From Casetext: Smarter Legal Research

Ex parte Temple

COURT OF CRIMINAL APPEALS OF TEXAS
Nov 23, 2016
NO. WR-78,545-02 (Tex. Crim. App. Nov. 23, 2016)

Opinion

NO. WR-78,545-02

11-23-2016

EX PARTE DAVID MARK TEMPLE, Applicant


ON APPLICATION FOR A WRIT OF HABEAS CORPUS IN CAUSE NO. 1008763-A FROM THE 178TH DISTRICT COURT OF HARRIS COUNTY YEARY, J., filed a concurring opinion. CONCURRING OPINION

This is a post-conviction application for writ of habeas corpus challenging Applicant's conviction for the murder of his wife, brought pursuant to Article 11.07 of the Texas Code of Criminal Procedure. TEX. CODE CRIM. PROC. art. 11.07. The habeas court conducted an extensive evidentiary hearing on the writ application, spanning twenty-four days over the course of two-and-a-half months, and has recommended that we grant Applicant a new trial. We filed and set the application on a number of claims, including: 1) claims predicated on Brady v. Maryland, 373 U.S. 83 (1963), upon which the habeas court recommends that we grant relief; 2) claims of ineffective assistance of counsel under Strickland v. Washington, 466 U.S. 668 (1984); and 3) a claim of actual innocence under this Court's opinion in Ex parte Elizondo, 947 S.W.2d 202 (Tex. Crim. App. 1996). Ultimately, I would not grant relief on the basis of Applicant's Brady claims, as the trial court has recommended and the Court does today. Nevertheless, I would sustain one of Applicant's claims of ineffective assistance of trial counsel. In explaining why, I will recount many of the circumstances giving rise to Applicant's Brady claims because they set the backdrop for my conclusion that trial counsel's deficient performance undermines confidence in the outcome of Applicant's trial.

I. PROCEDURAL HISTORY

On the afternoon of January 11, 1999, Belinda Temple was murdered by a single shot to the back of her head with a twelve-gauge shotgun as she knelt in the walk-in closet off of the master bathroom of her home in Katy. She was seven months' pregnant at the time. When investigators with the Harris County Sheriff's Office discovered that her husband, Applicant, was having an extramarital relationship with one of his co-workers at one of the local high schools, he became a major focus of their investigation. In March and April of 1999, prosecutors took the case to a grand jury. They produced a dozen witnesses, but they did not ask the grand jury to return an indictment at that time. And indeed, the evidence against Applicant, while it raised serious suspicions, was not overly compelling. There was some question whether he had been at home at the time of the murder, since he was seen with his and Belinda's three-year-old son, ET (hereinafter, "ET"), on the security videotape of a local supermarket either as, or at least soon after, the murder is most likely to have taken place. Moreover, the grand jury also heard testimony about the Temples' next-door neighbor, RJS, III (hereinafter, "RJS"), a sixteen-year-old boy who was a student at the high-school where Belinda worked as a tutor and special-needs instructor. Belinda had recently reported to RJS's parents that RJS had been habitually skipping classes, among other minor infractions, and RJS admitted to the grand jury that he had been home alone, asleep on the couch, at the time of the murder.

A firearms examiner testified at trial that the diameter of the wadding from the shotgun shell used to murder Belinda was "consistent with a 12-gauge more than any other gauge" of shotgun shell. It has not been seriously disputed at any stage of these proceedings that the murder weapon was a twelve-gauge shotgun.

The time-stamp on a security video from the supermarket indicated that Applicant arrived there with ET at 4:32 p.m.

Belinda worked at a different high school than Applicant.

Five years later, in the summer of 2004, the case was called to the attention of a prosecutor in the cold-case division of the Harris County District Attorney's office. By this time, Applicant had married Heather Scott, the object of his 1999 extramarital relationship. Without presenting any new evidence to the grand jury, the prosecutor obtained a murder indictment against Applicant and had him arrested. When the defense attorney hired by Applicant's family almost immediately requested an examining trial for Applicant, the prosecutor responded by closing the State's file to him, as was the Harris County District Attorney's unofficial office policy at that time. From that point on, the prosecutor provided defense counsel with nothing more than the limited discovery to which Applicant was minimally entitled under the law. As for Brady material, the tenor of the testimony of both the prosecutor and her second chair at the writ hearing was that they felt duty bound to disclose exculpatory or impeaching information only if they believed it to be true; if they felt confident that the sheriff's investigators had satisfactorily ruled out an alternative suspect, they did not believe they were obligated to disclose information about the investigation of that alternative suspect to the defense under Brady. Thus, while Applicant's family conveyed to defense counsel their belief that RJS was a viable suspect, the State did nothing at any time prior to Applicant's trial to alert defense counsel to the full extent of law enforcement's investigation into RJS's possible involvement in Belinda's murder.

Applicant knew that Belinda was pregnant, of course. As of 1999, however, the Legislature had not yet provided that the murder of a pregnant woman was a capital offense. See Acts 2003, 78th Leg., ch. 823, § 2.01, p. 2608, eff. Sept. 1, 2003 (amending Penal Code Section 1.07(26)'s definition of "individual" to include "an unborn child at every stage of gestation from fertilization until birth"); Lawrence v. State, 240 S.W.3d 912, 915 (Tex. Crim. App. 2007) (recognizing the 2003 legislative amendment authorizing capital prosecution for the murder of a mother and her fetus).

Prior to the passage of the Michael Morton Act in 2013, "Texas law gave a defendant the right to no more discovery than due process requires." Gerald S. Reamey, The Truth Might Set You Free: How the Michael Morton Act Could Fundamentally Change Texas Criminal Discovery, Or Not, 48 TEX. TECH. L. REV. 893, 898 (2016).

Asked whether the State should reveal information about an alternative suspect prior to trial, the second chair prosecutor replied, "Not if you run it down and there's no validity to it, then it is not something that needs to be disclosed." The first chair prosecutor testified similarly that she was not obligated to turn over information seemingly favorable to the defense if in her estimation it was "ridiculous." She went so far as to assert that it was her "job" to "stand up for" an alternative suspect like RJS if she believed he was being "wrongfully accused." She insisted that "[t]here is a line you have to draw in your own mind ethically where you quit accusing a 16-year-old boy of committing a capital murder."

Applicant's trial commenced in October of 2007, lasting approximately five weeks. In his opening statement to the jury before any evidence was presented, defense counsel made no mention of an alternative suspect. Instead, he emphasized that Applicant simply could not have had enough time to perpetrate the murder given the time-line he expected the evidence to reveal. The State's theory of the case was that Applicant was motivated to murder Belinda in order to pave the way to marry Scott; and that he staged a burglary to make it appear as if she had been killed by an intruder while he was out running errands with ET. Somewhere along the way he successfully disposed of the twelve-gauge shotgun he used to kill Belinda, according to the State's theory, and no weapon was ever found that was definitively shown to be the murder weapon. Although Applicant had owned a shotgun as a teenager, the parties disputed the gauge of that shotgun and, in any event, it was undisputed that Applicant no longer possessed that particular shotgun at the time of the offense. Also critical to the State's case was the inference that any other perpetrator besides Applicant would have caused the Temples' dog, Shaka, an aggressive chow, to raise a ruckus in the back yard (the purported point of entry having been the back door) and alert the whole neighborhood. Applicant rebutted this inference with substantial evidence, including his own testimony, that he had left Shaka in the stand-alone garage while out running errands—not in the back yard, as the State's theory assumed.

Because ET was running a low fever on the day of the offense, Belinda left school early to fetch him from his day care center and take him home. She then called Applicant to come tend to ET while she returned to school for an important meeting. She was murdered after she returned home from that meeting, but not before she stopped at her in-laws' house to pick up a batch of homemade chicken soup for ET. By the time she arrived home, ETwas feeling much better. At trial, the parties stipulated that a subsequent psychological examination revealed "no evidence that ET had been a witness to the murder."

Defense counsel began to discover the scope of the State's investigation of RJS early in the course of the trial, as sheriff's investigators testified and defense counsel was permitted to review their offense reports in preparing for cross-examination. He was also allowed to review the earlier grand jury testimony of those trial witnesses who had testified before the 1999 grand jury. By the seat of his pants, he began to develop a supplemental defensive theory (not inconsistent with his original defensive theory) that—as he ultimately argued to the jury in his final summation at the guilt stage—the evidence that RJS committed the offense, while admittedly sketchy, was nevertheless more substantial than the evidence of Applicant's guilt.

Most importantly, defense counsel was able to show that RJS had access to a Harrington & Richardson (hereinafter, "H & R") break-open twelve-gauge shotgun. This shotgun belonged to RJS's father. When it was recovered, it contained an expended shotgun shell. Before it had been fired, this shell had been "reloaded"—that is to say, it was not as originally manufactured, but had been re-packed by hand. The State's experts had concluded that the shotgun shell used to murder Belinda had been similarly reloaded. RJS's father possessed a number of these reloaded shells after the murder, but those were found to contain wadding that was different than the wadding from the reloaded shell used to kill Belinda (and recovered from the floor of the walk-in closet where her body was found). In short, while it was never demonstrated that the H & R shotgun was the actual murder weapon, defense counsel was able to argue to the jury that RJS had access to the closest thing to the murder weapon that any investigation had yet revealed. Having learned as well during trial that Belinda had reported RJS's truancy to his parents, defense counsel was also able to suggest to the jury that RJS had a substantial, albeit not particularly compelling, motive to kill her.

To a limited extent, defense counsel used the late-disclosed offense reports to impeach the testimony of the sheriff's detectives who claimed ultimately to have been "satisfied" that RJS was not the perpetrator. Moreover, defense counsel used information gleaned from both the offense reports and RJS's own earlier grand jury testimony to cross-examine RJS when he took the stand as the State's only rebuttal witness at the end of the guilt phase of evidence to deny having killed Belinda. The jury found Applicant guilty after deliberating for more than a full day. It later assessed him a life sentence.

These offense reports revealed that RJS was questioned by various sheriff's investigators as many as seven times, giving a number of oral statements and two written statements. He was subjected to two polygraph tests, both of which showed signs of deception when he was asked about his involvement in Belinda's murder. He eventually refused to submit to a third polygraph. In spite of these circumstances as detailed in his offense report, Detective Charles Leithner pronounced himself "satisfied" at trial that RJS did not kill Belinda. Defense counsel was prevented from introducing evidence of RJS's failed polygraphs, and Applicant did not complain of this ruling on appeal. But defense counsel was permitted to show that RJS was questioned repeatedly by the investigators and that, at one point, Leithner told RJS's parents that RJS could not be ruled out as a suspect until his story could be corroborated.

The docket sheet reflects that the jury was retired to deliberate at 2:47 p.m. on November 14, 2007, excused for the evening at 4:45 p.m., and then returned its verdict the next day at 4:25 p.m.

The major thrust of Applicant's direct appeal was to challenge the sufficiency of the evidence, both legally and factually, to show that he murdered his wife. While his appeal was pending, however, this Court eliminated factual sufficiency from the Texas criminal justice lexicon, to the dismay of two members of the Fourteenth Court of Appeals who were inclined to believe that the evidence against Applicant was factually insufficient. See Temple v. State, 342 S.W.3d 572 (Tex. App.—Houston [14th Dist.] 2010). The court of appeals concluded that the evidence was legally sufficient, over the dissent of one justice on denial of rehearing en banc who believed otherwise. On petition for discretionary review, this Court ultimately affirmed the court of appeals's judgment that the evidence was legally sufficient. Temple v. State, 390 S.W.3d 341, 363 (Tex. Crim. App. 2013).

Justice Seymore authored the panel opinion holding the evidence legally sufficient. Temple, 342 S.W.3d at 581-619. But he also authored a separate concurring opinion, decrying the panel's acquiescence to the plurality and concurring opinions of this Court in Brooks v. State, 323 S.W.3d 893 (Tex. Crim. App. 2010), which a few months earlier had abandoned factual sufficiency review.

Justice McCally, who was not a member of the original panel, authored an impassioned dissent to the denial of rehearing en banc, urging the court of appeals to hold that the evidence was legally insufficient. Temple, 342 S.W.3d at 628-46. Justice Seymore weighed in again, writing a separate dissent to the denial of en banc rehearing. He disagreed with Justice McCally's conclusion that the evidence was legally insufficient, but he opined that, for essentially the same reasons that Justice McCally concluded that the evidence was legally insufficient, he would hold that it was factually insufficient. Id. at 646-59. He was joined by Justice Anderson.

The majority opinion of the court of appeals panel, the separate opinions on the denial of rehearing en banc, and the opinion of this Court on discretionary review, all engaged in lengthy and meticulous recitations of the evidence. Though I have studied both the entire record of the appeal as well as the lengthy hearing on Applicant's writ application, I do not recite the facts in quite the same level of detail in this opinion. For present purposes, I must limit my fact recitation to those details most pertinent to Applicant's particular post-conviction claims. For the full flavor of how painstaking the legal sufficiency analyses were in this case, however, I refer the reader to those various recitations. Our discretionary review was limited to Applicant's sufficiency claims.

Applicant also complained on direct appeal about the late disclosure of the information with respect to the State's investigation of the alternative suspect, RJS. He argued that the delayed disclosure violated his right to due process under Brady. The court of appeals rejected this point of error for two reasons. First, it held that Applicant had procedurally defaulted this claim. Temple, 342 S.W.3d at 591. Because defense counsel did not formally seek a continuance until three weeks after the offense reports began to come to light, well into his presentation of defensive evidence and near the end of the guilt phase of the trial, the court of appeals held that he forfeited his right to complain, under this Court's precedent in Wilson v. State, 7 S.W.3d 136, 146 (Tex. Crim. App. 1999). Id. Alternatively, the court of appeals held that, because Applicant was able in any event to effectively present the untimely disclosed facts, there was not " a reasonable probability that the outcome of the trial would have been different had the State disclosed these facts earlier." Id. at 592. Applicant did not challenge this holding in his petition for discretionary review, and we had no occasion to address it in our opinion.

The Court's opinion today does not address this aspect of the court of appeals' holding.

Applicant also claimed on appeal that the first chair prosecutor engaged in numerous instances of prosecutorial misconduct during the course of the trial. The court of appeals agreed with many though not all of these claims but concluded that, on balance, her misconduct did not operate to deprive Applicant of a fair trial. Temple, 342 S.W.3d at 592-619. These numerous claims were not embraced within the scope of our discretionary review, which was limited to Applicant's sufficiency claims. To a large extent, Applicant's present post-conviction writ application continues to dwell on the prosecutor's conduct and her character. While I do not condone some of the prosecutor's actions in this case, I find Applicant's focus on her conduct and character largely to distract from the genuine issue under Brady, namely, whether Applicant was ultimately deprived of favorable evidence that might realistically have made a difference to the outcome of his trial. After all, the bedrock "principle" that undergirds Brady "is not punishment of society for the misdeeds of a prosecutor but avoidance of an unfair trial to the accused." Brady, 373 U.S. at 87 (citing Mooney v. Holohan, 294 U.S. 103 (1935)).

Applicant raises a number of issues in his post-conviction application for writ of habeas corpus, most prominent of which is his renewed Brady claim. Unfortunately, in neither his writ application nor during the extensive habeas evidentiary hearing has Applicant adequately distinguished information that he claims was belatedly disclosed during trial (hereinafter, "late-disclosed evidence") from information that he alleges was not disclosed until after trial (hereinafter, "undisclosed evidence"). To a significant extent, he continues to complain of the late disclosure rather than of any non-disclosure. Even though defense counsel reviewed many of the offense reports during trial and put much of it to effective use, Applicant contends that there were other bits of exculpatory or impeaching evidence embedded in them that defense counsel simply missed because of the conditions under which he was compelled to review them. In essence, Applicant argues that, while defense counsel was able to put to some good use the late-disclosed information, he was not able to put it to its optimal use, as he would have had the information been revealed to him prior to trial. Defense counsel was handicapped, Applicant concludes, by having to investigate even as he was having to litigate.

To the extent that Applicant is now complaining of late-disclosure of the same information that he complained about on direct appeal, the court of appeals' disposal of that issue, which we had no occasion to examine on discretionary review, has become law of the case. Thus, the court of appeals' holdings that Applicant's Brady complaint was procedurally defaulted and that, alternatively, the particular information he claimed on appeal that the State suppressed was not material, those holdings have become binding. See George E. Dix & John M. Schmolesky, 43B TEXAS PRACTICE: CRIMINAL PRACTICE AND PROCEDURE § 59:13, at 822 (3rd ed. 2011) ("The doctrine of the 'law of the case' clearly limits . . . efforts to use habeas to relitigate issues already resolved on appeal. The direct appeal and subsequent application for habeas relief are the same 'case' for purposes of this doctrine, and consequently the law of the case doctrine generally bars reconsideration of issues of law resolved in the appeal.") (footnotes omitted). Neither Applicant nor the Court today attempts to explain why the court of appeals' holdings with respect to procedural default and materiality should not be final—at least to the extent that Applicant continues to complain of suppression of the same evidence that formed the basis of his appellate complaint.

At the writ hearing, defense counsel testified, for example, that he was given about an hour to review the nearly 100 page offense report of Detective Leithner while sitting in the courtroom with one of the prosecutors hovering over him. Defense counsel later introduced Leithner's offense report as an exhibit in support of his motion for continuance, and the trial court admitted it for record purposes, so it was a part of the appellate record in the court of appeals.

In its proposed findings of fact and conclusions of law, in which it has recommended that we grant relief on the basis of Brady, the habeas court has likewise largely failed to distinguish late-disclosed evidence from undisclosed evidence. The habeas court has nonetheless concluded, albeit "not without some doubt," that Applicant was "denied a fair trial because of the State's failure to disclose or timely disclose favorable evidence; and had that evidence been disclosed or disclosed timely, the results of the trial would have been different." For reasons I will develop at some length, I share the habeas court's doubt. Ultimately, though, while I agree that we should grant Applicant a new trial, I would not base our ruling on his Brady claims. Instead, I would base our holding on one of Applicant's claims of ineffective assistance of counsel (though that ineffectiveness may very well have been a regrettable by-product of the fact that defense counsel was forced to investigate—even as he was trying to litigate—the case).

The judge who presided over the writ hearing is not the same judge who presided over Applicant's trial. After we filed and set this cause for submission, the State filed a motion in this Court to supplement the record with an affidavit from the judge who presided over the trial in 2007, expressing his disagreement with the habeas court judge's recommendation that we grant Brady relief in this cause. Applicant, in turn, filed a motion to strike the State's motion. We denied both motions on May 18, 2016.

I take this phrase verbatim from the habeas court's findings of fact and conclusions of law.

II. BRADY

A. The Legal Standard

The United States Supreme Court has expanded upon its 1963 decision in Brady to hold that a defendant suffers a due process violation if the State or one of its surrogates, whether willfully or not, 1) fails to disclose evidence that 2) is favorable to the defense (either because it is exculpatory or because it impeaches) and 3) is material in the sense that, had it been timely disclosed to the defense, there is a reasonable probability that the result of the proceeding would have been different. Strickler v. Greene, 527 U.S. 263, 280-82 (1999). "Reasonable probability" in this context does not mean "the defendant would more likely than not have received a different verdict with the evidence," but instead means that, having been deprived of the evidence, the defendant did not receive "a fair trial, understood as a trial resulting in a verdict worthy of confidence." Kyles v. Whitley, 514 U.S. 419, 434 (1995). It is important to bear in mind that the materiality inquiry "is not a sufficiency of evidence test." Id. "A defendant need not demonstrate that after discounting the inculpatory evidence in light of the undisclosed evidence, there would not have been enough left to convict." Id. It is enough to establish a reasonable probability of a different result "when the [State's] evidentiary suppression 'undermines confidence in the outcome of the trial.'" Id. (citing United States v. Bagley, 473 U.S. 667, 678 (1985)). As part of this consideration, "the reviewing court may consider directly any adverse effect that the prosecutor's [non-disclosure] might have had on the preparation or presentation of the defendant's case." Thomas v. State, 841 S.W.2d 399, 405 (Tex. Crim. App. 1992) (quoting Bagley, 473 U.S. at 683 (plurality opinion)). Moreover, materiality is to be assessed "collectively, not item by item." Kyles, 514 U.S. at 436.

As the United States Supreme Court recently emphasized, under this standard, an applicant "can prevail even if . . . the undisclosed information may not have affected the jury's verdict." Wearry v. Cain, 136 S.Ct. 1002, 1006 n.6 (2016).

We have said that, to constitute a Brady violation, the State's suppression must have resulted in the denial of evidence to the defense that "would have been admissible at trial." Ex parte Miles, 359 S.W.3d 647, 665 (Tex. Crim. App. 2012). We have noted, however, that "the analysis might not end there" because, as the Fifth Circuit has held, "if inadmissible evidence would give rise to the discovery of other admissible evidence or witnesses, the State does have a duty to disclose that evidence." Id. at 699 n.22 (citing United States v. Brown, 650 F.3d 581, 588 (5th Cir. 2011)).

We have also held that, if late-disclosed evidence "was turned over in time for the defendant to use it in his defense, the defendant's Brady claim would fail." Little v. State, 991 S.W.2d 864, 866 (Tex. Crim. App. 1999). So long as the defendant "received the material in time to use it effectively at trial, his conviction should not be reversed just because it was not disclosed as early as it might have and should have been." Id. (citing United States v. McKinney, 758 F.2d 1036, 1050 (5th Cir. 1999)). See also United States v. Valas, 822 F.3d 228, 237 (5th Cir. 2016) (same). On direct appeal in Applicant's case, in an alternative holding to its conclusion that Applicant procedurally defaulted his Brady claim, the court of appeals determined that defense counsel was able to put the late-disclosed materials to effective use at Applicant's trial. Temple, 342 S.W.3d at 591-92. We do not ordinarily entertain claims in a post-conviction application for writ of habeas corpus that were previously resolved against an applicant on direct appeal. Ex parte Acosta, 672 S.W.2d 470, 472 (Tex. Crim. App. 1984); Ex parte Brown, 205 S.W.3d 538, 546 (Tex. Crim. App. 2006).

This is not to say that Applicant should now be altogether barred from raising a Brady claim in his post-conviction writ application. As is the case with claims of ineffective assistance of trial counsel, a habeas applicant may be able to re-raise a Brady claim that was rejected on direct appeal if he can present new evidence of its validity. Cf. Ex parte Nailor, 149 S.W.3d 125, 131 (Tex. Crim. App. 2004) ("[I]f the appellate court rejects a claim of ineffective assistance of counsel because the record on direct appeal does not contain sufficient information to adequately address and resolve a particular allegation of counsel's deficient performance, the defendant may re-urge consideration of that specific act or omission in a later habeas corpus proceeding if he provides additional evidence to prove his claim."); Ex parte Bryant, 448 S.W.3d 29, 34-5 (Tex. Crim. App. 2014) (quoting Nailor and applying it to hold that the applicant could re-raise his ineffective assistance of counsel claim in a writ since he brought new evidence to support it); Brown, 205 S.W.3d at 547 n.26. (citing Nailor in support of the Court's decision to remand to the convicting court "to give applicant an opportunity to present whatever 'new' evidence he had in support of his 'old' allegation" of ineffective assistance of counsel).

Thus, Applicant may presently be able to raise Brady in either or both of two ways. First, if there is additional exculpatory or impeaching material beyond that which was disclosed for the first time at his trial—in short, undisclosed evidence—then of course Applicant may raise a Brady claim in post-conviction habeas proceedings, since this would essentially constitute a new Brady claim. Second, Applicant may be able to re-raise a previous Brady claim. But to the extent that Applicant continues to complain about the late-disclosed exculpatory or impeaching evidence that was rejected on direct appeal on the grounds of immateriality, he must produce additional evidence—beyond what is apparent from the appellate record—in order to establish incremental materiality. What that means is that Applicant must now show that there was exculpatory or impeaching value to the late-disclosed information beyond that which would have been apparent to the court of appeals from the appellate record. And he must also show that this additional exculpatory or impeaching value would have created a reasonable probability of a different outcome—even taking into account the uses to which defense counsel was able to put those late-disclosed materials during the trial. In other words, in my view, Applicant must present a record in these habeas corpus proceedings to establish that defense counsel could have used the late-disclosed materials, had they been timely disclosed, to substantially greater exculpatory or impeaching effect at trial than he actually did. It is doubtful, in my opinion, that Applicant has established sufficient incremental materiality here.

Because this would constitute a new Brady claim, we would not be bound by law of the case with respect to the Brady claim that Applicant raised on direct appeal.

Here I am giving Applicant the benefit of the doubt that, if he can present new evidence to show incremental materiality, he will not be bound by law of the case, and can make his claims in post-conviction habeas corpus proceedings notwithstanding the court of appeals' holdings with respect to procedural default and materiality.

B. Applicant's Allegations of Undisclosed Brady Evidence

Most of what Applicant developed at the extensive writ hearing was additional evidence pertaining to the materiality of the late-disclosed information contained in the various offense reports. But from my own review of both the trial and writ records, I have also been able to parse out some evidence that appears not to have been disclosed even at trial. None of this undisclosed evidence, however, strikes me as particularly momentous. I shall briefly discuss the most prominent examples that I have gleaned from the voluminous habeas record.

The discussion that follows is meant to be illustrative, not necessarily exhaustive.

The School Witnesses: Another investigating officer, Detective Tracy Shipley, interviewed several witnesses at Belinda's school with respect to certain matters, including what time Belinda had left the school to return home on the day she was murdered. Defense counsel was given a copy of Shipley's offense report at trial which summarized these interviews. None of these witnesses could say precisely what time Belinda drove off. The timing of Belinda's departure was critical. The later Belinda left the school, the later she would have arrived, first at her in-laws' home to pick up some homemade soup, and then at her own home. This would have narrowed Applicant's window of opportunity to have staged the burglary, killed Belinda, cleaned himself up afterwards from the inevitable blow-back from the shotgun blast, and then loaded ET into his truck to leave the scene.

The record is unclear whether Applicant was wearing the same clothes when captured on the supermarket video that he had been wearing earlier in the day, before Belinda could have been shot. But there was defensive testimony at trial that the shotgun blast, even with the muzzle touching the back of Belinda's head, would have caused some blow back and sprayed the shooter with blood. Sheriff's investigators did not detect blood on the clothes Applicant was wearing after Belinda's murder.

These witness interviews were tape-recorded, however, and defense counsel was not made aware of the audio recordings during trial. Telephone records that were admitted at trial showed that Belinda called Applicant at about 3:30 p.m. Applicant claims that the audio recordings reveal that two of the school witnesses could establish that Belinda was still in the parking lot of her school when she spoke to Applicant on her phone. This was exculpatory, Applicant claims, because it demonstrated that she had not yet left the school as late as 3:30, when the phone records showed that this conversation took place.

The record does not support Applicant's contention. The writ record contains the actual audio recording of one of these two witnesses, Courtney Ferguson, which I have listened to. Ferguson does not say at any point in the interview that she ever saw Belinda talking on her phone. The audio recording of her testimony lacks the exculpatory value Applicant attributes to it. The recording of the statement of the other witness, Margaret Christen, is not in the writ record, so I cannot presently tell whether she ever said that she saw Belinda on the phone. Additionally, Applicant claims that both Ferguson and Christen identified a third witness, Denise Lavoris, who was present and who (as the habeas court finds) "would have helped the defense timeline." But the record reveals nothing about what Lavoris might have had to say about whether Belinda was on the phone with Applicant while still in the school parking lot. It is thus pure speculation to say that she would have been helpful to the defense at trial.

In its recommended findings of fact, the habeas court calls this witness "Stacy" Ferguson, but she is identified in the offense report and in the audio recording as "Courtney" Ferguson. Courtney Ferguson's recorded interview is contained in Applicant's Exhibit 180—and then again in Applicant's Exhibit 178. See note 25, post.

Applicant has alleged, and the habeas court recommends that we find, that Christen "saw [Belinda] talking to [Applicant] on her cell phone between 3:20 and 3:30 pm on the day of her murder." From the citations to the writ record, it appears that the recording of Christen's interview is supposed to be contained in Applicant's Exhibit 178. But this exhibit does not contain Christen's interview. Instead, it is a duplicate of the same interviews (including Ferguson's) contained in Applicant's Exhibit 180. The Court has taken the trouble to order the original exhibits from the district clerk and I have verified that we still have not been provided with the audio recording of Christen's interview. None of these witnesses testified at the writ hearing. Thus, there is nothing in the record to substantiate the habeas court's recommended finding. At trial, Christen simply testified that Belinda's meeting at the school lasted until "about 3:20 to 3:30[,]" but Christen did not mention Belinda making a call on her cell phone during that time, and she did not see what time Belinda left the campus.

FBI Profiler Report: The habeas court recommends that we find that the State "never produced an FBI report which profiled the possible killer." In December of 2000, an FBI profiler prepared a report that listed the likely "offender characteristics" of the perpetrator of Belinda's murder. Applicant now argues that defense counsel could have used this FBI report to bolster his final argument at trial that RJS was at least as likely a suspect as Applicant was because RJS better fit the FBI profile. However, at the writ hearing the State introduced a transcript, gleaned from defense counsel's own case file, of a telephone conversation between defense counsel and the prosecutor that occurred more than two-and-a-half years before trial. In this phone conversation, the prosecutor alluded to the FBI profile report and offered defense counsel an opportunity to review it. While the writ record does not show whether defense counsel took her up on this offer, the phone conversation belies any claim that the FBI profile report was suppressed. The record does not support the habeas court's recommended finding.

RJS's Juvenile Probation Status: During the writ hearing, both defense counsel and one of his associates who sat second chair during Applicant's trial complained that they were never told before or during trial that RJS was on juvenile probation when Belinda was murdered. Defense counsel maintained that he could have used this information to bolster his argument at trial that RJS had a motive to murder Belinda—to avoid having his probation revoked. A Harris County appellate prosecutor confirmed during his writ-hearing testimony that RJS was on juvenile probation at the time of the offense. Even so, Applicant does not now argue, nor did the habeas court recommend that we find, that RJS's status as a juvenile probationer at the time of the offense constitutes undisclosed Brady evidence.

During questioning by one of the prosecutors at the writ hearing, the head of the appellate section in the Harris County District Attorney's Office, who personally handled Applicant's direct appeal, acknowledged as follows:

Q. Essentially, you wouldn't disagree that [defense counsel] knew that [RJS] was a juvenile delinquent?

A. Yes.

Q. For heavens sake, he was on juvenile probation, right?

A. Yes.


The record does not reveal why RJS was on juvenile probation. In 1999, as now, if RJS was on juvenile probation for nothing more than simple truancy, then the fact that Belinda could report him to the juvenile probation authorities for continuing to skip school could not possibly result in his being committed to the Texas Youth Commission. See TEX. FAM. CODE §§ 51.03(b)(2), 54.05(g) ("a disposition based solely on a finding that the child engaged in conduct indicating a need for supervision [which includes truant behavior, as opposed to a disposition based on a finding that the child engaged in "delinquent conduct"] may not be modified to commit the child to the Texas Youth Commission."). In short, it is unclear whether RJS's juvenile probation could actually be revoked, as defense counsel's complaints at the writ hearing presume. Thus, the present record does not reveal that RJS's status as a juvenile probationer would have provided him with a particularly compelling motive to commit murder, much less that he subjectively believed otherwise. Under these circumstances, I would not, sua sponte, fashion an argument that RJS's juvenile probationary status constituted favorable—and particularly, material—exculpatory or impeaching evidence that the State suppressed.

The Written Statements of the So-Called "Katy Boys": At trial, defense counsel obtained RJS's two written statements to the investigators as well as RJS's grand jury testimony prior to questioning RJS on cross-examination. He was also shown the offense reports of the particular detectives who testified at trial, which documented some, but not all, of the oral statements RJS made over the course of their investigation. Those same offense reports documented oral statements made by many of the so-called "Katy Boys," a group of teenage contemporaries of RJS who were investigated to some extent, mostly because of their relationship to RJS himself. But because none of the Katy Boys testified at trial, none of their written statements or grand jury testimony was turned over to the defense prior to or during trial. Applicant complains of the non-disclosure of two of RJS's oral statements and all of the written statements and grand jury testimony of the Katy Boys. The gist of his argument is that, had the State provided him with all of this information, he would have been significantly better equipped to develop and present his theory at trial that it was RJS, and perhaps two of his cohorts, CT and MG, who committed the offense.

"The Katy Boys" was the name the prosecutor gave to RJS's teenage contemporaries in her testimony at the writ hearing. Defense counsel had another, less polite, term for them.

Sheriff's investigators obtained written statements from CT, MG, CE, CC, and JP. I find no written statement from CG in the writ record. CT and MG testified before the grand jury in April of 1999.

It is unclear to me, however, that the various undisclosed statements and grand jury testimony supplies significant materiality. It is true that there are some inconsistencies among the various statements. But none of the Katy Boys (other than RJS) testified at trial, and so defense counsel would have had no occasion to use the inconsistencies among their statements to impeach them. To the extent that the various statements could have aided defense counsel in piecing together and presenting his alternative suspect theory, I am not inclined to believe it would ultimately have made much difference in the eyes of the jury, for reasons I will expand upon next in my discussion of the H & R shotgun.

Moreover, it is doubtful that would have elected to call any of the Katy Boys to the stand himself simply in order to impeach them.

The Recovery of the H & R Shotgun: A week or so before Belinda's murder, several of the Katy Boys burglarized a home belonging to the boyfriend of CG's mother. I glean the following facts from their various statements. Participating in the burglary were CG, CC, and CE. Several 12-gauge shotguns were taken during this burglary, though none was shown to be the murder weapon. Several days later, they took the shotguns out to shoot them and invited RJS to join them. RJS purloined his father's H & R shotgun and joined them on their shooting excursion. Afterwards, CE dropped RJS off at a car stereo business where RJS was to meet his father. Because RJS did not want his father to know he had taken the H & R shotgun, he left it with CE, who took it home and apparently hid it under his bed.

Defense counsel knew about the burglary of the home of the boyfriend of CG's mother prior to trial, having obtained an offense report of that offense in advance of trial. It was not until reviewing the offense reports pertaining to Belinda's murder at trial, however, that defense counsel learned that RJS had joined the Katy Boys in their shotgun-shooting excursion days before Belinda's murder. He questioned RJS about this event during his cross-examination of RJS at trial.

On the day of Belinda's murder, RJS and CE skipped their last class and left school. They briefly dropped by RJS's house, and then RJS drove CE home. RJS returned to his own home and called CT and MG, who came over. According to their various accounts, these three then drove down the street to the house of another acquaintance to try to obtain marijuana, then drove back to RJS's house. From there, they drove to a convenience store for cigarettes, then back once more to RJS's residence. MG and CT dropped RJS off so that MG could go pick up his mother from work. RJS claims he then fell asleep on the couch, to be awakened by his father at about 6 p.m., by which time emergency personnel and local constables were on the scene.

There is some discrepancy amongst the various statements with respect to the order of these events, but none is inconsistent with RJS's claim that he was home by about 4:30 p.m. Defense counsel was aware of at least the general outline of this narrative from the offense reports he reviewed at trial.

Three weeks later, at the end of January, the H & R shotgun was recovered. At trial, the testimony of both Detective Leithner and Detective Mark Schmidt both tended to suggest that it was recovered from RJS's residence a few weeks after the murder. One of the offense reports that defense counsel reviewed during trial indicated that the H & R shotgun was recovered by a Deputy Ramon Hernandez; but this offense report does not say where, or from whom, Hernandez recovered it. Hernandez's own supplemental offense report recounting his recovery of the H & R shotgun, if it ever existed at all, seems to have gone missing. Hernandez testified at the writ hearing, but by that time he could not remember from whom he recovered the shotgun in the absence of a supplemental offense report documenting its recovery. The mystery was apparently solved in 2012 when (in response to Applicant's budding actual innocence claim) one of the sheriff's investigators, Detective Holtke, re-interviewed CE. CE told Holtke that Hernandez had recovered the H & R shotgun from him. In none of CE's previous statements to investigators had he indicated that he had ever taken possession of the H & R shotgun.

Nobody at the writ hearing, including Hernandez himself, could say for sure that he had actually prepared such a supplement, but all agreed that, assuming he did, it has been lost.

Applicant makes much of the fact that it was never revealed to him at trial from whom or where Hernandez recovered the H & R shotgun. In its recommended findings of fact, the habeas court concludes that Hernandez's supplemental offense report was "lost, destroyed, or never prepared," and observes that the sheriff's investigators failed to question CE during their initial investigation "about his hiding the H & R shotgun." I fail to perceive how this non-disclosure of information pertaining to the recovery of the H & R shotgun deprived Applicant of favorable—much less material—evidence. That Applicant's jury may have gotten the false impression that the H & R shotgun was recovered from RJS's household seems to me to have militated to Applicant's benefit at trial, since it would have placed the shotgun within easy reach for RJS to use in murdering Belinda. Instead, Applicant spins an unlikely scenario in which he contends that he could have persuaded the jury, had the truth been timely revealed, that RJS took CE home from school on the day of the murder, retrieved the shotgun from CE at that time, but then later returned the shotgun to CE—to put back under CE's bed—after RJS used it to kill Belinda. There is no evidence in either the trial record or the writ record to support this theory.

Most of Applicant's present arguments for how he could have used the various undisclosed statements of RJS and the Katy Boys ultimately turn on the jury accepting this dubious proposition—that RJS retrieved the H & R shotgun from CE, used it (perhaps with the help of CT and MG) to kill Belinda, and then returned it to CE for safekeeping. I doubt that disclosure of either 1) the various statements and testimonies of the Katy Boys, or 2) the location of the H & R shotgun when it was recovered would have significantly enhanced defense counsel's ability to persuade the jury that RJS rather than Applicant was responsible for Belinda's murder. With respect to the latter, it might even have detracted.

Joe Sosa's Information: Three weeks after Belinda was killed, on February 4, 1999, Detective Schmidt returned a telephone call from Joe Sosa, a special education teacher at Katy High School. Sosa told Schmidt, among other things, that CT had told Sosa that CT had been "in [RJS's] home the night of the homicide along with [MG,]" and that CT had missed school the next day. Sosa also told Schmidt that, at some undisclosed time, MG "had made a comment that if you put a pillow up to a shotgun it will muffle the sound." This information appears in a supplemental offense report attributed to Schmidt that was randomly attached to the supplemental report of another officer, and, unlike Schmidt's other supplements, it is not numbered. I am frankly unable to tell whether defense counsel was aware of it during trial. Schmidt testified at the writ hearing that he conducted no follow-up investigation of this information that he had obtained from Sosa. Had defense counsel known of it before trial, he could have interviewed Sosa to try to put these statements of CT and MG in context. The record does not reveal what more Sosa might have been able to say.

There was no forensic evidence at trial to suggest that a pillow was used to muffle the shot that killed Belinda.

Joe Cadena's Statement: Defense counsel was never given access to an offense report that documented the statement of an across-the-street neighbor, Joe Cadena, who told investigators on January 25, 1999, that he had heard what he took to be a truck backfiring at about 4:30 p.m. on the day of Belinda's murder. The habeas court mentions this undisclosed information in its recommended findings as well. But the record also reveals that defense counsel was given access to a different offense report during trial that recorded an earlier statement that Cadena had given to canvassing officers on the night of the murder. In that earlier statement, Cadena asserted that, sometime between 4:30 and 5:30 p.m., he actually heard two backfires, which he attributed to a particular truck that he observed on the street. Although defense counsel's investigator spoke with Cadena during trial, defense counsel did not subpoena Cadena to testify.

It was shown at trial that several young boys who lived behind the Temple residence had heard what could have been a gunshot at around 4:30, at a time when Applicant must have been on his way to the supermarket. As the State hypothesized during the writ hearing, it is possible to argue that the boys heard the same backfire that Cadena thought he heard.

There is no evidence that the shotgun that killed Belinda was fired a second time.

C. Incremental Materiality of the Late-Disclosed Evidence

Perhaps marginally more convincing are Applicant's arguments that disclosure of much of the materials that the State revealed to defense counsel for the first time during trial could have been used to substantially greater effect had they been timely disclosed prior to trial. Some of Applicant's claims in this regard are more compelling than others. I shall highlight the most pertinent examples of this category as well.

Again, the discussion that follows is not exhaustive.

The Katy Boys' Failed Polygraphs: The Katy Boys were questioned more than once by sheriff's investigators and gave some oral statements in addition to their undisclosed written statements. As the habeas court notes in its recommended findings, many of them, including CE, CT, and MG, were subjected to multiple polygraph examinations, which they almost uniformly failed. Defense counsel learned of these facts when he reviewed the offense reports during trial. But the trial court refused to allow defense counsel to elicit any evidence with respect to the polygraph testing or results before the jury, and although defense counsel complained bitterly of this limitation at trial, Applicant did not challenge the exclusion of the polygraph evidence as an issue in his direct appeal. In any event, it is far from clear that the polygraph evidence was admissible under Texas law, or that it would have lead to admissible evidence that defense counsel would not otherwise have been alerted to investigate by the offense reports. For this reason, if no other, I cannot say that the State's late-disclosure of the polygraph evidence constituted a violation of Brady.

See Ex parte Bryant, 448 S.W.3d 29, 40 (Tex. Crim. App. 2014) ("Polygraph evidence is generally excluded from courtrooms because the reliability of such tests remains unproven and jurors could attach undue credibility to a test that purports to sort truth from fiction, a role for which a factfinder is more properly suited.") (citing Leonard v. State, 385 S.W.3d 570, 577 (Tex. Crim. App. 2012) ("[W]e have not once wavered from the proposition that the results of polygraph examinations are inadmissible over proper objection because the tests are unreliable.")).

The Parkers' Dog: Jim and Cynthia Parker lived in a house that was catty-corner to the Temple's property, divided from it by a fence. Defense counsel did not learn until he reviewed the offense reports at trial that a police canvass had disclosed that the Parker's dog had barked excitedly at the dividing fence at approximately 4:30 p.m. on the day of Belinda's murder—a time that coincides with his defensive theory for when the break-in and murder occurred (Applicant having been at the supermarket at about this time). Once defense counsel did learn about this evidence, however, he interviewed the Parkers one evening during the trial and then subpoenaed them, and they testified for the defense at trial. The habeas court recommends that we find defense counsel did not obtain this information until trial had commenced. The record substantiates this finding, but it makes no difference because Applicant cannot show that he suffered any disadvantage from the State's failure to divulge this information prior to trial.

See note 34, ante.

Shaka's "Access" to the Garage: The habeas court also finds that defense counsel did not learn about three other potential witnesses until he reviewed the offense reports at trial, all three of whom could have testified that Shaka, the Temple family's chow, ordinarily "had access to the garage." Other witnesses did testify to this fact at trial, however. And in any event, the fact that Shaka had "access" to the garage does not necessarily establish that he was in the garage at the time that Belinda was killed, and therefore could not have been expected to bark at any intruders. The testimony of these three additional witnesses would have added only quite modestly to any inference favorable to the defense, and not enough to establish any significant incremental materiality.

For example, Michael Ruggiero, a neighbor who lived across the street from the Temples, testified that the latch on their back gate was not catching properly, and that the Temples would therefore keep Shaka in the garage so he could not get out. His wife, Peggy, testified that she had observed Belinda arriving home on occasion and pulling into the garage. Belinda would honk the horn once the garage door had opened to signal for Shaka to move out of the way.

Applicant's Emotional Response: The habeas court recommends a finding that the late-disclosed offense reports also revealed two witnesses who could have attested that Applicant reacted emotionally at the scene to Belinda's death, sobbing with his head in his hands and appearing weak-kneed. Had he learned of the additional witnesses earlier, defense counsel may have had time to interview them and to subpoena them for trial. Other witnesses did testify at trial to Applicant's apparent emotional response, however, to counteract the testimony of sheriff's investigators who conveyed to the jury their impressions that Applicant seemed emotionless at the scene. One of Applicant's brothers testified at trial that Applicant appeared to be in shock. His mother testified that he had apparently been crying. Later, she maintained, after he was interrogated by sheriff's investigators and allowed to leave, Applicant was "completely distraught and broken." But these were family members, whose testimony could have been discounted by the jury as self-serving. Testimony of non-family witnesses with respect to Applicant's demeanor was favorable to the defense and defense counsel's failure to recognize their significance adds marginally to the incremental materiality.

RJS's Girlfriend: Although the habeas court does not mention it in its recommended findings of fact, there is another piece of information that was not revealed to defense counsel until he was allowed to review the offense reports at trial. Niki Biondo was RJS's girlfriend at the time of Belinda's murder. She told sheriff's investigators that sometime between 6:30 and 7:00 p.m. on the evening of the murder, RJS called her in a very emotional state ("crying") and told her that his next-door neighbor had been "shot"—not killed, but "shot." It is unclear how RJS could have learned this detail so soon after the fact. The sheriff's investigators did nothing to follow up on this lead. At the writ hearing, the State speculated that RJS could have heard this detail from bystanders on the street in the hour or so following Belinda's murder, notwithstanding that the sheriff's investigators would have taken pains, at least in theory, to conceal the specifics of the crime scene. A jury would not have been constrained to accept the State's speculation, however, and it would have been important for Applicant to discover Biondo's statement in time to track her down and interview her, to test the apparent exculpatory value of her story. She may have proven to be an important witness for the defense, and defense counsel's failure to recognize her potential importance in the limited time he had to review the offense reports also adds to the incremental materiality.

D. Collective Materiality

Because materiality is to be assessed "collectively, not item by item[,]" Kyles, 514 U.S. at 436, the question becomes: Is there sufficient undisclosed Brady evidence that, when taken together with Applicant's showing of incremental materiality of the late-disclosed evidence, would undermine our confidence in the result of Applicant's trial? I have found little that I regard as significant undisclosed Brady evidence, and not a great deal of incremental materiality in Applicant's claims of late-disclosed Brady evidence. In short, the record simply fails to reveal much Brady evidence—either undisclosed evidence with significant exculpatory or impeaching value or late-disclosed exculpatory or impeaching evidence that is incrementally material—to measure collectively for materiality.

For this reason, I would not grant relief on the basis of Brady, but would instead grant relief on the basis of ineffective assistance of counsel. Though the habeas court recommended that we reject all of Applicant's claims of ineffective assistance of counsel, it made few specific findings of fact with respect to any of these claims. In my view, Applicant has established by a preponderance of the evidence that defense counsel was deficient in at least one critical aspect, and in the context of this particular trial, that deficiency could well have proven to be a game-changer.

The habeas court simply concluded that Applicant's "current claim that trial counsel provided ineffective representation has not been shown to meet the Strickland requirements and relief is not justified." Such a perfunctory conclusion has little utility. When "the habeas judge's findings do not resolve the disputed fact issues, this Court must exercise its role as the ultimate finder of fact." Ex parte Flores, 387 S.W.3d 626, 635 (Tex. Crim. App. 2012). Fortunately, the factual development in the record is more than sufficient to provide a basis to glean the relevant facts ourselves and draw conclusions of law from them.

III. INEFFECTIVE ASSISTANCE OF COUNSEL

A. The Strickland Standard

In Strickland v. Washington, 466 U.S. 668 (1984), the United States Supreme Court encapsulated the Sixth Amendment standard for measuring the effectiveness of counsel:

A convicted defendant's claim that counsel's assistance was so defective as to require reversal of a conviction . . . has two components. First, the defendant must show that counsel's performance was deficient. This requires a showing that counsel made errors so serious that counsel was not functioning as the "counsel" guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.
Id. at 687. A habeas applicant is entitled to relief if he can demonstrate both deficient performance and prejudice by a preponderance of the evidence. Ex parte Moore, 395 S.W.3d 152, 157 (Tex. Crim. App. 2013).

With respect to the deficiency component of the Strickland standard, "[t]he proper measure of attorney performance remains simply reasonableness under prevailing professional norms." Strickland, 466 U.S. at 688. Counsel has "a duty to bring to bear such skill and knowledge as will render the trial a reliable adversarial testing process." Id. Appellate review of counsel's performance must be deferential, and "every effort [must] be made to eliminate the distorting effects of hindsight[.]" Id. at 689. An applicant for post-conviction habeas relief who claims his attorney performed deficiently must overcome the presumption "that counsel . . . rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment." Id. at 690.

In order to establish that his attorney's deficiency was prejudicial, a habeas applicant "must show that there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. at 694. A "reasonable probability" of a different result means more than that the error "had some conceivable effect on the outcome of the proceeding"; but the applicant "need not show that counsel's deficient conduct more likely than not altered the outcome in the case." Id. at 693. In short, "[a] reasonable probability is a probability sufficient to undermine confidence in the outcome." Id. at 694. To this end, "[s]ome errors will have had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture, and some will have had an isolated, trivial effect." Id. at 695-96. "Moreover, a verdict . . . only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support." Id. at 696.

While a reviewing court "normally looks to the 'totality of the representation' and 'the particular circumstances of each case' in evaluating the effectiveness of counsel, Ex parte Raborn, 658 S.W.2d 602, 605 (Tex. Cr[im]. App. 1983), the Court has also found that under some circumstances a 'single error of omission by . . . counsel [can] constitute[ ] ineffective assistance.' Jackson v. State, 766 S.W.2d 504 (Tex. Cr[im]. App. 1985), modified on other grounds on remand from the U.S. Supreme Court 766 S.W.2d 518 (Tex. Cr[im]. App. 1988)." Ex parte Felton, 815 S.W.2d 733, 735-36 (Tex. Crim. App. 1991). See also, Thompson v. State, 9 S.W.3d 808, 813 (Tex. Crim. App. 1999) ("[I]t is possible that a single egregious error of omission or commission by [applicant]'s counsel constitutes ineffective assistance.") (internal quotation marks omitted). The United States Supreme Court has likewise recognized that the Sixth Amendment "may in a particular case be violated by even an isolated error of counsel if that error is sufficiently egregious and prejudicial." Murray v. Carrier, 477 U.S. 478, 496 (1986). One mistake may, in some instances, prove momentous enough to justify the conclusion both that the attorney rendered constitutionally deficient performance and that the impact of that deficiency was such as to undermine appellate confidence in the result of the proceeding. "Although the [appellate] court must look to the level of counsel's overall performance, clearly negligent treatment of a crucial deficiency in the prosecution's case or an obvious strength of the defense will outweigh the adequate handling of a series of minor matters." 3 W. LaFave et al., CRIMINAL PROCEDURE § 11.10(c), at 1156-57 (4th ed. 2015).

E.g., Ex parte Scott, 581 S.W.2d 181 (Tex. Crim. App. 1979) (ineffective counsel at the punishment phase for failure to uncover the invalidity of a prior conviction that was used to enhance); May v. State, 722 S.W.2d 699 (Tex. Crim. App. 1984) (failure to file a sworn application for probation when the evidence demonstrated eligibility); Ex parte Zepeda, 819 S.W.2d 874 (Tex. Crim. App. 1991) (failure to request an accomplice-witness instruction requiring corroboration of the testimony of several accomplices as a matter of law); Vasquez v. State, 830 S.W.2d 948 (Tex. Crim. App. 1992) (failure to request instruction on the defense of necessity when the evidence raised the issue); Ex parte Varelas, 45 S.W.3d 627 (Tex. Crim. App. 2001) (failure to request limiting instruction/reasonable doubt instruction with respect to extraneous offenses that were integral to the State's proof); Ex parte Harrington, 310 S.W.3d 452 (Tex. Crim. App. 2010) (failure to investigate the validity of a prior conviction used to enhance punishment); Villa v. State, 417 S.W.3d 455 (Tex. Crim. App. 2013) (failure to request instruction on medical care defense when the evidence raised the issue); Ex parte Saenz, 491 S.W.3d 819 (Tex. Crim. App. 2016) (failure to impeach chief witness for the State with his prior inconsistent statements).

I believe that is the case here. Although the question of attorney deficiency in this case is a close one in light of defense counsel's overall performance, his mistake was a serious one. And there is a substantial basis to conclude, given the totality of this record, that the impact of that mistake, however isolated, could well have been profound.

B. Trial Counsel's Deficiency

The deficiency in this case centers on the trial testimony of Applicant's father, Charles Kenneth Temple, Jr., who testified as a defense witness at trial. On the night of Belinda's murder, Kenneth had given a written statement to the sheriff's investigators. Asked about the time that Belinda had dropped by his residence to pick up the homemade soup for the ailing ET on her way home, Kenneth maintained that he had gotten home from work at 3:30 p.m., "and Belinda arrived about fifteen minutes later at approximately 3:45 P.M." She "visited with us for a few minutes" and then, "I guess it was around 3:55 P.M. at the time she left." Testimony at trial indicated that the drive from Kenneth's residence to Applicant and Belinda's house takes about fifteen minutes. Thus, according to Kenneth's original estimate, Belinda could not have arrived home much earlier than 4:10 p.m. This would have left only a very narrow window of time—ten minutes or so—during which Applicant could have forced or coaxed Belinda into the walk-in closet, killed her, potentially changed his clothes, staged a burglary, hustled ET into his truck, and still arrived at the supermarket by 4:32 p.m. Defense counsel was aware of the content of Kenneth's written statement well in advance of trial.

The reporter's record at trial identified the witness as "Kenneth," not by his first name "Charles," and I conform to that designation in this opinion.

At the writ hearing, Detective Leithner confirmed that in his offense report he had indicated that it had taken him approximately sixteen minutes to drive the distance from Kenneth's house to Applicant's.

See note 23, ante.

This was the time that the security video showed Applicant and ET entering the supermarket.

When Kenneth testified before the grand jury in early April of 1999, he gave the same time estimates: Belinda arrived at his house "at 3:45," and they visited "for a few minutes standing there in the garage." "She probably was at my house from 3:45 to 3:55. I think she left about five minutes till 4:00." He confirmed that the drive from his house to Applicant and Belinda's home was "[a]bout 15 minutes." Defense counsel was provided with a copy of Kenneth's grand jury testimony in the middle of the State's cross-examination. In any event, he would have known at the time of trial from Kenneth's written statement to the sheriff's investigators that Kenneth's pre-trial estimate was quite favorable to his defensive posture, since it supports a time-line that would have made it even more problematic for a jury to conclude that Applicant could have murdered Belinda.

Inexplicably, Kenneth remembered the time-line differently at trial. He claimed once again that he got home at 3:30 p.m. Defense counsel asked him to give his "best estimate or if you looked at a clock, when did Belinda get there?" Kenneth told the jury, "3:32, or close to that." A short while later, defense counsel asked, "And then when did she leave." Kenneth answered, "In minutes. 3:45 at least." Defense counsel then asked Kenneth how long the drive was between the two houses. Kenneth answered:

A. 15 to -- 15 plus minutes.

Q. Okay. And she left there at 3:45?

A. Yes.
This estimate would have placed Belinda at home much closer to 4:00 o'clock, in conformance with the State's theory of the case. Though the prosecutor showed Kenneth selected portions of his grand jury testimony while she cross-examined him at trial, she never showed him his earlier testimony with regard to the time-line.

Kenneth testified again during the writ hearing. At this point, he reverted to his original account of the time-line, asserting once again that Belinda did not arrive at his house until 3:45, stayed for about ten minutes, and left "by 3:55." Kenneth maintained that he had given defense counsel a copy of his written statement to the sheriff's investigators on more than one occasion prior to trial. Even so, he maintained, defense counsel never advised him to review his written statement before testifying at trial. Kenneth reiterated that his grand jury testimony with respect to his time estimates was consistent with his written statement. When, on cross-examination at the writ hearing, the State's habeas counsel showed Kenneth a transcript of his trial testimony to confirm that he had given a different time estimate at trial, Kenneth seemed incredulous:

Q. Do you recall what you testified to during your son's trial back in 2007 as to what time Belinda Temple left her house?

A. I don't remember that specifically, no. I remember my written testimony [sic], not that.
Kenneth insisted that his testimony at trial had been inaccurate. He reiterated that defense counsel had not instructed him before his trial testimony to review his written statement, nor did he review it on his own accord "during the trial."

For his part, defense counsel made no excuses for this lapse when he testified at the writ hearing. He admitted that Kenneth had given him a copy of his written statement "soon after I became involved in the case." Applicant's counsel then asked defense counsel:

Q. Which of these two timelines are more beneficial to the defense, alibi defense for [Applicant]?

A. Well, the one of Ken Temple's timeline. The one that has Belinda arriving at the Temple home, Mr. and Mrs. Temple's home at 3:45 and then leaving at 3:55, because that would put her at [Applicant] and Belinda's home about 15 minutes later, which would make it after 4 o'clock.

Q. Can you think of any reason why you did not use that second timeline?

A. I have no explanation for it. I don't know.

* * *

Q. Can you imagine how you made that mistake?

A. No.

Q. Was it intentional?

A. No.

Q. Is there any strategic reason why you would not make at that time your only defense as close to 4:30 as possible?

A. No.


* * *

Q. Now, can you imagine any reason why you did not use [Kenneth's written statement] to refresh the memory of [Kenneth] when [he] said I got home -- she got to the house at 3:32?

A. No.

Q. And left at 3:45?

A. No. It was obviously different from his written statement, and I should have gone up and shown him his written statement and say "Does this refresh your recollection," but I did not.

Q. Have you done that before?

A. I have.

Q. On few or many occasions refreshing a witness' memory with a prior written statement?

A. Well, you know, that's second-year law school evidence. You can do that. I've done it a lot of times.

* * *
Q. . . . Mr. [defense counsel], is there any reason whatsoever for you, from a strategic perspective, not to use [Kenneth]'s timeline?

A. No.

Q. Did it harm your client not to use his timeline?

A. I now believe it did, yes.

Q. Why?

A. Well, because it gave more time to be explained. That is, more time to do what the State envisioned that [Applicant] did, that is, time to kill Belinda, get rid of the shotgun.

Strictly speaking, of course, defense counsel was mistaken to assume that he could have used Kenneth's written statement to refresh Kenneth's memory on the witness stand—at least not over an objection from the State. Kenneth did not purport to suffer from a lapse of memory while testifying at trial; he seemed to remember well enough. See Callahan v. State, 937 S.W.2d 553, 559 (Tex. App.—Texarkana 1996 no pet.) (predicate for using a document to refresh a witness's memory includes a showing that "his memory needed to be refreshed"). He just remembered differently than he had in the past. Nevertheless, a party may impeach his own witness. See TEX. R. EVID. 607 ("Any party, including the party that called the witness, may attack the witness's credibility). Defense counsel could have laid the predicate, under Rule of Evidence 613, to use Kenneth's written statement to elicit the fact that Kenneth had made an inconsistent statement in the past—in the hope that reminding Kenneth of his prior written statement would jog Kenneth's memory and cause him to revise the substance of his trial testimony. TEX. R. EVID. 613. Moreover, once he learned during Kenneth's cross-examination that Kenneth had testified to the grand jury consistently with his written statement, he could have impeached him with that as well, if necessary.

In Welch v. State, 576 S.W.2d 638, 641 (Tex. Crim. App. 1979), we explained:

A witness testifies from present recollection what he remembers presently about the facts in the case. When that present recollection fails, the witness may refresh his memory by reviewing memorandum made when his memory was fresh. After reviewing the memorandum, the witness must testify either his memory is refreshed or his memory is not refreshed. If his memory is refreshed, the witness continues to testify and the memorandum is not received as evidence. However, if the witness states that his memory is not refreshed, but has identified the memorandum and guarantees the correctness, then the memorandum is admitted as past recollection recorded.
(Emphasis supplied). All of this is contingent on a lapse of present recollection.

More to the point, defense counsel should have better prepared this witness for his critical trial testimony; at the very least, he should have asked Kenneth to review his written statement before taking the witness stand, to refresh his memory before trial. See, e.g., Perrero v. State, 990 S.W.2d 896, 899 (Tex. App.—El Paso 1999, pet. ref'd) (trial counsel provided ineffective assistance of counsel by putting his client on the witness stand without properly preparing him to testify). Perhaps defense counsel was distracted from his ordinary witness-preparation duties because he was busy exploring the many new evidentiary leads while in the process of trying his case—a product of the State's belated disclosure of so much information that was favorable to the defense.

Notwithstanding defense counsel's admission at the writ hearing, the State argues that his performance was not constitutionally deficient because failing to challenge Kenneth's trial testimony was nonetheless objectively reasonable. We have indeed held that trial counsel's conduct must be measured by an objective standard of reasonableness, and "a decision not motivated by strategy might be objectively reasonable." Ex parte Saenz, 491 S.W.3d 819, 829 (Tex. Crim. App. 2016). From this proposition, the State seems to argue that, because defense counsel might reasonably have chosen to stand on Kenneth's trial testimony, not challenge it, we may not find that his performance was constitutionally deficient. I ultimately disagree, for two reasons.

First of all, to the extent the State is suggesting that defense counsel actually made a strategic decision to let Kenneth's trial testimony stand unchallenged, the record presents little evidence to support that conclusion. Defense counsel denied it, insisting that he had simply made a mistake. "Whether a counsel's action or inaction is based on a strategic choice is a factual question, on which the defendant can offer evidence when the incompetency challenge is presented in a post-conviction proceeding (as often must be the case)." LaFave, CRIMINAL PROCEDURE § 11.10(c), at 1133. There is no compelling reason to reject defense counsel's testimony in this regard. See Saenz, 491 S.W.3d at 828 & n.9 (refusing to indulge the appellate presumption that counsel's decisions were strategically motivated in the context of a post-conviction habeas corpus proceeding at which "the record . . . was developed, and trial counsel was able to adequately respond"). I have no trouble believing that, presented for the first time in the middle of trial with an abundance of vital new information to support a new defensive theory—an alternative suspect—defense counsel lost his focus when executing his original defensive plan of alibi. As ultimate factfinder in post-conviction habeas corpus proceedings, we are free to accept trial counsel's assurance that his conduct was not based on trial strategy, even if the record presents some basis to believe that some objective strategy could have justified it. See Saenz, 491 S.W.3d at 829 ("[A]s the ultimate factfinder in habeas proceedings, we decline to adopt the habeas court's finding that trial counsel might have made a reasonable strategic decision . . . ."). Accordingly, I decline to automatically adopt the State's "post hoc rationalization of counsel's conduct" in place of counsel's own explanation. Wiggins v. Smith, 539 U.S. 510, 526-27 (2003).

When ineffective counsel is alleged on direct appeal, it is usually the case that the record is silent with respect to whether counsel's action or inaction was the product of strategy or mistake. Bone v. State, 77 S.W.3d 828, 833 & n.13 (Tex. Crim. App. 2002). "An ineffectiveness claim raised on direct appeal is limited to what the trial record reveals as to the grounding for counsel's actions, and here the appellate court commonly will assume a strategic motivation if any can possibly be imagined." LaFave, supra, at 1137. In the post-conviction context, however, where ineffectiveness of counsel has been alleged and the record has been developed with respect to counsel's actions, this appellate assumption no longer controls. Saenz, 491 S.W.3d at 828 & n.9. If trial counsel concedes that his challenged act or omission was not the product of strategy, the objective existence of a plausible strategic basis for the act or omission may provide a reason to doubt the genuineness of his concession. That would justify a finding of fact that his choice was, in fact, a strategic one notwithstanding his concession otherwise. But, as the ultimate fact-finder in post-conviction habeas corpus proceedings, we remain free to accept trial counsel's concession. Id. at *7. This is unlike in the appellate context, where appellate courts must defer to a trial court's finding with respect to the credibility of counsel's claim that an act or omission was inadvertent rather than strategic. See Okonkwo v. State, 398 S.W.3d 689, 694 & n.4 (Tex. Crim. App. 2013) (where trial counsel claimed lack of strategic motive for an omission at trial, appellate court "should have deferred to the trial court's implied finding that counsel's affidavit lacked credibility"). Of course, even in the post-conviction context, we usually defer to the recommended findings of the convicting court when they are supported by the record. E.g., Ex parte Van Alstyne, 239 S.W.3d 815, 817 (Tex. Crim. App. 2007). Here, however, the habeas court made only one finding of fact relevant to this ineffective counsel claim: "Defense counsel did not use Charles Kenneth Temple's written statement timeline." It made no recommended finding with respect to the credibility of defense counsel's concession that his failure to do so was an oversight, not the product of strategy. The only question that remains with respect to the performance prong of Strickland is whether that mistake was one that fell outside the bounds of reasonable professional standards.

Secondly, the State's asserted justification strikes me as less than objectively reasonable. The State argues that defense counsel was content to let Kenneth's trial testimony go uncorrected because it was consistent with the account he would later elicit from Applicant himself during Applicant's own trial testimony. Were I to regard this as a plausible tactical choice, it might present me with a reason to reject, as disingenuous, defense counsel's concession that his failure was a mistake, and to conclude instead that it was a reasonable tactical decision in keeping with his overall strategy in the case. But, for reasons I develop next, I find the State's proposed defensive strategy untenable.

In Applicant's own written statement to sheriff's investigators, also taken on the night of the murder, he estimated that Belinda "got home around 3:45 P.M." When he testified at trial, however, he claimed he did not know exactly what time Belinda had gotten home because he had taken his watch off to bathe ET. He guessed, however, that she had gotten home closer to 4:00 p.m. The State argues that defense counsel made a deliberate decision not to impeach Kenneth's earlier testimony because the estimate Kenneth had given at trial—that Belinda left his house at 3:45 p.m.—was consistent with Applicant's own (albeit revised) estimate that she had arrived home approximately fifteen minutes later, at 4:00 p.m. The State theorizes that defense counsel deliberately chose to forego corrective measures with respect to Kenneth's testimony in order to avoid any later contradiction of Applicant's own time estimate and to thereby maintain Applicant's credibility.

I find this theory questionable. Both Kenneth and Applicant were asked at trial to supply their best estimates of the time-line, and the jury was not necessarily expecting perfect precision. Defense counsel had nothing to lose and everything to gain by propounding the most favorable estimate available—regardless of whether it might conflict slightly with his client's. After all, under Applicant's own estimate, which placed Belinda home about 4:00, it would have been difficult, but not inconceivable, for him to have committed the offense and still arrived at the supermarket by 4:32 p.m. But under Kenneth's pre-trial estimate, placing Belinda home much closer to 4:10 p.m., it would have been practically impossible. I do not think that any reasonable defense lawyer would adopt the State's proposed strategy under these circumstances, and I decline its invitation to find that defense counsel in fact did.

This is not to say that the record is wholly devoid of any support for the State's argument. For instance, as part of a mock trial in preparation for Applicant's trial, the defense team apparently prepared a "Juror Notebook" containing a time-line which listed the time of Belinda's arrival home as "3:55 p.m." Similarly, during his opening statement to the jury at the beginning of trial, defense counsel told the jurors that the evidence would show that Belinda arrived home "between 3:45 and 4:00 o'clock sometime." This time-frame was consistent with Applicant's statement to sheriff's investigators (3:45), as later revised by his trial testimony (4:00).
But other excerpts from the trial record support defense counsel's assertions at the writ hearing that he simply made a mistake, not a strategic choice that he regrets in hindsight. For example, at one point during Applicant's direct examination at trial, in trying to establish what time Belinda must have arrived home, defense counsel asked Applicant:

Q. So if [Belinda] left [Kenneth's house] at 3:50, 3:55 she would have gotten [home] what time?
The prosecutor astutely objected that defense counsel's question "assumes facts not in evidence." Because Kenneth's trial testimony placed Belinda's departure at 3:45, not "3:50, 3:55," as defense counsel's question posited, the trial court correctly sustained this objection. Nevertheless, more than once defense counsel asked Applicant questions that seemed to assume that Kenneth's earlier trial testimony had been consistent with Kenneth's prior written statement. These exchanges strongly suggest to me that defense counsel simply did not realize that Kenneth had testified differently, and that his failure to try to correct Kenneth was a mistake, not a strategic choice. Defense counsel's second chair attorney confirmed that it had been the defensive strategy at trial "to try to get Belinda home as close to 4:32 as possible" because "the closer [to 4:32] that she arrives to the house, the more favorable it is to the defense." "Courts . . . readily find ineffective assistance when counsel's testimony at a post-conviction evidentiary hearing establishes that a failure to act on an important matter was a product of inattention in a setting where the missed option was obvious." LaFave, supra, at 1161.

Defense counsel failed to properly prepare Kenneth to testify consistently with his written statement with respect to a particular fact that was vital to the optimal presentation of his original defensive posture in the case. He was also unprepared to impeach Kenneth in the event that he persisted in testifying differently than his written statement. I would hold that these failures constituted an omission that fell below the standard of reasonable professional competence. I would hold that Applicant has satisfied the deficient-performance prong of Strickland, and I turn next to the question of prejudice.

C. Prejudice

The question before us on discretionary review in this case was legal sufficiency: whether a rational jury could convict Applicant on the basis of the facts presented. We held that the circumstantial evidence supported a rational jury verdict of guilty beyond a reasonable doubt. Temple, 390 S.W.3d at 363. But rational juries can acquit even when the evidence is legally sufficient to convict, and the question before us in gauging the prejudice prong of Strickland is not one of sufficiency of the evidence. Cf. Kyles v. Whitley, 514 U.S. at 434 (Brady's materiality standard, which is essentially the same as Strickland's prejudice prong, "is not a sufficiency of evidence test"). Given the evidence that Applicant's jury heard in this case, a rational jury might just as readily have acquitted him. Even the State's evidence presented some basis to doubt whether Applicant could possibly have had time to perpetrate the offense, and there is further evidence of a viable second suspect who was close at hand and who also arguably had a motive to commit Belinda's murder. In short, the jury's verdict, though rational, was hardly "one with overwhelming record support." Saenz, 491 S.W.3d at 833 (quoting Strickland, 466 U.S. at 696). When that is the case, reviewing courts may more readily conclude that deficient attorney performance results in prejudice. Id.

See Johnson v. Louisiana, 406 U.S. 356, 362 (1972) ("Jury verdicts finding guilt beyond a reasonable doubt are regularly sustained even though the evidence was such that the jury would have been justified in having a reasonable doubt[.]").

"Applying that principle here," id., it is not hard to imagine that defense counsel's mistake, isolated as it was, could have tipped the precarious balance the other way. Add to the existing record the evidence that the jury would have heard but for defense counsel's failure to properly prepare Kenneth to testify, and there is a reasonable probability that a rational jury would harbor a reasonable doubt that Applicant was the murderer. Had defense counsel been less distracted by late-disclosed evidence and therefore better focused to prepare his witness, there is little reason to doubt that Kenneth would have reverted to his original time estimate. And had the jury heard and credited Kenneth's original time estimate, it might more readily have concluded that Applicant could not have had time to kill Belinda, and it may therefore have given more credence to the alternative hypothesis—in some respects, better supported by the evidence—that RJS was the perpetrator.

As it is, relying on the estimate that Applicant originally gave to sheriff's investigators that Belinda arrived home as early as 3:45 p.m., the State argued during its final summation at the guilt phase of trial that Applicant enjoyed as much as 47 minutes to perpetrate the offense and get to the supermarket. Defense counsel was unequipped to refute this scenario in his own final argument. Had defense counsel properly prepared Kenneth to testify consistently with his original statement to the police and his grand jury testimony, then defense counsel would have been in a position to argue to the jury that Applicant had no more than ten minutes in which to commit the offense.

The Strickland standard does not require us to conclude that such a scenario is more likely than not before habeas corpus relief is appropriate. 466 U.S. at 693. "The result of a proceeding can be rendered unreliable, and hence the proceedings unfair, even if the errors of counsel cannot be shown by a preponderance of the evidence to have determined the outcome." Id. at 694. In view of the overall context in which this trial occurred, it is no great stretch to declare that our confidence in the result of Applicant's trial has been undermined, and that is enough to establish "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. I conclude that he has established by a preponderance of the evidence both deficient performance and prejudice, and thus he has established that he suffered a deprivation of his Sixth Amendment right to the effective assistance of counsel.

Indeed, as is the case with the Bagley test for materiality, under the Strickland test for prejudice, Applicant "can prevail even if" trial counsel's deficiency "may not have affected the jury's verdict." Wearry v. Cain, 136 S.Ct. at 1006 n.6.

IV. CONCLUSION

Accordingly, I concur in the Court's judgment to reverse Applicant's murder conviction and remand him to the custody of the Harris County Sheriff to answer to the indictment against him. FILED: November 23, 2016
DO NOT PUBLISH


Summaries of

Ex parte Temple

COURT OF CRIMINAL APPEALS OF TEXAS
Nov 23, 2016
NO. WR-78,545-02 (Tex. Crim. App. Nov. 23, 2016)
Case details for

Ex parte Temple

Case Details

Full title:EX PARTE DAVID MARK TEMPLE, Applicant

Court:COURT OF CRIMINAL APPEALS OF TEXAS

Date published: Nov 23, 2016

Citations

NO. WR-78,545-02 (Tex. Crim. App. Nov. 23, 2016)

Citing Cases

Hernandez v. State

Instead, "[t]here are three essential components of a Brady violation: (1) the evidence at issue must be…

In re State

Further, while the exclusion of evidence has been the remedy for violations of article 39.14, the Texas…