Sidney Clark, a state prisoner, has taken this appeal from the district court's denial of his petition for habeas corpus under 28 U.S.C. 2254, following exhaustion of state remedies, to vacate two state convictions. Clark was sentenced to five years imprisonment on each conviction. Apparently, the sentences are running concurrently.

Two constitutional questions are presented: (1) Whether Delaware denied the accused constitutionally adequate notice that his conduct violated a criminal statute and (2) whether pretrial publicity precluded a fair trial. Satisfied that the record justifies the district court's decision on the pre-trial publicity issue, we confine our discussion to the constitutional requirement of notice.

A grand jury indicted Sidney Clark in two bills which charged violations of 50 Del.Laws Ch. 299, 11 Del.C. 635 (1955), an embezzlement statute. The legislature had enacted that section in 1955 to replace former 11 Del.C. 635, 637-639, 641-643. In order to effect that desired substitution, the legislature repealed 637-639, 641-643, and 'amended' 635 by striking all of its former language and inserting the newly desired language as a new 635.

That new section provided as follows: 'Whoever embezzles money or other property which may be the subject of larceny, to the value of (One Hundred Dollars) $100.00 or more . . . is guilty of a felony . . ..' Clark, a lawyer, was charged with misappropriating funds of a client that had been entrusted to him.

Clark moved to dismiss the indictment, contending that new 635 was unconstitutionally vague for failure to define the term 'embezzles' and that there was no common law definition for this purely statutory crime. The Delaware trial court agreed that new 635 was void for vagueness, but did not dismiss the indictments. Instead, the judge held that, under established state law,1 where a subsequent statute 'amends' former statutory language, but in an unconstitutional fashion, then the earlier statute is revived if such a revival is consistent with the legislative purpose. Then, because he thought that the Delaware legislature surely intended to prohibit and punish the misappropriation of a client's funds, he deemed it more consistent with that intent to revive the old 635 and to require, as he did, that prosecution proceed under it.2

The state sought to appeal this interlocutory order, but the Delaware Supreme Court held itself without jurisdiction to decide the point at that stage. State v. Clark, Del.1970, 270 A.2d 371. The Court suggested that the state appeal the questioned ruling, if necessary, when the proceedings below were final.

Following Clark's trials and convictions the Delaware Supreme Court affirmed, Clark v. State, Del.1972, 287 A.2d 660, and the United States Supreme Court denied certiorari, 409 U.S. 812, 93 S. Ct. 139, 34 L. Ed. 2d 67. The Delaware Court rejected Clark's contention that trial and conviction under a conceptually revived statute after a preliminary ruling invalidating the predecessor statute under which the accused had been indicted, denied due process under the doctrine of Bouie v. City of Columbia, 1964, 378 U.S. 347, 84 S. Ct. 1697, 12 L. Ed. 2d 894. The court held that the revival doctrine served adequate notice that if new 635 were ruled unconstitutional old 635 would be effective. The state had not cross-appealed from the determination that new 635 was unconstitutional.

It is in this context that we consider the narrow question whether the notice requirement of the due process clause permits a state, after ruling one of its criminal statutes overly vague, to apply that statute's superseded predecessor to the defendant in the very case which ruled the successor statute unconstitutional. We hold that this was not constitutionally permissible. Our analysis follows.

The alleged offense occurred and the accused was indicted at a time when, pursuant to legislative enactment, the State of Delaware had published and was holding out the new 635 as its only proscription of such misconduct as the indictment charged. Delaware long ago recognized that embezzlement is not a crime at common law. State v. Brewington, 1910, 25 Del. 71, 78 A. 402. By force of the unreversed holding of the trial court, it is the rule of Clark'sprosecution that new 635 is void for vagueness. And by definition, an unconstitutionally vague statute is one that fails to give fair notice that particular conduct is proscribed by the state. See United States v. Harriss, 1954, 347 U.S. 612, 617; 74 S. Ct. 808, 98 L. Ed. 989; Connally v. General Construction Co., 1926, 269 U.S. 385, 391, 46 S. Ct. 126, 70 L. Ed. 322. Thus, Delaware's own interpretation of new 635 and its rejection of that section as a statutory basis for Clark's prosecution cause us to hold that new 635 did not provide constitutionally adequate notice. Cf. Bouie v. City of Columbia, supra.

It follows that this conviction can be upheld only if the old 635, the supercession of which had been legislatively declared and publicly announced, could continue to serve as notice of the criminality of Clark's conduct. To reach that conclusion, one would have to reason, first, that the new 635 on its face gave adequate public notice of its own invalidity and, second, that the public, thus informed, was then put on further notice that the officially announced statutory repeal or supercession of old 635 was legally ineffective.

In our view such reasoning is too tortured and too far removed from reality to satisfy the due process requirement that, at the time of an alleged offense, the accused shall have been on notice that his conduct was proscribed by the state's criminal law. Indeed, we are unable to surmount the first hurdle: that the new 635 could serve as notice of its own invalidity. And without that notice there could be no occasion to consider old 635 as possibly relevant.

The Supreme Court of Delaware was confronted with an analogous problem in State v. Dickerson, Del.1972, 298 A.2d 761. The controversy in that case arose as a result of Furman v. Georgia, 1972, 408 U.S. 238, 92 S. Ct. 2726, 33 L. Ed. 2d 346, which restricted the procedures which a state might use to impose the death sentence. Delaware's first-degree murder statute provided for only the death penalty, but a separate act, called the Mercy Statute, allowed a judge, at the discretionary recommendation of the jury, to impose life imprisonment. While Dickerson was awaiting trial, the trial judge certified to the Delaware Supreme Court the question of Furman's effect on Delaware's sentencing scheme. That court held that Furman had made the Mercy Statute unconstitutional, leaving the Delaware penalty for murder at mandatory execution. However, in the court's view, application of the mandatory execution provision could not be made to Dickerson himself, for at the time of his illegal conduct the Delaware Community was not on notice that, for a first-degree conviction, capital punishment was mandatory.

Had the Court adopted in Dickerson's case the analysis which is now used to sustain the present conviction, it would have reasoned that Dickerson was on notice at the time of the homicide that the Mercy Statute might be found unconstitutional and, in that event, he would risk a mandatory death sentence under the separate first degree statute. Indeed, in one respect, that analysis would have been easier there than here because it would not have required reviving any superceded legislation. We think that Delaware's refusal to find fair notice of the mandatory death penalty in Dickerson was correct and that it points to a similar result as to notice of the revival of old 635 here.

In addition, in this case the State of Delaware is in no position to argue that new 635 gave adequate notice of its own invalidity. For the state strongly urged in opposition to the motion to quash the indictment that the new statute was valid. Indeed, it took the above mentioned abortive interlocutory appeal from the ruling of invalidity, although on Clark's appeal from his conviction the state did not see fit to present this issue. An even more striking indication that new 635 gives no notice of its own invalidity is the fact that at least one judge of the Delaware Superior Court has held, after and contrary to the ruling at Clark's trial, that the new section is valid.3

The dissenting opinion eschews this analysis. However, because it approaches our problem in a different way, it is difficult to bring our disagreement into sharp focus. It seems to place reliance upon the general Delaware doctrine concerning possible 'revival' of an antecedent statute as principal justification for the conclusion that Delaware gave the public adequate notice that old 635 retained vitality. We have pointed out that one can reach this conclusion only by positing adequate public notice, contrary to the state's own representation, of the infirmity of new 635. But beyond that, the Delaware revival concept does not mandate automatic revival of the antecedent statute, as the dissent seems to assume. Rather, the Delaware revival doctrine is that, after finding an invoked statute invalid, a Delaware court is empowered to decide as a separate and additional question whether or not the statute's predecessor should be revived. Under the prior Delaware cases,4 which the Delaware court followed here, the judge is required to decide for or against revival by exercising his judgment as to which course the legislature would prefer.

Thus, in our view, the position of the dissenting opinion can be maintained only if the public had adequate notice both that new 635 was likely to be invalidated and that a judge would then choose to disinter old 635 rather than letting it rest in peace. Such speculative anticipation of a series of future judicial rulings is not the stuff out of which adequate notice can be fashioned.

The foregoing reasoning also suggests that the reliance of the dissent upon such cases as Chaplinsky v. New Hampshire, 1942, 315 U.S. 568, 62 S. Ct. 766, 86 L. Ed. 1031, is misplaced. In Chaplinsky, restrictive reading of a prohibitory statute was necessary to prevent it from infringing First Amendment rights. The highest court of the state had given the statute restrictive interpretation before Chaplinsky's alleged violation. And the statute, as theretofore authoritatively construed, clearly prohibited Chaplinsky's conduct. As concerns fair notice, the Chaplinsky decision shows, at most, that the public may reasonably be charged with notice not only of the language of a statute but also of prior judicial determination of its meaning.5 In contrast, at the time of Clark's misconduct the Delaware courts had not even intimated that new 635 was invalid, much less decided whether in that event, they would choose to revive old 635.

We recognize that the invalidation of this conviction will cause the release of a prisoner who must have been aware that the use of his client's funds for his own purposes-- though he may have intended to restore what he took-- was morally wrong and professionally reprehensible. But the state's decision in this case that new 635 is too vague to inform the public what it prohibits, has created a situation in which we cannot square this conviction with an essential procedural requirement of due process of law. Moreover, this is not a case in which a wrongdoer escapes severe punishment. The appellant has been disbarred, disgraced and imprisoned for more than a year, and his release will not change that.

The judgment will be reversed and the district court directed to grant a writ of habeas corpus.

I join that portion of the majority's opinion affirming the district court's rejection of Clark's contention relating to allegedly prejudicial pretrial publicity, but respectfully dissent from that aspect sustaining Clark's due process objections to the Delaware practice of reinstating an 'amended' statute once a successor statute has been declared invalid.

The factual and legal situation giving rise to this appeal is not complicated. Clark was indicted for alleged violations of 50 Del.Laws Ch. 299, 11 Del.C. 635 (1955), which in essence made it unlawful '(to embezzle) money or other property.' Clark, an attorney and one-time municipal court judge, was accused of misappropriating funds belonging to clients for whom he had secured substantial recoveries for personal injuries. On Clark's motion to dismiss the indictments, the Delaware trial court agreed with Clark's contention that the term 'embezzles,' by itself, in section 635 of the 1955 Act was an unconstitutionally vague definition of the substantive offense. Instead of dismissing the indictments, however, the trial court followed established Delaware law and considered as 'revived' the statute that section 635 'amended,' or, in effect, replaced. The constitutionality of the revived statute has not been previously questioned and is not questioned here. Rather, only its applicability to Clark is here in issue.

Clark asserted that the Delaware practice of reviving statutes which, although styled 'amended,' have in effect been interred, or at least held in abeyance, by subsequent legislation deprived him of fair warning of the conduct proscribed by the legislature. Clark's contention was rejected by the state courts, and also was spurned by the district court when it dismissed Clark's petition for habeas corpus.

The majority, however, reverses the judgment of the district court, and in so doing also repudiates the conclusion of the state courts. While expressly disavowing any implication that Clark was unaware that his activities were wrongful when he made off with the clients' funds,1 the majority, nevertheless, holds that 'the notice requirement of the due process clause (does not permit) a state, after ruling one of its criminal statutes overly vague, to apply that statute's superseded predecessor to the defendant in the very case which ruled that successor statute unconstitutional.'2 The majority appears to adopt the position that the proper approach to the fair warning element of the doctrine known by the alternative rubrics of void for vagueness and unconstitutional indefiniteness3 is to track the steps by which the public determines the nature and scope of the particular statutory restriction at issue. Then if, in a court's judgment, those steps appear unduly circuitous, it must strike down as unconstitutional the specific rule that requires those steps to be taken to determine the criminal provision governing the conduct.

Accordingly, the majority concludes that to sustain the constitutionality of the Delaware practice, it would be necessary 'to reason, first, that the new 635 on its face gave adequate public notice of its own invalidity and, second, that the public, thus informed, was then put on further notice that the officially announced statutory repeal or supercession of old 635 was legally ineffective.' This, according to the majority, 'is too tortured and too far removed from reality to satisfy' the due process clause.

My disagreement with the majority is twofold. First, assuming, but not conceding, that the majority's analytical approach is valid, it is difficult to reconcile its determination that Delaware's reviving practice is constitutionally deficient with repeated declarations by the Supreme Court that judicial narrowing of broad or vague statutory language may save a statute that on its face appears unconstitutional.4 Second, and perhaps more fundamental, the fair warning element of the doctrine of unconstitutional indefiniteness is not, in my judgment, so divorced from reality as to be devoid of subjective content. Admittedly, when other diverse interests promoted by the doctrine of constitutional indefiniteness are implicated, it may be appropriate, in some cases, to overlook that a particular defendant expected to be criminally prosecuted if his alleged wrongdoing happened to be detected. But in my view, none of these other interests operate in this case.

The first area of my difference with the majority centers upon the similarity between the Delaware judicial practice of reviving statutes and the not uncommon judicial practice of narrowing statutes when a literal reading of their terms suggests that they are unconstitutional because of vagueness or over-breadth.

In Chaplinsky v. New Hampshire,5 the Supreme Court was presented with a state statute arguably vague as well as overbroad in the sense that, because of the broad sweep or uncertainty of its language, it trenched upon or 'chilled' First Amendment freedoms. The statute, however, came to the Supreme Court 'authoritatively construed by the highest court of New Hampshire' in that very case as well as in earlier cases.6 The Supreme Court, in refusing to overturn Chaplinsky's conviction under the statute, stated:

We are unable to say that the limited scope of the statute as thus construed contravenes the constitutional right of free expression . . .. This conclusion necessarily disposes of appellant's contention that the statute is so vague and indefinite as to render a conviction thereunder a violation of due process. A statute punishing verbal acts, carefully drawn so as not unduly to impair liberty of expression, is not too vague for a criminal law.7

In effect, the Supreme Court was saying that absent the narrowing construction of the New Hampshire court, Chaplinsky could not have been convicted under the statute, but that in light of the narrowing construction placed on the statute, the conviction could stand.

Thus, the Supreme Court in Chaplinsky did not suggest that the procedure of judicial construction of ambiguous statutory language is a process that because of its likely unfamiliarity to the nonlawyer, in and of itself, creates vagueness problems. Nor has the Supreme Court so indicated since Chaplinsky was decided.8

Although the Delaware reviving practice appears somewhat unusual, and not one that I would especially espouse, it would seem no more vulnerable to vagueness objections than the practice of judicial construction of terms in a statute that are uncertain. The majority's implicit assertion that it is fanciful to expect that a person can recognize that a statute is constitutionally suspect on its face applies with equal force to the practice of judicial construction and the reviving practice.

Hence, if the majority's position is correct, the judicial narrowing of the statute, in Chaplinsky itself as well as earlier state court cases, should not have saved Chaplinsky's conviction.9 This is so because Chaplinsky, like Clark, would have had to recognize the surface unconstitutionality of the statute in order to be on notice that judicial constructions may achieve or may have achieved a constitutionally acceptable narrowing. But no such conclusion was reached by the Supreme Court in Chaplinsky. To permit Clark to rely on the apparent improbability of such recognition would, therefore, seem to conflict with the philosophy inherent in the Supreme Court's disposition of Chaplinsky's claim.

The majority, in pursuing its analysis, attempts to distinguish the present case from the typical case involving judicial narrowing by asserting that the new section 635, although unconstitutionally vague for the purposes of this case,10 is not so vague as to put Clark on notice of its invalidity. Delaware's attempt to avert successful constitutional challenge to the statute, according to the majority, amount to concessions that the statute was not vague, and thus does not give notice of its own invalidity.

I have substantial difficulty grasping the exact nature of the majority's approach in this respect for two related reasons. First, vagueness objections generally focus on the imprecise language of the statute. If would appear, therefore, untenable to contend that the language of a statute is unconstitutionally vague, as Clark did in the trial court, and yet not so vague as to leave Clark confused as to the scope of its proscriptions. Second, assuming that the majority does not mean that a statute may be both unconstitutionally indefinite and still fail to give notice of its invalidity, this portion of its opinion, Majority Opinion at 1082, seems implicitly to reject the determination of the Delaware trial court, which is the law for the purposes of this case, that new section 635 is, in fact, unconstitutionally vague.

Moreover, it is certainly not surprising that Delaware has claimed that section 635 is not unconstitutional. Indeed, it is, no doubt, the duty of the state officials to do so. The majority, however, seems to posit, on the basis of the state's position with respect to new section 635, the novel and puzzling proposition that a statute can be unconstitutionally vague and, at the same time, fail to give notice of its invalidity because of vagueness.

In addition, according to the majority's reasoning, to sustain the constitutionality of the judicial construction procedure, it would be necessary to assume that the public is, at least, on notice that an earlier judicial construction may have saved an otherwise unconstitutional statute. It would seem no less realistic to suppose that the public is on notice that the apparent surface unconstitutionality of a statute claimed to be vague or overbroad requires recourse to the predecessor statute for enlightenment as to the precise nature of the legislation that is, in fact, operative. Thus, if the majority's approach is the proper one to apply when a defendant challenges a juridical practice, the Supreme Court's approval of judicial narrowing might well be seriously undermined.

The majority does suggest that the existence of a prior judicial construction, alone, constitutes notice to a person that he must look further than the bare words of the statute. However, if a statute is not vague on its face, a person would be unlikely to search for judicial interpretations. It is difficult, therefore, to understand how a prior judicial construction itself can be deemed an essential part of the fair warning principle.

Further, if a prior judicial construction were an aspect of the fair warning element, as the majority contends, one would expect the Supreme Court expressly to disapprove applying a judicial narrowing of a statute in other than a prospective manner. Yet, despite some language suggesting the contrary,11 the Supreme Court 'has invariably allowed (penal statutes) the benefit of whatever clarifying gloss state courts may have added in the course of the litigation of the very case at bar.'12

Indeed, Bouie v. City of Columbia,13 a case upon which Clark heavily relies, does not support the majority's result. In dealing with a novel interpretation by the appellate court of South Carolina that broadened the scope of an otherwise precisely drawn statute, the Supreme Court pointed out that the statute in question would not have prohibited the defendants' conduct in view of the statute's clear and unambiguous language. If, prior to acting, the defendants in Bouie had read the statute at issue, they would have concluded, according to the Supreme Court, that there was no occasion to look to judicial interpretations to determine whether their contemplated conduct was proscribed. The limitations on conduct embodied in the South Carolina statute were clear from its terms.14 The Supreme Court in Bouie, therefore, did not undercut the accepted axiom, implicit in Chaplinsky and subsequent cases, that judicial construction may save a statute sufficiently uncertain on its face so as to raise substantial doubts regarding its constitutionality. Instead, the Supreme Court, in Bouie, prevented the state court from making conduct-- innocent when performed-- criminal by way of the device of subsequent and unexpected judicial expansion of statutory language.

The doctrine of unconstitutional indefiniteness serves as a bulwark against the destruction of salient constitutional and societal values. In Grayned v. City of Rockford,15 Justice Marshall offered the following succinct formulation of the concerns underpinning the doctrine of unconstitutional indefiniteness:

First, because we assume that a man is free to steer between lawful and unlawful conduct, we insist that laws give the person of ordinary intelligence a reasonable opportunity to know what is prohibited, so that he may act accordingly. Vague laws may trap the innocent by not providing fair warning. Second, if arbitrary and discriminatory enforcement is to be prevented, laws must provide explicit standards for those who apply them. A vague law impermissibly delegates basic policy matters to policemen, judges, and juries for resolution on an ad hoc and subjective basis, with the attendant dangers of arbitrary and discriminatory application. Third, but related, where a vague statute 'abut(s) upon sensitive areas of basic First Amendment freedoms,' it 'operates to inhibit the exercise of (those) freedoms.' Uncertain meanings inevitably lead citizens to "steer far wider of the unlawful zone' . . . than if the boundaries of the forbidden areas were clearly marked.'16

Thus, under the teachings of Grayned, it would appear necessary to inquire (1) whether a defendant, such as Clark, had fair warning of the criminal nature of his conduct, (2) whether the statute at issue lends itself to discriminatory and arbitrary enforcement, and (3) whether the exercise of First Amendment rights may be stultified.

The seeming variance between the majority's position in this case and the Supreme Court's refusal in cases like Chaplinsky to declare the practice of judicial narrowing improper derives from the majority's perhaps too subtle approach to the fair warning element of the doctrine of constitutional indefiniteness. It is unrealistic to assume that an individual usually surveys the applicable criminal code to determine whether contemplated conduct is, in fact, prohibited. The Supreme Court has not yet squarely held that the fair warning element possesses subjective content, but the language of the Supreme Court suggests that the test of fair warning is, at least to some extent, subjective.

In United States v. Harriss,17 the Court was presented with a vagueness challenge to the Federal Regulation of Lobbying Act. In upholding the statute, the Court stated:

The constitutional requirement of definiteness is violated by a criminal statute that fails to give a person of ordinary intelligence fair notice that his contemplated conduct is forbidden by the statute. The underlying principle is that no man shall be held criminally responsible for conduct which he could not reasonably understand to be proscribed.18

The Harriss language would seem to impart the notion that an individual who understood that his conduct, if detected, might subject him to criminal prosecution could not invoke the absence of fair notice to avoid conviction. The Court, in United States v. National Dairy Corporation,19 lent credence to this construction of its language in Harriss when it stated that 'in determining the sufficiency of the notice a statute must of necessity be examined in the light of the conduct with which a defendant is charged.'20

Apart from exposure to the language of criminal statutes, an individual acquires an understanding of the wrongfulness of classes of conduct in a variety of recondite ways. In fact, it is by informal enculturation that the typical member of society generally acquires knowledge of the standards of conduct to which he is expected to conform. It is implausible to suggest that significant numbers of those accused of criminal conduct have ever read the statutes pursuant to which they are charged. And in Connally v. General Construction Company,21 the Supreme Court indicated an awareness of the complexity of the process when it stated:

It will be enough for present purposes to say generally that the decisions of the court, upholding statutes as sufficiently certain, rested upon the conclusion that they employed words or phrases having a techical or other special meaning, well enough known to enable those within their reach to correctly apply them, or a well-settled common-law meaning, notwithstanding an element of degree in the definition as to which estimates might differ, or, as broadly stated by Mr. Chief Justice White 'that, for reasons found to result either from the test of the statutes involved or the subjects with which they dealt, a standard of some sort was afforded.'22

The majority concedes that Clark was, no doubt, aware of the nefarious nature of his conduct. Nevertheless, it seems to find some solace in the fact that embezzlement as such did not constitute a crime at common law. Embezzlement, however, has been a statutory crime in Delaware since 1879.23 Considering the type of activity-- purloining for personal purposes funds belonging to clients-- and its well established criminal character, it is not possible to conclude that Clark, a lawyer, was somehow unaware of the wrongfulness of his actions.

2. Discriminatory and Arbitrary Enforcement

Although Clark certainly expected that his conduct, if detected, would lead to criminal prosecution, we may still be required to strike down the Delaware practice if it implicates some of the other interests protected by the doctrine of constitutional indefiniteness. 'Although a statute may be neither vague, overboard, nor otherwise invalid as applied to the conduct charged against a particular defendant, he is permitted to raise its vagueness or unconstitutional over-breadth as applied to others.'24 When the uncertain language of a statute promotes discriminatory and arbitrary enforcement, a person clearly within the statute's bounds may be permitted to challenge its constitutionality.

In the present case, there is no representation, affirmatively or by implication, that the Delaware practice of reviving is employed on an arbitrary and discriminatory basis. Neither is it alleged that permitting a revived statute to operate in the very case in which its successor was held invalid was done in an arbitrary and discriminatory manner.

The majority, however, devotes considerable discussion to the Delaware Supreme Court's decision in State v. Dickerson.25 Although the reasons for the majority's attention to the Dickerson case are not specifically explained, the majority may be suggesting that the reviving procedure is not consistently invoked by the Delaware courts.

In Dickerson, the court refused to impose a mandatory death sentence even though it held that the provision allowing for discretionary imposition of a sentence other than death was unconstitutional.26 To order the death penalty in these circumstances, according to the Delaware court, would amount to the imposition of an increased penalty ex post facto as to Dickerson. But the majority here admits that the Delaware court in Dickerson did not regard the invalidation of the mercy provision and the concomitant establishment of the mandatory death sentence as an example of the reviving practice. Indeed, in a supplemental opinion in the Dickerson case itself, the Delaware court expressly used the reviving principal to confer on the Superior Court the power to require the mandatory life sentence and, in the process, cited the earlier state court decision in the case with which we are now concerned.27

Thus, if my understanding of Dickerson is correct, it does not stand for the proposition that the conditions under which the reviving practice comes into play are so vague that the practice is susceptible to employment in an arbitrary and discriminatory manner.

Neither can I regard the fact that the reviving practice is not automatically invoked once a statute is declared unconstitutional as a strong indication of discriminatory and arbitrary or even haphazard employment of the practice. The majority recognizes that a Delaware judge is directed to determine the intent of the legislature before deciding whether to revive the amended predecessor of a constitutionally infirm statute. Accordingly, a Delaware judge is not vested with absolute discretion, but must make a principled decision whether to revive or not to revive a statute.

Keeping in mind it is the Delaware reviving practice, not the statute so revived, that the majority finds objectionable, the practice is not equivalent to the shirking by the legislature of the responsibility of writing reasonably precise criminal statutes, an important ingredient of the vagueness doctrine. Rather, the reviving practice is of judicial origin, and it necessarily results in giving effect to a statute that is drawn precisely enough to obviate constitutional objections.

It cannot be suggested that the reviving practice has, by its own force, an inhibitory effect on the exercise of First Amendment rights. Since the reviving practice itself contains no proscriptions affecting First Amendment rights, it is only when the statute declared unconstitutional touches upon First Amendment rights or when an overbroad statute is revived pursuant to such practice that freedom of speech may be implicated.28 In that event, it would not be the reviving practice that would be subject to constitutional objection but, instead, the overbroad statute.

The foregoing analysis leads, it would seem, to the following conclusions. First, the majority's approach to the Delaware reviving practice suggests, when applied to the judicial narrowing of uncertain statutory terms, a conclusion at odds with that embodied in several Supreme Court decisions in the void for vagueness area. Second, the fair warning aspect of the doctrine of unconstitutional indefiniteness is not completely devoid of subjective content, and Clark's awareness of the wrongfulness of his conduct precludes his successful assertion of the absence of fair warning to void his conviction. Third, the reviving practice does not raise the specter of arbitrary and discriminatory enforcement nor does it, eo sponte, fetter the exercise of First Amendment rights-- evils that prompt the invocation of the doctrine of unconstitutional indefiniteness.

Because the considerations giving rise to the doctrine of unconstitutional indefiniteness do not, in my judgment, operate in this case and because Clark has been convicted of conduct which he necessarily knew to be illegal, I would affirm the judgment of the district court.