dissenting.
“The Sentencing Reform Act of 1984 revolutionized the manner in which district courts sentence persons convicted of federal crimes.” Burns v. United States, 501 U. S. 129, 132 (1991). The goals of rehabilitation and fairness served by individualized sentencing that formerly justified vesting judges with virtually unreviewable sentencing discretion have been replaced by the impersonal interest in uniformity and retribution.1 Strict mandatory rules have dramatically *160confined the exercise of judgment based on a totality of the circumstances. “While the products of the Sentencing Commission’s labors have been given the modest name ‘Guidelines,’ . .. they have the force and effect of laws, prescribing the sentences criminal defendants are to receive. A judge who disregards them will be reversed.” Mistretta v. United States, 488 U. S. 361, 413 (1989) (Scalia, J., dissenting).
Í — I
In 1970, during the era of individualized sentencing, Congress enacted the statute now codified as 18 U. S. C. §3661 to make it clear that otherwise inadmissible evidence' could be considered by judges in the exercise of their sentencing discretion. The statute, however, did not tell the judge how to weigh the significance of any of that evidence. The judge was free to rely on any information that might shed light on a decision to grant probation, to impose the statutory maximum, or to determine the precise sentence within those extremes. Wisdom and experience enabled the judge to give appropriate weight to uncorroborated hearsay or to evidence of criminal conduct that had not resulted in a conviction. Even if convinced that a jury had erroneously acquitted a defendant, the judge was not required to ignore the evidence of guilt. At the same time, however, he or she was free to discount the significance of that evidence if mitigating circumstances — perhaps the same facts that persuaded the jury that an acquittal was appropriate — were present. Like a jury in a capital case, the judge could exercise discretion “to dispense mercy on the basis of factors too intangible to write into a statute,” Gregg v. Georgia, 428 U. S. 153, 222 (1976) (White, J., concurring in judgment). cabined
Although the Sentencing Reform Act of 1984 has cabined the discretion of sentencing judges, the 1970 statute remains on the books. As was true when it was enacted, § 3661 does *161not speak to questions concerning the relevance or the weight of any item of evidence. That statute is not offended by provisions in the Guidelines that proscribe reliance on evidence of economic hardship, drug or alcohol dependence, or lack of guidance as a youth in making certain sentencing decisions. See Koon v. United States, 518 U. S. 81, 93 (1996). Conversely, that statute does not command that any particular weight — or indeed that any weight at all — be given to evidence that a defendant may have committed an offense that the prosecutor failed to prove beyond a reasonable doubt. In short, while the statute that introduces the Court’s analysis of these cases, ante, at 151, does support its narrow holding that sentencing courts may sometimes “consider conduct of the defendants underlying other charges of which they had been acquitted,” ante, at 149, it sheds no light on whether the district judges’ application of the Guidelines in the manner presented in these cases was authorized by Congress, or is allowed by the Constitution.
A closer examination of the interaction among §3661, the other provisions of the Sentencing Reform Act, and the Guidelines demonstrates that the role played by § 3661 is of a narrower scope than the Court’s opinion suggests. The Sentencing Reform Act was enacted primarily to address Congress’ concern that similar offenders convicted of similar offenses were receiving “an unjustifiably wide range of sentences.” S. Rep. No. 98-225, p. 38 (1983). It therefore created the Sentencing Commission and directed it to draft Guidelines that would cabin the discretion of all judges— those who were too harsh as well as those who were too lenient. See 28 U. S. C. § 991(b)(1)(B). While the abolition of parole indicates that the new rules were generally intended to increase the minimum levels of punishment, see 18 U. S. C. §§ 3624(a) and (b), they also confined the judges’ authority to impose the maximum sentences authorized by statute. The central mechanism that Congress promulgated to avoid disparate sentencing in typical cases is a require*162ment that for any sentence of imprisonment in the Guidelines, “the maximum of the range established for such a term shall not exceed the minimum of that range by more than the greater of 25 percent or 6 months,” 28 U. S. C. § 994(b)(2). The determination of which of these narrow ranges a particular sentence should fall into is made by operation of mandatory rules, but within the particular range, the judge retains broad discretion to set a particular sentence. the
By their own terms, the Guidelines incorporate the broadly inclusive language of §3661 only into those portions of the sentencing decision in which the judge retains discretion.
United States Sentencing Commission, Guidelines Manual § 1B1.4 (Nov. 1995) (USSG) provides:
“In determining the sentence to impose within the guideline range, or whether a departure from the guidelines is warranted, the court may consider, without limitation, any information concerning the background, character and conduct of the defendant, unless otherwise prohibited by law. See 18 U. S. C. §3661.”
Thus, as in the pre-Guidelines sentencing regime, it is in the area in which the judge exercises discretion that §3661 authorizes unlimited access to information concerning the background, character, and conduct of the defendant. When the judge is exercising such discretion, I agree that he or she may consider otherwise inadmissible evidence, including evidence adduced in a trial that resulted in an acquittal. But that practice, enshrined in §3661 and USSG §1B1.4, sheds little, if any, light on the appropriateness of the District Courts’ application of USSG §1B1.3, which defines relevant conduct for the purposes of determining the Guidelines range within which a sentence can be imposed.
II
The issue of law raised by the sentencing of Cheryl Putra involved the identification of the offense level that deter*163mined the range within which the judge could exercise discretion. Because she was a first offender with no criminal history, that range was based entirely on the offense or offenses for which she was to be punished. She was found guilty of aiding and abetting the intended distribution of one ounce of cocaine on May 8,1992, but not guilty of participating in a similar transaction involving five ounces of cocaine on May 9,1992. United States v. Putra, 78 F. 3d 1386, 1387 (CA9 1996). If the guilty verdict provided the only basis for imposing punishment on Ms. Putra, the Guidelines would have required the judge to impose a sentence of no less than 16 months in prison and would have prohibited him from imposing a sentence longer than 21 months.
If Putra had been found guilty of also participating in the 5-ounce transaction on May 9, 1992, the Guidelines would have required that both the minimum and the maximum sentences be increased; the range would have been between 27 and 33 months. As the District Court applied the Guidelines, precisely the same range resulted from the acquittal as would have been dictated by a conviction. Notwithstanding the absence of sufficient evidence to prove guilt beyond a reasonable doubt, the alleged offense on May 9 led to the imposition of a sentence six months longer than the maximum permitted for the only crime that provided any basis for punishment.2
*164In my judgment neither our prior cases nor the text of the statute warrants this perverse result. And the vigor of the debate among judges in the Courts of Appeals on this basic issue belies the ease with which the Court addresses it, without hearing oral argument or allowing the parties to fully brief the issues.3
*165III
The Court relies principally on three cases — Williams v. New York, 337 U. S. 241 (1949); McMillan v. Pennsylvania, 477 U. S. 79 (1986); and Witte v. United States, 515 U. S. 389 (1995) — to justify its outcome. In each instance, the reliance is misplaced.
For three reasons, Williams cannot support the result in these cases. First, it dealt with the exercise of the sentencing judge’s discretion within the range authorized by law, rather than with rules defining the range within which discretion may be exercised. Second, “[t]he accuracy of the statements made by the judge as to appellant’s background and past practices was not challenged by appellant or his counsel, nor was the judge asked to disregard any of them or to afford appellant a chance to refute or discredit any of them by cross-examination or otherwise.” 337 U. S., at 244. The precise question here — the burden of proof applicable to sentencing facts — was thus not before the Court in that case. Third, its rationale depended largely on agreement with an individualized sentencing regime that is significantly different from the Guidelines system. “Williams was decided in the context of a sentencing ‘system that focuse[d] on subjective assessments of rehabilitative potential. . . .’ Saltzburg, [Sentencing Procedures: Where Does Responsibility Lie?, 4 Fed. Sent. Rep. 248,250 (1992)].” United States v. Wise, 976 F. 2d 393, 409 (CA8 1992) (Arnold, C. J., concurring in part and dissenting in part). As this Court has acknowledged, see Burns, 501 U. S., at 132, the Guidelines wrought a dramatic change in sentencing processes, replacing the very system that justified Williams with a rigid system in which, “[f]or most defendants in the federal courts, sentencing is what the case is really about.” Wise, 976 F. 2d, at 409.
Even more than Williams, this Court, like all of the Circuits that have adopted the same approach as the District Courts in these cases, relies primarily on the misguided *1665-to-4 decision in McMillan v. Pennsylvania, 477 U. S. 79 (1986). For the reasons stated in my dissent in that case, ■id., at 96-104, I continue to believe that it was incorrectly decided and that its holding should be reconsidered. Even accepting its holding that the Constitution does not require proof beyond a reasonable doubt to establish a sentencing factor that increases the minimum sentence without altering the maximum, however, there are at least two reasons why McMillan does not dictate the outcome of these cases.
In McMillan, as in these cases, the defendant’s minimum sentence was enhanced on the basis of a fact proved by a preponderance of the evidence. But in McMillan, the maximum was unchanged; the sentence actually imposed was within the range that would have been available to the judge even if the enhancing factor had not been proved. In these cases, however, the sentences actually imposed were higher than the Guidelines would have allowed without evidence of the additional offenses. The McMillan opinion pointedly noted that the Pennsylvania statute had not altered “the maximum penalty for the crime committed” and operated “solely to limit the sentencing court’s discretion in selecting a penalty within the range already available to it without the special finding of visible possession of a firearm.” Id., at 87-88. Given the Court’s acknowledged “inability to lay down any 'bright line’ test” that would define the limits of its holding, id., at 91, and its apparent assumption that a sentencing factor should not be allowed to serve as a “tail which wags the dog of the substantive offense,” id., at 88, see also ante, at 156-157, n. 2, the holding should not be extended to allow a fact proved by only a preponderance to increase the entire range of penalties within which the sentencing judge may lawfully exercise discretion.4
*167Moreover, McMillan addressed only the constitutionality of a statute the meaning of which was perfectly clear. Nothing in the text of the Sentencing Reform Act of 1984 even arguably mandates the result that the District Courts reached in these cases. Indeed, as Justice Breyer points out in his separate concurrence, ante, at 159, the Sentencing Commission unquestionably has the authority to disallow the consideration of acquitted conduct. Similarly, the Commission could have chosen to set the burden of proof for sentencing proceedings at beyond a reasonable doubt without running afoul of the enabling legislation. Given the lack of a contrary command in the statute itself, as well as the complete absence of any pre-1984 precedent for establishing the range of a permissible sentence on the basis of a fact proved only by a preponderance of the evidence, the McMillan opinion which was announced in 1986 can shed no light on the meaning of the 1984 Act.
Nor does the Court’s decision in Witte v. United States, 515 U. S. 389 (1995), dictate the answer to the question presented by these cases. I continue to disagree with the conclusion reached by the Court in Witte, that the Double Jeopardy Clause does not prohibit convicting and sentencing an individual for conduct that has been decisive in determining the individual’s offense level for a previous conviction. But that is a different issue from the one here. The opinion in Witte, carefully and repeatedly, confined the Court’s holding to the double jeopardy context. Id., at 397 (defendant in this case “is punished, for double jeopardy purposes, only for the offense of which the defendant is convicted”); id., at 399 (disputed practice is not “punishment for that conduct within the meaning of the Double Jeopardy Clause”); id., at 404 (practice “constitutes punishment only for the offense of conviction for purposes of the double jeopardy inquiry”). What is *168at issue in these cases is not whether a defendant is being twice punished or prosecuted for the same conduct, but whether his or her initial punishment has been imposed pursuant to rules that are authorized by the statute and consistent with the Constitution.
IV
Putra's case involves “multiple offenses.” She was charged with several offenses and received a sentence that was based on the judge’s conclusion that she was guilty of each of these multiple offenses even though she had in fact been found guilty of only one offense. It is therefore appropriate to consider what the Sentencing Reform Act has to say about “multiple offenses.” the
say In 28 U. S. C. § 994(1) Congress specifically directed the Commission to ensure that the Guidelines included incremental sentences for multiple offenses. That subsection provides:
“The Commission shall insure that the Guidelines promulgated . .. reflect—
. .. “(1) the appropriateness of imposing an incremental penalty for each offense in a case in which a defendant is convicted of—
“(A) multiple offenses committed in the same course of conduct. .. .; and
.. “(B) multiple offenses committed at different times . . . .” (Emphasis added.)
It is difficult to square this explicit statutory command to impose incremental punishment for each of the “multiple offenses” of which a defendant “is convicted” with the conclusion that Congress intended incremental punishment for each offense of which the defendant has been acquitted.5 *169The Court, however, appears willing to read the statute’s treatment of multiple offenses as though it authorized an incremental penalty for each offense for which the defendant was indicted if she is convicted of at least one such offense. The fact that the text of the statute expressly authorizes such incremental punishment “for each offense” only when a “defendant is convicted of . . . multiple offenses” conveys « far different message to thoughtful judges.6
In my opinion the statute should be construed in the light of the traditional requirement that criminal charges must be sustained by proof beyond a reasonable doubt. That requirement has always applied to charges involving multiple offenses as well as a single offense. Whether an allegation of criminal conduct is the sole basis for punishment or merely one of several bases for punishment, we should presume that Congress intended the new sentencing Guidelines that it authorized in 1984 to adhere to longstanding procedural requirements enshrined in our constitutional jurisprudence. *170The notion that a charge that cannot be sustained by proof beyond a reasonable doubt may give rise to the same punishment as if it had been so proved is repugnant to that jurisprudence.
I respectfully dissent.
Compare Williams v. New York, 837 U. S. 241, 247-248 (1949) (“Reformation and rehabilitation of offenders have become important goals of criminal jurisprudence”), with 28 U. S. C. § 994(k) (rejecting rehabilitation as a goal of imprisonment) and 18 U. S. C. § 3553(a)(2) (stating that punish*160ment should serve retributive, deterrent, educational, and incapacitative goals).
The circumstances surrounding Vernon Watts’ sentencing were somewhat different from those involved in Putra’s sentencing. Watts was acquitted of the crime of using a firearm in relation to a drug offense, in violation of 18 U. S. C. § 924(c), but was found guilty of certain drug crimes. United States v. Waits, 67 F. 3d 790, 793 (CA9 1995). The sentencing judge enhanced Watts’ base offense level by two points, pursuant to USSG §2Dl.l(b)(l), after concluding that the defendant’s “possession” of the firearm in connection with the crime had been proved by a preponderance of the evidence. 67 F. 3d, at 797-798. Because the “use” of a firearm and its “possession” are not identical, the judge may not have relied on facts necessarily rejected by the jury in concluding that the sentencing enhancement was appropriate. I nevertheless believe that the enhancement *164was inappropriate because it was based on conduct that the judge found only by a preponderance of the evidence. Since Watts’ base offense level was increased by this evidence, I believe it should have been proved beyond a reasonable doubt.
Although the Court’s decision suggests that the approach taken by the Ninth Circuit in these cases breaks from settled law in every other Circuit, the opinion ignores the fact that respected jurists all over the country have been critical of the interaction between the Sentencing Guidelines’ mechanical approach and the application of a preponderance of the evidence standard to so-called relevant conduct. See, e. g., United States v. Silverman, 976 F. 2d 1502, 1519, 1527 (CA6 1992) (Merritt, C. J., dissenting); id., at 1533 (Martin, J., dissenting); United States v. Concepcion, 983 F. 2d 369, 389, 396 (CA2 1992) (Newman, C. J., concurring) (“A just system of criminal sentencing cannot fail to distinguish between an allegation of conduct resulting in a conviction and an allegation of conduct resulting in an acquittal”); United States v. Galloway, 976 F. 2d 414, 436 (CA8 1992) (Bright, J., dissenting, joined by Arnold, C. J., Lay, J., and McMillian, J.); United States v. Restrepo, 946 F. 2d 654, 663 (CA9 1991) (Pregerson, J., dissenting, joined by Hug, J.); id., at 664 (Norris, J., dissenting, joined by Hug, J., Pregerson, J., and D. W. Nelson, J.). Cf. United States v. Lanoue, 71 F. 3d 966, 984 (CA1 1995) (“Although it makes no difference in this case, we believe that a defendant’s Fifth and Sixth Amendment right to have a jury determine his guilt beyond a reasonable doubt is trampled when he is imprisoned (for any length of time) on the basis of conduct of which a jury has necessarily acquitted him”). See also Martin, The Cornerstone Has No Foundation: Relevant Conduct in Sentencing and the Requirements of Due Process, 3 Const. L. J. 25, 34-36 (1993); Beale, Procedural Issues Raised by Guidelines Sentencing: The Constitutional Significance of the “Elements of the Sentence,” 35 Wm. & Mary L. Rev. 147, 157-158 (1993); Herman, The Tail that Wagged the Dog: Bifurcated Fact-Finding under the Federal Sentencing Guidelines and the Limits of Due Process, 66 S. Cal. L. Rev. 289 (1992); Heaney, The Reality of Guidelines Sentencing: No End to Disparity, 28 Am. Crim. L. Rev. 161, 208-220 (1991).
I recognize that the shift from one Guideline range to a higher range does not produce a sentence beyond the statutory maximum. It does, however, mandate a sentence that is above the maximum that the judge *167would have had the legal authority to impose absent consideration of the “relevant conduct.”
Courts upholding the Guidelines’ relevant conduct provisions and their application in cases such as these have tended to focus their attention exclusively on those provisions in the statute that direct courts and the Commission to consider the “nature and circumstances of the offense” in *169determining an appropriate sentence. 18 U. S. C. § 3553(a)(1); see also 28 U. S. C. § 994(d). In § 994(d), Congress granted the Sentencing Commission the authority to “consider whether [certain enumerated factors], among others, have any relevance” in establishing Guidelines for offenses. Some courts have concluded that the inclusion of the qualifier “among others” in this provision indicated that Congress intended the Commission to include anything it felt was relevant to the sentencing decision. See, e. g., United States v. Galloway, 976 F. 2d, at 420-421; United States v. Thomas, 932 F. 2d 1085, 1089 (CA5 1991), cert. denied sub nom. Pullock v. United States, 502 U. S. 895, and Samuels v. United States, 502 U. S. 962 (1992).
But this provision cannot be read separately from the rest of the statute. The clear congressional directive concerning sentencing for “multiple offenses” must be read as an important limit on the “othe[r]” factors that can be considered relevant to determination of an offense level.
Some judges have concluded, in large part because of this provision, that the Guidelines’ relevant conduct rules are outside the scope of the authority Congress granted to the Commission. See Galloway, 976 F. 2d, at 430-431 (Beam, J., dissenting); United States v. Davern, 970 F. 2d 1490, 1507 (CA6 1992) (Merritt, C. J., dissenting).