Ronald C. ROBERTS, Senior Master Sergeant

U.S. Air Force, Appellant



No. 03-0109


Crim. App. No. 34236



United States Court of Appeals for the Armed Forces


Argued October 7, 2003


Decided March 23, 2004


ERDMANN, J., delivered the opinion of the Court, in which GIERKE, EFFRON and BAKER, JJ., joined.  CRAWFORD, C.J., filed an opinion concurring in the result.




For Appellant:  Captain David P. Bennett (argued); Colonel Beverly B. Knott, Major Terry L. McElyea, and Major Kyle R. Jacobson (on brief).


Appellee:  Major Shannon J. Kennedy (argued); Colonel LeEllen Coacher and Major Jennifer R. Rider (on brief).


Military Judge:  Michael B. McShane



This opinion is subject to editorial correction before final publication.



Judge ERDMANN delivered the opinion of the Court.

Appellant Senior Master Sergeant Ronald C. Roberts was convicted in a contested general court-martial of altering a public record, removing a public record, drafting and printing a false Air Force form and making a false statement, all in violation of Article 134, Uniform Code of Military Justice [UCMJ] 10 U.S.C. § 934 (2000).  The offenses related to the falsification of Roberts’ 1998 annual review, known as an Enlisted Performance Report (EPR).  Roberts was sentenced to 12 months’ confinement and reduction to the lowest enlisted grade.  The convening authority approved the confinement but lessened the grade reduction to senior airman.

Prior to Roberts’ court-martial, the defense submitted a motion requesting that the court compel the Government to disclose derogatory data regarding its witnesses, including the lead investigator, Air Force Office of Special Investigations (AFOSI) Special Agent (SA) “M.”  The military judge reviewed in camera a record of an internal investigation on SA M and denied the motion.

On appeal to the Air Force Court of Criminal Appeals, Roberts argued, inter alia, that the military judge’s denial of discovery regarding SA M was erroneous because the information withheld related to SA M’s veracity and therefore could have been used to impeach him.  The Air Force Court of Criminal Appeals affirmed the findings and sentence in an unpublished opinion.

We granted review of the following issue:


 We hold that although the military judge erred in not ordering the disclosure of the investigative file, the nondisclosure was harmless beyond a reasonable doubt and we therefore affirm the lower court decision.


The charges in this case arose out of the circumstances surrounding Roberts’ 1998 EPR.  Roberts’ wing commander declined to sign the original EPR that was prepared for his signature due to his concerns over a prior substantiated sexual harassment complaint against Roberts by a subordinate.  He instead directed his vice commander to sign the EPR.  According to testimony at trial, this lower level endorsement virtually guaranteed that Roberts would not receive a promotion to chief master sergeant.  The vice commander signed Roberts’ EPR.

Roberts was subsequently selected for promotion to chief master sergeant, which resulted in a complaint being filed with the Inspector General (IG).  The IG investigation revealed multiple inconsistent copies of Roberts’ 1998 EPR on file, which led to a criminal investigation.

In the course of the investigation by AFOSI, the true EPR signed by Roberts’ vice commander was never found.  However, two different falsified versions of the EPR were uncovered.  One version was purportedly signed by Roberts’ wing commander, while the other version was purportedly signed by Roberts’ vice commander.  The version with the vice commander’s purported signature contained glowing language that the vice commander testified he would not have approved, including a bullet which stated Roberts displayed “unmatched initiative, professionalism, and personal concern of [sic] subordinates.”

During an interview with defense counsel prior to trial, the lead AFOSI agent, SA M, revealed that he had previously been disciplined, but declined to provide any details.  Defense counsel subsequently requested that the Government provide the defense with copies of all disciplinary actions taken against SA M.  A Government attorney-adviser at AFOSI replied in a memorandum that he had reviewed records including those maintained on SA M, and found no information that had to be disclosed.  The attorney-adviser further stated that his review of the SA M investigation “did not reveal that SA [M] lied or falsely testified about the matter.”

Defense counsel then made a motion to compel discovery pursuant to, inter alia, Rule for Courts-Martial 701 [R.C.M.], which asked for all derogatory data against all prospective Government witnesses, and in the alternative, for an in camera review of that information.  The motion specifically referenced the disciplinary action against SA M.

The military judge reviewed the records concerning the investigation of SA M in camera.  The information provided to the military judge revealed that approximately three years before Roberts’ court-martial, while SA M was at a training course, he had sexual intercourse with another married AFOSI member.

The AFOSI investigative report of that incident contains a summary of an interview with SA M on February 13, 1997, in which he reportedly was given a rights advisement, and initially stated, “This is bullshit.  There is no improper relationship,” before he stopped talking.  SA M was subsequently given testimonial immunity and admitted to the intercourse with the married AFOSI member in a sworn statement.  He was not criminally charged, but he did receive an Unfavorable Information File.

After reviewing the file provided by the prosecution, including the summarized report of the February 13 interview, the military judge denied the defense motion, stating, “I have concluded that that file does not contain any information . . . which would be proper impeachment matters for the defense to have knowledge of and to use in their case.”


Roberts argues that the military judge erred in refusing to order disclosure of information which showed that SA M made a false official statement.  He claims he was entitled to the information under R.C.M. 701 and under Brady v. Maryland, 373 U.S. 83 (1963).1  Our review of discovery/disclosure issues utilizes a two-step analysis:  first, we determine whether the information or evidence at issue was subject to disclosure or discovery; second, if there was nondisclosure of such information, we test the effect of that nondisclosure on the appellant’s trial.

Nondisclosure of information pertaining to disciplinary action against SA M.


The right of an accused to obtain favorable evidence is established in Article 46, UCMJ, 10 U.S.C. § 846 (2000).  This statute is implemented in R.C.M. 701 which details the liberal discovery practice in courts-martial.  Rule for Courts-Martial 701 sets forth the rights and corresponding obligations of the parties to a court-martial.  Of particular importance in this case are the Government’s duties concerning disclosure of information requested by the defense which is “material to the preparation of the defense . . . ."  R.C.M. 701(a)(2)(A). 

Discovery practice under Article 46 and R.C.M. 701 “promote[s] full discovery . . . eliminates ‘gamesmanship’ from the discovery process” and is “quite liberal . . . .  Providing broad discovery at an early stage reduces pretrial motions practice and surprise and delay at trial.”  Manual for Courts-Martial, United States (2002 ed.), Analysis of Rules for Courts-Martial A21-32.  The military rules pertaining to discovery focus on equal access to evidence to aid the preparation of the defense and enhance the orderly administration of military justice.  To this end, the discovery practice is not focused solely upon evidence known to be admissible at trial.  See United States v. Stone, 40 M.J. 420, 422 (C.M.A. 1994)(citing United States v. Lloyd, 992 F.2d 348, 351 (D.C. Cir. 1993)).  The parties to a court-martial should evaluate pretrial discovery and disclosure issues in light of this liberal mandate.

Roberts made a proper request to compel discovery of information and the Government declined disclosure.  Thereafter, at Roberts’ request, the matter was submitted to the military judge to review in camera, pursuant to R.C.M. 701.  Under such circumstances, the military judge may review the information ex parte, in camera, and may order “that the discovery or inspection be denied, restricted, or deferred, or make such other order as is appropriate.”  R.C.M. 701(g)(2). 

An appellate court reviews a military judge’s decision on a request for discovery for abuse of discretion.  United States v. Morris, 52 M.J. 193, 198 (C.A.A.F. 1999).  A military judge abuses his discretion when his findings of fact are clearly erroneous, when he is incorrect about the applicable law, or when he improperly applies the law.  In this case, we are not dealing with any factual determinations.  We are reviewing the military judge’s determination whether this requested evidence was “material to the preparation of the defense” for purposes of the Government’s obligation to disclose under R.C.M. 701(a)(2)(A).  The military judge’s determination of materiality in this respect is a question of law that we review de novo.  Id.

Information about SA M’s denial of misconduct to which he subsequently confessed, whether or not it constituted proof of a false official statement, was probative of his truthfulness and could have been used in preparation of the defense to determine whether SA M could be impeached under Military Rule of Evidence 608(b) [M.R.E.].  Under M.R.E. 608(b), specific instances of the conduct of a witness, if probative of truthfulness or untruthfulness, may in the discretion of the military judge be inquired into on cross-examination of the witness.  In this context however, the question is not whether the military judge would or would not have permitted the cross-examination under M.R.E. 608(b), but whether the information was material to the defense’s preparation for trial.  See R.C.M. 701(a)(2)(A).  In light of the Government’s incorrect statement that the records of the SA M investigation “did not reveal that SA [M] lied or falsely testified about the matter”, the defense was left with no basis upon which to believe SA M’s veracity could be attacked. 

The defense had a right to this information because it was relevant to SA M’s credibility and was therefore material to the preparation of the defense for purposes of the Government’s obligation to disclose under R.C.M. 701(a)(2)(A).  In addition, the military judge improperly limited the scope of discovery when he apparently focused on admissibility, ruling that the “file does not contain any information . . . which would be proper impeachment matters for the defense . . . to have knowledge of and to use in their case.”  (Emphasis added).  We hold that the military judge erred as a matter of law when he denied the defense motion to compel discovery.  


Effect of Erroneous Nondisclosure.

Having determined that the information should have been disclosed during discovery, we now turn to the second phase of our analysis.  In this context, an appellate court reviews the materiality of the erroneously withheld information in terms of the impact that information would have had on the results of the trial proceedings.  Both phases of this analysis involve a determination of “materiality” but they are two distinct inquiries.  The first inquiry at the trial level is whether the information would be “material to the defense” in the preparation of their case and the second inquiry, at the appellate level, determines the materiality of the withheld information to the results of the trial.

This Court has adopted two appellate tests for determining materiality with respect to the erroneous nondisclosure of

discoverable evidence.  United States v. Hart, 29 M.J. 407, 410 (C.M.A. 1990).2 

The first test applies to those cases in which the defense either did not make a discovery request or made only a general request for discovery.  Once the appellant demonstrates wrongful nondisclosure under those circumstances, the appellant will be entitled to relief only by showing that there is a “reasonable probability” of a different result at trial if the evidence had been disclosed.  United States v. Bagley, 473 U.S. 667, 682 (1985); Hart, 29 M.J. at 410; see also Strickler v. Greene, 527 U.S. 263, 290 (1999).

The second test is unique to our military practice and reflects the broad nature of discovery rights granted the military accused under Article 46.  Where an appellant demonstrates that the Government failed to disclose discoverable evidence in response to a specific request or as a result of prosecutorial misconduct, the appellant will be entitled to relief unless the Government can show that nondisclosure was harmless beyond a reasonable doubt.  Hart, 29 M.J. at 410.    Roberts made a specific request for information about disciplinary actions involving SA M.  The requested information existed and the Government declined to disclose it.  The request was reviewed by the military judge who erroneously denied the motion to compel disclosure.  We will therefore use the “harmless beyond a reasonable doubt” standard in determining whether Roberts is entitled to relief.3 

The circumstantial evidence of Roberts' guilt was overwhelming.  Handwriting analyses showed that the signatures on both of the questioned EPRs had been traced.  Roberts’ vice commander testified that the signature found on the EPR purportedly signed by him was not his, and that he would not have approved the laudatory language in two of the six bullets under the “Rater’s Rater” section.  Roberts had a motive to fabricate his EPR, in order to get a promotion that he knew would otherwise have been denied him.  His fingerprint was found on one of the versions of the 1998 EPR next to the wing commander’s falsified signature.

Analysis of Roberts’ fingerprints showed evidence that the upper layer of skin on his hands had been deliberately altered.  An altered copy of Roberts’ 1997 EPR with a false signature was

found in his office desk.  It is well accepted that circumstantial evidence is sufficient to sustain a finding of guilt.  R.C.M. 918(c); see generally United States v. Lewis, 51 M.J. 376, 380 (C.A.A.F. 1999); United States v. Caballero, 37 M.J. 422, 425 (C.M.A. 1993).

Moreover, SA M, although an important witness, was far from the linchpin of the Government’s case.  He provided foundational testimony for the prosecution exhibits of the various EPR documents and media files.  He also testified about his role in securing Roberts’ fingerprints and in obtaining handwriting analyses of the signatures on the questioned EPRs and about his interview of Roberts.  Nine other witnesses testified to Roberts’ motive to forge the EPR, his access to the EPR at the critical periods, and the substantial evidence that the documents were in fact forged.

In light of the evidence in the entire record, we are satisfied that the nondisclosure was harmless beyond a reasonable doubt.


     The decision of the United States Air Force Court of Criminal Appeals is therefore affirmed.


1 Because we find that the military judge erred by not compelling disclosure pursuant to R.C.M. 701(a)(2)(A) we do not address any separate entitlement Roberts may have had to this information under Brady v. Maryland, 373 U.S. 83 (1963).


2 Since Hart, this Court has issued a number of decisions that deal with the materiality of undisclosed, discoverable evidence.  United States v. Mahoney, 58 M.J. 346, 349 (C.A.A.F. 2003); United States v. Morris, 52 M.J. 193, 197-98 (C.A.A.F. 1999); United States v. Stone, 40 M.J. 420, 422-23 (C.M.A. 1994); United States v. Green, 37 M.J. 88, 89-90 (C.M.A. 1993); United States v. Watson, 31 M.J. 49, 54-55 (C.M.A. 1990).  As these cases have sometimes used different terminology in stating the applicable tests, we take this opportunity to clarify the respective tests and burdens.


3 Although the military judge in this case conducted an in camera review of the disputed evidence under R.C.M. 701(g)(2), we review that ruling as a matter of law, giving no deference to that ruling under our de novo standard of review.  Similarly, the appellate standard of review for assessing the impact of improper nondisclosure is not deferential because we are not reviewing any trial level decision.  Our appellate assessment of impact is no different regardless of whether the discovery issue was ruled on by the military judge under R.C.M. 701(g)(2) or whether it arose from a Government decision to withhold certain evidence that was not discovered until after trial.  


     CRAWFORD, Chief Judge (concurring in the result):

     This case is yet another example of the majority selectively refusing to follow Supreme Court precedent and fashioning a different standard for the military without a showing of military necessity.  While I agree with the result in this case, I would apply the Supreme Court’s “reasonable probability” standard to the issue of wrongful nondisclosure as this Court did only last term in United States v. Mahoney, 58 M.J. 346 (C.A.A.F. 2003).  Also following the Supreme Court, I would apply the “reasonable doubt” standard where there has been perjured testimony.  United States v. Bagley, 473 U.S. 667, 679-80 (1985).  By abandoning Supreme Court precedent in the present case, the majority is inconsistent not only with this Court’s rulings as well as the Supreme Court’s, but it also undercuts the soundness of its adjudication.  While the end result may not change, the rationale employed by the majority will only serve to “undermine[] public confidence in the stability, and predictability of military justice.”  United States v. Kahmann, ___ M.J. ___ (C.A.A.F. 2004)(Crawford, C.J., concurring in the result).

     Over the years, the courts have sought to ensure that the accused’s right to a fair trial is not imperiled by the nondisclosure of evidence to the defense.  Accordingly, the Supreme Court’s standard of review for wrongful nondisclosure cases has evolved.  See, e.g., Strickler v. Greene, 527 U.S. 263 (1999); Kyles v. Whitley, 514 U.S. 419, 437 (1995); United States v. Bagley, 473 U.S. 667, 682 (1985); United States v. Agurs, 427 U.S. 97, 103 (1976).  In Agurs, the Supreme Court applied a “strict standard of materiality not just because [such cases] involve prosecutorial misconduct, but more importantly because they involve a corruption of a truth-seeking function of the trial process.”  Agurs, 427 U.S. at 104.  But, when there is “no reason to question the veracity” of the verdict, then the Agurs materiality test would not apply.  Id.  Our Court likewise has developed various tests to decide the issue of wrongful nondisclosure.  See, e.g., United States v. Hart, 29 M.J. 407, 410 (C.M.A. 1990)(applying a beyond a reasonable doubt standard); United States v. Eshalomi, 23 M.J. 12, 24 (C.M.A. 1986)(indicating that “we need not face [the Bagley] issue now because, even under the [Bagley] reasonable-probability test, reversal is required in this case”). 

     This evolution of the standard of review for wrongful nondisclosure stems from the question of which party bears the burden of proof.  Addressing this question in conjunction with standard of review, the Court stated in Strickler, 527 U.S. at 290 (citation omitted), that the issue is “whether the favorable evidence could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict.”  The burden is on the defense to show, first, that there has been non-disclosure; second, whether the evidence was material; and, third, whether “there [was] a reasonable probability that the jury would have returned a different verdict[.]”  Id. at 296.  See also Banks v. Dretke, ___ U.S. ___ (2004).  If the defense meets the “reasonable probability” standard, it generally would not be possible for the Government to prove beyond a reasonable doubt that the failure to disclose had no impact on the verdict.       

     Notwithstanding the burden of proof, the appropriate standard of review remains “reasonable probability” of a different verdict.  In keeping with this approach, this Court recently reiterated in Mahoney, as noted supra, that the “reasonable probability” test determines whether there has been wrongful nondisclosure of exculpatory evidence.  In so doing, we opined:

The constitutional guarantee of due process requires that “criminal defendants be afforded a meaningful opportunity to present a complete defense.”  California v. Trombetta, 467 U.S. 479, 485 (1984).  Accordingly, the prosecution must disclose to the defense “evidence favorable to an accused . . . where the evidence is material either to guilt or to punishment.”  Brady [v. Maryland, 373 U.S. 83, 87 (1963)].  “Favorable” evidence under Brady includes “impeachment evidence . . . that, if disclosed and used effectively, . . . may make the difference between conviction and acquittal.”  United States v. Bagley, 473 U.S. 667, 676 (1985) (citations omitted).  However, like other forms of exculpatory evidence, impeachment evidence is “material” to guilt or punishment “only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.”  Id. at 682.  Under the “reasonable probability” standard of materiality, “[t]he question is not whether the defendant would more likely than not have received a different verdict with the evidence, but whether in its absence he received a fair trial.”  Kyles, 514 U.S. at 434.  Therefore, “[a] ‘reasonable probability’ of a different result is

. . . shown when the government’s evidentiary suppression ‘undermines confidence in the outcome of the trial.’”  Id. (quoting Bagley, 473 U.S. at 678).    


Id. at 349.  As demonstrated through our citations in Mahoney, this approach is consistent with the Constitution and Supreme Court precedent.  Indeed, Bagley established a single standard of review, whether there is no request, a general request, or a specific request for the nondisclosed evidence.  Bagley, 473 U.S. at 682. 

     The “materiality” issue as applied by the majority and found in Rule for Courts-Martial 701(a)(2)(A) [hereinafter R.C.M.] is similar to the language employed in Brady v. Maryland, 373 U.S. 83 (1963), and its progeny:

We find the Strickland formulation of the Agurs test for materiality sufficiently flexible to cover the “no request,” “general request,” and “specific request” cases of prosecutorial failure to disclose evidence favorable to the accused:  The evidence is material only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.  A “reasonable probability” is a probability sufficient to undermine confidence in the outcome. 


Bagley, 473 U.S. at 682.  The materiality standard mentioned in R.C.M. 701 and the Supreme Court decisions does not require the demonstration by a preponderance of the evidence that the disclosure would have resulted in the defendant’s acquittal.

Bagley’s touchstone of materiality is a “reasonable probability” of a different result, and the adjective is important.  The question is not whether the defendant would more likely than not have received a different verdict with the evidence, but whether in its absence he received a fair trial, understood as a trial resulting in a verdict worthy of confidence.


Kyles, 514 U.S. at 434.  It is not a “sufficiency of the evidence test.”  As the Court noted, a Bagley error could not be treated as harmless because “a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different,” Bagley, 473 U.S. at 682, necessarily entails the conclusion that the suppression must have “had substantial and injurious effect or influence in determining the jury’s verdict.”  Brecht v. Abrahamson, 507 U.S. 619, 623 (1993)(quoting Kotteakos v. United States, 328 U.S. 750, 776 (1946)). 

     We should follow Supreme Court precedent which sets forth a bright line rule for nondisclosure and closely approximates the result this Court seeks to achieve today.  The “reasonable probability” rule ensures the rights of defendants and protects the interests of the Government.  It is a predictable and consistent rule rather than one that depends on predilections of the appellate courts in the future.  When the categories mentioned by the majority have to be further defined, I fear the end result will be further selective application of Supreme Court precedent in the future.

     Because the result in this case is the same regardless of which standard is applied, I concur in the result reached by the majority.

Home Page |  Opinions & Digest  |  Daily Journal  |  Scheduled Hearings  |  Search Site