Samra v. Warden, Donaldson Correctional Facility

Case: 14-14869 Date Filed: 09/08/2015 Page: 1 of 42 [DO NOT PUBLISH] IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT ________________________ No. 14-14869 ________________________ D.C. Docket No. 2:07-cv-01962-LSC MICHAEL BRANDON SAMRA, Petitioner - Appellant, versus WARDEN, DONALDSON CORRECTIONAL FACILITY, Respondent - Appellee. ________________________ Appeal from the United States District Court for the Northern District of Alabama ________________________ (September 8, 2015) Before ED CARNES, Chief Judge, and HULL and ROSENBAUM, Circuit Judges. PER CURIAM: Petitioner-Appellant Michael Brandon Samra was convicted and sentenced to death by an Alabama court in 1998 for the murders of four people, including Case: 14-14869 Date Filed: 09/08/2015 Page: 2 of 42 two children. Samra’s conviction and sentence were upheld on direct appeal, and the Alabama state courts rejected his claims for postconviction relief. Samra sought federal habeas relief under 28 U.S.C. § 2254, but the district court denied Samra’s federal petition. Samra now appeals raising two issues. First, Samra argues that his trial counsel was ineffective for failing to investigate evidence of brain dysfunction and for introducing and emphasizing evidence of Samra’s membership in a Satanic gang, which he contends strengthened the state’s aggravation case. Second, Samra asserts that his appellate counsel was ineffective for not raising an argument on appeal that Samra was entitled to pretrial notice of the specific statutory aggravating factor that the state intended to rely upon in pursuing the death penalty. After a thorough review of the record, and with the benefit of oral argument, we now affirm the denial of Samra’s federal habeas petition. I. A. The Criminal Offense Samra was convicted of capital murder, in violation of Alabama Code § 13A-5-40(a)(10), and he was sentenced to death for his role in the killings of Randy Duke, Dedra Hunt, Chelsea Hunt, and Chelisa Hunt. See Samra v. State (Samra Direct Appeal), 771 So. 2d 1108, 1111-12 (Ala. Crim. App. 1999); Samra v. Price (Samra § 2254 Proceeding), No. 2:07-cv-1962-LSC, 2014 WL 4452676, 2 Case: 14-14869 Date Filed: 09/08/2015 Page: 3 of 42 at *1 (N.D. Ala. Sept. 5, 2014). According to the evidence established at trial and by Samra’s own confession, Randy Duke’s sixteen-year-old son Mark Anthony Duke (“Duke”) devised the murder following an argument where Randy Duke refused to allow Duke to use a pickup truck. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. After planning the murder with Samra and two other friends, David Collums and Michael Ellison, the group obtained two guns and returned to Duke’s house. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. Samra and Duke entered the house while Collums and Ellison waited nearby. Id. Once inside, Duke went to the living room and shot his father, killing him. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. Meanwhile, Samra shot Dedra1 non-fatally in the cheek, and she fled upstairs, locking herself in the master bedroom’s bathroom with her six-year-old daughter Chelisa. Samra Direct Appeal, 771 So. 2d at 1111; Samra § 2254 Proceeding, 2014 WL 4452676, at *1. Duke broke down the bathroom door and shot Dedra to death. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. But because they had run out of bullets, Duke went downstairs to retrieve kitchen knives; he then slit Chelisa’s throat with a kitchen knife. Id. Dedra’s seven-year-old daughter Chelsea was hiding under a bed in another bedroom when Duke found her. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. According to Samra’s statement, Chelsea pled with Duke to 1 Because three members of the Hunt family were involved in these facts, we refer to each by first name to avoid confusion. 3 Case: 14-14869 Date Filed: 09/08/2015 Page: 4 of 42 stop and, as also evidenced by the defensive wounds on her body, vigorously fought for her life. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. Unable to kill her by himself, Duke held Chelsea down while Samra slit her throat. Samra § 2254 Proceeding, 2014 WL 4452676, at *1; Samra Direct Appeal, 771 So. 2d at 1112. According to the testimony of the medical examiner, both girls died as a result of drowning in their own blood. After committing the murders, Samra and Duke ransacked the house to make it appear as though a burglary had gone wrong. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. Duke later returned to his house on March 23, 1997, where he called 911 to report the murders. After a couple of days of investigating, the police determined that Duke, Samra, Collums, and Ellison were the perpetrators. Samra confessed his role in the crime during questioning and assisted police in recovering weapons. See Samra § 2254 Proceeding, 2014 WL 4452676, at *1. B. Trial Proceedings Samra was indicted for the four murders under § 13A-5-40(a)(10), Ala. Code, which makes it a capital crime when “two or more persons are murdered by the defendant by one act or pursuant to one scheme or course of conduct.” Samra Direct Appeal, 771 So. 2d at 1111. The indictment itself was a single paragraph listing the victims’ names and the weapons used to kill them. Later, the indictment 4 Case: 14-14869 Date Filed: 09/08/2015 Page: 5 of 42 was amended to add an aiding-and-abetting component and to clarify that Dedra Hunt was killed with a gun. The indictment did not describe any further details of the crime, and it did not specify any of the then-existing statutory aggravating circumstances under § 13A-5-49, Ala. Code, that would permit imposition of the death penalty. Samra was represented at trial and on appeal primarily by appointed counsel Richard Bell, an experienced attorney who devoted approximately 35 to 40 percent of his practice to criminal-defense work and had defended three capital cases prior to Samra’s. Samra § 2254 Proceeding, 2014 WL 4452676, at *2. Based on his initial investigations and interactions with Samra, Bell decided that his defense strategy needed to focus on Samra’s mental condition and the influence of gang membership on Samra’s actions. Samra § 2254 Proceeding, 2014 WL 4452676, at *2. Bell concluded, though, that Samra had only “a very small chance of winning the guilt phase,” and that the sentencing phase would be the main event. See Vol. 39 at 183. Before trial, Bell filed a motion to compel the state to disclose the aggravating circumstances upon which it intended to rely in seeking a death sentence, arguing that Samra needed to be informed of the aggravating factors that the state intended to prove in order to prepare for the sentence hearing. Samra § 2254 Proceeding, 2014 WL 4452676, at *2. The state argued that it was not 5 Case: 14-14869 Date Filed: 09/08/2015 Page: 6 of 42 required to reveal this information prior to the sentencing phase but that in any event, “the aggravating circumstances are very straight forward in the indictment.” Samra § 2254 Proceeding, 2014 WL 4452676, at *2. The court denied Samra’s motion, noting that the statute sets forth a very “limited” and “particularized” set of aggravating factors. With respect to Samra’s mental condition, Bell enlisted the expertise of Dr. Charles Scott, a forensic psychiatrist. Samra § 2254 Proceeding, 2014 WL 4452676, at *2. Dr. Scott prepared a 21-page report based on a six-hour psychiatric interview with Samra, an interview with Samra’s parents, and a review of Samra’s school and medical records. Id. Based on these interviews and records, Scott concluded that Samra was not suffering from a mental illness or defect that precluded him from distinguishing right from wrong and that he appreciated the wrongfulness of his conduct. In light of Samra’s history and Dr. Scott’s interactions with Samra, Dr. Scott recommended to Bell that Samra undergo “a complete neuropsychological evaluation, neurology consultation and brain imaging,” either through a SPECT or PET2 scan, or if those tests were unavailable, at least an MRI or x-ray. See Samra § 2254 Proceeding, 2014 WL 4452676, at 2 “SPECT” stands for single-photon emission computed tomography, a method for imaging the function of internal organs, including blood flow to the brain. Tests and Procedures: SPECT Scan, MAYO CLINIC (Feb. 20, 2014), http://www.mayoclinic.org/tests- procedures/spect-scan/basics/definition/prc-20020674. “PET” stands for positron emission tomography, a similar function-imaging procedure. Tests and Procedures: Positron Emission Tomography (PET) Scan, MAYO CLINIC (May 6, 2014), http://www.mayoclinic.org/tests- procedures/pet-scan/basics/definition/prc-20014301. 6 Case: 14-14869 Date Filed: 09/08/2015 Page: 7 of 42 *2.3 Dr. Scott’s report noted that his “preliminary opinion” of Samra’s culpability could change based on the results of these tests. In accordance with Dr. Scott’s recommendation, Bell obtained an MRI of Samra’s brain. It showed no structural abnormalities. Bell did not procure a SPECT or PET scan. 4 Samra § 2254 Proceeding, 2014 WL 4452676, at *2. Ultimately, Bell chose not to present Dr. Scott as a witness since he believed that Dr. Scott’s testimony would be unfavorable to Samra. Samra § 2254 Proceeding, 2014 WL 4452676, at *2. Because Bell had no evidence of brain dysfunction or mental disease and could not suppress Samra’s confession, he rested his guilt-phase trial strategy on Samra’s gang membership. As Bell explained, “Frankly . . . it appeared that that was the only thing I had.” Bell began laying the groundwork for this strategy during voir dire and in his opening statement, referring to Samra’s and Duke’s membership in the Forever Our Lord King Satan (“FOLKS”) gang. 3 Dr. Scott’s report listed as examples of types of brain imaging only MRI or PET scanning. Bell testified during the state postconviction hearing that Dr. Scott had also recommended to him obtaining a SPECT scan. The SPECT scan was also mentioned in an affidavit that Bell prepared on Dr. Scott’s behalf but was never signed by Dr. Scott. Although the state contends, correctly, that Dr. Scott’s report did not mention a SPECT scan, the record supports Samra’s claims that Dr. Scott did otherwise recommend a SPECT scan to Bell. 4 An MRI images anatomical structure, while a SPECT or PET scan images the function of organs. Magnetic Resonance Imaging (MRI), WEBMD, http://www.webmd.com/a-to-z- guides/magnetic-resonance-imaging-mri (last updated Sept. 9, 2014). 7 Case: 14-14869 Date Filed: 09/08/2015 Page: 8 of 42 During trial, the prosecution introduced photos and video recordings of the walls in Duke’s bedroom, which had various symbols and words carved into them. These etchings included Samra’s nickname, “Baby D.” At one point, the prosecutor described Duke’s room as having “gang-type writings” in it. During cross-examination of multiple police witnesses at trial, Bell emphasized the etchings and their affiliation with FOLKS and a “gang within a gang” known as the “Insane Gangster Disciples.” Photos of Samra’s tattoos were also admitted into evidence, although it is not clear if they were ever identified as gang-related. The defense called three witnesses during the guilt-phase of the trial. Dr. Kathleen Ronan, a clinical psychologist at the state’s Taylor Hardin Secure Medical Facility, testified that she had been ordered by the court to evaluate Samra’s competency to stand trial and his mental state at the time of the murders. Samra § 2254 Proceeding, 2014 WL 4452676, at *3. Dr. Ronan conducted a background interview, a mental status exam, a personality inventory, a discussion of the crime, and a trial-competency assessment. Vol. 13 at 19-20. Although Dr. Ronan did not perform a full IQ test, she testified that Samra possessed borderline intelligence based on her screening and the results of a test performed by another psychologist. She also opined that Samra displayed signs of depression and anxiety, had internal conflicts about being dependent on others, was “insecure in 8 Case: 14-14869 Date Filed: 09/08/2015 Page: 9 of 42 his interpersonal interactions,” and could become confused during periods of high stress. In addition, Dr. Ronan also testified that Samra told her he was affiliated with the FOLKS gang. The prosecutor then interrupted Bell’s questioning to point out that the gang question opened the door, allowing the prosecutor to present Dr. Ronan’s opinion that the murders had nothing to do with the gang membership, testimony the prosecutor conceded was otherwise inadmissible. Bell acknowledged that he understood that the prosecutor would ask that question, and the court also commented that it “underst[ood] all of that could be very appropriate strategy.” On cross examination, Dr. Ronan testified that she concluded Samra was competent to stand trial and that she found no evidence of any psychiatric disorder or mental disease that rendered him out of touch with reality or that impaired his sense of right and wrong. Samra § 2254 Proceeding, 2014 WL 4452676, at *3. She also testified that there was no evidence that Samra acted under duress or under the substantial domination of anyone at the time of the offense. The prosecutor asked Dr. Ronan what Samra told her about the connection between his gang membership and the murders, to which she answered that Samra said the 9 Case: 14-14869 Date Filed: 09/08/2015 Page: 10 of 42 killings had nothing to with the gang.5 Dr. Ronan also testified that Samra had a lack of emotionality when recounting the killings and that, based on the information available, she would “lean towards” that being indicative of antisocial personality disorder rather than emotional repression. The defense’s second witness was Dr. George Twente, a licensed psychiatrist. Samra § 2254 Proceeding, 2014 WL 4452676, at *3. Dr. Twente testified that he had studied the FOLKS gang, which he described as a close-knit drug-distribution organization that offered its members a sense of identity, belonging, and excitement. Dr. Twente also testified that to rise in the ranks of the gang, a member had to commit various illegal activities, and to rise to the highest level of “Set King,” he had heard a rumor that the member was required to kill his own mother. Dr. Twente confirmed that the symbols found in Duke’s bedroom were similar to symbols used by other FOLKS affiliates. According to Dr. Twente, if a member wanted to leave the gang he was told that he or a family member would be killed, but in Dr. Twente’s experience, nothing bad ever actually happened to members who left the gang. Dr. Twente asserted that usually gang killings were over territory or drugs, or they were retaliation for insults. Finally, 5 Before the prosecutor asked this question, he asked for a bench conference to find out if Bell would object. Bell said he would object because he did not think that asking about Samra’s gang affiliation opened the door to Samra’s statements about whether the killings were related to that affiliation. The court overruled the objection and permitted the state to ask Dr. Ronan a single question about what Samra told her about the killings’ relation to gang membership. 10 Case: 14-14869 Date Filed: 09/08/2015 Page: 11 of 42 Dr. Twente conceded that he had never spoken with Samra nor made any determination whether these killings were in any way gang related. Samra’s final witness was Sara Woodruff, a FOLKS member and friend of Samra, Duke, Collums, and Ellison. She commented that Samra was her “least favorite” because he wasn’t “all there,” although he could carry on normal conversations with her. Woodruff testified that Duke told her about the killings. On cross examination, she stated that Samra and Duke were good friends who got along well, and the local “gang” consisted basically of the four boys and her. With respect to the night that Duke told her about the killings, Woodruff observed no indication of hostility or threat between Duke and Samra, and she thought that they were friendly to each other. She also added that Duke said that the killings were about a dispute he had with his father and that he did not tell her it had anything to do with the gang, although Duke did make her swear an oath before he discussed the killings with her. On March 16, 1998, the jury returned a verdict of guilty on the charges of capital murder. Following a half-hour recess after receiving the verdict, the court began the sentencing phase of the trial. At this point, the state confirmed that it was pursuing the eighth statutory aggravating factor under § 13A-5-49, Ala. Code—that the offense was especially heinous, atrocious, or cruel—to justify imposition of a death sentence. 11 Case: 14-14869 Date Filed: 09/08/2015 Page: 12 of 42 The evidence introduced during the guilt phase was adopted for the sentencing phase. The state called a single witness, Thomas Hunt, the father of the two girls and ex-husband of Dedra, who testified as to the impact of their murders on him. During the penalty phase, Bell’s strategy shifted towards humanizing Samra. The defense called three members of Samra’s family: his aunt, his father, and his mother. The aunt testified that Samra was a loving and non-violent child. Samra’s father testified that, as a small child, Samra was developmentally slow and suffered tremors in his hands. His father testified that Samra was a good, obedient child until about 15 or 16 when he started using marijuana. His father then recounted various legal troubles Samra had concerning marijuana, until eventually he gave Samra an ultimatum: attend rehab for the marijuana use or leave the house. Samra moved out. Samra’s father stated that he suspected Samra was hanging around “gang-type people.” His father also recalled that Samra had been in special education most of his life and eventually dropped out of high school. Finally, he noted that Samra was never really capable of expressing his emotions, but could still be loving. Samra’s mother echoed his father’s testimony about Samra’s developmental difficulties and learning disabilities and his lack of emotionality. On the same day that the penalty phase began, March 16, 1998, the jury returned a unanimous verdict recommending death. Samra § 2254 Proceeding, 12 Case: 14-14869 Date Filed: 09/08/2015 Page: 13 of 42 2014 WL 4452676, at *4. The court sentenced Samra to death on May 7, 1998. The court found that the sole aggravating circumstance proved was that the crime was especially heinous, atrocious, or cruel as compared to other capital offenses. The court determined that two of seven statutory mitigating factors existed: a lack of significant criminal history and Samra’s age (nineteen) at the time of the offense. Ala. Code § 13A-5-51. In addition, the court concluded that the following non-statutory mitigating factors existed “to some degree” and were “worthy of consideration in the weighing of mitigating circumstances and aggravating circumstances”: Samra’s age and maturity; learning difficulties and disabilities; “family history and background and caring nature of Defendant”; “the effect of gang or group involvement upon Defendant”; Samra’s cooperation and truthfulness with law enforcement; Samra’s remorse; and the existence of only a solitary aggravating factor. The court nevertheless found that “when weighed against the many mitigating circumstances, both statutory and non-statutory, the aggravating circumstance substantially outweighs the mitigating circumstances,” and imposed a sentence of death by electrocution. 13 Case: 14-14869 Date Filed: 09/08/2015 Page: 14 of 42 C. Direct Appeals Samra raised several issues on direct appeal, including a claim of ineffective assistance of trial counsel6 for failing to investigate whether Samra “suffered from any neurological or organic mental disease or defect that would have rendered him unable to appreciate the nature and quality or wrongfulness of his acts at the time of the offense.” Samra Direct Appeal, 771 So. 2d at 1119. Samra also argued that counsel was ineffective for failing to adequately prepare a defense that Samra was not guilty by reason of mental disease or defect. Id. The Alabama Court of Criminal Appeals (“ACCA”) rejected the ineffective-trial-counsel claim, finding that Samra failed to support it with evidence but that, in any event, counsel “adequately investigated the appellant’s competence and sanity.” Id. at 1120. Samra’s counsel did not raise on direct appeal the argument rejected by the trial court that Samra was entitled to advance notice of which statutory aggravating factors the state planned to rely on to support the death penalty. The ACCA affirmed the death sentence. Samra Direct Appeal, 771 So. 2d at 1121-22. After finding no plain error and agreeing that the evidence supported the death sentence, the Alabama Supreme Court affirmed. Ex parte Samra, 771 So. 2d 1122, 1122 (Ala. 2000). The United States Supreme Court denied certiorari 6 Samra was represented by Bell on appeal with respect to all of his claims except the ineffective-assistance claim. The court appointed a separate attorney to argue that claim. See Samra Direct Appeal, 771 So. 2d at 1119 n.3. 14 Case: 14-14869 Date Filed: 09/08/2015 Page: 15 of 42 on October 10, 2000, making Samra’s conviction final. Samra v. Alabama, 531 U.S. 933, 121 S. Ct. 317 (2000) (mem.). D. State Collateral Proceedings Samra filed a state petition for postconviction relief under Rule 32 of the Alabama Rules of Criminal Procedure on October 1, 2001. The petition was amended three times, with the final petition being filed on August 16, 2002. Of the claims advanced in his Rule 32 petition, the following ones are relevant here: (1) Samra was denied effective assistance of trial counsel during the penalty phase because of counsel’s “failure to adequately investigate organic brain damage/brain dysfunction.” (2) Trial counsel during the penalty phase “was ineffective for presenting mitigating evidence that was actually aggravating,” citing counsel’s repeated references to Samra’s membership in a Satanic gang. (3) Trial counsel was ineffective for “failing to adequately object to the admission of pictures of wall etchings, ‘gang-type’ writings, and tattoos on Mr. Samra’s arms.” (4) Appellate counsel was ineffective for arguing that Samra’s death sentence “violates the due process clause of the Fourteenth Amendment because the petitioner did not receive notice of the actual statutory aggravating 15 Case: 14-14869 Date Filed: 09/08/2015 Page: 16 of 42 circumstance used at the sentencing phase, until after he had already been convicted of capital murder.” The Rule 32 court held a hearing on the brain-dysfunction claim. See Samra § 2254 Proceeding, 2014 WL 4452676, at *5. During the hearing, Samra called two medical witnesses and two attorney witnesses, including Bell. Bell testified that he did not obtain a SPECT test because he was told none were available in Birmingham, Alabama, and the closest machine was located in Nashville, Tennessee. Bell also submitted an affidavit stating that he did not request that Samra be transferred out of state for SPECT or PET testing. Although Bell recognized that a SPECT scan showing abnormalities could have been used as mitigating evidence, he testified that he did not obtain further neuropsychological testing on Samra because, “The neuropsychologist that we had contacted I believe stated that he would not or could not for some reason that I really don’t know what the reasons were that he could not do the testing of our client.” In addition, Bell attested that no further neuropsychological testing was conducted because Bell “believed Dr. Scott’s psychiatric examination covered this area.” Dr. Michael Gelbort, a clinical psychologist specializing in neuropsychology, also testified at the Rule 32 hearing. Dr. Gelbort performed a neuropsychological evaluation on Samra in 2002. His evaluation established that Samra possessed a verbal IQ of 79 (11-12th percentile), a nonverbal IQ of 87 (40th 16 Case: 14-14869 Date Filed: 09/08/2015 Page: 17 of 42 percentile),7 and a full-scale IQ of 81. Dr. Gelbort gave Samra the Categories Test, a test that measures “primarily frontal lobe and whole brain functioning.” Samra made 51 errors, placing him on the “cusp” between normal and brain impaired. As for Samra’s reading and math abilities, Dr. Gelbort found them to be in the 30th and 19th percentiles, respectively. On the Trailmaking test, which tests “cognitive flexibility and processing speed or efficiency,” Samra’s scores again placed him on the “cusp” between normal and slightly impaired. Samra’s scores on a memory test were also consistent with being on the line between average and mildly impaired. On the MMPI, which tests for gross psychopatholgy, Samra showed signs of mild depression but no signs of psychosis. Based on his clinical evaluations, Dr. Gelbort opined that Samra’s brain “would not be classified as normal or typical,” and that Samra possessed “some type of brain dysfunction.” Dr. Gelbort observed that this dysfunction “ha[d] more to do with verbally mediated skills as opposed to nonverbal or visual spatial skills.” Dr. Gelbort localized the dysfunction to the left side and frontal lobe, although he noted that there was no “focal damage” but “rather a diffuse pattern of dysfunction.” Ultimately, Dr. Gelbort concluded that Samra would have a more difficult time functioning than a normal person but that he was not “grossly impaired.” Dr. Gelbort also testified that he reviewed the testing results that the 7 This test was incomplete because the prison would not allow Dr. Gelbort to bring in certain equipment. 17 Case: 14-14869 Date Filed: 09/08/2015 Page: 18 of 42 state’s expert, Dr. Glen King, had obtained and found those to be consistent with his own testing. Besides these witnesses, Samra presented the testimony of Dr. James Mountz, a specialist in nuclear medicine and radiology. Between 1991 and 2003, Dr. Mountz was at the University of Alabama—Birmingham (“UAB”) where he “built up” the “functional brain imaging protocols.” Dr. Mountz testified that a dual-head camera SPECT machine was available at UAB at least by the time he arrived there in late 1990, and possibly as early as 1988. A PET scan was not available at UAB, however, until July 2001. Dr. Mountz testified that a SPECT scan, as relevant to this case, measures blood flow in the brain. As Dr. Mountz explained SPECT scans, a SPECT scan does not distinguish between a normal and an abnormal brain but rather between normal and abnormal blood flow. Dr. Mountz also distinguished a SPECT scan from an MRI, in that an MRI measures anatomical structure while the SPECT scan measures blood flow to those structures; “function as opposed to structure.” He further explained that an MRI may show a normal structure, but a decreased blood flow may lead to abnormal brain functioning. Dr. Mountz conducted a SPECT scan on Samra on August 1, 2002. From the scan, Dr. Mountz concluded that “[t]here were areas of decreased blood flow which were abnormal.” Dr. Mountz testified to two basic abnormalities: one that 18 Case: 14-14869 Date Filed: 09/08/2015 Page: 19 of 42 “falls into the milder category” and could just constitute normal variability, and a second “obvious abnormality in blood flow to the posterior frontal superior temporal region” of the right side of Samra’s brain. Dr. Mountz did not reach an opinion about Samra’s brain function but did opine that his brain blood flow was about one standard deviation below normal. He later characterized the blood flow as “low normal.” In a carefully worded answer, Dr. Mountz testified that the “abnormalities found in [Gelbort’s neuropsychological report] are not inconsistent with the brain SPECT scan” conducted on Samra. On cross examination, Dr. Mountz testified that a SPECT scan does not provide any insight into why a person commits murder or whether he can appreciate the wrongfulness of his conduct. He also testified that a more detailed analysis would be required to determine what functions the abnormal area controls in Samra’s specific case. Finally, Dr. Mountz also conceded that Samra’s brain scan could have looked much different in 1998. The state called two witnesses. The first, Dr. King, a clinical psychologist and attorney, conducted a neuropsychological evaluation of Samra. Dr. King performed an achievement test with Samra, where he scored Samra as reading at an eighth-grade level, spelling at a high-school level, and completing arithmetic at a sixth-grade level. According to Dr. King, an individual who has dropped out of school can be expected to score lower on these tests, as would an individual with a low IQ. 19 Case: 14-14869 Date Filed: 09/08/2015 Page: 20 of 42 Dr. King administered an outdated IQ test, but the corrected results placed Samra’s full-scale IQ in the high borderline range of around 79 and his verbal and nonverbal IQ scores were consistent with what Dr. Gelbort found. Samra performed well on some perceptual tests but poorly on others. His nondominant hand was weaker than expected, indicating “some lateralizing effect,” and his fine motor control was “below the cutoff.” Dr. King also concluded that Samra had some impairment in his visual-spatial area. Dr. King also administered the MMPI test. Consistent with Dr. Gelbort’s MMPI, Dr. King found that Samra suffered from mild depression and anxiety and immature interpersonal development but not any psychosis. Besides these tests, Dr. King administered the Categories Test, and Samra committed 52 errors, consistent with Dr. Gelbort’s testing and finding of mild impairment. Similarly, Samra scored slightly impaired on Dr. King’s Trailmaking test. As a result of his examinations, Dr. King concluded that Samra suffers from some impairment in his cognitive functioning. He further opined that at the time of the offense, Samra did not have any serious mental illness or mental defect that would have rendered him incapable of understanding the nature and consequences of his actions. While acknowledging a debate among experts about whether someone with borderline or retarded intellectual ability suffers from brain impairment, Dr. King ultimately concluded that Samra “has impairment of 20 Case: 14-14869 Date Filed: 09/08/2015 Page: 21 of 42 functioning that is consistent with what we would expect in someone who is in the borderline range of intellectual ability” but that this impairment did not impact Samra’s ability to appreciate the criminality of his conduct. Dr. Helen Mayberg, a neurologist, also testified for the state. She stated that the “generally accepted clinical uses of a SPECT technology are extremely limited,” and include diagnosing strokes, evaluating dementia, and identifying abnormalities associated with epilepsy, and identifying abnormalities following trauma. Like with any radiological procedure, Dr. Mayber explained, experience in evaluating SPECT scans is important in determining when a scan is “normal.” Dr. Mayberg reviewed Dr. Mountz’s report and found that it contained the typical elements of a SPECT report. According to Dr. Mayberg, though, one standard deviation was not abnormal but rather, still fell within the normal range. In Dr. Mayberg’s view, “brain damage” is too generalized a term, and the SPECT scan is too sensitive to blood-flow variation to make a good screening test for brain damage. Dr. Mayberg also reviewed Samra’s childhood records and speculated that he may have had a neurological problem as a child that improved over time. Finally, she agreed that Samra’s MRI from 1998 was normal. On January 12, 2005, the trial court denied Samra’s Rule 32 petition. See Samra § 2254 Proceeding, 2014 WL 4452676, at *9. With regard to the brain- dysfunction claim, the court held that Bell’s performance was not deficient as far 21 Case: 14-14869 Date Filed: 09/08/2015 Page: 22 of 42 as investigating brain damage because he had pursued some investigation and had settled on another strategy. The court also determined that Samra suffered no prejudice because the results of Dr. Gelbort’s and Dr. Mountz’s testing only affirmed the information of Samra’s borderline intellectual abilities already before the jury. The court also rejected the idea that Samra suffered from any organic brain dysfunction. Id. Nonetheless, the court added that even if Samra had reduced blood flow to his brain at the time of the offense, no evidence established that it would have affected Samra’s culpability, judgment, or insight. Further elaborating on prejudice, the trial court determined that no reasonable probability existed that the jury would have reached a different recommendation if Samra had presented Dr. Gelbort’s and Dr. Mountz’s evidence during the penalty phase of Samra’s trial. With respect to Samra’s argument that the gang evidence was more aggravating than mitigating, the Rule 32 court found that Bell’s strategy to portray Samra as gang-influenced was reasonable in light of the fact that Bell had no evidence of any other mental defect or prior history of violence. The court determined that Samra’s argument about Bell’s failure to object to the admission of gang etchings and tattoos was “wholly without merit” because the evidence complemented Bell’s own defense strategy. Apparently, the Rule 32 court did not reach the prejudice prong of the ineffective-assistance analysis with respect to 22 Case: 14-14869 Date Filed: 09/08/2015 Page: 23 of 42 these gang-related claims. The Rule 32 court also did not address the aspect of Samra’s due-process claim regarding notice of the aggravating factors; instead, it dealt with Samra’s related argument that Alabama’s sentencing procedure violated due process because it allows the judge to impose the sentence. On August 24, 2007, the ACCA affirmed the denial of postconviction relief. With respect to Samra’s due-process claim that he had not received notice of the aggravating factors, the ACCA determined that the claim failed on the merits because neither Apprendi v. New Jersey, 530 U.S. 466, 120 S. Ct. 2348 (2000)— decided before Samra’s conviction became final—nor Ring v. Arizona, 536 U.S. 584, 122 S. Ct. 2428 (2002)—decided after—“modified prior Alabama caselaw, ‘which holds that aggravating circumstances do not have to be alleged in the indictment.’” And to the extent that Ring could be read to support Samra’s argument, the ACCA noted that it was decided two years after his conviction became final and that appellate counsel could not have been ineffective for failing to anticipate changes in the law. The ACCA also affirmed the Rule 32 court’s ruling on the brain-dysfunction claim, finding that Bell employed a “well thought-out defense strategy,” and determining that Bell had no cause to investigate Samra’s organic brain function any further than he did. As for the evidence provided by Dr. Gelbort and Dr. Mountz, the ACCA concluded that it was not compelling and that, in any event, 23 Case: 14-14869 Date Filed: 09/08/2015 Page: 24 of 42 this additional evidence would not have “influenced the jury’s appraisal of Samra’s moral culpability.” After quoting the trial court’s discussions of whether the evidence was more aggravating than mitigating and whether Bell was ineffective for not objecting to the gang writings, the ACCA adopted and affirmed those conclusions without discussion. Following the ACCA’s affirmance of the denial of Samra’s Rule 32 petition, the Alabama Supreme Court denied certiorari on September 19, 2008. See Ex parte Samra, 34 So. 3d 737, 737 (Ala. 2008) (table decision).8 E. Federal Habeas Proceedings Samra filed a petition for habeas corpus under 28 U.S.C. § 2254 in the Northern District of Alabama on October 26, 2007, and an amended petition on February 21, 2014. See Samra § 2254 Proceeding, 2014 WL 4452676, at *10. The district court denied Samra’s § 2254 petition on September 5, 2014. Id. at *46. Applying the doubly deferential standard of Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052 (1984), and § 2254 to Samra’s ineffective-assistance-of- counsel claims, the district court determined that the ACCA did not misapply Strickland when it held that Bell’s investigation of Samra’s brain condition was not 8 After Duke’s death sentence was vacated in light of the United States Supreme Court’s decision in Roper v. Simmons, 543 U.S. 551, 125 S. Ct. 1183 (2005), because Duke was under eighteen at the time of the murders, Samra filed a successive Rule 32 petition arguing that his death sentence should be set aside because, essentially, it was unjust and disproportionate to execute Samra, who, although nineteen years old at the time of the offense, was a less culpable party than Duke. The state courts rejected this argument. The issues raised in Samra’s successive Rule 32 petition are not before us. 24 Case: 14-14869 Date Filed: 09/08/2015 Page: 25 of 42 deficient. Id. at *27. The district court noted that Bell had obtained Dr. Scott’s evaluation and an MRI, and he had investigated Samra’s background (including his tremors) and interacted with Samra. Id. Because nothing uncovered by Bell would have raised “red flags” as to an organic brain dysfunction, the district court deemed Bell’s investigation to be extensive and not deficient. Id. at *28-29. The district court also distinguished Samra’s reliance on the Fifth Circuit case of Lockett v. Anderson, 230 F.3d 695 (5th Cir. 2000). After observing that was it not a Supreme Court precedent, the district court concluded that “Samra has not presented any evidence that he has a mental illness or any connection between his purported ‘organic brain damage’ and his participation in these murders.” Id. at *29. Nor did the district court find that the prejudice determination was contrary to law. As the district court viewed the record, the murders were heinous, the SPECT test was not trustworthy, and all of the Rule 32 testing was merely cumulative of the evidence of Samra’s low IQ. Id. at *30-31. The district court similarly found Bell’s decision to emphasize Samra’s membership in the FOLKS gang as unassailable because Bell reached that decision after concluding that Samra had no mental defect or illness that would serve as a defense. Id. at *34. With regard to Bell’s presentation of Dr. Ronan’s testimony, the district court saw no error because her overall conclusion supported Bell’s defense theory, even if her testimony about the lack of gang involvement did not. 25 Case: 14-14869 Date Filed: 09/08/2015 Page: 26 of 42 Id. at *35. The district court likewise determined that Bell’s failure to object to the photographic gang-related evidence was in keeping with his defense strategy, so the state court did not misapply Strickland when it rejected Samra’s argument. Id. at *37. The district court did not discuss the prejudice prong of Strickland with respect to the gang-related claims. See id. at *31-37. Turning to Samra’s ineffective-assistance-of-appellate-counsel claim, the district court concluded that the ACCA “mischaracterized” the claim as a defective indictment issue rather than a notice issue. See id. at *38. And because the state courts did not address the merits of the notice claim, the district court determined that de novo review was appropriate rather than deferential review under § 2254. Id. at *39. Even applying de novo review, though, the district court determined that Samra’s claim failed. Id. at *39-40. The district court held that due process requires that a defendant receive notice of the charges against him only, not notice of the statutory aggravating factors that the state intended to use to justify a death sentence. Id. at *40. Because no due-process right existed, the district court reasoned, Bell was not deficient for failing to raise the argument on appeal. Id. Having concluded that Bell’s performance was not deficient, the court did not further address the prejudice prong of this inquiry. See id. 26 Case: 14-14869 Date Filed: 09/08/2015 Page: 27 of 42 Following the district court’s rejection of his federal habeas petition, Samra sought appellate review. We granted a certificate of appealability with respect to two issues: (1) Did the Alabama courts unreasonably apply Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052 (1984), when they determined that Samra’s trial counsel was not ineffective when trial counsel (a) failed to investigate and present evidence of brain impairment in mitigation of the death penalty; (b) introduced evidence during trial of Samra’s affiliation with a Satanic gang; and (c) did not object to the introduction at trial of Satanic markings found in his co-defendant’s bedroom and Samra’s own gang tattoos. (2) As to the claim that appellate counsel rendered ineffective assistance by failing to raise the issue of whether due process requires pretrial notice to a capital defendant of the specific statutory aggravating circumstances that the State intends to rely on in seeking a death sentence: (a) Is any component of this claim barred by the Teague 9 non-retroactivity doctrine? (b) Is 28 U.S.C. § 2254(d) deference due on any component of this claim? (c) Does due process require pre-trial notice to a capital defendant of which specific statutory aggravating circumstances the State intends to rely on in seeking a death sentence? (d) Was it ineffective assistance for the petitioner’s appellate counsel not to raise this issue on direct appeal? 9 Teague v. Lane, 489 U.S. 288, 109 S. Ct. 1060 (1989). 27 Case: 14-14869 Date Filed: 09/08/2015 Page: 28 of 42 II. A. General Habeas Standards Federal law permits a prisoner held “in custody pursuant to the judgment of a State court” to seek habeas relief “only on the ground that he is in custody in violation of the Constitution or laws or treaties of the United States.” 28 U.S.C. § 2254(a). Generally, a prisoner must first “fairly present” his federal claims to the state court and exhaust his state-court remedies before seeking federal habeas relief. Snowden v. Singletary, 135 F.3d 732, 735 (11th Cir. 1998). We review a district court’s denial of a § 2254 petition de novo. Sims v. Singletary, 155 F.3d 1297, 1304 (11th Cir. 1998). If the state courts do not address the merits of a fairly presented claim, a federal court’s review of that claim is de novo. See Davis v. Sec’y for the Dep’t of Corr., 341 F.3d 1310, 1313 (11th Cir. 2003) (per curiam). But when a state court has adjudicated a prisoner’s claim on the merits, a federal court may not grant habeas relief with respect to such a claim unless the state court’s adjudication (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). 28 Case: 14-14869 Date Filed: 09/08/2015 Page: 29 of 42 These standards are highly deferential and demand that we give state-court decisions the benefit of the doubt. Evans v. Sec’y, Dep’t of Corr., 703 F.3d 1316, 1325 (11th Cir. 2013) (en banc). A decision “is not ‘contrary to’ federal law unless it ‘contradicts the United States Supreme Court on a settled question of law or holds differently than did that Court on a set of materially indistinguishable facts.’” Id. (quoting Cummings v. Sec’y for Dep’t of Corr., 588 F.3d 1331, 1355 (11th Cir. 2009)). Nor is a state court’s decision “an ‘unreasonable application’ of federal law unless the state court ‘identifies the correct governing legal principle as articulated by the United States Supreme Court, but unreasonably applies that principle to the facts of the petitioner’s case, unreasonably extends the principle to a new context where it should not apply, or unreasonably refuses to extend it to a new context where it should apply.’” Id. The federal court does not ask whether the state decision is correct, but rather whether it is unreasonable. Id. B. Ineffective Assistance of Counsel To prevail under Strickland on a claim of ineffective assistance of trial counsel, a petitioner must show that (1) counsel’s performance was so deficient that “counsel was not functioning as the ‘counsel’ guaranteed” by the Sixth Amendment and (2) that counsel’s performance prejudiced the defense to the extent that the defendant was deprived of a fair, reliable trial. Strickland, 466 U.S. at 687, 104 S. Ct. at 2064. A court need not conduct this analysis in a particular 29 Case: 14-14869 Date Filed: 09/08/2015 Page: 30 of 42 sequence, and a court need not address both prongs if a petitioner fails to make a required showing on one of them. Id. at 697, 104 S. Ct. at 2069. Establishing deficient performance requires the petitioner to demonstrate that “‘counsel’s representation fell below an objective standard of reasonableness.’” Harrington v. Richter, 562 U.S. 86, 104, 131 S. Ct. 770, 787 (2011) (quoting Strickland, 466 U.S. at 688, 104 S. Ct. at 2064). A court applies a strong presumption that counsel’s representation fell within the wide range of reasonable professional conduct. Id. To show prejudice, a petitioner “must demonstrate ‘a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.’” Id. (quoting Strickland, 466 U.S. at 694, 104 S. Ct. at 2068). We evaluate claims of ineffective assistance of appellate counsel under the same Strickland standards. Philmore v. McNeil, 575 F.3d 1251, 1264 (11th Cir. 2009) (per curiam). III. Samra contends that he is entitled to federal habeas relief because his trial counsel was ineffective. Specifically relevant to this appeal, he asserts that his counsel failed to adequately investigate and present evidence of brain dysfunction in mitigation of the death penalty. He also contends that Bell’s strategy of emphasizing Samra’s involvement in a satanic gang, including Bell’s failure to 30 Case: 14-14869 Date Filed: 09/08/2015 Page: 31 of 42 object to evidence of certain gang-related drawings and tattoos was more aggravating than mitigating. For the reasons discussed in this section, we find that Samra has failed to establish prejudice with respect to his ineffective-trial-counsel claim. And because he has not shown prejudice, we neither reach nor offer any opinion on trial counsel’s performance. See Strickland, 466 U.S. at 697, 104 S. Ct. at 2069. A. Bell’s Investigation of Samra’s Brain Function Samra argues that the ACCA unreasonably applied federal law when it determined (a) that Bell’s investigation of Samra’s neuropsychological health was not deficient and (b) that the failure to introduce the evidence developed by Dr. Gelbort and Dr. Mountz during the Rule 32 proceedings did not prejudice Samra. With respect to the deficient-performance prong of the Strickland analysis, Samra contends that Bell was deficient for not pursuing functional brain testing, particularly a SPECT test, and also for not pursuing further neuropsychological testing when Bell’s own expert, Dr. Scott, had recommended those tests. Samra asserts that the failure to investigate prejudiced him because the postconviction evidence establishes that Samra does suffer from organic brain dysfunction and that such evidence is powerfully mitigating, thus undermining confidence in the death sentence. 31 Case: 14-14869 Date Filed: 09/08/2015 Page: 32 of 42 The state counters that Bell’s investigation of Samra’s brain function was adequate. In the state’s view, moreover, Samra was not prejudiced because the new evidence is cumulative and consistent with the evidence presented of Samra’s borderline intellectual ability, does not undermine Samra’s culpability, and would have had no chance of altering the jury’s balance of aggravating and mitigating factors in light of the brutal nature of the killings. We agree with the state that Samra has failed to establish that the state courts unreasonably applied Strickland when they found no prejudice. As we have noted, to demonstrate prejudice under Strickland, Samra must “show that there 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, 104 S. Ct. at 2068. “In assessing prejudice, we reweigh the evidence in aggravation against the totality of available mitigating evidence,” including all mitigating evidence produced at trial and developed during the collateral proceedings. Wiggins v. Smith, 539 U.S. 510, 534, 123 S. Ct. 2527, 2542 (2003); Williams v. Taylor, 529 U.S. 362, 397-98, 120 S. Ct. 1495, 1515 (2000). As described above, Dr. Gelbort’s examination, which was largely consistent with Dr. King’s examination, suggested some mild impairment of Samra’s brain’s functioning. But Dr. Mountz’s SPECT scan indicated just a different blood flow in one region of Samra’s brain approximately one standard 32 Case: 14-14869 Date Filed: 09/08/2015 Page: 33 of 42 deviation below the “normal” level of flow. And while Samra emphasizes what he describes as the abnormal nature of this blood flow, Dr. Mayberg opined that this result still fell within “normal” levels. Nor does Samra present any argument or evidence that his blood-flow level has had any cognitive or behavioral impact. In other words, there is no evidence that the blood flow level was so much an organic brain problem, as opposed to merely an organic brain anomaly, to the extent that it was even anomalous. Even Dr. Mountz testified, carefully, that the abnormal blood flow was “not inconsistent” with Dr. Gelbort’s diagnosis of mild impairment, not that the blood flow itself was indicative of impairment. And Dr. Mountz added that further analysis would be required to determine what the section of Samra’s brain with the different blood flow actually controls. While we recognize, of course, that any abnormality is admissible mitigation evidence, even assuming that the blood flow was abnormal, in the absence of any evidence explaining its effect on Samra, its mitigating impact is significantly reduced. Samra cites a number of cases that stand for the proposition that organic brain damage can be a significant mitigating factor in capital sentencing proceedings. And we agree and have recognized in the past that evidence of organic brain damage can be a powerful mitigating factor. See, e.g., Debruce v. Comm’r, Ala. Dep’t of Corr., 758 F.3d 1263, 1276 (11th Cir. 2014); Ferrell v. Hall, 640 F.3d 1199, 1234-35 & n.17 (11th Cir. 2011). But in those cases, the 33 Case: 14-14869 Date Filed: 09/08/2015 Page: 34 of 42 evidence established the existence of impairment—indeed, significant brain impairment. See Debruce, 758 F.3d at 1270 (petitioner suffered from “lingering emotional damage and social impairments associated with having been raised in a violent community,” as well as “blackout episodes consistent with seizures accompanied by periods of non-responsive staring and loss of memory”); Ferrell, 640 F.3d at 1203, 1234 (petitioner suffered from “extensive” and “disabling . . . organic brain damage to the frontal lobe, bipolar disorder, and temporal lobe epilepsy”). Even Lockett, the Fifth Circuit case upon which Samra heavily relies, dealt with a defendant who likely suffered from temporal-lobe epilepsy and paranoid schizophrenia. 230 F.3d at 713-14. Unlike the cases he cites, Samra possesses mild functional impairments and, based on his MRI scan, suffers from no structural brain abnormalities. Contrary to Samra’s argument on appeal that his counsel presented a paucity of mitigating evidence at his trial, the sentencing jury heard, and the sentencing judge found, several mitigating factors related to Samra’s mental health, including his borderline intelligence, schooling problems, substance abuse, childhood hand tremors, and lack of emotionality. And unlike in cases such as Wiggins, Debruce, or Ferrell, there is absolutely no evidence in Samra’s case of an abusive upbringing that could have contributed to a mental disorder. 34 Case: 14-14869 Date Filed: 09/08/2015 Page: 35 of 42 Ultimately, after weighing the evidence adduced in the postconviction proceedings from Dr. Gelbort and Dr. Mountz along with the evidence introduced during the guilt and sentencing phases, the state court concluded that Samra had not established a reasonable probability that this mitigating evidence undermines confidence in his unanimous death sentence. This was not an unreasonable application of Strickland’s prejudice analysis. Harrington, 562 U.S. at 100-01, 131 S. Ct. at 785. The fact remains that Samra participated in the gruesome murders of four people—including personally slitting the throat of a seven-year-old girl, causing her to drown in her own blood—all because his friend Duke’s father refused to let Duke use a pickup truck. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. None of the brain-impairment evidence Samra has provided leads us to conclude that a jury would have found that the killings were less heinous, atrocious, or cruel, or that Samra’s mild brain impairments outweighed the heinousness, atrociousness, or cruelty of the crime, had evidence similar to that presented at the Rule 32 hearing been introduced during the penalty phase of Samra’s trial. Absent the required showing of prejudice, we do not consider the sufficiency of Bell’s investigation. Samra’s claim that his trial counsel was ineffective for failing to adequately investigate Samra’s brain function must be denied. 35 Case: 14-14869 Date Filed: 09/08/2015 Page: 36 of 42 B. Bell’s Gang-Influence Strategy Samra contends that Bell’s strategy of emphasizing Samra’s membership in the FOLKS gang—including its connection to Satan, its violent character (including alleged matricide), and its association with drugs and prison—put evidence that was more aggravating than mitigating in front of the jury. In Samra’s view, Bell was deficient for choosing and presenting this strategy and for failing to object to the prosecution’s introduction of other photographic evidence of FOLKS-related markings found in Duke’s room and Samra’s tattoos. For its part, the state contends that Bell’s strategic decision to present a “substantial domination” defense is unassailable. While neither the state courts nor the district court reached the prejudice prong of this analysis, we find under a de novo review that, regardless of the competency of Bell’s chosen strategy, Samra has failed to establish a reasonable probability that it undermines confidence in the death sentence. We acknowledge, as we have previously, that evidence of gang membership and satanic worship has the potential to unduly prejudice a defendant. See United States v. Jernigan, 341 F.3d 1273, 1284-85 (11th Cir. 2003) (“[W]e do not wish to understate the prejudicial effect that evidence of a criminal defendant’s gang membership may entail. Indeed, modern American street gangs are popularly associated with a wealth of criminal behavior and social ills, and an individual’s membership in such 36 Case: 14-14869 Date Filed: 09/08/2015 Page: 37 of 42 an organization is likely to provoke strong antipathy in a jury.”); cf. McCorkle v. Johnson, 881 F.2d 993, 995 (11th Cir. 1989) (per curiam) (recognizing the “violence inherent in Satan worship”). But the facts of this crime belie this prejudice here. Even if we disregard the gang-related evidence and argument, the state presented overwhelming evidence— including Samra’s own confession—of the heinousness of this crime. By Samra’s own admission, after he assisted in killing three people, he slit the throat of a seven-year-old girl who was pleading and struggling for her life. Samra § 2254 Proceeding, 2014 WL 4452676, at *1. We find no reasonable probability that, absent evidence or discussion of Samra’s gang involvement, the jury would not have found these murders to be as especially heinous, atrocious, or cruel as it found them. Harrington, 562 U.S. at 104, 131 S. Ct. at 787. As a result, Samra’s claim that his trial counsel was ineffective for pursing a gang-related strategy and for failing to object to gang-related evidence must be denied.10 IV. Samra also contends that he is entitled to federal habeas relief because his appellate counsel was constitutionally ineffective. Specifically, Samra argues that he had a due-process right to be informed before his trial of the actual aggravating 10 As with the brain-dysfunction claim, we do not opine on whether counsel’s selection and implementation of this defense strategy was constitutionally sufficient. See Strickland, 466 U.S. at 697, 104 S. Ct. at 2069. 37 Case: 14-14869 Date Filed: 09/08/2015 Page: 38 of 42 factor or factors upon which the state intended to rely in seeking the death penalty and that his appellate counsel was deficient for not challenging the trial court’s rejection of this argument. Without deciding whether such a due-process right exists, we find that Samra’s counsel was not deficient in failing to raise the issue during Samra’s direct appeals. 11 Samra bases his due-process notice argument on two Supreme Court cases that were decided before his conviction became final: Jones v. United States, 526 U.S. 227, 119 S. Ct. 1215 (1999) and Apprendi v. New Jersey, 530 U.S. 466, 120 S. Ct. 2348 (2000). In Jones, the Supreme Court construed the federal carjacking statute, 18 U.S.C. § 2119, and concluded that the statutory provisions that enhanced a sentence based on a finding of bodily injury or death should not be viewed merely as sentencing factors but as elements of distinct offenses that must be charged in an indictment. See 526 U.S. at 251-52, 119 S. Ct. at 1228. Similarly, in Apprendi, the Court confirmed the principle expressed in Jones: “Other than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury, and proved beyond a reasonable doubt.” Apprendi, 530 U.S. at 490, 120 S. Ct. at 2362-63. Under Apprendi, sentencing factors that increase punishment beyond the 11 The state contends that the district court erred in considering this claim de novo after determining that the Alabama state courts had mischaracterized Samra’s claim. Instead, the state urges that we extend double deference under § 2254 and Strickland to the manner in which the state courts did decide this claim. We need not resolve this issue, however, because even under de novo review, Samra has failed to demonstrate deficient performance. 38 Case: 14-14869 Date Filed: 09/08/2015 Page: 39 of 42 statutory maximum are “the functional equivalent of an element of a greater offense than the one covered by the jury’s guilty verdict.” Id. at 494 n.19, 120 S. Ct. at 2365 n.19 (emphasis added). Because the statutory aggravating factors are required under Alabama law to increase the maximum punishment from life imprisonment to death, in Samra’s view, they constitute elements of the capital offense. And because due process generally requires advance notice of offense elements, Samra argues that he had a constitutional entitlement to advance notice of the specific aggravating factors that the state was pursuing in his case. Samra asserts that the performance of his appellate counsel was constitutionally deficient because counsel neglected to challenge the trial court’s refusal to provide this notice. But Samra cannot demonstrate that his appellate counsel was constitutionally deficient for failing to raise this due-process argument. We judge counsel’s performance from the perspective of an attorney operating at the time that the challenged decision was made. Strickland, 466 U.S. at 689, 104 S. Ct. at 2065. Here, Samra has cited no precedent—and we have found none—existing at the time of his direct appeal that required a state to provide advance notice of the specific aggravating factor it intended to prove. In fact, in our view, the circumstances existing at that time actually counseled that such an argument lacked merit. 39 Case: 14-14869 Date Filed: 09/08/2015 Page: 40 of 42 First, the Supreme Court had granted certiorari in Jones just a month before Samra filed his notice of direct appeal, and it did not decide Apprendi until after the Alabama Supreme Court upheld Samra’s death sentence. See Jones v. United States, 523 U.S. 1058, 118 S. Ct. 1405 (1998) (Mem.) (amending the grant of certiorari to two specific questions); Apprendi, 530 U.S. at 466, 120 S. Ct. at 2348; Ex parte Samra, 771 So. 2d at 1122. Samra suggests that his appellate counsel should have recognized from the grant of certiorari in Jones—a case involving the federal carjacking statute—that the issue of whether a state is required to provide capital defendants with advance notice of specific aggravating factors was an unresolved question of law. Although an exceptionally skilled or creative attorney may have anticipated the arguments and outcome of Jones (and later Apprendi) and sought to extend that rationale to the capital-sentencing context, we have repeatedly held that an attorney is not required to foresee changes in the law to provide constitutionally sufficient representation. See, e.g., LeCroy v. Sec’y, Fla. Dep’t of Corr., 421 F.3d 1237, 1261 n.27 (11th Cir. 2005) (“[A]ppellate counsel was not ineffective for failing to anticipate the change in the law”); Spaziano v. Singletary, 36 F.3d 1028, 1039 (11th Cir. 1994) (“We have held many times that ‘[r]easonably effective representation cannot and does not include a requirement to make arguments based on predictions of how the law may develop.’” (quoting 40 Case: 14-14869 Date Filed: 09/08/2015 Page: 41 of 42 Elledge v. Dugger, 823 F.2d 1439, 1443 (11th Cir.) (per curiam), modified in unrelated part, 833 F.2d 250, 250 (11th Cir. 1987) (per curiam)). Second, a substantial body of federal and state case law available to his appellate counsel at the time, while not necessarily foreclosing Samra’s due- process argument, strongly suggested such an argument would face an uphill battle with little chance of success. See, e.g., Walton v. Arizona, 497 U.S. 639, 647-49, 110 S. Ct. 3047, 3054-55 (1990) (expressly holding that capital aggravating factors are not elements of the offense, a holding that was specifically affirmed in Jones and Apprendi and not overruled until Ring), overruled by Ring, 536 U.S. at 589, 122 S. Ct. at 2432; Spenkelink v. Wainwright, 442 U.S. 1301, 1303-06, 99 S. Ct. 2091, 2092-94 (Rehnquist, Circuit Justice 1979) (denying stay of execution and asserting belief that no four Justices would agree to hear a claim that due process required advance notice of capital aggravating factors); Clark v. Dugger, 834 F.2d 1561, 1566 (11th Cir. 1987) (due process satisfied by statute listing potential aggravating factors and particularized notice of specific factors is not required); Knotts v. State, 686 So. 2d 431, 448-449 (Ala. Crim. App. 1995) (“A defendant has no right to advance notice of the state's intention to rely on any of the aggravating circumstances enumerated in § 13A–5–49.”). In light of this case law existing at the time that Samra’s appeal was filed, an attorney could not be faulted for declining to pursue the argument that capital 41 Case: 14-14869 Date Filed: 09/08/2015 Page: 42 of 42 sentencing factors were functionally equivalent to the elements of an offense and were required by due process to be disclosed in advance of trial. Philmore, 575 F.3d at 1264 (“In assessing an appellate attorney’s performance, we are mindful that ‘the Sixth Amendment does not require appellate advocates to raise every non- frivolous issue.’ Rather, an effective attorney will weed out weaker arguments, even though they may have merit.” (citation omitted) (quoting Heath v. Jones, 941 F.2d 1126, 1130-31 (11th Cir. 1991)). Thus, even if counsel had foreseen such an argument—and he was not required to have done so—counsel would not have been deficient in declining to pursue that argument. Because Samra’s appellate counsel’s performance was not deficient, Samra cannot prevail on his ineffective- appellate-counsel claim here. We therefore do not address the prejudice prong and the underlying merits of Samra’s due-process argument. See Strickland, 466 U.S. at 697, 104 S. Ct. at 2069; Philmore, 575 F.3d at 1264-65. V. For the reasons set forth above, we find that Samra has failed to make the required showings to prevail on his ineffective-assistance-of-counsel claims. Accordingly, we affirm the district court’s denial of Samra’s § 2254 petition. AFFIRMED. 42