View PDF Version

No. 01-687

In the Supreme Court of the United States






Solicitor General
Counsel of Record
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

In the Supreme Court of the United States

No. 01-687






Respondents contend that the error in this case-the imposition of an enhanced sentence in the absence of a grand jury finding on the enhancing fact-is jurisdictional, structural, and reversible under the harmless-error or plain-error standard. Each of those submissions is incorrect. The error here-like the error in imposing a conviction in the absence of a petit jury finding on an essential fact-is subject to harmless-error and plain-error review, and, under those standards, the sentences in this case should be sustained.


Respondents assert (Br. 12-21) that, because the district court lacked authority in light of Apprendi v. New Jersey, 530 U.S. 466 (2000), to impose a sentence that exceeds the otherwise-applicable statutory maximum based on a fact not alleged in the indictment, the court committed a "jurisdictional error." Respondents are mistaken. The error in this case, while a constitutional error under Apprendi, is not a jurisdictional error.1

1. The category of "jurisdictional errors," which always require reversal even if not preserved, is an exceedingly narrow one. A court commits a "jurisdictional error" in that sense only if it acts in the absence of subject-matter jurisdiction-an error that, unlike other errors in a criminal prosecution, cannot be waived by the parties in any circumstance. The right to indictment by a grand jury may be waived, as respondents appear to acknowledge (Br. 17). 2

As this Court has explained, "[j]urisdiction is the power to decide a justiciable controversy," United States v. Williams, 341 U.S. 58, 66 (1951), "and covers wrong as well as right decisions," Lamar v. United States, 240 U.S. 60, 64 (1916). Thus, "[o]nce subject-matter jurisdiction has properly attached, courts may exceed their authority or otherwise err without loss of jurisdiction." Prou v. United States, 199 F.3d 37, 45 (1st Cir. 1999) (district court's error in imposing enhanced sentence based on prior convictions not alleged in statutorily required information was not jurisdictional); see United States v. Wey, 895 F.2d 429, 431 (7th Cir. 1990) ("Courts may err, even offend against the Constitution, without losing subject-matter jurisdiction.").3

Clearly, the district court had subject-matter jurisdiction over this case by virtue of 18 U.S.C. 3231, which gives district courts "original jurisdiction * * * of all offenses against the laws of the United States." The superseding indictment alleged a complete federal felony offense under 21 U.S.C. 846. Accordingly, just as a court is not "oust[ed] * * * of jurisdiction" when it convicts a defendant under an unconstitutional statute, Williams, 341 U.S. at 66, the district court was not ousted of jurisdiction when it sentenced respondents under a procedure that was subsequently held to be unconstitutional.

Contrary to respondents' suggestion (Br. 19), the error in this case is unlike the error in cases such as In re Bonner, 151 U.S. 242 (1894), in which the court imposed a punishment that was not authorized by Congress. Here, Congress expressly authorized a maximum penalty of life imprisonment for drug trafficking offenses that involve at least 50 grams of cocaine base, as did respondents' offense in this case. See 21 U.S.C. 841(b)(1)(A)(iii). The error results from the absence of procedural prerequisites to the imposition of that sentence. Cf. Libretti v. United States, 516 U.S. 29, 56 & n.1 (1995) (Stevens, J., dissenting) (observing that, although "extrastatutory punishments implicate the very power of a court to act," "the court's power to act is not similarly implicated when it imposes a sentence that is arguably erroneous but nonetheless within the range authorized by statute").

2. Respondents rely (Br. 12, 15, 49) on Ex parte Bain, 121 U.S. 1, 13 (1887), which held that a court had erred in trying a defendant on a theory narrower than the one alleged in the indictment, and described that error as "jurisdiction[al]" in nature. As previously explained (U.S. Br. 40-42), Bain dates from an era in which the Court considered it "well settled" that its power to grant habeas corpus relief was limited to cases in which "the [lower] court had no jurisdiction to render the judgment which it gave." Bain, 121 U.S. at 3. As a consequence, the Court expanded the category of "jurisdictional" errors to include various constitutional and statutory errors made by courts that did not lack jurisdiction. See Wainwright v. Sykes, 433 U.S. 72, 79 (1977) (observing that the Court's use of "the concept of jurisdiction" to determine the availability of habeas relief ultimately became "more a fiction than anything else").

Since that era, while the Court has continued to find reversible error when an indictment did not adequately allege the offense of conviction, the Court has not continued to characterize such error as "jurisdictional." See, e.g., Silber v. United States, 370 U.S. 717 (1962) (per curiam); Russell v. United States, 369 U.S. 749 (1962); Stirone v. United States, 361 U.S. 212 (1960). Indeed, the Court twice described the error in Silber as one that a reviewing court "may" notice, 370 U.S. at 718, implying a degree of discretion inconsistent with the treatment of true jurisdictional errors. Bain itself has been overruled, United States v. Miller, 471 U.S. 130 (1985), and whatever survives of its reasoning need not be characterized as "jurisdictional." See id. at 142-143 (agreeing in dictum with the "aspect" of Bain that a court commits error by convicting a defendant of "an offense different from that which was included in the indictment," without suggesting that such an error is jurisdictional).

Reviving the expansive concept of "jurisdictional error" used in Bain would undermine the concerns for finality and judicial economy underlying the harmless-error and plain-error rules. Many errors in a criminal prosecution may, to the same extent as the error here, be characterized as a court's exceeding its constitutional or statutory authority- including the errors that were held not to require reversal in Johnson v. United States, 520 U.S. 461 (1997), and Neder v. United States, 527 U.S. 1 (1999), where the courts exercised authority that, under the Sixth Amendment, should have been exercised by the petit jury.4

3. Contrary to respondents' suggestion (Br. 20), Rule 12(b)(2) of the Federal Rules of Criminal Procedure does not preclude harmless-error or plain-error review in this case. Rule 12(b) specifies various defenses and objections that "must be raised prior to trial." Rule 12(b)(2) includes within that category "defenses and objections based on defects in the indictment or information (other than that it fails to show jurisdiction in the court or to charge an offense which objections shall be noticed by the court at any time during the pendency of the proceedings)." Read in context, the parenthethical phrase provides only that the described objections, even if not raised before trial, are not waived. See Fed. R. Crim. P. 12(f). It does not address the standard that a reviewing court is to apply to such objections, nor does it render inapplicable a rule "having apparently equal dignity," Rule 52 of the Federal Rules of Criminal Procedure. United States v. Vonn, 122 S. Ct. 1032, 1050 (2002).

In any event, the error in this case does not come within either the general category described by Rule 12(b)(2) or the two subcategories described in its parenthetical phrase. The superseding indictment showed subject-matter jurisdiction in the district court and charged an offense under 21 U.S.C. 846. See U.S. Br. 36 n.11. The error in this case occurred at sentencing when the district court imposed an enhanced sentence based on a finding of drug quantity in the absence of required procedural protections. That error is not in its essence based on a "defect" in the indictment, the subject of Rule 12(b)(2).


Respondents contend (Br. 22-38), principally based on this Court's decisions in Silber, Russell, and Stirone, that the district court committed a "structural" error in imposing an enhanced sentence based on a fact not alleged in the indictment. Those cases, however, predate this Court's comprehensive adoption of harmless-error analysis for constitutional errors in Chapman v. California, 386 U.S. 18 (1967). Consequently, the cases cannot be assumed to reflect the current distinction between errors that are, and are not, subject to harmless-error analysis. U.S. Br. 32, 34.

Equally important here, the Court has unanimously recognized that even structural errors, if not preserved, do not entitle the defendant to automatic reversal on plain-error review. See Johnson, 520 U.S. at 467. Indeed, the dissenting Justices in Neder, who would have held that the failure to submit an offense element to the petit jury was a structural error, recognized that the error would not automatically require reversal if not objected to at trial. See 527 U.S. at 34 (Scalia, J., concurring in part and dissenting in part). Respondents, who failed to make a constitutional objection at trial, could not prevail without satisfying the requirements of Rule 52(b), even if the error were structural. But the error here is not.

1. Since Chapman, when the Court has described the "very limited class of cases" involving errors that are exempt from harmless-error analysis, the Court has not included Silber, Russell, Stirone, or any other case in which a defendant was convicted or sentenced based on a fact not alleged in the indictment or information. See Neder, 527 U.S. at 8; Johnson, 520 U.S. at 468-469. Nor is such an error analogous to the errors in that "very limited class." As the Court has explained, when a defendant is tried without counsel or before a biased judge or on a charge returned by a grand jury from which one race was excluded, the error "infect[s] the entire trial process" and thus "necessarily render[s] a trial fundamentally unfair." Neder, 527 U.S. at 8. The Court held in Neder, however, that an error that involves a single discrete fact, which was not submitted to the petit jury, is not of that character. Here, the error likewise involves a single discrete fact, which was not submitted to the grand jury, and thus was not the sort of error that "infect[s] the entire trial process."

2. Respondents nonetheless suggest (Br. 24-25) that the error here is analogous not to the error in Neder, but to the defective reasonable doubt instruction in Sullivan v. Louisiana, 508 U.S. 275 (1993). As the Court explained in Neder, however, "the error [in Sullivan] was not subject to harmless-error analysis because it 'vitiates all the jury's findings,' and produces 'consequences that are necessarily unquantifiable and indeterminate.'" 527 U.S. at 11 (quoting Sullivan, 508 U.S. at 281, 282). In contrast, the error in Neder did not vitiate any of the petit jury's findings, but simply prevented the petit jury from making an additional finding on the materiality element. Ibid. Similarly, the district court's error in this case does not implicate any of the grand jury's (or petit jury's) findings on any of the elements of a drug conspiracy offense under 21 U.S.C. 846. The district court's error was in relying on one additional fact, which the grand jury (and the petit jury) had not been asked to find, in imposing sentence.

Nor does the error in this case, any more than the error in Neder, produce "necessarily unquantifiable and indeterminate" consequences. Sullivan, 508 U.S. at 281-282. The errors in both cases involved a district court's reliance on a fact that was not submitted to the correct fact-finder. In both cases, a reviewing court can reliably determine whether the error was prejudicial by inquiring whether the correct fact-finder, examining the entire record under the applicable standard of proof, would have found the fact, and whether the defendant had adequate notice of the fact and opportunity to contest it. See Neder, 527 U.S. at 17; Johnson, 520 U.S. at 468-469 (applying similar analysis under fourth component of plain-error standard).5

Respondents contend (Br. 24-25) that the omission of drug quantity from the superseding indictment affected the "framework" of the trial, and therefore qualifies as a "structural" error. But all parties to this case understood that the punishment scheme of 21 U.S.C. 841(b) made drug quantity a critical fact bearing on the maximum punishment available under the statute. All parties also understood that the district court (erroneously, in hindsight) would determine that fact at sentencing. The absence of an allegation in the indictment on the threshold drug quantities did not mislead respondents into thinking that drug quantity was irrelevant to the prosecution.

Apparently focusing on the petit jury's failure to find threshold drug quantities, respondents attempt (Br. 33) to distinguish Neder on the ground that the district court in that case found the materiality element "under the correct standard of proof," whereas the district court in this case found drug quantity by a preponderance of the evidence. But Neder's analysis of whether the error should be classified as "structural" did not turn on the standard of proof applied by the district court in finding a fact that should have been submitted to the petit jury. See 527 U.S. at 8-15. Indeed, the Court relied (id. at 10) on earlier cases applying harmless-error analysis to erroneous jury instructions where the omitted element was not found by the judge or jury to have been proved beyond a reasonable doubt. See, e.g., Rose v. Clark, 478 U.S. 570, 580 (1986). Here, moreover, the grand jury was not required to find threshold drug quantity beyond a reasonable doubt to return an indictment; all that was required was a finding of probable cause.6

3. Respondents suggest various reasons why the district court's imposition of a sentence based on a fact not submitted to the grand jury should be regarded as a structural error. But none of those reasons justifies distinguishing the grand jury from the petit jury with respect to structural error. Likewise, none justifies a per se rule of reversal on plain-error review for grand jury omissions, contrary to the approach taken in Johnson to petit jury omissions.

First, respondents contend (Br. 24) that the error in this case is structural, because the right to grand jury indictment "protect[s] not only the individual accused but also the system of justice." That is equally true of the right to trial by a petit jury. Yet the Court held in Neder that, if a district court violates that right by convicting a defendant based on a fact not submitted to the petit jury, the error is not structural. The same conclusion logically applies to a district court's error in sentencing a defendant based on a fact not submitted to the grand jury. The protection provided by the grand jury and the petit jury alike-to "the system of justice" as well as "the individual accused"-is against erroneous determinations of fact by a malicious or overzealous government official, whether a prosecutor or a judge. See, e.g., Neder, 527 U.S. at 19; Wood v. Georgia, 370 U.S. 375, 390 (1962). That protection is not undermined when a reviewing court is able to conclude, beyond a reasonable doubt, that a grand jury or a petit jury would have found the fact, if asked. See Neder, 527 U.S. at 19.

Second, respondents assert (Br. 25) that the error in this case implicates the "structural allocation of powers among the judiciary, the executive, and citizenry," and specifically the Framers' choice "to place the power to initiate a federal criminal proceeding with the citizens-not with the judge, not with the prosecutor." The Framers also chose to place the (at least) equally important power to decide guilt or innocence in a federal criminal proceeding with "the citizens," i.e., the petit jury. Yet, if a district court encroaches on that power by deciding an element that should have been decided by the petit jury, the error is not structural under Neder. The same conclusion applies when a district court decides an element or sentencing-enhancing fact that should have been, but was not, decided by the grand jury.7

Moreover, this case was initiated by a grand jury, just as Neder was decided by a petit jury. In Neder, the Court distinguished between the failure to obtain a petit jury finding on one element, an error that is subject to harmless-error analysis, and the failure to obtain a petit jury finding on any element, an error that is not subject to harmless-error analysis. See 527 U.S. at 10-11 (contrasting errors in Neder and Sullivan). Similarly, the Court could distinguish between the error in this case, which involved the district court's reliance at sentencing on a fact that was not submitted to the grand jury, from the more pervasive errors in other cases cited or hypothesized by respondents. See, e.g., United States v. Moreland, 258 U.S. 433 (1922) (defendant tried on felony charge without indictment). As the Court observed in Neder, "our course of constitutional adjudication has not been characterized by th[e] 'in for a penny, in for a pound' approach." 527 U.S. at 17 n.2.

Third, respondents argue (Br. 26) that any inquiry into whether the error in this case was harmless would be inappropriate, because "the grand jury's decision not to charge cannot be modified by a prosecutor, a petit jury, or a judge, whatever the evidence may be." A petit jury's decision to acquit is no less sacrosanct than a grand jury's decision not to charge. Indeed, an acquittal precludes any further prosecution of an offense, while a grand jury's rejection of an indictment does not. U.S. Br. 21. Yet, the Court has held that harmless-error analysis applies when the petit jury did not acquit, but instead found the defendant guilty, although without considering one element of the offense. It is equally appropriate to apply harmless-error analysis when the grand jury did not decline to charge, but instead charged the defendant with an offense, although without expressly considering one element or sentencing-enhancing fact.

Respondents rely (Br. 26) on the Court's observation in Vasquez v. Hillery, 474 U.S. 254, 263 (1986), that a "grand jury is not bound to indict in every case where a conviction can be obtained." That statement, however, which is not essential to the holding in that case, is in tension with the historical evidence, which indicates that grand juries traditionally have been expected to act in accordance with the law and to indict whenever the evidence establishes probable cause. See U.S. Br. 26-29, 30-31. Respondents do not address, much less refute, that historical evidence, which includes grand jury charges by earlier members of this Court instructing that the grand jury had a "duty" to indict whenever the evidence was sufficient.

To the contrary, respondents' own historical sources agree that the grand jury's role is to protect the "[i]nnocent," not to protect persons when there is probable cause to charge a crime. Br. 28 (quoting English Liberties, or the Free-born Subject's Inheritance 252 (Henry Care ed., 4th ed. 1719)); Br. 30 (quoting statement at Massachusetts ratifying convention). Respondents also point to a handful of politically sensitive cases in which grand juries declined to indict opponents of the Crown. See Br. 28 n.9, 29 n.10. At most, however, those cases stand for the proposition that grand juries, like petit juries, have "the power * * * to 'nullify' or exercise a power of lenity," as distinguished from "a right or something that a judge should encourage or permit if it is within his authority to prevent." United States v. Thomas, 116 F.3d 606, 615 (2d Cir. 1997).8

4. Respondents acknowledge (Br. 31) that the Fifth Amendment right to grand jury indictment, unlike the Sixth Amendment right to trial by a petit jury, has not been extended to the States through the Fourteenth Amendment as an essential component of a fair criminal proceeding. But respondents attempt to avoid the logical implication of the Court's different treatment of those two rights: that, if a district court's reliance on a fact not found by the petit jury is not so fundamental an error as to require reversal without regard to prejudice, then neither is a district court's reliance on a fact not found by the grand jury. Respondents quote (Br. 31) the Court's observation that, when a State has chosen to use the grand jury to initiate criminal prosecutions, the grand jury is "a central component of the criminal justice process." Campbell v. Louisiana, 523 U.S. 392, 398 (1998). But such observations provide no basis for placing the grand jury on a higher plane than the petit jury, which is similarly central to the criminal justice process, as the Court reaffirmed in Apprendi.

Respondents also contend (Br. 32) that, even in a State that proceeds by information, an error such as the one here would "violate[] due process and require[] reversal." The case on which respondents rely-Cole v. Arkansas, 333 U.S. 196 (1948)-is quite different from this case. In Cole, the information appeared to charge a violation of Section 2 of a state statute, and the defendants were tried and convicted under Section 2. See id. at 198-200. But the state supreme court affirmed the convictions as if the information had charged a violation of Section 1, which stated a "substantially different" offense. Id. at 202; see id. at 200. In reversing the convictions, the Court explained that, even assuming that the state supreme court correctly understood the information to charge a violation of Section 1, the defendants had been deprived of their due process right to "notice of the specific charge" and "a chance to be heard in a trial of the issues raised by that charge." Id. at 201. Cole thus involved an error that was recognized to have been prejudicial to the defendants. Cole does not speak to a case in which, although the trial court relied on a fact not alleged in the indictment or information, the defendant had notice that the fact was at issue and a meaningful opportunity to contest it. Consequently, Cole does not support a rule of automatic reversal for errors of the sort here.9

5. Respondents further maintain (Br. 35) that this case involves an additional structural error of "having a sentencing judge rather than a petit jury decide an element of the offense under the preponderance of the evidence standard." But that is precisely the sort of error that Neder subjects to harmless-error analysis (and Johnson to plain-error analysis). As explained above (at 9), although respondents seize on the fact that the district court in Neder found the materiality element under the reasonable doubt standard, that fact was not essential to the holding in the case. The courts of appeals, applying Neder, have held that a failure to submit a sentence-enhancing fact, such as drug quantity, to the petit jury is subject to harmless-error analysis, without regard to the standard of proof under which the district court found that fact. See, e.g., United States v. Vazquez, 271 F.3d 93, 103-105 (3d Cir. 2001) (en banc) (relying on Neder to hold that failure to submit drug quantity to the jury "is not a structural defect" and does not "affect substantial rights" where the defendant's "sentence would have been the same had the jury made the drug quantity finding"), petition for cert. pending, No. 01-9014 (filed Mar. 8, 2002).10


Respondents do not dispute that there was overwhelming and uncontrovertible evidence that their drug conspiracy offense involved at least 50 grams of cocaine base, the threshold quantity necessary to trigger a sentence of up to life imprisonment under 21 U.S.C. 841(b)(1)(A)(iii). Nor do respondents offer any plausible argument that they lacked notice that drug quantity, whether or not charged in the indictment, was a fact that would affect their statutory maximum sentence. Yet, respondents contend that, even under the harmless-error or plain-error standard, their sentences warrant reversal. That contention is incorrect.

1. Respondents argue that the error in this case necessarily "affects substantial rights," and thus required reversal under the harmless-error standard, because the error, if corrected, would result in a reduction in their sentences. Br. 38-39; see Br. 43-44 (similar argument with respect to plain-error standard). But it is almost invariably the case that, if an error in a criminal proceeding is found harmless, the defendant will remain subject to a conviction or sentence, or both, to which he would not be subject if the judgment were reversed. The relevant question is whether, if the defendant had been accorded the procedural protections that were erroneously denied, the outcome would have been the same. If so, the error is harmless.

Here, if the prosecutor, the grand jury, and the district court had not been operating on the erroneous (but prevailing) understanding that drug quantity did not have to be alleged in the indictment to permit an enhanced sentence, respondents would have received the same sentences that they actually received. U.S. Br. 44-48. It is clear, beyond a reasonable doubt, that the grand jury, if asked, would have found probable cause that respondents' offense involved the requisite drug quantity, and that respondents were on notice that drug quantity could affect their sentence.11

2. Respondents assert (Br. 39-41) that they lacked notice that the government would seek an enhanced sentence under 21 U.S.C. 841(b) based on the quantity of drugs involved in their drug conspiracy offense. That claim is without merit. Before this Court's decision in Apprendi, the courts of appeals uniformly treated drug quantity as a sentencing factor that did not have to be alleged in the indictment or proved to the petit jury beyond a reasonable doubt. U.S. Br. 15, 47; e.g., United States v. Dorlouis, 107 F.3d 248, 252 (4th Cir.), cert. denied, 521 U.S. 1126 (1997). Accordingly, respondents, like other defendants who were indicted, convicted, and sentenced before Apprendi, could not reasonably have assumed, based on the omission of drug quantity from the superseding indictment, that the government intended to limit its proof to establishing quantities of drugs that would support a sentence within the lowest statutory maximum for cocaine and cocaine base.

In this case, moreover, the superseding indictment made clear that the government had not, as respondents assert (Br. 40), "abandoned" proof that the conspiracy involved quantities of cocaine and cocaine base that would subject them to enhanced sentences. In light of the superseding indictment's detailed description of a multi-level drug trafficking conspiracy that maintained "stash houses" and employed numerous individuals to distribute drugs around the clock for a period of almost two years and its reference to transportation and delivery of "kilogram quantities of cocaine" (see U.S. Br. 46; J.A. 60-62), respondents' suggestion that they were misled by the absence of a drug quantity allegation into believing that the government intended to forgo proof of the threshold drug quantities required for enhanced sentences is incredible. If respondents were at all surprised at being sentenced under 21 U.S.C. 841(b)(1)(A)(iii), one would expect them to have expressed that surprise to the district court. But they did not.12

3. Respondents contend (Br. 41) that the plain-error standard is inapplicable in this case because they "objected to the quantity of drugs" at sentencing. Critically, however, respondents made no objection in the district court to being sentenced based on a quantity of drugs that was not alleged in the indictment (or proved to the petit jury beyond a reasonable doubt). Instead, respondents' sentencing objections merely contested the presentence reports' calculation of the quantity of drugs attributable to them. That did not preserve the constitutional claims that they advance here. See, e.g., United States v. Candelario, 240 F.3d 1300, 1304 & n.4 (11th Cir. 2001) ("a defendant's objection to the quantity of drugs that the Government attributes to him is not, on its own, a constitutional objection"), cert. denied, 533 U.S. 922 (2001).

Moreover, respondents' sentencing objections did not challenge the relevant factual determination: that their conspiracy offense involved at least 50 grams of cocaine base, a sufficient quantity to subject them to a maximum sentence of life imprisonment under 21 U.S.C. 841(b)(1)(A)(iii). As the government explained at the petition stage (U.S. Reply Br. 4 & n.2), respondents merely disputed the presentence reports' conclusion that their base offense level under the Sentencing Guidelines was 38, the level applicable to offenses involving at least 1.5 kilograms of cocaine base, and argued that the evidence would only support offense levels of 32, 34, or 36, each of which entails responsibility for at least 50 grams of cocaine base.

Respondents also seek (Br. 41-42) to avoid the application of the plain-error standard on the ground that, because Apprendi was not decided until after their trial, they could not be accused of deliberately failing to object in the district court in the hope of obtaining reversal on appeal. In Johnson, however, the Court made clear that plain-error review applies in all cases involving a forfeited error, not just those in which the defendant's failure to object might have been a trial tactic. Like this case, Johnson involved an intervening change in the law, so that "the law at the time of trial was settled and clearly contrary to the law at the time of appeal," 520 U.S. at 468, but the Court nonetheless applied plain-error analysis.

4. Respondents assert (Br. 43-47) that, even if the plain-error standard applies in this case, the imposition of an enhanced sentence based on a fact not alleged in the indictment always "affect[s] substantial rights" or "seriously affect[s] the fairness, integrity, or public reputation of judicial proceedings," and thus requires reversal on plain-error review. That contention rests on essentially the same arguments and cases on which respondents rely in arguing that the error is "structural" or "jurisdictional." Again, however, those arguments provide no persuasive reason for treating a district court's reliance on a fact not submitted to the grand jury, the error in this case, differently from a district court's reliance on a fact not submitted to the petit jury. Neder holds that the latter error does not affect substantial rights so long as the evidence was sufficiently strong that a rational petit jury would have found the fact under the appropriate standard of proof. And Johnson holds that the error does not, in those circumstances, seriously affect the fairness, integrity, or public reputation of judicial proceedings. Similarly, here, if a reviewing court concludes, after an examination of the entire record, that a rational grand jury (and petit jury) would have found that respondents' offense involved at least 50 grams of cocaine base and that respondents had notice that drug quantity was at issue, the district court's imposition of an enhanced sentence based on drug quantity did not affect respondents' substantial rights or the fairness, integrity, or public reputation of judicial proceedings.13

* * * * *

For the foregoing reasons and those stated in our opening brief, the judgment of the court of appeals should be reversed.

Respectfully submitted.

Solicitor General

APRIL 2002

1 A number of the cases cited by respondents (e.g., Br. 13) stand only for the proposition that an indictment must allege each element of an offense. Those cases are essentially irrelevant to the question here. This case concerns not whether the district court committed an error, but instead whether such an error requires automatic reversal, even if not raised in the district court.

2 Respondents observe (Br. 18) that a defendant cannot, by waiving his right to indictment, give a federal district court "jurisdiction over any state law violation charged in an information." But that is because a true absence of subject-matter jurisdiction cannot be waived, and federal courts generally lack jurisdiction over state crimes. It does not follow that other indictment errors or omissions are jurisdictional in character.

3 Williams and Lamar hold that a defendant's claim that an indictment does not allege any offense against the United States "goes only to the merits of the case," and not to the court's jurisdiction. Lamar, 240 U.S. at 65. See U.S. Br. 39-40. Respondents suggest (Br. 16) that those cases are distinguishable because they did not "involve[] an indictment that omitted elements of a crime," but instead involved indictments that alleged facts that, according to the defendants, did not satisfy the elements of a crime. But respondents do not explain why the error in omitting an element would be "jurisdictional," while the error in failing to charge any federal offense is not.

4 Respondents' reliance (Br. 13, 17) on other cases to support a claim of "jurisdictional" error is similarly misplaced. Albrecht v. United States, 273 U.S. 1, 8 (1927), mentioned the requirement of an indictment only in dictum (and cited only Bain); the issue was whether a warrant for arrest was invalid. Post v. United States, 161 U.S. 583 (1896), concerned the effect of a law requiring a case to be prosecuted in the division of the District of Minnesota where the offense was committed; the Court's holding that a different division lacked jurisdiction is irrelevant to the issue in this case.

5 Neder disavowed the "alternative" rationale in Sullivan, on which respondents rely (Br. 37), that it would be "pure speculation" to engage in harmless-error analysis of an error that "prevents the jury from rendering a 'complete verdict' on every element of the offense." Neder, 527 U.S. at 11. Neder explained that the absence of a "complete verdict" establishes only that an error occurred, not that the error requires reversal without regard to its prejudicial impact. Id. at 12-13.

6 Respondents also assert (Br. 33) that Neder and Johnson involved "pragmatic considerations that do not obtain here," apparently that a retrial in those cases would have demanded more judicial resources than would a resentencing in this case. But nothing in Neder or cases finding structural error turns on the burden on the judicial system of correcting the error. See, e.g., Vasquez v. Hillery, 474 U.S. 254, 262 (1986) (rejecting argument that harmless-error analysis should apply in cases of racial discrimination in grand jury selection because "requiring a State to retry a defendant, sometimes years later, imposes on it an unduly harsh penalty").

7 In contrast to other constitutional allocations of power, the allocation of power between judges, prosecutors, and citizens is subject to alteration by the defendant, who may waive his right to grand jury indictment or to trial by petit jury. See U.S. Br. 29.

8 The three cases specifically identified by respondents-the cases of the Earl of Shaftesbury and Stephen Colledge in 17th century England and the case of John Peter Zenger in colonial America-have historically been viewed as examples of grand jury independence. But it is not clear that those cases are also examples of grand jury nullification, i.e., a decision not to charge even when probable cause exists. The grand juries in those cases may simply have been doing their duty by refusing to indict innocent persons, despite pressure to do so from the Crown. See Leroy D. Clark, The Grand Jury 11 (1934) (observing that the Shaftesbury and Colledge cases "appear on the surface" to be ones in which "citizens sitting on the grand jury protected political enemies of the Crown from unfounded charges brought by a hostile government," although "[i]t is possible * * * that under the law at the time there was some basis for the charges"); id. at 18 (observing that the Zenger case "seem[s] to have been [an] arbitrary and calculated attempt[] to harass and silence the political opposition," although "it is possible that [Zenger] had technically violated British law as it stood at that time").

9 Nor does Dunn v. United States, 442 U.S. 100 (1979), on which respondents also rely (Br. 32-33), support such a rule. In Dunn, the defendant was indicted, tried, and convicted for making a false statement on one date. The court of appeals affirmed the conviction based on a different false statement on a different date. This Court reversed on the ground that the defendant had been denied his due process "right to be heard on the specific charges of which he is accused." Id. at 106. A defendant is not deprived of that right when the district court relies on a fact not alleged in the indictment, provided that the defendant otherwise has notice that the fact is at issue and an opportunity to contest it.

10 Respondents, joined by their amici, contend (e.g., Br. 32) that drug quantity is an element of an offense under Section 841 and Section 846, and thus that the superseding indictment failed to state the particular Section 846 offense for which they were sentenced. It makes no difference to the analysis in this case, however, whether drug quantity is viewed as an element of an aggravated offense under Section 841 and Section 846 or as a sentencing factor subject to the constitutional constraints of Apprendi. Cf. Neder, 527 U.S. at 8-15 (holding that harmless-error analysis applies where jury instructions did not charge complete offense); Johnson, 520 U.S. at 469-470 (same with respect to plain-error analysis). And there is no doubt, contrary to amici's submission, that Section 841(b) can constitutionally be implemented in accordance with Apprendi, regardless of how it is construed. See, e.g., United States v. Buckland, 277 F.3d 1173 (9th Cir. 2002) (en banc).

11 Contrary to respondents' assumption (Br. 44), the reviewing court's harmless-error or plain-error analysis would appropriately consider the entire record, including the evidence presented at trial and sentencing. See e.g., United States v. Mechanik, 475 U.S. 66 (1986); cf. Vonn, 122 S. Ct. at 1055 (reviewing court should consider entire record, not merely record of plea proceeding, to assess effect of violation of Federal Rule of Criminal Procedure 11).

12 The government mistakenly stated that at respondents' arraignments on both indictments, they were informed that the maximum penalty for the offense was life imprisonment. Pet 2; U.S. Br. 2 (citing Gov't C.A. Br. 45). The arraignments were not transcribed, and the government has recently listened to the tapes of the arraignments, which do not reflect that advice about penalties was provided. The tapes of the detention hearings for several respondents, however, show that they were informed after the superseding indictment that conviction would subject them to a maximum of life imprisonment or other enhanced penalties authorized by 21 U.S.C. 841(b)(1)(A). See L. Thomas Detention Hearing (6/8/98) (defense counsel states that he understands that Thomas is subject to maximum sentence of life imprisonment and mandatory minimum sentence of ten years, and that the case involves "allegations of multi-kilograms of crack cocaine being distributed"); L. Cotton Detention Hearing (7/16/98) (prosecutor states that Cotton was responsible for more than 1 1/2 kilograms of crack cocaine and 15 to 30 kilograms of powder cocaine, and that he is "looking at a life sentence" under the Guidelines); S. Hall, Jr., Detention Hearing (4/17/98) (prosecutor states that Stanley Hall, Jr., is likely to receive life sentence under the Guidelines because of quantity of crack cocaine involved in offense); Matilda Hall Detention Hearing (8/10/98) (prosecutor notes that Matilda Hall is facing mandatory minimum sentence of ten years).

13 Respondents urge (Br. 48-49) this Court to defer to the court of appeals' exercise of its discretion in this case. The court of appeals, however, did not merely exercise its discretion in a single case, but also prescribed a rule of automatic reversal for all cases in which a sentence was enhanced based on a fact not alleged in the indictment. See, e.g., United States v. Dinnall, 269 F.3d 418, 424 (4th Cir. 2001) (applying plain-error standard to reverse enhanced sentence imposed after guilty plea and reasoning that Cotton requires correction of such errors regardless of "quantum of evidence" of drug quantity). In an analogous context, this Court has required the exercise of discretion based on the entire record in the particular case. See Johnson, 520 U.S. at 469-470.