):
Mr. Kilday. Is it your feeling, in the original trial before the court-martial, [that the accused] has had a comparable advantage of a presumption of innocence that he would have had before a civil court?
Hearing of Subcommittee 11, Legal, of the House Armed Services Committee, To Amend the Articles of War to Improve the Administration of Military Justice, to Provide for More Effective Appellate Review, to Insure the Equalization of Sentences, and for Other Purposes 93 CR Part 3, pages 1955-1956 (April 14, 1947).
(2) Congress gave the Courts of Criminal Appeals the power to review cases for factual sufficiency because members of court-martial panels are conditioned to return whatever verdict they think the convening authority, their commander, wants.
The lack of confidence in courts-martial to overcome command influence was well documented in the December 13, 1946 Report of the War Department Advisory Committee on Military Justice. Specifically, page 3 of the Report noted “a disquieting absence of respect for the operation of the system” and “considerable indignation at some of the current and all too frequent breakdowns.” The primary causes of that lack of trust included:
-A serious deficiency of sufficiently qualified and trained men to act as members of the court or as officers of the court.
Report at 4. To fix those deficiencies, the War Department (now the Department of the Army) recommended:
The final review of all general court-martial cases should be . . . by such board or boards as shall be designated by The Judge Advocate General or the Assistant. This reviewing authority shall have the power to review every case as to the weight of the evidence, to pass upon the legal sufficiency of the record and to mitigate, or set aside, the sentences and to order a new trial.
Report at 8-9.
The fear that the War Department Report voiced as to the deleterious effect of command influence on the administration of military justice has proven prescient. The point is proven by cases like United States v. Riesback, 77 MJ 154 (C.A.A.F. 2018) (finding it “intolerable” that the convening authority “stacked” the court-martial panel with personnel trained as victim-advocates in a “weak” sexual assault case, and neither the trial judge nor the Service Court of Criminal Appeals corrected the error); United States v. Boyce, 76 M.J. 242 (C.A.A.F. 2017) (finding actions by the Secretary of the Air Force and the Chief of Staff of the Air Force “placed an intolerable strain on public perception” of the fairness of the military justice system) and United States v. Barry, 78 M.J. 70 (C.A.A.F. 2018) (finding that pressure exerted by the Deputy Judge Advocate General of the U.S. Navy against the convening authority constituted “an improper manipulation of the criminal justice process.”)
(3) Congress gave the Courts of Criminal Appeals the power to review cases for factual sufficiency because the ability of court-martial panels to reach reliable verdicts is diminished by the fact that they are not required to be unanimous in their convictions, and they are too small to engage in meaningful deliberations.
As mentioned above, The War Department’s 1946 Report was intended to address the “disquieting absence of respect” that service members felt for the military justice system. Included in that report was this recommendation from Captain Jack L. Oliver, credited by the Department as a “company commander with extensive combat experience, and JA officer":
A unanimous vote should be required for non-military offenses, but not for military offenses.
Report at 123. Captain Oliver’s inclination to require unanimity to convict was supported by half (3/6) of the judge advocates interviewed by the Committee who had served on Boards of Review (the predecessor to the modern-day Courts of Criminal Appeals). Report at 55.
This linking of factual sufficiency review on appeal to the fact that court-martial verdicts are not required to be unanimous has not been confined to the War Department’s 1946 Report. The Courts of Criminal Appeals themselves have drawn that linkage in exercising their factual sufficiency review authority:
In the military justice system, where servicemembers accused at court-martial are denied some rights provided to other citizens, our unique factfinding authority is a vital safeguard to ensure that every conviction is supported by proof beyond a reasonable doubt. This authority ‘provides a source of structural integrity to ensure the protection of service members’ rights within a system of military discipline and justice where commanders themselves retain awesome and plenary responsibility.’
Judges Brown, Hecker and Teller, United States v. Rivera, No. ACM 38649 (A.F.C.C.A., February 18, 2016) at 4-5 (citing Ex Parte Quirin for the military justice system’s lack of a right to trial by jury – which, for a federal court, would include the right to a 12-member unanimous verdict).
The concern that not requiring unanimity in verdicts increases the rate of wrongful convictions is borne out by the research. Citing multiple empirical studies, the Supreme Court has long noted that as juries get progressively smaller, they make for less reliable fact-finders, Ballew v. Georgia, 435 U.S. 223, 232 (1978), and “the additional authorization of nonunanimous verdicts” amplifies that deficiency, Burch v. Louisiana, 441 US 130 (1979). Further, the data shows that the lack of unanimity combined with a small panel size acts only “to the detriment of one side, the defense.” Ballew, 435 US at 236.
Recent Changes in the National Defense Authorization Act have refined the process for how Courts of Criminal Appeals conduct factual sufficiency review, but the concerns that prompted the creation of that power in the first place remain.
Professor Fidell’s assertion that adding lawyers and judges to the military justice process has mitigated some risk to the accused is probably true. There is no empirical research to support that contention, as there is with the size of jury panels and the effect of lacking unanimity, but the assertion that lawyers are better at such could be true – certainly, it’s no surprise that we lawyers think so. But, it does bear noting that, as this column discussed last week in Scholarship Saturday: We hear drums, drums in the deep, the value of having courts-martial presided over by military judges is diminished by the fact that those judges fundamentally lack independence from the government that appoints them, employs them, decides whether they get to remain employed, decides whether they are promoted, and decides whether their assignment after their time on the bench will be the American version of Siberia.
In all, the foundational concerns that drove Congress to give the Service Courts of Criminal Appeals the power of factual sufficiency review remain as pressing today as they were prior to 1950: Given the Congressional Research Service’s recent discussion of the renewed demands of great power competition on our forces, court-martial members detailed to weigh evidence are likely to be distracted by more important military duties piling up during their absence from their units. The cases cited earlier in this article show that courts-martial proceedings are still heavily burdened by unlawful command influence. And, the ability of courts-martial to engage in reliable fact-finding continues to be structurally hamstrung by the combination of lacking a unanimity requirement and having too few members to engage in meaningful deliberations or to benefit from diversity (at least as compared to a 12-member jury).
It is in that light – that there are still many trial-level structural deficits to be counterbalanced by factual sufficiency review on appeal – that Congress recently made substantive changes Article 66 of the UCMJ, which is the statute that gives the Courts of Criminal Appeals their factual sufficiency review power. This blog’s Editor-in-Chief, Professor Brenner Fissell discussed the changes to Article 66 last week in The New Factual Sufficiency Standard. In that article, he noted that, in Section 542 of the most recent NDAA Congress did two main things as regards the factual sufficiency review power. First, Congress has limited the Courts of Criminal Appeals to only conducting factual sufficiency review in cases where the accused makes a “specific showing of a deficiency in proof.” Second, Congress has limited the power to only those cases where “the Court is clearly convinced that the finding of guilty was against the weight of the evidence.”
The first measure – requiring the accused to make a showing of deficiency – does not change much in practice. Very few cases are reversed for factual insufficiency, so it stands to reason that even fewer of those cases would have come before the court without there first being a specific complaint of error from the accused. The second change can seem a little harder to parse. It limits the power to reverse a conviction for factual insufficiency to only those cases where the appellate judges are “clearly convinced” that the conviction “was against the weight of the evidence.” Given that the “weight of the evidence” required to convict is “proof beyond a reasonable doubt,” the most straight forward reading Congress’ words is that, to reverse a conviction, the Court of Criminal Appeals must be clearly convinced that there was not proof beyond a reasonable doubt. I’m hard-pressed to think of any case where a Court of Criminal Appeals has set aside a conviction where they were only sort-of convinced that the government failed to meet its burden at trial, or where they thought it was a coin toss as to whether that burden was satisfied. In practice, I suspect every case that was set aside under the old standard would be set aside under the new, as well.
Providing a bit more clarity on the standard the Courts of Criminal Appeals are supposed to apply when reviewing for factual sufficiency is not the only thing that Section 542 of the Fiscal Year 2021 NDAA accomplished. That provision also codified the U.S. Court of Appeals for the Armed Forces (CAAF) role in reviewing a Service Court of Criminal Appeals’ factual sufficiency determination. Specifically, the NDAA codified a decision from CAAF issued in 2005 holding:
It is within this Court’s authority to review [a Service Court of Criminal Appeals’] determination of factual sufficiency for application of correct legal principles. At the same time, this authority is limited to matters of law; we may not reassess a lower court’s fact-finding.
United States v. Leak, 61 MJ 234, 241(C.A.A.F. 2005).
The recent NDAA changes merely codify that holding. Specifically, Article 67(c) has been changed to read as follows (the NDAA deleted nothing from subparagraph (c), the underlining below signifies language the NDAA added):
(c)(1) In any case reviewed by it, the Court of Appeals for the Armed Forces may act only with respect to-
The addition of subparagraph (c)(1)(C) merely codifies the holding from Leak: Service Court of Criminal Appeals’ decisions on factual sufficiency may be reviewed by CAAF. However, that power of review is still constrained by subparagraph (c)(4)—limiting CAAF to “action only with respect to matters of law.” Put those two provisions together and you get Leak: CAAF may review the Service Court of Criminal Appeals’ factual sufficiency determinations to ensure they have employed a correct understanding of the applicable legal principles (including whether they properly interpreted Congress’ new “clearly convinced” limitation on their ability to overturn a conviction for factual insufficiency), but the application of that law to the facts remains a matter for the Service Courts of Criminal Appeals to resolve in their sole discretion.
In sum, the legislative history quoted and cited above shows that, essentially, the whole reason the Service Courts of Criminal Appeals were created was to conduct factual sufficiency reviews. The reasons they were given that power in the first place remain true today. Court-martial members continue to have more important military duties to attend to; commanders and their staff (including legal staff) continue to exert their influence over the system in a way that is hostile to ordered liberty; and structurally court-martial panels are still too small and lacking in unanimity to be considered reliable fact-finders, at least as regards convictions.
The new NDAA provisions provide a bit of structure to how the factual sufficiency review power will be exercised going forward. But the power itself remains. And, it remains powerfully important for Courts of Criminal Appeals to be on the look-out for appropriate cases in which to exercise that power. Otherwise, we may find ourselves back in the position the War Department found itself in around 1946, where military personnel apparently had little respect for the military justice system.