holding that § 2254(d) cannot be waived by the partiesSummary of this case from Canales v. Davis
Richard John Bourke, Louisiana Capital Assistance Center, New Orleans, LA, for Petitioner-Appellant. Cynthia S. Killingsworth, Carla Sue Sigler, Karen C. McLellan, District Attorney's Office for the Parish of Calcasieu, Lake Charles, LA, Elizabeth Baker Murrill, Esq., Assistant Attorney General, Colin Andrew Clark, Esq., Assistant Attorney General, Office of the Attorney General for the State of Louisiana, Andrea Barient, Louisiana Department of Justice, Baton Rouge, LA, Jeffrey Matthew Harris, Consovoy McCarthy Park, P.L.L.C., Arlington, VA, for Respondent-Appellee. Kyle Douglas Hawkins, Bill L. Davis, Assistant Attorney General, Office of the Attorney General, Office of the Solicitor General, Matthew Hamilton Frederick, Deputy Solicitor General, Office of the Solicitor General for the State of Texas, Austin, TX, for Amicus Curiae State of Texas.
Richard John Bourke, Louisiana Capital Assistance Center, New Orleans, LA, for Petitioner-Appellant.
Cynthia S. Killingsworth, Carla Sue Sigler, Karen C. McLellan, District Attorney's Office for the Parish of Calcasieu, Lake Charles, LA, Elizabeth Baker Murrill, Esq., Assistant Attorney General, Colin Andrew Clark, Esq., Assistant Attorney General, Office of the Attorney General for the State of Louisiana, Andrea Barient, Louisiana Department of Justice, Baton Rouge, LA, Jeffrey Matthew Harris, Consovoy McCarthy Park, P.L.L.C., Arlington, VA, for Respondent-Appellee.
Kyle Douglas Hawkins, Bill L. Davis, Assistant Attorney General, Office of the Attorney General, Office of the Solicitor General, Matthew Hamilton Frederick, Deputy Solicitor General, Office of the Solicitor General for the State of Texas, Austin, TX, for Amicus Curiae State of Texas.
Before STEWART, Chief Judge, JONES, SMITH, WIENER, DENNIS, OWEN, ELROD, SOUTHWICK, HAYNES, GRAVES, HIGGINSON, COSTA, WILLETT, HO, DUNCAN, ENGELHARDT, and OLDHAM, Circuit Judges.
ANDREW S. OLDHAM, Circuit Judge, joined by JONES, SMITH, OWEN, SOUTHWICK, WILLETT, HO, DUNCAN, and ENGELHARDT, Circuit Judges:
A Louisiana jury convicted Ricky Langley of second-degree murder. The state court overturned that conviction on direct appeal. So the State retried Langley and re-convicted him. Langley now seeks federal habeas relief. He argues his prior conviction should be construed as an implicit acquittal that bars the re-conviction and allows him to walk free. We disagree.
While on parole for a prior child-molestation conviction, Ricky Langley choked a six-year-old boy into unconsciousness and then, to ensure the child was dead, strangled him with a ligature and shoved a sock into the child’s mouth. Langley stuffed the boy’s corpse in a bedroom closet and lied to the child’s mother when she came looking for her son. Langley then waived his Miranda rights and repeatedly confessed on video to molesting and killing the boy. Police found the child’s body, wearing a t-shirt soaked in Langley’s semen, in the closet where Langley left him.
The State of Louisiana thrice tried and thrice convicted Langley for his heinous crime. The second and third trials lie at the heart of this case. But we explain all three for the sake of completeness.
Langley I. A Louisiana jury unanimously convicted Langley of first-degree murder and sentenced him to death. For reasons unrelated to this case, Langley’s first conviction was remanded on direct appeal in state court. See State v. Langley (Langley I ), 711 So. 2d 651, 675 (La. 1998) (per curiam) (granting rehearing in part and remanding); see also State v. Langley , 813 So. 2d 356, 358 (La. 2002) (quashing the indictment due to improper selection of the grand jury foreperson). So the State retried him for murder.
Langley II . At the second trial, the jury unanimously convicted Langley of murder once again. This time, however, the jury issued a verdict of second-degree murder. For reasons again unrelated to the appeal before us today, the second jury’s verdict was also overturned on direct appeal in state court. See State v. Langley (Langley II ), 896 So. 2d 200, 201 (La. Ct. App. 2004). So the State again retried Langley for murder.
Langley III. Before the third trial, however, the Louisiana Supreme Court held the second-degree murder conviction precluded the State from retrying Langley for first-degree murder. See State v. Langley (Langley III ), 958 So. 2d 1160, 1170 (La. 2007). The court based its holding on state law. Ibid. (citing LA. CONST . art. I, § 17 (A); LA. STAT. ANN . § 14:30.1 ; LA. CODE CRIM. PROC. ANN . arts. 598(A), 782(A), 841(A) ). But its holding accords with longstanding double jeopardy law because, "[h]istorically, courts have treated greater and lesser-included offenses as the same offense for double jeopardy purposes, so a conviction on one normally precludes a later trial on the other." Currier v. Virginia , ––– U.S. ––––, 138 S. Ct. 2144, 2150, 201 L.Ed.2d 650 (2018).
Therefore, at Langley’s third trial, the State charged him only with second-degree murder. Having lost before two juries, Langley decided to try his luck with a bench trial the third time around. Given the facts and his repeated videotaped confessions, however, the trial judge convicted him of second-degree murder. The court found as a matter of fact that Langley had specific intent to kill because, after their "sexual encounter," Langley thought death would "do this little boy a favor." The court again sentenced Langley to life in prison.
Langley again appealed. This time he argued the Double Jeopardy Clause should have prohibited the State from retrying him for second-degree specific-intent murder. That result is compelled, Langley said, by Ashe v. Swenson , 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970). Ashe identified a "collateral estoppel" "ingredient" in the Double Jeopardy Clause and held it precludes a retrial for any issue necessarily determined by a jury’s general verdict of acquittal. See id. at 442–45, 90 S.Ct. 1189. Of course, Langley was not acquitted of second-degree murder in Langley II ; he was convicted . Langley nonetheless argued Ashe should be extended to his facts. Langley reasoned the jury—which simply adjudged him "GUILTY," without specifying why—logically must have based its verdict on second-degree felony murder. If so, Langley hypothesized, the Langley II jury could’ve determined he lacked specific intent. And if all these hypotheses and deductions are true, Langley concluded, the State should be barred from retrying him for any offense that has specific intent as an element—including second-degree specific-intent murder.
The state courts rejected Langley’s effort to extend Ashe . See State v. Langley (Langley IV ), 61 So. 3d 747, 756–58 (La. Ct. App. 2011), cert. denied , 78 So. 3d 139 (La. 2012). The state appellate court first evaluated the record "to discern which facts were ‘necessarily determined’ " by the jury’s guilty verdict in Langley II . 61 So. 3d at 757. The only way to determine what the jury actually and necessarily determined is to evaluate what the jury actually and necessarily did—namely, convict Langley of second-degree murder. Although the state court recognized it was "possible that the jury verdict was based on a jury finding under the felony-murder rule," the court noted it was equally likely the jury based its verdict on second-degree specific-intent murder as an alternative to first-degree murder. Ibid. It was also possible the jury convicted Langley of second-degree murder as a "compromise verdict"—that is, a verdict that did not reflect the jury’s actual findings, but instead represented a compromise punishment of life in prison that was palatable to all jurors. Ibid . Because the jury could have reached its second-degree murder conviction without necessarily finding Langley lacked specific intent to kill, the Louisiana court held Langley "ha[d] not carried his burden of proving that the element of specific intent was actually decided [in his favor] in the previous trial" to preclude the relitigation of that issue in the third trial. Id. at 758.
Langley filed a federal habeas petition. The district court denied it. See Langley v. Prince , No. 2:13-cv-2780, 2016 WL 1383466, at *1 (W.D. La. Apr. 6, 2016). A panel of our Court, however, reversed and concluded not only that the state court’s opinion was wrong, but that it was "objectively unreasonable." Langley v. Prince , 890 F.3d 504, 521–23 (5th Cir. 2018). That decision would’ve allowed Langley to walk free. But we vacated it upon granting rehearing en banc .
This case implicates constitutional law, the equitable doctrine of estoppel, and statutory text. We address each in turn. We first explain the common-law and constitutional background of the Double Jeopardy Clause. Then we explain how Ashe and collateral estoppel fit into that background. Lastly, we explain how our application of Ashe is affected by the Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA"), Pub. L. No. 104-132, 110 Stat. 1214.
The Double Jeopardy Clause originates in the common-law plea autrefois acquit , meaning "prior acquittal," and the related plea autrefois convict . As Sir Edward Coke described it, "the maxim of the common law is, that the life of a man shall not be twice ... put in jeopardy for one and the same offence, and that is the reason and cause that auterfoits acquitted or convicted of the same offence is a good plea." Vaux’s Case (1591), 76 Eng. Rep. 992, 993; 4 Co. Rep. 44a, 45a (K.B.). But as far back as Vaux’s Case , the plea of prior acquittal was not always a get-out-of-jail-free card. Only some verdicts of acquittal in the first trial would effectively bar a second. See ibid. (discussing some qualifications to the plea); EDWARD COKE, THE THIRD PART OF THE INSTITUTES OF THE LAWS OF ENGLAND 214 (1st ed. 1644) (same); 2 MATTHEW HALE, HISTORIA PLACITORUM CORONÆ 393–95 (1st ed. 1736) (same).
Our Double Jeopardy Clause was framed against this background. James Madison’s first draft of that Clause stated: "No person shall be subject, except in cases of impeachment, to more than one punishment or one trial for the same offence." 1 ANNALS OF CONG . 451–52 (1789) (Joseph Gales ed., 1834). Representative Egbert Benson objected because the draft varied from "the right heretofore established" by the common law. Id. at 781. To cure the defect, Benson suggested striking the phrase regarding "one trial." Id. at 782. Representative Roger Sherman agreed. He reasoned, "if [the defendant] was convicted on the first [trial], and any thing should appear to set the judgment aside, he was entitled to a second, which was certainly favorable to him." Ibid. The House revised it accordingly, and the Senate concurred in the revision. See S. JOURNAL , 1st Cong., 1st Sess. 71 (1789).
As ratified, the Double Jeopardy Clause provides: "No person shall ... be subject for the same offence to be twice put in jeopardy of life or limb." U.S. CONST. amend. V. Madison’s initial phrasing ("more than one punishment or one trial") was thus replaced with a prohibition on putting a person in "jeopardy" more than once. Credit for that phrasing belongs to Blackstone. See United States v. Wilson , 420 U.S. 332, 341–42, 95 S.Ct. 1013, 43 L.Ed.2d 232 (1975) (noting the Fifth Amendment uses "language that tracked Blackstone’s statement of the principles of autrefois acquit and autrefois convict "); 4 WILLIAM BLACKSTONE, COMMENTARIES *335 ("[T]he plea of auterfoits acquit , or a former acquittal, is grounded on this universal maxim of the common law of England, that no man is to be brought into jeopardy of his life, more than once, for the same offence."). Thus, the historic core of the Double Jeopardy Clause generally bars re-trial of the "same offense" after a conviction or acquittal. See Currier , 138 S. Ct. at 2150.
The Framers adopted not only Blackstone’s language but also some English common-law exceptions to the pleas of prior acquittal and prior conviction. Most relevant here, the plea did not bar all attempts to retry a criminal defendant. The defendant could be retried, for example:
if the jury have been discharged without giving any verdict; or, if, having given a verdict, judgment has been arrested upon it,[ ] or a new trial has been granted in his favour; for, in such a case, his life or limb cannot judicially be said to have been put in jeopardy.
"An arrest of judgment was the technical term describing the act of a trial judge refusing to enter judgment on the verdict because of an error appearing on the face of the record that rendered the judgment invalid." United States v. Sisson , 399 U.S. 267, 280–81, 90 S.Ct. 2117, 26 L.Ed.2d 608 (1970) (plurality opinion). It bore some semblance to a motion for judgment notwithstanding the verdict, but the analogy is not precise. See Arrest of Judgment , Giles Jacob, A New Law-Dictionary (1st ed. 1729). The important point for present purposes is arrest of judgment was a post-conviction motion by the defendant challenging his conviction.
3 JOSEPH STORY, COMMENTARIES ON THE CONSTITUTION § 1781 (1st ed. 1833). Likewise, when an "attainder be reversed in a Court of Error," the defendant "may certainly be indicted again for the same offence, and the rule would be held to apply, that he had never been in jeopardy under the former indictment." Regina v. Drury (1849), 175 Eng. Rep. 516, 520; 2 Car. & K. 190, 199 (N.P.).
Attainder "[i]s when a Man hath committed Treason or Felony, and after Conviction[,] Sentence is passed on him." Attainder , Giles Jacob, A New Law-Dictionary (1st ed. 1729).
That is why it has long been true that a defendant can be retried after he successfully appeals his first conviction. See, e.g. , Ball v. United States , 163 U.S. 662, 672, 16 S.Ct. 1192, 41 L.Ed. 300 (1896) (citing Drury ). As the Supreme Court has explained:
While different theories have been advanced to support the permissibility of retrial, of greater importance than the conceptual abstractions employed to explain the Ball principle are the implications of that principle for the sound administration of justice. Corresponding to the right of an accused to be given a fair trial is the societal interest in punishing one whose guilt is clear after he has obtained such a trial. It would be a high price indeed for society to pay were every accused granted immunity from punishment because of any defect sufficient to constitute reversible error in the proceedings leading to conviction.
United States v. Tateo , 377 U.S. 463, 466, 84 S.Ct. 1587, 12 L.Ed.2d 448 (1964) ; accord Justices of Boston Mun. Court v. Lydon , 466 U.S. 294, 308, 104 S.Ct. 1805, 80 L.Ed.2d 311 (1984) ("The general rule is that the [Double Jeopardy] Clause does not bar reprosecution of a defendant whose conviction is overturned on appeal.").
The Supreme Court recently reminded us the line from Vaux’s Case to Ashe is a crooked one. See Currier , 138 S. Ct. at 2149–50 (noting Ashe "represented a significant innovation in our jurisprudence" that some say "sits uneasily with this Court’s double jeopardy precedent and the Constitution’s original meaning"). One reason why is, for the first 164 years of our Nation’s history, the prohibition on double jeopardy could not be vindicated in habeas proceedings by state prisoners.
From the Founding until after the Civil War, there was no such thing as federal habeas for individuals in state custody (with one limited exception). See Judiciary Act of 1789, ch. 20, § 14, 1 Stat. 73, 81–82 (providing federal habeas only for federal prisoners); Ex Parte Bollman , 8 U.S. (4 Cranch) 75, 98–99, 2 L.Ed. 554 (1807) (suggesting federal courts could issue only writs of habeas corpus ad testificandum for state prisoners). And although a federal prisoner had greater federal habeas privileges than a state prisoner, even the former could not use habeas proceedings to collaterally attack a conviction. During that time, a judgment in a criminal case was just like a judgment in any other case: It was res judicata . As Chief Justice Marshall put it:
The judgment of a court of record whose jurisdiction is final, is as conclusive on all the world as the judgment of this court would be. It is as conclusive on this court as it is on other courts. It puts an end to inquiry concerning the fact, by deciding it.
Ex parte Watkins , 28 U.S. (3 Pet.) 193, 202–03, 7 L.Ed. 650 (1830). In 1867, Congress extended the scope of federal habeas jurisdiction to state prisoners. See Habeas Corpus Act of 1867, ch. 28, § 1, 14 Stat. 385 (empowering federal courts "to grant writs of habeas corpus in all cases where any person may be restrained of his or her liberty in violation of the constitution, or any treaty or law of the United States"). Even then, however, the Supreme Court continued to interpret the scope of its habeas authority in a very limited way. If a prisoner was in jail under a state court judgment of conviction, the Court asked only whether that state court had jurisdiction over the defendant. See, e.g. , Pettibone v. Nichols , 203 U.S. 192, 206, 215–16, 27 S.Ct. 111, 51 L.Ed. 148 (1906) ; Harkrader v. Wadley , 172 U.S. 148, 168–70, 19 S.Ct. 119, 43 L.Ed. 399 (1898) ; see also Wright v. West , 505 U.S. 277, 285, 112 S.Ct. 2482, 120 L.Ed.2d 225 (1992) (plurality opinion) (describing this practice).
That made sense based on how the Great Writ developed at common law. King Charles I thought he could jail English subjects for any reason, or no reason at all. See Darnell’s Case (1627) (K.B.), in 3 T.B. Howell, A Complete Collection of State Trials 1–59 (5th ed. 1816); The Petition of Right, 3 Car. 1 c. 1, § 5 (1628). In response, the habeas writ became a tool for forcing the jailer to provide a lawful reason for the confinement. See An Act for the Better Secureing the Liberty of the Subject and for Prevention of Imprisonments Beyond the Seas, 31 Car. 2 c. 2 (1679). If he could not, the court could force the jailer to provide a trial or some other kind of process. See Hamdi v. Rumsfeld , 542 U.S. 507, 555, 124 S.Ct. 2633, 159 L.Ed.2d 578 (2004) (Scalia, J., dissenting). The habeas remedy (what we call the "privilege") did not apply to post-trial confinement because a criminal judgment, pursuant to a full-fledged criminal trial, was always a lawful basis for jailing a prisoner. See Bushell’s Case (1670), 124 Eng. Rep. 1006, 1009–10; Vaugh. 135, 142–43 (C.P.).
For almost a century following the 1867 Act, no prisoner (state or federal) could collaterally attack his conviction under the Double Jeopardy Clause. Take for example Ex parte Lange , 85 U.S. 163, 18 Wall. 163, 21 L.Ed. 872 (1873). In that case, a federal court sentenced the prisoner twice for one criminal offense of stealing mail bags. The government conceded the sentence violated the Double Jeopardy Clause. And the Court agreed: "For of what avail is the constitutional protection against more than one trial if there can be any number of sentences pronounced on the same verdict?" Id. at 173. Still, the Court held, that did not justify habeas relief. That’s because "[t]he judgment first rendered, though erroneous, was not absolutely void. It was rendered by a court which had jurisdiction of the party and of the offence." Id. at 174. And that was sufficient to deny relief.
It was not until 1953 that state prisoners could use federal habeas proceedings to relitigate free-standing constitutional claims after pressing and losing them in state court. See Brown v. Allen , 344 U.S. 443, 460–65, 73 S.Ct. 397, 97 L.Ed. 469 (1953) ; id. at 506–08, 73 S.Ct. 397 (opinion of Frankfurter, J.); see also Fay v. Noia , 372 U.S. 391, 460, 83 S.Ct. 822, 9 L.Ed.2d 837 (1963) (Harlan, J., dissenting) (describing Brown v. Allen as a "landmark decision[ ]" that "substantially expanded the scope of inquiry on an application for federal habeas corpus"); BRANDON L. GARRETT & LEE KOVARSKY, FEDERAL HABEAS CORPUS: EXECUTIVE DETENTION AND POST-CONVICTION LITIGATION 3 (Robert C. Clark et al. eds., 2013) (referring to Brown v. Allen as the "big bang"). And it was not until 1969 that the Supreme Court incorporated the Double Jeopardy Clause against the States. See Benton v. Maryland , 395 U.S. 784, 794, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969). On the same day it announced Benton , the Court held for the first time that state prisoners could raise Double Jeopardy claims in federal habeas. See North Carolina v. Pearce , 395 U.S. 711, 717–19, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969).
This is the backdrop for Ashe , which came the very next year. In Ashe , a group of masked men allegedly robbed six players at a poker game. 397 U.S. at 437, 90 S.Ct. 1189. Under the relevant state law, Ashe was guilty of robbery if he was one of the masked robbers, even if the State could not prove Ashe robbed any one particular poker player. Id. at 439, 90 S.Ct. 1189. The State tried Ashe for robbing the first player, but the jury acquitted him. Ibid. On their verdict form, the jury found Ashe "not guilty due to insufficient evidence." Ibid. Then the State attempted to try Ashe for robbing a second player. Ibid. The question was whether the Double Jeopardy Clause barred the second trial. Id . at 440–41, 90 S.Ct. 1189.
The Supreme Court held yes. Id . at 447, 90 S.Ct. 1189. The Court, however, did not base that holding on autrefois acquit , the common-law qualifications to that plea, or the original meaning of the Double Jeopardy Clause. Instead, the Court identified a collateral estoppel "ingredient" in that Clause. Id. at 442–44, 90 S.Ct. 1189. The Court then held the State was collaterally estopped from alleging Ashe was one of the robbers because the first jury (1) returned an acquittal and (2) necessarily determined there was insufficient evidence to prove Ashe was one of the robbers. Id. at 445–47, 90 S.Ct. 1189.
The Supreme Court therefore has made clear that Ashe has a different scope than the traditional protections of the Double Jeopardy Clause. "While ... Ashe ’s protections apply only to trials following acquittals, as a general rule, the Double Jeopardy Clause protects against a second prosecution for the same offense after conviction as well as against a second prosecution for the same offense after acquittal." Currier , 138 S. Ct. at 2150 (quotation omitted). That’s why the Court called Ashe "a significant innovation." Id. at 2149. Indeed, Ashe itself recognized the distinction between its collateral-estoppel rule and the rules that applied "at common law." 397 U.S. at 445 n.10, 90 S.Ct. 1189.
In response to Brown v. Allen —along with its progeny such as Ashe —Congress enacted AEDPA. See Schriro v. Landrigan , 550 U.S. 465, 473, 127 S.Ct. 1933, 167 L.Ed.2d 836 (2007) (noting AEDPA "changed the standards for granting federal habeas relief" from those in Brown v. Allen ). As relevant here, AEDPA prohibits a prisoner from raising any claim in federal court unless it was first exhausted in state court. See 28 U.S.C. § 2254(b) ; O’Sullivan v. Boerckel , 526 U.S. 838, 839–40, 119 S.Ct. 1728, 144 L.Ed.2d 1 (1999). After the state court adjudicates the claim, the prisoner must overcome "the relitigation bar imposed by AEDPA." Greene v. Fisher , 565 U.S. 34, 39, 132 S.Ct. 38, 181 L.Ed.2d 336 (2011) (citing 28 U.S.C. § 2254(d)(1) ). The statute thereby restores the res judicata rule Chief Justice Marshall recited in Ex parte Watkins and then modifies it. See Felker v. Turpin , 518 U.S. 651, 663–64, 116 S.Ct. 2333, 135 L.Ed.2d 827 (1996) (comparing AEDPA’s "modified res judicata rule" to Watkins ).
To overcome AEDPA’s relitigation bar, a state prisoner must shoehorn his claim into one of its narrow exceptions. As relevant here, he must show the state court’s adjudication of the claim "resulted in a decision that was  contrary to, or  involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States." 28 U.S.C. § 2254(d)(1).
The first exception to the relitigation bar—the "contrary to" prong—is generally regarded as the narrower of the two. A state-court decision is "contrary to" clearly established federal law only if it "arrives at a conclusion opposite to that reached by [the Supreme] Court on a question of law or if" it resolves "a case differently than [the Supreme] Court has on a set of materially indistinguishable facts." Terry Williams v. Taylor , 529 U.S. 362, 413, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). Langley identifies no Supreme Court precedent that is "opposite to" or "materially indistinguishable" from this case. So here, as in most AEDPA cases, the "contrary to" prong does not apply.
The only other exception to § 2254(d)(1) ’s relitigation bar—the "unreasonable application" prong—is almost equally unforgiving. The Supreme Court has repeatedly held that it is not enough to show the state court was wrong. See, e.g. , Renico v. Lett , 559 U.S. 766, 773, 130 S.Ct. 1855, 176 L.Ed.2d 678 (2010) ("[A] federal habeas court may not issue the writ simply because that court concludes in its independent judgment that the relevant state-court decision applied clearly established federal law erroneously or incorrectly." (quotation omitted)); Landrigan , 550 U.S. at 473, 127 S.Ct. 1933 ("The question under AEDPA is not whether a federal court believes the state court’s determination was incorrect but whether that determination was unreasonable—a substantially higher threshold."). Rather, the relitigation bar forecloses relief unless the prisoner can show the state court was so wrong that the error was "well understood and comprehended in existing law beyond any possibility for fairminded disagreement." Shoop v. Hill , ––– U.S. ––––, 139 S. Ct. 504, 506, 202 L.Ed.2d 461 (2019) (per curiam) (quotation omitted). In other words, the unreasonable-application exception asks whether it is "beyond the realm of possibility that a fairminded jurist could" agree with the state court. Woods v. Etherton , ––– U.S. ––––, 136 S. Ct. 1149, 1152, 194 L.Ed.2d 333 (2016) (per curiam); see also Sexton v. Beaudreaux , ––– U.S. ––––, 138 S. Ct. 2555, 2558, 201 L.Ed.2d 986 (2018) (per curiam) (asking "whether it is possible fairminded jurists could disagree that those arguments or theories are inconsistent with the holding in a prior decision of this Court" (quotation omitted)).
Overcoming AEDPA’s relitigation bar is necessary but not sufficient to win habeas relief. Even after overcoming the bar, the prisoner still must "show, on de novo review, that [he is] ‘in custody in violation of the Constitution or laws or treaties of the United States.’ " Salts v. Epps , 676 F.3d 468, 480 (5th Cir. 2012) (quoting 28 U.S.C. § 2254(a) ); see also Berghuis v. Thompkins , 560 U.S. 370, 390, 130 S.Ct. 2250, 176 L.Ed.2d 1098 (2010) ("[A] habeas petitioner will not be entitled to a writ of habeas corpus if his or her claim is rejected on de novo review [under] § 2254(a)."); Oral Argument at 13:30–13:59 (Langley’s acknowledgement that overcoming AEDPA’s relitigation bar is necessary but not sufficient to obtain habeas relief).
Langley’s claim fails under these demanding standards. We first explain that Langley cannot surmount AEDPA’s relitigation bar. Then we explain that the most-on-point Supreme Court precedent supports the State, not Langley. Lastly, even if we set aside AEDPA’s relitigation bar and review the claim de novo , Langley still cannot prove his second jury necessarily determined anything regarding his specific intent.
The first step in any case under AEDPA’s relitigation bar is to determine the "clearly established Federal law, as determined by the Supreme Court of the United States." 28 U.S.C. § 2254(d)(1). It is not enough to say, as the panel did, that the " Ashe doctrine" forms the relevant clearly established law, or that Ashe established the "governing principles." Langley , 890 F.3d at 516–18. The Supreme Court has repeatedly reversed the courts of appeals for identifying the relevant clearly established law at that level of generality.
Take for example Carey v. Musladin , 549 U.S. 70, 127 S.Ct. 649, 166 L.Ed.2d 482 (2006). In that case, the Ninth Circuit held California deprived the defendant of a fair trial by allowing a murder victim’s family members to sit in the front row of a jury trial wearing buttons with the victim’s photo. Musladin v. Lamarque , 427 F.3d 653, 654–55 (9th Cir. 2005). The Ninth Circuit identified the clearly established law as "the Williams test." Id. at 658. "The Williams test" referred to Estelle v. Williams , 425 U.S. 501, 503–06, 96 S.Ct. 1691, 48 L.Ed.2d 126 (1976), in which the Supreme Court held it would violate the defendant’s fair trial rights to compel him to appear at trial in prison garb. In reversing the Ninth Circuit, the Supreme Court held the clearly established law relevant under AEDPA’s relitigation bar is only the Supreme Court’s holdings, not its dicta. Musladin , 549 U.S. at 74, 127 S.Ct. 649. Therefore, Williams clearly established the law only as applied to prison garb—it could not be extended under AEDPA to vitiate state judgments for spectators’ buttons. Id. at 75–77, 127 S.Ct. 649. As the Supreme Court put it in a different but related context:
We have repeatedly told courts ... not to define clearly established law at a high level of generality. The dispositive question is whether the violative nature of particular conduct is clearly established. This inquiry must be undertaken in light of the specific context of the case, not as a broad general proposition.
In this case, our now-vacated panel opinion conflated the Supreme Court’s holding with its dicta in much the same way the Ninth Circuit did in Musladin . The Ashe Court had much to say about how or why collateral estoppel should apply in the criminal context—just as the Williams Court had much to say about how or why the State should not allow jurors to see unduly prejudicial things in the courtroom. But the holding in Ashe , like the holding in Williams , was narrower. The Ashe Court held only that a general verdict of acquittal for insufficient evidence that "petitioner was ... one of the robbers" precluded the State from "hal[ing] him before a new jury to litigate that issue again." 397 U.S. at 446, 90 S.Ct. 1189 ; see Yeager v. United States , 557 U.S. 110, 119, 129 S.Ct. 2360, 174 L.Ed.2d 78 (2009) (stating Ashe "held that the Double Jeopardy Clause precludes the Government from relitigating any issue that was necessarily decided by a jury’s acquittal in a prior trial").
The Supreme Court has found issue preclusion under Ashe only three other times. See Turner v. Arkansas , 407 U.S. 366, 369–70, 92 S.Ct. 2096, 32 L.Ed.2d 798 (1972) (per curiam); Harris v. Washington , 404 U.S. 55, 57, 92 S.Ct. 183, 30 L.Ed.2d 212 (1971) (per curiam); Simpson v. Florida , 403 U.S. 384, 386, 91 S.Ct. 1801, 29 L.Ed.2d 549 (1971) (per curiam). Turner , Harris , Simpson , and Ashe all involved blanket acquittals. See Turner , 407 U.S. at 367, 92 S.Ct. 2096 (noting jury returned "a general verdict of acquittal"); Harris , 404 U.S. at 55, 92 S.Ct. 183 (noting defendant "was acquitted by a jury" on a single charge); Simpson , 403 U.S. at 384–85, 91 S.Ct. 1801 (noting jury returned a "general" verdict of acquittal); Ashe , 397 U.S. at 439, 90 S.Ct. 1189 (noting jury returned one general verdict of acquittal: "not guilty due to insufficient evidence"). None of the four juries convicted the defendant of the charged crime.
In Yeager , the Court concluded "that acquittals can preclude retrial on counts on which the same jury hangs." 557 U.S. at 125, 129 S.Ct. 2360. The Court, however, did not find Ashe issue preclusion because it remanded the issue of "what the jury necessarily decided in its acquittals." Id. at 125–26, 129 S.Ct. 2360.
Therefore, none of these cases held issue-preclusion principles apply to a conviction. We asked the parties to identify any case extending Ashe to cases involving a conviction. The parties could not find a single Supreme Court case even hinting at that result. That’s unsurprising. As the Supreme Court recently acknowledged, " Ashe’ s protections apply only to trials following acquittals ." Currier , 138 S. Ct. at 2150 (emphases added). Thus, there is no "clearly established Federal law, as determined by the Supreme Court," explaining whether and to what extent a state court should find issue preclusion following a conviction.
After identifying the clearly established law, we move to step two—determining whether the state court decision "involved an unreasonable application of" that law. 28 U.S.C. § 2254(d)(1). To make this determination, we must ask whether any fairminded jurist could believe the "clearly established rule" does not apply to the "set of facts" at hand. White v. Woodall , 572 U.S. 415, 427, 134 S.Ct. 1697, 188 L.Ed.2d 698 (2014). "If such disagreement is possible, then the petitioner’s claim must be denied." Beaudreaux , 138 S. Ct. at 2558.
Langley loses at this step. A fairminded jurist could conclude the rule clearly established in Ashe does not apply to a conviction rather than a general acquittal. When a jury issues a general acquittal, it necessarily determines at least something in the defendant’s favor. It might be possible to identify that something and preclude the government from submitting it to a second jury. That task is obviously different—and more difficult—when the jury convicts the defendant on at least one count. In the face of a conviction on one count, it is not clear which issues if any the jury determined in the defendant’s favor on that same count. We may or may not find this distinction persuasive. That’s irrelevant. What matters is the last reasoned state court decision found it persuasive. See Langley IV , 61 So. 3d at 757–58 (last reasoned state court decision); Wilson v. Sellers , ––– U.S. ––––, 138 S. Ct. 1188, 1192, 200 L.Ed.2d 530 (2018) (requiring deference to that decision if reasonable). The state court recognized Ashe ’s applicability to a "general acquittal." Langley IV , 61 So. 3d at 757 (citing Ashe , 397 U.S. at 444, 90 S.Ct. 1189 ). By contrast, where the jury returns a conviction on "a lesser included offense," the state court found it’s "not always possible to determine" which issues if any should be precluded under Ashe . Ibid. The state court found it "possible" the jury made one of three determinations: (1) Langley was guilty of specific-intent murder, (2) Langley was guilty of something less than specific-intent murder, or (3) the jury avoided the specific-intent issue by rendering a "compromise verdict." Ibid. In the state court’s view, Langley’s argument that the jury found (2) to the exclusion of (1) and (3) was "clearly ... unsupported." Id. at 758.
The dissenters dispute this by confusing it. Imagine a two-count indictment (X and Y), where Y is a lesser-included offense of X. It is well settled that a conviction on Y bars retrial on X. See Price v. Georgia , 398 U.S. 323, 329, 90 S.Ct. 1757, 26 L.Ed.2d 300 (1970) (explaining the petitioner could be retried "for voluntary manslaughter after his first conviction for that offense had been reversed," but the Double Jeopardy Clause precluded retrial on the greater charge of murder regardless of "whether that acquittal is express or implied by a conviction on a lesser included offense when the jury was given a full opportunity to return a verdict on the greater charge" (footnote omitted)); Green v. United States , 355 U.S. 184, 190–91, 78 S.Ct. 221, 2 L.Ed.2d 199 (1957) (concluding petitioner could not be retried for first-degree murder after jury convicted him of second-degree murder). The Supreme Court has held this result is commanded by the Double Jeopardy Clause’s "historic core" protection, which applies to offenses, not issues. See Currier , 138 S. Ct. at 2150. It is undisputed the Louisiana Supreme Court correctly applied this principle here by holding the second jury’s conviction on the lesser-included offense (Y, second-degree murder) barred Louisiana from retrying Langley for the greater offense (X, first-degree murder). See Langley III , 958 So. 2d at 1170 (so holding). The dispute in this case is whether a conviction on Y can create issue preclusion on Y . The dissenters’ steadfast focus on X—which no one disputes and which has never been a part of this federal habeas proceeding—is tantamount to tilting at a windmill. See post at 183 (Higginson, J., dissenting) (arguing it’s "dangerous[ ]" to suggest Langley could be retried on X, even though everyone agrees Langley cannot be retried on X).
Even if we thought the state court committed "clear error" by so holding, we still could not grant relief. Woodall , 572 U.S. at 419, 134 S.Ct. 1697. After all, neither Ashe nor any other Supreme Court precedent mandates that a lesser-included-offense conviction—or to use the dissent’s preferred terminology, an "implicit acquittal"—be given issue-preclusive effect. And Supreme Court precedent does mandate caution in finding Ashe issue preclusion where the jury could have rendered a "compromise" or "lenity" verdict. See United States v. Powell , 469 U.S. 57, 65–66, 105 S.Ct. 471, 83 L.Ed.2d 461 (1984) ; Standefer v. United States , 447 U.S. 10, 22–23, 100 S.Ct. 1999, 64 L.Ed.2d 689 (1980) ; accord Bravo-Fernandez v. United States , ––– U.S. ––––, 137 S. Ct. 352, 363–64, 196 L.Ed.2d 242 (2016) (noting "the jurors in this case might not have acquitted on [certain] counts absent their belief that the ... convictions [on other counts] would stand"). Therefore, a fairminded jurist could find that Ashe ’s rule regarding general acquittals does not require issue preclusion for Langley’s conviction. Under AEDPA, that’s the end of the matter. See Woodall , 572 U.S. at 419–20, 134 S.Ct. 1697.
In the past, some federal courts mistakenly thought it was only the beginning. The Sixth Circuit, for example, faulted a state court for "unreasonably refus[ing] to extend" a Supreme Court precedent "to a new context where [the Sixth Circuit thought] it should apply." Woodall v. Simpson , 685 F.3d 574, 579 (6th Cir. 2012) (quoting Terry Williams , 529 U.S. at 407, 120 S.Ct. 1495 ). The Supreme Court emphatically reversed. The Court emphasized it "has never adopted the unreasonable-refusal-to-extend rule .... It has not been so much as endorsed in a majority opinion, let alone relied on as a basis for granting habeas relief." Woodall , 572 U.S. at 426, 134 S.Ct. 1697 ; see also ibid. (holding "we reject it"). That result is compelled by the text of the relitigation bar itself: " Section 2254(d)(1) provides a remedy for instances in which a state court unreasonably applies [the Supreme] Court’s precedent; it does not require state courts to extend that precedent or license federal courts to treat the failure to do so as error." Ibid. To the contrary:
"[I]f a habeas court must extend a rationale before it can apply to the facts at hand," then by definition the rationale was not "clearly established at the time of the state-court decision." AEDPA’s carefully constructed framework "would be undermined if habeas courts introduced rules not clearly established under
the guise of extensions to existing law."
Because a fairminded jurist could decide the clearly established rule does not cover this case, we’d have to extend Ashe to grant relief here. That is something AEDPA says we cannot do. See, e.g. , Woods v. Donald , ––– U.S. ––––, 135 S. Ct. 1372, 1377, 191 L.Ed.2d 464 (2015) (per curiam) ("Because none of our cases confront the specific question presented by this case, the state court’s decision could not be ‘contrary to’ any holding from this Court," nor an "unreasonable application" thereof. (quotation omitted)); Woodall , 572 U.S. at 427, 134 S.Ct. 1697 ("Perhaps the logical next step from [three previous Supreme Court cases] would be to hold that the Fifth Amendment requires a penalty-phase no-adverse-inference instruction in a case like this one; perhaps not. Either way, we have not yet taken that step, and there are reasonable arguments on both sides—which is all Kentucky needs to prevail in this AEDPA case."); Knowles v. Mirzayance , 556 U.S. 111, 122, 129 S.Ct. 1411, 173 L.Ed.2d 251 (2009) ("[T]his Court has held on numerous occasions that it is not ‘an unreasonable application of’ ‘clearly established Federal law’ for a state court to decline to apply a specific legal rule that has not been squarely established by this Court."); accord Teague v. Lane , 489 U.S. 288, 299–310, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989) (plurality opinion) (holding federal courts may not develop—and habeas petitioners may not seek—new legal rules on collateral review). As far as we can tell, the only other court of appeals to address this question agrees with us. See Owens v. Trammell , 792 F.3d 1234, 1246–50 (10th Cir. 2015) (holding AEDPA precluded awarding habeas relief based on Ashe following a conviction).
The dissenters appear to recognize that Louisiana’s courts have no obligation to extend Supreme Court precedent but paradoxically fault the state court for following that precedent. See post at 184–85 (Costa, J., dissenting). They suggest Louisiana violated the Constitution by allowing Langley to waive his right to a third jury trial after losing the first two. That is obviously wrong: The Supreme Court "hold[s] no constitutional doubts about the practice[ ], common in both federal and state courts, of accepting waivers of jury trial[s]." Duncan v. Louisiana , 391 U.S. 145, 158, 88 S.Ct. 1444, 20 L.Ed.2d 491 (1968). But even if arguendo the dissenters were right about the jury-trial right, AEDPA still would foreclose relief. If AEDPA protects a state court decision refusing to extend Supreme Court precedent, it certainly protects a state court decision refusing to contradict that precedent.
Extending Ashe in these circumstances would also conflict with other clearly established law. That’s because the Supreme Court has confronted similar facts before and rejected the prisoner’s Double Jeopardy claim. See Schiro v. Farley , 510 U.S. 222, 114 S.Ct. 783, 127 L.Ed.2d 47 (1994). If anything, Schiro was a harder case.
The jury convicted Schiro of felony murder (count II) but did not return a verdict on intentional murder (count I). Id. at 225–26, 114 S.Ct. 783. "Thereafter, in a separate sentencing hearing, the same jury unanimously concluded that Schiro did not deserve the death penalty, presumably because he had not intended to kill." Id. at 239, 114 S.Ct. 783 (Stevens, J., dissenting) (footnote omitted). And that presumption appeared well grounded because "[t]he principal issue at trial was Schiro’s mental condition." Id. at 240, 114 S.Ct. 783 ; see also ibid. ("No one disputed that he had caused his victim’s death, but intent remained at issue in other ways. Five expert witnesses—two employed by the State, one selected by the court, and two called by the defense—testified at length about Schiro’s unusual personality, his drug and alcohol addiction, and his history of mental illness." (citations omitted)). Moreover, both the jury instructions and Indiana state law permitted the jurors to return a guilty verdict on every count on which they had unanimity—which could imply the jury intended to acquit the defendant on each count they failed to return (like count I, intentional murder). See id. at 233–34, 114 S.Ct. 783 (majority opinion); id. at 246–47, 114 S.Ct. 783 (Stevens, J., dissenting) (arguing jury’s failure to return a verdict on intentional murder implicitly acquitted on that count). Nonetheless, the State argued Schiro’s intent was an aggravating factor that justified the court in sentencing him to death. The Supreme Court held the jurors’ failure to return a verdict on intentional murder did not collaterally estop the State from so arguing. Id. at 232–36, 114 S.Ct. 783 (majority opinion); see also Sattazahn v. Pennsylvania , 537 U.S. 101, 113–15, 123 S.Ct. 732, 154 L.Ed.2d 588 (2003) (holding the Double Jeopardy Clause does not bar reprosecution for capital murder after prisoner successfully appeals judgment for life sentence).
The principal dissent says Schiro is distinguishable because it involved a single (albeit bifurcated) trial as opposed to two successive trials. Post at 181–82 (Higginson, J., dissenting). That’s irrelevant. The Supreme Court assumed Ashe applied identically in both circumstances and then held Schiro’s Ashe claim failed. See Schiro , 510 U.S. at 232, 114 S.Ct. 783. The Schiro Court’s pure Ashe holding is fully applicable here.
Louisiana law makes this case easier than Schiro . Under Louisiana law, "the jury must be given the option to convict the defendant of the lesser offense, even though the evidence clearly and overwhelmingly supported a conviction of the charged offense." State v. Porter , 639 So. 2d 1137, 1140 (La. 1994). And the jury was given that option. The Langley II jury was repeatedly told—orally and in writing—that "[t]he responsive lesser offenses to the charge of First Degree Murder are Second Degree Murder and Manslaughter." Neither the dissenters nor Langley’s able appellate attorney has ever disputed that the evidence supported every element of the first-degree murder count against Langley, including specific intent. And a rational jury could have credited that overwhelming evidence and still —in accordance with the instructions and the law—returned a verdict for the lesser-included offense of second-degree specific-intent murder.
As far as the Schiro opinion reveals, the jury in that case received no such option. To the contrary, Indiana law at least arguably required Schiro’s jury to return a verdict on count I (intentional murder) if they agreed the State proved it. See 510 U.S. at 240–42, 114 S.Ct. 783 (Stevens, J., dissenting). And to the extent Schiro’s jury instructions were ambiguous on that score, Langley’s were even more so. See infra at 169–70. If the jury’s failure to return a verdict on intentional murder did not trigger collateral estoppel in Schiro , it certainly does not do so here.
Finally, it bears emphasis that Schiro was a pre-AEDPA death-penalty case. Even after Schiro’s jury potentially acquitted him of intentional murder by returning only a verdict of felony murder, the trial judge rejected the jury’s recommended sentence and held the State proved intent beyond a reasonable doubt for purposes of sentencing the defendant to death. Id. at 226–27, 114 S.Ct. 783 (majority opinion). The Supreme Court upheld that result even without AEDPA’s relitigation bar.
In contrast, Langley’s jury did not return a verdict of felony murder. It returned a verdict of "second-degree murder," which could mean Langley was convicted of specific-intent murder or felony murder. Langley also faces the additional burden of AEDPA. If Indiana could prevail in Schiro , then Louisiana must prevail on easier facts and a much more favorable legal standard. See Alvarado , 541 U.S. at 664, 124 S.Ct. 2140.
At the en banc argument, Langley suggested it matters whether the state court (or the state’s lawyer at the panel stage) cited Schiro . It doesn’t. Federal courts must apply § 2254(d) in light of controlling Supreme Court holdings regardless of whether the state court or the state’s lawyer cites them.
First, it doesn’t matter whether the state court cited Schiro . The Ninth Circuit once refused to apply AEDPA’s relitigation bar because "the state court ‘failed to cite ... any federal law, much less the controlling Supreme Court precedents.’ " Early v. Packer , 537 U.S. 3, 8, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002) (per curiam) (alteration in original) (quoting Packer v. Hill , 291 F.3d 569, 578 (9th Cir. 2002) ). The Supreme Court unanimously and summarily reversed: "Avoiding [vitiation of a state judgment in federal court] does not require citation of our cases—indeed, it does not even require awareness of our cases, so long as neither the reasoning nor the result of the state-court decision contradicts them." Ibid. ; see also Mitchell v. Esparza , 540 U.S. 12, 16, 124 S.Ct. 7, 157 L.Ed.2d 263 (2003) (per curiam).
Second, it also doesn’t matter whether the State’s panel-stage appellate lawyer cited Schiro . The relitigation bar constrains our ability to award habeas relief regardless of what counsel cites or does not cite. See 28 U.S.C. § 2254(d) ("An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court [unless statutory exceptions are satisfied]." (emphasis added)). Every court of appeals to consider the question—including ours—has held a State’s lawyers cannot waive or forfeit § 2254(d) ’s standard. That likewise means a State’s lawyers cannot waive or forfeit the applicable "clearly established law." See, e.g. , Thompson v. Runnels , 705 F.3d 1089, 1097–1100 (9th Cir. 2013) ; BRYAN R. MEANS, FEDERAL HABEAS MANUAL § 3:97 (2019). Schiro rejected a stronger Double Jeopardy claim under harder facts and without the added hurdle of § 2254(d) ’s relitigation bar. Schiro thus provides an independent basis for denying Langley’s claim.
See Winfield v. Dorethy , 871 F.3d 555, 560–63 (7th Cir. 2017) (holding State’s lawyer cannot waive applicability of AEDPA and emphasizing "the general principle that waiver does not apply to arguments regarding the applicable standard of review"), cert. denied , ––– U.S. ––––, 138 S. Ct. 2003, 201 L.Ed.2d 262 (2018) ; Wilson v. Mazzuca , 570 F.3d 490, 500 (2d Cir. 2009) ("The standard of review set forth in AEDPA is not conditional. It is stated in mandatory terms—habeas relief ‘shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings.’ " (emphases omitted) (quoting 28 U.S.C. § 2254(d) )); Gardner v. Galetka , 568 F.3d 862, 879 (10th Cir. 2009) ("[T]he correct standard of review under AEDPA is not waivable. It is, unlike exhaustion, an unavoidable legal question we must ask, and answer, in every case."); Brown v. Smith , 551 F.3d 424, 428 n.2 (6th Cir. 2008) (holding "a party cannot ‘waive’ the proper standard of review [under AEDPA] by failing to argue it"), abrogated on other grounds , Cullen v. Pinholster , 563 U.S. 170, 131 S.Ct. 1388, 179 L.Ed.2d 557 (2011) ; Valdez v. Cockrell , 274 F.3d 941, 950 (5th Cir. 2001) ("The word ‘shall’ is mandatory in meaning. Thus, we lack discretion as to the operation of this section." (citation omitted)). One court—the Ninth Circuit—stepped out of line, and it was GVR’d by the Supreme Court. See James v. Ryan , 679 F.3d 780, 802 (9th Cir. 2012) (finding waiver of argument that state courts adjudicated the ineffective assistance of counsel claims on the merits and that AEDPA’s relitigation bar thus applied to them), cert. granted, judgment vacated, and remanded by Ryan v. James , 568 U.S. 1224, 133 S.Ct. 1579, 185 L.Ed.2d 572 (2013).
Wilson v. Sellers is not to the contrary. Wilson requires us to "look through" to the last reasoned state court decision and apply AEDPA’s relitigation bar to it. 138 S. Ct. at 1192 ; see supra Part III.A.2 (doing so). But Wilson does not purport to overrule Packer or Esparza . Nor does Wilson say the state court must cite a Supreme Court decision to trigger AEDPA’s strictures. See Meders v. Warden, Ga. Diagnostic Prison , 911 F.3d 1335, 1350 (11th Cir. 2019) (explaining Wilson "was not about the specificity or thoroughness with which state courts must spell out their reasoning to be entitled to AEDPA deference"); Hebert v. Rogers , 890 F.3d 213, 221 (5th Cir. 2018) (explaining "the brevity of a state court’s opinion is immaterial" and noting "a state court’s decision does not need to be thorough or directly address [the] Supreme Court’s cases").
Here, as in Schiro , the last-reasoned state court decision held the prisoner failed to prove the jury necessarily determined the specific-intent issue in his favor. Compare Schiro , 510 U.S. at 232–36, 114 S.Ct. 783, with Langley IV , 61 So. 3d at 757–58. Schiro thus illustrates that the last-reasoned state court decision was a reasonable application of Supreme Court precedent, including some holdings the state court did not cite. Nothing in AEDPA, Wilson , or any other relevant authority requires the state court to cite Schiro — or any other specific Supreme Court case—to insulate its decision from vitiation in federal court. Wilson likewise does not prohibit this Court from considering Supreme Court cases not cited when evaluating the reasonableness of the state court’s reasoning. Indeed, we are often compelled to do so to determine "clearly established Federal law." 28 U.S.C. § 2254(d)(1).
The principal dissent takes issue with our application of AEDPA. Even if the dissent’s arguments were well taken and AEDPA’s relitigation bar did not apply, Langley would not automatically be entitled to habeas relief. Instead, he would still need to show—under a de novo review standard—"that he is in custody in violation of the Constitution ... of the United States." 28 U.S.C. § 2254(a) ; see Thompkins , 560 U.S. at 390, 130 S.Ct. 2250 ; Salts , 676 F.3d at 480. Langley cannot do so because he cannot prove the Langley II judgment triggered collateral estoppel of the specific-intent issue.
Collateral estoppel—or, as we call it today, issue preclusion—originates in the law of civil judgments. See, e.g. , Cromwell v. County of Sac , 94 U.S. 351, 354, 24 L.Ed. 195 (1876). As with other preclusion doctrines (like res judicata ), the idea is that an issue definitively settled once is "forever settled as between the parties." Baldwin v. Iowa State Traveling Men’s Ass’n , 283 U.S. 522, 525, 51 S.Ct. 517, 75 L.Ed. 1244 (1931) ; see also DAVID L. SHAPIRO, CIVIL PROCEDURE: PRECLUSION IN CIVIL ACTIONS 48 (2001) ("[A]n issue once decided is settled, at least as between the parties."). In civil cases, the Supreme Court "regularly turns to the Restatement (Second) of Judgments for a statement of the ordinary elements of issue preclusion." B&B Hardware, Inc. v. Hargis Indus., Inc. , ––– U.S. ––––, 135 S. Ct. 1293, 1303, 191 L.Ed.2d 222 (2015). Under the Restatement, in turn, a civil judgment can generate issue preclusion if and only if it meets certain essential prerequisites. See RESTATEMENT (SECOND) OF JUDGMENTS § 27 (Am. Law Inst. 1982). Three of those prerequisites are relevant here: (a) The issue must be "actually ... determined" by the judgment, (b) the issue must be "essential to the judgment," and (c) the judgment must be "valid and final." Ibid.
In civil cases, the availability of appellate review of the judgment in the first case is particularly important to its issue-preclusive effect in a second case. See id. § 28. That’s because, as noted above, a civil judgment generates issue preclusion only when it’s "valid and final." And the "valid[ity]" of a judgment is suspect if it cannot be reviewed. Therefore, the Restatement concludes, "the availability of review for the correction of errors has become critical to the application of preclusion doctrine." Id. § 28 cmt. a; see also Bravo-Fernandez , 137 S. Ct. at 358 ("In civil litigation, where issue preclusion and its ramifications first developed, the availability of appellate review is a key factor."). Correlatively, once a civil judgment is reversed on appeal, it’s obviously no longer "valid" and retains zero preclusive effect. See RESTATEMENT (SECOND) OF JUDGMENTS § 27 cmt. o; 18A CHARLES ALAN WRIGHT ET AL., FEDERAL PRACTICE AND PROCEDURE § 4432 (3d ed. 2018) [hereinafter WRIGHT & MILLER ] ("Reversal and remand for further proceedings on the entire case defeats preclusion entirely until a new final judgment is entered by the trial court or the initial judgment is restored by further appellate proceedings.").
The principal reason issue preclusion is narrower in criminal cases than in civil ones is the limited availability of appellate review for the former. Criminal issue preclusion attaches to a general verdict of acquittal, and "the Government is precluded from appealing or otherwise upsetting such an acquittal by the Constitution’s Double Jeopardy Clause." Powell , 469 U.S. at 65, 105 S.Ct. 471. This "absence of appellate review of acquittals ... calls for guarded application of preclusion doctrine in criminal cases." Bravo-Fernandez , 137 S. Ct. at 358 ; see also Currier , 138 S. Ct. at 2152 (plurality opinion) ("We think that caution remains sound.").
Take for example Standefer . In that case, the defendant was indicted for bribing an IRS official. 447 U.S. at 11, 100 S.Ct. 1999. While that indictment was pending, the IRS official was acquitted of accepting three bribes from Standefer. Id. at 12–13, 100 S.Ct. 1999. Standefer argued the IRS official’s acquittal should trigger nonmutual collateral estoppel against the government’s prosecution of Standefer. Id. at 13–14, 21–22, 100 S.Ct. 1999. The Supreme Court rejected that argument because the government did not have a full and fair opportunity to litigate the acquittal of the IRS agent. Id. at 22, 100 S.Ct. 1999. For example, the government could not seek a new trial because the acquittal is contrary to the evidence, nor could it appeal the acquittal. The Supreme Court explained:
The absence of these remedial procedures in criminal cases permits juries to acquit out of compassion or compromise or because of their assumption of a power which they had no right to exercise, but to which they were disposed through lenity. It is of course true that verdicts induced by passion and prejudice are not
unknown in civil suits. But in civil cases, post-trial motions and appellate review provide an aggrieved litigant a remedy; in a criminal case the Government has no similar avenue to correct errors. Under contemporary principles of collateral estoppel, this factor strongly militates against giving an acquittal preclusive effect.
Id. at 22–23, 100 S.Ct. 1999 (quotations and citations omitted). Time and again—from Powell and Standefer to Currier and Bravo-Fernandez —the Supreme Court has repeatedly admonished lower courts to carefully apply issue preclusion in criminal cases.
Our now-vacated panel opinion misapplied these principles. It ignored the Supreme Court’s admonition regarding "guarded application of preclusion doctrine in criminal cases." Bravo-Fernandez , 137 S. Ct. at 358. In its place, the panel substituted a rigid logic game, complete with numbered "conditions" that could be "fulfilled" or negated according to "the rules of logic." 890 F.3d at 519–20. That not only contravenes the Supreme Court’s warnings in cases like Currier , Bravo-Fernandez , Standefer , and Powell , but it also contravenes Ashe itself. Ashe emphasized "the rule of collateral estoppel in criminal cases is not to be applied with the hypertechnical and archaic approach of a 19th century pleading book, but with realism and rationality." 397 U.S. at 444, 90 S.Ct. 1189.
Under a proper understanding of collateral estoppel principles, Langley cannot demonstrate Langley II precluded the specific-intent issue. That’s for three reasons.
First, Langley cannot prove the jury "actually determined" the issue of specific intent even under the (broader) rules of civil judgments. See RESTATEMENT (SECOND) OF JUDGMENTS § 27 (one prerequisite of preclusion is the issue was "actually ... determined" in the first civil action); SHAPIRO , supra , at 176 ("[T]he first precondition for the application of issue preclusion [is] that the issue have been ‘actually litigated and determined’ ... in the prior action."). Here is what the Langley II jury actually determined:
2 MAY 16, 2003 WE, THE JURY IN THE ABOVE CAPTIONED MATTER, FIND THE DEFENDANT, RICKY JOSEPH LANGLEY, GUILTY OF SECOND DEGREE MURDER ON OR ABOUT FEBRUARY 7, 1992. [REDACTED\] REPRESENTATIVE
We turn then to the jury instructions. The judge orally instructed the jury it could premise its second-degree murder conviction on a finding of specific intent. During its deliberations, the jury sent a note asking for "the instruction sheet" on "specific intent" (among other things). The judge provided the jury with written instructions that again told the jury it could convict Langley of second-degree murder based on specific intent. Langley never objected to any of this at trial. To the contrary, counsel for the State and the defense had a colloquy with the trial judge over this exact instruction. And everyone agreed the jury should be instructed on second-degree specific-intent murder. Then at oral argument before our en banc Court, Langley’s counsel conceded the jury was given the option of returning a legally valid conviction of second-degree specific-intent murder. See Oral Argument at 9:08–9:29.
We are aware of no case from any court that would allow us to infer a jury "irrationally" chose a concededly valid option offered in the instructions. It was therefore wrong to hold, as the panel did, that no "rational jury could have convicted Langley of specific intent second degree murder." Langley , 890 F.3d at 521 (quotation and alteration omitted). Under de novo review, we hold the state court was objectively correct to find "[i]t is possible that the jury convicted the defendant of specific intent second degree murder." Langley IV , 61 So. 3d at 757. Langley therefore cannot prove the jury "actually determined" the issue of specific intent in his favor.
Second, and for similar reasons, Langley cannot prove the issue of specific intent was "necessary" or "essential to the judgment" even under the (broader) civil preclusion rules. See RESTATEMENT (SECOND) OF JUDGMENTS § 27 (one prerequisite of preclusion is the issue was "essential to the judgment"); SHAPIRO , supra , at 177 (same). Under Louisiana law, a jury can find a defendant overwhelmingly guilty of first-degree murder and still choose to convict of second-degree murder. See Porter , 639 So. 2d at 1140 ; LA. CODE CRIM. PROC. ANN . art. 814(A)(1) (responsive verdicts to "First Degree Murder" include "Guilty of second degree murder"). In accordance with this law, the jury was repeatedly instructed it could find every element of first-degree murder—including specific intent—and still choose to return a verdict of second-degree murder. The jury also was instructed it could convict of second-degree murder without finding specific intent. That means the jury could return its lawful second-degree murder conviction after (a) finding specific intent, (b) finding no specific intent, or (c) declining to consider the question of specific intent. To infer why the Langley II jury convicted him only on second-degree murder "would require speculation into what transpired in the jury room," and would require us to "scrutinize" the jury’s "failures to decide" rather than its actual decision. Yeager , 557 U.S. at 122, 129 S.Ct. 2360. We cannot do that. The existence of three possibilities for the actual verdict means the issue of specific intent was not essential to the judgment. And since there could be no preclusion even under the broader civil preclusion rules, there certainly can be no issue preclusion under Ashe . Again, the state court’s judgment would survive de novo review. See Langley IV , 61 So. 3d at 758.
Moreover, the instructions gave the jury a rational reason not to decide the issue. If the jury wanted to reconvene for a punishment hearing to sentence Langley to death, it would have to confront the specific-intent issue, find it, and convict him of first-degree murder. But if the jury chose second-degree murder, it could convict without deciding the specific-intent issue, avoid a separate sentencing hearing, and ensure Langley would spend the rest of his life behind bars. The jury instructions were explicit to that effect: "Whoever commits the crime of Second Degree Murder shall be punished by life imprisonment at hard labor without the benefit of probation, parole, or suspension of sentence." And Langley’s lawyer used these instructions to plead for the jury’s mercy. The record suggests the jury might’ve chosen second-degree murder for precisely this reason. See Langley II Sentencing Tr. at 15–16 (May 22, 2003). The state court therefore was objectively correct to conclude the jury could have avoided deciding the specific-intent issue by reaching a "compromise verdict" that sentenced Langley to life in prison. Langley IV , 61 So. 3d at 757.
This possibility makes the panel’s grant of habeas relief all the more untenable. The point of the preclusion doctrines is to protect verdicts against collateral attacks by "multiple lawsuits" and to enforce "repose." Montana v. United States , 440 U.S. 147, 153–54, 163, 99 S.Ct. 970, 59 L.Ed.2d 210 (1979). That means "a losing litigant deserves no rematch after a defeat fairly suffered." Astoria Fed. Sav. & Loan Ass’n v. Solimino , 501 U.S. 104, 107, 111 S.Ct. 2166, 115 L.Ed.2d 96 (1991). It would be the height of irony for Langley to convince the jury to choose second-degree murder so he could spend the rest of his life in jail—only to demand a rematch on the basis of issue preclusion, collaterally attack the conviction, and walk free. And whatever else might be said about treating the jury like a pawn in this way, it hardly respects "the fundamental role of juries." Post at 184 (Costa, J., dissenting).
Third and finally, Langley cannot prove the issue of specific intent was decided in a "valid and final" judgment even under the (broader) civil preclusion rules. See RESTATEMENT (SECOND) OF JUDGMENTS § 27 (one prerequisite of preclusion is the issue was decided in a "valid and final" civil judgment); SHAPIRO , supra , at 176 ("In addition to the requirement of ‘validity,’ a judgment must be ‘final’ to be entitled to recognition."). When a judgment is partially reversed on appeal, "[t]here is no preclusion as to the matters vacated or reversed." 18A WRIGHT & MILLER , supra , § 4432 ; cf. Aguillard v. McGowen , 207 F.3d 226, 229 (5th Cir. 2000) ("A conviction overturned on appeal cannot constitute a final judgment for purposes of collateral estoppel."). And the preclusive effect of the remainder of the judgment "is controlled by the actual appellate disposition." 18A WRIGHT & MILLER , supra , § 4432.
Here, the Louisiana intermediate appellate court reversed the Langley II judgment and remanded for retrial on everything. See Langley II , 896 So. 2d at 212. The Louisiana Supreme Court agreed the "trial error require[d] a reversal of Langley’s conviction and sentence," but held, under Louisiana law, the jury’s conviction for the lesser-included offense of second-degree murder precluded retrying Langley for the greater offense of first-degree murder. See Langley III , 958 So. 2d at 1170 (citing LA. CODE CRIM. PROC. ANN . art. 598(A) ). That was the entirety of its preclusion decision; it remanded everything else for retrial. See id. at 1171. This "actual appellate disposition" means, even under the ordinary rules applicable to civil judgments, the State would not be issue-precluded from retrying Langley for second-degree specific-intent murder. 18A WRIGHT & MILLER , supra , § 4432. And we know one thing with confidence: The Double Jeopardy Clause’s issue-preclusion ingredient cannot sweep more broadly than the equitable doctrine that has governed civil cases for centuries. See Bravo-Fernandez , 137 S. Ct. at 357–58. 3.
This is not to say the Langley II judgment "could have no preclusive effect on the [Langley III trial]." Post at 184 (Higginson, J., dissenting); see also id. at 174–75 (arguing our "Restatement-based analysis sows doubt that any part of the [Langley II ] verdict was a valid final judgment"); post at 185–86 (Costa, J., dissenting) (similar). The Louisiana Supreme Court held the Langley II conviction for second-degree murder barred retrial for the greater offense of first-degree murder. See Langley III , 958 So. 2d at 1169–70 (citing, inter alia , Green , 355 U.S. at 188, 193, 78 S.Ct. 221 ). We sow no doubt about that, as we’ve already explained. See supra n.5. The U.S. Supreme Court has repeatedly reminded us, however, that offense preclusion under Green and issue preclusion under Ashe are different. See, e.g. , Currier , 138 S. Ct. at 2150. We agree with the state court that Langley II does not trigger Ashe issue preclusion. This case no more threatens the "unassailable" finality of Langley II than do other cases to which Ashe does not apply. See, e.g. , Sattazahn , 537 U.S. at 113–15, 123 S.Ct. 732 (holding finality concerns do not bar second prosecution for capital murder even after first trial yielded judgment of life imprisonment).
The dissenters offer four responses to our de novo rejection of Langley’s claim. The first is confusing. The second is imaginary. The third is irrelevant. And the fourth is unfortunate.
First, the confusion: The dissenters excoriate our reliance on the Restatement (Second) of Judgments as somehow constituting a "doctrinal innovation" in issue-preclusion law. See, e.g. , post at 174, 182–83 (Higginson, J., dissenting). But as noted above, the Supreme Court itself "regularly turns to the Restatement (Second) of Judgments for a statement of the ordinary elements of issue preclusion." B&B Hardware , 135 S. Ct. at 1303 ; see also, e.g. , Herrera v. Wyoming , ––– U.S. ––––, 139 S. Ct. 1686, 1696–97, 203 L.Ed.2d 846, 2019 WL 2166394, at *7 (May 20, 2019) ; id. at *16–20 (Alito, J., dissenting); Kircher v. Putnam Funds Tr. , 547 U.S. 633, 646–47, 126 S.Ct. 2145, 165 L.Ed.2d 92 (2006) ; New Hampshire v. Maine , 532 U.S. 742, 748–49, 121 S.Ct. 1808, 149 L.Ed.2d 968 (2001) ; Arizona v. California , 530 U.S. 392, 414, 120 S.Ct. 2304, 147 L.Ed.2d 374 (2000) ; Baker v. Gen. Motors Corp. , 522 U.S. 222, 233 n.5, 118 S.Ct. 657, 139 L.Ed.2d 580 (1998) ; United States v. Stauffer Chem. Co. , 464 U.S. 165, 171, 104 S.Ct. 575, 78 L.Ed.2d 388 (1984) ; Montana v. United States , 440 U.S. 147, 153–54, 99 S.Ct. 970, 59 L.Ed.2d 210 (1979). Unsurprisingly, the Supreme Court also relies on the Restatement to determine the bounds of Ashe issue preclusion. See, e.g. , Bravo-Fernandez , 137 S. Ct. at 357–58 ; Bobby v. Bies , 556 U.S. 825, 834, 837, 129 S.Ct. 2145, 173 L.Ed.2d 1173 (2009). There is nothing remotely "innovati[ve]" about our reliance on the Restatement here. Post at 182 (Higginson, J., dissenting).
The dissenters say these cases are irrelevant because they rejected Ashe claims by deciding Ashe did not "apply in a given situation" rather than "actually adjudicating" the claims. Post at 182 n.19 (Higginson, J., dissenting); see also id. at 180. This purported distinction—between Ashe ’s applicability and Ashe ’s application—has no substance. If a doctrine does not "apply in a given situation," then a claim based on that doctrine fails. And a court so holding has "actually adjudicat[ed]" the claim. Whatever the dissent means, it cannot dispute that in both cases the Court considered an Ashe claim, relied on the Restatement to analyze that claim, and then rejected the claim on the merits.
Equally baffling is the dissenters’ concern over whether the state courts relied on the Restatement. E.g. , post at 174 (Higginson, J., dissenting). Under AEDPA’s relitigation bar, the state court’s reasoning can matter. See, e.g. , Wilson , 138 S. Ct. at 1191–92. But we’re not discussing the Restatement to determine whether the relitigation bar protects the state court’s judgment. We’re discussing it to hold that—even without the bar—the state court was correct under de novo review to find no issue preclusion. Supreme Court precedents (and our own) specifically authorize us to deny a state prisoner’s habeas claim under either the relitigation bar or de novo review. See Thompkins , 560 U.S. at 390, 130 S.Ct. 2250 ; Mirzayance , 556 U.S. at 123–24, 129 S.Ct. 1411 ; Salts , 676 F.3d at 480. We previously discussed the former; here we’re discussing the latter. The dissenters appear confused over which standard applies where.
Their second response is imaginary. The dissenters posit a hypothetical jury trial with instructions that were never actually given. It’s simply not true the judge instructed the jurors "to begin with the single charge of first degree murder and ... work their way down through the list of responsive verdicts." Post at 182 (Higginson, J., dissenting). Nor did the court instruct the Langley II jury that it could consider second-degree murder "only if it were not ... convinced" of specific intent. Id. at 176. The actual jury instructions said the exact opposite: The court instructed the jury it could find Langley guilty of "SECOND DEGREE MURDER" based on a finding "THAT THE DEFENDANT ACTED WITH SPECIFIC INTENT TO KILL." The dissenters cannot find issue preclusion by ignoring the instructions given to the jury and imagining others that were not.
Their third response is irrelevant. The dissenters make much of the jury instruction that said, "[i]f you are convinced beyond a reasonable doubt that [Langley] is guilty of first degree murder, your verdict should be ‘guilty.’ " Post at 176 (Higginson, J., dissenting) (second alteration in original). The dissenters say this instruction prohibited the jury from returning a verdict for second-degree specific-intent murder. Of course, that ignores the other instructions that empowered the jury to return a "SECOND DEGREE MURDER" verdict based on a finding "THAT THE DEFENDANT ACTED WITH SPECIFIC INTENT TO KILL." It ignores Langley’s agreement—at trial and here—that the jury could return a verdict for second-degree specific-intent murder. See supra at 165–66. And it would require holding the jury instructions violated Louisiana law. See supra at 166 (noting, under Porter and Article 814(A)(1), the jury could find specific intent and choose second-degree murder). "We do not think that a federal court can presume so lightly that a state court failed to apply its own law." Bell v. Cone , 543 U.S. 447, 455, 125 S.Ct. 847, 160 L.Ed.2d 881 (2005) (per curiam).
But even if Langley could misconstrue the instructions as violating state law, it would still be irrelevant. Schiro holds that issue preclusion does not attach where "[t]he jury instructions on the issue of intent to kill were ... ambiguous." 510 U.S. at 234, 114 S.Ct. 783. If we agree with the dissenters on anything, it’s that one instruction very clearly told the Langley II jury it could convict of second-degree specific-intent murder. And if we spot the dissenters all of their points arguendo , the absolute most they can prove is that a second jury instruction told the jury it could not convict of second-degree specific-intent murder. That ambiguity would put the case on all fours with Schiro . And it would compel the denial of habeas relief—with AEDPA or without it.
Fourth and finally, the unfortunate: The dissenters accuse us of "dangerously disregard[ing] Supreme Court precedent," "eras[ing] constitutional protections," and tearing "many pages ... from the United States and Federal Reporters." Post at 175, 183–84 (Higginson, J., dissenting). Worse, they question whether our real motivation is to underrule Ashe because we "disagree strongly with [its] foundations." Id. at 175 n.5. Worse still, they say we have bartered away our legal principles "wholesale" to reach a preferred policy result. Id. at 174–75. This sort of rhetoric is regrettable.
We will not respond in kind. But we will make our motivation patently clear: It is the law. Ashe , Turner , and every other Supreme Court case finding issue-preclusion under the Double Jeopardy Clause involved a general acquittal. This one does not. If we were state judges, we’d obviously still disagree with the dissenters about whether issue preclusion attaches to Langley’s conviction. That much is obvious from our de novo review of the issue-preclusion question and the dissenters’ very different approach to it.
But of course, we are not state judges. And we are bound by AEDPA. Under AEDPA’s relitigation bar, the very existence of reasonable disagreement forecloses relief. See, e.g. , Musladin , 549 U.S. at 76–77, 127 S.Ct. 649. Yet the dissenters do not acknowledge this standard, let alone explain how their analysis would be any different with or without it. See Harrington v. Richter , 562 U.S. 86, 101, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011) (reversing the Ninth Circuit because "it is not apparent how the Court of Appeals’ analysis would have been any different without AEDPA").
* * *
The principal Founding-era concern regarding the scope of Article III was that it could empower federal judges to run roughshod over state courts. See, e.g. , Brutus, Essay I (Oct. 18, 1787), in 2 THE COMPLETE ANTI-FEDERALIST 363, 366–67 (Herbert J. Storing ed. 1981). Few things bring this concern into sharper relief than using logic games in federal habeas to set free from state custody a thrice-convicted child-murderer.
The dissenters suggest we should not care about the Anti-Federalists because they "lost." Post at 184 (Costa, J., dissenting). But Judge Costa’s "winners" cared about the Anti-Federalists—so much so they wrote an entire book to respond to the Anti-Federalists’ views. See generally The Federalist (Clinton Rossiter ed. 1961); see specifically id. NO. 81, at 486 (responding to Brutus I); see also The Essential Antifederalist xiv (W.B. Allen & G. Lloyd eds., 2d ed. 2002) ("The Federalist should be read in light of the Antifederalist critique and not the other way around. As [George] Washington himself implied, if it were not for the Antifederalists, The Federalist would not be as good as it is."). And many of the reasons that compelled Madison, Jay, and Hamilton to care about the Anti-Federalists are still valid today. See Andrew S. Oldham, The Anti-Federalists: Past as Prologue , 12 N.Y.U. J. L. & Liberty (forthcoming 2019).
Judgment AFFIRMED. Habeas DENIED.
JENNIFER WALKER ELROD and CATHARINA HAYNES, Circuit Judges, joined by CARL E. STEWART, Chief Judge, concurring:
We concur in the judgment of the en banc court in this case. We write separately because we conclude that this case is resolvable based solely on the limitations on federal court habeas review as a result of AEDPA and the narrowness of Ashe v. Swenson , 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970), as well as the nuances of Louisiana law.
As is well established, and as the majority opinion explains, our review of legal decisions by state courts in this context is limited to decisions "contrary to" or involving "an unreasonable application of ... clearly established Federal law, as determined by the Supreme Court of the United States ." Majority Op. at 155 (emphasis added) (quoting 28 U.S.C. § 2254(d)(1) ). The Supreme Court, as recently as a few months ago, emphasized these limitations in Shoop v. Hill , where it vacated a Sixth Circuit decision that relied on Supreme Court precedent to conclude that habeas relief was warranted. ––– U.S. ––––, 139 S. Ct. 504, 507, 509, 202 L.Ed.2d 461 (2019) (per curiam). The problem was that the Supreme Court authority on which the Sixth Circuit relied postdated the state court’s decision. Id. at 507. The Supreme Court rejected the Sixth Circuit’s reasoning that the subsequent precedent was "merely an application of what was clearly established" by earlier Supreme Court case law. Id. at 508. So, even an extension the Supreme Court itself has made is out of bounds if that extension came after the state court decision. As the majority opinion points out, the original panel opinion in this case did and would have to extend Ashe , something we cannot do. Majority Op. at 156–58. Ashe involved different facts—namely, an explicit acquittal instead of an implied acquittal based on a conviction for a lesser offense. See Ashe , 397 U.S. at 439, 445–46, 90 S.Ct. 1189. Thus, the AEDPA analysis ends there. The principal dissenting opinion’s suggestion that the state court cited the right cases and perhaps reached a debatable result but applied the wrong reasoning is contrary to our limited AEDPA review particularly where, as here, the state court was examining state law and ruling on whether the jury’s determinations in Langley II precluded Langley’s conviction in Langley III . See Majority Op. at 163.
Although the above is enough, another straightforward basis supports affirmance: Even if, as the dissenting opinions argue, we were to accept that applying Ashe to an implied acquittal when there was an actual conviction is somehow not an extension of precedent, the Louisiana court’s conclusion under Ashe was objectively correct. See State v. Langley , 61 So. 3d 747, 757–58 (La. Ct. App. 2011) (holding that Langley had "not carried his burden of proving that the element of specific intent was actually decided" because "[i]t is possible that the jury convicted [Langley] of specific intent second degree murder"). Under Louisiana law, it is not only possible but also entirely permissible that the Langley II jury convicted Langley of second degree specific intent murder. After all, in a Louisiana criminal trial, "the jury must be given the option to convict the defendant of the lesser offense, even though the evidence clearly and overwhelmingly supported a conviction of the charged offense." State v. Porter , 639 So. 2d 1137, 1140 (La. 1994). While perhaps unique, this statutory "responsive verdict" right has existed in Louisiana law "[s]ince before the turn of the century[.]" Id.
The principal dissenting opinion overlooks this critical anomaly in Louisiana law when it concludes that the jury necessarily decided the issue of specific intent in Langley’s favor. See Principal Dissenting Op. at 179. As the majority opinion observes, the trial court explicitly instructed the Langley II jury that it could convict Langley of second degree murder based on a finding of "SPECIFIC INTENT TO KILL." Majority Op. at 169. The jury was further instructed, in line with Louisiana’s responsive verdict rule, that second degree murder was a "responsive lesser offense[ ]" to first degree murder. Majority Op. at 161. Thus, under Louisiana law as explained in the jury instructions, even if the jury found that the evidence supported a conviction for first degree murder, it could nonetheless vote to convict Langley of second degree specific intent murder. This, then, is the logical flaw in the principal dissenting opinion: it assumes that, in returning a verdict of second degree murder, the jury must have determined that the evidence was insufficient for a first degree murder conviction. But Louisiana law tells us that is simply not so. In sum, while the principal dissenting opinion emphasizes that we must "focus on ‘the actual instructions given the jury’ and assume the jury ‘would have been obligated’ to follow [them,]" it fails to assess the totality of those instructions, particularly the instruction that the jury was not required to answer Question 1 on the verdict form before proceeding to Question 2. Principal Dissenting Op. at 172–73 (quoting Turner v. Arkansas , 407 U.S. 366, 369, 92 S.Ct. 2096, 32 L.Ed.2d 798 (1972) ). The dissenting opinion’s conclusion that the jury failed to find specific intent was based on the trial judge’s oral instruction that the jury’s "verdict should be ‘guilty’ " if it was "convinced beyond a reasonable doubt that [Langley was] guilty of first degree murder[.]" Principal Dissent at 176. But this ignores the written jury instructions stating that second degree murder was a proper responsive verdict as well. The dissenting opinion’s failure to consider the jury instructions as a whole leads it to draw inferences about the jury’s verdict that do not logically follow from the totality of the circumstances. Taken as a whole, the jury instructions actually undercut those inferences.
This is not the same thing as jury nullification. True, both concepts involve a jury declining to convict the defendant of an offense that the evidence supports, which is perhaps the reason for the Louisiana Supreme Court’s observation that they are "similar." Porter , 639 So. 2d at 1140. But similarity is not "equivalence." Principal Dissent at 183 n.21. While a nullifying jury acts outside the bounds of the law, a jury convicting of a lesser offense under the Louisiana rule provides a lawful responsive verdict. Compare id. (noting that responsive verdicts are a "statutory right"), with 75A Am. Jur. 2d Trial § 667 (2019 update) ("Jury nullification refers to the jury’s power to disregard the rules of law and evidence in order to acquit the defendant[.]").
Second degree felony murder is the offense that the majority obscures with the paraphrase "something less than specific intent murder" in its explanation of the Louisiana Court of Appeal’s decision.
The principal dissenting opinion construes the Louisiana court’s jury instructions like ordinary federal jury instructions and in doing so disregards a significant nuance in Louisiana law. This runs counter to AEDPA’s goal of advancing "comity, finality, and federalism" and threatens the "mutual respect and common purpose existing between the States and the federal courts." Williams v. Taylor , 529 U.S. 420, 436, 120 S.Ct. 1479, 146 L.Ed.2d 435 (2000). To preserve that careful balance, we should adhere to Louisiana’s long-established responsive verdict rule and afford the Langley II jury’s verdict the high level of respect that it is due.
Simply put, Louisiana’s Third Circuit Court of Appeal did not unreasonably apply clearly established federal law and, based on its superior understanding of the way responsive verdicts work in Louisiana, its conclusion was objectively correct. Accordingly, the district court correctly denied relief. We therefore join in the judgment of affirmance of the district court’s denial of habeas relief.
STEPHEN A. HIGGINSON, Circuit Judge, joined by WIENER, DENNIS, GRAVES, and COSTA, Circuit Judges, dissenting:
The majority concludes that the Louisiana Third Circuit Court of Appeal reasonably rejected Ricky Langley’s argument that Ashe v. Swenson , 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970), precluded the State of Louisiana retrying the issue of Langley’s specific intent to kill. The majority says that the panel, in its now-vacated decision granting relief, enforced an unduly rigid conception of Ashe and impermissibly faulted the state court for not extending Ashe . But the majority’s opinion, ostensibly an effort to set Supreme Court precedent straight, never explains that precedent. It does not, because it cannot. To say what the governing law actually requires is to pull a thread that unravels the majority’s analysis.
Ashe preclusion operates at the level of issues ––that is, elements of an offense, rather than offenses in toto . Ashe requires reviewing courts to decide "whether a rational jury could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration." 397 U.S. at 444, 90 S.Ct. 1189. We are to decide that question by "examin[ing] the record of a prior proceeding, taking into account the pleadings, evidence, charge, and other relevant matter." Id. In identifying what the jury "necessarily determined," we are to assume a rational jury, not to speculate about "what transpired in the jury room." Yeager v. United States , 557 U.S. 110, 122, 129 S.Ct. 2360, 174 L.Ed.2d 78 (2009). We are to focus on "the actual instructions given the jury" and assume the jury "would have been obligated" to follow those instructions where they lead. Turner v. Arkansas , 407 U.S. 366, 369, 92 S.Ct. 2096, 32 L.Ed.2d 798 (1972). When the "only logical conclusion" is that the jury necessarily decided the issue in the defendant’s favor, the Double Jeopardy Clause precludes retrying the defendant on that issue. Id. at 369–70, 92 S.Ct. 2096. What the majority derides as "logic games" is thus no more, and no less, than the test that the Supreme Court has directed us to apply.
Langley faced three possible offenses of conviction at his 2003 trial that are relevant here: the charged offense, first degree specific-intent murder; and two responsive verdicts, second degree specific-intent murder and second degree felony murder.1 The jury’s verdict, in accordance with state law, was "guilty of second degree murder," not specifying the type.
The majority uses a stylized X & Y illustration in describing the issue, but it erases the difficulty by cutting the number of possible offenses from three to two.
The Louisiana Court of Appeal suggested three explanations for the jury’s verdict, concluding that it could not say whether the jury had necessarily decided the issue of Langley’s specific intent. State v. Langley , 61 So. 3d 747, 757–58 (La. Ct. App. 2011). Properly applied, Ashe ’s principles foreclose two of the three explanations, just as they compel the remaining one: Langley’s specific intent to kill was the only element of murder disputed in 2003; his lawyers successfully disputed it; the jury acquitted him of first degree specific-intent murder; hence the jury convicted him of second degree felony murder.
The Louisiana Court of Appeal’s other explanations, avoiding Ashe protection, were that the jury may have chosen second degree specific-intent murder or may simply have reached a "compromise verdict" regardless of specific intent. Langley , 61 So. 3d at 757–58. Both contravene Ashe and Turner ’s directions to assume a rational jury that follows the facts where its instructions lead. If the State had proved Langley’s specific intent, a rational jury following the instructions given here would have convicted him of first degree specific-intent murder.
Thus, to say what Ashe requires is to see that it leaves just one explanation for Langley’s 2003 conviction: acquittal on the issue of specific intent. In 2007, the Louisiana Supreme Court, relying on both state and federal law, ruled that Langley’s 2003 verdict acquitted him of first degree murder, barring retrial on that charge. State v. Langley , 958 So. 2d 1160, 1170 (La. 2007). The Louisiana Code of Criminal Procedure operated to make the jury’s verdict of second degree murder an acquittal of first degree murder. Id. (citing La. Code Crim. Proc. Ann. art. 598(A) ). As the Louisiana Supreme Court recognized, this was also in keeping with long-standing United States Supreme Court precedent recognizing implied acquittals. Id. at 1169 (citing Green v. United States , 355 U.S. 184, 193, 78 S.Ct. 221, 2 L.Ed.2d 199 (1957) ).
It is this acquittal to which Ashe issue preclusion attaches. Langley’s argument is straightforward and grounded in Supreme Court precedent: Ashe , which is a half-century old, and Green , which is even older. This Supreme Court precedent entitles Langley to habeas relief.
As the panel stated, the State may re-prosecute Langley for second degree murder under La. R.S. 14:30.1 —or for any other crime—on a theory that does not have as an essential element proof of Langley’s specific intent to kill or harm.
If the majority dealt squarely with Langley’s argument, we could perhaps have avoided much length and complication in our combined opinions. The majority does acknowledge that Langley was acquitted of first degree specific-intent murder in 2003. But the majority is unable to explain why that acquittal can bar retrial on the charge, yet not on the charge’s elements. And so the majority attempts to rationalize the state court’s decision in other ways.
In Part III(A)(2), the majority suggests that the Louisiana Court of Appeal refused to extend Ashe to implied acquittals on the theory that the law did not clearly establish that it was required to do so. But no extension was required, and the state court plainly believed that Ashe applied. It explained that the "Double Jeopardy Clause protects against successive prosecutions following acquittal or conviction" and stated the correct Ashe standard. 61 So. 3d at 757.
Next, in Part III(B)(1), the majority rationalizes the Louisiana Court of Appeal’s decision with reference to Schiro v. Farley , 510 U.S. 222, 114 S.Ct. 783, 127 L.Ed.2d 47 (1994). But Schiro simply had different facts than this case. Moreover, Schiro never appeared in the state court’s decision, in name or in substance, and it has never played a part in the State’s opposition to Langley’s habeas petition.
By relying on post hoc rationalizations that cannot be squared with what the state court actually said, the majority departs from the Supreme Court’s recent direction on review of reasoned state-court decisions: "a federal habeas court simply reviews the specific reasons given by the state court and defers to those reasons if they are reasonable ." Wilson v. Sellers , ––– U.S. ––––, 138 S. Ct. 1188, 1192, 200 L.Ed.2d 530 (2018) (emphasis added). The obligation to search for supportive reasoning obtains only when a state court issues a decision unaccompanied by any reasoning from itself or a lower state court. Id. at 1195 ; see Harrington v. Richter , 562 U.S. 86, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011). Richter ’s "could have supported" framework does not apply otherwise. Wilson , 138 S. Ct. at 1195.
The majority also departs from the Supreme Court’s constitutional command in Ashe . The majority imports extended discussion, far more than of Ashe itself, from the Second Restatement of Judgments, which the Supreme Court has never used to adjudicate an Ashe claim. In place of the straightforward Ashe inquiry explained above, the majority develops a novel set of "essential prerequisites," analyzing Langley’s claim under a framework that played no part in the Louisiana Court of Appeal’s decision or in the State’s arguments at any stage in this litigation.
In turn, the majority’s wholesale substitution of principles, embraced without either district court or adversary treatment, broadly threatens double jeopardy doctrine. Rather than dealing squarely with Langley’s argument that Ashe preclusion flows from Langley’s acquittal of first degree murder in 2003, the majority’s Restatement-based analysis sows doubt that any part of the 2003 verdict was a valid final judgment. In the process, the majority threatens a double jeopardy pillar: the "unassailable" finality of acquittals, even when "based upon an egregiously erroneous foundation." Yeager , 557 U.S. at 122–23, 129 S.Ct. 2360 (quoting Fong Foo v. United States , 369 U.S. 141, 143, 82 S.Ct. 671, 7 L.Ed.2d 629 (1962) ). Perhaps the majority’s clash with basic doctrine is inadvertent, but the risk of such clashes is why we follow precedent, rather than developing novel bodies of law without adversary treatment, in the face of contrary Supreme Court commands.
Why this avoidance of Supreme Court precedent, both old and new? Perhaps because its correct application yields an unthinkable result due to the horror of Langley’s crime. The majority accurately describes the gruesome details, which shock and disgust. As it happens, Langley’s 2003 jury had been instructed on predicate offenses for felony murder that were not enumerated in the felony murder statute at the time of Langley’s offense. The State discovered this error on the eve of the 2009 trial, which appeared to close off the felony murder route to a new second degree murder conviction. That left the State in a bind: charge lesser offenses or retry the specific intent issue decided in Langley’s favor in 2003. The State chose the latter, and here we are.
Or perhaps because several Justices have recently intimated doubts about Ashe . See Currier v. Virginia , ––– U.S. ––––, 138 S. Ct. 2144, 2149–50, 201 L.Ed.2d 650 (2018). But it remains the Supreme Court’s "prerogative alone to overrule one of its precedents," even when circuit judges disagree strongly with their foundations. State Oil Co. v. Khan , 522 U.S. 3, 20, 118 S.Ct. 275, 139 L.Ed.2d 199 (2001).
Though rejecting the State’s choice may seem unthinkable, the monstrosity of Langley’s crime does not put him beyond constitutional protection. The Constitution protects all, including the least and worst among us. Indeed, its safeguards against the profound deficiencies that marred Langley’s first two trials are the reason that the majority is able to call Langley "thrice-convicted." If commission of serious crime suffices to erase constitutional protections, many pages must be torn from the United States and Federal Reporters. But it is not in our power to abrogate constitutional law announced by the Supreme Court, nor should we do so indirectly.
The vacated panel opinion recounts this case’s long history in detail, Langley v. Prince , 890 F.3d 504, 508–14 (5th Cir. 2018), and the majority’s opinion notes the relevant points. A brief review is sufficient here. Committing the offense in 1992, Langley first stood trial in 1994, and his conviction of first degree murder was then set aside due to a flaw that, while substantial, is not significant here. See State v. Langley , 813 So. 2d 356 (La. 2002).
In 2003, the trial relevant to our Ashe inquiry took place. The State charged Langley again with first degree murder, and Langley pleaded not guilty as well as not guilty by reason of insanity. His counsel conceded that Langley had killed the victim, a boy six years old. The defense focused instead on Langley’s state of mind. Contrary to the majority’s assertion that evidence of Langley’s specific intent was overwhelming, defense counsel argued that Langley could not form the specific intent to kill because his mental illness, history of trauma, and exposure to a toxic prenatal environment had rendered him unable to understand or intend the consequences of his actions.
For example, the jury heard testimony regarding Langley’s history of mental breakdowns; his family trauma; his significant pre-natal exposure to medical drugs, alcohol, and x-rays (because months of his early gestation occurred while his mother was hospitalized after a car accident, put in a body cast, and treated intensively by doctors who did not know of the pregnancy); expert opinion on the permanent brain damage Langley may have incurred from his toxic pre-natal environment; and expert opinion on Langley’s mental illness and state of mind at the time of the killing.
The trial judge––whose misconduct would cause this conviction to be set aside ––instructed the jury on first degree murder, which consisted of (1) killing a human being (2) with specific intent to kill or to inflict great bodily harm (3) with one or more aggravating factors. See La. R.S. 14:30(A). The State pursued two possible aggravators––either that Langley was committing second degree kidnapping or that the victim was under the age of twelve. See id. 14:30(A)(5). Because the fact of the killing and the age of the victim were not contested, the State needed only to prove that Langley had the requisite specific intent.
The judge left the courtroom for significant portions of the proceedings, cut off the defense’s closing argument, refused to entertain certain objections, and generally "failed to maintain order and decorum" in the courtroom. See State v. Langley , 896 So. 2d 200, 203–07 (La. Ct. App. 2004).
Crucially for our Ashe inquiry, the trial judge instructed the jury to begin with first degree murder, the charged offense: "[I]f you are convinced beyond a reasonable doubt that [Langley] is guilty of first degree murder, your verdict should be ‘guilty.’ " The jury could then proceed to considering a lesser offense only if it were not so convinced. The judge then instructed the jury on the lesser offenses that Louisiana has deemed responsive to a charge of murder. The judge explained that second degree murder consists of either: (1) killing a human being (2) with specific intent to kill or inflict great bodily harm ("specific-intent second degree murder"), see La. R.S. 14:30.1(A)(1), or (1) killing a human being (2) while committing or attempting certain enumerated felonies ("second degree felony murder"), see id. 14:30.1(A)(2). As to second degree felony murder, the judge instructed the jury that the relevant felonies were second degree kidnapping, see id. 14:44.1, and cruelty to juveniles, see id. 14:93. The judge then told the jury: "If you are not convinced that [Langley] is guilty of first degree murder, but you are convinced beyond a reasonable doubt that [he] is guilty of second degree murder, the form of your verdict should be ‘guilty of second degree murder.’ " Thus, under these instructions, and as the Louisiana Supreme Court would later determine, a second degree murder verdict was an acquittal of first degree murder.
The judge’s oral instructions erroneously defined specific-intent second degree murder as the killing of a human being "with or without specific intent to kill or to inflict great bodily harm." (emphasis added). During deliberations, the jury received a written corrected instruction, with the consent of both parties.
The just-quoted instruction was the basis for the Louisiana Supreme Court’s conclusion that the jury acquitted Langley of first degree specific-intent murder. Langley , 958 So. 2d at 1170.
Consistent with state law, the verdict form listed the possible responsive verdicts—"guilty," "guilty of second degree murder," "guilty of manslaughter," "not guilty by reason of insanity," and "not guilty"—and instructed the jury to return one and only one of them. The jury returned a verdict finding Langley guilty of second degree murder and, by operation of state law, acquitting him of first degree murder.
The Louisiana Court of Appeal then reversed and remanded for a new trial due to the trial judge’s misconduct. See State v. Langley , 896 So. 2d 200, 212 (La. Ct. App. 2004). Significantly, the court believed that the judge’s misconduct was structural error, rendering the verdict "an absolute nullity" and permitting the State to re-try Langley for first degree murder. Id. at 210–12. The State attempted to do just that, and Langley’s motion to quash the new indictment brought the issue to the Louisiana Supreme Court, which ruled:
The instructions admonished jurors that if they were not convinced beyond a reasonable doubt "that the defendant is guilty of First Degree Murder, but you are convinced beyond a reasonable doubt that the defendant is guilty of second degree murder the form of your verdict should be guilty of Second Degree Murder." Jurors then returned a lawful, unanimous verdict convicting Langley of second degree murder. Second degree murder is a crime under the laws of Louisiana and is a responsive verdict to a charge of first degree murder.
Under these circumstance[s], and by operation of longstanding double jeopardy law, we hold that the unanimous verdict of guilty of second degree murder returned by Langley’s jury in [Langley’s second trial] implicitly acquitted him of first degree murder.
State v. Langley , 958 So. 2d 1160, 1170 (La. 2007) (citations omitted). This ruling relied on both state and federal law. The Louisiana Supreme Court read the jury instructions as requiring the jury to acquit on first degree murder before considering second, and Louisiana law provides that "[w]hen a person is found guilty of a lesser degree of the offense charged, the verdict ... is an acquittal of all greater offenses charged in the indictment." Id. at 1169–70 (quoting La. Code Crim. Proc. Ann. art. 598(A) ). The court also cited United States Supreme Court precedent that, "when a defendant is convicted of a lesser included offense and that conviction is overturned on appeal, the conviction operates as an implied acquittal of the charged crime, prohibiting the State from retrying the defendant on the original charge." Id. at 1169 (citing Green , 355 U.S. at 193, 78 S.Ct. 221 ).
A bench trial followed in 2009, with first degree murder removed from the indictment. Raising the Ashe issue, Langley’s counsel argued that specific-intent second degree murder should also be removed, because the 2003 verdict could be rationally explained only as an acquittal on the issue of Langley’s specific intent. Second degree felony murder would be left as the most serious charge. But the trial judge rejected Langley’s argument, so the indictment contained both varieties of second degree murder. The next day, however, the State orally withdrew the felony murder charge, having realized that its preferred predicate offenses, second degree kidnapping and cruelty to juveniles, were not enumerated in the felony murder statute at the time of Langley’s offense. Specific-intent second degree murder, already under the cloud of Ashe , became the State’s only route to a murder conviction.
The judge ultimately found Langley guilty of second degree murder. The ruling explicitly stated that "[t]he issue of specific intent ... is necessary for the determination of guilt," and found that the requisite specific intent was present. Langley’s counsel renewed the Ashe objection in a post-trial motion, but the judge stood by his earlier ruling. The judge then imposed the mandatory sentence of life imprisonment without parole.
On appeal, the Louisiana Third Circuit Court of Appeal issued the ruling in question here. It recognized that the Double Jeopardy Clause applied "following acquittal or conviction." Langley , 61 So. 3d at 757 (quotation omitted). It acknowledged Ashe "prohibits the state from relitigating an issue of ultimate fact that has been determined by a valid and final judgment." Id. (citing Ashe , 397 U.S. at 443, 90 S.Ct. 1189 ). It then correctly quoted the Ashe standard. Id. ("[T]o determine which facts were ‘necessarily decided’ by the general acquittal in the first trial, it is necessary to examine the record of the prior proceeding in order to determine ‘whether a rational jury could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration.’ ") (quoting Ashe , 397 U.S. at 444, 90 S.Ct. 1189 ). The court’s Ashe analysis, in full, is as follows:
When a lesser included offense to the crime charged is returned by a jury it is not always possible to determine why that verdict was reached. It is possible that the jury convicted the defendant of specific intent second degree murder. It is possible that the jury verdict was based on a jury finding under the felony-murder rule, and the jury determined there was no specific intent to kill. It is equally plausible that, given the nature of the case, the verdict was, in fact, a compromise verdict. Regardless of the jury’s thought process in this particular case, clearly the argument that the issue of specific intent was "necessarily determined" is unsupported. The defendant has not carried his burden of proving that the element of specific intent was actually decided in the previous trial.
Id. at 757–58. The Louisiana Supreme Court then declined discretionary review, 78 So. 3d 139 (La. 2012), leading Langley to federal habeas and ultimately to our court.
Ashe tells courts how to identify the issues that a jury necessarily determined, and its method is directed squarely at deciphering general verdicts:
Where a previous judgment of acquittal was based upon a general verdict, as is usually the case, this approach requires a court to examine the record of a prior proceeding, taking into account the pleadings, evidence, charge, and other relevant matter, and conclude whether a rational jury could have grounded its verdict upon an issue other than that which the defendant seeks to foreclose from consideration.
397 U.S. at 444, 90 S.Ct. 1189. We are to assume a rational jury, not to speculate about "what transpired in the jury room." Yeager , 557 U.S. at 122, 129 S.Ct. 2360. Indeed, relief under Ashe depends on the assumption that the jury acted rationally. When the jury’s verdict is "irreconcilably inconsistent"––for instance, convicting on a compound offense but acquitting on one of its predicates––the verdict has no preclusive effect. See Bravo-Fernandez v. United States , ––– U.S. ––––, 137 S. Ct. 352, 356–57, 196 L.Ed.2d 242 (2016). A court applying Ashe also assumes that the jury believed any "substantial and uncontradicted evidence of the prosecution on a point the defendant did not contest." Ashe , 397 U.S. at 444 n.9, 90 S.Ct. 1189 (quotation omitted). Finally, a court applying Ashe assumes that the jury followed its instructions. This principle is implicit in Ashe ’s concept of a rational jury, and it is explicit in Turner v. Arkansas , 407 U.S. 366, 92 S.Ct. 2096, 32 L.Ed.2d 798 (1972). There, the Supreme Court focused on "the actual instructions given to the jury" and assumed that the jury "would have been obligated" to follow them where the facts in evidence led. Id. at 369, 92 S.Ct. 2096.
When there is just a "single rationally conceivable issue in dispute before the jury," Ashe , 397 U.S. at 444, 90 S.Ct. 1189, as there is here, this can be a straightforward inquiry. At trial in 2003, the jury was instructed on three offenses relevant here: first degree specific-intent murder; second degree specific-intent murder; and second degree felony murder. The two degrees of specific-intent murder shared two elements: the killing of a human being and the specific intent to kill or inflict great bodily harm. First degree differed from second only by specifying the age of the victim––under twelve––an element not in dispute. The fact of the killing was not disputed either. Specific intent was thus the single rationally conceivable issue in dispute before the jury. If specific intent had been proven, a rational jury following the instructions given here would have been obligated to choose first degree murder. The jury did not, indicating that it had necessarily decided the issue of specific intent in Langley’s favor. As such, the jury’s choice of second degree murder can be rationally explained only as a felony murder verdict.
As noted, the Louisiana Code of Criminal Procedure undergirds these instructions. A responsive verdict of "guilty of second degree murder" operates to acquit a defendant charged with first degree murder of that offense while also convicting him of second degree murder. See La. Code Crim. Proc. Ann. arts. 598(A), 814(A)(1). That verdict is also the only mechanism by which a jury can achieve that result. If the jury had returned a verdict of "not guilty of first degree murder," the judge would have been required to reject it. Id. art. 813. By following this state law in interpreting the 2003 verdict, I do only what the Louisiana Supreme Court did in 2007. I thus cannot agree with the concurring opinion’s view of the jury instructions.
As noted at the outset, the Ashe analysis forecloses the two other possibilities suggested by the Louisiana Court of Appeal: that the jury convicted Langley of specific-intent second degree murder, or that the jury reached a compromise verdict. 61 So. 3d at 757–58. The jury instructions, if rationally followed, rule out both. The jury was told to start with first degree murder, two elements of which were uncontested. If the State had proved specific intent, the remaining element, the jury would have been obligated to convict Langley of first degree murder, not second degree murder. Likewise, the jury instructions did not suggest or permit a compromise verdict on a lesser offense despite convincing evidence of a greater offense. The Louisiana Court of Appeal’s speculation about a compromise cannot be squared with Turner ’s teaching to treat juries as "obligated" to follow their instructions. Consequently, the Louisiana Court of Appeal’s alternative explanations were objectively unreasonable applications of Ashe and its progeny.
There is thus only one rational explanation of the jury verdict’s acquittal of first degree murder and conviction of second degree murder: the jury acquitted on the issue of specific intent, hence convicted Langley of felony murder. Langley’s retrial in 2009 should not have been allowed to proceed on the charge of second degree specific-intent murder. The resulting conviction therefore violates the Double Jeopardy Clause, entitling Langley to habeas relief. III
The majority’s reasons for not disturbing the Louisiana Court of Appeal’s decision depend either on new rationales not employed by the state court or on avoidance of what Ashe requires. Each move the majority makes is therefore a wrong step on the landscape of Supreme Court precedent.
The majority begins by framing the panel’s ruling as requiring an extension of Ashe : "A fairminded jurist could conclude the rule clearly established in Ashe does not apply to a conviction rather than a general acquittal." Supra , Part III(A)(2). This statement is puzzling at first glance, because the Louisiana Supreme Court ruled that Langley was acquitted of first degree specific-intent murder. Langley , 958 So. 2d at 1170. The 2009 trial proceeded on that ruling, and any possible preclusion would attach to that acquittal. The majority means to say that Langley received an implied acquittal of first degree murder alongside his conviction of second degree murder, and the law is not clearly established that Ashe preclusion may arise from such a verdict.
The majority is quite right to hedge that "[w]e may or may not find this distinction persuasive." But it is quite wrong to say that "the last reasoned state court decision found it persuasive." On the contrary, the Louisiana Court of Appeal plainly believed that Ashe applied. 61 So. 3d at 757. It cited Ashe , quoted the standard, and asked the right question––albeit a question it answered unreasonably. If the state court had any doubt that Ashe applied, it did not say so. Consequently, the majority has contrived a rationale for the state court’s decision that is incompatible with the reasoning that the state court actually gave.
The Supreme Court’s decisions give us no license to conduct AEDPA review this way. Following Wilson v. Sellers , the mode of our analysis under the "unreasonable application" prong of 28 U.S.C. § 2254(d)(1) depends on whether the state-court decision is accompanied by any reasoning. 138 S. Ct. at 1191–92. If no reasoning accompanies the decision, we are to "determine what arguments or theories ... could have supported[ ] the state court’s decision." Richter , 562 U.S. at 102, 131 S.Ct. 770. But if any reasoning does, whether from the issuing court or a lower state court, the "federal habeas court simply reviews the specific reasons given by the state court and defers to those reasons if they are reasonable ." Wilson , 138 S. Ct. at 1192 (emphasis added).
A nuance not relevant here is the possibility that an unexplained merits decision by a higher state court rested on reasoning different from that expressed by a lower state court. Wilson addresses this possibility through a rebuttable-presumption framework. 138 S. Ct. at 1196–97. That nuance does not arise in this case. The Louisiana Supreme Court’s decision was a discretionary denial of review, so the Louisiana Court of Appeal’s decision was the last decision on the merits in the state system.
Wilson bears on two issues that had divided the circuits. The first issue is the proper object of a federal habeas court’s focus when the last state court to adjudicate the merits of a post-conviction claim did not explain its reasoning but a lower state court did. Wilson squarely answers the question: "the federal court should ‘look through’ the unexplained decision to the last related state-court decision that does provide a relevant rationale." 138 S. Ct. at 1192. The second issue is the method of reviewing reasoned state-court decisions under the "unreasonable application" prong of AEDPA. Wilson brought clarity to this second issue. As noted, Wilson tells us that AEDPA review of reasoned decisions is a "straightforward inquiry when the last state court to decide a prisoner’s federal claim"––as here––"explains its decision on the merits in a reasoned opinion." Id. "[The] federal habeas court simply reviews the specific reasons given by the state court and defers to those reasons if they are reasonable ." Id. (emphasis added).
Compare, e.g. , Dennis v. Sec., Pa. Dept. of Corr. , 834 F.3d 263, 281–82 (3rd Cir. 2016) (en banc) ("[F]ederal habeas review does not entail speculating as to what other theories could have supported the state court ruling when reasoning has been provided, or buttressing a state court’s scant analysis with arguments not fairly presented to it."), with, e.g. , Evans v. Davis , 875 F.3d 210, 216 (5th Cir. 2017) ("[We consider] not only the arguments and theories the state habeas court actually relied upon to reach its ultimate decision but also all the arguments and theories it could have relied upon.").
The Supreme Court issued Wilson in April 2018, after the panel heard oral argument in this case but before publishing its opinion. The panel acknowledged Wilson ’s likely impact on AEDPA review but applied Richter ’s "could have supported" framework in an abundance of caution. Langley , 890 F.3d at 515 & n.15. The full court subsequently requested and received supplemental briefing on Wilson .
Another panel of this court has also recognized Wilson ’s significance. See Thomas v. Vannoy , 898 F.3d 561, 568 (5th Cir. 2018) (acknowledging that the "continued viability" of the Fifth Circuit’s approach was "uncertain" after Wilson ). See also Meders v. Warden, Ga. Diagnostic Prison , 911 F.3d 1335, 1349 (11th Cir. 2019) ("What [Wilson ] means is we are to focus not merely on the bottom line ruling of the decision but on the reasons , if any, given for it.") (emphasis added). Commentators have recognized its significance as well. See Brian R. Means, Federal Habeas Manual § 3:70 (2018) ("[T]he Supreme Court [has] apparently settled the matter: the ‘fill the gaps’ aspect of Richter —considering grounds that could have supported the state court’s decision—does not extend beyond unexplained rulings to reasoned state court decisions."); Leading Case, Antiterrorism and Effective Death Penalty Act—Habeas Corpus—Scope of Review of State Proceedings— Wilson v. Sellers, 132 Harv. L. Rev. 407, 412–13 (2018) ("[T]he Wilson Court limited one of the harshest pieces of Richter ’s legacy––the practice of courts imagining all possible bases for denying relief––to Richter ’s specific procedural posture, thus sparing habeas petitioners from a burden that AEDPA need never have imposed on them.").
That direction governs us here. The Louisiana Court of Appeal explained its reasoning for denying relief. That reasoning unreasonably applied Ashe and its progeny. Our analysis should then proceed to de novo review of the petitioner’s claim. See Salts v. Epps , 676 F.3d 468, 480 (5th Cir. 2012). By instead interposing a new rationale not given by the state court and not compatible with the reasons it did give, the majority runs afoul of Wilson ’s direction.
The majority’s lengthy discussion of Schiro v. Farley , 510 U.S. 222, 114 S.Ct. 783, 127 L.Ed.2d 47 (1994), is likewise out of place. The Louisiana Court of Appeal never employed any reasoning that could be said to flow from Schiro . The State did not brief Schiro below or to the panel, and at oral argument before the panel, counsel for the State made no use of Schiro when given the chance. The majority’s discussion is thus another effort to supply novel reasoning, contra Wilson , in support of the state court’s decision. Under de novo review of Langley’s claim, the majority’s reliance on Schiro also fails to provide meaningful support to the state court’s decision. In part, this is because Schiro ’s facts simply differ in determinative ways. For one, this case concerns issue preclusion between successive trials; Schiro concerned issue preclusion between the guilt and sentencing phases of a single trial. 510 U.S. at 225–26, 114 S.Ct. 783. For another, the jury instructions in this case directed jurors to begin with the single charge of first degree murder and, as described above, work their way down through the list of responsive verdicts. In Schiro , three counts of murder were charged, and the instructions did not clearly direct jurors to proceed from greater to lesser offenses as the instructions did here. Id. at 233–34, 114 S.Ct. 783. Finally, the jury instructions in this case were clear that the jury could convict Langley for second degree murder without finding specific intent, via the felony murder option. The jury instructions in Schiro were ambiguous on that very point. Id. at 234, 114 S.Ct. 783.
See Oral Argument at 30:54, Langley v. Prince , 890 F.3d 504 (5th Cir. 2018), http://www.ca5.uscourts.gov/OralArgRecordings/16/16-30486_10-4-2017.mp3.
The majority says its use of Schiro to uphold the state court’s decision does not contravene Wilson because state courts are not required to cite the correct case names. That is hardly the issue here.
Schiro also creates trouble for the majority’s other post hoc rationalization of the state court’s decision. As noted, the majority rests its holding on the idea, never espoused by the state court, that Ashe ’s application to implied acquittals accompanying convictions is not clearly established. But Schiro suggests that Ashe does apply. In its discussion, the Court first cited long-standing precedent on implied acquittals. 510 U.S. at 236, 114 S.Ct. 783 ("We have in some circumstances considered jury silence as tantamount to an acquittal for double jeopardy purposes.") (citing Green , 355 U.S. at 190–91, 78 S.Ct. 221 ; Price v. Georgia , 398 U.S. 323, 329, 90 S.Ct. 1757, 26 L.Ed.2d 300 (1970) ). It then added that "[t]he failure to return a verdict does not have collateral estoppel effect, however, unless the record establishes that the issue was actually and necessarily decided in the defendant’s favor." Id. Indeed, the Court conducted an Ashe analysis in Schiro . Id. at 234–36, 114 S.Ct. 783. It simply ruled that Schiro had failed to carry his burden. Id. at 236, 114 S.Ct. 783. Consequently, if Schiro adds anything here, its weight belongs on Langley’s side of the scale.
Finally, there is the majority’s issue-preclusion analysis. Supra , Part III(C). It is here that the majority’s refusal to explain the Ashe analysis required by Supreme Court precedent is most glaring. Rather than look to Ashe , Yeager , or other Supreme Court law, the majority instead imports the Second Restatement of Judgments. From the Restatement, the majority derives new "essential prerequisites" for issue preclusion to obtain, which debut in the majority’s opinion without any adversarial treatment at any stage in this litigation. While the Supreme Court has cited the Restatement’s issue-preclusion principles in various contexts, it has never employed the novel framework advanced by the majority to adjudicate an Ashe claim. The majority’s misbegotten doctrinal innovation cloaks the majority’s departure from governing law, disrupts settled double jeopardy doctrine, and is likely to confuse the state and federal judges of this circuit as they adjudicate Ashe claims in the future.
To justify its creation of a novel Restatement-based framework, the majority cites Bravo-Fernandez v. United States , ––– U.S. ––––, 137 S. Ct. 352, 196 L.Ed.2d 242 (2016). There, the Restatement appeared in the Court’s discussion of general principles, but the Court concluded Ashe did not apply, so there was no Ashe claim to adjudicate. Id. at 357–58, 362–33. The majority also cites Bobby v. Bies , 556 U.S. 825, 129 S.Ct. 2145, 173 L.Ed.2d 1173 (2009). That case concerned the effect of a change in the law on a single, concluded proceeding; double jeopardy protection did not arise. Id. The majority is blurring the distinction between Supreme Court decisions about whether Ashe should apply in a given situation and its decisions actually adjudicating an Ashe claim. The majority uses the former decisions to obscure what the latter decisions require us to do in this case.
To start, the majority misstates the fundamental question as being what the jury "actually determined," citing the Restatement, rather than as what it "necessarily decided." See Yeager , 557 U.S. at 119–20, 129 S.Ct. 2360. This difference is subtle but significant, because the latter formulation trains the reviewing court’s attention on a rational jury adhering to its instructions, and not on speculation about "what transpired in the jury room." Id. at 122, 129 S.Ct. 2360. The majority makes the very error condemned by the Supreme Court in Yeager when it speculates that Langley’s jury chose second degree murder because it heard defense counsel’s plea for mercy and because it wanted to avoid a capital punishment hearing.
Similarly, choosing novel Restatement-based standards permits the majority to deploy a misrepresentation of Louisiana responsive verdict law without acknowledging the Supreme Court precedent that would rule it out. The majority describes the specific-intent second degree murder instruction as a "concededly valid option." Indeed, like many states, Louisiana recognizes "that a defendant, when charged with a crime for which the Legislature has provided a responsive verdict, has the statutory right to have the jury characterize his conduct as the lesser crime." State v. Porter , 639 So. 2d 1137, 1140 (La. 1994). Louisiana treats "the jury’s prerogative to return a responsive verdict similar to the jury’s power of nullification," available to the jury "even though the evidence clearly and overwhelmingly supported a conviction of the charged offense." Id.
See, e.g. , Tex. Code Crim. Proc. § 37.09 (lesser included offenses); Wortham v. State , 412 S.W.3d 552, 557–58 (Tex. Crim. App. 2013) (holding that "[a]nything more than a scintilla of [relevant] evidence entitles the defendant to [a jury instruction on] the lesser charge").
This equivalence drawn by Porter between a jury’s choice of a responsive verdict and jury nullification rebuts the concurring opinion’s attempt to distinguish the two.
But the existence of responsive verdicts does not affect the Ashe analysis, which assumes a rational jury that follows its instructions. Given the secrecy of the jury room, the possibility of a nullification verdict is ever-present. Accounting for it in the Ashe analysis would make it impossible to say what a jury "necessarily determined," and so would effectively eliminate Ashe , as our court has long recognized. See United States v. Tran , 433 F. App'x 227, 231 (5th Cir. 2011) ("[I]f we consider jury nullification as a basis on which the jury might have acquitted ... we would in effect be eliminating the entire doctrine of collateral estoppel and greatly weakening the protection against double jeopardy.") (quoting United States v. Leach , 632 F.2d 1337, 1341 n.12 (5th Cir. 1980) ). The majority’s indulgence of that possibility runs counter to Ashe ’s rational-jury assumption, Turner ’s assumption that juries adhere to instructions, and Yeager ’s direction to avoid speculation about what transpired in the jury room.
The majority’s use of the Restatement causes still more mischief. Avoiding Langley’s argument that his first degree murder acquittal is the source of his relief under Ashe , the majority suggests that the reversal of Langley’s 2003 second degree murder conviction, due to trial judge misconduct, rendered the result of the 2003 trial not a "valid and final" judgment. It is of course true that a conviction vacated due to trial error does not preclude retrial on the same offense. But the majority dangerously disregards Supreme Court precedent, old and new, by suggesting that the 2003 verdict could have no preclusive effect on the 2009 trial. It is a pillar of double jeopardy doctrine that the finality of an acquittal is "unassailable" even if it is "based upon an egregiously erroneous foundation." Yeager , 557 U.S. at 122–23, 129 S.Ct. 2360 (quoting Fong Foo , 369 U.S. at 143, 82 S.Ct. 671 ). The Court’s recent decision in Bravo-Fernandez drives the point home. ––– U.S. ––––, 137 S. Ct. 352 (2016). Defendants Bravo and Martinez were convicted of a bribery offense but acquitted of conspiring to commit and traveling to commit that bribery, a classic "irreconcilably inconsistent" verdict. Id. at 362–64. Instructional error infected the conviction, so it was reversed. Due to the inconsistency in the verdict, the Court rejected Bravo and Martinez’s argument that their conspiracy and travel acquittals should preclude retrial for bribery. But the Court was clear that the Double Jeopardy Clause "forever bars" retrial on the acquitted charges, no matter the trial error. Id. at 365–66. The majority’s doubts about the finality of the 2003 verdict cannot be reconciled with Bravo-Fernandez and the well-established law to which it adhered.
The majority does acknowledge what it cannot avoid: the Louisiana Supreme Court’s ruling that the 2003 verdict impliedly acquitted Langley of first degree murder, barring retrial on that charge. But the majority is unable to explain why the implied acquittal can bar retrial on that charge but not the charge’s elements. Langley’s specific intent was the "single rationally conceivable issue in dispute before the jury," Ashe , 397 U.S. at 444, 90 S.Ct. 1189, and so the jury’s acquittal of first degree murder barred retrial on that element of the charge, just as it barred retrial on the charge itself. The majority cannot or will not say this, and the price of the majority’s avoidance is a blow dealt to the edifice of Supreme Court law.
* * *
Under the Double Jeopardy Clause, the verdict rendered by the jury in 2003 prohibited the State of Louisiana retrying the issue of Langley’s specific intent to kill or inflict great bodily harm. Langley’s 2009 conviction for specific-intent second degree murder therefore should not stand. Accordingly, I would reverse the district court’s judgment and remand this case with instructions to grant Langley’s petition for a writ of habeas corpus, leaving the State free to retry Langley on charges that do not require proof of his specific intent. Because the majority sidesteps numerous Supreme Court precedents and clashes with others in order to avoid that result, I dissent.
GREGG COSTA, Circuit Judge, joined by WIENER and HIGGINSON, Circuit Judges, dissenting:
I had thought the Anti-Federalists lost. But see Maj. Op. at 170. What is more, it is ironic to invoke their rejected constitutional vision in defense of a decision that undermines one of the Anti-Federalists’ most fervent beliefs: the fundamental role of juries. MICHAEL J. KLARMAN, THE FRAMERS’ COUP 350 (2016); HERBERT J. STORING, WHAT THE ANTI-FEDERALISTS WERE FOR 18–19 (1981). As a leading Anti-Federalist inveighed, "jury trials, which have so long been considered the surest barrier against arbitrary power, and the palladium of liberty, with the loss of which the loss of our freedom may be dated, are taken away by the proposed form of government." The Antifederalist No. 83 (Luther Martin), in THE ANTIFEDERALIST PAPERS 241, 241 (Morton Borden ed., 1965 ). One took it even further: "O! my fellow citizens, think of this while it is yet time, and never consent to part with the glorious privilege of trial by jury, but with your lives." Essay of A Democratic Federalist (Oct. 17, 1787), in 5 THE FOUNDERS’ CONSTITUTION 354, 355 (Philip B. Kurland & Ralph Lerner eds., 1987). And in contrast to the Anti-Federalists’ unsuccessful criticisms of the independence of federal judges and their power to review state court rulings, see Maj. Op. at 170 (citing Brutus Essay I), the Anti-Federalists’ campaign for jury rights was a success: not in defeating the Constitution, but in amending it. See U.S. CONST. amends. VI, VII.
So important was the jury right the Anti-Federalists fought for that, until the early twentieth century, a defendant charged with serious crimes could not be "tried in any other manner than by a jury of twelve men." Home Ins. Co. of New York v. Morse , 87 U.S. 445, 451, 20 Wall. 445, 22 L.Ed. 365 (1874) (citing Cancemi v. People , 18 N.Y. 128 (1858) ); see also Patton v. United States , 281 U.S. 276, 298, 50 S.Ct. 253, 74 L.Ed. 854 (1930) (reversing course and allowing a defendant to waive the jury). As the first Justice Harlan explained in rejecting the view that a defendant could agree to waive the requirement of a full jury, "the wise men who framed the constitution of the United States and the people who approved it were of the opinion that life and liberty, when involved in criminal prosecutions, would not be adequately secured except through the unanimous verdict of twelve jurors." Thompson v. Utah , 170 U.S. 343, 353, 18 S.Ct. 620, 42 L.Ed. 1061 (1898). A leading modern scholar reaches the same conclusion about the original understanding: A jury had to decide felony trials; bench trials were not allowed. See AKHIL AMAR, THE BILL OF RIGHTS 104–08 & nn. 97, 102 (1998) (emphasizing the mandatory Article III language that "trial of all crimes ... shall be by jury" as well as the writings of both Federalist and Anti-Federalists who viewed the jury guarantee as a structural provision and not just an individual right); see also Cancemi , 18 N.Y. at 138 (rejecting defendant’s ability to waive 12-member jury because that would also allow a defendant to agree to "trial committed to the court alone," which the common law did not permit); Recent Development, Accused in Multiple Prosecution Held to Have Absolute Right to Waive Jury Trial , 59 COLUM. L. REV . 813, 814 (1959) ("Until shortly after the turn of the century, federal courts and most state courts applied the common law rule that a jury trial can not be waived in a felony case in which the defendant enters a plea of not guilty."); Note, Waiver of Constitutional Right to Twelve Jurors , 9 HARV. L. REV . 353 (1895) (similar).
Yet the majority opinion lets a judge’s finding of specific intent override a jury’s earlier determination that this required mens rea was not proven. That undermines both the right to a jury and the protection against double jeopardy. As the Anti-Federalists recognized, the latter is essential to the former. See Brutus Essay XIV (Feb. 28, 1788), in THE ANTIFEDERALIST PAPERS , supra , at 234, 235 (lamenting the possibility of "a second hearing" on appeal after acquittal by a jury); see also AMAR , supra , at 96 (explaining that the Double Jeopardy Clause "dovetails with the Sixth Amendment jury right" because it protects "the integrity of the initial petit jury’s judgment"). If the state can keep retrying someone until it achieves its desired result, then the jury right that both the Federalists and Anti-Federalists cherished, see U.S. CONST. art. III (guaranteeing jury in criminal cases); Federalist No. 83, at 467 (Alexander Hamilton) (Clinton Rossiter ed., 1999) ("The friends and adversaries of the plan of the convention, if they agree in nothing else, concur at least in the value they set upon trial by jury ...."), is no right at all.