OPINION OF THE COURT
WOLD and RABY, Senior Judges:On 9 June 1983, appellant was tried by a military judge sitting alone as a general court-martial. He was convicted, in accordance with his pleas, of one specification alleging wrongful possession of hashish and of two specifications alleging wrongful distribution of hashish, in violation of Article 134, Uniform Code of Military Justice [hereinafter cited as UCMJ], 10 U.S.C. § 934 (1982). He was sentenced to a dishonorable discharge, confinement at hard labor for sixteen months, forfeiture of all pay and allowances, and reduction to Private E-l. The convening authority approved the sentence.
Appellant alleges for the first time before us that he suffered illegal pretrial confinement and punishment in violation of Article 13, UCMJ, 10 U.S.C. § 813, and that the actions of his brigade-level commander, Colonel Leslie L. Beavers, had an impermissible chilling effect which denied appellant a “fair forum” for the disposition of his case. We will discuss the effect of Colonel Beavers’ actions on the referral process, on witnesses, and on appellant’s pleas, as well as possible violations of Articles 13 and 55, UCMJ, 10 U.S.C. § 855.
I. BACKGROUND1
During the early months of 1983, an investigation by agents of the Criminal Investigation Command (CID) revealed large-scale drug abuse and distribution within Pinder Barracks, Federal Republic of Germany, the home of the Division Artillery (DIVARTY), 1st Armored Division. One particular DIVARTY unit, the 6th Battalion, 14th Field Artillery (6/14th FA), had a major drug problem. Urine testing had revealed that about one-fourth of that battalion’s personnel used illicit drugs. On 24 March 1983, agents of the CID met with Colonel Beavers, his servicing judge advocate, and other members of his staff to obtain authorization to apprehend a large number of DIVARTY soldiers. Colonel Beavers decided on a mass apprehension at a unit formation. His decision was based in part on advice from his servicing judge advocate that such an event would be legal if properly conducted.
On 25 March 1983, as planned, approximately 1200 troops attended a post-wide formation at Pinder Barracks. There were *876a total of thirteen company-sized units present, including a military police company, a maintenance company, the Headquarters and Headquarters Battery of DIVARTY, and the ten batteries of the 6/14th FA and the 1st Battalion, 22nd Field Artillery.
Colonel Beavers, who was not only the DIVARTY commander but also the Pinder Barracks installation commander and a special court-martial convening authority, addressed the formation. He spoke generally regarding leadership, discipline, command climate, and the need for readiness. He stated that drug abuse and drug trafficking had an adverse impact on the readiness of the command and consequently would not be tolerated in DIVARTY. He announced that some of the soldiers present at the formation met the standards of neither the Army nor Pinder Barracks and should be removed from their units. Some individuals claim that Colonel Beavers specifically called these soldiers “criminals”, while others claim that he either used that term in a general sense or stated that criminals would not be tolerated in-DIVARTY. Some individuals claim that Colonel Beavers called them “bastards” during the mass apprehension, although the statements are in conflict on this point. For the purposes of this decision, we will assume that Colonel Beavers used both terms.
While Colonel Beavers was speaking, military police, CID agents, and German police entered through the gate and surrounded the parade field. As Colonel Beavers read the names of forty individuals, starting with the most junior, those individuals were told to report to him at the front of the formation. Enlisted members were escorted to the front by their battery commanders and first sergeants. Noncommissioned officers were escorted by their battalion commanders and command sergeants major. The one officer involved was escorted by his battalion commander. The unit crests were removed from the uniforms of a number of these individuals, including appellant, before they reported.2 When the soldiers reported to Colonel Beavers and saluted, their salutes were not returned. (It is an Army custom not to salute prisoners. See, e.g., Army Regulation 190-47, Military Police: The United States Army Correctional System, para. 4-8 (1 Oct. 1978).)
The soldiers were formed to the side of the platform on which Colonel Beavers stood. They were then turned over to the CID agents and military police. Within the view of the remaining troops in the formation, the soldiers were required to spread their hands and feet and lean against a building while they were searched. They were then handcuffed, marched to a waiting bus, and transferred to the district headquarters of the CID for processing and questioning.
The German police and the military police, using drug detection dogs, then assisted in health and welfare inspections conducted through the ranks of the formation, inside the barracks, in other facilities, and in parking areas where some privately owned vehicles were searched. These inspections revealed additional drugs.
As the forty individuals who had been apprehended completed their interviews at the CID office, they were returned to their units. Thirty-five of the forty were members of the 6/14th FA. (Appellant was a member of Battery A, 6/14th FA.) The battalion commander of the 6/14th FA arranged to billet those who lived on post in a bay area on the top floor of the battalion headquarters. Those who lived off post were allowed to return home. The individuals billeted in the bay area continued to live there until charges were preferred. Each person was provided with a bed and a wall locker. Although there was no direct heat in the bay area, it was heated by convection from the floors below and the occupants were permitted to use privately supplied heaters. In contrast, other bar*877racks had no heat at all because of heating plant renovations.
Once charges were preferred, each individual was told he could move back to his battery if he wished. The battalion commander expressly informed the soldiers that they had not been placed in any form of restraint but that they could, if they desired, stay in the bay area for their own protection. Approximately twenty-seven of the soldiers (including appellant) elected to stay at the battalion headquarters until their cases were disposed of. These soldiers were formed into a platoon which held formations, performed post support activities, prepared for an impending annual general inspection, and participated in training, supply, and maintenance functions. The platoon was called the “Peyote Platoon.” Although it is unclear who first coined the phrase, it may have been the battalion’s command sergeant major.3 There is also a dispute as to whether the so-called “Peyote Platoon” was forced to march to the cadence of “peyote, peyote, peyote.” Again, for the purposes of this decision, we will assume that it was.
Of the 35 soldiers apprehended from the 6/14th FA, 29 were charged with one or more incidents of wrongful distribution of hashish, and the remaining six were charged with wrongful use of hashish. Fifteen of the 35 (including appellant) were referred to general courts-martial by the Commander, 1st Armored Division. Eleven of these were convicted on pleas of guilty, three were convicted after contesting the charges, and one was administratively discharged in lieu of trial under the provisions of Army Regulation 635-200, Personnel Separations: Enlisted Personnel, Chapter 10 (107, 4 June 1982). The division commander also referred four members of appellant’s battalion to special courts-martial empowered to impose bad-conduct discharges. One of these accused pled guilty and three received administrative discharges in lieu of trial. Additionally, nine soldiers were convicted and two were acquitted at summary courts-martial. The remaining five soldiers received nonjudicial punishment. None of the 35 were referred to trial by Colonel Beavers. Appellant’s battery commander preferred and forwarded the charges against appellant, but made no recommendation as to the appropriate level of court-martial. (The commander of the 6/14th FA has stated that he had reserved the disposition of all drug cases to himself. This is currently authorized by Rules for Courts-Martial [hereinafter cited as RCM] 306(a) and 601(f). See also, Manual for Courts-Martial, United States, 1969 (Revised edition) [hereinafter cited as MCM, 1969], paragraphs 32¿ and 32e.) Appellant’s battalion commander and Colonel Beavers both recommended trial by general court-martial.4
In the case at bar, we have been offered evidence of only one commissioned officer and one noncommissioned officer, among the many from whom sworn statements have been taken, who say that they understood from these events that they were being directed not to testify or to dispose of a case in a certain way.
Sergeant First Class (Retired) Locke C. Brayboy stated that after the mass appre*878hension his battery commander, Captain Brian E. Burks, held a meeting of his entire battery, at which Captain Burks told those present “not to associate” with the apprehended soldiers “until after their court-martial or until their cases had been disposed of in some manner.” Because of the “tone” of this meeting, Sergeant Bray-boy concluded that “it would be in our best interests not to testify for these individuals.” Sergeant Brayboy was not, however, aware of anyone who had suffered for testifying for an accused, nor had he heard of Colonel Beavers or any other officer becoming upset because anyone had testified.
Captain Burks, the commander of Service Battery, 6/14th FA, stated that following the mass apprehension he believed that the “judgment of what level of administrative or judicial action” to use “had been taken out of [his] hands.” Captain Burks subsequently stated that, “[A]t no time did I feel pressure from my superiors affected ... my judgment in recommending a specific level of disposition of the case.”
As noted above, appellant was a member of Battery A, 6/14th FA. Neither Captain Burks nor Sergeant Brayboy played any part in the disposition of appellant’s case. The evidence which has been offered to this court does not indicate that either of these two men or any other member of Captain Burks’ battery was in possession of any information which would have qualified him to testify for appellant.
All of the following have sworn that they are aware of no attempts by anyone to discourage testimony favorable to those who had been apprehended: Colonel Beavers and the DIVARTY command sergeant major; appellant’s battalion commander, battalion executive officer, and battalion command sergeant major; and appellant’s battery commander and acting first sergeant. The DIVARTY command sergeant major and appellant’s battery commander and acting first sergeant all testified on behalf of other soldiers apprehended on 25 March 1983.
We have judicially noted the following information from the records of trial in the fifteen cases which resulted in punitive discharges and which are consequently before this court for review. First, in a majority of those cases, commissioned or noncommissioned officers (or both) from DIVARTY testified for the accused. Second, some 16 commissioned and noncommissioned officers (including two battery commanders and two first sergeants) testified for the accused in one or more of these eases.5
At trial, appellant did not indicate that he had had any difficulty obtaining witnesses,6 but offered no favorable character testimony. The record of trial reflects that appellant’s battery commander had evaluated appellant as follows: “His performance has been marginal, as he has displayed a marked lack of motivation and leadership ability.” Also, appellant had failed his last Skill Qualification Test prior to trial. (This is one of the primary indicia of professional competence for enlisted soldiers.) As a general rule in trials by court-martial, submission of favorable character evidence by the accused “opens the door” to rebuttal with unfavorable character evidence, which is otherwise inadmissible.
II. THE NATURE AND SCOPE OF JUDICIAL CONCERN
A. Unlawful Command Influence
The legislative history of the UCMJ clearly demonstrates Congress’ historical *879commitment to the creation of a military justice system which is free of unlawful command influence. During hearings convened to consider enactment of the UCMJ, witnesses testified regarding the subtle nature of unlawful command influence. For example:
As to command influence: I would certainly like to second Mr. Spiegelberg’s statement that there is absolutely no way of proving an officer guilty of article 377 unless he is a hopeless idiot. In the first place, you cannot prevent an officer from discussing a case at the dinner table ... or even at a staff meeting—
Now ... what happens? The latrine rumor keeps moving down the line. The man who was at the table talked to his friend. I have never seen anything spread faster than what the commanding officers [sic] wants in a military unit____ So that the officers appointed to the court, if you have that type of thing going on, realize that the Old Man wants an example made.
I have seen that same thing carried out simply by the device of the commanding general’s putting an officer in arrest by a special order____ He would like to emphasize the fact that he considers this a very serious offense.
Hearings on S. 857 and H.R. 4080 Before a Subcomm. of the Comm, on Armed Services, 81st Cong., 1st Sess. 90 (1949) [hereinafter cited as Hearings] (statement of Arthur E. Farmer, Chairman, Committee on Military Law, War Veterans Bar Association). See also Hearings on H.R. 2498 Before a Subcomm. of the Comm, on Armed Services, 81st Cong., 1st Sess. 652 (1949).
In an attempt to prevent unlawful command influence, Congress enacted Article 37, UCMJ, and provided for the punishment of violations of this article under Article 98, UCMJ, 10 U.S.C. § 898. Congress also relied heavily on the UCMJ’s appellate court system for protection from unlawful command influence, as indicated by this exchange between Senator Leverett Salton-stall and Mr. E.M. Morgan, Professor of Law, Harvard University:
Senator Saltonstall: Mr. Chairman, may I say this? ... We have provided this very high court on the law; we have provided a board of review of the facts— Professor Morgan: That is right.
Senator Saltonstall: And if we have done those things, is not the accused amply protected from any influence?
Professor Morgan: That is exactly the way our committee felt about it____
Hearings at 307. This reliance is also reflected in the following statement by Professor Morgan:
On the question of restriction of command control, we felt that when the board of review [predecessor of the Court of Military Review] in the Office of the Judge Advocate General, which is so far removed from any control of the convening authority, had power to handle law, fact and sentence, that that eliminated a great part of the evils of command control.
Hearings at 45.
In the discharge of this responsibility, the United States Court of Military Appeals and the several Courts of Military Review have recognized that unlawful command influence poses a special threat to military due process of law by introducing into judicial proceedings powerful external factors which can skew results. “[A]n allegation of unlawful command influence is a serious matter,” United States v. Alexander, 19 M.J. 614, 616 (A.C.M.R.1984), for which we have shown “special solicitude,” United States v. Blaylock, 15 M.J. 190,193 (C.M.A.1983). Recent experience has demonstrated that the problems which motivated Congress over thirty years ago still demand close attention and concern. See, e.g., United States v. Treakle, 18 M.J. 646 *880(A.C.M.R.1984) (en banc), pet. granted, 20 M.J. 131 (C.M.A.1985).
B. Actuality and Appearance
1. Origins
A concern with the appearance as well as the actuality of unlawful command influence was brought to the attention of Congress when the UCMJ was first enacted:
[A]re you going to stop with the theory that a fair trial is all that is required or are you going to go further and say that it is necessary to the welfare of the armed services that their personnel believe that they are getting a fair trial as a help to the maintenance of morale.
It seems to me that, first, you must insure a fair trial, and second, you must maintain a belief in a fair trial if you are to have a fighting army, and a fighting army and the ability to win wars is the thing upon which command has based its argument that it must control the courts.
Hearings at 87 (statement of Mr. Farmer). See also Hearings at 82 (statement of Senator Wayne Morse that the military was “lucky that public resentment [had] not reached higher proportions ... in regard to military justice.”).
In the Court of Military Appeals, a similar concern with the appearance as well as the actuality of unlawful command influence surfaced in United States v. Navarre, 17 C.M.R. 32 (C.M.A.1954), which was decided during the first wave of command influence cases considered by the court. The majority found no unlawful command influence from a commander’s pretrial “instructions” to an audience which included some of the members of Navarre’s court-martial. Judge Paul W. Brosman, dissenting, would have directed a rehearing based on his finding that actual unlawful command influence had affected the trial. (“[The] ‘instructions’ distinctly did impinge on the court-martial’s function____”) Nevertheless, in his dissent Judge Brosman laid the cornerstone of the appearance doctrine with the following:
We are concerned here with much more, I believe, than the protection of an accused person named Navarre. In United States v. Walters [16 C.M.R. 191 (CMA 1954)], this Court said: “... [T]he court’s actions and deliberations must not only be untainted, but must also avoid the very appearance of impurity____ When such an unhappy appearance is present, proper judicial administration often requires reversive action____”
A judicial system operates effectively only with public confidence — and, naturally, this trust exists only if there also exists a belief that triers of fact act fairly and without undue influence.
17 C.M.R. at 43 (citations omitted).
A month later, writing for the majority in United States v. Stringer, 17 C.M.R. 122 (C.M.A.1954), Judge Brosman returned to the public confidence theme in another context. Stringer involved the interpretation of the double jeopardy provision in Article 44, UCMJ, 10 U.S.C. § 844. The first trial had been terminated after a member suggested that the court might “hang the man innocently” unless the government did a better job of presenting its case. This comment elicited the following:
[The member’s comment] — differently worded — could have injured the legal rights of no interest [as opposed to “party”].
The manner of phrasing that request was, however, so misleading as to bring into operation values transcending the parties before the court____ Such considerations are indeed akin to those underlying the doctrines of “general prejudice” and “military due process”____
17 C.M.R. at 134 (emphasis added) (citations omitted).
Appearances were next mentioned in United States v. Fowle, 22 C.M.R. 139 (C.M.A.1956). There, the trial counsel read to the members an instruction from the Secretary of the Navy promulgating a policy to discharge, inter alia, thieves. Fowle had been convicted of larceny. The Court of Military Appeals accepted the convening authority’s post-trial determination that the substantial rights of the accused had in *881fact been prejudiced. In dicta, Judge Homer Ferguson said, “[The] appearance of impartiality cannot be maintained in a trial unless the members of the court are left unencumbered from powerful external influences.” Id. at 142.
United States v. Kitchens, 31 C.M.R. 175 (C.M.A.1961), is sometimes cited as a case decided on the basis of the appearance of unlawful command influence. See, e.g., United States v. Ibbetson, 35 C.M.R. 814, 816 (A.F.B.R.1965). In fact, it is not. The court, considering a letter which was critical of courts-martial which had failed to impose discharges, implicitly found that the evidence had placed the burden of persuasion on the government and that the government had failed to sustain that burden. “We cannot answer that question [whether the accused was prejudiced] with any degree of assurance; and since we cannot, we must accord the accused the benefit of the doubt____” 31 C.M.R. at 180.8
In United States v. Johnson, 34 C.M.R. 328 (C.M.A.1964), Chief Judge Quinn, speaking for the court, said:
[T]he apparent existence of “command control” ... is as much to be condemned as its actual existence. As a matter of principle, any doubt in the matter must be resolved in favor of the accused. United States v. Kitchens, [31 C.M.R. 175 (CMA 1961) ]; see also United States v. Mamaluy, [27 C.M.R. 176 (CMA 1959)]....
The appearance, or the existence, of command influence provides a presumption of prejudice; but the presumption is rebuttable.9
34 C.M.R. at 331. The case, however, was decided on the basis of a finding of actual unlawful command influence. 34 C.M.R. at 330 n. 1; id. at 333 (“Under the circumstances, we cannot say with conviction that the court-martial was uninfluenced ... in its deliberation on the sentence.”).
In United States v. Rosser, 6 M.J. 267 (C.M.A.1979), the court emphasized the importance of considering the totality of the circumstances of a case when assessing “the appearance of fairness and freedom from command influence mandated by Congress and by our decisions____” 6 M.J. at 272.10 The court in Rosser found several discrete bases for reversing the trial judge’s denial of a mistrial; arguably, one *882of those grounds was the naked appearance of unlawful command influence. If so, Rosser was the first and remains to date the only case in which the Court of Military Appeals has based its remedy on a finding of the appearance of unlawful command influence without finding (sometimes tacitly) that actual unlawful command influence had occurred.
The most recent reference by the Court of Military Appeals to the appearance doctrine came in United States v. Grady, 15 M.J. 275 (C.M.A.1983). Grady was a mirror image of Fowle, in that it was Grady’s trial defense counsel rather than the prosecutor who put the fox among the chickens. Once again, the members were confronted during the trial with a service policy of discharging certain offenders — this time drug offenders. Once again, the court condemned the appearance of unlawful command influence (liberally quoting from Fowle), and once again the court found actual unlawful command influence and decided the case on that basis. “[T]he mention of such policy before the 'members improperly invade[d] the province of their sentencing deliberations ... and the accused was prejudiced____” 15 M.J. at 276-277.
2. Thesis
Command influence law addresses two different questions which must be considered from two different points of view. The first question is whether the accused was prejudiced by actual unlawful command influence. The second question is whether there will exist in the minds of the public the appearance that he was.
3. “The Public”
By “the public” we mean not only the civilian population, but also the rank and file of the services. Since the vast majority of servicemembers have no direct contact with the military justice system, their perspective is very similar to that of the civilian public and for purposes of this analysis may be regarded as essentially the same.
We do not understand the appearance doctrine to be limited in its application to situations which have already been publicized. We believe that the appearance doctrine was devised to insure that public confidence in the military justice system would not be undermined by the appearance that the accused was prejudiced by unlawful command influence in a given case if that case were subjected to public attention. Cf. United States v. Berry, 20 C.M.R. 325, 330 (A.B.R.1956); United States v. Thompson, 14 C.M.R. 38, 41 (A.B.R.1954).
Finally, as noted below, in determining whether a given factual situation has created an appearance of unlawful command influence, we are concerned with an opinion held by a substantial segment of the reasonable members of the public.
4. Analysis
Because the actuality and the appearance of unlawful command influence present different problems and threaten different values, certain important differences in the legal doctrines which address them must necessarily exist.
First, the issue of actual unlawful command influence must be considered from inside the military justice system: the court considers evidence which has been offered in the form of exhibits or testimony; the court subjects this'body of evidence to critical analysis and determines in detail what actually happened; the court evaluates the case in the light of its expertise in military affairs, including its perception of the integrity of the military justice system as shaped by its intimate familiarity with that system. On the other hand, the appearance that an appellant has been prejudiced by unlawful command influence must be considered from outside the military justice system: the court looks at the case through the eyes of reasonable members of the public; the court looks at the information which the public can reasonably be expected to have (primarily information supplied through the public news media and reported cases); the court does not expect the public to analyze this data with technical precision, but rather to draw broad conclusions from it; the court takes into account the unfamiliarity of the public with the military justice system and the resulting natural tendency to distrust it.
*883Second, the perception of a participant in the military justice system as to what he has been told by his commander is relevant to the existence of actual unlawful command influence. For example, if a witness mistakenly believes he has been told not to testify, and complies, then the case has to that extent actually been affected, not merely apparently affected. Thus, the perception of a participant in the system is not to be confused with the appearance of unlawful command influence. The former has actual effects within the system; the latter is an opinion in the minds of those outside the system about what has happened inside the system.11 Stated another way, actual unlawful command influence can arise from an inaccurate perception of a commander’s desires by a participant in the system;12 this may or may not give rise to an appearance of unlawful command influence, but in either event is not the same thing as the appearance of such influence.
Third, when resolving an issue of actual unlawful command influence, a court is performing its function of adjudicating the rights of the respective parties; on the other hand, when it turns to the appearance of unlawful command influence, a court is fulfilling its own responsibility to safeguard the military justice system from a loss of public confidence.13
Finally, the origins and objectives of the appearance doctrine indicate to us that the public need not be unanimous in bestowing its confidence on the system, nor need that confidence be absolute — what is required is that a substantial majority of the reasonable members of the public have confidence in the integrity of the system. A court must have a higher degree of confidence that the findings and sentence are free from prejudice when actual unlawful command influence is at issue.
5. Conclusions
Two conclusions clearly follow from these premises.
First, the appearance of unlawful command influence is not merely a lower threshold for raising the issue of actual unlawful command influence or for finding the existence of actual unlawful command influence. If it were, there would be no point in considering whether actual unlawful command influence had occurred, yet that inquiry continues to be the focal point of the cases which deal with unlawful command influence. In fact, the appearance of unlawful command influence — this opinion in the minds of reasonable members of the public which we have described above — will normally become relevant only in the absence of actual unlawful command influence. This is because the process of public*884ly determining whether the accused has been prejudiced by actual unlawful command influence, and providing a remedy if he has, will normally remove the appearance of prejudice as well.
Second, the appearance of unlawful command influence does not affect the actual fairness of a trial; the appearance of unlawful command influence and actual unlawful command influence are two different things. The law and the courts concern themselves with the appearance of unlawful command influence because of its deleterious effect on public confidence in the military justice system, not because an accused has any legitimate claim to relief where he has in fact suffered no prejudice but only appears to have. It is the interests of the military justice system itself which the appearance doctrine was designed to protect, since it is the military justice system itself which is harmed by the loss of public confidence.
We now turn to a closer examination of the doctrines of actual and apparent unlawful command influence and their application to the case at bar. We begin with the doctrine of actual unlawful command influence.
III. ACTUAL UNLAWFUL COMMAND INFLUENCE
A. The Appellate Model
1. The Presumption of Regularity
Omnia praesumuntur solemniter esse acta. This principle provides that when a decision is rendered by a court of record or a court of general jurisdiction, a rebuttable presumption exists as to the correctness of the decision and as to the regularity of the proceedings of that court. See Key v. State, 147 Ga.App. 800, 250 S.E.2d 527, 530 (1978); People v. Schomer, 64 Ill.App.3d 440, 21 Ill.Dec. 127, 131, 381 N.E.2d 62, 66 (1978); State v. Williams, 296 N.C. 693, 252 S.E.2d 739, 742 (1979); State v. Hashimoto, 46 Hawaii 183, 377 P.2d 728, 731 (1962); People v. Calvert, 82 Ill.App.3d 350, 37 Ill.Dec. 636, 638, 402 N.E.2d 638, 640 (1980); Dickerson v. State, 151 Ga.App. 429, 260 S.E.2d 535, 536 (1979). See generally 5 Am.Jur.2d Appeal and Error § 704 (1962). This presumption is not limited to civil cases but is equally applicable to criminal cases. See 5 Am. Jur.2d Appeal and Error § 867 (1962), and cases cited therein. We believe that the proceedings of general courts-martial and special courts-martial empowered to adjudge bad-conduct discharges are entitled to the same rebuttable presumption of correctness and regularity as civilian courts of record or general jurisdiction because these military courts are so constituted, and proceedings in them are so conducted, as to produce judgments equally as reliable.14 Certainly this has been true since the Military Justice Act of 1968 created the office of military judge and vested the Courts of Military Review with judicial status. The Military Justice Act of 1983, which pro*885vides an opportunity for direct Supreme Court review of those court-martial cases which have been reviewed by the Court of Military Appeals, adds further cogent grounds for this conclusion.15
2. The Burden of Persuasion
Thus, a case arrives at this court for review with a presumption of correctness and regularity. At that point, the appellant must produce sufficient evidence of an error affecting the validity of the findings or sentence in his case to shift the burden of persuasion to the government. Such evidence may come from the material already in the “entire record,” see Article 66(c), UCMJ, 10 U.S.C. § 866(c), or it may consist of admissible extra-record evidence. If the appellant fails to produce or point out such evidence, and the court itself does not find such evidence in the course of its review, the government has nothing to answer and the findings and sentence are affirmed.
What evidence is needed to shift the burden of persuasion to the government when actual unlawful command influence is alleged?
First, we believe that when actual unlawful command influence is addressed at the appellate level, it is appropriate to consider the totality of the circumstances based on the entire body of evidence before the court. See United States v. Littrice, 13 C.M.R. at 51; United States v. Calley, 46 C.M.R. 1131, 1155 (A.C.M.R.1973), affirmed, 48 C.M.R. 19 (C.M.A.1973); see also United States v. Hawthorne, 22 C.M.R. at 88.16 This is necessary in order to avoid the useless exercise of determining whether the court has before it evidence which could reasonably sustain a conclusion that the accused suffered prejudice, when the court also has before it evidence which shows that he in fact did not. At the trial level, when the judge considers only the evidence favorable to the proponent in deciding whether a matter has been placed in issue, he does so in order to avoid taking from the jury its task of deciding the ultimate issues in the case. At the appellate level it is the court itself which must decide whether the harm in question actually occurred as well as deciding whether the evidence is sufficient to shift the burden of persuasion. Since we have no occasion to divide the responsibility for decision-making, we have no occasion for dividing the evidence before us.
Second, we believe that when actual unlawful command influence is addressed for the first time on appeal, the same standard should apply as if the issue were being addressed at the trial level. When there are sufficient reasons to dispense with the normal rules of waiver for an issue like unlawful command influence and thus permit it to be raised for the first time on appeal, see, e.g., United States v. Hawthorne, 22 C.M.R. at 93, those same reasons support application of the same standard on appeal as at the trial level. At the trial level, the standard for a party who is asserting a collateral matter is evidence *886sufficient to render reasonable a conclusion in favor of the allegation asserted. Husbands v. Pennsylvania, 395 F.Supp. 1107, 1139 (E.D.Pa.1975), citing 9 Wigmore, Evidence § 2494 (Chadbourn rev. 1981). This is the same quantum of evidence required to entitle a party to have his case go to the jury. Id.17 Evidence from which only an unreasonable conclusion could be drawn is manifestly insufficient to shift the burden of persuasion. To employ such a standard would simply invite frivolous complaints.
Third, the ultimate question in a case is whether the appellant should be granted relief and, if so, what relief. The answer to this question in a. ease of actual unlawful command influence depends on whether the appellant has suffered specific prejudice. See, e.g., United States v. Johnson, 34 C.M.R. 328; United States v. Wood, 32 C.M.R. 217 (C.M.A.1962); United States v. Littrice, 13 C.M.R. at 51.18 Obviously, this means that the prohibited activity must cause some harm in the appellant’s case in order to merit relief, cf. United States v. Hudson, 20 M.J. 607, 609 (A.F.C. M.R.1985). Similarly, “generalized, unsupported claims of ‘command control’ will not suffice to create a justiciable issue.” Green v. Widdecke, 42 C.M.R. 178, 181 (C.M.A.1970). Cf. United States v. Calley, 46 C.M.R. at 1160 (“Influence in the air, so to speak, is a contradiction in terms. An object and effect upon the object must be identified for influence to exist.”).
Thus, in order to shift the burden of persuasion to the government, appellant must produce or point out evidence which, considering the totality of the circumstances, is sufficient to allow a reasonable person to conclude that actual unlawful command influence affected appellant’s case.
B. Application of the Model
1. Sufficiency of Appellant’s Evidence
Appellant has come to us without such evidence. He has merely presented evidence which shows that it is possible that his chain of command was deprived of their discretion in the disposition of his case, or that it is possible that he was deprived of favorable testimony. This is not enough to shift the burden of persuasion to the government. Mere possibilities cannot overcome the general presumption of regularity; to hold that they can would be to hold that the presumption itself does not exist, because in an infinite universe anything is merely possible.19 *887An appellant will normally be able to obtain and present the necessary evidence, if it exists. (In the event that an appellant asserts that unusual circumstances thwart his compliance, we will, of course, consider the merits of his claim on a case-by-case basis.)
2. Assistance of Counsel
Every military accused has available to him before trial, during trial, and on appeal, the assistance of competent counsel who in turn have access to all reasonable means necessary to represent him. If there were any substantial reason to question the professional competence or institutional capability of the defense bar in cases of actual unlawful command influence, the court might arguably be justified in assuming an inquisitorial function. In the absence of such legitimate necessity, a departure from our adjudicative role would sow the seeds of disaster. The genius of the adversary system lies in the unique capability of a competent defense lawyer, armed with the confidences and the cooperation of his client, to present his client’s case. No outsider, working without the advantages which flow from the attorney-client relationship, can hope to serve a defendant’s cause effectively. In addition, for this court to attempt to perform on appellant’s behalf the investigative and advocacy functions which are properly the responsibility of his counsel would set a precedent which would actually work against a strong and active defense bar by diminishing that bar’s responsibility for its clients’ welfare. We are unwilling to incur such costs unless compelled to do so by the prospect of genuine and fundamental harm to the military justice system. We can at this point envision such harm only in two circumstances: (a) a breakdown in the adversary process, or (b) the clear threat of a loss of public confidence in the military justice system. The former could be established by a showing in a particular ease that defense counsel are for some reason actually unable to effectively perform their duties on behalf of their client. As to the latter, see section IV, THE APPEARANCE OF UNLAWFUL COMMAND INFLUENCE, infra.
3. Access to Evidence
Certainly appellant in the case at bar has access to the sources of the necessary evidence, and that evidence is at least as readily available to appellant as to the government. Indeed, any appellant is in a far better position than the government to determine the identities of the potential witnesses in his case. For example, an appellant and his trial defense counsel alone know the details of his trial strategy and an appellant has far better information than anyone else about his associates and what they (and his military superiors) might be willing to state on his behalf at trial.
Thus we require of appellant nothing which is unreasonable or unfair. For example, credible evidence that a person who had some particular knowledge relevant to an accused’s case reasonably understood that Colonel Beavers or some other commander had told him not to testify would, unless rebutted by the government, trigger a presumption that the witness had complied with the commander’s order. United States v. Treakle, 18 M.J. at 657. This, coupled with a showing that the evidence in question was relevant to some material aspect of appellant’s case and that its absence caused substantial harm, would shift the burden of persuasion to the government. (An appellant might, for example, make such a showing by submitting his or his trial defense counsel’s affidavit or by referring to the trial transcript alone.) The government would then be obliged to produce (or point out already existing) “clear and positive” evidence that appellant was not in fact deprived of such testimony or *888persuade the court that no specific prejudice resulted if he was. Id.20
In many cases, the presumption that the recipient of unlawful pressure succumbed and complied will, as a practical matter, have the same effect as a presumption of specific prejudice. This comes about when the nature of the pressure and the identity of the recipient are such that if the recipient did as he was told, the accused was, by definition, denied a fair trial. A ease in which a court member is presumed to have voted in favor of a predetermined verdict or sentence because he has been told to do so is an obvious example. See United States v. Johnson, 34 C.M.R. 328; United States v. Kitchens, 31 C.M.R. 175. The same is true of cases involving actual (as opposed to potential) witnesses. United States v. Rosser, 6 M.J. 267. In other cases, such as those involving potential (rather than actual) witnesses, there is a gap between the presumption flowing from a finding that unlawful pressure influenced the recipient and a finding of any prejudicial effect in the appellant’s case. In the potential witness situation, what is needed to fill the gap is a showing that the witness would have affected the case or some decision regarding the case or would even have been called.
In the case at bar, appellant has failed to provide us with even that evidence which is most readily available to him. He has failed to produce statements from his chain of command, or from the people to whom he would normally have gone in search of favorable testimony, stating how they interpreted the events in question, if they were affected, and if so how. What is more, he has failed to reveal the identities of the people he believes were in possession of relevant information which would have qualified them to testify in his behalf.
4. The Problem of Stealth
Appellant argues that to require him to produce additional evidence is unrealistic because unlawful command influence tends to be done in secret and to be concealed by both the perpetrator and those whom he succeeds in influencing. “[UJnlawful command influence may assume many forms, may be difficult to uncover, and affects court members in unsuspecting ways.” United States v. Karlson, 16 M.J. 469, 474 (C.M.A.1983). This is a legitimate concern, and one to which we and the other courts created by the UCMJ have given the most serious consideration. The answer to this problem is not, however, to overreact by employing the same procedural responses or attaching the same consequences to a spurious, frivolous, or baseless claim as to an allegation which has genuine merit. This is the result which would follow from placing the burden on the government to prove a negative based on a predicate of merely possible prejudice. Instead, the sensible course is simply to recognize that the natural human tendency to conceal that which is forbidden is a valuable piece of circumstantial evidence which can help an appellant satisfy his burden. This factor should be given the full credit to which it is due in the particular circumstances of a given case but should not, in our view, automatically relieve appellant of his rightful burden.
In the case at bar, the acts of which appellant complains were done in public and in the presence of appellant himself. The exhibits which we have been offered do not indicate that Colonel Beavers or anyone else engaged in any surreptitious activity relevant to appellant’s contentions. The possible consequences were clearly foreseeable by appellant and his trial defense counsel. The only aspect of the situation which is subject to the danger of *889concealment just mentioned is the impact of a perceived improper directive on the recipient. This concern is fully met by the availability of the presumption, noted above, that if the recipient perceived such a directive, he complied with it.
5. Holding
Appellant has said a great deal to us about the evils of unlawful command influence. We agree, but we cannot dispose of his case merely by declaring that unlawful command influence is a bad thing; what we are dealing with is the question whether unlawful command influence caused specific prejudice in his case. In order to allocate its finite resources effectively, our system of justice has placed a reasonable burden on appellant to show initially that his case is one which requires additional attention. Since appellant has failed to satisfy this burden, he has no legitimate claim to further proceedings or relief.
IV. THE APPEARANCE OF UNLAWFUL COMMAND INFLUENCE
A. Introduction
As noted above, the appearance of unlawful command influence creates dangers for the military justice system as a whole just as actual unlawful command influence creates dangers for an individual accused. Therefore our determination that appellant’s claim of actual unlawful command influence merits no relief does not end our responsibility. Whenever actual unlawful command influence is an issue in a case, we must also consider whether the appearance of unlawful command influence exists and, if so, what remedial action is required. United States v. Rosser, 6 M.J. at 271-272.21
B. Rights and Remedies
Since it is the interests of the military justice system rather than those of the appellant which are endangered by the appearance of unlawful command influence, the remedy must relate to the interests of the system rather than those of the appellant. Similarly, when the appearance of unlawful command influence is at issue the procedural rights which the appellant would enjoy if it were his interests which were endangered are no longer relevant. United States v. Karlson, 16 M.J. 469, deals with an appellant’s rights during litigation which involves an appellant’s interests; that case is therefore inapplicable to this aspect of the court’s inquiry. Cf. United States v. Williams, 17 M.J. 207 (C.M.A.1984). In addition, since a court which is addressing the appearance of unlawful command influence is no longer engaged in fact-finding of the kind contemplated in United States v. Ray, 43 C.M.R. 171 (C.M.A.1971), United States v. Dubay, 37 C.M.R. 411 (C.M.A.1967), or United States v. Allen, 25 C.M.R. 8 (C.M.A.1957), those precedents are also inapplicable (although the court may find such proceedings desirable in a given case, as noted below). The proper procedures in appearance cases are those which the court itself finds necessary and useful in fulfilling its responsibilities.
The remedy to be applied for the appearance of unlawful command influence logically should be tailored to the restoration of public confidence under the particular circumstances of the case at hand; should avoid unnecessary expenditure of scarce resources; and should not create an actual injustice in place of an apparent one.
The most effective and satisfactory remedies center on the process of appellate review, which is itself a powerful force for public confidence. The mere fact that an appellate court has examined a case and affirmed the results is sufficient in the vast majority of court-martial cases to satisfy the public that justice was done by the trial court. There are three principal reasons for this. First, the fact that another body has reviewed the matter is a source of *890reassurance, on the principle that two heads are better than one. Second, the weight of the court’s reputation for competence and fairness is thrown into the balance. Third, and perhaps most important when applicable, the relevant facts are frequently laid out in an appellate court’s opinion as a matter of public record.
It is axiomatic that the best way to dispel the appearance of evil is to publish the truth about the situation. The information already in a trial record is usually sufficient to demonstrate the truth of the facts relied on by the appellate court. If not, the addition of evidence supplied by the parties at the appellate level often suffices. In some cases, it may be necessary to employ additional fact-finding techniques in order to accomplish this end. The available techniques form a stepped array, ranging, for example, from affidavits through depositions to evidentiary hearings. Each technique has advantages and disadvantages which must be considered in the context of a given case, as well as a differing degree of visible reliability as to the information it produces. The employment of court-ordered fact-finding techniques entails costs as well as benefits and the costs can be very high. A given fact-finding technique should be employed when it is both necessary and suitable for the restoration of public confidence in the circumstances of the case at hand, but not otherwise. Each ease involving the appearance of unlawful command influence presents a unique situation which requires a carefully tailored response.
Reversal of findings or sentence is an unmerited windfall to an appellant who has not suffered actual prejudice, although it may be required as a last resort when no other feasible course of action will restore public confidence. We believe that reversal is clearly not required in all cases where the appearance of unlawful command influence exists, but should be invoked only when other measures have been tried and found wanting or are obviously inadequate.
C. Application and Holding
What will reasonable members of the public believe in light of the mass apprehension and the “Peyote Platoon”? In our judgment, they will believe that the members of Colonel Beavers’ command understood that Colonel Beavers was sending them the following message:
Illegal drugs are a serious threat to this unit and its mission and so are soldiers who use or distribute them. I am not going to put up with illegal drugs in this command. Every effort will be made to detect soldiers who abuse or distribute drugs. I will do my best to see that those who are caught are vigorously prosecuted. To prove that I am serious, watch what happens to these soldiers who have been caught committing drug offenses.
Will reasonable members of the public believe that, as a result, the members of Colonel Beavers’ command altered or withheld testimony or disposed of a given case in a way they thought was wrong?22 Whatever the answer might have been based only on the mass apprehension and the “Peyote Platoon,”23 we have now published the fact that the resulting cases were disposed of in the broadest variety of ways, ranging from nonjudicial punishment *891to general court-martial. This information plainly indicates that the commanders involved, with the possible exception of the commander of Service Battery, 6/14th FA, exercised their own discretion. We have also published the fact that commissioned and noncommissioned officers testified for most of the soldiers tried in even the most serious cases, and the fact that there were obvious reasons which would account for appellant’s failure to offer similar testimony at his trial. This information plainly indicates that witnesses were available to testify for the soldiers who were apprehended. With all of this information now available to the public, together with the court’s judicial analysis of the circumstances, we are satisfied that the answer to the above question is “No,” and that any appearance of unlawful command influence which may have existed has been dispelled.24
Since we are satisfied that there is no appearance of unlawful command influence, we have no occasion for further action on that account.25V. USE OF THE MODEL
This model provides a systematic methodology, constructed from the principles laid down by the Court of Military Appeals, for the resolution of unlawful command influence cases. We believe that it is sufficiently flexible to allow effective resolution of unlawful command influence issues over the broad range of possible scenarios.
As noted above, the appearance of unlawful command influence will normally become relevant only in the absence of actual unlawful command influence because the process of publicly determining whether the appellant has been prejudiced by actual unlawful command influence, and remedying it if he has, will normally remove the appearance of unlawful command influence as well.
However, as an example of an alternate means of applying the model, a *892case may occur in which the appearance of unlawful command influence is so aggravated and so ineradicable that no remedy short of reversal of the findings and sentence will convince the public that the accused has been fairly tried. When such a situation is obvious from the beginning, the case should be approached through the appearance side of the model rather than wasting time and effort on an attempt to determine whether actual unlawful command influence has affected the case, since the remedy dictated by such an appearance would moot the actuality issue.
On the other hand, addressing appearances before reality carries with it a danger. One can, by confusing appearances with perceptions or otherwise, inadvertently label and dispose of as an appearance case what is really an instance of actual unlawful command influence in which the burden has shifted to the government but has not been met. Cf. United States v. Rosa, 46 C.M.R. 480 (N.B.R.1972). We are concerned that such confusion will blur the important distinction between appearance and reality and may result in the application of inappropriate remedies as well.
VI. OTHER ISSUES
A. Voluntariness of Guilty Plea
In an affidavit dated 25 October 1983, appellant stated that he had suffered great stress because of Colonel Beavers’ actions. He averred that he became frightened by the manner in which he was apprehended and “readily accepted a pretrial agreement for 16 months instead of possibly 30 years confinement at hard labor.” The implication is that appellant’s pleas of guilty were coerced by a threat from Colonel Beavers to deprive appellant of witnesses and otherwise interfere with appellant’s right to a fair trial.
Certainly a coerced plea of guilty cannot form the basis of a valid conviction. If we believed that appellant’s pleas were not voluntary, we would set aside the findings and the sentence forthwith. However, we must consider the following information from the record of trial. Appellant, a non-commissioned officer, twice distributed illegal drugs to a military police investigator posing as a field soldier, and the government had a strong case against him. These transactions occurred at an overseas training center where artillery units were engaged in live fire exercises. Appellant was able to obtain a favorable plea bargain which very significantly reduced his sentence vulnerability from thirty years’ confinement to sixteen months’. Appellant was 26 years old and a Sergeant with six years of military service, not an inexperienced or impressionable youth. Although appellant had the assistance of competent civilian counsel and full knowledge of the facts, he made no motion for a change of venue or other appropriate relief. During the inquiry into the providence of appellant’s pleas, appellant repeatedly stated that his pleas were voluntary and the trial judge so found. For these reasons, we do not believe appellant’s assertions, either the explicit or the implicit ones.
B. Article 13 Violations
Appellant alleges for the first time on appeal that he was subjected to pretrial punishment in violation of Article 13, UCMJ. We hold that appellant has waived this issue.
Normally, it is inappropriate to raise such issues for the first time on appeal. See United States v. Johnson, 41 C.M.R. 49 (C.M.A.1969); United States v. Martinez, 19 M.J. 744, 747 (A.C.M.R.1984), and cases cited therein. See also United States v. Palmiter, 20 M.J. at 100 (Everett, C.J., concurring) (“The conclusion that a servicemember is free to waive his protections under Article 13 is especially persuasive because sometimes there are obvious benefits for him to derive from such a waiver.”). Such appeals have been permitted, however, where the pretrial treatment of the appellant was “of such serious proportions” as to warrant appellate consideration. United States v. Johnson, 41 C.M.R. at 50 (C.M.A.1969); United States v. Peacock, 19 M.J. 909 (A.C.M.R.1985), pet. denied, No. 49621 (C.M.A. 10 May *8931985); United States v. Franklin, 41 C.M.R. 431 (A.C.M.R.), pet. denied, 19 USCMA 597 (C.M.A.1969).
This exception has thus far been invoked only where the appellant was a pretrial confinee who was compelled to live and work with sentenced prisoners.26 In the case at bar, appellant was not commingled with sentenced prisoners. The evidence offered in support of appellant’s allegations indicates that he was placed in the “Peyote Platoon” for a brief time to await processing of charges. After charges were preferred, he was told he could return to his unit but elected to stay and to work with the other members of the group. This evidence does not bespeak compulsion.
Even if we assume that appellant’s retention in the “Peyote Platoon” was involuntary, we reach no different result. Evaluating the situation on its own merits against the standard provided in United States v. Johnson, 41 C.M.R. 49, it is clear that the conditions under which appellant lived and worked were not so onerous as to be of “serious proportions.”
C. Article 55 Violations
In view of the broad language of Article 55, UCMJ, and in an abundance of caution, we have considered whether the removal of appellant’s unit crests and his affiliation with the “Peyote Platoon” constituted the infliction of “any other cruel or unusual punishment.” Assuming, without deciding, that this treatment was unauthorized, we find that it was neither cruel nor unusual under contemporary standards of decency. See generally Trop v. Dulles, 356 U.S. 86, 101, 78 S.Ct. 590, 598, 2 L.Ed.2d 630 (1958); Gregg v. Georgia, 428 U.S. 153, 173, 96 S.Ct. 2909, 2925, 49 L.Ed.2d 859 (1976). Neither did this treatment even remotely approach an “unquestioned and serious deprivation of basic human needs.” Rhodes v. Chapman, 452 U.S. 337, 347, 101 S.Ct. 2392, 2999, 69 L. Ed.2d 59 (1981) and cases cited therein. Finally, the failure to raise this issue at trial is strong evidence that appellant was not subjected to cruel or unusual punishment. Cf. United States v. Palmiter, 20 M. J. at 97. (Appellant’s failure to raise the issue at trial may also constitute waiver, cf. United States v. Martinez, 19 M.J. at 747, but we need not decide that issue in this ease.) Accordingly, we find that Article 55, UCMJ, was not violated.
D. Multiplicity
We hold that Specification 1 of the Charge (possession of hashish) is multiplicious for findings purposes with Specifications 2 and 3 of the Charge (distribution of hashish). United States v. Zubko, 18 M.J. 378 (C.M.A.1984). We are satisfied appellant suffered no prejudice as to his sentence.
VII. DECREE
The finding of guilty of Specification 1 of the Charge is set aside and that specification is dismissed. The remaining findings of guilty and the sentence are affirmed.
Chief Judge SUTER, Senior Judges MARDEN, McKAY and YAWN, and Judges FELDER, LYMBURNER, WERNER and WALCZAK concur. Judge WATKINS did not participate in the decision of this case. Judges COMEAU, GILL, WILLIAMS, CARMICHAEL, KENNETT, and ROB*894BLEE were assigned to this court after oral argument was held in this case and did not participate in the decision.. Except where otherwise indicated, the following reflects the contents of extra-record sworn statements which have been offered to this court on appeal. We have considered all of the material offered by both parties.
. Colonel Beavers has stated that he issued no blanket order to remove unit crests, but that when one commander asked about removing unit crests, he told that commander, “Yes, do it."
. The term "Peyote Platoon” may have derived from the fact that the covert drug operation which resulted in the mass apprehension was code-named "Peyote.” In addition, it appears to have been a common practice to use Native American names within the battalion for various events and activities. The official description of the unit crest states that the fourteen dots on the crest represent "a peyote, a species of small cactus which is a sacred symbol of the Kiowa Chieftain, Santanta.”
. The disposition of a military case is ordinarily determined from the bottom up. That is, the commander at the lowest level has the first opportunity to take action. If the action he believes appropriate is within his authority, he proceeds and the case is completed. If not, the commander must forward the case up the chain of command with his recommendation for disposition. Company-level commanders have the power to dispose of a case with nonjudicial punishment. The battalion is normally the first level at which the commander can convene a court-martial — ordinarily limited to a summary court-martial in tactical units. A brigade-level commander can convene special courts-martial, and a division commander can convene general courts-martial and special courts-martial empowered to impose bad-conduct discharges. Each level of command also has the powers of the levels below it.
. See United States v. Thomas, CM 444139; United States v. Winans, CM 444230; United States v. Jones, CM 444460; United States v. Rodriguez-Nieves, CM 444461; United States v. Swanson, CM 444462; United States v. Newton, CM 444466; United States v. Mowers, CM 444469; United States v. Davis, CM 444990. The accused in two of these cases made post-trial affidavits stating that their friends had been pressured not to speak up for them. The sole commissioned officer among the accused alleged he could not obtain witnesses at his Article 32, UCMJ, investigation. See United States v. Corriere, CM 444764. None of these three raised the issue at trial.
. This fact has probative value. See United States v. Yslava, 18 M.J. 670, 673 (A.C.M.R.1984) (en banc), pet. granted, 19 M.J. 281 (C.M.A. 1985); cf. United States v. Palmiter, 20 M.J. 90 (C.M.A.1985).
. Article 37, UCMJ, 10 U.S.C. § 837. This article forbids both "censure, reprimand, or admonition]” of any member, judge or counsel of a court-martial and “attempt[s] to coerce or by any unauthorized means, influence the action of a court-martial ... or any member thereof____"
. The confusion about the basis of the holding in Kitchens may stem from too cursory a reading of the case. Another case which may be subject to the same kind of confusion is United States v. Littrice, 13 C.M.R. 43 (C.M.A.1953), which was also decided on the basis of actual unlawful command influence. Id. at 51 (“Under the peculiar facts of this case, we find command coercion which was prejudicial to the accused.”). For an interesting reference to Lit-trice, see United States v. Hawthorne, 22 C.M.R. 83, 87-88 (C.M.A.1956). The following are further examples of cases decided on the basis of the presence or absence of actual unlawful command influence; United States v. Johnson, 34 C.M.R. 328, 330 n. 1, 333 (C.M.A.1964); United States v. Olson, 29 C.M.R. 102, 105 (C.M.A.1960); United States v. Ferguson, 17 C.M.R. 68, 81 (C.M. A. 1954).
. The statement that “any doubt ... must be resolved in favor of the accused" does not require absolute certainty, of course. The cases cited in Johnson indicate the standard intended by the court. "[W]e cannot answer [the question whether the appellant was prejudiced] with any degree of assurance; and since we cannot, we must accord the accused the benefit of the doubt____” United States v. Kitchens, 31 C.M.R. at 180 (emphasis supplied). "[W]e must ... determine whether there is a fair risk [that the appellant was prejudiced] and, if there is doubt in our minds, we resolve the uncertainty in his favor.” United States v. Mamaluy, 27 C.M.R. at 181 (emphasis supplied). Later, in United States v. Rosser, 6 M.J. 267, 272 (C.M.A.1979), the court held that "clear and positive” evidence could rebut the “presumption of prejudice" which the Johnson court had posited. See United States v. Treakle, 18 M.J. 646 (A.C.M.R.1984). Read together, these cases reveal that the intended standard is the absence of a "fair risk” of prejudice, or, stated conversely, "clear and positive” evidence that no prejudice is present.
. One excellent reason for considering the totality of the circumstances is that this approach focuses attention on the commander’s disposition of the case and on any resulting trial proceedings, which is where any prejudicial impact would by definition be felt. See, e.g., United States v. Johnson, 34 C.M.R. at 331-332; see also United States v. Wright, 37 C.M.R. 374, 376 (C.M.A.1967); United States v. Albert, 36 C.M.R. 267, 271 (C.M.A.1966).
. By "those outside the system" we mean the general public. As noted above, this includes the rank and file of military personnel. Of course, a soldier who is a participant in a particular case (such as a witness) does not thereby lose his identity as a member of the general public. Nevertheless, it is the opinion of the public in general which constitutes the appearance of unlawful command influence, not the opinion of a particular member of the general public. Therefore, a particular soldier’s perception that he has been commanded to withhold testimony does not, in itself, constitute an appearance of unlawful command influence.
. This is true only if that perception is reasonable. United States v. Johnson, 34 C.M.R. at 331; United States v. Ferguson, 17 C.M.R. 68, 81 (C.M.A.1954); United States v. Knudson, 16 C.M.R. 161, 171 (C.M.A.1954); United States v. Treakle, 18 M.J. at 653.
. In order to act in a case, a court created under Article III of the Constitution normally must have before it a "case or controversy.” Mills v. Green, 159 U.S. 651, 653, 16 S.Ct. 132, 40 L.Ed. 293 (1895); see also Southern Pacific Terminal Co. v. ICC, 219 U.S. 498, 514, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911). Perhaps it is because of this doctrine that the Court of Military Appeals and the Courts of Military Review have sometimes proceeded in ways which might suggest that the existence of an appearance of unlawful command influence, standing alone, gives rise to a justiciable issue between the accused and the government. There are, of course, legitimate exceptions to the "case or controversy” requirement. See, e.g., Southern Pacific Terminal Co. v. ICC, 219 U.S. at 515, 31 S.Ct. at 283; Moore v. Ogilvie, 394 U.S. 814, 816, 89 S.Ct. 1493, 1494, 23 L.Ed.2d 1 (1969). Which exception, based on what rationale, is applicable to the appearance doctrine need not detain us; the Court of Military Appeals has posited the appearance doctrine and it is our responsibility to apply it.
. Although we regard this question as more fundamental than whether such courts-martial are "courts of record or courts of general jurisdiction", we do view such courts-martial as "courts of record” and this fact lends support to our conclusion. A court of record is "[a] court that is required to keep a record of its proceedings, and that may fine or imprison." Black’s Law Dictionary 319 (5th ed. 1979). A verbatim transcript is made of the proceedings of these courts-martial, the accused is provided with a free authenticated copy, see, e.g., Articles 19 and 54, UCMJ, 10 U.S.C. §§ 819, 854; RCM 1103; MCM, 1969, para. 82, and the original transcript is filed as a permanent, retrievable public record, see, e.g., Army Regulation 27-10, Legal Services: Military Justice, para. 5-36 (15 Mar 1985); Army Regulation 340-18, Office Management: The Army Functional Files System, paras. 404-02, 404-03 (15 Dec 1984). Such courts-martial are empowered to adjudge fines and confinement and, in the case of general courts-martial, the death penalty. See, e.g., Articles 18, 19, 56, and 118, UCMJ, 10 U.S.C.A. §§ 818, 819, 856, 918; RCM 1003; Manual for Courts-Martial, United States, 1984 [hereinafter cited as MCM, 1984], part IV; MCM, 1969, paras. 126 and 127.
We also note that Article 59(a), UCMJ, 10 U.S.C. § 859(a), requires a showing that an "error materially prejudiced] the substantial rights of the accused” before a court-martial proceeding can be “held incorrect on the ground of an error of law.” This statutory provision is wholly compatible with the existence of a rebuttable presumption of the correctness and regularity of court-martial proceedings.
. Although few military cases have commented explicitly on the status to be accorded the judgments and proceedings of courts-martial, we find the presumption of regularity implicit in virtually every opinion of the Court of Military Appeals and the Courts of Military Review. For the cases which contain explicit support for the proposition, see United States v. Weaver, 1 MJ. 111, 115 (C.M.A.1975); United States v. West, 17 MJ. 627 (N.M.C.M.R.1983), pet. denied, 18 MJ. 22 (C.M.A.1984); cf. United States v. Brown, 48 C.M.R. 827, 830 (A.C.M.R.1974) ("Courts have long followed the presumption of regularity in the conduct of governmental affairs”). There is a line of cases which must be carefully distinguished to avoid confusion. These are the cases, such as United States v. Alston, 11 M.J. 656 (A.F.C.M.R.1981), and United States v. Newton, 22 C.M.R. 534 (A.C.M.R.1956), which hold that because each court-martial must be called into existence by the act of a convening authority, no presumption exists that any given trial was held before a properly constituted court-martial. This is a very different matter than the rule we posit, which presumes the regularity of the judgments and proceedings of a court-martial once it has been established that that court-martial was properly called into existence. See United States v. Newton, 22 C.M.R. at 542-543.
. We believe this principle also applies when the appearance of unlawful command influence is involved.
. As an alternative, the standard could be elevated when unlawful command influence is first addressed on appeal, by analogy to the treatment of the issue of mental responsibility when addressed for the first time on appeal. When mental responsibility is addressed at the trial level, the “some evidence” standard applies. RCM 916(k)(3)(A); United States v. Cockerell, 49 C.M.R. 567 (A.C.M.R.1974). When it is addressed for the first time on appeal, the standard is higher, requiring evidence which, when considered with any relevant evidence in the “entire record", can reasonably be expected to result in a different outcome at a rehearing. United States v. Triplett, 45 C.M.R. 271, 277 (C.M.A.1972). This elevation of the standard when the issue of mental responsibility is addressed for the first time on appeal was accomplished in Triplett despite the court’s explicit recognition that the issue of mental responsibility occupies a favored position in the law. Id. at 273. We believe that the issue of unlawful command influence is even more sensitive and more favored and therefore decline to elevate the standard on appeal.
. “Whether an appellate finding of unlawful command influence mandates reversal of the findings and sentence . without particularized prejudice to an accused” is an issue which is presently before the United States Court of Military Appeals. United States v. Yslava, 19 MJ. 281 (C.M.A.1985). Unless and until that court decides otherwise, we are bound by the rule of specific prejudice currently in effect as a result of that court's previous decisions. In light of the Congressional mandate in Article 59(a), UCMJ, and in order to avoid Draconian or Procrustean remedies, we believe that a rule of general prejudice is inappropriate in cases involving unintentional violations. The court’s disposition of the pending issue may or may not resolve whether unlawful command influence of the sort which would constitute a violation of Article 98, UCMJ, calls for application of a rule of general prejudice. Since the rule to be applied in cases of “knowing or intentional” violations of Article 37 is a question which is not presented by the case at bar, we reserve comment in that respect.
. Cf. NLRB v. Ohio New and Rebuilt Parts, Inc., 760 F.2d 1443 (6th Cir.1985) (mere possibility that public official will be biased in performance of quasi-judicial duties not sufficient to *887overcome presumption of impartiality in performance of official duties). Appellant has eloquently argued that the appropriate standard is an "objectively reasonable possibility." (Colonel William G. Eckhardt at oral argument, 28 March 1985.) The elements of objectivity and reasonableness are prominent features of the standard we apply. See subsection III A, The Appellate Model, supra. What remains is the question whether a possibility will suffice. For the reasons stated, we hold that it will not.
. For instance, the following are cases in which the burden clearly had shifted to the government: United States v. Johnson, 34 C.M.R. at 331 (“As a matter of principle, any doubt in the matter must be resolved in favor of the accused.”); United States v. Kitchens, 31 C.M.R. at 180 ("We cannot answer that question with any degree of assurance; and since we cannot, we must afford the accused the benefit of the doubt____”); United States v. Toon, 48 C.M.R. 139, 143 (A.C.M.R.1973) (“Under the circumstances, we cannot say with conviction that the court-martial was uninfluenced ... in its deliberation on the sentence.").
. For a definition of the appearance of unlawful command influence, see subsection II B 2, Thesis and II B 4, Analysis, supra.
. This is a question which does not appear to have occurred to Colonel Beavers or his servicing judge advocates; consequently, thousands of hours of effort have been expended toward resolving the resulting controversy.
. From that information alone, reasonable members of the public might conclude that some of Colonel Beavers’ audience drew the same inferences that Sergeant Brayboy and (perhaps) Captain Burks did. Applying United States v. Johnson, 34 C.M.R. at 331, we would then attribute to those members of the public the belief that such soldiers did as they thought they had been told to do. This segment of the public might then believe that enough of the audience was affected to cause harm in our appellant’s case. When a court finds that a substantial segment of the public would have such a belief, appropriate corrective action is required based on the appearance of unlawful command influence alone. (See subsection IV B, Rights and Remedies, supra.)
. In addition, the case at bar simply does not involve or even mimic the immediate and invasive events which prompted the result in United States v. Brice, 19 MJ. 170 (C.M.A.1985).
. Id. The commingling of sentenced and unsentenced prisoners has historically been held to be a per se violation of Article 13. United States v. Pringle, 41 C.M.R. 324 (C.M.A.1970); United States v. Nelson, 39 C.M.R. 177 (C.M.A. 1969); United States v. Bayhand, 21 C.M.R. 84 (C.M.A.1956). These cases form the underpinnings of the exception articulated in United States v. Johnson, 41 C.M.R. 49. However, a recent opinion by Judge Cox in United States v. Palmiter, 20 M.J. at 96, stating that “commingling per se without more does not constitute punishment,” represents a challenge to the viability of Pringle, Nelson, and Bayhand. Should Judge Cox’s view prevail, the Johnson exception could arguably be undercut as well. Pending the outcome of that debate, however, we are clearly governed by Johnson and seek only to apply it as was intended by the Johnson court.