IN THE COURT OF CRIMINAL APPEALS OF TEXAS NOS. WR-19,518-13, -14
EX PARTE RAY HIGHTOWER JR., Applicant
ON APPLICATIONS FOR WRITS OF HABEAS CORPUS IN CAUSE NOS. 00353-D & 00354-D IN THE 350TH DISTRICT COURT FROM TAYLOR COUNTY
HERVEY , J., filed a concurring opinion in which RICHARDSON and NEWELL, JJ., joined.
CONCURRING OPINION
It’s almost the norm now that an applicant who raises an Article 11.073 junk-science
claim will also raise a false-evidence claim. I write separately because I think it would be
worthwhile to address the relationship between the claims.
I. ARTICLE 11.073
An applicant can obtain relief under Article 11.073 if he can show by a preponderance of
the evidence that he would not have been convicted had the new scientific evidence been
available at the time of trial. TEX . CODE CRIM . PROC. art. 11.073(b). Scientific evidence is
unavailable at trial if it could not have been discovered through the exercise of reasonable Hightower–2
diligence. Id. art. 11.073(b)(1)(A). To determine if “relevant scientific evidence was not
ascertainable through the exercise of reasonable diligence . . . the court shall consider whether the
field of scientific knowledge, a testifying expert’s scientific knowledge, or a scientific method on
which the relevant scientific evidence is based has changed . . . .” Id. art. 11.073(d). One example
of a change in a field of science that affects trial evidence can be found in our decision in Ex
parte Chaney. Ex parte Chaney, 563 S.W.3d 239 (Tex. Crim. App. 2018). In that case, this Court
held that the field of scientific study regarding human-bitemark identification had evolved in a
way since trial that almost completely discredited the State’s bitemark evidence, which was an
essential part of its case. Id. at 260. (The mark was identified as a human bitemark, was
determined to have been inflicted at the time of the murders, and was “matched” to the applicant
to the exclusion of all others.) Id. at 262. An example of a change in expert knowledge affecting
a case is found in our Ex parte Robbins decision. Ex Parte Robbins, 478 S.W.3d 678 (Tex. Crim.
App. 2014) (Robbins II). There, the applicant’s postconviction evidence showed that the medical
examiner, who testified for the State at trial that the infant victim died of intentional asphyxiation
by compression, would revise her testimony based on further experience as a practicing forensic
pathologist and would now testify that the cause of death was undetermined. Id. at 692.
II. FALSE EVIDENCE
a. Background
“To prove a false-evidence habeas corpus claim, a claimant must first show that the
evidence in his or her case was false and then that the false evidence was material.” Ex parte
Barnaby, 475 S.W.3d 316, 323 (Tex. Crim. App. 2015). An applicant can show materiality by
establishing by a preponderance of the evidence that there is a reasonable likelihood that the error Hightower–3
contributed to his conviction or punishment. Giglio v. United States, 405 U.S. 150, 153 (1972);
Ex parte Chabot, 300 S.W.3d 768, 771 (Tex. Crim. App. 2009); Ex parte Fierro, 934 S.W.2d
370 (Tex. Crim. App. 1996).
Under our false-evidence jurisprudence, an applicant must prove falsity by a
preponderance of the evidence by presenting “[d]efinitive or highly persuasive evidence.” Ex
parte De La Cruz, 466 S.W.3d 855, 867 (Tex. Crim. App. 2015). The testimony need not be
strictly false. See id. at 866. It is sufficient if the testimony, when considered in its entirety,
misled the jury.1 Id. An applicant can rely on evidence showing that the expert testimony was
false at the time of trial, or they can rely on evidence showing that the trial testimony is now
known to be false in hindsight.2 We have found testimony to be false when a “witness omitted or
glossed over pertinent facts.” Ex parte Robbins, 360 S.W.3d 446, 462 (Tex. Crim. App. 2011)
(Robbins I). But we have found expert witness testimony was not false when the witness
“testified openly” about their professional opinion, “did not omit pertinent details,” and the
testimony was not “entirely refuted by any expert.” Id. False-evidence claims in this context tend
to fall into four categories:
(1) scientists who intentionally make false statements at trial about the field of scientific knowledge,
1 To me, this is an important distinction because it means that accurate statements made by a scientist at trial might later be found to be“false” through no fault of the scientist. 2 Chaney, 563 S.W.3d at 264 (“one to a million” testimony at trial was false when given); see Ex parte Henderson, 246 S.W.3d 690 (Tex. Crim. App. 2007) (new evidence that short- distance falls could have caused the infant victim’s head injuries was a material exculpatory fact undermining expert testimony at trial that short-distance falls could not cause such injuries). Although Henderson was not a false-evidence case per se, Dr. Bayardo’s testimony was effectively determined to be false in light of intervening developments in the scientific community’s understanding of short-distance falls. Hightower–4
(2) scientists who unintentionally make false statements at trial about the field of scientific knowledge,
(3) scientists who make statements at trial about the field of scientific knowledge that were correct based on the science as it was understood at the time but which have since been undermined (rendered misleading) by intervening scientific developments, and
(4) scientists who make statements at trial about the field of scientific knowledge that were correct based on their scientific knowledge but which have since been undermined (rendered misleading) because the scientist would now revise those statements after acquiring additional knowledge about the field.
b. Precedent
In Ex parte Henderson, 384 S.W.3d 833 (Tex. Crim. App. 2012) (per curiam)
(Henderson II), the infant victim suffered catastrophic head injuries, and the State argued that the
applicant intentionally caused those injuries. The applicant claimed that the injuries were caused
when she accidentally dropped the victim a few feet onto a concrete floor. Id. at 833–34. Dr.
Bayardo, an assistant medical director, testified for the State. Id. at 832. He said that the cause of
death was “homicide” and that applicant’s claim about the short-distance fall was “false” and
“impossible” because the victim’s type of injuries could not have been caused by a short-distance
fall. Id. at 833–34. During postconviction proceedings, however, Dr. Bayardo said that he would
revise his trial testimony based on intervening developments in the scientific field. Id. He would
now testify that the cause of death was “undetermined” because “the medical profession ha[d]
gained a greater understanding of pediatric head trauma” and the types of injuries the infant
victim suffered could be caused by a short-distance fall onto concrete. Id. at 839 (Cochran, J.,
concurring). Calling Dr. Bayardo’s “re-evaluation of his 1995 opinion, which he states is based
upon new scientific developments, . . . a material exculpatory fact,” we remanded the cause for Hightower–5
further proceedings. Id. at 834.
In Chavez v. State, No. AP-76,291, 2010 WL 4638619 (Tex. Crim. App. Nov. 17, 2010)
(not designated for publication),3 a medical examiner testified for the State that unknown hair
found at the victim’s residence shared physical characteristics of a hair from the appellant, but
DNA testing excluded the appellant as the donor of the unknown hair. Id. at *1. We concluded
that the examiner’s testimony was not false, reasoning that she “clearly stated that she could not
positively identify the unknown hair as [the appellant’s]” and testified only that “physical
characteristics of those hairs were consistent with the physical characteristics of [the appellant’s]
hair.” Id. at *7. The examiner’s testimony about the physical similarities, we noted, was not
refuted. Id.
In Robbins I, the medical examiner who performed the autopsy of the infant victim
testified that the victim died of asphyxiation due to the compression of her chest and abdomen.
Robbins I, 360 S.W.3d at 448. On cross-examination, the examiner testified that, while she could
not exclude other causes that could reasonably explain the victim’s death, she believed beyond a
reasonable doubt that the infant victim was killed by asphyxiation. Id. at 451. During
postconviction proceedings, the examiner said that she would revise her trial testimony and that,
now, she would say that the cause of death is “undetermined,” although she still thought that the
death was suspicious. Id. at 454. She based her revised opinion on additional experience as a
medical examiner and additional information she learned about the investigation since trial. Id.
We concluded that the trial testimony was not false because it “did not result in a false
3 I cite this case even though it was unpublished because much of the reasoning appears to have been adopted by the Court in Robbins I. Hightower–6
impression of the facts,” noting that the examiner testified openly about her opinion on the cause
of death, and her reasoning for drawing that conclusion. Id. at 462. She also did not omit
pertinent details. Id. We also analogized the case to Chavez, noting that the medical examiner’s
trial testimony was not completely refuted because she did not definitively testify that the cause
of death was intentional asphyxiation, only that she believed beyond a reasonable doubt that it
was, and none of the new evidence persuasively showed that the victim could not have been
intentionally asphyxiated. Id. Further, we distinguished Henderson on the basis that Dr.
Bayardo’s testimony was attributable to intervening scientific developments, whereas the medical
examiner’s testimony in Robbins I was based on a re-evaluation of “the same autopsy findings
and other information that she had at the time of trial.” Id. at 463.
Judge Cochran filed a dissenting opinion (joined by two judges), in which she argued that
the issue was whether the scientific evidence was so unreliable that the applicant’s trial was
fundamentally unfair and the risk of an inaccurate verdict too high, not whether the medical
examiner’s trial testimony was false. Id. at 471 (Cochran, J., dissenting). Judge Alcala also filed a
dissenting opinion (no judges joined), and she argued that the medical examiner’s testimony was
false because it left the jury with a false impression when considered in its entirety. Id. at 478
(Alcala, J., dissenting). Judge Alcala acknowledged that the examiner’s trial testimony had not
been wholly refuted, but she thought that it was sufficiently refuted to be considered false given
that the examiner’s “present position acknowledges that the cause and manner of death could
possibly be natural causes or homicide and that both are equally likely.” Id.
In Henderson II, 384 S.W.3d at 833, the Court again addressed the testimony of Dr.
Bayardo and his conclusion that the infant victim was intentionally killed and that the victim’s Hightower–7
catastrophic head injuries could not have been caused by the accidental short-distance fall as
claimed by the applicant. We issued a per curiam opinion in which we accepted the trial court’s
recommendation that the applicant should be given a new trial without specifying why. Id. at
834. In side opinions, judges struggled with how to characterize the claim. Judge Price argued in
his concurring opinion (no judges joined) that Dr. Bayardo’s testimony was unintentionally false.
Id. at 834 (Price, J., concurring). Judge Cochran argued in her concurring opinion (joined by
three judges) that the applicant was entitled to a new trial because, like the medical examiner in
Robbins I, Dr. Bayardo’s testimony had been rendered unreliable by intervening developments in
science, undermining faith in the accuracy of the verdict. Id. at 844 (Cochran, J., concurring).
Judge Alcala joined Judge Cochran, but she also wrote her own concurring opinion (no judges
joined) in which she argued that the key difference between Robbins I and Henderson II is that
the former was about whether the medical examiner’s testimony was false, but the latter was
about whether the applicant was entitled to a new trial based on new scientific understandings
about short-distance falls and the types of injuries they can cause in infants.
Judge Keasler (joined by Presiding Judge Keller and myself), wrote a dissenting opinion
in which he argued that Dr. Bayardo’s testimony was not false because his revised opinion that
the cause of death was uncertain did not exclude the possibility that the cause of death was
homicide. Id. at 851 (Alcala, J., dissenting). I wrote a dissenting opinion (joined by Judge
Keasler and Presiding Judge Keller), in which I agreed with Judge Cochran that the issue was
about the accuracy of the verdict, not whether the testimony was false, but I argued that Dr.
Bayardo’s trial testimony was not so unreliable as to call into question the accuracy of the
verdict. Id. at 861 (Hervey, J., dissenting). I also pointed out that even if there was a due process Hightower–8
violation, the Court did not address harm. Id. at 860.
c. What our precedent suggests
Our jurisprudence suggests a few things. First, the definitiveness of the scientific
testimony is of paramount importance. If the scientific trial testimony was definitive, and the new
evidence wholly refutes that testimony, that testimony was false. For example, if a medical
examiner testified that “Thing A” caused the victim’s death, but it is later determined that “Thing
B” caused the victim’s death, the trial testimony and the new understanding are mutually
exclusive and cannot both be true. Most scientific testimony, however, is not definitive. When
the trial testimony is equivocal, the first question seems to be whether the universe of possible
explanations given by the scientist at trial included the possibility supported by the new evidence.
For example, a medical examiner might testify that they believe beyond a reasonable doubt that
“Thing A” caused the victim’s death, but they could not rule out the possibility that “Thing B”
caused the victim’s death. We have held that such trial testimony did not mislead the jury so long
as the examiner openly testified about his professional opinion, did not gloss over or omit
pertinent facts, and the examiner’s opinion has not been “entirely refuted by any expert.” On the
other hand, a medical examiner could testify that they believed that “Thing A” or “Thing B”
caused the victim’s death, but new scientific evidence might show that “Thing C” caused the
victim’s death. Although equivocal, the testimony is nonetheless misleading because the
examiner misled the jury by suggesting that only “Cause A” or “Cause B” could have caused the
victim’s death.
d. This case
Here, the State alleged that Applicant kidnapped the victim. Police recovered unidentified Hightower–9
hairs and a wig from Applicant’s truck. Police thought that Applicant might have worn a wig
when he kidnapped the victim because, while the victim described Applicant as having hair, he
was actually balding. As part of its case, the State presented a hair-comparison expert. The expert
told the jury that, unlike fingerprinting, hair comparison was not an exact science and that he
could not determine whether a particular hair came from a specific person. He explained that he
examines hair using a microscope to compare the individual characteristics of the hairs, including
the medulla, pigmentation, coloration, width, and length of the hairs, and that he uses a
“polarmetric type microscopy” test to “verify” his results. According to the expert, one of the
hairs found in the truck matched a known sample from the victim “in all respects.” The expert
meant that the hairs “matched up” in their physical characteristics, not that the unidentified hair
positively came from the victim. While testifying he used phrases like, “same as,” “same,” “same
characteristics,” “matched,” “matches,” and “same match.” When asked what the likelihood was
that hair from two people would “match,” the expert testified that “from the studies that have
been made and that I’ve read and literature in—on this subject, I believe the best analysis is that
one in 20,000 people have the same type of hair.” He later clarified that it would be 1 in 20,000
Caucasians with blonde or light brown hair. Applicant did not say and was not asked what
studies or literature he relied upon.
In addition to “matching” the victim’s hair to a hair found in Applicant’s truck, the expert
also testified that two of the five hairs “could have come from wig hair.” He explained that he
went to wig stores and “picked up numerous wig hairs samples from wig stores” to
microscopically examine, and according to him, all of the hairs on older wigs “showed the same
type of characteristic breakage or damage as the two hairs.” He said that “the hairs appeared very Hightower–10
damaged” and that “the breaks in them were very rough, and sharp . . .” Notwithstanding that, he
expressly declined to positively state that he knew the two hairs came from a wig. On cross-
examination, the expert said it was just a visual comparison and was not based on any objective
testing, only his experience.
Applicant argues that the follow statements by the expert were false:
(1) Only 1 in 20,000 Caucasians had hair with characteristics similar to Applicant’s hair was not supported by the science at the time of trial,
(2) The expert bolstered his “matching” conclusions by “verifying” them with polarized light microscopy (PLM)—even though hair comparison scientists have never used that test for verification, and
(3) The hair from Applicant’s truck was from a wig was based on his novel theories of hair morphology, which appear to have no basis in the literature or scientific studies, even at the time of trial.
For the reasons I explain, I think that Applicant is right about the falsity of the statistical
testimony but not the PLM or wig testimony. Applicant agrees that the hair-microscopy
community believed that statistically significant associations could be supported by
microscopically comparing hairs at the time of trial, but he argues that the community now
recognizes that the science cannot support the statistics used at trial. To the contrary, the new
evidence indicates that the community now appears to believe that, at most, an examiner can
testify that the contributor of a known hair could be included in a pool of people of unknown
size, which has little probative value. Even though the expert testified openly, this seems to
establish that intervening developments in the hair-microscopy community’s understanding of
microscopic hair comparisons has changed in a way that renders the expert’s trial testimony on
the issue misleading. Hightower–11
There is also another component to the statistical testimony. Applicant argues that,
although some statistics were thought to be supported by the science at the time of trial, the
expert’s “1 in 20,000 Caucasians” statistic was never supported by the science. Applicant’s
experts cite a 1974 study, which “estimated a probability of 1 in 4,500 . . . scalp hairs” possibly
having common physical characteristics. That statistic apparently applied to all people with scalp
hair, not only Caucasians with blonde to light brown hair. This issue is made more difficult
because the judge overruled a defense objection asking the expert to disclose the literature he
relied upon. Without knowing the basis of the testimony, it might be near impossible for an
applicant to show that trial testimony has been undermined or was false. Here, Applicant’s
experts, although aware of the 1974 study, were not aware of any study or literature that would
support the “1 in 20,000” or the “1 in 20,000 Caucasians with blonde to light brown hair”
statistics. As Applicant points out, even if the expert had testified that 1 in 20,000 people have
hair with the same physical characteristics, he still overstated the likelihood of a match orders of
magnitude above what the science at the time of trial supported. The statistical testimony is
problematic.
Applicant also argues that the expert unintentionally used false evidence when he claimed
to verify his microscopic comparisons using PLM. According to Applicant’s experts, “there is
little use for PLM in a microscopical hair comparison and it certainly does not verify the results
of the other microscopical observations or techniques used during the comparison process.” But
Applicant has not shown that the expert’s testimony that he could use PLM to verify that the
hairs “matched” in their physical characteristics was false. His experts limit their discussion of
PLM to a few sentences, none of which explain, nor do they cite to studies or literature that Hightower–12
explain, what PLM is used for or why it would not be scientifically helpful to use a PLM to
corroborate microscopic-hair comparison results. The burden in postconviction proceedings is on
the applicant. I also note that Applicant’s argument is built on the presumption that the expert’s
“matching” testimony has been discredited, but as I already explained, the expert testified only
that the physical characteristics of the hair matched, not that the hairs came from the same
person. Applicant has provided no evidence refuting that the hairs have the same physical
characteristics.
Applicant argues that the expert’s wig testimony was unintentionally false because it is
now known that his conclusion that one of the hairs from the truck had the same physical
characteristics of wig hairs he had obtained from various stores can no longer be supported by the
science. Applicant relies on a report from the FBI about common flaws in microscopic-hair
comparison testimony. According to him, the expert committed a Type III error (as defined by
the FBI report) which is when,
the examiner cites the number of cases or hair analyses worked in the lab and the number of samples from different individuals that could not be distinguished from one another as a predictive value to bolster the conclusion that a hair belongs to a specific individual. This type of testimony exceeds the limits of the science.
FBI, MICROSCOPIC HAIR COMPARISON ANALYSIS 1 (Nov. 9, 2019). What the expert actually
explained was that he examined the hairs recovered from the truck and compared them to hairs
from various wigs obtained from different stores to determine if any of the hairs appeared like
they might have come from a wig. He did not purport to identify a specific wig that the hairs
from the truck came from, like the wig recovered by police. The expert was also candid that his
comparison was objective and based only on his own experience. On this evidence, I do not find Hightower–13
these statements by the expert to be false for purposes of a due-process false-evidence analysis.
III. CONCLUSION
Applicant makes three allegations of false evidence. I agree that the statistical testimony
is false because it has been rendered misleading over time, but I believe that Applicant failed to
meet his burden of proof to show falsity as to the PLM testimony and the expert’s wig testimony.
Despite the problematic statistical testimony, it was nonetheless immaterial in light of the other
overwhelming, inculpatory evidence. As this analysis shows, determining whether scientific
evidence is false can be a difficult issue to resolve.
With these comments, I concur in the disposition of the Court.
Filed: May 12, 2021
Publish