Eric Cathey v. Lorie Davis, Director

• Court: Court of Appeals for the Fifth Circuit
• Decided: May 11, 2017
• Docket: 16-70015
• Status: Published

Opinion

Case: 16-70015 Document: 00513989592 Page: 1 Date Filed: 05/11/2017

IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT United States Court of Appeals Fifth Circuit No. 16-20312 FILED May 11, 2017 Lyle W. Cayce In re: ERIC DEWAYNE CATHEY, Clerk

Movant, ------------------------------------------- CONSOLIDATED WITH 16-70015

ERIC DEWAYNE CATHEY,

Petitioner - Appellant

v.

LORIE DAVIS, DIRECTOR, TEXAS DEPARTMENT OF CRIMINAL JUSTICE, CORRECTIONAL INSTITUTIONS DIVISION,

Respondent - Appellee

Appeals from the United States District Court for the Southern District of Texas

Before HIGGINBOTHAM, DENNIS, and ELROD, Circuit Judges. PER CURIAM: Eric Dewayne Cathey filed a habeas petition raising an Atkins claim in the Southern District of Texas. 1 The district court concluded that Cathey’s

1 In Atkins v. Virginia, the Supreme Court held that the execution of criminals who are intellectually disabled violates the Eighth Amendment. 536 U.S. 304, 321 (2002). The term “intellectual disability” has replaced the term “mental retardation.” See Brumfield v. Cain, 135 S. Ct. 2269, 2274 n.1 (2015). “Yet, because the term mental retardation is used by Case: 16-70015 Document: 00513989592 Page: 2 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 petition was successive and transferred it to this Court. Cathey appeals the district court’s transfer order. Alternatively, he asks this Court for authorization to file a successive habeas application. We AFFIRM the district court’s transfer order and GRANT the motion for authorization. I. Eric Dewayne Cathey was convicted of capital murder and sentenced to death in Texas state court. On direct appeal, the Texas Court of Criminal Appeals (“CCA”) affirmed Cathey’s conviction and sentence, 2 and the United States Supreme Court denied his petition for a writ of certiorari. 3 Cathey then filed a state habeas petition, which the CCA also denied. On April 2, 2004, Cathey filed a federal habeas petition in the Southern District of Texas. Relevant here, this petition did not include an Atkins claim. The district court denied Cathey’s petition, and this Court declined to grant a Certificate of Appealability (“COA”). 4 In November 2008, on the eve of his scheduled execution, Cathey filed a second state habeas petition raising an Atkins claim. The CCA granted a stay and remanded to the state trial court for a hearing on the petition. 5 Following a five-day hearing, the state trial court signed Cathey’s proposed findings of fact and conclusions of law and recommended that the CCA grant relief. On November 5, 2014, the CCA rejected this recommendation and denied Cathey’s second state habeas petition. 6 Thereafter, Cathey filed in this Court a motion for authorization to file a successive habeas petition raising an Atkins claim. Less than two months later, Cathey asked for permission to withdraw this

both the parties and relevant legal authority, we use mentally retarded throughout our opinion.” Ladd v. Stephens, 748 F.3d 637, 639 n.1 (5th Cir. 2014). 2 Cathey v. State, 992 S.W.2d 460, 461 (Tex. Crim. App. 1999). 3 Cathey v. Texas, 528 U.S. 1082 (2000). 4 See Cathey v. Dretke, 174 F. App’x 841, 841–42 (5th Cir. 2006). 5 Ex parte Cathey, 451 S.W.3d 1, 3 (Tex. Crim. App. 2014) (citation omitted). 6 See Ex parte Cathey, 451 S.W.3d at 3–4.

2 Case: 16-70015 Document: 00513989592 Page: 3 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 motion because he no longer believed that his planned habeas petition qualified as successive. We granted Cathey’s request. Soon after, Cathey filed a petition for habeas corpus raising an Atkins claim in the Southern District of Texas. The State moved to dismiss Cathey’s petition, urging that it was successive. The district court agreed and transferred Cathey’s petition to this Court. 7 Cathey now appeals the district court’s transfer order. Alternatively, he again moves this Court for authorization to file a successive habeas petition. Consistent with our recent guidance, 8 the clerk’s office consolidated Cathey’s two appeals. II. Cathey first challenges the district court’s conclusion that his habeas petition is “second or successive.” Under 28 U.S.C. § 2244(b)(2), “[a] claim presented in a second or successive habeas corpus application . . . that was not presented in a prior application shall be dismissed unless” the petitioner can satisfy one of two narrow exceptions. 9 “In the usual case, a petition filed second in time”—such as Cathey’s petition—“and not otherwise permitted by the terms of § 2244 will not survive AEDPA’s ‘second or successive’ bar.” 10 But “AEDPA uses the phrase ‘second or successive’ as a ‘term of art.’” 11 That is, “[t]he phrase does not encompass all ‘applications filed second or successively in time.’” 12 Rather, “AEDPA’s bar on second or successive petitions only applies

7 See also In re Epps, 127 F.3d 364, 365 (5th Cir. 1997) (per curiam). 8 See United States v. Fulton, 780 F.3d 683, 688 (5th Cir. 2015). 9 Emphasis added. 10 Panetti v. Quarterman, 551 U.S. 930, 947 (2007). 11 In re Lampton, 667 F.3d 585, 587–88 (5th Cir. 2012) (quoting Magwood v. Patterson,

561 U.S. 320, 332 (2010)). 12 Id. at 588 (quoting Magwood, 561 U.S. at 332); see also Panetti, 551 U.S. at 944

(“The Court has declined to interpret ‘second or successive’ as referring to all § 2254 applications filed second or successively in time . . . .”). 3 Case: 16-70015 Document: 00513989592 Page: 4 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 to a later-in-time petition that challenges the same state-court judgment as an earlier-in-time petition.” 13 In Magwood v. Patterson, the Supreme Court applied this rule to a second-in-time habeas petition challenging a death sentence. The petitioner, Magwood, had been sentenced to death in Alabama state court. Following an unsuccessful direct appeal, Magwood filed a federal habeas petition. The district court upheld Magwood’s conviction, but vacated his death sentence and remanded for a new sentencing hearing. After the hearing was conducted in state court, Magwood was again sentenced to death. Magwood then filed a second federal habeas petition challenging his death sentence. 14 Although this petition was filed second in time, the Court held that it was not “second or successive” because it was the “first application” to challenge the “intervening judgment” entered after the second sentencing hearing. 15 That is, it was the first petition to challenge Magwood’s new death sentence. 16 Cathey argues that the same analysis applies here. As he recounts the facts, the state trial court found that Cathey was intellectually disabled “and that his sentence should be commuted to life.” In rejecting these findings and conclusions, so the argument goes, the CCA effectively resentenced him to death and entered a new judgment. Consequently, Cathey claims that his current petition challenges this new judgment entered by the CCA—not the judgment entered when he was originally convicted. And just as in Magwood, he urges that this second-in-time habeas petition is the “first application” to challenge the “intervening judgment” and death sentence entered by the CCA.

13 In re Lampton, 667 F.3d at 588. 14 See 561 U.S. at 324–39. 15 See id. at 339. 16 Id.

4 Case: 16-70015 Document: 00513989592 Page: 5 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 The State disagrees, arguing that “Cathey’s 1997 death sentence has never been disturbed.” It asserts that in Texas, only the CCA has the authority to grant habeas relief, and it did not do so here. Because Cathey was denied relief, the State contends, he was never resentenced. The State thus concludes that “no new or intervening judgment has been entered in Cathey’s case since the time he filed his first federal habeas petition in 2004.” The determinative question is whether the 2014 CCA decision constitutes a new judgment under Magwood such that Cathey’s present habeas petition is not “second or successive” under § 2244. “While Magwood establishes that a habeas application challenging a ‘new judgment’ is not second or successive, it does not define the term ‘new judgment.’” 17 This Court has explained, “[w]hether a new judgment has intervened between two habeas petitions, such that the second petition can be filed without this Court’s permission, depends on whether a new sentence has been imposed.” 18 There was no formal resentencing here. “Texas trial courts only make recommendations to the Court of Criminal Appeals but do not rule on habeas petitions.” 19 Indeed, they lack the authority to grant relief on an Atkins claim or vacate a death sentence. 20 Cathey acknowledges there was no formal resentencing, but asks this Court to take a functional approach in determining whether there was a new sentence. In this vein, Cathey argues that the Texas

17 United States v. Jones, 796 F.3d 483, 485 (5th Cir. 2015). 18 In re Lampton, 667 F.3d at 588 (citation omitted). 19 Hatten v. Quarterman, 570 F.3d 595, 599 n.3 (5th Cir. 2009) (citing Ex parte Brown,

205 S.W.3d 538, 546 (Tex. Crim. App. 2006)); accord Moore v. Texas, 137 S. Ct. 1039, 1044 (2017) (“Under Texas law, the CCA, not the court of first instance, is ‘the ultimate factfinder’ in habeas corpus proceedings.” (citations omitted)). 20 See Ex parte Alexander, 685 S.W.2d 57, 60 (Tex. Crim. App. 1985) (“It is well

established that only the Court of Criminal Appeals possesses the authority to grant relief in a post-conviction habeas corpus proceeding where there is a final felony conviction. The trial court is without such authority.” (citations omitted)); accord Padieu v. Court of Appeals of Tex., Fifth Dist., 392 S.W.3d 115, 117 (Tex. Crim. App. 2013) (per curiam). 5 Case: 16-70015 Document: 00513989592 Page: 6 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 procedure “affords the petitioner the same ability to present new evidence in the trial court supporting his claim as a resentencing hearing would.” However, Magwood instructs that “[a] § 2254 petitioner is applying for something: His petition ‘seeks invalidation (in whole or in part) of the judgment authorizing the prisoner’s confinement.’” 21 Cathey argues that the 2014 CCA decision is the “most recent order by which Cathey’s punishment is currently authorized.” But if Cathey convinced a court to invalidate that decision, he would in all likelihood continue to be detained as Texas trial courts do not have authority to grant habeas petitions. We are thus persuaded that there was no intervening judgment here because the 2014 CCA decision is not the one authorizing his confinement. The district court’s transfer order is affirmed.

III. In the alternative, Cathey seeks permission to file a successive habeas petition. We find that he has made the requisite prima facie showing and therefore grant his motion for authorization to file a successive petition. Paramount to this decision is the standard of review at this stage and the process that follows. 22 Our grant of a motion to file a successive petition is “tentative in the following sense: the district court must dismiss the motion that we have allowed the applicant to file, without reaching the merits of the motion, if the court finds that the movant has not satisfied the requirements for the filing of such a motion. The district court then is the second gate through which the petitioner must pass before the merits of his or her motion are heard.” 23 “The district court must conduct a thorough review to determine if

21 Magwood, 561 U.S. at 332 (quoting Wilkinson v. Dotson, 544 U.S. 74, 83 (2005)). 22 See In re Morris, 328 F.3d 739, 740–41 (5th Cir. 2003) (per curiam). 23 Id. at 741 (citations and internal quotation marks omitted).

6 Case: 16-70015 Document: 00513989592 Page: 7 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 the motion conclusively demonstrates that it does not meet AEDPA’s second or successive motion requirements.” 24 Under 28 U.S.C. § 2244(b)(2): A claim presented in a second or successive habeas corpus application under section 2254 that was not presented in a prior application shall be dismissed unless-- (A) the applicant shows that the claim relies on a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable; or (B)(i) the factual predicate for the claim could not have been discovered previously through the exercise of due diligence; and (ii) the facts underlying the claim, if proven and viewed in light of the evidence as a whole, would be sufficient to establish by clear and convincing evidence that, but for constitutional error, no reasonable factfinder would have found the applicant guilty of the underlying offense. Cathey argues that his Atkins claim satisfies subsection (A). To satisfy this subsection, Cathey must make a “prima facie showing” that: (1) “his Atkins claim was not presented” in a prior application; (2) his Atkins claim “relies on a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable”; and (3) his Atkins claim has merit. 25 “Our court has adopted the following definition of prima facie showing: We understand it to be simply a sufficient showing of possible merit to warrant a fuller exploration by the district court.” 26 “If we determine that it appears ‘reasonably likely’ that the motion and supporting documents

24 Id. (citations and internal quotation marks omitted). 25 In re Campbell, 750 F.3d 523, 530 (5th Cir. 2014) (internal quotations and citations omitted); accord In re Henderson, 462 F.3d 413, 415 (5th Cir. 2006) (per curiam). 26 In re Campbell, 750 F.3d at 530 (quoting In re Morris, 328 F.3d at 740); accord In

re Wood, 648 F. App’x 388, 390 (5th Cir. 2016) (per curiam) (unpublished) (“In other words, this court should not, at this stage, rule on the merits, but merely determine whether [the petitioner’s] claim deserves further exploration by the district court.”). 7 Case: 16-70015 Document: 00513989592 Page: 8 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 indicate that the application meets the ‘stringent requirement’ for the filing of a successive petition, then we must grant the filing.” 27 Cathey has made a sufficient showing to proceed to a fuller review, though “[w]e express no view on whether [Cathey] will or ultimately should prevail on his claim.” 28 The State does not dispute the first element, that Cathey’s Atkins claim was “not presented” in his prior federal habeas petition. 29 It does, however, dispute that Cathey’s claim relies on a previously unavailable new rule of constitutional law and that Cathey’s claim has merit. A. Disputes over the second element with respect to Atkins claims are infrequent because “[t]here is no question that Atkins created a new rule of constitutional law . . . made retroactive to cases on collateral review by the Supreme Court.” 30 And in the typical case, the petitioner has filed his habeas petition before Atkins was decided, making an Atkins claim “previously unavailable.” 31 Here, however, Atkins was decided on June 20, 2002, yet Cathey filed his original federal habeas petition (without raising an Atkins claim) on April 2, 2004. Cathey acknowledges that Atkins had been decided by the time he filed his first federal petition, but argues that the rule was “previously unavailable to [him] under the circumstances.” He identifies two circumstances that rendered Atkins practically unavailable in April 2004: Courts did not consider the Flynn Effect until at least 2005. Between February and April 2004, as Mr. Cathey was filing his

27 In re Woods, 155 F. App’x 132, 135 (5th Cir. 2005) (per curiam) (unpublished) (citations omitted). 28 In re Mathis, 483 F.3d 395, 399 (5th Cir. 2007). 29 See In re Campbell, 750 F.3d at 530 (citation omitted). 30 Id. (citations omitted). 31 See, e.g., id. (citations omitted); In re Henderson, 462 F.3d at 414–15; In re Brown,

457 F.3d 392, 396 (5th Cir. 2006). 8 Case: 16-70015 Document: 00513989592 Page: 9 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 first federal habeas petition, Texas first articulated its standards for Atkins claims. At the time, a 77 IQ test score was not perceived to be within the range for Atkins-level intellectual functioning. Indeed, a review of published cases reveals that only one applicant in the country with an unadjusted IQ score above 75 brought an Atkins claim before Mr. Cathey filed his first federal habeas petition in 2004—a claim that was unsuccessful. What is more, it was not until after Mr. Cathey filed his second state habeas petition in 2008 that the State disclosed other evidence in its possession suggesting that Mr. Cathey’s true IQ was at most 73 and, in fact, well within Atkins range. 32 The Flynn Effect “is a phenomenon positing that, over time, standardized IQ test scores tend to increase with the age of the test without a corresponding increase in actual intelligence in the general population. Those who follow the Flynn effect adjust for it by deducting from the IQ score a specified amount for each year since the test was normalized.” 33 Cathey avers that, in 2004, he did not know about the “problem of aging norms” nor “the State’s evidence of a lower IQ score,” and thus “had no reason to pursue an Atkins claim that nobody else had won and only one person had even tried.”

32 Footnotes and citations omitted. 33 Wiley v. Epps, 625 F.3d 199, 203 n.1 (5th Cir. 2010), as revised (Nov. 17, 2010) (citation omitted). This Court has routinely declined to address Flynn Effect arguments, typically reciting some version of the following: “the Flynn Effect ‘has not been accepted in this Circuit as scientifically valid.’” E.g., Gray v. Epps, 616 F.3d 436, 446 n.9 (5th Cir. 2010) (quoting In re Mathis, 483 F.3d at 398 n.1). Importantly, however, nor has the Flynn Effect been rejected. In Brumfield v. Cain, this Court agreed it had “not recognized the Flynn effect,” but found it “not necessary to decide whether to recognize the Flynn effect in this case.” 808 F.3d 1041, 1060 n.27 (5th Cir. 2015), cert. denied, 136 S. Ct. 2411 (2016) (mem) (citations omitted). Similarly, in Wiley v. Epps, this Court declined to address “what, if any, impact [the Flynn Effect] has in this case.” 625 F.3d at 210; accord In re Salazar, 443 F.3d 430, 433 n.1 (5th Cir. 2006) (per curiam) (assuming for argument that the Flynn Effect is valid, but “express[ing] no opinion as to whether this is actually the case”). At this juncture, we follow that pattern. It is not necessary to accept or reject the Flynn Effect’s validity to determine whether Atkins was previously unavailable to Cathey in 2004. We also note the Eleventh Circuit’s recent conclusion that district courts, upon their consideration of the expert testimony, may apply or reject the Flynn Effect, which is a finding of fact reviewed for clear error. See Ledford v. Warden, Georgia Diagnostic & Classification Prison, 818 F.3d 600, 640 (11th Cir. 2016); see also Walker v. True, 399 F.3d 315, 322–23 (4th Cir. 2005) (directing district court to consider Flynn Effect evidence). 9 Case: 16-70015 Document: 00513989592 Page: 10 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 Cathey further argues that denying his motion would encourage “kitchen-sink petitions,” in that “it would require habeas petitioners to raise any and all imagined or theorized legal bases for habeas relief—whether grounded in fact or not—out of fear that those claims would be later foreclosed even in the light of developments in the law or facts.” The State responds that this Court’s decision in Mathis v. Thaler 34 forecloses Cathey’s claim. The State argues that, like Cathey, “Mathis filed his first federal habeas petition in April 2003, after Atkins was decided in June 2002, yet the petition did not include an Atkins claim.” 35 Several years later, Mathis sought authorization to file a successive petition raising an Atkins claim. Although Mathis did not file his first petition until after the issuance of Atkins, he argued that the Atkins rule was “previously unavailable” to him. This Court disagreed and concluded that Mathis “offer[ed] no cogent argument to excuse his failure to include his Atkins claim in his first federal petition when that claim was available to him for nine months after Atkins was decided.” 36 The State presses us to do the same here. Cathey responds that Mathis was a case of “intentional withholding of a viable and available ground” for relief, whereas Cathey “had little basis under the law as it existed at the time he filed his first federal habeas petition to conclude that he had a claim that arose under the Supreme Court’s Atkins doctrine.” Further, with respect to the Flynn Effect, the State argues that the legal availability of a claim “does not depend on its prior success in lower courts.” The State, citing psychology articles, asserts that, in any event, the Flynn Effect was recognized for at least twenty years before Cathey’s first federal petition, so Cathey could have learned of it and premised his Atkins claim on

34 616 F.3d 461 (5th Cir. 2010). 35 Id. at 467. 36 Id. at 473.

10 Case: 16-70015 Document: 00513989592 Page: 11 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 it in his first federal petition. The State points to Rivera v. Dretke, 37 which, in 2006, mentioned the issue of rising IQ scores in relation to an Atkins claim. Moreover, though the State acknowledges prison worksheets that reference an IQ of “below 73,” it notes that the documents do not explain the basis for that score. The State also disputes that it withheld the worksheets, suggesting that they were available upon the proper request. This Court has had few occasions to analyze whether a rule of constitutional law was “previously unavailable” for purposes of a successive habeas petition when the pertinent Supreme Court decision was published at the time of the petitioner’s initial habeas petition. In the recent unpublished case of In re Wood, Wood filed his initial federal habeas petition on May 6, 2002, before Atkins, but amended his petition on October 2, 2002, after Atkins. 38 However, “Wood did not raise an Atkins claim in the amended petition, nor did he seek to amend the petition a second time to include an Atkins claim.” 39 Only later when Wood attempted to file a successive writ did he raise an Atkins claim, asserting it was previously unavailable to him. 40 This Court considered “whether a rule was ‘available’ if, as in Wood’s case, it was announced while a defendant’s first federal habeas petition was pending.” 41 After noting that courts faced with this issue had not adopted categorical rules, this Court “adopt[ed] the [Eleventh Circuit’s] feasibility standard,” which “takes into account the particular circumstances of the previous habeas proceeding: ‘[i]f the new rule was announced while the original § 2254 petition was pending the applicant must demonstrate that it was not feasible to amend

37 No. Civ. B-03-139, 2006 WL 870927 (S.D. Tex. Mar. 31, 2006), aff'd in part, vacated in part sub nom. Rivera v. Quarterman, 505 F.3d 349 (5th Cir. 2007). 38 In re Wood, 648 F. App’x at 389. 39 Id. at 390. 40 See id. at 389. 41 Id. at 391.

11 Case: 16-70015 Document: 00513989592 Page: 12 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 his or her pending petition to include the new claim.’” 42 Applying this standard, this Court found that Wood had “not made a prima facie showing of unavailability,” 43 because he had “not demonstrated that his representation in the initial federal habeas proceedings was so deficient as to render the Atkins claim functionally unavailable; nor ha[d] he given any other explanation that could excuse his failure to amend his petition to include an Atkins claim and seek a stay and abeyance thereof.” 44 The Wood decision essentially adopted a rebuttable presumption that a new rule of constitutional law was previously available if published by the time a district court ruled on a petitioner’s initial habeas petition. Mathis v. Thaler supports this understanding. In Mathis, this Court rejected Petitioner Mathis’s arguments that Atkins was “previously unavailable” when Atkins was published before Mathis filed his initial habeas petition. 45 Mathis argued “that had he attempted to exhaust his Atkins claim in state court prior to filing his first federal habeas application, he risked forfeiting federal review of his previously exhausted claims . . . [and that] if he had pursued his Atkins claim in a successive state habeas petition, he also risked forfeiting federal review of his exhausted claims” because of uncertainty in the law as it applied to AEDPA’s statute of limitations. 46 The Court rejected both arguments. In doing so, it stated that “Mathis offer[ed] no cogent argument to excuse his failure to include his Atkins claim in his first federal petition[.]” 47 Similar to Wood, the Mathis Court was guided by the dates of the Supreme Court decision and initial habeas filing, but did not endorse a strict

42 Id. (quoting In re Everett, 797 F.3d 1282, 1288 (11th Cir. 2015) (other citation omitted)). 43 Id. at 392 (citations omitted). 44 Id. 45 Mathis, 616 F.3d at 467, 473. 46 Id. at 467–68 (footnote omitted). 47 Id. at 473.

12 Case: 16-70015 Document: 00513989592 Page: 13 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 rule—likely recognizing the potential for a gray area of previous unavailability despite technical availability. This case falls into that gray area. At this preliminary stage, we find that Cathey has presented sufficiently “cogent argument[s]” 48 that Atkins was previously unavailable at the time of his first petition and its disposition. At that time, Cathey believed his IQ score to be 77—outside of the range that was then understood to satisfy the subaverage intellectual functioning prong of an Atkins claim. In Atkins, the Supreme Court held that the Eighth Amendment prohibited the execution of intellectually disabled criminals. 49 The Atkins Court noted that 70 to 75 or lower is “typically considered the cutoff IQ score for the intellectual function prong of the mental retardation definition.” 50 However, it left “to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.” 51 The Texas courts “follow[ed] an American Association on Mental Retardation (AAMR) definition of mental retardation, adopted by [the Texas Court of Criminal Appeals] in Ex parte Briseno,” which required “‘significantly subaverage’ general intellectual functioning . . . usually evidenced by an IQ ‘of about 70’ or below[.]” 52 This Court has observed this baseline score of 70 when analyzing Atkins claims by Texas petitioners. 53 Indeed, the Briseno court

48 Id. 49 See Atkins, 536 U.S. at 321. 50 Id. at 309 n.5 (citation omitted). 51 Id. at 317 (internal quotation marks, citation, and footnote omitted). 52 Ex parte Woods, 296 S.W.3d 587, 589 (Tex. Crim. App. 2009) (citation and footnote

omitted); accord Ex parte Hearn, 310 S.W.3d 424, 430 (Tex. Crim. App. 2010) (“[T]his court interprets the ‘about 70’ language of the AAMR’s definition of mental retardation to represent a rough ceiling, above which a finding of mental retardation in the capital context is precluded.” (citations omitted)); Ex parte Sosa, 364 S.W.3d 889, 891 (Tex. Crim. App. 2012); Busby v. Davis, No. 15-70008, 2017 WL 396549, at *3 (5th Cir. Jan. 27, 2017) (unpublished) (“Texas developed its Atkins framework in Ex parte Briseno.” (citing Ex parte Briseno, 135 S.W.3d 1 (Tex. Crim. App. 2004) abrogated by Moore, 137 S. Ct. 1039)). 53 In re Mathis, 483 F.3d at 397 (“Typically, a person’s IQ must be measured at 70 or

below to qualify as mentally retarded.” (citing In re Hearn, 418 F.3d 444, 447 n. 4 (5th Cir. 13 Case: 16-70015 Document: 00513989592 Page: 14 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 ultimately found that the petitioner’s IQ scores of 72 and 74 did not satisfy the “significantly subaverage general intellectual functioning” prong. 54 Briseno was published two months before Cathey filed his initial federal habeas petition. 55 At that point in time, Cathey had no reason to believe his known score of 77—outside even the 70 to 75 range cited by the Atkins Court and sometimes others 56—would satisfy an Atkins claim. 57 Even the State calls an IQ score of 77 “a score above the range considered indicative of intellectual disability.” The State nevertheless suggests that Cathey could have raised an Atkins claim based on a Flynn Effect-argument. Cathey in turn argues that the Flynn Effect was not recognized as viable in the courts at that time. By our review, the first mention of “Flynn Effect” in the case law—and the only mention before Cathey filed his initial habeas petition—was in a 2003 Western District of Virginia district court case. 58 In Walton v. Johnson, the petitioner argued the Flynn Effect inflated his IQ scores. 59 The district court explained the Flynn Effect, but ultimately noted that, even if it applied it, the petitioner’s IQ score would still be too high. 60 We are not persuaded that this single case out of the

2005))); Blue v. Thaler, 665 F.3d 647, 657 (5th Cir. 2011); Morris v. Dretke, 413 F.3d 484, 489 (5th Cir. 2005). 54 Ex parte Briseno, 135 S.W.3d at 14. 55 Briseno was published on February 11, 2004. 56 Ex parte Modden, 147 S.W.3d 293, 298 (Tex. Crim. App. 2004) (noting “the 70–75

score that generally indicates subaverage general intellectual functioning”). 57 Nor was there any reason to believe otherwise before December 23, 2004, when the

district court denied habeas relief, so he could amend his petition as per Wood. See 648 F. App’x at 392. 58 Walton v. Johnson, 269 F. Supp. 2d 692, 699 n.5 (W.D. Va. 2003), judgment vacated

by 407 F.3d 285 (4th Cir. 2005), affirming judgment of district court en banc, 440 F.3d 160 (4th Cir. 2006). 59 Id. The petitioner had received full-scale IQ scores on the Wechsler Adult

Intelligence Scale-Revised (WAIS-R) test of 90, 77, and 69. Id. at 695. 60 The district court applied the Flynn Effect to his score of 90, the only score before

his eighteenth birthday, and found his adjusted score of 85 would still be too high. Id. at 699 n.5. We find the next mention out of the Eleventh Circuit in In re Hicks, 375 F.3d 1237 (11th 14 Case: 16-70015 Document: 00513989592 Page: 15 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 Western District of Virginia demonstrates that an Atkins argument based on the Flynn Effect was “available” to Cathey, who believed his IQ score to be 77. Not only was the Flynn Effect discussion in Walton relegated to a footnote, but whether Cathey was intellectually disabled for Atkins purposes was, and is, to be informed by Texas law, not Virginia law. 61 We find the first mention of “Flynn Effect” out of a Texas court, or federal court applying Texas law, in In re Salazar. 62 In Salazar, the petitioner’s expert opined that the petitioner’s IQ score may have been inflated by the Flynn Effect. 63 This Court noted that the expert did “not indicate what effect it would have had on [the petitioner’s] score in particular or even whether it is appropriate to adjust an individual’s score based on this theory.” 64 It then found that the petitioner’s IQ score was still too high even applying the Flynn Effect. 65 The State argues that Salazar “did not render Atkins newly available, and . . . does not represent a new rule of constitutional law recognized by the Supreme Court.” The State is correct. But Cathey does not argue that Salazar contains a new rule of constitutional law for § 2244(b)(2) purposes; Cathey points to Salazar to support his contention that the courts in April of 2004 had not considered the Flynn Effect, thus rendering any Atkins claim premised on that effect unviable. Rivera v. Dretke, 66 which the State cites to, does not alter the landscape. In Rivera, a district court decision noted the phenomenon of the Flynn Effect, 67 but the decision was released after Salazar. Moreover, in

Cir. 2004) (by the court). In Hicks, “Flynn Effect” is mentioned in evidence cited to by the dissenting judge. Id. at 1242–43 (Birch, J., dissenting). 61 See Atkins, 536 U.S. at 317. 62 443 F.3d 430, 433 (5th Cir. 2006). 63 Id. The petitioner scored an 87 on the WAIS-R test. Id. 64 Id. (footnote omitted). 65 Id. at 433 n.1 (noting “readjusted score would range from 80.7 to 74.4, both of which

are above the cutoff score of 70”). 66 2006 WL 870927. 67 2006 WL 870927, at *14.

15 Case: 16-70015 Document: 00513989592 Page: 16 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 Rivera, the petitioner presented an IQ score of 68, 68 whereas Cathey was sitting on an IQ score of 77. The State argues that Cathey “provides no support for excusing the failure to properly raise an available claim simply because the claim is meritless.” This argument assumes the conclusion: that claims are “available” despite being meritless. We think a claim must have some possibility of merit to be considered available. In the same way we would not expect someone who, based on evidence, believed he was nineteen-years-old at the time of his crime to bring a Roper claim, 69 we cannot expect someone who, based on evidence, believed his IQ was 77 to bring an Atkins claim two years after Atkins was decided in a state that had declared 70 as the benchmark IQ score, 70 even accounting for a five-point margin of error. 71 For similar reasons, the State’s contention that a claim’s legal availability “does not depend on its prior success in lower courts” is not sound in the context of this particular Atkins claim. That Atkins was “previously unavailable” is bolstered by evidence that came to light in 2010 that suggests Cathey’s IQ is “below 73.” Citing to the Atkins trial transcript, Cathey explains that a “Service Investigation Worksheet” indicating an IQ score of “below 73” was brought to his counsel’s attention by the State. At the hearing, Captain Steven Bryant, of the Polunsky Unit of the Texas Department of Criminal Justice (“TDCJ”) Correctional Institution Division, testified that when a prisoner first arrives, an IQ test is

68 Id. at *14, *26. 69 In Roper v. Simmons, the Supreme Court held that the death penalty was unconstitutional for juvenile offenders under the age of 18. 543 U.S. 551, 568, 578 (2005). 70 Ex parte Hearn, 310 S.W.3d at 428 (“Determining whether one has significantly

subaverage intellectual functioning is a question of fact. It is defined as an IQ of about 70 or below.” (footnote omitted)). 71 Blue, 665 F.3d at 658 (“[Ex parte Hearn, 310 S.W.3d 424 (Tex. Crim. App. 2010)]

establishes that, under Texas law, the lack of a full-scale IQ score of 75 or lower is fatal to an Atkins claim.” (citations omitted)). 16 Case: 16-70015 Document: 00513989592 Page: 17 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 given. 72 Upon being given the worksheet from 1998, Captain Bryant agreed that the form indicated Cathey’s IQ was below 73. In fact, there are two such worksheets. The State explains that “[t]he documents . . . are TDCJ Service Investigation Work Sheets, which are completed when an inmate is charged with a prison disciplinary infraction.” The worksheets, from 1998 and 2006, indicate Cathey was appointed counsel to assist him in the disciplinary process because, among other reasons, his “IQ [was] below 73.” Unable to dispute the existence of these worksheets, the State instead argues that the worksheets do not state the basis for the “below 73” comment. The parties also dispute whether the State withheld the documents, but both agree that they came to light in 2010 around the time of the state court hearing. Ultimately, Cathey has presented evidence that Cathey’s IQ was “below 73,” which, if true, would make an Atkins claim available where it previously was not. “Below 73” puts Cathey in the margin of error of 70. 73 Indeed, his IQ score may be even lower should a reviewing court ultimately find merit in the Flynn Effect. 74 This, in combination with the courts’ treatment of the Flynn Effect in 2004, is sufficient to make a prima facie case of previous unavailability that merits fuller

72 We note that other cases confirm that IQ tests are performed in prisons. Busby, 2017 WL 396549, at *4 n.25 (prison records of Texas inmate “show that he was administered an ‘unknown’ IQ test in 2001”); Rivera, 2006 WL 870927, at *2 (noting evidence including “[p]rison records indicating that [petitioner] had been given [IQ] tests while in prison (for a previous unrelated conviction)”). 73 Ex parte Hearn, 310 S.W.3d at 428 (“There is ‘a measurement error of approximately

5 points in assessing IQ,’ which may vary from instrument to instrument. Thus, any score could actually represent a score that is five points higher or five points lower than the actual IQ.” (citations omitted)). 74 In Ledford, the Eleventh Circuit surveys the state of Flynn Effect arguments in

various circuits. 818 F.3d at 635–37. It also notes that the DSM-V references the Flynn Effect, stating, “[i]n its only reference to the Flynn effect, the DSM–V provides: ‘Factors that may affect [intelligence] test scores include practice effects and the “Flynn effect” (i.e., overly high scores due to out-of-date test norms).’” Id. at 638 (citing American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (5th ed. 2013) at 37). 17 Case: 16-70015 Document: 00513989592 Page: 18 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 exploration in the district court. 75 In the face of this evidence, we are not prepared to foreclose Cathey’s petition before the district court’s second, “thorough review.” 76 As per Wood, Cathey has made a prima facie showing that it was not feasible to amend his initial petition to include an Atkins claim, 77 and has provided an “explanation that could excuse his failure to amend his petition to include an Atkins claim and seek a stay and abeyance thereof.” 78 We further note that Mathis v. Thaler, to which the parties frequently cite, concerned whether Mathis satisfied § 2244(b)(2)’s requirements, 79 not the preliminary issue we face now whether a prima facie case has been shown. This distinction is important. In order to file a successive writ in the district court, this Court

75 See In re Morris, 328 F.3d at 741 (Higginbotham, J., concurring) (expressing doubt but joining in grant of motion to file successive writ given “tentative” process). We recognize the difficulty in pinpointing precisely when Cathey’s Atkins claim became available, but we need not determine that. The statute only requires that the constitutional rule was “previously unavailable” at the time of Cathey’s initial habeas finding. See Mathis, 616 F.3d at 467 (“The issue before us is whether Mathis has demonstrated that his Atkins claim was ‘previously unavailable’ at the time he filed his first federal habeas application.” (citing 28 U.S.C. § 2244(b)(2); In re Salazar, 443 F.3d at 431–32. Cathey filed his first federal habeas petition on April 2, 2004. 76 In re Morris, 328 F.3d at 740–41 (internal quotation marks and citations omitted).

This Court has held “that a petitioner cannot bring a successive claim” under subsection (B) “where he does not assert that the newly discovered evidence would negate his guilt of the offense of which he was convicted, i.e., capital murder.” In re Webster, 605 F.3d 256, 257 (5th Cir. 2010); accord In re Sparks, 657 F.3d 258, 260 n.2 (5th Cir. 2011) (per curiam); Turner v. Epps, 460 F. App’x 322, 330 (5th Cir. 2012) (per curiam) (unpublished) (applying Webster to § 2244(b)(2)). Cathey does not—and cannot—allege that the IQ scores establish that he is not guilty of capital murder. The argument that “eligibility” for capital punishment ought not be governed by the rules on successive writs, though not without purchase, appears to be foreclosed. See In re Webster, 605 F.3d at 259–60 (Wiener, J., concurring). However, notwithstanding the distinction sometimes drawn between a legal and factual basis of a claim, In re Davis, 121 F.3d 952, 956 (5th Cir. 1997), to our knowledge, this Court has not held that a petitioner is barred from moving to file a successive claim under subsection (A) where evidence supports his argument that a new rule of constitutional law is available where it previously was not, which is what Cathey does here. 77 In re Wood, 648 F. App’x at 391. 78 Id. at 392. 79 See Mathis, 616 F.3d at 464–65.

18 Case: 16-70015 Document: 00513989592 Page: 19 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 must first determine whether the petitioner has made “a sufficient showing of possible merit to warrant a fuller exploration by the district court.” 80 “If we grant the motion, the district court must conduct its own independent review of whether or not [Cathey] has met the requirements of § 2244(b). The district court is, therefore, the ‘second gate through which the petitioner must pass before the merits of his or her motion are heard.’” 81 This Court granted Mathis’s motion for authorization upon finding he made the prima facie showing. 82 Only after returning to the district court did the district court dismiss Mathis’s petition for failing to satisfy § 2244(b)(2), 83 which this Court then affirmed. 84 All we decide now is that Cathey has made “a sufficient showing of possible merit” 85 that Atkins established a new rule of constitutional law previously unavailable to him; he must still prove it in the district court. B. We now turn to the third and final element—whether Cathey’s Atkins claim has merit. 86 In Texas, intellectual disability was defined “as a disability 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.” 87 This formulation originated in Ex parte Briseno, which “adopted the definition of, and standards for assessing,

80 In re Campbell, 750 F.3d at 530 (quoting In re Morris, 328 F.3d at 740); accord In re Arnick, 826 F.3d 787, 789 (5th Cir. 2016) (per curiam) (Elrod, J., dissenting) (describing this Court’s review of successive motions as “modest”). 81 In re Mathis, 483 F.3d at 397 (citations omitted). 82 Id. at 399–400. 83 Mathis, 616 F.3d at 466. 84 Id. at 473. 85 In re Campbell, 750 F.3d at 530 (quoting In re Morris, 328 F.3d at 740). 86 See id.; In re Henderson, 462 F.3d at 415. 87 Hunter v. State, 243 S.W.3d 664, 666 (Tex. Crim. App. 2007) (citing Ex parte Briseno,

135 S.W.3d at 7). 19 Case: 16-70015 Document: 00513989592 Page: 20 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 intellectual disability contained in the 1992 (ninth) edition of the American Association on Mental Retardation (AAMR) manual, predecessor to the current [American Association on Intellectual and Developmental Disabilities (AAIDD)]-11 manual. Briseno incorporated the AAMR–9’s requirement that adaptive deficits be ‘related’ to intellectual-functioning deficits.” 88 In order to determine relatedness, courts looked to “seven evidentiary factors”—which came to be known as the Briseno factors. 89 In Moore v. Texas, the Supreme Court reviewed Texas’s method for determining intellectual disability, with a focus on the Briseno factors. In June of 2016, Cathey moved for a stay pending the outcome in Moore. We denied his motion, noting that “[i]f—during our review of the filings—we determine that Moore may prove controlling, we retain the right to hold our decision and wait for the Supreme Court’s guidance.” This is what we did. In Moore, the CCA had rejected a Texas habeas court’s recommendation “that the CCA reduce Moore’s sentence to life in prison or grant him a new trial on intellectual disability.” 90 “In the CCA’s view, the habeas court erroneously employed intellectual-disability guides currently used in the medical community rather than the 1992 guides adopted by the CCA in Ex parte Briseno.” 91 However, the Supreme Court vacated the CCA’s judgment, finding that the Briseno factors—“[n]ot aligned with the medical community’s information, and drawing no strength from [the Court’s] precedent”—could not be used to restrict someone from being deemed intellectually disabled. 92

88 Moore, 137 S. Ct. at 1046 (internal citations and footnote omitted). 89 Id. (internal quotation marks omitted); accord Ex parte Briseno, 135 S.W.3d at 8 (“There are . . . some other evidentiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder[.]”). 90 Moore, 137 S. Ct. at 1046 (citation omitted). 91 Id. (citations omitted). 92 Id.

20 Case: 16-70015 Document: 00513989592 Page: 21 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 Indeed, although he would have affirmed the CCA’s decision on other grounds, Chief Justice Roberts observed that the Court unanimously agreed that the Briseno “factors are an unacceptable method of enforcing the guarantee of Atkins, and that the CCA therefore erred in using them to analyze adaptive deficits.” 93 The Court did not announce what should replace the Briseno factors; however, it observed that “[t]he medical community’s current standards supply one constraint on States’ leeway in [enforcing Atkins’s holding],” and faulted the CCA for “rejecting the habeas court’s application of medical guidance.” 94 The state habeas court used “the generally accepted, uncontroversial intellectual-disability diagnostic definition” of “(1) intellectual-functioning deficits . . . ; (2) adaptive deficits (‘the inability to learn basic skills and adjust behavior to changing circumstances[]’); and (3) the onset of these deficits while still a minor.” 95 These same factors guide our determination of whether Cathey’s Atkins claim has merit. Vacating the decision below, the majority in Moore: (1) held that the Briseno factors adopted by the Texas Court of Criminal Appeals for evaluating an Atkins claim are based on “superseded [medical] standards,” 96 that “creat[e] an unacceptable risk that persons with intellectual disability will be executed,” in violation of the Eighth Amendment, and thus “may not be used . . . to restrict qualification of an individual as intellectually disabled,” 97 (2) reiterated “that the Constitution ‘restrict[s] . . . the State’s power to take the life of’ any intellectually disabled individual” and that “[e]xecuting intellectually disabled individuals . . . runs up against a national consensus against the practice . . .

93 Id. at 1053 (Roberts, C.J., dissenting). 94 Id. at 1053 (majority opinion). 95 Id. at 1045 (internal citations omitted). 96 Id. at 1052. 97 Id. at 1044.

21 Case: 16-70015 Document: 00513989592 Page: 22 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 and creates a ‘risk that the death penalty will be imposed in spite of factors which may call for a less severe penalty,’” 98 (3) explained that, although “‘the task of developing appropriate ways to enforce’ the restriction on executing the intellectually disabled” is left to the States, “States’ discretion is not . . . ‘unfettered,’” and “[e]ven if ‘the views of medical experts’ do not ‘dictate’ a court’s intellectual-disability determination, . . . the determination must be ‘informed by the medical community’s diagnostic framework,’” 99 (4) stated that Supreme Court precedent precludes “disregard of current medical standards,” 100 (5) held that “[t]he CCA’s conclusion that Moore’s IQ scores established that he is not intellectually disabled is irreconcilable with Hall,” which instructs that, “where an IQ score is close to, but above, 70, courts must account for the test’s ‘standard error of measurement,’” and that “[b]ecause the lower end of Moore’s score range falls at or below 70, the CCA had to move on to consider Moore’s adaptive functioning” 101 and (6) held that “[t]he CCA’s consideration of Moore’s adaptive functioning . . . deviated from prevailing clinical standards and from the older clinical standards the court claimed to apply.” 102 “A prima facie showing of mental retardation is simply a sufficient showing of possible merit to warrant a fuller [exploration] by the district court.” 103 Importantly, “the state court findings concerning the Atkins claim are wholly irrelevant to our inquiry as to whether [the petitioner] has made a prima facie showing of entitlement to proceed with his federal habeas

98 Id. at 1048 (quoting Atkins, 536 U.S. at 320–21). 99 Id. (quoting Hall, 134 S. Ct. at 1998, 2000). 100 Id. at 1049. 101 Id. (quoting Hall, 134 S. Ct. at 1995, 2001). 102 Id. at 1050. 103 In re Hearn, 418 F.3d at 445 (internal quotation marks and citations omitted).

22 Case: 16-70015 Document: 00513989592 Page: 23 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 application, which is an inquiry distinct from the burden that [the petitioner] must bear in proving his claim in the district court.” 104 1. The first prong, intellectual-functioning deficits, is typically “indicated by an IQ score ‘approximately two standard deviations below the mean’—i.e., a score of roughly 70—adjusted for ‘the standard error of measurement[.]” 105 Briseno and its progeny used the same numerical baseline. 106 Adjusting for the five-point standard error of measurement, as we must, 107 Cathey’s known IQ score of 77 results in a range of 72 to 82. However, the Hall 108 Court reminded that “[i]ntellectual disability is a condition, not a number. Courts must recognize, as does the medical community, that the IQ test is imprecise.” 109 Cathey has offered evidence that suggests such imprecision in his score of 77. First, Cathey has offered evidence of the consequences of the Flynn Effect, an accepted scientific phenomenon, 110 regardless of its ultimate status in the courts and its application to specific IQ scores. 111 Cathey urges that correcting for the Flynn Effect results in “a true IQ of 71.6,” which yields a range of 66.6 and 76.6 after accounting for the standard error of measurement. We need not

104 In re Wilson, 442 F.3d 872, 878 (5th Cir. 2006) (by the court) (citation omitted); accord In re Henderson, 462 F.3d at 415; In re Mathis, 483 F.3d at 397. 105 Moore, 137 S. Ct. at 1045 (quoting AAIDD–11, at 27). 106 See Ex parte Briseno, 135 S.W.3d at 7 n. 24; Ex parte Hearn, 310 S.W.3d at 428. 107 Moore, 137 S. Ct. at 1049 (“Hall instructs that, where an IQ score is close to, but

above, 70, courts must account for the test’s ‘standard error of measurement.’” (citation omitted)). 108 Hall v. Florida, 134 S. Ct. 1986 (2014). 109 Id. at 2001 (citation omitted). 110 See Ex parte Cathey, 451 S.W.3d at 12 (“The trial judge heard extensive evidence

concerning the ‘Flynn Effect,’ including testimony from Professor Flynn himself. It was generally agreed by all of the experts that the ‘Flynn Effect’ does exist and is valid.”); Moore, 137 S. Ct. at 1049 (“Hall indicated that being informed by the medical community does not demand adherence to everything stated in the latest medical guide. But neither does our precedent license disregard of current medical standards.”). 111 See supra note 33.

23 Case: 16-70015 Document: 00513989592 Page: 24 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 credit this specific output to acknowledge that the Flynn Effect raises the inference that Cathey’s score of 77 from a 1996 IQ test last normed in 1978 may be susceptible to inflation. This inference is viable not because we today decide to accept the Flynn Effect—we need not reach that issue—but because Cathey has also presented documents from 1998 and 2006 that note his IQ is “below 73.” This inference of inflation is buttressed by results from achievement tests administered by Dr. Yohman, and school records that reflect low to failing grades in ninth grade and dropping out the following year. The Hall Court warned that “[a] State that ignores the inherent imprecision of [IQ] tests risks executing a person who suffers from intellectual disability.” 112 That risk lurks here. The State contends that in Blue v. Thaler, this Court rejected an argument to adjust IQ scores of 76 and 77 down based on the Flynn Effect. But Blue never mentions “Flynn Effect,” and further that case considered whether to issue a COA, not whether to allow a successive petition. 113 Further still, whereas “Blue [did] not produce[] an IQ score within the parameters serving as a precursor to a diagnosis of mental retardation[,]” 114 Cathey has produced evidence suggestive of an IQ score “below 73.” The State also argues that “the only issue in this case is whether it was unreasonable for the state court to decline to adjust Cathey’s IQ score to account for the Flynn Effect.” But this is incorrect. As previously stated, “the state court findings concerning the Atkins claim are wholly irrelevant to our inquiry as to whether [the petitioner] has made a prima facie showing of entitlement to proceed with his federal habeas application, which is an inquiry

112 Hall, 134 S. Ct. at 2001 (citation omitted). 113 Blue, 665 F.3d at 652. 114 Id. at 661 (citations and internal quotation marks omitted).

24 Case: 16-70015 Document: 00513989592 Page: 25 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 distinct from the burden that [the petitioner] must bear in proving his claim in the district court.” 115 Finally, the State avers that “[t]his Court has consistently denied relief . . . ‘when an inmate has IQ scores both under and over 70.’” But none of the cases the State cites to were at the same successive writ procedural posture as here, 116 and all of them predate the Supreme Court’s decision in Hall. In Hall, “the Florida Supreme Court ha[d] . . . held that a person whose test score is above 70, including a score within the margin for measurement error, does not have an intellectual disability and is barred from presenting other evidence that would show his faculties are limited.” 117 But the Supreme Court found that this “strict IQ test score cutoff” 118 was unconstitutional. 119 The Court found it “disregard[ed] established medical practice . . . [by] tak[ing] an IQ score as final and conclusive evidence of a defendant’s intellectual capacity, when experts in the field would consider other evidence[,]” and by “rel[ying] on a purportedly scientific measurement of the defendant’s abilities . . . while refusing to recognize that the score is, on its own terms, imprecise.” 120 The Supreme Court was focused on the standard error of measurement, not the Flynn Effect, but nevertheless indicated skepticism of IQ cut-offs. Given the procedural posture and evidence presented, we will not now resort to such a cut-off. Cathey has made “a sufficient showing of possible merit [of significant

115 In re Wilson, 442 F.3d at 878. 116 Pierce v. Thaler, 604 F.3d 197, 199 (5th Cir. 2010) (affirming denial of habeas relief and evidentiary hearing on Atkins claim); Eldridge v. Quarterman, 325 F. App’x 322, 329 (5th Cir. 2009) (unpublished) (denying COA); Williams v. Quarterman, 293 F. App’x 298, 316 (5th Cir. 2008) (per curiam) (unpublished) (affirming denial of habeas relief on Atkins claim). 117 Hall, 134 S. Ct. at 1994 (citation omitted). 118 Id. 119 Id. at 2000. 120 Id. at 1995. The Court noted that its analysis did not include states that used

cutoffs of 75 or greater, as the petitioner did not challenge those rules. Id. at 1996. 25 Case: 16-70015 Document: 00513989592 Page: 26 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 subaverage intellectual functioning] to warrant a fuller exploration by the district court.” 121 2. We also find Cathey has made a prima facie showing of adaptive deficits. As recently explained by the Supreme Court, “Briseno adopted the definition of, and standards for assessing, intellectual disability contained in the 1992 (ninth) edition of the American Association on Mental Retardation (AAMR) manual, predecessor to the current AAIDD–11 manual.” 122 In Briseno, the CCA noted that “[i]mpairments in adaptive behavior” were defined as “significant limitations in an individual’s effectiveness in meeting the standards of maturation, learning, personal independence, and/or social responsibility that are expected for his or her age level and cultural group, as determined by clinical assessment and, usually, standardized scales.” 123 However, the Briseno Court found “[t]he adaptive behavior criteria [to be] exceedingly subjective” and thus offered “some other evidentiary factors which factfinders in the criminal trial context might also focus upon in weighing evidence as indicative of mental retardation or of a personality disorder[.]” 124 These evidentiary factors came to be known as the “Briseno factors.” 125 Further, “Briseno incorporated the AAMR–9’s requirement that adaptive deficits be ‘related’ to intellectual-functioning deficits.” 126

121 In re Campbell, 750 F.3d at 530 (quoting In re Morris, 328 F.3d at 740). 122 Moore, 137 S. Ct. at 1046 (citing Briseno, 135 S.W.3d, at 7; other citation omitted)). 123 Ex parte Briseno, 135 S.W.3d at 7 n.25 (citing American Association on Mental

Deficiency at 11). 124 Id. at 8. These include, for example, “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?” and “Has the person formulated plans and carried them through or is his conduct impulsive?” Id. 125 See, e.g., Ex parte Sosa, 364 S.W.3d at 892–93. 126 Moore, 137 S. Ct. at 1046 (citation and footnote omitted).

26 Case: 16-70015 Document: 00513989592 Page: 27 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 In Moore, the Supreme Court criticized the CCA for its continued reliance on the standard set out in Briseno: The CCA . . . fastened its intellectual-disability determination to “the AAMR’s 1992 definition of intellectual disability that [it] adopted in Briseno for Atkins claims presented in Texas death- penalty cases.” By rejecting the habeas court’s application of medical guidance and clinging to the standard it laid out in Briseno, including the wholly nonclinical Briseno factors, the CCA failed adequately to inform itself of the “medical community's diagnostic framework[.]” 127

The Moore Court noted that “current manuals offer ‘the best available description of how mental disorders are expressed and can be recognized by trained clinicians’” 128 and also cited the latest AAIDD manual for its definition of intellectual disability. It found that “[i]n determining the significance of adaptive deficits, clinicians look to whether an individual’s adaptive performance falls two or more standard deviations below the mean in any of the three adaptive skill sets (conceptual, social, and practical).” 129 Turning to those standards here, in the category of conceptual skills, Cathey points to evidence suggesting his difficulties with language, using money, and reading and writing. In the category of social skills, Cathey points to evidence of his gullibility, his lack of self-esteem, and his difficulties with relationships. And in the category of practical skills, Cathey points to his challenges completing chores, impairments in assessing risks, and difficulties keeping steady work. Furthermore, Cathey highlights evidence from Dr. Jack Fletcher, who administered the Vineland Adaptive Behavior Scales test to

127 Id. at 1053 (internal citations omitted). 128 Id. (quoting DSM–5, at xli; other citations omitted). 129 Id. at *6 (citing AAIDD–11, at 43); accord Ladd, 748 F.3d at 645–46 (“With respect

to limitations in adaptive functioning, the CCA has explained that ‘three adaptive-behavior areas are applicable to determining mental retardation: conceptual skills, social skills, and practical skills.’” (quoting Ex parte Hearn, 310 S.W.3d at 428)). 27 Case: 16-70015 Document: 00513989592 Page: 28 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 analyze Cathey’s adaptive behaviors. As part of this test, Dr. Fletcher interviewed Cathey’s older sister and former wife. From these interviews, Dr. Fletcher concluded that Cathey’s adaptive behavior composite score was in the first percentile or less. 130 The State responds that its expert, Dr. Tim Proctor, disagreed with aspects of Dr. Fletcher’s administration of the Vineland. For instance, Dr. Proctor noted concerns with applying the Vineland retrospectively, and noted concerns with the subjects’ credibility. These may be valid critiques, but at this stage they are insufficient for us to completely discount Dr. Fletcher’s conclusions. The State also argues that Dr. Proctor “reviewed Cathey’s prison correspondence and testified that Cathey’s correspondence demonstrated, inter alia, an awareness of his Atkins claim, an ability to plan, an understanding of current events, an ability to manage money, and an ability to think abstractly.” The State specifically notes Dr. Proctor’s testimony “that an inmate’s prison behaviors are relevant in assessing the inmate for intellectual disability.” However, Moore signaled restraint in using such evidence: “the CCA stressed Moore’s improved behavior in prison. Clinicians, however, caution against reliance on adaptive strengths developed ‘in a controlled setting,’ as a prison surely is.” 131 Ultimately, we find that Dr. Fletcher’s testing and conclusions regarding Cathey’s adaptive deficits, in combination with the affidavits and school records, satisfy a prima facie case of adaptive deficits.

130 From his interview with Cathey’s sister, Dr. Fletcher determined a composite score of 66, and scored 66 in socialization, 68 in daily living, and 69 in communication. From his interview with Cathey’s ex-wife, Dr. Fletcher determined a composite score of 59, and scored 61 in communication and daily living, and 60 in socialization. 131 Moore, 137 S. Ct. at 1050 (citations omitted).

28 Case: 16-70015 Document: 00513989592 Page: 29 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 3. Finally, Cathey has presented sufficient evidence to show the onset of these deficits prior to age 18. Cathey argues he “suffered numerous serious head traumas during his childhood, a risk factor for intellectual disability.” Cathey also contends that “[i]mpaired care-giving and adult non- responsiveness are also risk factors for mental retardation.” 132 Cathey notes his “traumatic environment” growing up, including drug-dealing, gun battles, and prostitution taking place in the home. Cathey urges that these are risk factors identified by the American Association on Intellectual and Developmental Disabilities. Cathey concludes that there is “[n]o evidence of an intervening cause after age 18 . . . that could account for [his] limitations in intellectual and adaptive functioning.” The State responds that Cathey was not in special education classes, and that his “poor school performance was attributable to factors other than his intelligence.” The State asserts that before Atkins, no one thought Cathey was intellectually disabled, pointing again to the CCA opinion denying Cathey’s Atkins claim. The State argues that much of Cathey’s evidence is “readily contradicted” by trial testimony. However, “[a]t his pre-Atkins trial, [Cathey] had little reason to investigate or present evidence relating to intellectual disability. In fact, had he done so at the penalty phase, he ran the risk that it would ‘enhance the likelihood . . . future dangerousness [would] be found by the jury.’” 133 Given the evidence Cathey presented and that there is no identifiable intervening cause, Cathey has made a sufficient showing of onset before the

132 See also Moore, 137 S. Ct. at 1051 (“[T]raumatic experiences . . . count in the medical community as ‘risk factors’ for intellectual disability. Clinicians rely on such factors as cause to explore the prospect of intellectual disability further, not to counter the case for a disability determination.” (citations omitted)). 133 Brumfield, 135 S. Ct. at 2281 (quoting Atkins, 536 U.S., at 321).

29 Case: 16-70015 Document: 00513989592 Page: 30 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 age of 18 to proceed. “Cases in which this court has denied a motion to file a successive habeas claim based on Atkins usually involve sparse records, where no expert has diagnosed the movant with retardation.” 134 Here, Cathey has put forth a variety of evidence, including Dr. Fletcher’s test result and conclusions, family member affidavits, and school and hospital records. “[F]rom our vantage, the evidence shows more than sufficient ‘possible merit to warrant a fuller exploration by the district court.’” 135 IV. “[T]he Constitution ‘places a substantive restriction on the State’s power to take the life’ of a mentally retarded offender.” 136 Cathey has offered sufficient evidence that he may be excluded from the death penalty for this reason. At this stage, he has made a prima facie showing that his successive petition satisfies 28 U.S.C. § 2244(b)(2)(A), and we accordingly grant his motion for authorization. We remind that such a grant is “tentative” and is to be followed by the district court’s “thorough review.” 137 In addition to undertaking a thorough review on the question of successiveness, there are procedural questions concerning the timeliness of Cathey’s claim that the district court has not yet decided, and should decide in its review. Like the petitioner in Campbell, Cathey first raised an Atkins claim several years after Atkins was actually decided. 138 Cathey argues that his

134 In re Mathis, 483 F.3d at 398 (citation omitted); accord In re Brown, 457 F.3d at 396–97 (failing to make prima facie showing of intellectual disability when report relied upon indicates an IQ “significantly above the range of [intellectual disability],” and report “does not demonstrate deficits in specific areas of adaptive functioning” (citations omitted)); In re Johnson, 334 F.3d 403, 404 (5th Cir. 2003) (per curiam) (two letters from forensic psychologist referring to “multiple areas of concern,” prior evaluation that “did not clearly reflect mental incapacitation,” and a “belief” of verbal intelligence levels, along with seventh grade transcript reflecting academic failure insufficient). 135 In re Campbell, 750 F.3d at 532 (citations omitted). 136 Atkins, 536 U.S. at 321 (quoting Ford v. Wainwright, 477 U.S. 399, 405 (1986)). 137 In re Morris, 328 F.3d at 741 (citations and quotation marks omitted). 138 In re Campbell, 750 F.3d at 532.

30 Case: 16-70015 Document: 00513989592 Page: 31 Date Filed: 05/11/2017

No. 16-20312 Cons w/ No. 16-70015 motion is nevertheless timely because the State withheld the “below 73” IQ evidence, and alternatively based on equitable tolling. As in Campbell, “[t]he delay gives us pause,” but further factual development is needed. 139 The district court is best positioned to initially resolve the statute of limitations issue. 140 For the reasons stated above, we AFFIRM the district court’s transfer order and GRANT the motion for authorization.

139 Id. at 533. 140 See id. at 534; see also In re Henderson, 462 F.3d at 417. 31