Ambrose, Cynthia

IN THE COURT OF CRIMINAL APPEALS
OF TEXAS
NO. PD-0143-15

THE STATE OF TEXAS

v.

CYNTHIA AMBROSE, Appellee

ON APPELLEE’S PETITION FOR DISCRETIONARY REVIEW
FROM THE FOURTH COURT OF APPEALS
BEXAR COUNTY

A LCALA, J., delivered the opinion of the Court in which K ELLER, P.J., M EYERS,
J OHNSON, K EASLER, H ERVEY, R ICHARDSON, and N EWELL, J.J., joined. Y EARY, J., did
not participate.

OPINION

This case addresses accomplice-witness jury-charge error raised in a motion for new

trial. Cynthia Ambrose, appellee, is a former kindergarten teacher at Salinas Elementary

School who was convicted of the misdemeanor offense of official oppression after a jury trial

at which the testimony of a purported accomplice, Ramirez, was presented. After the jury

found her guilty, Ambrose filed a motion for new trial alleging that the jury instructions were
 Ambrose - 2

erroneous in that they failed to instruct the jury that Ramirez’s accomplice-witness testimony

had to be corroborated. See T EX. C ODE C RIM. P ROC. art. 38.14. She further asserted that she

was egregiously harmed by the error. The trial court agreed with Ambrose, and it made what

it characterized as findings of fact and conclusions of law in support of its ruling. On appeal,

the court of appeals assumed without deciding that the trial court properly determined that

the jury instructions were erroneous, but, as to the matter of harm, it disregarded the trial

court’s findings and conclusions and it instead determined that Ambrose was not egregiously

harmed under the substantive application of the Almanza factors. State v. Ambrose, 457

S.W.3d 154, 162 (Tex. App.—San Antonio 2015). In her three grounds in her petition for

discretionary review, Ambrose’s first two grounds contend that the appellate court erred by

applying the Almanza egregious-harm standard and by failing to defer to the trial court’s

findings of fact made after the motion-for-new-trial hearing. See Almanza v. State, 686

S.W.2d 157 (Tex. Crim. App. 1984) (op. on. reh’g). Ambrose’s third ground for review

asserts that the appellate court erred in concluding that the error was not egregiously harmful

to her.1 We conclude that the court of appeals properly held (1) that the Almanza harm

1
The three grounds in Ambrose’s petition state:

(1) When a trial judge issues findings of fact and conclusions of law that find a
defendant suffered egregious harm from unobjected to jury charge error, does
applying the Almanza egregious harm standard on appellate review violate and
conflict with Texas (Ex parte Wheeler, 203 S.W.3d 317 (Tex. Crim. App. 2006)) and
United States Supreme Court (Oregon v. Kennedy, 456 U.S. 667 (1982)) precedent
that a reviewing court must defer to a lower court’s factual findings?

(2) Under the egregious harm standard, does an appellate court violate Texas (Ex
 Ambrose - 3

standard applies to jury-charge error reviewed on appeal, even when the error was addressed

in a motion for new trial, (2) that it was not required to defer to the trial court’s factual

findings in this case, and (3) that the record failed to show that Ambrose was egregiously

harmed by the error in the charge. We, therefore, affirm the judgment of the court of appeals.

I. Background

Ramirez was a fellow kindergarten teacher at the same school as Ambrose. One of

Ramirez’s students, A.N., was behaving disruptively in class and had hit another student.

Pursuant to a school policy that directs a teacher to take a disruptive child into another

teacher’s classroom, Ramirez took A.N. into Ambrose’s classroom. Ramirez asked Ambrose

if A.N. could stay with her “[b]ecause he was bullying the other student.” Ambrose

instructed A.N. to sit right beside her and asked the questions “Why are you bullying other

students?” and “How would you like for other students to bully you?” Ambrose then said

to the students in her classroom, “Come on, boys and girls, let’s line up and let’s bully

[A.N.].” About seven students struck A.N. before the assault ended when one girl struck

parte Wheeler, 203 S.W.3d 317 (Tex. Crim. App. 2006)) and United States Supreme
Court (Oregon v. Kennedy, 456 U.S. 667 (1982)) precedent when it ignores a trial
court’s factual findings and substitutes its own view of the evidence for that of the
trial [court]?

(3) If the egregious harm standard does apply on direct review in this case, did the
appellate court correctly apply the egregious harm standard when it only considered
the testimony that supported the State’s case and not “the entire jury charge, the state
of the evidence, including the contested issues and weight of probative evidence, the
argument of counsel and any other relevant information revealed by the record of the
trial as a whole” as required by Almanza v. State, 686 S.W.2d 157 (Tex. Crim. App.
1984)?
 Ambrose - 4

A.N. harder than the other students had, prompting Ambrose to say, “Okay, not that hard.”

Ramirez then left Ambrose’s classroom and later sent one of her other students to retrieve

A.N. About two weeks later, Ramirez reported the incident to Jeffrey Large, the school’s

principal, and Gerrie Spellmann, the vice-principal. Large and Spellmann investigated the

accusation, which included having a discussion with Ambrose about the events.

Ambrose was charged with the offense of official oppression under Texas Penal Code

Section 39.03 for “intentionally subject[ing] [A.N.] . . . to mistreatment, to wit: directing and

allowing others to strike the complainant, knowing it was unlawful.” 2 At Ambrose’s trial,

other than Ramirez’s testimony, the jury heard testimony from Large and Spellmann, in

addition to other witnesses. Large and Spellmann each recited statements made by Ambrose

admitting that she had instructed students to hit A.N. and that she then observed two or three

students hit A.N. before she stopped the assault on A.N. According to Spellmann, Ambrose

“let us know that she had instructed her students to hit [A.N.] on the arm, but not too hard

so that the student would know how it felt to be bullied.” Spellmann further testified,

“[Ambrose] stated that—we asked how many students had [ ] struck [A.N.] and she stated

2
Section 39.03 of the Texas Penal Code states in relevant part,

(a) A public servant acting under color of his office of employment commits an
offense if he:
(1) intentionally subjects another to mistreatment . . . that he knows is
unlawful. . . .
(b) For purposes of this section, a public servant acts under color of his office of
employment if he acts or purports to act in an official capacity. . . .

TEX . PENAL CODE § 39.03.
 Ambrose - 5

only about two or three. But the next student hit too hard and that she had it stopped.” Like

Spellmann, Large testified that, during the investigation of the incident, Ambrose stated that

“she instructed the students to hit the other student but not hard. She said that I believe two

or three students hit the student and then a fourth student hit too hard and then she stopped

it immediately after that.”

In her defense at trial, Ambrose acknowledged the essence of the events as they had

been described by Ramirez, Large, and Spellmann, but she minimized her culpability by

claiming that she did not intend for any of the students to hit A.N. and that she immediately

stopped the assault when the first student suddenly struck A.N. Ambrose testified that

Ramirez brought A.N. into her classroom and said that he was being a bully. Ambrose asked

the other students what A.N.’s consequences should be, and they made different suggestions,

one of which was that the child A.N. hit should be allowed to hit him back. Ambrose asked,

“Does anybody want to show him what it feels like?” Before she knew what was happening,

one student struck A.N., and Ambrose immediately intervened, telling the other students to

“get away from him.” When the prosecutor asked, “Did you ever intend for one student to

hit another student?” Ambrose responded, “No, that wasn’t my intention.”

After the close of evidence, the trial court’s instructions to the jury did not include an

accomplice-witness instruction, and Ambrose did not object to its absence. The jury found

Ambrose guilty. The court assessed her punishment at one year of confinement, probated for

two years, with a 30-day jail sanction as a condition of her community supervision.
 Ambrose - 6

Following her conviction, Ambrose filed a motion for a new trial that alleged, among

other complaints, that the jury instructions were erroneous based on the trial court’s failure

to instruct the jury regarding the law of accomplice-witness testimony. Ambrose claimed

that Ramirez was an accomplice to the offense and that the jury should have been instructed

that Ambrose could not be convicted based solely on Ramirez’s uncorroborated testimony.

The trial court granted the motion for a new trial based on its determination that the

absence of the accomplice-witness instruction was erroneous and that Ambrose was

egregiously harmed by the error. The trial court made findings of fact and conclusions of law

in support of its ruling.3 The State appealed. On appeal, the State contended that the trial

court erred by granting Ambrose’s motion for new trial. The State alternatively claimed that

3
In its findings of fact and conclusions of law in support of its ruling, the trial court
summarized the substance of Ramirez’s testimony, but it did not make any finding regarding the
credibility of that testimony. The trial court further found that Ramirez was provided transactional
immunity in exchange for her testimony. In addition, based on evidence that was presented in the
post-trial proceedings, the trial court found it to be true that Judson ISD police initially intended to
file charges against Ramirez “for official oppression and [for] failing to report this event,” but later
decided to pursue charges only against Ambrose. The findings further state, “When asked by this
Court why Ms. Ramirez received immunity during Ms. Ambrose’s sentencing, the prosecutor for the
State responded that [Ramirez’s] testimony was necessary for the State’s case because she was the
only other adult to witness the events.” The findings indicate that Ramirez was an accomplice as
a matter of law, or, “[a]t the very least,” she was an accomplice as a matter of fact. The findings note
that, although the trial court had initially prepared jury instructions that included an accomplice-
witness instruction, that instruction was not ultimately included in the instructions that were
submitted to the jury. The findings state that “[n]o request for a jury instruction on accomplice as
a matter of fact or as a matter of law was requested in this case,” and they further state that
Ambrose’s trial counsel testified at the hearing on the motion for new trial that he “ultimately
decided [not to] request an instruction for his own strategic purposes.”

In its conclusions of law, the trial court summarized both the law on accomplice-witness
testimony and the general law governing errors in the charge. Applying that law, the court concluded
that the failure to include the accomplice-witness instruction had caused Ambrose egregious harm.
 Ambrose - 7

the jury instructions were correct because Ambrose was not entitled to an accomplice-witness

instruction for Ramirez’s testimony, and that even if the instructions were erroneous on that

basis, the record failed to show that Ambrose was egregiously harmed by the absence of the

instruction. The court of appeals did not address the State’s first alternative argument, but

it agreed with the second alternative under the assumption that error had occurred.

The court of appeals held that the trial court erred by granting Ambrose’s motion for

new trial because the record failed to show that she was egregiously harmed by the absence

of an accomplice-witness instruction for Ramirez’s testimony. Ambrose, 457 S.W.3d at 162.

The court of appeals applied the Almanza egregious-harm standard because Ambrose had not

objected to the omission of the accomplice-witness instruction. Id. at 160. In analyzing the

evidence for egregious harm, the court of appeals observed that “Ambrose agreed during her

testimony that she was a public servant acting under the color of state law and that subjecting

a child to corporal punishment through other students would constitute mistreatment. Thus,

the only contested element of the charged offense was whether Ambrose intended to subject

A.N. to being struck by other students.” Id. The court of appeals therefore focused its

analysis “on evidence ‘tending to connect’ Ambrose’s intent to the intent alleged in the

charge—that Ambrose intended that other students strike A.N.” Id. at 160-61. The court of

appeals reviewed the record and observed that the “non-accomplice corroborative evidence

. . . is relatively strong.” Id. at 161. In particular, it noted that Ambrose “admitted to saying

something that prompted students in her classroom to get up and strike A.N.” Id. The court
 Ambrose - 8

observed that, although Ambrose denied explicitly directing her students to hit A.N., “her

own testimony tended to connect her intent to the charged intent of subjecting A.N. to

mistreatment by directing and allowing her students to strike him.” Id. The court also took

note of the testimony from Large and Spellmann, both of whom indicated that Ambrose had

admitted to instructing her students to hit A.N. Id. Regarding the jury charge and the State’s

arguments, the court reasoned that these factors “emphasized the strength of the

corroborating evidence and the weakness of Ambrose’s contradictory testimony.” Id. The

court concluded that “[t]he inclusion of an accomplice-witness instruction would not have

rendered the State’s case clearly and significantly less persuasive,” and, thus, it held that

Ambrose was not “deprive[d] [ ] of a fair and impartial trial” as a result of the omission of

the accomplice-witness instruction from the charge. Id. at 162.

II. Procedural Standards for Jury-Charge Error Raised in a Motion for New Trial

Ambrose’s first two grounds for review challenge the procedural standards employed

by the court of appeals. In her first ground, she contends that the court of appeals erred by

applying the Almanza egregious-harm standard on appellate review rather than applying the

abuse-of-discretion standard usually employed by an appellate court reviewing a trial court’s

ruling on a motion for a new trial. See State v. Thomas, 428 S.W.3d 99, 103-04 (Tex. Crim.

App. 2014). In her second ground, she contends that the court of appeals impermissibly

substituted its own view of the evidence for that of the trial court by ignoring the trial court’s

factual finding that Ambrose suffered egregious harm. After discussing the general
 Ambrose - 9

requirements for accomplice-witness instructions, we address these two grounds in turn.

A. General Requirements for Accomplice-Witness Instructions

Because Ambrose complains about the absence of the accomplice-witness instruction,

we begin by briefly reviewing that law. “A conviction cannot be had upon the testimony of

an accomplice unless corroborated by other evidence tending to connect the defendant with

the offense committed; and the corroboration is not sufficient if it merely shows the

commission of the offense.” T EX. C ODE C RIM. P ROC. art. 38.14. Therefore, if an accomplice

to the offense testifies for the State, the accomplice’s testimony must be corroborated by non-

accomplice evidence that tends to “connect the accused to the offense.” Smith v. State, 332

S.W.3d 425, 442 (Tex. Crim. App. 2011). Such evidence may be either direct or

circumstantial. Cathey v. State, 992 S.W.2d 460, 462 (Tex. Crim. App. 1999) (“It is not

necessary that the corroborating evidence directly connect the defendant to the crime or that

it be sufficient by itself to establish guilt; it need only tend to connect the defendant to the

offense.”). Depending on the circumstances, an accomplice-witness instruction may direct

the jury that an accomplice is one as a matter of law or one as a matter of fact for the jury’s

determination. Zamora v. State, 411 S.W.3d 504, 510 (Tex. Crim. App. 2013) (“A proper

accomplice-witness instruction informs the jury either that a witness is an accomplice as a

matter of law or that he is an accomplice as a matter of fact. The evidence in each case will

dictate the type of accomplice-witness instruction that needs to be given.”) (citations

omitted).
 Ambrose - 10

Here, like the court of appeals, we will assume that the trial court erred by failing to

include an accomplice-witness instruction for Ramirez’s testimony. When the issue is raised

by the evidence, the trial court must instruct the jury on the accomplice-witness rule because

it is the “law applicable to the case.” Id. at 513; T EX. C ODE C RIM. P ROC. art. 36.14.

B. Almanza Framework for Analyzing Harm Applies in a Motion for New Trial

Because Ambrose’s complaint is one of jury-charge error, the Almanza framework for

assessing harm applies.

Under Almanza, the degree of harm required for reversal depends on whether
the error was preserved in the trial court. Ngo v. State, 175 S.W.3d 738, 743
(Tex. Crim. App. 2005); Almanza, 686 S.W.2d at 171. Where, as here, the
defendant did not raise a timely objection to the jury instructions, reversal is
required only if the error was fundamental in the sense that it was so egregious
and created such harm that the defendant was deprived of a fair and impartial
trial.

Villarreal v. State, 453 S.W.3d 429, 433 (Tex. Crim. App. 2015). This “basic framework”

for analyzing preserved and unpreserved claims of error in the charge is “not a court-made

rule; it is based on this Court’s interpretation of Article 36.19.” Posey v. State, 966 S.W.2d

57, 60 (Tex. Crim. App. 1998).4

4
Article 36.19 of the Code of Criminal Procedure states,

Whenever it appears by the record in any criminal action upon appeal that any
requirement of Articles 36.14, 36.15, 36.16, 36.17 and 36.18 has been disregarded,
the judgment shall not be reversed unless the error appearing from the record was
calculated to injure the rights of defendant, or unless it appears from the record that
the defendant has not had a fair and impartial trial. All objections to the charge and
to the refusal of special charges shall be made at the time of the trial.

TEX . CODE CRIM . PROC. art. 36.19.
 Ambrose - 11

This Court has held that Almanza is the appropriate standard for conducting a harm

analysis in cases reviewing jury-charge error, even when the error was first asserted in a

motion for a new trial. Igo v. State, 210 S.W.3d 645, 647 (Tex. Crim. App. 2006). In Igo,

as in this case, the defendant had complained of an error in the charge for the first time in a

motion for new trial. Id. at 646. After the trial court denied his motion, Igo challenged this

ruling on appeal, and the court of appeals applied the Almanza egregious-harm standard and

concluded that the error was harmless. Id. On discretionary review in this Court, Igo

asserted that Almanza was the improper standard for reviewing the trial court’s ruling on his

motion for new trial and that the ruling should instead be reviewed for abuse of discretion.

Id. In rejecting Igo’s complaint, this Court explained that, “[e]ven though [Igo] characterized

his claim as error in denying him a new trial, this case presents error in the charge,” and thus,

“the proper standard is Article 36.19, as construed in Almanza.” Id. at 647. This Court

further reasoned that “a statute cannot be superceded by a rule,” and when a statute “directs

what treatment an appellate court must give to a particular type of error, a rule of appellate

procedure cannot be employed to circumvent the statutory requirement.” Id. Further, as a

policy consideration, the Court observed that, were it to accept Igo’s argument, “defendants

would no longer be required to preserve a jury-charge error at trial so long as the issue was

raised in a motion for new trial because any error in the charge could be said to ‘misdirect’

the jury.” Id. Such a rule would “contradict[ ] the policy of encouraging the timely

correction of errors, which is embodied both in Article 36.19 and in our own rules of
 Ambrose - 12

appellate procedure.” Id. Moreover, the Court observed that Igo’s proposed rule would

“essentially eviscerate the two-tiered harm analysis required by [Article 36.19] and do away

with the requirement that egregious harm be shown when the defendant has failed to timely

urge an objection.” Id. For these reasons, the Court concluded that Igo could not “use a

motion-for-new-trial appellate rule to circumvent the harm analysis imposed by Article

36.19.” Id. Thus, under Igo and Almanza, unpreserved jury-charge error does not require a

new trial, even when the error is complained of in a motion for new trial, unless the error

causes “egregious harm.” See id.; Almanza, 686 S.W.2d at 171.

Although she recognizes that this Court has already held in Igo that the Almanza

egregious-harm standard, as opposed to an abuse-of-discretion standard, applies under these

circumstances, she contends that Igo is distinguishable. In particular, she states that Igo

involved a trial court’s denial of a motion for new trial, whereas this case involves a trial

court’s granting of a new trial. This argument is without merit. Nothing in the language of

Igo suggests that its holding is limited to review of rulings denying motions for new trial, and

we can conceive of no reason why it should be so limited. Moreover, in a case subsequent

to Igo that involved an appeal of a trial court’s ruling granting a motion for new trial, this

Court granted the petition, vacated the court of appeals’s judgment, and remanded that case

to the court of appeals for it to reconsider its ruling in light of Igo. See State v. McKnight,

213 S.W.3d 915 (Tex. Crim. App. 2007) (per curiam) (broadly describing the holding of Igo

as being that “the Almanza harm analysis does apply to jury-charge errors presented in a
 Ambrose - 13

motion for new trial,” and remanding to the court of appeals “in light of our opinion in Igo”).

We, therefore, disagree with Ambrose’s contention that Igo is inapplicable here.

In addition to suggesting that Igo is distinguishable, Ambrose implicitly suggests that

Igo was wrongly decided, at least insofar as it would apply to a situation in which the trial

court made findings of fact and conclusions of law in support of its new-trial ruling.

Specifically, she suggests that applying the Almanza egregious-harm standard on appeal

under these circumstances is inconsistent with the rule that an appellate court must defer to

a trial court’s factual findings made in support of its ruling, particularly when those findings

involve evaluations of witness credibility and demeanor. See, e.g., Guzman v. State, 955

S.W.2d 85, 89 (Tex. Crim. App. 1997). As we explain further below in our analysis of

Ambrose’s second ground for review, we do not find any inherent inconsistency between an

appellate court’s application of the Almanza egregious-harm standard under these

circumstances and the rule that an appellate court owes deference to a trial court’s factual

findings. The problem with Ambrose’s assertion in this regard is that a trial court’s

egregious-harm determination is not a purely factual matter; rather, it is a mixed question of

law and fact to which deference is afforded only some of the time. See id. (explaining that

rule of deference applies to trial court’s resolution of mixed question of law and fact only

when the resolution of that ultimate question “turns on an evaluation of credibility and

demeanor”). Finding that Ambrose has presented us with no compelling reason to revisit our

holding in Igo on the basis of any conflict between the egregious-harm standard and the rule
 Ambrose - 14

requiring deference to a trial court’s factual findings, we decline to do so. We hold that Igo

is controlling here, and on that basis, we conclude that the court of appeals properly applied

the Almanza egregious-harm standard on appeal. Therefore, Ambrose’s first ground for

review alleging that the court of appeals erred by applying the Almanza standard is overruled.

C. Deference to Trial Court Not Required on Conclusion About Egregious Harm

Ambrose’s second ground for review contends that, in deciding whether there is

egregious harm in a case in which that question was considered by the trial court and the trial

court made findings of fact and conclusions of law addressing it, appellate courts should

defer to a trial court’s factual findings. In support, she cites cases indicating that a reviewing

court is required to defer to the trial court’s factual findings that are supported by the record.

See, e.g., Oregon v. Kennedy, 456 U.S. 667 (1982); Ex parte Wheeler, 203 S.W.3d 317, 325-

26 (Tex. Crim. App. 2006). In general, we agree with Ambrose to the extent that a question

of fact or a mixed question of law and fact that turns on an evaluation of credibility and

demeanor should be reviewed with deference. Guzman, 955 S.W.2d at 87-89 (explaining

that a reviewing court “should afford almost total deference to a trial court’s determination

of the historical facts that the record supports especially when the trial court’s findings are

based on an evaluation of credibility and demeanor”; deference is also warranted as to trial

court’s resolution of mixed questions of law and fact that “turn[ ] on an evaluation of

credibility and demeanor”). However, as we explain more fully below, the ultimate question

of egregious harm in this case is a mixed question of law and fact that does not turn on an
 Ambrose - 15

evaluation of credibility and demeanor; thus, contrary to Ambrose’s suggestion, the appellate

court was not required to defer to the trial court’s egregious-harm determination.5

An appellate court owes no deference to a trial court’s determinations when they are

actually determinations as to questions of law or mixed questions of law and fact that do not

turn on an evaluation of credibility and demeanor. See Martinez v. State, 348 S.W.3d 919,

923 (Tex. Crim. App. 2011) (holding that a de novo standard of review applied to “mixed

questions that do not depend on credibility determinations”); see also Loserth v. State, 963

S.W.2d 770, 772 (Tex. Crim. App. 1998) (holding that whether an in-court identification has

been tainted is a mixed question of law and fact that does not turn on an evaluation of

credibility and therefore is subject to de novo review). Applying that rule here, we conclude

that the court of appeals properly declined to defer to the trial court’s determination that

Ambrose was egregiously harmed by the absence of the accomplice-witness instruction.

Despite Ambrose’s characterization of the trial court’s determination of egregious harm as

a factual finding, it was actually a mixed question of law and fact that did not turn on an

evaluation of witness credibility and demeanor in this case. We also note that, regardless of

how a trial court labels its findings of fact and conclusions of law, an appellate court must

5
Because we conclude here that the question of whether there is egregious harm is a mixed
question of law and fact that does not turn on an evaluation of witness credibility and demeanor, we
decide that the appellate court was not required to defer to the trial court’s egregious-harm
determination in this case. We do not reach the more difficult question of whether a determination
of egregious harm could ever be a mixed question of law and fact that does turn on an evaluation of
credibility and demeanor, and, if so, whether under those circumstances the appellate court would
be bound to defer to some or all of the trial court’s factual findings or its conclusions of law that turn
on an evaluation of credibility and demeanor.
 Ambrose - 16

examine the substance of the findings and conclusions and treat them by their substance

rather than by their label. See State v. Sheppard, 271 S.W.3d 281, 291-92 (Tex. Crim. App.

2013).

The trial court made its determination of egregious harm in its “Findings of Fact and

Conclusions of Law.” The document consisted of two subheadings, each with several

subparagraphs. Under the first subheading, “Findings of Fact,” the trial court listed what

Ramirez testified about, that Ramirez had initially been charged with official oppression, that

the State had provided Ramirez with transactional immunity, that Ramirez was an accomplice

as a matter of law, and that Ambrose had not requested an accomplice-witness instruction.

The trial court did not make any factual determinations related to the credibility or demeanor

of Ramirez or any of the other witnesses. Under the second subheading, “Conclusions of

Law,” the trial court summarized the law regarding who qualifies as an accomplice, when

the trial court is required to provide the accomplice-witness instruction, and it indicated that

the defendant must show egregious harm when the omission of the instruction is unobjected

to. In the final subparagraph, the document stated, “This Court was present for all the

testimony in this case, has reviewed the legal arguments presented by the State and Defense,

and has reviewed the law on this issue. This Court concludes that the failure to include the

jury instruction on the accomplice witness rule caused Ms. Ambrose egregious harm, thus

requiring a new trial.” Here, there were no factual findings or credibility determinations as

to which a reviewing court should defer. Rather, here the purported factual findings were
 Ambrose - 17

in actuality mixed questions of law and fact that did not turn on the credibility of the

evidence. We, therefore, hold that the trial court’s determination that Ambrose suffered

egregious harm is a legal conclusion to which no deference is required by the appellate court.

Martinez, 348 S.W.3d at 923. Therefore, we overrule Ambrose’s second ground of error.

III. Analysis of Whether Record Demonstrates Egregious Harm

Ambrose’s third ground for review alleges that the court of appeals misapplied the

Almanza standard in its analysis of whether the omission of the accomplice-witness

instruction egregiously harmed Ambrose. The court of appeals held that the trial court’s

conclusion that Ambrose was egregiously harmed was in error. We agree that Ambrose was

not egregiously harmed, although we reach that conclusion through slightly different

reasoning than that employed by the court of appeals.

“Jury-charge error is egregiously harmful if it affects the very basis of the case,

deprives the defendant of a valuable right, or vitally affects a defensive theory.” Marshall v.

State, 479 S.W.3d 840, 843 (Tex. Crim. App. 2016) (citation omitted). Under the relevant

standard, we have traditionally considered (1) the entirety of the jury charge, (2) the state of

the evidence, (3) counsel’s arguments, and (4) any other relevant information revealed by the

entire trial record. Id. “[C]ourts are required to review the relevant portions of the entire

record to determine whether [a defendant] suffered actual harm, as opposed to theoretical

harm, as a result of the error.” Id. (citing Arline v. State, 721 S.W.2d 348, 351-52 (Tex.

Crim. App. 1986)). “Under the egregious harm standard, the omission of an accomplice
 Ambrose - 18

witness instruction is generally harmless unless the corroborating (non-accomplice) evidence

is ‘so unconvincing in fact as to render the State’s overall case for conviction clearly and

significantly less persuasive.’” Herron v. State, 86 S.W.3d 621, 632 (Tex. Crim. App. 2002)

(quoting Saunders v. State, 817 S.W.2d 688, 692 (Tex. Crim. App. 1991)). In assessing the

strength of the non-accomplice evidence, we examine (1) its reliability or believability, and

(2) the strength of its tendency to connect the defendant to the crime. Id.

In determining whether the omission of the accomplice-witness instruction

egregiously harmed Ambrose, the court of appeals conducted its harm analysis by examining

the non-accomplice evidence for corroboration of Ramirez’s assertion that Ambrose intended

for the other students to strike A.N. The court of appeals determined that “the only contested

element of the charged offense was whether Ambrose intended to subject A.N. to being

struck by other students,” and it therefore focused its analysis on “evidence tending to

connect Ambrose’s intent to the intent alleged in the charge—that Ambrose intended that

other students strike A.N.” Ambrose, 457 S.W.3d at 160-61. However, the appropriate focus

of a harm analysis in this case is to examine the non-accomplice evidence for corroboration

“tending to connect” Ambrose to the commission of the offense. Casanova v. State, 383

S.W.3d 530, 539 (Tex. Crim. App. 2012). The corroboration requirement in Article 38.14

does not apply separately to each element of the offense charged or to each aspect of the

accomplice’s testimony. If that were so, there would be no value in the testimony at all. See

Holladay v. State, 709 S.W.2d 194, 199 (Tex. Crim. App. 1986) (“It is clearly unnecessary
 Ambrose - 19

that the accomplice should be confirmed in every circumstance which he details in evidence;

for there would be no occasion to use him at all as a witness if his narrative could be

completely proved by other evidence free from all suspicion.”). In Vasquez v. State, this

Court observed that Article 38.14, by its very terms, requires only that there “be some non-

accomplice evidence tending to connect the defendant to the crime, not to every element of

the crime.” 56 S.W.3d 46, 48 (Tex. Crim. App. 2001) (holding that Article 38.14 does not

require corroboration of accomplice-witness testimony regarding a deadly-weapon finding).

In Joubert v. State, this Court similarly explained that the “corroborating evidence under

38.14 need not be sufficient, standing alone, to prove beyond a reasonable doubt that a

defendant committed the offense”; “[t]here need be only some non-accomplice evidence

tending to connect the defendant to the crime, not to every element of the crime.” 235 S.W.3d

729, 731 (Tex. Crim. App. 2007) (holding that Article 38.14 does not require corroboration

as to defendant’s role in an offense as a principal or as a party). Thus, our analysis to

determine whether Ambrose suffered egregious harm from the omission of the accomplice-

witness instruction is distinct from that of the court of appeals because we do not focus on

the intent element, but rather look only for evidence tending to connect Ambrose to the

charged offense.

Because there was strong corroborative evidence tending to connect Ambrose to the

offense, the totality of the record fails to show that she was egregiously harmed by the

omission of the accomplice-witness instruction. Casanova, 383 S.W.3d at 539 (“As the
 Ambrose - 20

strength of the corroborating evidence increases, however, a reviewing court may no longer

be able to declare that the lack of an accomplice-witness instruction resulted in egregious

harm.”). In this case, the record has strong corroborative evidence of Ramirez’s testimony

from the testimony by principal Large, assistant-principal Spellmann, and Ambrose herself.

Spellmann testified, “[Ambrose] let us know she had instructed her students to hit the student

on the arm, but not too hard so that the student would know how it felt to be bullied.” Large

testified, “[Ambrose] told us that she instructed the students to hit the other student but not

too hard.” Spellmann and Large each recounted that Ambrose stated that two or three

students hit A.N. after Ambrose had instructed them to hit A.N. Furthermore, by her own

testimony, Ambrose asked, “Does anybody want to show him what it feels like [to be

bullied]?” Also according to Ambrose’s own testimony, after she asked that question, a child

struck A.N. In light of the testimony from Large, Spellmann, and Ambrose herself that she

either instructed or asked the students to hit A.N. and that A.N. was hit one or more times

after that, there is an adequate amount of non-accomplice testimony that tends to connect

Ambrose to the offense of official oppression. Unlike the court of appeals’s analysis, we do

not additionally determine whether there is an adequate amount of non-accomplice testimony

that tends to connect Ambrose to the specific element of her intent to commit the offense

because it is unnecessary to corroborate every element of an offense for the evidence to be

adequate. Although we reach the same conclusion as the court of appeals by a slightly

different analysis, we agree with that court that the totality of the record fails to show
 Ambrose - 21

egregious harm from the omission of the accomplice-witness instruction. We overrule

Ambrose’s third ground for review.

IV. Conclusion

We hold that the court of appeals properly applied the Almanza egregious-harm

standard in this case in which the jury-charge error was raised in a motion for new trial. We

also hold that the court of appeals properly declined to defer to the trial court’s fact findings

in this case because those findings were in actuality matters of law or mixed questions of law

and fact that did not turn on the credibility of the evidence or demeanor of the witnesses. We

further hold that the court of appeals correctly concluded that Ambrose did not suffer

egregious harm from the omission of the accomplice-witness instruction. Therefore, we

overrule all three of Ambrose’s grounds for review and affirm the judgment of the court of

appeals.

Delivered: April 27, 2016

Publish