Mark Allen Jenkins v. Commissioner, Alabama Department of Corrections.

Opinion

[PUBLISH]

IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT

No. 17-12524

D.C. Docket No. 4:08-cv-00869-VEH MARK ALLEN JENKINS, Petitioner - Appellant,

versus

COMMISSIONER, ALABAMA DEPARTMENT OF CORRECTIONS, Respondent - Appellee.

Appeal from the United States District Court for the Northern District of Alabama

(June 29, 2020)

Before WILSON, BRANCH, and TJOFLAT, Circuit Judges. BRANCH, Circuit Judge:

We sua sponte vacate our previous opinion and substitute the following in lieu thereof. 1

* * *

Mark Allen Jenkins, an Alabama prisoner sentenced to death for the 1989 murder of Tammy Ruth Hogeland, appeals the district court’s denial of his petition for a writ of habeas corpus, 28 U.S.C. § 2254. Before us are Jenkins’s arguments that he received ineffective assistance of counsel during the penalty phase of his trial and that he is intellectually disabled and therefore ineligible for the death penalty. After careful consideration, and with the benefit of oral argument, we affirm the denial of Jenkins’s habeas petition on both grounds.

I. BACKGROUND

A. The Crime and Arrest We summarize the following background narrative from opinions of the Alabama Court of Criminal Appeals and our own review of the record. See Jenkins v. State, 627 So. 2d 1034, 1037–40 (Ala. Ct. Crim. App. 1992); Jenkins v. State, 972 So. 2d 111, 119–20 (Ala. Ct. Crim. App. 2004). The events leading up to the murder began the evening of April 17, 1989. Jenkins was at the home of his

1 We note that appellant has a pending petition for rehearing en banc along with a supporting amicus brief. We will still consider appellant’s petition in light of this substituted opinion, and therefore he need not refile. However, appellant may file a ten-page supplemental petition based on the new opinion if desired within the standard time frame for filing rehearing petitions.

3

acquaintance Christine Nicholas. He had met her several months earlier at the Omelet Shoppe restaurant where she worked. Jenkins was very intoxicated and tried to seduce her. When she resisted, Jenkins got “real mad” and asked her several times what she would do if someone came up from behind her and grabbed her.

At around 1 a.m., Jenkins and Nicholas drove to the Riverchase Omelet Shoppe. Jenkins went inside and talked with one of the waitresses, Frieda Vines. The manager, Douglas Thrash, recognized Jenkins as a regular customer who knew all of the waitresses. Thrash overheard part of Jenkins’s conversation with Vines and heard mention of the Omelet Shoppe location near the Birmingham Airport. Jenkins and Nicholas then returned to Nicholas’s home. Around 2 a.m., Nicholas’s mother asked Jenkins to leave. He did so, falling down some steps and ramming his car into another vehicle in the process.

Meanwhile, 23-year-old Tammy Hogeland was making her way to work at the Omelet Shoppe. Her sister Wendy, along with Tammy’s young son, picked Tammy up from her college classes and drove her to the Riverchase Omelet Shoppe a little after 9:30 p.m. Sometime after 10:00, Tammy was sent to work at the airport Omelet Shoppe on Tenth Avenue, since that location was unexpectedly shorthanded. Tammy was wearing her watch, a necklace with the words “special sister,” her class ring, and her diamond engagement ring. Tammy was working as a

4

cook that night and was wearing a blue apron, black pants, a white shirt, and white shoes. She and Sarah Harris were the only employees working at the airport Omelet Shoppe that night.

At about 2 a.m. on April 18, Harris saw Jenkins—whom she did not know—

drive up in a red sports car. His arrival was memorable to Harris because the car nearly jumped the curb and crashed through the restaurant’s glass wall. Jenkins got out of the car and came into the restaurant, appearing to be intoxicated. He walked over to Hogeland and began talking to her. Harris later saw Jenkins and Hogeland drive away together in the red sports car. That was the last time anyone who knew Hogeland ever saw her. Hogeland left behind her cigarettes, lighter, purse, and paycheck. She left without telling anyone, which she had never done before. Although Harris saw them drive off, she could not later say whether Hogeland left with Jenkins willingly or was instead abducted.

At around 5 a.m. that morning, Bobby and Geraldine Coe had stopped to buy gas at a Chevron station on I-59 northeast of Birmingham when they saw Jenkins drive up in the red sports car. The Coes noticed a female who appeared to be “passed out” in the front passenger seat, but they could not say whether she was alive or dead. While Bobby Coe was pumping gas, Jenkins asked him for some cigarettes and said, “Looks like it’s been a long night and it looks like it’s going to be a long day.” Jenkins then said “God bless you” before asking directions to I-

5

459. Coe gave him directions and got back in his car. As Coe pulled out onto the interstate, he saw Jenkins follow him in the red sports car. Coe then saw the car flash its lights, slow down, and pull to the side of the road between mile markers 151 and 152.

At 8 a.m., Wendy Hogeland learned that Tammy was not at the restaurant, and their mother called the police. Meanwhile, Jenkins went to the home of Steve Musser, who noticed that he was wearing the same clothes as the day before. Jenkins told Musser that his car had been stolen the night before and asked him if he would say that he had been with him all night. Musser refused. Christine Nicholas also saw Jenkins in a grocery store that morning. He was looking at a newspaper, making a phone call, and attempting to sell his Buick. Nicholas loaned him $4 for gas. At around 10 o’clock that morning, Jenkins sold his car to Michael Brooks, a mechanic at a local Chevron station. Jenkins had explained that he needed the money so he could visit his sick mother in California. Another mechanic at the service station drove Jenkins to the Greyhound bus station later that morning.

Jenkins awoke the next day on the Greyhound bus in Houston and was ejected from the bus because his fare was used up. He then hitchhiked from Houston to Phoenix, then to San Diego and Los Angeles. Jenkins was first identified as a suspect in Hogeland’s disappearance by Officer Mike Weems of the

6

Hoover Police Department on April 19. Weems ate dinner at the Riverchase Omelet Shoppe just about every day and had talked with Hogeland just as often. He also knew Jenkins from the restaurant. Weems learned of Hogeland’s disappearance from the other waitresses. Remembering how Jenkins often talked to Hogeland and passed her notes, 2 he gave Jenkins’s name to the missing persons investigator. The investigator also learned that a red Mazda RX-7 sports car, which had been reported stolen from the service station where Jenkins worked, had been recovered on I-459.

The afternoon of April 21, a truck driver who had happened to stop on I-59 near mile marker 151 discovered Hogeland’s body off the side of the road. She was naked, wearing only a watch, and was so badly decomposed that the body had to be identified by dental records. From the fractured hyoid bone, it was determined that she was manually strangled to death. Also found at the scene were Hogeland’s apron, shoes, bra, panties, pants, and hair net, as well as some beer cans, a Mazda RX-7 owner’s manual, and other items later determined to have come from the red Mazda. Her necklace and rings were never found.

Alabama authorities issued a warrant for Jenkins’s arrest and also obtained a federal fugitive warrant. An FBI agent arrested Jenkins on May 1 in front of his

2 The jury was not permitted to hear, and the record does not reveal, what exactly Jenkins had said to Hogeland that made Officer Weems suspect him in her disappearance.

7

uncle’s house in Wilmington, California. Jenkins’s uncle later gave the police three bags of Jenkins’s personal effects. Fibers recovered from Hogeland’s body and clothing matched those of the Mazda, and hairs matching Hogeland’s were found in the Mazda. Similarly, fibers from Jenkins’s clothing placed him in the Mazda, and fibers of his jeans were found on Hogeland’s apron. A bootprint near Hogeland’s body matched a combat boot from among the belongings Jenkins’s uncle gave to the police, and Jenkins’s roommate said the boots looked just like those that Jenkins wore.

A business card like those that had been in the Mazda’s glove box, belonging to an owner of the Mazda, was recovered from Jenkins’s wallet after his arrest in California. Later, Jenkins’s cellmate in the Alabama jail reported that Jenkins told him “he had done the crime” and was worried that the couple from the gas station would identify him or that the police would find his fingerprint on a beer can at the scene. A St. Clair County grand jury indicted Jenkins for capital murder.

B. The Trial and Sentencing Jenkins was represented at trial in 1991 by attorneys Douglas Scofield and Stan Downey. Scofield, the lead attorney, had been practicing criminal defense in Birmingham since 1984 but had never before tried a capital murder case. Jenkins’s landlord and his grandmother had arranged the referral and representation before

8

the court appointed him. Scofield agreed to the appointment on the condition that a St. Clair County lawyer also be appointed. That lawyer was Downey, who had more experience than Scofield but whose capital experience was limited to one case that had not gone to trial. The two attorneys had agreed that Downey would be responsible for jury selection, but Scofield eventually took over during the voir dire. Per their agreement, Scofield was responsible for the guilt-phase trial, although Downey participated by interviewing Jenkins several times. Scofield also met with Jenkins around a dozen times. Jenkins had told him about his background and childhood, including the fact that he had had a difficult childhood and had lived on the streets since age 11.

Throughout the weeklong trial, Scofield pursued an outright acquittal. In particular, he challenged Harris’s identification of Jenkins as the man who drove away with Hogeland, and he challenged the medical examiner’s opinions about the time and cause of Hogeland’s death. In his closing argument, Scofield took a wide- ranging approach to undermining the State’s wholly circumstantial case against Jenkins. He attacked the State’s timeline of the crime, questioning whether Jenkins in his intoxicated state would have had the time and the ability to leave Nicholas’s house, steal the Mazda, and abduct and kill Hogeland. He reminded the jury of the weakness of Harris’s identification of Jenkins and the Coes’ identification of the woman in the car. He pointed out that no physical evidence definitively linked

9

Jenkins to the Mazda, to Hogeland’s body, or to her jewelry, which was never recovered. And he argued that the evidence provided no plausible motive for Jenkins to have kidnapped, robbed, or killed Hogeland. In sum, he urged the jury to find reasonable doubt and acquit Jenkins. At the very least, he argued, they should find that the State failed to prove Jenkins’s intent to rob and kidnap Hogeland, which were necessary to convict Jenkins of capital murder. He further argued that, if the jury thought Jenkins killed Hogeland as a crime of passion, they could convict him only of manslaughter. The jury deliberated for three hours and fifteen minutes before convicting Jenkins of capital murder, Ala. Code § 13A-5- 40(a)(1) & (2), i.e., murder committed in the course of kidnapping and robbery.

The penalty phase took place that same afternoon. A week before trial, Scofield and Downey had discussed the penalty phase, and Scofield understood that Downey would be responsible for handling the penalty phase. Nonetheless, the witness who was present to testify on Jenkins’s behalf during the penalty phase was better acquainted with Scofield, so Downey suggested that afternoon that Scofield conduct the direct examination. Scofield urged Downey to interview the witness quickly and prepare to examine him, and Downey did so.

Downey handled the penalty presentation to the jury. The State did not present any additional aggravation evidence, resting on the evidence presented during the guilt phase. One witness, Lonnie Seal, testified on Jenkins’s behalf in

10

order, in Downey’s words, to “reveal another side of Mark Jenkins to you that you don’t know anything about.” Seal had met Jenkins in Fontana, California, in 1988 when they both worked at the same garage. Seal and his wife became friends with Jenkins, and he visited in their home. When the Seals decided to move to Alabama a few months later, Jenkins agreed to help with their move and to live with the Seals in Alabama for a time. It took Lonnie Seal several weeks to find a job, but Jenkins found a job in two days and contributed toward the family’s rent and groceries. Jenkins continued to visit the Seals after he moved into his own home, and he would always bring a small gift for and play with the Seals’ infant son.

Downey’s closing argument to the jury requested a sentence of life without parole. Downey reminded the jurors that they had agreed during voir dire that death was not always the appropriate punishment for homicide. Quoting the Bible,3 he reminded the jury that the Old Testament permitted capital punishment only when supported by the testimony of two or three eyewitnesses, whereas in this case there had been only circumstantial evidence. He asserted from Seal’s testimony that “Mark had a side of him that was generous, that was giving, and that was

3 “If anyone kills a person, the murderer shall be put to death on the evidence of witnesses. But no person shall be put to death on the testimony of one witness.” Numbers 35:30 (ESV). Downey also paraphrased Deuteronomy 17:6 as “On the evidence of two witnesses or three witnesses, he is [sic] shall be put to death. He shall not be put to death on the evidence of one witness.”

11

kind,” and argued that the deterrent and retributive purposes of capital punishment would not be served in this case.

The jury deliberated for 50 minutes before returning its nonbinding recommendation of death by a vote of 10 to 2. See Ala. Code § 13A-5-46(e) (1981) (amended 2017 to make jury’s verdict binding). After a separate hearing, the court found the two statutory aggravating factors for which the State had argued: that the murder was committed during the commission of a robbery, and that it was committed during the commission of a kidnapping. See Ala. Code § 13A-5-49(4). The court also found two statutory mitigating factors: that Jenkins had no significant history of prior criminal activity, and his age at the time of the crime (21). See id. § 13A-5-51(1), (7). The court considered but rejected the statutory mitigation that, due to his consumption of alcohol, Jenkins’s capacity to appreciate the criminality of his conduct or to conform his conduct to the requirements of law was substantially impaired. See id. § 13A-5-51(6). The court found that the aggravating circumstances outweighed the mitigating circumstances and, in accordance with the jury’s recommendation, imposed a sentence of death.

C. The State and Federal Post-Conviction Proceedings On direct appeal, the Alabama Court of Criminal Appeals affirmed Jenkins’s conviction and death sentence. Jenkins v. State, 627 So. 2d 1034 (Ala. Ct. Crim. App. 1992), aff’d, 627 So. 2d 1054 (Ala. 1993), cert. denied, 511 U.S. 1012

12

(1994). In 1995, Jenkins filed his first petition for state postconviction relief, Ala. R. Crim. P. 32, 4 alleging, among many other things, that he received ineffective assistance of counsel during his penalty phase. The Rule 32 trial court conducted a three-day hearing at which evidence was presented about the performance of Jenkins’s attorneys.

Scofield testified about the representation that he and Downey provided. In relevant part, Scofield asserted that he and Downey had “decided that [developing mitigation for the penalty phase] would be something that Stan [Downey] would handle.” Downey later told Scofield that he hadn’t done any such investigation, but Scofield admitted that, “[a]part from what he has told me, I don’t know what he has done.” Scofield testified that Downey “didn’t appear prepared” for the penalty phase, based on Downey’s last-minute suggestion that Scofield examine Seal since he had talked with him before.

In preparation for the penalty phase, Scofield himself did not interview any member of Jenkins’s family, but he did not know whether Downey interviewed any family members. Scofield did interview Jenkins’s landlord and some other acquaintances but did not do so specifically for the purpose of developing

4 “Subject to the limitations of Rule 32.2, any defendant who has been convicted of a criminal offense may institute a proceeding in the court of original conviction to secure appropriate relief on the ground that . . . [t]he constitution of the United States or of the State of Alabama requires a new trial, a new sentence proceeding, or other relief. . . .” Ala. R. Crim. P. 32.1.

13

mitigation. He had met with Jenkins around a dozen times, and Jenkins had told him his life story including his experience of abuse as a child. In hindsight, Scofield said he had a “lack of appreciation of mitigating circumstances.” He repeatedly denied in his Rule 32 testimony that his omissions during the penalty phase were strategic: “I didn’t understand what I was doing. . . . I just didn’t understand what I needed to do.”

For reasons that remain unclear, Downey—who is since deceased—did not testify or submit an affidavit in the Rule 32 proceeding. Scofield answered, “Obviously not,” when asked whether he had personal knowledge of everything Downey may or may not have done to prepare for the penalty phase. The record of the extent of Downey’s pretrial preparation is thus limited to Downey’s detailed fee declaration, which he submitted to the State of Alabama in documentation of the compensation due him as appointed counsel. See Ala. Code § 15-12-21. In it, Downey reported that he spent 71 hours 5 on out-of-court preparation for trial. The report details 35 hours of interviews with Jenkins in 18 separate visits to the jail over the course of 17 months, as well as one substantive conversation with Jenkins’s grandmother. 6

5 In Jenkins’s state postconviction appeal, the Alabama Court of Criminal Appeals erroneously stated that Downey’s fee declaration reports 171 hours of pretrial preparation. See Jenkins, 972 So. 2d at 137. We discuss the effect of this discrepancy infra note 7. 6 The state appellate court also noted that Downey had moved for a continuance in October 1989 so that he could conduct “[f]urther discovery and investigation (including a possible trip to

14

The Rule 32 trial court also heard testimony from four family members and one friend of Jenkins’s. The family members presented for the first time a stark depiction of Jenkins as an unloved and abused child in California. Jenkins’s older brother Michael Jenkins testified that all of his siblings have different fathers and that his parents often used drugs and fought. Michael testified that Jenkins was beaten frequently as a child by his stepfather, if not every day then every other day or every three days. Jenkins was beaten all over his body with a variety of implements. Michael was beaten too. Jenkins’s stepfather would lock Jenkins in his bedroom for a couple of hours at a time, sometimes without dinner. Some of the beatings were prompted by Jenkins’s bedwetting or soiling. Michael testified that Jenkins’s stepfather sometimes made Jenkins wear his soiled underwear on his head and even forced him to eat his own feces a few times. Michael testified that Jenkins’s bladder and bowel control problems began after a camping trip he took with his stepfather and stepgrandfather.

Michael testified that, as children, he and Jenkins were made to work on car engines late into the night and to shovel horse manure daily. The family moved frequently and the children attended many different schools as a result. Michael testified that he was Jenkins’s only friend as a child and that Jenkins did poorly in

California).” Id. We note that Scofield had orally suggested to the trial court in September 1989 that more time might be needed with respect to “evidence about the defendant” and witnesses in California, as well as possible psychological evaluation.

15

school. Jenkins began running away from home as a young teenager and lived with various other family members or on the streets. Michael said that Jenkins was a good brother to him and had asked him to testify at his trial, but he never heard from his attorneys or found out when the trial was. Michael admitted to having cognitive and memory problems due to a head injury years earlier.

Jenkins’s cousin Tammy Pitts testified that, as an infant, Jenkins’s mother beat and neglected him, leaving him dirty and in soiled diapers. His stepfather also beat Jenkins daily until he left home at age 13, sometimes bruising him so badly that he was laid up in bed for days or weeks. Jenkins had problems with bedwetting and soiling, and his stepfather made him wear soiled clothes to school. Jenkins would be locked in his filthy bedroom around the clock except to attend school and do chores. He was usually not allowed to eat dinner with the rest of the family and subsisted on scraps and dog food. The entire family belittled Jenkins because his biological father was Hispanic and his complexion was therefore darker. Jenkins’s parents drank a lot and used drugs every day. Pitts knew these things because she had daily contact with the family and lived with them off and on, and Jenkins lived with her for a time after he ran away from home. No one had contacted her about testifying at Jenkins’s trial.

Jenkins’s second cousin Betty DeLavega testified that her relationship with Jenkins was like that of an aunt. She lived with the Jenkins family for five months

16

when Jenkins was 10 or 11. She saw Jenkins’s stepfather slap Jenkins “[q]uite a few times,” and once, he made Jenkins eat his own feces. DeLavega testified that Jenkins’s parents used drugs nearly every day and withheld affection from Jenkins and Michael. No one contacted her about Jenkins’s trial; Jenkins’s grandmother had asked her to testify in the Rule 32 hearing about any abuse she had witnessed.

Jenkins’s grandmother Doris Wagoner testified that while Jenkins’s mother was pregnant with him, her husband was in prison and she drank and took drugs. When Jenkins was born, his mother initially put him up for adoption, then changed her mind weeks later. Jenkins was a lethargic infant and his mother cared more about partying than about her children. Wagoner testified that she heard of Jenkins’s mistreatment as a child but didn’t see it. She could tell from the way that Jenkins feared his stepfather that he was being abused, and she knew he slept in a filthy bed. After Jenkins was arrested for murder, Wagoner spoke with Scofield, who asked for money relating to his representation of Jenkins. She did not remember whether she talked with Scofield about Jenkins’s background. She could not remember why she didn’t testify at the trial; she might have been ill, but she was probably busy.

Sherry Seal, the wife of Lonnie Seal, who had testified in the penalty phase of the trial, testified that Jenkins was respectful and generous and that she trusted him with her children. She related how he supported her family after moving to

17

Alabama, giving them “pretty much” his entire paycheck. She said she would still feel safe around Jenkins. Two jailers also testified that Jenkins was a respectful, courteous, model inmate who never complained or caused any problems.

Jenkins also introduced medical, school, and juvenile court records from his childhood. The medical records document his premature birth and his mother’s intention to give him up for adoption. The school records reflect a struggling student who was repeatedly promoted to the next grade because of his age, despite serious academic difficulties, chronic truancy, and frequent school changes. The juvenile court records document Jenkins’s placement in juvenile hall starting at age 14, following repeated acts of grand theft auto and running away.

The Rule 32 hearing concluded with the testimony of two psychologists about Jenkins’s history of psychological trauma stemming from his childhood abuse. The defense expert, Dr. David Lisak, had met with Jenkins three times and interviewed a dozen of his family members, family friends, and associates. Dr. Lisak testified that the pervasive abuse and neglect in Jenkins’s childhood impaired his ability to cope with trauma as an adult. He opined that Jenkins’s childhood abuse adversely affected his cognitive and emotional development and other abilities. He described how Jenkins had been neglected as an infant and sexually abused by his grandfather at age four, which led to a lifetime of bladder and bowel dysfunction. He also discussed the hypothesis that children who are abused are at

18

greater risk for perpetrating violence later in life, but he acknowledged there would have been other causative factors in Jenkins’s case. Dr. Lisak described Jenkins’s intellectual capacity as “somewhat borderline,” though he did not perform any testing of his own.

The State’s expert, Dr. Karl Kirkland, had met with Jenkins for several hours and administered several tests. On the Wechsler Adult Intelligence Scale, Jenkins “scored in the range of borderline intellectual functioning which is between mild mental retardation and low average intellectual functioning” with an overall IQ of 76. Dr. Kirkland agreed that Jenkins had been “reared in a chaotic dysfunctional family” and generally agreed with Dr. Lisak’s assessment of the negative effects of childhood trauma.

The court considered all of this evidence before denying Rule 32 relief in full. As relevant here, it found that some of the evidence about the severity of abuse that Jenkins suffered as a child was not credible. It further found that Jenkins suffered no prejudice from any deficient performance by his penalty-phase counsel because the aggravating circumstances outweighed any additional mitigating evidence Downey might have adduced.

The Alabama Court of Criminal Appeals affirmed the denial of Rule 32 relief. Jenkins v. State, 972 So. 2d 111 (Ala. Ct. Crim. App. 2004). In relevant part, the appeals court rejected Jenkins’s ineffective assistance claim, finding that

19

counsel reasonably chose to pursue a penalty-phase strategy of residual doubt and good character. Id. at 147. It explained that the new evidence was self- contradictory in that it asserted that childhood abuse would have made Jenkins a violent adult yet also asserted that Jenkins was meek and mild. Id. The appeals court also agreed with the lower court’s conclusion that Jenkins suffered no prejudice from any error by his counsel because the aggravating circumstances still outweighed any new mitigating ones. Id. at 148. The court also noted that evidence about Jenkins’s good behavior in jail awaiting trial was “minimally mitigating” and would not have affected the sentence. Id. at 149.

Because the Supreme Court’s decision in Atkins v. Virginia7 had issued during the pendency of the Rule 32 appeal, the Alabama appeals court ordered supplemental briefing on the possible impact of Atkins upon Jenkins’s sentence. Jenkins asked the court to stay the proceedings and conduct an evidentiary hearing after the Alabama legislature enacted a statute about executing the mentally retarded or, in the alternative, to vacate his sentence and remand after the Alabama legislature acted. He later asserted that he is mentally retarded and therefore ineligible for the death penalty. The appeals court proceeded to reject Jenkins’s Atkins claim on the merits because the record established that his IQ was 76 and

7 536 U.S. 304, 321 (2002) (holding that the execution of mentally retarded offenders violates the Eighth Amendment). The rule announced in Atkins is retroactive to cases on collateral review. In re Holladay, 331 F.3d 1169, 1172–73 (11th Cir. 2003).

20

that he maintained relationships and employment. Id. at 155. The Alabama Supreme Court summarily affirmed, in relevant part. Ex parte Jenkins, 972 So. 2d 159, 165 (Ala. 2005), remanded to 972 So. 2d 165 (Ala. Ct. Crim. App. 2005) (juror misconduct claim), cert. denied, 552 U.S. 1167 (2008).

In 2008, Jenkins filed his first petition for a writ of habeas corpus in federal court pursuant to 28 U.S.C. § 2254. The federal proceedings were stayed for several years while Jenkins pursued additional, unrelated Rule 32 relief in the state courts, ultimately unsuccessfully. In the district court, Jenkins moved for an evidentiary hearing on his Atkins claim. After reviewing the state court record, with a focus on the Rule 32 testimony of Drs. Lisak and Kirkland, the district court denied the motion in 2015. Noting that our Circuit has accepted the State of Alabama’s definition of mental retardation following Atkins, the court found that Jenkins is not intellectually disabled under that standard. In a 347-page order, the district court also denied relief on all of Jenkins’s claims including ineffective assistance of counsel during the penalty phase and affirmed as reasonable the state court’s conclusion that Jenkins failed to establish prejudice from any error by his counsel. The district court later denied reconsideration and a certificate of appealability.

Jenkins now appeals. Our Court granted Jenkins a certificate of appealability, 28 U.S.C. § 2253(c), on two issues:

21

1) Whether the district court erred in denying Appellant’s claim that his trial counsel rendered ineffective assistance of counsel in the investigation and presentation of mitigating evidence during the penalty phase of Appellant’s 1991 trial; and

2) Whether the district court erred in denying Appellant’s claim that he is intellectually disabled and ineligible for the death penalty under Atkins v. Virginia, 536 U.S. 304 (2002).

II. STANDARDS OF REVIEW We review the denial of a § 2254 petition for a writ of habeas corpus de novo. Morrow v. Warden, Ga. Diagnostic Prison, 886 F.3d 1138, 1146 (11th Cir. 2018). Yet, the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA) also governs this appeal, which establishes a “highly deferential standard for evaluating state-court rulings, [and] demands that state-court decisions be given the benefit of the doubt.” Cullen v. Pinholster, 563 U.S. 170, 181 (2011) (quoting Woodford v. Visciotti, 537 U.S. 19, 24 (2002) (per curiam)). Thus, under AEDPA, our review of a final state habeas decision is greatly circumscribed, and we cannot grant a state petitioner habeas relief on any claim that was adjudicated on the merits in state court unless the state court’s adjudication of the claim:

(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or

(2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.

28 U.S.C. § 2254(d)(1)–(2).

22

“[C]learly established Federal law” means “the holdings, as opposed to the dicta, of [the Supreme] Court’s decisions as of the time of the relevant state-court decision.” Williams v. Taylor, 529 U.S. 362, 412 (2000). “[W]hen the last state court to decide a prisoner’s federal claim explains its decision on the merits in a reasoned opinion . . . a federal habeas court simply reviews the specific reasons given by the state court and defers to those reasons if they are reasonable.” Wilson v. Sellers, 138 S. Ct. 1188, 1192 (2018). For a state-court decision to be “contrary to” clearly established federal law, the state court must have “applie[d] a rule different from the governing law set forth in [Supreme Court] cases, or if it decides a case differently than [the Supreme Court] ha[s] done on a set of materially indistinguishable facts.” Bell v. Cone, 535 U.S. 685, 694 (2002). An “unreasonable application” of federal law occurs “if the state court correctly identifies the governing legal principle from [the Supreme Court’s] decisions but unreasonably applies it to the facts of the particular case.” Id. To be clear, “an unreasonable application of federal law is different from an incorrect application of federal law.” Williams, 529 U.S. at 410. “Indeed, ‘a federal habeas court may not issue the writ simply because that court concludes in its independent judgment that the relevant state-court decision applied clearly established federal law erroneously or incorrectly.’” Renico v. Lett, 559 U.S. 766, 773 (2010) (quoting Williams, 529 U.S. at 411). Rather, the state court’s application of federal law “must be ‘objectively

23

unreasonable.’ This distinction creates ‘a substantially higher threshold’ for obtaining relief than de novo review.” Id. (quotations omitted); White v. Woodall, 572 U.S. 415, 419 (2014) (explaining that, for purposes of § 2254(d)(1), the State’s court’s application of clearly established federal law must be “‘objectively unreasonable,’ not merely wrong; even ‘clear error’ will not suffice” (quoting Lockyer v. Andrade, 538 U.S. 63, 75 (2003))).

A state court’s decision is reasonable “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) (quoting Yarborough v. Alvarado, 541 U.S. 652, 664 (2004)). “[E]ven a strong case for relief does not mean the state court’s contrary conclusion was unreasonable.” Id. at 102. Rather, a prisoner must show that the state court’s ruling “was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Id. at 103.

In addition, “a determination of a factual issue made by a State court shall be presumed to be correct,” and the prisoner bears “the burden of rebutting the presumption of correctness by clear and convincing evidence.” 28 U.S.C. § 2254(e)(1). “This deference requires that a federal habeas court more than simply disagree with the state court before rejecting its factual determinations. Instead, it must conclude that the state court’s findings lack even fair support in the record.”

24

Smith v. Comm’r, Ala. Dep’t of Corr., 924 F.3d 1330, 1337 (quoting Rose v. McNeil, 634 F.3d 1224, 1241 (11th Cir. 2011)). When a federal habeas court makes a factual finding as part of its habeas determination, we review for clear error, which is a “highly deferential” standard of review. Thomas v. Allen, 607 F.3d 749, 752 (11th Cir. 2010) (quoting Holton v. City of Thomasville Sch. Dist., 425 F.3d 1325, 1350 (11th Cir. 2005)). “A finding is ‘clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” United States v. U.S. Gypsum Co., 333 U.S. 364, 395 (1948).

III. DISCUSSION

A. Ineffective Assistance of Counsel A prisoner alleging that he received ineffective assistance of counsel in violation of the Sixth Amendment must establish two elements. Strickland v. Washington, 466 U.S. 668, 687 (1984). “First, the defendant must show that counsel’s performance was deficient.” Id. Review of counsel’s actions is “highly deferential” and “a court must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Id. at 689. Additionally, “every effort [must] be made to eliminate the distorting effects of hindsight, to reconstruct the circumstances of counsel’s challenged conduct, and to evaluate the conduct from counsel’s perspective at the time.” Id. In other words,

25

the petitioner must show “that counsel made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment.” Id. at 687.

“Second, the defendant must show that the deficient performance prejudiced the defense.” Id. Prejudice occurs when there is a reasonable probability that, “but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694. “When a defendant challenges a death sentence . . . the question is whether there is a reasonable probability that, absent the errors, the sentencer—including an appellate court, to the extent it independently reweighs evidence—would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Id. at 695. “The likelihood of a different result must be substantial, not just conceivable.” Harrington, 562 U.S. at 112. In determining whether there is a reasonable probability of a different result, a court must “consider ‘the totality of the available mitigation evidence—both that adduced at trial, and the evidence adduced in the habeas proceeding’—and ‘reweig[h] it against the evidence in aggravation.’” Porter v. McCollum, 558 U.S. 30, 41 (2009) (quoting Williams, 529 U.S. at 397–98).

“Surmounting Strickland’s high bar is never an easy task.” Harrington, 562 U.S. at 105 (quoting Padilla v. Kentucky, 559 U.S. 356, 371 (2010)). Thus, “[e]stablishing that a state court’s application of Strickland was unreasonable

26

under § 2254(d) is all the more difficult” because “[t]he standards created by Strickland and § 2254(d) are both ‘highly deferential’ and when the two apply in tandem, review is ‘doubly’ so.” Id. (internal citations omitted).

To be clear, “whether defense counsel’s performance fell below Strickland’s standard” is not the question before a federal habeas court reviewing a state court’s decision under § 2254. Id. at 101. As the Supreme Court has cautioned:

Were that the inquiry, the analysis would be no different than if, for example, this Court were adjudicating a Strickland claim on direct review of a criminal conviction in a United States district court. Under AEDPA, though, it is a necessary premise that the two questions are different . . . [for] [a] state court must be granted a deference and latitude that are not in operation when the case involves review under the Strickland standard itself.

Id. Accordingly, where, as here, “§ 2254(d) applies, the question is not whether counsel’s actions were reasonable. The question is whether there is any reasonable argument that counsel satisfied Strickland’s deferential standard.” Id. at 105 (emphasis added). Consequently, “[f]ederal habeas courts must guard against the danger of equating unreasonableness under Strickland with unreasonableness under § 2254(d).” Id. With these reviewing principles in mind, we discuss each of Strickland’s elements in turn.

1. Deficient Performance Jenkins argues that his counsel performed deficiently by failing to “investigate compelling mitigating evidence or develop a penalty-phase theory.”

27

The state court disagreed. It quoted decisions of our Court that hold that “there is no per se rule that evidence of a criminal defendant’s troubled childhood must always be presented as mitigating evidence.” Jenkins, 972 So. 2d at 144 (quoting Marek v. Singletary, 62 F.3d 1295, 1300 (11th Cir. 1995)). Noting that “Downey was in charge of the penalty phase” but “we do not have the benefit of Downey’s testimony as to what occurred and why,” it reviewed the trial record to find that “counsel argued residual doubt and Jenkins’s good character at the penalty phase.” Id. at 145–46. It concluded that “Downey’s decision to concentrate on reasonable doubt and to portray Jenkins as a good person was reasonable under the circumstances.” Id. at 147.

Jenkins insists, however, that counsel’s failure to develop mitigating evidence about his childhood abuse was “not the product of strategy, but rather arose from ingnorance and inexeperiance [sic].” The problem for Jenkins is that, like the state courts, this Court also does not know what Downey did or why. The record is silent as to his thoughts and intentions as he prepared for the penalty phase. Given that Jenkins bore the burden of overcoming the strong presumption of competence and of proving Downey’s deficient performance, we are perplexed why Jenkins’s new counsel did not have Downey testify or submit an affidavit in the Rule 32 proceeding. See Strickland, 466 U.S. at 687, 691 (“The defendant must show that counsel’s performance was deficient” and “inquiry into counsel’s

28

conversations with the defendant may be critical to a proper assessment of counsel’s investigation decisions, just as it may be critical to a proper assessment of counsel’s other litigation decisions.” ). Nor did Jenkins testify to the substance of his conversations with his counsel. “The reasonableness of a trial counsel’s acts, including lack of investigation or excluding character witnesses from the sentencing phase, depends ‘critically’ upon what information the client communicated to counsel.” Chandler v. United States, 218 F.3d 1305, 1324 (11th Cir. 2000) (en banc) (quoting Strickland, 466 U.S. at 691). There is simply no record evidence to support Jenkins’s central assertion that Downey made no tactical choices in his preparation for the penalty phase.

Jenkins instead relies on Scofield’s Rule 32 testimony about how he and Downey prepared for the trial and sentencing. We acknowledge that Scofield repeatedly testified that he did not really know what he was doing and that he failed to make strategic choices with respect to mitigating circumstances and sentencing. But Scofield was equally clear that Downey—the more experienced lawyer—was in fact responsible for the penalty phase. And Scofield also repeatedly testified that he did not know what Downey may or may not have done to prepare. Furthermore, his lack of knowledge about Downey’s preparation was understandable. He and Downey were not colleagues who shared an office and

29

might reasonably be expected to know what the other was doing much of the time. Scofield and Downey did not even work in the same city.

This case is unlike those in which we have found deficient performance where the record showed that counsel investigated insufficiently because he “waited until the eleventh hour” to begin preparing for the penalty phase while faced with “overwhelming evidence of guilt.” See, e.g., Johnson v. Sec’y, D.O.C., 643 F.3d 907, 932 (11th Cir. 2011). The limited record that Jenkins developed establishes that Downey spent at least 71 hours preparing for trial, spread out over 17 months. The state court interpreted this sparse record as reflecting a penalty- phase strategy of residual doubt, the logical extension of Downey’s guilt-phase investigation of the evidence and Scofield’s pursuit at trial of an acquittal based on reasonable doubt. See Jenkins, 972 So. 2d at 147. In such a situation, “[a] lawyer’s time and effort in preparing to defend his client in the guilt phase of a capital case continues to count at the sentencing phase.” Tarver v. Hopper, 169 F.3d 710, 715 (11th Cir. 1999).

We decline to dictate how many hours or how many sessions an attorney must spend interviewing his counsel in order to provide effective assistance. “No absolute rules dictate what is reasonable performance for lawyers.” Chandler, 218 F.3d at 1317 (citing Strickland, 466 U.S. at 688–89). But even this limited record undermines Jenkins’s assertion that Downey prepared inadequately. Downey’s 71

30

hours of preparation included 35 hours of jailhouse conversations with Jenkins during 18 separate meetings. 8 According to Downey’s notes, Scofield was present for only five of those meetings (12 hours’ worth), leaving 23 hours over 13 meetings between Jenkins and Downey alone. Downey’s notes reveal little about the content of those conversations beyond broad notations such as “evidence, witnesses, general status,” and discussions with Jenkins of the evidence and testimony against him were surely relevant to Downey’s penalty-phase strategy of residual doubt.

But the record does not reveal that Downey knew or should have known that Jenkins’s childhood merited investigation beyond that which he may have done. The record shows that Downey developed enough of a rapport with Jenkins that, according to his fee declaration, he “g[ave a] Christmas gift [and] discuss[ed] how much to tell his family.” But how much Downey knew about Jenkins’s family background can only be guessed from Scofield’s Rule 32 hearing testimony. Scofield testified that he discussed Jenkins’s background during his dozen or so jailhouse meetings with him, and that Downey was present for some of those meetings. Scofield—and possibly Downey—learned that Jenkins experienced

8 We note that the state court clearly erroneously stated that Downey’s fee declaration documented 171 hours of pretrial preparation. See Jenkins, 972 So. 2d at 137. However, the state court accurately stated that “he spent over 25 hours talking with Jenkins in more than 10 visits to the jail.” Id. That statement was not an unreasonable determination of the facts.

31

abuse as a child, that his stepfather was brutal, and that he lived on the streets and in juvenile institutions because of his brutal home life. It is against this backdrop that Downey’s presumptive “decision not to investigate must be directly assessed for reasonableness in all the circumstances, applying a heavy measure of deference to counsel’s judgments.” Strickland, 466 U.S. at 691.

That assessment depends a great deal on what Jenkins told his counsel about his background. See id. But we do not know whether Downey was present when Scofield learned about Jenkins’s childhood abuse, nor what else Downey may have learned during his 13 solo discussions with Jenkins. Therefore, “given the absence of evidence in the record, we must assume counsel carried out his professional responsibility and discussed mitigation with his client.” Chandler, 218 F.3d at 1324; Burt v. Titlow, 571 U.S. 12, 17 (2013) (“It should go without saying that the absence of evidence cannot overcome the ‘strong presumption that counsel’s conduct [fell] within the wide range of reasonable professional assistance.’” Id. (alteration in original) (quoting Strickland, 466 U.S. at 689)). We will not “turn that presumption on its head by giving [Jenkins] the benefit of the doubt where it is unclear what [Downey] did or did not do.” Williams v. Head, 185 F.3d 1223, 1235 (11th Cir. 1999). We will not assume that Jenkins told Downey more than he told Scofield, and Scofield denied remembering that Jenkins told him about the severity or frequency of his stepfather’s physical abuse. He denied remembering Jenkins

32

telling him about being beaten daily, being locked in his room 24/7, or being forced to eat his feces. And there is no suggestion in the record that counsel knew or had reason to investigate the allegation of sexual abuse by his stepgrandfather. The record simply does not reflect that Jenkins told either of his lawyers about the severe abuse that some of the Rule 32 witnesses later described. We will not fault those lawyers for failing to present as mitigation this limited evidence that they may have reasonably believed to be unremarkable. See Strickland, 466 U.S. at 691.

To be sure, counsel’s failure “to discover all reasonably available mitigating evidence” may constitute deficient performance in some cases. Wiggins v. Smith, 539 U.S. 510, 524 (2003) (quoting ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases § 11.4.1(C) (1989)). Our Court has identified deficient performance in several instances where counsel had good reason to investigate the defendant’s background but failed to do so. See Morrow, 886 F.3d at 1147 (collecting cases). But we assess “a particular decision not to investigate . . . for reasonableness in all the circumstances, applying a heavy measure of deference to counsel’s judgments.” Strickland, 466 U.S. at 691. Thus, we have rejected a hard rule that counsel’s “strategic decisions can be considered reasonable only if they are preceded by a ‘thorough investigation.’” Williams, 185 F.3d at 1237 (quoting Rogers v. Zant, 13 F.3d 384, 387 (11th Cir. 1994)). Because we simply do not know why Downey chose not to investigate Jenkins’s childhood

33

more thoroughly, Downey is entitled to the presumption that this strategic decision was reasonable. See Chandler, 218 F.3d at 1314 n. 15 (en banc) (“An ambiguous or silent record is not sufficient to disprove the strong and continuing presumption.”).

And though we have also rejected “a per se rule of ineffective assistance where counsel does not consult family members,” Holladay v. Haley, 209 F.3d 1243, 1252 (11th Cir. 2000), we note that Downey did not fail altogether to consult family members. Downey had at least one substantive conversation with Jenkins’s grandmother, and we know from her Rule 32 hearing testimony that she never witnessed Jenkins’s childhood abuse. Given the silence of the record and the strong presumption of competence to which Downey is entitled, we decline to hold as unreasonable the state court’s decision not to fault Downey for not doing more when Jenkins and his grandmother did not and could not, respectively, provide a basis for doing so. Cf. United States v. Pease, 240 F.3d 938, 941–92 (11th Cir. 2001) (“we cannot say as a matter of law that reliance on a client’s statements is per se deficient performance”). Nor did Jenkins’s counsel have an absolute duty to put on abusive-childhood evidence, even if they knew of it. See Chandler, 218 F.3d at 1319 (“Counsel is not required to present every nonfrivolous defense; nor is counsel required to present all mitigation evidence, even if the additional mitigation evidence would not have been incompatible with counsel’s strategy.”);

34

Waters v. Thomas, 46 F.3d 1506, 1511 (11th Cir. 1995) (en banc) (“Our decisions are inconsistent with any notion that counsel must present all available mitigating circumstance evidence.”).

Instead, we affirm as a reasonable application of Strickland the state court’s conclusion that Downey’s strategic decision to emphasize residual doubt was reasonable. See Jenkins, 972 So. 2d at 147. In reaching this conclusion, the State court explained that residual doubt had been recognized as an effective strategy in the penalty phase of capital cases, citing cases from this Circuit, law review articles, and the American Law Institute’s model penal code. These sources provided guides (i.e. “prevailing professional norms”) from which the state court could measure whether Downey’s performance fell within the wide range of reasonable professional assistance contemplated by Strickland. See Knowles, 556 U.S. at 123 (“[B]ecause the Strickland standard is a general standard, a state court has even more latitude to reasonably determine that a defendant has not satisfied that standard.”); see also Renico, 559 U.S. at 776 (“Because AEDPA authorizes federal court’s to grant relief only when state courts act unreasonably, it follows that ‘[t]he more general the rule’ at issue—and thus the greater the potential for reasoned disagreement among fair-minded judges—‘the more leeway [state] courts have in reaching outcomes in case-by-case determinations.” (quoting Yarborough, 541 U.S. at 664))). Indeed, we have held that “residual doubt is perhaps the most

35

effective strategy to employ at sentencing.” Chandler, 218 F.3d at 1320 n.28. “[E]specially when—as in this case—the evidence of guilt was not overwhelming, we expect that petitioners can rarely (if ever) prove a lawyer to be ineffective for relying on this seemingly reasonable strategy to defend his client.” Id. at 1320 (footnote omitted); see also Tarver, 169 F.3d at 715–16 (collecting law review articles on the merits of residual-doubt strategy). Not only can such a strategy be constitutionally adequate, in some cases it may be “the most effective performance in defense to the death penalty.” Tarver, 169 F.3d at 716.

Accordingly, we cannot say that the state court’s determination that Downey’s manifest strategy of residual doubt satisfied Strickland’s deferential standard was objectively unreasonable under the circumstances. Jenkins’s case is precisely the sort where reminding the jury of the misgivings it may have held about the State’s case could have been effective. Scofield’s guilt-phase closing argument from that very morning, which explicitly emphasized doubt and the weakness of the State’s evidence, would have been fresh in the minds of the jury.9 Although we are not questioning the jury’s guilty verdict or the sufficiency of the evidence, which are beyond the scope of our habeas review, we acknowledge that the State’s case against Jenkins was—as Downey reminded the jury during the

9 During the penalty phase of a capital case, “[e]vidence presented at the trial of the case may be considered insofar as it is relevant to the aggravating and mitigating circumstances without the necessity of re-introducing that evidence at the sentence hearing.” Ala. Code § 13A-5-45(c).

36

penalty phase—entirely circumstantial. Indeed, Scofield maintained post- conviction that the case for reasonable doubt was “[v]ery strong.” No one saw Jenkins commit either the murder or the robbery and kidnapping that elevated Hogeland’s murder to a capital crime. Scofield’s guilt-phase closing argument challenged the inferential leaps the State was asking the jury to make in finding that Jenkins committed these capital-predicate offenses and that he formed the intent to kill. This situation was not one where a guilt-phase strategy of acquittal— and therefore a penalty-phase strategy of residual doubt—was unreasonable in light of the evidence. Cf. Johnson, 643 F.3d at 932–33 (deficient sentencing investigation where defendant had confessed to two murders).

Thus, although Downey’s penalty-phase closing argument did not specifically invoke lingering doubt, it clearly reminded the jury of the entirely circumstantial nature of the State’s guilt-phase case. See Chandler, 218 F.3d at 1320 & n.6 (a strategy of “focus[ing] on obtaining an acquittal and then, at sentencing, on lingering doubt” is reasonable, even if counsel does not use the words “lingering doubt” or “residual doubt”). Downey closed this reminder by explaining to the jury how a foundational ancient legal system proscribed capital punishment in the absence of eyewitness testimony—the exact situation here. The state court found that Downey’s penalty-phase closing argument reflected a strategic “decision to concentrate on reasonable doubt,” see Jenkins, 972 So. 2d at

37

147, a tactical choice entitled to a “strong presumption of correctness,” Marek, 62 F.3d at 1300 & n.3.

Jenkins suggests other theories of mitigation that Downey might have more effectively pursued,10 but our deferential review under Strickland does not ask whether counsel could possibly or ideally have been more effective. “The test for ineffectiveness is not whether counsel could have done more.” Waters, 46 F.3d at 1518. We do not ask whether an attorney’s representation “deviated from best practices or common custom,” and we should resist the temptation to second-guess an attorney with the benefit of our hindsight. Richter, 562 U.S. at 105. Rather, Strickland asks only “whether, in light of all the circumstances, the identified acts or omissions were outside the range of professionally competent assistance.” Strickland, 466 U.S. at 690. And, because § 2254(d) applies, we ask only “whether there is any reasonable argument that counsel satisfied Strickland’s deferential standard.” Harrington, 562 U.S. at 105.We conclude that the state court did not

10 In particular, Jenkins argues that his counsel should have presented “model prisoner” evidence. The state court noted, “Good conduct during pretrial incarceration is not necessarily a mitigating circumstance,” before finding that Jenkins had not met his burden of establishing that his counsel even had information about his conduct in jail. Jenkins, 972 So. 2d at 149. We agree, and we decline to impose a requirement of sua sponte investigation of this kind in every capital case. Cf. Williams, 529 U.S. at 396 (finding deficient performance based in part on not returning a call from a prison ministry volunteer who offered to vouch for defendant’s good behavior in prison). “Surely, counsel is not required to call a witness to testify to facts such as lack of violent nature when the jury has rejected such an approach and has found that the defendant is guilty of murder.” Griffin v. Wainwright, 760 F.2d 1505, 1512 (11th Cir. 1985), vacated on other grounds, 476 U.S. 1112 (1986), reaff’d, 874 F.2d 1397 (11th Cir. 1989).

38

unreasonably determine that Jenkins failed to establish objectively incompetent performance by his counsel during the penalty phase of his trial. Our inquiry is not “what the best lawyers would have done. Nor is the test even what most good lawyers would have done. We ask only whether some reasonable lawyer at the trial could have acted, in the circumstances, as defense counsel acted at trial. . . . We are not interested in grading lawyers’ performances.” White v. Singletary, 972 F.2d 1218, 1220–21 (11th Cir. 1992). And “[i]f fairminded jurists could reasonably disagree” about the reasonableness of counsel’s performance, “then habeas relief is due to be denied.” Johnson, 643 F.3d at 932.

2. Prejudice

Even if Jenkins had demonstrated that his counsel performed as no reasonable lawyer could have, he must also demonstrate prejudice—that is 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. Given that the jury here recommended a sentence of death by the narrowest possible vote, 10 to 2, Jenkins need establish only “a reasonable probability that at least one juror would have struck a different balance” between life and death. Wiggins, 539 U.S. at 537. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Strickland, 466 U.S. at 694. As explained above, in measuring that probability here, we will “evaluate the totality of the available

39

mitigation evidence—both that adduced at trial, and the evidence adduced in the habeas proceeding in reweighing it against the evidence in aggravation.” Williams, 529 U.S. at 397–98.

We begin our analysis by clarifying what exactly Jenkins is arguing about prejudice. We do not know precisely which aggravating and mitigating factors the jury may have implicitly found before making its recommendation of death. The trial court found at sentencing two statutory aggravating factors: the robbery and the kidnapping of the victim. See Ala. Code § 13A-5-49(4). Because none of the mitigation evidence offered in the Rule 32 hearing bears on those factors, we see no reasonable probability that the aggravating factors would have been determined differently by the jury or the court. The court also found at sentencing two statutory mitigating factors: no significant history of prior criminal activity and age. See id. § 13A-5-51(1), (7). There was also strong evidence of Jenkins’s intoxication, which was a key part of Jenkins’s guilt-phase defense and could have supported a third mitigating factor, impaired capacity. See id. § 13A-5-51(6).11

11 The jury had heard evidence about all three of these statutory mitigators during the guilt phase of the trial. Downey’s penalty-phase opening statement reviewed for the jury all seven of the possible statutory mitigating factors, and the court also recited them in its jury instructions, reminding the jury that they could consider guilt-phase evidence as well as non-statutory mitigation. Jenkins has not established that there is a substantial probability that the jury’s recommendation would have been different had it been more explicitly reminded of Jenkins’s age, criminal history, and intoxication, given that the trial court, the Rule 32 trial court, and the state appellate court all explicitly considered those mitigators and found them outweighed by the aggravators.

40

Jenkins does not argue that the jury or the court would have found other statutory mitigators in light of the Rule 32 evidence. Rather, he contends that the jury might have found non-statutory mitigation based on Jenkins’s abusive childhood, low intelligence, or mental health problems that would have outweighed the aggravating factors. 12 The Rule 32 trial court found that Jenkins had not demonstrated prejudice even assuming that counsel had performed deficiently. See Jenkins, 972 So. 2d at 138. It concluded that he had not shown a reasonable probability that the sentencer “would have concluded that a weighing of the aggravating and mitigating circumstances did not warrant death” upon a reweighing of the evidence. See id. It emphasized the depravity of the crime and the aggravating circumstances and found “that none of [Jenkins’s] witnesses was credible and that they had exaggerated the level of abuse that Jenkins had been exposed to when he was [a] child.” See id. at 138, 144. The state appellate court affirmed, declining to disturb the credibility determinations of the trial court. Id. at 142–44. Independently

12 Alabama law lists seven statutory mitigating circumstances, but states that mitigation “shall . . . not be limited to” those seven circumstances. Ala. Code § 13A-5-51. Further, “mitigating circumstances shall include any aspect of a defendant’s character or record and any of the circumstances of the offense that the defendant offers as a basis for a sentence of life imprisonment without parole instead of death, and any other relevant mitigating circumstances which the defendant offers as a basis for a sentence of life imprisonment without parole instead of death.” Id. § 13A-5-52. A jury may recommend a sentence of death only if it finds that “one or more aggravating circumstances . . . exist and that they outweigh the mitigating circumstances, if any.” Id. § 13A-5-46(e)(3).

41

reweighing the evidence, it concluded that, “[g]iven the aggravating circumstances that were proven by the State and the facts surrounding Hogeland’s murder, we, like the circuit court, are confident that death was the appropriate punishment for Jenkins’s actions.” Id. at 148.

In light of the state court’s credibility determination and our doubly deferential standard of review, the state court’s determination that Jenkins failed to establish prejudice was not contrary to, or an unreasonable application of, Strickland. Our review of this evidence does not show that it is likely that the jury would have concluded that death was not the appropriate punishment if it had heard the Rule 32 evidence. Much of the proffered mitigation evidence is, in our view, much less mitigating than Jenkins asserts. In particular, as the state court noted, the testimony of Jenkins’s older brother Michael was likely to have been a double-edged sword. See Evans v. Sec’y, Dep’t of Corr., 703 F.3d 1316, 1327 (11th Cir. 2013) (en banc) (“it is reasonable to conclude that a defendant was not prejudiced when his mitigation evidence was a two-edged sword” (quoting Ponticelli v. Sec’y, Fla. Dep’t of Corr., 690 F.2d 1271, 1296 (11th Cir. 2012)). Michael was the closest relative to testify at the Rule 32 hearing and his testimony about the abuse Jenkins experienced should have been the most reliable, since he experienced much of the same physical abuse himself. The brothers shared the same filthy bedroom, received beatings from the same paddle, and were frequently

42

absent from school together. But Michael’s testimony contained some discrepancies, and he admitted to cognitive and memory problems. Perhaps Michael’s testimony would have made Jenkins appear more sympathetic in the eyes of the jury—but perhaps not. While Jenkins ran away from home and eventually committed murder, Michael responded to the same background by running away from home, graduating from high school, and providing health care for both the terminally ill and people with substance use disorders. It is possible that the jury could even have found Jenkins to be more culpable for not overcoming his difficult upbringing and to pose more of a danger of future violence.

We also find reasonable the state court’s identification of other problems with the mitigating evidence Jenkins argues should have been presented. Hearing that evidence firsthand in the Rule 32 proceeding, the trial court found the testimony of the cousins less credible in light of the lack of physical abuse documented in the school health records. See Jenkins, 972 So. 2d at 139 (noting that school records observed a rash and gingivitis). It faulted Pitts in particular for purporting to have witnessed ongoing, severe abuse but taking only minimal, ineffective steps to obtain outside help. See id. It further noted that DeLavega and Wagoner actually testified that they never saw Jenkins beaten, and it noted all of the witnesses’ interest in getting Jenkins off death row. All of these factors would

43

likely have caused a reasonable jury to discount their testimony. See id. at 139–42. We must accept the trial court’s credibility determinations of the Rule 32 witnesses. Consalvo v. Sec’y for Dep’t of Corrs., 664 F.3d 842, 845 (11th Cir. 2011) (“Determining the credibility of witnesses is the province and function of the state courts, not a federal court engaging in habeas review. Federal courts have ‘no license to redetermine credibility of witnesses whose demeanor has been observed by the state trial court, but not by them.’” (quoting Marshall v. Lonberger, 459 U.S. 422, 434 (1983))). The credibility of a witness is a question of fact entitled to a presumption of correctness under AEDPA. Id. “In the absence of clear and convincing evidence, we have no power on federal habeas review to revisit the state court’s credibility determinations.” Bishop v. Warden, 726 F.3d 1243, 1259 (11th Cir. 2013) (citing Lonberger, 459 U.S. at 434).

The state court also noted that Wagoner testified that she was not available to testify at the trial, and it found that Seal’s testimony would have been cumulative to that of her husband. Thus, counsel’s failure to call those two witnesses could not have been prejudicial to Jenkins. See id. at 141, 142. Neither of these conclusions involves an unreasonable determination of the facts. We agree that the Rule 32 testimony of Sherry Seal was substantially similar to the testimony of her husband Lonnie Seal that the jury heard during the penalty phase. Both testified about meeting Jenkins in California, Jenkins’s assistance with their move

44

to Alabama, and his residence with them in their Alabama home. Sherry added only that Jenkins sometimes wet his bed. It was not unreasonable for the state court to conclude that Sherry’s testimony was cumulative of her husband’s, and “a petitioner cannot satisfy the prejudice prong of the Strickland test with evidence that is merely cumulative of evidence already presented.” Rose v. McNeil, 634 F.3d 1224, 1243 (11th Cir. 2011).

Further, the school, medical, and juvenile court records that Jenkins introduced in the Rule 32 proceeding have their own problems. As noted, they do not document or even corroborate the severe level of abuse alleged by some of the witnesses. Though these documents tell a sad story, they do not create a substantial likelihood that the jury would have recommended life. Although they portray Jenkins as a victim of truly unfortunate circumstances, they also show his unsavory delinquent behavior. That delinquency included a history of stealing cars, and given that the jury likely inferred that Jenkins stole the red Mazda, new information about his history of theft may not have been especially mitigating. Similarly, we agree with the state court that evidence of Jenkins’s compliant conduct in pretrial detention is “minimally mitigating” where there had been no evidence of jail misconduct as an aggravator. See Jenkins, 972 So. 2d at 149; cf. Skipper v. South Carolina, 476 U.S. 1, 5 & n.1 (1986) (good jail behavior is

45

“potentially mitigating” especially “[w]here the prosecution specifically relies on a prediction of future dangerousness in asking for the death penalty”).

Jenkins argues that allowing the jury to see his experience of mental, emotional, and cognitive difficulties, compounded by years of abuse and neglect and considered alongside his mild character and good jail behavior as an adult, would have convinced them that Hogeland’s murder was an “aberration” that deserved a punishment less severe than death. That interpretation of Jenkins’s life story may be conceivable. But more than mere conceivability is required to establish prejudice: “The likelihood of a different result must be substantial, not just conceivable.” Richter, 562 U.S. at 112 (emphasis added). We do not find the state court’s decision that a different result was not substantially likely to be an unreasonable application of Strickland. As we have noted, we think it equally likely that the jury could have rejected this alternate narrative out of hand in light of having just convicted him of the violent murder of a young mother. We also think it possible that the jury could have viewed the murder as just one step in a logical progression of degeneracy that began with behavioral problems at school, continued through the juvenile court system, and escalated past grand theft auto into kidnapping and murder. Taken separately as well as together, the proffered evidence does not compel us to conclude that a verdict of life would have been substantially likely. Because fairminded jurists could disagree about the state

46

court’s conclusion about Strickland prejudice, we affirm the denial of habeas relief on this ground.

B. Intellectual Disability Jenkins argues that the district court erred when it denied his claim that he is intellectually disabled and ineligible for the death penalty under Atkins v. Virginia, 536 U.S. 304 (2002). The Supreme Court held in Atkins that executing an offender who has mental retardation 13 violates the Eighth Amendment’s ban on excessive punishments. Id. at 321. It defined mental retardation by quoting the American Association on Mental Retardation:

Mental retardation refers to substantial limitations in present functioning. It is characterized by significantly subaverage intellectual functioning, existing concurrently with related limitations in two or more of the following applicable adaptive skill areas: communication, self-care, home living, social skills, community use, self-direction, health and safety, functional academics, leisure, and work. Mental retardation manifests before age 18.

Id. at 308 n.3. Alabama courts have followed that tripartite definition, adopting what they consider the “broadest” definition of mental retardation for Atkins purposes:

[A] defendant, to be considered mentally retarded, must have significantly subaverage intellectual functioning (an IQ of 70 or below), and significant or substantial deficits in adaptive behavior.

13 In 2002, the Atkins Court followed prevailing convention in referring to “mentally retarded” offenders. We will use that now-disfavored term only when quoting or discussing older sources. Since 2014, the Supreme Court has exclusively used the terms “intellectually disabled person” and “person with intellectual disability.” See, e.g., Hall v. Florida, 572 U.S. 701, 704 (2014); Moore v. Texas, 139 S. Ct. 666, 672 (2019).

47

Additionally, these problems must have manifested themselves during the developmental period (i.e., before the defendant reached age 18).

Ex parte Perkins, 851 So. 2d 453, 456 (Ala. 2002).

The clinical evidence adduced during the Rule 32 hearing with respect to Jenkins’s intellectual ability was that his IQ was 76, reflecting “borderline intellectual functioning.” The record also included the report of the Lunacy Commission 14 which found that Jenkins was competent to stand trial, as well as Jenkins’s school and juvenile court records. The court also considered all the testimony it had heard about Jenkins and reviewed the transcript of his murder trial before rejecting Jenkins’s claim of mental retardation.

On appeal, after reciting the Perkins standard, the Alabama Court of Criminal Appeals explained that Jenkins had not established mental retardation:

Dr. Kirkland testified that he performed psychological tests on Jenkins and that Jenkins’s IQ was 76. There was evidence presented at Jenkins’s trial indicating that Jenkins maintained relationships with other individuals and that he had been employed by P.S. Edwards Landscaping Company, Cotton Lowe 76 Service Station, and Paramount Painting Company. The record fails to show that Jenkins meets the most liberal view of mental retardation adopted by the Alabama Supreme Court in Perkins.

14 Alabama law requires a capital defendant to be “committed to the Department of Mental Health and Mental Retardation for examination” if there is a question about his capacity to stand trial or if there will be a mental-disease-or-defect defense. Ala. Code § 15-16-22(a). “The assigned mental health professional(s) shall examine the defendant with respect to determining the presence of any mental disease or defect which, if determined to be present, would affect the capacity of the defendant to proceed or continue to trial or which would affect the defendant’s criminal responsibility at the time of the commission of the crime.” Id.

48

Jenkins, 972 So. 2d at 155. In this habeas appeal, Jenkins raises arguments about all three components of the intellectual disability showing under Atkins and Perkins, which we address in turn.

1. Intellectual Function The first component of intellectual disability under Perkins is “significantly subaverage intellectual functioning (an IQ of 70 or below).” Perkins, 851 So. 2d at 456. Jenkins argues that the use of a strict IQ cutoff of 70 was contrary to or an unreasonable application of Atkins.

We disagree. The state court’s decision that Jenkins failed to establish mental retardation in this respect was a reasonable application of the record and of Atkins. As a threshold matter, we note that the state court did not say it was applying a strict cutoff of 70; it merely stated that Dr. Kirkland had measured Jenkins’s IQ as 76. Jenkins, 972 So. 2d at 155. Under our deferential review, the question is whether the decision that Jenkins had not established the intellectual functioning aspect of intellectual disability was an unreasonable application of Atkins. The Supreme Court explained in Atkins that it was leaving “to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.” Atkins, 536 U.S. at 317 (quoting Ford v. Wainwright, 477 U.S. 399, 416–17 (1986)). In line with that invitation, our Circuit has accepted Alabama’s judicial definition of intellectual disability as laid out in

49

Perkins. 15 See Thomas, 607 F.3d at 752–53. Jenkins objects, however, that the Perkins standard, which he characterizes as articulating a “strict cutoff” of 70, conflicts with more recent Supreme Court precedent that instead requires a “clinical” definition that goes beyond a single IQ score.

In particular, Jenkins points to Hall v. Florida, in which the Supreme Court held that when an offender’s IQ is near but greater than 70, courts must take into account the IQ test’s standard error of measurement as well as non-IQ evidence of intellectual function. See Hall v. Florida, 572 U.S. 701, 723 (2014) (“This Court agrees with the medical experts that when a defendant’s IQ test score falls within the test’s acknowledged and inherent margin of error, the defendant must be able to present additional evidence of intellectual disability, including testimony regarding adaptive deficits.”). But, as Jenkins concedes, our Circuit has specifically held that Hall is not retroactive to cases on collateral review. See In re Henry, 757 F.3d 1151, 1161 (11th Cir. 2014) (explaining that Hall does not supply a new substantive rule but rather “merely provides new procedures for ensuring that States do not execute members of an already protected group”). We also note that Alabama courts have found the Perkins standard to comport with Hall in that it does not prevent an offender from presenting the standard error of measurement

15 Notwithstanding the urging of the Alabama Supreme Court, see Perkins, 851 So. 2d at 455 n.1, the Alabama legislature has not enacted a statutory definition of intellectual disability.

50

or non-IQ evidence of intellectual disability. See Reeves v. State, 226 So. 3d 711, 728–29 (Ala. Ct. Crim. App. 2016). Here, the state court did not unreasonably apply Atkins when it analyzed the intellectual functioning component of intellectual disability.

Jenkins also argues that the state court’s decision on this component in his Rule 32 proceeding resulted from an unreasonable determination of the facts. He asserts that his IQ score of 76, when combined with the test’s standard error of measurement and considered alongside the Flynn effect (which we discuss below), does show that he has significantly subaverage intellectual function. For the reasons that follow, we disagree.

Most fundamentally, neither of the clinicians who testified about Jenkins’s intellectual ability in the Rule 32 hearing opined that he was mentally retarded. The closest thing Jenkins can point to is the testimony of Dr. Kirkland that his IQ is two standard deviations below the mean. But his entire testimony, in context, did not establish intellectual disability. To the contrary: Dr. Kirkland testified that Jenkins “scored in the range of borderline intellectual functioning which is between mild mental retardation and low average intellectual functioning.” Furthermore, Jenkins’s own expert Dr. Lisak fully reviewed Dr. Kirkland’s clinical findings. He agreed with those findings, concluding that the IQ results show “borderline intelligence.” It is tremendously significant that no clinical assessment

51

in the entire record of these proceedings has found that Jenkins has mental retardation or intellectual disability.

Jenkins nonetheless argues that his IQ score of 76 establishes the intellectual component of intellectual ability when the standard error of measurement and the Flynn effect are considered alongside that score. It was not unreasonable for the state court to disagree. As we have discussed, Jenkins is not retroactively entitled to have the state courts take into account the margin of error of the IQ test. That statistical margin of error—5 points in this case—takes into account characteristics of the test and variations due to chance in order to state how confident the tester is that the measured result accurately states Jenkins’s true, unknowable IQ. A result of 76 with a standard error of measurement of 5 simply means that the tester is 68% confident that Jenkins’s true IQ falls between 71 and 81, and 95% confident that his true IQ falls between 66 and 86. See David H. Kaye & David A. Freedman, Reference Guide on Statistics, in Reference Manual on Scientific Evidence 211, 243–44 (3d ed. 2011); see also Hall, 572 U.S. at 738–39 (Alito, J., dissenting). Of course, Jenkins would like us to consider only the possibility that the test result overstated his IQ—but the standard error of measurement is a two-way street. “[T]he standard error of measurement is a bi-directional concept that does not carry with it a presumption that an individual’s IQ falls to the bottom of his IQ

52

range.” Ledford v. Warden, Ga. Diag. & Classification Prison, 818 F.3d 600, 641 (11th Cir. 2016).

Even if Jenkins were somehow entitled to subtract the 5-point margin of error from his score and thereby lower his “true” IQ to 71, he still would not satisfy the clinical intellectual component of intellectual disability. Thus, he sought in the district court to introduce evidence about the Flynn effect in order to bring his IQ down to 70 or below. As our Court has explained, the Flynn effect “recognizes the fact that IQ test scores have been increasing over time”:

The Flynn effect acknowledges that as an intelligence test ages, or moves farther from the date on which it was standardized, or normed, the mean score of the population as a whole on that assessment instrument increases, thereby artificially inflating the IQ scores of individual test subjects. Therefore, the IQ test scores must be recalibrated to keep all test subjects on a level playing field.

Thomas, 607 F.3d at 753. We have said that “[a]n evaluator may also consider the ‘Flynn effect’” in assessing an offender’s possible intellectual disability, id. (emphasis added), but neither we nor the Alabama Supreme Court nor the U.S. Supreme Court have said that a court must consider it in order to reasonably apply Atkins. See, e.g., Reeves, 226 So. 3d at 739 (“This Court has repeatedly held that a circuit court is not required to accept, consider, or apply the ‘Flynn Effect’ in determining intellectual disability.”).

More problematically, Jenkins did not raise the Flynn effect issue in the Alabama state courts. Jenkins asserts that he implicitly raised it as part and parcel

53

of his Atkins claim during his Rule 32 appeal, but we do not agree that the Alabama courts had a full and fair opportunity to consider this issue, because it was not explicitly presented to them. See Kelley v. Sec’y for the Dep’t of Corr., 377 F.3d 1317, 1349–50 (11th Cir. 2004) (holding that one ineffective-assistance argument was unexhausted despite the petitioner’s general presentation of a Strickland claim in the state courts). Thus, under AEDPA, Jenkins is procedurally barred from raising this unexhausted claim in his federal habeas petition. See 28 U.S.C. § 2254(b)(1)(A), (c); Mason v. Allen, 605 F.3d 1114, 1119 (11th Cir. 2010). We will allow a prisoner to overcome a procedural bar if he establishes cause for his default and prejudice therefrom. See, e.g., Smith v. Jones, 256 F.3d 1135, 1138 (11th Cir. 2001). But Jenkins cannot establish prejudice from being barred from asserting this claim now because the Supreme Court has not “clearly established” that a court must consider the Flynn effect in assessing intellectual disability. Thus, under AEDPA, it could not form the basis for a federal habeas claim. See 28 U.S.C. § 2254(d)(1). We therefore conclude that the state court did not unreasonably determine the facts or unreasonably apply Atkins with respect to the intellectual component of intellectual disability.

2. Adaptive Function The second component of intellectual disability is “significant or substantial deficits in adaptive behavior.” Perkins, 851 So. 2d at 456. Jenkins argues that the

54

state court, which mentioned only Jenkins’s successes in employment and social skills, see Jenkins, 972 So. 2d at 155, failed to consider his deficits. He adds that even the mention of employment success does not tell the whole story. Jenkins insists that his jobs were menial, his employers took advantage of him, and he held the jobs for only a few months each.

But the state court’s terse discussion of Jenkins’s adaptive behavior did not involve an unreasonable determination of the facts or an unreasonable application of Atkins. To the contrary, its decision on this issue was reasonable in light of both the record and Atkins.

Atkins elaborated that the adaptive-skill component of intellectual disability requires substantial present limitation in at least two of the following areas: “communication, self-care, home living, social skills, community use, self- direction, health and safety, functional academics, leisure, and work.” Atkins, 536 U.S. at 308 n.3 (quoting definition of American Association on Mental Retardation). The record supports the state court’s judgment that Jenkins does not have substantial deficits in at least two areas. To be sure, we will assume that Jenkins may be substantially limited in functional academics; Dr. Kirkland assessed his reading, spelling, and arithmetic skills at a third-grade level. But the record does not compel a finding of substantial present deficits in the areas of communication, self-care, home living, and work. Although Jenkins emphasizes

55

his transient employment, the record shows he was able to find work quickly, to work hard to support the Seals’ household until Lonnie Seal found a job, and, later, to move into his own home. That record evidence is inconsistent with a substantial deficit in the area of work.

As the State observes, the facts of the crime also fail to show significant deficits in the areas of communication, self-care, community use, and self- direction. Jenkins was able to communicate well enough to solicit an alibi and to sell his car, including writing out a bill of sale. He then was able to arrange for his flight from Alabama, including purchasing a bus ticket and hitchhiking across the country. Indeed, throughout his trial and postconviction proceedings, the people who knew Jenkins as an adult consistently described a man who did not have serious difficulties in communicating, forming relationships, working, and caring for himself. No witness mentioned these kinds of difficulties in his or her testimony about Jenkins. There was some evidence that Jenkins was “not articulate” and that his house was messy, but that testimony does not rise to the level of overwhelming evidence of substantial functional deficits that would support a conclusion that the state court unreasonably determined the facts.

Of course, the record also contains evidence of Jenkins’s childhood academic and social deficits, attributable at least in part to his parents’ neglect, but Jenkins’s childhood is not directly relevant to our consideration of his present

56

limitations. Although we acknowledge that the Supreme Court has noted that such “risk factors” may support further exploration of the possibility of intellectual disability, see Moore v. Texas, 137 S. Ct. 1039, 1051 (2017), the state court record contains the results of that exploration. Overall, that record supports the state court’s conclusion that Jenkins does not have substantial deficits in adaptive behaviors.

3. Juvenile Onset

The third and final component of intellectual disability is manifestation of the first and second components “before the defendant reached age 18.” Perkins, 851 So. 2d at 456. Jenkins asserts, the State assumes, and we agree that the state court did not rule upon this issue, so there is nothing to which we must defer under AEDPA. See, e.g., Brumfield v. Cain, 135 S. Ct. 2269, 2282 (2015). Thus, we will review the district court’s legal conclusions on this issue de novo, and its factual findings for clear error. See, e.g., Grossman v. McDonough, 466 F.3d 1325, 1335 (11th Cir. 2006). Jenkins argues that he has satisfied the age-of-onset component by showing “a wide range of evidence of deficits in both intellectual and adaptive functioning throughout Jenkins’s childhood.”

In fact, the ample record of Jenkins’s childhood, as developed in the Rule 32 proceeding, does not point to intellectual disability before age 18. School assessments at age 12 reported an IQ score of 83, an adaptive behavior rating of

57

“low,” and noted overall an intellectual capacity of “average,” a diagnosis of dyslexia, and “poor functional skills.” The psychologist attributed Jenkins’s difficulties to a combination of “[f]requent moves, changes in schools and family instability” with “[l]ow ability coupled with a specific learning disability” and “[l]ack of responsibility for self control and poor use of time.” Although Jenkins failed several grades of school, teachers additionally attributed his difficulties to his behavioral problems and his high number of absences from school.

Similarly, IQ tests ordered by the juvenile court at age 14 gave scores of 81 and 86 on different tests, in the “[d]ull–[n]ormal” range and reflecting “a definite learning disability.” These records all show a child with serious academic deficits and some intellectual and adaptive deficits, but they do not clearly show an intellectually disabled child. In particular, Jenkins’s childhood IQ scores of 83, 81, and 86 fail to point to intellectual disability or mental retardation. We are not left with the definite and firm conviction that the district court erred when it found that “Jenkins’s [childhood] intelligence scores and adaptive functioning skills were attributable to his learning disability and other circumstances in his life, rather than mental retardation.”

58

In sum, we agree with the district court that the decision of the Alabama Court of Criminal Appeals that Jenkins does not have intellectual disability was not contrary to or an unreasonable application of Atkins v. Virginia. 16

16 In light of our determination that Jenkins has not met his burden of demonstrating that the state court’s decision was contrary to, or an unreasonable application of, Atkins under § 2254(d), his request for an evidentiary hearing is denied. The dissent asserts that we are attempting to “rewrite history” and refusing to consider Jenkins’s challenge to the district court’s denial of his request for an evidentiary hearing, but that is simply not the case. Even accepting the dissent’s contentions that the COA granted in this case could be construed as encompassing the district court’s denial of the evidentiary hearing, the inescapable fact remains that Jenkins did not challenge the denial of the evidentiary hearing in this appeal. Instead, he asserted that:

“[o]nce a petitioner has shown the state court merits adjudication was unreasonable under § 2254(d), § 2254(e)(2) allows for an evidentiary hearing so long as the petitioner was diligent in state court and pleads facts that, if true, would warrant relief. . . . Because § 2254(d)’s limitations have been overcome . . .

and Jenkins alleged facts sufficient to show he meets all three Atkins prongs, . . .

to the extent this Court finds the record insufficient to grant relief, it should remand for an evidentiary hearing.”

With this statement, Jenkins is clearly not challenging the district court’s denial of his request for an evidentiary hearing. Indeed, at no point in any of his filings before this Court did Jenkins ever argue that the district court abused its discretion in denying his request for an evidentiary hearing. In other words, our prior opinion erroneously addressed an issue that was not properly before this Court, which must be corrected, despite the fact that the dissent, inexplicably, would prefer that we manufacture and address an argument that was never presented. To be clear, Jenkins’s only discussion in this Court regarding the matter of an evidentiary hearing was his request that (A) in the event that we determined that he was entitled to relief under § 2254(d)— i.e., if we concluded that the state court’s decision was contrary to, or an unreasonable application of, Atkins—and (B) we found the record otherwise insufficient to grant him relief upon de novo review of his claim, then (C) we should remand for an evidentiary hearing, pursuant to § 2254(e)(2). Jenkins framed his evidentiary hearing request in this manner because, pursuant to Cullen v. Pinholster, 563 U.S. 170, 185 (2011), a § 2254 petitioner is precluded from receiving an evidentiary hearing in the district court on a claim that was adjudicated on the merits by a state court, unless he first demonstrates an entitlement to relief under § 2254(d). Thus, for the reasons previously set forth in this opinion, because we conclude Jenkins has not met his burden of demonstrating that the state court’s decision was contrary to, or an unreasonable application of, Atkins, under § 2254(d), Supreme Court precedent precludes an evidentiary hearing on his claim. Id. In other words, if, and only if, the state habeas petitioner clears § 2254(d)’s hurdles, may an evidentiary hearing be granted, provided that an evidentiary hearing is not otherwise barred by the limitations set forth in § 2254(e)(2). Id. at 185–86; Pope

59

IV. CONCLUSION

The denial of Jenkins’s petition for a writ of habeas corpus is AFFIRMED.

v. Sec’y, Fla. Dep’t of Corrs., 752 F.3d 1254, 1263 (11th Cir. 2014) (“[Pinholster] rests on a clear, emphatic rule: if a state court has adjudicated the claim on the merits, then a petitioner must satisfy § 2254(d)(1) based only on the record before that state court. . . . It is also true that, under Pinholster, § 2254(d) must be satisfied before a federal habeas court may consider any § 2254(e)(2) evidence.”). Accordingly, because Jenkins did not overcome § 2254(d)’s hurdles, we do not reach the point advanced by the dissent as to whether Jenkins otherwise satisfied § 2254(e)(2)’s requirements for an evidentiary hearing.

60

WILSON, Circuit Judge, dissenting:

Mark Jenkins must prove that the Alabama state courts acted unreasonably in denying him post-conviction collateral relief. 28 U.S.C. § 2254(d)(1). Jenkins has satisfied that burden as to his ineffective assistance of counsel claim, contrary to the majority’s conclusion. As for his Atkins claim, the record is insufficient to decide whether Jenkins is entitled to relief, and the matter should be remanded for an evidentiary hearing. The majority, however, refuses to consider Jenkins’s request for a hearing and relies on pre-Atkins evidence to deny him relief. Because I disagree with the majority’s conclusions, I dissent.

I. Ineffective Assistance of Counsel at Penalty Phase Jenkins first asserts that his trial counsel’s failure to investigate his abusive upbringing and to present any mitigating evidence at the penalty phase of his trial constituted ineffective assistance of counsel. The majority rejects this contention, concluding that the Alabama state courts did not act unreasonably in denying Jenkins’s ineffective assistance of counsel claim on collateral review. I disagree and maintain that Jenkins established both deficient performance and prejudice, entitling him to habeas relief. See Strickland v. Washington, 466 U.S. 668, 687 (1984).

61

A. Deficient Performance The Alabama state courts concluded that Jenkins’s counsel, Stan Downey, did not perform deficiently at the penalty phase. The majority finds that conclusion reasonable primarily because Downey (1) was not obligated to investigate Jenkins’s abusive childhood and (2) reasonably pursued a residual doubt strategy at the penalty phase. But further review of the record requires me to part ways with the majority’s reasoning and, ultimately, persuades me that the Alabama courts’ conclusion was indeed unreasonable. See 28 U.S.C. § 2254(d)(1) (providing that a writ of habeas corpus should not be granted to a state prisoner unless the state court adjudication involved “a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States”).

Jenkins offers numerous examples of how his counsel performed deficiently in the penalty phase of trial. First, counsel failed to thoroughly investigate Jenkins’s abusive childhood—an error that constitutes deficient performance. In Williams v. Allen, we held that counsel’s failure to investigate mitigating evidence at the penalty stage constituted ineffective assistance of counsel because counsel drew upon only three sources of information—a psychologist’s report, a presentence investigation report, and an interview with the appellant’s mother. 542 F.3d 1326, 1340 (11th Cir. 2008) (Williams v. Allen); see also Debruce v. Comm’r,

62

Ala. Dep’t of Corr., 758 F.3d 1263, 1270, 1273–74 (11th Cir. 2014) (finding counsel’s performance deficient where counsel interviewed the petitioner and his mother, but failed to contact expert witnesses or other family members). In that case, we concluded that counsel’s lack of investigation resulted in “an incomplete and misleading understanding of Williams’ life history.” Williams v. Allen, 542 F.3d at 1340.

Jenkins’s counsel did even less than the attorneys in Williams v. Allen and Debruce. That is, Jenkins’s counsel conducted no investigation. Downey did not consider Jenkins’s poor school records, mention Jenkins’s juvenile history, produce documentary evidence that Jenkins suffered from severe depression, or pursue records demonstrating Jenkins’s significant deficits in cognitive functions. We have previously found that a counsel’s failure to investigate these rudimentary examples of mitigating evidence constitutes ineffective assistance of counsel. See Daniel v. Comm’r, Ala. Dep’t of Corr., 822 F.3d 1248 (2016) (reversing and remanding the district court’s dismissal of the ineffective assistance of counsel claim and noting how counsel should have more thoroughly investigated Daniel’s history of physical and sexual abuse, his poor school records, and evidence of a potential intellectual disability).

Given the lack of preparation, it is not surprising that Downey failed to call any witnesses to testify at the penalty stage about Jenkins’s childhood, thus failing

63

to give a complete depiction of Jenkins’s “excruciating life history.” See Wiggins v. Smith, 539 U.S. 510, 537 (2003) (finding deficient performance where counsel did not adequately investigate petitioner’s life history for mitigating evidence); see also Debruce, 758 F.3d at 1270, 1273–74; Johnson v. Sec’y, Dep’t of Corr., 643 F.3d 907, 932 (11th Cir. 2011) (“No reasonable attorney, after being told by his client that he had an abusive upbringing, would fail to interview members of his client’s family who were readily available and could corroborate or refute the allegations of abuse.”).1 If Downey had effectively investigated—or investigated at all—he would have been aware of several available and informed witnesses. Ultimately, only one witness testified at the penalty phase: Lonnie Seal.2 Downey met with Seal for only 15 minutes before the penalty phase began. That 15 minutes represents the extent of Downey’s preparation for the penalty phase—a wholly inadequate amount of time under Supreme Court precedent. See Williams v. Taylor, 529 U.S. 362, 395 (2000) (Williams v. Taylor) (finding mitigation investigation deficient where “counsel did not begin to prepare for that phase of the proceeding until a week before the trial”).

1 The majority goes to great pains to maintain that Downey did not know about Jenkins’s abusive childhood. This assertion is directly refuted by the record. Douglas Scofield, Jenkins’s lead attorney, testified that Jenkins described his traumatic upbringing at meetings at which Downey was present. 2 It is unclear whether Downey contacted Seal and requested that he testify during the penalty phase of trial. Given that Downey urged Scofield to conduct Seal’s direct examination because Scofield was better acquainted with Seal, it is likely that Scofield made the arrangements.

64

Despite ample evidence of deficient performance, the Alabama Court of Criminal Appeals (CCA) concluded that Downey performed reasonably. The CCA’s decision rested on three explanations: (1) because Downey did not testify, courts must presume he acted reasonably; (2) based on the record, Downey played an active role investigating and preparing for the penalty phase of trial; and (3) Downey’s decision to pursue a residual doubt strategy was reasonable.3 Because each of these findings constituted an unreasonable determination of the facts or an unreasonable application of clearly established federal law, Jenkins is entitled to habeas relief. See 28 U.S.C. § 2254(d)(1)–(2).

The CCA’s first rationale—presuming Downey acted reasonably because Downey did not testify at the Rule 32 hearing—was based on both an unreasonable determination of facts and an unreasonable application of clearly established law. The CCA stated that “the record is virtually silent as to what actions were or were not taken or what was or was not done by Mr. Downey at trial and why” and therefore assumed that Downey’s actions “could have been reasonable and strategic under the circumstances.” The majority likewise assumes Downey acted reasonably. This conclusion is based, in large part, on Downey’s failure to testify at the Rule 32 hearing. According to the majority, “[t]he problem for Jenkins is that, like the state courts, this Court also does not know what Downey did or why.”

3 The majority opinion endorsed the CCA’s first and third explanations.

65

But the record clearly illustrates Downey’s failure to investigate and overall lack of preparation. First, Downey’s fee declarations—which the majority concedes were “detailed”—showed no investigation of mitigating circumstances, and no preparation for the penalty phase. Importantly, Downey’s fee declarations were exceptionally detailed, recounting his work in five-minute increments. The single reference to the “mitigating” phase of trial in the declarations concerned a meeting with Scofield that occurred only a week before trial. Other than that vague entry, no notation indicates that Downey did anything to prepare for the penalty phase of trial during the nearly two years he represented Jenkins.

Second, and more persuasively, Douglas Scofield, Jenkins’s lead attorney, explicitly testified that Downey told him that Downey had not investigated potential mitigating evidence. Both the Alabama courts and the majority ignore that fact and instead focus on Scofield’s testimony that, “[a]part from what he has told me, I don’t know what he has done.” This statement, to which the majority assigns great weight, is hardly illuminating. Of course Scofield knows only what Downey told him. And what Downey told him was this: Downey did not investigate potentially mitigating evidence before the penalty phase.

In the face of this direct evidence of deficient performance, the Alabama courts and the majority pivot to Downey’s failure to testify at the Rule 32 hearing. Without his testimony, the majority reasons, we must assume that Downey’s

66

decision not to investigate was strategic and therefore his representation reasonable. I disagree, and I have reservations about the practical effects of the majority’s understanding of the law. The majority suggests (if not holds) that the absence of testimony from an allegedly deficient attorney per se means that the attorney’s actions were reasonable. It is true that “counsel is strongly presumed to have rendered adequate assistance,” Strickland, 466 U.S. at 690. But a petitioner may overcome that burden with evidence of inadequate assistance. And that evidence may take various forms. To satisfy the burden, it is not necessary for the allegedly deficient counsel to testify and admit wrongdoing. Nor is it necessary for the record to reflect counsel’s “thoughts and intentions as he prepared for the penalty phase,” as the majority seems to require. When evaluating whether a counsel’s performance was deficient, we ask whether his performance “fell below an objective standard of reasonableness.” Strickland, 466 U.S. 687–88. We do not contemplate what counsel thought or intended. We consider whether his performance was objectively unreasonable. See id. Because the Alabama courts did not apply the proper standard, their conclusion was an unreasonable application of clearly established law. See id.

The CCA’s second reason for denying Jenkins relief was also based on an unreasonable application of clearly established federal law. The CCA concluded that Downey had adequately investigated and prepared for the penalty phase of

67

trial, emphasizing that Downey met with Jenkins on several occasions and spoke with Jenkins’s grandmother. Based on these findings, the CCA determined that Downey provided adequate assistance of counsel during the penalty phase. This is an unreasonable application of clearly established law that “[a]n attorney has a duty to conduct a reasonable investigation, including an investigation of the defendant’s background, for possible mitigating evidence.” Porter v. Singletary, 14 F.3d 554, 557 (11th Cir. 1994).

As previously discussed, it is clear from the record that Downey knew of Jenkins’s horrible upbringing yet still failed to investigate potentially mitigating evidence. “A reasonable investigation . . . should have included, at a minimum, interviewing other family members who could corroborate the evidence of abuse and speak to the resulting impact on [Jenkins].” Williams v. Allen, 542 F.3d at 1340. But Downey did not conduct a single interview—of a family member or otherwise—in preparation of the penalty phase. 4 Because Downey did not conduct an investigation, let alone a reasonable investigation, the CCA’s conclusion that Downey adequately investigated potentially mitigating evidence was an unreasonable application of clearly established law.

4 Downey’s fee declaration reflects one telephone call he had with Jenkins’s grandmother, Doris Wagoner. But that call occurred 18 months before trial and, according to Wagoner, concerned a request for money.

68

The CCA’s third rationale for its decision—that Jenkins’s counsel reasonably and strategically decided to focus on residual doubt—was both an unreasonable application of clearly established law and an unreasonable determination of the facts. In Williams v. Taylor, the Supreme Court held that otherwise reasonable trial strategies cannot be justified unless counsel first conducted a constitutionally adequate investigation. 529 U.S. at 395–96. And here, Downey conducted no such investigation. It cannot be said, therefore, that Downey’s decision—if one can even call it that—to pursue a residual doubt strategy was strategically reasonable. See id. The CCA’s conclusion to the contrary was an unreasonable application of clearly established law.

Likewise, the conclusion was an unreasonable determination of the facts.

On direct appeal, the Alabama CCA noted that “[t]he evidence overwhelmingly pointed to the guilt of the appellant.” Jenkins v. State, 627 So. 2d 1034, 1041 (Ala. Crim. App. 1992). Then, on appeal from Jenkins’s Rule 32 petition, the CCA again acknowledged how “[t]he evidence establishing Jenkins’s guilt was overwhelming.” Jenkins v. State, 972 So. 2d 111, 157 (Ala. Crim. App. 2004); see also id. at 156 (“The amount of evidence incriminating Jenkins . . . was overwhelming.”). Despite the “overwhelming” evidence of Jenkins’s guilt, the CCA concluded that Jenkins’s counsel acted reasonably in presenting a residual doubt defense during the penalty phase of trial. But no reasonable attorney would

69

employ a residual doubt strategy in the face of such overwhelming evidence. See Johnson, 643 F.3d at 932–33. The Alabama courts’ contrary conclusion, therefore, was an unreasonable determination of the facts.

B. Prejudice Jenkins’s Rule 32 petition also pleaded sufficient facts to show that his counsel’s deficient performance prejudiced his defense, satisfying the second Strickland prong. To establish prejudice, a petitioner must show 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. Here, the jury voted in favor of the death penalty by the slimmest margin permitted under Alabama law: ten votes in favor and two votes against. Jenkins need only establish, then, “a reasonable probability that at least one juror would have struck a different balance.” Wiggins, 539 U.S. at 537 (emphasis added). In measuring that probability, we “evaluate the totality of the available mitigation evidence— both that adduced at trial, and the evidence adduced in the habeas proceeding in reweighing it against the evidence in aggravation.” Williams v. Taylor, 529 U.S. at 397–98.

The testimony elicited at the Rule 32 hearing was horrendous. Four family members offered detailed, consistent testimony about Jenkins’s traumatic childhood. The majority describes that testimony, highlighting some of the most

70

appalling facts elicited at the Rule 32 hearing. For example, Jenkins’s stepfather would beat Jenkins on a semi-daily basis, would lock him in a room for hours without food, and even made Jenkins eat his own feces. In addition to the mitigating facts detailed by the majority, the following came to light during the Rule 32 hearing. Jenkins’s mother had an affair while her husband, Jenkins’s stepfather, was in prison. When Jenkins’s stepfather was released, he took out his frustrations about the affair on Jenkins. Because Jenkins’s biological father was Hispanic, Jenkins’s stepfather referred to Jenkins in demeaning names like “piece of shit,” “Mexican trash,” “Puerto Rican puke,” and “bastard.”

The Rule 32 testimony also detailed the alcohol substance abuse, drug addiction, and incessant domestic violence that surrounded Jenkins’s childhood. And, most shockingly, the testimony described how Jenkins’s grandfather sexually abused Jenkins when Jenkins was just four years old. As a result of that sexual abuse, Jenkins, suffered bladder and bowel movement issues. Jenkins’s incontinence became a trigger for his stepfather: when Jenkins had an accident, his stepfather would beat him or subject him to severe degradation. For example, Jenkins’s stepfather would hang Jenkins’s soiled sheets in the front yard for all the neighbors to see, make Jenkins wear his soiled clothes on his head, and, most

71

disturbingly, make Jenkins eat his own feces. The only way to escape the abuse was by running away. So, by age twelve, Jenkins lived on the streets. 5 Jenkins also introduced medical, school, and juvenile court records from his childhood, which shed further light on his difficult childhood and intellectual deficiencies. Moreover, two psychologists testified about Jenkins’s history of psychological trauma stemming from his childhood abuse—weighty evidence that is not even considered in the majority’s discussion. Finally, two prison guards testified that Jenkins was the ideal prisoner during pretrial detention. One of the guards even described Jenkins as the best inmate he had ever supervised.

Given this compelling mitigating evidence, it is not difficult to conclude that Jenkins established “a reasonable probability that at least one juror would have struck a different balance” between life and death. Wiggins, 539 U.S. at 537; see also Daniel, 822 F.3d at 1276 (listing several factors that support a prejudice finding, including (1) evidence of impaired intellectual functioning; (2) evidence of childhood sexual abuse; and (3) the fact that the jury voted only 10–2 in recommending a death sentence).

5 The CCA found parts of the family members’ testimony not credible, though it did not explicitly discredit testimony describing some of the worst abuse. Given the grotesque nature of the family members’ testimony, and the fact that their testimony was consistent, there is reasonable probability that at least one more juror would have been compelled to vote against a death sentence.

72

II. Intellectual Disability Jenkins also argues that he is entitled to habeas relief under Atkins v.

Virginia, 536 U.S. 304 (2002). He contends that he satisfies the three requirements necessary to establish intellectual disability under Alabama law—(1) subaverage intellectual functioning; (2) deficits in adaptive behavior; and (3) onset before the age of 18, see Ex parte Perkins, 851 So. 2d 453, 456 (Ala. 2002)—thus, exempting him from implementation of the death penalty. Alternatively, Jenkins argues that we should remand for an evidentiary hearing on the Atkins issue if we determine that the record is insufficient to decide the claim.

The majority concludes that the CCA did not act unreasonably in concluding that Jenkins failed to satisfy any of the Perkins prongs. In doing so, it relies on the evidence that Jenkins presented before the Supreme Court decided Atkins. In a previous opinion that it now vacates, the majority also addressed Jenkins’s request for an evidentiary hearing and determined that he was not entitled to one because he had not diligently attempted to develop the factual basis for his Atkins claim in state court. Jenkins v. Comm’r, Ala. Dep’t of Corr., 936 F.3d 1252, 1280 (11th Cir. 2019). Alternatively, the majority concluded that Jenkins would not be able to prove that he was intellectually disabled at such a hearing, given the “ample evidence about intellectual ability” that Jenkins had already adduced. Id. The majority now concludes that Jenkins did not properly appeal the evidentiary-

73

hearing issue because it was not explicitly specified in the certificate of appealability (COA) or properly asserted in his initial brief.

I disagree with the majority’s decision that the evidentiary-hearing issue is not squarely before this court on appeal for several reasons discussed below. And I also disagree with the majority’s ultimate conclusion that the CCA did not act unreasonably in concluding that Jenkins failed to satisfy any of the Perkins prongs, as the record is insufficient to undergo a substantive Atkins analysis and the matter should be remanded for an evidentiary hearing.

A. Appellate Review of Jenkins’s Request for an Evidentiary Hearing The majority attempts to rewrite history and now claims that Jenkins did not properly appeal the denial of an evidentiary hearing, apparently because the COA does not explicitly refer to an evidentiary hearing. I disagree because the basis of the district court’s denial of an evidentiary hearing, the record of the appeal, and the nature of the appeal—a constitutional claim by a death row petitioner—support construing the COA to encompass the evidentiary hearing issue.

First, the COA should be construed to encompass the denial of an evidentiary hearing because the district court’s denial of Jenkins’s Atkins claim and its denial of his request for a hearing were one and the same. In the order denying Jenkins’s motion for an evidentiary hearing, the district court stated that Jenkins would first have to prevail under § 2254(d) to be entitled to an evidentiary hearing.

74

The court then addressed Jenkins’s Atkins claim at great length and concluded that he failed to show that the CCA’s decision was unreasonable under § 2254(d). Then it summarily stated that “[b]ecause he cannot prevail on his Atkins claim, Jenkins is not entitled to an evidentiary hearing on that claim,” without further discussion.

Second, the majority’s decision to not consider Jenkins’s request for an evidentiary hearing is contrary to the record of this appeal. Jenkins specifically appealed the denial of his motion for an evidentiary hearing in the Notice of Appeal. His motion for a COA requested review of the district court’s determination that his Atkins claim failed—the basis of its denial of an evidentiary hearing. The panel granted a COA on the issue of “[w]hether the district court erred in denying Appellant’s claim that he is intellectually disabled and ineligible for the death penalty under Atkins” without explicitly specifying his request for an evidentiary hearing.

Nonetheless, in the eyes of the panel and the parties—at least initially—the COA encompassed Jenkins’s request for an evidentiary hearing. In his initial brief, Jenkins asserts that the state court record shows that he is intellectually disabled and, alternatively, requests the court remand for an evidentiary hearing if the court finds the record insufficient to grant relief. Jenkins even states that he “has not had an opportunity to develop and present his claim post-Atkins.” The government

75

does not challenge the propriety of addressing Jenkins’s request for an evidentiary hearing—it argues that he is not entitled to one. And the majority squarely addressed Jenkins’s request for an evidentiary hearing in its previous opinion, but it agreed with the government that Jenkins was not entitled to one. See Jenkins, 936 F.3d at 1275, 1279–80 (stating “[Jenkins] also appeals the denial of an evidentiary hearing on this issue” and “Jenkins argues that the district court erred when it denied his request for an evidentiary hearing on his Atkins claim”).

Finally, for the majority to now change course is particularly troubling because Jenkins is a death row petitioner raising an Atkins claims, and he has never had a post-Atkins hearing on his claim. See Burgess v. Comm’r, Ala. Dep’t of Corr., 723 F.3d 1308, 1316, 1318 (11th Cir. 2013) (holding that the state court unreasonably relied on the pre-Atkins record, where evidence “was presented in an entirely different context and without the benefit of any explanation of how it would or would not be consistent with” an intellectual disability).

To be sure, this court’s review is limited to the issues specified in the COA.

McClain v. Hall, 552 F.3d 1245, 1254 (11th Cir. 2008). But we also “construe the issue specification in light of the pleadings and other parts of the record.” Murray v. United States, 145 F.3d 1249, 1251 (11th Cir. 1998) (per curiam). Here, the district court did not make an independent determination concerning an evidentiary hearing; it determined that Jenkins’s Atkins claim failed and, therefore, that Jenkins

76

was not entitled to either habeas relief or an evidentiary hearing. Jenkins specifically appealed the district court’s denial of his request for an evidentiary hearing, and he has maintained on appeal that an evidentiary hearing is warranted if the court finds the record insufficient to determine his Atkins claim. Moreover, the parties and the court were all in agreement—until the majority’s abrupt change of mind—that the scope of the appeal included the district court’s denial of an evidentiary hearing.

I see no reason why the majority now turns a blind eye to these parts of the record and withholds appellate review from Jenkins on his request for an evidentiary hearing.

B. Jenkins’s Atkins Claim I also disagree with the majority’s conclusion that Jenkins is not entitled to relief on his Atkins claim, as it is not possible to undergo a substantive Atkins analysis based on the record before us. For one, the CCA’s evaluation of Jenkins’s Atkins claim was a mere three sentences. And, more significantly, Jenkins never had an opportunity to present evidence on his Atkins claim. To deny Jenkins’s claim, therefore, the majority relies on evidence from unrelated parts of the record. Such an approach is improper because, “[p]rior to Atkins, [the petitioner] could not have been expected to necessarily present evidence sufficient to support an Atkins claim because such evidence constituted a [double-edged] sword.” Burgess, 723

77

F.3d at 1320. The evidence was a double-edged sword, we explained, because prior to Atkins, evidence of intellectual disability could suggest that a defendant would be dangerous in the future; therefore, “a defendant could reasonably decide not to highlight his mental retardation.” Id. at 1318. For that reason, courts should not rely on pre-Atkins evidence—evidence that was presented in a wholly distinct context—to support a finding that a petitioner is not intellectually disabled. So, here, instead of attempting to make a medical diagnosis based on an insufficient record, we should remand for an evidentiary hearing.

The government argues that Jenkins did not diligently attempt to present his Atkins claim in the state courts, and so he should be precluded from an evidentiary hearing on this claim now.6 I disagree.

“[I]f the petitioner was diligent in developing the record in the state habeas proceedings, ‘a federal court may grant an evidentiary hearing.” Id. at 1320; see 28 U.S.C. § 2254(e)(2). “‘Diligence depends on whether the prisoner made a reasonable attempt, in light of the information available at the time, to investigate and pursue claims in state court,’ and ‘will require in the usual case that prisoner, at a minimum, seek an evidentiary hearing in state court.’” Burgess, 723 F.3d at 1320 (alteration omitted) (emphasis omitted). Given the unusual procedural

6 The majority refused to grant Jenkins an evidentiary hearing in its original opinion for this reason. See Jenkins, 936 F.3d at 1279–80.

78

posture of Jenkins’s claim, Jenkins made a reasonable attempt to investigate and pursue his Atkins claim in state court.

The Supreme Court issued Atkins v. Virginia while Jenkins’s petition for post-conviction collateral relief was pending before the CCA. The CCA, therefore, requested supplemental briefing on the effect, if any, Atkins had on Jenkins’s petition. At that point, neither the Alabama Supreme Court nor the Alabama legislature had interpreted the groundbreaking case, making Jenkins’s counsel’s next step unclear. Jenkins’s counsel nonetheless made a reasonable attempt to pursue Jenkins’s Atkins claim—he requested that the CCA “vacate his death sentence and remand the case back to the lower court for further proceedings.” The request appears even more reasonable when considered alongside procedurally analogous cases decided around the same time.

At the time Jenkins’s counsel was preparing the supplemental briefing, there had been three other post-conviction petitions for relief that shared Jenkins’s unusual procedural posture. See Clemons v. State, 55 So. 3d 314 (Ala. Crim. App. 2003); Wood v. State, 891 So. 2d 398 (Ala. Crim. App. 2003); Tarver v. State, 940 So. 2d 312 (Ala. Crim. App. 2004). These three cases are similar to Jenkins’s petition in that (1) the petitioners were sentenced to death; (2) the petitioners filed Rule 32 petitions; (3) their Rule 32 petitions were dismissed at the trial court level; (4) while their appeals before the CCA were pending, Atkins was decided; and

79

(5) given the Atkins decision, each petitioner submitted supplemental briefing. In their supplemental briefs, Clemons, Wood, and Tarver each argued that he was intellectually disabled under Atkins and thus it would be unconstitutional to execute him. None of the petitioners asked for an evidentiary hearing on their Atkins claim or requested a remand for further proceedings. Despite the lack of request for a hearing, the CCA remanded for an evidentiary hearing in each case.

With these cases in mind, it would have been reasonable for Jenkins to think that broadly stating his Atkins claim would constitute “diligent” pursuit of his claim. Nonetheless, Jenkins did more than these other petitioners: he specifically requested that the CCA “vacate his death sentence and remand the case back to the lower court for further proceedings.” Thus, given “the information available at the time,” Jenkins made a “reasonable attempt” to pursue his claim.

Additionally, in Alabama, post-conviction relief petitioners with factually sufficient claims are entitled to an evidentiary hearing. Ala. R. Crim. P. 32.9. So, if the CCA had granted Jenkins’s request to remand “for further proceedings,” Jenkins would have received an evidentiary hearing. Jenkins’s request that the CCA “remand for further proceedings,” then, was effectively equivalent to a request for an evidentiary hearing.

80

III. Conclusion

Jenkins’s Rule 32 evidentiary hearing spanned three days. Four family members testified about his horrific upbringing. Two psychiatrists testified about how that upbringing affected Jenkins as an adult. Two prison guards testified about Jenkins’s exemplary behavior while in pretrial detention. And one friend testified about Jenkins’s character. Evidence introduced also documented Jenkins’s academic failings and his juvenile record. In short, Jenkins presented an abundance of compelling mitigating evidence. But the jury never heard any of that evidence. Downey failed to introduce it at the penalty phase. In fact, Downey failed to investigate any mitigating evidence whatsoever. Downey’s failures cannot be attributed to strategic decisions. Nor can they be called reasonable. His performance can only be described as deficient. And that deficient performance was surely prejudicial—there is more than a reasonable probability that the horrifying and detailed Rule 32 testimony would have persuaded one juror to vote against sentencing Jenkins to death. Jenkins is therefore entitled to habeas relief on his ineffective assistance of counsel claim.

Jenkins is also entitled to a hearing on his Atkins claim. Though Jenkins presented ample mitigating evidence at the Rule 32 hearing, he never had an opportunity to present evidence concerning (1) his intellectual functioning, (2) his adaptive deficits, or (3) whether his potential intellectual disability manifested

81

before he reached 18—the three requirements necessary to establish intellectual disability under Alabama law and thus exempt an individual from a death sentence. The CCA nonetheless decided to evaluate Jenkins’s Atkins claim, apparently believing that the record, which included no direct evidence concerning whether Jenkins was intellectually disabled, was sufficient to resolve the complex medical issue. In just three sentences, the CCA arrived at a diagnosis: not intellectually disabled.

Despite the insufficient factual record, the majority fails to remand for an evidentiary hearing and relies on unrelated, pre-Atkins evidence to deny Jenkins’s Atkins claim. Equally troubling, the majority now refuses to consider Jenkins’s request for an evidentiary hearing, which he has maintained in the district court and this court, and which is properly before this court on appeal. I therefore dissent.