Allen, Ex Parte Billy Frederick

IN THE COURT OF CRIMINAL APPEALS

OF TEXAS

NOS. AP-75,580/75,581

EX PARTE BILLY FREDERICK ALLEN, Applicant

ON APPLICATION FOR A WRIT OF HABEAS CORPUS FROM THE 195TH JUDICIAL DISTRICT COURT

DALLAS COUNTY H OLCOMB, J., announced the judgment of the Court in an opinion in which J OHNSON and C OCHRAN, JJ., joined. P RICE and C OCHRAN, JJ., each filed a concurring opinion.

K ELLER, P.J., filed a dissenting opinion. M EYERS, J., filed a dissenting opinion in which K ELLER, P.J., joined. H ERVEY, J., filed a dissenting opinion in which K ELLER,

P.J., joined. W OMACK, J., concurred. K EASLER, J., did not participate.

OPINION

This is a post-conviction habeas corpus proceeding, brought under Article 11.07 of the Texas Code of Criminal Procedure, in which applicant Billy Frederick Allen presents a Schlup1-type claim of actual innocence as a procedural gateway through which to raise his otherwise barred constitutional claim of ineffective assistance of trial counsel.2 We hold that applicant is entitled to relief.

Background

Applicant was separately charged with the murders3 of Raven Dannelle Lashbrook and James Perry Sewell. The two cases were tried together on September 14, 1983. The record shows the following facts pertinent to the issue before us. On April 9, 1983, somewhere between 4:00 and 4:30 a.m., Officer Curtis Clary, of the University Park Police Department, was dispatched to the location of a possible shooting. Upon his arrival, he found a man (later identified as Sewell) gagged, handcuffed, and covered with blood, standing near a duplex. Sewell told Clary that he had been shot in the back of the head twice, cut and robbed, and that his girlfriend had been kidnapped. Clary seated Sewell in the front yard. Clary’s partner had arrived separately by then, and the two officers went inside the duplex to check for suspects and additional victims. They found none. Meanwhile, some University Park firemen/paramedics arrived at the scene. They attended to Sewell and then placed him inside the ambulance to take him to the hospital. Before they left, however, Clary came out of the house, approached Sewell from the side of the ambulance, and asked him if he knew who had attacked him. According to Clary’s testimony, Sewell stated that it was “Billy Allen, Bert [whose last name he did not know] and another white male he didn’t know at all.”

After Sewell was taken to the hospital, the officers searched the house again and discovered the body of a woman (later identified as Lashbrook, the girlfriend that Sewell had apparently thought had been kidnapped) in a car parked in the carport. Another officer arrived shortly and dusted the area for fingerprints. He found four identifiable prints, two of which were taken from the house and belonged to Sewell himself. The third print was taken from one of Sewell’s cars and was never identified. The fourth print was taken from the roof of the car in which Lashbrook’s body was found and was determined to be that of applicant’s left palm.

At trial, the State relied primarily on this palm print and Sewell’s statement (as related by Officer Clary, that Sewell identified “Billy Allen” as one of his attackers) to show applicant’s complicity in the two murders.4 The defense, in turn, relied on the testimony of the State’s own witnesses – who acknowledged the victims’ involvement in drug-trafficking5 and agreed that the ransacked condition of Sewell’s house at the time of the crime suggested that someone might have been looking for drugs6 – to show that the murders were more likely committed by some of the victims’ own criminal associates. The defense was the first to admit that applicant knew Sewell and visited him frequently. But it tried to show that applicant was not one of Sewell’s criminal associates who were more likely to have killed Sewell because applicant’s meetings with Sewell were only in connection with applicant’s legitimate business, i.e., to sell Sewell the scraps of gold that applicant received at his used-goods store. The defense also tried to explain the presence of applicant’s palm print on Sewell’s car through the testimony of applicant’s wife. According to this testimony, applicant had met Sewell in a café, just a few days before the murder, to negotiate the price of the scraps of gold he had sold Sewell the night before at Sewell’s house. After the meeting, applicant went to Sewell’s car and stood leaning against it, apparently with his hand on the roof of the car, as Sewell sat inside counting the money to pay him for the scraps of gold.

The jury, during their deliberations, sent the trial court two notes, asking the court to reread Officer Clary’s testimony regarding Sewell’s statement as to who had assaulted him. The court supplied the jury with the relevant portion of the testimony and the jury, eventually, found applicant guilty of both murders. The punishment phase of the trial followed, and the jury assessed applicant’s punishment at 99 years imprisonment in both cases. Applicant appealed, and the court of appeals affirmed the convictions. Allen v. State, Nos. 05-83-01297-CR, 05-83-01298-CR (Tex. App.— Dallas, February 7, 1985, pet. ref’d).

Applicant filed several writs challenging these convictions. In the first writ, filed pro se in October, 1984, he raised newly discovered evidence (showing that he had been misidentified and therefore wrongfully convicted) as his sole ground for relief. That writ was denied without written order because actual innocence based on newly discovered evidence was not the proper subject of habeas review at that time. See Ex Parte Binder, 660 S.W.2d 103 (Tex. Crim. App. 1983).7

In the second writ, filed with the trial court in April, 1992, applicant raised four issues, only two of which are relevant for our purposes because they are similar to the ones raised in the present application: (1) the “State suppressed relevant, material and exculpatory evidence from the defense,” and (2) applicant was “denied his right to effective assistance of counsel relative to a motion for new trial proceeding.” The trial court held an evidentiary hearing, found that the application had merit because of the newly discovered evidence, and recommended granting a new trial “in the interest of justice.”8

The second application was filed in this Court in April, 1993. Applicant filed a supplement to that application in April, 1994, presenting “factual innocence” as an additional ground for relief. The trial court, in its supplemental findings, “remain[ed] of the opinion” that applicant was entitled to a new trial, and that the application had merit and should be granted. We recognized that “the trial judge believes that fairness requires a new trial,” but found that he had made no specific findings as to the trial counsel’s effectiveness, nor otherwise provided a “legal basis” for his recommendations. We therefore remanded the matter to the trial court to specifically determine whether, in its opinion, the trial counsel’s failure to raise the newly discovered evidence in his motion for new trial constituted ineffective assistance of counsel.

Upon remand, the trial court specifically found that the trial counsel “could have discovered the contents of [the newly discovered evidence] before trial”; and that, “[a]s a matter of law,” the trial counsel’s “failure to investigate constituted a lack of diligence that is not recognized as a basis for granting a new trial.” Yet, the trial court concluded that the trial counsel’s failure to raise the newly discovered evidence in the motion for new trial did not constitute ineffective assistance of counsel because he had “correctly concluded that a motion for new trial would be rejected” because of his own lack of due diligence in discovering that evidence. Thus, the trial court concluded that the application had “no merit with regard to the claim of ineffective assistance of counsel.” The trial court also failed to find any merit in applicant’s supplemental claim of “factual innocence” under the actual-innocence standard set out in State ex rel. Holmes v. Third Court of Appeals, 885 S.W.2d 8

Although this case was remanded to the trial court several times and the trial court consequently issued a number of findings, this was the only evidentiary hearing held. Therefore, all references to an evidentiary hearing are to that held on December 9 and 10, 1992. 389 (Tex. Crim. App. 1994).9 Pursuant to these findings, the trial court recommended that relief be denied; and this Court denied the application, without written order, in January 1995.

Applicant filed a third application pro se, in February, 1996, alleging ineffective assistance of trial counsel for his failure to adequately investigate the case and to timely discover the exculpatory evidence and present it to the jury during the guilt phase of the trial. We dismissed this application as a subsequent application in July, 1996.

In May 2004, applicant filed the instant application pro se, alleging, inter alia, ineffective assistance of counsel and denial of a fair trial (the due process claim). We recognized, in our December 7, 2005 order, that this was the first writ that applicant had filed since our decision in Elizondo. We also noted that “[i]n Elizondo this Court held for the first time that claims of actual innocence in a non-capital case are cognizable by a court in a post-conviction habeas corpus proceeding.” We therefore concluded that applicant’s claims were not barred by the Texas Code of Criminal Procedure, Article 11.07, Section 4, “[b]ecause Applicant’s prior actual innocence claims were denied at a time when such claims were not cognizable on habeas review.”

As we noted, however, the trial court had “again recommend[ed] granting Applicant a new trial, but again [made] no specific findings as to the basis for [that] recommendation.”10 We therefore remanded the matter to the trial court to resolve the issues. The trial court issued supplemental findings, but again did not make any specific findings regarding the issues in this case. Rather, it only concluded that (1) the trial counsel had not been ineffective, but that (2) applicant should be granted a new trial because claims of actual innocence were now cognizable in habeas corpus proceedings, and (3) because the newly discovered evidence of innocence was “so strong that the Court cannot have confidence in the outcome of the trial.” We filed and set this case to resolve the issues involved in applicant’s actual-innocence claim.

Discussion

In the case at bar, applicant asserts a Schlup-type claim of actual innocence11 based on new evidence discovered after trial. In Schlup, the Supreme Court pointed out the main difference between the Schlup-type and the Herrera12-type claims of actual innocence. As the Court noted, Herrera had presented his claim of innocence to support “a novel substantive constitutional claim.” 513 U.S. at 314 (emphasis added). Thus, under that theory, Herrera’s innocence would have rendered his execution a “‘constitutionally intolerable event,’” even if the proceedings that led to his conviction had been “entirely fair and error free.” Id. (citation omitted). In contrast, Schlup’s claim of innocence was “procedural, rather than substantive.” Id. Thus, “[h]is constitutional claims [were] based not on his innocence, but rather on his contention” that the procedural violations in the proceedings leading to his conviction “denied him the full panoply of protections afforded to criminal defendants by the Constitution.” Id. In other words, Schlup’s claim of innocence was “‘not itself a constitutional claim, but instead a gateway through which a habeas petitioner must pass to have his otherwise barred constitutional claim considered on the merits.’” Id. at 315 (quoting Herrera, 506 U.S. at 404). This “gateway” has been codified in our post-conviction habeas corpus statute as a requirement for relief for a subsequent application. See TEX . CODE CRIM . PROC. art. 11.07, § 4(a)(2).13

The Court used this important difference between the two types of actual-innocence claims to justify the requirement of a different burden of proof in the two cases. Thus, the Court considered it “appropriate to apply an ‘extraordinarily high’ standard of review” in a Herrera-type claim, in which “a petitioner has been ‘tried before a jury of his peers, with the full panoply of protections that our Constitution affords criminal defendants,’” id. at 315-16 (citations omitted), but not in the case of a Schlup-type claim, in which some aspect of the procedure itself is at issue. Id. at 316. In the latter case, the Court held that even “if the habeas court were merely convinced that those new facts raised sufficient doubt about [the petitioner’s] guilt to undermine confidence in the result of the trial without the assurance that that trial was untainted by constitutional error, [the petitioner’s] threshold showing of innocence would justify a review of the merits of the constitutional claims.” Id. at 317. In other words, “the petitioner must show that it is more likely than not that no reasonable juror would have convicted him in the light of the new evidence.” Id. at 327 (emphasis added).

In short, we keep in mind this important distinction between the two types of actual- innocence claims, and the separate standards of review applicable to them, as we examine the case before us, in which applicant “accompanies his [Schlup-type] claim of innocence with an assertion of constitutional error at trial.”14 Specifically, applicant alleges that his trial counsel was ineffective.15 We find the test for reviewing such claims in Strickland v. Washington, 466 U.S. 668 (1984). We note that Strickland does not provide a mechanical formula for such review. Rather, it states that “[t]he benchmark for judging any claim of ineffectiveness must be whether counsel’s conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result.” Id. at 686. Our “scrutiny of counsel’s performance must be highly deferential.” Id. at 689. Thus, we “must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Id. But the defendant can rebut the presumption by showing that (1) “counsel’s performance was deficient” and (2) “the deficient performance prejudiced the defense.” Id. at 687.

The Court did not give any specific guidelines for determining deficient performance. It simply advised that counsel is expected to provide “reasonably effective assistance,” and that such “reasonableness” should be measured under “prevailing professional norms.” Id. at 687-88. Essentially, counsel is obliged to fulfill “certain basic duties,” including “a duty to bring to bear such skill and knowledge as will render the trial a reliable adversarial testing process.” Id. at 688. Strickland provides greater guidance, however, in determining whether the deficient performance found in the first step prejudiced the defense. Specifically, Strickland requires the defendant to “show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different,” defining a “reasonable probability” as “a probability sufficient to undermine confidence in the outcome.” Id. at 694.

In the case before us, applicant argues that the trial counsel rendered ineffective assistance of counsel for failing to: (1) properly investigate the case and discover that the victim Sewell had named someone other than applicant as his attacker; (2) move for a continuance at trial to investigate the matter when he was surprised by Officer Clary’s testimony regarding Sewell’s statement to Clary in the ambulance; and (3) raise in his motion for new trial the issue of the newly discovered evidence that he discovered when he did have the matter investigated after trial.

We agree. We note at the outset that “a verdict or conclusion only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support.” Strickland, 466 U.S. at 696. This rule is especially significant in the present case in which the record shows that: (1) there were no eye witnesses (other than the victim Sewell himself) to any part of the crime; (2) no witnesses testified as to any pre-existing animosity between applicant and either murder victim; (3) no accomplice witness testified; (4) applicant made no confession nor any other self-incriminating statement; (5) none of the property stolen from the victims’ residence at the time of the murders was recovered from applicant’s possession; and (6) the State made no showing of a motive on applicant’s part for these murders. In fact, there was not even any physical evidence connecting applicant to the crime, except for that one palm print found on the roof of the car in which Lashbrook’s body was found. In short, the State, as it indicated in its closing argument,16 relied primarily on the palm print and Sewell’s statement, as reported by Officer Clary, that a “Billy Allen” had attacked him. But it is this very statement that was so strongly controverted by the newly discovered evidence that it caused the habeas court to lose confidence in the outcome of the trial and to repeatedly recommend granting applicant a new trial.

The trial counsel testified as follows at the evidentiary hearing. He had hired a private investigator17 to investigate the State’s case against applicant. The investigator had been a police officer himself with the Dallas Police Department for almost thirty years before he retired and became a private investigator. It was apparently because of this service that he knew the chief of the University Park Police and met with him to discuss the State’s case against applicant. The police chief did not know the details of the case himself and summoned Detective Holman, the lead investigator in the case, to share those details with the defense investigator.18 Detective Holman purportedly gave the defense investigator all the details, including a two-page police offense report by Officer Clary detailing the events of the night of the crime. The defense investigator then gave the trial counsel both his own report of his conversation with Detective Holman and the two-page police report that Detective Holman had given to him. But neither Detective Holman nor the two- page police report had mentioned anything about Sewell’s telling Clary in the ambulance that “Billy Allen” was one of his assailants. The trial counsel was, therefore, surprised to hear Officer Clary’s testimony to that effect and to receive a four-page police offense report from the State, at trial. It was on the third page of the report, which he had never before received, that Clary had reported Sewell’s telling him in the ambulance that “Billy Allen” had attacked him.

In spite of this surprise, however, the record shows that the trial counsel never asked for a continuance and pressed on with the trial, using his closing argument to suggest that the last two pages of the police offense report were added after the fact to strengthen Officer Clary’s testimony which in turn, counsel seemed to suggest, was a lie. It was only after the trial that counsel sent his investigator to find out if there were any other witnesses to Sewell’s possible statement in the ambulance. The investigator found two such witnesses, both firemen-paramedics from the University Park Fire Department, who were present in the ambulance when Clary had spoken to Sewell. According to the investigator’s interviews with these paramedics, and the paramedics’ own affidavits, the supervisor paramedic, Sonny Kvapil, recalled that Sewell did state “Billy Allen,” but that he had also mentioned a middle name. Kvapil was too busy at the time, however, to clearly hear what it was. But the other paramedic, Phil Castle, who was attending to Sewell at the time, did clearly hear Sewell name a “Billy Wayne Allen” as his attacker.19 Indeed, Castle has consistently told everyone that has spoken with him on the subject (for example, the trial counsel’s investigator, the Dallas County District Attorney’s Office investigator, and the 1992 habeas counsel’s investigator), submitted an affidavit, and also testified at the evidentiary hearing that “Mr. Sewell said over and over . . . that Billy Wayne Allen shot him”; that Sewell clearly named a Billy Wayne Allen “[a]t least five or six” times; and that Castle will “never forget that.” On cross examination, the State asked Castle if there was “any possibility that [he] could have misheard, that [Sewell] could have said some other name besides Wayne?” Castle replied, “No, ma’am.”

We note that later investigation by the habeas counsel’s investigator revealed that there was a Billy Wayne Allen, who even lived in the Dallas-Fort Worth area (the same area in which the victims lived) and had a criminal record with drug and robbery convictions. Indeed, the record shows that this Billy Wayne Allen was even a police suspect on the day of the murders. As Detective Holman testified at the evidentiary hearing,

I radioed back to the station when I left Presbyterian Hospital and gave them what information that I had at that time. They began checking any and all possibilities of a Billy Allen and came up with several at that time. One being a Billy Wayne Allen, the other one being Billy Frederick Allen. As a matter of fact on Billy Wayne Allen I showed that the computer check was made on him on the date of the offense at approximately 8:30 in the morning so I was full aware of a Billy Wayne Allen at that time.

Later in the testimony, Detective Holman stated that in fact he had three (not “several”) Billy Allens as suspects in this case and that Billy Wayne Allen was one of them. He explained that he chose to focus on Billy Frederick Allen (applicant) rather than Billy Wayne Allen because of the palm print found on the roof of the car in which Lashbrook’s body was found.20

The habeas counsel’s investigator also testified at the evidentiary hearing. He explained how he independently ascertained the existence of this Billy Wayne Allen. According to his testimony, he examined court records, public records and the affidavits submitted by the paramedics and other people who had tried to contact the police identifying themselves as associates of the two victims and claiming to have information regarding the murders.21 He also talked to Paramedic Castle as well as some of the people who had submitted the affidavits. Unlike the police, the habeas investigator even carried a picture of Billy Wayne Allen to ascertain which “Billy Allen” was being discussed.22 Based on his investigation, the investigator found that there was only one Billy Wayne Allen in the Dallas-Ft. Worth area that fit the “profile” at issue in the case, i.e., someone with a record of drug abuse and violence “the one everyone seemed to know.”

In addition to the testimony from the police, the trial prosecutor and the defense investigator acknowledging both the existence of this Billy Wayne Allen and his criminal record, the habeas court also received affidavits from people stating that this Billy Wayne Allen was well-acquainted with Sewell, that he was also seen in possession of some of the property stolen from the victims on the night of their murder and that he had even admitted to some people that he was the one responsible for Sewell’s murder.

In short, the habeas court heard a lot of testimony and received a lot of evidence at the evidentiary hearing that tended to show the existence of a Billy Wayne Allen who was more likely to have been the murderer than applicant who seemed to have no motive at all. Indeed, the fact that Sewell, even though he expressly claimed to have been shot in the head twice, cut and stabbed, and was visibly bleeding when Officer Clary arrived at the scene, still went to the trouble of naming the middle name (“Wayne”) “at least five or six” times such that the Paramedic Castle would “never forget” it, tends to confirm that Sewell in fact knew both Billy Frederick Allen (applicant) and Billy Wayne Allen and that he was doing his best to identify the right “Billy Allen” as his assailant.23

The trial counsel admitted receiving the information about the paramedics (regarding Sewell’s identification of his assailant ) from his investigator after the trial and before he had filed the motion for new trial on applicant’s behalf. Yet, he did not follow up on this information, as the habeas investigator did, and thus failed to discover all the information that the habeas investigator found.24 Indeed, the trial counsel failed to even raise the newly discovered evidence from the paramedics in that motion for new trial. As noted earlier, he explained this failure on the ground that the motion for new trial would not have been granted because the trial court would have found that the evidence in question could have been obtained earlier but for the the trial counsel’s lack of diligence. Indeed, the habeas court in 1994 specifically found that the trial counsel “could have discovered the contents of Mr. Castle’s statements before trial” and that as “a matter of law, . . . [the trial counsel’s] failure to investigate constituted a lack of diligence that is not recognized as a basis for granting a new trial.” Nevertheless, the habeas court was reluctant to find that the trial counsel’s failure to raise the claim of newly discovered evidence in the motion for new trial constituted ineffective assistance of counsel because the trial counsel had “correctly concluded that a motion for new trial would be rejected on grounds of due diligence.”

But as a well-known criminal defense attorney stated in his expert-witness affidavit,

[The trial counsel] was obligated to raise this issue in the motion for new trial and develop the record at the hearing. The trial court may have excused his lack of diligence, if any, and granted a new trial as a matter of fundamental fairness. Had the trial court done so, the State could not have appealed under the law then in effect. Assuming that the trial court had denied the motion, the issue could have been presented on direct appeal, and the appellate court could have found diligence under the circumstances or excused the lack thereof.

In other words, as that attorney added, the trial counsel “should have acted as an advocate, rather than a judge, and raised this issue without regard to his opinion of the outcome.” The attorney concluded with his belief that a “reasonably competent counsel, rendering reasonably effective assistance, would have included and developed this issue in a motion for new trial, and if denied, would have raised the issue on direct appeal.”

We agree.25 The trial counsel himself admitted his “lack of diligence” when he explained at the evidentiary hearing that he did not raise the newly discovered evidence at the motion for new trial because such evidence would have been rejected on the grounds of due diligence. The habeas court agreed with his assessment of the probable outcome of that motion, and we do not disagree with that probability. But such “lack of diligence” was precisely what deprived applicant of the opportunity to present critical evidence to the jurors and possibly obtain a favorable verdict at his trial, or at the very least, preserve an important issue for appeal: the testimony of an independent witness, who was found to be credible at least by the habeas court,26 that the victim repeatedly and clearly named someone other than applicant as his assailant, just hours before slipping into a coma that lasted until he died more than two months later.

Indeed, the importance of this evidence is underscored by the fact that the jurors, during their deliberations in the guilt phase of the trial, sent two notes, both of which sought to have the trial court clarify the name of the assailant that Clary reported Sewell had told him in the ambulance. Finally, the trial counsel does not claim that his decision not to raise the issue of the newly discovered evidence in the motion for new trial was based on some trial strategy. On the contrary, he agreed with his fellow criminal defense attorney, the expert witness, that he should have raised the newly discovered evidence claim in the motion for new trial. In short, we find that the “lack of diligence” admitted by the trial counsel, and recognized by the habeas court as “a matter of law,” was clearly deficient performance under the Strickland ineffective-assistance-of-counsel analysis.

The prejudice to the defense from this deficient performance is also clear. If the jury had disregarded Officer Clary’s suggestion on the stand that the “Billy Allen” that Sewell identified in the ambulance was applicant, Billy Frederick Allen, and instead believed Paramedic Castle’s testimony that Sewell had specifically and repeatedly identified a “Billy Wayne Allen” (and Paramedic Kvapil’s testimony, if the defense had put him on the stand, that Sewell did mention a middle name that Officer Clary never acknowledged in his testimony), then the only other critical evidence against applicant, according to the State’s own closing argument, was applicant’s palm print found on the roof of the car in which Lashbrook’s body was found.27 But the State’s own fingerprint expert testified that the print could have been left there two days before the murders. The defense then put on its own fingerprint expert (who had actually been the supervisor of the State’s expert when the latter was still in training), who testified that the print could have been left there several days, if not weeks, before the murders. The State did not even cross-examine this witness or try to rebut this evidence. The defense also offered an innocent explanation of how applicant’s palm print was left on the car, through the testimony of applicant’s wife who described how applicant had stood leaning against Sewell’s car, as the latter sat inside counting the money to give to applicant, just a day or two before the murders.

In short, it is unlikely that the State would have been able to gain applicant’s conviction based on the palm print alone. In light of this state of the evidence, it is not surprising that the habeas court found that the newly discovered evidence “of innocence [was] so strong that [it could not] have confidence in the outcome of the trial.” We agree; and because it was the trial counsel’s deficient performance that prevented such evidence from being presented to the jury in the first place, we hold that such deficient performance prejudiced applicant’s defense.

Conclusion

We hold that the trial counsel provided ineffective assistance to the applicant in this case. We, therefore, grant applicant relief from the judgment and remand him to the custody of the Sheriff of Dallas County.

DELIVERED: February 4, 2009.

DO NOT PUBLISH