Williams v. Quarterman

293 F. App'x 298

• Court: Court of Appeals for the Fifth Circuit
• Decided: September 19, 2008
• Docket: 07-70006
• Status: Unpublished

Opinion

PER CURIAM: *

Petitioner-Appellant Jeffrey Demond Williams (“Williams”) appeals the district court’s denial of his petition for a writ of habeas corpus. For the following reasons, we affirm the district court’s denial of Williams’s habeas petition and remand for the district court to consider whether to grant a certificate of appealability (“COA”) on Williams’s post-judgment motions.

I. FACTUAL BACKGROUND AND PROCEDURAL HISTORY

A. Factual Background

The district court set forth the relevant facts in this case:

On the night of May 19, 1999, Officer Tony Blando drove around the parking lot of the Roadrunner Inn [in Houston, Texas] looking for stolen vehicles. Blando was in plainclothes and drove an unmarked car. Blando saw Williams driving a Lexus. He inquired about the car [via a computer search in his car], and learned that the car was stolen in an aggravated robbery. As Williams stepped out of the car, Blando approached him with his weapon drawn. Trial testimony explained that it was departmental practice for officers to draw their weapons when approaching a suspect in a stolen vehicle. Blando and Williams tussled, and Williams pulled out a gun and fatally shot Blando.

Two witnesses also testified that they saw the shooting. One was staying at the Roadrunner Inn, and the other at a hotel next door. One of the witnesses testified that she heard two men arguing in the parking lot. When she looked out the window, she saw a white man and a black man arguing, and heard the black man telling the white man not to handcuff him. After the white man got one handcuff on the black man, the black man spun around and shot the white man. Williams was arrested near the scene a short while later. He was wearing one handcuff at the time.

After his arrest, Houston Police Officer Alen Brown read Williams his Miranda warnings and interviewed him. Williams acknowledged that he understood his rights and gave two statements in which he admitted shooting Blando. At trial the medical examiner testified that the cause of Blando’s death was a perforating gunshot wound to the chest.

Police recovered shell casings fired from .380 caliber, .40 caliber, and 9 millimeter weapons. Blando fired a .380. At least some of the other bullets were fired by Williams’ gun. Williams’ fingerprints were on the Lexus and on Blando’s Jeep Cherokee.

The defense presented no witnesses during the guilt-innocence phase. In his statement to the police, however, Williams contended that he did not know Blando was a police officer, but thought that Blando was trying to rob him, and shot in self-defense. One of the officers who responded to Blando’s call for help, however, testified that Blando was wearing his badge around his neck. The jury found Williams guilty of capital murder for the murder of Officer Blando.

During the penalty phase the State presented the testimony of Jennifer Null, the owner of the stolen Lexus. On May 10, 1999, nine days before the shooting, the car was stolen from Null at gunpoint. The thief attempted to force Null into the car, but she refused. Officer Allen Brown testified that Williams admitted to committing several robberies and expressed no remorse over the Blando shooting. Two witnesses also identified Williams as the gunman in a robbery-shooting in January of 1999. The victim of that shooting was Ezzard McCowan. The same gun was used in the McCowan and Blando shootings. Blando’s wife offered victim impact testimony and testified that Blando was a Boy Scout leader.

(citations omitted).

B. State Court Proceedings

A Houston jury found Williams guilty of capital murder and concluded that he posed a future danger to commit criminal acts of violence that would constitute a continuing threat to society. The jury also found insufficient mitigating evidence to warrant a life sentence. The trial court therefore sentenced Williams to death. The Texas Court of Criminal Appeals (“TCCA”) affirmed the conviction and sentence, Williams v. State, No. 73,796 (Tex.Crim.App.2002), and denied Williams’s application for postconviction relief, Ex Parte Williams, No. 50,662-01 (Tex.Crim.App.2003).

On June 17, 2003, Williams filed a successive state habeas petition, claiming that he is mentally retarded and therefore that the Eighth Amendment, as interpreted by the Supreme Court in Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002), bars his execution. Williams also argued that the Sixth Amendment bars his execution because the jury did not make a determination on his mental retardation claim. 1 Finding that Williams failed to make a prima facie case of mental retardation, the TCCA dismissed the petition as an abuse of the writ. Ex Parte Williams, No. 50,662-02 (Tex.Crim.App.2003).

C. Federal Court Proceedings

On July 20, 2004, Williams filed the present federal habeas petition. In it, Williams raises the same issues as in his state petition. 2 The district court found that the state court’s conclusion that Williams failed to plead a prima facie case of mental retardation “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), and accordingly found Williams’s petition was not barred by this threshold inquiry under AEDPA. The district court therefore ordered an evidentiary hearing on the question of Williams’s mental retardation.

The district court dismissed Williams’s petition with respect to his various other claims. On Williams’s ineffective assistance of counsel (“IAC”) claim, the court concluded that Williams had failed to raise this argument in state court, rendering that claim unexhausted and procedurally barred. The court found Williams’s various other claims either procedurally defaulted or unexhausted.

A magistrate judge held a seven-day evidentiary hearing on the issue of mental retardation. After responsive briefing, the magistrate judge issued a 78-page memorandum and recommendation summarizing the evidence adduced from the evidentiary hearing and recommending that the district court deny Williams’s petition. However, the magistrate judge also recommended that the district court grant a COA. The district court adopted the magistrate judge’s memorandum and recommendation and dismissed Williams’s habe-as petition with prejudice, granting a COA on the mental retardation claim and denying a COA on Williams’s other claims (although Williams had not sought a COA on any of his claims).

Subsequently, the district court denied two motions made by Williams. First, the court considered Williams’s Motion to Alter or Amend Judgment Pursuant to Rule 59(e) of the Federal Rules of Civil Procedure. In this motion, Williams made (1) an actual innocence claim based on newly discovered evidence and (2) an IAC claim. On the actual innocence claim, the court reasoned that even if the claim were substantiated, it would not entitle Williams to relief. See Herrera v. Collins, 506 U.S. 390, 400, 113 S.Ct. 853, 122 L.Ed.2d 203 (1993) (“Claims of actual innocence based on newly discovered evidence have never been held to state a ground for federal habeas relief absent an independent constitutional violation occurring in the underlying state criminal proceeding.”).

Although the district court previously held Williams’s IAC claim to be unexhaust-ed and procedurally defaulted, Williams argued that new evidence of actual innocence provided a basis to excuse that default. He also claimed the evidence of his mental retardation adduced at the eviden-tiary hearing was further proof of ineffective assistance for his trial counsel’s failure to present that information as mitigating evidence. The district court concluded that even if Williams could overcome the procedural default bar, the evidence adduced at the evidentiary hearing did not support an IAC claim because “the substance of this evidence was already before the jury [and] there is no reasonable probability that this evidence would have led to a different result.”

Second, the district court considered Williams’s motion for relief from judgment under Rule 60(b). In his Rule 60(b) motion, Williams alleged that the government failed to comply with Federal Rule of Criminal Procedure 45(b)(l)’s requirement that opposing counsel be served with copies of subpoenas duces tecum. At the evidentiary hearing, the magistrate judge had recognized that Rule 45(b) violations occurred but nevertheless allowed the subject documents into evidence. Concluding that Williams failed to show how these violations prejudiced his ability to present his case, the district court denied the Rule 60(b) motion.

D. Evidentiary Hearing

The evidentiary hearing produced significant evidence supporting and refuting Williams’s claims of mental retardation.

i IQ tests and achievement tests

Williams scored a 70 or 71 in three different IQ test administrations. Williams’s first IQ score was recorded at age 16. On July 7, 1992, the Houston Independent School District (HISD) administered to Williams the Wechsler Intelligence Scale for Children — III (“WISC-III”). Williams scored a verbal IQ of 79, a performance IQ of 65, and a full scale IQ of 70. Dr. Joyce Hood, who administrated these tests, interpreted the results:

Jeffrey’s Full Scale score is in the Borderline range, two standard deviations below the mean, and places him in the 2nd percentile relative to children his age in the normal population.

Jeffrey’s verbal score is in the Borderline range, more than one standard deviation below the mean, and in the 8th percentile relative to children his age in the normal population.

Jeffrey’s performance score is in the Mentally Retarded range, more than two standard deviations below the mean, and in the 1st percentile relative to children his age in the normal population.

Williams was not considered mentally retarded at the time because HISD’s cutoff for that determination was a full scale IQ score of 69.

At the same testing session, Williams took several achievement tests. These tests indicated that his academic achievement in reading, math, and written language was above his assessed IQ level. As the district court noted, “while [Williams’s] IQ showed that he was in the second percentile intellectually, he functioned academically in between the twenty-first and fifty-eighth percentiles.” Dr. Hood offered no explanation for this discrepancy. She did, however, state her opinion that “Jeffrey Williams is mentally retarded.”

Williams next took an IQ test after his murder conviction, in a test 3 administered by the defense expert, Dr. Gilda Kessner. His verbal IQ score was 70, with a 95 percent confidence interval of 66 4 to 76. His performance IQ score was 77, with a 95 percent confidence interval of 72 to 85. The full scale score was 71, 5 with a full scale confidence interval between 68 and 76. 6 Dr. Kessner also administered the Mini Mental State Examination, a test for functionality and concentration, and the REY 15, a memory test. Dr. Kessner concluded that Williams made an adequate effort on those tests and was not malingering.

Dr. Kessner also administered two achievement tests. 7 Williams scored in the non-mentally-retarded range on both. Dr. Kessner testified that in her opinion, Williams’s scores on various academic achievement tests were not inconsistent with his IQ scores because the confidence intervals for the achievement tests overlapped with confidence intervals on the IQ tests. However, the district court concluded that Dr. Kessner presented “no credible explanation” for this discrepancy. Another expert testifying for Williams, Dr. Richard Garnett, stated that IQ testing is the best measure of intelligence because it is the broadest measure of the diffuse elements of intelligence, unlike achievement tests, which measure a narrower portion of intelligence.

The government’s expert, Dr. Thomas Allen, administered the Stanford-Binet V (“SB-5”) Test of Intelligence. Williams scored a non-verbal IQ of 70, a verbal IQ of 75, and a full scale IQ of 71. Dr. Allen also used Green’s Word Memory Test (“GWMT”) to assess Williams’s level of effort on the IQ test. Dr. Allen testified that he believed Williams’s low IQ score was the result of his failure to put forth a good effort on the test. Williams’s score on a portion of the GWMT was lower than if he had been randomly guessing. On another section, Williams scored so low that in Dr. Allen’s opinion, Williams was deliberately giving the wrong answers.

Williams most recently took an intelligence test when he arrived on death row. A licensed professional counselor working for the Texas Department of Criminal Justice administered the TONI-3, a test of nonverbal intelligence. Williams scored a nonverbal IQ of 83, a score in the thirteenth percentile and too high to qualify Williams for a place in the counselor’s mental health case load.

The district court also considered Williams’s scores in routine standardized tests administered while he was in school. In his sixth, seventh, and eighth-grade years, Williams took the Metropolitan Achievement Test (“MAT”), a national standardized achievement test. In three sittings of the exam, he scored below grade level in some subjects and above grade level in others. In a different achievement test administered in the seventh grade, the Texas Educational Assessment of Minimum Skills (“TEAMS”), Williams’s scores reflected subject mastery in mathematics, reading, and writing. In the ninth grade, Williams passed the Texas Assessment of Academic Skills (“TAAS”) test in reading and writing, but he failed the mathematics portion.

ii Other expert testimony

Dr. Garnett testified about Williams’s adaptive deficits. He opined that Williams showed significant adaptive limitations in the areas of self care, home living, social and personal skills, work, and leisure. With respect to self care, Dr. Garnett cited Williams’s failure to dress appropriately for the weather. In the area of home care, Dr. Garnett noted that during the only time Williams lived alone for a brief period, he ended up losing his apartment, becoming homeless, and his car broke down because he failed to maintain it. In the area of social skills, Dr. Garnett noted that Williams was beaten up as child and that he had difficulty interacting with people. In the area of work, Dr. Garnett noted that Williams had held five or six jobs in his life and was fired from all of them. In the area of leisure, Dr. Garnett noted testimony that Williams could not play video games or basketball, and that he had no independent hobbies or other leisure activities.

As further evidence of adaptive limitations, Dr. Garnett cited Williams’s reluctance to bathe as an adolescent, inability to play sports and games, becoming homeless after losing his job, and failure to maintain his cars. Dr. Garnett also testified that mentally retarded people often adopt strategies to function in everyday life, such as relying on “ambassadors” — caretakers or enablers to help them. According to Dr. Garnett, Williams adopted such strategies.

Dr. Garnett admitted that these deficits could be explained by other diagnoses, such as a conduct disorder, oppositional defiance, or learning disabilities. He testified that Williams did not suffer from antisocial personality disorder based on testimony that Williams was empathetic and joked with friends. Dr. Garnett also testified that Williams did not have ADHD because Williams could focus at times.

in. Williams’s school records

During middle school and at the beginning of high school, Williams was in a regular classroom. In April 1992, in Williams’s tenth grade year, he was expelled from high school after he was blamed for a series of locker thefts, setting a trash can on fire, and truancy. He then was admitted to West Oaks psychiatric hospital, where he received counseling and attended group therapy sessions. Doctors at West Oaks diagnosed Williams with a conduct disorder and hyperactivity and prescribed medication for short attention span and impulsivity. Williams completed the second semester of his tenth-grade eoursework at a community service campus of West Oaks. He passed all of his classes except Algebra.

He returned to his high school in the fall. In November 1992, a committee at Williams’s high school decided that he qualified for special education accommodations because of emotional disturbance. He took some classes in the regular classroom and others in a special education environment. All classes were taught at the tenth-grade level. Required modifications included frequent breaks, defined physical space, cooling-off periods, positive reinforeers, and a discipline management plan. The following April, the school modified Williams’s education plan to mainstream him into some regular education classes at the eleventh-grade level in the fall. In the 1992-1993 academic year, Williams passed all his classes and earned several A’s. The following year, Williams took eoursework at the twelfth-grade level, earning passing scores but with a deterioration in his behavior in the spring term. He earned grades in the 70 to 71 range for his regular English and government classes and scores in the 80s for his special education classes. Williams’s high-school grades were generally low but passing. At the times he received his lowest grades, he also had a large number of recorded absences. Williams graduated from high school with a grade point average of 2.19 and a class rank of 326 out of 480.

James Claypool, the principal at one of Williams’s high schools, recounted Williams’s self-destructive behavior (as summarized by the magistrate judge):

Claypool described Williams as soft-spoken and respectful, but prone to getting into trouble when not supervised by an authority figure. Claypool described Williams as being fascinated by crime and criminals and gravitated toward those peers who engaged in criminal activities. Claypool stated that Williams often informed school officials of his friends’ illegal activities and then returned to his friends and told them what he had just done. As a result, Williams was frequently beaten up by his Mends. Claypool often counseled Williams about this attention-getting/self-destructive behavior, but Williams was unable to stop the cycle of informing on his friends and then informing on himself. This behavior continued until Williams was moved to a different classroom.

(footnotes omitted).

iv. Character testimony

The evidentiary hearing included testimony of people who knew Williams at various stages of his life. The court considered, for example, an assessment of Williams at age 16 by his own parents:

Jeffrey is not very talkative and is oppositional at times. He has been a problem since kindergarten. He is a habitual liar, has cut teachers’ tires, and has stolen from stores in the mall. He runs with a bad crowd and is easily influenced by them. He shows no remorse for things he does.

Williams’s high school principal, Claypool, spoke further of Williams’s abilities:

Claypool testified that he did not interact with Williams as if he were mentally retarded. Rather, Claypool considered him to be “bright,” “socially skilled with adults,” and “manipulative.” Williams was not considered to be mentally retarded academically and was not placed in classes for the mentally retarded. Claypool considered Williams “above average” in his dress and hygiene. Clay-pool testified that, while a teacher noted that Williams often wore a long coat to school, he did not observe that type of clothing when he saw Williams. The teacher’s note dated from a time when Williams was breaking into lockers. Claypool believed that the long coat was worn to hide stolen items.

Williams’s childhood friends testified about odd behavior. For example, Williams refused to play by the normal rules of childhood games: in a game of kickball, he threw the ball over the fence; in a game of basketball, he would kick the ball. Several witnesses testified that Williams often wore inappropriate clothing, such as an overcoat with a hood during the summer (in Houston) or t-shirts and shorts in the winter. Several witnesses testified about what appeared to be attention-getting behavior. Williams would often yell out inappropriate remarks when he was unable to keep up with conversation.

Other witnesses spoke of Williams’s self-destructive behavior, poor hygiene, and inability to cook. Several witnesses testified that they treated Williams differently because they believed he was “slow.”

v. Williams’s job history

Williams enlisted in the Naval Reserve during his senior year in high school. He reported for active duty in July 1994. Several witnesses testified that Williams was slower in academies than other soldiers. Williams also was reluctant to do menial chores or meet the Navy’s standards for neatness. On an assignment as helmsman of a ship, Williams caused the ship to “roll” by turning the wheel too quickly. Williams’s supervisor testified that he lacked “military bearing” and was frequently late. His supervisor ranked Williams “ ‘towards the bottom’ in terms of understanding and carrying out his other responsibilities among the hundreds of seamen” he had supervised. Others testified that Williams was capable of hard work, but that he had a bad attitude toward authority.

Williams’s Navy disciplinary record recounts episodes of insubordination, failure to maintain a neat bunk area, and excessive alcohol consumption. He was coun seled and punished for such behavior on numerous occasions. In March 1996, a disciplinary board recommended that Williams be separated from military service. Williams received a general discharge from the Navy in October 1996.

After leaving the Navy, Williams worked briefly as a deck hand on a casino boat. Williams also worked as a stocker at a grocery store and a delivery truck driver for an auto parts store.

vi Williams’s writing skills

While in the Navy, Williams wrote a letter — with a few suggestions and corrections from another sailor — to the Administrative Separation Board to discourage it from discharging him from the Navy. The letter explained that in the Navy, Williams was able to escape the atmosphere of drugs and death that he had lived in while growing up. He expressed his wish to take a Navy exam to allow him to move to another division. Williams explained that if he stayed in the Navy, he would have a better chance of finding a job in the civilian world later. He also wrote of hoping to buy a car and a house in Houston. Dr. Garnett assessed this letter as written at the seventh-grade level. Other death row inmates housed near Williams testified that they helped Williams in writing letters and commissary orders, although a prison official testified that security procedures would have made it very difficult for inmates to exchange written notes.

The district court rejected Williams’s Atkins claim. However, acknowledging that it was a “close case,” the court, sua sponte, granted a certificate of appealability for this court to review the district court’s finding of no mental retardation. Because the district court granted a COA on the issue of mental retardation, we have jurisdiction to review the district court’s denial of habeas relief on Williams’s Atkins claim. 28 U.S.C § 2253.

II. DISCUSSION

A. Mental Retardation Claim

The primary claim in this appeal is that Williams is mentally retarded and therefore ineligible for execution under Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002). In Atkins, the Supreme Court held that under the Eighth Amendment’s “evolving standards of decency” review, “death is not a suitable punishment for a mentally retarded criminal.” Id. at 321, 122 S.Ct. 2242. The Atkins Court did not adopt specific criteria for determining mental retardation. Rather, the Court left to the states “the task of developing appropriate ways to enforce the constitutional restriction____” Id. at 317, 122 S.Ct. 2242.

Since then, the TCCA has developed standards by which it evaluates an inmate’s claim of mental retardation. See Ex parte Briseno, 135 S.W.3d 1, 8 (Tex.Crim.App.2004). The TCCA adopted the framework established by the American Association on Mental Retardation (“AAMR”) in conjunction with standards contained in the Texas Persons with Mental Retardation Act (“PMRA”), Tex Health & Safety Code Ann. § 591.003(13). According to the AAMR definition,

Mental retardation refers to substantial limitations in present functioning. It is characterized by significantly subaver-age 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.

Atkins, 536 U.S. at 309 n. 3, 122 S.Ct. 2242 (quoting Mental Retardation: Definition, Classification, and Systems of Supports 5 (9th ed.1992)). Under the PMRA, “ ‘mental retardation’ means significantly subaverage general intellectual functioning that is concurrent with deficits in adaptive behavior and originates during the developmental period.” Tex. Health & Safety Code Ann. § 591.003(13). Combining and summarizing these definitions, Texas courts define mental retardation as characterized by “(1) ‘significantly subaverage’ general intellectual functioning; (2) accompanied by ‘related’ limitations in adaptive functioning; (3) the onset of which occurs prior to the age of 18.” Briseno, 135 S.W.3d at 7 (footnotes omitted). To obtain habeas relief, Williams must show that he meets all three elements. See Clark v. Quarterman, 457 F.3d 441, 444 (5th Cir. 2006). We consider each of these elements in turn.

i. Standard of review

In a habeas corpus appeal, this court reviews the district court’s findings of fact for clear error and its conclusions of law de novo. Ramirez v. Dretke, 396 F.3d 646, 649 (5th Cir.2005). This court reviews mixed questions of law and fact under a de novo standard by independently applying the law to the facts found by the district court, as long as those facts are not clearly erroneous. Id. The determination of whether Briseno’s three prongs have been met is a factual finding that we review for clear error. See Rivera v. Quarterman, 505 F.3d 349, 361-63 (5th Cir.2007). “ ‘A finding is clearly erroneous only if it is implausible in the light of the record considered as a whole.’” Id. (quoting St. Aubin v. Quarterman, 470 F.3d 1096, 1101 (5th Cir.2006)). “If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently.” Anderson v. City of Bessemer City, 470 U.S. 564, 573-74, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985). “Where there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous.” Id.

ii Significantly subaverage intellectual functioning

a. Applicable law

In determining intellectual functioning, the Briseno court focused on IQ: “Significantly subaverage intellectual functioning is defined as an IQ of about 70 or below (approximately 2 standard deviations below the mean).” Briseno, 135 S.W.3d at 7 n. 24 (quoting DSM-IV 8 at 39). Mental health professionals apply this standard flexibly, and “sometimes a person whose IQ has tested above 70 may be diagnosed as mentally retarded while a person whose IQ tests below 70 may not be mentally retarded.” Id. In Briseno, for example, the petitioner’s IQ test results ranged from a low of 67 to a high of 88. Id. at 14 n. 53. In that case, the court nevertheless concluded that recent IQ scores of 72 and 74 were a more accurate reflection of the petitioner’s intelligence. Id. at 14. Therefore, the petitioner was not mentally retarded. Id. at 18.

b. Magistrate judge’s memorandum and recommendation

The evidence adduced at the evidentiary hearing amounts to a battle of the experts. The experts agree that Williams achieved a full scale IQ of 70 or 71 on three separate tests. The experts disagree on whether those scores are a valid measure of Williams’s intellectual functioning. The government claims these low scores are the result of malingering — deliberately achieving low scores. Williams disagrees, arguing that it would have been impossible to achieve such consistent results over time by malingering.

The district court credited expert testimony that Williams’s low IQ scores were lower than if he had been randomly guessing. In other words, he must have intentionally given wrong answers. The court also contrasted Williams’s low IQ scores with his academic competence: scores on academic achievement tests in middle school and high school; grades earned in high school; and Williams’s writing ability, as evidence by the letter he wrote to the Administrative Separation Board while serving in the Navy. The court concluded that “Williams’[s] strengths are simply too strong to support the conclusion that he has an IQ of 70 or below, or is otherwise significantly subaverage in intellectual functioning.”

The court found that the root of Williams’s problems in school was behavioral, not intellectual, and his academic problems also could have been related to alcohol and drug abuse. The court also considered testimony of multiple witnesses who did not believe that Williams was mentally retarded.

The district court concluded that Williams has an IQ substantially higher than the threshold of 70. The magistrate judge noted that while Williams’s Ml scale IQ scores were consistent over time, the two components of the IQ test — verbal and performance — fluctuated markedly.

The district court ultimately sided with the government’s expert in concluding that “Williams’ performance on academic achievement tests taken in middle-school, high-school, and as an adult are uniformly inconsistent with his purported IQ.” The court also cited Williams’s high school grades (“Williams took classes at the twelfth-grade level, earned an 80 and 89 in two semesters of Chemistry, graduated with a C average and had a class rank of 326 out of 480”) and his writing ability (as evidenced by several letters in the hearing record) as inconsistent with a finding of mental retardation.

c. Analysis

In this appeal, Williams challenges the district court’s conclusion that his IQ scores are not an accurate indication of his intellectual functioning. Williams argues that (1) his IQ scores were too consistent over time to be inaccurate or intentionally low, (2) there is no direct evidence of malingering in two of his three IQ tests, and (3) other factors relied on by the district court should not override his IQ scores.

In applying Briseno, the Fifth Circuit counsels a flexible approach to reading IQ scores, warning that “courts should not rigidly consider an IQ score to be determinative of the defendant’s intellectual functioning.” Clark, 457 F.3d at 444-5; see also Briseno, 135 S.W.3d at 7 n. 24 (“Psychologists and other mental health professionals are flexible in their assessment of mental retardation.”).

Superficially, Williams’s IQ scores are remarkably consistent. Dr. Kessner, Williams’s expert, testified about the unlikelihood that a test-taker could intentionally achieve such consistent scores:

Q How would you estimate the difficulty of being able to attain intentionally a target score on three tests over a course of 14 or 15 years with three different administrators?

A I think the probability of that is as low or lower than any probability of anything I could think of. I mean

Q Like winning the Powerball?

A Extremely limited. I hesitate to say a number, but will say I think it’s almost impossible.

Dr. Kessner added, “I cannot for the life of me figure out how someone could malinger so elegantly as to get the same scores on these tests over time.” Nevertheless, the components underlying these scores undermine the consistency argument. In 1992, Williams’s full scale score of 70 was derived from a verbal IQ of 79 and a performance IQ of 65. In Dr. Kessner’s administration of the IQ test, Williams’s full scale score was 71, but his verbal score dropped to 70 while his performance score rose to 77. In Dr. Allen’s administration, the full scale score remained 71 but the component scores flip-flopped again, with a non-verbal score of 70 and a verbal IQ of 75. Thus, in different administrations, Williams was capable of scoring as high as a 77 on the performance section and a 79 on the verbal — both outside the mental retardation range. Thus, the district court did not commit clear error in concluding that these IQ scores do not support a finding of mental retardation because the district court’s finding is plausible in light of the record.

Williams also challenges the district court’s reliance on Dr. Allen’s finding of malingering. The district court credited Dr. Allen’s testimony on the results of the GWMT, which revealed that Williams was not putting forth a good effort during that test administration. According to Dr. Allen, Williams’s score on one portion of the GWMT — “immediate recall” — was worse than chance; in other words, Williams must have intentionally missed questions. In another portion of that test — “delayed recall” — Williams’s score was lower than a score expected from a person hospitalized with advanced dementia. Dr. Allen testified that the GWMT is “the most widely researched test of effort in literature and on the market.” Williams failed this test while groups of mentally retarded children and adults have passed it. Adjusting Williams’s IQ score for poor effort, Dr. Allen estimated that Williams’s true IQ score is “in the 90 range.”

Dr. Kessner administered a different test for effort, the REY 15. Based in part on Williams’s performance on that test, Dr. Kessner believed that he put forth adequate effort. However, the district court found this testimony less persuasive than Dr. Allen’s testimony on effort. Further, Drs. Kessner and Garnett conceded that the REY 15 is not an effective test of effort — they both testified there is no instrument that can be used to reliably test effort in institutionalized individuals that has been published, researched, and supported in scientific literature. Dr. Allen testified at length about the research underlying the GWMT and various cutoff scores for effort. By contrast, Dr. Kess-ner’s testimony about Williams’s effort was highly subjective. It therefore was not clear error for the district court — given this contradictory testimony — to credit Dr. Allen’s opinion. 9

Finally, Williams argues that the district court erred in using achievement test scores to override IQ scores that suggested borderline mental retardation. To the extent that this argument urges us to credit IQ scores to the exclusion of all other measures of intelligence, we decline to do so. Dr. Allen testified that achievement tests and IQ scores are highly correlated, and he discredited at least one of Williams’s IQ scores on the basis of malingering. Dr. Kessner could provide no rationale for the discrepancy between Williams’s IQ scores and his academic achievement. Ultimately, the district court found Dr. Allen’s explanation to be “more credible.” Such a credibility determination is a finding of fact that this court reviews for clear error. See Lewis v. Dretke, 355 F.3d 364, 368 (5th Cir.2003); see also Canal Barge Co. v. Torco Oil Co., 220 F.3d 370, 375 (5th Cir.2000) (‘“The burden of showing that the findings of the district court are clearly erroneous is heavier if the credibility of witnesses is a factor in the trial court’s decision.’ ” (quoting Dunbar Med. Sys., Inc. v. Gammex Inc., 216 F.3d 441, 453 (5th Cir.2000))). In light of the record before us, we cannot conclude that the district court clearly erred.

The district court was persuaded that the evidence as a whole — including Williams’s performance in the classroom and on standardized achievement tests, his writing ability, and the opinions of numerous witnesses who did not believe that Williams was retarded — was inconsistent with a finding of mental retardation. Nothing in the Briseno standard compels a blind adherence to IQ. In fact, Briseno itself recognizes that IQ alone is not determinative of mental retardation. Briseno, 135 S.W.3d at 7 n. 24 (quoting DSM-IV at 39). The mental retardation standard is designed to be flexible: “sometimes a person whose IQ has tested above 70 may be diagnosed as mentally retarded while a person whose IQ tests below 70 may not be mentally retarded.” Id. Furthermore, we have previously observed that there is “no binding authority that requires an IQ test specifically, that is, entirely alone, at the core, or as any singular threshold, to provide the basis for a finding of mental retardation.” Morris v. Dretke, 413 F.3d 484, 497 (5th Cir.2005).

Because Williams carries the burden of proving mental retardation, failure to prove the first Briseno prong should end our inquiry. However, because the district court considered this issue to be a “close call,” we continue to the second prong to illustrate that, even if Williams has significantly subaverage intellectual functioning, he did not exhibit the type of adaptive deficits required under Briseno. 10

Hi. Related adaptive limitations

The AAMR definition of mental retardation requires a showing of “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.” Atkins, 536 U.S. at 309 n. 3, 122 S.Ct. 2242. The Texas PMRA similarly requires deficits in “adaptive behavior,” meaning “the effectiveness with or degree to which a person meets the standards of personal independence and social responsibility expected of the person’s age and cultural group.” Tex Health & Safety Code Ann. § 591.003(1).

Briseno noted that “[t]he adaptive behavior criteria are exceedingly subjective, and undoubtedly experts will be found to offer opinions on both sides of the issue in most cases.” 185 S.W.3d at 8. Therefore, the TCCA offered a list of other “eviden-tiary factors” to consider in deciding whether evidence suggests mental retardation or a personality disorder:

• Did those who knew the person best during the developmental stage — his family, friends, teachers, employers, authorities — think he was mentally retarded at that time, and, if so, act in accordance with that determination?

• Has the person formulated plans and carried them through or is his conduct impulsive?

• Does his conduct show leadership or does it show that he is led around by others?

• Is his conduct in response to external stimuli rational and appropriate, regardless of whether it is socially acceptable?

• Does he respond coherently, rationally, and on point to oral or written questions or do his responses wander from subject to subject?

• Can the person hide facts or lie effectively in his own or others’ interests?

• Putting aside any heinousness or gruesomeness surrounding the capital offense, did the commission of that offense require forethought, planning, and complex execution of purpose?

Id. at 8-9.

b. Magistrate judge’s memorandum and recommendations

The district court considered evidence of Williams’s bizarre and antisocial conduct. The court also noted Dr. Garnett’s testimony that Williams showed adaptive limitations in the areas of self care, home living, social and personal skills, work, and leisure. However, the court was not persuaded that these limitations were caused by mental retardation. Dr. Garnett conceded, in cross examination, that those deficits could all be explained by other diagnoses such as conduct disorder, oppositional defiance, or learning disabilities. The district court found more credible Dr. Allen’s testimony that Williams’s strange behavior was due to anti-social personality rather than mental retardation.

In contrast to Dr. Garnett’s subjective evaluation, the state’s expert, Dr. Allen, administered the Vineland Adaptive Behavior test to measure Williams’s adaptive deficits. The only area where Williams showed an adaptive deficit was in socialization, and that deficit could be explained by antisocial personality disorder rather than mental retardation.

Further, the district court noted Williams’s substantial adaptive strengths, finding that Williams rented an apartment, bought at least two vehicles, washed his clothes, and could conform his conduct to social norms when he wanted. Although some witnesses reported that Williams did not understand the rules of sports and games, other testimony suggested that Williams did play sports and understand the rules. Ultimately, the court concluded that “Williams’s eccentric behaviors are just as easily seen as attention-getting behaviors as they are evidence of mental retardation.” If Williams does suffer from some adaptive deficits, the court concluded, they are not “significant.”

As in our analysis of the first Briseno prong, we give great weight to the district court’s credibility determinations. 11 At the evidentiary hearing, the court heard from many witnesses who testified about Williams’s abilities throughout his life, including numerous examples of bad behavior and antisocial conduct. Although Williams’s expert, Dr. Garnett, testified that this evidence supports a finding of mental retardation, the state’s expert, Dr. Allen, gave compelling testimony attributing Williams’s bizarre behavior to diagnoses other than mental retardation.

Before this court, Williams does little more than recount testimony adduced at the evidentiary hearing, charging that the district court discounted evidence of his “repeated failures to live a normal life,” including “his inability to keep a job, maintain a car, live on his own, have a girlfriend, or maintain his hygiene.” Williams argues that witnesses from each stage of his life testified that he “could not live like a normal person for his age.” Williams’s brief includes a two-and-a-half-page chart summarizing the testimony of various witnesses about his adaptive deficits.

However, the only specific point of error Williams raises is an argument that the district court should not have considered adaptive strengths in evaluating Williams’s adaptive deficits. The experts spent some time battling over this point in the evidentiary hearing. Dr. Garnett testified that the AAMR’s standard is a deficits-only model. In other words, a person’s strengths should be ignored in making a diagnosis. Dr. Garnett cited the DSM manual’s “exclusionary clause,” which dictates that once a subject meets minimum thresholds for deficits, the diagnosis of mental retardation holds regardless of other adaptive strengths. Dr. Garnett did concede, however, that strengths in specific adaptive areas should be balanced against deficits in those areas to determine if the deficits are “significant.”

Dr. Alen criticized the deficits-only model for its tendency to produce confirmation bias. As he explained, if an expert evaluates a subject with a borderline IQ, the deficits-only model encourages the expert to look for deficits which tend to confirm a diagnosis of mental retardation while ignoring contrary evidence. Thus, the deficits-only model is too likely to confirm a suspected diagnosis of mental retardation if evidence of adaptive abilities is not considered. 12 To be sure, the Briseno definition does reference adaptive limitations without reference to strengths. However, the standard also requires that these limitations in certain adaptive categories be “significant.” As Williams’s expert testified, weighing strengths against weaknesses is a crucial part of determining whether purported adaptive limitations are “significant.”

Furthermore, this court has previously held that courts are not barred from con sidering adaptive strengths in evaluating mental retardation claims. See Clark v. Quarterman, 457 F.3d 441, 447 (5th Cir. 2006). In Clark, the petitioner argued that evidence of his adaptive skills did not support the state court’s determination that he was not mentally retarded because this was evidence of strengths and not limitations. Id. The court reasoned that

[t]his is incorrect — evidence of a strength in a particular area of adaptive functioning necessarily shows that the defendant does not have a weakness in that particular area. Even if, as [the petitioner] argues, adaptive limitations rather than strengths often define mental retardation, the evidence in this case shows primarily adaptive strengths and does not show limitation in any significant area. The evidence in this case showed that prior to being incarcerated, [the petitioner] functioned normally across a broad range of adaptive behaviors.

Id.

Similarly, the district court in the instant case concluded that the evidence showed significant adaptive strengths — -especially in the area of functional academics — and adaptive weaknesses that were either insignificant or could be more credibly explained by other behavioral diagnoses. To find clear error, we must do more than merely disagree with the district court’s conclusion. Anderson, 470 U.S. at 573-74, 105 S.Ct. 1504. We therefore find no clear error in the district court’s crediting the state’s expert testimony over Williams’s expert on the issue of adaptive deficits. Williams exhibited numerous adaptive strengths to support a conclusion that he lacked significant adaptive weaknesses. Williams made long-distance calls with calling cards, bought plane tickets unassisted, traveled out of state, rented an apartment, applied for unemployment benefits on his own, and washed his own clothes. Although the magistrate judge observed that several witnesses noted Williams’s “various odd behaviors,” the court concluded that Williams “could conform his conduct to expected social norms when he wanted to” and that his “eccentric behaviors are just as easily seen as attention-getting behaviors as they are evidence of mental retardation.” We find that the district court did not clearly err in reaching that conclusion.

B. Rule 59(e) Motion and Rule 60(b) Motion

Williams challenges the district court’s denial of his post-judgment motions under Federal Rules of Civil Procedure 59(e) and Rule 60(b), both filed after the district court issued its order denying habeas corpus relief. In his 59(e) motion, Williams sought (1) the opportunity to pursue a claim of actual innocence based on newly discovered evidence; (2) the opportunity to pursue an ineffective assistance of counsel claim based on the evidence of mental retardation presented at the evidentiary hearing but not at his sentencing; and (3) a stay and abeyance to allow him to exhaust state court remedies. The district court denied this motion. In his Rule 60(b) motion, Williams alleged a series of violations of Federal Rule of Civil Procedure 45(b)(1) — which imposes a notice requirement on the issuance of subpoenas duces tecum — by the state’s counsel. The district court also denied Williams’s Rule 60(b) motion. Williams did not seek a COA to review the denial of these motions, either from this court or from the district court. The district court did not grant or deny a COA on these issues.

As a threshold matter, we consider whether this court has jurisdiction to hear an appeal on these motions. A COA is required to appeal “the final order in a habeas corpus proceeding in which the detention complained of arises out of process issued by a State court.” 28 U.S.C. § 2258(c)(1)(A). Although the circuits are split on whether a COA is required to appeal a district court’s denial of a Rule 60(b) motion, Gonzalez v. Crosby, 545 U.S. 524, 535 n. 7, 125 S.Ct. 2641, 162 L.Ed.2d 480 (2005), the Fifth Circuit requires a COA only if a Rule 60(b) motion seeks “to amend or alter the judgment of a first habeas proceeding, in which case the Rule 60(b) motion should be treated as a successive habeas petition.” Dunn v. Cockrell, 302 F.3d 491, 492 n. 1 (5th Cir.2002). However, the court has read this exception to the COA requirement narrowly, concluding that Dunn’s exception that “a COA is not required to appeal the denial of a Rule 60(b) motion applies only when the purpose of the motion is to reinstate appellate jurisdiction over the original denial of habeas relief.” Ochoa Canales v. Quarterman, 507 F.3d 884, 888 (5th Cir.2007) (per curiam). In Dunn, the court allowed a petitioner to appeal the district court’s denial of his Rule 60(b) motion because the motion merely sought to reinstate his final judgment so he could file a timely notice of appeal. Dunn, 302 F.3d at 493. The petitioner in that case did not seek to appeal the merits of his habeas petition via the Rule 60(b) motion. Id. at 492.

In Ochoa Canales, by contrast, the petitioner sought relief from a district court judgment that his earlier habeas petition was barred for failure to exhaust state remedies. Ochoa Canales, 507 F.3d at 886. Citing a “narrow interpretation” of Dunn, the court held that the petitioner was attempting to use Rule 60(b) to alter the judgment in a habeas proceeding and therefore was required to obtain a COA. Id. at 888.

Williams seeks specific relief — an order compelling the government to disclose documents and granting additional time for supplemental briefing — and thus seeks relief other than the mere “reinstate[ment of] appellate jurisdiction over the original denial of habeas relief’ sought in Dunn. See Ochoa Canales, 507 F.3d at 888. If the exception to the COA requirement articulated in Dunn is indeed limited to the facts of Dunn, a COA is required in this case. Compare Hunt v. Quartennan, 270 Fed.Appx. 357, 357 (5th Cir.2008) (per curiam) (“[The petitioner’s] Rule 60(b) motion did not merely seek authorization for out-of-time appeal. Rather, he raised several claims related to the judgment dismissing his § 2254 petition. Consequently, the COA requirement applies to his appeal from the judgment denying his Rule 60(b) motion.”) with Davis v. King, 270 Fed.Appx. 355, 356 (5th Cir.2008) (per curiam) (“As [the petitioner] was not attempting to use his motion for reconsideration to alter the judgment in his underlying habeas petition, but instead attempting to vacate the judgment and have it reentered so that he could file a timely notice of appeal, a COA is not necessary.”).

Williams also must obtain a COA to appeal his Rule 59(e) motion. Although we have not previously addressed this issue, we find the Second Circuit’s reasoning persuasive:

[T]he evident Congressional intent of § 2253(c)(1) is to weed out unmeritorious appeals taken from denials of habeas petitions; that it would be rather anomalous for Congress to have intended that appeals be allowed from the denial of Rule 60(b) motions without the same scrutiny; and that the interest served by the COA requirement — relieving the state and the court system of the burdens resulting from the litigation of insubstantial appeals — is served by the ap plication of that requirement to appeals from orders denying post-judgment relief. These rationales apply with equal force to Rule 59(e) motions. We therefore conclude that the § 2253(c)(1) COA requirement applies to an appeal from an order denying a Rule 59(e) motion when the underlying judgment is a denial of a § 2254 petition.

Jackson v. Albany Appeal Bureau Unit, 442 F.3d 51, 54 (2d Cir.2006) (internal quotation marks, citations, and footnote omitted).

Normally, we may construe a petitioner’s notice of appeal as a request for a COA. See Fed. R.App. P. 22(b). However, Williams never sought a COA from the district court on the denials of his Rule 60(b) and 59(e) motions. Even if we were to construe his notice of appeal as a request for a COA from this court, a petitioner must first request a COA from the district court before requesting it from an appellate court. See Sonnier v. Johnson, 161 F.3d 941, 945-46 (5th Cir.1998); Whitehead v. Johnson, 157 F.3d 384, 387-88 (5th Cir.1998). “Compliance with the COA requirement of 28 U.S.C. § 2253(c) is jurisdictional, and the lack of a ruling on a COA in the district court causes this court to be without jurisdiction to consider the appeal.” Whitehead, 157 F.3d at 388 (citing Muniz v. Johnson, 114 F.3d 43, 45 (5th Cir.1997)). Therefore, we will not consider the merits of Williams’s 60(b) and 59(e) motions.

We have previously remanded a case to the district court where a petitioner failed to seek a COA from the district court before appealing the denial of Rule 59(e) and 60(b) motions. See, e.g., United States v. Arzola-Amaya, 247 F.3d 240, 2001 WL 43527, at *1 (5th Cir.2001) (per curiam) (unpublished table decision). Because our precedent requires a district court to first consider whether to grant a COA, we remand these two issues to the district court to make that determination.

III. CONCLUSION

For the reasons stated above, we affirm the district court’s denial of habeas corpus relief on Williams’s claim of mental retardation. However, because we lack jurisdiction to consider Williams’s appeal of his post-judgment motions, we remand to the district court to consider whether to grant or deny a COA on those issues.

AFFIRMED IN PART AND REMANDED.

* Pursuant to 5th Cir. R. 47.5, the court has determined that this opinion should not be published and is not precedent except under the limited circumstances set forth in 5th Cir. R. 47.5.4.

1 . Williams's argument that the district court erred in not requiring the state to prove a lack of mental retardation beyond a reasonable doubt is foreclosed by our prior precedent. See In re Johnson, 334 F.3d 403, 405 (5th Cir.2003) (“[N]either Ring and Apprendi nor Atkins render the absence of mental retardation the functional equivalent of an element of capital murder which the state must prove beyond a reasonable doubt."); see also Ex parte Briseno, 135 S.W.3d 1, 9 (Tex.Crim.App. 2004) ("Ring and Atkins do not require a post-conviction jury determination of applicant’s claim of mental retardation.”).

2 . The Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA”) bars federal habe-as relief on issues that were adjudicated on the merits in state court unless the state court’s decision was (1) "contrary to, or involved an unreasonable application of, clearly established federal law, as determined by the Supreme Court of the United States” or (2) "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).

3 . Williams completed the Wechsler Adult Intelligence Scale, Third Edition (“WAIS-III").

4 . The magistrate judge’s memorandum lists the bottom of this range as 68, but the transcript of Dr. Kessner’s testimony reveals a range of 66 to 76.

5 . The magistrate judge’s memorandum lists the full scale score as 70, but the transcript of Dr. Kessner’s testimony reveals a score of 71.

6 . Again, the magistrate judge mixed up these numbers. The magistrate judge lists the range as 66 to 76, but Dr. Kessner’s testimony was that the 95 percent confidence interval was 68 to 76.

7 . She administered the Wide Range Achievement Test-Revision 3 ("WRAT-3”) and the Kaufman Functional Academic Skills Test ("K-FAST"). On the K-FAST, Williams scored a 78 in arithmetic and an 82 in reading with a composite score of 79. Dr. Kess-ner noted that the confidence intervals of Williams’s K-FAST and WISC-3 scores overlapped. The 95 percent confidence interval for Williams's WISC-3 IQ test was 68 to 76, and the 95 percent confidence interval for his F-KAST scores was 69 to 89. Of course, these are only 95 percent confidence intervals, meaning that, according to Dr. Kessner, "[f]ive times out of a hundred, you might test him and he might get outside of this [confidence interval] range.”

8 . American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (4th ed. 2000) ("DSM-IV”).

9 . In Williams's first IQ test, Dr. Hood did not administer a standardized test for effort. In her notes, Dr. Hood wrote that Williams appeared to put forth a good effort, but he did not persist when questions became more difficult.

10 . We do not address the third prong — onset before age eighteen — because Williams took an IQ test at age 16. If we were to find that test determinative, the third Briseno prong would be satisfied. However, because we conclude that the district court did not err in discrediting that and other IQ tests, Williams’s intelligence is not subaverage and we need not address the issue of onset.

11 . Indeed, as we have noted previously, the subjectiveness of evaluating the district court's analysis of the "adaptive deficits” prong counsels significant deference:

As the CCA has noted, the adaptive behavior criteria are "exceedingly subjective.” The district courts each held extensive evidential hearings, which included testimony on this issue from, among other witnesses, [the petitioner’s] family members and teachers, and multiple experts. That this case presents, in Judge Hanen’s words, “a close call” strengthens the need to be mindful of the district court’s conclusions. Judge Hanen, having actually presided over the second evidentiary hearing, is in a better position than this court to judge and weigh the credibility of the witnesses who testified on the extent, duration, and causes of [the petitioner’s] adaptive functioning limitations.

Rivera v. Quarteman, 505 F.3d 349, 363 (5th Cir.2007) (footnotes omitted).

12 . Dr. Allen explained that the best scientific method is to begin with the “null hypothesis” — in this case, a diagnosis of no mental retardation — then assess the evidence to determine whether it meets the burden of proving mental retardation.