The habeas corpus petitioner in this case was convicted of a state crime by a Massachusetts judge under the Massachusetts "two-tier" criminal trial procedure. As is his right under that procedure, the petitioner then chose to have a completely new trial by jury. He argued, however, to the new trial court, to the Massachusetts Supreme Judicial Court, and then to the federal district court, that he must be released, rather than undergo a new trial, because his bench trial conviction rested on insufficient evidence and any new trial would therefore place him in double jeopardy, in violation of the Fifth Amendment to the United States Constitution.

The federal district court reviewed the trial record and found that the evidence at the bench trial was constitutionally insufficient for conviction. The court further found that the state had not obtained defendant's informed consent to run a risk of double jeopardy harm, though it was not (nor in the future would it be) difficult for the state to seek such informed consent from criminal defendants. It held that a second trial was therefore barred by the Double Jeopardy Clause. Lydon v. Justices of Boston Municipal Court, 536 F.Supp. 647 (D.Mass.1982). We agree with the district court and affirm its decision.

* The Massachusetts two-tier system, like others in nearly half the states, allows a defendant to choose either the standard route of a jury trial with appellate review, or the two-tier route. See Comment, Double Jeopardy Problems Presented by Two-Tier Systems, 69 Geo.L.J. 1525, 1527 (1981). If he chooses the latter alternative, he first receives a bench trial. If he is convicted, he then has an absolute right to choose a second trial to take place de novo, either before a jury, or (if the defendant waives the jury) before another judge. Mass.Gen.Laws ch. 218, Secs. 26, 26A, 27A. The original system did not permit the initial choice--all defendants went through both tiers of the system. See Ludwig v. Massachusetts, 427 U.S. 618, 96 S.Ct. 2781, 49 L.Ed.2d 732 (1976). The present system is the product of the Court Reorganization Act, Chapter 478 of the Act of 1978, see Note, Double Jeopardy and the De Novo System, 16 New England L.Rev. 303, 308-09 (1981), following recommendations made by the Governor's Select Committee on Judicial Needs (the "Cox Committee") in 1976.

Prior to 1978, it would have been difficult to see how, from a constitutional perspective, this system could work to a defendant's disadvantage. For one thing, he is free to disregard the system by choosing an initial trial by jury. For another, if he chooses the two-tier alternative, he is entitled to a totally new trial on request. It is as if the state automatically accepts any complaint a defendant may have about his first trial; a new trial is always granted. See Ludwig v. Massachusetts, 427 U.S. at 631, 96 S.Ct. at 2788. At the same time, the system saves both the state and the defendant time and expense, because many defendants, presumably after seeing the evidence against them, do not seek the automatic new trial to which they have a right. Thus, the Massachusetts Supreme Judicial Court wrote that only about 9 percent of the criminal defendants in Boston chose a jury trial in the first instance, and "thousands of complaints were finally disposed of by convictions at bench trials because the affected defendants did not seek a jury trial." Lydon v. Commonwealth, 381 Mass. 356, --- n. 5, 409 N.E.2d 745, 748 n. 5 (1980).

In 1977 and 1978, however, the Supreme Court broadened the scope of defendant's double jeopardy protection in a way perhaps not foreseen by the Massachusetts system's reformers. In Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978), the Court overturned a prior line of cases and held that once "a reviewing court has determined that in a prior trial the evidence was insufficient to sustain the verdict of the jury," the Double Jeopardy Clause forbids a retrial. 437 U.S. at 5, 98 S.Ct. at 2144 (record review made by a federal appellate court); accord, Hudson v. Louisiana, 450 U.S. 40, 101 S.Ct. 970, 67 L.Ed.2d 30 (1981) (record review made by a state trial court). Then in Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979), the Supreme Court held that a federal habeas court must look to see whether a state conviction rests upon enough evidence to convince a rational trier of fact to find guilt beyond a reasonable doubt. See In re Winship, 397 U.S. 358, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970) (a conviction failing to meet that standard violates the Fourteenth Amendment "due process" clause). Petitioner believes that, under Jackson, the federal habeas court here properly determined that the evidence at the first trial was constitutionally insufficient and, hence, under Burks, he cannot be retried--he is entitled to unconditional release from custody.

Before evaluating petitioner's arguments, we shall describe the procedural context in which this case arises. The petitioner, Lydon, was charged with possessing implements designed to break open a "depository" (here, an automobile) in order to steal money or property from it. He chose to participate in the two-tier system. He was convicted at the first tier. The trial judge rejected his claim that the prosecution had not presented evidence of one key element of the offense charged, namely, an intent to steal money or property from the car.

Lydon asked for a trial de novo by a jury. He then asked the jury-trial judge to dismiss the charges on double jeopardy grounds. The motion was denied. A single justice of the Massachusetts Supreme Judicial Court, however, stayed the second trial and heard Lydon's claim. The justice was willing to assume the evidence was inadequate, and, on that assumption, asked the full court in the exercise of its "superintendence powers," Mass.Gen.Laws ch. 211, Sec. 3, to answer two questions:

1. Is it a denial of a defendant's right not to be placed in double jeopardy to require him to go through a jury trial, ... when the evidence at the bench trial was insufficient to warrant a conviction?

2. Assuming that a jury trial in such an instance would be a denial of a defendant's right not to be placed in double jeopardy, may the issue of the sufficiency of the evidence at the bench trial be considered again at the trial court level ...?

Lydon v. Commonwealth, 381 Mass. 356, --- n. 3, 409 N.E.2d at 747 n. 3.

The Supreme Judicial Court answered the first of these questions in the negative. It reasoned, first, that the initial trial judge had held that the evidence against Lydon was sufficient; since the state system did not provide for any review of that holding, no double jeopardy claim could arise. Second, regardless of whether the evidence was sufficient, there was no constitutional injury. Lydon chose to enter a system where a new trial would be his for the asking; in return, he must accept the fact that his only remedy would be a new trial. The Double Jeopardy Clause protects against governmental oppression; here, there was no such oppression because Lydon had volunteered to participate in the two-tier system.

The Supreme Judicial Court went on to answer the second question affirmatively, but added that its answer was "largely academic." Since Lydon had no double jeopardy right, the second trial court would not pass upon the adequacy of the evidence at the first trial.

Lydon sought certiorari, which was denied, and then sought habeas corpus. The federal district court, in the habeas proceeding, reviewed the original trial transcript and found the evidence to be constitutionally insufficient. The district court concluded that retrial was barred by the Double Jeopardy Clause. It ordered Lydon released and prohibited his retrial.

The Commonwealth has appealed. It obtained a stay of the district court's order pending this appeal. The Commonwealth has stated it will not retry the petitioner until the appeal is decided. Lydon is currently released on "personal recognizance."

We must first decide whether the district court had authority to review the record of Lydon's bench trial to determine whether the evidence was sufficient for conviction. Otherwise, the district court, faced with an authoritative state court finding that the evidence was sufficient, would have seen no double jeopardy problem. We believe that the district court had this authority.

1. Contrary to the Commonwealth's contention, Lydon met the technical "custody" requirements of the two habeas corpus statutes relevant here, 28 U.S.C. Sec. 2241(c)(3) and 28 U.S.C. Sec. 2254. The first of these statutes authorizes the writ to be issued on behalf of "a prisoner ... in custody in violation of the Constitution ... of the United States ...." The second authorizes its issuance on behalf of "a person in custody pursuant to the judgment of a state court" on the ground, (among others) "that he is in custody in violation of the Constitution ...."

It is now well established that release on personal recognizance is "custody" within the meaning of both of these habeas corpus statutes. Hensley v. Municipal Court, 411 U.S. 345, 93 S.Ct. 1571, 36 L.Ed.2d 294 (1973). Personal recognizance subjects an individual to restraints "not shared by the public generally," for "state judicial officers may demand his presence at any time," and "[d]isobedience is itself a criminal offense." 411 U.S. at 351, 93 S.Ct. at 1574. In this instance, if Lydon failed to appear for his jury trial, his original sentence could be reimposed. Mass.Gen.Laws ch. 278, Sec. 24. Moreover, although the state characterizes Lydon's choice of a new trial as a choice that "wipes out" the first court's judgment, the fact is that Lydon is under special restraint because he was found guilty at the bench trial; otherwise, he would have been released. Lydon is essentially in the same position as the defendant in Hensley, who "remained at large under an order of the state court staying execution of his sentence." 411 U.S. at 347, 93 S.Ct. at 1573. We therefore believe that Lydon is in "custody" under the less restrictive language of Sec. 2241(c)(3) and also under the more restrictive language of Sec. 2254, which requires that the custody be "pursuant to the judgment of a state court." Cf. Thames v. Justices of the Superior Court, 383 F.Supp. 41, 43 (D.Mass.1974) (release on bail after a mistrial is "custody" under both Sec. 2254 and Sec. 2241).

2. Once the technical custody problems are met, at least as a general matter, a habeas court can review a petitioner's claim that a state court conviction rests upon insufficient evidence. Jackson v. Virginia, supra. The constitutional basis for such review arises from the fact that if, "after viewing the evidence in the light most favorable to the prosecution," no "rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt," the state conviction violates the Fourteenth Amendment's "due process" requirements. 443 U.S. at 319, 99 S.Ct. at 2789. See Johnson v. Louisiana, 406 U.S. 356, 362, 92 S.Ct. 1620, 1624, 32 L.Ed.2d 152 (1971); In re Winship, supra. And, "assuming that state remedies have been exhausted, and that no independent and adequate state ground stands as a bar, it follows that such a claim is cognizable in a federal habeas corpus proceeding." Jackson v. Virginia, 443 U.S. at 321, 99 S.Ct. at 2790 (citations omitted).

3. While we recognize potential practical problems related to the proper timing of habeas "evidence sufficiency" review, cf. Stevens v. Circuit Court of Milwaukee County, 675 F.2d 946 (7th Cir.1982), we believe that here review of Lydon's claim before state court retrial was appropriate. For one thing, Lydon's sufficiency of the evidence claim is part of his double jeopardy attempt to prevent the retrial with which he is directly and immediately threatened. He seeks habeas relief now to avoid the very psychological and material hardships of a second trial that the Double Jeopardy Clause seeks to prevent. See Green v. United States, 355 U.S. 184, 187-88, 78 S.Ct. 221, 223, 2 L.Ed.2d 199 (1957). Unless habeas review is available before the second trial, a meritorious "evidence insufficiency" claimant would have to suffer those hardships before winning his case. Benson v. Superior Court Department of the Trial Court of Massachusetts, 663 F.2d 355, 359 (1st Cir.1981). This consideration led the Supreme Court to characterize a double jeopardy ruling in a federal trial court as a "collateral order," allowing an immediate appeal under Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949). Abney v. United States, 431 U.S. 651, 97 S.Ct. 2034, 52 L.Ed.2d 651 (1977). This consideration does not settle the timing question, for Lydon has a legitimate double jeopardy claim only if he can reverse the earlier ruling that the evidence was sufficient. And it is the timing of this latter review that is at issue. See, e.g., United States v. Becton, 632 F.2d 1294 (5th Cir.1980), cert. denied, 454 U.S. 837, 102 S.Ct. 141, 70 L.Ed.2d 117 (1981). Nonetheless, this factor argues in favor of speedy habeas review.

For another thing, all state remedies on the evidence sufficiency claim have been exhausted, thus satisfying the explicit exhaustion requirement of Sec. 2254 and the similar implicit requirement of Sec. 2241. See Braden v. 30th Judicial Circuit Court of Ky., 410 U.S. 484, 490, 93 S.Ct. 1123, 1127, 35 L.Ed.2d 443 (1972). None of the state courts to which Lydon brought his claim actually ruled on the sufficiency of the evidence, and, under the theory announced by the Massachusetts Supreme Judicial Court, the first-tier judge's determination that the evidence was sufficient will never be reviewed anywhere within the Massachusetts system. See Lydon v. Commonwealth, 381 Mass. at ---, 409 N.E.2d at 752.

Moreover, we are aware of no "independent and adequate state ground" now barring review. Jackson v. Virginia, 443 U.S. at 321, 99 S.Ct. at 2790. See Estelle v. Williams, 425 U.S. 501, 96 S.Ct. 1691, 48 L.Ed.2d 126 (1976); Francis v. Henderson, 425 U.S. 536, 96 S.Ct. 1708, 48 L.Ed.2d 149 (1976); Fay v. Noia, 372 U.S. 391, 430-34, 83 S.Ct. 822, 844-45, 9 L.Ed.2d 837 (1963). Indeed, Lydon's conviction at the second trial would not, we should think, bar a federal habeas court from examining the sufficiency of the evidence at the first trial. Cf. Fay v. Noia, 372 U.S. at 433-34, 83 S.Ct. at 846. Jackson mandates a federal "evidence sufficiency" examination; and, as we shall discuss further in Part IV below, Burks invalidates a second trial where the first conviction rests upon insufficient evidence.

Further, Massachusetts has not sought to postpone habeas review until after the second trial. Rather, the Supreme Judicial Court delayed Lydon's retrial while it considered his double jeopardy claim. The Commonwealth did not attempt to proceed with the second trial pending resolution of Lydon's federal claim. It has not argued that a speedy retrial is important.

Finally, in principle we deal here with habeas review where there is no state appellate review. In principle, habeas then involves only one set of review procedures. Since most states typically provide one such review procedure prior to a retrial--where, for example, a defendant appeals a conviction--the additional delay that habeas may bring here is likely to be no greater than in the typical state case with state appellate review before retrial.

At the same time, we note two factors that militate against allowing habeas review prior to retrial. First, were such review generally available, it might create practical problems when a defendant convicted in a state court appeals, alleging "evidence insufficiency" as one of a number of grounds for reversal. The state appellate courts may well reject the "evidence insufficiency" claim, yet order a retrial on other grounds. Can the defendant in such a case then force the state to run the gauntlet of federal habeas proceedings on the "evidence insufficiency" issue prior to his state retrial? Can he obtain delay of the retrial, regardless of the merits of his claim, and thereby, through elapsed time and fragile memory, reduce his chances of subsequent conviction?

Second, to hold that defendants generally seek federal habeas "evidence insufficiency" review prior to a state retrial may seem anomalous when contrasted with Fourth, Fifth and Ninth Circuit cases denying a federal defendant appellate review of a similar claim prior to federal retrial. United States v. Ellis, 646 F.2d 132 (4th Cir.1981); United States v. Rey, 641 F.2d 222 (5th Cir.), cert. denied, 454 U.S. 861, 102 S.Ct. 318, 70 L.Ed.2d 160 (1981); United States v. Becton, supra; United States v. Carnes, 618 F.2d 68 (9th Cir.), cert. denied, 447 U.S. 929, 100 S.Ct. 3028, 65 L.Ed.2d 1124 (1980). In each of these cases, jury deadlock resulted in a mistrial. The defendant's motion for acquittal on the basis of the insufficiency of the evidence was denied, and a new trial was ordered. The defendant then moved to dismiss on the grounds of double jeopardy, in turn, based on a claim of "evidence insufficiency." The appellate court in each case dismissed the appeal, on the grounds that it was interlocutory and not appealable under the collateral order exception to the finality rule. The essence of the court's position was that the defendant was merely appealing the trial court's decision that the evidence was constitutionally sufficient. Such an issue, each court held, was not "collateral" under the Cohen/Abney test; but rather was directly involved in the merits of the case. Therefore, the appellate court would not review that decision, even to reach the double jeopardy issue. There are other circuits taking a different view. See United States v. McQuilkin, 673 F.2d 681 (3d Cir.1982) (interlocutory appeal on "evidence insufficiency" allowed in federal case); Delk v. Atkinson, 665 F.2d 90 (6th Cir.1981) (habeas review of evidence insufficiency allowed prior to state retrial). We do not choose between the views of the Fourth, Fifth and Ninth Circuits on the one hand, and the Third and Sixth on the other. Rather, we simply note that it would be odd if, under the view of most circuits, a federal defendant were denied interlocutory appeal, while a state defendant was entitled to receive federal habeas review before his retrial.

While we recognize the force of these latter two sets of arguments, we do not find them determinative here. For one thing, the practical problem of interfering with ongoing state proceedings might be mitigated through wise use of the habeas corpus court's power to stay those proceedings, see 22 U.S.C. Sec. 2251, reserving use of that power for those cases that present a strong showing of likely success on the merits, and otherwise taking appropriate account of the "comity" principles discussed in Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971). Compare Robinson v. Wade, 686 F.2d 298, 302-03 n. 7 (5th Cir.1982), and Baker v. Metcalfe, 633 F.2d 1198, 1200 n. 3 (5th Cir.), cert. denied, 451 U.S. 974, 101 S.Ct. 2055, 68 L.Ed.2d 354 (1981), with Stevens v. Circuit Court of Milwaukee County, supra, and Sparks v. Garrison, 446 F.Supp. 649 (M.D.N.C.1978), aff'd 612 F.2d 1310 (4th Cir.1979). Insofar as the habeas court takes account of the need for speedy retrials, the sense of anomaly with the federal "interlocutory appeal" cases (which themselves rest on the need for workable procedures) is diminished.

For another thing, the case before us is one in which, for reasons previously set out, the need for postponement of habeas review is minimal. We therefore restrict our holding to such a case. We conclude that habeas review of "evidence insufficiency" was proper at this time in the case before us. But, we specifically decline to extend our holding to other cases not characterized by the several factors favoring review listed here. We postpone to another day any authoritative decision about "habeas timing" in such other cases.

The district court reviewed the record and concluded that the evidence was, in fact, insufficient. The Commonwealth's complaint alleged that Lydon possessed implements designed to break open a "depository" (here an automobile), with the intent to steal money or property from it, in violation of Mass.Gen.Laws ch. 266, Sec. 49. The evidence at trial was as follows. A plainclothes police officer testified that late one night he observed the defendant and another man in the process of breaking into the right front window of a car. When the two had opened the car, both of them climbed into it. As the witness and other policemen moved in, an alarm went off in the car and both men climbed back out. On searching the defendants, the police found a bottle of whiskey. Nearby, they found a dentpuller and a screwdriver. The witness stated that the defendants had no time to take anything from the car. The evidence permitted an inference that the defendant intended to steal the car itself, that he intended to take it for a short ride, or that he merely intended to go to sleep in it. But under Massachusetts law, it was impossible to say that the Commonwealth had shown beyond a reasonable doubt that the defendant intended to steal money or property from the car.

This was the view of a Massachusetts court in an almost identical case. The court stated,

Breaking into an automobile merely with the intent to use it is not so 'remote according to the ordinary course of events' as to allow us to discount such an intent in this case and permit the jury to conclude that there was an intent to steal from the automobile merely from the breaking.

Commonwealth v. Armenia, 4 Mass.App.Ct. 33-39, 340 N.E.2d 901, 904-05 (Mass.App.1976) (citation omitted). Because the Commonwealth had not shown "an intent to steal from the automobile," the court set the jury's verdict aside.

Having examined these facts and the Massachusetts authority interpreting the statute, we must conclude that the federal district court's determination that the evidence was not constitutionally sufficient to support conviction is not erroneous. This is not to say that defendant was innocent of all crimes. But one can be convicted only of the crime charged. And, as to that crime, an intent to steal from the automobile was not shown.

Once a habeas court has decided that the evidence at Lydon's first trial was constitutionally insufficient--that no rational trier of fact could have found him guilty as charged beyond a reasonable doubt--the Double Jeopardy Clause, as interpreted in Burks, forbids his retrial. As classically stated:

The underlying idea, one that is deeply ingrained in at least the Anglo-American system of jurisprudence, is that the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense, thereby subjecting him to embarrassment, expense and ordeal and compelling him to live in a continuing state of anxiety and insecurity, as well as enhancing the possibility that even though innocent he may be found guilty.

Green v. United States, 355 U.S. 184, 187-88, 78 S.Ct. 221, 223, 2 L.Ed.2d 199 (1957). In applying this purpose, the Supreme Court has attached special weight to judgments of acquittal, Tibbs v. Florida, --- U.S. ----, ----, 102 S.Ct. 2211, 2218, 72 L.Ed.2d 652 (1982); see United States v. DiFrancesco, 449 U.S. 117, 129, 101 S.Ct. 426, 433, 66 L.Ed.2d 328 (1980); United States v. Scott, 437 U.S. 82, 91, 98 S.Ct. 2187, 2193, 57 L.Ed.2d 65 (1978); United States v. Martin Linen Supply Co., 430 U.S. 564, 571, 97 S.Ct. 1349, 1354, 51 L.Ed.2d 642 (1977). "A verdict of not guilty whether rendered by the jury or directed by the trial judge, absolutely shields the defendant from retrial." Tibbs v. Florida, 102 S.Ct. at 2218; see, e.g., United States v. Martin Linen Supply Co., 430 U.S. at 564, 97 S.Ct. at 1351; United States v. Ball, 163 U.S. 662, 666-71, 16 S.Ct. 1192, 1193, 41 L.Ed. 300 (1896). And, in Burks, the Court specifically held that a reversal of a conviction because the evidence was constitutionally insufficient has the same effect as an acquittal, for it means that no rational fact finder could have found the defendant guilty. Burks v. United States, 437 U.S. at 16, 98 S.Ct. at 2149; Tibbs v. Florida, 102 S.Ct. at 2217-18.

The underlying rationale, explained fully in Burks, is that the prosecution gets one full and fair opportunity to muster its evidence. If it nonetheless produces a case "so lacking that it should not have even been submitted to the jury," then it ought not to have another opportunity to supply additional evidence.

Since we necessarily afford absolute finality to a jury's verdict of acquittal--no matter how erroneous its decision--it is difficult to conceive how society has any greater interest in retrying a defendant when, on review, it is decided as a matter of law that the jury could not properly have returned a verdict of guilty.

437 U.S. at 16, 98 S.Ct. at 2149 (emphasis in original). Tibbs v. Florida, 102 S.Ct. at 2217-18; Hudson v. Louisiana, 450 U.S. at 43, 101 S.Ct. at 972. To allow the prosecution that opportunity--to learn from the first trial and to produce a more adequate case the second time--is to allow the type of oppression the Double Jeopardy Clause was designed to guard against. Burks v. United States, 437 U.S. at 11, 98 S.Ct. at 2147; Green v. United States, supra.

Where "evidence insufficiency" is found, the reasons that typically allow retrial when an appellate court finds trial error are not present. As the Court explained in Burks,

reversal for trial error, as distinguished from evidentiary insufficiency, does not constitute a decision to the effect that the government has failed to prove its case. As such, it implies nothing with respect to the guilt or innocence of the defendant. Rather, it is a determination that a defendant has been convicted through a judicial process which is defective in some fundamental respect, e.g., incorrect receipt or rejection of evidence .... When this occurs, the accused has a strong interest in obtaining a fair readjudication of his guilt free from error, just as society maintains a valid concern for insuring that the guilty are punished....

The same cannot be said when a defendant's conviction has been overturned due to a failure of proof at trial, in which case the prosecution cannot complain of prejudice, for it has been given one fair opportunity to offer whatever proof it could assemble.

Burks v. United States, 437 U.S. at 15-16, 98 S.Ct. at 2149 (footnote omitted).

We have quoted at some length from the Supreme Court's recent double jeopardy cases because both the Court's language in these cases and the rationale that it embodies make clear that the Double Jeopardy Clause forbids retrial here. The prosecution had a full and fair opportunity to present its evidence. The evidence was such that no rational trier of fact could have found Lydon guilty of the crime charged. Had the trier of fact acquitted Lydon, Massachusetts law would have barred his retrial. The process of judicial review has conveniently pinpointed the evidence which was lacking, and retrial simply gives the prosecutor another opportunity to supply it. In these respects, it is difficult to see any significant difference between this case and the situation in Burks of a trial followed by appellate reversal for insufficient evidence.

The Supreme Court has twice rejected double jeopardy attacks on the constitutionality of the Massachusetts, and other similar two-tier systems. However, neither of these cases affect our conclusion here. Colton v. Kentucky, 407 U.S. 104, 92 S.Ct. 1953, 32 L.Ed.2d 584 (1972), concerned the question of the state's power to impose an enhanced sentence at the second trial; it is not directly relevant to the issue before us. Ludwig, decided prior to Burks, holds that, as a general matter, the Double Jeopardy Clause does not forbid the retrial de novo of one previously convicted. Its reasoning, however, supports a finding of unconstitutionality here.

The Ludwig Court reached its conclusion primarily by drawing a parallel between a "two-tier" defendant seeking a new trial after conviction, and an ordinary convicted defendant seeking a new trial on appeal. The Court wrote in 1975:

The Massachusetts system presents no danger of prosecution after an accused has been pardoned; nor is there any doubt that acquittal at the first tier precludes reprosecution. Instead, the argument appears to be that because the appellant has been placed once in jeopardy and convicted, the State may not retry him when he informs the trial court of his decision to "appeal" and to secure a trial de novo.

Appellant's argument is without substance. The decision to secure a new trial rests with the accused alone. A defendant who elects to be tried de novo in Massachusetts is in no different position than is a convicted defendant who successfully appeals on the basis of the trial record and gains a reversal of his conviction and a remand of his case for a new trial. Under these circumstances, it has long been clear that a State may reprosecute. United States v. Ball, 163 U.S. 662 [16 S.Ct. 1192, 41 L.Ed. 300] (1896). The only difference between an appeal on the record and an appeal resulting automatically in a new trial is that a convicted defendant in Massachusetts may obtain a "reversal" and a new trial without assignment of error in the proceedings of the first trial.

427 U.S. at 631-32, 96 S.Ct. at 2788. This parallel between the "two-tier" trial and the ordinary trial plus appeal supported the Massachusetts system's constitutionality prior to Burks. But Burks carved out an exception to the general rule that retrial is permissible. Since Burks requires release when the evidence was constitutionally insufficient at the first trial, given the analogy upon which Ludwig rested, Ludwig requires the same result within the two-tier system.

The single respect in which this case differs from both Ludwig and the ordinary instance considered in Burks is that, under the modification made in the Court Reorganization Act of 1978, a defendant now may freely choose whether to subject himself to the system's first tier. If he wishes, he can proceed directly to the second. The Massachusetts Supreme Judicial Court believed this fact critical. It apparently felt that a defendant who had voluntarily chosen to subject himself to the first tier thereby waived whatever rights he may have had to any relief other than a new trial. Such a waiver would be reasonable in light of the unusual benefit the system grants him, namely a new trial for the asking.

While we agree that it would be reasonable to waive a Burks-type right to obtain this benefit, we do not agree that the defendant did so here. It is clear, that where constitutional rights are at issue, there must be an "intentional relinquishment of a known right or privilege" before they can be waived. Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 1023, 82 L.Ed. 1461 (1938). Although prior to Burks the existence of a constitutional right may have been in doubt, after that case, it is apparent that the harm at issue--the prosecutor's second chance to supply constitutionally sufficient evidence--has a constitutional dimension. And Burks itself indicates that a waiver of the defendant's right to avoid that harm is not to be implied lightly. That case specifically denied that a defendant's decision to seek a new trial on appeal automatically waived his right (if the evidence proved insufficient) to avoid a new trial altogether.

There was no intentional relinquishment of the right at issue in this case. There is no indication that the defendant clearly understood he was waiving any right to potential freedom from retrial; to the contrary, if anything, he believed that the state could not retry him again should the evidence at the first trial prove constitutionally insufficient. And, he has vigorously asserted this claim throughout.

There is no special circumstance present here that suggests that anything less than an explicit waiver is appropriate. Amici argue that the district court's ruling would significantly impair the workings of the present system. But that is not so. The present system fully informs defendants of their choice and its implications for their right to a jury trial. Indeed, the trial must "be by a jury of six, unless the defendant files a written waiver and consent to be tried by the court without a jury, subject to his right of [trial de novo]." Mass.Gen.Laws. ch. 218, Sec. 26A. Rule 21(A) of the Initial Rules of Criminal Procedure for the District Courts provides:

The oral notice of right to first-instance jury trial shall include a statement to the defendant that if he or she waives this right, he or she will receive a trial before a judge and if not satisfied with the results of that trial will be able to appeal for a new trial before a jury.

The Massachusetts Supreme Judicial Court has itself suggested that consideration "be given to enlarging the consent form so that the defendant acknowledges that his only avenue of relief from any error in the bench trial is a trial de novo." Lydon v. Commonwealth, 381 Mass. at --- n. 13, 409 N.E.2d at 751 n. 13. We see no reason to indulge in the legal fiction of "implied consent" or "constructive waiver" when the reality could be so readily obtained. A fully informed defendant could proceed to a full, ordinary single level jury trial if he did not choose to waive the right here at issue. It may be that many defendants would waive the Burks-type right, for it would come into play only in a very limited circumstance, namely, when a first-tier judge wrongly concludes that the evidence was constitutionally sufficient to convict and a later court reverses the ruling. But, whether or not to waive is a decision that an informed defendant ought to make.

For these reasons, we believe that the district court properly issued a writ of habeas corpus. We note the Commonwealth's argument that the district court lacks the additional power specifically to enjoin a second trial. But, rather than explore the merits of this technical question of law, we shall simply follow the path of the court in Delk v. Atkinson, 498 F.Supp. 1282 (M.D.Tenn.1980), rev'd on other grounds, 665 F.2d 90 (6th Cir.1981) which found it "unnecessary to order specific injunctive relief barring state retrial ..., because the Court assumes that the State will act in good faith in light of the Court's decision and dismiss the charges against the petitioner." 498 F.Supp. at 1284. The appropriate relief is simply issuance of the writ providing Lydon his freedom from being required to answer any summons to a second trial. We have no reason to believe the Commonwealth will seek to retry him unless this decision is reversed.

Affirmed.

LEVIN H. CAMPBELL, Circuit Judge (dissenting).

The court may be right. Certainly Judge Breyer's opinion deals scrupulously with the issues, and I do not question the intelligence and logic reflected in my colleagues' decision. Yet I see this as one of those cases where the relentless application of secondary precepts developed in other, very different settings leads ultimately to a foolish and wrong result--a result not dictated by the Constitution, and one which has the unfortunate effect of undermining a useful and fair state procedure.

My colleagues are sensitive to the latter problem. The court's opinion reflects its deep concern over the harm to the Massachusetts two-tier system which may now occur, as a result of our judgment, from frivolous petitions and appeals alleging insufficiency of evidence and double jeopardy. Whether there are ways around this, through more elaborate waiver forms or other mechanisms, I do not know. I do know that the Massachusetts system is likely to become more costly and less coherent as a result of our decision--at a time when virtually every commentator, including the Chief Justice of the United States, is decrying the diminishing ability of both state and federal judicial systems to cope with the caseload. Moreover, I fear that decisions such as this reflect an ongoing, perhaps irreversible tendency of federal courts to treat subsidiary principles announced in constitutional decisions as if they were propositions of plane geometry, and to extend and apply them rigidly, with the unfortunate result that healthy experimentation and diversity at state and local levels is made increasingly less possible.

Here, building on Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978), our court interrupts the practical functioning of Massachusetts' two-tier system--a fair and useful system which the state offers as an elective option to criminal defendants charged with misdemeanors and lesser felonies. In Burks, the Court held that the double jeopardy clause prohibits a second trial following the reversal of a conviction obtained with insufficient evidence. Here, we fasten on the fact that a defendant convicted at a first-tier bench trial has no way to avoid a second-tier trial if he wishes to challenge the state's case either on the facts or the law: thus he must be "tried again" even if the earlier proceeding at which he was convicted was lacking in sufficient evidence. To avoid this, we insist that he and other first-tier convicted defendants must be allowed to raise their insufficiency claims before the second trial so they can be discharged if they prevail. We thus throw a substantial monkey wrench in the two-tier system whose distinguishing feature and, indeed, supreme virtue is that it offers a second trial, not an appeal on the law.

I would contend that the reasonable way to analyze the double jeopardy issue in this case is to look at the Massachusetts system as a whole, not in isolated fragments. If this is done, I see little ground for serious concern that the policy underlying the double jeopardy clause is being offended.

To begin with, the two-tier system was not imposed on the defendant. Massachusetts defendants may elect to receive an initial jury trial followed by a conventional appeal. There is thus no issue of Massachusetts forcing Lydon to accept a system which provides, in the second stage, for a de novo trial rather than a conventional appeal.

Second, the two-tier procedure Lydon elected affords all the most basic double jeopardy protection that would be afforded in a more conventional system. As Massachusetts has long recognized, a judgment of acquittal at the first tier precludes later prosecution at the second level. And it is well established that trial de novo following a first-tier conviction does not ordinarily offend the double jeopardy clause. Ludwig v. Massachusetts, 427 U.S. 618, 96 S.Ct. 2781, 49 L.Ed.2d 732 (1976).

Third, the two-tier system affords distinct benefits to a criminal defendant not present in the more conventional system. These benefits flow from the very same aspect of the system which Lydon criticizes. Under the two-tier system, a defendant receives two full opportunities to be acquitted on the facts. The conventional American criminal system, unlike that in some European countries, permits virtually no appellate review of factual determinations. The opportunities now afforded to suppress evidence on legal and constitutional grounds do not help on this score: indeed, the most worrisome ground for a possible miscarriage of justice--misidentification by a witness--remains largely untouched by the jurisprudence of the past generation. The Massachusetts two-tier system, however, allows the possibly innocent defendant two opportunities for acquittal on the facts. This is a benefit of no small dimension. For defendants generally, it is far more meaningful than retaining a right to appeal from the first trial on grounds of insufficient evidence. In foregoing the latter, convicted defendants gain the right not to be convicted finally unless the prosecution twice proves their guilt beyond a reasonable doubt.

Fourth, while the de novo trial allows the prosecution to put in a different and perhaps stronger case, the same is of course true for the defendant. Bearing in mind that, in the usual case, the first-tier court can be expected to acquit where the evidence is insufficient, it will be the rare case where a convicted defendant with a valid insufficiency claim is forced to stand trial for a second time. The far more usual case will be that of the defendant convicted at the first tier on legally sufficient (if factually close) evidence. Under the Massachusetts system, that defendant gets the distinct benefit of a brand new trial, in which he has all the tactical benefits of knowing the prosecution's case and shifting his defenses to meet it.

Fifth, in the rare case of someone like Lydon who was convicted at the first tier on insufficient evidence, the defendant has no lack of opportunities for later vindication: first, he may be acquitted on motion after the prosecution puts in its case in the de novo trial, assuming the judge feels the prosecution's evidence is insufficient. Second, the jury may acquit him at the second trial. Third, he may later win on an insufficiency claim relative to the second trial in a Massachusetts appellate court. While technically he is "tried again", the second stage proceeding can be regarded as but an enlarged, fact-sensitive part of a single, continuous course of judicial proceedings during which, sooner or later, a defendant receives more--rather than less--of the process normally extended to criminal defendants in this nation. To deny a state the right so to structure its optional criminal process is to approach saying that the Constitution imposes for all time a single model of criminal process.

Sixth, Lydon argues that the second-tier trial is unfair because it allows the prosecution to improve on its case--to put in more evidence. As noted, any disadvantage would seem to be more than outweighed by the advantage to most defendants of being able to improve on their defense in the second trial. But beyond this, the likelihood of the prosecution's obtaining additional evidence would seem slight. The prosecution has every incentive to put in its best evidence in the first tier. If it does not, and acquittal results, it can never proceed again. Moreover, the stronger the showing, the more likely the defense is to accept a first-round conviction as final, without seeking a de novo trial. It follows that appellate review of the second conviction will commonly provide an effective remedy in the rare instance in which both trial courts and the jury improperly assess the sufficiency of the evidence. With these safeguards in place, there is every reason to believe that the two-tier system--if permitted to function without interruption--will vindicate claims of evidentiary insufficiency.

Lydon's predicament illustrates the point. The court concludes that the evidence introduced at Lydon's trial failed to pass muster under the rule of Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979). Yet the two-tier system, were it permitted to run its course, would surely correct the error. Of course, part of the process involves Lydon's submission to a second trial. But the second trial typically poses little threat that the Commonwealth will present new, more substantial evidence; indeed, a review of the record suggests that there is probably no effective way to supply evidence of intent. And even if this were to occur in isolated cases, the result is hardly so unfair, when weighed against the system's countervailing benefits to defendants as a whole, as to warrant restructuring the whole system.

Finally, I return to the point first made--that the two-tier system was not imposed on Lydon. Lydon's submission to a second trial is part of the bargain he struck with the Commonwealth when he chose the two-tier system with its distinct advantages as well as this one disadvantage. The Commonwealth requires a defendant who wishes to be tried in the two-tier system to sign a written waiver of his right to trial by jury. At the time Lydon signed the waiver he was represented--as the Commonwealth requires1--by an attorney whose competence is undisputed. Any attorney, and indeed most clients, would know that in choosing the two-tier system, a defendant surrenders his right to challenge legal errors committed in the initial proceeding; in place of this right, he obtains the more important right to require a trial de novo, after which, of course, there is the conventional appellate process. See Mann v. Commonwealth, 359 Mass. 661, 664, 271 N.E.2d 331, 333-34 (1971). Lydon has never asserted that he lacked knowledge of the limited right of appeal in the two-tier system. Thus, we may assume that he had a sufficient practical understanding of the consequences of his choice.

This court nonetheless holds that Lydon's intelligent, informed decision to be tried in the two-tier system did not constitute a waiver of his right to challenge the sufficiency of the evidence at his bench trial. The court does not conclude that the right is so fundamental as to defy any form of waiver. Rather it holds that Lydon's decision did not constitute an "intentional relinquishment of a known right." Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 1023, 82 L.Ed. 1461 (1938). In so concluding, the court implies that a written waiver of the right would pass constitutional muster.

Thus, the court invites the Commonwealth to adopt a procedure that insofar as I can determine, would have no practical impact on a defendant's choice of the two-tier system. Under the suggested written waiver procedures, a defendant would rely heavily on the advice and expertise of his attorney in deciding whether to waive his right to challenge evidentiary sufficiency. In making this strategic choice, the attorney would be called on to weigh precisely the same factors that Lydon's attorney presumably weighed in this case. The effect of the present opinion is to create the possibility of a host of new technicalities which may vitiate the practical usefulness of the de novo system. I would hold Lydon to the consequences of his choice.

In conclusion, I fear that the court's decision is a further step in locking the states into rigid, formalistic criminal procedures, without regard to the actual justice or injustice of a given system, merely to satisfy refined, logical principles derived from prior cases decided in other contexts. Such an approach provides employment for attorneys, increases the expense of justice, and enlarges the possibility that a guilty person may get off on a technicality. It does little to advance the cause of justice. The Massachusetts two-tier system must be viewed as a whole, not forced through a strainer developed from cases dealing with a different system. If so viewed, I believe that it does not create double jeopardy. What it does do is provide a practical trial option with many advantages to both sides.