Summary
denying habeas relief on AEDPA grounds after observing that a federal habeas court may choose to take a "shortcut" to Brecht in denying habeas relief
Summary of this case from Davenport v. MacLarenOpinion
Case No. 19-1588
02-24-2020
LASHON TERREL HOLLMAN, Petitioner-Appellant, v. SCOTT SPRADER, Warden, Respondent-Appellee.
NOT RECOMMENDED FOR PUBLICATION
File Name: 20a0115n.06 ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF MICHIGAN BEFORE: DAUGHTREY, KETHLEDGE, and THAPAR, Circuit Judges.
THAPAR, Circuit Judge. A jury found that Lashon Hollman murdered his neighbor. The state appeals court ruled that some of Hollman's statements to police shouldn't have reached the jury but found the error harmless beyond a reasonable doubt. The district court found that determination reasonable and denied federal habeas relief. We affirm.
I.
Margaret Torres became worried when her friend Cassandra Nelson didn't answer her calls or texts for more than two days. Torres went to Nelson's apartment and found no answer there either. As she banged on the doors and called her friend's name, Torres grew more certain that something was wrong. So she called the police.
As Torres anxiously awaited the police's arrival, she saw a young man walking down the street. She stopped him to ask whether he had seen Nelson lately. He replied that he hadn't seen her in a month or two and then walked away.
When the police arrived, they pried Nelson's door open and confirmed Torres's worst fears. Nelson was inside, lying in a pool of blood, dead from upwards of fifty stab wounds and blunt force trauma to the head. Her body was nude from the waist down.
Nelson's next-door neighbor, Hollman, soon became the prime suspect. The first time law enforcement spoke to him, Hollman said that he'd last been in Nelson's house a few weeks before her death. He also promised to take a polygraph test later that week. But he quickly stopped cooperating and never showed up for the polygraph.
The next time Hollman was interviewed, the detectives brought him down to the station in cuffs and read him his rights. See Miranda v. Arizona, 384 U.S. 436 (1966). When they made it clear that they thought Hollman was guilty, he said that he wanted a lawyer. At that point, the officers stopped questioning him and instead started inspecting him for injuries. But a few minutes later, the conversation turned back to the investigation as if nothing had happened.
As the conversation went on, Hollman admitted to the detectives that he had lied: he was in Nelson's house the night she was killed. But he insisted that he wasn't the killer. In this new story, Hollman, Nelson, and a man known as "LB" were drinking in Nelson's apartment on the night in question. ("LB" was the nickname of Lionell Beckom, an acquaintance of Nelson's whom the police had already interviewed.) At a certain point, Hollman ran home to grab some cigarettes and to use the restroom. When he returned to Nelson's place, he found LB pacing around with a knife. Hollman asked LB what he had done; LB dashed out of the apartment. Hollman then found Nelson lying on the floor with her injuries and stayed with her until she lost consciousness. He didn't call the police because he was afraid of being implicated in the crime.
Later, Hollman moved to suppress these statements, arguing that the police illegally reinitiated questioning after he had invoked his right to counsel. But the trial court allowed the interview to go to the jury with only minor redactions. After a four-day trial, the jury found Hollman guilty of Nelson's murder (plus related torture and weapons charges).
Hollman appealed the verdict. The Michigan Court of Appeals found that Hollman had invoked his right to counsel and that any statements after that should have been suppressed. But the court affirmed his convictions anyway because it found that any error was harmless beyond a reasonable doubt. See Chapman v. California, 386 U.S. 18, 24 (1967). The Michigan Supreme Court declined to review the case.
Hollman then took his claim to federal court. The district court denied habeas relief because it did not think that the state court's harmlessness analysis was unreasonable or that Hollman's challenged statements substantially influenced the jury's verdict. See 28 U.S.C. § 2254(d); Brecht v. Abrahamson, 507 U.S. 619, 623 (1993); Kotteakos v. United States, 328 U.S. 750, 776 (1946). This appeal followed.
II.
The first step in getting the right answer is asking the right question. But here we have a choice of prompts. This choice flows from the intersection of the Antiterrorism and Effective Death Penalty Act ("AEDPA") with two harmless-error tests: the Chapman reasonable-doubt test and the Brecht/Kotteakos substantial-influence test.
When a state court has found an error harmless beyond a reasonable doubt (the Chapman test), a federal habeas court has two options. The first is to ask whether the state court's Chapman analysis was "reasonable." Ruelas v. Wolfenbarger, 580 F.3d 403, 413 (6th Cir. 2009); see also Mitchell v. Esparza, 540 U.S. 12, 16-17 (2003) (per curiam). If the answer is yes, the case is over because AEDPA requires federal courts to defer to the reasonable decisions of state courts. See 28 U.S.C. § 2254(d). If the answer is no, the federal court must then decide for itself whether the error was harmless under the more government-friendly Brecht standard. Ruelas, 580 F.3d at 412 (citing Johnson v. Acevedo, 572 F.3d 398, 404 (7th Cir. 2009)).
The other option—a shortcut of sorts—is to leapfrog AEDPA and jump directly to Brecht. See Davis v. Ayala, 135 S. Ct. 2187, 2198-99 (2015) (citing Fry v. Pliler, 551 U.S. 112, 120 (2007)). But here the parties have focused mainly on whether the state court's harmlessness analysis was reasonable. So we follow their lead and ask whether the state court's application of Chapman was objectively unreasonable under AEDPA. In other words, we ask: assuming that the trial court erred by admitting Hollman's statements after he asked for counsel, did the appellate court then obviously err—so obviously that "there could be no fairminded disagreement on the question"—by finding that error harmless beyond a reasonable doubt? White v. Woodall, 572 U.S. 415, 427 (2014) (cleaned up). To that question, we must answer no.
The Michigan Court of Appeals weighed the evidence and reached the reasonable conclusion that Hollman's challenged statements didn't taint the verdict. As the court recognized, the statements weren't a confession to killing Nelson—they were the opposite. To be sure, they tended to incriminate Hollman. But they did so indirectly, not directly. The challenged statements were incriminating in two basic ways: (1) they established that Hollman was with Nelson the night of the murder; and (2) they showed that he felt like he had something to hide. The problem for Hollman is that other evidence amply established both points.
That other evidence falls into four categories: (1) DNA evidence; (2) incriminating statements Hollman made to friends; (3) evidence that Hollman took Nelson's cell phone from the crime scene; and (4) lies Hollman told the police and others.
The DNA. Hollman was a conclusive DNA match for an opened beer can that was found a foot and a half from Nelson's body. He was also a 1-in-1,786 match for the dominant DNA profile on a bread knife that was found stashed under Nelson's mattress, covered in blood. This blade was consistent with Nelson's knife wounds. It also fit the pattern of a bloody object that had been pressed against Nelson's back and left an imprint.
The statements. A friend of Hollman's, Quamay Henne, testified that Hollman said he had gone to a girl's house, got in an altercation, and "stabbed the s--- out of her." R. 6-3, Pg. ID 301-02. Hollman also showed Henne some scratches on his arm that he had received during the struggle. Henne's then-girlfriend, Candace Parish, testified that she too had seen the scratches on Hollman's arm and that Hollman had chalked them up to "a fight with a girl the night before." R. 6-13, Pg. ID 635. And Candace Parish's mother, Cynthia, testified that she overheard part of Hollman and Henne's conversation "about someone being killed," further corroborating Henne's testimony. Id. at 642.
The phone. The weekend before she died, Nelson went with Hollman to an electronics store where she bought a new Samsung cell phone with a sliding keyboard. The police didn't find that phone (or any phone) when they searched Nelson's home. But Candice Parish testified that she saw Hollman with the same kind of phone the day after Nelson's death. That was strange because Hollman's everyday phone was a cheap flip model, not one with a sliding keyboard. Not only that, Candice Parish remembered Hollman making a strange remark about how he "shouldn't have" the phone, as well as something about taking out the battery. Id. at 635. In the same vein, Henne testified that Hollman was carrying a slider phone the day after Nelson's death, but a different and less "fanc[y]" flip phone after her murder became public knowledge. R. 6-3, Pg. ID 304.
The lies. Last (but far from least), Hollman lied often. He told the police that neither he nor his girlfriend had a cell phone. (They both did.) He told them that he couldn't meet for a polygraph because he was watching his child. (The detectives had just seen the child with Hollman's girlfriend.) He told them that the last time he was in Nelson's house was two weeks before her death. (The DNA evidence made that extremely implausible, plus a neighbor saw Hollman leave Nelson's house with her on the night of her death.) And Torres identified Hollman as the young man on the street who had said the last time he saw Nelson had been a month or two earlier. (There was video footage of their trip to the electronics store just a few days earlier.)
The bottom line: taken together, all this evidence makes the state court's finding of harmless error reasonable. It's not just that the other evidence was enough to convict Hollman. See Kotteakos, 328 U.S. at 765, 767. The key point is that the other evidence supported substantially the same inferences as the challenged statements. The DNA evidence alone almost certainly would have persuaded the jury that Hollman was in Nelson's house the night she died. And that would have proved that he was lying when he told the police that he hadn't been there for several weeks (among other indisputable lies). Knowing Hollman for a liar, the jury would have had no choice but to consider his possible motives for lying: what was he trying to hide? The simplest answer: he was trying to hide the fact that he had killed Nelson, just as he had reportedly confessed to Henne.
It's true that the simplest answer was not the only conceivable one. It's also true that a jury could have disbelieved Henne's and the Parishes' damning testimony. But this jury evidently believed them, and Hollman hasn't explained how the admission of his interrogation statements might have made the difference in that credibility judgment. Hollman argues that the challenged statements corroborated Henne's testimony. But that just isn't so: the story Hollman gave the police contradicted the one he allegedly told Henne.
Hollman's other arguments fare no better. He contends, for instance, that his statements were "offered to the jury as a confession, an emotional reaction conveying an acceptance of responsibility and guilt." Appellant Br. at 23. Again, not so. To be sure, the interview showed Hollman upset. But as he presented it, he was upset because he was terrified of going to jail for a crime he didn't commit. From first to last, he steadfastly denied killing Nelson. Although the jurors might have believed (and apparently did believe) that Hollman was lying the whole time, they could not possibly have construed his statements as a "confession" to the crime.
Finally, Hollman argues that the statements were made harmful by the state's closing argument, in which the prosecutor allegedly told the jury that Hollman's admission to being present at Nelson's death was sufficient to establish aiding-and-abetting liability. But that's not what the prosecutor said. He merely explained (correctly) that on the off-chance the jury thought Hollman might have not killed Nelson alone, Hollman would still be guilty if he had participated or had helped the killer. The jury could not possibly have construed Hollman's statements as an admission to even that much. Nor is it plausible that the jury convicted on an aiding-and-abetting theory. The state didn't pursue such a theory and no evidence suggested it. Tellingly, the jury convicted Hollman of one charge (carrying a dangerous weapon with unlawful intent) for which the instructions didn't contain any complicity theory.
In the end, the state court's Chapman analysis was reasonable, so AEDPA bars relief. See Ayala, 135 S. Ct. at 2198-99. As a result, any error was harmless under Brecht as well. See Fry, 551 U.S. at 120; Ruelas, 580 F.3d at 413.
We affirm.