Go to:	Part II	Case Menu	Briefs Menu

ARGUMENT
I. Section 2254(d) is a modified rule of res judicata.

A. Plain Meaning.

The place to begin, as Senator Biden noted in the debate, is at the top. See 141 Cong. Rec. S7842 (daily ed. June 7, 1995). "An application for a writ of habeas corpus on behalf of a person in custody pursuant to a judgment of a State court shall not be granted with respect to any claim adjudicated on the merits in State court proceedings unless . . . ." 28 U. S. C. §2254(d) (emphasis added).

This states the general rule. What follows the "unless" are exceptions to the general rule. This rule is a prohibition against granting relief to a party who has already litigated and lost the same claim. This is a rule of prior adjudication. It falls into the same family of rules as the doctrines of res judicata, law of the case, successive petitions, and Stone v. Powell, 428 U. S. 465 (1976). "The general principle in this language in the Hatch bill is that Federal courts shall not grant a claim that was adjudicated in State court proceedings. That is what is at the top." 141 Cong. Rec. S7842 (daily ed. June, 7, 1995) (statement of Sen. Biden); Scheidegger, Habeas Corpus, Relitigation, and the Legislative Power, 98 Colum. L. Rev. 888, 946 (1998).⁽²⁾

The attempts to pound the square peg of §2254(d) into the round hole of retroactivity typically focus on the exceptions and ignore the rule. For example, Liebman & Ryan, "Some Effectual Power": The Quantity and Quality of Decisionmaking Required of Article III Courts, 98 Colum. L. Rev. 696, 866-868 (1998) quote the main body of subsection (d) along with paragraph (1), but then never mention the general rule in their statutory analysis. See also Brief for Petitioner 26-39 (same); Brief for the American Civil Liberties Union as Amicus Curiae 4-13 (same) ("ACLU Brief"). These arguments are mistaken on their own terms. More fundamentally, though, it would be exceedingly odd for Congress to have expressed the basic nature of the rule in the exceptions rather than the main rule.

If Congress were going to codify or modify Teague and enact a rule of retroactivity, one would expect the topics of retroactivity, time, and choice of governing law to be placed front and center, i.e., something like this:

"(a) RETROACTIVE APPLICATION OF NEW RULES--

     .     .     .     .     .

" 'Sec. 2257. Law applicable

" 'In cases subject to this chapter, all claims shall be governed by the law as it was when the petitioner's conviction became final. A court considering a claim under this chapter shall consider intervening decisions by the Supreme Court of the United States which establish fundamental constitutional rights.'." S. 1657, 103d Cong., 1st Sess. §304(a) (1993).

One would further expect that when such a bill reached the floor, the Teague decision and other retroactivity cases would be prominent in the debate. That is what happened in the debate on S. 1657, supra, which really was a retroactivity provision. See 139 Cong. Rec. 29,444-29,449 (1993). That is not what happened in the debate over S. 735. Teague was not even mentioned on the floor. The same Senators (Hatch, Biden, and Specter) who debated Teague and retroactivity in considering S. 1657 debated the present measure at length without mentioning Teague. 141 Cong. Rec. S7831-S7849 (daily ed. June 7, 1995). The best the other side can point to is a few ambiguous impromptu responses by witnesses to a question in a committee hearing. See ACLU Brief 27, n. 47. A weaker example of legislative history is difficult to imagine.

The word actually used on the floor was not "retroactivity," but "deference." "As one commentator accurately recounts, in both houses of Congress section 2254(d) 'was called a "deference" standard by every member who spoke on the question, opponents as well as supporters.' " Matteo v. Superintendent, 171 F. 3d 877, 890 (CA3 1999) (en banc). The word "deference" is important, because that is the kind of rule that the concurring Justices in West emphatically declared that Teague was not. Wright v. West, supra, 505 U. S., at 304 (O'Connor, J., concurring in the judgment); id., at 307 (Kennedy, J., concurring in the judgment). Both the supporters and opponents understood that this bill took the step that the Court considered, but did not take, in West. In light of the case law under the prior statute, and Congress's arguable ratification of it in 1966, see id., at 294-295 (plurality opinion), the step was properly one for Congress rather than the Court to take. See id., at 305-306 (O'Connor, J., concurring in the judgment); see also Patterson v. McLean Credit Union, 491 U. S. 164, 172-173 (1989) (statutory interpretation precedents have special weight as stare decisis, precisely because Congress can abrogate them); Scheidegger, supra, 98 Colum. L. Rev., at 888-889.

The language of subsection (d) closely tracks the language of subsection (b), the exhaustion rule. That is because they are similar rules. Subsection (a) states the exclusive ground for granting federal habeas relief to a state convict: "that he is in custody in violation of the Constitution or laws or treaties of the United States." Subsections (b) and (d) state reasons for not reaching the underlying claim, based on what did or did not happen in state court. Section (b) forbids relief if the prisoner has not fairly presented his claims to state courts, with certain exceptions. See Castille v. Peoples, 489 U. S. 346, 351 (1989). Subsection (d) forbids relief if the state courts have decided the claims against the prisoner, with certain exceptions.

In both subsections, the exceptions deal with gross deficiencies in the state process. If state process is absent or ineffective, it need not be exhausted. 28 U. S. C. §2254(b)(1)(B). If a state court fails to apply a clearly established Supreme Court precedent or applies it unreasonably, its judgment does not bar relief. §2254(d)(1); Neelley v. Nagle, 138 F. 3d 917, 924 (CA11 1998). But if the state provides effective process, the state court recognizes the correct rule of law, and the state court reasonably applies the rule to the facts, then that decision stands, subject only to review by higher state courts and by this Court on certiorari. This rule is the natural result of principles recognized by this Court. "Direct review is the principal avenue for challenging a conviction . . . . 'The role of federal habeas proceedings, while important . . . , is secondary and limited.' " Brecht v. Abrahamson, 507 U. S. 619, 633 (1993) (quoting Barefoot v. Estelle, 463 U. S. 880, 887 (1983)).

"We may mystify any thing. But if we take a plain view of the words of the [statute], and give to them a fair and obvious interpretation, we cannot fail in most cases of coming to a clear understanding of its meaning. We shall not have far to seek. We shall find it on the surface, and not in the profound depths of speculation." Ex parte Siebold, 100 U. S. 371, 393 (1880).

The nature of this rule is plain and found on the surface. It forbids granting a claim previously rejected by a coordinate court, unless an exception applies. It is a rule of prior adjudication, not a rule of retroactivity or choice of law.

B. Tradition--Generally.

The traditional general rule is res judicata. State court judgments have res judicata effect in federal court. That has been the rule, if not from day one, at least from year two. The first Congress, in its second session, enacted the Full Faith and Credit Act. See Baker v. General Motors Corp., 522 U. S. 222, 232, n. 4 (1998); Act of May 26, 1790, ch. 11, 1 Stat. 122; 28 U. S. C. §1738; Scheidegger, supra, 98 Colum. L. Rev., at 912. This statute applies to federal courts as much as state courts. Mills v. Duryee, 7 Cranch (11 U. S.) 481, 485 (1813). It applies to federal questions as well as state-law questions, and it does not permit reexamination of the merits. "That the adjudication of federal questions by the [state court] may have been erroneous is immaterial for purposes of res judicata." Angel v. Bullington, 330 U. S. 183, 187 (1947); see also Allen v. McCurry, 449 U. S. 90, 103 (1980) (no right to relitigate in federal court).

In cases where res judicata does not apply, relitigation of an issue may be barred by the "law of the case" doctrine. "Under this doctrine, a court should not reopen issues decided in earlier stages of the same litigation." Agostini v. Felton, 521 U. S. 203, 236 (1997). This "doctrine applies as much to the decisions of a coordinate court in the same case as to a court's own decisions." Christianson v. Colt Industries Operating Corp., 486 U. S. 800, 816 (1988); Scheidegger, supra, 98 Colum. L. Rev., at 914, and n. 158.

The doctrine is more flexible than res judicata, however, Arizona v. California, 460 U. S. 605, 618 (1983), and it does not stop a court from reconsidering an issue if the prior decision "is clearly erroneous and would work a manifest injustice." Id., at 618, n. 8 (emphasis added); Agostini, supra, 521 U. S., at 236. The term "clearly" is important here. Upon finding the decision of the first court to be plausible, the inquiry of the coordinate court ends. Christianson, supra, 486 U. S., at 819.

Unlike res judicata, law of the case depends on the relative status of the courts in question. A higher court hearing an appeal is not bound by a decision of a lower court. See id., at 817. Conversely, a lower court is not entitled to declare the higher court decision clearly erroneous and disregard it. See Vendo Co. v. Lektro-Vend Corp., 434 U. S. 425, 427-428 (1978).

On one point, amicus ACLU is exactly right: "Inferior federal courts and state courts are co-equals; both answer only to this Court." ACLU Brief 4; see also Scheidegger, supra, 98 Colum. L. Rev., at 898-899. The problem is that since Brown v. Allen, supra, 344 U. S. 443 (1953), they have not been treated as co-equals. Congress has acted to correct that anomaly, based squarely on the premise that "[s]tate courts, in many respects, are just as good, if not better, than the Federal courts--in these areas, just as good." 141 Cong. Rec. S7846 (daily ed. June 7, 1995) (statement of Sen. Hatch).

Congress did not exactly adopt the coordinate-court law-of-the-case doctrine in §2254(d). It crafted a new rule. Yet the rule is closely analogous to law of the case, and it proceeds on a similar premise. In essence, Congress has told the federal district and circuit courts to stop treating the state courts as lower courts and start treating them as coordinate courts.

C. Tradition--Habeas.

The subsequent history has been discussed many times and need not be repeated here. Suffice it to say that before Brown v. Allen, supra, this Court had developed two discretionary rules of prior adjudication. First, contrary to the common law rule, a judge presented with a successive petition could, but need not, consider the prior denial, even to the point of giving it controlling weight. Salinger v. Loisel, 265 U. S. 224, 231 (1924); see McCleskey, supra, 499 U. S., at 481-482.

Second, Darr v. Burford, 339 U. S. 200, 214 (1950), overruled on other grounds, Fay v. Noia, supra, 372 U. S., at 435-436, acknowledged that the adjudication on direct review would not be res judicata. However, Darr held unequivocally that following state review and denial of certiorari, the habeas court "may decline to examine further into the merits because they have already been decided against the petitioner. [Footnote citing Salinger.] Thus there is avoided . . . repeated adjudications of the same issues by courts of coordinate powers." Id., at 215 (emphasis added).

Words can hardly be more clear. Darr was a rule of prior adjudication, albeit a flexible, discretionary one. The persistent myth that denial of habeas relitigation in this era was based on a substantive theory of due process rather than a rule of prior adjudication, see, e.g., Peller, In Defense of Federal Habeas Corpus Relitigation, 16 Harv. C.R.-C.L. L. Rev. 579, 663 (1982), is just that--a myth.

The Salinger rule of successive petitions evolved from one of amorphous discretion to the very lax rule of Sanders v. United States, 373 U. S. 1, 17-18 (1963), to the more structured rule of McCleskey, supra, 499 U. S., at 493, and finally into the codified rule of 28 U. S. C. §2244(b). The effect of the prior state adjudication of the claim has followed a similar, though not parallel, evolution. Brown, supra, effectively reduced the state decision to a mere precedent from another jurisdiction, albeit a particularly pertinent one. See Wright v. West, 505 U. S. 277, 305 (1992) (O'Connor, J., concurring in the judgment). In §2254(d), Congress has done with Darr just what McCleskey and §2244(b) did with Salinger; it has restored the spirit of the pre-1953 discretionary rule but replaced it with a structured and more tightly limited rule.

Throughout American history, the availability of federal habeas for state prisoners has varied as a function of confidence in state courts relative to federal courts. It was forbidden in the beginning, when federal courts were a feared innovation. See Scheidegger, supra, 98 Colum. L. Rev., at 932. It was expansive during Reconstruction, see Forsythe, The Historical Origins of Broad Federal Habeas Review Reconsidered, 70 Notre Dame L. Rev. 1079, 1117-1118 (1995), retracted in the late nineteenth century, see, e.g., In re Wood, 140 U. S. 278, 285-286 (1891), expanded again in the 1950s and 1960s during the civil rights struggle, see Brown, supra; Fay v. Noia, supra, and retracted again in the last quarter of this century in Stone v. Powell, 428 U. S. 465 (1976), Teague, supra, Coleman, supra, and other cases.

The enactment of a qualified prior adjudication bar is the next logical step in this evolution. It takes us back to the principle that relitigation is the exception and finality of judgments is the rule. See Ex parte Hawk, 321 U. S. 114, 118 (1944) (per curiam) ("not ordinarily re-examine"). It does so based on the recognition that state courts today are far different from what they were in 1953 or 1963, that they warrant more confidence, and that their judgments deserve more respect.

D. Teague Survives.

Breard v. Greene, 523 U. S. 371, 377 (1998) (per curiam) sets that claim to rest. Teague remains as an independent limitation on the scope of habeas relief. Where the state court has ruled on the merits, Teague will overlap with §2254(d)(1). There are, however, a number of situations when Teague will continue to have independent force.

In Breard, petitioner claimed novelty as cause for his default. Ibid.; cf. Reed v. Ross, 468 U. S. 1, 14-15 (1984). Teague effectively shuts down this route, unless one of its exceedingly narrow exceptions applies. Teague also has independent force when the federal court has decided, rightly or wrongly, that a state's procedural default rule is not adequate and independent. Claims defaulted on appeal in the nation's largest state are presently in this status. See Morales v. Calderon, 85 F. 3d 1387, 1393 (CA9 1996), cert. denied sub nom. Calderon v. Morales, 519 U. S. 1001 (1996). The class of claims to which Teague applies but §2254(d)(1) does not remains significant.

Congress could not possibly have intended to withdraw Teague's protection from this entire class of claims. That would be contrary to the entire thrust and purpose of the bill. The judicially created limitation of retroactivity and the legislatively created limitation of prior adjudication stand as independent grounds for decision. In many cases both apply, but each reaches claims that the other does not.

	Beginning of this file	Part II	Case Menu

2. This article is cited throughout this brief out of necessity, not hubris. To stay within the page limit, the discussion is thin in spots, referring to the article for a more complete discussion.    
 
August 1999