Derman v. United States

SELYA, Circuit Judge. In this appeal, petitioner-appellant Herbert Derman challenges the district court’s denial of his application for post-conviction relief. He argues that the Supreme Court’s holding in Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000) — announced after we rejected his direct appeal but before the expiration of the time within which he was eligible to apply for a writ of certiorari — • demands that we set aside his conviction. Although his appeal is ably argued and raises intricate issues (including one that has divided our sister circuits and another that is of first impression at the appellate level), it is in the end unavailing. Consequently, we affirm the district court’s dismissal of the application for post-conviction relief. I. BACKGROUND We limn those facts pertinent to the instant appeal, referring readers who crave more exegetic detail to our opinion affirming the petitioner’s conviction. See United States v. Derman, 211 F.3d 175 (1st Cir.2000). Beginning in 1984, Marcel Rosenzweig oversaw an underground greenhouse on property owned by the petitioner. This facility housed a huge marijuana-growing operation. The venture prospered for several years. When word of a large-scale marijuana grow leaked in 1991, police officers visited the site. They were thrown off the scent by a legal above-ground greenhouse that Rosenzweig and his cohorts ran to conceal the illegal activities below. Because the officers did not realize what lay beneath, their search revealed only trace amounts of marijuana. In an abundance of caution, Rosenzweig moved the enterprise to a different locus. The culprits continued growing and distributing marijuana until the federal government cracked the case four years later. A federal grand jury sitting in the District of Massachusetts soon indicted the petitioner and six other persons (including Rosenzweig). The indictment charged the defendants with, inter alia, conspiring to manufacture and distribute marijuana. See 21 U.S.C. §§ 841(a)(1), 846. As part of the conspiracy charge, the indictment specifically mentioned a statutory provision mandating a ten-year minimum sentence for conspiracies involving at least 1,000 marijuana plants. See id. § 841(b)(1)(A). *38 All the defendants, save only the petitioner, pleaded guilty to the charges. The petitioner maintained his innocence, asserting that he had no knowledge of either the underground greenhouse or its unlawful contents. The district court instructed the jury on the standard issues, but not on the question of drug quantity. The jury found the petitioner guilty on the conspiracy count and on a related forfeiture count. 1 The jury returned its verdict on July 29, 1998. The district court denied the petitioner’s post-trial motions and scheduled the disposition hearing-to take place on March 5, 1999. Drug quantity was a contested issue. See United States v. O’Cam-po, 973 F.2d 1015, 1026 (1st Cir.1992) (“[T]he base offense level of a co-conspirator at sentencing should reflect only the quantity of drugs he reasonably foresees it is the object of the conspiracy to distribute after he joins the conspiracy.”). The relevant measure of drug quantity in this case was the number of marijuana plants involved. Leaving recidivism to one side, a defendant convicted of participation in a conspiracy that involves fewer than fifty plants can receive an inearcerative sentence of no more than five years. 21 U.S.C. § 841(b)(1)(D). If the conspiracy involved fifty plants or more, the maximum sentence is twenty years. Id. § 841(b)(1)(C). For a conspiracy of 100 plants or more, the maximum sentence is forty years. Id. § 841(b)(1)(B). Finally, for a conspiracy of 1,000 plants or more, the maximum sentence is life. Id. § 841(b)(1)(A). The number of marijuana plants also can dictate a mandatory minimum sentence: five years for 100 plants or more, id. § 841(b)(1)(B), and ten years for 1,000 plants or more, id. § 841(b)(1)(A). In this instance, the probation department prepared a presentence investigation report (the PSI Report) concluding that the petitioner was responsible for 213,000 marijuana plants. The petitioner objected, claiming that he could not reasonably have foreseen the vast amounts of marijuana grown underneath his property and elsewhere. On that basis, he argued that his sentence should not be more than five years. See id. § 841(b)(1)(D) (establishing a five-year default statutory maximum for a quantity of marijuana less than fifty plants). The district court flatly rejected the petitioner’s contention. Focusing on the fact that 1,000 plants was the number of marijuana plants needed to trigger a ten-year mandatory minimum sentence, see id. § 841(b)(1)(A), the court stated: “I would have to find that the sky was green to conclude that there weren’t at least a thousand plants that were foreseeable in this conspiracy at the time that [the petitioner] entered into it.... ” That said, the court proceeded to calculate the total number of plants attributable to the petitioner. The court presumed that the petitioner reasonably could have foreseen an output of 200 plants a month (a total of 26,400 plants over the eleven-year span of the conspiracy). The court then sentenced the petitioner to an inearcerative term of 121 months (one month above the applicable mandatory minimum). The petitioner appealed his conviction, but not his sentence. We rejected his direct appeal on May 5, 2000. Derman, 211 F.3d at 177. In the course of that appeal, he assigned no error implicating either the jury instructions or the lower court’s assessment of drug quantity. *39 The petitioner had ninety days from the date of entry of our judgment to file a petition for a writ of certiorari to the Supreme Court of the United States. See Sup.Ct. R. 13(1). The Supreme Court decided Apprendi on June 26, 2000-well within that ninety-day window. Appren-di `s core holding is that "[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt." 530 U.S. at 490, 120 S.Ct. 2348. The petitioner's trial seemingly flouted this principle: after all, the district court did not instruct the jury on a fact-drug quantity-that increased the maximum penalty for the petitioner's crime from five years (the default statutory maximum) to life imprisonment. Compare 21 U.S.C. § 841(b)(1)(D) with id. § 841(b)(1)(A). The petitioner claims that he instructed his appellate counsel to file a certiorari petition on this ground during the period in which that opportunity was available to him, but that counsel neglected to comply. So ended direct review of the petitioner's conviction and sentence. The petitioner thereafter retained a new lawyer. On April 27, 2001, he launched a collateral attack in the district court. See 28 U.S.C. § 2255. He asserted that the court had sentenced him in violation of Apprendi, and that he had received ineffective assistance of counsel because his appellate attorney had ignored his instructions to file a petition for a writ of certiorari The district court held a hearing on October 9, 2001. Ruling from the bench, the court dismissed the petition. Three days later, the court elaborated on its reasoning in a written rescript. To summarize, the court rejected the petitioner's Ap-prendi claim for three reasons. It held that Apprendi could not be applied retroactively to the petitioner's habeas claim; that the petitioner had waived any and all objections to his sentence by failing to challenge the sentence on direct appeal; and that, in all events, the district court's failure to submit the drug quantity question to the jury was harmless because the record contained overwhelming evidence that the number of marijuana plants involved in the plot exceeded the number necessary to bring an elevated maximum sentence into play. On much the same basis the court also rejected the petitioner's ineffective assistance claim. This appeal ensued. We granted a certificate of appealability to decide the Ap-prendi issues. See 28 U.S.C. § 2258(c). II. FINALITY We begin our analysis by determining the applicability of Apprendi to this petition-a determination that requires us to decide when the petitioner's conviction became "final" within the meaning of the relevant statute. Whether a convicted defendant may find refuge in a rule of criminal procedure newly announced by the Supreme Court depends in large part on timing. If the conviction is not yet final when the Supreme Court announces the rule, then inferior courts must apply that rule to the defendant's case. Griffith v. Kentucky, 479 U.S. 314, 322, 107 S.Ct. 708, 93 L.Ed.2d 649 (1987). If, however, the conviction is already final, then the defendant ordinarily may not avail himself of the newly announced rule. Teague v. Lane, 489 U.S. 288, 310, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989). 2 *40 But words often are chameleons, taking on different shades of meaning in different contexts. The question, then, is how to define the word "final" in this setting. At the time the Court decided Teague, the definition seemed clear: a conviction becomes final when "a judgment of conviction has been rendered, the availability of appeal exhausted, and the time for a petition for certiorari [has] elapsed or a petition for certiorari [filed and] finally denied." Griffith, 479 U.S. at 321 n. 6, 107 S.Ct. 708. But the law, by its nature, evolves over time, and the Antiterrorism and Effective Death Penalty Act (AEDPA), Pub.L. No. 104-132, 110 Stat. 1214 (1996), casts a shadow on the Griffith definition. In that statute, Congress for the, first time established time limits applicable to the filing of habeas petitions: a prisoner (state or federal) has one year from the date on which his conviction becomes "final" within which to seek federal habeas relief. 3 See 28 U.S.C. §~ 2244(d)(1), 2255. In discussing finality in this context, Congress employed slightly different terminology in regard to the time limits applicable to state as opposed to federal prisoners. The one-year period for filing a petition from a state court conviction begins to run on "the date on which the judgment became final by the conclusion of direct review or the expiration of the time for seeking such review." 28 U.S.C. § 2244(d)(1)(A). The one-year period for filing a petition challenging a federal court conviction is de~ scribed more tersely; that period begins to run from and after "the date on which the judgment of conviction becomes final Id. § 2255. In Gendron v. United States, 154 F.3d 672 (7th Cir.1998) (per curiam), the Seventh Circuit determined that the linguistic differences in the two provisions meant that Congress intended divergent versions of finality for state and federal convictions. The court reasoned that "[w]here Congress includes particular language in one section of an act but omits it in another section of the same act, . . . the language will not be implied where it has been excluded." Id. at 674. Since Congress used the phrase "expiration of the time for seeking such review" only in regard to state prisoners,, the court concluded that "federal prisoners who decide' not to seek certiorari with the Supreme Court will have the period of limitations begin to run on the' date [the court of appeals] issues the mandate in their direct criminal appeal." Id. In Kapral v. United States, 166 F.3d 565 (3d Cir.1999), the Third Circuit took issue with this interpretation. The court expressed concern that Gendron invited a series of unfortunate results. Taken at face value, the Gendron rationale means that a federal prisoner can file a collateral attack in the district court and then continue his pursuit of direct review by filing a certiorari petition. Id. at 570-71. Moreover, under that rationale, finality will be defined differehtly for limitation purposes than for purposes of the Teague "new rule" analysis. Id. at 572. Last-but far from least-the Gendron rationale creates an artificial distinction between state and federal prisoners-a distinction that Congress, given the common origin and purpose of the two AEDPA provisions, would have had no reason to draw. Id. at 575. In light of these realities, the court deter *41 mined that Congress intended sections 2244 and 2255 to operate on the same time line. Id. Thus, under the Third Circuit’s reading of the AEDPA, a conviction— whether state or federal — does not become final until “the later of (1) the date on which the Supreme Court affirms the conviction and sentence on the merits or denies the defendant’s timely filed petition for certiorari, or (2) the date on which the defendant’s time for filing a timely petition for certiorari review expires.” Id. at 577. Although the Fourth Circuit has followed Gendron, see United States v. Torres, 211 F.3d 836, 839-41 (4th Cir.2000), a clear majority of the circuit courts that have addressed the question have marched in lockstep with Kapral. See, e.g., Kaufmann v. United States, 282 F.3d 1336, 1339 (11th Cir.2002); United States v. Garcia, 210 F.3d 1058, 1060 (9th Cir.2000); United States v. Gamble, 208 F.3d 536, 536-37 (5th Cir.2000) (per curiam); United States v. Burch, 202 F.3d 1274, 1279 (10th Cir.2000). We too find the Kapral formulation the more persuasive. While it would be pleonastic to rehearse the Third Circuit’s analysis, we feel obliged to emphasize a factor that has special relevance to the instant appeal: the relationship between the finality of the defendant’s conviction and the announcement of new rules of criminal procedure by the Supreme Court. Five years after Teague was decided, the Court made pellucid that “[a] state conviction and sentence become final for purposes of [the Teague analysis] when the availability of direct appeal to the state courts has been exhausted and the time for filing a petition for a writ of certiorari has elapsed or a timely filed petition has been finally denied.” Caspari v. Bohlen, 510 U.S. 383, 390, 114 S.Ct. 948, 127 L.Ed.2d 236 (1994). Accordingly, were we to follow Gendron’s lead and distinguish between the finality of state and federal convictions, we would be forced to deny federal prisoners the benefit of a new rule available to similarly situated state prisoners. Such a result is counter-intuitive, and we do not think that Congress intended the nature of the particular sovereign detaining the prisoner to have such a powerful impact on which defendants may avail themselves of a new rule. Indeed, we can discern no sensible reason why Congress might wish to classify prisoners in this peculiar manner — and under the best of circumstances, such a taxonomy would be difficult to administer. Certainly, the minor variation in language between section 2244 and section 2255, respectively, does not demand such an awkward result. Section 2255 directs that a federal court judgment be “final”— and that term has a well-defined meaning in federal law. See, e.g., Griffith, 479 U.S. at 321 n. 6, 107 S.Ct. 708; Hanover Ins. Co. v. United States, 880 F.2d 1503, 1509 (1st Cir.1989). Section 2244 contains the same directive with respect to state court convictions, but finality is a concept that has differing meanings under the laws of the several states. Compare, e.g., Warren v. State, 833 S.W.2d 101, 102 (Tenn.Crim. App.1992) (“If one convicted of a crime takes no action to perfect his right to appeal, the statute of limitations begins to run from the date of final conviction”), with, e.g., Mont.Code. Ann. § 46-21-102(1) (defining finality for purposes of state ha-beas proceedings in a manner similar to 28 U.S.C. § 2244(d)(1)(A)). Thus, Congress chose to ensure uniformity by explaining what “final” means in the context of a federal habeas petition that seeks to challenge a state court conviction. No similar need existed with respect to federal court convictions (and, thus, the explanatory language was omitted in section 2255). *42 We will not paint the lily. We hold that a conviction for a federal defendant who fails to file a petition for a writ of certiora-ri becomes final when the period in which he seasonably might have dpne so expires. This levels the playing field as between state and federal prisoners desirous of seeking federal habeas relief. Our construction of section 2255 resolves the threshold question presented in this appeal. Because the Supreme Court decided Apprendi while the petitioner still had breathing room within which to file a petition for certiorari, his conviction was not yet final at that time. It follows inexorably that the Apprendi rule applies to the petitioner's case. Griffith, 479 U.S. at 322, 107 S.Ct. 708. What remains to be seen is whether Apprendi aids the petitioner's cause. III. THE MERITS We next trace the contours of the Ap-prendi error that occurred during the petitioner's trial and decide whether it necessitates vacation of the petitioner's sentence. A. The Apprendi Error. The government concedes that the petitioner's case suffers from a strain of Apprendi error. In hindsight, the trial court should have asked the jury to determine, beyond a reasonable doubt, whether the underlying conspiracy involved a drug quantity sufficient to trigger a sentence higher than the five-year default statutory maximum. See United States v. Barnes, 244 F.3d 172, 177-78 (1st Cir.) (explaining the genesis of the default statutory maximum in an analogous context), cert. denied, - U.S. , 122 S.Ct. 379, 151 L.Ed.2d 289 (2001); see also 21 U.S.C. § 841(b)(1)(D) (setting five-year maximum for a conspiracy involving fewer than fifty marijuana plants). Absent such a determination, the sentence imposed by the district court-ten years and one month-is open to question. The petitioner tries to define the district court's error more broadly. He asserts that the court transgressed Apprendi by determining the drug quantity that was reasonably foreseeable to him rather than tendering that question to the jury. In other words, he claims that the court should have submitted to the jurors not only the qi~estion of drug quantity vis-à-vis the conspiracy but also the individualized question of what drug quantity was attributable to him as a coconspirator. We do not agree: the Apprendi error is far narrower in scope than the petitioner suggests. In Edwards v. United States, 523 U.S. 511, 118 S.Ct. 1475, 140 L.Ed.2d 703 (1998), the Supreme Court held that, as long as (1) the jury finds beyond a reasonable doubt that a defendant participated in a conspiracy, and (2) the Court sentences him within the statutory maximum applicable to that conspiracy, the court may "determine both the amount and the kind of `controlled substances' for which [the] defendant should be held accountable- and then ... impose a sentence that varies depending upon amount and kind." Id. at 513-14, 118 S.Ct. 1475. Apprendi, decided two years later, did not purport to overrule Edwards, and the two decisions are easily harmonized: in a drug conspiracy case, the jury should determine the existence vel non of the conspiracy as well as any facts about the conspiracy that will increase the possible penalty for the crime of conviction beyond the default statutory maximum; 4 and the judge should deter *43 mine, at sentencing, the particulars regarding the involvement of each participant in the conspiracy. See Edwards, 523 U.S. at 514, 118 S.Ct. 1475. This means that once the jury has determined that the conspiracy involved a type and quantity of drugs sufficient to justify a sentence above the default statutory maximum and has found a particular defendant guilty of participation in the conspiracy, the judge lawfully may determine the drug quantity attributable to that defendant and sentence him accordingly (so long as the sentence falls within the statutory maximum made applicable by the jury's conspiracy-wide drug quantity determination). Cf. United States v. Eirby, 262 F.3d 31, 37 (1st Cir.2001) (holding, post-Apprendi, that when a sentence falls within the statutory maximum, "judicial determination of drug quantity under a preponderance-of-the-evidence standard remains a viable option"). The rule, then, is that the government need only allege and prove to the jury the bare facts necessary to increase the statutory sentencing maximum for the conspiracy as a whole (e.g., that the conspiracy involved at least 1,000 marijuana plants). See United States v. Patterson, 292 F.3d 615, 623 (9th Cir.2002) (upholding conviction based on jury finding that the defendant was guilty of manufacturing 100 or more plants, although the jury was not instructed to determine the exact amount). The decisions cited by the petitioner in support of a contrary rule do not withstand scrutiny. He relies chiefly on United States v. Nordby, 225 F.3d 1053 (9th Cir.2000), but no extended discussion of that decision is warranted. It suffices to say that the Ninth Circuit severely limited Nordby `s reach in United States v. Buckland, 289 F.3d 558, 567-68 (9th Cir.2002) (en banc), and then issued a decision in Patterson that coheres with the decision we reach today. See Patterson, 292 F.3d at 623. Our recent opinion in United States v. Bailey, 270 F.3d 83 (1st Cir.2001), does not advance the petitioner's position. There, the defendant argued that the jury-and not the judge-should have decided the quantity of drugs attributed to him at sentencing. Id. at 85-86. We agreed that, because the defendant had been sentenced to a term above the default statutory maximum, an Apprendi error had occurred. Id. at 89. We then considered whether the error was harmless and concluded that it was not; the evidence needed to boost the statutory maximum to the next level was sketchy and permitted a reasonable factfinder to decide the drug quantity issue either way. Id. at 89-90. Consequently, we vacated the sentence and remanded for resentencing. Id. at 90. The decisive factor in Bailey, however, was the government's inability to prove beyond a reasonable doubt that the conspiracy involved the requisite drug quantity necessary to elevate the applicable statutory maximum to the next level. That weakness is not present here. See infra Part 111(B). To summarize, we conclude that an Apprendi error occurred, but that the trial court's determination that the petitioner reasonably could have foreseen that the conspiracy would encompass at least 26,-400 marijuana plants should not be regarded as part and parcel of that error. See Edwards, 523 U.S. at 513-14, 118 S.Ct. 1475. Instead, the Apprendi error relates to the failure to have the jury determine the number of plants involved in the con *44 spiracy. It is to the effect of that error that we now turn. B. The Effect of the Error. The next step in the pavane requires that we determine whether the discerned error invalidates the petitioner's sentence. On this point, the petitioner faces an uphill climb. The fact that Apprencli is available to him in theory (because his conviction was not yet final when Apprencli was decided) does not mean that he can take advantage of that decision in practice; the rub is that he failed either to object to the jury instructions or to contest the trial court's sentencing determination on direct appeal. These omissions transgress the general rule that a criminal defendant must seasonably advance an objection to a potential constitutional infirmity in order to preserve the point for collateral attack. 5 See Burks v. Dubois, 55 F.3d 712, 716 (1st Cir.1995). The rationale behind the rule is straightforward: A contemporaneous objection enables the record to be made with respect to the constitutional claim when the recollections of witnesses are freshest, not years later in a federal habeas proceeding. It enables the judge who observed the demeanor of those witnesses to make the factual determinations necessary for properly deciding the federal constitutional question. Wainwright v. Sykes, 433 U.s. 72, 88, 97 S.Ct. 2497, 53 L.Ed.2d 594 (1977). Moreover, the rule prevents "sandbagging' on the part of defense lawyers, who may take their chances on a verdict of not guilty in a state trial court with the intent to raise their constitutional claims in a federal habeas court if their initial gamble does not pay off." Id. at 89, 97 S.Ct. 2497. The petitioner concedes that he failed to raise and preserve an Apprendi objection at his trial, but he nonetheless seeks to avoid any penalty for this procedural default on the ground that Apprencli changed the traditional method of determining drug quantity for sentencing purposes, and, thus, constituted a watershed decision. Whether or not this characterization of Apprendi is apt, the petitioner's argument is misguided. The inquiry into the applicability of the procedural default rule is, for the most part, black or white: either the defendant proffered a timely objection or he did not. While there are a few exceptions, see, e.g., Bousley v. United States, 523 U.S. 614, 621-22, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998) (discussing exception for claims that could not be presented without further factual development, e.g., a claim that a guilty plea was coerced), the subsequent announcement of a Supreme Court ruling-whether or not it blazes new trails-is not one of them. 6 *45 Bousley illustrates this point. That case involved a habeas application prosecuted by an individual who had pleaded guilty to a federal firearms offense that was later circumscribed by the Court. Id. at 616-18, 118 S.Ct. 1604 (citing Bailey v. United States, 516 U.S. 137, 143-50, 116 S.Ct. 501, 133 L.Ed.2d 472 (1995)). Even though the defendant entered his plea without knowing that the Court subsequently would curtail the statute in a way that might have affected the outcome of his case, the Court performed an archetypical procedural default analysis when it considered his habe-as petition. See id. at 621-22, 118 S.Ct. 1604. We see no principled distinction here. We thus proceed to the question of whether the petitioner's procedural default is excused. A defendant can surmount this hurdle in one of two ways. First, he can offer evidence sufficient to prove that he is actually innocent of the underlying charge. Id. at 622, 118 S.Ct. 1604. The petitioner makes no such claim. The second way in which a defendant can clear the procedural default hurdle is by showing good cause for the default and actual prejudice resulting therefrom. Burks, 55 F.3d at 716. We use the conjunctive purposefully because the defendant must carry the devoir of persuasion as to both cause and prejudice. Id. Because the petitioner in this case has not sufficiently demonstrated prejudice, see text i'r&fra, we need not inquire into the question of cause, 7 The showing of prejudice needed to cure a procedural default generally requires a habeas petitioner to demonstrate "that `there is a reasonable probability' that the result of the trial would have been different" absent the error. Strickler v. Greene, 527 U.s. 263, 289, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999) (quoting Kyles v. Whitley, 514 U.s. 419, 433, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995)). The question is not whether the petitioner, qua defendant, would more likely have received a different verdict had the error not occurred, but whether he received a fair trial, understood as a trial worthy of confidence, notwithstanding the bevue. See Kyles, 514 U.S. at 434, 115 S.Ct. 1555; see also Prou v. United States, 199 F.3d 37, 49 (equating the prejudice standard for ineffective assistance cases with the standard for showing case and prejudice under United States v. Frady, 456 U.S. 152, 102 S.Ct. 1584, 71 L.Ed.2d 816 (1982)). To be sure, the Supreme Court has not elaborated the precise definition of the cause and prejudice standard for all claims. See Amadeo v. Zant, 486 U.S. 214, 221, 108 S.Ct. 1771, 100 L.Ed.2d 249 *46 (1988). 8 Still, any, error that results in unfairness so patent as to violate the Due Process Clause will necessarily satisfy the Strickler and Kyles standards. See Murray v. Carrier, 477 U.S. 478, 494, 106 S.Ct. 2639, 91 L.Ed.2d 397 (1986). We analyze the issue of prejudice based on an examination of the record as a whole. See Frady, 456 U.S. at 169, 102 S.Ct. 1584. Here, our inquiry focuses on the likelihood that the jury, had it been asked the question, would have found that the underlying conspiracy involved the manufacture and distribution of at least fifty marijuana plants. See 21 U.S.C. § 841(b)(1)(C); see also Harris v. United States, - U.S.-,-, 122 S.Ct. 2406, 2418, 153 L.Ed.2d 524, (2002). The burden rests with the petitioner to show that there is a reasonable probability that the jury would have reached a different, more favorable conclusion. Coleman v. Thompson, 501 U.S. 722, 750, 111 S.Ct. 2546, 115 L.Ed.2d 640 (1991); Burks, 55 F.3d at 716. This inquiry need not detain us. The evidence is commanding that the conspiracy of which the appellant was a member involved far more than fifty plants, and that any rational jury would have found as much. After all, during the course of the trial the government presented detailed evidence about the size, scope, and inner workings of the criminal cabal. During the seven years that the coconspirators operated the underground greenhouse on the petitioner's property, marijuana was grown in three rooms. Two rooms were 104 feet long and 36 feet wide, and the other 72 feet long and 38 feet wide. The facility was equipped with special air-conditioning and heating systems, and was powered by an independent generator. At its high point, the underground greenhouse housed as many as 20,000 plants. Individual harvests yielded as many as 5,000 plants. The alternate site was on the same order of magnitude; the government seized 5,600 plants when it raided the premises in 1995. In a nutshell, no reasonable juror could have found that the conspiracy involved fewer than several thousand marijuana plants. The evidence of the size and duration of the operation was copious. So too was the evidence of the petitioner's participation in the enterprise (indeed, he does not now contest that the government adequately tied him to the marijuana-growing operation). Fairly viewed, the record as a whole does not lend credence to the petitioner's plaint that the outcome might have been different but for the Apprendi error. In an effort to blunt the force of this conclusion, the petitioner argues that, had the district court consigned the drug-quantity issue to the jury, he would have contested that issue more vigorously. This argument rings hollow. The indictment explicitly stated that the petitioner was charged with conspiracy in the manufacture and distribution of 1,000 or more marijuana plants, and he was thus on notice that he faced a mandatory minimum sentence of ten years unless he could cast doubt upon that allegation. That was incentive enough to mount as robust a challenge as possible. And to cinch matters, the petitioner has alluded to no evidence *47 upon which he could have based a credible claim that the conspiracy involved only a small number of marijuana plants. The petitioner also posits that he did not know of the breadth of the conspiracy, and, thus, could not have foreseen the number of plants attributed to him at sentencing. That argument is moot. As we have said, once it was established that the petitioner was a participant in a conspiracy that involved at least fifty marijuana plants, the district court was free, under Edwards and Apprendi, to determine foreseeability and sentence the petitioner within the elevated statutory maximum. The court did so — and the petitioner eschewed a timely challenge to that determination. 9 IV. CONCLUSION We need go no further. Although the jury instructions in this case did not anticipate Apprendi, that unpreserved error did not result in cognizable prejudice. Neither the petitioner’s trial nor his sentence were fundamentally unfair. Under the circumstances, the district court appropriately denied the application for post-conviction relief. Affirmed. Footnotes [1] The jury acquitted the petitioner on several other charges, including money laundering. Because the acquitted conduct is not relevant to the issues on appeal, we do not elaborate on those counts. [2] To be sure, there are certain circumstances in which a newly announced rule may be applied retroactively to a conviction that became final beforehand. See Teague, 489 U.s. at 310, 109 S.Ct. 1060. Given the timing *40 here, see text infra, we need not explore these exceptions. [3] This one-year limit is subject to certain periods of tolling. See, e.g., Neverson v. Bissonnette, 261 F.3d 120, 124 (1st Cir.2001). These refinements are not relevant here. [4] This does not mean that a jury need return a special verdict describing the precise amount of drugs involved in the conspiracy. It is enough that the jury supportably deter *43 mines, beyond a reasonable doubt, that the conspiracy involves a drug quantity that surpasses the threshold amount needed to trigger the relevant (higher) statutory maximum. See, e.g., United States v. Patterson, 292 F.3d 615, 623 (9th Cir.2002) (finding no Apprendi error when jury returned a guilty verdict after being instructed on only a threshold quantity). *45 (8th Cir.2001), cert. denied, - U.S. -, 122 S.Ct. 848, 151 L.Ed.2d 725 (2002); United States v. Sanders, 247 F.3d 139, 145-46 (4th Cir.), cert. denied, - U.S. -, 122 S.Ct. 573, 151 L.Ed.2d 445 (2001); United States v. Smith, 241 F.3d 546, 548-49 (7th Cir.), cert. denied, - U.S. -, 122 S.Ct. 267, 151 L.Ed.2d 195 (2001). We recognize, however, that most of these decisions produced vociferous dissents, see, e.g., McCoy, 266 F.3d at 1272-74 (Barkett, J., dissenting); Moss, 252 F.3d at 1005-06 (R. Arnold, 3., dissenting), and, given the utter absence of any prejudice here, it would serve no useful purpose to dive gratuitously into these murky waters. [5] This rule was first developed in the context of examinations by federal habeas courts into independent and adequate grounds for procedural default in state trials. E.g., Wainwright v. Sykes, 433 U.s. 72, 86-89, 97 S.Ct. 2497, 53 L.Ed.2d 594 (1977). The Supreme Court has extended the rule to federal convictions. E.g., Bousley v. United States, 523 U.s. 614, 622, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998). [6] This is not to say that how one categorizes Apprendi is irrelevant. Whether Apprendi may properly be classified as a watershed opinion and whether the petitioner could have foreseen the Court's ruling are factors that go to the "cause" prong of the procedural default analysis. See Bousley, 523 U.S. at 623, 118 S.Ct. 1604. Those courts of appeals that have grappled with these questions to date have held unanimously that Apprendi was not a watershed decision, that the Court's opinion was foreseeable to criminal defendants, and, therefore, that no cause existed sufficient to excuse a procedural default. E.g., McCoy v. United States, 266 F.3d 1245, 1258-59 (11th Cir.2001), cert. denied, - U.S. 122 S.Ct. 2362, 153 L.Ed.2d 183 (2002); United States v. Moss, 252 F.3d 993, 100 1-03 [7] This means, of course, that we need not dwell on the petitioner's allegation that his appellate counsel was ineffective in failing to file a petition for certiorari. It bears mention, however, that the Supreme Court has explicitly held that the failure to file a petition for discretionary review cannot constitute cognizable ineffective assistance of counsel, and, thus, such a failure cannot constitute cause sufficient to excuse a procedural default. Coleman v. Thompson, 501 U.S. 722, 752-53, 111 S.Ct. 2546, 115 L.Ed.2d 640 (1991). [8] The Court recently held, however, that a defendant's failure to object to an Apprendi error at trial engenders plain error review on direct appeal. United States v. Cotton, - U.S. -.-, 122 S.Ct. 1781, 1785, 152 L.Ed.2d 860 (2002), This is a rigorous standard-under it, a court will reverse a conviction only if, among other things, the error seriously affected the fairness, integrity, or public reputation of the trial-but the showing of prejudice required to excuse a procedural default is even more demanding. Frady, 456 U.S. at 166, 102 S.Ct. 1584. [9] Even were the issue of foreseeability still open, the district court's findings seem unim-pugnable. Given the sheer size of the marijuana-growing operation, its duration, and the mass of evidence indicating that the petitioner played an integral role in it, there is no realistic possibility, let alone a reasonable probability, that a jury would have attributed fewer than fifty marijuana plants to him.