(dissenting).
Today we uphold obviously abusive and punitive criminal sanctions. The sentences approved by this court shock the human conscience and by legal standards stun the judicial conscience as well. But perhaps more relevant here, they were imposed illegally.
Disclosure of the Presentence Reports
Without bothering to give any reason for the harsh penalties or to reveal to defendants or their counsel the contents of their presentence reports, the district court imposed five-year maximum sentences, to be served consecutively, on these first offenders. There exists no possible reason for such severe punishment other than the district court’s reliance on uncorroborated hearsay contained within the presentence reports. The defendants pled guilty to three counts of mail fraud. It is true that the government indicted them on 23 additional counts, but these were dismissed.1 The presentence reports reveal that the postal inspectors were hampered in their investigation since the insurance company files were lost; nevertheless, the reports state that postal officials “feel” these defendants participated in many more fraudulent claims than they were charged with. This is the rankest sort of hearsay which deserves to fester in the open. If the defendants are to be sentenced for other crimes, the government has the burden of proving them. The defendants cannot lawfully be sentenced for crimes not proven. Yet it seems clear, as made manifest by the majority opinion and the presentenee reports, that this is what occurred. This illegal procedure places an almost intolerable burden on a defendant. How can the defendants rebut unfounded hearsay, unproven charges, when they are not allowed to see the presentence reports or even to know the judge’s reasons for imposing a particular sentence ? 2
This case is no different than United States v. Weston, 448 F.2d 626 (9th Cir. 1971), cert. denied, 404 U.S. 1061, 92 S.Ct. 748, 30 L.Ed.2d 749 (1972), in which the Ninth Circuit vacated the defendant’s sentence and remanded for resentencing. On resentencing, the district court was instructed not to “rely upon the information contained in the presentence report unless it is amplified by information such as to be persuasive of the validity of the charges there made.” Id. at 634. The court summed up the situation as follows:
In essence, then, what we have is a conviction at a trial providing all of the safeguards required by the Constitution, of an offense warranting, in the opinion of the trial judge, the minimum sentence of five years. This is followed by a determination, based on unsworn evidence detailing otherwise unverified statements of a faceless informer that would not even support a search warrant or an arrest, and without any of the constitutional safeguards, that Weston is probably guilty of additional and far more serious crimes, for which she is then given an additional sentence of fifteen years. Cf. Specht v. Patterson, 1967, 386 U.S. 605, 87 S.Ct. 1209, 18 L.Ed. *9022d 326; Kent v. United States, 1966, 383 U.S. 541, 553-554, 86 S.Ct. 1045, 16 L.Ed.2d 84. To us, there is something radically wrong with a system of justice that can produce such a result.
Id. at 630-631.
To say that disclosure of presentence reports lies within the discretion of the district court and that the court’s exercise of discretion will not be disturbed except for abuse is one thing; to refuse to require disclosure where abuse has been shown is another. We observed in United States v. Carden, 428 F.2d 1116, 1118 (8th Cir. 1970), that the Advisory Committee under Rule 32(c) of the Federal Rules of Criminal Procedure “did not contemplate an absolute rule of nondisclosure without relevancy to the particular circumstance of a case.” And we added:
It is always advisable for the trial judge to at least state on the record the various factors he has taken into consideration in rendering his sentence. . . . Such a procedure serves as a checkmate to the danger of misinformation being placed in the hands of the court.
Id. at 1118 (emphasis added).
See also Hess v. United States, 496 F.2d 936 (8th Cir. 1974).
Today, the majority chooses to ignore those holdings. Today’s holding additionally defies the standards established by the National Advisory Commission on Criminal Justice Standards and Goals, in its Report on Corrections, Standard 5.16, as well as the holdings of most of the courts of appeals. See, e. g., United States v. Espinoza, 481 F.2d 553 (5th Cir. 1973); United States v. Brown, 470 F.2d 285 (2d Cir. 1972); United States v. Janiec, 464 F.2d 126 (3rd Cir. 1972); United States v. Picard, 464 F.2d 215 (1st Cir. 1972); United States v. Miller, 495 F.2d 362 (7th Cir. 1974).
The presentence reports do not rely on any confidential sources which might “dry up” if disclosed. There is no derogatory information about the defendants themselves which might retard their rehabilitation. In sum, there exist no relevant factors here to support nondisclosure of the reports.
Even assuming that nondisclosure were appropriate, if there are justifiable reasons for the sanctions imposed, these reasons should be made known to the defendants and to the reviewing court. When the district judge is free to impose whatever sentence he chooses, without explanation, this court has no way of knowing whether or not he properly exercised his discretion, or whether he in fact exercised it at all. When the defendants are not provided with an explanation, the likelihood that they will feel they have been unfairly treated is greatly magnified. This only fosters disrespect for the law and certainly cannot assist in the defendants’ rehabilitation. See M. Frankel, Criminal Sentences 39-49 (1973).3
*903 Sentence Review
It must be acknowledged that the scope of federal appellate review of sentencing is still unsettled. Nevertheless, this circuit has recognized the injustice of unreviewed abusive sentences. In Woosley v. United States, 478 F.2d 139 (8th Cir. 1973) (en banc), we held that an appellate court had the power to “consider whether the imposition of a maximum sentence was in itself an abuse of discretion . . . .’’Id. at 146. The Woosley court further noted that “the record completely fails to justify, nor has the district judge undertaken to explain, the imposition of a maximum penalty . . . .’’Id. at 147. After reviewing the facts in Woos-ley, the court concluded:
The broad and unreviewable discretion possessed by federal district courts in matters of sentencing does not extend to the meting out of punishment manifestly disproportionate to the nature of the crime and the character of the criminal.
Id. at 148.
Federal courts review the excessiveness of civil damages, when abuse of discretion is shown by judge or jury. It is an anachronism not to afford similar review of sentences in criminal cases.4
One of the alleged difficulties of sentence review is the lack of manageable judicial standards. There exists the continued concern that appellate judges will simply substitute their “guess” for that of the trial judge. But the trial judge does not, or at least should not, act out of personal benevolence any more than he should act punitively outside of the law. Discretion lawfully exercised must represent a legal judgment, not a mere human reaction to choice. The same standards which move a trial judge to a particular sentencing decision mark out the boundaries by which lawful exercise of judgment must be measured. To review the exercise of judicial discretion in the sentencing process is not that difficult a task — the appellate process involves similar appraisals in many areas *904of the law. I suggest the main difficulty in reviewing lawful discretion on appeal comes not in measuring the exercised judgment against the marked boundaries of the field, but in the all too human predilection of judges to routinely “rubber stamp” the questioned discretion of a district judge without making a qualitative review of the actual decision. It is as erroneous to substitute conclusory agreement as it is to voice opinionated disagreement.
In order to avoid charges that appellate review of sentences “reflects no more than the substitution of one guess about the proper disposition for another,” Standard 1.2 of the ABA Project on Minimum Standards for Criminal Justice, Standards Relating to Appellate Review of Sentences, establishes general objectives of sentence review:
The general objectives of sentence review are:
(i) to correct the sentence which is excessive in length, having regard to the nature of the offense, the character of the offender, and the protection of the public interest;
(ii) to facilitate the rehabilitation of the offender by affording him an opportunity to assert grievances he may have regarding his sentence;
(iii) to promote respect for law by correcting abuses of the sentencing power and by increasing the fairness of the sentencing process; and
(iv) to promote the development and application of criteria for sentencing which are both rational and just.
Id. at 21.
Proper review of the sentences in the present cases would accomplish all four of the above objectives. Although there are no fixed rules to bind the district courts in imposing sentences, it is apparent that defendants’ sentences run counter to all modern guidelines on the proper imposition of sanctions. The Report on Corrections of the National Advisory Commission on Criminal Justice Standards and Goals recommends imprisonment only where there is no suitable alternative:
The traditional sanction imposed for violation of the criminal law in this country was confinement. Any other sentence that in some way moderated the nature or extent of confinement was considered a lenient act on the part of the state. It is more clearly recognized today than confinement is unnecessary and inappropriate in a large number of cases. Supervision in the community has been demonstrated to be at least as effective in insuring future law-abiding conduct without the human degradation, hostility, and public expense associated with confinement. Confinement thus should be considered the alternative to be used when no other disposition will protect the public safety.
Id. at 152.
In its Standard 5.2, on Sentencing the Nondangerous Offender, the National Advisory Commission provides the following guidelines to assist the sentencing court in its decision:
Criteria should be established for sentencing offenders. Such criteria should include:
1. A requirement that the least drastic sentencing alternative be imposed that is consistent with public safety. The court should impose the first of the following alternatives that will reasonably protect the public safety:
a. Unconditional release.
b. Conditional release.
c. A fine.
d. Release under supervision in the community.
e. Sentence to a halfway house or other residential facility located in the community.
f. Sentence to partial confinement with liberty to work or participate in training or education during all but leisure time.
g. Total confinement in a correctional facility.
*9052. A provision against the use of confinement as an appropriate disposition unless affirmative justification is shown on the record. Factors that would justify confinement may include :
a. There is undue risk that the offender will commit another crime if not confined.
b. The offender is in need of correctional services that can be provided effectively only in an institutional setting, and such services are reasonably available.
c. Any other alternative will depreciate the seriousness of the offense.
3. Weighting of the following in favor of withholding a disposition of incarceration:
a. The offender’s criminal conduct neither caused nor actually threatened serious harm.
b. The offender did not contemplate or intend that his criminal conduct would cause or threaten serious harm.
e. The offender acted under strong provocation.
d. There were substantial grounds tending to excuse or justify the offender’s criminal conduct, though failing to establish defense.
e. The offender had led a law-abiding life for a substantial period of time before commission of the present crime.
f. The offender is likely to respond affirmatively to probationary or other community supervision.
g. The victim of the crime induced or facilitated its commission.
h. The offender has made or will make restitution or reparation to the victim of his crime for the damage or injury which was sustained.
i. The offender’s conduct was the result of circumstances unlikely to recur.
j. The character, history, and attitudes of the offender indicate that he is unlikely to commit another crime.
k. Imprisonment of the offender would entail undue hardship to dependents.
l. The offender is elderly or in poor health.
m. The correctional programs within the institutions to which the offender would be sent are inappropriate to his particular needs or would not likely be of benefit to him.
Report on Corrections, supra at 150-151.
The work of the ABA Project on Minimum Standards for Criminal Justice, Standards Relating to Sentencing Alternatives and Procedures, reflects the same views on sentencing as the Report of the National Advisory Commission. Standard 2.2 of the ABA Project provides :
The sentence imposed in each case should call for the minimum amount of custody or confinement which is consistent with the protection of the public, the gravity of the offense and the rehabilitative needs of the defendant.
Id. at 61.
Similarly, Standard 2.3(c) provides that “[a] sentence not involving confinement is to be preferred to a sentence involving partial or total confinement in the absence of affirmative reasons to the contrary.” Id. at 64.5
*906Both the National Advisory Commission and the ABA Project are also opposed to consecutive sentences, except in the case of dangerous offenders. The Report of the National Advisory Commission recommends a presumption “in favor of concurrent sentences for multiple offenses.” Id. at 166. The Report finds little justification for consecutive sentences, other than to extend the period of confinement:
Such an extension, if based solely on the fact that more than one offense was committed, regardless of the needs of the particular offender or the requirements of public safety, amounts to the imposition of sanction purely for punishment purposes.
Id.
The Report concludes that only dangerous offenders should ever be subjected to more than five years imprisonment.
The ABA Standards Relating to Sentencing Alternatives and Procedures are also generally opposed to consecutive sentences. Standard 3.4(b) holds that “[cjonsecutive sentences are rarely appropriate.” Id. at 171.
In the Commentary to this Standard, the Advisory Committee states that “[t]he consecutive sentence should be the exception rather than the rule . . . . [T]he basis of a consecutive sentence, like the basis of any other sentence which extends for an unusually long period, ought to be a finding by the court that the defendant poses such an exceptional risk to the public that long-term incarceration is necessary as a protective measure.” Id. at 179.
In addition to running counter to modern penological thinking, the sentences imposed by the district court are far in excess of the average sentence for defendants convicted of similar crimes. In 1971 (the most recent year for which data are available), 496 defendants were convicted of mail fraud offenses. Of this total, less than 50% were sentenced to any time in prison. Among those who were incarcerated, the average sentence was 33.2 months. Only 37 defendants out of the total of 496 were sentenced to five or more years. Administrative Office of the United States Courts, Federal Offenders in the United *907States District Courts: (1973). 1971, at 96
With these guidelines in mind, examination of the facts in the present cases shows that the sentences imposed were completely unwarranted. The district court chose the most severe sentencing alternative available, yet there are no aggravating circumstances in either case requiring such extreme measures.
Defendant Dace is 46 years old; has been married for 21 years; has three children, ages 15, 11 and 3; and had been employed by the Hartford Insurance Group for ten years. The extent of his past criminal record was a misdemeanor charge in St. Louis County, which resulted in a suspended imposition of sentence. The trial court was advised that Dace has insufficient funds to support his family while confined; that his wife is now working, but cannot afford to keep all three children at home; and that she has been denied welfare assistance. The court was further informed that Dace has employment available to him upon his release from custody. The presentence report shows that Dace cooperated fully with law enforcement authorities during their investigation.
Defendant Schneider is 52 years old; has been married for 27 years; has two children in college; and had been steadily employed. He served his country as a combat pilot in Korea, suffered a disability as a result of such service, and was decorated three times. He also served as a volunteer police officer and has no prior criminal record of any sort. With regard to the three counts to which Schneider pleaded guilty, his counsel pointed out that Schneider had made full restitution to the insurance company.6 Schneider also cooperated fully with law enforcement officers. Schneider’s counsel produced a letter from a Mr. Reeves, offering Schneider employment upon his release from custody, and several other letters of recommendation from various people concerning Schneider’s general character.
In summary, Schneider has no prior criminal record at all; Dace has had only one relatively minor scrape with the law. Neither defendant could be considered the least bit dangerous to the public. Their briefs indicate that Dace, after confinement at Leavenworth Penitentiary, was quickly transferred to the minimum security prison camp at Leavenworth, Kansas, and that Schneider has been recommended for the same assignment. Schneider has made at least partial restitution to the insurance company. Neither defendant is likely to benefit from rehabilitative training programs available in prison. These are not unskilled defendants who, if released from custody, might well be forced to commit another crime out of economic necessity. On the contrary, both defendants have jobs available to them immediately upon their release. In the meantime, their confinement is working a severe hardship on their families, who may soon have to look to the state for support.
Keeping Schneider and Dace in prison for an extended period of time is far more likely to have a negative rehabilitative effect. By the time they are released, their employment prospects may well have disappeared. They are likely to be so bitter against society that their readjustment into the community will be impossible. I would vacate both sentences, remand to the district court and require that the court disclose the pre-sentence report to each defendant, and that the court thereafter resentence each defendant, taking into consideration time already served. The district court should be directed to consider immediate probation. If some justifiable reason can be demonstrated on the record to deny probation, then the court could require further imprisonment based solely on the charges to which the defendants have pleaded guilty and no others.
ORDER ON REHEARING
Upon consideration of appellants’ petition for rehearing we deem it *908appropriate to emphasize that where the presentence report is not disclosed to a defendant, “[i]t is always advisable for the trial judge to at least state on the record the various factors he has taken into consideration in rendering his sentence. * * * Such a procedure serves as a checkmate to the danger of misinformation being placed in the hands of the court.” United States v. Carden, 428 F.2d 1116, 1118 (8th Cir. 1970) (emphasis added). Failure to do so may well require a vacation of sentence and remand for resentencing under the teachings of Townsend v. Burke, 334 U.S. 736, 68 S.Ct. 736, 92 L.Ed. 1690 (1948); and United States v. Tucker, 404 U.S. 443, 92 S.Ct. 589, 30 L.Ed.2d 592 (1972).
Our refusal to vacate and remand for resentencing in the instant case follows a careful examination of the presen-tence reports which, as we have noted, do not include unreliable information that a district judge would normally rely on in rendering sentence.
The petition for rehearing is therefore denied.
LAY, Circuit Judge, dissents from the denial of rehearing for all the reasons stated in his original dissent.
. Defendant Schneider was named in only seven counts of the original 26-count indictment.
. The majority states that examination of the presentenee reports by this court is sufficient to insure that no misinformation was relied upon by the trial court. At 900. I do not feel, however, that this is a viable sub-statute for disclosure to defendants and their counsel in the first instaaace. This court is in no better position to assess the accuracy of the reports than the sentencing court. Defendants, still in the dark as to the contents of the report, have never been afforded the opportunity to rebut the charges contained therein. Cf. Townsend v. Burke, 334 U.S. 736, 68 S.Ct. 1252, 92 L.Ed. 1690 (1948).
. A short statement of reasons relied upon by the district court for the imposition of a particular sentence would go a long way toward curbing abuses of discretion in sentencing :
The key to limiting abuse of judicial discretion is not to be found in reducing the sentencing judge’s options, but rather in providing definitive guidelines to specify what circumstances are aggravating and which mitigating. To leave these terms undefined is to invite their interpretation according to personal bias. Should this approach be adopted, a sentencing judge would be required to ground his decision on something less amorphous than his limitless and unreviewable discretion. Furthermore, the sentencing judge would be forced to incorporate as a part of the record those aggravating and mitigating circumstances upon wlich he relied in rendering a particular sentence. Without such a provision, personal bias might still prevail.
Zumwalt, The Anarchy of Sentencing in the Federal Courts, 57 Judicature 96, 104 (1973).
Similarly, Judge Frankel observes:
The point, anyhow, is that the parties (especially the loser) are, on deep principles, not merely entitled to a decision; they are entitled to an explanation. And this serves for more than the satisfaction of aesthetic or purely spiritual needs, though *903I disclaim any belittling of these. The duty to give an account of the decision is to promote thought by the decider, to compel him to cover the relevant points, to help him eschew irrelevancies — and, finally, to make him show that these necessities have been served. The requirements of reasons expressly stated is not a guarantee of fairness. The judge or other official may give good reasons while he acts upon outrageous ones. However, given decision-makers who are both tolerably honest and normally fallible, the requirement of stated reasons is a powerful safeguard against rash and arbitrary decisions.
M. Frankel, Criminal Sentences 40-41 (1973).
. The ABA Project on Minimum Standards for Criminal Justice, Standards Relating to Appellate Review of Sentences, states in pertinent part:
Finally, it should be observed that the Advisory Committee has had the benefit of a detailed study of appellate review of sentences in England, where sentence review has been the practice since 1907. The conclusion of that study is worth noting .here:
English lawyers and judges find it odd that the question whether sentences should be subject to appellate review is a matter of debate in the United States. The time before which . . . [the Court of Criminal Appeal] was established with authority to review sentences lies beyond the memory of the active members of the English legal profession today; they have never known another system. So to them sentencing review is taken for granted. In addition, though, everyone consulted believes it to be a good thing. Most think its chief virtue is that it serves as a protection against erratic sentences, the occasional peaks of excessiveness. A number of thoughtful persons also believe that it contributes greatly to sentencing policy and to an improved atmosphere for the sentencing process.
MEADOR REPORT, Appendix C, p. 130, infra. One would think that a country which prides itself for its protections against state invasion of liberty would also take pride in providing that protection where it often is needed the most, where the difference to the victim may be counted in years.
Id. at 6.
Bee also M. Frankel, Criminal Sentences 75-85 (1973) ; Zumwalt, The Anarchy of Sentencing in the Federal Courts, 57 Judicature 96, 104 (1973).
. The Commentary to this Standard explains the presumption in favor of probation:
The Advisory Committee believes that the starting point for every sentence should be probation or some other sentence not involving commitment or confinement, and that the extent to which commitment or confinement is employed in a given case should turn on the appearance of specific reasons which seem to call for that disposition. See § 2.2, supra and §§ 2.4(c) and 2.5(c), infra.
*906The results tend to support the conclusion that for many judges incarceration is the automatic sentencing response. More harm than good can be caused by such an attitude. Often institutionalization will result in little more than education of the offender in more sophisticated methods of engaging in criminal conduct, and certainly a great deal less than a realistic program of rehabilitation. Particularly in the case of first offenders, there is a much greater chance in most cases of avoiding a subsequent offense by helping the offender adjust to society than by removing him from it. See § 2.1, comment f, and § 2.2, comment a, supra.
In addition to the fact that probation often offers the best possibility of meaningful rehabilitation, it is considerably less expensive than institutionalization. Figures supplied to the Advisory Committee by the Administrative Office of the United States Courts indicate that in fiscal 1964 the per capita cost of probation in the federal system was 59 cents a day, while the cost of housing a prisoner in a federal institution was $6.35 per day. The annual figures were $215 and $2,318, respectively. The Administrative Office reports in addition that probationers during the period in question earned $62 million. Intangible costs in terms of impact on a family deprived of its breadwinner, including possible welfare payments, add to the cost of incarceration, not to mention the cost per inmate of the construction of facilities for housing prisoners. Compare Herlands, When and How Should a Sentencing Judge Use Probation, 35 F.R.D. 487, 498 (1964). Crime Commission figures disclose that the average state spends approximately $3400 a year, excluding capital expenditures, to keep a youth in a state training school, while probation for the same period costs approximately one-tenth of that amount. Construction costs are estimated to be in excess of $20,000 per bed in a correctional institution. While the Commission correctly points out that necessary increases are warranted in order to strengthen probation services, the expense of this sanction will still remain considerably below institutionalization. See President’s Comm’n, Corrections 28, 175-76.
Id. at 72-73.
. Apparently, Schneider even repayed some monies kept by his confederates, in addition to that which he had gained for himself.