(with whom MOORE, Circuit Judge, concurs), dissenting :
While I agree with Judge KAUFMAN that Gideon v. Wainwright requires that defendants in state felony proceedings be advised of their right to counsel whether they plead guilty or stand trial, I cannot agree that that principle must or should be applied now to the pleas of guilty entex’ed by Ripple in 1931, by Moore in 1943, by Dxxrocher in 1950, and by Brown in 1954.
By its action in Doughty, the Supreme Coux't did look back to 1959. But the Court gave no reasons in support of its summary 22-wox-d reversal. Until the Court specifically invalidates pleas of guilty entered prior to 1959, it seems to me that we should not look backwards to any earlier year. There are compelling reasons why we should not subject the states to the hazards of reprosecuting cases where the events in question are more than five years old.
*315Gideon clearly establishes that states must now provide counsel to indigent defendants faced with felony charges unless such defendants knowingly waive their right to counsel. Failure to provide counsel, in the absence of such a waiver, will invalidate any conviction. Thus Gideon will have served its primary purpose in assuring indigent defendants the aid of counsel from this time henceforth.
What then do the federal courts accomplish if they choose to apply Gideon retroactively without limit of time? Until the Gideon decision of March 18, 1963, the states operated under the rule of Betts v. Brady, 316 U.S. 455, 62 S.Ct. 1252, 86 L.Ed. 1595 (1942), that state convictions in non-capital cases were invulnerable to attack on the ground that the defendant had not been represented by counsel unless he could establish that lack of counsel had resulted in fundamental unfairness.
While to many of us the continued failure of the states to provide counsel to the indigent defendant in felony cases seemed a black chapter in the history of American law1 —and as time goes by will seem even more inexplicable — there were many reasonable men who thought that the states were doing substantial justice and, in any event, that the federal courts should intervene only upon a positive showing that the defendant had been unfairly treated.
The wonder is that Gideon did not come sooner. The question must have been raised in scores of the many hundred pro se applications from state prisoners which deluged the Supreme Court and other federal courts in the years prior to March 1963. By 1963 there were in the prisons of most of the fifty states thousands of defendants who had pleaded guilty without offer of counsel and without benefit of counsel.
But the fact that some change in the law was necessary and indeed long overdue does not justify in itself the unlimited retroactive application of that change when finally effected. Throughout this country considerably more than 80 per cent of those accused of serious crimes plead guilty whether or not represented by counsel. Thus, there can be no doubt that state prisoners serving sentences imposed after pleas of guilty did, in fact, commit the crimes for which they are now imprisoned as they have admitted this by their plea of guilty. The petitions of Ripple, Moore, Durocher, and Brown maké it crystal clear that they had in fact committed the acts as charged in the indictment to which they pleaded guilty. While counsel might have advised or negotiated some mitigation of plea or sentence, an overwhelming percentage of unrepresented indigent defendants would have borne the stigma of conviction in any event; assistance of counsel could have resulted at best in only a very small percentage of acquittals or dismissals.
Unlimited retroactive application of Gideon would thus compel the release of thousands of felons who can no longer be tried because witnesses and evidence are no longer available. Indeed it is the more hardened criminals who profit most by unlimited retroactivity, for they are still serving lengthy sentences imposed long ago at a time when the states could properly assume that fundamental fairness was the applicable test. As to the overwhelming majority of such prisoners there can be no doubt that they received substantially fair treatment. The majority opinion would not only let them out of jail but it would permit them to have the slate wiped clean by reason of the passage of time which renders the state unable now to prove its case. To require the release of thousands of state prisoners who undoubtedly are guilty of crimes serious enough to justify long terms of imprisonment strikes a severe blow to enforcement of law and the administration of criminal justice.
On the other hand, were we to limit the retroactive effect of Gideon and Doughty to the period of time neces*316sarily comprehended in those cases, the states might still prosecute most of such defendants and bring them to trial if new pleas of guilty were not forthcoming. Fortunately, or not, the Supreme Court has chosen to act in cases where the criminal charges are relatively recent in point of time, much more so than in the cases before us. Gideon was tried in 1961 on charges that he had committed an offense earlier that same year. The state did in fact retry him, albeit unsuccessfully. The charge against Doughty was for rape allegedly committed in 1958, and his plea of guilty was entered in 1959.
We do not know whether the Supreme Court thus meant to leave open the question of retrospective application to convictions prior to 1959. While we must now go as far as that Court went, so that other defendants who were unrepresented as were Gideon and Doughty are not denied the equal protection of those decisions, it seems to me that retrospective effect prior to 1959 is fraught with so many unfavorable consequences which far outweigh the little good which might possibly flow from such action that I would do no more than what the Supreme Court has already chosen to do in these two eases.
We have been admonished many times that federal courts ought not to interfere with the administration of criminal justice by the states. Mr. Justice Harlan, writing for a majority of the Supreme Court, said in Hoag v. New Jersey, 356 U.S. 464 at 468, 78 S.Ct. 829 at 833, 2 L.Ed.2d 913 (1958): “For it has long been recognized as the very essence of our federalism that the States should have the widest latitude in the administration of their own systems of criminal justice.” See also Twining v. New Jersey, 211 U.S. 78, 29 S.Ct. 14, 53 L.Ed. 97 (1908). Surely any federal court of less stature than the Supreme Court ought not go one step further than is necessary to carry out the plain and expressed mandate of that Court where to do so will certainly amount to a wholesale jail delivery from many state prisons in this country. At the very least it would seem that the Court has left open the question of how far it will go into the past to give retrospective effect to Gideon and Doughty. See Justice Harlan’s dissent in Pickelsimer v. Wainwright, 375 U.S. 2, 3, 84 S.Ct. 80, 11 L.Ed.2d 41 (1963). And no one would doubt the power of the Court to say how far back into the past a newly made rule, even one declaring constitutional rights, must be applied. See Griffin v. Illinois, 351 U.S. 12, 25-26, 76 S.Ct. 585, 100 L.Ed. 891 (1956) (Frankfurter, J., concurring); Bender, Retroactive Effect of an Overruling Constitutional Decision: Mapp v. Ohio, 110 U.Pa.L.Rev. 650 (1962) ; Traynor, Mapp v. Ohio at Large in the Fifty States, 1962 Duke L. J. 319, 338-42; Note, Prospective Overruling and Retroactive Application in the Federal Courts, 71 Yale L.J. 907, 909-33 (1962); Snyder, Retrospective Operation of Overruling Decisions, 35 Ill.L.Rev. 121 (1940).
For the above reasons I would affirm the orders of the district court in all four cases.
. The writer expressed this opinion in 1952. 37 Cornell Law Quarterly 555.