in Re: The Commitment of Gregory A. Jones

In the
Court of Appeals
Second Appellate District of Texas
at Fort Worth
___________________________
No. 02-18-00019-CV
___________________________

IN RE: THE COMMITMENT OF GREGORY A. JONES

On Appeal from the 89th District Court
Wichita County, Texas
Trial Court No. 185,786-C

Before Sudderth, C.J.; Kerr and Pittman, JJ.
Opinion by Chief Justice Sudderth
 OPINION

In six issues, Appellant Gregory A. Jones appeals the trial court’s order that he

be committed as a sexually violent predator. See Tex. Health & Safety Code Ann. §

841.003. We hold that the evidence presented at trial was legally sufficient, but we

reverse the trial court’s order of commitment and remand these proceedings to the

trial court for a new trial because the trial court erred in rejecting Jones’s request for a

jury instruction.

Background

I. Testimony at trial

Three witnesses testified at trial: psychologist Jason Dunham, psychiatrist Sheri

Gaines, and Jones.

Dunham and Gaines testified to their evaluations of Jones and their opinions

that Jones suffered from a behavioral abnormality that made him likely to commit a

sexually violent offense. Each expert met with Jones individually and interviewed him

for about two and a half hours. Additionally, Dunham administered two actuarial

tests, the Static-99R and the Psychopathy Checklist Revised (PCLR).

Both experts explained how they reviewed records related to Jones’s criminal

history, including indictments, police reports, witness statements, victim statements,

and prison materials including parole case summaries, medical records, and

disciplinary records; and that they reviewed each other’s deposition and Jones’s

deposition. The remainder of Dunham’s and Gaines’s testimony can be summarized
2
into four categories: (A) Jones’s criminal history, (B) other risk factors, (C) protective

factors, and (D) diagnoses and conclusions.

A. Jones’s criminal history

Both Dunham and Gaines explained that the details of a sex offender’s past

offenses—both sexual and nonsexual in nature—were important to consider in their

evaluations, as well as the person’s lifestyle before and after he committed the crimes.

Details of Jones’s past offenses emerged largely through Dunham’s testimony, which

conveyed to the jury the details of the offenses as Dunham knew them from the

criminal records. Those offenses included convictions for attempted sexual assault,

assault, attempted aggravated sexual assault with a deadly weapon, and burglary with

intent to commit assault. We have summarized the details of Jones’s convictions as

testified to primarily by Dunham based upon his review of Jones’s criminal records.

Where appropriate, we have incorporated Jones’s trial testimony regarding the

convictions.

1. February 1996: Attempted sexual assault

The first sexual-offense conviction described at trial was Jones’ February 1996

attempted sexual assault of a 24-year-old college student. Jones attacked the woman

in the early morning hours when she was returning to her apartment with her one-

year-old son after taking him to the emergency room. Dunham testified,

[Jones] came up behind her and grabbed her from behind and she was
holding her baby, but he put his hand over her mouth when she tried to
scream and . . . she said he had a latex glove on - - on the hand that he
3
 covered her mouth with. He - - he pushed her against the wall. He tried
to - - he pulled her pants down. He tried to rip her underwear off. She
fell. She hit her head . . . against a - - a brick wall and the baby fell to the
ground and Mr. Jones began dragging her across the grass and she was
kicking and screaming and some neighbors came out and - - and started
to yell at him and then he took off and he ran off and then he turned
around and came back and he picked up something off the ground and it
was like a ski mask that he picked up off the ground and he got away.

In his interviews with Dunham and Gaines and in his testimony at trial, Jones

denied any wrongdoing and offered his own version of events. His explanations

asserted that he was visiting a friend in that apartment complex when he bumped into

a woman and pushed her because he was startled. In his testimony at trial, Jones

claimed that he tried to help her up, but “she became frightened and started to scoot

back on the grass and as she was scooting back on the grass, her - - her pants started

to come down.” Jones denied noticing a baby but thought he heard a baby crying in

the distance.

Jones pleaded guilty to attempted sexual assault related to the incident and was

sentenced to 10 years’ incarceration. During the commitment trial, Jones denied his

guilt and claimed that he only pleaded guilty to avoid a longer sentence.

Dunham testified that he believed the victim’s story. Characterizing the

incident as a “callous sexual act,” Dunham distinguished it from “a typical sexual

assault or attempted sexual assault” because Jones attacked his victim while she was

holding a baby. Dunham testified, “To me that kind of raises sort of the

psychopathic type of nature of it to a point where you would attempt to sexually

4
assault, you know, a woman holding a one-year old, you know, in her arms, but that

did not deter him.”

Dunham emphasized both the public and planned nature of the offense. It

occurred in public, indicating to Dunham an “inability to control urges.” Jones was

also prepared—with a ski mask and latex glove—and it appeared that he had been

following the woman prior to the attack. Dunham also remarked on the nature of the

victim as a stranger to Jones and explained, “A stranger situation elevates somebody’s

risk” to reoffend.

2. 1997: “Offensive or Provocative” Assault

Dunham testified that in September 1997 Jones was convicted of three assault1

charges that occurred in Arlington. Dunham described the events leading to the

convictions as follows:

• Jones approached a woman who was trying to open her apartment door, lifted
up her dress or skirt, and grabbed her buttocks. When the woman turned
around, he smiled at her and then left.

• Jones approached a woman who was walking down the sidewalk and pulled on
the bottom of her shorts.

• Jones approached a woman at her apartment complex mailboxes and lifted her
dress as she walked by.

1
Dunham refers to these convictions and suspected offenses as “offensive and
provocative conduct,” which is also how they are referred to in the municipal court
records of conviction. See Tex. Penal Code Ann. § 22.01(a)(3) (“A person commits an
offense if the person . . . intentionally or knowingly causes physical contact with
another when the person knows or should reasonably believe that the other will
regard the contact as offensive or provocative.”).

5
 Dunham also testified that, based on the records he reviewed, Jones was

suspected of committing several more assaults but was never charged, although Jones

supposedly admitted to one of them. Dunham testified that, in all, Jones was a

suspect in 12 separate Arlington cases, three of which resulted in the above-described

convictions. Those cases in which he was not charged or convicted included an

instance when a man sexually assaulted a woman at knifepoint in an apartment

laundry room early in the morning and then made her walk back to her apartment

naked, an instance when a man grabbed a woman’s buttocks while she was walking

past an apartment-complex swimming pool, and an instance when a man fondled a

woman over her clothes while she was taking boxes out of her vehicle outside her

apartment.

According to police records, Jones told Arlington police that he enjoyed

grabbing women because he liked to see the surprised looks on their faces. Dunham

found this admission to be clinically important because it indicated that Jones was

aroused by the “shock value” and “seeing the fear in [the women’s] faces.”

At trial, Jones admitted that he pleaded guilty to the three assault charges

committed in Arlington in 1997, but he denied committing the crimes and provided

dubious explanations for his conduct. For instance, he claimed that his watch caught

on a woman’s dress and that he did not raise her dress on purpose. He further denied

making a statement contained in a 1997 police record admitting to police that he had

committed other offenses for which he was never caught.
6
 3. February 1998: Attempted aggravated sexual assault with a
deadly weapon

In February 1998, Jones approached a woman in an apartment laundry room

one evening, held a gun to her head, threatened to kill her, unfastened her belt, and

put his hand up her shirt and fondled her breasts. He was interrupted when another

woman walked into the laundry room, at which point he put the gun into his coat

pocket and then fled the scene. Dunham pleaded guilty to attempted aggravated

assault with a deadly weapon and was sentenced to 20 years’ imprisonment, which he

was serving at the time of these civil-commitment proceedings.

Again, despite pleading guilty, Jones denied committing this offense and

claimed that he pleaded guilty only to avoid a longer sentence.

Dunham described this offense as “callous” in light of the weapon and also

viewed this offense as an escalation of Jones’s sexually-offensive behavior against

adult females with whom Jones had no relationship. Dunham testified that this

second serious offense established a pattern and separated Jones from “[m]ost sexual

offenders,” as he testified that most sexual offenders “don’t do it multiple times.”

4. May 1998: Attempted sexual assault

Jones assaulted another woman a mere three months later. Around 1:00 or

2:00 a.m. one night, Jones broke into a woman’s apartment by climbing through a

front window. The woman woke up and tried to call 9-1-1, but Jones approached

her, held a gun to her head, and told her to hang up. Jones proceeded to steal some

7
of the woman’s property and then sexually assaulted her—he forced the woman to

disrobe, fondled her genitals, and said, “[Y]ou know, it’s too bad that you called 9-1-1.

I was going to - - I would have had sex with you, but I’ve got to go because you called

the police.” Jones then left the apartment.

Jones pleaded guilty to the offense and was sentenced to 20 years, to be served

concurrently with the February 1998 offense sentence. Again, despite his guilty plea,

Jones denied his guilt. He testified that his friend lived in the complex and his friend

had given Jones permission to stay in the apartment even if the friend was not home.

Jones claimed that he went into the wrong window and pulled his gun on the woman

inside. He told Dunham and Gaines that he put his hand in the woman’s underwear

to check for a gun; at trial, he testified that he “patted in the front of her panties to

see if did she have . . . maybe a razor blade, box cutter, knife.” He also testified that

the woman gave him a necklace and a VCR, which he then pawned. Unsurprisingly,

Dunham and Gaines did not believe Jones’s version of events, a story Gaines

described as “just really not even believable.”

5. August 1998: Burglary with intent to commit assault

Continuing this pattern, a few months later in August 1998, Jones assaulted

another woman in an apartment complex. The woman had noticed someone peering

through her window around 6:45 a.m. and when her boyfriend left for work about 15

minutes later, leaving the door unlocked, Jones entered the apartment. The woman

reported that Jones claimed to be a maintenance man there to fix a leaky faucet but
8
that he quickly grabbed her arm, held a screwdriver to her throat, and told her to be

quiet. The woman was able to escape out of a back window and Jones fled.

Jones again denied committing the offense despite pleading guilty to burglary

with the intent to commit assault. For this offense, Jones received another concurrent

20-year sentence.

Despite the fact that this offense was not sexually violent, Dunham considered

it to be significant. Dunham explained that he considered it to be a sex-related

offense because the victim related that she felt Jones was about to rape her and it fit

Jones’s pattern of sexual offenses.

6. Arrest for 1998 offenses

Dunham testified Jones was not arrested until December 1998 when he was

identified as a suspicious person loitering in an apartment complex. Jones was

arrested with an unregistered weapon in his car, and Dunham also believed that the

police found a screwdriver in his car. While Jones was being held in jail for unlawfully

carrying a firearm, the victims identified him in a photo lineup and he was charged

with the 1998 offenses.

7. Nonsexual offenses

Dunham also reported Jones’s nonsexual criminal history, explaining why it

was important to determining his antisocial tendencies. Jones had been convicted of

assaulting his ex-wife and causing bodily injury for which he received a six-month

term of probation. Additionally, Jones’s nonsexual criminal history included a DWI,
9
driving without a license, failure to pay traffic tickets, and driving without insurance.

Dunham explained that even though these were nonsexual and—with the exception

of the assault of his ex-wife—nonviolent, it was irresponsible behavior that later

escalated to the multiple sexual assaults. And in Gaines’s view, Jones’s nonsexual

criminal history was indicative of antisocial personality traits.

B. Other risk factors

Dunham and Gaines identified a number of other risk factors including:

• The number of victims.

• The fact that his victims were strangers.

• The force and weapons used.

• The locations of the offenses in public places. Gaines testified that
forensic psychiatry literature “indicate[s] that offending in a public place
poses a risk factor for re-offending because it’s brazen.”

• Jones’s hypersexual nature. Dunham testified that Jones’s ex-wife had
described Jones as sexually promiscuous and that Jones had admitted to
hiring a prostitute.

• Jones’s denial of the offenses.

• Jones’s lack of remorse.

• Jones’s denial that he is a sex offender.

• Jones’s refusal to acknowledge his need for sex-offender treatment and
to avoid high-risk situations.

10
 C. Protective factors

Dunham and Gaines also identified protective factors that statistically reduce

an offender’s level of risk. These included:

• Jones’s age. At the time of trial, Jones was 52 and both experts testified
that an individual’s risk of reoffending decreases as they age. Dunham
noted, however, that Jones appeared “much younger” than 52 and
Jones’s sex drive still appeared to be high.

• Jones’s good behavior in prison. Dunham even noted that this was
surprising considering Jones’s escalating antisocial behavior before
prison.

• Jones’s stable childhood, family support system, and positive
employability. But Dunham also explained that Jones’s stable childhood
and his positive adjustment to prison indicated that Jones does well in a
structured environment and did not start committing crimes until he was
in an unstructured environment. This led Dunham to believe that the
fact that Jones would be required to register as a sex offender may not
be sufficient to deter future violent sexually-offensive behavior by Jones,
especially because Jones had committed sexual offenses while on parole.

• According to Dunham’s observations, there was no evidence at the time
of trial that Jones had a current sexual preoccupation.
D. Dunham’s and Gaines’s diagnoses and conclusions

1. Static-99R and PCLR results

Dunham administered two actuarial tests as part of his evaluation of Jones: the

Static-99R, which is designed to measure a person’s risk of committing another sex

offense in the future, and the PCLR, which Dunham described as “a measure of how

psychopathic somebody is.” Dunham scored Jones as a five on the Static-99R,

placing Jones in the “above average risk category” in comparison to “the average sex

11
offender.” Dunham testified that the average score of a “typical sex offender” is from

one to three, the average risk range. He also explained that Jones would have been a

six when he was younger, placing him in the range of a well-above-average risk.

Dunham scored Jones as a 20 on the PCLR, indicating a “[m]oderate level of

psychopathic characteristics.” Gaines took into account these test results but also

criticized placing too much reliance on such tests because they do not take into

account the individual facts of an offender’s crimes.

2. Sexual Sadism

Both Dunham and Gaines acknowledged Jones’s history of a pattern of violent

sexual offenses over multiple years. And both experts testified regarding Jones’s

sadistic behavior, but their opinions differed slightly regarding whether he could be

diagnosed with a sexually sadistic disorder.

As Gaines explained, sexual sadism disorder involves deriving sexual

gratification from someone else’s suffering or humiliation or from putting someone

else down. According to Gaines, to warrant a diagnosis, the disorder must last at least

six months and be problematic, i.e., cause the person problems in life, such as in

relationships, the person’s job, or result in criminal convictions. Gaines identified

several signs of sexual sadism in Jones’s sexual offenses—attacking a woman holding

a crying infant, ripping off a woman’s pants in public to humiliate her, threatening a

woman, “Don’t make a move or I’m going to kill you,” and assaulting two women at

gunpoint. Gaines diagnosed Jones as a full-blown sexual sadist because she
12
concluded that Jones derived intense sexual arousal from his victims’ physical and

psychological suffering based upon his continued pattern of behavior of approaching

women sexually and grabbing their bodies.

Dunham, on the other hand, stopped short of diagnosing Jones as a sadist

because he was unsure of the source of Jones’s arousal. Dunham testified, “The

reason I lean towards the sexual sadistic side is his comment to the officers saying that

he liked to enjoy the surprised look on [the women’s] faces when he was describing

why he was groping women and lifting up the[ir] skirts.” Despite his hesitancy to

fully diagnose Jones as a sadist, Dunham nevertheless testified that this paraphilic

disorder affected Jones’s emotional and volitional capacity in the past and made him a

menace to the health and safety of other people.

3. Antisocial behavior

Dunham and Gaines both identified Jones as having certain antisocial

personality traits. As Dunham testified, “Antisocial means kind of criminal mindset,

law-breaking behavior, violating the norms of society. . . . Adult antisocial behavior

indicates that . . . as an adult he has gotten into trouble, law-breaking behavior,

violating societal norms[.]” Gaines emphasized Jones’s lack of remorse, his denial of

his offenses, and his tendency to lie as an antisocial personality trait; she described

Jones’s renditions of his past offenses as “absurd” and “pathological” lies that could

even be indicative of some degree of psychopathy.

13
 4. Substance abuse

Dunham and Gaines also diagnosed Jones with a history of cannabis and

alcohol abuse. According to Gaines, Jones took a course about substance abuse “a

long time ago” in prison but he could not remember anything he learned in the

course. Gaines also expressed concern about Jones’s admission that before he was

incarcerated he had regularly smoked marijuana in combination with prescription

medication.

5. The experts’ conclusions

Dunham and Gaines both concluded that Jones suffers from a behavioral

abnormality that makes him likely to commit predatory acts of sexual violence. In

addition to the above observations, Dunham expressed his concern that Jones had

not received any sex-offender treatment and that Jones did not believe he needed any

such treatment. Dunham described Jones’s attitude as “a poor - - poor appraisal of

his own risk. It just shows that he doesn’t - - there’s a difference between denial and

actually not understanding one’s risk and I - - I don’t think he understands his risk.”

Dunham testified that Jones was “clearly” a high-risk offender when he entered

prison and expressed doubt that Jones’s age or the controlled environment of prison

had any effect of reducing that high risk. Dunham continued, “[Jones’s] current

beliefs and attitudes about it[,] that has not changed. He hasn’t had any type of

intervention to help him understand his triggers, understand his high-risk situations,

14
understand his offending cycle. And he’s about to be released without any - - without

any knowledge of - - of how to encounter those . . . temptations in the future.”

II. The jury’s verdict

The jury found that Jones is a sexually violent predator and the trial court

entered an order of commitment.

Discussion

Jones brings six issues on appeal. In his first two issues, Jones argues that the

trial court erred by allowing Dunham to testify to “basis” evidence that described the

underlying facts related to Jones’s convictions. In his third and fourth issues, Jones

attacks the legal and factual sufficiency of the evidence supporting the jury’s finding

that he is a sexually violent predator. In his fifth issue, Jones complains of the trial

court’s submission of an Allen charge to the jury. And in his sixth issue, Jones argues

that the trial court erred by denying his request that the jury be instructed that it could

render a verdict in his favor by a 10-2 vote.

We will first address the legal sufficiency of the evidence to support the order

of commitment because, if sustained, it would afford Jones the greatest relief. See

Tex. R. App. P. 43.3; Bradleys’ Elec., Inc. v. Cigna Lloyds Ins. Co., 995 S.W.2d 675, 677

(Tex. 1999).

15
I. Sufficiency of the evidence

A. Legal sufficiency

We review sexually-violent-predator civil commitment proceedings for legal

sufficiency of the evidence using the appellate standard of review applied in criminal

cases. In re Commitment of Short, 521 S.W.3d 908, 911 (Tex. App.—Fort Worth 2017,

no pet.). We assess the evidence in the light most favorable to the verdict to

determine whether any rational trier of fact could find the statutory elements required

for commitment beyond a reasonable doubt. Id.

To have an offender civilly committed, the State must show beyond a

reasonable doubt that the person (1) is a repeat sexually violent offender, and (2)

suffers from a behavioral abnormality that makes him likely to engage in a predatory

act of sexual violence. See Tex. Health & Safety Code Ann. §§ 841.003 (defining

“Sexually Violent Predator”), 841.062(a) (imposing a “beyond a reasonable doubt”

burden of proof). Jones does not dispute his status as a sexually violent offender

because he has more than one conviction for a sexually violent offense. See id. §

841.003 (defining “repeat sexually violent offender” as a person who has been

convicted of more than one sexually violent offense).

Jones argues that the evidence supporting the jury’s verdict is legally

insufficient because Dunham’s and Gaines’s expert opinions relied upon facially

unreliable evidence and therefore constituted no evidence. To support his argument,

Jones points to supreme court holdings that a conclusory, unsupported expert
16
opinion cannot be considered probative evidence, regardless of whether there is no

objection. But reliance on these cases for support of this argument is misplaced. City

of San Antonio v. Pollock, 284 S.W.3d 809, 818 (Tex. 2009).

Pollock involved claims for negligence and nuisance related to a child’s in utero

exposure to harmful gases from a closed municipal landfill. Id. at 812. An expert

testified to the recorded gas levels in a sealed monitoring well on the child’s family’s

property and concluded that the family was exposed to the same gas level elsewhere

on their property, outside of the sealed monitoring well. Id. at 818. The supreme

court held that the expert’s opinion suffered from a fatal analytical gap because there

was no evidence allowing an inference that the gas concentration in the ambient air of

the home and yard would have been remotely close to the gas concentration within

the sealed monitoring well. Id. at 818–19. The expert’s “naked conclusion” therefore

constituted no evidence. Id. at 819.

But here Jones does not demonstrate a similar analytical gap within Dunham’s

or Gaines’s opinions. Instead, Jones argues that their opinions constituted no

evidence because, in his view, they relied upon “facially unreliable evidence” gathered

from Jones’s criminal records. Pollock and other cases recognize the difference in this

argument: “When a scientific opinion is admitted in evidence without objection, it

may be considered probative evidence even if the basis for the opinion is unreliable.” Id. at

818 (emphasis added); see also Coastal Transp. Co., Inc. v. Crown Cent. Petroleum, 136

S.W.3d 227, 233 (Tex. 2004) (distinguishing between challenges to an expert’s
17
scientific methodology and “no evidence challenges where, on the face of the record,

the evidence lacked probative value”) (citing Maritime Overseas Corp. v. Ellis, 971

S.W.2d 402, 412 (Tex. 1998), cert. denied, 525 U.S. 1017).

Furthermore, to preserve his argument that these opinions were based on

unreliable evidence, Jones was required to lodge an objection to such opinion

testimony at trial. Coastal, 136 S.W.3d at 233 (concluding that “when a reliability

challenge requires the court to evaluate the underlying methodology, technique, or

foundational data used by the expert, an objection must be timely made so that the

trial court has the opportunity to conduct this analysis”). Jones’s only objection to

Dunham’s and Gaines’s testimony related to the contents of the criminal records was

based on hearsay. He therefore failed to preserve any objection that the expert

opinions were based upon supposedly unreliable evidence. See id.; see also Tex. R. App.

P. 33.1(a).

Even if we consider Jones’s argument as one that Dunham’s and Gaines’s

opinions lacked any probative value because they were unsupported by the record, it

fails. Dunham’s and Gaines’s opinions are not conclusory or speculative in such a

way as to render them non-probative. Both experts explained the basis of their

opinions, which included, among other things, a review of Jones’s criminal history,

including his convictions and uncharged behavior. Based upon their reviews of his

criminal history—which included multiple sexual offenses that escalated from lifting

strange women’s skirts to sexually assaulting women at gunpoint—and other risk
18
factors, both experts concluded that Jones suffers from a behavioral abnormality that

makes him likely to commit sexual offenses in the future. Their opinion testimonies

were not bare conclusions but instead were based upon research and reason. Their

opinions were legally sufficient to support the jury’s verdict.

We overrule Jones’s third issue.

II. Nonunanimity jury instruction

In his sixth issue, Jones argues that the trial court erred by declining to issue a

jury instruction that the jury could render a verdict in Jones’s favor nonunanimously,

by a vote of 10-2.2 Whether a jury can render a “No” verdict upon agreement of ten

jurors appears to be an issue of first impression.3

2
Jones requested the following instruction be included in the general
instructions for signing the verdict certificate:

1. A “no” answer means that at least 10 jurors agree the answer
to the question is “no[.”] If at least 10 jurors agree that the answer to
the question is “no[,”] those jurors must sign the verdict.

2. A “yes” answer must be unanimous. This means all 12 jurors
must agree [the] answer to the question is “yes[.”]
3
The Austin court of appeals is currently considering the issue in two cases, In
re Commitment of R.G., 03-18-00331-CV, and In re Commitment of A.G., 03-18-00332-CV.
In those cases, the juries each rendered a verdict that the sex offender was not a
sexually violent predator and did so with votes of 10 jurors. The State, as the
appellant in those cases, has argued that the jury must be unanimous in finding that a
person is not a sexually violent predator. Notably, the State declined to make those
arguments on appeal in the instant case, instead jumping straight to an argument that
Jones has failed to show that any error was harmful.

19
 This is an issue of statutory construction, a matter of law that we review de

novo. Melden & Hunt, Inc. v. East Rio Hondo Water Supply Corp., 520 S.W.3d 887, 893

(Tex. 2017). When the text of a statute is clear, as it is here, we do not resort to rules

of construction or extrinsic aids to construe the text; we give effect to the ordinary or

common meaning of each word of the statute. Tex. Gov’t Code Ann. § 312.002(a);

Melden, 520 S.W.3d at 893.

Section 841.062 provides:

(a) The judge or jury shall determine whether, beyond a reasonable
doubt, the person is a sexually violent predator…
(b) A jury determination that the person is a sexually violent predator
must be by unanimous verdict.
Tex. Health & Safety Code Ann. § 841.062.

This statute unambiguously requires that the jury must be unanimous in its

determination that a person is a sexually violent predator. But by its plain words, the

statute clearly does not require that the jury be unanimous in a determination that a

person is not a sexually violent predator. Consequently, the rules of civil procedure

apply and such a determination only requires the agreement of 10 jurors. Tex. Health

& Safety Code Ann. § 841.146(b) (“Except as otherwise provided by this subsection, a

civil commitment proceeding is subject to the rules of procedure and appeal for civil

cases.”); Tex. R. Civ. Proc. 292(a) (“a verdict may be rendered in any cause by the

concurrence . . . of the same ten or more members of an original jury of twelve”).

20
 The contrast between the wording of subsections (a) and (b) further supports

this conclusion. The legislature chose to write subsection (a) as requiring the judge or

jury to “determine whether . . . the person is a sexually violent predator.” Id. §

841.062(a). If the legislature intended to require a unanimous jury to find that a

person is not a sexually violent predator, it could have similarly worded subsection (b),

to wit: A jury determination of whether the person is a sexually violent predator must

be by unanimous verdict. Or it could have omitted subsection (b) altogether and

stated in subsection a, “The judge or jury, by unanimous verdict, shall determine whether

. . . the person is a sexually violent predator.” See, e.g., Tex. Fam. Code Ann. § 54.03(c)

(requiring juvenile-delinquency jury determinations to be unanimous: “Jury verdicts

under this title must be unanimous”). Instead, the legislature specifically directed the

unanimous verdict requirement in subsection (b) to a determination “that the person is

a sexually violent predator.” Id. (b) (emphasis added). We presume that the

legislature meant what it said and that it intended to require unanimity only for a

determination that a person is a sexually violent predator, not for a determination that

a person is not. See Fitzgerald v. Advanced Spine Fixation Sys., Inc., 996 S.W.2d 864, 866

(Tex. 1999) (“[I]t is a fair assumption that the Legislature tries to say what it means,

and therefore the words it chooses should be the surest guide to legislative intent.”).

At trial, the State argued that the legislature has “never . . . contemplated in any

type [of] trial for it to be unanimous for one [jury answer] and ten to two for another

[jury answer].” The State was wrong; the legislature has contemplated exactly that in
21
the closely analogous situation of exemplary damage awards. By statute, “[e]xemplary

damages may be awarded only if the jury was unanimous in regard to finding liability

for and the amount of exemplary damages.” Tex. Civ. Prac. & Rem. Code Ann. §

41.003(d); see also Tex. R. Civ. P. 292(b) (“A verdict may be rendered awarding

exemplary damages only if the jury was unanimous in finding liability for and the

amount of exemplary damages.”). Conversely, only 10 jurors need to agree that

exemplary damages are not warranted. See Tex. R. Civ. P. 226a, comment (“You are

instructed that in order to answer ‘Yes’ to [the question of liability for exemplary

damages], your answer must be unanimous. You may answer ‘No’ to [the question of

liability for exemplary damages] only upon a vote of 10 . . . or more jurors.”); see also

Comm. on Pattern Jury Charges, State Bar of Texas, Texas Pattern Jury Charges:

General Negligence, Intentional Personal Torts & Workers’ Compensation PJC 4.2C,

7.11B, 10.14C (2018).4 We therefore hold that the trial court erred by declining to

instruct the jury that it may render a “no” finding with 10 jurors’ concurrence.

Having found error, we must determine harm. To obtain a reversal of a

judgment on the basis of trial-court error in civil cases, the appellant must show that

the error occurred and that it probably caused rendition of an improper judgment or

probably prevented the appellant from properly presenting the case to this court.

4
The legislature has also seen fit to allow such verdicts in criminal cases when
the death penalty is considered. See Tex. Code Crim. Proc. Ann. art. 37.071(d)(2),
(f)(2) (requiring unanimity to impose the death penalty but a ten-juror vote to answer
“no” to death-penalty special issues).

22
Tex. R. App. P. 44.1(a); Romero v. KPH Consolidation, Inc., 166 S.W.3d 212, 225 (Tex.

2005). Charge error is generally considered harmful if it relates to a “contested,

critical issue.” Transcon. Ins. v. Crump, 330 S.W.3d 211, 225 (Tex. 2010) (quoting

Columbia Rio Grande Healthcare, L.P. v. Hawley, 284 S.W.3d 851, 856 (Tex. 2009)).

Unless the appellate court is reasonably certain that the jury was not significantly

influenced by issues erroneously submitted to it, the error is reversible. Romero, 166

S.W.3d at 227–28.

The State argues that Jones has failed to show harm because there is no

evidence in the record of how the jurors’ vote may have been split. This cannot be

the test. To accept the State’s argument would create an insurmountable barrier to

any defendant’s ability to show harm in such circumstances, given that polling a

potentially deadlocked jury is generally considered improperly coercive and therefore

error. See Lowenfield v. Phelps, 484 U.S. 231, 232, 108 S. Ct. 546, 548 (1988) (discussing

Brasfield v. United States, 272 U.S. 448, 47 S. Ct. 135 (1926) and the impropriety of

polling a jury prior to the delivery of a verdict); Stevens v. Travelers Ins. Co., 563 S.W.2d

223, 229 (Tex. 1978) (approving of the use of verdict-urging instructions in civil cases

but warning of the inherent “danger of coercion” in their use); Barnett v. State, 161

S.W.3d 128, 134 (Tex. App.—Fort Worth 2005) (“Courts have found jury coercion

23
when . . . the trial court continues to poll the jury after a lack of unanimity is

revealed.”), aff’d, 189 S.W.3d 272 (Tex. Crim. App. 2006).5

Although we do not know precisely how the vote among the jurors was split

during deliberations, we know that a split existed. The jury issued four notes

concerning the substance of the case: three requested portions of the record to review

testimony or definitions of certain terms and one, issued four-and-a-half hours after

the jury began deliberating, declared that the jury was deadlocked. In response, a

modified Allen charge was delivered by the trial court, urging the jury to continue

deliberating. A unanimous verdict was delivered about an hour and fifteen minutes

later. Because we know that a split existed, a 10-2 instruction could have had a

significant impact on this situation.

It is beyond argument that the requested charge relates to a contested

element—indeed, the central question for the jury to resolve—and is therefore

generally considered harmful. See Transcon, 330 S.W.3d at 226 (holding that charge

error relating to question of causation in worker’s compensation case related to the

sole issue in the case and was therefore harmful). Considering this, the jury’s

temporary deadlock, and the record as a whole, we cannot be reasonably certain that

5
The failure of the jury to reach either a ten-person or unanimous agreement
would result in a retrial. See Tex. R. Civ. Proc. 289 (permitting the discharge of a jury
that cannot agree to a verdict and instructing that the cause shall be retried at the
direction of the trial court), 292(a) (requiring a minimum of ten jurors to agree to a
verdict in a civil case).

24
the verdict was not significantly influenced by the trial court’s error. See Romero, 166

S.W.3d at 227–28. Having found harmful error, we sustain Jones’s sixth issue.

Conclusion

Having held that the trial court committed reversible error by denying Jones’s

request for a 10-2 jury instruction, we reverse the trial court’s order of commitment

and remand this case to the trial court for a new trial.6 See Transcon, 330 S.W.3d at 227

(reversing and remanding case for a new trial where trial court committed harmful

error by rejecting defendant’s requested jury instruction).

/s/ Bonnie Sudderth
Bonnie Sudderth
Chief Justice

Delivered: February 28, 2019

6
We therefore do not reach Jones’s remaining four issues. See Tex. R. App. P.
43.3; Bradley’s Elec., 995 S.W.2d at 677.

25