249 F.3d 584
7th Cir.
2001

Lead Opinion

EASTERBROOK, Circuit Judge.

Ever since his conviction of drug offenses in 1997, Wayne Brannigan has been carrying on a rear-guard action. We affirmed his conviction and sentence (along with those of his co-conspirators) in an unpublished order. United States v. Jones, No. 97-2262, 1998 WL 234516 (7th Cir. May 7, 1998). Brannigan filed a collateral attack under 28 U.S.C. § 2255. After the district court denied his petition, we declined to issue a certificate of appeal-ability. Brannigan v. United States, No. 00-1628 (7th Cir. July 21, 2000). The ink was hardly dry on that order before Bran-nigan asked this court for permission to file a second collateral attack. See 28 U.S.C. § 2255 ¶ 8. We denied that application. Branigan [sic] v. United States, No. 00-3075 (7th Cir. Sept. 1, 2000). Now Brannigan has filed a successive application for permission to commence collateral litigation. That application, too, was denied with prejudice last month, thus complying with the deadline in 28 U.S.C. § 2244(b)(3)(D), in a short order noting that we would later issue an opinion with a more complete explanation. (Section 2255 ¶ 8 incorporates § 2244(b).) This opinion fulfils that promise.

Brannigan is among the many prisoners who believes that Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000), requires every sentence for a drug offense to be reduced. We warned in Talbott v. Indiana, 226 F.3d 866, 869 (7th Cir.2000), that hasty action on this belief may be costly, because an unfounded petition invoking Ap-prendi may squander the prisoner’s opportunity to file one collateral attack as of right (subjecting future challenges to the gatekeeping provisions of § 2244(b) and § 2255 ¶ 8), or may lead to problems under § 2244(b)(1): “A claim presented in a second or successive habeas corpus application ... that was presented in a prior application shall be dismissed.” A bad Apprendi argument in one application thus may scuttle a better Apprendi argument later. That is exactly what has happened to Brannigan. His earlier application for permission to file a second collateral attack contended that Apprendi foreclosed the district court’s decision to *587add two levels to his offense seriousness under the Sentencing Guidelines for possessing a weapon in the course of his drug dealing. (The extra levels led to his sentence of life imprisonment rather than, say, 360 months.) Brannigan contended that this increase was improper because he had not been convicted of a firearms offense. Sometimes we protect prisoners from the consequences of ill-considered Apprendi contentions by dismissing their applications without prejudice, for the Supreme Court has not declared Apprendi to be retroactively applicable on collateral attack. See Hernandez v. United States, 226 F.3d 839 (7th Cir.2000). But when an argument invoking Apprendi would fail even if that case turns out to be fully retroactive, we deny it on the merits in order to forestall a further round of litigation if the Supreme Court later should decide in favor of retroactivity. That was the fate of Brannigan’s initial application. Apprendi does not require facts pertinent to application of the Sentencing Guidelines to be determined under an elevated burden of persuasion; it holds only that circumstances affecting the statutory maximum punishment must be established beyond a reasonable doubt to the satisfaction of the trier of fact. See Apprendi, 120 S.Ct. at 2362-63; United States v. Patterson, 241 F.3d 912 (7th Cir.2001); Talbott, 226 F.3d at 869. See also Edwards v. United States, 523 U.S. 511, 118 S.Ct. 1475, 140 L.Ed.2d 703 (1998).

Brannigan’s current argument— that the jury rather than the district judge should have determined how much cocaine the conspirators distributed — is closer to the holding of Apprendi. He errs in thinking that Apprendi requires the full weight of drugs, which the district judge determined to be more than 1.5 kilograms of crack, to be ascertained beyond a reasonable doubt by the jury. Distributing even 50 grams of crack exposes a person to life imprisonment. 21 U.S.C. § 841(b)(1)(A)(iii). Once the trier of fact concludes beyond a reasonable doubt that the defendant distributed 50 grams of crack, the district judge decides by a preponderance of the evidence the full extent of relevant conduct for purposes of sentencing. Still, Brannigan has the makings of an Apprendi claim because the jury was not asked to resolve the 50-gram question. Brannigan’s claim is weak, not only because the district judge found that the conspiracy entailed at least 1.5 kilograms of crack (making it unlikely that a jury would have balked at finding a mere 50 grams), but also because, even if Apprendi turns out to be retroactive, to prevail on any claim first raised on collateral attack the petitioner must establish “cause” and “prejudice,” which is more difficult than establishing “plain error.” See United States v. Frady, 456 U.S. 152, 162-66, 102 S.Ct. 1584, 71 L.Ed.2d 816 (1982); United States v. Smith, 241 F.3d 546 (7th Cir.2001). Yet Brannigan has at least the kernel of an Apprendi argument, one whose resolution we would defer under Hernandez but for the operation of § 2244(b)(1).

Section 2244(b)(1) says that a “claim” presented in a prior application is forever closed. What is a “claim” as § 2244(b)(1) uses that word? The answer is elusive. Defining the “claim” for purposes of preclusion in civil litigation has been a complex process, and it is tempting to borrow the answer' — that a single set of facts producing a single injury is one “claim” no matter how many legal theories can be invoked in support of relief. E.g., Herrmann v. Cencom Cable Associates, Inc., 999 F.2d 223 (7th Cir.1993). But this can’t be the right way to understand “claim” in § 2244(b)(1), for then one crime would produce one “claim” no matter how *588many things had gone wrong. This would imply that every successive collateral attack on a single conviction and sentence must be dismissed. Yet § 2244(b) and § 2255 ¶ 8 suppose that multiple collateral attacks are possible. Thus it is essential to define the “claim” as a challenge to a particular step in the case, such as the introduction of a given piece of evidence, the text of a given jury instruction, or the performance of counsel. That’s essentially how Bennett v. United States, 119 F.3d 470, 471-72 (7th Cir.1997), understands it. If, for example, the defendant invokes the fourth amendment to protest the introduction of one item of evidence, a later contest to the same evidence based on the fifth or sixth amendment is just another iteration of the same claim. “A rehashed claim is not a new claim.” Bennett, 119 F.3d at 472. In this respect the civil definition of a “claim” remains instructive; in both civil and criminal practice it is the underlying events, rather than the legal arguments advanced to obtain relief from those events, that demarcate a “claim.” The criminal practice differs because for collateral attack the “events” in question are what transpired in court, rather than the out-of-court events that precipitated the litigation. But the principle that new legal arguments about the same events do not amount to a new claim remains. Accord, Babbitt v. Woodford, 177 F.3d 744, 746 (9th Cir.1999); McDonald v. Bowersox, 125 F.3d 1183, 1185-86 (8th Cir.1997). (In re Medina, 109 F.3d 1556, 1565 (11th Cir.1997), assumes, to the contrary, that every new legal argument is a new “claim” for purposes of § 2244(b)(1), but the court did not give a reason, and the subject apparently had not been debated by the litigants.)

Brannigan’s sentence was determined by the table in the Sentencing Guidelines (U.S.S.G. § 5A) that combines and specifies the effect of all calculations that go into the offense level and criminal history. His current position, however, depends on disgregating that process. In his previous application Brannigan complained about one of the adjustments (two levels for possessing a weapon); now he uses Apprendi to complain about the relevant-conduct calculation (that is, the quantity of cocaine involved). Each petition concerns the same sentence, and the legal theory used to challenge that sentence is the Apprendi principle. It would cut matters entirely too fine to divide into separate “claims” each element of the calculation under the Sentencing Guidelines. That would fracture a single sentence into dozens of “claims,” one for each prior conviction that affects the criminal history level plus one for each offense-severity level. For a calculation that in general is not supposed to allow any collateral attack, see Scott v. United States, 997 F.2d 340 (7th Cir.1993) (the Sentencing Guidelines are not “laws” for purposes of § 2255, so errors in calculating the sentence generally cannot be raised on collateral attack), this would be overboard. Section 2244(b)(1), like the Anti-terrorism and Effective Death Penalty Act (aedpa) of which it is a part, is designed to promote finality, not to illuminate a route to pursuing scores of collateral attacks.

Brannigan’s best argument would be that, although the last time around he disputed the calculation under the Sentencing Guidelines, now he is disputing the district court’s assumption that the maximum lawful sentence is life imprisonment rather than, say, the 20-year maximum for distributing any detectable quantity of cocaine. See 21 U.S.C. § 841(b)(1)(C). This is not a distinction on which Brannigan relies — recall that he thinks Apprendi applicable to the whole of the Guidelines calculation, and that contention, the one he actually presents, is the same “claim” as *589before. Moreover, making the application of § 2244(b)(1) turn on whether a petitioner makes a good legal argument (here, using Apprendi to contest the determination of the statutory maximum) or a bad legal argument (in the former application, using Apprendi to contest a two-level increase for firearms) would sap that statute of effect. Section 2244(b)(1) supposes that the first application was bad and the second better; otherwise the first would not have been dismissed and it would not be necessary to invoke principles of preclusion to dismiss the second. To carve up a trial, conviction, and sentence in such a way that sound and unsound legal arguments are necessarily different “claims” would be to nullify § 2241(b)(1). It is better to conclude that all variations of Apprendi-based challenges to a single sentence are a single “claim.”

Section 2244(b)(2)(A) provides an independent reason for denying a successive application based on Apprendi. A court of appeals must deny an application that presents a claim omitted from a prior application, unless that claim was “previously unavailable” to the prisoner. If Brannigan’s latest use of Apprendi is indeed a “claim” different from the contest to the extra levels for possessing a weapon, then it had to be raised at the same time. Section 2244(b)(2)(A) prevents a prisoner from filing a series of applications, all based on the same decision, to challenge different events in his prosecution, trial, and sentence. Cf. Burris v. Parke, 95 F.3d 465 (7th Cir.1996) (en banc); In re Page, 179 F.3d 1024 (7th Cir.1999).

Thus Brannigan loses either way: under § 2244(b)(1) if he has successively presented different aspects of a single claim, and under § 2244(b)(2)(A) if he has two genuinely different claims based on the same opinion of the Supreme Court. That is why we denied Brannigan’s latest application with prejudice.





Concurrence Opinion

CUDAHY,

concurring in the judgment.

The issue here is whether the application should be dismissed with prejudice or without. Of course, as the majority reiterates, the purpose of 28 U.S.C. § 2244 is to preclude (in the most comprehensive way imaginable) any extended collateral litigation challenging criminal convictions. In general, the section prescribes a “one petition and out” format, so that, excepting extraordinary circumstances, successive petitions never survive for consideration on the merits. This format — and our practice in applying it — presents three possible outcomes for Brannigan’s successive Ap-prendi-based petition. One possible outcome, arising under § 2244(b)(1), is that Brannigan’s claim must be dismissed with prejudice because it was presented in a prior application. An identical outcome is dictated by § 2244(b)(2), under which it is our practice to dismiss with prejudice a claim not presented in a prior application unless it relies on “a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable .... ” Only if Brannigan has presented a new claim that relies on a new rule of constitutional law that was “previously unavailable” to him, will we dismiss his application without prejudice under § 2244(b)(2) because the Supreme Court has not yet declared Apprendi to be retroactive, although there is a possibility that it will in the future.

Judge Easterbrook, the author of today’s majority opinion, has elsewhere characterized the effect of § 2244 as follows: “If the claim has been presented before, it has to be dismissed. If it’s never been presented before, it has to be dismissed.” Constitutional Law Scholars Attempt to Distill Recent Supreme Court Term, 65 *590U.S.L.W. 2274, 2287 (1996). It is thus not surprising that in its haste to ensure that Brannigan never mounts an Apprendi-based challenge to his sentence in this court again, the majority attempts to show that both § 2244(b)(1) and (2) require that Brannigan’s application be dismissed with prejudice. The majority primarily and extensively argues that Brannigan has presented a claim that is identical to the claim presented in his initial application, and that Brannigan’s successive application should thus be dismissed with prejudice under § 2244(b)(1). In the alternative, the majority argues (in a final paragraph that appears to be little more than an afterthought) that Brannigan’s claim, although possibly new, relies on Apprendi — a rule that was not previously unavailable to him. Thus, § 2244(b)(2) also dictates, that Bran-nigan’s application be dismissed with prejudice. But, by devoting its primary effort to the issue of whether Brannigan is now presenting a new claim or an old one, the majority may be pursuing an issue that may have broader ramifications in other contexts. I will therefore give principal attention to the majority’s argument about the meaning of “claim” and bring a secondary focus to its comments on the alternative ground for decision, which may turn on the meaning of “unavailable.”

With respect to the question whether we have two claims or one, I believe that Brannigan’s arguments — the first challenging a weapons enhancement and the second challenging a drug quantity determination' — present different claims. A claim, specifically in the context of the federal habeas statute, is “a set of facts giving rise to a right to a legal remedy.” See Bennett v. United States, 119 F.3d 470, 471-72 (7th Cir.1997) (Posner, C.J.). A claim is therefore distinguished by its facts (specifically, by its “nucleus of operative facts”), not just by the legal principle that it invokes or the body of law from which it derives. In principle, the majority would seem to agree since it correctly notes that “in both civil and criminal practice it is the underlying events, rather than the legal arguments advanced to obtain relief from those events, that demarcate a ‘claim.’ ” Op. at 587-88. The majority continues along the same line of analysis by declaring that “it is essential to define the ‘claim’ as a challenge to a particular step in the case ....” Id. However, the majority then flees the implications of this analysis by declaring that “[i]t would cut matters entirely too fine to divide into separate ‘claims’ each element of the calculation under the Sentencing Guidelines.” Id. at 588.

In my view, the majority’s position simply rejects the straightforward definition of a “claim” as being distinguished by its facts (specifically, its nucleus of operative facts). The facts surrounding a weapons enhancement are obviously quite different from the facts surrounding a drug quantity determination. Therefore, allegations relating to one set of facts would in normal parlance constitute a “claim” separate and distinct from allegations relating to another set of facts.

The majority’s position that Brannigan may not disaggregate the process of U.S.S.G. § 5A, which, in prescribing a sentence, “combines and specifies the effect of all calculations that go into the offense level and criminal history,” op. at 588, also results in a definition of “claim” that might encourage applicants to challenge their sentence in a manner that is contrary to generally accepted pleading requirements. Because the majority refuses to disaggregate the process of § 5A, under the majority’s reasoning an Apprendi-based challenge to an applicant’s sentence might require no more specificity than a general allegation that the sentencing court somehow violated Apprendi when it *591calculated the inmate’s sentence. This, after all, would be the appropriate event, under the majority’s theory, to be raised by the petition. But, were an inmate to file an application containing such a con-clusory allegation, I have little doubt that we would dismiss it for failure to allege a sufficiently specific claim. See, e.g., Dellenbach v. Hanks, 76 F.3d 820, 822 (7th Cir.1996) (noting generally that petitions are dismissed when the petitioner makes conclusory, rather than specific factual, allegations); Aleman v. United States, 878 F.2d 1009, 1012-13 (7th Cir.1989) (conclu-sory allegation that two individuals were government informants required dismissal of § 2255 petition). To avoid dismissal for failure to state a specific claim, we would require the inmate to specify what aspect or aspects of his sentencing calculation ran afoul of Apprendi. This strongly suggests that the same specific aspects constitute the “claim” as denoted by § 2244(b)(2).

Thus, I believe that, at least with regard to Apprendi claims, disaggregation of the sentencing calculation is required when looking at the facts that form such a claim. That is what Brannigan has done here, alleging a violation of Apprendi because of his weapons enhancement in the first application, and because of the drug quantity ascribed to him in the second application. These claims each rely on a different nucleus of facts, which is specific enough to avoid dismissal as conclusory.

Perhaps realizing the difficulties of its primary argument, the majority also invokes a parade of horribles that raises the possibility of challenging a single sentence multiple times based on one challenge for each of “dozens” of prior convictions affecting the criminal history level, “plus one for each offense-severity level.” Slip op. at 5. But such a scenario poses no threat in the real world, for it is hard to imagine that a Supreme Court decision would provide a non-frivolous basis for invalidating each one of “dozens” of prior convictions. And if an inmate files a barrage of frivolous claims, based on Apprendi or whatever, they can be quashed regardless of the provisions of § 2244 because this court has an arsenal of weapons to employ against serial filers of frivolous claims — whether convicted inmates or others.

Accordingly, it is entirely reasonable and consonant with the plain meaning of the statute to regard a claim based on a firearms enhancement to be quite different from one based on drug quantity. The AEDPA requires that we deal differently with a claim “presented in a prior application” from one “not presented in a prior application.” Brannigan’s drug quantity claim has not been presented in a prior application and it should be dealt with on that basis.

In its final (almost “throw-away”) paragraph of argument, the majority presents an entirely different rationale for dismissing with prejudice. Under § 2244(b)(2)(A), we will dismiss a new claim with prejudice unless the claim “relies on a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable .... ” Crucial here is the answer to the question whether Ap-prendi qualifies as a new rule that was “previously unavailable” to Brannigan. It may, of course, be argued that Apprendi is no longer a new rule from Brannigan’s perspective because it was decided prior to Brannigan’s previous habeas corpus application. See Bennett v. United States, 119 F.3d 470, 472 (7th Cir.1997); In re Medina, 109 F.3d 1556, 1565 (11th Cir.1997). It seems to me, however, that the language of § 2244(b)(2)(A) ought to be read as a whole and that, as long as a rule remains “unavailable,” it must correspondingly be regarded as “new.” Cf. Hernandez v. *592United States, 226 F.3d 839, 841 (7th Cir.2000) (“[A] new rule that is retroactive for purposes of collateral attack is not ‘available’ for a § 2256 motion until the Supreme Court has clearly ruled that this is the case.”). In other words, a rule remains “new” as long as it is “unavailable.” Since the Supreme Court has not made Apprendi retroactive to cases on collateral review, the rule remains unavailable to the applicant and functionally occupies the same position as if it were literally “new.” Whether the applicant knew of the existence of the rule or not — whether it was “new” to him — he could not take advantage of it; it was not available to him.

However, we have the additional circumstance here that, when he filed his earlier application, the applicant knew of Appren-di and attempted to rely on its (then unavailable) rule. Significantly, however, the panel then considered his claim as if Ap-prendi had been made retroactive to cases on collateral review by the Supreme Court. Cf. Hernandez, 226 F.3d at 841. Roughly speaking, that earlier decision was “on the merits” in that it applied the Apprendi rule even though Apprendi was then “unavailable.” It may not be stretching things too far to see this earlier decision as affording Brannigan all the consideration due him under the statute. On this basis, I see the factors arguing for and against dismissal with prejudice as being in equipoise. A tie, perhaps, goes to the majority, and on that basis, I can concur in the judgment.