The sole, inexorable issue presented by this appeal requires us to predict whether the Supreme Court of the United States will apply its decision in O’Callahan v. Parker, 395 U.S. 258, 89 S.Ct. 1683, 23 L.Ed.2d 291 (1969), to comparable proceedings of military courts which reached a stage of complete finality prior to June 2, 1969, the date that decision was announced. The court below reasoned that O’Callahan should be denied retroactive application, 305 F. Supp. 1186. We affirm.
It all started for James Roy Gosa on August 13, 1966. He was then serving as a member of the United States Air Force stationed at Warren Air Force Base in Wyoming, and on the night in question, he was officially off-duty and with permission of his superior officers had left the military post dressed in mufti. Around midnight Gosa allegedly raped a civilian in Cheyenne, Wyoming. The asserted victim was not on any type of military duty and had no direct or indirect relationship with the military establishment. Although Gosa was arrested by Cheyenne civilian authorities for prosecution in their courts, he was subsequently released from their detention when the complaining party failed to appear. He thereupon was immediately taken into military custody and was charged with violation of Article 120 of the Uniform Code of Military Justice (U.C.M.J.) 10 U.S.C.A. § 920, which provides that any person subject to the Code who commits an act of rape may be punished as a court-martial may direct. Pursuant to the provisions of sub-chapters IV and V, U.C.M.J. (10 U.S. C.A. §§ 816-829), a general court-martial was duly convened which tried petitioner and, on December 2, 1966, found him guilty as charged. All of the multiple review procedures provided by the U.C. M.J. were accorded.1 On July 11, 1967 Gosa petitioned the Court of Military Appeals for a grant of review under Article 67, U.C.M.J. (10 U.S.C.A. § 867). All direct review procedures were exhausted and Gosa’s conviction became final in law on August 16, 1967 when the Court of Military Appeals denied review.
On August 21, 1969, Gosa filed his application for a writ of habeas corpus in the court below and on the 6th of November, 1969, filed with the United States Court of Military Appeals a motion to vacate his sentence and conviction. Both the application and the mo*755tion were based upon assertions that Gosa’s confinement was invalid in the light of the decision in O’Callahan that the general court-martial which tried him lacked jurisdiction. Both the application for habeas relief and the motion to vacate, which the Court of Military Appeals treated as a petition for reconsideration, were denied.2
O’CALLAHAN — COMPARISON AND ANALYSIS
The Supreme Court granted a petition for certiorari review of a Tenth Circuit case styled Relford v. Commandant U. S. Disciplinary Barracks, Ft. Leavenworth, 409 F.2d 824, for the limited purpose of deciding the retroactivity and scope of O’Callahan. See 397 U.S. 934, 90 S.Ct. 958, 25 L.Ed.2d 114 (1970). However, when Relford came on to be heard on its merits the Court determined that because Relford’s offense had been perpetrated within the geographical boundary of a military post, it had a service connection which O’Callahan lacked. Thus, a decision on retroactivity was deemed inappropriate. 401 U.S. 355, 91 S.Ct. 649, 28 L.Ed.2d 102 (1971). Rel-ford enumerated 12 factors which, if present, deprive a military court-martial of jurisdiction to try a member of the Armed Forces otherwise under the jurisdiction of that court by the congressional mandate of Article 2, U.C.M.J. (10 U.S. C.A. § 802).3 Each of these factors is unquestionably present in Gosa’s case; indeed the only distinction, locale — the Territory of Hawaii vis-a-vis the State of Wyoming — if effective at all, makes Gosa’s case stronger. Indubitably, had O’Callahan been rendered prior to these events in Gosa’s case, that decision would have deprived the general court-martial which tried Gosa of jurisdictional authority to hear or determine that cause.4 We cannot avoid deciding the scope of its applicability as precedent. We therefore must analyze it.
In Relford, the Court capsuled its prior holding in O’Callahan thus:
[B]y a five-to-three vote, the Court held that a court-martial may not try a member of our armed forces charged with attempted rape of a civilian, with housebreaking, and with assault with intent to rape, when the alleged offenses were committed off-post on American territory when the soldier was on leave, and when the charges could have been prosecuted in a civilian court.
*756Looking in greater detail to the opinion itself, we first note that certiorari in O’Callahan was limited to the question:
Does a court-martial, held under the Articles of War, Tit. 10, U.S.C. § 801 et seq., have jurisdiction to try a member of the Armed Forces who is charged with commission of a crime cognizable in a civilian court and having no military significance, alleged to have been committed off-post and while on leave, thus depriving him of his constitutional rights to indictment by a grand jury and trial by a petit jury in a civilian court? 393 U.S. 822, 89 S.Ct. 177, 21 L.Ed.2d 93.
After reciting the unlimited grant of congressional authority “To make Rules for the Government and Regulation of land and naval Forces” contained in Article I, Section 8, Clause 14 of the Constitution, and the Bill of Rights language which excepted only cases arising in the land or naval forces, and excepted those cases only from the Fifth Amendment’s requirement of grand jury presentment or indictment, Mr. Justice Douglas, speaking for the majority, pointed out that Congress had developed a system of military justice with fundamental differences from civilian courts. He stated the issue in these words:
If the case does not arise ‘in the land or naval forces,’ then the accused gets first, the benefit of an indictment by a grand jury and second, a trial by jury before a civilian court as guaranteed by the Sixth Amendment and by Art. Ill, § 2, of the Constitution which provides, in part:
‘The Trial of all Crimes, except in Cases of Impeachment, shall be by Jury; and such Trial shall be held in the State where the said Crimes shall have been committed; but when not committed within any State, the Trial shall be at such Place or Places as the Congress may by Law have directed.’
Those civil rights are the constitutional stakes in the present litigation. (395 U.S. 262, 89 S.Ct. 1685) (Final sentence emphasis supplied.)
Then, after a discussion of pre- and post-Constitution military court history, the conclusion of the decision was put in this language:
We have concluded that the crime to be under military jurisdiction must be service connected, lest ‘cases arising in the land or naval forces, or in the militia, when in actual service in time of war or public danger,’ as used in the Fifth Amendment, be expanded to deprive every member of the armed services of the benefits of an indictment by a grand jury and a trial by a jury of his peers. (395 U.S. at 272, 89 S.Ct. at 1690).
Clearly then, grand and petit jury protections were the core rights sought to be vouchsafed. Since the opinion also spoke of other procedural aspects of the military system and compared some of these to civilian court processes, we cannot state with absolute assurance that the Court will later hold that only these two Bill of Rights protections were involved. However, this uncertainty is not critical to our conclusion.
Our analysis of O’Callahan must also center upon determining whether the Court decided that military tribunals lacked adjudicatory power over servicemen’s offenses which were not “service-connected”. Did the opinion hold that courts-martial lacked power over the subject matter and person of such a soldier because Congress had no constitutional authority to vest it, or did O’Callahan decide that the lack of grand and petit jury procedures (and perhaps other civilian court protections) resulted in the loss of jurisdiction otherwise within the control of congressional grant? In United States ex rel. Flemings v. Chafee, 330 F.Supp. 193 (E.D.N.Y. 1971), a most thorough and scholarly judicial determination, Judge Wein-stein comes to the somewhat guard*757ed conclusion that lack of adjudicatory power was the rationale of the decision. He also notes that other jurists have reached the opposite conclusion. See the decision here on appeal, Gosa v. Mayden, 305 F.Supp. 1186 (N.D.Fla.1969); Mercer v. Dillon, 19 U.S.C.M.A. 264, 41 C.M.R. 264 (1970); Schlomon v. Moseley (1971) [Civ.No. L-1003, May 19, 1971]; and the opinion of the Board of Review [Art. 66, U.C.M.J.] in United States v. King, ACM 20361 (July 30, 1969), review denied U.S.C.M.A., 40 C.M.R. 327 ( ).5
Despite the weight of authority to the contrary, we find the reasoning of Flem-ings persuasive on this issue. Read with an open mind, O’Callahan’s foundation, framework and structure deny to the legislation which breathed the breath of judicial life into the forum that tried Sgt. O’Callahan, the necessary basis in constitutional power to reach his type of case. It declares that because of the Bill of Rights, Article I, Section 8 cannot be read to enable Congress to authorize the military courts to try a peacetime soldier who, freed of military responsibility, albeit temporarily, is charged with a crime (a) cognizable in a civilian court and (b) having no military significance. It placed O’Callahan in the same status as a discharged serviceman,6 a civilian employed by the Armed Forces overseas,7 or a civilian accompanying the military service overseas.8 Although the language of the opinion does not say it in so many words, it holds that a member of the Armed Forces has an area of off-duty life wherein his general serviceman status is an insufficient nexus to bring his actions under the constitutional range of military “Government and Regulation”. RETROACTIVITY — NEED IT BE TESTED
The threshold problem we face in the instant appeal presents a completely novel issue, for O’Callahan did not overturn a prior precedent of the Court. It invalidated a part of a law made by Congress. We must determine whether the Supreme Court’s doctrines of retroac-tivity are applicable to a decision which undoes congressional action in a context where the Act involved has a half-century background of at least tacit judicial approval.
Of course it is illogical to assign shadings or degrees of nullity to acts which may be classed as void. But, with equal certitude, logic would assert that if a court decision reversing a prior judicial error of fundamental constitutional law must be tested for its retroactive impact, a court decision reversing a long established, judicially recognized legislative rule ought to be entitled to the same testing. For how could one assert that a proceeding which is invalid because it is in excess of constitutional right is less vacuous because its nullity results from judicial error rather than legislative action? Equally obviously, it is no more illogical to apply the rules for determining retroactivity to losses of liberty or property resulting from unconstitutional judicial precedent than to unconstitutional legislative action.
With these axioms in mind, we look to see how the doctrine of judicial determination of retroactivity came into existence. Norton v. Shelby County, Tennessee, 118 U.S. 425, 6 S.Ct. 1121, 30 L.Ed. 178 (1886), declared that an unconstitutional enactment conferred no right, imposed no duty, afforded no protection and was, in legal contemplation, *758as inoperative as though it had never been passed.9 In Chicot County Drainage Dist. v. Baxter State Bank, 308 U.S. 371, 60 S.Ct. 317, 84 L.Ed. 329 (1940), the Court retracted the broad statements of Norton and declared that those statements had to be taken with qualifications. It reasoned that the actual existence of a law prior to the determination of unconstitutionality is an operative fact and may have consequences which cannot justly be ignored. Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601 (1965), the progenitor of the Court’s precedents in this field, imported the rationale of Chicot into the criminal law. Linkletter is controlling precedent here for the assertion that there is no basis for distinction between legislation and judge-made law in reasoning retroactivity vel non.
Under our view set out above, O’Callahan presents another significant first in the field of retroactive adjudication because it involved a determination that the Constitution would not support a law which invested a judicial system with jurisdiction over the person and subject matter of the action tried. It has always been the law that proceedings of a court which is without jurisdiction of the subject matter are void, but does this inevitably lead to the necessity for full retrospective application of the court decision which first discovers and announces the jurisdictional deficit? We hold it does not.
Indeed, the question of jurisdiction lies at the second level in the analysis of the problem at hand. For though O’Callahan determines a lack of jurisdiction, the determination is the result of a new adjudication of constitutional right. Being taught as we will later show, that we can reject as inconsequential the specific provision of the Constitution on which a new precedent rests, as well as the relative value of the constitutional guarantee involved, we are impelled to the conclusion that this more than half-decade of precedent for selective retro-activity may not be ignored. In the light of this reasoning, no decision can arbitrarily be assigned full retroactivity solely on the basis that it operates on jurisdictional rather than proscribing the denial of a fundamental constitutional right that relates to another area of the processes of the criminal law.
Finally, one distinction remains to be made. The case at bar is not like United States v. United States Coin & Currency, 401 U.S. 715, 91 S.Ct. 1041, 28 L.Ed.2d 434 (1971). There the Court prohibited a forfeiture of property which had its basis in the refusal of a citizen to incriminate himself, which is, of course, a form of conduct that could not have been constitutionally punished at any time from and after the date the Bill of Rights was adopted. Such right to refuse to incriminate oneself was not first confected in the decisions which declared that gambling registration forms would constitute self-incrimination and could not be required. Since there was no right to arrest the gambler who refused to file such forms, there was no rig! „ at all to seize his gambling equipment at the time of his illegal arrest. The opinion closes on this cogent note:
In the case before us, however, even the use of impeccable factfinding procedures could not legitimate a verdict decreeing forfeiture, for we have held that the conduct being penalized is constitutionally immune from punishment. (401 U.S. 715, 91 S.Ct. at 1046) 10
It begs the question to assert that because the issue in O’Callahan is pure jurisdiction that Coin & Currency is analogous. No citizen can rightfully be jailed for exercising his religious free*759doms or petitioning his government for a redress of grievances, even though some new litigation situation could conceivably arise in which such deprivations of liberty would be expressly voided. His right to be free of such restraints is clearly established in the Constitution itself and not in decisional precedent. This is just not at all the same as the new view of the right of Congress to regulate military jurisdiction over Sgt. O’Callahan which the Court announced on June 2, 1969. This latter decision is in the same category as Bloom v. State of Illinois, 391 U.S. 194, 88 S.Ct. 1477, 20 L.Ed.2d 522 (1968), or any other new or altered adjudication that changes fundamental rights of those accused of crime. One of the O’Callahcm-Relford criteria, see note 3 supra, is that the crime involved must be cognizable in a civilian court. The crime of rape, for which O’Callahan, Relford and Gosa were prosecuted, was equally contrary to the laws of the civilian jurisdictions involved and the U.C.M.J. The issue for those cases and Gosa’s as well is not whether the accused could be tried at all, but which forum had the right to conduct the proceedings. Not whether, but where.
In sum, we hold that there is no arbitrary or simplistic basis for deciding retroactivity.11 The heart of the doctrine is a reasoned application of new constitutional precedent. We must follow the course those decisions dictate in the case at bar.
RETROACTIVITY — GENERAL PRINCIPLES
The historical background against which the Supreme Court of the United States first explicated a set of rules to govern whether new court-enunciated constitutional principles of criminal law were to be invested with prospective, limited or fully retroactive application is detailed in Mr. Justice Clark’s opinion in Linkletter v. Walker, supra. The intervening decisions which touched on this principle, as well as the comments of legal scholars who have explored the subject matter, have just been collated in Mr. Justice White’s opinion in Williams v. United States, 401 U.S. 646, 91 S.Ct. 1148, 28 L.Ed.2d 388 (1971).12 We need *760not attempt to duplicate this effort. However, the problem in this case is not discovering precedent but determining how it should be applied.13 We therefore deem more than bare citations necessary to develop the rationalization of our disposition of the vital and legally complex issue presented.
Linlcletter denied full retroactive application to the rule of Mapp v. Ohio, 367 U.S. 643, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961). Mapp changed prior de-cisional law by holding that the Fourteenth Amendment operated to require State courts to exclude evidence from criminal trials when it had been obtained in searches and seizures which violated the Fourth Amendment. The majority opinion epitomized its holding thus:
[W]e believe that the Constitution neither prohibits nor requires retrospective effect. * * *
Rather than ‘disparaging’ the [Fourth] Amendment we but apply the wisdom of Justice Holmes that ‘[t]he life of the law has not been logic: it has been experience.’ Holmes, The Common Law (Howe ed. 1963) * * *
In short, we must look to the purpose of the Mapp rule; the reliance placed on the Wolf doctrine; and the effect on the administration of justice of a retrospective application of Mapp. * * *
All that we decide today is that though the error complained of might be fundamental it is not of the nature requiring us to overturn all final convictions based upon it. (Emphasis supplied)
Tehan v. United States ex rel. Shott, 382 U.S. 406, 86 S.Ct. 459, 15 L.Ed.2d 453 (1966), refused retroactivity to Griffin v. California, 380 U.S. 609, 85 S.Ct. 1229, 14 L.Ed.2d 106 (1965), which denied to the states that did not already proscribe it, the right to comment on the failure of the defendant to testify in a criminal proceeding. Tehan differentiated the self-incrimination protections of the Fifth Amendment from (i) the denial of the assistance of counsel, (ii) the burdening of the opportunity of in di gents to appeal and (iii) the use of coerced confessions, all of which had been given fully retroactive application, by pointing out that these latter three processes infected a criminal proceeding with “the clear danger of convicting the innocent.” By contrast, it classed the privilege against self-incrimination not as an adjunct to the ascertainment of truth but as a protection of the right of an individual to be let alone. By application of the tripartite test of Linlcletter —purpose, reliance and effect — the Court reasoned that a prospective application of Griffin’s rule best served the interests of justice.
In a majority opinion by Chief Justice Warren, Johnson v. State of New Jersey, 384 U.S. 719, 86 S.Ct. 1772, 16 L.Ed.2d 882 (1966) refused retroactive effect to Escobedo v. State of Illinois, 378 U.S. 478, 84 S.Ct. 1758, 12 L.Ed.2d 977 (1964) and Miranda v. State of Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966), which had defined and delineated the rights of those accused of crime during the arrest and investigatory period and had rendered inadmissible statements and confessions received when such accused were without counsel and had not been warned and advised of their rights. In the course of refining and applying Linlcletter, Johnson14 stated:
We here stress that the choice between retroactivity and nonretroaetivity in no way turns on the value of the constitutional guarantee involved. * * We also stress that the retroactivity *761or nonretroaetivity of a rule is not automatically determined by the provisions of the Constitution on which the dictate is based. Each constitutional rule of criminal procedure has its own distinct functions, its own background of precedent, and its own impact on the administration of justice, and the way in which these factors combine must inevitably vary with the dictate involved. * * *
Finally, we emphasize that the question whether a constitutional rule of criminal procedure does or does not enhance the reliability of the fact-finding process at trial is necessarily a matter of degree. * * *
We are thus concerned with a question of probabilities and must take account, among other factors, of the extent to which other safeguards are available to protect the integrity of the truth-determining process at trial. (384 U.S. at 728-729, 86 S.Ct. at 1778-1779) 15
Assuming that there were past injustices which could have been averted by having counsel present at the time, Mr. Justice Brennan, speaking for the Court in Stovall v. Denno, 388 U.S. 293, 87 S.Ct. 1967, 18 L.Ed.2d 1199 (1967), refused retroactive application to United States v. Wade, 388 U.S. 218, 87 S.Ct. 1926, 18 L.Ed.2d 1149 (1967) and Gilbert v. State of California, 388 U.S. 263, 87 S.Ct. 1951, 18 L.Ed.2d 1178 (1967), which had rendered inadmissible in State and Federal prosecutions pretrial identification procedures handled in the absence of counsel. It is this opinion which announced the most frequently quoted epitome for retroactivity determinations :
(a) the purpose to be served by the new standards, (b) the extent of the reliance by law enforcement authorities on the old standards, and (c) the effect on the administration of justice of a retroactive application of the new standards. (388 U.S. at 297, 87 S.Ct. at 1970)
Since most other cases declaring a constitutional right to counsel have been granted full retroactive effect and since Stovall speaks of the Wade and Gilbert rights as ones which prevent dangers and unfairness in the fact-determining process and enhance the integrity and reliability of trials, it is of considerable moment to the ease at bar that Stovall expressly gave the reliance and effect factors overriding significance in restricting the effect of the decisions to Wade and Gilbert alone.
Even though it is a short per curiam with two justices concurring on different grounds and two dissenting, DeStefano v. Woods, 392 U.S. 631, 88 S.Ct. 2093, 20 L.Ed.2d 1308 (1968) is perhaps the most significant precedent to the case at bar. There, the Court refused retroactive application to Duncan v. State of Louisiana, 391 U.S. 145, 88 S.Ct. 1444, 20 L.Ed.2d 491 (1968), which held that States cannot deny a request for jury trial in serious criminal cases, and to Bloom v. State of Illinois, supra, which held that the right to jury trial extends to trials for serious criminal contempts. Based upon the (a) (b) (c) test that had been distilled from prior decisions in Stovall, the Court reasoned Duncan should not be applied retroactively because:
[I]n the context of the institutions and practices by which we adopt and apply our criminal laws, the right to jury trial generally tends to prevent arbitrariness and repression. As we stated in Duncan, ‘We would not assert, however, that every criminal trial — or any particular trial — held before a judge alone is unfair or that a defendant may never be as fairly treated by a judge as he would be by a jury.’ *762(392 U.S. 633-634, 88 S.Ct. 2095) (Emphasis supplied.)
Of greater moment for Gosa’s ease was this application of retrospectivity standards to Bloom:
The considerations are somewhat more evenly balanced with regard to the rule announced in Bloom v. State of Illinois. One ground for the Bloom result was the belief that contempt trials, which often occur before the very judge who was the object of the allegedly contemptuous behavior, would be more fairly tried if a jury determined guilt. Unlike the judge, the jurymen will not have witnessed or suffered the alleged contempt, nor suggested prosecution for it. However, the tradition of non-jury trials for contempts was more firmly established than the view that States could dispense with jury trial in normal criminal prosecutions, and reliance on the cases overturned by Bloom v. State of Illinois was therefore more justified. Also, the adverse effects on the administration of justice of invalidating all serious contempt convictions would likely be substantial. Thus, with regard to the Bloom decision, we also feel that retroactive application is not warranted. (392 U.S. at 634, 88 S.Ct. at 2096.) (Emphasis supplied.)
On April 5, 1971, in the combined eases of Williams v. United States and Elkanich v. United States, supra, the Court refused retroactive application to Chimel v. The State of California, 395 U.S. 752, 89 S.Ct. 2034, 23 L.Ed.2d 685 (1969), which denied to states the right to admit certain evidence seized incident to an arrest. In a four-judge majority opinion announced by Mr. Justice White,16 the Court gave us this most current assay of how it applied the (a) part test of Stovall:
Since * * * [Linkletter], we have held to the course that there is no inflexible constitutional rule requiring in all circumstances either absolute retro-activity or complete prospectivity for decisions construing the broad language of the Bill of Rights.
Where the major purpose of new constitutional doctrine is to overcome an aspect of the criminal trial which substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts in past trials, the new rule has been given complete retroactive effect. Neither good-faith reliance by state or federal authorities on prior constitutional law or accepted practice, nor severe impact on the administration of justice has sufficed to require prospective application in these circumstances. It is quite different where the purpose of the new constitutional standard proscribing the use of certain evidence or a particular mode of trial is not to minimize or avoid arbitrary or unreliable results but to serve other ends. (401 U.S. 646 at 651-653, 91 S.Ct. at 1151-1153.)
Two other opinions, rendered the same day as Williams and Elkanich, dealt with the application of Marchetti v. United States, 390 U.S. 39, 88 S.Ct. 697, 19 L.Ed. 2d 889 (1968), and Grosso v. United States, 390 U.S. 62, 88 S.Ct. 709, 19 L.Ed.2d 906 (1968), which prohibited the prosecution of gamblers who failed to register and pay a tax imposed by federal law on the grounds that the registration requirement violated Fifth Amendment privileges against self-incrimination. In Mackey v. United States, 401 U.S. 667, 91 S.Ct. 1160, 28 L.Ed.2d 404 (1971), four justices concluded that Marchetti and Grosso should not be applied retroactively since there was no threat to the reliability of the fact-finding process involved in Mackey’s trial for income tax evasion because of the evidentiary use of the Marchetti-Grosso proscribed wa*763gering tax forms.17 United States v. United States Coin & Currency, supra, involved forfeiture of property because of a refusal to file a self-incriminating form. It has been discussed above.18
This full, but hopefully not fulsome, development of the elements which form today’s doctrine for court control of the reach of precedential effect of new pronouncements of constitutional rights in the field of criminal law, sets the stage for testing O’Callahan for this case. The tests will be applied in the (a) (b) (c) formula format of Stovall with due regard for the explanations and emphasis supplied to each criterion by subsequent decisions.
RETROACTIVITY — APPLICATION OF THE TESTS
(A) Purpose of the New Standard.
The purpose of O’Callahan may be stated in two ways. Affirmaitvely, it secured the constitutional right of grand jury presentment or indictment and petit jury trial to servicemen on active duty who were accused of crimes having no service connection.19 Negatively phrased, O’Callahan denied military jurisdiction which exceeded the least possible power which the Constitution reposed in Congress and did so to avoid numerous incidents and functions of military justice considered less satisfactory to the determination of guilt than procedures available in civilian courts that would occupy the jurisdictional vacuum.
Under the affirmative statement of the test, DeStefano, supra, particularly since it applies Bloom prospectively, predicts that the Supreme Court will hold O’Callahan should not be applied retroactively. In the words of the Court, the purposes of that decision were to require jury trials (a) in serious criminal cases because such jury trials generally tend “to prevent arbitrariness and repression” which even impartial judges might exhibit, and (b) in serious criminal contempt cases because juries could “more fairly” try alleged eontemnors than could a judge who had been the object of the contemptuous act. The same purposes underlie O’Callahan, even given its stern view of the faults of military courts which we detail below.
Obviously the negative statement gives a much broader sweep to O’Callahan and requires an independent analysis of purpose in the light of the Court’s prior inquiries into the reliability of guilt determination — the fairness of the trial — the very integrity of the fact-finding process.
Candor, rather than even a hint of disrespect, compels the observation here that this particular facet of testing for retro-activity deals almost entirely in subjective judge-conceived notions based in no part on tangible evidence developed by an adversary process or otherwise, but rather upon feelings and concepts which are the product of each individual jurist’s experiences and readings. Thus, with no deepseated assurance that the question *764ought not be certified to the High Court,20 we reason to the following conclusions.
No analysis of the wider purpose of O’Callahan would be correct that did not weigh the critical ingredient of reliability of the fact-finding processes which it altered. Likewise, no test of this factor would be objective which overlooked the critical, indeed deprecatory, terms which the majority opinion applied to the general system of military justice.
Quoting from United States ex rel. Toth v. Quarles, 350 U.S. at 22-23, 76 S.Ct. at 8, it states:
There are dangers lurking in military trials which were sought to be avoided by the Bill of Rights and Article III of our Constitution. Free countries of the world have tried to restrict military tribunals to the narrowest jurisdiction deemed absolutely essential to maintaining discipline among troops in active service.
Then Mr. Justice Douglas’ own words proclaim:
A court-martial is not yet an independent instrument of justice but remains to a significant degree a specialized part of the overall mechanism by which military discipline is preserved. * * * expansion of military discipline beyond its proper domain carries with it a threat to liberty. (395 U.S. at 265, 89 S.Ct. at 1686)
A court-martial is tried, not by a jury of the defendant’s peers which must decide unanimously, but by a panel of officers empowered to act by a two-thirds vote. The presiding officer at a court-martial is not a judge whose objectivity and independence are protected by tenure and undiminishable salary and nurtured by the judicial tradition, but is a military law officer. Substantially different rules of evidence and procedure apply in military trials. Apart from those differences, the suggestion of the possibility of influence on the actions of the court-martial by the officer who convenes it, selects its members and the counsel on both sides, and who usually has direct command authority over its members is a pervasive one in military law, despite strenuous efforts to eliminate the danger. (395 U.S. at 263-264, 89 S.Ct. at 1686)
While the Court of Military Appeals takes cognizance of some constitutional rights of the accused who are court-martialed, courts-martial as an institution are singularly inept in dealing with the nice subleties of constitutional law. (395 U.S. at 265, 89 S.Ct. at 1687)
A civilian trial * * * is held in an atmosphere conducive to the protection of individual rights, while a military trial is marked by the age-old manifest destiny of retributive justice.7
As recently stated: “None of the travesties of justice perpetrated under the UCMJ is really very surprising, for military law has always been and continues to be primarily an instrument of discipline, not justice.” Glasser, Justice and Captain Levy, 12 Columbia Forum 46, 49 (1969). (395 at 266, 89 S.Ct. at 1687) 21
While commentators on both sides of the O’Callahan retroactivity issue have been critical of these statements,22 and other authorities have stated more fa*765vorable conclusions about the fundamental fairness of the system,23 we hasten to assert that, as an inferior tribunal, we have no prerogative and it is not our purpose to dispute any of O’Callahan’s language in the slightest degree. Our direct quotations here are solely for the purpose of demonstrating that, demeaning of military justice as these remarks may be, the opinion does not formulate its new constitutional restriction on congressional power for the purpose of preventing or undoing the conviction of innocent men. There was no determination that the UCMJ carried a clear danger of convicting the innocent, nor was it adjudicated that Congress had ordained a truth-determining process which lacked integrity or which was infected with procedures which substantially impaired the truth-finding function.
Civilian court bias which might tend to protect local citizens and their property from the troops was not put into the scales for comparison, nor was the worth of the military system tested at the general court-martial level where Gosa was tried. There the serviceman receives many procedural rights which are even more conducive to fact accuracy than most civilian forums accord.24 The Court has told us that the extent to which other safeguards are available is also a pertinent consideration. As the history of Gosa’s case amply demonstrates, general courts-martial receive several direct reviews of fact and law. The civilian staffed United States Court of Military Appeals, whose judges have fifteen-year tenure at a salary equal to our own, has both direct25 and habeas review power.26 Also, the federal courts have long been available for a collateral attack upon court-martial proceedings to secure basic constitutional guarantees and a full, fair hearing on all allegations raised.27
We conclude that O’Callahan ultimately decides no more on this subject than that there is a belief that a civilian court trial with grand and petit jury protections would tend to prevent arbitrariness and repression and be fairer. This belief is insufficient under DeStefano standards to warrant retroactivity if other criteria point strongly to prospective application. Indeed, if the military court system as a whole were procedurally deficient, the attack and the holding in O’Callahan would certainly have condemned such lack of procedural and substantive due process equally with jurisdiction. Otherwise, the decision itself would stand as a denial of equal protection to those it left included in O’Callahan and, more importantly, in Relford. The latter case, according to its note 14 (91 S.Ct. at 658), extended the jurisdictional reach of military courts to about 80% of those servicemen who might otherwise have been excluded if a narrower definition of O’Callahan service connection had been adopted.
(B) Justified Reliance on the Old Standard.
We have been unable to find any authority or comment indicating that O’Cal*766lahan was foreshadowed in other opinions28 To the contrary, Kinsella v. United States ex rel. Singleton, swpra n. 8, gave an interpretation to the power of Congress under the Constitution to constitute a system of military justice and to infuse it with jurisdiction which was clearly wide enough to encompass O’Callahan. The announcement there was:
The test for jurisdiction, it follows, is one of status, namely, whether the accused in the court-martial proceeding is a person who can be regarded as falling within the term ‘land and naval Forces’. (361 U.S. at 241, 80 S.Ct. at 301)
No more need be said to demonstrate that this criterion clearly favors prospectivity.
(C) Effect on the Administration of Justice of Retroactive Application.
Here is another point that is free from doubt. If O’Callahan is held to be entitled to full retroactivity, the impact upon military justice, and upon the federal court system too, will be veritably staggering. In response to the request of this Court, the Department of the Air Force has advised that its court-martial systems have processed 475,349 cases since 1949 and although the sheer bulk of analysis prevented a case by case examination, a two-year sampling indicated to the Solicitor General of the United States that 5% would constitute a reasonable working hypothesis of the number of cases that could raise a retroactive O’Callahan issue. This calculates to be 23,767 trials in this branch of the service alone.29 As the Court of Military Appeals has observed in Mercer v. Dillon, supra, peacetime court-martial jurisdiction over O’Callaharirtype cases has been in existence since 1916. That opinion also pointed out that in the year 1968 there were 74,000 special and general courts-martial in the Army, the Navy and the Air Force. With a 55-year history, which includes several years in which the Armed Forces were swollen by the manpower demands of two major “world” wars, the total number of cases involved may reasonably be expected to number in the hundreds of thousands. Out of these possibilities, the numbers which could present issues still subject to review can only be rankly conjectured because of the variety of issues that could be raised. Again alluding to the language of Mercer v. Dillon:
The range of relief could be extensive, involving such actions as determinations by the military departments of whether the character of discharges must be changed, and consideration of retroactive entitlement to pay, retired pay, pensions, compensation, and other veterans’ benefits. Among the difficulties would be the necessity of reconstructing the pay grade that a member of the armed forces would have attained except for the sentence of the invalidated court-martial, a task complicated by the existence of a personnel system involving selection of only the best qualified eligibles and providing for the elimination of others after specified years of service.30
Indeed, it seems to this Court that a major justification for resolving its acknowledged doubts as to the correctness of the decision in this case in favor of prospectivity is the tremendous effect which a holding of retroactivity could produce solely within this Circuit, and only within the brief interim between the time our decision was announced and the time the Supreme Court could finally determine the merits of the case. An *767erroneous determination of retrospectivity could inundate this already overloaded system and the military forums with claims that would ultimately have to be reversed or dismissed. On the other side of the coin, if we err in holding for prospective application, while we acknowledge that we surely have added burdens to those wrongly imprisoned or deprived of their rights, the dimensions of the error would be infinitely less. The relative consequences to the administration of justice clearly indicate that we should pursue the more cautious course.31
CONCLUSION
We have set down in detail some will doubtless deem unnecessary the processes by which we have reasoned our decision. We have done so because the issue with which we have dealt is one of the greatest moment, involving as it does not only Gosa’s freedom but potentially the freedom and property rights of many other citizens. If our reasoning is faulty, it is laid bare — its error will be plain. Using the best lights we are given, we believe it to be correct. We determine that the decision of the District Court that James Roy Gosa was not entitled to ha-beas corpus relief is correct and that decision is Affirmed.
. The convening authority referred the record to his Staff Judge Advocate for review as required by Article 61, U.C.M.J. (10 U.S.C.A. § 861). In what Gosa himself acknowledges was a comprehensive review of the record, the Staff Judge Advocate submitted his written opinion to the convening authority, that the findings and sentence of the general court-martial were correct in law and in fact. Subsequently the convening authority approved the findings and sentence as required by Article 64, U.C.M.J. (10 U.S.C.A. § 864), and pursuant to Article 65, U.C.M.J. (10 U.S.C.A. § 865) the convening authority followed this final action by sending the entire record, his action and the opinion of his Staff Judge Advocate to the Judge Advocate General of the United States Air Force where it was reviewed by a Board of Review pursuant to Article 66, U.C. M.J. (10 U.S.C.A. § 866). This Board affirmed both the finding of guilt and the sentence.
. 305 F.Supp. 1186 (N.D.Fla.1969) ; 19 U.S.C.M.A. 327, 41 C.M.R. 327 (March 20, 1970).
. 1. The serviceman’s proper absence from the base.
2. The crime’s commission away from the base.
3. Its commission at a place not under military control.
4. Its commission within our territorial limits and not in an occupied zone of a foreign country.
5. Its commission in peacetime and its being unrelated to authority stemming from the war power.
6. The absence of any connection between the defendant’s military duties and the crime.
7. The victim’s not being engaged in the performance of any duty relating to the military.
8. The presence and availability of a civilian court in which the case can bo prosecuted.
9. The absence of any flouting of military authority.
10. Tlie absence of any threat to a military post.
11. The absence of any violation of military property.
One might add still another factor implicit in the others :
12. The offense’s being among those traditionally prosecuted in civilian courts.
See also Articles 17 & 18, U.C.M. J. (10 U.S.C.A. §§ 817 & 818).
. Not only did Gosa’s alleged crime occur prior to O’Callahan, his conviction and sentence became final to the date of that decision; thus, there is no occasion for us to take any position on the issue of partial retroactivity accorded to the decision by the United States Court of Military Appeals which has, since the date of that decision, applied O’Callahan to void every court-martial action affecting a serviceman similarly situated whose conviction was still on direct review on June 2, 1969. Mercer v. Dillon, 19 U.S.C.M.A. 264, 41 C.M.R. 264 (March 6, 1970).
. In what must be considered dicta since the serviceman did not come within O’Callahan’s ambit, another district court also reasons against retroactivity without reaching this point. Thompson v. Parker, 308 F.Supp. 904 (M.D.Penn.1970).
. United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955).
. McElroy v. United States ex rel. Guag-liardo, 361 U.S. 281, 80 S.Ct. 305, 4 L.Ed.2d 282 (1960) ; Grisham v. Hagan, 361 U.S. 278, 80 S.Ct. 310, 4 L.Ed.2d 279 (1960).
. Reid v. Covert, 354 U.S. 1, 77 S.Ct. 1222, 1 L.Ed.2d 1148 (1957) ; Kinsella v. United States ex rel. Singleton, 361 U.S. 234, 80 S.Ct. 297, 4 L.Ed.2d 268 (1960).
. There is earlier, though non-judicial, precedent. In 1701, even before we had our present Constitution, the Colonial orator and patriot, James Otis, in his Argument Against the Writs of Assistance, stated: “An act against the Constitution is void; an act against natural equity is void.” See also Nelson and Westbrook, infra n. 11.
. See also Harrington v. United States, 444 F.2d 1190 (5th Cir. 1971).
. We note tlmt the opinions of commentators which have come to the attention of the Flemings court and to ours are in disagreement on this point. Nelson and Westbrook, Court-Martial Jurisdiction Over Servicemen for “Civilian Offenses”, an Analysis of O’Callalmn v. Parker, 54 Minn.L.Rev. 1, 45; Note: Court-Martial Jurisdiction Limited to “Service-Connected” Cases. 44 Tulane L.Rev. 417; and Note : Constitutional Law — Retroactivity —Should O’Callahan be Applied Retroactively?, 2 Tex.Tech.L.Rev. 106; all three reason that the structure of the opinion in terms of jurisdictional deficit does not preclude testing the rule’s application by the retroactivity standards applied to other new applications of constitutional fundamentals and that the statements critical of the fairness of military justice do not overcome the persuasiveness of the reliance and effect criteria of the Supreme Court’s retroactivity rulings discussed infra. Whereas, Note: Denial of Military Jurisdiction over Servicemen’s Crimes Having No Military Significance and Cognizable in Civilian Courts, 64 Nw.U.L.Rev. 930, and Note: O’Callahan v. Parker, A Military Jurisdictional Dilemma, 22 Baylor L.Rev. 64, make the ipse dixit assertion that the jurisdictional tenor of the opinion obviously requires retroactivity.
. The only other Supreme Court eases which our research discloses have dealt with the issue, even tangentially, are Fuller v. Alaska, 393 U.S. 80, 89 S.Ct. 61, 21 L.Ed.2d 212 (1968), which refused retroactive application to Lee v. Florida, 392 U.S. 378, 88 S.Ct. 2096, 20 L.Ed. 1166 (1968), forbidding State use of evidence obtained in violation of the Federal Communications Act; North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969), which gave retroactive application to Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969), declaring the Fifth Amendment’s double jeopardy protections applicable to State criminal procedures. [See Ashe v. Swenson, 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970)3; DeBacker v. Brainard, 396 U.S. 28, 90 S.Ct. 163, 24 L.Ed.2d 148 (1969), which applied the non-retroactive holding of DeStefano v. Woods to a minor who had been denied the right of trial by jury; and the other *760cases decided the same day Williams was handed down which are discussed infra.
. See Mr. Justice Harlan’s dissenting opinion in Williams, supra, 401 U.S. at 676, 91 S.Ct. at 1172, 28 L.Ed.2d at 411 (1971), in which he observed : “ * * * Linkletter became almost as difficult to follow as tlio tracks of a beast of prey in search of his intended victim.”
. Which viewed Escobedo and Miranda as containing Fifth rather than Sixth Amendment rights.
. Three years later Mr. Justice Warren also delivered the opinion of the Court in Jenkins v. Delaware, 395 U.S, 213, 89 S.Ct. 1077, 23 L.Ed.2d 253 (1909), which denied the application of Miranda exclusion standards to evidence that had been secured prior to the date that de-cisión was announced in retrials which occurred subsequent to the date of Miranda.. This opinion places strong emphasis on the reliance criterion,
. Justices Brennan and Marshall concurred in the denial of retroactivity in separate opinions. Mr. Justice Black concurred in the result, but on the ground that Chimel had been wrongly decided. Mr. Justice Harlan filed a lengthy and thought-provoking dissent in . Williams and Elkanich and a doubtful concurrence in Mackey, infra.
. Again the Court was badly split in its reasoning. Considering that matters of procedure rather than substance were involved, Mr. Justice Harlan concurred. Justices Brennan and Marshall concluded that the bar of the Fifth Amendment did not extend to the enforcement of income tax laws applicable to those in the business of accepting wagers, and that Marchetti and Grosso could be distinguished. Justices Black and Douglas dissented. See also United States v. Scaglione, 446 F.2d 182, 5th Cir. 1971.
. On the same day, Hill v. California, 401 U.S. 797, 91 S.Ct. 1106, 28 L.Ed.2d 484 (1971), involving the retroactivity of Ghimel, was decided. It was not expressly reasoned but merely relied on Williams, supra. That day also, the Court handed down United States v. White, 401 U.S. 745, 91 S.Ct. 1122, 28 L.Ed.2d 453 (1971), adhering to its decision of nonretroactivity announced in Desist v. United States, 394 U.S. 244, 89 S.Ct. 1030, 22 L.Ed.2d 248 (1969). This case likewise did not contain an explication of principles.
. The record before us does not disclose and no claim is advanced that Gosa’s military trial was held away from the vicinage of the crime or that he was in any way impeded in securing other constitutional protections. We, therefore, do not reach these issues which were discussed in Flemings, supra.
. A procedure available under 28 F.S.C.A. § 1254(3) (1966).
. Footnote 7 in this quote refers to the entire system ns one of “so-called military justice”.
. See e. g., Nelson and Westbrook, supra, n. 11, in favor of prospectivity, which states:
Considered as a whole, the majority opinion is persuasive only to those who were already persuaded. Faced with a problem involving many complex variables and requiring the delicate balancing of competing interests, the Court responds with dogmatic assertions about military justice. (55 Minn.L.Rev. 1, 63)
while the comment in Baylor L.Rev., supra, n. 1, which opts for retroactivity, observes:
Through a theme of the military as historical bandits of freedom for its *765members, the Court sets the tone for its construction.
(22 Baylor L.Bov. at 67)
. See Mercer v. Dillon, supra; Chief Justice (then Judge) Burger dissenting in United States ex rel. Guagliardo & McElroy, 104 U.S.App.D.C. 112, 259 F.2d 927, 940, referred to the USMJ as having received universal recognition as “affording the basic elements of fairness” ; Chief Justice Warren, The Bill of Bights and the Military, 37 N.Y.U.L.Rev. 181, 188-189; Quinn, Some Comparisons Between Courts-Martial and Civilian Practice, 15 U.C.L.A.L.Rev. 1240; Everett, O’Callalian v. Parker, Milestone or Millstone in Military Justice, Duke L.J. 853; see also Senator Ervin’s anthology of significant instances in which courts-martial accorded procedural protections before civilian courts, 115 Cong.Rec. S 7174, 7145 (June 25, 1969).
. See Quinn, 15 U.C.L.A.L.Rev. 1240, supra, n. 23 and Comment, 22 Baylor L.Rev. 64, supra, n. 11.
. Art. 67 U.C.M.J. (10 U.S.C.A. § S67).
. See Levy v. Resor, 17 U.S.C.M.A. 135, 37 C.M.R. 399 (1967).
. Burns v. Wilson, 346 U.S. 137, 73 S.Ct. 1045, 97 L.Ed. 1508 (1953). See also Noyd v. Bond, 395 U.S. 683, 89 S.Ct. 1876, 23 L.Ed.2d 631 (1969).
. Cf. Roberts v. Russell, 392 U.S. 293, 88 S.Ct. 1921, 20 L.Ed.2d 1100 (1968).
. Doubtless this has been somewhat restricted by Relford.
. That same court also observed: In many of the courts-martial of earlier years, jurisdictional facts could have been developed on the record if there had been any reason to predict the need for doing so. The practical effect of voiding earlier convictions will often be to grant immunity from prosecution as a result of State statutes of Imitations having run> witnesses having been scattered, and memories having been taxed beyond per. missible limits.
. Judge Godbold’s dissent suggests a middle ground tactic wliicli would opt for retroactivity, then stay our holding pending High Court review. Such a course would have little if any ameliorating effect upon the anticipated impact of such a circuit ruling on the district courts, who would bear the entire brunt anyway during the relatively brief period between our decision and the Supremo Court’s ultimate determination.