Tyrik McCLELLAN, Petitioner–Appellee, v. Lloyd RAPELJE, Respondent–Appellant.
In this Michigan criminal case, the Petitioner, Tyrik McClellan, at age 19, was convicted of first degree murder of the victim, Nathan Auld, in connection with a barroom conflict between two groups. McClellan received a mandatory life sentence without parole and a consecutive two-year sentence for a firearm conviction. The district court granted a writ of habeas corpus on the ground that McClellan's defense lawyer did not provide the effective assistance of counsel required by the Sixth Amendment because the lawyer did not interview numerous eyewitnesses who would have testified that McClellan acted in self defense. On appeal, we consider three sets of issues: first, on the merits, the State asserts that the defense lawyer did not violate the ineffective assistance of counsel standard set out in Strickland v. Washington, 466 U.S. 668 (1984); second, the State asserts that the district court should not have reached the merits because McClellan defaulted his claim by not raising it in a procedurally correct way; and third, the State asserts that the Antiterrorism and Effective Death Penalty Act, “AEDPA,” and recent Supreme Court cases interpreting it, Harrington v. Richter, 131 S.Ct. 770 (2011), and Cullen v. Pinholster, 131 S.Ct. 1388 (2011), so limit the federal courts' authority to decide habeas cases on the merits that McClellan's petition must be dismissed. We will take up these three sets of issues in the order followed by the district court. After considering these three sets of issues, we agree with the district court decision and affirm.
I. The Merits of the Ineffective Assistance of Counsel Claim
In a thorough opinion, the district court recited the evidence introduced at the original trial in 2002 and the countervailing evidence offered at the evidentiary hearing in federal district court on April 6, 2011. The shooting death occurred outside a bar in Detroit in the early morning hours of December 20, 2001. There was a heated altercation inside the bar between two groups, both of whom were asked to leave. Outside the bar, McClellan shot Auld. At the trial, he claimed self defense, but he neither testified nor offered any proof. Six of Auld's companions and two employees of the bar testified for the prosecution. There was confusion in the testimony as to whether Auld had rushed McClellan. The prosecution's theory, apparently accepted by the jury, was that the shooting was premeditated because the evidence showed that McClellan had gone to his car after leaving the bar and retrieved his gun. Defense counsel interviewed none of the witnesses and put on no witnesses in McClellan's defense. The jury rendered its verdict based entirely on the prosecution's case.
The evidence introduced at the federal habeas evidentiary hearing ten years later, if believed, disproved the State's theory of premeditation. McClellan and eight witnesses testified that McClellan had taken the gun into the bar and did not get it from his car after leaving the bar. This evidence disproves the State's theory of premeditation. The witnesses corroborated McClellan's story that Auld, who was 280 pounds and over six feet tall, rushed him with a chrome object in his hand that appeared to be a gun and that McClellan then pulled out his gun and fired at Auld. The defense lawyer at the original trial, Jeffrey Edison, also testified at the evidentiary hearing that he did not interview the witnesses or call McClellan because he believed he could undermine the prosecution's case through cross-examination of the prosecution witnesses.
In issuing the writ, the district court found that at the original trial no witness testified “that they saw Auld with a weapon.” However, at the federal hearing several witnesses “testified that Auld was armed” both before he was shot and afterward, and the district court found they would have made effective witnesses for McClellan if they had been interviewed and called at trial. Opinion, p. 17.
Based on the evidence, the district court found that “the record demonstrates that defense counsel made a decision that he could prevail on his self-defense theory with only the prosecutor's witnesses without ever conducting any investigation.” Id. at 18. There was strong evidence presented by “the defense witnesses' testimony about how they snuck Petitioner into the bar with a handgun.” Id. at 22. The district court found that the evidence, if offered at trial, “would have undermined the prosecutor's argument that Petitioner premeditated the murder because he retrieved the handgun after he left the bar.” Id.
On the ineffective assistance of counsel issue, the district court concluded:
But when, as here, it turns out that the accused's attorney never bothered to investigate what members of his client's group would say, and it turns out that their version would have significantly supported a claim of self defense, it follows that the attorney was not acting as “counsel” as contemplated by the Sixth Amendment and that his failings undermine confidence in the outcome of the trial. Therefore, Petitioner has demonstrated entitlement to habeas relief based on his claim of ineffective assistance of trial counsel.
Id. at 24.
The State argues on appeal that McClellan has not demonstrated deficient performance because “Edison made strategic decisions on how to proceed in this case—deciding that it was wiser to hold the prosecution to its burden rather than calling witnesses to the stand, including McClellan himself.” Resp. brief, p. 59. The State also argues that “[d]eficient performance—if it occurred—did not lead to prejudice on the facts of this case,” Id. at 65. Citing the fact that defense counsel was “an experienced criminal defense attorney, with 35 years of practice, and 90% of his practice focused on criminal defense,” the State seems to suggest that this experience adds to the deference we should pay to counsel's strategy of not interviewing or calling any defense witnesses. We agree with the reasoning of the district court that trial counsel was grossly deficient in his failure to interview defense witnesses.1 See Wiggins v. Smith, 539 U.S. 510, 521 (2003) (“[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation. In other words, counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” (internal quotation marks omitted)); English v. Romanowski, 602 F.3d 714, 728 (6th Cir.2010) (“Yet English's attorney admitted that he failed to interview any of the witnesses outside of the courthouse or prior to trial. Counsel could not have developed a reasonable trial strategy since he based his decisionmaking on what counsel guessed the witnesses might say in the absence of a full investigation, and not on what investigation reveals witnesses will actually testify to.” (internal quotation marks and alterations omitted)).
II. Procedural Default by Appellate Counsel
In addition to its argument that trial counsel was not deficient in his representation, the State defended the trial verdict under the doctrine of “procedural default.” The State initially asserted as an independent, non-merits, state procedural ground for dismissing the habeas petition that the state courts themselves “did not reach the merits” of the Sixth Amendment ineffectiveness claim because appellate counsel for McClellan did not raise the issue on direct appeal or in an alternative Michigan proceeding sometimes referred to as a “Ginther ” evidentiary hearing, a procedure created under Michigan Court Rule 7.211(C)(1). The district court ruled against the State on grounds that McClellan had established the technical requirements of “cause” and “prejudice” to overcome procedural default under the “cause and prejudice” doctrine established in Murray v. Carrier, 477 U.S. 478, 488–89 (1986).
Again in a thorough decision, the district court explained its reasoning that “procedural default”—based on appellate counsel's failure to raise ineffective assistance of counsel—does not bar habeas review on the merits of McClellan's ineffective assistance of counsel claim. The district court explained that appellate counsel, Mr. Simon, “had no independent recollection of Petitioner's appeal. But both Petitioner and his aunt testified that they sent letters to Simon informing him of the basis for investigating a possible ineffective assistance of trial counsel claim.” Opinion, p. 13. The district court found their testimony to be credible and supported by copies of the letters they sent to Simon. The court also found “that Simon completely ignored this lead—all of the defense witnesses testified that they were never contacted by Simon.” Id.
The court found prejudice as well because “there is a reasonable probability that had the state court been presented with the evidence that was presented to this Court, Petitioner's conviction for first-degree murder would have been reversed.” Id. at 14. We agree that once a reasonable judge found out about how this case was handled by counsel in the Michigan trial and appellate courts, it would find prejudice. Both trial and appellate counsel offered the Michigan courts nothing significant in the way of self defense, when in fact there were many witnesses whose testimony could raise a reasonable doubt in the case. Reasonable judges, if they had known about the problem, would not have allowed this kind of gross miscarriage of justice to pass unnoticed and unremedied.
III. The State's Post-hearing Motion for Reconsideration
The State reversed its position after initially losing its case in the district court. It made a new claim in its motion for reconsideration that there was, in fact, no “procedural default,” but rather that the Michigan Court of Appeals reached and decided the merits of McClellan's Sixth Amendment ineffective assistance of counsel claim. The state trial court dismissed McClellan's habeas petition on “procedural default” grounds. People v. McClellan, No. 02–3088 (Mich. Cir. Ct. for Wayne Cnty. July 19, 2006) [hereinafter Trial Court Opinion]. The State's argument now is that the Michigan Court of Appeals said in its order of August 30, 2007, that McClellan's application for leave to appeal the procedural default ruling is denied “for lack of merit in the grounds presented.” People v. McClellan, No. 279129 (Mich.Ct.App. Aug. 30, 2007). In reversing its position on procedural default, the State argued in its motion for reconsideration that under the recent case of Harrington v. Richter, 131 S.Ct. 770 (2011), the federal court must hold that the Michigan Court of Appeals denied the ineffectiveness claim on the merits and that the recent case of Cullen v. Pinholster, 131 S.Ct. 1388 (2011), prevents a federal evidentiary hearing that would allow facts to be considered that are contrary to the facts before the state court.
The district court denied reconsideration because the Michigan Court of Appeals clearly did not reach the merits of the Sixth Amendment ineffectiveness claim. It said that the ground presented on appeal had to be about procedural default, not the merits of the ineffective assistance claim, because procedural default was the ruling appealed from. The district court explained:
There is reason to think that the Michigan Court of Appeals did not reach the merits of Petitioner's claim. As both parties acknowledged in their post-hearing briefs, the trial court explicitly found Petitioner's ineffective assistance of counsel claim to be procedurally defaulted. Under Ylst [Ylst v. Nunnemaker, 500 U.S. 793 (1991) ]—the very case the Richter court directed courts to consider when deciding whether to apply its presumption of a merits adjudication—if the last state court judgment adjudicating a claim contains no reasoning, a federal habeas court must look to the last reasoned state court judgment rejecting the federal claim and apply a presumption that later unexplained orders upholding the judgment or rejecting the same claim rested upon the same grounds. Id. at 803. Under Ylst, therefore, there is a presumption that the Michigan Court of Appeals' unexplained order rests on the same non-merits basis relied on by the state trial court.
Opinion and Order Denying Respondent's Motion for Reconsideration, pp. 3–4, July 11, 2011. The difficulty with the State's new argument is that the State concedes forthrightly that the Michigan Court of Appeals panel “did not have the lower court record when it rendered its decision.” Resp.'s Supplemental Brief, at p. 1 (Oct. 25, 2012).
We recently explained that “absent some indication or Michigan procedural principle to the contrary, we must presume that an unexplained summary order is an adjudication on the merits for AEDPA purposes.” Werth v. Bell, 692 F.3d 486, 493 (6th Cir.2012) (internal quotation marks and alterations omitted). There are thus two avenues by which a petitioner can overcome this presumption. One avenue, whether there is a procedural principle to the contrary, concerns the procedural framework established by the state with respect to postconviction review generally. At issue here is the same phrase utilized by the Michigan courts that was reviewed in Werth. Because we concluded in Werth that no Michigan procedural principle by itself allows us to construe the phrase “for lack of merit in the grounds presented” as a denial on procedural grounds, McClellan cannot overcome the presumption on this ground. Id. at 493–94. Our holding in Werth does not, however, automatically foreclose the second avenue of relief for McClellan, as it requires analysis of the procedural history and factual underpinnings of each individual petitioner's case. Id. at 494. Unlike in Werth, McClellan offers us a “reason to think some other explanation for the state court's decision is more likely,” and thus we must consider this reason and determine whether it overcomes the presumption. Id. (internal quotation marks omitted).
Two Michigan state courts reviewed McClellan's post-conviction petition, and the available evidence establishes that each of these courts denied the petition on procedural-default grounds. The Michigan trial court, the first to review McClellan's petition, stated this explicitly in its order. Trial Court Opinion at 1–2. The trial court began its analysis as follows: “Before considering the validity of Defendant's legal arguments, the court must determine whether the defendant has satisfied the two-pronged ‘cause and prejudice’ requirement of” the Michigan procedural-default statute. Id. at 1. The trial court, after engaging in a brief analysis, concluded that “Defendant has failed to show either ‘good cause’ or ‘actual prejudice’ as required by” the Michigan procedural-default statute. Id. at 2 (internal quotation marks omitted).
The second court to review McClellan's petition, the Michigan Court of Appeals, “did not have the lower court record when it rendered its decision,” as conceded by the State. Resp.'s Supplemental Br. at 1.2 In other words, the Michigan Court of Appeals reviewed McClellan's petition, and the trial court opinion, which explicitly denied relief on procedural-default grounds.3 The only reasonable inference to be drawn from the plain language of the documents and the State's concession is that the Michigan Court of Appeals could not have denied McClellan's petition on the merits of the ineffectiveness claim.4 Applying the analysis outlined in Werth, McClellan has provided sufficient evidence to rebut the presumption and demonstrate that, in this instance, the Michigan Court of Appeals in using the words “for lack of merit in the grounds presented” did not reach the merits of the Sixth Amendment ineffectiveness claim.
We therefore disagree with the dissent's statement that the “grounds presented” in the Michigan Court of Appeals was the ineffective assistance of counsel issue now presented to us. We agree with the district court and the Michigan Attorney General's initial view that the issue there was procedural default, not the decision of any federal constitutional issues on the merits. It seems obvious that a factual question cannot be considered as “a ground presented” to and decided by an appellate court when it was (1) not decided by the trial court, (2) not the subject of any findings of fact made by the trial court, and (3) not included in any trial court record filed or presented to the appellate court. A factual issue cannot be created for an appellate court out of thin air without any trial court record concerning the issue before it and without any reference to the issue in the appellate court's decision. It cannot be made up like a will-o'-the-wisp in a fairy tale. As many courts have concluded: “Whether a claim has been adjudicated on the merits is a case specific inquiry” and depends on “any relevant factual findings made by the state courts.” Winston v. Pearson, 683 F.3d 489, 496 (4th Cir.2012); accord Sellan v. Kuhlman, 261 F.3d 303, 312–13 (2d Cir.2001) (collecting cases).
There were no relevant factual findings in the Michigan courts on the issue before us, and hence for the reasons conceded by the State in its supplemental brief and the reason given by the district court, we believe that the Michigan Court of Appeals in using the words “for lack of merit in the grounds presented” did not reach the merits of the Sixth Amendment ineffectiveness claim. The State was correct in its initial submission to the district court that the state courts declined to reach the merits because they found procedural default based upon the failure of appellate defense counsel to even raise the Sixth Amendment issue. Since there was no decision on the merits concerning ineffective assistance of counsel, the Cullen v. Pinholster case does not prohibit the consideration of evidence on the merits in a later federal evidentiary hearing. Pinholster only applies to limit consideration of additional evidence when the state court has previously decided the same merits issue later presented to the federal court.
Accordingly, the judgment of the district court is affirmed.
The district court here erred in three respects. First, by concluding that the Michigan Court of Appeals did not adjudicate McClellan's ineffective assistance of counsel claim on the merits; second, by considering additional evidence on that claim from a hearing held by the district court; and third, by concluding that McClellan's claim warranted relief.
The Michigan Court of Appeals' denial of McClellan's claims “for lack of merit on the grounds presented” was a merits adjudication subject to AEDPA deference. Accordingly, under Cullen v. Pinholster, the federal district court was limited to considering the record before the Michigan Court of Appeals at the time that court rendered its decision—this of course does not include evidence adduced at a hearing held by the federal district court several years after the state court decision.
Applying AEDPA deference here, I conclude that McClellan's counsel was not ineffective and that McClellan was not prejudiced by his counsel's actions. Because the majority concludes otherwise, I respectfully dissent.
Supreme Court and Sixth Circuit precedent dictate that the Michigan Court of Appeals' one sentence decision denying McClellan relief “for lack of merit in the grounds presented” was an adjudication on the merits.
In Harrington v. Richter, 131 S.Ct. 770, 784–85 (2011), the Supreme Court held that “when a federal claim has been presented to a state court and the state court has denied relief, it may be presumed that the state court adjudicated the claim on the merits in the absence of any indication or state-law procedural principles to the contrary.” The Court further explained, “․ [w]here a state court's decision is unaccompanied by an explanation, the habeas petitioner's burden still must be met by showing there was no reasonable basis for the state court to deny relief.” Id. at 784. The Court also acknowledged that the presumption that the decision was on the merits “may be overcome when there is reason to think some other explanation for the state court's decision is more likely.” Id. at 785 (citing Ylst v. Nunnemaker, 501 U.S. 797, 803 (1991)).
Recently, in Werth v. Bell, ––– F.3d ––––; No. 10–2183 (6th Cir. August 28, 2012), a panel of this Court held that under Richter, a Michigan Court of Appeals' order denying relief “for lack of merit in the grounds presented” was presumed to be an adjudication on the merits, and thus “AEDPA deference applies ․ absent some ‘indication or [Michigan] procedural principle to the contrary.’ “ Werth, slip op. at *5.
The application of Richter and Werth to the present matter is straightforward. The Michigan Court of Appeals' denial of McClellan's ineffective assistance claim “for lack of merit on the grounds presented” is an adjudication on the merits subject to AEDPA deference, unless there is “reason to think some other explanation for the state court's decision is more likely.” Richter, 131 S.Ct. at 785; Werth, slip op. at *5.
The majority here incorrectly concludes there are other reasons to believe the state court of appeals' decision was on procedural grounds rather than on the merits. First, the majority argues that “the available evidence establishes” that the state courts' decisions denied McClellan's petition as procedurally defaulted rather than on the merits. The evidence cited by the majority includes the language in the state trial court opinion denying McClellan's motion for relief from judgment and whether the court of appeals had the trial court record before it when it decided McClellan's application for leave to appeal.
This evidence includes information produced by the state after additional briefing by the parties.1 The State's supplemental brief stated that it contacted the Chief Clerk of the Michigan Court of Appeals who, according to the State, advised that “the docket events [in the case] reveal that the panel deciding McClellan's appeal from the denial of his motion ․ did not have the lower court record when it rendered its decision.” The majority inappropriately concludes from this that in fact the judges on the state court of appeals panel did not have access to or review the trial court record. The fact that a court employee did not docket receipt of the transcripts does not necessarily mean the judges did not review them. The Chief Clerk acknowledged the uncertainty about what the judges actually review when deciding a case precisely because “judges do not record their deliberative process in docket entries.”
Further, it is not unheard of for an appellate court to access lower court records even when those records are not on the court of appeals or lower court docket. For example, the federal courts' Pacer system—or for that matter a brief phone call—allows us to access many lower court documents that have not otherwise been docketed in the court of appeals or that may be sealed. The bottom-line is there really is no way of knowing what the Michigan Court of Appeals judges were considering when they denied McClellan's application. This is precisely the kind of “theoretical possibility” and “pure speculation” the Supreme Court admonished habeas courts to avoid in an effort to circumvent the deference owed to state court summary orders. Richter, 131 S.Ct. at 785. With this in mind, surely, the Richter Court would not agree with the majority's reliance on the Chief Clerk's assertions as a basis for concluding that the state court of appeals decision was not on the merits.
Apart from speculation about what materials the state court of appeals judges actually considered in denying relief, there is no reason here to conclude that there is some other non-merits basis for the court of appeals decision. In Ylst, the Supreme Court asserted that “where ․ the last reasoned opinion on the claim explicitly imposes a procedural default, we will presume that a later decision rejecting the claim did not silently disregard that bar and consider the merits.” 501 U.S. at 803. In Ylst, the last reasoned opinion on the issue in question was a California Court of Appeal opinion in which “the sole basis for its rejection of the ․ claim was [a] state procedural rule.” Id. at 799.
But in this case, in denying McClellan's motion for relief from judgment, the trial court discussed both merits and procedure, thus making it difficult to conclude that the subsequent state court of appeals' summary decision must rest on purely procedural grounds. The trial court's opinion, a succinct nine paragraphs, spent nearly half of its space discussing the merits of McClellan's ineffective assistance claims, including that McClellan's counsel's performance was not objectively unreasonable for failing to call defense witnesses to support McClellan's self-defense claim, and that McClellan could not establish actual prejudice from these trial decisions. Specifically, the opinion stated:
The decision of what witnesses to call at trial is legal strategy. The court will not substitute its own judgment for that of defense counsel. Any actions by defense counsel, which can be attributed to legal strategy, will not be grounds for ineffective assistance of counsel. Therefore, counsel forgoing the opportunity to call witnesses to support that Defendant acted in self-defense is not grounds for ineffective assistance of counsel.
(Order Den. Mot. for Relief, Page ID # 1020). This is clearly merits language.2
Additionally, while the trial court stated in conclusion that McClellan failed to show cause and prejudice for his procedural default, it also stated that McClellan's motion was denied, “[f]or all the reasons stated.” In other words, it is not at all clear from the opinion that the “sole basis” for rejecting McClellan's claim was the state procedural rule. In Werth, we noted that “pointing only to the ambiguous language of the Michigan courts' orders, and Michigan cases and court rules ․” is not enough to support an explanation that some other reason for the court of appeals summary denial is more likely. Werth, slip op. at *7 (citing Richter, 131 S.Ct. at 785). Further, this Court's prior cases leave no doubt that as long as the state court put forward a merits-based ground for denying post-conviction relief, its mentioning of procedural default as an alternative or even primary ground for denying relief does not preclude AEDPA deference. Hoffner v. Bradshaw, 622 F.3d at 487, 505 (6th Cir.2010); see also Brooks v. Bagley, 513 F.3d 618, 624–25 (6th Cir.2008) (confirming this rule and noting that “[a]ll of the circuit courts that have considered th[is] question” have held the same).
Next, with respect to McClellan's subsequent leave application to the Michigan Court of Appeals, the majority improperly concludes that because the court of appeals did not have the trial court record at the time it rendered its decision, “the only reasonable inference to be drawn” is that the court of appeals “could not have denied McClellan's petition on the merits[.]” The majority's conclusion is not only speculative, but it is also based on an inaccurate premise.
First, McClellan's primary claim in his appeal from the denial of his motion for relief from judgment was not that he had overcome a procedural default, but rather that he was denied effective assistance of counsel in that his attorney “failed to investigate and present his self-defense claim,” and “failed to request disclosure of discovery.” McClellan's procedural default argument was relegated to his third point on appeal. Further, McClellan's application also included a fact section detailing the events that transpired the night of the shooting, as well as relevant trial testimony with citations to the trial court record.3
Second, the majority also fails to discuss the fact that the Michigan Court of Appeals had before it several sworn affidavits from McClellan and the witnesses he thought should be investigated and presented at trial. These affidavits were attached in the appendix to McClellan's leave application, and his application cites to them multiple times. The affidavits are intended to establish that McClellan was acting in self-defense in that they allege McClellan's victim picked up an object before approaching McClellan and that it appeared as though the victim was carrying a gun. The affidavits also assert that McClellan was not searched as he entered the club and that he did not leave the club to retrieve a weapon. McClellan's ineffective assistance claim rested entirely on what he claimed these witnesses would have testified to at his trial had they been called.
In other words, the state court of appeals had the merits of McClellan's ineffective assistance claim front and center. Not only were the underlying facts of the shooting and the testimony at trial set forth in McClellan's application, but the court was also able to consider the testimony that McClellan claimed was so crucial to his defense and thus to his ineffective assistance claim.
The conclusion that the court of appeals decision was on the merits and that the Richter presumption has not been rebutted is bolstered by two other considerations. First, Michigan courts have “consistently held that denial of an application ‘for lack of merit in the grounds presented’ is a decision on the merits of the issues raised.” People v. Collier, 2005 WL 1106501, at *1 (Mich.Ct.App.2005). Second, the State asserted at oral argument that the court of appeals' denial here was atypical and that generally, in denying appeals from motions for relief from judgment, in “85–90%” of cases, the court issues an order stating the appellant failed to establish grounds for relief under Mich. Ct. R . 6.508(D). That rule includes the standard for procedural default, Mich. Ct. R. 6.508(D)(3), and is therefore likely to have been the rule that the court of appeals would have relied on had its denial of McClellan's application been procedural.
In sum, the merits of McClellan's claims were before the court of appeals, the court was able to consider the underlying factual circumstances and trial testimony, as well as the affidavits from several of the witnesses that McClellan argued would have changed the outcome in his trial, and as a matter of Michigan law, the court's statement that it was denying the application on the merits is a decision on the merits. All of this reinforces the Richter presumption that the state court decision was on the merits, and undermines the majority's conclusion that there is some other reason to conclude the decision was not merits based. Here, there is not enough in the record to establish there is a “reason to think some other explanation for the state court's decision is more likely.” Richter, 131 S.Ct. at 785; Werth, slip op. at *5.4
Because the Michigan Court of Appeals' decision denying McClellan's leave application “for lack of merit in the grounds presented” adjudicated McClellan's ineffective assistance of counsel claim on the merits, it is entitled to AEDPA deference under 28 U.S.C. § 2254(d). Werth, slip op. at *6. McClellan therefore has to show that the state court's decision was contrary to or an unreasonable application of clearly established federal law as decided by the Supreme Court, or that it resulted in a decision that was based on an unreasonable determination of the facts in light of the record before the state court. 28 U.S.C. § 2254(d)(1) and (2).
In Cullen v. Pinholster, 131 S.Ct. 1388, 1398 (2011), the Supreme Court held “that review under § 2254(d)(1) is limited to the record that was before the state court that adjudicated the claim on the merits.” This Court has agreed that “[§ 2254(d)(1) ] review is ․ ‘limited to the record that was before the state court.’ “ Bray v. Andrews, 640 F.3d 731, 737 (6th Cir.2011). In Pinholster, the Supreme Court made sure to clarify that this limitation was not a wholesale bar on federal evidentiary hearings, and that § 2254(e)(2) would “continue[ ] to have force where § 2254(d)(1) does not bar federal habeas relief,” such as when the state court decision was not an adjudication on the merits. 131 S.Ct. at 1401. But here, the state court decision was on the merits, and thus Pinholster's limitations apply. As such, to the extent the district court reviewed McClellan's claims under § 2254(d)(1), it was error for the district court to consider evidence outside of the state court record.5
McClellan has failed to establish on the record before the state court of appeals that its decision was contrary to or an unreasonable application of Supreme Court precedent. In assessing the reasonableness of an unexplained merits adjudication by a state court under § 2254(d), a habeas court must “determine what arguments or theories supported or ․ could have supported the state court's decision; and then it must ask whether it is possible fairminded jurists could disagree that those arguments or theories are inconsistent with the holding in a prior decision of [the Supreme Court].” Richter, 131 S.Ct. at 786. Where the state court did not elucidate the reasons for its decision, “the habeas petitioner's burden still must be met by showing there was no reasonable basis for the state court to deny relief.” Id. at 784. This “highly deferential standard” requires that determinations made in state court “be given the benefit of the doubt.” Pinholster, 131 S.Ct. at 1398 (internal quotation marks omitted).
McClellan's claims rest on the effectiveness of his counsel at trial. Accordingly, to prevail on his claim, McClellan must demonstrate that his counsel's performance was deficient and that he suffered prejudice as a result. Strickland v. Washington, 466 U .S. 668, 687 (1984); Foust v. Houk, 655 F.3d 524, 533 (6th Cir.2011). To show deficiency, McClellan must overcome the “strong[ ] presum[ption]” that his counsel “rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment.” Strickland, 466 U.S. at 690. To demonstrate prejudice, McClellan must “show that there is a reasonable probability that, but for, counsel's unprofessional errors, the result of the proceeding would have been different.” Id. at 694. This means he must show a “ ‘substantial,’ not just ‘conceivable,’ likelihood of a different result.” Pinholster, 131 S.Ct. at 1403 (quoting Richter, 131 S.Ct. at 791).
When analyzing a Strickland claim under § 2254(d), this Court's review is “ ‘doubly deferential.’ “ Pinholster, 131 S .Ct. at 1403 (quoting Knowles v. Mirzayance, 556 U.S. 111, 129 (2009)). Accordingly, the key question “ ‘is whether there is any reasonable argument that counsel satisfied Strickland's deferential standard.’ “ Foust, 655 F.3d at 533–34 (emphasis added) (quoting Richter, 131 S.Ct. at 788).
McClellan argues that his trial counsel was ineffective for failing to conduct a full investigation and interview witnesses who may have corroborated his self-defense claims and negated the argument that his actions were premeditated. But McClellan has failed to establish that his counsel's strategy was objectively unreasonable or that he was prejudiced by his counsel's strategy.
McClellan's trial counsel argued throughout the trial that McClellan acted in self-defense. He did so in his opening and closing remarks and also by attacking the testimony of the prosecution's witnesses and arguing that it was the prosecution's burden to prove that McClellan did not act in self-defense. As part of this strategy, defense counsel elicited testimony that some of the witnesses heard someone in the victim's group had a gun, that McClellan was back-pedaling and pushed up against a wall when the victim charged at him, and that the shooting occurred in an atmosphere of threats and tension. Accordingly, the jury was instructed on first and second degree murder, voluntary manslaughter, and self-defense.
McClellan's trial counsel clearly made the case for self-defense. Accordingly, McClellan's ineffective assistance claim is essentially that certain witnesses, had they been investigated and called, could have made his self-defense claim stronger. The affidavits in the Michigan court of appeals' record show that much of the additional testimony would have merely reiterated what defense counsel was able to elicit from the prosecution's witnesses—specifically, that McClellan was being assaulted at the time the shots were fired and that someone in the victim's group had a gun.
Additionally, as the Supreme Court recently noted in Richter: “To support a defense argument that the prosecution has not proved it is sometimes better to try to cause pervasive suspicion of doubt than to strive to prove a certainty that exonerates. All that happened here is that counsel pursued a course that conformed to the first option ․ In light of the record here there was no basis to rule that the state court's determination was unreasonable.” 131 S.Ct. at 790. On the record before the state court here, there is no reason to conclude that McClellan's trial counsel's performance was deficient.
McClellan's argument that he was prejudiced by his counsel's alleged deficiency is also without merit. As noted in Pinholster, for a defendant to show a “reasonable probability” of a different result due to counsel's deficient performance requires a “substantial” not just “conceivable” likelihood of a different result. 131 S.Ct. at 1403 (quoting Richter, 131 S.Ct. at 791). McClellan cannot show that had his counsel investigated or called his family and friends to testify that there would be a “substantial” likelihood of a different result on his claim that the shooting was in self-defense or that it was not premeditated.
One of the prosecution's witnesses, a club security guard, testified that the victim was not carrying a weapon at the time of the shooting and that McClellan shot the victim after he was already down on the ground. He also testified that McClellan was one of two people he saw coming from a parking lot area after leaving the club but prior to joining the altercation. He further testified that it appeared as though the victim was falling backward before he had even been shot. Additionally, the medical testimony about the victim's four bullet wounds did not support McClellan's self-defense claim because there were no wounds going from front to back in a manner that would justify concluding he used the gun in self-defense case.
It is remotely possible that McClellan's family members' would have bolstered his self-defense claim, but his family members may also have been seen as biased and thus their testimony possibly discredited, whereas the security guard and medical testimony is relatively objective.
The majority also asserts that McClellan's witnesses would have testified that he had the weapon in the club and did not retrieve it from his car, and that this evidence “disprove[s] the State's theory of premeditation.” And in fact, the affidavits considered by the state court of appeals show that McClellan's witnesses would have testified consistent with this theory. But there was ample other evidence at trial to support a finding of premeditation under Michigan law.
In Michigan, “premeditation and deliberation require sufficient time to allow the defendant to take a second look.” People v. Schollaert, 486 N. W.2d 312, 318 (Mich.Ct.App.1992); see also People v. Tilley, 273 N.W.2d 471, 473–74 (Mich.1979) (“the interval between initial thought and ultimate action should be long enough to afford a reasonable man time to subject the nature of his response to a second look.”) (internal quotation omitted). At trial, the prosecutor presented the alternative argument that McClellan premeditated when he had the gun at his side for some time prior to shooting the victim, and several witnesses testified to this account. Accordingly, there was evidence in the record to support a claim that McClellan premeditated even if he had not left the club to go get a gun from a vehicle. It is not substantially likely that if McClellan's friends had testified McClellan had the gun when he entered the bar that their biased testimony would have been enough to create a different result.
Based on the record before the Michigan Court of Appeals, there are several reasonable arguments that counsel satisfied Strickland's deferential standard. Foust, 655 F.3d at 533–34. As such, its merits determination was not contrary to or an unreasonable application of clearly established federal law and to the extent that the district court and majority conclude otherwise, those conclusions are erroneous.6
For the foregoing reasons, I dissent. I would reverse the decision of the district court conditionally granting the writ of habeas corpus.
1. The dissent treats our discussion of McClellan's failure-to-investigate claim as a failure-to-call-witnesses claim, which is governed by a distinct standard. As we are granting McClellan's petition based on his failure-to-investigate claim, any analysis by the dissent relating to the failure-to-call-witnesses doctrine is irrelevant. Because we review a grant of a habeas petition de novo, the district court's chosen wording is immaterial to our holding. In any event, the dissent mischaracterizes the district court's discussion by piecing together segments of sentences that are paragraphs apart in the district court order. The district court's ultimate determination is as follows: “Rather, the evidence presented at the hearing shows that Petitioner's counsel's deficient failure to investigate sufficiently undermines the Court's confidence in the outcome of the trial.” McClellan v. Rapelje, No. 2:08–CV–10735, 2011 WL 2447999, at *13 (E.D. Mich. June 14, 2011).
2. Given that the State has conceded that the lower court record was not in front of the Michigan Court of Appeals, it is unclear why the dissent is attempting to put in play a point that remains undisputed by the parties. Labeling the State's position as “speculation” does not transform a party's concession into a live issue for the court to decide.At oral argument, we requested that the parties submit supplemental briefing stating their positions as to what the Michigan Court of Appeals had before it when it rendered its decision. The portion of oral argument in which this request was made is transcribed in its entirety as follows:It would be useful for this Court to know in a letter brief from each of you what your position is as to what was the record in front of the Michigan Court of Appeals. And while we're on that topic, we would also like to have you submit letter briefs by October 25 on the subject of whether there was invited error and what we should do about that and any cases that pertain to that.Oral Argument at 31:30–32.00, McClellan v. Rapelje (No. 11–1841). Prior to this request, the panel asked the State to clarify that the record was “a physical, tangible record, as opposed to an electronic record,” given the date of the Michigan Court of Appeals decision. Id. at 31:15. After oral argument, the case manager sent a letter to the parties summarizing her interpretation of the panel's request. The letter stated as follows:The briefing letter is pursuant to the panel's request at today's oral argument that the parties file a letter brief addressing the following:-what physical record did the Michigan Court of Appeals have before it, and-the issue of invited error.The letter briefs are due from counsel by October 25, 2012.Letter from Louise Schwarber, Case Manager, McClellan v. Rapelje (No. 11–1841). The State's concession, then, that the Michigan Court of Appeals did not have the trial record in front of it reflects the position of the State as requested by the panel. It does not, as suggested by the dissent, reflect the position of the clerk of the Michigan Court of Appeals.It is further important to note that we have disregarded any unsolicited additional information provided by the State in its Supplemental Brief concerning the non-public processes of the Michigan Court of Appeals. In fact, it is the dissent that hypothesizes about the ways in which the public docket entries might inaccurately reflect the actual record in the very way it warns the Richter Court would reject.
3. Once again, the dissent argues a position that has been expressly conceded by the State in an attempt to strengthen its own reasoning. In its brief, the State made clear its position on the trial court order. “McClellan raised his ineffective assistance of counsel claims in a motion for relief from judgment, which the state trial court denied on procedural grounds.” Appellant Br. at 13. It is hard to imagine a clearer statement of the State's position, and we decline to evaluate arguments conceded by a litigant, as the dissent would have us do.
4. It is unclear why the dissent places significant weight on the presence of an unsubstantiated fact section included in McClellan's delayed application for leave to appeal before the Michigan Court of Appeals. In so doing, the dissent appears to be asserting that the Michigan Court of Appeals would reach a decision on the merits of a claim without consulting the evidence in the record. The fact that the application included citations to the record indicates nothing more than that counsel was comprehensive in compiling the brief. We further find it curious that the dissent considers as relevant the order of the claims in McClellan's delayed application for leave to appeal before the Michigan Court of Appeals. McClellan in no way represented that his ineffective-assistance-of-counsel claim was his primary claim on appeal, and it is nonsensical to imply that the organization of a party's brief dictates the basis upon which a court rules. Finally, the State's bald assertion at oral argument that 85–90% of orders use different language is wholly unsubstantiated by any evidence in the record.Additionally, the Michigan Court of Appeals has made clear that it does not consider extra-record evidence: “Appeals to the Court of Appeals are heard on the original record,” which is defined as “the original papers filed in [the lower court] or a certified copy, the transcript of any testimony or other proceedings in the case appealed, and the exhibits introduced.” Mich. Ct. R. 7.210(A)(1). Therefore, the presence of the affidavits attached to the application for leave to appeal is immaterial.
1. This information arose in response to several questions posed during oral argument, including from me, and in response to additional briefing requested of the parties. In my view, this line of inquiry into the state court's decision making process was probably inappropriate because, as explained herein, disclosure of such information creates a risk that matters extrinsic to the public record will improperly influence the reviewing court's decision.
2. The majority's characterization of McClellan's claim as strictly a failure to investigate claim undermines their argument that the state trial court's denial of the motion for relief from judgment is evidence rebutting the Richter presumption. McClellan only presented a failure to call claim to the trial court, not a failure to investigate claim, and the trial court only discussed the merits of his failure to call claim. It did not even discuss a failure to investigate claim in the context of default or otherwise. It was not until his subsequent leave application that McClellan squarely presented a Michigan court with his failure to investigate claim. Additionally, the federal district court's conditional grant of habeas relief appeared to rest on both failure to investigate and failure to call claims. It stated, “The Court held an evidentiary hearing on Petitioner's claim that his trial counsel was ineffective for failing to call a number of witnesses at trial.” In its subsequent discussion, the court found that McClellan's counsel was deficient for “failing to investigate the possibility of calling the men ․ as defense witnesses at trial,” and that counsel “did not make an objectively reasonable decision not to call the defense witnesses․” McClellan v. Rapelje, 2011 WL 2447999, at *11 (E.D. Mich. June 14, 2011).
3. The majority argues that because I rely in part on the fact section of McClellan's briefing to the court of appeals, I am asserting that a court could decide the merits of a claim without considering evidence in the record. It is important to note, that we are starting from the presumption that the decision was on the merits. Thus, I am attempting to show that McClellan has failed to rebut this presumption. Certainly, the majority is not taking the position that reviewing McClellan's leave application or the attachments therein is inappropriate in attempting to discern whether there is some other non-merits basis for the court of appeals summary order. Additionally, my conclusion that McClellan failed to rebut the Richter presumption is not only based on the fact section of his leave application, but I also point to several other things that lead clearly to the conclusion that some other non-merits reason for the state court of appeals decision has not been shown to be “more likely.” Richter, 131 S.Ct. at 785. Further, the majority's reliance on MICH CT. R. 7.210(A)(1) (strictly defining the record on appeal as the transcript and exhibits from the lower court), if read as completely unambiguous, suggests that an appellate court could not even consider the parties' briefs in deciding a case. Clearly, that reading is incorrect. Moreover, as stated above, before even considering whether the state court of appeals looked at McClellan's affidavits, we are already presuming the decision was on the merits. It is McClellan's burden to show that the decision was not on the merits, and he has not carried that burden here.
4. Moreover, the fact that the State initially argued to the federal district court that McClellan's claims were procedurally defaulted and then later argued that there was a decision on the merits cannot be grounds for concluding there is some other non-merits basis for the state court of appeals decision. Whether AEDPA deference or de novo review applies is a question for the courts to answer-not the parties. See K & T Enters., Inc. v. Zurich Ins. Co., 97 F.3d 171, 175 (6th Cir.1996) (“The parties ․ cannot determine this court's standard of review ․ [s]uch a determination remains for this court to make for itself.”). And certainly, a party's litigating position does not bind a court to a particular outcome. The fact that the State pressed a particular argument at different points in the proceedings does not necessarily mean that the state court of appeals' decision was consistent with that position.
5. Pinholster was decided before the district court issued the writ, and the State moved for reconsideration based on Richter and Pinholster, but the district court did not mention Pinholster in either of its written decisions.
6. The district court rejected the State's claim that the Michigan Court of Appeals' order was on the merits, but addressed the § 2254(d) analysis in the alternative.
MERRITT, J., delivered the opinion of the court, in which MOORE, J., joined. McKEAGUE, J. (pp. 11–22), delivered a separate dissenting opinion.