PUBLISHED OPINION OF THE COURT
BOOKER, Senior Judge:A military judge sitting as a general court-martial convicted the appellant, after mixed pleas, of burglary, conduct unbecoming an officer and a gentleman, fraternization, and indecent assault, respectively violations of Articles 129, 133, and 134, Uniform Code of Military Justice, 10 U.S.C. §§ 929, 933, and 934. The convening authority (CA) approved the announced sentence of confinement for three years, forfeiture of all pay and allowances, and a dismissal from the U.S. Marine Corps. We set aside the guilty findings with respect to the sole Article 133 offense and one of the indecent assault allegations, but otherwise affirmed the findings and, after reassessment, the sentence. United States v. Lee, No. NMCCA 200600543, 2007 WL 1890683, 2007 CCA LEXIS 233, unpublished op. (N.M.Ct.Crim.App. 26 Jun. 2007).
Proceedings After Initial NMCCA Action
The Court of Appeals for the Armed Forces granted review of an allegation by the *536appellant that his defense counsel at trial, Captain (Capt) Reh, failed to disclose a conflict of interests, namely, his transfer into prosecution duties during the course of the litigation, thereby resulting in an uninformed selection of counsel. The Court of Appeals for the Armed Forces set aside our 2007 decision and returned the record to the Judge Advocate General (JAG) to order a hearing under United States v. DuBay, 37 C.M.R. 411, 1967 WL 4276 (C.M.A.1967), focusing on nine “factual issues ... in need of resolution,” with further instructions to our court to conduct an Article 66, UCMJ, 10 U.S.C. § 866 review after receiving the Du-Bay record. United States v. Lee, 66 M.J. 387, 390 (C.A.A.F.2008).1 The CA duly ordered a DuBay proceeding to address the concerns articulated.
When we received the record of the Du-Bay proceeding, we determined that the judge who presided over the hearing was disqualified from doing so, and we therefore returned the record to the JAG for remand to an appropriate CA who then ordered another hearing.2 United States v. Lee, No. 200600543, 2009 WL 3747173, 2009 CCA LEXIS 385, unpublished op. (N.M.Ct.Crim.App. 10 Nov. 2009). We received the results of the second hearing along with additional pleadings from the parties and briefs from amici curiae. After hearing argument on the matter, we determined that several substantial questions remained unanswered, and accordingly we returned the record so that the CA could reopen the DuBay hearing so that we might receive, among other things, evidence, not a mere affidavit untested by the adversarial process, from the appellant regarding the questions from the Court of Appeals for the Armed Forces.3 We now have the results of that hearing along with further pleadings from the parties.
The testimony at the DuBay hearings establishes that Capt Reh’s transfer to prosecution duties was disturbingly not uncommon, as it allowed defense counsel who were leaving the service to wind down their cases, and it also allowed those defense counsel who wished to serve a tour in Iraq or Afghanistan to wind down their cases and to gain some prosecutorial perspective before reporting to the theater of operations. It appears in Capt Reh’s case that he was initially reassigned to prosecution duties simply to ease his release from active duty, but when an opportunity to deploy to Afghanistan arose due to another officer’s unavailability, he was able to extend his active-duty time and take advantage of the deployment opportunity. By the time the record of trial and the staff judge advocate’s recommendation were available for defense review, Capt Reh had deployed to Afghanistan and a substitute defense counsel assembled a clemency request and the response to the staff judge advocate’s recommendation.
As the appellant correctly points out in his brief, the passage of time has affected witnesses’ ability to remember critical facts, and additionally critical case files have been lost. We appreciate the DuBay judge’s difficult task of assessing credibility and, in that regard, we note the consistency of the appellant’s position, first expressed in Appellate Exhibit XXX and repeated in his DuBay testimony, as it stands in distinction to the *537somewhat shifting narratives provided by other witnesses in this case. It is a disturbing day in military justice, and it weighs heavily upon our assessment of the proper way forward in this case, when we, like the DuBay judge, determine that the testimony of a convicted felon is more credible than that of the two sworn officers of the court who testified.
Factual Background to this Appeal
The appellant was represented at trial by a uniformed defense counsel, Capt Reh, and a civilian defense counsel, Mr. Sheldon. Both counsel had been involved with his case since the pretrial investigation stage, and charges were referred to a general court-martial in September 2004. Some time shortly after the charges were referred for trial, Capt Reh learned that he would be transferred to duties in the prosecution office at Camp Le-jeune.
Capt Reh formally changed positions at the end of the year, carrying over several special and general courts-martial as a defense counsel. The appellant’s trial on the merits concluded in early May 2005, that late trial date dictated in part by defense requests for continuances to conduct further investigation. Capt Reh at that point carried a respectable load of special and general courts-martial in his new capacity as trial counsel. Capt Reh’s new supervisor was Major (Maj) Keane, the Military Justice Officer at Camp Lejeune. Maj Keane served as the prosecutor in the appellant’s trial and was responsible for preparing a fitness report on Capt Reh for the period that included the appellant’s trial, although he refrained from substantive comment on Capt Reh’s defense work.
While Capt Reh and Mr. Sheldon testified that they did, at some point before the move, inform the appellant that the prosecution duties may have posed a “potential conflict,” neither defense attorney revealed that Capt Reh would be working directly for the same officer who was prosecuting the appellant; furthermore, we cannot on the record before us conclude that Capt Reh fully revealed his professional predicament to his co-counsel, as evidenced in AE XXX. The combination of the appellant’s later testimony and his corn-plaint in AE XXX tell us that he had but an imperfect understanding of the arrangement. It was only while the appellant was serving his term of confinement, reading a book provided by an acquaintance from Camp Le-jeune about an officer prosecution going on contemporaneously with the appellant’s, that he fully understood the relations among the counsel. To illustrate even more starkly the knotty relations, Capt Reh and Maj Keane, on opposite sides of the courtroom in the appellant’s case, teamed up as trial counsel in that other case, while the other Marine officer was represented by Maj Stackhouse, the Senior Defense Counsel who concurrently provided narrative portions for Capt Reh’s fitness report for his performance on his remaining defense cases. Maj Stackhouse, coincidentally, was the substitute defense counsel for post-trial matters for the appellant.
Issues Specified by the Court of Appeals for the Armed Forces
In his findings of fact and conclusions of law, the DuBay judge addressed the nine questions posed by the Court of Appeals for the Armed Forces. Regarding questions 1 and 2, which deal generally with the “command climate” at the installation, we are concerned that the arrangements depended to a greater or lesser extent on the cordial professional and personal relations within the legal community at the installation where this all occurred and the belief that the uniformed defense counsel would not “pull his punches” in his representation of the appellant. We do not see cordiality or professionalism as in any way a detriment to fair trials and due process, but we are concerned with the idea that the interests- of justice are subject to the variable of interpersonal relations.
It might be unfair to say that the supervisors were complicit in the problem, but the record does indicate that Capt Reh was obviously occupying space in the prosecution wing of the law center and that the senior defense counsel would occasionally have to track him down to meet with the appellant.
We note, with some disbelief and disapproval, that none of the counsel involved *538brought this potential problem to the attention of the military judge, e.g., Record at 1164, the independent person in the best position to gather facts and to make rulings in the interest of justice. See generally United States v. Breese, 11 M.J. 17, 20 (C.M.A.1981). The record reveals a troubling lack of transparency before the tribunal and an even more troubling failure to recognize the conflicts, real or perceived, occasioned by Capt Reh’s assignment as a direct subordinate of the prosecutor in this case. Indeed, the record reveals that while the supervisory attorneys, both those performing defense functions and those performing trial functions, realized that there might be a conflict of interests, not a single supervisory attorney recommended to either the defense counsel or the trial counsel that the military judge be informed.
Finally, we adopt without qualification the DuBay judge’s finding that the appellant has not identified any effects on the representation resulting from the claimed conflicts of interests. Our difficulty in accepting this finding is that, while the record supports it, the length of time that has passed, the inconsistent accounts from hearing to hearing, and the reluctance of the various participants to lay bare the facts, cause us to wonder whether the whole truth can ever be known. Nothing in any of the testimony offered before the DuBay judge suggests any obvious deficiency of performance. See Strickland v. Washington, 466 U.S. 668, 687, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Echoing our misgivings expressed above, we may never know what questions the defense did not ask, what tangents they did not consider, due to the professional conflict.
The appellant has now added to his pleadings a claim that the uniformed counsel was ineffective because he did not prevail on witness or continuance requests, but he leaves to speculation what could or should have been done differently. He does not point to any prejudice with respect to the outcome of the trial, as his civilian counsel returned to Ireland (albeit at the appellant’s expense) to conduct further investigations. Because we are concerned with a systemic problem, however, we note that not all persons accused at court-martial have the resources that the appellant apparently had; it well could have been the case that the travel funding requests were critical to a fair trial to another, less financially capable, appellant.
Ineffective Assistance, Structure, Process
How we resolve the issue posed by the Court of Appeals depends, in part, on how we frame the question presented to us. By choosing the framework we essentially choose the outcome. In performing our analysis, we take to heart the pronouncement that the Constitutional and statutory rights to counsel are accorded “not for [their] own sake, but because of the effect [they have] on the ability of the accused to receive a fair trial.” United States v. Cronic, 466 U.S. 648, 658, 104 S.Ct. 2039, 80 L.Ed.2d 657 (1984). Cf. United States v. Wiesen, 56 M.J. 172, 176-77 (C.A.A.F.2001)(citing need for fairness and public perception of fairness of military justice system).
We must consider, as the Court of Appeals for the Armed Forces directed with its final “factual issue in need of resolution,” the effect on the litigation at the trial court. Although not perhaps in so many words, the Court of Appeals is requiring an analysis consistent with ineffective assistance of counsel jurisprudence, the typical framework for conflict cases, as noted by the minority opinion in Lee. 66 M.J. at 390 (Ryan, J., dissenting), citing Cuyler v. Sullivan, 446 U.S. 335, 350, 100 S.Ct. 1708, 64 L.Ed.2d 333 (1980).
If we view this matter as one of ineffective assistance of counsel, then we are bound to conclude that the appellant has not met his burden and we must affirm the findings and sentence. In performing an ineffective assistance of counsel analysis, courts are guided by Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). An appellant (or habeas corpus petitioner) who attacks the result of a criminal proceeding is required to show “that counsel made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel’s errors were so serious as to deprive the defendant of a fair trial, a *539trial whose result is reliable.” Id. at 687, 104 S.Ct. 2052.
Ethical breaches can, but frequently do not, form a basis for concluding that the counsel made the serious error required by Strickland. See Nix v. Whiteside, 475 U.S. 157, 165, 106 S.Ct. 988, 89 L.Ed.2d 123 (1986). Even if Capt Reh did breach his duty of loyalty to Capt Lee, a decision which we need not reach in this case, the DuBay hearings support a conclusion that the appellant suffered no prejudice due to any possible ineffective assistance.
We could also view this as a matter of waiver of conflict-free counsel and conclude, now that we have evidence from the appellant as well as his two counsel, that he did not knowingly and voluntarily waive such counsel. Reported cases in this area, some of which pre-date Strickland, include instances when one attorney undertakes representation of two or more persons who are accused of a joint offense and instances when an attorney represents one client who has imparted confidential information to the attorney that may be necessary during the course of representing another client. Compare Sullivan with Holloway v. Arkansas, 435 U.S. 475, 98 S.Ct. 1173, 55 L.Ed.2d 426 (1978). Those cases are inapposite given the facts of this case, but they do inform the analytical approach.
In Sullivan, the respondent was one of three co-actors accused of a murder. The three were represented in separate trials by the same two attorneys, and it was clear from the statement of facts in the ease that Sullivan knew full-well of the multiple representation and potential conflict, yet he did not object at trial. Sullivan, 446 U.S. at 337-38, 100 S.Ct. 1708. As the Court noted in its opinion, nothing in the circumstances in the case indicated that the trial court knew or should have known of a potential conflict, as none of the participants brought the matter to the attention of the trial judge. Id. at 348, 100 S.Ct. 1708. The court held that a “potential conflict of interest” was insufficient to warrant relief, but instead that “[i]n order to demonstrate a violation of his Sixth Amendment rights, a defendant must establish that an actual conflict of interest adversely affected his lawyer’s performance.” Id. at 350, 100 S.Ct. 1708.
Holloway, in contrast to Sullivan, involved a timely objection to multiple representation because of divergent interests of the co-defendants. In that case, the Court created an “automatic reversal” rule when an attorney was forced, over objection, to represent conflicting interests. Holloway, 435 U.S. at 489-90, 98 S.Ct. 1173. In sharp distinction to our case, Holloway’s counsel brought the conflict before the court for resolution.
The court’s latest decision in this area is in Mickens v. Taylor, 535 U.S. 162, 122 S.Ct. 1237, 152 L.Ed.2d 291 (2002), another habeas corpus case. Mickens was represented by appointed counsel who had previously represented a person whom Mickens was accused of killing. The court held that even though the state trial judge knew, or should have known, of a potential conflict, it would not craft an “automatic reversal” rule if the trial court failed to inquire into, and obtain, a waiver of conflict-free counsel. Id. at 174, 122 S.Ct. 1237. The court reiterated that it was still necessary in these cases to demonstrate an adverse effect on the representation. Id.
The “conflicted counsel” cases seem to share a requirement with the “ineffective counsel” cases that the client demonstrate some effect on the performance of the attorney or some effect on the outcome of the proceedings. Our review of the record and of the DuBay proceedings, in light of these approaches, would compel us, under Article 59, UCMJ, 10 U.S.C. § 859 to affirm the findings and the sentence.
The appellant and the legal ethicists serving as amici curiae would have us declare that the arrangement here constitutes a structural error, that is, an error that strikes so close to the heart of due process that no assessment of prejudice is necessary. See, e.g., United States v. Brooks, 66 M.J. 221, 224 (C.A.A.F.2008). They argue that the appellant was denied his “counsel of choice,” urging that once Capt Reh became conflicted he was no longer the “chosen” counsel. Leaving for another day whether “counsel of choice” jurisprudence even applies in the context of a *540detailed uniformed counsel, we simply cannot accept the expansion of the term to the extent that the appellant would have us do in his case.
We also must distinguish this case from the decision this term in United States v. Hutchins, 69 M.J. 282 (C.A.A.F.2011). The appellant in that case was represented by a team of four attorneys, three uniformed, one retained. One of the uniformed attorneys requested to resign from active duty; the resignation took effect shortly before trial. The military judge in the case did not formally relieve the attorney from his duties to the appellant, nor did he obtain the appellant’s consent to sever the relationship; instead, the military judge viewed the severance as a “done deal” to be acquiesced in by the appellant. We presumed prejudice due to the complementary role that each attorney played in the case. The Court of Appeals reversed, holding that the continuous presence of retained counsel preserved the counsel “structure” of the 6th Amendment such that we could test for prejudice. Hutchins, 69 M.J. at 292 (C.A.A.F.2011)(erroneous severance of one of several counsel on defense team is not “structural error” in the circumstances of that particular case but is instead testable for prejudice).
In this case, however, where the retained counsel was misinformed by Capt Reh of the nature of the conflict and therefore did not take the proper prophylactic action, we are not comfortable “assessing the defense team as a whole.” Cf. United States v. Boone, 42 M.J. 308, 313 (C.A.A.F.1995). On this basis, therefore, we are closer to structural error than the Court of Appeals ruled we were in our Hutchins decision, but consistent with that court’s holding in Hutchins and the cases it cites in reaching that holding, we still do not conclude a structural error infected this appellant’s case.
We reject alternative analytical approaches in favor of one which focuses on the chain-of-command relationship between the uniformed defense counsel, Capt Reh, and the prosecutor in this case, Maj Keane, and the failure of any attorney, either within the defense technical chain or the prosecution technical chain, to bring the matter to the attention of the military judge for discussion on the record. Taking that approach, which is similar to that adopted in United States v. Whidbee, 28 M.J. 823 (C.G.C.M.R.1989), leads us to our conclusion to afford relief to the appellant under Article 66, UCMJ.
We note initially that Capt Reh was not disqualified to act as the appellant’s counsel in this case. See RULE FOR COURTS-MARTIAL 502(d)(4), MANUAL FOR COURTS-MARTIAL, UNITED STATES (2008 ed.) (this provision is identical to that which was in effect at the time of the appellant’s trial). Such a disqualification would have created the structural error that the appellant urges existed at his trial, notwithstanding the presence of a retained attorney, and would lead to reversal.
Military jurisprudence has for some time declined to consider a command relationship between opposing counsel to be prejudicial per se. See United States v. Hubbard, 43 C.M.R. 322, 325, 1971 WL 12786 (C.M.A.1971); see also United States v. Nicholson, 15 M.J. 436, 439 (C.M.A.1983). It is only by distinguishing those cases and Whidbee that we reach the conclusion that the findings and sentence must be set aside.
In Nicholson, the defense team moved to disqualify the assistant trial counsel in that case because the latter was responsible for drafting fitness reports on the detailed defense counsel. After receiving the assurance of the defense counsel that he did not feel in the least intimidated by the prospect of litigating against his supervisor, the military judge questioned the Sailor himself about the conflict and was satisfied that the Sailor voluntarily and intelligently waived the conflict. 15 M.J. at 437-38.
Nicholson cited the earlier decision of Hubbard as regards the command relationship. Hubbard was apparently unaware that his detailed counsel was subordinate to the trial counsel, as there was no motion to disqualify and no colloquy on the record. The Court of Military Appeals concluded, however, that prejudice per se was not the proper approach to the case and, because “no basis exists for questioning the adequacy of de*541fense counsel’s performance,” 43 C.M.R. at 324-25, affirmed.
The facts in Whidbee are closer to those before us, but even that case must be distinguished. The command relationship — that the trial counsel was the rating officer for the detailed defense counsel — was disclosed, albeit perhaps incompletely, to the client. 28 M.J. at 827. Tellingly, neither counsel brought the information before a military judge for a full explication of the right to conflict-free counsel and a possible election by the accused of a different counsel. Id. at 828. The court held in that case that the Government bore a heavy burden of proof that the accused had voluntarily waived conflict-free counsel and that the Government had failed to meet that burden, thus resulting in the findings and sentence being set aside. Id. at 830.
What distinguishes Hubbard, Nicholson, and Whidbee from the appellant’s case is that in those other cases, the command relationship existed before the offenses occurred and remained static during the course of the proceedings. One clear impact of those existing relationships is that the military judges in Hubbard, Nicholson, and Whidbee would know, or should know, of the potential conflict, and would be prepared to conduct the necessary inquiry. See Mickens, 535 U.S. at 173, 122 S.Ct. 1237. The same cannot be said of the arrangement in the appellant’s case, as Capt Reh’s duties and loyalties shifted mid-trial.
A more apt analogy, but still distinguishable on a critical issue, is Garcia v. Bunnell, 33 F.3d 1193 (9th Cir.1994). The approach in Garcia is particularly instructive, as in that case Mr. Garcia’s attorney at trial had just accepted a position with the very prosecutor’s office that was prosecuting Mr. Garcia. The colloquy at the trial level made it clear that Mr. Garcia understood his right to conflict-free counsel and waived it. Id. at 1195-97. We reiterate that we do not have that same sort of colloquy here, and we join the DuBay judge in concluding that the appellant did not make an intelligent decision to waive conflict-free counsel; additionally, because of the incompleteness of the disclosure and the failure of counsel for either party to bring the matter to the attention of the military judge, the holding in Garcia does not influence our decision.
At oral argument, the appellant’s counsel remarked upon the apparent comfort, perhaps through habituation, of the supervisory attorneys at the law center with the process of transferring counsel from defense to prosecution. To point out the ethical perils of this practice, we extend the rationale of Whidbee and hold that, when a defense counsel is assigned duties that place him in the rating chain of the trial counsel, defense counsel must advise the client and any co-counsel of the potential conflicting interests and then arrange for the client to be advised by a disinterested party as to the necessity for a waiver. Defense counsel must notify the military judge of the potential conflict; failing that, trial counsel, as an officer of the court, must do so. Only when the military judge is satisfied that the client understands the right to conflict-free counsel and waives any disability may the trial progress; however, in the interests of justice, the military judge may consider other remedies such as disqualification of the trial counsel from further participation, or alteration of the rating chain of the defense counsel. See United States v. Biagase, 50 M.J. 143, 148 (C.A.A.F.1999). See also United States v. Gore, 60 M.J. 178, 186 (C.A.A.F.2004) (judge’s powers generally).
Conclusion
The appellant has been unable to point to any effect on his trial arising from any ineffectiveness on the part of his defense team. Both the DuBay judge and we are unable to find any prejudicial impact, and Article 59 would normally require affirmation of our earlier decision. See United States v. Tardif, 57 M.J. 219, 223 (C.A.A.F.2002). On the other hand, and recognizing that it is not our statutory duty to enforce the Canons of Legal Ethics, it is occasionally necessary to “apply needed prophylaxis” in cases, such as the one before us, where application of the ineffective assistance of counsel test is inadequate to assure vindication of an accused service member’s rights to counsel and to military due process. See generally Mick*542ens, 535 U.S. at 176, 122 S.Ct. 1237. Accord Tardif, 57 M.J. at 224 (service Courts of Criminal Appeals have authority, under Article 66, to determine what findings and sentence should be approved, even if legal error is not prejudicial, if they deem relief is appropriate under Article 66) (citing United States v. Collazo, 53 M.J. 721, 727 (Army Ct.Crim.App.2000)).
As the Court of Appeals for the Armed Forces directed in this case, see Lee, 66 M.J. at 390, we have performed a fresh review of this case under Article 66. Under Article 66, we “may affirm only such findings of guilty and the sentence ... as [we find] correct in law and fact and determine[ ], on the basis of the entire record, should be approved.” 10 U.S.C. § 866(c)(emphasis added). This duty has been described as an “ability to protect an accused,” see United States v. Parker, 36 M.J. 269, 271 (C.M.A.1993), and it has been said that “[a] clearer carte blanche to do justice would be difficult to express.” United States v. Claxton, 32 M.J. 159, 162 (C.M.A.1991).
We are bound to conclude that the system of identifying and resolving professional conflicts of interests failed the appellant in this case. We find that his guilty pleas were improvident as they were made without a clear understanding of the burden under which his counselors labored when advising him to enter these pleas as part of the overall trial strategy.
We find that the appellant further did not knowingly or intelligently waive conflict-free counsel for the trial on the merits. We are unable to conclude, however, for reasons discussed above, whether this legal error materially prejudiced the appellant’s substantial rights. Ordinarily, then, we would be compelled to affirm the findings on the contested charges under Article 59.
As we noted previously, however, the system failed this appellant. His counsel were laboring under a professional disability that he did not fully understand; it may be that his counsel themselves did not fully understand the disability. With the passage of time, moreover, Capt Reh became more fully ensconced in the prosecution function as his remaining defense cases were resolved. The appellant did not benefit from the sober and detached perspective of the military judge whom our system empowers to hear and resolve professional conflicts-as all counsel connected with this litigation failed to inform the trial judge of the inherent conflict in this case. The result of a trial in which the uniformed defense counsel is simultaneously working as a prosecutor under the officer who prosecuted the appellant might not appear to the general public to be “fair”.
Employing, therefore, our “awesome, plenary, de novo power of review,” United States v. Cole, 31 M.J. 270, 272 (C.M.A.1990), it is our judgment, based on the entire record, that the findings and the sentence in this court-martial should not be approved. The findings and the sentence are set aside. A rehearing is authorized on all charges and specifications except for Charge II and its sole specification (unbecoming conduct between 9 and 12 January 2004) and Specification 8 of Charge III (indecent assault).
Senior Judge MAKSYM and Judge PERLAK concur.. Summarized, the questions are (1) the circumstances, including start date, of the new detailing; (2) consideration given to counsel’s active defense cases; (3) "need” for the reassignment; (4) detailed counsel's actions as a trial counsel during representation of the appellant; (5) any supervisory relationship between detailed defense counsel and any prosecutor; (6) nature of disclosures to the appellant; (7) the appellant’s understanding of disclosures; (8) civilian counsel’s role in the matter; (9) impact on representation.
. Because we determined the first DuBay hearing to have been faulty, we will not refer to it at all. All references to "DuBay ” hearings in this opinion will refer to the second and third hearings collectively, both of which were presided over by the same military judge, namely, the Circuit Military Judge for the Southern Judicial Circuit.
.Our order specifically noted that the appellant had made a colorable claim of ineffective assistance of counsel (through breach of the duty of loyalty) and that this claim effectively waived the attorney-client privilege under MILITARY RULE OF EVIDENCE 502, MANUAL FOR COURTS-MARTIAL, UNITED STATES (2008 ed.). See, e.g., United States v. Dupas, 14 M.J. 28, 30 (C.M.A.1982). Our order neither directed nor contemplated receipt of any incriminating testimony from the appellant.