Freddie Owens v. Bryan Stirling

967 F.3d 396

Opinion

PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 18-8

FREDDIE OWENS, Petitioner − Appellant,

v.

BRYAN P. STIRLING, Commissioner, South Carolina Department of Corrections; WILLIE D. DAVIS, Warden of Kirkland Correctional Institution,

Respondents – Appellees.

Appeal from the United States District Court for the District of South Carolina, at Rock Hill. Terry L. Wooten, Senior District Judge. (0:16-cv-02512-TLW)

Argued: December 11, 2019 Decided: July 22, 2020

Before WILKINSON, KEENAN and DIAZ, Circuit Judges.

Affirmed by published opinion. Judge Diaz wrote the opinion, in which Judge Wilkinson and Judge Keenan joined.

ARGUED: Michael Francis Williams, KIRKLAND & ELLIS, LLP, Washington, D.C., for Appellant. Melody Jane Brown, OFFICE OF THE ATTORNEY GENERAL OF SOUTH CAROLINA, Columbia, South Carolina, for Appellees. ON BRIEF: Robert Lee, VIRGINIA CAPITAL REPRESENTATION RESOURCE CENTER, Charlottesville, Virginia, for Appellant. Alan Wilson, Attorney General, Donald J. Zelenka, Deputy Attorney General, J. Anthony Mabry, Senior Assistant Attorney General, OFFICE OF THE ATTORNEY GENERAL OF SOUTH CAROLINA, Columbia, South Carolina, for

Appellees.

3

DIAZ, Circuit Judge:

Freddie Eugene Owens was sentenced to death three times by a South Carolina jury for the 1997 murder of Irene Graves during an armed robbery of the Speedway convenience store where she worked. He appeals the district court’s grant of summary judgment in favor of the respondent state officials Bryan P. Stirling and Willie D. Davis (collectively “the State”) on his petition for writ of habeas corpus, brought pursuant to 28 U.S.C. § 2254.

Owens argues that counsel in his third capital sentencing trial provided ineffective assistance by failing to thoroughly investigate and present available mitigating evidence and to object on Confrontation Clause grounds to the trial court’s admission of state records summarizing a number of his disciplinary infractions while incarcerated. He also contends that counsel in both his third capital sentencing trial and his initial postconviction proceeding were ineffective by failing to develop evidence of frontal lobe abnormalities in his brain through comprehensive neuroimaging.

The state court rejected Owens’s first two claims on the merits in 2013 and his third claim, which Owens didn’t raise in his initial petition, as procedurally defaulted in 2017. The district court thereafter dismissed Owens’s § 2254 petition, holding that the state court reasonably applied clearly established Supreme Court law in rejecting Owens’s two exhausted claims under Strickland v. Washington, 466 U.S. 668 (1984), and that Owens failed to demonstrate cause under Martinez v. Ryan, 566 U.S. 1 (2012), to excuse the procedural default of his third Strickland claim.

We agree with the district court on all fronts. Emphasizing our deferential standard of review in regard to the exhausted claims, we decline to disturb the state court’s holdings

4

that counsel thoroughly investigated and presented Owens’s mitigating evidence and that the Confrontation Clause didn’t apply to a business record not prepared specifically for use at trial. With respect to the defaulted claim, we accentuate Martinez’s high procedural bar, which Owens fails to meet because his underlying claim is insubstantial. In so concluding, we exercise our discretion to reconsider an issue that we implicitly resolved in Owens’s favor by granting his certificate of appealability. Accordingly, we affirm.

I.

We first sketch the long chain of events giving rise to this capital habeas action, which encompass a criminal trial, three sentencing trials, three rounds of direct appeal, and two rounds of state postconviction proceedings, in addition to the proceedings in the district court.

A.

Owens’s conviction for murder, armed robbery, conspiracy to commit armed robbery, and use of a firearm in the commission of a violent crime traces back to October 31, 1997. Early that Halloween evening, Owens was driving around his hometown of Greenville, South Carolina with his three co-defendants—Andre Golden, Nakeo Vance, and Lester Young—when the foursome conspired to hold up a series of local businesses. After “casing” several options, two of them (sans Owens) robbed the Prestige Cleaners on Lauren’s Road at around 6:45 p.m. Later in the evening, all four robbed the Conoco Hot Spot convenience store on Augusta Road.

5

After midnight on November 1, the four men conferred on Owens’s front porch about splitting up and robbing two more businesses near the intersection of Lauren’s Road and I-85. One was a Waffle House, which was assigned to Vance and Young (but which they found too crowded to carry out their plan). The other was the Speedway convenience store, which was assigned to Owens and Golden.

Security footage from inside the store showed two men wearing makeshift disguises over their heads—one a ski mask and the other a pair of panty hose—entering at around 4:00 a.m. The masked men accosted Graves (a single mother of three who was working as many jobs), removed what turned out to be $37.29 from the register, and led Graves at gunpoint to the back of the store, where the safe was located. When Graves couldn’t open the safe because she didn’t know the combination, the man in the ski mask shot her in the right side of the head with a .32 caliber pistol, killing her instantly.

B.

Owens and his companions were indicted the following October, and the State promptly filed a notice of intent to seek the death penalty for Graves’s murder. Owens was tried alone for that crime beginning on February 8, 1999. Lacking forensic evidence to connect Owens to the scene, the State’s case rested largely on witness testimony. Golden (who had since pleaded guilty) testified to the events described above, including that Owens was the man in the ski mask who pulled the trigger. Vance also testified for the State, adding that Owens took credit for having “shot that bitch in the head” after hopping in Vance’s getaway car. J.A. 1329. Owens’s then-girlfriend testified that he had confessed to having shot the clerk to her, too. Owens also confessed to a detective and an investigator

6

who had been assigned to the case and who likewise testified for the State. On this evidence, a jury returned a guilty verdict on all counts on February 15.

The sentencing phase of Owens’s trial began two days later. This phase was separate from the guilt phase pursuant to South Carolina law, which provides that “the court shall conduct a separate sentencing proceeding . . . upon [the] conviction or adjudication of guilt of a defendant of murder” in all cases where “the State seeks the death penalty.” S.C. Code Ann. § 16-3-20(B). If the sentencing jury (or the judge in non-jury cases) unanimously finds at least one statutory aggravating circumstance beyond a reasonable doubt and recommends that a sentence of death be imposed, “the trial judge shall sentence the defendant to death.” Id. § 16-3-20(C). The statutory aggravating circumstances include, as relevant here, that the murder was committed during an armed robbery or during an armed larceny. See id. § 16-3-20(C)(a)(1)(e), (f). The statute also sets forth a nonexclusive list of mitigating circumstances that the jury (or judge) must be allowed to consider in reaching a sentencing verdict. See id. § 16-3-20(C)(b).

On the morning of trial, defense counsel notified the court that they had been served with a Supplemental Notice of Aggravation the previous afternoon, which indicated the State’s intent to introduce—as evidence of Owens’s future dangerousness and inability to adapt to incarceration—a statement Owens made that morning confessing to the murder of a fellow inmate named Christopher Lee. Defense counsel moved for a brief continuance to investigate the circumstances of the confession, but the court denied the motion. The sentencing hearing began, evidence of Lee’s murder was introduced, and the jury returned

7

a unanimous verdict finding the statutory aggravating circumstance of murder while in the commission of armed robbery and recommending that Owens be sentenced to death.

On appeal, the Supreme Court of South Carolina affirmed Owens’s conviction but vacated his capital sentence and remanded for resentencing, finding that the trial court had abused its discretion in refusing to grant a continuance. State v. Owens, 552 S.E.2d 745, 759 (S.C. 2001). The high court reasoned that in light of “the capital nature of the proceeding” and “the timing of [Owens’s] statement” regarding Lee’s death, “due process necessitated a brief . . . continuance to allow defense counsel the opportunity to interview the inmates and personnel at the detention center.” Id.

C.

Owens elected a bench trial for his second capital sentencing proceeding. The trial court sentenced Owens to death anew, but South Carolina’s high court again vacated the sentence and remanded for another resentencing. See State v. Owens, 607 S.E.2d 78, 80 (S.C. 2004). This time, the Supreme Court found that Owens’s waiver of his right to a jury trial wasn’t voluntary under South Carolina law because the trial court had impermissibly injected its personal opinion into Owens’s decision. Id. at 79–80.

D.

Owens’s third capital sentencing trial forms the basis of the ineffective-assistance claims raised in his § 2254 petition. Owens was represented in that proceeding by Everett P. Godfrey, Jr. and Kenneth C. Gibson (collectively “Sentencing Counsel”), who were appointed on February 2, 2006, for a trial scheduled to begin on October 2. Godfrey took immediate responsibility for Owens’s mitigating evidence, while Gibson took charge of

8

the guilt- or crime-related evidence. Closer to trial, once they realized that there was “a whole lot more” mitigating evidence than guilt-related evidence to cover, Gibson took on secondary responsibility for presenting Owens’s mitigation case. J.A. 2095.

Counsel spent much of the first six months of their appointment on preliminary tasks, such as performing legal research, reviewing the case files from Owens’s previous trials and lawyers, and, closer to summer, performing general trial preparation. They also twice visited Owens at Lieber Correctional Institute, outside of Charleston, and Godfrey visited Owens a third time in early October, after the trial had been pushed back.

In mid-August, Godfrey began to assemble a team of specialists to help investigate and present Owens’s mitigating evidence. The team, which came together by September, consisted of five members: clinical social worker Marjorie Hammock, neuropsychologist Dr. Tora Brawley, forensic psychiatrist Dr. Donna Schwartz-Watts, and mitigation investigators Paige Tarr and Carolyn Graham.

The centerpiece of Owens’s mitigation case, Hammock had testified in each of his previous sentencing trials; Godfrey selected her not only because she “knew the case” and “knew the facts,” J.A. 2195, but because Owens’s previous sentencing counsel “was very happy with her” presentation, J.A. 2157. Hammock’s role was to “give the jury an understanding . . . of [the] things that had gone on” in Owens’s life, especially with respect to his troubled “family background,” that “led up to” his offense conduct. J.A. 2102. Tarr and Graham served as Hammock’s boots on the ground, “go[ing] out” to “find witnesses, interview witnesses, [and] identify issues . . . that . . . might mitigate the circumstances” of Owens’s conviction. J.A. 2108.

9

As for Dr. Schwartz-Watts, Godfrey sought her out because they had worked together on numerous cases in the past, and he liked the work she had done in them. The two discussed at length the need to bring on a neuropsychologist, and Dr. Schwartz-Watts recommended Dr. Brawley, whom she knew from previous cases as well.

Like Godfrey, Dr. Schwartz-Watts knew that the role of neuropsychologist had been fulfilled in the previous sentencing trial by Dr. James Evans, who had evaluated Owens by means of a Quantitative Electroencephalogram (or “qEEG” for short)—a diagnostic tool that analyzes behavioral-cognitive function by measuring the brain’s electrical activity— and testified that Owens has “mild brain dysfunction” in his frontal lobe. J.A. 76. But Dr. Schwartz-Watts preferred Dr. Brawley’s more “conservative” approach, which she found more “reliable” than Dr. Evans’s. J.A. 2382.

Godfrey agreed that Dr. Evans’s use of a qEEG to assess Owens’s cognitive state—

which had to be sent “out west, like to California,” for diagnosis—“would [not] play in front of a local jury,” and that it would be prudent to take a more cautious approach. J.A. 2196. Accordingly, when Dr. Brawley performed her neuropsychological evaluation of Owens on September 12, her “battery” of tests, J.A. 2404, didn’t include a qEEG, which she too found to be “controversial” and “experimental,” J.A. 2402.

For her part, Dr. Schwartz-Watts met with Owens three times before trial. These interviews taught her that Owens had witnessed and experienced “a lot of abuse” over the years, though Owens denied having ever been sexually abused. J.A. 2383. Dr. Schwartz- Watts also learned that Owens was “receiving treatment for a presumed bipolar disorder” and taking “powerful psychiatric medications,” including an anticonvulsant called

10

Depakote, while incarcerated at Lieber. J.A. 2440. Because “this new evidence” meant that “some of the mitigation that had been completed in prior trials [would] need to be revamped,” she urged Godfrey to request a continuance. Id. He did so, and on October 2 (the date for which the trial was originally scheduled), the trial court delayed the trial by five weeks, rescheduling for November 6.

Owens’s third capital sentencing trial took place from Monday, November 6 to Saturday, November 11, 2006. The State called a dozen witnesses, including various medical experts, several Greenville county sheriffs and special investigators, Nakeo Vance, Owens’s girlfriend at the time of Graves’s murder, and a state official from Lieber. As in Owens’s previous sentencing trials, the State’s evidence touched equally on his offense conduct as his future dangerousness and inability to adapt to prison.

Among other things, evidence of the latter included testimony about Owens’s killing of his fellow inmate on the eve of his first sentencing trial—the “elephant in the room,” as Godfrey later referred to it. J.A. 2147. It also included an official record describing twenty- eight of the numerous disciplinary infractions that Owens had received while in the custody of the South Carolina Department of Corrections (SCDOC),which had been prepared by officials “under a duty to report” the incidents pursuant to state law. J.A. 1530.

For the defense, Godfrey’s opening remarks stated that the “evidence in mitigation”

would show “how it is that Freddie grew up,” especially with regard to “the violence that he ha[d] lived through his entire life,” such that the jury would “come to know him as a person.” J.A. 1194. Godfrey highlighted that Owens suffered from “an impulse control disorder,” J.A. 1196, but that “within th[e] past year” he had begun “asking for help . . . .

11

to control his actions,” J.A. 1194. And while Godfrey warned the jury that the mitigating evidence often wasn’t “pretty,” he asserted that Owens’s difficult upbringing, together with his underlying and long-untreated disorders, revealed that life was “the proper sentence.” J.A. 1196.

Counsel called five witnesses, starting with Hammock. While her presentation was more succinct than in Owens’s previous sentencing hearings, Hammock covered her psychosocial assessment in similar terms. Hammock testified that she had interviewed Owens, his mother, his sisters, one of his brothers, his stepfather, staff from the South Carolina Department of Juvenile Justice (SCDJJ) when he was incarcerated there, various medical experts, and others yet; and had reviewed many records.

Hammock explained that Owens was born to an eighteen-year-old single mother “who had very limited resources” and “difficulty caring for her family,” J.A. 1552; that there was “a great deal of violence” in both his family—largely emanating from his father and stepfather, and often directed at his mother, his siblings, and himself—and neighborhood alike, id.; that he and his siblings were put in foster care when he was four due to “abuse and neglect,” which didn’t cease after his mother regained custody, id.; that there was “a considerable amount” of incarceration, mental illness, and substance abuse in his family, J.A. 1554; and that he suffered from “significant learning disabilities” and “educational deficits,” J.A. 1553. She also stated that Owens was taught to fight “at a very early age” and developed “the kind of coping strategies that made him always on the defensive.” J.A. 1553–54. She concluded by remarking that such troubled upbringings have been shown to increase the likelihood that a child will later engage in violent crime.

12

Counsel’s second witness was Fain Cones Maag, Owens’s third-grade teacher. She testified about Owens’s “sparse” and violence-ridden home environment as well, J.A. 1562, relating that bullies would often chase him around the neighborhood and that, in response, his stepfather would lock him out in order to teach him that he “had to fight,” J.A. 1563. Maag also provided more detail about Owens’s “learning deficiencies,” including his “tremendous trouble” learning to read, J.A. 1562, while adding that Owens “had some real gifts,” J.A. 1564.

Third up was Dr. Brawley. She testified about the “battery” of neuropsychological tests she had conducted to assess “how the different areas of [Owens’s] brain [were] functioning” and the conclusions she had drawn from them. J.A. 1573. In this respect, Dr. Brawley testified that Owens had “a history of lifelong problems with brain function” and related “psychiatric issues,” such as impulsivity, irritability, and depression; “some select areas of deficit,” particularly in the verbal areas of the brain; and a history of head injury. J.A. 1581. But on cross-examination, Dr. Brawley clarified that, in her professional opinion, Owens’s cognitive defects didn’t “rise to the level of . . . diagnosing him with . . . any kind of brain syndrome” or “mental illness.” J.A. 1582–83. She also stated that she didn’t “put a whole lot of clinical significance” on Owens’s head injuries. J.A. 1582.

Dr. Brawley was followed by Dr. Thomas Cobb, a forensic psychiatrist with SCDOC who had been treating Owens at Lieber since August 2005. Dr. Cobb had diagnosed Owens with an unspecified anxiety disorder; an unspecified impulse control disorder, meaning “an inability to not react to something [in a way] that may cause . . . harm” to oneself or another; and Antisocial Personality Disorder, which he attributed to

13

Owens’s stunted emotional development, and to which he attributed Owens’s impulsivity and irritability. J.A. 1593. Dr. Cobb explained that he had prescribed Owens the powerful psychiatric medications that Dr. Schwartz-Watts had noticed in order to treat these conditions, including Depakote, which he was using in combination with Risperdal (a drug commonly used to treat disorders like bipolarism and schizophrenia) to stabilize Owens’s mood and give his brain “time to think.” J.A. 1595. He concluded by opining that these and other medications had worked some improvement in Owens’s behavior.

Counsel’s fifth and final witness was Dr. Schwartz-Watts. She opined that Owens’s “long history of illegal behaviors” was rooted in his well-documented impulsivity. And because Owens had “a history of head injury” as well, Dr. Schwartz-Watts considered whether his impulsivity was the result of brain damage. But instead of brain damage, she concluded that Owens’s impulsivity was the product of three disorders: Attention Deficit Disorder, Dysthymic Disorder (i.e., chronic depression), and Antisocial Personality Disorder (in agreement with Dr. Cobb).

Dr. Schwartz-Watts also addressed how the circumstances of Owens’s “developmental history” as a young child, including his experiences with abuse, neglect, and violence, had contributed to these disorders, and therefore to Owens’s impulsiveness and proclivity to violence. Id. Finally, Dr. Schwartz-Watts echoed Dr. Cobb in opining that Owens had improved since taking the medications that Dr. Cobb had prescribed, and could continue to receive those medications while in prison.

Owens’s third capital sentencing trial wrapped up shortly after Dr. Schwartz-

Watts’s testimony, and that same day, the jury returned a verdict unanimously finding both

14

of the statutory aggravating circumstances that the State had charged—“[t]hat the murder was committed while in the commission” both “of robbery while armed with a deadly weapon” and “of larceny with the use of a deadly weapon”—and recommending that Owens be sentenced to death. J.A. 1758, 1768, 1771; cf. S.C. Code Ann. § 16-3- 20(C)(a)(1)(e), (f). The trial court entered judgment accordingly.

This time on direct appeal, the Supreme Court of South Carolina affirmed Owens’s capital sentence. State v. Owens, 664 S.E.2d 80, 82 (S.C. 2008). The United States Supreme Court then denied Owens’s petition for writ of certiorari. Owens v. South Carolina, 555 U.S. 1141 (2009).

E.

Owens then filed a pro se petition for post-conviction relief in state court. Keir M.

Weyble (of Cornell Law School), who had worked on Owens’s petition for certiorari, and Emily C. Paavola (then a fellow of Cornell’s Death Penalty Project) (collectively “Initial Postconviction Counsel”) were appointed to represent Owens, and ultimately filed two amended petitions on his behalf. The operative petition asserted over a dozen separate grounds of ineffective assistance, several of which have been consolidated into the first two claims raised in Owens’s § 2254 petition.

The state postconviction court held an evidentiary hearing on Owens’s petition. All but one of the witnesses called at the hearing had participated in one or more of Owens’s sentencing trials, including Godfrey and Gibson, Drs. Schwartz-Watts and Brawley, Owens’s previous counsel, and a mitigation investigator named Drucy Glass, who had worked on the first two trials.

15

The other witness, called by Initial Postconviction Counsel, was Dr. James Garbarino, an expert in child development. Dr. Garbarino, who had interviewed Owens for four hours, testified about children’s vulnerability to trauma and other “risk factors” that have been shown to have a negative impact on a child’s development and to increase the likelihood that a child will later engage in violent crime. J.A. 2256. He opined that Owens “had almost all of the risk factors that [he had] ever read about.” J.A. 2267. He also related that Owens had revealed during their interview that he had been sexually abused several times as a child, including during his time in SCDJJ custody, though his official records didn’t document the abuse.

The state court denied Owens’s petition, for reasons that we discuss below (to the extent relevant here). See Owens v. State, No. 2009-CP-23-0741 (S.C. Com. Pl. Feb. 13, 2013). The Supreme Court of South Carolina denied Owens’s petition for writ of certiorari, and Owens declined to seek a writ of certiorari from the United States Supreme Court.

F.

Owens then commenced this federal habeas action by moving in the district court to stay his execution and appoint new counsel. The district court granted the motions, and Owens’s new counsel (“Federal Habeas Counsel”) filed the operative petition for writ of habeas corpus pursuant to 28 U.S.C. § 2254.

Federal Habeas Counsel also filed a second petition for postconviction relief in state court, raising eight new ineffective-assistance claims, including the defaulted claim raised in this appeal. While these new claims would likely be (and later were) denied on procedural grounds due to Owens’s failure to raise them in his initial postconviction

16

petition, Owens could (and does) attempt to excuse the procedural default in federal court by showing cause and prejudice under Martinez and Coleman v. Thompson, 501 U.S. 722 (1991).

In further pursuit of Owens’s neuroimaging claim, Federal Habeas Counsel obtained a comprehensive neurobehavioral assessment of his brain from Dr. Ruben C. Gur, an expert in brain behavior. Dr. Gur’s assessment comprised structural and functional neuroimaging using both Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) techniques.

Dr. Gur concluded that the neuroimaging showed “abnormalities indicating brain damage” in regions of the brain “important for regulating emotions and behavior,” J.A. 4182, suggesting that Owens’s frontal lobe was “unable to do its job and act as the brakes on the primitive emotional impulses” emanating from his “hyper-activated” amygdala, J.A. 4183. Dr. Gur’s assessment was in turn reviewed by Dr. Stacey Wood, a forensic neuropsychologist, whose independent review and evaluation concurred that Owens “has significant brain impairment” in his frontal lobe, J.A. 4200, resulting in “neuropsychological deficits related to . . . executive functioning,” J.A. 4201.

After receiving the results of Drs. Gur’s and Wood’s assessments, Federal Habeas Counsel filed an amended § 2254 petition in the district court, adding (among others) the defaulted neuroimaging claim. They then moved to stay the § 2254 action pending the state court’s resolution of Owens’s second petition, which the district court granted. The state court ultimately denied Owens’s second petition as procedurally defaulted, and Owens didn’t appeal that denial.

17

G.

Owens thereafter resumed proceedings in the district court, where his operative § 2254 petition asserted five exhausted grounds and seven unexhausted grounds for relief. The State filed a (second) motion for summary judgment, which was initially considered by a magistrate judge. The magistrate judge issued a Report and Recommendation (“R&R”) advising that the district court grant summary judgment in favor of the State and deny Owens’s petition and related motions for an evidentiary hearing. Owens v. Stirling, No. 0:16-CV-2512-TLW-PJG, 2018 WL 3104276, at *1 (D.S.C. Jan. 12, 2018). Owens objected to the R&R, but the district court adopted it in full. Owens v. Stirling, No. 0:16- CV-02512-TLW, 2018 WL 2410641, at *9, *45 (D.S.C. May 29, 2018). The court then denied Owens’s motion to alter or amend the judgment. Owens v. Stirling, No. 0:16-CV- 02512-TLW, 2018 WL 5720445, at *3 (D.S.C. Nov. 1, 2018).

H.

We docketed Owens’s case on November 16, 2018. On November 27, we granted Owens’s motion to substitute counsel. Counsel thereafter sought a certificate of appealability (“COA”), pursuant to 28 U.S.C. § 2253(c)(1), on the three claims raised here. We granted the certificate as requested, indicating that Owens had “demonstrate[d] ‘a substantial showing of the denial of a constitutional right’” with respect to each claim. See Miller-El v. Cockrell, 537 U.S. 322, 327 (2003) (quoting 28 U.S.C. § 2253(c)(1)). This appeal followed.

18

II.

Owens’s § 2254 petition presents two exhausted claims of ineffective assistance, meaning that the state court rejected them on the merits. First, he claims that Sentencing Counsel provided ineffective assistance by failing to adequately investigate and present available mitigating evidence about his upbringing, family background, and general social history. Second, he claims that counsel was ineffective by failing to raise a readily viable Confrontation Clause objection to the trial court’s admission of the State’s record summarizing twenty-eight of his disciplinary infractions while in custody.

We consider these claims in turn, “reviewing de novo the district court’s denial of [Owens’s] petition” with respect to each. Gray v. Zook, 806 F.3d 783, 790 (4th Cir. 2015). But we review the state postconviction court’s denial of these claims only to the extent of determining whether it involved an unreasonable application of the Supreme Court’s clearly established precedent. And we conclude that it didn’t.

A.

Our standard of review derives from the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), which circumscribes a federal court’s ability to issue a writ of habeas corpus “on behalf of a person in custody pursuant to the judgment of a State court.” 28 U.S.C. § 2254(d). Under AEDPA, a federal court may not grant habeas relief on a claim that the state postconviction court rejected on the merits unless that court’s determination “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding,” id. § 2254(d)(2), or “was contrary to, or involved an

19

unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States,” id. § 2254(d)(1).

Owens relies on § 2254(d)(1), contending that the state court’s decision involved an unreasonable application of clearly established federal law under the Supreme Court’s Strickland line of cases. A state court’s decision involves an unreasonable application of such clearly established law when the court “identifies the correct governing legal rule from th[e Supreme] Court’s cases but unreasonably applies it to the facts of the particular state prisoner’s case.” Williams v. Taylor, 529 U.S. 362, 407 (2000) (O’Connor, J., delivering the opinion of the Court with respect to Part II).

By “clearly established,” § 2254(d)(1) “refers to the holdings, as opposed to the dicta, of th[e Supreme] Court’s decisions as of the time of the relevant state-court decision.” Id. at 412. And to be “unreasonable,” the state court’s application of that law must be “objectively unreasonable,” not simply incorrect. Barnes v. Joyner, 751 F.3d 229, 238–39 (4th Cir. 2014); see also Williams, 529 U.S. at 412 (“[A]n unreasonable application of federal law is different from an incorrect or erroneous application of federal law.”). Otherwise stated, “[a] state court’s determination that a claim lacks merit precludes federal habeas relief so long as fairminded jurists could disagree on the correctness of the state court’s decision.” Harrington v. Richter, 562 U.S. 86, 101 (2011) (cleaned up).

In enacting AEPDA, Congress thus recognized that federal courts “owe state tribunals significant deference” with respect to their determination that a state prisoner isn’t entitled to habeas relief. Bennett v. Stirling, 842 F.3d 319, 323 (4th Cir. 2016). Indeed, AEDPA “reflects the view that habeas corpus is a guard against extreme malfunctions in

20

the state criminal justice systems, not a substitute for ordinary error correction.” Harrington, 562 U.S. at 102 (cleaned up). While our standard of review by no means “preclude[s] relief” or “impl[ies] abandonment or abdication of judicial review,” Miller- El, 537 U.S. at 340, it does mean that we may not “second-guess the reasonable decisions of state courts,” Renico v. Lett, 559 U.S. 766, 779 (2010).

Our deference in this case contains an additional layer. For where, as here, a state prisoner claims ineffective assistance of counsel as the basis for habeas relief, we must review the claim through the “highly deferential” lens of Strickland as well. Richardson v. Branker, 668 F.3d 128, 139 (4th Cir. 2012) (cleaned up). AEDPA and Strickland thus provide “dual and overlapping” lenses of deference, which we apply “simultaneously rather than sequentially.” Id. And because “[s]urmounting Strickland’s high bar is never an easy task,” it is “all the more difficult” to establish “that a state court’s application of Strickland was unreasonable . . . under § 2254(d).” Morva v. Zook, 821 F.3d 517, 528 (4th Cir. 2016) (cleaned up). “This double-deference standard effectively cabins our review” to determining “whether there is any reasonable argument that counsel satisfied Strickland’s deferential standard.” Id. (cleaned up).

To establish ineffective assistance of counsel under Strickland, a defendant must satisfy two standards: (1) “that counsel’s performance was deficient,” and (2) that counsel’s deficient performance “prejudiced the defense.” 466 U.S. at 687. The first prong, deficient performance, requires a showing “that counsel’s representation fell below an objective standard of reasonableness,” as measured by “prevailing professional norms” and in light of “all the circumstances” of the representation. Id. at 688. While such professional norms

21

may be “reflected in American Bar Association [ABA] standards and the like,” such guides are just that—“only guides”—for determining what constitutes reasonable representation in a given case, id. at 688, and no fixed set of rules may “take account of the variety of circumstances faced by defense counsel,” id. at 688–89.

In assessing counsel’s performance, our scrutiny “must be highly deferential.” Id.

at 689. Because “[i]t is all too tempting for a defendant to second-guess counsel’s assistance after conviction or adverse sentence,” and “all too easy for a court, examining counsel’s defense after it has proved unsuccessful, to conclude that a particular act or omission . . . was unreasonable,” Strickland cautions that “[a] fair assessment of attorney performance requires that every effort be made to eliminate the distorting effects of hindsight . . . and to evaluate the conduct from counsel’s perspective at the time.” Id. We must therefore “indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Id. In all, the “critical question” is whether counsel’s performance “amounted to incompetence under prevailing professional norms, not whether it deviated from best practices.” Winston v. Pearson, 683 F.3d 489, 504 (4th Cir. 2012) (cleaned up).

Once a defendant has established that counsel’s performance was deficient, he must then prove that counsel’s deficient performance prejudiced his defense under Strickland’s second prong. In the effective assistance context, prejudice means “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694. A reasonable probability, in turn, is one “sufficient to undermine confidence in the outcome.” Id. In the capital sentencing context,

22

“the question is whether there is a reasonable probability that, absent the errors, the sentencer . . . would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Id. at 695.

With these dually deferential standards of review firmly in mind, we turn now to consider Owens’s claim of ineffective assistance.

B.

Owens contends that Sentencing Counsel was ineffective by failing to adequately investigate and present available mitigating evidence about his life history. Most of his arguments here focus on counsel’s allegedly “less than complete investigation,” such that their choices about which mitigating evidence to present could not have been the products of “reasonable professional judgments.” See id. at 691. Together with evidence that counsel allegedly possessed but didn’t present, he asserts that a reasonably complete “mitigation story” would have “neutralized the State’s aggravating evidence by explaining [his] behavior through the lens of his past experiences.” Pet’r’s Br. 48.

The state postconviction court rejected Owens’s claim, reasoning that counsel “conducted a thorough investigation into potential mitigating evidence and chose to present evidence that [they] thought would favor Owens at trial.” J.A. 3704. While the court observed that counsel didn’t discover or present every iota of available mitigating evidence, it concluded that they developed “a cogent mitigation case through the testimony of Hammock, Schwartz-Watts, Cobb, Brawley, and Maag,” which substantially covered all of the evidence that Owens claims to have been left out. J.A. 3707. The court also concluded that any mitigating evidence that counsel failed to investigate or present wasn’t

23

prejudicial because the additional evidence presented in the postconviction proceeding “would [not] have struck a different balance” between the total aggravating and mitigating evidence. J.A. 3705 (quoting Gray v. Branker, 529 F.3d 220, 238 (4th Cir. 2008)).

We think that the state court reasonably applied the Supreme Court’s holdings in determining that counsel adequately investigated (and competently presented) Owens’s mitigating evidence. Of course, the Court has longed recognized that capital sentencing counsel have an “obligation to conduct a thorough investigation of the defendant’s background,” Williams, 529 U.S. at 396, in an effort “to discover all reasonably available mitigating evidence,” Wiggins v. Smith, 539 U.S. 510, 524 (2003) (cleaned up). But the issue turns on how “thorough” the mitigation investigation must be. And here, the Court’s cases indicate that the investigation need only be reasonably thorough. See Strickland, 466 U.S. at 690–91 (equating capital sentencing counsel’s duty of “thorough investigation” to “a duty to make reasonable investigations” and noting that “[i]n any ineffectiveness case, a particular decision not to investigate must be directly assessed for reasonableness in all the circumstances”); cf. Wiggins, 539 U.S. at 533 (“Strickland does not require counsel to investigate every conceivable line of mitigating evidence . . . .”).

In Williams, for instance, the Court found (for the first time in the capital sentencing context) the state court’s denial of post-conviction relief unreasonable under § 2254(d)(1) on the ground that counsel had “failed to conduct” any investigation into the defendant’s background. See 529 U.S. at 395. There, counsel overlooked even the most basic available mitigating evidence about the defendant’s “nightmarish childhood,” id., including with respect to his abusive and neglectful parents, his time in foster care and the juvenile justice

24

system, his borderline intellectual disability, and his failure to advance past sixth grade. Id. at 395–96. Instead, “the sole argument in mitigation” that counsel advanced was that the defendant had “turned himself in.” Id. at 398. And that clearly wasn’t enough.

The Court reached the same conclusion in Wiggins, on the basis that capital sentencing counsel had abandoned their investigation of the defendant’s background after reviewing just two sources—the presentence report and certain records from the Baltimore City Department of Social Services—and thus failed to discover any evidence of his history of severe sexual abuse, his diminished mental capacity, and his frequent periods of homelessness. 539 U.S. at 524–35. Similarly, in Rompilla v. Beard, the Court found it “obvious” that counsel’s performance “fell below the level of reasonable performance,” where they had failed “to look at any part” of the defendant’s conviction file—“a public document[] readily available for the asking at the very courthouse where [he] was to be tried”—despite notice that the state intended to introduce portions of the file as aggravating evidence. 545 U.S. 374, 383–84 (2005); see also Andrus v. Texas, --- S. Ct. ----, No. 18- 9674, 2020 WL 3146872, at *1 (U.S. June 15, 2020) (per curiam) (finding deficient performance where counsel “failed even to look for” evidence of the defendant’s “grim” life history); Porter v. McCollum, 558 U.S. 30, 40 (2009) (per curiam) (same where counsel “failed to uncover and present any evidence of [the defendant’s] mental health or mental impairment, his family background, or his military service”).

In Bobby v. Van Hook, by contrast, the Supreme Court found reasonable the state court’s determination that capital sentencing counsel had adequately investigated the defendant’s mitigating evidence under circumstances similar to those present here. There,

25

the Court emphasized that counsel had interviewed the defendant’s parents, aunt, and “a family friend whom [the defendant] visited immediately after the crime,” id. at 9; consulted with two expert witnesses (though not “an independent mental-health expert,” id. at 6) “more than a month before trial,” id. at 9; and reviewed the defendant’s military and medical records, id.

The Court also underscored the evidence that counsel discovered (and presented at trial) from their investigation, including that the defendant’s parents were heavy drinkers and that the defendant himself “started drinking as a toddler,” id. at 10; that he “grew up in a combat zone” and “watched his father beat his mother weekly,” id. (cleaned up); that he “attempted suicide five times,” id.; and that he suffered from borderline personality disorder, id. at 11. In light of such extensive mitigating evidence, the Court reasoned that counsel’s failure “to find more” wasn’t clearly deficient. Id. at 11–12.

This court has reached similar conclusions as well. In Morva, for example, we found reasonable the state court’s “decision on deficient performance” where capital sentencing counsel had “hired a mitigation expert,” interviewed many of the petitioner’s family members, and presented thirteen witnesses (including several mental-health experts) who testified about the defendant’s absent and neglectful parents, his “nomadic lifestyle and homelessness as a young adult,” his “ongoing health problems,” and his “odd . . . beliefs and behavior.” 821 F.3d at 529–30. We held the same in DeCastro v. Branker, where counsel had investigated the defendant’s “personal history” by interviewing the defendant, his mother, and his aunt; had reviewed his “school and criminal records”; had “retained an investigator”; and had obtained a psychiatric evaluation. 642 F.3d 442, 456

26

(4th Cir. 2011). And conversely, we’ve gone the other way in cases where counsel’s investigation was significantly lacking, such as where counsel altogether “failed to investigate for mental health evidence.” See Gray, 529 F.3d at 229.

In light of these cases, the state postconviction court’s conclusion that Sentencing Counsel’s mitigation investigation was reasonably thorough doesn’t warrant relief under our dually deferential standards of review. Indeed, counsel’s efforts to discover mitigating evidence far exceeded the efforts made in Williams, Wiggins, Rompilla, and Porter, more closely resembling (if not exceeding) those made in Bobby.

Recall that counsel retained two investigators to help discover Owens’s mitigating evidence, a clinical social worker to help assemble Owens’s social history, and two medical experts to help evaluate Owens’s cognitive functioning. Altogether, this team interviewed the majority of Owens’s immediate family members, in addition to others (such as his third- grade teacher) who knew him during his formative years; consulted numerous other experts, including his treating psychiatrist; and reviewed extensive family, school, medical, incarceration, and other records. And it yielded testimony from five witnesses, who painted a clear picture of Owens’s impoverished, neglectful, and abusive childhood; his pervasive exposure to violence, substance abuse, and incarceration; and his various learning disabilities and cognitive disorders.

Moreover, counsel undertook their investigation with the benefit of two previous trials’ worth of mitigating evidence to draw from, comprising several case files and including an array of additional expert perspectives. In light of all of this, the state court reasonably determined that “[t]his is not a case in which the defendant’s attorneys failed to

27

act while potentially powerful mitigation evidence” was available, but rather one where their “decision not to seek more . . . than was already in hand fell within the range of professionally reasonable judgments.” See Bobby, 558 U.S. at 11–12 (cleaned up).

We are also mindful that “the more general the federal rule” in question, “the more leeway state courts have in reaching outcomes in case-by-case determinations.” Bennett, 842 F.3d at 322 (quoting Yarborough v. Alvarado, 541 U.S. 652, 664 (2004) (alterations adopted)). Whereas the proper application of a “specific” legal rule “may be plainly correct or incorrect” in many cases, the contours of “more general” rules or standards “must emerge in application over the course of time,” thereby demanding “a substantial element of judgment” in many cases. Yarborough, 541 U.S. at 664. Here, we think it plain that the Supreme Court’s thoroughness standard is general enough that state postconviction courts must use substantial judgment in applying it to cases where counsel’s investigation is neither clearly adequate nor clearly inadequate. We are therefore confident that the state court’s judgment is entitled to deference in this case.

We reject Owens’s counterarguments. In particular, Owens complains that his lawyers waited too long to begin investigating and delegated too much of the responsibility to their team members. Yet the nearly two months that counsel and their team spent investigating far exceeded the one week at issue in Williams, see 529 U.S. at 363, and roughly equaled the amount of time in Bobby, see 558 U.S. at 9–10. And especially because counsel had the benefit of two trials’ worth of mitigating evidence to build from, the amount of time they devoted to additional investigation was reasonably sufficient.

28

Nor is this a case in which counsel “abdicated their responsibility” for investigating Owens’s mitigating evidence. See Winston, 683 F.3d at 505. As our cases recognize, capital sentencing counsel “often must necessarily rely on . . . investigators, experts, and other members of the defense team to gather essential evidence.” Winston v. Kelly, 784 F. Supp. 2d 623, 632 (W.D. Va. 2011), aff’d sub nom., Winston v. Pearson, 683 F.3d 489 (2012). The corollary is that counsel themselves must “be familiar with readily available documents,” among other evidence, “necessary to an understanding of [the defendant’s] case.” Winston, 683 F.3d at 505 (cleaned up).

That’s precisely what happened here. While counsel employed a team of specialists to gather Owens’s available mitigating evidence, they closely supervised the team’s efforts and familiarized themselves with the findings. Gibson’s and Godfrey’s time sheets indicate that they each spent many hours, especially during the two months before trial, studying their team members’ files (in addition to the files from Owens’s previous sentencing trials), discussing trial strategy with them, and preparing them for examination. This case is thus unlike Winston, on which Owens relies and in which counsel failed to discover that the defendant had an IQ of 66 because they altogether “neglect[ed] to review” the school records obtained by their boots on the ground. See id.

We likewise reject Owens’s complaint that counsel failed to discover an array of available mitigating evidence. Though Owens contends that counsel neglected to obtain the social chronology prepared by Drucy Glass (a mitigation investigator who had worked with Hammock) during Owens’s previous sentencing trial, the record indicates the contrary; indeed, Hammock’s testimony about Owens’s family history and time in foster

29

care touched upon many of the details contained in Glass’s chronology and accompanying records. Owens similarly mischaracterizes the record in asserting that counsel failed to investigate the marginal economic conditions in which his family survived and the violence that riddled his neighborhood. To the contrary, counsel elicited testimony from multiple witnesses about both of these circumstances, which permeated their presentation.

The record similarly belies Owens’s contention that Sentencing Counsel failed to look into the horrific conditions of confinement during his period of incarceration at SCDJJ or the sexual abuse he experienced there. Though (as we discuss below) counsel declined to have Dr. Schwartz-Watts present much testimony about Owens’s time in juvenile detention, it’s clear that she investigated the topic by interviewing Owens and reviewing his SCDJJ records, including mental-health records from his treating psychiatrist. Dr. Schwartz-Watts and Godfrey were also both aware of the high-profile class-action lawsuit over the conditions of confinement during Owens’s time there, which resulted in a judgment against SCDJJ. 1 And Godfrey himself went to one of the facilities not only to discuss the lawsuit with the very lawyer who filed it, but also to observe the “horrible” conditions for himself. J.A. 2199.

1 The district court ultimately held that SCDJJ’s facilities were marred by numerous “constitutional and statutory deficiencies” and ordered the state to adopt a remedial plan. Alexander S. ex rel. Bowers v. Boyd, 876 F. Supp. 773, 804 (D.S.C. 1995), as modified on denial of recons. (Feb. 17, 1995); cf. Davis v. Boyd, No. 96-2540, 1997 WL 355626, at *1 (4th Cir. June 27, 1997) (per curiam) (discussing some of the remedial measures that the district court approved on June 28, 1995).

30

As to sexual abuse, Drs. Schwartz-Watts and Brawley alike testified that they “specifically” asked Owens whether he had ever experienced it, whether at SCDJJ or elsewhere, J.A. 2402, and Dr. Schwartz-Watts also looked for evidence of sexual abuse in Owens’s records. Yet Owens “denied” having been sexually abused to each of them, J.A. 2383, and none of his records, including his SCDJJ records, contained evidence of such abuse. Counsel can thus hardly be faulted for not discovering evidence that didn’t come to light until Owens revealed it during his postconviction interview with Dr. Garbarino.

Owens fares no better in complaining that his lawyers failed to consult with appropriate experts. Specifically, Owens suggests that counsel should have retained an expert like Dr. Garbarino to testify about the relation between Owens’s traumatic childhood and his violent behavior. But the Court has never held that counsel is obligated to consult experts, let alone a particular kind of expert, in developing a mitigation case. While the Court has “assumed that in some cases counsel would be deemed ineffective for failing to consult or rely on experts” at all, it has cautioned that “[r]are are the situations” in which counsel’s “wide latitude . . . in making tactical decisions will be limited to any one technique or approach.” Harrington, 562 U.S. at 106 (cleaned up).

Here, the state postconviction court reasonably found that counsel consulted an appropriate array of mental-health experts—at least two of whom, in any event, did testify (to some extent) about the link between Owens’s background and his behavior. Even assuming it would have been “best practices” to consult a child-development expert as well, professional norms didn’t compel that approach. See Winston, 683 F.3d at 504.

31

To the extent Owens complains that counsel failed to present mitigating evidence within their possession, we reject this argument as well. Owens advances two points in this respect. First, he contends that counsel should have presented evidence of the conditions of confinement during his time in SCDJJ custody. But the state postconviction court reasonably concluded that counsel’s decision not to present such evidence was the product of reasonable professional judgment. Godfrey testified that he didn’t introduce evidence about the conditions at SCDJJ for two reasons: because it would have “opened the door” to evidence of Owens’s “numerous” disciplinary violations, J.A. 2201; and because its probative value would have been “difficult . . . to . . . sell a jury on,” J.A. 2199. While others might have judged differently, we think counsel judged with reasonable competence in avoiding such “double-edged” evidence. See Gray, 529 F.3d at 239.

Second, and more generally, Owens contends that counsel should have presented his mitigation story in far greater detail. But regardless of whether such detail would have made for a more compelling narrative, the fact that it would have been “merely cumulative to the evidence actually heard by the jury . . . undercuts [Owens’s] claim for deficient performance.” See Morva, 821 F.3d at 530. The Supreme Court has likewise rejected the notion that counsel must tell a defendant’s life history with elaborative detail, reasoning that where (as here) counsel put forth “substantial mitigation evidence,” any cumulative evidence about the same circumstances heard by the jury offers “an insignificant benefit, if any at all.” See Wong v. Belmontes, 558 U.S. 15, 23 (2009) (per curiam); cf. Bobby, 558 U.S. at 11 (“[T]here comes a point at which evidence from more distant relatives can reasonably be expected to be only cumulative . . . .”).

32

The cases on which Owens relies in this respect are distinguishable because none rested on merely “superficial” or “generalized” presentations. See Pet’r’s Br. 43. Rather, in each of them, our sister circuits emphasized that the jury had “heard nothing” at all about substantial mitigating evidence. Outten v. Kearney, 464 F.3d 401, 421 (3d Cir. 2006); see also Powell v. Collins, 332 F.3d 376, 399–400 (6th Cir. 2003) (counsel “fail[ed] to make even a limited investigation”); Cargle v. Mullin, 317 F.3d 1196, 1221–22 (10th Cir. 2003) (counsel overlooked “significant mitigating information,” including that the defendant had learning difficulties and that his father was abusive); Neal v. Puckett, 286 F.3d 230, 240 (5th Cir. 2002) (en banc) (counsel failed to discover similar “readily available” mitigating facts because they had no “time or money” to do so); Jermyn v. Horn, 266 F.3d 257, 308– 09 (3d Cir. 2001) (counsel “fail[ed] to investigate the circumstances of [the defendant’s] childhood,” including “allegations of childhood abuse”). The same cannot be said here.

Finally, we think the state postconviction court also reasonably concluded that any deficiency in Sentencing Counsel’s investigation or presentation wasn’t prejudicial in light of the total balance of aggravating and mitigating evidence. As the court observed, the State presented overwhelming evidence, not only that Owens satisfied both statutory aggravating circumstances charged, but also of his future dangerousness and inability to adapt to incarceration. Weighed against such aggravating evidence, the additional mitigating evidence that Owens contends counsel should have discovered and presented— nearly all of which, as noted, was either doubled-edged or cumulative—would indeed “have offered an insignificant benefit, if any at all.” See Wong, 558 U.S. at 23. We thus defer to the state court on this ground as well.

33

C.

Owens next contends that Sentencing Counsel were ineffective by failing to object on Confrontation Clause grounds to the trial court’s admission of the State’s record summarizing twenty-eight of his disciplinary infractions while in SCDOC custody. Owens argues that counsel should have so objected because the Supreme Court clearly established his right to confront the authors of his disciplinary record in Crawford v. Washington, 541 U.S. 36 (2004), Davis v. Washington, 547 U.S. 813 (2006), and Melendez-Diaz v. Massachusetts, 557 U.S. 305 (2009). And he asserts that the state court unreasonably applied these cases in concluding to the contrary.

The Confrontation Clause provides that “[i]n all criminal prosecutions, the accused shall enjoy the right . . . to be confronted with the witnesses against him.” U.S. Const. amend. VI. The “ultimate goal” of this safeguard being “to ensure reliability of evidence,” the Confrontation Clause “commands, not that evidence [necessarily] be reliable, but that reliability be assessed in a particular manner: by testing in the crucible of cross- examination.” Crawford, 541 U.S. at 61. Under the Supreme Court’s seminal decision in Crawford, the Confrontation Clause accomplishes this goal by barring the admission of “[t]estimonial” out-of-court statements made by declarants unavailable for trial unless the defendant “had a prior opportunity to cross-examine” the declarant. Id. at 59.

The state court held that Owens was “unable to establish deficient performance” on this claim because the disciplinary record was a “non-testimonial business record[]” that did “not implicate the Confrontation Clause.” J.A. 3700–01. And the court reasoned that the business record was nontestimonial because it wasn’t “prepared in anticipation of

34

producing testimony at [Owens’s] trial, but rather in accordance with South Carolina statutory law for the administration of prison affairs.” J.A. 3701–02 (citing, inter alia, S.C. Code Ann. § 24-21-70). The court also held that any deficiency in counsel’s failure to object wasn’t prejudicial in light of the “overwhelming evidence of Owens’[s] future dangerousness, bad character, and inability to adapt to prison life,” including testimony about many of the same disciplinary infractions. J.A. 3703.

We believe the state court’s decision reasonably applied the standard that a plurality of the Supreme Court articulated in Melendez-Diaz for determining whether a business record implicates the Confrontation Clause. Melendez-Diaz was the first case in which the Court applied Crawford’s novel “testimonial” standard to a document and potential business record. The records at issue there were certain “certificates of analysis,” which the plurality likened to affidavits, “showing the results of [a] forensic analysis” performed on a white substance (which proved to be cocaine) that police had seized from the defendant’s vehicle. See Melendez-Diaz, 557 U.S. at 308 (plurality opinion). In holding that there was “little doubt” these affidavits fell within the “‘core class of testimonial statements’” first outlined in Crawford, id. at 310 (quoting 541 U.S. at 51), the Court emphasized that their “sole purpose” under state law “was to provide prima facie evidence” in the defendant’s criminal trial, id. at 311 (cleaned up); see also United States v. Cabrera- Beltran, 660 F.3d 742, 752 (4th Cir. 2011) (noting that the Court’s holding “relied heavily on the fact that the affidavits at issue were specifically created for trial purposes”).

The limited effect of Melendez-Diaz’s holding with respect to business records is evident where the plurality rejected the defendant’s argument that the affidavits were

35

categorically nontestimonial because they “qualif[ied] as traditional . . . business records” under Fed. R. Evid. 803(6). See 557 U.S. at 321. In this respect, the plurality held that even assuming the affidavits were business records, such records are nonetheless “testimonial” for purposes of the Confrontation Clause “if the regularly conducted business activity is the production of evidence for use at trial.” Id. In other words, while business records “are generally admissible absent confrontation . . . because—having been created for the administration of an entity’s affairs and not for the purpose of establishing or proving some fact at trial—they are not testimonial,” that isn’t so where the business records were “prepared specifically for use at petitioner’s trial.” Id. at 324. 2 In light of that reasoning, we have understood Melendez-Diaz to establish the principle that a business record is “testimonial” only if the record was created primarily for the purpose of “proving some fact at trial.” See Cabrera-Beltran, 660 F.3d 752 (quoting Melendez-Diaz, 557 U.S. at 324). Our sister circuits have tended to agree. See, e.g., United States v. Garcia, 887 F.3d 205, 213 (5th Cir. 2018); United States v. Lorenzo-Lucas, 775 F.3d 1008, 1010 (8th Cir. 2014); United States v. James, 712 F.3d 79, 94–96 (2d Cir. 2013); United States v. Cameron, 699 F.3d 621, 640 (1st Cir. 2012); United States v. Smith, 640

2 Before Crawford, the Court’s prevailing interpretation of the Confrontation Clause exempted extrajudicial statements by unavailable declarants as long as they bore “adequate indicia of reliability.” See Ohio v. Roberts, 448 U.S. 56, 66 (1980) (cleaned up), abrogated by Crawford, 541 U.S. 36. Because adequate indicia of reliability could be established where the statements fell “within a firmly rooted hearsay exception,” id., the right of confrontation didn’t attach to business records under Roberts, cf. Fed. R. Evid. 803(6) (setting forth the traditional business records exception to the rule against hearsay).

36

F.3d 358, 363 (D.C. Cir. 2011); United States v. Yeley-Davis, 632 F.3d 673, 679 (10th Cir. 2011); United States v. Orozco-Acosta, 607 F.3d 1156, 1163 (9th Cir. 2010).

Indeed, even an apparent majority of the Supreme Court—including the late Justice Scalia, who authored the plurality opinion in Melendez-Diaz—has since adhered to the view that business records “are testimonial and require confrontation” only when they were “prepared specifically for use at a criminal trial.” See Michigan v. Bryant, 562 U.S. 344, 392 (2011) (Scalia, J., dissenting); id. at 358, 359 (plurality opinion) (noting that “when a statement is not procured with a primary purpose of creating an out-of-court substitute for trial testimony,” its admissibility “is the concern of state and federal rules of evidence, not the Confrontation Clause”); see also Bullcoming v. New Mexico, 564 U.S. 647, 669–70 (2011) (Sotomayor, J., concurring in part and concurring in the judgment) (“Melendez– Diaz explained that . . . documents kept in the regular course of business may ordinarily be admitted at trial despite their hearsay status, except if the regularly conducted business activity is the production of evidence for use at trial.” (cleaned up)).

In light of this broad consensus, the state court reasonably applied the Supreme Court’s holdings in stating that a business record isn’t testimonial unless it was “prepared in anticipation of producing testimony at trial” rather than for the “administration” of the entity’s “affairs.” J.A. 3701–02. The court also reasonably determined that Owens’s disciplinary record was prepared “for the administration of prison affairs” rather than “in anticipation of producing testimony” in any of his trials. Id.

Specifically, South Carolina law provides that, for any prisoner confined in the state penitentiary, the Department of Corrections “must keep a record of the industry, habits,

37

and deportment of the prisoner, as well as other information requested by the board [of probation] or the director [of the department] and furnish it to them upon request.” S.C. Code Ann. § 24-21-70. As Owens concedes, the record was thus prepared in the ordinary course of the prison’s business, pursuant to its obligation under state law “to furnish upon request records of an inmate’s deportment.” Pet’r’s Br. 63. And while Owens points out that the record could be admitted in a criminal trial, see State v. Whipple, 476 S.E.2d 683, 687–88 (S.C. 1996), that alone doesn’t suffice to trigger the protection of the Confrontation Clause under the Supreme Court’s clearly established law. 3 Owens’s counterarguments are unavailing. Above all, Owens is wrong to assert that Melendez-Diaz clearly established the principle that any declaration “made under circumstances which would lead an objective witness reasonably to believe that the statement would be available for use at a later trial” is testimonial. See 557 U.S. at 310 (quoting Crawford, 541 U.S. at 52). While the plurality did quote that broader “‘formulation[] of th[e] core class of testimonial statements’” from Crawford in passing, id. (quoting 541 U.S. at 51), its holding (to reiterate) rested on the fact that the affidavits were prepared for “the sole purpose” of being used at trial, id. at 311. And while some judges have suggested that the broader formulation should apply to business records, see James, 712 F.3d at 108 (Eaton, J., concurring), such a view is far from clearly established.

3 Because we uphold the state postconviction court’s reasoning under § 2254(d)(1), we needn’t address the parties’ additional dispute over whether the Confrontation Clause applies to capital sentencing trials such as Owens’s, in which the jury is invited to find additional aggravating facts that would support a death sentence.

38

Owens further misses the mark in attributing this broader definition to Crawford itself. To the contrary, Crawford referred to statements “made under circumstances which would lead an objective witness reasonably to believe that the statement would be available for use at a later trial,” 541 U.S. at 52 (cleaned up), as merely one of the “[v]arious formulations of th[e] core class of ‘testimonial’ statements” proposed in that case by the parties and their amici, id. at 51. But Crawford declined to adopt any of those formulations as law, reasoning that the statements at issue “qualif[ied] under any definition.” Id. at 52.

Similarly, in contending that Davis stands for the principle that any statement whose “primary purpose . . . is to establish or prove past events potentially relevant to a later prosecution” is testimonial, 547 U.S. at 822, Owens divorces the Court’s language from its narrow context. Like Crawford, Davis involved the kind of statements that traditionally implicated the Confrontation Clause—those made by an eyewitness to a crime “in response to police interrogation.” See id. Read in this context, the language on which Owens relies serves only to differentiate amongst such statements based on whether or not they were made “to enable police assistance to meet an ongoing emergency.” See id. at 828.

Indeed, because Crawford and Davis both dealt only “with ordinary witnesses,”

neither case clearly established more than the proposition “that formal statements made by a conventional witness—one who has personal knowledge of some aspect of the defendant’s guilt—may not be admitted without the witness appearing at trial to meet the accused face to face.” See Melendez-Diaz, 557 U.S. at 330–31 (Kennedy, J., dissenting). Accordingly, because none of these cases hold that Owens’s disciplinary record, as a

39

business record, implicates the Confrontation Clause, the state court reasonably concluded that Sentencing Counsel’s failure to raise such an objection wasn’t deficient.

We likewise conclude that the state court reasonably determined that any deficiency in counsel’s failure to so object wasn’t prejudicial in light of the overwhelming evidence of Owens’s future dangerousness and inability to adapt to incarceration. Such evidence included testimony not only about Owens’s killing of his fellow inmate on the eve of his first sentencing trial, but also about many of the same infractions summarized in the disciplinary record. In this respect, several of counsel’s own witnesses, especially Dr. Cobb, opened the door to testimony about Owens’s ongoing disciplinary problems upon cross-examination by testifying about his improvements. It’s thus debatable at best that Owens’s disciplinary record prejudiced the outcome of his third sentencing trial.

To sum up, the district court properly concluded that the state postconviction court reasonably applied the Supreme Court’s precedents in denying each of Owens’s exhausted Strickland claims. We turn now to consider his unexhausted Strickland claim.

III.

Owens’s procedurally defaulted claim of ineffective assistance brings us back to the investigation of his mitigation case. Here, Owens requests an evidentiary hearing in the district court on the merits of his claim that Sentencing Counsel was ineffective by failing to obtain a comprehensive neuroimaging evaluation, which revealed evidence of structural and functional brain damage in 2016. Yet because Owens neglected to present this claim in his initial postconviction petition, he must demonstrate cause and prejudice under

40

Martinez and Coleman to obtain a hearing. And despite implying the contrary conclusion when we granted the COA, we agree with the district court that Owens fails to demonstrate cause because his underlying claim is insubstantial. We therefore deny his request.

A.

We start by describing the cause-and-prejudice standard, which reflects the well-

established principle that “[f]ederal habeas courts reviewing the constitutionality of a state prisoner’s conviction and sentence are guided by rules designed to ensure that state-court judgments are accorded the finality and respect necessary to preserve the integrity of legal proceedings within our system of federalism.” See Martinez, 566 U.S. at 9. Chief among such rules is “the doctrine of procedural default, under which a federal court will not review the merits of claims, including constitutional claims, that a state court declined to hear because the prisoner failed to abide by a state procedural rule.” Id. As a result, a state court’s “invocation of a [state] procedural rule to deny a prisoner’s claims precludes federal review of the claims” so long as the state procedural rule is “adequate to support the judgment,” is “firmly established,” and has been “consistently followed.” Id.

Of course, “[t]he doctrine barring procedurally defaulted claims from being heard is not without exceptions.” Id. at 10. As relevant here, in Coleman the Supreme Court held that where “a state prisoner has defaulted his federal claims in state court pursuant to an independent and adequate state procedural rule,” a federal habeas court may nonetheless entertain the claim if “the prisoner can demonstrate cause for the default and actual prejudice as a result of the alleged violation of federal law.” 501 U.S. at 750. The Court also held that a state prisoner may establish cause by establishing that his attorney’s

41

assistance was “constitutionally ineffective under the standard established in Strickland v. Washington.” Id. at 752 (cleaned up). But the Court proceeded to curtail the effect of these rules by further holding that ineffectiveness on the part of a state prisoner’s counsel in a postconviction proceeding doesn’t qualify as cause, reasoning that because the Sixth Amendment doesn’t obligate states to provide counsel beyond direct review, the prisoner “bears the risk” of any errors by postconviction counsel. Id. at 754.

In Martinez, however, the Court found it necessary to “qualif[y]” Coleman’s holding that postconviction counsel’s errors don’t qualify as cause to excuse a procedural default. 566 U.S. at 9. After all, not only do such errors literally cause the default, but there would otherwise be no remedy available to review the prisoner’s claims. See id. at 10–11. Martinez thus announced the following “narrow exception,” id. at 9: “Where, under state law, claims of ineffective assistance of trial counsel must be raised in an initial-review collateral proceeding, a procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if, in the initial-review collateral proceeding, there was no counsel or counsel in that proceeding was ineffective,” id. at 17.

Otherwise stated, a state prisoner may establish cause under Martinez by showing (1) that the defaulted ineffective-assistance-of-trial-counsel claim is “substantial,” (2) that counsel in the initial state collateral-review proceeding was ineffective or absent, and (3) that state law required the ineffective-assistance-of-trial-counsel claim to be raised in the initial collateral-review proceeding as opposed to on direct review. See Trevino v. Thaler, 569 U.S. 413, 423 (2013). A “substantial claim” is one that has “some merit,” a standard that the Martinez Court likened to the one that governs the issuance of COAs under 28

42

U.S.C. § 2253(c)(2). See Martinez, 566 U.S. at 14 (citing Miller-El, 537 U.S. 322); cf. Miller-El, 537 U.S. at 327 (noting that a prisoner must “demonstrate ‘a substantial showing of the denial of a constitutional right’” (quoting 28 U.S.C. § 2253(c)(2)).

A few of our sister circuits have remarked an apparent incongruity lurking in the Martinez standard. To establish that postconviction counsel’s errors “caused prejudice under Strickland,” a state prisoner would have to show that counsel “could have obtained a different result had he presented the now-defaulted ineffective-assistance-of-trial-counsel claim.” Workman v. Superintendent Albion SCI, 915 F.3d 928, 938 (3d Cir. 2019). To do that, however, a state prisoner would have to show that the defaulted claim is itself meritorious. Id. at 938–39; accord Brown v. Brown, 847 F.3d 502, 513 (7th Cir. 2017). In other words, Martinez appears to require a state prisoner to prevail on the merits of his underlying claim merely “to excuse the procedural default” and “obtain consideration on the merits.” Workman, 915 F.3d at 939.

These sister circuits have understandably “rejected that notion.” See id. at 940.

Instead, they have reasoned that a state prisoner satisfies Martinez by showing, first, that initial postconviction counsel performed deficiently, under the first prong of Strickland, by failing to exhaust the underlying ineffective-assistance-of-trial-counsel claim, but not that said counsel’s deficient performance was prejudicial, under the second prong of Strickland; and second, that the underlying claim is substantial, or has some merit, with respect to both prongs of Strickland. Id.; accord Brown, 847 F.3d at 513 We agree with our sister circuits that this rule “is sensible, workable, and a proper reading of Martinez.” Workman, 915 F.3d at 941. Accordingly, to establish cause to

43

excuse the procedural bar to his underlying ineffective-assistance-of-Sentencing-Counsel claim, Owens must show (1) that the underlying claim is substantial and (2) that Initial Postconviction Counsel’s failure to raise it was deficient. 4 We turn now to this inquiry.

B.

With respect to the first of these showings, Owens echoes the petitioner in Brown by contending that we “have already determined that his defaulted ineffective assistance of trial counsel claim is substantial under Martinez” by granting his COA with respect to it. See 847 F.3d at 515. If so, that begs the question of whether we are bound by that prior determination under the “law of the case” doctrine, which “posits that when a court decides upon a rule of law, that decision should continue to govern the same issues in subsequent stages in the same case,” Christianson v. Colt Indus. Operating Corp., 486 U.S. 800, 816 (1988) (cleaned up), including with respect to “an earlier decision of a panel of an appellate court,” United States v. Houser, 804 F.2d 565, 567 (9th Cir. 1986).

Our sister circuit sidestepped this issue in Brown, finding that Martinez didn’t elucidate the nature of its substantiality standard and noting that other circuits have offered “limited further guidance.” See 847 F.3d at 515; cf. Martinez, 566 U.S. at 21 n.2 (Scalia, J., dissenting) (“The Court does not explain where this substantiality standard comes from

4 We recently held that South Carolina law satisfies the remaining element of Martinez by requiring ineffective-assistance claims to be raised on collateral review, Sigmon v. Stirling, 956 F.3d 183, 198 (4th Cir. 2020), and the State doesn’t argue otherwise here. Additionally, because we find that Owens fails to establish cause under Martinez, we needn’t address whether he satisfies the prejudice prong of Coleman. Cf. Martinez, 566 U.S. at 18 (stating that “the question of prejudice” remained open on remand).

44

. . . .”). The court observed that, while Martinez cited to Miller-El in discussing its substantiality standard, it did so with an ambiguous “cf.” signal, and without clarifying how the two standards relate to each other. See Brown, 847 F.3d at 515. So the court opted to “conduct a separate and deeper review of the record, beyond [its] grant of a [COA],” and reaffirm thereby that the underlying claim was “substantial” under Martinez. Id.

The only other circuit that appears to have identified this issue likewise declined to resolve it. See Dansby v. Hobbs, 766 F.3d 809, 840 n.4 (8th Cir. 2014) (assuming the two standards are identical but positing that the denial of certain underlying claims as insubstantial under Martinez “may be construed as the revocation of the COA as to those claims”). Yet while other circuits haven’t squarely addressed it, several have suggested that Martinez incorporated the standard for issuing a COA under § 2253(c)(2) into its definition of substantiality. See Workman, 915 F.3d at 937–38; Flores v. Stephens, 794 F.3d 494, 505 (5th Cir. 2015).

For our part, we think the Supreme Court’s cases with respect to § 2253(c)(1) bear out Owens’s assertion that the Martinez substantiality standard is identical to the one we implicitly resolved in his favor by granting the COA. As the Court “reiterate[d]” in Miller- El, a prisoner satisfies § 2253(c)(1)’s substantiality standard “by demonstrating that jurists of reason could disagree with the district court’s resolution of his constitutional claims.” 537 U.S. at 327 (citing Slack v. McDaniel, 529 U.S. 473, 484 (2000)). And in Slack, the Court held how that standard (which derives from the Court’s pre-AEDPA decision in Barefoot v. Estelle, 463 U.S. 880 (1983)) applies to a claim that the district court “dismissed on procedural grounds.” 529 U.S. at 484. In that instance, a COA may issue only if the

45

prisoner shows “that jurists of reason would find it debatable” both “whether the district court was correct in its procedural ruling” and “whether the petition states a valid claim of the denial of a constitutional right.” Id. In other words, Slack requires a prisoner to demonstrate a “substantial underlying constitutional claim[]” to warrant a COA under § 2253(c)(1). Id. (emphasis added).

That the underlying constitutional claim “is a substantial one” is precisely the showing that Owens must now make (again) to show cause under Martinez. See 566 U.S. at 14. Simply put, both standards turn on whether the claim underlying the procedural default is “substantial.” And while Martinez phrases that inquiry in terms of the claim’s having “some merit,” see id., whereas Slack phrases the inquiry in terms of “reasonable jurists” being able to debate the claim’s validity, see 529 U.S. at 484, we see little daylight between those formulations. If reasonable jurists could debate the merits of the underlying claim, then it must have (at least) some merit.

That said, we don’t think this apparent redundancy between the Martinez and § 2253(c)(1) substantiality standards precludes us from reconsidering the merit of Owens’s underlying claim at this stage of the case. As the Supreme Court has recognized, the law of the case doctrine “merely expresses the practice of courts generally to refuse to reopen what has been decided, not a limit to their power.” Messenger v. Anderson, 225 U.S. 436, 444 (1912). The doctrine is therefore “discretionary” rather than “mandatory,” and admits of a variety of exceptions. See CNF Constructors, Inc. v. Donohoe Const. Co., 57 F.3d 395, 398 n.1 (4th Cir. 1995); accord Houser, 804 F.2d at 567. In the context of an earlier decision of a panel of an appellate court, our sister circuits have acknowledged the ability

46

to reconsider a ruling “on the same issue presented in the same action if a showing is made which compels us to reconsider our prior decisions.” See Houser, 804 F.2d at 568. We too have done so in the context of “a prior ruling of a motion panel,” which is essentially what we have here. See CNF Constructors, 57 F.3d at 398 n.1.

In the Martinez context, we think ourselves warranted in reconsidering the substantiality of the underlying claim for three reasons. First and foremost, the substantiality standard implicates our very “jurisdiction to consider [Owens’s] appeal,” an issue that circuit courts have tended to view as immune from the law of the case doctrine in light of our “duty . . . to dismiss whenever it becomes apparent that we lack jurisdiction.” See Houser, 804 F.3d at 568–69 (cleaned up); accord CNF Constructors, 57 F.3d at 397 n.1; Johnson v. Burken, 930 F.2d 1202, 1205 (7th Cir. 1991); EEOC v. Neches Butane Prods. Co., 704 F.2d 144, 147 (5th Cir. 1983); Green v. Dep’t of Commerce, 618 F.2d 836, 839 n.9 (D.C. Cir. 1980). True, the standard is jurisdictional in the technical sense only with respect to the Slack iteration, since § 2253(c)(1) provides that “an appeal may not be taken” from a final order dismissing a § 2254 petition “[u]nless a circuit justice or judge issues a certificate of appealability.” 28 U.S.C. § 2253(c)(1). Yet the Martinez iteration might be viewed similarly as “quasi-jurisdictional” insofar as no federal court may consider the merits of a defaulted claim until the procedural bar is excused. See Martinez, 566 U.S. at 9–10; cf. Champagne v. Schlesinger, 506 F.2d 979, 982 (7th Cir. 1974) (noting that “exhaustion is a quasi-jurisdictional problem”). And whether framed as rescinding the COA (as suggested by our sister circuit in Dansby) or simply denying the claim under Martinez, our ability to reconsider the issue leads to the same result.

47

Second, as is often the case with the ruling of a “preliminary” panel, “practical realities” required us to rule on the COA “without the benefit of full briefing and oral argument,” which prevented the relevant arguments from being “fully present[ed]” to us. See CNF Constructors, 57 F.3d at 397 n.1. Indeed, we granted the COA based solely on Owens’s opening brief, without the benefit of any briefing by the State (not to mention oral argument). See Local R. App. P. 22. And reflecting the “provisional” nature of that ruling, our order was not only unpublished but also “without opinion,” which furthers weighs against its tying our hands under Martinez. See EEOC, 704 F.2d at 147. Thus, while a merits panel “does not lightly” revisit an earlier panel decision “during the course of the same appeal,” we have more leeway to do so here, where the issue didn’t receive a “[f]ull review” until after the COA had issued. See Houser, 804 F.2d at 568.

And third, the very fact that Martinez directs us to address an issue we already confronted when deciding whether to grant a COA leaves us no doubt that we may reconsider the substantiality of the underlying claim. By citing Miller-El in discussing its cause standard, Martinez demonstrated (at the very least) the Court’s awareness that a prisoner must first, before any argument as to procedural default, make the requisite showing to obtain a COA. See 566 U.S. at 14. Yet Martinez nowhere suggests that cause is predetermined as soon as a COA is granted. Indeed, because such an effect would render the first step of the Martinez standard superfluous, the law of the case doctrine seems antithetical to its operation. We therefore view our duty under Martinez to consider the substantiality of the underlying claim, even after granting a COA, as affording us another (and fuller) opportunity to do so.

48

Accordingly, we turn now to reconsider—without regard to our grant of a COA—

whether Owens’s underlying claim is substantial under Martinez. 5 C.

Upon reconsideration, we are satisfied that the district court properly found that Owens’s underlying claim—that Sentencing Counsel rendered ineffective assistance by failing to have him neuroimaged for evidence of structural and functional brain damage— is insubstantial. See Owens, 2018 WL 2410641, at *33–35.

As noted with respect to Owens’s very first claim, Sentencing Counsel’s obligation to conduct a thorough investigation of his mitigating evidence certainly extended to evidence that could “cast light on [his] mental condition.” See Rompilla, 545 U.S. at 382; see also Wiggins, 539 U.S. at 534–35 (counsel “failed to discover and present” evidence of Wiggins’s “diminished mental capacities”); Williams, 529 U.S. at 396 (counsel “failed to introduce available evidence that Williams was borderline mentally retarded” (cleaned up)). But the issue again turns on how thorough the investigation had to be under prevailing professional norms. And here, while counsel may have rendered ineffective assistance had they failed “to investigate [Owens’s] mental condition as a mitigating factor” altogether, see Caro v. Woodford, 280 F.3d 1247, 1254 (9th Cir. 2002) (cleaned up), or failed “to provide mental health experts with information needed to develop an accurate profile of

5 Our holding in this respect is consistent with our decisions in two recent cases, where we found (without being called to address the issue resolved here) that an underlying Strickland claim was insubstantial under Martinez despite having granted a COA. See Sigmon, 956 F.3d at 193, 199; Moore v. Stirling, 952 F.3d 174, 181, 185–86 (4th Cir. 2020).

49

[his] mental health,” see id., or even overlooked “red flags pointing up a need to test further,” see Rompilla, 545 U.S. at 392 (cleaned up), on this record, there is no merit to the claim that they were ineffective by not obtaining a comprehensive neuroimaging evaluation.

Our analysis of Owens’s first claim illustrates that counsel thoroughly investigated his mental condition as a mitigating factor; indeed, along with his impoverished and violence-ridden upbringing, it was the predominant factor in their mitigation case. Recall that counsel began their representation with the benefit of two previous trials’ worth of mitigating evidence, including myriad mental health records and expert opinions. They then enlisted their own team of experts—including a neuropsychologist, forensic psychiatrist, and clinical social worker—to take that information, find more of it, and independently assess Owens’s mental condition. These experts, who were well aware of the indicia of mental impairment, probed the mental health histories of Owens and his family members, consulted with treating doctors and other medical experts, and conducted their own suite of evaluations. And finally, counsel presented the opinions of their mental health experts, who testified collectively to the effect that, while Owens suffered from a variety of disorders and deficiencies that made him more prone to engage in violent crime, he didn’t suffer from brain damage or mental illness.

Owens fails to show that reasonable jurists could debate whether such a thorough investigation of his mental condition fell short of an objective standard of reasonableness. Not only did counsel take extensive measures to investigate Owens’s behavioral cognition for mitigating evidence, their mental health experts reached conclusions that belied the

50

need for comprehensive neuroimaging—which no expert suggested during Owens’s case until his federal habeas proceedings.

And much as Owens might disagree with those conclusions in retrospect, he provides no reason to believe that they were professionally unreasonable. Thus, because Strickland “does not require counsel to investigate every conceivable line of mitigating evidence,” Wiggins, 539 U.S. at 533, the fact that Sentencing Counsel didn’t see fit to pursue a comprehensive neurobehavioral assessment doesn’t amount to “incompetence under prevailing professional norms,” Winston, 683 F.3d at 504 (cleaned up).

Relatedly, Owens falls short of establishing that counsel missed any red flags pointing to the need to obtain neuroimaging or otherwise further investigate for evidence of structural and functional brain damage. Owens’s reliance on his prescription for Depakote in this respect—the only red flag he identifies on appeal—is misplaced.

Specifically, Owens relies on Dr. Wood’s opinion that neuroimaging “should have at the very least been considered,” J.A. 4203, due to the “history of seizure disorders” indicated by Owens’s “treatment with Depakote,” J.A. 4202. He also points to Dr. Brawley’s affidavit swearing that, had she “been given information regarding previous seizure activity” while preparing for trial, this “would have cause[d] [her] to recommend a full neurological evaluation.” J.A. 4231. Yet the record belies the premise that Owens’s treatment with Depakote at Lieber was indicative of seizure activity. To the contrary, as Dr. Cobb testified himself in Owens’s third sentencing trial, he had prescribed Depakote (in combination with Risperdal) only to help stabilize Owens’s moods and “slow [his] brain down.” J.A. 1594–95. Dr. Cobb also testified that Depakote is commonly prescribed to

51

treat mood disorders like bipolarism, for which Dr. Schwartz-Watts herself discovered that Owens had been receiving treatment at Lieber, and thus not only seizure disorders. In light of these explanations, we discern no indication that Sentencing Counsel overlooked evidence of seizure activity.

Finally, we agree with the district court that Sentencing Counsel’s decision not to call Dr. Evans as a witness also bears on the analysis. As the neuropsychology expert in the previous sentencing trial, Dr. Evans testified that Owens had mild brain dysfunction in his frontal lobe. That testimony generally echoes the results of Dr. Gur’s comprehensive neuroimaging assessment. And the technique that Dr. Evans used to reach his results (qEEG) bears similarity to the techniques used by Dr. Gur (PET and MRI); all three may even be considered subsets of “neuroimaging” technologies. See Laura Stevens Khoshbin & Shahram Khoshbin, Imaging the Mind, Minding the Image: An Historical Introduction to Brain Imaging and the Law, 33 Am. J.L. & Med. 171, 176–77 (2007).

Sentencing Counsel thus already had “similar evidence” at their fingertips to that which Owens contends they should have developed through neuroimaging. See Owens, 2018 WL 2410641, at *35. But when presented with the claim (which Owens didn’t raise in his § 2254 petition) that counsel was ineffective by declining to present Dr. Evans’s testimony, the state court properly denied it on the ground that counsel’s decision to go with Dr. Brawley’s testimony reflected a professionally reasonable strategy to take a more conservative approach to the neurological mitigating evidence.

As Godfrey testified during Owens’s initial postconviction proceeding, he was concerned that avant-garde diagnostic techniques wouldn’t play in front of a local

52

Greenville jury. He was also aware that any evidence of brain damage would be double- edged and might well do more harm than good for Owens’s mitigation case, because it would bespeak his inability to become less violent. Since this “sound trial strategy” would apply with comparable force to similar evidence based on Dr. Gur’s neuroimaging evaluation, see Strickland, 466 U.S. at 689 (quoting Michel v. Louisiana, 350 U.S. 91, 101 (1955)), it presents an additional reason why Owens’s underlying claim is insubstantial. 6

* * *

For the foregoing reasons, we hold that the state postconviction court reasonably denied Owens’s exhausted Strickland claims on the grounds that capital sentencing counsel thoroughly investigated and presented his available evidence in mitigation, and didn’t neglect a viable Confrontation Clause objection to his disciplinary record. We also hold that the district court properly denied Owens’s defaulted Strickland claim as insubstantial under Martinez. Therefore, the judgment of the district court is AFFIRMED.

6 Because we conclude that Sentencing Counsel didn’t perform deficiently in failing to obtain comprehensive neuroimaging, we needn’t address whether Owens’s underlying claim is substantial with respect to the prejudice prong of Strickland. We also needn’t address whether Initial Postconviction Counsel performed deficiently in failing to exhaust Owens’s underlying claim.