Opinion
19-2212
07-21-2021
MARK MCQUEEN, Petitioner-Appellant, v. O'BELL T. WINN, Warden, Respondent-Appellee.
NOT RECOMMENDED FOR PUBLICATION
ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF MICHIGAN
BEFORE: COLE, BUSH, and NALBANDIAN, Circuit Judges.
OPINION
JOHN K. BUSH, CIRCUIT JUDGE
In 2011, a Michigan jury convicted Mark McQueen of first-degree criminal sexual conduct involving his eleven-year-old daughter. The trial court sentenced him to 25 to 40 years' imprisonment. After a failed direct appeal to the Michigan Court of Appeals and the Michigan Supreme Court, and multiple unsuccessful post-conviction motions, McQueen filed a writ of habeas corpus in federal court. The district court denied his petition. We affirm.
I.
A. Factual History
The charge against McQueen followed from the accusation that he sexually assaulted his eleven-year-old daughter, A.M., in late 2009. Five months after the alleged assault, in April 2010, A.M. testified at a probable cause hearing. She told the court that on the Wednesday before Thanksgiving 2009, McQueen picked up her and her brother, D.M., from their grandmother's house, and that she and D.M. stayed at McQueen's house for the reminder of Thanksgiving break. A.M. said that on Sunday night, while she was asleep in the living room, McQueen called out to her, told her that he was lonely, and asked that she come into his room. She testified that she went into the room, laid on the bed, and that McQueen told her that he was going to "touch" her. A.M. allegedly responded by telling her dad "No." But according to A.M., McQueen told his daughter that if she didn't cooperate, he would "beat" her. A.M. also testified that McQueen touched her inside her "private spot" with his hand. She said nothing to her mother at first. But after about three weeks A.M. told her mother what McQueen had done to her. Based on that account the court bound the case over for trial, which took place the following year.
At trial, A.M.'s testimony featured some minor gaps and inconsistencies with her statements during the probable cause hearing. For example, she could not recall what day the alleged assault took place until counsel reminded her of her testimony at the probable cause hearing. She also said that McQueen had told her he would "kill" her if she didn't let him have his way, not that he would "beat" her as she previously testified. She also contradicted her prior testimony by saying at trial that she had seen McQueen after the assault (previously she said she had not), and by saying that McQueen had assaulted her more than once (previously she had only reported the one assault). Generally, though, she told the same story at trial that she did at the probable cause hearing. She further noted at trial that her mom-whose later testimony reinforced A.M.'s account-took her to the hospital for an examination, to the police station to file a report, and then to a place called Care House for a special interview.
At the hospital, a pediatric nurse examined A.M. and recorded A.M.'s account of what happened. The nurse then testified at trial about what A.M. told her-essentially, that A.M.'s father had touched her genital area with his finger one night when she was visiting him over the Thanksgiving holiday period. The nurse also reported that A.M. did in fact show signs of vaginal trauma, though the nurse could not specify exactly what caused the trauma or when it occured.
At the police station A.M.'s mother filed a report, and a detective was assigned to the case. The detective testified at trial that he spoke with A.M. and her mother. The detective did not, however, formally interview A.M. because protocol dictated that children be interviewed at Care House.
At Care House, a forensic interviewer named Margo Moltmaker interviewed A.M. The prosecution called Moltmaker as an expert witness during the trial to buttress A.M.'s testimony. Moltmaker told the jury that during the forensic interview, A.M. seemed to be telling the truth.
Forensic interviewers are trained to gather information about incidents of alleged child abuse in a manner that is supposed to yield accurate information from the child.
Prior to trial, McQueen's attorney had moved for funds to hire an expert to rebut Moltmaker's testimony at trial. At the hearing on the motion, trial counsel said that he might be considered ineffective if he failed to hire an opposing expert-he also noted that he had left a phone message with an expert, Dr. Katherine Okla, who left a return message, but that the two had not formally discussed any plans. The court granted the motion, providing counsel with $1,500.00 to hire Dr. Okla as an expert. But at trial, counsel did not call Dr. Okla to testify, and it does not appear that he consulted with her beyond leaving the initial message.
Counsel did, however, cross-examine Moltmaker. He questioned Moltmaker about her procedure, and why she failed to follow standard protocol which dictated that child interviews be videotaped so that experts can adequately review the film. He also inquired about some of the inconsistencies between A.M.'s account in the forensic interview (which had been transcribed), and A.M.'s testimony at the probable cause hearing and at trial.
In addition, counsel called Tawanna Patterson, McQueen's girlfriend, as an alibi witness. Patterson told the jury that she and McQueen took A.M. and D.M. to their grandmother's house Sunday evening-when A.M. alleged that the two were at McQueen's house. She also said that after dropping A.M. off, she and McQueen went home and spent the rest of the night together, alone.
McQueen had also wanted his counsel to call his cousin, Archie McQueen, as an alibi witness. McQueen says that he told his trial counsel that Archie could have told the jury that the two were at a rap party until two or three in the morning on the night in question, so McQueen could not have been at the house with A.M. as alleged. Trial counsel did not call Archie to testify.
The jury credited A.M.'s account of the facts, convicting McQueen of criminal sexual conduct in the first degree. And because this was McQueen's fourth felony offense, the court sentenced him to a mandatory 25 to 40 years' imprisonment.
B. Procedural History
On direct appeal to the Michigan Court of Appeals, McQueen's appellate counsel argued that his trial counsel had been constitutionally ineffective for failing to object to certain hearsay statements that Moltmaker made. In accord with a special Michigan state court rule, McQueen also filed two supplemental pro se briefs alleging several other trial counsel deficiencies. Neither McQueen nor his appellate counsel raised the claims relevant to McQueen's habeas petition now before us.
The Michigan Court of Appeals considered the briefing before it and affirmed McQueen's conviction. People v. McQueen, No. 306317, 2013 WL 3814349 (Mich. Ct. App. July 23, 2013) (per curiam). The Michigan Supreme Court denied McQueen's application for leave to appeal. People v. McQueen, 840 N.W.2d 350 (Mich. 2013) (mem.).
McQueen then filed a pro se post-conviction motion for relief from judgment in the lower court. He raised the three claims relevant to the habeas petition now before the panel, among others. He argued that his trial counsel provided ineffective assistance both by failing to consult and hire Dr. Okla to rebut Moltmaker's testimony, despite having the funds and the court's authorization to do so, and by failing to call Archie as an alibi witness. He also argued that his appellate counsel provided ineffective assistance by failing to raise those two claims on direct appeal. The court denied McQueen's postconviction motion for relief from judgment on the briefs without holding an evidentiary hearing.
After McQueen received notice that his motion had been denied, he filed two new motions: a motion to reconsider and a motion to amend or supplement his motion for relief from judgment. In the motion to amend, McQueen may have added an argument that his appellate counsel was ineffective for failing to request an evidentiary hearing to review certain deficiencies of his trial counsel. He also moved for a post-conviction evidentiary hearing. The court denied both motions in a single order.
It first denied the motion for reconsideration on procedural grounds, determining that McQueen could have presented the new claims raised in his motion for reconsideration in his initial motion for relief from judgment. It also found that the new claims lacked merit. Then, because it had already denied McQueen's motion for relief from judgment, the court construed McQueen's motion to amend as a second motion for relief from judgment. And it denied the second motion under Michigan Court Rule 6.5(G) because it was not based on a retroactive change in law, or a claim of new evidence. Finally, the court denied McQueen's request for a post-conviction evidentiary hearing "in light of the above rulings."
McQueen sought leave to appeal those rulings and again moved for an evidentiary hearing. The Michigan Court of Appeals denied leave, citing Michigan Court Rule 6.508(D)(3), which prohibits Michigan courts from hearing claims that could have been but were not raised on direct appeal. It also denied McQueen's motion for an evidentiary hearing.
McQueen filed an application for leave to appeal in the Michigan Supreme Court, but that too was denied under Rule 6.508(D).
Lastly, McQueen filed his petition for habeas relief in federal court. He raised sixteen claims, which largely consisted of the claims he raised before the state court in the motions discussed above. The district court denied all of the claims and denied McQueen a certificate of appealability. We, however, granted McQueen a certificate to appeal three discrete claims: (1) whether trial counsel performed ineffectively by failing to interview or call Dr. Katherine Okla as an expert witness, (2) whether trial counsel performed ineffectively by failing to interview or call Archie McQueen as an alibi witness, and (3) whether appellate counsel performed ineffectively by failing to raise either of those claims on direct appeal. This appeal followed.
II.
A. Standard of Review
Normally we review a district court's legal conclusions de novo and its findings of fact for clear error. Miller v. Genovese, 994 F.3d 734, 741 (6th Cir. 2021). But where, as here, the district court did not conduct an evidentiary hearing and relied on the state-court record, we review both the legal conclusions and the factual findings de novo. Barton v. Warden, S. Ohio Corr. Facility, 786 F.3d 450, 460 (6th Cir. 2015) (per curiam).
Of course, in habeas appeals, we also must determine whether AEDPA deference applies to the state court's resolution of the relevant claims. To do so, we look to the last reasoned state- court decision. See Ylst v. Nunnemaker, 501 U.S. 797, 803 (1991). If that decision is "on the merits," AEDPA deference applies. 28 U.S.C. § 2254; Reiner v. Woods, 955 F.3d 549, 556 (6th Cir. 2020). If it is not, we apply de novo review. Stermer v. Warren, 959 F.3d 704, 721-22 (6th Cir. 2020). And as relevant in this appeal, habeas review is precluded entirely if a state court dismisses a petitioner's claims because he "has failed to meet a state procedural requirement that is independent of the federal question and adequate to support the judgment," unless the petitioner can show cause and prejudice for the procedural default. Biros v. Bagley, 422 F.3d 379, 386-87 (6th Cir. 2005).
B. Ineffective Assistance of Trial Counsel
To prove his claims of ineffective assistance of trial counsel, McQueen must show that his attorney's performance was deficient and that he was prejudiced as a result. See Strickland v. Washington, 466 U.S. 668, 687 (1984). McQueen contends that his trial counsel fell short in two ways: first, counsel failed to consult with or hire Dr. Katherine Okla as an expert witness to rebut the testimony of Margo Moltmaker, the state's forensic interviewer; and second, counsel failed to call his cousin Archie McQueen as an alibi witness.
The parties agree that the last reasoned decision on McQueen's ineffective assistance of trial counsel claims is the Michigan Court of Appeals's denial of his application for leave to appeal the denial of his post-conviction motions. They also agree that in citing MCR 6.508(D)(3) to explain its decision, the Michigan Court of Appeals applied an independent and adequate state ground sufficient for procedural default-declining to review McQueen's trial-counsel claims because he failed to raise them on direct appeal. See Amos v. Renico, 638 F.3d 720, 733 (6th Cir. 2012). Accordingly, we can only review McQueen's two trial-counsel claims if he demonstrates both that he had cause for his failure to raise them on direct appeal and that he would suffer prejudice if we did not forgive his procedural default. See Clifton v. Carpenter, 775 F.3d 760, 767-68 (6th Cir. 2014). McQueen claims that his appellate counsel provided ineffective assistance under Strickland which, if correct, would meet that standard. See Chase v. MaCauley, 971 F.3d 582, 592, 595-96 (6th Cir. 2020). So to determine whether McQueen demonstrates cause and prejudice for his procedural default, we must discuss the merits of McQueen's appellate-counsel claim.
Although there are two prongs in the Strickland analysis, to answer the inquiry in this case, we focus largely on the first: whether appellate counsel's performance was deficient. Counsel's performance was deficient if it "fell below an objective standard of reasonableness." Strickland, 466 U.S. at 688. That is a difficult standard to satisfy. For one, "counsel has no obligation to raise every possible claim" on appeal. McFarland v. Yukins, 356 F.3d 688, 710 (6th Cir. 2004). And "the decision of which among the possible claims to pursue is ordinarily entrusted to counsel's professional judgment." Id. Finally, "[c]ounsel's performance is strongly presumed to be effective." Id. (quoting Scott v. Mitchell, 209 F.3d 854, 880 (6th Cir. 2000)).
On direct appeal, McQueen's appellate counsel raised one claim regarding his trial counsel's alleged ineffectiveness. That claim related to trial counsel's failure to object to the forensic interviewer's alleged hearsay testimony about what A.M. had told her out of court. The Michigan Court of Appeals correctly noted in dismissing the claim on direct appeal that counsel's failure to object was sound trial strategy-the jury had already heard the same testimony from A.M. herself, so even if made and sustained, the objection would not have affected the jury's decision.
To evaluate whether McQueen's appellate counsel performed deficiently by raising only that one claim instead of the two McQueen raises here, we must consider the merits of the two trial-counsel claims that his appellate counsel failed to raise. See Henness v. Bagley, 644 F.3d 308, 317 (6th Cir. 2011). If either of the claims satisfy Strickland's two prongs, it is likely that McQueen's appellate counsel was deficient for failing to raise it. So to determine whether appellate counsel was deficient we must discuss the merits of McQueen's trial-counsel claims. 1. Failure to call Dr. Katherine Okla
McQueen first claims that his trial counsel was ineffective for failing to call Dr. Katherine Okla to testify or to consult with her about his case. We disagree. McQueen's trial counsel did not perform deficiently by failing to call Dr. Okla, and even if he did, his performance did not prejudice McQueen.
a. Deficient Performance
According to McQueen, Dr. Okla's expert testimony was necessary to rebut the testimony of the State's forensic interviewer, Margo Moltmaker, whose expert testimony buttressed A.M.'s testimony at trial. "But Strickland does not enact Newton's third law for the presentation of evidence, requiring for every prosecution expert an equal and opposite expert from the defense." Harrington v. Richter, 562 U.S. 86, 111 (2011). "[S]trategic decisions-including whether to hire an expert-are entitled to a 'strong presumption' of reasonableness." Dunn v. Reeves, 141 S.Ct. 2405, 2410 (2021) (per curiam) (quoting Richter, 562 U.S. at 104). "In many instances cross- examination will be sufficient to expose defects in an expert's presentation." Richter, 562 U.S. at 111.
McQueen does not provide us with an affidavit from Dr. Okla describing what she might testify to. Our existing precedent is unclear about whether such information is required where the state court did not adjudicate the claim on the merits. Compare Clinkscale v. Carter, 375 F.3d 430, 444 (6th Cir. 2004) (noting that the district court's holding that petitioner needed an affidavit from his alleged alibi witness to satisfy Strickland was "contrary to . . . law") with Clark v. Waller, 490 F.3d 551, 557 (6th Cir. 2007) (noting that petitioner "cannot show that he was prejudiced" by counsel's failure to call a witness where the only evidence for the witnesses alleged testimony was petitioner's own assertions). We have no occasion to resolve the apparent conflict; even if no affidavit is required, McQueen is not entitled to habeas relief. And, of course, if the state court had adjudicated this claim on the merits, we would have been limited to the state court record. See Cullen v. Pinholster, 563 U.S. 170, 185 (2011).
That appears to be the approach trial counsel took in defending McQueen. After Moltmaker took the stand, McQueen's trial counsel cross-examined her about the forensic protocols she employed-the precise topic Dr. Okla would allegedly have testified about. Specifically, counsel asked Moltmaker if she followed the best practices listed by the prevailing forensic interviewing protocols. She replied: "Yes." So counsel asked Moltmaker if she had complied with the protocol's suggested best practice of videotaping the forensic interview she conducted with A.M. She had not. Counsel then proceeded to confirm the potential for inaccuracy inherent in Moltmaker's technique, as McQueen alleges Dr. Okla would have done if she had testified. He asked about the purpose of videotaping: to permit experts to evaluate the interviewer's "sublime body language suggestions . . . nodding, shaking, any making of faces." Counsel also confirmed with Moltmaker that the protocol was created to make sure that the interviewer's conduct conforms with professional norms, and that without a videotape, the interviewer's body language could not be reviewed.
But that is not all. Counsel questioned Moltmaker about how children like A.M. might be manipulated by police or their guardians in these types of situations and made known that A.M. had spoken with a few relatives, the police, and a nurse prior to speaking with Moltmaker.
Lastly, counsel questioned Moltmaker about some seeming discrepancies in A.M.'s testimony based on the forensic interview notes. In that regard, he asked Moltmaker about A.M.'s representation during the interview that the sexual criminal conduct took place over "Christmas time," given that A.M. testified later in the interview, at the probable cause hearing, and at trial, that the conduct took place on a Sunday evening over Thanksgiving. He also asked about A.M.'s representation during the interview that her underwear was ripped, given that she denied that representation at trial. And he asked about A.M.'s statement during the interview that McQueen was wearing pants at the relevant time, given her trial testimony that he was only wearing underwear. Counsel further pursued that line of question during re-cross-examination.
In lieu of a video, one of Moltmaker's team members took verbatim notes of the forensic interview.
Trial counsel also conducted an extensive cross-examination of A.M., which included a few questions about Moltmaker's interview with the child.
Trial counsel's cross-examination and recross-examination touched on most of what McQueen alleges Dr. Okla would have testified to. The cross was also relatively comprehensive, and it displayed counsel's knowledge of the required protocols, their purpose, Moltmaker's failure to follow them, and the inconsistencies in A.M.'s testimonies. And because there was no video of Moltmaker's interview for Dr. Okla to deploy her expertise and review, it would have been difficult for Dr. Okla to analyze the interview and rebut Moltmaker's testimony. Counsel's evident preparation for the cross, and the lack of a recording for Dr. Okla to analyze, weigh in favor of counsel's decision to not consult with Dr. Okla. Considering all of this, we cannot say that counsel was constitutionally deficient for not calling Dr. Okla to the stand. See Richter, 562 U.S. at 111 ("[I]t is difficult to establish ineffective assistance when counsel's overall performance indicates active and capable advocacy."); see also Samatar v. Clarridge, 225 Fed.Appx. 366, 372 (6th Cir. 2007) (noting that the decision to rely on cross-examination instead of opposing expert testimony "does not constitute ineffective assistance of counsel . . . even if the wisdom of such an approach is debatable." (internal citations and quotation marks omitted)).
Of course, McQueen's counsel did argue at a motions hearing before the trial court that he might be considered ineffective if he did not hire and call Dr. Okla to the stand to rebut Moltmaker's trial testimony. And the trial court provided counsel with $1,500.00 to hire Dr. Okla as an expert (which counsel failed to use). But the Strickland standard is objective, not based on one attorney's subjective view. See Strickland, 466 U.S. at 687-88. When making his arguments at the hearing, counsel cited to People v. Owens, No. 288074, 2010 WL 4320396 (Mich. Ct. App. Nov. 2, 2010) for the proposition that he might be considered ineffective if he did not hire Dr. Okla. But Owens does not speak so broadly. The court in Owens held that trial counsel was ineffective for failing to call the same Dr. Okla as an expert witness to raise doubts about the accuracy of the forensic interview conducted in that case. Id. at *2. But its conclusion was based in part on trial counsel's admission that "he did not know anything about Michigan's forensic interviewing protocol" and because he was "unprepared" when discussing the issue before the court. Id. at *3. Here, trial counsel's cross- and recross-examination of Moltmaker demonstrates that he was not similarly ignorant or unprepared. Ultimately, McQueen bears the burden to overcome the strong presumption that his trial counsel was reasonably competent. See Strickland, 466 U.S. at 689. He fails to satisfy that burden here.
He also fails to demonstrate that counsel was deficient for failing to consult with Dr. Okla in the first place. Counsel was reasonably prepared at trial to question Moltmaker on the issues Dr. Okla would have allegedly discussed. It cannot be said that he did "next to nothing to determine if the State's . . . [evidence] was impervious to attack," or that he was "unconcerned about the State's [testimonial] evidence." Richey v. Bradshaw, 498 F.3d 344, 362 (6th Cir. 2007). His decision to proceed with cross-examination to rebut Moltmaker's testimony could be deemed a reasonable decision that made particular investigation into Dr. Okla's expert opinion unnecessary. Strickland, 466 U.S. at 691; see also Samatar, 225 Fed.Appx. at 372.
Finally, McQueen's arguments concerning the lack of evidence available to the court do not counsel in favor of granting habeas relief. "[T]he absence of evidence cannot overcome the strong presumption that counsel's conduct fell within the wide range of reasonable professional assistance." Burt v. Titlow, 571 U.S. 12, 23 (2013) (internal quotation marks and brackets omitted) (quoting Strickland, 466 U.S. at 689). And "even if there is reason to think that counsel's conduct was 'far from exemplary, '" we still cannot grant relief because "'[t]he record does not reveal' that counsel took an approach that no competent lawyer would have chosen." Dunn, 141 S.Ct. at 2411 (quoting Titlow, 571 U.S. at 23-24).
b. Prejudice
To establish prejudice, a petitioner "must show that there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome." Strickland, 466 U.S at 694.
Even if counsel's failure to call Dr. Okla to the stand or to consult with her before trial was deficient, it did not prejudice McQueen. Assuming that Dr. Okla could provide the information McQueen alleges, there is not a reasonable probability that her testimony or consultation would have changed the outcome of the trial. Dr. Okla's testimony likely would have been aimed at discrediting Moltmaker's testimony. And the primary purpose of Moltmaker's testimony was bolstering A.M.'s account of the facts. But Moltmaker was not the only witness that buttressed A.M.'s account. The State also called A.M.'s mother, the nurse who examined A.M., and the detective in charge of the case. Even McQueen acknowledges that each of those witnesses corroborated the "broad elements" of A.M.'s account. Moreover, the jury heard directly from A.M. at trial. It was thus able to evaluate A.M.'s credibility for itself.
To the extent that McQueen argues that Dr. Okla's testimony would have raised doubts about the testimonies provided by A.M., her mother, the nurse, and the detective, his point is well taken. McQueen suggests that Dr. Okla might have pointed out that A.M. could have been manipulated by those parties. But McQueen's trial counsel raised that very issue during his cross-examination of Moltmaker. [[R. 8-11, PageID 690] So the jury heard that possibility; it just did not buy it.
Beyond that, the purpose of Dr. Okla's testimony was more or less fulfilled by trial counsel's cross-examination. As described previously, the cross was in-depth and targeted the very matters Dr. Okla would have touched on. The evidence simply does not demonstrate that Dr. Okla's alleged testimony would have cast significantly more doubt on Moltmaker's or A.M.'s testimony than trial counsel's cross-examination of the two. See Samatar, 225 Fed.Appx. at 372. Perhaps the jury would have given Dr. Okla's testimony more weight than counsel's cross examination. But for the reasons discussed above, that possibility alone does not mean that there is a reasonable probability that the result of the trial would have come out differently had counsel called Dr. Okla to testify or consulted with her before trial.
2. Failure to call Archie McQueen
Next, McQueen claims that his trial counsel was ineffective for failing to call his cousin Archie McQueen as an alibi witness. Here too, we disagree. McQueen's trial counsel did not perform deficiently by failing to call Archie as an alibi witness. Even if he did, his performance did not prejudice McQueen.
a. Deficient Performance
According to McQueen, Archie would have testified that he and McQueen were at a rap party until two or three in the morning on the night in question. Here, McQueen fails to overcome the presumption that trial counsel's decision not to call Archie as an alibi witness was sound trial strategy. See Strickland, 466 U.S. at 689. For one, counsel already had alibi testimony from Tawanna Patterson. Patterson testified that she and McQueen took A.M. to her grandmother's house Sunday evening (when A.M. alleged the two were at McQueen's house). She also testified that after dropping A.M. off, she and McQueen went home and spent the rest of the night together. So this is not a case where counsel's failure to investigate or call an alibi witness left a defendant high and dry, with no defense. See Hendrix v. Palmer, 893 F.3d 906, 922-23 (6th Cir. 2018).
Again, McQueen has presented no affidavit indicating what his proposed witness would testify to; all we have is McQueen's own affidavit saying he was with Archie at a rap party until two or three AM, and an affidavit from McQueen's aunt-filed during McQueen's post-conviction proceedings-stating that McQueen and Archie were "out at an affair" on Sunday evening. As noted previously, we need not decide whether the lack of an affidavit is detrimental to McQueen's claim. Even assuming Archie would have testified as alleged, McQueen is not entitled to habeas relief.
These circumstances are much more like those found in cases where we have approved counsel's decision not to call a witness when calling that witness might have exposed the defendant to more risk. See Davis v. Lafler, 658 F.3d 525, 537-38 (6th Cir. 2011). For example, here, Archie's testimony would have contradicted Patterson's testimony. Archie would allegedly have testified that he was with McQueen at a rap party on Sunday evening. But Patterson testified that she and McQueen were alone at home during that time. Not calling an alibi witness whose account would contradict another alibi witnesses's testimony is just common sense. McQueen has also not provided any evidence that trial counsel failed to adequately consider calling Archie as a potential alibi witness. So we "assume that counsel adequately considered the possibility, but ultimately decided that the best strategy was not to present [Archie's]" testimony. See Davis, 658 F.3d at 537-38.
What's more, Archie's alleged testimony would not have done much, if anything, for McQueen's defense. See id. (stating that failure to call a witness whose testimony would not have helped the defendant's cause did not amount to deficient performance). A.M.'s testimony placed McQueen at his home on Sunday evening, alone in bed. Archie's testimony would have placed McQueen at a party much later in the evening, leaving open the possibility that McQueen was with A.M. at his house sometime earlier. It was Patterson's testimony that could have convinced the jury that McQueen was not alone in bed with A.M. earlier Sunday evening.
And finally, again, McQueen's assertions that we need more evidence does not help his cause. See, e.g., Titlow, 571 U.S. at 23.
b. Prejudice
To the extent that Archie's testimony might have had any effect on the jury, it would have undermined Patterson's testimony. Calling Archie to the stand thus might have increased the probability of McQueen's conviction, not decreased it. So even if counsel was somehow deficient for failing to call Archie to the stand, that decision certainly did not prejudice McQueen.
Having determined that neither of McQueen's trial counsel claims has merit, we hold that his appellate counsel did not perform ineffectively either. "[B]y definition, appellate counsel cannot be ineffective for a failure to raise an issue that lacks merit." Greer v. Mitchell, 264 F.3d 663, 676 (6th Cir. 2001). Therefore, we cannot excuse McQueen's procedural default, and federal habeas review of McQueen's trial-counsel claims is precluded. Biros, 422 F.3d at 386.
C. Ineffective Assistance of Appellate Counsel
In addition to arguing that his appellate counsel's ineffectiveness excuses his procedural default on his trial-counsel claims, McQueen raises a standalone claim of ineffective assistance of appellate counsel. He argues that his state appellate counsel was ineffective for failing to raise the two trial-counsel claims discussed at length above. On the merits, this claim is exactly the same as the one we just reviewed, so it fails for the same reasons. But it raises thorny procedural issues which relate to whether our review is de novo or subject to AEDPA. Though wading through that procedural morass is not necessary to our disposition, we offer a few dicta on the point to guide future litigants.
1. Adjudication on the Merits
McQueen argues that the last state court to issue a reasoned opinion, the Michigan Court of Appeals, did not adjudicate his standalone appellate-counsel claim on the merits. He contends that it applied a procedural bar, so the panel's review should be de novo. The government argues that the Michigan Court of Appeals' unexplained denial of McQueen's request for an evidentiary hearing counts as an adjudication on the merits, so the panel's review should be subject to AEDPA deference. Neither argument is correct.
Orders from Michigan courts that cite to Rule 6.508(D)(3), like the order relevant in this case, do not always refer to procedural default. See Guilmette, 624 F.3d at 290-92. "Rule 6.508(D) has both a procedural and a substantive component." Id. at 291. "In some cases, the context of a brief order citing Rule 6.508(D) clearly indicates that the state appellate court is affirming the lower court's determination that a petitioner's claims are procedurally defaulted." Id. at 290 (citing Ivory v. Jackson, 509 F.3d 284, 292-93 (6th Cir. 2007)). In other cases, brief orders citing that rule are best considered adjudications on the merits. Id. at 290-91 (citing People v. Jackson, 633 N.W.2d 825, 826 (Mich. 2001) (per curiam)). And in the odd case, orders citing to the rule are ambiguous. Id. at 291. That was the situation we faced in Guilmette; it is also the situation here.
In Guilmette, after an unsuccessful direct appeal, the petitioner sought state post-conviction review, arguing for the first time that his state trial and appellate counsel had provided him ineffective assistance. Id. at 289. The Michigan trial court denied the claims on the merits, and both the Michigan Court of Appeals and the Michigan Supreme Court denied leave to appeal, issuing similar orders stating that the petitioner had failed to meet the burden of establishing entitlement to relief under Rule 6.508(D)(3). Id. Sound familiar?
We held there that "[t]he procedural-default rule stated by Rule 6.508(D)(3) applies only to claims that could have been brought on direct appeal, and thus-by necessity-it does not apply to claims of ineffective assistance of appellate counsel." Id. at 291. And we concluded that Michigan's citation to Rule 6.508(D)(3) to deny petitioner's ineffective assistance claims was thus ambiguous and deemed the order unexplained and subject to the look-through principle of Ylst v. Nunnemaker, 501 U.S. 797, 803 (1991). Id.
Here, the Michigan Court of Appeals also cited Rule 6.508(D)(3), ostensibly to deny McQueen leave to appeal the post-conviction court's adjudication of his claims, including his appellate-counsel claim. We reject McQueen's contention that the Michigan Court of Appeals thereby applied a procedural bar, regardless of its legitimacy, and that we should thus review his appellate-counsel claim de novo. We have "an independent duty to scrutinize the application of state rules that bar our review of federal claims." Cone v. Bell, 556 U.S. 449, 468 (2009). And as discussed, the Michigan Court of Appeals' citation to Rule 6.508(D)(3) could not have applied to McQueen's appellate-counsel claims. Guilmette, 624 F.3d at 291. As in Guilmette then, here, we consider the Michigan Court of Appeals' citation to Rule 6.508(D)(3) as ambiguous and deem its order unexplained and subject to the look-through principle of Ylst and its progeny, Wilson v. Sellers, 138 S.Ct. 1188, 1195 (2018). See Guilmette, 624 F.3d at 290-92.
Under Ylst and Wilson, we presume that "[w]here there has been one reasoned state judgment rejecting a federal claim, later unexplained orders upholding that judgment or rejecting the same claim rest upon the same ground." Wilson, 138 S.Ct. at 1194 (quoting Ylst, 501 U.S. at 803). Where "an earlier opinion 'fairly appear[s] to rest primarily upon federal law,' we will presume that no procedural default has been invoked by a subsequent unexplained order that leaves the judgment or its consequences in place." Ylst, 501 U.S. at 803 (citation omitted) (quoting Coleman v. Thompson, 501 U.S. 722, 740 (1991)); Wilson, 138 S.Ct. at 1194.
The earlier opinion from the post-conviction court held that McQueen's "contention involving ineffective assistance of appellate counsel amounts to a mere difference over strategy." It rested that conclusion on the oft-cited principle from Jones v. Barnes, 463 U.S. 745, 751-52 (1983) that an advocate's decision to "winnow out weaker arguments and focus on those more likely to prevail is not evidence of ineffective assistance." The court also cited to Strickland throughout its opinion. That is an adjudication on the merits. So the last reasoned decision adjudicated McQueen's appellate-counsel claim on the merits, and AEDPA applies. See 28 U.S.C. § 2254(d).
The government's argument that the Michigan Court of Appeals adjudicated McQueen's appellate-counsel claim on the merits because it denied his motion for an evidentiary hearing without explanation is incorrect. According to the government, we have previously held that a Michigan Court of Appeals' unexplained denial of an evidentiary hearing constitutes an adjudication on the merits. But the cases it cites for that proposition say no such thing. They simply apply the presumption from Harrington v. Richter: a state court has adjudicated a matter on the merits when it issues an unexplained order and there is no indication or state-law procedural principles to the contrary. See Nali v. Phillips, 681 F.3d 837, 852 (6th Cir. 2012); Hendrix, 893 F.3d at 918; Marion v. Woods, 663 Fed.Appx. 378, 382 (6th Cir. 2016). Unlike in this case, however, each of those cases dealt with a Michigan Court of Appeals decision on direct appeal- so there was no earlier state court opinion to look through to. The Supreme Court clarified in Wilson that Richter's presumption does not apply where a court can, as here, "look through" to the reasoned opinion of a lower court. Wilson, 138 S.Ct. at 1195.
2. AEDPA review "Under AEDPA, a federal court can grant habeas relief for a state court's legal error only if it '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.'" Miller, 994 F.3d at 741 (quoting 28 U.S.C. § 2254(d)(1)). Not much discussion is needed here. If McQueen cannot make out a claim of ineffective assistance of appellate counsel under de novo review, he certainly cannot do so under AEDPA. See Jackson v. Houk, 687 F.3d 723, 740-41 (6th Cir. 2012) (recognizing that AEDPA review of a state court's adjudication of ineffective assistance is "'doubly' deferential" (quoting Richter, 562 U.S. at 105)). Therefore, McQueen's standalone claim of ineffective assistance of appellate counsel does not entitle him to habeas relief.
III.
In sum, we affirm the district court's denial of McQueen's habeas petition. McQueen's trial-counsel claims were procedurally defaulted, and he fails to demonstrate cause and prejudice for the default. And his appellate-counsel claim lacks merit under de novo review, which means that it cannot pass muster under AEDPA.