WOMACK, J.,
delivered the opinion for a unanimous Court.
These are post-conviction applications for writs of habeas corpus under Article 11.07 of the Code of Criminal Procedure. The applicant alleges that the State knowingly used false testimony at his trial, in violation of his Fourteenth Amendment right to due process. We agree with the convicting court’s recommendation to grant relief.
I. Background
A. Factual Background and Trial
The applicant, then 22 years old, began an online relationship with L.S., then 13
years old, in January 2004. The applicant professed his love and sexual desire for L.S., who said that, although she was a virgin, she wanted to have sex with the applicant. The applicant and L.S. met in person on January 30. The applicant picked up L.S. and her friend J.R.(a 14-year-old girl) from the middle school they attended. He drove the girls to his apartment, where they all consumed alcohol and took Xanax pills. Eventually, the applicant led L.S. to his bedroom where she immediately fell asleep on the bed. J.R. fell asleep on the couch in the living room.
At trial, in July 2006, each girl recounted waking up at some point in the night with her pants and underwear removed and the applicant engaged in vaginal intercourse with her. Each girl said that she was heavily intoxicated, went back to sleep almost immediately, and did not remember many details. When she awoke the next morning, L.S. asked the applicant if they had had sex, and he confirmed that they had. He told J.R. that she and he had “messed around.”
The applicant took the girls, who were still under the effects of Xanax and alcohol, to their middle school. Witnesses described the girls’ behavior and physical condition on that day.
L.S.’s mother, Michelle, testified that L.S. had undergone extensive and continuing psychiatric treatment since the night at the applicant’s apartment. Michelle testified that L.S. had been repeatedly committed to intensive treatment facilities because of continuing behavior problems, including a drug overdose. After the State rested, the applicant presented no evidence, and the jury found him guilty of the sexual assault of J.R. and the aggravated sexual assault of L.S.
During the punishment phase, L.S.’s father, William, testified that after the assault, L.S. was “crying, sobbing, seemingly numb, seemingly somewhat in shock.” He “immediately” noticed changes in L.S.’s behavior; she became withdrawn and ceased being an outgoing person. After “a few months,” L.S. left her school because she was being taunted by her classmates. L.S. transferred to a different school, and spent a week in a psychiatric hospital. William testified that L.S.’s therapist suggested that L.S. be sent away for intensive treatment at boarding schools, and, though he first resisted this suggestion, “eventually [he and his wife] found out that the therapist was right.” Though William’s testimony gave no details of the specific timeline of L.S.’s treatments, he said that L.S. was sent away for “ten to eleven months,” first to treatment at a “wilderness school,” then to a boarding school, then back to the “wilderness school.” William estimated that the treatment had cost around $140,000, with insurance covering only $5,500. William suggested that 25 years would be an appropriate sentence for the applicant.
J.R.’s father then testified about the ways that J.R.’s behavior had changed since the applicant’s assault. The defense did not present any punishment evidence. The State asked for “something in the thirty to forty-year range when determining the punishment in this case.” The jury assessed the maximum punishment, 20 years, for the sexual assault of J.R., and 28 years for the aggravated sexual assault of L.S. The convictions and sentences were affirmed on appeal.
B. Habeas Proceedings
1. The July Police Report
Through a Public Information Act
request, the applicant’s habeas attorney secured the district attorney’s file on these cases. In the file was a folder labeled “Work Product,” which contained a police report.
The reporting officer detailed being dispatched to L.S.’s home on July 27, 2005. L.S. and her parents were having an argument because one of L.S.’s coworkers had told William that L.S., then 15 years old, was having a relationship with a 25-year-old man, Davis.
William told the police officer that L.S. had confessed to him that the relationship was sexual. When the officer spoke with L.S. alone, L.S. said that she had met Davis in September 2004, started having sex with him that month, and since then had been having sex with him every one or two weeks. According to the report, L.S. “stated that [Davis] treated her nicely but he did deal a large amount of drugs, sometimes in front of her.”
The police report also detailed later attempts to contact L.S. and her family, concluding with a September 2, 2005 entry noting that Michelle had informed the officer that L.S. had been sent away to a wilderness boot camp for rehabilitation. The report concludes: “Case will exceptionally cleared/closed [sic ] for lack of information of the [complainant].”
Also contained in the “Work Product” folder were emails showing that Jamie Harris, the assistant district attorney then prosecuting the case, knew of the July 27 incident on July 29. The copy of the police report contained in the “Work Product” folder was printed on November 16, 2005.
2. Habeas Hearing
The applicant applied for writs of habeas corpus, arguing that the State unconstitutionally suppressed the July police report
and presented false testimony in violation of the Fourteenth Amendment. The applicant argues that the State gave the jury the misleading impression that all of L.S.’s psychological treatment was the result of the applicant’s assault, but that the relationship between L.S. and Davis could have been partly responsible for L.S.’s need for treatment.
The convicting court held a fact-finding hearing on October 1, 2009.
Former chil
dren’s court advocate Patty Smith testified that in 2005 she spoke with William and Michelle about L.S.’s relationship with Davis. Smith knew then that L.S. was not sent away from home for therapy until sometime after the July 27 incident. Smith had told Harris about Davis and about her suspicions that L.S. was selling drugs and may have been initiated into a gang. Smith stated her belief, based in part on conversations with William and Michelle, that the drugs, gang initiation, and relationship with Davis were the result of the trauma caused by the applicant’s assault. Smith did not believe that the Davis relationship was a “traumatic” experience for L.S.
Dorian Cottar, the lead prosecutor during the trial, testified that the July police report was in the “Work Product” folder when he took over the case from Harris. Cottar read the report and determined that it was irrelevant to the applicant’s case. The only additional investigation he performed into the matter was to run a criminal background search on Davis; Cot-lar said he could not remember whether the search turned up anything, but he did not think it did. Cottar said that he was not aware of Davis being prosecuted for any offense related to L.S. He speculated that the “cleared/closed for lack of information of the [complainant]” notation means that the case was never shown to a prosecutor.
Cottar cast doubt on whether L.S. had actually had sex with Davis, pointing out that L.S.’s statement came during “a heated argument” with her parents. Cottar said that after being assaulted by the applicant, L.S. “lied to her parents about a lot of things.” When asked whether he intended to leave the jury with the impression that the applicant was the sole cause of L.S.’s psychological problems, Cottar responded: “I don’t know. I suppose. I don’t know.”
Cottar speculated that giving the jury evidence of the Davis relationship would have been harmful to the applicant because he considered the relationship a symptom of the psychological harm caused by the applicant’s assault. Cottar explained that he did not introduce the evidence at trial because he did not believe he needed it to get the verdict and punishment he wanted, and introducing the evidence would have unnecessarily exposed more of L.S.’s private life.
John Parras, the applicant’s attorney during the trial, testified that at the time of trial he was unaware of L.S.’s relationship with Davis and the allegations of drug use and gang membership. After his testimony, the court ordered the parties to present arguments on this application at a later date, and then recessed.
3. The August Police Reports
While preparing for arguments, Alycia Harvey, the prosecutor handling this writ application, looked up two police report numbers listed as “Related Cases” on the July police report. According to these new-found reports, written throughout the month of August, 2005, L.S. ran away from home on July 31, 2005 — four days after the incident described in the July police report. William and Michelle told police that they believed L.S. was with Davis.
The police were not able to find Davis, but did find his mother. In the presence of the reporting officer, Davis’s mother called Davis, who said that he did not know where L.S. was, and that L.S. had represented herself as being 17 years old. Going to Davis’s supposed address and tracking down his supposed car produced no information regarding L.S.’s location.
L.S. called home in the early afternoon of August 3 and said that she was waiting in a Wal-Mart parking lot and was ready to come home. Michelle called the police, who then met Michelle at the Wal-Mart where L.S. was waiting. The officer “could tell by [L.S.’s] actions and her eyes that she was on some kind of drugs.” L.S. told the officer that she was on “zan-bars [benzodiazepine], ex [‘ecstasy,’ an amphetamine], coke [cocaine], dust [phencycli-dine], and possibly other drugs.” L.S. said that she could not recall whether she had had sexual relations with anyone. An ambulance took L.S. to a hospital to have a rape test performed.
Appended to the August police reports is a laboratory report from the Harris County Medical Examiner’s office stating that vaginal swabs performed on L.S. on August 3, 2005 tested positive for blood and semen. The laboratory report is dated July 31, 2006.
The copy of these police reports contained in the record shows that they were printed on October 6, 2009. Harvey turned the August police reports over to the convicting court, which then ordered the State to turn them over to the applicant.
4. The Convicting Court’s Findings of Fact and Conclusions of Law
After the parties presented arguments, the convicting court largely adopted the applicant’s proposed findings of fact and conclusions of law. The convicting court found that the July police report was material to the - case, favorable to the applicant, and suppressed by the State in violation of
Brady.
The convicting court found that the State elicited testimony that falsely conveyed the impression that all of L.S.’s psychological treatment was made necessary by the applicant’s assault. The convicting court concluded that
[t]he prosecution’s suppression of this evidence and presentation of testimony creating the misleading impression that applicant was solely to blame for L.S.’s physical, emotional, and psychological problems undermine confidence in the sentences.
The convicting court recommended that we grant the applicant a new punishment hearing. We ordered that the case be filed and set in order to determine whether the State violated
Brady,
and, if so, whether the result of the trial would have been different had the State disclosed the suppressed evidence.
II. Analysis
A. The Applicant’s Ground for Relief
The applicant presents us with a single ground for relief, asserting that the State failed to disclose favorable evidence and “presented [L.S.’s] parents’ misleading testimony creating the false impression that her physical, emotional, and psychological problems resulted solely from her sexual encounter with applicant.”
False-testimony claims and suppressed-evidence claims are legally discrete.
In this case, the suppressed evidence was not inherently exculpatory, nor would it have generally impeached the credibility of any of the State’s witnesses. The July police report would have aided the applicant only by showing that specific testimony from Michelle and William was misleading. The applicant has not argued that the July police report was relevant for any other purpose.
The essence of the applicant’s claim, then, is that the State knowingly presented false testimony; the police report is merely evidence that the testimony was false (and that the State knew it was false). Both the State and the applicant have presented arguments regarding both the
Brady
and false-testimony claims. The convicting court made findings of fact addressing both claims, and its recommendation for relief was based on both claims combined. Because we believe that the false-testimony claim is the underlying issue here, we will address this matter as a false-testimony claim.
B. Use of False Testimony
1. False Testimony
A conviction procured through the use of false testimony is a denial of the due process guaranteed by the Federal Constitution.
The use of false testimony at the punishment phase is also a due-process violation.
A due-process violation may arise not only through false testimony specifically elicited by the State, but also by the State’s failure to correct testimony it knows to be false.
“It does not matter whether the prosecutor actually knows that the evidence is false; it is enough that he or she should have recognized the misleading nature of the evidence.”
Though the case law in this area frequently refers to “perjured” testimony, there is no requirement that the offending testimony be criminally perjurious.
“It is sufficient if the witness’s testimony gives the trier of fact a false impression.”
These rules are not aimed at preventing
the crime of perjury — which is punishable in its own right
— but are designed to ensure that the defendant is convicted and sentenced on truthful testimony.
2. The State’s Knowledge
To constitute a violation of due process under Federal precedent, the State must knowingly use false testimony.
This Court allows applicants to prevail on due-process claims when the State has unknowingly used false testimony.
Even under this expanded notion of due process, however, the State’s knowledge is still a relevant factor to determine the standard we use for reviewing an applicant’s habeas claim.
3. Materiality
The knowing use of false testimony violates due process when there is a “reasonable likelihood” that the false testimony affected the outcome.
We have characterized this as a requirement that the false testimony must have been material.
This standard is more stringent
(ie.,
more likely to result in a finding of error) than the standard applied to
Brady
claims of suppressed evidence, which requires the defendant to show a “reasonable probability” that the suppression of evidence affected the outcome.
The “reasonable likelihood” standard is equivalent to the standard for constitutional error, which “requires] the beneficiary of a constitutional error to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.”
C. Application to This Case
In an 11.07 writ proceeding, this Court is the ultimate factfinder.
The convicting court is the original factfin-der, and we will generally defer to and accept that court’s findings of fact and conclusions of law when they are supported by the record.
1. Was the Testimony False?
Michelle testified that the “biggest problem” with L.S.’s behavior after the applicant’s assault was that she always felt dirty and took two showers a day.
Michelle said that L.S. gained weight, required therapy, and “wasn’t the same socially.” The State then asked whether
there was “anything else specifically that you’ve noticed different about [L.S.] since January 31st of 2004 that we haven’t talked about.” Michelle said there was not. She mentioned only one intervening event between the applicant’s assault and L.S.’s intensive therapy: a drug overdose, about which she gave no details. William testified that after the applicant’s assault, he and Michelle “eventually ... found out” that L.S.’s original therapy was not helping and that she needed more intensive therapy, such as the wilderness camp and boarding school. He gave no information regarding intervening events.
The habeas record shows that Michelle and William sent L.S. away for intensive therapy nineteen months after the applicant’s assault. In that time, L.S. had a job at a restaurant and (she said) engaged in a long-term, intimate relationship with a 25-year-old drug dealer. Members of the prosecution team also believed that during this time L.S. was selling drugs and was initiated into a gang. Shortly after Michelle and William learned all this, L.S. ran away from home for three days. She came home on drugs and with evidence of sexual assault still inside her vagina, though she did not remember who assaulted her. Only after these events was L.S. sent away for the intensive therapy that Michelle and William described in their testimony.
The State argues that this testimony did not create a misleading impression, because the behavior problems detailed in the July and August police reports were themselves the result of the psychological trauma L.S. suffered from the applicant’s assault. The State cites cases in which mental-health professionals testified that promiscuity and drug use are symptoms of sexual assault,
and to a fact sheet from the Children’s Assessment Center stating that promiscuity is a symptom displayed by 38% of child sexual assault victims.
The convicting court did not accept this argument, and we shall defer to that determination if there is support for it in the record. The convicting court found that the gap in the evidence presented to the jury was so significant that Michelle and William’s testimony “creat[ed] a misleading impression of the facts.” Based on the gravity of the events that were left out of their testimony, the fact that the testimony glossed over the lengthy period of time between the applicant’s assault and L.S.’s intensive therapy, and the fact that the testimony attributed all of L.S.’s psychological treatment to harm done by the applicant, we believe there is support in the record for the convicting court’s finding.
2. Did the State Know the Testimony was False?
The convicting court found that the State had knowledge of the July police report, and that the State was aware that L.S. was sent away for intensive therapy only after her parents learned of the Davis relationship. The State does not contest these findings, and they are supported by the record.
There are no findings of fact related to the August police reports, nor is there any testimony from the habeas hearing regarding them. The record could support a finding that the State should have known of the incident detailed in the August police reports,
but the habeas court did not
consider the issue. The habeas court’s recommendation is supported by its findings that the knowledge of the July police report and the timing of L.S.’s intensive therapy was sufficient to alert the State that the testimony from Michelle and William was misleading.
3. Was the False Testimony Material?
The convicting court found that the misleading testimony “undermine[s] confidence in the sentences,” a materiality standard it derived from
United States v.
Bagley
and
Kyles v.
Whitley.
This is the materiality standard for
Brady
claims, which is less likely to find error than the standard applicable to false testimony claims. If the convicting court found that the applicant met the
Bagley
standard, then it necessarily found that the applicant met the false-testimony standard.
For the aggravated sexual assault of L.S., cause # 1030953-A, the possible sentence ranged from 5 to 99 years, or life, and the applicant received 28 years. For the sexual assault of J.R., cause
#
1030954-A, the possible sentence ranged from 2 to 20 years, and the applicant received 20 years.
The main evidence the State presented at the punishment phase came from the girls’ fathers,
who described how the girls reacted to the applicant’s assault. The misleading testimony in this case amplified the impact that the applicant’s actions had on L.S. We accept the convicting court’s finding that there is a reasonable likelihood that the false testimony resulted in a harsher punishment in cause 1030953-A.
The false testimony did not directly bear on the sexual assault of J.R., but that does not end our analysis of whether the false testimony affected the applicant’s sentence in her case (cause 1030954-A). First, we note that J.R.’s behavior did not change so markedly as L.S.’s. J.R. began sleeping on the floor at the foot of her parents’ bed, was the victim of taunting at school, and performed uncharacteristically poorly at school. She underwent two years of therapy. In her guilt-phase testimony, J.R. said that she had gone to group-therapy sessions once per week for two years after the applicant’s assault. She also described thinking of suicide in the days after the
applicant’s assault. We also note, as the State did in its guilt-phase jury argument, that J.R. played no role in soliciting sex with the applicant and seems to have gone with the applicant to keep L.S. company.
The applicant received the maximum sentence for J.R.’s case, which was still less than the sentence he received for cause L.S.’s case. Had it been otherwise
{e.g.,
had the applicant not received the maximum for J.R.’s case, or had the punishment for J.R.’s case exceeded that for L.S.’s case), then we might accept an argument that the false testimony regarding L.S.’s treatment did not affect the sentence the applicant received for J.R.’s case. In this circumstance, however, we cannot.
In its punishment-phase jury argument, the State made no mention of what the punishment should be in cause 1030954-A. Instead, the State referred to William’s testimony that 25 years was an appropriate punishment for the aggravated sexual assault of L.S. The State then argued that 25 years was insufficient to protect society, and that the jury should “look in the thirty to forty-year range when determining the punishment in this case.” All of these numbers exceeded the maximum possible punishment for cause 1030954-A. At the habeas hearing, the applicant argued that the jury assessed 28 years for cause 1030953-A and used that as a baseline from which to assess 20 years for cause 1030954-A.
There is evidence in the record to support the convicting court’s conclusion that there is a reasonable likelihood that the false testimony that affected the applicant’s sentence in cause 1030953-A also affected the applicant’s sentence in cause 1030954-A.
D. Harmless Error
To obtain relief on habeas review, in addition to showing that a due-process violation occurred, an applicant may be required to show that the due-process violation was not harmless.
In two types of situations, this Court has applied a harmless-error standard to habeas claims alleging the use of false testimony.
1.
Ex parte Fierro
The applicant in
Ex parte Fierro
complained that the State knowingly used false testimony at a pre-trial suppression hearing. Mexican police had held Fierro’s parents in custody while the El Paso police interrogated Fierro. The police officer to whom Fierro confessed testified at a pretrial suppression hearing that he did not know of the actions taken by the Mexican police. Evidence showing that the officer’s testimony was false was available, to Fierro at the suppression hearing, but he seems to have made no use of it until his habeas application.
We determined that the State’s knowing use of false testimony violated due process, but was subject to harmless-error analysis.
Not only did the applicant have to meet the “reasonable likelihood” materiality standard to show that the false testimony violated due process, he also had “to prove by a preponderance of the evidence that the [violation] contributed to his conviction or punishment.”
We limited this holding, however, to situations where the applicant could have raised the issue on direct appeal. We “expressed] no opinion” whether the preponderance-of-the-evidence standard would apply if an applicant could “demonstrate that he had no oppor
tunity to raise the issue on direct appeal.”
2.
Ex parte Chabot
The applicant in
Ex parte Chabot
complained that the State had unknowingly used false testimony at his trial.
An accomplice witness, Pabst, testified that Chabot raped and killed a woman while he, Pabst, was in a different room. Pabst specifically denied sexually assaulting the victim. Twenty years after the trial, new tests revealed Pabst’s DNA on samples taken from the victim’s vagina, proving that Pabst’s testimony was false. Neither the State nor the applicant knew at the time of Chabot’s direct appeal that Pabst’s testimony was false.
We determined that the State’s unknowing use of false testimony was subject to the same preponderance-of-the-evidence standard we applied in
Fierro.
Chabot
did not address the question left open in
Fierro,
of whether the preponderance-of-the-evidence standard should apply to the State’s knowing use of false testimony when the applicant could not have raised the issue on direct appeal.
3. State’s Knowing Use of False Testimony Unknown to the Applicant
Here we are faced with a situation left unaddressed by
Fierro
and
Chabot.
The State used Michelle’s and William’s testimony, which created the false impression that L.S.’s psychological treatment was solely attributable to the applicant’s assault. Yet the State was aware of evidence that intervening events may have caused Michelle and William to seek treatment for L.S. The applicant did not have access to this information at trial, and learned of the July police report only when his habeas attorney gained access to it through a Public Information Act request.
When the State knowingly uses false testimony, the determinative factor in whether the defendant can raise the issue on direct appeal is, frequently, how well the State hides its information. In
Fierro,
for example, the State turned over to the applicant, at the pre-trial suppression hearing, a police report showing that testimony at the suppression hearing was false. Fierro failed to raise this matter at trial, in a motion for new trial, or on direct appeal. When the claim came to us on habeas review, Fierro had already had the opportunity to present the claim in circumstances where it would not have been subject to the preponderance-of-the-evidence standard, but only to the reasonable-likelihood standard of materiality.
In contrast, the applicant in this case had no opportunity to present his claim on direct appeal, in large part because the State concealed information suggesting that the testimony was misleading. Under the Harris County District Attorney’s open-file policy applicable at the time of the applicant’s trial, defense counsel was free to examine most documents in the prosecutor’s file, but not documents in ‘Work Product” folders. The lead prosecutor knew of the July police report, knew that it was in a Work Product” folder, and intended that the applicant not have access to it because the prosecutor did not consider it relevant to the case.
When a habeas applicant has shown that the State knowingly used false, material testimony, and the applicant was unable to raise this claim at trial or on appeal, we will grant relief from the judgment that was obtained by that use.
We grant relief from the judgments as to the punishments in these cases. The convicting court may hold a new punishment hearing in both causes.