point made by Professor Wechsler in a talk on appellate review of sentences at a Second Circuit Judicial Conference:

"The strategy of the Department of Justice in my time has been . . . to increase the alternatives available to district courts in imposing sentence. That strategy has reached, I think, a really superb culmination in the existing legislation, and there's only one thing in the existing legislation that seems to me really unreasonable, and that is the fixed sentence. I think the fixed sentence is unreasonable because the judge may not separately address himself to the question of the minimum and the maximum. That seems to me patently unreasonable because different considerations obviously bear on the minimum and the maximum sentence." Appellate Review of Sentences, A Symposium for the Judicial Conference of the United States Court of Appeals for the Second Circuit, 32 F.R.D. 249, 291 (1962) (remarks of Professor Wechsler).

I would agree that "different considerations obviously bear on the minimum and the maximum sentence." The main function of the maximum sentence is to state that period of time beyond which, no matter how bad a risk the defendant may seem at the time, he simply may not be detained on the basis of the criminal act for which he has been sentenced. The maximum, in other words, expresses the essentially moral judgment that society is not justified in incarcerating an individual beyond the point fixed by its limit for the particular criminal conduct involved.

The minimum sentence on the other hand-or the time at which the defendant becomes eligible for parole, to state the same thing another wayserves wholly different objectives. Reassurance of the community and deterrence are the two which are most commonly cited. See ABA Standards, Sentencing Alternatives, pp. 146-47.

The bill under discussion is drafted on the premise that the first judgment— what the maximum sentence shall be-is properly a matter for individual determination by the sentencing judge, who is on the scene and therefore is better able to make a judgment about the particular risks posed by the particular defendant involved. Indeed, maximum sentences are determined in this manner for most federal crimes. Yet-inconsistently and erroneously, it seems to me when it comes to the question of the minimum sentence, the judge no longer has discretion to make his individualized judgment aimed at the particular offender, and is deprived of the power to make a separate assessment of the validity of the objectives of a minimum term for imposition in the particular case. It seems to me that judicial discretion, separately exercised on both issues, is the better approach in principle.

In addition, there are adverse consequences of a practical nature that I would see as the result of the bill as presently drafted. In the first place, judges may well be reluctant to make use of the bill because of the necessity of imposing a long minimum term. Closely related is the fact that even if they do use the bill, the judges are likely to frame their maximum sentence with an eye to how long they think to be a realistic minimum. The fact that the maximum and the minimum are inflexibly tied together, in other words, will mean that neither will be independently determined in accordance with the demands of the particular case. A judge may think that thirty years is quite appropriate as an outside limit on the time a particular offender should be held, but simply refuse to impose such a sentence when it means that he must be detained for twenty. Similarly, a judge may think that ten years is an appropriate outside limit because, for example, of the defendant's age, and yet be required to impose a longer sentence in order to achieve the minimum which he also thinks desirable. It is clear from the way judges now deal with minimum sentences, I think, that an attempt to deny the direct exercise of discretion over this issue will result in the use of another outlet for the expression of the discretion sought to be denied. The tendency, I believe, will be to force maximum sentences down in order to achieve a more realistic minimum and in some cases to discourage use of the enhanced penalty at all. This, of course, is counterproductive to the major ends of the bill, which I see as primarily directed at the availability and use of long-term maximum sentences as one line of attack against organized crime.

I would therefore conclude that it would be much better to model the parole eligibilty provisions of this bill after the existing federal law, which as Professor Wechsler points out is in the general quite excellent on this point. What I personally would do is make parole eligibility immediate, unless the sentencing judge took the initiative to impose a minimum term. I would then permit, as the federal law now does under 18 U.S.C. § 4208 (a), the imposition of a minimum term (or parole eligibility date) at any point up to one-third of the maximum sentence imposed. This practice is supported not only by the present

federal law (though I have reversed the emphasis; now the judge must act affirmatively in order for a minimum sentence of one-third not to be imposed), but also by the recommendations of the Model Penal Code and the ABA. See ABA Standards, Sentencing Alternatives § 3.2 & pp. 142-60.

3. Disclosure of the presentence report.-Before addressing myself to the particulars involved in the present bill, let me state at the outset that it seems to me to be completely inconsistent with fair procedures to tell a sentencing judge that he may choose any term between one and twenty years (as many of our statutes do, both state and federal), and at the same time to say that he need not even notify the defendant of the facts on which he is basing his judgment. It seems to me plainly silly to get as technical as we do about the precise language in the indictment, the exact contours of the hearsay rule and the minutia of jury instructions, and at the same time to permit a decision with the consequences of the sentencing decision, and with such a wide range of possibilities in most cases, to be made without even giving the defendant the opportunity to speak to the factual basis on which the judge is acting. We surely would not tolerate the assessment of damages in a civil action in such a manner. Nor can I think of any other issue of comparable significance which is so summarily treated."

But whatever the proper resolution of the disclosure issue in the run-of-the-mine case, it seems to me doubly clear that disclosure should be required in a statute of this sort. The procedural provisions of this statute seem in general to be quite good: the defendant is specifically advised at the time of indictment that he may be subject to an enhanced term; he is then advised with specificity of what the basis for such action may be; he is not prejudiced at his trial by the disclosure to the jury of the charge or its basis; and he is given a special hearing at the time of sentencing at which he is afforded an opportunity to meet the charge.

It seems to me inconsistent with these procedures, and extremely vulnerable to constitutional attack in this context, then to say that the judge may make his findings either on the basis of the evidence submitted at the hearing or on the basis of an undisclosed presentence report. Would it then be possible for a judge to find on the basis of the secret information in the presentence report that the defendant was indeed a professional offender or did indeed commit the two prior offenses as charged? If the defendant offers evidence that he did not commit the prior offenses charged by the prosecutor, would it then be possible for the judge to respond that "what you say is contradicted by the information in the presentence report, but I am not going to tell you what that information is?"

It seems to me that the constitutional problems which cases like Specht, Robinson, and others indicate are in store for any legislation of this type at the least indicate that obvious problems of fairness of this sort be attended to. I am not, of course, advocating that a full adversary proceeding with all the trappings of the criminal trial be imported into the sentencing stage. It does not seem to me that that is necessary, or wise. But it does seem to me that notice to the defendant of the facts on the basis of which he is being held subject to this extraordinary sentence should be an elemental requirement. Again, this is an issue to which the American Bar Association has addressed itself, and again the point I am urging upon you is one to which it has come in its published report. See ABA Standards, Sentencing Alterntaives § 5.5(b) (iii) & pp. 213-28, 257, 262-66.

One argument that might be expected in response to my contention is that it is in these very cases, particularly when dealing with professional criminals, that non-disclosure of the presentence report is most warranted. For it is in this context that real fears will be generated in those who might be asked to supply information, and that the normally unpersuasive argument that the sources of information may be dried up by disclosure might really be true.

I can only answer such an argument by reference to a point I have already made. Though overworked, the image that may lead to a holding that this legislation is unconstitutional is not related to the star chamber. It may well be possible under this act to convict a defendant of a minor felony carrying only a two-year maximum sentence, charge him at the same time with being a professional offender, find him to be such an offender on the basis of information to which he does not have access, and sentence him to thirty years. As I commented above, such a procedure already has the look of an attempt to secure a long sentence through the back door, perhaps motivated by an inability to prove in open court the facts which in substance form the basis for the whole proceeding. The part which non-disclosure of the presentence report plays in such a spectacle may, indeed, not be the sine qua non that leads to a holding

For a general treatment of the disclosure issue, see ABA Standards, Sentencing Alternatives, pp. 216-24. The reference is to Robinson v. California, 370 U.S. 660 (1962).

of unconstitutionality. But when added to the other images which are invoked, it surely may play a substantial role.

The answer to the argument that this is a very case in which non-disclosure is necessary, in other words, is that this very fact is what makes disclosure critical. The whole proceeding smacks of one which is motivated by an inability to prove beyond a reasonable doubt to a jury in open court the facts on which the sentence is based. The "need for secrecy" argument on which disclosure might be withheld thus reinforces the basic constitutional defect from which the whole proceeding might well be held to suffer.

4. Review of the sentence.-As I read § 3577, either the defendant or the United States may initiate an appeal looking to review of the sentence. The Circuit Courti is authorized in turn either to reduce or increase the sentence. Moreover, it does not appear that the power to increase is tied to an appeal by the United States. It would therefore seem that even if the defendant alone initiates an appeal seeking a reduction of the sentence, the Court may nevertheless disagree with his contentions and impose an increased term. I will comment separately on two aspects of this structure.

(a) Increase on an appeal by the defendant. As you are undoubtedly aware, two cases very much related to the issues posed be this situation are now on the docket of the United States Supreme Court. See North Carolina v. Pearce, No. 413, 1968 Term; Simpson v. Rice, No. 418, 1968 Term. These cases present the issue of whether a defendant may be given an increased sentence after his first sentence has been set aside for one reason or another. More broadly speaking, the question is whether there is constitutional defect in a procedure which opens to a defendant an avenue by which he can assert an error in the proceedings against him, but which at the same time holds out the possibility that if he takes advantage of the procedure, he may wind up in worse shape than when he began. From this perspective, it would appear that the same issue posed by the pending cases is involved in the provision under discussion here.

It may therefore be that the Court in these two decisions will answer the constitutional questions against providing the power to increase the sentence if the defendant alone has initiated the appeal. if that is the case, then I would certainly suppose that the power should not be included in the present statute.

* This could happen on any one of a number of grounds. The main possibilities would appear to be the three arguments listed below: (1) The first is based on the so-called "unconstitutional conditions" doctrine. See Comment. Another Look at Unconstitutional Conditions, 117 U. Pa. L. Rev. 144 (1968); Note, Unconstitutional Conditions, 73 Harv. L. Rev. 1595 (1960). That doctrine suggests that It is unconstitutional to hold out a benefit which can only be taken advantage of by the relinquishment of a constitutional right. There are several ways this doctrine can be made relevant here, but the main one would appear to be the following: If the defendant did not appeal, he would clearly be protected by the doctrine of double jeopardy from ever receiving a higher sentence. The condition on which he is permitted to appeal is that he give up this constitutional protection. He therefore has been given a benefit (the right to appeal) only upon the condition that he relinquish a constitutional right (the right not to have his sentence increased). For an alternative formulation of the same argument, see Commonwealth v. Littlejohn, 4 Crim. L. Rep. 2359 (Pa. January 24, 1969); 80 Harv. L. Rev. 891, 895 (1967). See also United States v. Walker, 346 F. 2d 428 (4th Cir. 1965). (2) The second argument is based on the double jeopardy clause itself and is derived from the cases of Green v. United States, 355 U.S. 184 (1957) and Henderson v. California, 386 P. 2d 677 (1963). Green held that it offended double jeopardy to reconvict an offender of first degree murder after he had successfully appealed a conviction for second degree murder. Henderson extended the principle of the case to cover the situation where a defendant was sentenced to death for a first degree murder conviction after he appealed a first degree conviction at which he was sentenced to life. While it is true that the Supreme Court went the other way when last it was confronted with a case like Henderson (see Stroud v. United States, 251 U.S. 15 (1919)), that case is fifty years old, the argument continues, and has been undercut by Green. Even the dissenters in Green recognized that the cases were indistinguishable.

(3) The third argument is based on equal protection. As pointed out in the Harvard Law Review, it has at least two aspects: the first is that it is unfair discrimination to permit a higher sentence only in the case of those defendants who choose to appeal. There undoubtedly are many among the class of defendants who do not appeal and who equally are deserving of higher sentences; to use as the criterion for eligibility for a higher sentence those who choose to assert the right to appeal is to pick an irrelevant measure. The second prong of the equal protection attack is based on unfairness in the availability of the appeal: those defendants who have been sentenced to the maximum can appeal freely; those who have lesser sentences are substantially deterred from appealing because of the possibility of an increase. This distinction too is without any relevance to the possibility of error or to any penological objectives which would be served by permitting an increase. See 80 Harv. L. Rev. 891, 895 (1967). See also Commonwealth v. Littlejohn, 4 Crim. L. Rep. 2359 (Pa. January 24, 1969); Rice v. Simpson, 2 Crim. L. Rep. 2068-69 (U.S.D.C. Ala. Sept. 26, 1967).

It is of course also possible that the case may be decided in favor of the prisoners on grounds that are not relevant to the discussion here. There are due process arguments, for example, that would seem to have particular applicability to the Pearce and Rice situations but may not control the issue posed by this bill. See, e.g., 80 Harv. L. Rev. 891, $95 (1967).

It is my very firm opinion, however, that even if the Court upholds the procedures involved in Pearce and Rice against constitutional attack, it would not be sound as a matter of legislative policy to permit the appellate courts to increase a sentence upon the defendant's appeal alone. Most of my reasons for holding this view have been spelled out in detail in ABA Standards, Appellate Review of Sentences, pp. 57-63 (Tent. Draft, April 1967). It may be helpful, however, if I attempt to summarize them here.

The arguments in favor of power in the appellate court to increase a sentence if the defendant appeals basically come down to two: that in principle it is not sound to place limitations on a reviewing court which may prevent the doing of justice in a concrete case; and that without some deterrent to appeals against sentence, a defendant with nothing to lose will automatically appeal, thereby flooding the courts with appeals, most of which will be without merit anyway. The second contention is the easiest to answer. I would make three points. First, it is erroneous in principle because it reflects the adoption of a remedy that has nothing to do with the problem which is sought to be cured. No one would advance the "deterrent" theory as a reason why a defendant who is entitled to a reduction should not appeal. The kind of appeal that is sought to be deterred is the frivolous appeal, the appeal that is without merit and that therefore is thought to take up the court's time without purpose. But the solution of permitting an increase is just as likely to deter the meritorious appeal as it is the appeal that is without merit; it operates with equal force as a deterrent to good appeals as well as bad ones. If too many frivolous appeals is the problem, other methods of attack-more streamlined appellate court procedures, for example should be tried before resort is had to overkill.

The deterrence theory also suffers from another serious defect. To say on the one hand that our conceptions of justice require that decisions be subject to a reviewing process, and on the other that we are so afraid that defendants will take advantage of the process that we must build in serious incentives not to take advantage of it, reflects a kind of hypocrisy which I think we could well do without. The image of justice is not advanced by our present practice of permitting a single individual to make a decision of enormous consequence often involving a range of choice between probation and twenty or thirty years and at the same time taking the position that he need not state his reasons and that no one can undo the decision he has made. To correct this image by saying that we will provide a check against abuse only if the defendant is willing to take the risk that his sentence will be doubled seems to me to compound rather than cure the genuine issues of fairness which I see in this situation.

I would add finally on this point that in the particular class of cases with which this legislation deals, it seems to me that the deterrence argument will have even less force than it might normally have. I would be very surprised if the average professional offender or organized crime offender is going to be content with a long sentence, or even a short one, without initiating an appeal of some sort. If either guilt or the sentence was contested at the trial level, it is unrealistic to assume that appeals will not be taken on the merits, on the procedures, and on anything else that legal ingenuity can come up with. And once an appeal has been taken on any other issue, it seems to me that no added burden of significance is placed on the appellate court by virtue of the fact that the case also includes an attack on the sentence.10

This issue was thoroughly debated in the course of developing the ABA Standards on Appellate Review of Sentences. The original Advisory Committee, though split 7-4, recommended against the power to increase upon an appeal by the defendant. Its reasons are contained in the passage cited above. The Special Committee, essentially an overseeing group, split 8-4 the other way. See ABA Standards, Appellate Review of Sentences, p. 2 (Proposed Revisions of Standards, December 1967). The issue went to the floor of the ABA's House of Delegates in this posture, where after debate the Special Committee's view was adopted by a vote of 95-76. See 2 Crim. L. Rep. 2421 (Feb. 28, 1968). The result of the ABA project is thus a viewpoint with which I personally disagree.

10 Several other points perhaps deserve footnote mention. First, a defendant who is of the opinion that his sentence is too long is likely to appeal his conviction if he feels that he would unduly expose himself to appeal against his sentence. And a court that agrees with him might very well, as some appellate courts have, reverse the conviction in such a context rather than let what it thinks to be an unjust sentence stand. Second, no one would suggest that a defendant receive an increase in his sentence because he has taken a frivolous appeal; yet the image will undoubtedly get abroad that this is "what is really going on" if a court increases a sentence, even for legitimate reasons. The effect will be, as I believe could be documented by the effect in England, Massachusetts, Maine, Connecticut and Maryland, a sharp reduction in the number of appeals shortly after an increase is imposed, with a gentle rise until the next time, all of which, of course, is utterly unrelated to the reason review is authorized and to the merits of the cases that do not get appealed.

This is not an answer, however, to the contention that if justice in the particular case at the appellate level indicates that an increase is in order, the court should be empowered to go ahead and impose one. I admit that there is force to this contention, and it seems clear to me that it is on this point-not deterrence— that the issue should turn.

My response to this justification for authorizing an increase in the sentence is that I think the issue more properly addressed as one of whether the prosecution should be entitled to appeal. If the answer there is no, then it seems to me that the force of the principle that the court should be entitled to do justice in the particular case is measurably reduced. In cases where only the defendant has appealed-and a fortiori if the prosecutor is not entitled to an appealthere is an element of unfairness in responding to a request to reduced by imposing an increase. There is no other context in which this is permitted. The defendant does not expose himself, for example, to an increase in the degree of the crime or to conviction of a more serious offense by appealing his conviction. And yet the argument seems equally convincing to me in such instances or perhaps I should say equally unconvincing-that the court, since it has the case, should be empowered to do justice, even if justice includes a decision that the defendant is more culpable than the court below thought he was. It would undoubtedly be deemed unfair" in a civil case for the court to increase the award of damages where the appellant sought a reduction and no cross-appeal were taken. It seems to me that the same point should control here. If the prosecutor wants an increase, then he should ask for it, and should put the defendant on notice that he will have to defend himself against an increase. If there are reasons why the prosecutor should not be entitled to take an appeal against the sentence as I believe there are then those same reasons seem equally convincing to me in the situation where the defendant alone has appealed. And in fact, as I am trying to suggest and is suggested by the constitutional arguments advanced in the Pearce and Rice cases, it is worse even more unfair to the defendant-to authorize an increase if the defendant alone appeals than it is to permit the prosecutor to initiate the appeal.

(b) Appeal by the prosecutor. There are two inquiries which must be made in an evaluation of the question of whether the prosecutor should be allowed to initiate an appeal in which he seeks an increase in the sentence. The first involves an exploration of the constitutional issues. The second approaches the question as a matter of legislative judgment.

Constitutional issues. It may be helpful at the outset simply to catalogue the most relevant constitutional precedents. Kepner v. United States, 195 U.S. 100 (1904), seems a logical starting point. In that case, a lawyer was charged with embezzlement of his client's funds. He was acquitted by the Philippine court of first instance. Upon an appeal by the United States of the acquittal, the Philippine Supreme Court reversed the lower court, entered a judgment of conviction and imposed a jail sentence. The case is complicated by the fact it involves the understandings of the Executive and the Congress in the establishment of ground rules for the government of the Philippines, but it seems clear both from the opinion and subsequent cases that the Court treated the issue before it as substantially involving the double jeopardy clause of the Fifth Amendment.12

12

The clear result of the Kepner case, in spite of a persuasive dissent by Justice Holmes, was that the government cannot appeal a judgment of acquittal:

"The court of first instance, having jurisdiction to try the question of the guilt or innocence of the accused, found Kepner not guilty; to try him again upon the merits, even in an appellate court, is to put him a second time in

11 And perhaps unconstitutional too, though on jury trial grounds. See Dimick v. Schiedt, 293 C.S. 474 (1935). But see Jehl v. Southern Pacific Co., 427 P. 2d 988 (Cal. 1967); Fisch v. Manger, 130 A. 2d 815 (N.J. 1957).

12 I can think of at least four arguments that the Court meant its opinion to be taken as a construction of the Fifth Amendment: the Court explicitly said so (see 195 U.S. at 124); by so holding the Court was able to avoid the issue of whether the double jeopardy clause, together with the rest of the Bill of Rights, applied of its own force to the Philippines (see 195 U.S. at 124-25); Justice Holmes in dissent was clearly basing his criticism on the view that the majority was interpreting the constitution (see 195 U.S. at 134); and the Court explicitly said in Trono v. United States, 199 U.S. at 529, decided but a year later, that Kepner was a constitutional decision. It therefore seems to me that the attempt in Green v. United States, 355 U.S. 184, 195-98 (1957) to suggest that these cases were not interpretations of the double jeopardy clause is in error. Whether they were correct interpretations of the double jeopardy clause is of course another matter.