Cobb, Travis

IN THE COURT OF CRIMINAL APPEALS
OF TEXAS

NO. WR-95,984-01

EX PARTE TRAVIS COBB

ON APPLICATION OF WRIT OF HABEAS CORPUS
CAUSE NO. 2018CR11604-W1 IN THE 399TH DISTRICT COURT
BEXAR COUNTY

RICHARDSON, J., filed a concurring opinion.

CONCURRING OPINION

In this case, the State and the habeas court agree with the applicant that he is entitled

to actual innocence relief. Indeed, the State actually knows who actually committed the

crime based on new DNA evidence. Despite that, in an unsolicited crusade now lasting

almost a decade, a member of this Court continues to advocate that the Elizondo standard

is not rigorous enough to identify the “actually innocent” from the hundreds of non-

meritorious cases this Court has received. The 30-year established burden is a "Herculean"

standard with the burden resting solely on the applicant. Yet, Judge Yeary's familiar

concurrence continues to advocate that Elizondo is not rigorous enough. Is it now time to
raise the bar and make it more difficult for applicants to obtain that relief? One might ask

why now after 30 years. Has either side asked us to do that in this case? That answer is no.

In the 30 years applicants have been required to meet that herculean Elizondo standard, has

any county attorney, district attorney, the State Prosecuting Attorney (SPA), or AG asked

us to revisit that standard? Again, that answer is no. In this case the State is not asking this

court to raise the bar for actual innocence, nor is it even challenging the Elizondo standard,

they agree this applicant did not commit the crime and they know who did. When is it

proper for Judges on this court to be stronger advocates for the State than ... well, the

State? 1

Applicant, Travis Cobb, was convicted of an aggravated robbery in San Antonio,

Texas and sentenced to twenty-five years of incarceration. After spending five years in

prison, new exculpatory evidence, including DNA tests from a cup the true culprit was

recorded drinking from, clearly and convincingly shows Applicant is actually innocent of

the crime he was convicted of. I join the Court in that finding. I write separately to address

the claim—though not a genuine issue in controversy—that Applicant should not be called

“actually innocent” because he has not proven it beyond a reasonable doubt, and that the

Elizondo standard is insufficiently rigorous.

Facts of the case

1
It violates Due Process when a judge fails to be impartial and “hold the balance nice,
clear, and true between the state and the accused.” Caperton v. A.T. Massey Coal Co., Inc., 556
U.S. 868, 878 (2009) (quoting Tumey v. Ohio, 273 U.S. 510, 532 (1927)).
2
In the July of 2017, a suspect entered a Walmart in San Antonio and began

concealing merchandise in his backpack. The suspect was a white male with a beard and

wearing a medical boot on his right leg. The suspect’s visit was recorded in a series of

Walmart’s surveillance video recordings. Though somewhat grainy and pixelated,

surveillance footage shows that the suspect was completely bald to the point his head was

shiny and wearing shorts and a short-sleeved shirt. It also recorded him drinking from a

Styrofoam cup as he moved about the store. Surveillance video then shows the suspect

wearing a backpack and moving to exit the store without paying for the acquired

merchandise. As the suspect neared the entrance, David Gamez, a loss-prevention officer

employed by Walmart, began pursuing him. Gamez managed to confront the suspect just

outside the exit and grabbed onto his backpack. The suspect then wiggled out of the

backpack harness and used a bladed object to cut Gamez causing significant injury. Gamez

was hospitalized for the severe wound he suffered.

The backpack, the stolen merchandise, and the Styrofoam cup were recovered. No

usable fingerprints were recovered from the cup and it was not tested for DNA. However,

a still-photograph of the back of the suspect as he was leaving the store was generated from

the surveillance video and distributed to police officers. Officer Preikschat of the San

Antonio Police Department saw the image told one of the investigating detectives that it

resembled Applicant because he had seen Applicant wearing a “leg brace” after Applicant

had fallen off a ladder sometime in the past. (3 RR 23). He further claimed that Applicant

was homeless matching characteristics associated with the suspect. A photo lineup of six

3
photographs including both bearded and non-bearded males was generated by Officer

Gallegos and shown to Gamez in a car outside his residence. Out of the six photographs,

Gamez positively identified Applicant as the perpetrator. On the photo-identification form,

he handwrote that he was certain of his identification of Applicant based on “facial features

such as facial hair and the subject’s skin color.”

Applicant was tried and subsequently convicted for aggravated robbery which was

affirmed on appeal. He was sentenced to twenty-five years of incarceration to which he

served five years before being released on bond during the pendency of his writ application

because the trial court believed him to be likely actually innocent.

The record on writ now reveals new facts that affirmatively point towards

Applicant’s innocence. We now know that Officer Preikschat’s identification of Applicant

based on a still-photo of the back of the suspect was inaccurate and misleading. Although

Officer Preikschat did not lie about knowing Applicant suffered a leg injury and that he

had seen him wearing a “leg brace,” he omitted that he had not seen Applicant in years and

did no subsequent follow-up to confirm that his information was still true. 2 We now know

that, by the time of the offense, roughly five years had passed since Applicant’s leg injury.

Officer Preikschat’s information misled others to believe he had seen Applicant wearing a

2
At the time of Applicant’s arrest, Applicant had an ongoing civil rights lawsuit against
the SAPD and the City of San Antonio for the use of excessive force arising from an unrelated
incident. The lawsuit eventually ended in a settlement. See Cobb v. San Antonio P.D., et al., No.
5:16-CV-00739-FB (W. D. Tex.—San Antonio Div. 2018).

4
“leg brace” fairly recent to the time of the offense. Medical records contemporaneous to

the time of the crime, including previously undisclosed medical records from the Bexar

County Jail, show that there was nothing wrong with either of Applicant’s legs.

Officer Preikschat’s claim that Applicant was homeless—a detail consistent to what

was known about the actual perpetrator at the time—was false. Applicant had been living

in an apartment provided by the U.S. Department of Veteran Affairs. He was also regularly

visited by a social worker employed by that agency. The social worker confirmed that

Applicant had no leg injury and had never seen Applicant wearing a cast or brace in any

visit leading up to the crime. Additionally, those who knew him attested that he never wore

shorts or short sleeved shirts.

We also know that the photograph lineup conducted in a car outside Gamez’s house

was flawed. First, without any explanation as to why, it was not presented in a double-blind

format as normally required under SAPD regulations. And although all parties likely acted

in good faith, Gamez clearly stated he identified Applicant in the lineup based on his beard

and skin color. This was in violation of the instructions which warned witnesses not to base

their identification based on changeable features such as “head/facial hair” and

“complexion colors.” SAPD Photo Identification Instructions, Writ R. at *480.

Dr. Trent Terrell, a veteran experimental psychologist specializing in eyewitness

memory and identification, also reviewed the photo-identification lineup for its reliability.

5
Having been consulted in over 80 criminal cases, he concluded that Gamez’s positive

identification was not reliable for the following reasons:

1. The witness provided no description of the suspect beyond his age, race, and
facial hair. His clothing and description of the leg cast have no bearing on a
facial identification. With such limited memory of a person to work with, the
reliability of a subsequent photo identification is certainly in question.
2. Latency, stress, weapon focus, and cross-racial identification likely further
reduce the reliability of the identification.
3. The witness identified the suspect based on a changeable feature (facial hair).
Had the suspect not been bearded in the mug shot photo, he likely would not
have been identified. This strongly calls into question the validity of the
identification.
4. The identification procedure was not recorded, despite the officer having the
means to do so. This raises questions about how that procedure was carried
out. We have no way of knowing how the procedure unfolded, or how certain
the witness was of his choice.

Aff. Dr. Trent Terrell (signed Oct. 7, 2020), Writ R. at *698.

The final, and perhaps most important, pieces to the puzzle were revealed in

Applicant’s post-conviction DNA testing of the Styrofoam cup that the perpetrator of the

attack was recorded drinking from. The test revealed the DNA profiles of two subjects—

neither of which were consistent with Applicant. The first DNA profile was a CODIS match

to Lisa Marie Rojas who was questioned extensively at the scene of the crime during the

original investigation. She was questioned at the time because she was seen entering the

Walmart with the suspect and talking with him. At the time of the original investigation,

Rojas claimed she had no knowledge of and was unrelated to the perpetrator in any way.

6
The record on post-conviction writ now reveals this was a lie. She was in a dating

relationship with Dustin Cody McCall. At the time of the post-conviction writ, McCall was

detained at the Bexar County Jail for unrelated offenses. A DNA sample was collected

from McCall under court order. It was matched successfully to the second DNA profile

found on the Styrofoam cup. Consistent with the DNA profile evidence, McCall’s physical

attributes are more consistent to the image from the surveillance videos. Moreover, further

investigation revealed a witness who knew both Rojas and McCall. During a writ hearing,

the witness testified that he heard McCall brag about committing the aggravated robbery

in detail and how another person was arrested for it.

Based on the new evidence revealed in the writ record, I join the majority of the

Court in finding Applicant actually innocent. I write separately to address the dissenter

who challenge, unsolicited, that finding and question whether our standard in Elizondo, is

rigorous enough to the task of finding someone actually innocent.

Guilt shall not escape nor innocence suffer
In our criminal justice system, those who serve the State are charged with the

“twofold aim of which is that guilt shall not escape nor innocence suffer.” Berger v. United

States, 295 U.S. 78, 88 (1935) (emphasis added). In the mass majority of cases, these

objectives do not conflict with each other because to punish the guilty is to preserve and

defend the innocent.

7
In criminal cases, our founders established that the government must prove their

case “beyond a reasonable doubt” lest they condemn an innocent person. 3 And in cases

where an innocent person is wrongfully condemned, the “Great Writ of Habeas Corpus”

was maintained to right what would otherwise be a travesty of justice. Hence, for those

who are truly innocent and wrongfully convicted, our case law in its current form (though

not perfect) allows relief from injustice under a “Herculean” burden standard of proof. And

this burden is rightfully intended to be set at a level of difficulty high enough so as to

separate out those who frivolously claim innocence knowing it is false. But it leaves the

gate sufficiently ajar to allow the worthy to pass. Thus, the burden is said to be Herculean—

but not Sisyphean. 4 Ex parte Cook, 691 S.W.3d 532, 591 n. 88. (Tex. Crim. App. 2024).

The challenge to Elizondo
In Ex parte Elizondo, this Court held that for an Applicant to be entitled to actual

innocence, he must demonstrate “by clear and convincing evidence that no rational jury

3
Our founders adhered to Blackstone’s Ratio (“[I]t is better that ten guilty persons escape
than that one innocent suffer”) from William Blackstone’s Commentaries to which our founders
adhered to when drafting the Constitution. John Adams, Adam’s Argument for the Defense,
Wemms Tr. 148-178 (Dec. 3-4, 1770); Letter from Benjamin Franklin to Benjamin Vaughan (Mar.
14, 1785); 4 William Blackstone, Commentaries on the Laws of England 352 (Clarendon Press
1769); see Dist. of Columbia v. Heller, 554 U.S. 570, 593-94 (2008) (“Blackstone, whose works,
we have said, constituted the preeminent authority on English law for the founding
generation, . . . .” (J. Scalia writing for the majority)). And though this ratio was not intended to
be mathematically precise, its wisdom is clear: if and when they conflict, the call to preserve the
innocent is more sacred than condemning the guilty.
4
In Greek mythology, Sisyphus was assigned a laborious task of futility as punishment for
offending Olympus. He was sentenced to roll a giant boulder up a steep hill for all eternity.
Whenever he neared the top of the hill, the boulder was cursed to roll all the way back to the
bottom.

8
would convict him” in light of newly discovered evidence affirmatively proving his

innocence when balanced against the entire record. Ex parte Elizondo, 947 S.W.2d 202,

205-07, 210 (Tex. Crim. App. 1996). This standard has stood the test of time without

serious attack by the State as insufficiently rigorous since its inception at the end of the last

century.

Nevertheless, in the face of this Herculean standard, the challenging judge takes the

position that we must raise the bar of scrutiny to “beyond a reasonable doubt” for applicants

to be found actually innocent. 5

All of this is in spite of the fact that our legislative, executive, and judicial branches

have not expressed any concerns over the use of that term, or when it is granted. The Texas

Supreme Court has adopted the term “actual innocence” in determining when and if an

exoneree should be compensated by the state for years of wrongful incarceration. 6 The

Legislature passed the Michael Morton Act in honor of a man who was wrongfully

5
This argument seems inconsistent with itself. Under this argument, “factual or historical
innocence” can only exist when all possibility and suspicion of guilt is removed from the equation.
But that can only happen if the standard is held at “beyond all doubt” and not “beyond a reasonable
doubt.” And to balance the system on the trial end, we would need to require that the jury convicts
under the same “beyond all doubt.” This would be completely unworkable. Even the certainty
underlying DNA profile comparisons (typically expressed as a fraction like 1 in 10 billion) is a
non-zero number. Thus, it cannot satisfy J. Yeary’s strict “threshold of proof” that an applicant
“manifestly did not commit the offense” because there will always be a non-zero possibility that
applicant is the true culprit. Ex parte Cacy, 543 S.W.3d 802, 804 (Tex. Crim. App. 2016) (Yeary,
J., concurring).
6
See In re Allen, 366 S.W.3d 696, 706 (Tex. Crim. App. 2012) (acknowledging that actual
innocence habeas relief granted by the Court of Criminal Appeals entitled Applicant to
compensation from the State under the Timothy Cole Act (TEX. CIV. PRAC. & REM. CODE §
103.001)).

9
convicted and incarcerated for over 20 years. See TEX. CODE CRIM. PROC. art. 39.14. This

act was prompted by the State’s failure to turn over exculpatory evidence that could have

exonerated Morton decades earlier from his wrongful incarceration. During his

incarceration, his young child grew up being falsely told that his father murdered his

mother. Passage of the act was bipartisan and was signed by Governor Rick Perry.

The Texas Tech University campus bears a statue honoring Timothy Cole, a veteran

and student at the university, who was wrongfully convicted and imprisoned for the 1985

sexual assault of another student. And even though another inmate (a serial rapist)

confessed to the crime, Cole’s protestations of innocence went ignored. Cole tragically

died in prison of an asthma attack that the State’s evidence at the time asserted he could

not have had. As a result of DNA testing after his death, Cole was posthumously exonerated

by Governor Perry and the Timothy Cole Act was passed into law. 7

However, we now find ourselves again engaged in the unsolicited argument of

whether one is only innocent, actually innocent, or in some legal purgatory black-hole of

7
TEX. CIV. PRAC. & REM. CODE § 103.001 states:

(a) A person is entitled to compensation if:
(1) the person has served in whole or in part a sentence in prison under the laws
of this state; and
(2) the person:
***
(B) has been granted relief in accordance with a writ of habeas corpus that
is based on a court finding or determination that the person is actually
innocent of the crime for which the person was sentenced; . . . .

(emphasis added).
10
never having been found guilty of anything. The Texas Code of Criminal Procedure,

Article 38.03 and the Texas Penal Code, Chapter 2, Section 2.01 states that everyone is

presumed innocent until proven guilty beyond a reasonable doubt. Even to the die-hard

originalist that language seems patently clear – you are innocent until the State proves you

guilty. As a result, it stands to reason if no jury would find you guilty beyond a reasonable

doubt following a successful writ application, one is at least entitled the presumption of

innocence. Moreover, if you are found not guilty in our system you have the right to have

your record expunged, which is as though the arrest and offense never occurred. 8

For most other reversible errors, the applicant only bears the burden to show some

type of harm that undermines the confidence in the verdict. See, e.g., Thomas v. State, 841

S.W.2d 399, 404 (Tex. Crim. App. 1992); see also TEX. R. APP. PROC. 44.2(b). If he

8
TEX. CODE CRIM. PROC. art. 55A.002 states:

A person to whom this subchapter applies is entitled to have all records and files relating
to the arrest expunged if the person is:
(1) tried for the offense for which the person was arrested; and
(2) acquitted by the trial court, except as provided by Article 55A.151.

TEX. CODE CRIM. PROC., art 55A.003 states:

A person who to whom this subchapter applies is entitled to have all records relating to
the arrest expunged if:
(1) the person is:
(A) tried and convicted for the offense for which the person was arrested; and
(B) subsequently pardoned or otherwise granted relief on the basis of actual
innocence with respect to that offense; and
(2) the applicable pardon or court order clearly indicates on its face that the pardon or
order was granted or issued on the basis of the person’s actual innocence.

(emphasis added)
11
succeeds, he typically can only earn the relief of a new trial. Granted, there might be cases

that prevent one from being found guilty beyond a reasonable doubt where evidence is

excluded, or unavailable on constitutional grounds. Some of those instances have occurred

in the series of Goines cases from Houston. In those rare cases we have simply granted

relief by setting aside those convictions and remanding the case to the trial court for further

proceedings, but we have not granted actual innocence.

Our standard for actual innocence in Elizondo demands far more. The applicant

must offer new evidence that when combined with the existing evidence and viewed in its

totality “unquestionably establishes the applicant’s innocence.” Ex parte Brown, 205

S.W.3d 538, 547 (Tex. Crim. App. 2006). And although each case is evaluated on a case-

by-case basis, the applicant must typically undermine each piece of the State’s evidence

such that no rational jury would convict him. Ex parte Elizondo, 947 S.W.2d at 210. And

even then, this Court still has the power and does deny relief when it becomes obvious that

the relief sought might simply be a new trial.

It speaks volumes for how difficult this standard is to achieve when there is no

known record of a defendant being retried in our system after having been declared actually

innocent—even though the State has the right to do so. Both concurring opinions make it

sound like some sort of grave injustice is committed if by chance we erroneously declare a

person actually innocent. However, they ignore the fact that the State may try that person

again, even in light of an actual-innocence finding by this Court. However, there is just no

record that it has ever happened.

12
A pattern of concurring (and dissenting) opinions

Nevertheless, this disagreement is not confined to semantics nor the technicalities

of a legal standard. This is because Judge Yeary’s “concurring/dissenting” history suggests

something more. It suggests that true “actual innocence” relief, in his view, ought to be

effectively unattainable in even the most shamefully wrongful convictions. In each of these

cases the State recommended granting relief by a new trial and/or actual innocence.

Ex parte Cacy, No. WR-85,420-01, 2016 WL 6525721 (Tex. Crim. App. Nov. 2,
2016 (not designated for publication)

The pattern of keeping actual innocence relief unattainable begins with Sonia Cacy.

In 1990, Sonia Cacy lived with her uncle after her parents tragically died. Her uncle, a

heavy smoker, was universally recognized by all members of the family to be extremely

reckless with fire and cigarettes. Findings, Ex parte Cacy, WR-85,420-01 at 41. Family

members testified that Cacy’s uncle required the use of oxygen tank due to health issues.

He often lit multiple cigarettes after forgetting that he had others already lit. He carelessly

dropped cigarettes on the floor. Additionally, Cacy’s uncle often used a blowtorch to roast

hotdogs and marshmallows. Id.

One morning, the house caught on fire and burned down. Cacy escaped by crawling

out her window while her uncle died in the blaze. She attempted to go back into the home

to rescue him, but was restrained by the firemen at the scene. Cacy was accused of murder

by arson after a forensic lab analyst tested the remains of her uncle’s clothing and found

the presence of gasoline. This was the only physical evidence to support the prosecution’s

13
theory that Cacy poured gasoline on her uncle before setting him on fire. There were no

other pour patterns identified in the house as typically found in other arson cases. The State

provided no motive for Cacy to kill him or burn down the house. 9 (footnote –). Cacy was

found guilty and sentenced to 55 years in prison. However, she received a new sentencing

hearing on appeal where her sentence was increased to 99 years.

In 2012, it was revealed that the Texas State Fire Marshal’s Office (a separate

agency independent of the State, Defense, and Judge) concluded in an independent

investigation of Cacy’s case that there was no evidence of arson. New developments in

arson investigation discovered that the testing methodology for accelerants were yielding

false positives. A review of the gas chromatogram evidence generated from Cacy’s uncle’s

clothing found no gasoline present. Id. at 57. It was also exposed that the forensic lab

analyst had omitted the fact that the laboratory was contaminated with accelerant residue

that could have resulted in a false positive for gasoline at the time of his testing. In an

affidavit, another former analyst and whistleblower for irregularities at that same

laboratory attested that a prior test of the uncle’s clothing by another analyst yielded no

finding of gasoline.

Because there was no evidence of arson or motive (along with many other reasons),

the State Prosecuting Attorney’s Office (SPA) and the habeas judge recommended granting

Cacy relief on actual innocence. This Court agreed and issued a per curiam opinion finding

9
In fact, after Cacy’s parents passed away, Cacy was homeless and her Uncle allowed her
to live with him. Thus, Cacy did not have any motive to set her residence on fire.
14
Cacy actually innocent. And even though the entirety of the evidence in the State’s case

had been invalidated, Judge Yeary, in a published side opinion, refused to label Cacy as

“actually innocent” because she had not proven that she “manifestly did not commit the

offense.” Ex parte Cacy, 543 S.W.3d 802, 804 (Tex. Crim. App. 2016) (Yeary, J.,

concurring).

Ex parte Chaney, 563 S.W.3d 239 (Tex. Crim. App. 2018)

This view that actual innocence relief should be unattainable continued in Chaney.

There, a couple known to sell cocaine were found brutally stabbed numerous times and

their throats slashed. Bloody shoeprints from two sets of shoes, blood smears, and blood

spatter were all over the apartment. Additionally, a human bitemark was found on the male

victim’s arm. It was universally agreed that the killer would have been covered in blood

from the knees to his feet. After what appeared to be a drug ledger connected Chaney to

the couple and showed he owed the victims money, 10 investigators visited Chaney at his

work and saw him wearing tennis shoes consistent with one of the bloody shoeprints at the

crime scene. A partial thumbprint belonging to Chaney was also found near the kitchen

entrance.

The shoes were tested for blood. Although no blood was visible, the testing resulted

in a presumptive positive for the presence of blood on the shoes. Two odontologists also

compared the bitemark on the victim’s arm to Chaney’s teeth. Both testified that they were

10
Testimony from the State’s star witness later showed that Chaney had cleared his debt
to the victims and thus had no motive to kill them. Chaney, 536 S.W.3d at 247.
15
a “match.” Because the bitemark comparison evidence was the State’s only evidence

establishing the identity of the perpetrator, the State spent nearly the entire second closing

argument explaining how “only one in a million could have possibly made that bite mark.”

Chaney, 563 S.W.3d at 254. Although testimony from multiple defense witnesses showed

he was somewhere else at the time of the murders, the jury found Chaney guilty based on

the strength of the bitemark comparison evidence and sentenced him to life in prison.

Three decades later, new developments in the field of odontology revealed that

using bitemark comparison evidence was scientifically unreliable in establishing the

identity of suspects. And although the thumbprint placed Chaney at the apartment at some

point, it was consistent with his past visits to the home. Because these new developments

completely undermined the State’s identification of Chaney as the perpetrator, the case was

re-investigated and witnesses reinterviewed.

During this second investigation, it was revealed that forensic testing of Chaney’s

tennis shoes found absolutely no blood on them. Furthermore, during the original

investigation, detectives had searched Chaney’s residence and clothing only to conclude

that there was no blood evidence to be found there. None of this exculpatory evidence was

revealed to Chaney’s attorneys leading up to and during Chaney’s trial.

DNA testing was also conducted on the female victim’s fingernails to reveal the

presence of three male contributors. But none of those profiles were originating from

Chaney. And consistent with these exculpatory results, Chaney was also excluded as the

16
producer of some hairs found in one of the female victim’s hands. In fact, “[t]he State

characterized its investigation as uncovering a ‘vast amount of evidence’ to support

Chaney’s actual innocence claim.” Id. at 275.

This Court found that Chaney had managed to invalidate every piece of the State’s

evidence. Thus, the Court concluded that “no reasonable juror would have convicted

[Chaney]” under the totality of all the new and old evidence. However, Judge Yeary

disagreed with the habeas court, the State, and this Court’s majority. Though he agreed

Chaney was entitled to relief under Article 11.073, “I do not agree that [Chaney] has even

established entitlement to relief under Elizondo, much less that he is ‘actually innocent’ in

any definitive sense.” Id. at 288 (Yeary, J., concurring). He further claimed that it lies

outside our judicial authority to exonerate convicted persons. 11 Claiming that Chaney had

yet to prove himself “factually, historically, and morally” innocent, he accused the majority

of “lack[ing] a humility appropriate to our true station when we declare to the public that

an applicant is actually innocent.” Id. (Yeary, J., concurring).

Ex parte Cook, 691 S.W.3d 532 (Tex. Crim. App. 2024)

In Ex parte Cook Judge Yeary would have denied all relief. The same State agency,

who was found to have engaged in multiple Brady and ethical violations by this Court and

the U.S. Supreme Court, at least conceded that Cook was entitled to have his conviction

set aside. Only one judge on this Court disagreed with that recommendation.

11
Id. at 287 (Yeary, J., concurring); but see TEX. CIV. PRAC. & REM. CODE § 103.001;
TEX. CODE CRIM. PROC., art 55A.003.
17
Cook was wrongfully convicted of the capital murder of a young woman in Tyler,

Texas and sentenced to death. After three trials, several appellate reversals including one

by the United States Supreme Court, and a fourth prosecution, this Court found Cook

actually innocent after discovering new exculpatory evidence. This evidence included

revelations that agents of the State had violated multiple Brady rules and had lied about a

deal made with a jailhouse snitch that was their star witness in the first trial. Additionally,

DNA tests were linked to the alternate suspect after Cook had been on death row for over

20 years and the State had used that suspect as a key witness multiple times under oath.

The DNA discovered in the victim’s panties the night she was murdered was actually

linked to that witness/suspect. Almost 40 years later, the alternate suspect admitted, with

a full grant of immunity, that he had committed perjury multiple times by claiming he had

not had sex with the victim for weeks prior to her murder, and had in fact been with the

victim the night of the murder. He also admitted after decades of denial that had knowing

possession of a book about sexual homicides with graphic pictures of other murder scenes

that bore striking similarities with the victim. In addition to these new revelations, the

physical evidence showed that it was practically impossible for Cook to have committed

the crime. These revelations were just the tip of the iceberg that are found in this Court’s

106-page decision granting actual innocence relief.

And in spite of successfully invalidating almost every piece of evidence against him,

including the new exculpatory DNA evidence, the revelation of the prosecutorial

18
conspiracy to fabricate evidence, other previously undisclosed evidence, and the revelation

of perjury by multiple material witnesses, Judge Yeary dissented saying:

Wow! I hope they are right since their judgment wins the day in this Court.
***

In all humility, I may be wrong about my assessment of the evidence in this
case, but I remain convinced today that, even being put on notice of the false
evidence pointed to by [Cook], on balance, a rational jury would still have
found [Cook] guilty.

Ex parte Cook, 691 S.W.3d 532, 592 (Tex. Crim. App. 2024) (Yeary, J., dissenting). And

in declaring so, Judge Yeary would not have even offered Cook a new trial.

The pattern in other cases and the instant case

More than this, we have not been able to find any case where Judge Yeary agreed

that a successful claimant under Elizondo was “actually innocent” in the rough decade of

sitting on this Court—even claimants he agrees deserve relief under Elizondo.12

Furthermore, despite his implied fears to the contrary, it has never happened (that we’re

aware of) where a convicted person declared actually innocent was later found to be

12
Ex parte Sanchez, 690 S.W.3d 679 (Tex. Crim. App. 2024) (Yeary, J. dissenting); Ex
parte Cook, 691 S.W.3d 532 (Tex. Crim. App. 2024) (Yeary, J. dissenting); Ex parte Martin, 689
S.W.3d 325 (Tex Crim. App. 2024) (Yeary, J. dissenting); Ex parte Santillan, 666 S.W.3d 580
(Tex Crim. App. 2023) (Yeary, J., concurring); Ex parte Johnson, No. WR-94,542-01, 2023 WL
2995187 (Tex Crim. App. 2023) (Yeary, J., concurring); Ex parte Hawkins, 664 S.W.3d 181 (Tex
Crim. App. 2022) (Yeary, J., dissenting); Ex parte Mallet, 602 S.W.3d 922 (Tex Crim. App. 2020)
(Yeary, J., concurring); Ex parte Chaney, 563 S.W.3d 239 (Tex Crim. App. 2018) (Yeary, J.,
concurring); Ex parte Cacy, 543 S.W.3d 802 (Tex Crim. App. 2016) (Yeary, J., concurring).

19
objectively guilty of the crime from which he was exonerated. If anything, we have only

confirmed the opposite. 13 See, e.g., Timothy Cole.

But perhaps the final damning aspect of Judge Yeary’s proposed heightened

standards for actual innocence is in its application to the current Applicant’s case. By all

accounts, this is another clearcut case of actual innocence: First, even the State concedes a

mistake was made and he is actually innocent. Second, the State has evidence showing

another individual is actually guilty of committing the crime. Applicant has convincingly

proven, under every meaning of that phrase, to be actually innocent of the aggravated

robbery. And even though he was innocent, he spent five years of his life incarcerated by

the State. If Judge Yeary’s proposed heightened standards cannot be satisfied in such a

clear case such as this, then the standard he proposes is far more problematic than corrective

in assuring that the guilty do not escape punishment while preserving the innocent. And

because it would allow the oppression of innocents in even non-controversial cases (“Better

to condemn a hundred innocent people than let one truly guilty person go free”), 14 adopting

it would be heretical to the inherited principles of our constitutional founders. 15 Whatever

flaws our actual innocence case law may face, Judge Yeary’s answer is no solution at all

because it renders the burden impossible and futile in all but name.

13
Exploring the actual innocence standard under the Texas Constitution (as opposed to the
requirements of the Federal Due Process Clause) may be worthwhile as Presiding Judge Schenck
suggests. However, it is admittedly not an issue in controversy in this case.
14
Short, Philip, Pol Pot: Anatomy of a Nightmare 299 (Henry Holt & Co. 2006).
15
See supra note 3 (discussing Blackstone’s Ratio: “[I]t is better that ten guilty persons
escape than that one innocent suffer.”).
20
Accomplishing the “twofold aim” here

The great irony here in this case is that this Court does not have to choose between

“guilt shall not escape nor innocence suffer.” Berger v. United States, 295 U.S. 78, 88

(1935). This Court can do both—remedy the unjust conviction of an innocent man and seek

punishment against the one identified as the true culprit—which, today, it correctly does.

And all parties, including both the prosecutors for the State and the habeas court, agree that

the evidence here clearly and convincingly shows that it is righteous to grant relief to

Applicant because he is actually innocent. Only the opinions disagreeing with the actual

innocence finding from this Court—and not those who are actually charged with

prosecuting offenders of the law—advocates differently.

Conclusion

Thomas Jefferson once wrote: “It is more honorable to repair a wrong than to persist

in it.” It comes with no surprise that our criminal justice system will make mistakes and

reach unjust outcomes despite our honest best efforts. After all, we are only human. But

our saving grace is in our aspiration to become more perfect. And achieving that relies not

only in our successes when justice is done, but also in maintaining a robust ability to

recognize and correct our failures to do justice. Making the Elizondo standard substantially

more rigorous would undermine that.

Is the Elizondo standard perfect? Probably not. Consistent with our Texas

Constitution, could it use some refinement in the types of cases previously mentioned in

this opinion? Possibly—but those issues are not presented to us in this case. Is raising the

21
bar to make it virtually impossible to obtain actual innocence the solution? Nothing in our

jurisprudence suggests that such an approach is necessary, it is already a high burden. Our

Elizondo standard may not be perfect, but here it is perfectly applied to find Applicant

actually innocent, because he is. With these concurring thoughts, I join the majority.

Filed: March 26, 2025

Publish

22