Go to:	Part III	Case Menu	Briefs Menu

II. The Courts of Appeals have unanimously rejected the thesis that review of mixed questions is unchanged.

Now petitioner, with the support of the ACLU and others, would have this Court believe that the long, bitter debate over this Act was much ado about not very much, that supporters and opponents were both mistaken, and that the Act does little more than tinker at the edges of Teague v. Lane, 489 U. S. 288 (1989). See Brief for Petitioner 32-33; ACLU Brief 2; Liebman & Ryan, supra, 98 Colum. L. Rev., at 867-873.

The most important area of dispute involves the "mixed question" situation, where the applicable rule is clearly established but general in its terms, and the dispute involves the application of that rule to specific facts. Petitioner claims that the AEDPA left this area unchanged, and he claims five circuits in support. Brief for Petitioner 39, n. 20. Actually, this thesis has been unanimously rejected by the circuits which have considered it.

As petitioner acknowledges, the Fourth, Fifth, Seventh, and Eleventh Circuits are contrary to his position. See Brief for Petitioner 40; Drinkard v. Johnson, 97 F. 3d 751, 767-768 (CA5 1996) ("unreasonable application" clause applies to mixed questions of law and fact); Lindh v. Murphy, 96 F. 3d 856, 870 (CA7 1996), rev'd on other grounds, 521 U. S. 320 (1997) (same); Neelley v. Nagle, 138 F. 3d 917, 924 (CA11 1998) (following Drinkard); Green v. French, 143 F. 3d 865, 870 (CA4 1998) (largely the same, with different analysis for factually indistinguishable cases).

Other circuits have taken different analytical approaches. However, when we look carefully at how those approaches are applied to decide actual cases, we see that the differences are more apparent than real. Most importantly, all the circuits with a defined position agree that when the state court applies the correct general rule to specific facts, its decision stands if it is reasonable, and mere disagreement is insufficient for collateral attack.

Petitioner relies on Ayala v. Speckard, 89 F. 3d 91 (CA2 1996) for the Second Circuit's position in favor of "plenary" relitigation of such claims. However, the Second Circuit subsequently granted rehearing en banc. The en banc court rendered its decision after Lindh v. Murphy, 521 U. S. 320 (1997), and hence did not need to discuss the AEDPA. See Ayala v. Speckard, 131 F. 3d 62 (CA2 1997). The panel decision in Ayala therefore has no value as precedent, and the Second Circuit has not since taken a definite stand. See Smalls v. Batista, No. 98-2526, text accompanying n. 5 (CA2 July 30, 1999) (noting but not taking position in circuit split).

The first court of appeals case to expressly disagree with Drinkard and Lindh was the First Circuit opinion in O'Brien v. Dubois, 145 F. 3d 16 (CA1 1998). O'Brien held that "the habeas court asks whether the Supreme Court has prescribed a rule that governs the petitioner's claim." Id., at 24. If so, the court proceeds with the "contrary to" prong and otherwise goes to the "unreasonable application" prong. Ibid.

Despite its superficial similarity to the Liebman approach, in actual application the First Circuit is closer to Drinkard and Lindh. Its "contrary to" analysis is limited to specific rules, id., at 25, and if "no Supreme Court precedent is dispositive of a petitioner's claim," ibid., the First Circuit proceeds to "unreasonable application." "This reduces to a question of whether the state court's derivation of a case-specific rule from the [Supreme] Court's generally relevant jurisprudence appears objectively reasonable." Ibid.

Applying the standard to O'Brien's Confrontation Clause claim of restricted recross-examination, the court found the general rule of effective cross-examination insufficiently specific for the "contrary to" branch. The state court's resolution was plausible, and that was sufficient to deny relief under §2254(d)(1). Id., at 27.

Bui v. DiPaolo, 170 F. 3d 232 (CA1 1999) eliminates any doubt that the First Circuit does not support petitioner's thesis. On Bui's Confrontation Clause claim, the state and federal courts agreed that the relevant legal principles reduced the question to whether Bui had made a sufficient foundational showing of witness bias to entitle him to cross-examination on the point. The question of whether evidence is sufficient to meet a particular legal standard is the quintessential "mixed question," the kind involved in Wright v. West, 505 U. S. 277 (1992). Bui examined the state court's determination for reasonableness. 170 F. 3d, at 244. The court expressly rejected the notion that the AEDPA standard was "one of plenary review," holding that such a transformation of the standard would amount to defiance of Congress's will. Id., at 243.

"When, as now, a petitioner can show only that rational minds might differ over how to apply certain general constitutional principles to the specific circumstances of his case, the current habeas corpus standard of review does not allow a federal court to invalidate a state conviction." Id., at 246 (emphasis added). The First Circuit has thus fully rejected the notion that review of mixed questions remains "plenary."

Most recently, the First Circuit applied its O'Brien approach to an ineffective assistance claim in Vieux v. Pepe, No. 98-1864 (CA1 July 19, 1999). The court held that "the petitioner may succeed under the 'contrary to' clause only if Supreme Court caselaw directly governs the claim and the state court got it wrong." Id., part III, para. 6. The federal court noted that the state court "conducted a proper Strickland analysis." Id., part IV, para. 2. That is, it correctly recognized the governing rule. See ibid. In this case, unlike the typical Strickland case, the assessment of the attorney's performance involved a question of "pure" law regarding the federal wiretapping statute. With no Supreme Court precedent on the latter point, Vieux evaluated the state decision under "unreasonable application." Id., part IV, paras. 9-10. In a footnote, the court reiterates that nearly all Strickland claims will pass the "contrary to" stage. Id., n. 2. Hence, the main question in such cases will be the "unreasonable application" analysis.

The circuit position generally most favorable to habeas petitioners is that of the Third Circuit in Matteo v. Superintendent, 171 F. 3d 877 (CA3 1999) (en banc). Yet even the Third Circuit is closer to the state's interpretation than it is to the petitioner's.

Matteo largely follows O'Brien, supra. It asks, under the "contrary to" prong, whether "applicable Supreme Court precedent . . . resolves the petitioner's claim." Id., at 888 (emphasis added). Still following the First Circuit, Matteo rejects the idea that " 'a general standard that covers the claim' " is sufficient for "contrary to" review. Ibid. (quoting O'Brien, 145 F. 3d, at 24). West-type claims, applying general standards to specific facts, are thus excluded from "contrary to" review and move over to "unreasonable application."

The court quotes Senator Specter in the floor debate saying, " 'Unless it is unreasonable, a State court's decision applying the law to the facts will be upheld.' " Id., at 890 (emphasis added); see 142 Cong. Rec. 7799 (1996). Matteo also specifically rejects the notion that "contrary to" review is "plenary" as petitioner uses that term. "This standard precludes granting habeas relief solely on the basis of simple disagreement with a reasonable state court interpretation . . . ." Matteo, 171 F. 3d, at 888. For claims evaluated under "unreasonable application," which will be most of them, Matteo reaffirms that mere disagreement is not enough. "To hold otherwise would resemble de novo review, which we believe is proscribed by the statute." Id., at 889 (emphasis added).

For its standard, Matteo settles on this: "whether the state court decision, evaluated objectively and on the merits, resulted in an outcome that cannot reasonably be justified." Id., at 891. Matteo says its standard will result in the granting of more petitions than the First Circuit's "outside the universe of plausible, credible outcomes." Id., at 889. Amicus believes that the reality is similar to the situation which existed in ineffective assistance cases before Strickland v. Washington, 466 U. S. 668, 696-697 (1984), when the differing formulations of the performance standard rarely changed the outcome of the case.

The Matteo court's approach to the issue in the case illustrates the extent of the change wrought by the AEDPA. Matteo's claim involved a phone call he made from jail, which was taped by police with the consent of the other party, Lubking. 171 F. 3d, at 882-884. The elements of Matteo's Sixth Amendment claim were established in Supreme Court precedent: "(1) the right to counsel must have attached at the time of the alleged infringement; (2) the informant must have been acting as a 'government agent'; and (3) the informant must have engaged in 'deliberate elicitation' of incriminating information from the defendant." Id., at 892 (citing Maine v. Moulton, 474 U. S. 159, 170-171 (1985) and United States v. Henry, 447 U. S. 264, 269-270 (1980)); see also Kuhlmann v. Wilson, 477 U. S. 436, 459 (1986).

If the Third Circuit followed petitioner's view of the statute, it would have conducted "plenary 'contrary to' review," because "Supreme Court law had specifically designed a rule for the claim when the state court acted . . . ." Brief for Petitioner 38. There is no "absence of prescribed law" here, cf. id., at 39, and hence, according to petitioner, no basis for "unreasonable application" review.

The Matteo court did just the opposite. It declared that the rule quoted above was not specific enough "to merit 'contrary to' review" and proceeded to "unreasonable application." 171 F. 3d, at 893. Yet the rule in question is surely no less specific than the rules in the pre-AEDPA "mixed question" cases where de novo review was applied. See Strickland, supra, 466 U. S., at 698 (ineffective assistance of counsel); Wright v. West, supra, 505 U. S., at 295 (lead opinion) (sufficiency of the evidence; de novo standard assumed and applied); Thompson v. Keohane, 516 U. S. 99, 102 (1995) (Miranda, question of when suspect is in "custody"). Indeed, the issue before the court in Matteo was governed by a rule more specific than the rules in Strickland and West, and comparable to the one in Thompson. If specificity is the criterion for choosing between "contrary to" and "unreasonable application," as the First and Third Circuits hold, and if the rule at issue in Matteo falls on the nonspecific side of the line, then all of the recent de novo/mixed question cases would also fall on that side.

Petitioner cites Canales v. Roe, 151 F. 3d 1226, 1229, n. 2 (CA9 1998) for the proposition that the Ninth Circuit requires plenary review of Strickland claims. Brief for Petitioner 39, n. 20. That decision merely states that the Strickland test is "clearly established," a proposition never in dispute.⁽³⁾ A far better example of the Ninth Circuit position on this issue is Furman v. Wood, 169 F. 3d 1230 (CA9 1999). Furman reaffirmed the Ninth Circuit's acceptance of the Drinkard/Lindh analysis, i.e., that mixed questions are governed by the "unreasonable application" prong. Id., at 1232. Applying the statute to the state court's decision on the Strickland claim, Furman reviewed the state court's holdings on both performance and prejudice and found them reasonable. Id., at 1235. That is sufficient to bar relief under the statute. Ibid.

In the Tenth Circuit, petitioner relies on Miller v. Champion, 161 F. 3d 1249, 1253 (CA10 1998). Brief for Petitioner 39, n. 20. Miller is a curious opinion. It quotes the new standard, 161 F. 3d, at 1253, but never applies it. For the proposition that ineffective assistance claims are mixed questions and hence reviewed de novo, Miller simply cites Parker v. Champion, 148 F. 3d 1219, 1221 (CA10 1998), which does not mention the AEDPA at all. 161 F. 3d, at 1254. There is no discussion of §2254(d)(1) in either Parker or Miller and no attempt to distinguish or criticize decisions from other circuits holding that the statute had changed the rule. The actual holding of Miller was that the petitioner had tried to develop the facts in state court, was denied the opportunity to do so, and hence was entitled to an evidentiary hearing in federal court. Id., at 1253, 1259. Given that circumstance, §2254(d)(1) is only tangential to the case, so the lack of discussion may be understandable.

Other Tenth Circuit cases before and since Miller have addressed the statute, and they paint a different picture. Houchin v. Zavaras, 107 F. 3d 1465, 1470 (CA10 1997) declared that the "AEDPA increases the deference to be paid by the federal courts to the state court's factual findings and legal determinations." LaFevers v. Gibson, No. 98-6302 (CA10 June 16, 1999) interprets §2254(d)(1) as allowing an exception to claim preclusion only "if: (1) the state court decision is in square conflict with Supreme Court precedent which is controlling on law and fact or (2) if its decision rests upon an objectively unreasonable application of Supreme Court precedent to new facts." Id., part III, para. 2 (emphasis added). Manifestly, Wright v. West-type cases applying established standards to case-specific facts would fall under the second prong.

The Sixth Circuit has declined to take sides while noting, consistently with the position in this brief, that there is less to the circuit split than meets the eye. In Nevers v. Killinger, 169 F. 3d 352, 357-362 (CA6 1999), the court reviewed the positions of the other circuits. Nevers then held that fact-intensive, case-specific applications of established rules fall under "unreasonable application" under either approach. Id., at 360-361.

Finally, in Long v. Humphrey, No. 98-3409 (CA8 July 14, 1999), the Eighth Circuit took an approach similar to Nevers. The "manifest necessity" standard for ordering a mistrial over the defendant's objection is a fact-intense, case-specific mixed question, and it falls under the "unreasonable application" prong under either Drinkard or O'Brien. Id., para 7.

In summary, then, ten of the eleven numbered circuits have ruled on the question of how to evaluate, under §2254(d), claims involving clearly established but general rules as applied to case-specific facts, when no Supreme Court case is factually indistinguishable. The circuits have unanimously concluded that such cases fall under the "unreasonable application" prong.

	Beginning of this file	Part III	Case Menu

3. The actual holding of Canales is that petitioner's claim, a variation on the claim presently before this Court in Roe v. Ortega, No. 98-1441, was not clearly established. 151 F. 3d, at 1231. Hence there was no need to go into a detailed examination of how to apply the statute to a straight Strickland claim.    
 
August 1999