| Syllabus
|
Opinion
[Ginsburg] |
Concurrence
[Breyer] |
Dissent
[Alito] |
|---|---|---|---|
| HTML version
PDF version | HTML version
PDF version | HTML version
PDF version | HTML version
PDF version |
MICHAEL GREENLAW, aka MIKEY, PETITIONER v. UNITED STATES
on writ of certiorari to the united states court of appeals for the eighth circuit
Justice Alito, with whom Justice Stevens joins, and with whom Justice Breyer joins as to Parts I, II, and III, dissenting.
I respectfully dissent because I view the cross-appeal requirement as a rule of appellate practice. It is akin to the rule that courts invoke when they decline to consider arguments that the parties have not raised. Both rules rest on premises about the efficient use of judicial resources and the proper role of the tribunal in an adversary system. Both are sound and should generally be followed. But just as the courts have made them, the courts may make exceptions to them, and I do not understand why a reviewing court should enjoy less discretion to correct an error sua sponte than it enjoys to raise and address an argument sua sponte. Absent congressional direction to the contrary, and subject to our limited oversight as a supervisory court, we should entrust the decision to initiate error correction to the sound discretion of the courts of appeals.
I
Before laying out my view in more detail, I must first address the question whether federal courts have subject-matter jurisdiction to enlarge an appellee’s judgment in the absence of a cross-appeal. Because the Court would not recognize any exceptions to the cross-appeal requirement when the defendant appeals his sentence, it does not decide that question. See ante, at 7. I must confront it, though I do not regard it as a substantial question. The cross-appeal requirement seems to me a prime example of a “ ‘rule of practice,’ subject to exceptions, not an unqualified limit on the power of appellate courts.” El Paso Natural Gas Co. v. Neztsosie, 526 U. S. 473, 480 (1999) . While a court should generally enforce the cross-appeal requirement, a departure from it would not divest the court of jurisdiction.
This Court has never addressed whether an appellate court’s jurisdiction to enlarge a judgment in favor of an appellee is contingent on a duly filed cross-appeal. The majority’s contention that “[o]ur own opinions contain statements supporting” the “ ‘jurisdictional’ ” characterization of the requirement, ante, at 6, relies on a misreading of that precedent. The Court may have previously characterized the cross-appeal requirement as limiting “ ‘[t]he power of an appellate court to modify a decree,’ ” ibid. (quoting Morley Constr. Co. v. Maryland Casualty Co., 300 U. S. 185, 187 (1937) ), but it does not follow that jurisdiction is conditioned on a properly filed cross-appeal. A court may lack the power to do something for reasons other than want of jurisdiction, and a rule can be inflexible without being jurisdictional. See Eberhart v. United States, 546 U. S. 12, 19 (2005) (per curiam).
The jurisdiction of the courts of appeals is fixed by Congress. See Bowles v. Russell, 551 U. S. ___, ___ (2007) (slip op., at 6); Ankenbrandt v. Richards, 504 U. S. 689, 698 (1992) (“ ‘[T]he judicial power of the United States … is (except in enumerated instances, applicable exclusively to this Court) dependent for its distribution and organization, and for the modes of its exercise, entirely upon the action of Congress’ ” (quoting Cary v. Curtis, 3How. 236, 245 (1845))). If Congress wants to withhold from the courts of appeals the power to decide questions that expand the rights of nonappealing parties, it may do so. See U. S. Const., Art. III, §1 (authorizing Congress to establish the lower courts and, by corollary, to fix their jurisdiction); Kontrick v. Ryan, 540 U. S. 443, 452 (2004) (“Only Congress may determine a lower federal court’s subject-matter jurisdiction”). The jurisdictional question thus reduces to whether Congress intended to make a cross-appeal a condition precedent to the appellate court’s jurisdiction to enlarge a judgment in favor of a nonappealing party.
As always with such questions, the text of the relevant statute provides the best evidence of congressional intent. The relevant statute in this case is 18 U. S. C. §3742 (2000 ed. and Supp. V). Section 3742(a) authorizes a criminal defendant to “file a notice of appeal” to review a sentence that was, among other possibilities, “imposed in violation of law.” E.g., §3742(a)(1). Section 3742(b) provides parallel authority for the Government to “file a notice of appeal” to review unlawful sentences. E.g., §3742(b)(1). The statute conditions the Government’s authority to further prosecute its appeal on “the personal approval of the Attorney General, the Solicitor General, or a deputy solicitor general designated by the Solicitor General.” §3742(b).
Nothing in this language remotely suggests that a court of appeals lacks subject-matter jurisdiction to increase a defendant’s sentence in the absence of a cross-appeal by the Government. In fact, the statute does not even mention cross-appeals. It separately authorizes either party to “file a notice of appeal,” but it never suggests that the reviewing court’s power is limited to correcting errors for the benefit of the appealing party. If anything, it suggests the opposite. Without qualifying the appellate court’s power in any way, §3742(e) instructs the court to determine, among other things, whether the sentence was “imposed in violation of law.” §3742(e)(1). And while §3742(f)(2) limits the action that a court of appeals can take depending on which party filed the appeal, compare §3742(f)(2)(A) (sentences set aside as “too high” if defendant filed) with §3742(f)(2)(B) (sentences set aside as “too low” if Government filed), no such limitation appears in §3742(f)(1). That paragraph requires a court of appeals simply to set aside any sentence “imposed in violation of law or imposed as a result of an incorrect application of the sentencing guidelines.”
II
Since a cross-appeal has no effect on the appellate court’s subject-matter jurisdiction, the cross-appeal requirement is best characterized as a rule of practice. It is a rule created by the courts to serve interests that are important to the Judiciary. The Court identifies two of these interests: notice to litigants and finality. Ante, at 14; see also Neztsosie, supra, at 480. One might add that the cross-appeal requirement also serves a third interest: the appellate court’s interest in being adequately briefed on the issues that it decides. See Fed. Rule App. Proc. 28.1(c) and Advisory Committee’s Notes, 28 U. S. C. App., pp. 615–616. Although these are substantial interests in the abstract, I question how well an inflexible cross-appeal requirement serves them.
Notice. With respect to notice, the benefits of an unyielding cross-appeal requirement are insubstantial. When the Government files a notice of cross-appeal, the defendant is alerted to the possibility that his or her sentence may be increased as a result of the appellate decision. But if the cross-appeal rule is, as I would hold, a strong rule of practice that should be followed in all but exceptional instances, the Government’s failure to file a notice of cross-appeal would mean in the vast majority of cases that the defendant thereafter ran little risk of an increased sentence. And the rare cases where that possibility arose would generally involve errors so plain that no conceivable response by the defendant could alter the result. It is not unreasonable to consider an appealing party to be on notice as to such serious errors of law in his favor. And while there may be rare cases in which the existence of such a legal error would come as a complete surprise to the defendant or in which argument from the parties would be of assistance to the court, the solution to such a problem is not to eliminate the courts of appeals’ authority to correct egregious errors. Rather, the appropriate response is for the court of appeals to request supplemental briefing or—if it deems that insufficient—simply to refuse to exercise its authority. Cf. Irizarry v. United States, 553 U. S. ___, ___ (2008) (slip op., at 9). In short, the Court’s holding does not increase the substance of the notice that a defendant receives; it merely accelerates that notice by at most a few weeks in a very small number of cases.
The Court contends that “[g]iven early warning, [the defendant] can tailor his arguments to take account of [the risk of a higher sentence] … [o]r he can seek the Government’s agreement to voluntary dismissal of the competing appeals.” Ante, at 14 (citing Fed. Rule App. Proc. 42(b)). But the Court does not explain how a notice of cross-appeal, a boilerplate document, helps the defendant “tailor his arguments.” Whether the cross-appeal rule is ironclad, as the Court believes, or simply a strong rule of practice, a defendant who wishes to appeal his or her sentence is always free to seek the Government’s commitment not to cross-appeal or to terminate a cross-appeal that the Government has already taken. Fed. Rule App. Proc. 42(b).
Finality. An inflexible cross-appeal rule also does little to further the interest of the parties and the Judiciary in the finality of decisions. An appellate court’s decision to grant a nonappealing party additional relief does not interrupt a long, undisturbed slumber. The error’s repose begins no earlier than the deadline for filing a cross-appeal, and it ends as soon as the reviewing court issues its opinion—and often much sooner. Here, for example, the slumber was broken when the Government identified the error in its brief as appellee. See Brief for United States 5.
Orderly Briefing. I do not doubt that adversarial briefing improves the quality of appellate decisionmaking, but it hardly follows that appellate courts should be denied the authority to correct errors that seriously prejudice nonappealing parties. Under my interpretation of the cross-appeal rule, a court of appeals would not be obligated to address errors that are prejudicial to a nonappealing party; a court of appeals would merely have the authority to do so in appropriate cases. If a court of appeals noticed such an error and concluded that it was appropriate to address the issue, the court could, if it wished, order additional briefing. If, on the other hand, the court concluded that the issue was not adequately addressed by the briefs filed by the parties in the ordinary course and that additional briefing would interfere with the efficient administration of the court’s work, the court would not be required to decide the issue. Therefore, I do not see how the courts of appeals’ interest in orderly briefing is furthered by denying those courts the discretionary authority to address important issues that they find it appropriate to decide.
Indeed, the inflexible cross-appeal rule that the Court adopts may disserve the interest in judicial efficiency in some cases. For example, correcting an error that prejudiced a nonappealing defendant on direct review might obviate the need for a collateral attack. Cf. Granberry v. Greer, 481 U. S. 129, 134 (1987) (allowing the Court of Appeals to address the merits of an unexhausted habeas corpus petition if “the interests of comity and federalism will be better served by addressing the merits forthwith [than] by requiring a series of additional state and district court proceedings before reviewing the merits of the petitioner’s claim”); Munaf v. Geren, 553 U. S. ___, ___ (2008) (slip op., at 13) (recognizing “occasions … when it is appropriate to proceed further and address the merits” of a habeas corpus petition rather than reverse and remand on threshold matters). Because the reviewing court is in the best position to decide whether a departure from the cross-appeal rule would be efficient, rigid enforcement of that rule is more likely to waste judicial resources than to conserve them.
In sum, the Court exaggerates the interests served by the cross-appeal requirement. At the same time, it overlooks an important interest that the rule disserves: the interest of the Judiciary and the public in correcting grossly prejudicial errors of law that undermine confidence in our legal system. We have repeatedly stressed the importance of that interest, see, e.g., United States v. Olano, 507 U. S. 725, 736–737 (1993) ; Press-Enterprise Co. v. Superior Court of Cal., Riverside Cty., 464 U. S. 501, 507 (1984) ; New York Central R. Co. v. Johnson, 279 U. S. 310, 318 (1929) , and it has justified departures from our traditional adversary framework in other contexts. The Court mentions one of those contexts, see ante, at 5 (pro se litigation), but there are others that deserve mention.
The most well-known is plain-error review. Federal Rule of Criminal Procedure 52(b) authorizes reviewing courts to correct “[a] plain error that affects substantial rights … even though it was not brought to the court’s attention.” Although I agree with the Court that this Rule does not independently justify the Eighth Circuit’s decision, see ante, at 8–9, I believe that the Rule’s underlying policy sheds some light on the issue before us. We have explained that courts may rely on Rule 52(b) to correct only those plain errors that “ ‘seriously affec[t] the fairness, integrity or public reputation of judicial proceedings.’ ” Olano, supra, at 736 (quoting United States v. Atkinson, 297 U. S. 157, 160 (1936) ). We have thus recognized that preservation of the “fairness, integrity or public reputation of judicial proceedings” may sometimes justify a departure from the traditional adversarial framework of issuepresentation.
Perhaps the closest analogue to the cross-appeal requirement is the rule of appellate practice that restrains reviewing courts from addressing arguments that the parties have not made. Courts typically invoke this rule to avoid resolving a case based on an unaired argument, even if the argument could change the outcome. See, e.g., Santiago v. Rumsfeld, 425 F. 3d 549, 552, n. 1 (CA9 2005); United States v. Cervini, 379 F. 3d 987, 994, n. 5 (CA10 2004). But courts also recognize that the rule is not inflexible, see, e.g., Santiago, supra, at 552, n. 1, and sometimes they depart from it, see, e.g., United States Nat. Bank of Ore. v. Independent Ins. Agents of America, Inc., 508 U. S. 439, 448 (1993) (“After giving the parties ample opportunity to address the issue, the Court of Appeals acted without any impropriety in refusing to accept what in effect was a stipulation on a question of law” (citing Swift & Co. v. Hocking Valley R. Co., 243 U. S. 281, 289 (1917) )); United States v. Moyer, 282 F. 3d 1311, 1317–1318 (CA10 2002); Dorris v. Absher, 179 F. 3d 420, 425–426 (CA6 1999).
A reviewing court will generally address an argument sua sponte only to correct the most patent and serious errors. See, e.g., id., at 426 (concluding that the error, if overlooked, would result in “a miscarriage of justice”); Consumers Union of U. S., Inc. v. Federal Power Comm’n, 510 F. 2d 656, 662 (CADC 1974) (balancing “considerations of judicial orderliness and efficiency against the need for the greatest possible accuracy in judicial decisionmaking”). Because the prejudicial effect of the error and the impact of error correction on judicial resources are matters best determined by the reviewing court, the court’s decision to go beyond the arguments made by the parties is committed to its sound discretion. See United States Nat. Bank of Ore., supra, at 448 (reviewing an appellate court’s decision to address an argument sua sponte for abuse of discretion).
This authority provides a good model for our decision in this case. The Court has not persuaded me that the interests at stake when a reviewing court awards a nonappealing party additional relief are qualitatively different from the interests at stake when a reviewing court raises an issue sua sponte. Authority on the latter point recognizes that the interest of the public and the Judiciary in correcting grossly prejudicial errors of law may sometimes outweigh other interests normally furthered by fidelity to our adversarial tradition. I would recognize the same possibility here. And just as reviewing courts enjoy discretion to decide for themselves when to raise and decide arguments sua sponte, I would grant them substantial latitude to decide when to enlarge an appellee’s judgment in the absence of a cross-appeal.1
III
The approach I advocate is not out of step with our precedent. The Court has never decided whether the cross-appeal requirement is “subject to exceptions [or] an unqualified limit on the power of appellate courts.” Neztsosie, 526 U. S., at 480. That question was reserved in Neztsosie, ibid., even as the Court recognized that lower courts had reached different conclusions, see id., at 480, n. 2. I would simply confirm what our precedent had assumed: that there are exceptional circumstances when it is appropriate for a reviewing court to correct an error for the benefit of a party that has not cross-appealed the decision below.
Indeed, the Court has already reached the very result that it claims to disavow today. We have long held that a sentencing court confronted with new circumstances may impose a stiffer sentence on remand than the defendant received prior to a successful appeal. See Chaffin v. Stynchcombe, 412 U. S. 17, 23 (1973) ; North Carolina v. Pearce, 395 U. S. 711, 719–720 (1969) , overruled on other grounds, Alabama v. Smith, 490 U. S. 794 (1989) . The Court makes no effort to explain the analytical difference between those cases and this one. If a sentencing court may rely on new circumstances to justify a longer sentence on remand, why cannot one of the new circumstances be the court’s discovery (by dint of appellate review) that its first sentence was based on an error of law?2
Even today, the Court refuses to decide whether the cross-appeal requirement admits of exceptions in appropriate cases. While calling the rule “ ‘inveterate and certain,’ ” ante, at 6 (quoting Morley Constr. Co., 300 U. S., at 191), the Court allows that “there might be circumstances in which it would be proper for an appellate court to initiate plain-error review,” ante, at 9; see also ante, at 5, n. 2. The Court’s mandate is limited to a single class of cases—sentencing appeals, and then only when the appeal is brought by the Government.
The Court justifies the asymmetry in its decision by pointing to 18 U. S. C. §3742(b), which provides that “[t]he Government may not further prosecute [the] appeal without the personal approval of the Attorney General, the Solicitor General, or a deputy solicitor general designated by the Solicitor General.” According to the majority, “[i]t would severely undermine Congress’ instruction were appellate judges to ‘sally forth’ on their own motion to take up errors adverse to the Government when the designated Department of Justice officials have not authorized an appeal from the sentence the trial court imposed.” Ante, at 7 (citation omitted).
The problem with this argument is that §3742(b) does not apportion authority over sentencing appeals between the Executive and Judicial Branches. By its terms, §3742(b) simply apportions that authority within an executive department. It provides that the “[t]he Government” may not “prosecute” the appeal without approval from one of the listed officials. It says nothing about the power of the courts to correct error in the absence of a Government appeal. Had Congress intended to restrict the power of the courts, the statute would not stop “[t]he Government” from “prosecut[ing]” unauthorized appeals; instead, it would stop “the Court of Appeals” from “deciding” them.
The design that the Court imputes to the drafters of §3742(b) is inconsistent with the text in another important respect. Suppose that the District Court imposes a sentence below the range set forth in the Federal Sentencing Guidelines, and the Government files an authorized appeal on the ground that the sentence is unreasonable. Suppose further that the reviewing court discovers, to the surprise of both parties, that the District Court made a further error by overlooking a mandatory minimum to which the defendant was subject. The mandatory minimum would raise the defendant’s sentence beyond what even the Government had wanted. Under the majority’s theory, see ante, at 7, the reviewing court should not remand for imposition of the mandatory minimum, since the decision to seek the higher sentence belonged to the Government alone. But that conclusion is plainly at odds with the text of the statute, which imposes no limits on sentencing review once the named officials have signed off on the appeal.
Section 3742(b)’s limited effect on sentencing review implies that the statute was not designed to prevent judicial encroachment on the prerogatives of the Executive. It is more likely that Congress wanted to withhold from the Executive the power to force the courts of appeals to entertain Government appeals that are not regarded as sufficiently important by the leadership of the Department of Justice. Allowing the courts of appeals, in their discretion, to remedy errors not raised in a cross-appeal in no way trenches on the authority of the Executive. Section 3742(b) may have also been designed to serve the Executive’s institutional interests. Congress may have wanted to ensure that the Government maintained a consistent legal position across different sentencing appeals. Or perhaps Congress wanted to maximize the impact of the Government’s sentencing appeals by giving high-level officials the authority to nix meritless or marginal ones. These institutional interests of the Executive do not undermine the Judiciary’s authority to correct unlawful sentences in the absence of a Government appeal, and they do not justify the Court’s decision today.
IV
For the reasons given above, I would hold that the courts of appeals enjoy the discretion to correct error sua sponte for the benefit of nonappealing parties. The Court errs in vacating the judgment of the Eighth Circuit, and I respectfully dissent.3
Notes
1 The Court argues that petitioner’s original sentence was neither so fundamentally unfair nor so harmful to our system of justice as to warrant sua sponte correction by the Court of Appeals. Ante, at 16, n. 9. But these considerations, which may well support a conclusion that the Court of Appeals should not have exercised its authority in this case, cf. n. 3, infra, surely do not justify the Court’s broad rule that sua sponte error correction on behalf of the Government is inappropriate in all cases.
2 The Court finds it “hard to imagine a case in which a district court, after a court of appeals vacated a criminal sentence, could properly increase the sentence based on an error the appeals court left uncorrected because of the cross-appeal rule.” Ante, at 16, n. 8. Happily, we need not imagine such cases, since they come before our courts every day. For examples, we have no further to look than the sentencing cases remanded en masse following our recent decision in United States v. Booker, 543 U. S. 220 (2005) . In Booker’s wake, it was common for newly convicted defendants to appeal their sentences, claiming that they received enhancements that they would not have received under the advisory guidelines. Many of those cases were remanded for resentencing, and some defendants wound up with even longer sentences on remand. See, e.g., United States v. Singletary, 458 F. 3d 72, 77 (CA2) (affirming a sentence lengthened by 12 months following a Booker remand), cert. denied, 549 U. S. 1047 (2006) ; United States v. Reinhart, 442 F. 3d 857, 860–861 (CA5 2006) (affirming a sentence lengthened from 210 months to 235 months following a Bookerremand). These cases represent straightforward applications of the cross-appeal rule: The Government had not cross-appealed the sentence, so the reviewing court did not order the defendant’s sentence lengthened. And yet the sentence was ultimately lengthened when the error was corrected on remand. The Court fails to explain the conceptual distinction between those cases and this one. If the Court permits sentencing courts to correct unappealed errors on remand, why does it not permit the courts of appeals to do the same on appeal?
3 Neither the parties nor our amicus have addressed whether, under the assumption that the Court of Appeals enjoys discretion to initiate error correction for the benefit of a nonappealing party, the Eighth Circuit abused that discretion in this case. As framed by petitioner, the question presented asked only whether the cross-appeal requirement is subject to exceptions. Because the parties have not addressed the fact-bound subsidiary question, I would affirm without reaching it. See United States v. International Business Machines Corp., 517 U. S. 843 , n. 3 (1996).