View PDF Version

In the Supreme Court of the United States

STATE OF WASHINGTON, PETITIONER

v.

ARTURO R. RECUENCO

ON WRIT OF CERTIORARI
TO THE SUPREME COURT OF WASHINGTON

BRIEF FOR THE UNITED STATES
AS AMICUS CURIAE SUPPORTING PETITIONER

PAUL D. CLEMENT
Solicitor General

Counsel of Record

ALICE S. FISHER
Assistant Attorney General

MICHAEL R. DREEBEN
Deputy Solicitor General

KANNON K. SHANMUGAM
Assistant to the Solicitor General
Department of Justice
Washington, D.C. 20530-0001
(202) 514-2217

QUESTION PRESENTED

Whether an error in enhancing a sentence above the otherwise-applicable statutory maximum based on a finding by the judge rather than the jury is harmless error when it is clear beyond a reasonable doubt that the jury would have made the same finding if asked.

In the Supreme Court of the United States

No. 05-83

STATE OF WASHINGTON, PETITIONER

v.

ARTURO R. RECUENCO

ON WRIT OF CERTIORARI
TO THE SUPREME COURT OF WASHINGTON

BRIEF FOR THE UNITED STATES
AS AMICUS CURIAE SUPPORTING PETITIONER

INTEREST OF THE UNITED STATES

This case presents the issue whether an error in enhancing a sentence above the otherwise-applicable statutory maximum based on a finding by the judge rather than the jury is harmless error when it is clear beyond a reasonable doubt that the jury would have made the same finding if asked. The resolution of that question will affect federal prosecutions in which a court erroneously fails to submit a statutory sentencing factor to the jury. See, e.g., 21 U.S.C. 841(b) (2000 & Supp. II 2002) (establishing graduated penalties, based on amount, for manufacture, distribution, or possession of illegal drugs). The United States therefore has a substantial interest in this case.

STATEMENT

1. On September 18, 1999, respondent was arrested at his home in King County, Washington, following a domestic dispute with his wife, Amy. At respondent's trial, Mrs. Recuenco testified that respondent had yelled at her for not preparing dinner for his sisters, smashed a glass stove top with a metal pipe, and then retrieved a gun from a filing cabinet and pointed it at her. When Mrs. Recuenco called the police, respondent violently yanked on the phone cord, ripping the jack out of the wall. After arriving at the scene, officers heard Mrs. Recuenco shouting that respondent had a gun and was going to kill her. Mrs. Recuenco subsequently showed one of the officers where the gun was located; the gun was found to be loaded. Pet. App. 10a.

As relevant here, respondent was charged with second-degree assault, in violation of Wash. Rev. Code Ann. § 9A.36.021 (West 2000 & Supp. 2005).1 Respondent was potentially subject to one of two different sentencing enhancements. If respondent was "armed with a firearm" during a Class B felony (such as second- degree assault), the trial court was required to increase his sentence by three years. Id. § 9.94A.533(3) (West 2003 & Supp. 2005). If, however, respondent was "armed with a deadly weapon other than a firearm," the trial court was required to increase his sentence by only one year. Id. § 9.94A.533(4) (emphasis added). A separate statute provided that, if "there has been a special allegation and evidence establishing that the accused * * * was armed with a deadly weapon," the jury was required to "find a special verdict as to whether or not the defendant * * * was armed" with that weapon. Id. § 9.94A.602 (West 2003). Critically, that statute, unlike the statute containing the sentencing enhancements, did not distinguish between firearms and other deadly weapons. The information contained a special allegation that respondent had been "armed with a deadly weapon, to-wit: a handgun, under the authority of" both statutes. J.A. 3.

2. Respondent was tried in January 2000, before this Court's decisions in Apprendi v. New Jersey, 530 U.S. 466 (2000), which held that any fact, other than a prior conviction, which increases the penalty for a crime beyond the statutory maximum must be submitted to the jury and proved beyond a reasonable doubt, and Blakely v. Washington, 542 U.S. 296 (2004), which made clear, in the specific context of Washington's sentencing regime, that the "statutory maximum" for purposes of Apprendi is the maximum sentence that a judge could impose based on the facts admitted by the defendant or reflected in the jury's verdict. At the conclusion of trial, the judge submitted a special-verdict form to the jury with the following question: "Was [respondent] armed with a deadly weapon at the time of the commission of the crime of Assault in the Second Degree?" J.A. 13. Neither party asked that the jury be instructed to determine more specifically whether, if respondent was armed with a deadly weapon, the weapon was a firearm; to the contrary, respondent proposed the same question that was submitted to the jury. J.A. 6.

Both parties appear to have recognized that the relevant "deadly weapon" was a gun (which plainly qualified as a "firearm" under Washington law, see Wash. Rev. Code Ann. § 9A.04.110(6) (West 2000)). At one point, the judge stated that "there is no dispute in this case that we are talking about a gun." J.A. 16. Although respondent denied that he had pointed the gun at his wife, Pet. App. 10a, counsel for respondent sought an instruction on the allegedly lesser included offense of "aiming a firearm," Pet. App. 11a; see Wash. Rev. Code Ann. § 9.41.230 (West 2003). The jury was actually instructed that "[t]he term 'deadly weapon' includes any firearm, whether loaded or not," J.A. 7, and that "[a] pistol, revolver, or any other firearm is a deadly weapon whether loaded or unloaded," J.A. 8. The jury found respondent guilty of, inter alia, second-degree assault, and it found that respondent had been armed with a deadly weapon during the commission of the assault. Pet. App. 4a.

At sentencing, respondent contended that, because the jury had not specifically found that he had been armed with a firearm, he was subject only to the one-year enhancement for being armed with a deadly weapon other than a firearm (and not the three-year enhancement for being armed with a firearm). Pet. App. 4a. Counsel for respondent acknowledged, however, that "the allegation and the basis on which this case was tried was under the theory of a firearm," J.A. 30, and conceded that the fact that respondent had been armed with a firearm had been "pleaded and argued to the jury and evidently, perhaps obviously, proven to the jury," J.A. 37. The judge rejected respondent's argument and imposed the three-year enhancement, thereby increasing respondent's sentence from three months (the base sentence) to 39 months. Pet. App. 4a.2

3. In a post-Apprendi but pre-Blakely decision, the Washington Court of Appeals affirmed. Pet. App. 9a-19a. The court reasoned that, even assuming that the jury was specifically required to find that respondent had been armed with a firearm before the firearm enhancement could validly be applied, any error was harmless. Id. at 18a. Quoting a Washington Supreme Court decision that in turn quoted this Court's decision in Neder v. United States, 527 U.S. 1 (1999), the court explained that an error was harmless when the appellate court could conclude beyond a reasonable doubt that the verdict would have been the same absent the error. Pet. App. 18a. The court noted that the information had specified that the deadly weapon that respondent had used was a handgun (and thus a firearm); that the prosecution had argued that respondent had committed the assault with a firearm; that the jury had found that respondent had committed the assault with a deadly weapon; and that no other weapon had been mentioned. Ibid.

4. In the wake of this Court's decision in Blakely, the Washington Supreme Court, considering the case together with other cases presenting Blakely issues, reversed and remanded for resentencing. Pet. App. 1a-8a. The court held that the trial court had committed a Blakely error by applying the firearm enhancement, see id. at 6a-7a, and that the error was not subject to harmless-error analysis under the court's simultaneous decision in State v. Hughes, 110 P.3d 192 (Wash. 2005). Pet. App. 8a.

In Hughes, the Washington Supreme Court held that Blakely errors could never be harmless. Pet. App. 20a-27a. In so doing, the court relied heavily on this Court's decision in Sullivan v. Louisiana, 508 U.S. 275 (1993), which held that a constitutionally defective reasonable-doubt instruction could never give rise to harmless error. Pet. App. 21a-23a. The court reasoned that, where an invalid reasonable-doubt instruction had been given, "there was nothing upon which to apply the harmless error analysis to conclude that, but for the error, the result would have been the same." Id. at 23a. That situation, the court explained, was "directly analogous" to the instant situation, because "there was no jury finding beyond a reasonable doubt of aggravating factors warranting an enhanced sentence." Ibid. "It would be illogical," the court concluded, "to perform harmless error analysis on the absence of those findings." Ibid.

The Washington Supreme Court further determined that Hughes was not analogous to Neder v. United States, 527 U.S. 1 (1999), which held that the omission of an element of the offense from the jury's instruction could constitute harmless error. Pet. App. 26a-27a. The court reasoned that, "[a]lthough Neder involved the situation where a jury did not find facts supporting every element of the crime, it still returned a guilty verdict." Id. at 27a. "Like traditional harmless error analysis cases," the court continued, "the reviewing court could ask whether but for the omission in the jury instruction, the jury would have returned the same verdict." Ibid. "Where Blakely violations are at issue, however, the jury necessarily did not return a special verdict or explicit findings on the aggravating factors supporting the exceptional sentence." Ibid. Instead, the court reasoned, "[t]he reviewing court asks whether but for the error, the jury would have made different or new findings." Ibid. Because the court found the better analogy to be to Sullivan, not Neder, it concluded that "[h]armless error analysis cannot be conducted on Blakely Sixth Amendment violations." Ibid.3

SUMMARY OF ARGUMENT

A trial court's error in enhancing a sentence above the otherwise-applicable statutory maximum based on a finding by the judge rather than the jury is harmless error when it is clear beyond a reasonable doubt that the jury would have made the same finding if asked. This Court has held that an error is intrinsically harmful, or "structural," only in a limited number of contexts in which the error infects the entire trial process or otherwise renders a trial fundamentally unfair. In a series of other cases, by contrast, this Court has held that various errors in jury instructions concerning elements of the charged offense can be harmless. Most notably, in Neder v. United States, 527 U.S. 1 (1999), the Court held that the omission of a jury instruction on an offense element does not constitute structural error.

It logically follows from Neder's holding-that the failure to obtain a jury finding on an element of an offense is not structural error-that the failure to obtain a finding on a sentence-enhancing fact, in violation of Apprendi and its progeny, is not structural error either. In both cases, the jury has not made a necessary factual finding to support the judgment. And in both cases, a reviewing court can review the evidence to determine whether it is clear beyond a reasonable doubt that the jury would have made the requisite finding if it had been asked to do so. There is no basis for distinguishing the error in this case from the error in Neder.

In Sullivan v. Louisiana, 508 U.S. 275 (1993), this Court held that an error in jury instructions is structural when it undermines all of the jury's findings, not when it merely prevents the jury from making an additional finding on a single issue. The strand of reasoning in Sullivan that harmless-error analysis cannot be conducted at all when the jury has not made a finding on a particular issue was expressly rejected in Neder, 527 U.S. at 11-13, and it cannot support a finding of structural error here. Nor can Neder be distinguished on the ground that the jury here returned a verdict only for the original offense and not the "enhanced" offense. The appropriate inquiry for purposes of harmless-error analysis is not whether the verdict would have been the same, but rather whether the overall outcome of the proceedings (here, the sentence) would have been the same absent the challenged error.

Finally, the error in this case cannot be deemed automatically harmful simply because a defendant necessarily would have received a lower sentence if the judge had not erroneously imposed a higher sentence based on a fact not found by the jury. Although an error of this type is realized only at sentencing, the essence of the constitutional violation is reliance on a finding by the judge rather than the jury on the sentence-enhancing fact. Because the Sixth Amendment issue turns on the absence of the jury finding, that omission-and its impact-is critical to whether the error caused prejudice. In this case, because the relevant fact was established by overwhelming and "uncontroverted" evidence, Neder, 527 U.S. at 18, it is clear beyond a reasonable doubt that the jury would have found that fact if it had been asked to do so, and the error is therefore harmless.

ARGUMENT

I. THE ABSENCE OF A JURY FINDING ON A FACT USED TO ENHANCE A SENTENCE BEYOND THE OTHERWISE-APPLICABLE MAXIMUM IS SUBJECT TO HARMLESS-ERROR REVIEW

In Blakely v. Washington, 542 U.S. 296 (2004), this Court made clear that it is constitutional error for a judge to impose a sentence above the otherwise-applicable maximum based on a fact, other than a prior conviction, not reflected in the jury's verdict.4 Like most constitutional error, however, a Blakely error is subject to harmless-error analysis. Such an error should be found harmless when a reviewing court concludes that it is clear beyond a reasonable doubt that the jury would have found the omitted fact if it had been asked to do so, such that the outcome would have been the same absent the error.

A. The Failure To Submit An Element Of The Offense To The Jury Is Subject To Harmless-Error Review

1. The widespread adoption of harmless-error rules in American jurisprudence occurred in the early twentieth century, arising out of concern that appellate courts were reversing criminal convictions on the basis of mere technical errors. See Kotteakos v. United States, 328 U.S. 750, 759-760 (1946). This Court later held that at least some constitutional errors, like non-constitutional errors, may be harmless and thus do not require reversal. See Chapman v. California, 386 U.S. 18, 22 (1967). It is now recognized that "most constitutional errors can be harmless." Arizona v. Fulminante, 499 U.S. 279, 306 (1991). In cases involving constitutional error, the harmless-error doctrine, which applies when the defendant made a timely objection at trial, requires an appellate court to disregard the error where it is clear beyond a reasonable doubt that the error did not affect the outcome of trial proceedings. See, e.g., Neder v. United States, 527 U.S. 1, 7 (1999).5

This Court has identified a narrow class of fundamental constitutional errors as so intrinsically harmful that they require reversal without inquiry into whether they had an effect on the outcome. Errors are intrinsically harmful, or "structural," only when they "infect the entire trial process," Brecht v. Abrahamson, 507 U.S. 619, 630 (1993); "necessarily render a trial fundamentally unfair," Rose v. Clark, 478 U.S. 570, 577 (1986); or affect "[t]he entire conduct of the trial from beginning to end" and "the framework within which the trial proceeds," Fulminante, 499 U.S. at 309, 310. In Johnson v. United States, 520 U.S. 461, 468-469 (1997), and Neder, 527 U.S. at 8, this Court listed six examples of structural error: (1) a biased trial judge, see Tumey v. Ohio, 273 U.S. 510 (1927); (2) the complete denial of counsel, see Gideon v. Wainwright, 372 U.S. 335 (1963); (3) the denial of self-representation at trial, see McKaskle v. Wiggins, 465 U.S. 168 (1984); (4) the denial of a public trial, see Waller v. Georgia, 467 U.S. 39 (1984); (5) racial discrimination in selection of a grand jury, see Vasquez v. Hillery, 474 U.S. 254 (1986); and (6) the administration of a defective reasonable-doubt instruction, see Sullivan v. Louisiana, 508 U.S. 275 (1993).

Consistent with its view that errors are structural "only in a very limited class of cases," Johnson v. United States, 520 U.S. at 468, this Court has concluded that most constitutional errors are susceptible of harmless-error analysis. See, e.g., Fulminante, 499 U.S. at 306-307 (listing examples); cf. Rose, 478 U.S. at 579 (noting that, "if the defendant had counsel and was tried by an impartial adjudicator, there is a strong presumption that any other [constitutional] errors that may have occurred are subject to harmless-error analysis"). In a number of cases, the Court has held that errors in jury instructions concerning an element of the charged offense can be harmless. See, e.g., Yates v. Evatt, 500 U.S. 391 (1991) (erroneous rebuttable presumption); Carella v. California, 491 U.S. 263 (1989) (per curiam) (erroneous conclusive presumption); Pope v. Illinois, 481 U.S. 497 (1987) (misstatement of element); Rose v. Clark, supra (erroneous rebuttable presumption). In California v. Roy, 519 U.S. 2 (1996) (per curiam), the Court held that harmless-error analysis could apply where the trial court erroneously failed to instruct the jury that it could convict the defendant on a theory of aiding and abetting only if it found that the defendant had the intent of aiding the principal's crime. Id. at 3. After noting that the error could equally be characterized as the "misdescription" of an element of the crime (i.e., the element of mens rea) or the "omission" of an element (i.e., the element of intent to aid the principal), the Court concluded that the error was subject to harmless-error analysis, under the less stringent standard applicable to constitutional claims raised in federal habeas proceedings. Id. at 5-6; see p. 10, note 5, supra.

2. In Neder, the Court held that the failure to instruct the jury on an element of the offense does not constitute structural error. 527 U.S. at 8-15. The error at issue in Neder was the failure to submit the element of materiality to the jury in a tax-fraud case. Id. at 4. The Court reasoned that "[t]he error at issue here * * * differs markedly from the constitutional violations we have found to defy harmless-error review." Id. at 8. "Unlike such defects as the complete deprivation of counsel or trial before a biased judge," the Court explained, "an instruction that omits an element of the offense does not necessarily render a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence." Id. at 9. To the contrary, the Court noted that "[the defendant] was tried before an impartial judge, under the correct standard of proof and with the assistance of counsel," and "a fairly selected, impartial jury was instructed to consider all of the evidence and argument in respect to [the defendant's] defense against the tax charges." Ibid. The Court reasoned that its holding was "dictate[d]" by its earlier decisions in cases such as Pope, Carella, and Roy, which held that errors in jury instructions concerning an element of the charged offense could be harmless. Id. at 13.

The Court further concluded that the error at issue was in fact harmless. Neder, 527 U.S. at 15-20. The Court held that such an error was harmless when it was "clear beyond a reasonable doubt that a rational jury would have found the defendant guilty absent the error." Id. at 18. That standard was met, the Court concluded, because the record evidence supporting the existence of the omitted element was "so overwhelming" that the defendant had not even contested it. Id. at 16. In reaching that conclusion, the Court rejected the argument that a reviewing court would effectively serve as a "second jury" by engaging in harmless-error analysis of the failure to instruct on an element of the offense. Id. at 19. Instead, the Court noted, a reviewing court, "in typical appellate-court fashion, asks whether the record contains evidence that could rationally lead to a contrary finding with respect to the omitted element." Ibid. "If the answer to that question is 'no,'" the Court explained, "holding the error harmless does not reflect a denigration of the constitutional rights involved." Ibid. (citation and internal quotation marks omitted). The Court likewise rejected the slippery-slope argument that treating the failure to instruct the jury on a single element as harmless error would effectively endorse directed verdicts in criminal cases. Id. at 17 n.2. The Court reasoned that "our course of constitutional adjudication has not been characterized by this 'in for a penny, in for a pound' approach." Ibid.

B. Reliance On A Sentence-Enhancing Fact That Was Not Submitted To The Jury Is Constitutionally Equivalent To Failing To Submit An Element Of The Offense To The Jury

1. The error at issue in this case (i.e., reliance on a fact found by the judge, rather than the jury, to enhance a sentence beyond the statutory maximum) is closely analogous to the error at issue in Neder (i.e., entry of a judgment of conviction after a failure to instruct the jury to find an element of the offense). The main difference-that the jury's findings here, unlike in Neder, were sufficient to support a conviction for an offense-is not material. In both settings, the trial court has failed to secure a finding by the jury, the body that is constitutionally empowered to make it, on one of the issues necessary to support the judgment. And in both settings, the error is harmless when the record reveals that the same outcome would have resulted if the issue had been submitted to the jury under proper instructions.

a. Like the error in Neder, the sentencing court's finding of a sentence-enhancing fact that should instead have been found by the jury is not a structural defect. The error does not "infect the entire trial process" or "necessarily render a trial fundamentally unfair." Neder, 527 U.S. at 8 (citations omitted). Like the defendant in Neder, respondent had counsel and was tried before an impartial tribunal. Id. at 9.6 The challenged error does not affect the jury's verdict, but instead affects only the finding that supported the enhanced sentence. The type of error at issue here thus bears no relation to the pervasive and fundamental errors that this Court has held to be structural. See pp. 10-12, supra. Instead, the error should be treated like the error at issue in Neder. This Court has suggested that a sentence-enhancing fact is the "functional equivalent" of an element of the offense. Apprendi v. New Jersey, 530 U.S. 466, 494 n.19 (2000). It follows that an error in relying on a judge-found, rather than a jury-found, sentence-enhancing fact is, for constitutional purposes, the "functional equivalent" of entering judgment on a verdict where the judge, rather than the jury, has decided an element of the offense. In Neder, this Court held that the absence of a jury finding on an offense element does not constitute structural error; the same reasoning applies to the absence of a jury finding on a sentence-enhancing fact.7

As with the error in Neder, a reviewing court can readily apply harmless-error principles in evaluating the imposition of a sentence that exceeds the otherwise-applicable maximum on the basis of a fact not found by the jury. This Court has repeatedly noted that the task of a reviewing court in applying harmless-error analysis is to review the entire record. See, e.g., Yates, 500 U.S. at 409; Delaware v. Van Arsdall, 475 U.S. 673, 681 (1986); United States v. Hasting, 461 U.S. 499, 509 n.7 (1983); cf. United States v. Vonn, 535 U.S. 55, 75-76 (2002) (noting that reviewing court should consider entire record in assessing effect of violation of Fed. R. Crim. P. 11). If the reviewing court determines that the record does not "contain[] evidence that could rationally lead to a contrary finding," Neder, 527 U.S. at 19, and concludes that it is clear beyond a reasonable doubt that the jury would have found the omitted fact if it had been asked to do so, the error is harmless, because the defendant would have received the same sentence if the right to a jury finding on the sentence-enhancing fact had been observed.

As in Neder, applying harmless-error review to the imposition of sentence without securing a jury finding on a sentence-enhancing fact appropriately balances "society's interest in punishing the guilty" against the constitutional right to a trial by jury. Neder, 527 U.S. at 18 (quoting Connecticut v. Johnson, 460 U.S. at 86 (plurality opinion)). The jury's function of "guard[ing] against a spirit of oppression and tyranny on the part of rulers" is not "fundamentally undermine[d]," id. at 19, where the jury finds all of the facts essential to the defendant's enhanced sentence, with the exception of a single fact, and where the reviewing court concludes that any rational jury would have found that fact as well. In that context, the harmless-error doctrine "serve[s] a very useful purpose insofar as [it] block[s] setting aside [the defendant's sentence] for small errors or defects that have little, if any, likelihood of having changed the result of the trial." Chapman, 386 U.S. at 22.

b. This Court's decision in United States v. Cotton, 535 U.S. 625 (2002), supports the application of harmless-error analysis to Blakely error. At issue in Cotton was the failure to charge or to secure a jury finding on drug quantity, which was used to enhance the defendants' sentences under 21 U.S.C. 846. Because the defendants in Cotton did not preserve their objection at trial, the case involved the federal plain-error doctrine, which applies when the defendant failed to make a timely objection in the district court. Fed. R. Crim. P. 52(b). Under that doctrine, a reviewing court asks whether (1) there is error; (2) the error is plain; (3) the error affects substantial rights; and (4) the error seriously affects the fairness, integrity, or public reputation of judicial proceedings. Johnson v. United States, 520 U.S. at 466-467. The third component of the plain-error inquiry-i.e., whether the error affects substantial rights-largely tracks the harmless-error inquiry applicable when a defendant did make a timely objection at trial.8

In Cotton, the Court held that the failure to charge in the indictment and prove to the jury a fact that was used to enhance the sentence beyond the statutory maximum, in violation of Apprendi, did not constitute reversible plain error. 535 U.S. at 631-634. Although the Court did not pass specifically on the question whether the third component of the plain-error inquiry was satisfied, it concluded that the fourth component was not satisfied because any error did not seriously affect the fairness, integrity, or public reputation of judicial proceedings. Id. at 632-633. In so concluding, the Court noted that the evidence concerning the sentence-enhancing fact was "overwhelming" and "essentially uncontroverted." Id. at 633.

The Court's holding in Cotton that Apprendi error does not seriously affect the fairness, integrity, or public reputation of judicial proceedings strongly suggests that Apprendi (or Blakely) error is not structural as well. Indeed, in Neder, the Court relied on its earlier decision in Johnson v. United States, which similarly held that the error at issue did not satisfy the fourth component of the plain-error inquiry, in determining that the error was not structural. Neder, 527 U.S. at 9. The Court noted that, "[a]lthough reserving the question whether the omission of an element ipso facto 'affect[s] substantial rights,' we concluded [in Johnson] that the error did not warrant correction in light of the 'overwhelming' and 'uncontroverted' evidence supporting materiality." Ibid. (citations omitted). So too here, the Court's decision in Cotton-specifically, its determination that an Apprendi error does not seriously affect the fairness of judicial proceedings where the evidence concerning the omitted fact was "overwhelming" and "essentially uncontroverted," 535 U.S. at 633-"cuts against the argument that [the error] at issue will always render [the outcome] unfair." Neder, 527 U.S. at 9.

2. The Washington Supreme Court concluded that a Blakely error was "readily distinguishable" from the type of error at issue in Neder. Pet. App. 27a. There is no valid basis, however, for that distinction.

a. In holding that a Blakely error was not subject to harmless-error analysis, the Washington Supreme Court relied heavily on this Court's decision in Sullivan v. Louisiana, supra. Pet. App. 21a-23a. In Sullivan, the Court unanimously held that a defective reasonable-doubt instruction gave rise to structural error. 508 U.S. at 281-282. The Court reasoned that such an instruction, unlike an instruction that merely "erect[ed] a presumption regarding an element of the offense," "vitiate[d] all the jury's findings" and thereby produced "consequences that are necessarily unquantifiable and indeterminate." Id. at 280, 281, 282. In a particular line of reasoning on which the Washington Supreme Court relied, however, this Court further stated that, when such an instruction is given, "there has been no jury verdict within the meaning of the Sixth Amendment"; thus, "[t]here is no object, so to speak, upon which harmless-error scrutiny can operate." Id. at 280. The Court added that "[t]he Sixth Amendment requires more than appellate speculation about a hypothetical jury's action * * *; it requires an actual jury finding of guilty." Ibid.

Standing alone, that line of reasoning from Sullivan would seemingly support the conclusion that any instructional error concerning an offense element (or a sentence-enhancing fact) is structural, insofar as the reasoning suggests that it is never appropriate for a reviewing court to determine what a jury would have found in the absence of an error. See Pet. App. 23a (stating that this Court has held that "speculating on the jury's verdict * * * is never allowed"). But such a reading of Sullivan could not be reconciled with a host of this Court's other precedents, as this Court pointed out in Neder. 527 U.S. at 11-15; see Sullivan, 508 U.S. at 282-285 (Rehnquist, C.J., concurring). While the Court in Neder endorsed the reasoning in Sullivan that an error was structural where it "vitiat[ed] all the jury's findings," Neder, 527 U.S. at 11 (quoting Sullivan, 508 U.S. at 281), the Court determined that the error at issue in Neder did not satisfy that standard, because it merely prevented the jury from making a finding on a single element of the offense. Id. at 10-11. The Court then proceeded expressly to disavow the "alternative reasoning" of Sullivan, on which the defendant in Neder had relied. Id. at 11. The Court explained that, "[a]lthough this strand of the reasoning in Sullivan does provide support for [the defendant's] position, it cannot be squared with our harmless-error cases." Ibid. Specifically, the Court cited its earlier decisions in Pope, Carella, and Roy, which held that errors in jury instructions concerning offense elements could be harmless. Id. at 11-13. The Court reasoned that "the absence of a 'complete verdict' on every element of the offense establishes no more than that an improper instruction on an element of the offense violates the Sixth Amendment's jury trial guarantee"-not the further proposition that such an error was structural. Id. at 12.

In Mitchell v. Esparza, 540 U.S. 12 (2003) (per curiam), the Court reiterated that understanding of Sullivan. In Mitchell, a defendant convicted of capital murder filed a federal habeas petition, claiming, inter alia, structural error because the indictment failed to charge him as a "principal offender." Id. at 14. The Court unanimously held that the defendant's claim lacked merit. Id. at 19. The Court rejected the court of appeals' reasoning that the failure to charge the defendant as a "principal" was the "functional equivalent" of "dispensing with the reasonable doubt requirement." Id. at 16 (citation omitted). In so doing, the Court noted that, in Neder, it had "explicitly distinguished Sullivan because the error in Sullivan * * * '"vitiat[ed] all the jury's findings,"' whereas the trial court's failure to instruct the jury on one element of an offense did not." Ibid. (citations omitted). "Where the jury was precluded from determining only one element of an offense," the Court concluded, "we held that harmless-error review is feasible." Ibid.

Under a proper understanding of Sullivan, a Blakely error-like the error in Neder-is not structural. The error in Neder did not "vitiate[] all the jury's findings," Sullivan, 508 U.S. at 281, but instead simply prevented the jury from making an additional finding on a single offense element. Similarly, the trial court's error in this case does not undermine any of the findings that the jury actually made; instead, the trial court's error was in relying on one additional fact, which the jury had not been asked to find, in imposing sentence. Moreover, the error in Neder did not produce "consequences that are necessarily unquantifiable and indeterminate," Neder, 527 U.S. at 11 (quoting Sullivan, 508 U.S. at 282), because a reviewing court can reliably determine whether the error was prejudicial by examining the record and assessing whether it is clear beyond a reasonable doubt that the jury would have found the omitted element. The trial court's error in this case can be analyzed for harmlessness in the same way as in Neder: a reviewing court can examine the record and assess whether it is clear beyond a reasonable doubt that the jury would have found the omitted fact. Under Sullivan, therefore, a Blakely error does not constitute a structural error necessitating automatic reversal.

b. The Washington Supreme Court suggested that the error in this case was distinguishable from the error in Neder because the Neder jury "still returned a guilty verdict" for the offense in question, even though the jury "did not find facts supporting every element" of that offense, and "the reviewing court could ask whether[,] but for the [error], the jury would have returned the same verdict." Pet. App. 27a. In this case, by contrast, the jury "necessarily did not return a * * * verdict" for the "enhanced" offense. Ibid. Accordingly, the Washington Supreme Court concluded, "there is no basis upon which to conduct a harmless error analysis." Ibid.; see Br. in Opp. 6 (suggesting that, in this case, the judge "necessarily is making its own determination as to proper verdict" and "[is] setting aside the jury's verdict").9

That reasoning is fundamentally flawed, because the appropriate "basis upon which to conduct a harmless error analysis" in this case is not the jury's verdict, but rather the judge's sentence. In cases involving trial errors, this Court has frequently stated that the test for harmless-error analysis is whether it is clear beyond a reasonable doubt that, but for the error, the verdict would have been the same. See, e.g., Neder, 527 U.S. at 17 (stating that, "where a reviewing court concludes beyond a reasonable doubt that * * * the jury verdict would have been the same absent the error, the erroneous instruction is properly found to be harmless"); Chapman, 386 U.S. at 24 (requiring "the beneficiary of a constitutional error to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained"). In other cases, however, the Court has framed the relevant inquiry as whether the error affected the "outcome" or "result" of proceedings in the trial court more generally. See, e.g., United States v. Olano, 507 U.S. 725, 734 (1993); United States v. Montalvo-Murillo, 495 U.S. 711, 722 (1990); Bank of Nova Scotia v. United States, 487 U.S. 250, 256 (1988); United States v. Mechanik, 475 U.S. 66, 72 (1986); Berkemer v. McCarty, 468 U.S. 420, 444 (1984). That more generic formulation is correct, because this Court has applied harmless-error analysis (or has analogously applied the third component of plain-error analysis) not only to pure trial errors (with respect to which it is natural to ask whether the verdict would have been the same), see, e.g., Neder, 527 U.S. at 17; Yates, 500 U.S. at 405; Chapman, 386 U.S. at 25-26; Kotteakos, 328 U.S. at 764-765; but also to errors in grand jury proceedings, see Bank of Nova Scotia, 487 U.S. at 263; errors at detention hearings, see Montalvo-Murillo, 495 U.S. at 722; errors in plea proceedings, see United States v. Dominguez Benitez, 542 U.S. 74, 83 (2004); and, most critical for present purposes, sentencing errors, see, e.g., Richmond v. Lewis, 506 U.S. 40, 49 (1992); Williams v. United States, 503 U.S. 193, 203 (1992); Parker v. Dugger, 498 U.S. 308, 319 (1991). The correct inquiry here, therefore, is whether, if the judge had secured a finding from the jury rather than relying on a fact found by the court, the sentence that respondent received- and thus the outcome of proceedings below-would have been the same.

By focusing on whether "the jury would have returned the same verdict" but for the error, Pet. App. 27a, the Washington Supreme Court's rule would lead to different results in functionally indistinguishable cases. Suppose, for example, that the jury in this case had been permitted to return a verdict finding an "enhanced offense" under Wash. Rev. Code Ann. § 9.94A.533(3) (West 2003 & Supp. 2005) (which contains the firearms enhancement), but had not been instructed specifically to determine whether respondent was armed with a firearm. If the jury had returned a verdict finding the defendant guilty of the "enhanced offense" and the judge had imposed the enhanced sentence, the Blakely error (in failing to secure a necessary factual finding from the jury) would presumably be subject to harmless-error analysis under the Washington Supreme Court's reasoning. That would be the case simply because the jury returned a complete verdict on the "enhanced" offense, rather than, as here, a verdict only on the "unenhanced" offense of second-degree assault simpliciter. There is no basis for drawing such an arbitrary and formalistic distinction between those two cases. In each case, the reviewing court should instead consider whether the sentence, although imposed without the requisite jury finding, would have been the same absent the challenged error.10

c. Some defendants challenging Blakely (or Apprendi) errors have contended that, even if the appropriate analysis is whether the sentence would have been the same absent the challenged error, such an error is automatically not harmless. See, e.g., United States v. Vazquez, 271 F.3d 93, 101 (3d Cir. 2001) (en banc) (rejecting that claim), cert. denied, 536 U.S. 963 (2002). Those defendants contend that the error in question occurs only at the time of sentencing, when a judge imposes a sentence above the otherwise-applicable maximum based on a fact not found by the jury. If that error had not occurred at that juncture, the reasoning goes, the sentence necessarily would not have been the same, because a defendant would not have received an enhanced sentence at all.

That argument, however, fundamentally misapprehends the nature of a Blakely error. A Blakely error does constitute "sentencing error" insofar as it is fully realized only at sentencing, when a judge imposes a sentence above the otherwise-applicable maximum based on a fact not found by the jury. A Blakely error, however, critically depends on the failure to secure a jury finding on the sentence-enhancing fact.11 The "trial" and "sentencing" components of a Blakely error are inextricably intertwined. "On the one hand, the trial error exists only because of the sentencing error[;] [o]n the other hand, the sentencing error cannot occur without the trial error." Vazquez, 271 F.3d at 101.

It would be wholly artificial to focus solely on the "sentencing" component of a Blakely error, rather than on the entirety of the error, in determining whether the error is harmless-particularly because the Sixth Amendment issue arises only because of the absence of a jury finding to support the sentence.12 When a Blakely error is committed, the core of the constitutional violation is the failure to submit to the jury a sentence-enhancing fact that the jury should have found on proof beyond a reasonable doubt. And when it is clear beyond a reasonable doubt that the jury, if asked, would have found the sentence-enhancing fact, and thus that the defendant would have received the same sentence after being found guilty of the "enhanced" offense, any error in failing to submit that fact to the jury (and imposing an enhanced sentence) was harmless. Just as a Blakely error is not structural, therefore, so too is it not automatically harmful.

II. THE ABSENCE OF A JURY FINDING ON THE FACT THAT RESPONDENT HAD BEEN ARMED WITH A FIREARM WAS HARMLESS

Should the Court conclude that Blakely errors are subject to harmless-error analysis, it would be appropriate for this Court to determine that the error at issue in this case was harmless as a matter of first instance. This Court has noted that, "[a]lthough our usual practice in cases like this is to reverse and remand for a new determination under the correct standard, we have the authority to make our own assessment of the harmlessness of a constitutional error in the first instance." Yates, 500 U.S. at 407. In Neder, the Court determined that the failure to instruct the jury to find the element of materiality was harmless, noting that "[t]he evidence supporting materiality was so overwhelming * * * that [defendant] did not argue to the jury-and does not argue here-that his false statements of income could be found immaterial." 527 U.S. at 16.

In this case, because there was "overwhelming" and "essentially uncontroverted" evidence (Cotton, 535 U.S. at 633) that respondent had been armed with a firearm during the assault for which he was convicted, it is clear beyond a reasonable doubt that the jury would have found that fact if it had been asked to do so. The information in this case specifically alleged that the deadly weapon that respondent had used was a handgun (and thus a firearm). J.A. 3. Although neither party asked that the jury be instructed to determine whether, if respondent was armed with a deadly weapon, the weapon was a firearm, both parties appear to have operated on the assumption that the deadly weapon at issue was a gun. Counsel for respondent sought an instruction on the allegedly lesser included offense of "aiming a firearm," Pet. App. 11a. While the jury was not instructed to determine whether the deadly weapon at issue was a firearm, it was instructed that "[a] pistol, revolver, or any other firearm is a deadly weapon whether loaded or unloaded." J.A. 8. Finally, counsel for respondent acknowledged that "the allegation and the basis on which this case was tried was under the theory of a firearm," J.A. 30, and conceded that the fact that respondent had been armed with a firearm had been "pleaded and argued to the jury and evidently, perhaps obviously, proven to the jury," J.A. 37. Because respondent did not "contest[] the omitted [fact]" or "raise[] evidence sufficient to support a contrary finding," Neder, 527 U.S. at 19, any error in imposing an enhanced sentence without securing a jury finding on that fact was harmless.

CONCLUSION

The judgment of the Washington Supreme Court should be reversed.

Respectfully submitted.

<

PAUL D. CLEMENT
Solicitor General

ALICE S. FISHER
Assistant Attorney General

MICHAEL R. DREEBEN
Deputy Solicitor General

KANNON K. SHANMUGAM
Assistant to the Solicitor General

DECEMBER 2005

1 Some of the statutory provisions involved in this case have been renumbered, without material revision, since respondent was charged. For convenience, all citations refer to the current versions of the relevant provisions.

2 Although petitioner was also convicted on other counts, the sentence on those counts was suspended. Pet. App. 4a.

3 The Washington Supreme Court appears to have applied federal law in holding that Blakely errors could never be harmless. At a minimum, federal law establishes a floor for determining whether particular federal constitutional errors can be subject to harmless-error review, see Connecticut v. Johnson, 460 U.S. 73, 81 n.9 (1983) (plurality opinion); Chapman v. California, 386 U.S. 18, 21 (1967), and for determining when those errors are in fact harmless, see id. at 24 (applying "beyond a reasonable doubt" standard); cf. State v. Linehan, 56 P.3d 542, 549 (Wash. 2002) (applying same standard), cert. denied, 538 U.S. 945 (2003).

4 In Apprendi v. New Jersey, 530 U.S. 466 (2000), this Court first held that any fact, other than a prior conviction, which increases the penalty for a crime beyond the statutory maximum must be submitted to the jury and proved beyond a reasonable doubt. Blakely involved the "appl[ication]" of the general rule expressed in Apprendi to the specific context of statutory sentencing enhancements, see 542 U.S. at 301, and, for present purposes, Apprendi and Blakely errors are materially indistinguishable.

5 Where the error at issue is not of constitutional dimension, or where the error is of constitutional dimension but is being reviewed in the context of a federal habeas proceeding, the reviewing court is instead required to determine whether the error had a "substantial and injurious effect or influence in determining the jury's verdict." Kotteakos, 328 U.S. at 776; see Brecht v. Abrahamson, 507 U.S. 619, 638 (1993) (applying Kotteakos standard to federal habeas review); cf. Fed. R. Crim. P. 52(a) (stating, without distinguishing between constitutional and non-constitutional error, that "[a]ny error * * * that does not affect substantial rights must be disregarded"). In the federal system, when the defendant has failed to preserve the claim of error, the defendant is required to carry the burden of establishing plain error. United States v. Cotton, 535 U.S. 625, 631-634 (2002); Johnson v. United States, 520 U.S. 461, 466 (1997); United States v. Olano, 507 U.S. 725, 734-735 (1993).

6 Here, of course, unlike in Neder, the relevant statute permited the judge to find the relevant fact. After Blakely, it is clear that the Constitution renders such judicial factfinding insufficient. But the fact that the factfinding proceeded as permitted by statute certainly cannot render the error here any more structural than the error in Neder.

7 As this Court noted in Neder, 527 U.S. at 17 n.2, the degree of constitutional error is relevant in determining whether an error can be harmless. Compare United States v. Cronic, 466 U.S. 648, 659 (1984) (holding that complete denial of counsel mandates reversal in all cases), with Strickland v. Washington, 466 U.S. 668, 695 (1984) (holding that ineffective assistance of counsel requires reversal only upon showing of "reasonable probability that, absent [counsel's] errors, the factfinder would have had a reasonable doubt respecting guilt"). But this case, like Neder, involves the absence of a jury verdict on one of the facts necessary to support the judgment; it does not involve a total deprivation of a jury verdict. Indeed, in this case, like a hypothetical case identical to Neder except that the omission of the element would result in a verdict that would support a conviction for a lesser included offense, the verdict covers every element of a criminal offense.

8 The primary difference between the harmless-error inquiry and the third component of the plain-error inquiry is that "the tables are turned on demonstrating the substantiality of any effect on a defendant's rights." Vonn, 535 U.S. at 62-63. In the plain-error inquiry, the defendant bears the burden of proving that the error does affect his substantial rights. Olano, 507 U.S. at 734-735; see United States v. Dominguez Benitez, 542 U.S. 74, 83 (2004).

9 In Neder, the defendant contended that there was no valid verdict because the jury had failed to make findings on all of the elements of the relevant offense. 527 U.S. at 11. The Court, however, rejected that contention on the ground that it had previously applied harmless-error analysis in similar circumstances. Id. at 11-13.

10 At one point in its analysis, the Washington Supreme Court suggested that it would be inappropriate for a reviewing court engaging in harmless-error analysis to ask whether, but for the error, "the jury would have made different or new findings," because such a court would be "speculat[ing] on what [the jury] would have done if [it] had been asked to find different facts." Pet. App. 27a. In Neder, however, the Court endorsed precisely such an inquiry: namely, whether the jury would have found an element of the offense if an instruction concerning that element had been given. See 527 U.S. at 19.

11 Lower courts have consistently held that an objection to a Blakely (or Apprendi) error is timely if it is made at sentencing. See, e.g., United States v. Stewart, 306 F.3d 295, 309-313 (6th Cir. 2002), cert. denied, 537 U.S. 1138, 537 U.S. 1146, and 538 U.S. 1036 (2003); United States v. Candelario, 240 F.3d 1300, 1304-1305 (11th Cir.), cert. denied, 533 U.S. 922 (2001); United States v. Garcia-Guizar, 234 F.3d 483, 488 (9th Cir. 2000), cert. denied, 532 U.S. 984 (2001); United States v. Doggett, 230 F.3d 160, 165 (5th Cir. 2000), cert. denied, 531 U.S. 1177 (2001). Those courts, however, have also held that a defendant could preserve his objection to a Blakely error by making it when the jury is being instructed-although a defendant may have no incentive to do so under a regime in which the defendant need not object until sentencing. See, e.g., Candelario, 240 F.3d at 1305 (noting that "a defendant's constitutional objection will be timely if made at any time prior to sentencing").

12 That artificiality is underscored by comparing a Blakely error to the omission of an element that marks the boundary between greater and lesser included offenses. In both cases, the consequences of the omission may not be felt until sentencing, but Neder's analysis is fully applicable to a case in which the omitted element makes a difference between a greater and lesser included offense. See, e.g., United States v. Riley, 250 F.3d 1303, 1307 (11th Cir. 2001) (per curiam) (applying harmless-error analysis to error under Castillo v. United States, 530 U.S. 120 (2000), in failing to obtain jury finding on firearm-type element of a greater offense under 18 U.S.C. 924(c)); United States v. Shea, 211 F.3d 658, 672 (1st Cir. 2000) (applying harmless-error analysis to error under Jones v. United States, 526 U.S. 227 (1999), in failing to obtain jury finding on bodily-injury element of a greater offense under the carjacking statute, 18 U.S.C. 2119), cert. denied, 531 U.S. 1154 (2001). It likewise should be fully applicable here.