This case and two other cases before this court1 raise important questions regarding a criminal defendant’s constitutional right to a speedy trial.2 We had occasion recently to deal extensively with the content of that right in Glasgow v. State, 469 P.2d 682 (Alaska 1970), and we are now called upon to clarify and expand upon the principles announced therein.

I.FACTUAL BACKGROUND

The relevant facts in this case are as follows: On April 23, 1969, five days after the death of her husband, appellant herein was arrested, charged with first degree murder, and released on bail. Following a preliminary hearing, appellant was bound over to the grand jury on June 23, 1969, on the charge of second degree murder. A month later, on July 24, 1969, the grand jury returned an indictment on that charge. Appellant was arraigned July 29, 1969, and on August 12, 1969, she entered a not guilty plea.

The case remained completely dormant from August 12, 1969, until June 9, 1970, when the superior court notified appellant’s counsel by letter that trial would be held July 6, 1970.3 The latter date was fourteen months and thirteen days after the original complaint, arrest and arraignment. Appellant’s subsequent motion to dismiss the indictment with prejudice based on her constitutional right to a speedy trial was denied after a hearing on July 2, 1970. It is this ruling which appellant seeks to have reviewed here.

At the trial, appellant was convicted of involuntary manslaughter, given a five-year suspended sentence, and placed on probation for a period of five years.

II. DETERMINATION OF SPEEDY TRIAL VIOLATION

The speedy trial guarantee has been seen as fulfilling at least three purposes: (1) to prevent harming the defendant by a weakening of his case as evidence and memories of witnesses grow stale with the passage of time; (2) to prevent prolonged pre-trial incarceration; and (3) to limit the infliction of anxiety upon the accused because of long-standing charges.4 In postulating speedy trial standards, it is *948necessary to be cognizant of each of these purposes.

In Alaska, contrary to many states, we have no statutory provisions which implement speedy trial guarantees by setting a definite period within which trial must be held.3 Whether a speedy trial violation has occurred depends on the facts of each case and general constitutional principles.

In Spight v. State, 450 P.2d 157, 158 (Alaska 1969), we drew upon Medina v. People, 154 Colo. 4, 387 P.2d 733 (1963), cert. denied, 379 U.S. 848, 85 S.Ct. 88, 13 L.Ed.2d 52 (1964), in deciding that speedy trial is a “relative concept in that the circumstances of each case determine whether it had been afforded an accused.” 387 P. 2d at 736.

In Spight, the defendant had requested several continuances, which were granted, but later continuances were granted at the request of the prosecution because of the inability to locate a witness. The total delay due to the prosecution’s own requests, and the granting of a mistrial, amounted to about three months. We held that since appellant had made no showing that the delay was “purposeful or oppressive or that [it] prejudiced her defense in any manner. * * * [a]ppellant’s constitutional right to a speedy trial was not violated.”

In Glasgow v. State, 469 P.2d 682 (Alaska 1970), this court again balanced society’s interest in the conviction of those who violate its criminal laws against a defendant’s constitutional interest in a speedy trial. We found that a delay of approximately fourteen months attributable to the state was excessive:

We find this to be an excessive amount of time within which to bring an accused to trial and, as such, a deprivation of the constitutional right to a speedy trial. In arriving at this conclusion, we rely on the persuasive reasoning of [the opinions of the United States Supreme Court in Smith v. Hooey, 393 U.S. 374, 89 S.Ct. 575, 21 L.Ed.2d 607 (1969) and Klopfer v. North Carolina, 386 U.S. 213, 87 S.Ct. 988, 18 L.Ed.2d 1 (1967)] and on our interpretation of section 11, article 1 of the Alaska Constitution.

It should be noted that in maintaining a proper balance between the needs of the accused and the commitments of the judicial process, we approve the statement of countervailing principles as set forth in United States v. Ewell * * * 6

We then went on to quote the following language from the Ewell opinion, 383 U.S. 116, 120, 86 S.Ct. 773, 776, 15 L.Ed.2d 627, 630-631 (1966):

[B]ecause of the many procedural safeguards provided an accused, the ordinary procedures for criminal prosecution are designed to move at a deliberate pace. A requirement of unreasonable speed would have a deleterious effect both upon the rights of the accused and upon the ability of society to protect itself. * * * ‘[T]he essential ingredient is orderly expedition and not mere speed.’ Smith v. United States, 360 U.S. 1, 10, 79 S.Ct. 991, 997, 3 L.Ed.2d 1041 [1048],

We held, however, that a fourteen-month delay was an improper balance to strike.

With this framework of purpose and countervailing interests as background, we now turn to the question of what factors, other than the length of delay, are relevant in passing on a motion to dismiss based on an asserted speedy trial violation.

*949The state argues that the absence of proof of demand for trial, lack of prosecuA torial fault,7 and no showing of actual prejudice should be determinative here. However, our opinion in Glasgow v. State, supra, casts serious doubt on the relevancy of any of these criteria where lengthy delay has occurred:

The degree to which the earlier case law on speedy trial is still sound has been thrown into some doubt by the Klopfer and Hooey decisions [supra]. Both stress the burden resting upon the state to bring the accused promptly to trial. Neither case resorts to undue technicalities about the necessity of showing prejudice to the defense, nor are waiver doctrines apparently relied upon.

As a result of Hooey and Klopfer it is apparent that a positive duty rests upon the courts and prosecutors to see that criminal defendants are afforded a speedy trial. The requirement of a demand, the showing of prejudicial harm to the defendant’s case, and the implication of a waiver from the defendant’s failure to make a demand arguably are no longer determinative of the claim to a speedy trial. While the presence of a demand or a showing of prejudice to one’s case can only help the claim, their absence alone will not necessarily frustrate the right to a speedy trial, including the right to a dismissal of the charges with prejudice when there has been a clear denial of this constitutional right. We reach this conclusion on the basis of our interpretation of article I, section 11, of the Alaska Constitution rather than upon any dispositive holding in Hooey and Klopfer. * * * (footnotes omitted.) 8

Additional discussion may be necessary, however, both in order to further clarify our holding in Glasgow and to answer the state’s request that we recede from the above portion of the Glasgow opinion.

A. Waiver

Our decision in Glasgow to reject the doctrine that failure of an accused to demand a speedy trial constitutes a waiver was based on a growing body of legal thought which holds that waiver of constitutional rights can ordinarily only be done knowingly and intentionally.9 In this regard, the position taken by the Supreme Court of the United States is particularly illustrative.

The United States Supreme Court in Johnson v. Zerbst, 304 U.S. 458, 464, 58 S. Ct. 1019, 1023, 82 L.Ed. 1461, 1466 (1938), defined “waiver” as “an intentional relinquishment or abandonment of a known right or privilege.” The Supreme Court further stated that every presumption should be indulged against waiver, Aetna Ins. Co. v. Kennedy, 301 U.S. 389, 393, 57 S.Ct. 809, 81 L.Ed. 1177, 1180 (1937), and that the Court would not presume acquies*950cence in the loss of fundamental rights. Ohio Bell Telephone Co. v. Public Utilities Comm’n, 301 U.S. 292, 307, 57 S.Ct. 724, 81 L.Ed. 1093, 1103 (1937).

Against this backdrop, the cases of Escobedo v. Illinois, 378 U.S. 478, 84 S.Ct. 1758, 12 L.Ed.2d 977 (1964); Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L. Ed.2d 694 (1966), and Stovall v. Denno, 388 U.S. 293, 87 S.Ct. 1967, 18 L.Ed.2d 1199 (1967), were decided and held that the prosecution had a duty to inform the accused of particular rights, and that there could be no waiver in the absence of an explanation of those rights. There is respectable authority in the federal courts that waiver of a right to a speedy trial is similarly not to be inferred from mere silence unless the silence is for some tactical purpose.10

Many of these authorities were alluded to in Glasgow and provide a sound foundation for the opinion. The state argues, however, that another factor should be considered in weighing the effect of silence. The failure to demand a speedy trial can be a tactical ploy by a guilty defendant. Most defense counsel are aware that delay may be beneficial. Witnesses, memories, etc., may be lost, and the case may be forced from the judicial process and justice thwarted.

However, we feel it is too much to expect of human nature that a defendant must demand a speedy trial to receive one when he has no assurance as to the outcome of trial. The uncertainty as to the verdict can bring indecision to the innocent as well as the guilty. While trial should vindicate the innocent, such a defendant may nevertheless be apprehensive about the outcome, for even his innocence has not prevented him from reaching the trial stage of the criminal process. Thus, the innocent defendant is presented with two unhappy alternatives: either not to demand trial, hoping the state will not prosecute, or to demand trial, hoping to vindicate himself in spite of his fear of being found guilty.11

Furthermore, under our system of criminal justice, it is the prosecution which initiates a case and which has the power of going forward with it. In the exercise of this power, it is the duty of the public prosecutor to observe the constitution. To condition that duty on a defendant’s having demanded constitutional compliance is, in our view, an unacceptable misallocation of the burden of insuring a speedy trial. We, therefore, reaffirm the position announced in Glasgow that the burden is upon the state to give a speedy trial or be denied the power to prosecute.12

B. Cause of the Delay

The state, while admitting that a “congested court calendar is not the fault of this appellant,” argues that since this is “not an example of a case where delay has been obtained by the prosecution,” there is no violation of appellant’s right to a speedy trial. We disagree.

We are unable to see how this offers any comfort to appellant, who suffers the same anxiety and lessened ability to defend herself either way. Why there should be a different rule for harm imper*951sonally inflicted as opposed to harm directly attributable to the prosecution is unexplained in logic. The result is the same in both cases. It makes no difference what causes the loss of constitutional rights to a defendant so long as they are gone without his assent.

C. Prejudicial Effect of the Delay

While appellant was not incarcerated during the period between arrest and trial and may not be able to prove actual harm to her ability to defend herself, this does not mean that she suffered no injury because of the delay. The appellant was charged with the most serious of our crimes, murder. The public humiliation and scorn attended with such a charge are of substantial consequence. The difficulty of maintaining normal patterns of life under such conditions is readily apparent.13 The need to explain such a charge to all but a very few makes normal public expressions and circulation virtually impossible. One must virtually wait for trial to become truly free again.

The entire focus of Klopfer v. North Carolina, supra, is based upon the public humiliation and scorn to which an indicted person is subjected during the delay in bringing him to trial:

We, too, believe that the position taken by the court below was erroneous. The petitioner is not relieved of the limitations placed upon his liberty by this prosecution merely because its suspension permits him to go ‘whithersoever he will.’ The pendency of the indictment may subject him to public scorn and deprive him of employment, and almost certainly will force curtailment of his speech, associations and participation in unpopular causes. By indefinitely prolonging this oppression, as well as the ‘anxiety and concern accompanying public accusation,’ the criminal procedure condoned in this case by the Supreme Court of North Carolina clearly denies the petitioner the right to a speedy trial which we hold is guaranteed to him by the Sixth Amendment of the Constitution of the United States.

386 U.S. at 221-222, 87 S.Ct. at 992, 18 L. Ed.2d at 6-7 (footnote omitted).

A very real problem is also created with the necessity of marshaling the facts and maintaining the memories of witnesses during the period of the delay. At what point do day-by-day events encroach upon memory to erase the subtle shades of meaning which may make the differences between the various degrees of criminal offenses? There are too many uncertainties to answer such a broad question. However, the potential for substantial prejudice is • undeniably present despite the difficulties in proving actual harm.

Since a basic purpose of the speedy trial guarantee is to insure that the verdict returned represents the justice of the factual' situation, any delay which results in a-distortion of the facts presented is undesirable. Extensive delays threaten the basis of the trial itself and the integrity of the verdict.

We therefore hold that in a case where the time between the charge14 and *952the trial is as long as in this case,15 we shall presume prejudice to the accused.16

The delay in bringing appellant to trial here exceeded the amount of time condemned in Glasgow as excessive. We again announce that such a delay is an improper balance to strike and hold that the appellant’s right to a speedy trial under the Alaska Constitution has been violated.

III. RETROACTIVE APPLICATION OF GLASGOW17

In the recent case of Judd v. State, 482 P.2d 273 (Alaska 1971), we rejected the notion that all new constitutional interpretations must be considered to have always been the law, and that they must therefore necessarily be given complete retroactive effect. That rejection was not original with this court.18 Nor were the criteria we adopted as guideposts in the analysis of retroactivity questions jurispru-dentially unique:

The criteria guiding resolution of the question of retroactivity which apparently have been agreed upon by all authorities are as follows: (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.19

We are again called upon to apply these criteria to a decision of this court announcing a new constitutional doctrine. We hold that Glasgow v. State, supra, must be given effect in this case.

A review of the decisions of the Supreme Court of the United States dealing with retroactivity questions indicates that the starting point in analysis is the purpose criterion. Where the purpose of the new rule is primarily related to the integrity of the verdict, the application thereof has generally been extended to all cases.20 This result was recently noted by Justice White writing for a plurality of the Court in Williams v. United States, 401 U.S. 646, 91 S.Ct. 1148, 1152, 28 L.Ed.2d 388, 397 (1971) :

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 *953reliance 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.21

On the other hand, where the purpose of a new constitutional standard is not to minimize arbitrary or unreliable fact findings, but to serve other ends, retroactive application has generally been denied.22 Thus, in Judd v. State, supra, we held that Chimel v. California, 395 U.S. 752, 89 S.Ct. 2034, 23 L.Ed.2d 685 (1969), and Fresneda v. State, 458 P.2d 134 (Alaska 1969), which limited the permissible scope of search incident to arrest, would not be given retroactive effect. The problem in search and seizure cases is not the reliability of the evidence obtained, but rather the constitutional propriety of the methods used by the investigating authority to obtain it.

Difficult analytical problems arise, however, and necessitate reference to the other two criteria, because with many new constitutional doctrines, “[t]he extent to which a condemned practice infects the integrity of the truth-determining process at trial is a ‘question of probabilities.’ ” Stovall v. Denno, 388 U.S. 293, 298, 87 S.Ct. 1967, 1970, 18 L.Ed.2d 1199, 1204 (1967), quoting Johnson v. New Jersey, 384 U.S. 719, 729, 86 S.Ct. 1772, 16 L.Ed.2d 882, 890 (1966). It is these penumbral cases, which only marginally or secondarily concern the integrity of the fact-finding process, which require a careful balancing analysis.23

*954The right to a speedy trial serves several purposes,24 only one of which is concerned with the accuracy of verdicts in criminal trials. The existence of these other purposes, however, in no way diminishes the importance of speedy trial as a guarantee of basic procedural fairness to the accused. For protracted delay in bringing a criminal case to trial may substantially impinge upon the reliability of the fact-finding process. The purpose of our holding on the issue of speedy trial in Glasgow is thus significantly, though not exclusively, related to the integrity of the fact-finding process. Under these circumstances, it is appropriate to consider in balance the remaining criteria which bear on the question of retroactivity. We must weigh against the purpose criterion the extent of reliance on pre-Glasgow law by law enforcement authorities, and, further, the probable effect on the administration of justice of a retroactive application of Glasgow. We find neither of these factors countervailing.

The extent of reliance on this court’s decision in Goss v. State, 390 P.2d 220 (Alaska 1964), where we held the right of an accused to speedy trial may be waived by failure to assert it, overruled by Glasgow, is somewhat difficult to analyze. The state has advanced no reason for the delay in bringing the appellant to trial from which reliance could be inferred. Indeed, it is difficult to reach any other conclusion than that the delay here was caused by the indifference of the superior court as to whether the state or the defendant desired a speedy trial. The state concedes that in most cases in which indictments had been returned during this period it moved for speedy trial because the court would not itself set criminal cases for trial.25 While official reliance on a prior decision is a valid consideration in deciding whether a new rule is to be applied retroactively, mere indifference as to whether an individual is afforded his constitutional right is not.

Moreover, reliance upon the holding in Goss would be a valid consideration only to the extent that such reliance was justified. Prior to our decision in Glasgow, the United States Supreme Court, ruling in Klopfer *955and Hooey, supra, substantially undermined the theoretical validity of the Goss holding. Although Glasgow was decided on state constitutional grounds, we relied heavily upon the language of Klopfer and Hooey. In this sense, the United States Supreme Court decisions laid the groundwork for Glasgow, and, at the same time, cast substantial doubt upon the continuing vitality of our holding in Goss. It is further appropriate to note the general trend in judicial decisions rejecting the doctrine of waiver of constitutional rights by silence.26 Thus, it is simply difficult to see how any actual reliance upon Goss could be considered to have been justifiable.

The effect of a retroactive application of Glasgow in this case on the administration of justice is difficult to guage. Since the remedy we selected in Glasgow was dismissal of the indictment with prejudice, repros-ecution will be barred herein. However, we are simply not persuaded that the effect of the application of Glasgow in this case would be to empty the jails or otherwise be “so devastating as to need no elaboration.” Tehan v. United States ex rel. Shott, 382 U.S. 406, 419, 86 S.Ct. 459, 467, 15 L.Ed.2d 453, 461 (1966). 27

It must be remembered that the rights we protect today are those of the innocent as well as the guilty. While the immediate benefit of our decision may appear to inure to the guilty, we cannot say that justice .'will suffer if, by upholding the constitutional right 6f all persons to a speedy trial, we would protect those who may be guilty in order to insure the rights of those who are innocent. “The constitutional requirement that guilt in criminal cases be proven beyond a reasonable doubt serves to limit, but cannot eliminate, the number of criminal defendants found guilty who are in fact innocent.” Williams v. United States, 401 U.S. 646, 91 S.Ct. 1148, 28 L.Ed.2d 388, 402 (1971) (Brennan, J., concurring). Where the defendant has asserted his innocence by pleading not guilty, and has subsequently been found guilty at a trial whose integrity is now questioned, we are not in a position to attack the legitimacy or sincerity of the claim of innocence.

One of the cornerstones of our legal system is that the law assumes the innocence of all persons charged with crime until such time as they are convicted in a manner consistent with the principles embodied in our constitutional system. And without speedy trial, neither this court nor the state is able to establish with the certainty required by our law that the verdict is just.

The judgment and conviction of the superior court are reversed, and the case is remanded to the superior court with in*956structions to dismiss the indictment herein with prejudice.

joining, dissenting in part.

While the extent of the majority holding is unclear, it applies Glasgow v. State, 469 P.2d 682 (Alaska 1970), retroactively, at least to the case at bar. Although the majority would leave the matter thus — in apparent uncertainty — I believe that in fairness to all concerned this court should take a position on the extent of Glasgow’s retroactive application. It is for this reason that although I concur in Parts I and II of the majority opinion, I cannot subscribe to its retroactive application of Glasgow and must, therefore, dissent from Part III and the court’s judgment.

In deciding whether a decision of this court will be applied retroactively, three factors must be weighed: “(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.” Judd v. State, 482 P.2d 273, 278 (Alaska 1971). On this my brothers and I are agreed. It is only in the use of these criteria that we differ.

I

The standards announced in Glasgow comprehend several purposes. The integrity of the truth-finding process is only one of these purposes, and it is hardly the paramount consideration in enforcing the right to speedy trial.1 Other important goals are the prevention of prolonged pretrial incarceration and the minimizing of the anxiety and insecurity inflicted upon accused persons by long-standing charges.2 An additional purpose, not mentioned in Glasgow as such, is to check oppressive action by public officials which may occur through deliberately withholding a criminal trial. Of these purposes, only one could relate to the integrity of the conviction, and even then not in every case.

A failure to honor the other purposes underlying the speedy trial guarantee may be deleterious on distinct public policy bases, but a conviction obtained after such a tardy trial may be just as relible as one which occurs through prompt adjudication.

A lack of speedy trial does not in all cases affect the reliability of the conviction. In some cases, of course, it may have that effect. In numerous cases it would be impossible to determine. In other cases the critical evidence of guilt may consist of documentary evidence, objectively developed scientific evidence, or circumstantial evidence which has been amassed through meticulous procedures. Even the direct testimony of witnesses may be verifiable because their memories were recorded contemporaneously with the observed events. In all of these cases the mere lapse of time is of no discernible importance.

The possibility of faded memory is, therefore, only one of the many aspects presented by the lack of a speedy trial. In some cases the time of trial may have no appreciable effect upon the fairness of the truth-finding process. In still other instances delay may actually work in favor of the accused, not to his detriment.

It should be remembered that the state must prove its case beyond a reasonable doubt. Delay of trial may not only weaken memory, it may render witnesses less appealing, harder to locate, and less inclined to implicate the accused. For these reasons, defendants frequently find it advan*957tageous to accede to prosecutorial delay. Nor is the accused entirely without means for protecting his case from the effects of delay. He and his counsel can interview witnesses and otherwise gather evidence while the charges are still fresh.

In short, it cannot safely be assumed that mere lapse of time will necessarily render convictions unreliable. This is particularly true in the case before us, where little over a year elapsed between charge and trial. In countless similar cases convictions have been obtained without the suggestion being advanced that delay alone rendered the integrity of the truth-finding process questionable.

II

Reliance by law enforcement officials and the judiciary on pre-Glasgow doctrine was probably much greater than the majority opinion implies. One examining the authorities might well have concluded, in good faith, that Goss v. State, 390 P.2d 220 (Alaska 1964), and the principles set forth therein were still good law. Goss was, of course, the law of this state until it was overruled in relevant part by Glasgow on May 29, 1970. Moreover, under Knudsen v. City of Anchorage, 358 P.2d 375, 377-380 (Alaska 1960) it was reasonable to conclude that this court felt itself bound in interpreting Article I, Section 11 of the Alaska Constitution 3 to give an accused no greater rights than had previously been given by the United States Supreme Court in construing the Sixth Amendment to the Federal Constitution.4 This holding of Knudsen was likewise overruled in Glasgow.5

Some eight months previously this court had partially overruled Knudsen in holding a defendant had the right to the presence of his counsel while police took a handwriting sample from him, a step the United States Supreme Court had been unwilling to take. Roberts v. State, 458 P.2d 340, 342 (Alaska 1969).6 Yet, considering what had been the peculiarly dormant nature of the right to speedy trial, even a particularly perceptive legal scholar might well have failed to anticipate the degree to which this court would make the right to speedy trial more meaningful in the Glasgow decision. One might still have believed Goss and the discernible federal authorities binding. Moreover, the language in Roberts is permissive, not mandatory. The opinion merely holds that we may, not that we necessarily will, employ more rigorous standards than required by the United States Supreme Court.

The right to speedy trial had received scant attention from the United States Supreme Court prior to Klopfer v. North Carolina, 386 U.S. 213, 87 S.Ct. 988, 18 L. Ed.2d 1 (1967).7 As Mr. Justice Brennan *958stated recently, “[W]e have yet even to trace its contours.” Dickey v. Florida, 398 U.S. 30, 40-41, 90 S.Ct. 1564, 26 L.Ed.2d 26, 34 (1970) (concurring opinion). Klop-fer, which in 1967 for the first time applied the right to speedy trial through the Due Process Clause of the Fourteenth Amendment to the states, concerned the constitutionality of a particular application of an unusual North Carolina criminal procedural device denominated a “nolle prose-qui with leave.” 8 In this state there is no procedure even remotely analogous. The only other case decided by the United States Supreme Court prior to Glasgow was Smith v. Hooey, 393 U.S. 374, 89 S.Ct. 575, 21 L.Ed.2d 607 (1969). That case concerned a six-year delay during which the defendant, imprisoned in another state, made repeated demands to be brought to trial in the forum state. The facts of that case were sufficiently unlike those in Glasgow as to provide little guidance for judges or counsel. It is only by looking to the spirit of Klopfer, Smith v. Hooey, and other significant constitutional decisions, discussed in Glasgow, that one can piece together the various sources which underlie the Glasgow decision.

One examining the pre-Glasgow law on this subject might well have been lulled into a false sense of security about the urgency of the speedy trial guarantee in Alaska. I would, therefore, hold that there was considerable probability of justifiable reliance by courts and law enforcement officials on pre-existing principles, abandoned or overruled in Glasgow v. State.

Ill

If Glasgow v. State is interpreted as fully retroactive on the precedential basis of this case, the most serious aspect of the majority’s opinion will be its implications for the administration of justice.9 The results of a retroactive application of Glasgow are not hard to imagine. The effect would be to bestow an unexpected windfall upon a number of ill-deserving convicted felons. If the lack of a speedy trial under Glasgow standards goes to the integrity of the guilt determination, there is little alternative but to grant post-conviction relief to anyone who was not in times past afforded a constitutionally speedy trial. In applying the results of today’s holding, we may end up reaching back many years, to a time when the speedy trial guarantee was ill-defined, and, relatively speaking, often slighted. The consequence of today’s holding may well be to free various persons who were legally convicted many years ago under constitutional standards then obtaining. Moreover, this gratuitous liberation would be accomplished without regard to the actual guilt of the defendants or the strength and cogency of the evidence produced against them at trial. I cannot join in the creation of such a doctrine.

While Gideon v. Wainwright, 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963) is indeed fully retroactive, see, e. g., Kitchens v. Smith, 401 U.S. 847, 91 S.Ct. 1089, 28 L.Ed.2d 519 (1971), this has resulted only in new trials for indigents previously denied their right to counsel. While, to be sure, in some cases in which the trials occurred long ago, retrial will be impossible and dismissal will be the inevitable result, even this is a far cry from the foreseeable consequences of today’s holding. In Glasgow the remedy selected was dismissal of the indictment, with prejudice. Reprosecu- *959' tion was totally barred. The effect of such a holding is clearly more drastic, in terms of freeing convicted felons on grounds totally irrelevant to the adequacy of the evidence against them, than was Gideon.

It is for such reasons that “the effect on the administration of justice” is one of the recognized criteria for determining retro-activity in constitutional adjudication. If the law is to grow, new principles must be nurtured and older ones must be abandoned. The technique of prospective overruling permits orderly progress in the law. The unsettling effect of new doctrine is cushioned by limiting the retroactive operation of newly formed rules. In the case before us I can find no reason for possibly vacating, on a haphazard basis, the criminal convictions of numerous persons who in fact may have suffered scant, if any, injury.

For these reasons I must dissent from Part III of the majority decision. Since the vast majority of the delay in this case occurred prior to the date of the decision in Glasgow v. State, supra, I would affirm appellant’s conviction.