Opinion for the Court filed by Circuit Judge WILKEY.
Dissenting opinion filed by Circuit Judge GINSBURG.
WILKEY, Circuit Judge:The resolution by this and other courts of the question posed by this case may significantly affect the prospects for ongoing reform of the nation’s parole systems.
Like most of its state counterparts, the United States Board of Parole operated for many years under a statutory charter granting it almost unlimited and unreviewable discretion in parole decisionmaking. Because the Board was composed of less than ten members, its day-to-day decisions were of necessity largely made by its hearing examiners throughout the country. These decisions, undertaken without the protection of much in the way of either procedural safeguards or substantive guideposts, were often criticized as erratic and unfair and the system was widely perceived as ripe for reform.
As a consequence, the Board instituted major reforms which Congress subsequently ratified in 1976 by passing the Parole Commission and Reorganization Act. The new system not only operates with much enhanced procedural regularity, but enables the Board, now renamed the Parole Commission, to direct and channel the exercise of discretion by its hearing examiners by means of substantive guidelines which suggest the period of confinement the Board considers usually to be appropriate for convicts sharing similar criminal and community backgrounds.
The appellant here challenges the operation of this system by suggesting that the manner in which the parole authorities exercise their discretion is so central a determinant of his punishment that the ex post facto clauses of the Constitution prohibit parole decisions on any basis other than that used by the Parole Board at the time of his crime. In particular, the appellant protests the application of the new guidelines, promulgated well after his crime, conviction and sentencing, to assist the Commission in deciding whether he should be reparoled after he was found guilty of fresh crimes committed while he was on parole, and then reincarcerated.
A ruling favorable to the appellant would imply that the guidelines now being used by the Commission are annexed to the sentence of prisoners convicted now and in the future. This would mean that, contrary to the will of Congress, the guidelines could not be altered unless the Commission were willing to operate under a multiplicity of standards applicable to individual convicts depending on the dates of their crimes.
We do not believe the ex post facto clause demands such rigid application.
I. THE FACTS AND PROCEDURAL BACKGROUND
After pleading guilty to armed bank robbery, Charles R. Warren was sentenced in *185March 1969 to twenty years in federal prison, but seven years later, in April 1976, was released on parole. While at liberty he not only committed another armed bank robbery but was also convicted of shoplifting. So in July 1977 his parole was revoked, and he returned to prison. He now appeals the district court’s summary dismissal of a habeas corpus petition he brought in forma pauperis. The district court invoked 28 U.S.C. § 1915(d), which authorizes such a disposition whenever the court is “satisfied that the action is frivolous or malicious.”
In his petition to the district court Warren raised three issues. Two are without substance; they were properly dismissed by the district court. One of these two claims — that certain parole hearings were held “out-of-time” — Warren has not pursued on appeal. The other — that Warren was denied due process when he was not permitted a fellow inmate’s assistance throughout one of his parole hearings — runs afoul of a recent Supreme Court case establishing that an inmate does not have a right to assistance at a parole release hearing.1 To escape the force of that authority Warren has attempted to argue that the parole hearing in question, although it was held more than a year after the revocation of his parole, somehow constituted a revocation hearing rather than a release hearing. We find no merit in this contention, and pursue it no further here.
Warren’s remaining claim is his ex post facto claim, which is hardly “frivolous.” It is based on the fact that when Warren was first convicted of bank robbery in 1969 the United States Board of Parole was still exercising its discretion without reference to any explicit standards other than its statutory mandate;2 not until 1973 did the Board begin using a set of guidelines distilled from its prior practice.3 But those guidelines, adopted four years after Warren was first convicted, are now being used to determine his eligibility for rerelease after the revocation of his parole. Warren claims that the application to his case of guidelines promulgated after his first conviction violates the federal ex post facto clause of the Constitution.4
Warren’s appeal raises this question regarding the reach of the ex post facto prohibition for the first time in this Circuit. The Second,5 Sixth,6 Seventh,7 and Ninth Circuits 8 have rejected similar claims, but in a somewhat analogous ease the Third *186Circuit ruled otherwise.9 The Supreme Court granted certiorari to consider the Third Circuit decision, but ultimately remanded the case without reaching the issue. Thus Warren’s claim is certainly at least colorable. We conclude that the district court acted improperly in dismissing the appellant’s petition under the authority of 28 U.S.C. § 1915(d). Were material facts at issue, we would be required to remand this case to the district court for the required hearing. That step is unnecessary, however, because we conclude that the petitioner’s claim, though not frivolous, is without merit as a matter of law. We therefore affirm.
II. THE EX POST FACTO GUARANTEE
Before determining whether the ex post facto clause applies to Warren’s case, it is helpful briefly to consider the nature of the ex post facto guarantee. The Constitution prohibits both Congress and the various states from enacting ex post facto laws.10 The scope of these provisions was clarified in 1798 when the Supreme Court first considered the reach of the ex post facto clauses in the much cited case of Calder v. Bull.11 Our understanding of the basis of the prohibition against ex' post facto laws has changed remarkably little since that time.
In Calder, Justice Chase, noting that the expression “ex post facto” “had been in use long before the revolution, and had acquired an appropriate meaning, by legislators, lawyers and authors,”12 summarized what manner of laws he understood to be “within the words and the intent of the prohibition”:13
1st. Every law that makes an action done before the passing of the law, and which was innocent when done, criminal; and punishes such action. 2d. Every law that aggravates a crime, or makes it greater than it was, when committed. 3d. Every law that changes the punishment, and inflicts a greater punishment, than the law annexed to the crime, when committed. 4th. Every law that alters the legal rules of evidence, and receives less, or different testimony, than the law required at the time of the commission of the offence, in order to convict the offender.14
Having thus set forth his understanding of the scope of the ex post facto prohibition,15 Justice Chase indicated why the *187clauses were included in the Constitution. The prohibition against the ex post facto laws, he wrote,
very probably arose from the knowledge, that the parliament of Great Britain claimed and exercised a power to pass such laws .... These acts were legislative judgments; and an exercise of judicial power. ... With very few exceptions, the advocates of such laws were stimulated by ambition, or personal resentment and vindictive malice. To prevent such and similar acts of violence and injustice, I believe, the federal and state legislatures were prohibited from passing any ... ex post facto law.16
Justices Paterson and Iredell, in their own separate opinions in Calder v. Bull, made clear their agreement with Justice Chase that the ex post facto provisions were motivated by a dreary history of legislative abuses. In Justice Paterson’s words: “The historic page abundantly evinces, that the power of passing such laws should be withheld from legislators; as it is a dangerous instrument in the hands of bold, unprincipled, aspiring and party men, and has been too often used to effect the most detestable purposes.”17 Justice Iredell enlarged at some length on the evils traceable to ex post facto enactments:
The history of every country in Europe will furnish flagrant instances of tyranny exercised under the pretext of penal dispensations. Rival factions, in their efforts to cross each other, have superseded all the forms, and suppressed all the sentiments of justice; while attainders, on the principle of retaliation and proscription, have marked all the vicissitudes of party triumph. The temptation to such abuses of power is unfortunately too alluring for human virtue; and therefore, the framers of the American constitutions have wisely denied to the respective legislatures, federal as well as state, the possession of the power itself: they shall not pass any ex post facto law; or, in other words, they shall not inflict a punishment for any act, which was innocent at the time it was committed; nor increase the degree of punishment previously denounced for any specific of-fence.18
From the outset, then, the ex post facto clauses have been understood to have been principally aimed at curtailing legislative abuses.19 A reading of Calder v. Bull as well as other early authority suggests, however, that the ex post facto provisions were meant to serve other ancillary functions as well. One such derivative purpose was neatly summarized by Justice Chase in Calder v. Bull itself: “the very nature of our free republican governments,” he wrote, implies “that no man should be compelled to do what the laws do not require; nor to refrain from acts which the laws *188permit.”20 Justice Chase thus expressed a libertarian ideal appropriate for a pluralist society in which universal consensus will rarely exist as to the immorality of lawful acts.21 Because an individual in such a society cannot be charged with knowledge that his lawful acts are immoral in the absence of an existing criminal enactment, the element of mens rea cannot be assumed to exist.22 The legislature therefore is prohibited from retroactively imposing criminal penalties for the perpetration of lawful acts perhaps considered immoral only by some. In a society committed to liberty and not governed by orthodoxy the presumption must be that acts not specifically prohibited are permitted. Such a presumption guarantees that the citizenry may feel secure in acting in reliance on existing law and assures that fair notice will be given of any change.
The elements of fair notice and reasonable reliance are closely associated with another basis for the ex post facto proscriptions. Because an ex post facto law fails to provide fair warning, it cannot serve the core purpose of the criminal law, to regulate behavior by threatening unpleasant consequences should an individual commit a harmful act. Obviously, when a law is enacted after the fact, the time for threats has already passed. Quoting Blackstone, Justice Paterson seems to allude to this failure in Calder v. Bull.23 An early Massachusetts case sets forth the objection more explicitly:
The reason why these laws are so universally condemned is, that they overlook the great object of all criminal law, which is, to hold up the fear and certainty of punishment as a counteracting motive, to the minds of persons tempted to crime, to prevent them from committing it. But a punishment prescribed after an act is done, cannot, of course, present any such motive.24
Of course, the other goals of the modern criminal law — rehabilitation of the criminal, retribution, incapacitation of the criminal so that, at least for a time, he can perpetrate no further harm, and the education and general deterrence of the public at large from committing similar acts — can all be satisfied to some degree by ex post facto legislation. The constitutional ban on ex post facto laws, however, suggests that the framers considered the possibility of special deterrence a prerequisite to the imposition of specifically criminal penalties. The framers’ position may have been based on *189the idea that, because special deterrence is so central to the criminal law, enactment of a criminal statute that cannot serve this function raises a strong presumption that the legislature’s motives are impermissible. Since judicial inquiry into the motives of the legislature is difficult and unseemly, the framers may have considered it the better course to ban such legislation from the start.
In sum, there appear to be three broad bases for the constitutional prohibition of ex post facto legislation. First, and undoubtedly the most significant, is the invitation to legislative abuse ex post facto powers entail. Second is the frustration of reasonable reliance on existing laws and the consequent insecurity and infringement of liberty to which the citizenry would be subject were ex post facto legislation permitted. Finally, the ban on ex post facto legislation assures that the legislature can make recourse to the stigmatizing penalties of the criminal law only when its core purpose of deterrence could thereby possibly be served.25
III. THE PAROLE SYSTEM: THEN AND NOW
With an understanding of the purposes of the ex post facto clauses in mind, we can turn to consider the question whether the application of the new parole guidelines to Warren’s case somehow violates those prohibitions. In order to do so, however, we must first examine both how the parole system worked at the time Warren committed the crime for which he is now incarcerated and how it has been changed since then.
It is well to begin with some background.26 Our present parole system is the product of more than a century of ongoing reform in the post-conviction treatment of criminals. In the early nineteenth century the purpose of imprisonment generally was understood to be deterrence and retribution alone. Punishment was designed to fit the crime and legislatures specified with precision the period of incarceration appropriate to each, leaving the sentencing judge no latitude for variation. In time, it became obvious that this system of fixed sentences was unduly rigid in that it did not permit the sentencing judge to tailor the punishment to the concrete particulars of each defendant’s case, with due regard for individual mitigating and aggravating circumstances. As a result, legislatures began to give judges a range of possible sentences from which to choose. Deterrence and retribution nonetheless remained the principal justifications for incarceration.
Beginning in the latter part of the nineteenth century, however, the view that punishment should fit not the crime but the criminal gained currency. Raw vengeance and straightforward deterrence gradually were replaced by rehabilitation as the principal theoretical basis for imprisonment. Adherents of this new position argued that the convict’s experience during his imprisonment could be structured to motivate him to abandon his criminal disposition for that of a law-abiding citizen. Once this metamorphosis was accomplished, the aim of incarceration would be fulfilled, and further *190confinement could serve no purpose. Until rehabilitation had been achieved, on the other hand, a potential recidivist could not responsibly be returned to society.
Under this penal theory the appropriate period of incarceration could not be determined at sentencing, because the sentencing judge is usually not in a position to predict the term of correctional “therapy” necessary to rehabilitate a criminal to the point that his release would not pose an unacceptable risk to society. A corollary of the rehabilitative model in its purest form, then, is the indeterminate sentence, which leaves the task of determining when it is safe to release an offender to trained correctional and parole personnel.27 Under this approach a quickly rehabilitated murderer could be released rather promptly, while an incorrigible petty thief might languish in jail forever.
Even at the peak of its fashion, however, the rehabilitative model was not wholly accepted in most jurisdictions, but instead was moderated' to the extent that legislatures and judges retained their power to set maximum and minimum sentences. The legislature, operating at the level of greatest generality, would typically undertake to prescribe only a broad range of possible sentences. Then, the sentencing judge, confronting a particular criminal convicted for committing a specific harmful act, would narrow the range by selecting a maximum, and perhaps also a minimum, term from the broader interval permitted by the legislature, with an eye both to the mitigating and aggravating circumstances of the crime and the criminal’s prospects for rehabilitation. Finally, a parole authority would periodically review the criminal’s progress toward rehabilitation, releasing him when his imprisonment had served its purpose.
A system of this type was in use in the federal jurisdiction when Warren was sentenced in 1969.28 Under the sentence most commonly used at that time, the so-called “regular adult” sentence, a federal inmate became eligible for parole after serving one-third of the sentence imposed by the judge.29 Once eligible, an inmate’s application would be considered by a hearing examiner 30 who was guided only by the rather vague and general statutory directive that a prisoner could be released on parole if (1) “there [was] a reasonable probability that such prisoner [would] live and remain at liberty without violating the laws,” and (2) “such release [was] not incompatible with the welfare of society.” 31
*191The hearing examiner reached his discretionary decision from seeing and talking with the inmate, perhaps detecting signs of repentance and reform, and from an ad hoc examination of the reports of the offender’s institutional adjustment, prior record, and so forth. After arriving at his conclusion, the examiner would forward his report to the Washington, D. C., offices of the Board where the members of the Board would vote without discussion on the case. At no point in the process was the inmate permitted representation, the information being used by the examiner to make the decision revealed, or an explanation of the basis of the decision provided. “The inmate was merely to appear, to be scrutinized, to answer the questions put and ask no others, and to assume the attitude he or she believed best indicated ‘readiness’ for parole.”32
At the conclusion of this process, judicial review was almost never available. As the Second Circuit put it in a widely cited case decided the year after Warren was convicted:
[The prisoner] is entitled only to be released after full service of his sentence less good time earned during incarceration. The Board is given absolute and exclusive discretion to decide whether or not to initiate parole release proceedings and, if so, whether parole should be granted to him. [The prisoner] has been constitutionally deprived of his right to liberty for the period of his sentence. Like an alien seeking entry into the United States (as distinguished from a lawful resident alien) he does not qualify for procedural due process in seeking parole.33
The system thus had little if anything in the way of procedural safeguards, and called for an unstructured exercise of discretion that could easily produce arbitrary results. To remedy both these defects the Board began a series of experimental programs and pilot projects in the early part of the last decade.34 This effort culminated in the Parole Commission and Reorganization Act of 1976,35 which provided a statutory basis for a variety of procedural protections not relevant here, as well as for a structuring of the exercise of discretion by parole officials by means of guidelines promulgated by the Parole Commission. It is these guidelines with which we are concerned in this case.
Under the guideline system, hearing examiners no longer render their recommendations at large, but instead by reference to a set of guideposts which indicate the customary time to be served by convicts meeting certain offense severity and parole prognosis characteristics.36 An inmate being considered for parole is first assigned a “salient factor score” statistically designed to aid in predicting the risk of parole violation by reference to nine weighted personal attributes, such as employment and drug use history. The salient factor score is then used to determine the inmate’s “parole prognosis” on a scale from “poor” to “very good.” Next, the inmate’s crime is rated on *192an “offense severity scale” ranging from “low” to “very high.” The recommended range of time to be served for the resulting combination of parole prognosis and offense severity scores is then found in a table of guidelines.
When an inmate is returned to prison after having committed a new crime while on parole, his term as a parole violator is also determined by reference to the guideline table, but using the offense severity of his fresh crime, with his previous salient factor score overridden by the fact of his parole violation, so that he is automatically assigned to the “poor risk” column of the table.37
The guidelines for reparole, like the guidelines for parole, however, are just that —guidelines. In the words of the regulation itself: “A decision outside these guidelines (either above or below) may be made when circumstances warrant. For example, violations of an assaultive nature or by a person with a history of repeated parole failure may warrant a decision above the guidelines.”38 The legislative history of the 1976 Act also sets out additional examples of circumstances in which deviation from the guidelines would be appropriate. Incarceration for a period in excess of the guideline recommendation is suggested for those convicted of offenses requiring unusual degrees of sophistication and planning or involving participation in a large scale conspiracy or continuing criminal enterprise, while a prisoner’s adverse family or health situation is suggested as an indication that a below-guideline term might be appropriate.39 And, in order to ensure individual consideration beyond the mechanical application of the guidelines, Congress enacted a special section of the Act detailing the additional information the Commission is to consider, including the recommendations of the prison staff, presentence investigation reports, and reports of physical, mental and psychiatric examinations of the offender, and “such additional relevant information concerning the prisoner (including information submitted by the prisoner) as may be reasonably available.” 40
By its practice the Commission has demonstrated its willingness to deviate from the guidelines in exercising its discretion.41 Nonetheless, the guidelines have considerable force, and the Commission apparently generally adheres to them, at one time complying with the guidelines in a little over ninety percent of the cases before it.42 This is entirely proper under the statute, which specifies that a deviation from the guidelines requires “good cause,” which must be explained to the inmate in a written notice stating “with particularity” the reasons for the deviation.43 “Good cause” is defined in the legislative history as “substantial reason” including “only those grounds put forward by the Commission in good faith and which are not arbitrary, irrational, unreasonable, irrelevant or capricious.” 44
The requirement of “good cause” is consistent with the intent of Congress that the guidelines “serve as a national parole policy which seeks to achieve both equity between individual cases and a uniform measure of justice.”45 Because overly frequent deviations from the guidelines would undercut these purposes, Congress has indicated that the Commission should reevaluate the guidelines if it finds itself forced with ex*193cessive frequency to deviate from them,46 and the Commission in its regulations has expressly retained this authority.47
In summary, when Warren committed the crime for which he is now incarcerated, the United States Parole Board was exercising its discretion on an ad hoc basis, in the absence of any explicitly articulated guidelines. Now, its successor agency, the United States Parole Commission, exercises its discretion in a more structured fashion by reference to a framework of guidelines indicating the length of incarceration recommended for classes of offenders defined by offense severity and parole prognosis criteria. Warren claims the application of these guidelines subjects him to a constitutionally impermissible ex post facto increase in his punishment.
IV. ANALYSIS
We are now prepared to undertake the task of determining whether the ex post facto clause forbids the Parole Commission from referring to its guidelines when it exercises its discretion whether to reparole Warren.
A. Has Warren’s Punishment Been Increased?
To offend the ex post facto clause the guidelines must somehow increase Warren’s punishment. The existence of the guidelines, however, may not in fact augment Warren’s punishment, either actually or potentially. First, the Parole Commission may choose not to follow the guidelines in Warren’s case. It may reparole him earlier than suggested by the guidelines. Or it may refuse to reparole him in spite of guideline recommendations to the contrary. Whether to follow the guidelines or to change the guidelines is, after all, at the Commission’s discretion.
But even if the Commission chooses to follow the guidelines, it cannot be said that Warren is thereby worse off than he would have been under the old system; no one knows what the old Parole Board would have done in Warren’s case. Had it continued to operate, the Board might have reacted with great disfavor to the news that Warren had committed two crimes while on parole and refused ever to reparole him. If that had been its decision, Warren would have had no grounds to complain and no recourse elsewhere.48
Furthermore, we cannot say even that the category of prisoners to which Warren belongs (based on his salient factor and offense severity scores) has been disadvantaged by the introduction of the guideline system. The Commission’s guideline table grew directly out of the Parole Board’s past practice. The offense severity index, for example, is simply an average of the subjective offense severity evaluations of individual hearing examiners and Parole Board members.49 Similarly, the salient factor score is based on the Board’s experience with the success rate on parole of groups of inmates with similar characteristics.50 As a result, the guidelines embody what may well have been the Board’s practice anyway.
The adoption of the guidelines has had one effect of which we can be fairly certain, however. Great deviations from the average term must now be far less frequent than before. Warren must now be relatively unlikely to receive a reparole term much below — or above — the average. But it is unclear that this smaller random “scatter” disadvantages Warren, because the reduction in his chance of receiving a far below-*194average reparóle term has been compensated by a corresponding decrease in his chance of receiving a far above-average term.
Thus the guidelines may not offend the ex post facto clause in Warren’s case simply because they may not have worsened his position. But we do not base our resolution of this case on this ground alone. For the purposes of this case we can assume that the existence of the guidelines somehow does operate to Warren’s detriment.
B. Are the Policies of the Ex Post Facto Clause Implicated?
There is another point to be made at the outset of our analysis: Warren’s case implicates the policies behind the ex post facto clause hardly at all. This is because the reparole guidelines were promulgated after Warren committed his first crime but before Warren was released on parole and before he committed his fresh offenses. This means, as a consideration of the aims of the ex post facto prohibition demonstrates, that the use of the guidelines in Warren’s case cannot offend the policies behind the ex post facto clause.
For after the promulgation of the guidelines, Warren was not powerless to avoid their impact. He knew — or is charged with notice51- — of the new guidelines and the consequences that would await him should he commit another crime while on parole. Moreover, the Parole Commission, when it promulgated the guidelines, had no way to know which convicts then incarcerated, if any, would commit crimes while on parole in the future and thereby subject themselves to the force of the new system. For these reasons none of the principal evils the ex post facto clause was designed to prevent is threatened by the use of the guidelines to aid consideration of Warren’s case.
First, the adoption of the guidelines could not be used to perpetrate legislative abuses of the type the clause was principally designed to prevent, because at the time the guidelines were promulgated the parole authorities had no way to determine whom the guidelines would affect. Thus no one in Warren’s position could have been unfairly singled out for vindictive or malicious condemnation. Second, Warren was provided fair notice on which he could rely regarding the consequences of prohibited conduct while he was released on parole. And finally, assuming hypothetically that the guidelines do in fact add to an inmate’s punishment for crimes committed on parole, the imposition of the guidelines serves the core purpose of the criminal law, special deterrence of future crime. There is thus no danger the guidelines could be used inappropriately to invoke the drastic stigma and focused penalties of the criminal law in circumstances in which the central aim of the criminal law could not possibly be achieved. In sum, as applied to Warren and others like him, the guidelines do not offend the policies behind the ex post facto clause.
Support for this position can be found in Gryger v. Burke.52 In that case the Supreme Court upheld a recidivist statute calling for augmented penalties for subsequent convictions for the same offense, as applied to a criminal whose first offense occurred before the statute was passed. The recidivist statute in Gryger enhanced the penalty to be imposed on past wrongdoers for future misconduct, just as the reparole guidelines hypothetically augment the penalty for the future crimes of past wrongdoers committed while on parole. Both, upon enactment or promulgation, affected past wrongdoers, but neither runs afoul of the aims of the ex post facto clause because the operation of each can be triggered only by future crimes. To put this in other words, in Warren’s case the reparole guidelines would not have affected him but for his *195subsequent criminal convictions.53 Gryger thus appears to dispose of Warren’s case on the basis that the guidelines — if they punish at all — punish acts committed after they were promulgated.
For formal, technical reasons, however, Gryger may not control. The problem is this: Warren is incarcerated for a federal crime but the crimes he committed while on parole were state crimes. The United States Parole Commission is without authority to punish Warren for his state offenses; whatever action it takes in cognizance of those crimes can only be part of the punishment for Warren’s federal crimes. That is, strictly speaking, when the Parole Commission judges Warren unfit for reparole on the basis that he committed state crimes while on parole, it is not actually punishing him for those state crimes— that is for the state authorities — but merely using the fact of those crimes as a basis for prediction of Warren’s reparole prognosis. Not until Warren is turned over to the state authorities will his punishment for his state offenses begin. Formally, then, the guidelines are an incident of Warren’s punishment for his first crime, which was committed before the guidelines were promulgated. Thus, while the reparole guidelines may operate for ex post facto purposes very much like the recidivist statute in Gryger, such functional considerations may be formally inapposite.54
C. Being at the Mercy of the Parole Commission is Part of the Punishment
Thus, while it is uncertain that the guidelines in fact work to Warren’s detriment, and while the functional basis of the ex post facto clause is not substantially implicated in Warren’s case, we do not rest our decision on those grounds alone. What clinches the case is that Warren was sentenced to be held in federal prison at the discretion — howsoever exercised — of the parole authorities, so that guidelines which merely rationalize the exercise of that discretion do not offend the ex post facto clause.
The key here is that Warren was not sentenced under a law fixing a definite term of imprisonment, or even an indefinite but more or less predictable term. Warren was convicted under a statute committing the determination of his period of confinement to the almost unfettered discretion of the Parole Board to be exercised at some future date. In fact, at the time of Warren’s crime and conviction, the law failed to entitle Warren even to an explanation of the Board’s action in his case. Moreover, because the Board’s actions were utterly within its discretion, the law did not provide any meaningful recourse to review elsewhere. A prisoner seeking parole was like “an alien seeking entry into the United States.”55
A change merely in the manner in which the Board, now the Commission, exercises its discretion thus cannot offend the ex post facto clause. The punishment prescribed for Warren’s armed bank robbery was service of a minimum term in prison plus whatever additional term (up to a maximum set at the time of sentencing) as the *196Board, in its sole discretion, should later determine. The power the Parole Commission now has over Warren is an essential part of the punishment for the crime he committed. The existence of that power of discretion in the Commission has not been altered since Warren committed his crime. So, although the manner in which the Commission exercises its discretion has changed — perhaps for the better — Warren’s punishment has not been augmented since his conviction. No promise, real or implied, was ever held out that the Board would exercise its discretion in some set fashion. The Board was free to act as it saw fit, and such was or could have been known to Warren at the time of his plea bargain.
Warren is charged with the knowledge at the time of his crime and conviction that any number of possible factors could enter into the Board’s future exercise of its discretion, with many of those factors susceptible to change over time. One factor, for example, could be the extent of overcrowding in federal prisons. A high level of overcrowding when Warren came up for parole might tilt the Board in the direction of releasing him to make room for even more dangerous criminals or to prevent his progress from actually being impeded by further confinement in an overcrowded institution. On the other hand, a situation in which federal prisons had only partial occupancy, perhaps as a result of an utterly unpredictable spurt in the construction of new facilities, might propel the Board to confine Warren for a longer duration. Either way, Warren cannot complain that a change in the Board’s calculations after the date of his crime and conviction has violated the ex post facto clause. Nor, for example, could he complain that, shortly before his case came up for review, a photocopy of a scholarly article espousing the view that men like Warren rarely can be rehabilitated had been passed from Board member to Board member. And so forth.
In short, Warren was not sentenced to be considered for parole and reparole by a Board acting as the Board would have acted had it met to pass on his case at the time of his conviction. He was sentenced to be considered for parole at some much later time, when the nation’s circumstances, the Board’s membership, and the prevailing views of penologists all could have shifted against him. Under the penal theory behind the parole system, Warren’s sentence was deliberately designed to be indeterminate within a broad range so that the precise date of his release could be determined by the best professional judgment available at the time of his release as to his prospects for a law-abiding life, among other things. It is contrary to the spirit of that theory to freeze the exercise of discretion by the parole authorities at the moment of Warren’s crime. The punishment prescribed for Warren was to be held after his minimum term at the mercy of parole authorities exercising their judgment as best they could. The mere fact that the Parole Commission’s mercy is now channeled, structured and rationalized by a formal system of guidelines does not worsen Warren’s position. He was sentenced to be held at their discretion. He is being held at their discretion. Such are the wages of crime.
It is important to point out that, precisely because Warren was sentenced to be held at the discretion of the parole authorities, under the ex post facto clause he is entitled to an opportunity to have that discretion exercised; anything less would impermissibly augment his penalty.56 But an entitlement to have discretion exercised does not imply an entitlement to have it exercised in a particular way; the essence of discretion is the absence of fixed rules. *197We thus hold that the application of the Parole Commission’s reparole guidelines to Warren does not offend the ex post facto clause of the Constitution.57
We note that we are joined in this conclusion by all Circuits that have considered the question save one.58
Affirmed.
. Greenholtz v. Inmates of the Nebraska Penal and Correctional Complex, 442 U.S. 1, 99 S.Ct. 2100, 60 L.Ed.2d 668 (1979).
. In 1969 the United States Board of Parole made the decision to release an eligible offender in its discretion subject only to a broad statutory mandate, now repealed, specifying that:
If it appears to the Board of Parole from a report by the proper institutional officers or upon application by a prisoner eligible for release on parole, that there is a reasonable probability that such prisoner will live and remain at liberty without violating the laws, and if in the opinion of the Board such release is not incompatible with the welfare of society, the Board may in its discretion authorize the release of such prisoner on parole.
18 U.S.C. § 4203(a) (1970) (repealed by section 2 of the Parole Commission and Reorganization Act of 1976, Pub.L.No.94-233, 90 Stat. 219 (codified at 18 U.S.C. §§ 4201-4218 (1976)).
. The guidelines were published at 38 Fed.Reg. 31,942 — 45 (1973). In 1976 Congress approved the use of the guidelines by enacting the Parole Commission and Reorganization Act which requires the Commission to promulgate such guidelines and authorizes the release of prisoners on parole pursuant to the guidelines. 18 U.S.C. §§ 4201—1218 (1976). The guidelines are described in Project, Parole Release Decision-making and the Sentencing Process, 84 Yale L.J. 810 (1975).
. The Constitution forbids the enactment of ex post facto laws by either the states, U.S.Const. art. 1, § 10, cl. 1, or Congress, id. § 9, cl. 3.
. Shepard v. Taylor, 556 F.2d 648 (2d Cir. 1977) (dictum).
. Ruip v. United States, 555 F.2d 1331 (6th Cir. 1977).
. Zeidman v. United States Parole Comm’n, 593 F.2d 806 (7th Cir. 1979).
. Rifai v. United States Parole Comm’n, 586 F.2d 695 (9th Cir. 1978).
. Geraghty v. United States Parole Comm’n, 579 F.2d 238 (3d Cir. 1978), vacated and remanded on other grounds, 445 U.S. 388, 100 S.Ct. 1202, 63 L.Ed.2d 479 (1980). The Supreme Court has deferred consideration of similar ex post facto claims not only in Geraghty, 445 U.S. at 390 n.1, 100 S.Ct. at 1205 n.1, but also earlier in United States v. Addonizio, 442 U.S. 178, 184, 99 S.Ct. 2235, 2239, 60 L.Ed.2d 805 (1979).
. See note 4 supra.
. 3 U.S. (3 Dall.) 386, 1 L.Ed. 648 (1798). Opinions were delivered by Justices Chase, Paterson, and Iredell.
. Id. at 391.
. Id. at 390.
. Id. A quite similar summary was provided by the Court in 1977:
It is settled, by decisions of this Court so well known that their citation may be dispensed with, that any statute which punishes as a crime an act previously committed, which was innocent when done; which makes more burdensome the punishment for a crime, after its commission, or which deprives one charged with crime of any defense available according to law at the time when the act was committed, is prohibited as ex post facto.
Dobbert v. Florida, 432 U.S. 282, 292, 97 S.Ct. 2290, 2297, 53 L.Ed.2d 344 (1977) (quoting Beazell v. Ohio, 269 U.S. 167, 169-70, 46 S.Ct. 68, 69, 70 L.Ed. 216 (1925)).
. The question in Calder v. Bull did not concern the application of the ex post facto clause in a criminal case; at issue instead was whether the ex post facto prohibition extended to civil legislation. Two of the three justices writing in Calder, Justices Paterson and Iredell, explicitly concluded that retroactive civil laws are not prohibited. Justice Chase left open the question whether retroactive civil legislation might ever come within the ambit of the ex post facto clause, concluding only that the act of the Connecticut legislature in question in Calder was not prohibited. There is considerable historical support for the view that the framers intended the ex .post facto clauses to prohibit the passage of all retroactive laws. See Crosskey, The Ex-Post-Facto and the Contracts Clauses in the Federal Convention: A *187Note on the Editorial Ingenuity of James Madison, 35 U.Chi.L.Rev. 248 (1968); Crosskey, The True Meaning of the Constitutional Prohibition of Ex-Post-Facto Laws, 14 U.Chi.L.Rev. 539 (1947). Nonetheless, modern courts have applied the ex post facto prohibition solely to criminal enactments. E. g., Harisiades v. Shaughnessy, 342 U.S. 580, 594, 72 S.Ct. 512, 521, 96 L.Ed. 586 (1952) (“It has always been considered that that which [the ex post facto clause] forbids is penal legislation which imposes or increases criminal punishment for conduct lawful prior to its enactment. Deportation, however severe its consequences, has been consistently classified as a civil rather than a criminal procedure. Both of these doctrines .. . have been considered closed for many years and a body of statute and decisional law has been built upon them.”) (footnotes omitted).
. 3 U.S. (3 Dall.) at 389.
. Id. at 396 (Paterson, J., concurring).
. Id. at 399-400 (Iredell, J., concurring).
. See Note, Ex Post Facto Limitations on Legislative Power, 73 Mich.L.Rev. 1491, 1500-01 (1975). Ex post facto punishments had evidently been used by the newly independent states prior to the adoption of the Constitution. See Reppy, The Spectre of Attainder in New York, 23 St. John’s L.Rev. 1 (1948); Thompson, Anti-loyalist Legislation During the American Revolution, 3 Ill.L.Rev. 81 (1908); The Federalist No. 44, at 351 (J. Hamilton ed. 1868) (J. Madison). In Calder, Justice Chase cited a number of acts of the British Parliament that were notorious and undoubtedly known to the framers. 3 U.S. (3 Dall.) at 389.
. 3 U.S. (3 Dall.) at 388.
. See Note, Ex Post Facto Limitations on Legislative Power, 73 Mich.L.Rev. 1491, 1500 n.34 (1975).
. This principle is pushed to the limit when applied to war crimes. Cf. In re Goering [1946] Ann.Dig. 203, 208 (No. 92) (International Military Tribunal, Nuremberg) (“To assert that it is unjust to punish those who in defiance of treaties and assurances have attacked neighboring states without warning is obviously untrue, for in such circumstances the attacker must know that he is doing wrong, and so far from it being unjust to punish him, it would be unjust if his wrong were allowed to go unpunished.”).
The important purpose served by this principle in less extreme cases should nonetheless be clear. Consider, for example, the evils that would attend retroactive criminal legislation concerning subject matter on which no broad contemporary consensus exists, such as, say, abortion.
One commentator has suggested that:
Perhaps the strictness with which an ex post facto limitation should be applied in a society should vary inversely with the extent to which that society’s needs are recognized and goals are shared by its members. Thus, disciplinary systems within such institutions as military organizations, schools, and professional associations need not describe conduct for which sanctions will be imposed with the same precision as those peopled by more diffuse groups.
Note, Ex Post Facto Limitations on Legislative Power, 73 Mich.L.Rev. 1491, 1500 n.34 (1975).
. “[I]t is impossible, that the party could foresee, that an action, innocent when it was done, should be afterwards converted to guilt, by a subsequent law; he had, therefore, no cause to abstain from it; and all punishment for not abstaining, must, of consequence, be cruel and unjust.” Calder v. Bull, 3 U.S. (3 Dall.) 386, 395, 1 L.Ed. 648 (1798) (Paterson, J., concurring) (quoting 1 W. Blackstone, Commentaries *46).
. Jacquins v. Commonwealth, 63 Mass. (9 Cush.) 279, 281 (1852).
. For a cogent, useful summary of the fuñetional basis of the ex post facto prohibition see Note, Ex Post Facto Limitations on Legislative Power, 73 Mich.L.Rev. 1491 (1975). See also Crosskey, The Ex-Post-Facto and the Contracts Clauses in the Federal Convention: A Note on the Editorial Ingenuity of James Madison, 35 U.Chi.L.Rev. 248 (1968); Hochman, The Supreme Court and the Constitutionality of Retroactive Legislation, 73 Harv.L.Rev. 692 (1960); Slawson, Constitutional and Legislative Considerations in Retroactive Lawmaking, 48 Calif. L.Rev. 216 (1960); Crosskey, The True Meaning of the Constitutional Prohibition of Ex-Post-Facto Laws, 14 U.Chi.L.Rev. 539 (1947); Smead, The Rule Against Retroactive Legislation: A Basic Principle of Jurisprudence, 20 Minn.L.Rev. 775 (1936); Field, Ex Post Facto in the Constitution, 20 Mich.L.Rev. 315 (1922). The Supreme Court has recently referred briefly to the purposes of the ex post facto clauses in Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 964-65, 67 L.Ed.2d 17 (1981).
. The brief summary which follows is, drawn from the opinions for the Court in United States v. Grayson, 438 U.S. 41, 45—48, 98 S.Ct. 2610, 2613-2614, 57 L.Ed.2d 582 (1978), and in Williams v. New York, 337 U.S. 241, 246-51, 69 S.Ct. 1079, 1082-1085, 93 L.Ed. 1337 (1949).
. See, e. g., Lewis, The Indeterminate Sentence, 9 Yale L.J. 17, 27 (1899) (“convicts [regardless of the nature of their crime] can never be rightfully imprisoned except upon proof that it is unsafe for themselves and for society to leave them free, and when confined can never be rightfully released until they show themselves fit for membership in a free community”) (quoted in United States v. Grayson, 438 U.S. 41, 46, 98 S.Ct. 2610, 2613, 57 L.Ed.2d 582 (1978)).
. See e. g., United States v. Grayson, 438 U.S. 41, 47, 98 S.Ct. 2610, 2614, 57 L.Ed.2d 582 (1978) (referring to 18 U.S.C. § 4205 (1976)).
. See 18 U.S.C. § 4202 (1970); 28 C.F.R. § 2.2 (1974). In fiscal year 1972, nearly 70 percent of all adult federal inmates were given regular adult sentences. [1970-1972] U.S. Board of Parole Biennial Rep. 14-15. Warren himself was paroled shortly after serving one-third of his twenty year sentence; he was sentenced on 26 March 1969, granted parole on 5 November 1975, and released on 5 April 1976. Brief for Appellant at 4.
Besides the regular adult sentence, the federal judge could at his option impose either an “(a)(2)” sentence, 18 U.S.C. § 4208(a)(2) (1970); 28 C.F.R. § 2.3 (1974), in which case the prisoner would become eligible for parole at the Parole Board’s discretion rather than after serving one-third of his sentence, or an “(a)(1)” sentence, 18 U.S.C. § 4208(a)(1) (1970); 28 C.F.R. § 2.3 (1974), under which the sentencing judge would set a date for parole eligibility at any time before one-third of the full sentence had been served.
. For a discussion of the parole decisionmaking process in effect at the time of Warren’s conviction see Project, Parole Release Decisionmaking and the Sentencing Process, 84 Yale L.J. 810, 820-22 (1975).
. 18 U.S.C. § 4203(a) (1970) (repealed by Parole Commission and Reorganization Act of 1976, Pub.L.No.94-233, 90 Stat. 219 (codified at 18 U.S.C. §§ 4201-4218 (1976))).
. Project, Parole Release Decisionmaking and the Sentencing Process, 84 Yale L.J. 810, 821 (1975).
. Menechino v. Oswald, 430 F.2d 403, 408-09 (2d Cir. 1970) (an action by a state prisoner claiming to have a constitutionally protected right to procedural due process at a parole release hearing), cert. denied, 400 U.S. 1023, 91 S.Ct. 588, 27 L.Ed.2d 635 (1971). See also, e. g., Scarpa v. United States Bd. of Parole, 477 F.2d 278, 283 (5th Cir.) (en banc) (“Due process rights do not attach at such proceedings. In the absence of evidence of flagrant, unwarranted, or unauthorized action by the Board, it is not the function of the courts to review such proceedings.”), remanded for consideration of question of mootness, 414 U.S. 809, 94 S.Ct. 79, 38 L.Ed.2d 44 (1973), vacated as moot, 501 F.2d 992 (5th Cir. 1973). Scarpa was cited with approval in the legislative history of the 1976 Parole Commission and Reorganization Act. H.R.Rep.No.838, 94th Cong., 2d Sess. 28 (conference committee report), U.S.Code Cong. & Admin.News 1976, p. 335.
. Project, Parole Release Decisionmaking and the Sentencing Process, 84 Yale L.J. 810, 822 (1975).
. 18 U.S.C. §§ 4201—4218 (1976).
. For the current parole guidelines see 28 C.F.R. § 2.20 (1980).
. The application of the guidelines to reparole is specified in 28 C.F.R. § 2.21 (1980).
. 28 C.F.R. § 2.21(c) (1980). Cf. id. § 2.20(c) (parole guidelines).
. H.R.Rep.No.838, 94th Cong., 2d Sess. 27.
. 18 U.S.C. § 4207 (1976).
. See, e. g., United States v. Addonizio, 442 U.S. 178, 182 n.5, 99 S.Ct. 2235, 2238 n.5, 60 L.Ed.2d 805 (1979).
. Geraghty v. United States Parole Comm’n, 579 F.2d 238, 267 & nn. 150-51 (3d Cir. 1978), vacated on other grounds, 445 U.S. 388, 100 S.Ct. 1202, 63 L.Ed.2d 479 (1980).
. 18 U.S.C. § 4206(c) (1976).
. H.R.Rep.No.838, 94th Cong., 2d Sess. 27, U.S.Code Cong. & Admin.News 1976, p. 359.
. Id.
. Id.
. 28 C.F.R. § 2.20(g) (1979) (“The Commission shall review the guidelines, including the salient factor score, periodically and may revise or modify them at any time as deemed appropriate.” ).
. See text and accompanying notes 31-34 supra.
. See Project, Parole Release Decisionmaking and the Sentencing Process, 84 Yale L.J. 810, 823-24 (1975).
. Id. A salient factor score of 6 to 8, for example, correlates to a range of successful parole between 72 and 79 percent. Id. at 824 n.69.
. See United States v. Casson, 434 F.2d 415 (D.C.Cir.1970) (application to burglar of criminal statute signed into law before burglary but announced contemporaneously with burglary upheld against attack as ex post facto enactment because burglar could have known law was about to be signed by President alter passage by Congress).
. 334 U.S. 728, 68 S.Ct. 1256, 92 L.Ed. 1683 (1948).
. The operation of the recidivist statute in Gryger and the reparole guidelines in the present case is to be carefully distinguished from the operation of a statute that deprives a prisoner of an opportunity to earn good time credit which was previously available. For such a statute clearly does violate the ex post facto clause because it disadvantages a prisoner in a way he cannot control. Furthermore, at the time such a statute is passed, the legislature can know which current prisoners will be thereby so deprived, so the risk of legislative abuse exists. And, of course, the fact that the prisoner is deprived merely of an opportunity rather than of some “vested” interest does not imply that he has not suffered a detriment cognizable under the ex post facto clause. See Weaver v. Graham, 450 U.S. 24, 101 S.Ct. 960, 67 L.Ed.2d 17 (1981).
. See Greenfield v. Scafati, 277 F.Supp. 644 (D.Mass.1967) (cited in Weaver v. Graham, 450 U.S. 24, 32, 34, 101 S.Ct. 960, 966, 967, 67 L.Ed.2d 17 (1981), aff'd mem., 390 U.S. 713, 88 S.Ct. 1409, 20 L.Ed.2d 250 (1968).
. Menechino v. Oswald, 430 F.2d 403, 408-09 (2d Cir. 1970), cert. denied, 400 U.S. 1023, 91 S.Ct. 588, 27 L.Ed.2d 635 (1971).
. Adverse changes in the frequency with which a prisoner may be considered for (as distinguished from granted) parole, or in the time at which a prisoner first becomes eligible for consideration for parole, have been found to violate the ex post facto clauses. See, e. g., United States ex rel. Graham v. United States Parole Comm’n, 629 F.2d 1040 (5th Cir. 1980); Rodriguez v. United States Parole Comm’n, 594 F.2d 170 (7th Cir. 1979); Love v. Fitzharris, 460 F.2d 382 (9th Cir. 1972), vacated on other grounds, 409 U.S. 1100, 93 S.Ct. 896, 34 L.Ed.2d 682 (1973).
. The dissent contends that we should remand to the District Court for “the development of some kind of record . . . before reaching the merits of Warren’s claim.” Dissenting Opinion (Dis. Op.) at 199. We perceive no need for any “record” additional to Warren’s habeas petition if he did not and cannot allege a judicially remediable claim. Warren asserted therein merely that the reparole guidelines violate the ex post facto clause simply because they were promulgated subsequent to commission of the offense for which he is serving time. See Appellant’s Motion for a Writ of Habeas Corpus at 5-8, Appendix to Brief for Appellee at 21-24. We have held that this theory fails as a matter of law in view of the requirement— embedded in both the guidelines and the authorizing statute — that discretion be exercised in applying the guidelines to individual cases.
The dissent states, however, that “[o]n appeal, counsel appointed by this court . .. urged that the Parole Commission applies the guidelines inflexibly, thereby subjecting Warren and others similarly situated to enhanced penalties,” and suggests a remand on that score. Dis. Op. at 198. It is far from clear that a mechanical administrative application of the guidelines — indeed, an ultra vires application — could implicate the ex post facto clause, which historically has been confined to legislative abuses. See, e. g., text accompanying notes 16-19 supra; Frank v. Mangum, 237 U.S. 309, 343-44, 35 S.Ct. 582, 593-594, 59 L.Ed. 969 (1915) (prohibition does not intercept judicial decisions); Ross v. Oregon, 227 U.S. 150, 160-64, 33 S.Ct. 220, 222-223, 57 L.Ed. 458 (1913) (clause does not reach even judicial construction of ambiguous statute). At any rate, we deem counsel’s oblique references to inflexible application of the guidelines, unaccompanied by any profession of ability to produce at least some acceptable evidence thereof, insufficient ground for a remand to enable exploration for either a constitutional or a statutory violation. Cf. United States ex rel. Joseph v. Cannon, 642 F.2d 1373, 1386 & n.112 (D.C.Cir.1981).
. See cases cited notes 5 to 9 supra.