concurring in the judgment:
I too would remand this case for resen-tencing but for different reasons than those expressed by the majority.
I.
My reservations about the majority’s reasoning are twofold. First, although ostensibly reviewing the district court's exercise of discretion under Fed.R.Crim.P. 35, the majority conducts what appears to be akin to proportionality review. The majority claims that the district court did not exercise its sentencing discretion in any “dis-cernable or meaningful way,” At 1004, and that it weighed “only one side of the discretionary scale.” Id. at 1006. In truth, the district court did exercise its discretion and found the nature of the unlawful conduct such as to outweigh any factors adduced in mitigation of it. The decision to overturn the exercise of such discretion propels federal appellate courts on the basis of little more than personal preference into what is indisputably a trial court function. In its apparent belief that the sentence here is disproportionately severe relative to the offense committed, the appellate court has effectively supplanted the district court’s discretion with its own.
Second, the majority emphasizes Gugliel-mi’s prison record, personal character, and individual family circumstances and criticizes the district court for failing to accord these factors significant weight. I question whether such factors can provide a proper ground for finding an abuse of discretion, particularly in light of the Sentencing Guidelines which explicitly de-empha-size such considerations in favor of offense-based sentencing. See 28 U.S.C. §§ 991(b)(1)(B), 994(e). Although this case involves pre-Guidelines sentencing, the Guidelines surely cannot be irrelevant to our inquiry. The district court’s approach with respect to personal factors is similar to that adopted by Congress and I find it difficult to reverse the district court for abuse of discretion when its emphasis upon the nature of the offense rather than upon the characteristics of the defendant has since become the approved approach to sentencing.
II.
Nevertheless, I cannot accept the sentence in this ease. The severity of a sentence ought not to be purely idiosyncratic but should be pursuant to a legislative mandate. See United States v. Smith, 686 F.2d 234, 239 (5th Cir.1982) (“It is for Congress to say what shall be a crime and how that crime shall be punished; it is not the prerogative of this Court.”); United States v. Holmes, 838 F.2d 1175, 1178 (11th Cir.1988). Capital punishment, the most severe penalty society can visit upon a wrongdoer, has been imposed pursuant to the clear mandate of many legislatures and *1009generally by the community speaking through a jury. Similarly, the Guidelines may appear strict at times, but they are authorized by Congress and set broad parameters for the exercise of discretion. One of the reasons the Guidelines were promulgated was to insure that punishment would actually reflect society’s sentiments and not a personally idiosyncratic view; the framers “sought reasonable uniformity in sentencing by narrowing the wide disparity in sentences imposed for similar criminal offenses committed by similar offenders.” United States Sentencing Commission, Guidelines Manual, Ch. 1, Pt. A, § 3, p.s. (Nov. 1990) [hereinafter U.S.S.G.].
Two reasons in combination lead to the conclusion that the sentence here is indeed idiosyncratic and beyond the scope of any legislative mandate. First, the length of the sentence is the consequence of an artificial construct. The assigned twenty-five year imprisonment was based on five consecutive shipments of films that led to five consecutive five-year sentences. The number of shipments, and thus the magnitude of the possible sentence, could easily be manipulated. This aspect is not in itself fatal because any undercover operation necessarily entails some discretion in the frequency and the amount of undercover purchases. The difficulty lies, however, in the apparent lack of any limiting principle. Legislative will can be distorted if a sentence is simply run up by a continuing series of orders. Indeed, the Sentencing Commission was aware of that possibility and designed the Guidelines “with an eye toward eliminating unfair treatment that might flow from count manipulation.” U.S.S.G. Ch. 1, Pt. A, § 4(a), p.s. “For example, the guidelines treat a three-count indictment, each count of which charges sale of 100 grams of heroin or theft of $10,000, the same as a single-count indictment charging sale of 300 grams of heroin or theft of $30,000.” Id.; see id. § 3D1.2(b), (d). The artificial nature of Gu-glielmi’s sentence, while not solely determinative, surely must be one factor in the inquiry.
The second indication that the sentence here is idiosyncratic is the Guidelines’ computation of punishment for Guglielmi’s conduct. The Guidelines mandate a sentence of eight to fourteen months for someone with no criminal record, enhanced potentially to eighteen to twenty-four months if the obscene films also portrayed “sadistic or masochistic conduct or other depictions of violence.” U.S.S.G. § 2G3.1(b)(2). The sentence Guglielmi received was twenty-five years. We cannot presume that Gu-glielmi will receive parole after serving eight years and four months or at any subsequent time; such an assumption would be beyond our province and would constitute mere speculation on our part. The Guidelines, of course, are not binding on the district court in this case. Nevertheless, they remain the best indication of society’s views as expressed through legislative authorization, and they are quite at odds with what has happened here. The district court’s sentence is, in fact, the sort of abuse that led to the promulgation of the Guidelines in the first place.
Setting aside a sentence imposed by a conscientious district judge should be a rare thing. But the wrong done an individual by purely idiosyncratic punishment may be so apparent as to constitute one of those rare instances where recourse may be had to an appellate court. Because the sentence here strayed far from any legislative sanction, I concur in the judgment.