Butler, Steven Anthony

Opinion

IN THE COURT OF CRIMINAL APPEALS

OF TEXAS

NO. WR-41,121-02 EX PARTE STEVEN ANTHONY BUTLER, Applicant

ON APPLICATION FOR A WRIT OF HABEAS CORPUS

IN CAUSE NO. 511112 FROM THE

185^TH DISTRICT COURT OF HARRIS COUNTY

Price, J., filed a dissenting statement in which Johnson, J., joined.

DISSENTING STATEMENT

On June 27, 2007, we denied post-conviction habeas corpus relief to this applicant, ⁽¹⁾ rejecting his Atkins claim that he cannot be executed consonant with the Eighth Amendment because he is mentally retarded. ⁽²⁾ Without explicitly adopting the recommended findings of fact and conclusions of law of the convicting court, we nevertheless agreed with the lower court that the application should be denied on the merits, and we did so without further comment. ⁽³⁾ The applicant filed a federal petition for writ of habeas corpus in federal district court. That court, in a published opinion, rejected the applicant's challenge to our judgment in the state habeas proceedings, under the highly deferential criteria of the Antiterrorism and Effective Death Penalty Act [AEDPA], but was sufficiently troubled to grant the applicant a certificate of appealability to the Fifth Circuit. ⁽⁴⁾

The State's expert at the Atkins state habeas writ hearing was Dr. George Denkowski, a licensed psychologist with experience in diagnosis and treatment of mental retardation. It was largely on the basis of Dr. Denkowski's input that the convicting court was able to recommend finding against the applicant with respect to all three of the diagnostic criteria for mental retardation: general intellectual functioning, adaptive functioning, and onset before age 18. ⁽⁵⁾ Although we did not adopt those findings, this Court's summary rejection of the applicant's Atkins claim was surely based in equal measure on Denkowski's testimony because, without it, it is fairly clear that the applicant presented evidence that would have established mental retardation at least by a preponderance of the evidence. At the evidentiary hearing, the applicant presented ample evidence, including expert testimony, that would have served to establish all three prongs of the diagnostic criteria. Having reviewed the transcript of the 2006 evidentiary hearing, I, for one, would readily have found that the applicant demonstrated mental retardation to the requisite level of confidence--but for Denkowski's rebuttal testimony. ⁽⁶⁾

Since we rejected the applicant's Atkins claim in 2007, Denkowski's diagnostic practices have come under considerable professional scrutiny. In April of last year, he entered into a settlement agreement, in proceedings that were brought against him by the Texas State Board of Examiners of Psychologists with the State Office of Administrative Hearings, in which he agreed to discontinue forensic evaluations for mental retardation in Atkins cases. ⁽⁷⁾ The applicant subsequently persuaded the Fifth Circuit to stay his appeal of the federal district court's denial of his federal habeas petition to allow him to seek reconsideration in this Court of the denial of relief in view of the settlement agreement. While the Texas Rules of Appellate Procedure do not contemplate the filing of a motion for rehearing following the denial of a post-conviction application for writ of habeas corpus, ⁽⁸⁾ we are authorized, and on occasion have exercised our authority, to revisit final judgments in such matters, on our own motion, under extraordinary circumstances. ⁽⁹⁾ We did so in this case, remanding the cause to the convicting court "to allow it the opportunity to re-evaluate its initial findings, conclusions, and recommendation in light of the Denkowski Settlement Agreement." ⁽¹⁰⁾

The convicting court has apparently refused to avail itself of this opportunity. Our remand order invited the convicting court to "order affidavits or hold a live hearing if warranted." ⁽¹¹⁾ Accordingly, the applicant offered new affidavits and requested a hearing. The convicting court made no ruling on these matters. According to the applicant (although we have no official record of it), the convicting court simply announced from the bench that it would not "reconsider its Atkins ruling." ⁽¹²⁾ Instead, the convicting court "directed the District Attorney to submit a new set of proposed findings of fact and conclusions of law." ⁽¹³⁾ Apparently, the convicting court did not invite the applicant to do the same. ⁽¹⁴⁾ The convicting court then signed the State's Proposed Findings of Fact and Conclusions of Law on Remand "without a single change." ⁽¹⁵⁾ Noting that the Denkowski settlement agreement "does not contain a finding that Dr. Denkowski improperly evaluated the applicant for mental retardation[,]" the recommended findings and conclusions on remand make reference to the settlement agreement "only for historical background of the applicant's case and . . . not . . . for purpose of relitigating the issue of the applicant's claim of mental retardation." Otherwise, what we have now received is an only-slightly-reworked version of the original forty-eight page recommended findings of fact and conclusions of law that we refused to adopt back in 2007. At practically every point in the original findings of fact at which the convicting court had previously made a recommended finding expressly "based on Dr. Denkowski's credible affidavit," the recommended findings of fact on remand now simply delete the word "credible." ⁽¹⁶⁾ What are we to make of this? Does the convicting court now recommend that we once again make these numerous specific findings of fact on the basis of Denkowski's affidavit--even though the convicting court apparently no longer recommends that we regard that affidavit as "credible"? Beyond these amendments, the convicting court has done little more than to add a phrase toward the end of its recommended findings of fact by which it proposes that we find that the applicant has failed to establish the three diagnostic criteria for mental retardation "even absent the testimony and evidence elicited from Dr. George Denkowski during the habeas proceedings[.]" Neither the amendments themselves nor the process by which they were made inspire confidence.

General Intellectual Functioning: With respect to the first prong of the diagnostic criteria for mental retardation, Denkowski persuaded the convicting court that the results of several IQ tests, including one that he himself had conducted, ⁽¹⁷⁾ should be discounted in favor of an IQ test conducted when the applicant was a teenager that reaped an IQ score of 80, less than two standard deviations below the mean of 100. Contrary to the testimony of the applicant's experts, Denkowski maintained that even application of the so-called "Flynn Effect" would not lower this particular score to below two standard deviations below the mean, disagreeing with the applicant's experts (and, apparently, with diagnostic convention) with respect to the precise numerical reduction that ought to apply (.13 of a point per year, as opposed to .3, times the number of years since the particular testing instrument was last normed). The convicting court now recommends that we find that "neither the Fifth Circuit Court of Appeals nor" this Court has "recognized the Flynn Effect as scientifically valid." But neither has this Court rejected the validity of the Flynn Effect and, like the federal district court, we are therefore "left with the evidentiary record in this case." ⁽¹⁸⁾ Here, all of the experts, including Denkowski, acknowledged the diagnostic legitimacy of the Flynn Effect, disagreeing only with respect to how it should apply to adjust the applicant's scores. As the federal district court observed, "[o]n this point the literature did not support Dr. Denkowski's reduced discount factor," and did support that of the applicant's experts. ⁽¹⁹⁾

Denkowski also convinced the convicting court that the applicant's score on a certain achievement test indicated that the higher IQ score of 80 was the more reliable, despite testimony from the applicant's experts that the data support a judgment that individuals with IQ scores in the mild mental retardation range are capable of comparatively high scores on achievement tests and that there is no direct correlation between IQ scores and achievement test scores. Finally, he convinced the convicting court--again, against convention, if the applicant's experts are to be credited--that the best measure of IQ is not always the full-scale IQ score at all, but may instead be something called the Perceptual Organization Index, which is apparently a sub-test of the WAIS-III IQ test that Denkowski himself had administered to the applicant to obtain a full-scale IQ score of 69. In Ex parte Hearn, we held that the kind of clinical judgment that Denkowski seems to have utilized to inflate the applicant's IQ cannot serve "as a replacement for full-scale IQ scores in measuring intellectual functioning." ⁽²⁰⁾

The federal district court was clearly troubled by the convicting court's reliance on Denkowski's "heavily disputed opinions." ⁽²¹⁾ It deferred to those opinions only because Denkowski was, after all, "qualified as an expert in mental retardation," and because the applicant had not satisfied the onerous burden to refute his opinions by clear and convincing evidence as required by the AEDPA. ⁽²²⁾ That Denkowski has agreed to discontinue his Atkins assessments in light of the many professional complaints that have been lodged against him suggests to me that we ought to discount the convicting court's new recommended findings of fact and conclusions of law with respect to the applicant's general intellectual functioning precisely to the extent that they continue to rely, explicitly or implicitly, on Denkowski's "heavily disputed opinions." ⁽²³⁾

Adaptive Deficits: The same may be said for the convicting court's reliance on Denkowski's testimony with respect to the second prong of the diagnostic criteria for mental retardation, adaptive deficits. There are at least two aspects of Denkowski's input with respect to adaptive deficits that we should reject outright. First, the convicting court found that two of the standardized instruments that were utilized in this case--one by the applicant's main expert and the other by Denkowski himself--should not be taken into account because they are not normed for the long-term incarcerated such as the applicant. That left only the Vineland Adaptive Behavior Scales. ⁽²⁴⁾ This instrument was administered by the applicant's expert to yield a composite score of 39, with the mean being 100. ⁽²⁵⁾ The convicting court discounted this score, however, on the basis of Denkowski's assurances that a score this low would indicate an individual with profound mental retardation, which the applicant clearly is not. ⁽²⁶⁾ In accepting Denkowski's judgment in this regard, the convicting court flatly ignored testimony from several of the applicant's experts to the effect that the Vineland is a very good instrument for discriminating those with borderline adaptive abilities from those with at least mild mental retardation, but that it has far less power to distinguish between levels of mental retardation itself. The applicant's main expert readily conceded that the applicant's score of 39 was "probably a little too low," but he nevertheless believed it to be a more than reliable indication that the applicant had at least two areas of significant adaptive deficits. A second of the applicant's experts readily agreed. We should reject the convicting court's recommendation to credit Denkowski's testimony over that of the applicant's experts in the wake of the subsequent controversy surrounding Denkowski's diagnostic practices.

Second, also with respect to the second prong, the convicting court expressly credited Denkowski's testimony that evidence of criminal adaptability may properly be taken into account in arriving at an ultimate clinical judgment with respect to adaptive deficits. But, according to the applicant's experts, this is decidedly not the conventional view. In any event, in rejecting the applicant's substantial evidence of adaptive deficits, both Denkowski and the convicting court focused inordinately on the applicant's relative strengths rather than ruling out manifest weaknesses in at least two adaptive skills areas. As I argued two years ago, albeit in an unpublished dissent, this emphasis on adaptive strengths rather than adaptive weaknesses runs contrary to standard diagnostic protocol, which I believe the courts are obliged to follow in implementing Atkins. ⁽²⁷⁾

Onset Before 18: As for the third prong, onset before age 18, the convicting court simply concluded that, because the applicant could not establish either of the first two prongs at all, he had obviously failed to establish timely onset. By definition, this conclusion with respect to the third prong is every bit as suspect as the convicting court's conclusions with respect to the first two.

The evidence that remains after discounting Denkowski's influence convinces me that the applicant has established all three of the diagnostic criteria for mental retardation by a preponderance of the evidence and cannot constitutionally be executed under Atkins. Under these circumstances, I do not believe that the State's usual weighty interest in the repose and finality of its verdicts is sufficiently compelling to overcome the Eighth Amendment's aversion to executing a mentally retarded offender. ⁽²⁸⁾ With the controversy surrounding Denkowski's forensic methods, I do not think that, as the court of return and the ultimate arbiter of both the facts and the law in capital post-conviction habeas corpus proceedings, we can be sufficiently confident of our original judgment to allow it to stand. Nor can we rely on the federal courts to be our backstop--under the AEDPA, the federal courts are required to pay almost insurmountable deference to our credibility determinations. ⁽²⁹⁾ The applicant will get no opportunity to present additional evidence in federal court. ⁽³⁰⁾ I would reject the convicting court's recommended findings of fact and conclusions of law as insupportable and instead find that the applicant has satisfied his burden to establish that he is a person suffering mental retardation and is therefore immune from execution under the Eighth Amendment. Failing that, if the Court is determined to pay deference to the convicting court's recommendations, we should at least take up the applicant's suggestion that we remand the case once again to the convicting court for a more searching reevaluation than it conducted on the original remand. Because the Court does neither, I respectfully dissent.

FILED: June 27, 2012

PUBLISH

1. Ex parte Butler, No. WR-41,121-02, 2007 WL 1847377 (Tex. Crim. App., delivered June 27, 2007) (not designated for publication).

2. Atkins v. Virginia, 536 U.S. 304 (2002).

3. See Ex parte Butler, supra, at *1 ("The trial court held a hearing and made findings of fact and conclusions of law recommending that this application be denied because Applicant has failed to show that he is mentally retarded. We have reviewed the record of the hearing and the trial court's findings of fact and conclusions of law and agree that this application should be denied.")

4. Butler v. Quarterman, 576 F. Supp. 2d 805, 816-17 (S.D. Tex. 2008).

5. E.g., Ex parte Hearn, 310 S.W.3d 424, 428 (Tex. Crim. App. 2010).

6. In the context of post-conviction habeas corpus, the convicting court is the "original" fact-finder, and we ordinarily pay great deference to that court's findings of fact and conclusions of law when supported by the record. But that deference is not boundless, and we do not simply rubber-stamp the convicting court's recommendations. This Court is the "ultimate" fact-finder, with the prerogative to reject the convicting court's recommendations on those rare occasions when we deem it appropriate, even when they are supported by the record, if we think another disposition is manifestly better supported by the record. Ex parte Spencer, 337 S.W.3d 869, 879-80 n.1 (Tex. Crim. App. 2011) (Price, J., concurring); Ex parte Robbins, 360 S.W.3d 446, 467 n.14 (Tex. Crim. App. 2011) (Price, J., concurring).

7. See Maldonado v. Thaler, 625 F.3d 229, 234 (5^th Cir. 2010) (noting the proceedings brought against Denkowski in the State Office of Administrative Hearings); Pierce v. Thaler, 604 F.3d 197, 213 (5^th Cir. 2010) (noting that this Court had recently found Denkowski's credibility to be lacking in Ex parte Plata, No. AP-75,820, 2008 WL 151296 (Tex. Crim. App., delivered Jan. 16, 2008) (not designated for publication)).

8. Tex. R. App. P. 79.2(d).

9. Ex parte Moreno, 245 S.W.3d 419, 427-29 (Tex. Crim. App. 2008).

10. Ex parte Butler, No. WR-41,121-02, 2011 WL 6288411, at *1 (Tex. Crim. App., delivered Dec. 14, 2011) (not designated for publication).

11. Id.

12. Applicant's Objections to the District Court's Proposed Findings of Fact and Conclusions of Law (hereinafter "Objections") at 1.

13. Id. at 2.

14. Under Article 11.071, Sections 8 and 9, both parties "shall" file proposed findings of fact and conclusions of law for the convicting court to consider. Tex. Code Crim. Proc. art. 11.071, §§ 8(b) & 9(e). See Jefferson v. Upton, 130 S.Ct. 2217, 2223 (2010) ("Although we have stated that a court's verbatim adoption of findings of fact prepared by prevailing parties should be treated as findings of the court, we have also criticized that practice." (citation and internal quotation marks omitted)).

15. Objections at 2.

16. By my count, there are nineteen specific findings of fact that the convicting court originally recommended we make based on "the credible affidavit of Dr. Denkowski" that it now recommends we make based simply on "the affidavit of Dr. Denkowski[.]" In addition, two of the convicting court's originally recommended conclusions of law with respect to the first prong of the diagnostic criteria for mental retardation (general intellectual functioning) were based on, inter alia, "the credible affidavit of Dr. Denkowski[.]" The convicting court continues to recommend these conclusions of law to us, but has altogether removed any reference to Denkowski's affidavit.

17. Denkowski administered the Wechsler Adult Intelligence Scale, 3^rd edition (WAIS-III) to the applicant in 2006, obtaining a full-scale IQ score of 69.

18. Butler v. Quarterman, supra, at 814.

19. Id. at 814-15.

20. 310 S.W.3d at 431.

21. Butler v. Quarterman, supra, at 816.

22. Id.

23. Id.

24. In Ex parte Briseno, we acknowledged that adaptive behavior should be measured "by clinical assessment and, usually, standardized scales." 135 S.W.3d 1, 7 n.25 (Tex. Crim. App. 2004) (quoting American Association on Mental Deficiency (AAMD), Classification in Mental Retardation 1, 11 (Grossman ed. 1983)). The Vineland is one of the accepted scales. American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) 42 (4^th ed. 2000).

25. A significant limitation in adaptive functioning may be "defined by a score of at least two standard deviations below either (1) the mean in one of the three adaptive behavior skills areas or (2) the overall score on a standardized measure of conceptual, social, and practical skills." Ex parte Hearn, supra, at 428.

26. In its finding of fact number 157 on remand, the convicting court continues to find "that the applicant's score of 39 is equivalent to an IQ score of 39[.]" Going to the particular page of the habeas hearing record to which the convicting court cites, however, I find that Denkowski made no such assertion. Instead, he simply said that a composite score of 39 on the Vineland should be "interpreted just like an IQ score." By this I take Denkowski simply to mean that, in scoring the Vineland, the mean is 100 (just as the mean in an IQ test is 100), such that a score of 39 would fall more than four standard deviations below the mean--hence, the basis for agreement among all of the experts at the applicant's writ hearing that it was a very low score.

27. See Lizcano v. State, No. AP-75,879, 2010 WL 1817772, at *37 (Tex. Crim. App., delivered May 5, 2010) (Price, J., joined by Johnson & Holcomb, JJ., concurring and dissenting) (not designated for publication) (the presence of strengths in one or more adaptive skills areas does not counter-indicate mental retardation so long as weaknesses are identified in at least two others).

28. Ex parte Moreno, supra, at 429.

29. See 28 U.S.C. § 2254(d)(2) (federal habeas petition "shall not be granted" unless state court adjudication of the claim "was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding").

30. Cullen v. Pinholster, 131 S.Ct. 1388, 1401 (2011).