Case Information

UNITED STATES of America, Plaintiff-Appellee, v. Juan SANCHEZ-CERVANTES, aka Hugo Quirox, Quiroc, Quiroz, Quiroz Trejo, and Quiroz Tapia, Defendant-Appellant.

No. 98-35897

United States Court of Appeals, Ninth Circuit

March 1, 2002

Argued and Submitted Nov. 5, 2001. As Amended March 15, 2002.

282 F.3d 664

respectfully dissent from the part that results in the granting of this petition.

William S. Labahn, Law Offices of William S. Labahn, P.C., Eugene, OR, for the defendant-appellant.

Michael A. Rotker, Assistant United States Attorney, Washington, DC, and Christopher L. Cardani, Assistant United States Attorney, Eugene, OR, for the plaintiff-appellee.

Before: HUG, T.G. NELSON, and GOULD, Circuit Judges.

T.G. NELSON, Circuit Judge.

Juan Sanchez-Cervantes appeals from the district court's denial of his initial petition for relief pursuant to 28 U.S.C. § 2255. Sanchez-Cervantes argues that his conviction and sentence for federal drug violations should be vacated because of the rule announced in Apprendi v. New Jersey1 and because he received ineffective assistance of counsel at trial. We hold that the new rule of criminal procedure announced in Apprendi does not apply retroactively on initial collateral review, and Sanchez-Cervantes' counsel did not render ineffective assistance. Therefore, we affirm the district court's dismissal of Sanchez-Cervantes' habeas petition.

I.

On March 18, 1993, Juan Sanchez-Cervantes was indicted on one count of possession with intent to distribute methamphetamine, one count of possession with intent to distribute cocaine, one count of conspiracy to distribute controlled sub-stances, all in violation of 21 U.S.C. § 841(a)(1), and one count of illegal reentry into the United States in violation of 8 U.S.C. § 1326(a) and (b)(2). At trial, the Government introduced evidence that Sanchez-Cervantes conducted numerous drug deals and that he illegally reentered the country, as well as evidence of Sanchez-Cervantes' prior drug convictions. Sanchez-Cervantes testified at trial and admitted to illegally reentering the United States and to being a small-time drug dealer. However, he denied any involvement in a conspiracy. Of the seven co-defendants on trial, Sanchez-Cervantes was the only one to testify.

The jury convicted Sanchez-Cervantes on all counts, but made no findings as to drug quantities. After determining, based on the presentence drug report, that Sanchez-Cervantes was responsible for 280.6 grams of methamphetamine, 1,387.3 grams of cocaine, and 176 grams of marijuana, the judge sentenced Sanchez Cervantes to 235 months' imprisonment and a five-year term of supervised release.

Sanchez-Cervantes appealed his conviction and sentence, which we affirmed on April 26, 1996. He then filed a pro se § 2255 petition, which the district court denied. He appealed the denial of his § 2255 petition. We granted his motion to sever his appeal from the appeals of his co-defendants and remanded to the district court to consider his ineffective assistance of counsel claim, which was based on trial counsel's advice encouraging Sanchez-Cervantes to testify at trial. The district court granted Sanchez-Cervantes' application for a court-appointed attorney, and his new attorney moved to sever his petition from those of his co-defendants. The district court severed Sanchez-Cervantes' petition from those of his co-defendants and agreed to hear the ineffective assistance of counsel claim.

While the case was pending in the district court, the Supreme Court decided Apprendi. Sanchez-Cervantes sought to amend his petition, arguing that his sentence violated the ruling in Apprendi because the court did not submit the drug quantity determination to the jury to be found beyond a reasonable doubt. The district court allowed Sanchez-Cervantes to amend his petition in light of Apprendi.

On November 21, 2000, the district court held an evidentiary hearing on the ineffective assistance of counsel claim. After the hearing, the district court denied Sanchez-Cervantes' petition as to both the ineffective assistance of counsel claim and the Apprendi claim. The court ruled that Sanchez-Cervantes made a knowing and voluntary decision to testify and that his counsel's advice was based on a strategic decision that was not objectively unreasonable. The court also held that Apprendi cannot be applied retroactively to cases on initial collateral review. Sanchez-Cervantes filed this appeal. We review a district court's decision to deny a § 2255 petition de novo.2

II.

At the time of Sanchez-Cervantes' trial and sentencing, all of the circuits in the country allowed a judge to determine the drug quantity for which the defendant was responsible by a preponderance of the evidence. After Sanchez-Cervantes filed his § 2255 petition, the Supreme Court ruled in Apprendi 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."3 Sanchez-Cervantes asserts that by allowing the judge to determine the drug quantities, his sentence violated the rule established in Apprendi.4 Before we can reach the merits of his claim, we must determine whether the ruling in Apprendi applies retroactively to initial petitions for collateral review. We decide that it does not. Accordingly, we affirm the district court's decision and do not reach the merits of Sanchez-Cervantes' claim.

In Teague v. Lane,5 the Supreme Court held that new constitutional rules of criminal procedure that had not been announced at the time the defendant's conviction became final cannot be applied retroactively on collateral review unless they fit within one of two narrow exceptions.6 These exceptions exist if a new rule (1) "places certain kinds of primary private individual conduct beyond the power of the criminal law-making authority to proscribe," or (2) "requires the observance of those procedures that ... are implicit in the concept of ordered liberty."7 Thus, in order to apply the rule of Apprendi retroactively, we must determine that Apprendi is a new rule of criminal procedure that fits into one of Teague's exceptions.

Before we apply the Teague analysis, we must address Sanchez-Cervantes' argument that Teague does not apply in this case because Teague involved a state prisoner's § 2254 petition, not a federal prisoner's § 2255 petition. This argument fails. Although we have not specifically held that Teague applies to § 2255 petitions, we have applied it in that context.8 To clarify our position, we now hold, along with the Second, Fourth, Seventh, and Tenth Circuits, that Teague does apply to federal prisoners.9 The rule against retroactive application of new laws supports important interests of finality that pertain to both the federal system and the state system.10 It would seem inequitable to impose a federal/state dichotomy onto the Teague analysis and allow Teague's general rule against retroactivity to deny use of a new rule to state prisoners but allow such use to federal prisoners because Teague does not apply to them. The history of the Supreme Court's jurisprudence supports this conclusion. Teague was based on earlier opinions by Justice Harlan that considered the retroactivity analysis in the context of § 2255 petitions.11 Thus, the Teague retroactivity doctrine applies to both § 2254 and § 2255 habeas petitions.

In Jones v. Smith,12 we applied the Teague retroactivity analysis to the rule established in Apprendi.13 In Jones, an element of the crime was omitted from the state court information, but the jury instructions were proper as to all the elements.14 In analyzing whether Apprendi was retroactive, we held that Apprendi was a new rule of criminal procedure, thus satisfying the first step of the analysis.15 We went on to hold that the Apprendi rule, "at least as applied to the omission of certain necessary elements from the state court information," did not fit into either Teague exception.16 Thus, we declined to apply Apprendi retroactively. Because Jones limited its analysis and holding regarding the Teague exceptions to the facts of that case, it guides but does not control our decision here. We must now examine whether Apprendi, by requiring the jury to make drug quantity findings beyond a reasonable doubt, fits within one of Teague's exceptions. We hold that it does not.

[4] Pursuant to Teague's first exception, the first category of rules that will be applied retroactively include those placing "certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe."17 Apprendi neither decriminalized drug possession or drug conspiracies nor placed such conduct beyond the scope of the state's authority to proscribe. Thus, the first exception does not apply here. The other circuits that have addressed this exception in the context of the drug statutes agree.18 The matter rests on whether requiring the jury to make drug quantity determinations beyond a reasonable doubt falls within Teague's second exception.

The second exception in Teague allows courts to apply certain "watershed rules of criminal procedure" retroactively.19 These rules must be applied retroactively if a failure to adopt them will result in an impermissibly large risk that the innocent will be convicted and if the procedure at issue implicates the fundamental fairness of the trial.20 The Supreme Court has stated that to qualify under Teague, these rules must not only improve the accuracy of the trial but also alter our understanding of the "bedrock procedural elements essential to the fairness of a proceeding."21 The sweeping rule announced in Gideon v. Wainwright,22 that counsel shall be provided in all criminal trials for serious offenses, is the prototypical example of a watershed rule.23

We do not believe that requiring the jury to make drug quantity determinations beyond a reasonable doubt will greatly affect the accuracy of convictions. Nor is this rule a bedrock procedural element. Our view is consistent with the three circuits that have already ruled on this specific issue.24 In the cases facing those courts, as with the case before us, the existence of a drug violation was established beyond a reasonable doubt. The alleged Apprendi error only concerns an enhancement of the defendant's sentence based on a drug quantity finding by the judge. Therefore, the accuracy of the underlying conviction is not at issue. Moreover, most sentences will not be affected by Apprendi because they fall within the statutory maximum of twenty years allowed for convictions based on any measurable amount of drugs. If the jury convicted the defendant of a drug violation, even with no finding of a particular drug quantity, a sentence of twenty years or less would not violate Apprendi.25 Therefore, the rule established in Apprendi would apply only in a limited number of cases and is not the "sweeping rule" necessary to fall within Teague's second exception.26 Finally, the judge could still make a drug quantity finding and set the sentence without violating Apprendi as long as the sentence did not exceed twenty years for each count.27 Sending the drug quantity determination to the jury cannot be fundamental to the fairness of the proceeding if the judge is allowed to make such findings in some circumstances.

Our decisions that subjected Apprendi claims to harmless error analysis or plain error review lend additional support to our determination that Apprendi is not a bedrock procedural rule.28 In these cases, we did not consider Apprendi errors to be structural. A structural error is one that necessarily renders a trial fundamentally unfair and therefore invalidates the conviction.29 We only review for plain error or assess whether an error is harmless when the error is not structural; in those circumstances, the court must determine whether any substantial rights were prejudiced by the error.30 By applying harmless error analysis or plain error review to Apprendi claims, we have necessarily held that Apprendi errors do not render a trial fundamentally unfair.31 Therefore, it would seem illogical to hold that such an error is a watershed rule that "implicate[s] the fundamental fairness of the trial."32 In addition, the Supreme Court noted in Tyler v. Cain that not all structural-error rules fit into Teague's second exception.33 This implies that Teague's second exception is even narrower than the category of structural-error rules. From these holdings, it follows that the new Apprendi rule is not so fundamental as to fit within Teague's second exception.

Sanchez-Cervantes argues that Apprendi must be retroactive because the cases upon which it relies had previously been given retroactive effect by the Supreme Court. The reasoning in Apprendi stems primarily from In re Winship,35 where the Supreme Court held that a defendant cannot be convicted of a crime unless there is proof beyond a reasonable doubt of every element of the offense.36 The Supreme Court later held that the rule established in Winship must be given retroactive effect.37 Then, in Mullaney v. Wilbur,38 the Court extended Winship by holding that the state was required to prove the absence of provocation in a homicide beyond a reasonable doubt.39 This rule was also made retroactive.40 Sanchez-Cervantes submits that because Apprendi is based on the rule of "surpassing importance" established in Winship, and extends that rule like Mullaney did, it must be retroactive too.

Sanchez-Cervantes' argument is flawed because not every extension of Winship is necessarily a watershed rule of criminal procedure. The rules announced in Winship and Mullaney were given retroactive effect because they were to "overcome an aspect of the criminal trial that substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts[.]"41 The application of Apprendi only affects the enhancement of a defendant's sentence once he or she has already been convicted beyond a reasonable doubt. Therefore, it does not rise to the level of importance of Winship or Mullaney. Allowing the judge to determine the quantity of drugs for sentencing purposes does not impair the jury's ability to find the truth regarding whether the defendant possessed, distributed, or conspired to distribute some amount of drugs.42

Finally, our holding in Jones v. Smith43 that Apprendi did not fit into Teague's second exception lends itself to a similar holding in this case.44 Although the facts and circumstances for applying Apprendi differ between the cases, we should not apply Teague on a piecemeal basis. A new rule should be retroactive as to all cases or as to none to avoid inconsistencies and unnecessary litigation. The decision that Apprendi is not a watershed rule should not vary with each unique fact scenario but should hold constant because we are assessing whether the new rule is fundamental, not whether its application is fundamental in varying situations.45 For all of the reasons stated above, we hold that Apprendi does not apply retroactively to cases on initial collateral review.

III.

We now turn to Sanchez-Cervantes' claim of ineffective assistance of counsel. In assessing claims of ineffective assistance of counsel, we must follow the guidelines set forth in Strickland v. Washington.46 To establish that his counsel rendered ineffective assistance, the defendant must show that counsel's performance was deficient, and that the deficiency prejudiced the defendant.47 To be deficient, counsel's actions must be objectively unreasonable. Courts indulge in a strong presumption that conduct "falls within the wide range of reasonable pro-fessional assistance."48 To be prejudicial, a reasonable probability must exist that, but for counsel's conduct, the result of the trial would have been different.49 If either prong is not met, we must dismiss the claim.

Sanchez-Cervantes claims that his counsel's performance was deficient because he advised Sanchez-Cervantes to testify at trial and admit to illegally reentering the United States and conducting small-time drug deals. By putting Sanchez-Cervantes on the stand, his counsel also exposed Sanchez-Cervantes to cross-examination, which risked revealing past drug convictions and destroying his credibility. Sanchez-Cervantes argues that advising him to testify was a "desperation move" that relieved the Government of its burden of proof.

Sanchez-Cervantes cannot prevail on this claim because his counsel's performance was not deficient and did not result in any prejudice to Sanchez-Cervantes. As the district court noted, Sanchez-Cervantes voluntarily accepted his counsel's advice to testify. Sanchez-Cervantes admitted at the evidentiary hearing that he freely agreed to testify after his counsel explained the risks to him. Testimony from the evidentiary hearing also established that the advice was part of counsel's strategy to try to avoid a conviction for conspiracy. The Government had substantial evidence linking Sanchez-Cervantes to many drug deals as well as concrete evidence that Sanchez-Cervantes had illegally reentered the country. Counsel stated that he believed he could not win an acquittal on all the charges but that he could produce reasonable doubt on the conspiracy charge if Sanchez-Cervantes testified. He acknowledged that putting his client on the stand would open the door for admitting the prior convictions. He explained that the convictions would come in as evidence anyway, so putting Sanchez-Cervantes on the stand would not disadvantage him.

Advising Sanchez-Cervantes to testify was not objectively unreasonable in these circumstances. His counsel had a valid reason for doing so and proceeded to examine Sanchez-Cervantes with that objective in mind. Sanchez-Cervantes' testimony was consistent with his being a small-time, solo drug dealer who was not connected to the other defendants. It is not the role of the courts to second-guess an attorney's tactical decisions. Accordingly, we find that Sanchez-Cervantes' attorney's conduct was not deficient.

Although we need not determine prejudice after finding that counsel's performance was not deficient, we conclude that Sanchez-Cervantes' ineffectiveness claim would also fail the second prong of the Strickland test. Regardless of whether Sanchez-Cervantes testified, the Government produced ample evidence to convict him. The Government established that a large amount of cocaine was seized from a barn behind Sanchez Cervantes' house and several witnesses testified that they had purchased illegal drugs from Sanchez-Cervantes. While the fact that all the other defendants were convicted without testifying does not prove that such evidence would have convicted Sanchez-Cervantes, it makes a strong case that his testimony did not affect the outcome of the proceeding. Sanchez-Cervantes cannot establish that, but for the advice to testify, there is a reasonable probability that the result of the trial would have been different. Thus, there was no ineffective assistance of counsel.

CONCLUSION

We hold that Apprendi does not apply retroactively to cases on initial collateral review, and therefore, Sanchez-Cervantes' Apprendi claim is barred. His ineffective assistance of counsel claim fails because his counsel's performance was not deficient and did not prejudice Sanchez-Cervantes. Therefore, we affirm the district court's denial of Sanchez-Cervantes' § 2255 petition.

AFFIRMED.

HUG, Circuit Judge, concurring.

I concur in the opinion because I believe it is compelled by our en banc decision in United States v. Buckland, 277 F.3d 1173 (9th Cir. 2002) (en banc). However, were it not for the majority opinion in Buckland, I would see the case differently. Sanchez-Cervantes was indicted only for violating 21 U.S.C. § 841(a)(1) with no quantity of drugs specified. Thus, the sentence under § 841(b)(1)(C) was applicable. The quantity of drugs found by the judge exceeded the amount the jury could have found under the indictment and instructions to the jury for a violation of § 841(a)(1). In order for the jury to find these quantities the indictment would have to charge violations of § 841(b)(1)(A) or (B) with an appropriate instruction to the jury. As I contended in my concurring and dissenting opinion in Buckland,1 this would constitute charging and proving to the jury elements of separate crimes. In that circumstance, the Teague analysis would be quite different.

T.G. NELSON

UNITED STATES CIRCUIT JUDGE

Footnotes

1 530 U.S. 466, 120 S. Ct. 2348, 147 L. Ed. 2d 435 (2000).

2 United States v. Chacon-Palomares, 208 F.3d 1157, 1158 (9th Cir. 2000).

3 Apprendi, 530 U.S. at 490.

4 The maximum sentence for cases like this one, in which the jury does not find a specific quantity of drugs, is 240 months. 21 U.S.C. § 841(b)(1)(C). Nevertheless, Sanchez-Cervantes was sentenced to 235 months on the basis of the drug quantity found by the judge during sentencing.

5 489 U.S. 288, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989).

6 Id. at 310-11.

7 Id. at 307 (internal quotation marks and citation omitted).

8 See United States v. Judge, 944 F.2d 523, 525 (9th Cir. 1991) (denying claim in § 2255 petition because new Supreme Court rule did not fit within Teague's second exception and therefore was not retroactive on collateral review); United States v. Garcia, 210 F.3d 1058, 1059-60 (9th Cir. 2000) (applying general rule of Teague to determine timeliness of § 2255 petition).

9 Gilberti v. United States, 917 F.2d 92, 94-95 (2d Cir. 1990); United States v. Martinez, 139 F.3d 412, 416 (4th Cir. 1998); Van Daalwyk v. United States, 21 F.3d 179, 181-83 (7th Cir. 1994); Daniels v. United States, 254 F.3d 1180, 1193-94 (10th Cir. 2001). All of these cases cite concerns of finality and consistency, as well as the fact that Teague was based on Justice Harlan's approach, which was developed in the context of § 2255 petitions.

10 See Teague, 489 U.S. at 309 ("Application of constitutional rules not in existence at the time a conviction became final seriously undermines the principle of finality which is essential to the operation of our criminal justice system.").

11 See Mackey v. United States, 401 U.S. 667, 675 (1971) (Harlan, J., concurring in judgment in part and dissenting in part); Desist v. United States, 394 U.S. 244, 256 (1969) (Harlan, J., dissenting).

12 231 F.3d 1227 (9th Cir. 2000).

13 Id. at 1236-38. Jones was a § 2254 petition so it did not decide the issue addressed in the previous paragraph.

14 Id. at 1232.

15 Id. at 1236.

16 Id. at 1238 (holding that, regarding Teague's second exception, "the omission of particular key words from the written information neither increases the risk that an innocent person will be convicted nor hinders the fundamental fairness of the trial").

17 Teague, 489 U.S. at 307.

18 See United States v. Sanders, 247 F.3d 139, 148 (4th Cir. 2001) ("The first exception clearly does not apply here because Apprendi did not place drug conspiracies beyond the scope of the state's authority to proscribe."); United States v. Moss, 252 F.3d 993, 997 n. 3 (8th Cir. 2001) (noting that first exception is not relevant in this type of case); McCoy v. United States, 266 F.3d 1245, 1256-57 (11th Cir. 2001) (agreeing that the first exception is inapplicable to the rule announced in Apprendi).

19 Jones, 231 F.3d at 1237 (citing Teague, 489 U.S. at 312).

20 See id.

21 Sawyer v. Smith, 497 U.S. 227, 242 (1990) (internal citation, quotation marks, and emphasis omitted).

22 372 U.S. 335 (1963).

23 See O'Dell v. Netherland, 521 U.S. 151, 167 (1997). In Tyler v. Cain, 533 U.S. 656 (2001), the Supreme Court noted the extreme narrowness of Teague's second exception. It stated that "it is unlikely that any of these watershed rules has yet to emerge." Id. at 667 n. 7 (internal quotation marks and citation omitted). It also stated that not all new rules relating to due process or the fundamental requirements of due process alter our understanding of bedrock procedural elements. Id. In fact, the Supreme Court has not found any rule to qualify under the second exception since Teague came out. See Sanders, 247 F.3d at 148.

24 Sanders, 247 F.3d at 149-51; Moss, 252 F.3d at 999-1000; McCoy, 266 F.3d at 1258.

25 United States v. Buckland, 277 F.3d 1173, 1184 (9th Cir. 2002) (en banc) (holding that 21 U.S.C. § 841(a) was constitutional and stating that, under Apprendi, a jury conviction without a drug quantity finding exposes the defendant to a maximum term of twenty years for each count).

26 See O'Dell, 521 U.S. at 167 ("Unlike the sweeping rule of Gideon ... the narrow right of rebuttal that Simmons [v. South Carolina, 512 U.S. 154 (1994)] affords to defendants in a limited class of capital cases has hardly altered our understanding of the bedrock procedural elements essential to the fairness of a proceeding." (internal quotation marks, citation and emphasis omitted)). In addition, Buckland established that in cases involving multiple counts of drug violations, such as this one, the judge is required to impose consecutive sentences under U.S. Sentencing Guideline § 5G1.2(d). 277 F.3d at 1184-85. Therefore, many sentences that violate the Apprendi rule would be upheld because of mandatory stacking under the Sentencing Guidelines, further limiting the applicability of Apprendi.

27 See Buckland, 277 F.3d at 1185 ("Apprendi does not alter the authority of the judge to sentence within the statutory range provided by Congress.").

28 Id. at 1183-87 (upholding sentence under plain error review); United States v. Garcia-Guizar, 234 F.3d 483, 488-89 (9th Cir. 2000) (finding that Apprendi error was harmless); United States v. Antonakeas, 255 F.3d 714, 727 (9th Cir. 2001) (holding that, under plain error review, defendant's substantial rights were not affected); United States v. Saya, 247 F.3d 929, 942 (9th Cir. 2001) (subjecting Apprendi claim to plain error review).

29 See Arizona v. Fulminante, 499 U.S. 279, 309-10 (1991).

30 Jones v. United States, 527 U.S. 373, 389 (1999); Neder v. United States, 527 U.S. 1, 15 (1999).

31 Our conclusion that Apprendi errors do not create a fundamentally unfair trial is supported by the Supreme Court's decision in Neder. In that case, the Court held that, unlike a jury instruction that was defective as to all the elements of a crime, an instruction that omitted one element of the offense did not create a structural error. Neder, 527 U.S. at 15.

32 Teague, 489 U.S. at 312.

33 533 U.S. 656 (2001).

34 Id. at 667 n. 7.

35 397 U.S. 358 (1970).

36 Id. at 364.

37 Ivan V. v. City of New York, 407 U.S. 203, 205 (1972).

38 421 U.S. 684 (1975).

39 Id. at 704.

40 Hankerson v. North Carolina, 432 U.S. 233, 242-44 (1977).

41 Hankerson, 432 U.S. at 243.

42 Tyler v. Cain lends support to the conclusion that not all extensions of Winship are automatically retroactive. The Supreme Court held in Tyler that it had not yet made Cage v. Louisiana, 498 U.S. 39 (1990), which was a straightforward extension of Winship, retroactive to cases on collateral review. Tyler, 533 U.S. at 667.

43 231 F.3d 1227 (9th Cir. 2000).

44 Id. at 1238.

45 We express no opinion as to whether the Supreme Court's holding in Arizona v. Ring, 25 P.3d 1139 (Ariz. 2001), cert. granted, 534 U.S. 1103 (2002) (considering whether allowing a judge to impose a death sentence violates Apprendi), would apply retroactively if the petitioner in that case prevails.

46 466 U.S. 668 (1984).

47 Id. at 687.

48 Id. at 689.

49 Id. at 694.

1 See United States v. Buckland, 277 F.3d 1173 (9th Cir. 2002) (en banc) (Hug, J., concurring in part and dissenting in part).