Jacob Douglas Ellison v. State

In the
Court of Appeals
Second Appellate District of Texas
at Fort Worth
___________________________

No. 02-18-00056-CR
___________________________

JACOB DOUGLAS ELLISON, Appellant

V.

THE STATE OF TEXAS

On Appeal from the 355th District Court
Hood County, Texas
Trial Court No. CR13588

Before Sudderth, C.J.; Birdwell and Womack, JJ.
Memorandum Opinion by Justice Womack
 MEMORANDUM OPINION

Appellant Jacob Douglas Ellison appeals his conviction for aggravated sexual

assault. In his first issue, he contends that the trial court wrongly refused a question

during voir dire. Controlling precedent shows that the refusal was proper. In his

second issue, Ellison argues that he received ineffective assistance when counsel advised

him to plead guilty. A lack of record support for his allegations requires us to conclude

otherwise. In a third issue, Ellison complains of the State’s closing argument. This

complaint is not preserved for our review. We therefore affirm.

I. BACKGROUND

In 2016, Diane Isim1 lived on the outskirts of town in a trailer with her four

children and Ellison, who was the father of two of her children. Diane’s daughter Nora

had a different father. One morning in June 2016, Nora told her mother that Ellison

had molested her. An investigation began, and on October 12, 2016, Ellison was

indicted on multiple counts. Count three charged him with aggravated sexual assault

of a child younger than fourteen. See Tex. Penal Code Ann. § 22.021(a)(1)(B)(i),

(a)(2)(B). Ellison pleaded guilty to count three as charged, and the State moved to

dismiss the remaining counts. Ellison elected to have punishment tried to a jury.

1
We use pseudonyms to protect the complainant’s identity. See Daggett v. State,
187 S.W.3d 444, 446 n.3 (Tex. Crim. App. 2005).

2
 At the trial on punishment, Diane testified about the morning when Nora told

her, “Daddy Jacob does bad things to me.” According to Diane, Nora described how

Ellison would remove her underwear and assault her. Diane took Nora to Cook

Children’s Medical Center. Nurse Christi Thornhill testified that during an interview at

the hospital, Nora explained that Ellison had come into her room and put his genitals

or his fingers into her “[m]ore than one time, maybe six” times. By Thornhill’s account,

Nora reported that the abuse began when she was five; she was six at the time of the

interview. Nora also testified concerning the abuse, confirming many of these details.

Investigator Robert Young of the district attorney’s office testified that he took Diane’s

report and, later, interviewed Ellison. According to Young, Ellison admitted being

aroused around Nora, explained that he felt disgusted with himself, and said he believed

Nora’s story.

During the defense’s case, the jury heard from Ellison’s employer, who testified

that he had been a model employee, was forthcoming about the charges against him,

and was sorry for what he had done. Ellison’s brother Andrew testified that he did not

observe anything out of the ordinary when he stayed with Ellison and Diane; there were

no signs that Nora was afraid of Ellison. Ellison also presented evidence that because

he had pleaded guilty, he would be subject to stringent sex offender registration

requirements for the rest of his life.

3
 After the conclusion of the evidence, the trial court assessed punishment at the

statutory maximum—ninety-nine years’ confinement and a $10,000 fine. See id. § 12.32.

Ellison appeals.

II. LIMITATION ON VOIR DIRE

In his first issue, Ellison challenges the trial court’s refusal of a question during

voir dire. Ellison requested permission to ask the venire members whether they could

be impartial and consider the evidence before making a decision if the victim was under

the age of ten. The trial court denied Ellison’s request but gave him latitude to ask a

slightly different question: whether they could remain “fair and impartial” when “this

charge is an aggravated sexual assault of a child who’s under the age of 14.”

On appeal, Ellison complains that the trial court improperly restricted voir dire.

He asserts that his proffered question would have better shed light on any bias against

those charged with sex crimes against young children, which he maintains is a proper

subject of inquiry.

Ellison acknowledges that the court of criminal appeals rejected a substantially

similar question in Barajas v. State, 93 S.W.3d 36 (Tex. Crim. App. 2002). There, defense

counsel desired to ask whether the venire members could be impartial in an indecency

case involving a victim who was eight to ten years old or, in the alternative, a victim

who was nine years old. Id. at 38. The appellate court laid out the standards for judging

the limitation of voir dire questioning as follows:

4
 The trial court has broad discretion over the process of selecting a jury.
The main reason for this is that voir dire could go on forever without
reasonable limits. We leave to the trial court’s discretion the propriety of
a particular question[,] and the trial court’s discretion will not be disturbed
absent an abuse of discretion. A trial court’s discretion is abused only
when a proper question about a proper area of inquiry is prohibited.
A question is proper if it seeks to discover a juror’s views on an issue
applicable to the case. An otherwise proper question is impermissible,
however, if it attempts to commit the juror to a particular verdict based
on particular facts. In addition, a voir dire question that is so vague or
broad in nature as to constitute a global fishing expedition is not proper
and may be prevented by the trial judge.
Id. at 38–39 (citations omitted). The Barajas court held that the defendant’s proposed

question concerning the victim’s age was, at once, both an improper commitment

question and so vague that it would lead to an inappropriate fishing expedition. Id. at

39–40.

Ellison says that Barajas is distinguishable. That is true, perhaps, but not in any

way that benefits Ellison.

First, Ellison points out that the defendant in Barajas advanced a more specific

age range for the hypothetical victim: in Barajas, between ages eight and ten; here, under

age ten. But if the question in Barajas was too broad and vague, we fail to see how

offering an only somewhat looser description of the victim’s maturity works to Ellison’s

advantage. See Green v. State, 191 S.W.3d 888, 892 (Tex. App.—Houston [14th Dist.]

5
2006, pet. ref’d) (holding that the case was controlled by Barajas when defendant

proposed a slightly vaguer description of the victim’s age range).2

Second, Ellison observes that in Barajas, the defendant went to the jury for a

determination of guilt as well as punishment, whereas Ellison had a jury trial only for

punishment. 93 S.W.3d at 38. Ellison does not explain the significance of this

distinction, and Barajas shows that there is none. According to the Barajas court, the

complainant’s age might be relevant to three aspects of the trial: the guilt phase, the

determination of the victim’s credibility as a witness, and the assessment of punishment.

Id. at 39. But the court held that as it was phrased, the proposed question was fatally

flawed with regard to all three aspects. Id. at 39–40. Here, because no trial was had on

guilt, the victim’s age could be relevant to only two of the three aspects mentioned in

Barajas—credibility and punishment—but the same flaws persist.

According to the Barajas court, one of these flaws was the framework in which

the inquiry was housed. “The question ‘can you be fair and impartial under a given set

of facts?’ can be repeated to include every fact in a given case.” Id. at 41. “This ‘fair

and impartial’ question is a license to go fishing, without providing any concrete

information for the intelligent use of peremptory or for-cause challenges.” Id. This

2
As we explain below, the trial court’s proposed age range—under the age of
fourteen—better complied with the rules governing commitment questions.

6
broad question framework, the court said, was liable to lead to “fishing expeditions

during voir dire that may extend jury selection ad infinitum.” Id. at 42.

Which brings us to a third distinction, the format of the proposed question. In

Barajas, the defendant proposed to ask if the venire members “could be fair and

impartial” given the victim’s age. Here, Ellison’s proposed age inquiry differed in its

phrasing: would the victim’s age “affect their ability to . . . remain impartial and consider

the evidence before making a decision?” Despite the minor difference between these

premises, the two are equally broad and indefinite. Thus, Ellison’s proposed inquiry

was contained in a fishing boat of a slightly different build, but a fishing boat

nonetheless according to the logic of the Barajas majority.3 Ellison’s proposed inquiry

could have been, but was not required to be, disallowed on this basis alone, and

exclusion of Ellison’s proposed question would not exceed the court’s “broad

discretion” over how voir dire should be conducted. See id. at 39 (providing that a vague

question “may” be excluded). And Ellison does not complain on appeal that the trial

court improperly allowed an alternative question in a similar format. See id. at 39

(explaining that a court abuses its discretion only when “a proper question about a

3
Ellison also argues that Barajas was wrongly decided, citing the well-considered
dissent in that case. Barajas v. State, 93 S.W.3d 36, 45 (Tex. Crim. App. 2002) (Meyers,
J., dissenting). But the soundness of Barajas is not a question for this court. See Sell v.
State, 488 S.W.3d 397, 399 (Tex. App.—Fort Worth 2016, pet. ref’d); Lydia v. State,
117 S.W.3d 902, 913 (Tex. App.—Fort Worth 2003, pet. ref’d) (Dauphinot, J.,
concurring) (acknowledging our duty to follow Barajas, but respectfully encouraging the
court of criminal appeals to reconsider that opinion).

7
proper area of inquiry is prohibited” (emphasis added)). Usually we may not reverse a

trial court’s ruling on a basis not raised by the appellant. Martinez v. State, 91 S.W.3d

331, 336 (Tex. Crim. App. 2002).

As previously mentioned, the Barajas court found another flaw in the proposed

inquiry: it was an improper commitment question. Commitment questions are those

that commit a prospective juror to resolve, or to refrain from resolving, an issue a

certain way after learning a particular fact. Standefer v. State, 59 S.W.3d 177, 179 (Tex.

Crim. App. 2001). For a commitment question to be proper, one of the possible

answers to that question must give rise to a valid challenge for cause. Id. at 182. “If a

venire member stated that she would resolve the appellant’s guilt on the basis of the

victim’s age, that venire member would be challengeable for cause.” Barajas, 93 S.W.3d

at 39. But a commitment question may also be improper if it introduces “facts beyond

those necessary to establish a challenge for cause.” Lydia v. State, 109 S.W.3d 495, 498

(Tex. Crim. App. 2003). For example, the State may properly ask a general commitment

question as to whether the venire members could follow a law that holds a person guilty

of possession even though the possession involves only a residue amount of drugs,

because such a question would include only that quantum of factual content necessary

to frame a proper commitment. Standefer, 59 S.W.3d at 182. But it would cross the line

if the State supplied additional facts within its question, such as that the defendant was

arrested and found with a crack pipe in his pocket. Id.

8
 Ellison contends that his proposed question was a proper commitment question

about a proper subject of inquiry and that the trial court therefore abused its discretion

by disallowing it. However, Ellison’s question would have supplied the jury with an

additional, case-specific fact—a more specific age-range for the victim—beyond what

was necessary to frame any commitment. The trial court therefore did not abuse its

discretion in rejecting it. See id.

This suggests a fourth point of distinction. In Barajas, the trial court excluded

any questions relating to the victim’s age whatsoever. 93 S.W.3d at 42. Here, the trial

court denied Ellison’s proposed question, but the court allowed a compromise solution,

permitting Ellison to ask about a victim under age fourteen. While the trial court chose

a different path than that discussed in Barajas, the court nonetheless adhered to that

case’s directive not to introduce any additional facts beyond what was necessary to

frame the commitment. The factual content of the question was restricted to the basic

elements of the offense as set out in the statute.4 See Standefer, 59 S.W.3d at 181

(approving of a commitment question when it exposed the venire members to no

additional facts beyond the necessary “factual elements . . . listed in the statute”); Lee v.

State, 176 S.W.3d 452, 461 (Tex. App.—Houston [1st Dist.] 2004) (approving of a

commitment question because it described the victim’s age only as “under the age of

A person commits the offense of aggravated sexual assault if the person causes
4

the penetration of the sexual organ of a child younger than fourteen years of age by any
means. Tex. Penal Code Ann. § 22.021(a)(1)(B)(i), (a)(2)(B).

9
seventeen,” as set out in the statutory elements, without disclosing the victim’s specific

age), aff’d, 206 S.W.3d 620 (Tex. Crim. App. 2006); see also Rodriguez-Flores v. State,

351 S.W.3d 612, 622–23 (Tex. App.—Austin 2011, pet. ref’d) (similar).

By limiting Ellison’s question to a factual basis that tracked the statute, the trial

court (1) allowed Ellison to probe for prejudice regarding age but (2) shielded the venire

members from premature exposure to the facts of the case and, in doing so, (3) limited

the risk of an improper commitment question. See Jacobs v. State, 560 S.W.3d 205, 207,

213 (Tex. Crim. App. 2018) (approving the trial court’s middle-path solution—rejecting

the defendant’s proposed question about bias against a defendant previously convicted

of a “sexual offense” but allowing a more general question concerning prior “assaultive

offenses”—because it allowed defendant the opportunity to ferret out bias but avoided

the risk that a more specific question would expose the venire to the facts of the case).

Despite this distinction, then, the trial court remained within its rights under Barajas.

None of these four distinctions serve Ellison’s purpose, and after considering

them all, we remain convinced that Barajas is squarely on point. We therefore follow

that case’s command and overrule Ellison’s first issue.

III. VOLUNTARINESS OF PLEA

In his second issue, Ellison contends that he received ineffective assistance of

counsel. He asserts that the trial court erred by denying his motion for new trial on that

account.

10
 A claim of ineffective assistance of counsel may be raised in a motion for new

trial. Smith v. State, 286 S.W.3d 333, 340 (Tex. Crim. App. 2009). We review a trial

court’s denial of a motion for new trial on the basis of ineffective assistance under an

abuse of discretion standard. Burch v. State, 541 S.W.3d 816, 820 (Tex. Crim. App. 2017).

This is a deferential standard of review that requires appellate courts to view the

evidence in the light most favorable to the trial court’s ruling. Id. We do not substitute

our own judgment for that of the trial court, and we must uphold the trial court’s ruling

if it is within the zone of reasonable disagreement. Id.

A defendant is entitled to effective assistance of counsel in the guilty-plea

context. Ex parte Evans, 537 S.W.3d 109, 111 (Tex. Crim. App. 2017). To establish

ineffective assistance, an appellant must prove by a preponderance of the evidence that

his counsel’s representation was deficient and that the deficiency prejudiced the

defense. Strickland v. Washington, 466 U.S. 668, 687, 104 S. Ct. 2052, 2064 (1984); Nava

v. State, 415 S.W.3d 289, 307 (Tex. Crim. App. 2013); Hernandez v. State, 988 S.W.2d 770,

770 (Tex. Crim. App. 1999). The record must affirmatively demonstrate that the claim

has merit. Thompson v. State, 9 S.W.3d 808, 813 (Tex. Crim. App. 1999).

In evaluating counsel’s effectiveness under the deficient-performance prong, we

review the totality of the representation and the particular circumstances of the case to

determine whether counsel provided reasonable assistance under all the circumstances

and prevailing professional norms at the time of the alleged error. See Strickland,

466 U.S. at 688–89, 104 S. Ct. at 2065; Nava, 415 S.W.3d at 307; Thompson, 9 S.W.3d at

11
813–14. Our review of counsel’s representation is highly deferential, and we indulge a

strong presumption that counsel’s conduct was not deficient. Nava, 415 S.W.3d at 307–

08.

An appellate court may not infer ineffective assistance simply from an unclear

record or a record that does not show why counsel failed to do something. Menefield v.

State, 363 S.W.3d 591, 593 (Tex. Crim. App. 2012); Mata v. State, 226 S.W.3d 425, 432

(Tex. Crim. App. 2007). Trial counsel “should ordinarily be afforded an opportunity to

explain his actions before being denounced as ineffective.” Menefield, 363 S.W.3d at

593. If trial counsel did not have that opportunity, we should not conclude that counsel

performed deficiently unless the challenged conduct was “so outrageous that no

competent attorney would have engaged in it.” Nava, 415 S.W.3d at 308. Direct appeal

is usually inadequate for raising an ineffective-assistance-of-counsel claim because the

record generally does not show counsel’s reasons for any alleged deficient performance.

See Menefield, 363 S.W.3d at 592–93; Thompson, 9 S.W.3d at 813–14.

Strickland’s prejudice prong requires a showing that counsel’s errors were so

serious that they deprived the defendant of a fair trial—that is, a trial with a reliable

result. Strickland, 466 U.S. at 687, 104 S. Ct. at 2064. In other words, an appellant must

show a reasonable probability that the proceeding would have turned out differently

without the deficient performance. Id. at 694, 104 S. Ct. at 2068; Nava, 415 S.W.3d at

308. In the guilty-plea context, the focus of the prejudice inquiry is on whether a

defendant has shown that but for counsel’s errors, there is a reasonable probability that

12
he would not have pleaded guilty and would have insisted on going to trial. Ex parte

Torres, 483 S.W.3d 35, 43 (Tex. Crim. App. 2016) (citing Hill v. Lockhart, 474 U.S. 52,

59, 106 S. Ct. 366, 370 (1985)).

As we read his brief, Ellison offers three theories as to how his trial counsel was

deficient under the first prong of Strickland. First, Ellison says he was “unaware the

juries in Hood County routinely give out 99-year sentences on guilty pleas for

aggravated sexual assault, and . . . he should have been told this information by his trial

counsel.”

However, the record contains no proof that sentences of this kind were routine.

To defeat the presumption of reasonably effective assistance, any allegation of

ineffectiveness must be firmly founded in the record. Prine v. State, 537 S.W.3d 113, 117

(Tex. Crim. App. 2017). Ellison introduced no evidence of sentencing patterns in the

locality, let alone evidence by offense type and plea status. As such, the presumption

of effective assistance remains intact. See id.

Second, Ellison asserts that trial counsel was ineffective in that he made an

improper and coercive promise under Brady v. United States, 397 U.S. 742, 753, 90 S. Ct.

1463, 1471 (1970). According to Ellison’s testimony at the hearing on his motion,

counsel advised him that pleading guilty was “probably the only way that [he was] ever

seeing the light of day again.” Ellison contends this advice runs afoul of Brady’s

pronouncement that to be admissible, “a confession must be free and voluntary: that

is, must not be extracted by any sort of threats or violence, nor obtained by any direct

13
or implied promises, however slight, nor by the exertion of any improper influence.”

Id., 90 S. Ct. at 1471 (quotations omitted). By Ellison’s reasoning, trial counsel’s dire

prediction was an improper and coercive promise, and because he pleaded guilty based

on this promise, his plea was involuntary under Brady.

Ellison’s reliance on Brady is misplaced. He refers to a portion of Brady that

concerns the constitutional dangers that arise when “agents of the State . . . produce a plea

by actual or threatened physical harm or by mental coercion overbearing the will of the

defendant.” Id. at 750, 90 S. Ct. at 1470 (emphasis added). More specifically, it discusses

ill-gotten “promises” by the “prosecutor[].” Id. at 755, 90 S. Ct. at 1472. Ellison cannot

contend that his trial counsel was a state actor.5

Rather than casting defense counsel as a villain, Brady speaks of counsel as a

guardian of Fifth Amendment freedoms, saying that the harmful effects of coercive

5
Ellison has not directed our attention to any other forms of coercive state action
that might render his plea involuntary. To the contrary, the record shows that state
actors took steps to ensure that Ellison’s plea was knowing and voluntary. Before
accepting his guilty plea, the trial court explained the range of punishment; verified that
Ellison was not pleading guilty due to fear, threats, undue persuasion, or promises;
admonished him that he would be required to register as a sex offender and ensured
that he understood the ramifications thereof; and reminded Ellison of his right to a jury
trial on guilt. The trial court confirmed that Ellison was competent; that he was satisfied
with counsel’s representation; and that he was freely, voluntarily, knowingly, and
intelligently pleading guilty only because he was “in fact, guilty of the offense.” See Tex.
Code Crim. Proc. Ann. art. 26.13. Only after Ellison confirmed that he understood and
agreed to all of the above did the trial court accept his guilty plea. The trial court
substantially fulfilled its duty to admonish Ellison, which creates a prima facie case that
his plea was knowing and voluntary. Martinez v. State, 981 S.W.2d 195, 197 (Tex. Crim.
App. 1998). It was then Ellison’s burden to demonstrate otherwise. Id.

14
state action may be “dissipated by the presence and advice of counsel.” Id. at 754, 90

S. Ct. at 1472. And Brady does not address ineffective assistance; it assumes a defendant

who is already “competently counseled” and “correctly advised with respect to the then

existing law as to the possible penalties.” Id. at 757, 90 S. Ct. at 1473. Brady stands for

the proposition that when he is correctly counseled, the defendant cannot later

complain that his plea was involuntary when subsequent developments in the law

change the applicable range of punishment. Id., 90 S. Ct. at 1473. Ellison has not

identified any change in the law of punishment, and Brady holds that even if he had, he

could not complain. Thus, Brady is wholly inapposite.

Nonetheless, Ellison’s argument might more generally be understood as a

complaint that trial counsel’s prediction of Ellison’s chances was not consistent with

the state of the evidence. Thus, in what we understand as his third theory of

ineffectiveness, Ellison reasons that counsel was inadequate in advising him to plead

guilty because there was evidence suggesting that he was innocent. For example, at the

hearing on the motion for new trial, Ellison’s employer testified that he heard Diane

say that she believed Ellison was innocent, and multiple witnesses attested that Diane

often brought the children to see Ellison at work while the charges were pending. A

friend of Ellison’s named Jimmy Hill related that he had frequently seen Ellison playing

with the children at the trailer park, but he had never witnessed Ellison do anything to

hurt the children. Similarly, Ellison’s mother Sandra swore that he was a good father

who loved his children dearly and that he would never do anything to hurt them. As

15
for Ellison himself, he professed his innocence and testified that Diane had attempted

to drop the charges, but the State would not comply. In view of this evidence, Ellison

argues that trial counsel was ineffective for urging him to plead guilty.

However, trial counsel was not asked to testify at the hearing on the motion for

new trial. Again, trial counsel should ordinarily be afforded an opportunity to explain

his actions before being denounced as ineffective, but counsel was never given the

chance to explain his strategic thinking. Menefield, 363 S.W.3d at 593. In the absence

of such an explanation, an appellate court commonly will assume a strategic motivation

if any can possibly be imagined. Ex parte Miller, 330 S.W.3d 610, 616 n.9 (Tex. Crim.

App. 2009). We can imagine at least one valid strategic motivation for this advice: in

light of the seriousness of the offense and the compelling evidence that Ellison was

guilty of it, counsel could have reasonably determined that Ellison’s best option was to

seek leniency by throwing himself upon the mercy of the jury. See Ex parte Scott,

541 S.W.3d 104, 122 (Tex. Crim. App. 2017) (recognizing that “accepting responsibility

for his behavior and pleading for mercy from the jury” was a valid defensive strategy).

The fact that this strategy ultimately proved unsuccessful does not render counsel

ineffective. Garza v. State, 261 S.W.3d 361, 367 (Tex. App.—Austin 2008, pet. ref’d).

“That appellant was ultimately assessed the maximum punishment means only that the

risk did not pay off,” not that counsel’s representation was constitutionally

“unacceptable.” Delrio v. State, 840 S.W.2d 443, 447 (Tex. Crim. App. 1992). Thus,

based on this undeveloped record, Ellison has not shown that counsel’s prediction and

16
advice were so outrageous that no competent attorney would have issued them, see

Nava, 415 S.W.3d at 308, especially given the inherent difficulty in predicting how a jury

will assess punishment. See Ex parte Rich, 194 S.W.3d 508, 514 n.12 (Tex. Crim. App.

2006) (“[I]t is impossible to predict with any certainty what sentence he would have

received . . . .”); Hanson v. State, No. 11-09-00278-CR, 2011 WL 704639, at *4 (Tex.

App.—Eastland Jan. 13, 2011, pet. ref’d) (mem. op., not designated for publication)

(“Predicting a sentencing decision is an inexact science at best.”); cf. Ex parte Chandler,

182 S.W.3d 350, 359 (Tex. Crim. App. 2005) (noting, in the Strickland context, that “a

bar card does not come with a crystal ball”).

Viewing the evidence in the light most favorable to the trial court’s ruling, we

conclude that Ellison has failed to satisfy the first prong of Strickland. See Burch,

541 S.W.3d at 820. It is therefore unnecessary to examine the second prong. See Lopez

v. State, 343 S.W.3d 137, 144 (Tex. Crim. App. 2011). We overrule Ellison’s second

issue.

IV. JURY ARGUMENT

In his third issue, Ellison complains that the State made improper closing

arguments. He maintains that the State inserted inflammatory personal appeals and

matters outside the record when the State argued as follows: “With your verdict, you’re

protecting your own children, your own families, you’re protecting my kids and all of

the kids who none of . . . us have even met that live here in Hood County.” Similarly,

moments later, the State urged the jury to issue a stiff punishment in order “to tell me

17
and each other that this is never going to happen to my kids or your kids by this

defendant.”

However, Ellison did not object at any point during the State’s closing argument.

When a defendant fails to timely object, he will forfeit a complaint concerning improper

jury argument. Hernandez v. State, 538 S.W.3d 619, 622 (Tex. Crim. App. 2018); Cockrell

v. State, 933 S.W.2d 73, 89 (Tex. Crim. App. 1996); see Tex. R. App. P. 33.1(a)(1). Ellison

has therefore not preserved this claim for our review.

We overrule Ellison’s third and final issue.

V. CONCLUSION

Having overruled Ellison’s three issues, we affirm the judgment of the trial court.

/s/ Dana Womack

Dana Womack
Justice

Do Not Publish
Tex. R. App. P. 47.2(b)

Delivered: August 22, 2019

18