Larry Jean Hart v. the State of Texas

In the Court of Criminal
Appeals of Texas
══════════
No. PD-0677-22
══════════

LARRY JEAN HART,
Appellant
v.

THE STATE OF TEXAS

═══════════════════════════════════════
On Appellant’s Petition for Discretionary Review
From the Fifth Court of Appeals
Dallas County
═══════════════════════════════════════

YEARY, J., filed a dissenting opinion in which KELLER, P.J., and
KEEL, J., joined.

The Court’s opinion abandons our usual posture of deference to a
trial court’s broad discretion in admitting or excluding evidence. Clearly
the Court disagrees with the trial court’s decision. But the Court’s
HART – 2

opinion fails to demonstrate that the trial court’s decision was outside
“the zone of reasonable disagreement.” Montgomery v. State, 810 S.W.2d
372, 391 (Tex. Crim. App. 1990) (op. on reh’g). That is because it can’t—
because the trial court’s decision wasn’t outside that zone.
In Montgomery, the Court explained that, when reviewing a trial
court’s decision to admit or exclude evidence under Rule 403 of the Texas
Rules of Evidence, an “appellate court should not conduct a de novo
review of the record with a view to making a wholly independent
judgment whether the probative value of evidence . . . is substantially
outweighed by the danger of unfair prejudice.” Id. at 392. Instead,
Appellate courts “should reverse the judgment of the trial court ‘rarely
and only after a clear abuse of discretion.’” Id. (quoting United States v.
Maggitt, 784 F.2d 590, 597 (5th Cir. 1986)). But the Court dispenses
with both of those admonishments today. Although it pays lip-service to
the abuse of discretion standard, in effect, the Court performs a de novo
review of the trial court record and concludes that, because it would not
have admitted the evidence at issue, the trial court abused its discretion
to do otherwise—and it was error for the court of appeals to affirm the
trial court’s decision.
I am convinced the court of appeals correctly upheld the trial
court’s decision to admit Appellant’s rap videos and Facebook posts over
Appellant’s Rule 403 objection. At the very least, the trial court’s
decision fell well within the “zone of reasonable disagreement.” 1 Id. at

1 The Court concludes that the trial court’s admission of the rap videos

alone warrants reversal of the court of appeals’ judgment in affirming that
decision. However, under the Montgomery factors, I would hold that admission
HART – 3

391. Both the State’s decision to offer the evidence and the trial court’s
decision to admit it were fair responses to Appellant’s decision to
introduce a fact question about whether he was too naïve to have
formulated the mental state required to commit the offense alleged. I
would therefore affirm the judgment of the court of appeals.
I. THE MONTGOMERY FACTORS
As the Court’s opinion notes, in Montgomery, we articulated
factors that a trial court should consider in deciding whether the
probative value of a piece of evidence is “substantially outweighed by a
danger of . . . unfair prejudice[.]” TEX. R. EVID. 403; Montgomery, 810
S.W.2d at 389–91. Those factors include: (1) the “inherent
probativeness” of the evidence, (2) the evidence’s “potential . . . to
impress the jury in some irrational but nevertheless indelible way[,]” (3)
the time required by the proponent to develop the evidence, and (4) the
proponent’s need for the evidence. Id. Let us focus here, then, on those
factors identified by the Court in Montgomery.
A. Inherent Probative Value
The “inherent probative force” of an item of evidence refers to
“how strongly it serves to make more or less probable the existence of a
fact of consequence to the litigation[.]” Gigliobianco v. State, 210 S.W.3d
637, 641 (Tex. Crim. App. 2006). Addressing this factor, the court of
appeals observed that the “primary disputed issue at trial” was
Appellant’s intent to participate in a capital murder. Hart v. State, No.
05-19-01394-CR, 2022 WL 3754537 at *8 (Tex. App.—Dallas Aug. 30,

of both the rap videos and Appellant’s Facebook posts fell well within the zone
of reasonable disagreement.
HART – 4

2022) (mem. op., not designated for publication). In light of his claimed
naïveté and lack of language comprehension skills, the issue crystalized
into whether those conditions prevented him from forming the required
mental state, or mens rea. Id.
Citing the State’s appellate brief, the court of appeals noted that
the rap evidence, and the rap videos particularly, demonstrated
Appellant’s “ability to easily communicate with words” and his
“familiarity with criminal subject matter.” Id. Similarly, in contrast to
his claims that he did not own or have a gun and that he saw no one
with guns on the night of the murder, Appellant’s Facebook posts
suggested a familiarity with guns, an understanding of illicit uses of
them, and an awareness of the need to hide them and to not “get
kaught.” 2 The court of appeals correctly decided that the “trial court

2In addition to the two rap videos, which the Court describes, see
Majority Opinion at 4–6, the State introduced four posts from Appellant’s
Facebook page under the name “Block FrBndz Hart” as follows:

• State’s Exhibit 80: “You know I draw down you draw attention
slime.”

• State’s Exhibit 81: “Pull-up with them straps on me like Steve
Urkel!!”

• State’s Exhibit 82: “[G]otta hide the blicky.”

• State’s Exhibit 83: “Best advice I kan give my lill niggas dont get
kaught[.]”

The State failed to ask Appellant if “Block FrBndz Hart” was his account but
had earlier asked Appellant, “Is there a reason why you have Block on all your
social media?” Appellant responded: “It’s just a name I came up with.”
The Court asserts that State’s Exhibits 80, 82, and 83 are lyrics by other
rappers. Id. at 7 n.4. The Court may be right, but it errs to cite facts outside
HART – 5

reasonably could have concluded that the probative value of the evidence
[was] high.” Id. I agree with the court of appeals: the probative value of
the evidence with respect to Appellant’s friendliness, naïveté, and
sophistication was high enough that the trial court could have, without
difficulty, reasonably concluded that its value was not substantially
outweighed by any risk of unfair prejudice. And the Court does not
dispute the court of appeals’ judgment with respect to this factor. See
Majority Opinion at 14. So far, so good.
B. Improper Basis of Decision
The court of appeals observed that, while “the [rap] evidence did
have [some] potential to impress the jury ‘in some irrational but
nevertheless indelible way,’ [citation omitted] we cannot say the trial
court’s balancing determination was a ‘clear abuse of discretion.’” Hart,
No. 05-19-01394 at *8 (quoting Montgomery, 810 S.W.2d at 390). It
seems to me that the court of appeals was focused on the right things.
The logical probative force of the evidence at issue in this case
was that Appellant’s willingness and ability to lip-sync, post, and write
(or at least learn) rap lyrics about crime made it more likely that, when
he was asked to be the driver in a burglary, he was actually aware that
he was agreeing to facilitate a burglary and should have anticipated the
resulting capital murder. The risk of unfair prejudice posed by this
evidence was that the jury might conclude that Appellant’s mere
familiarity with the criminal subject matter of the lyrics made it more
likely that he participated in this crime—as a straightforward character-

the record in support of that assertion. See Johnson v. State, 624 S.W.3d 579,
585 (Tex. Crim. App. 2021) (“An appellate court cannot consider an item [of
evidence] that is not a part of the record on appeal.”).
HART – 6

conformity inference. Counterbalancing that risk, however, was: (1)
defense counsel’s opportunity to rebut any impermissible inference
during Appellant’s re-direct examination, (2) his opportunity to address
the same in closing argument, and (3) his opportunity to request a
limiting jury instruction—which Appellant did not do here, but could
have if he had truly thought it was necessary.
The Court sees things differently. In its view, “this factor
weigh[ed] heavily in favor of exclusion.” Majority Opinion at 19. In
reaching that conclusion the Court appears to largely adopt Appellant’s
argument that rap lyrics are so inherently inflammatory that, absent
some factual nexus between the lyrics and the crime alleged, the risk of
unfair prejudice will necessarily outweigh the probative value of the
lyrics as evidence. Id. at 15–19; Appellant’s Brief at 30. This approach
is flawed for a number of reasons.
First, the Court’s arguments suggest a fundamental
misunderstanding of the reason the State sought to admit the rap
evidence in this case. In response to Appellant’s testimony on direct
examination, the State argued, outside of the presence of the jury, that
it had a right to rebut Appellant’s own evidence suggesting that he was
just friendly and unsophisticated. Specifically, it argued:
Your Honor, we believe that the Defendant in his
testimony opened the door to character evidence. He
previously testified that he’s a friendly person.
We also believe that he opened the door to his level
of sophistication. He said that he is not good with
literature. Specifically, we know him to be a rapper where
he does fluently form sentences and phrases and frequently
writes raps. And we believe that that directly relates to his
ability to understand what people are communicating to
HART – 7

him and form his own opinions about things.

The State did not offer the rap evidence in this case to prove
Appellant was a criminal because he sang rap songs about crime. It
offered that evidence only to rebut Appellant’s evidence—evidence that
Appellant was not compelled to offer—that he was so naïve about and
unfamiliar with crime, and so cognitively and linguistically limited, that
he could not have intended to participate in a burglary on the night of
the murder. Put simply, the rap evidence offered by the State here is
just the kind of rebuttal evidence that Rule 404(a)(2)(A) explicitly
permits. TEX. R. EVID. 404(a)(2)(A) (“In a criminal case, a defendant may
offer evidence of the defendant’s pertinent trait, and if the evidence is
admitted, the prosecutor may offer evidence to rebut it.”).
Contrary to what the Court seems to argue, the probative value
of this kind of evidence—Rule 404(a)(2)(A) rebuttal evidence—does not
depend on any factual nexus between the rap lyrics and Michael
Gardner’s murder. See Majority Opinion at 15–19. 3 Whether Appellant
was rapping about true events is beside the point. The point is that the
lyrics tended to show that Appellant had the ability to communicate and
understand sufficiently to form the intent to participate in the crime at
issue.

3As Professor Imwinkelried has explained, at least as a general
proposition when assessing the admissibility of extraneous offenses under Rule
404(b), “[t]he test should be logical relevance rather than similarity. The better
view is that the judge should demand proof of similarity only if the proponent’s
theory of logical relevance assumes similarity.” 1 Edward J. Imwinkelried,
Uncharged Misconduct Evidence § 2:13 (2023). I see no reason why the same
considerations should not apply with equal force to the admissibility of rebuttal
evidence under Rule 404(a)(2)(A).
HART – 8

The courts that have held rap lyrics inadmissible have done so
when those lyrics were offered to prove “motive, opportunity, intent,”
etc., under Rule 404(b)(2). See, e.g., United States v. Bey, No. CR 16-290,
2017 WL 1547006 at 1* (E.D. Pa. Apr. 28, 2017) (“The government
moves to admit this music video under Rule 404(b)(2) to prove
knowledge, intent, and absence of mistake[.]”); 4 TEX. R. EVID. 404(b)(2).
But this case is not like those cases. To illustrate, if the State had sought
to admit the rap lyrics at issue during its own case-in-chief to prove that,
because Appellant sang about gun violence, he intended to participate
in the murder of Michael Gardner, then requiring a closer link between
the actions the lyrics described and the facts of the crime would make
some sense. But when, as here, the evidence is only offered during cross-
examination as trait-rebuttal evidence under Rule 404(a)(2)(A), any
similarity (or lack of similarity) between the rap lyrics and the facts
surrounding the charged crime is irrelevant. Consequently, the fact that
the trial court did not require the State to prove some factual similarity

4 See also Baker v. State, 899 S.E.2d 139, 141 (Ga. 2024) (State
introduced “a 33-second-long portion of a [rap music] video” to establish
Baker’s identity, motive, and opportunity with respect to malice murder
charge); United States v. Gamory, 635 F.3d 480, 488 (11th Cir. 2011)
(Government introduced rap video to prove identity and Gamory’s connection
to co-conspirators in federal drug and money-laundering charges); State v.
Skinner, 95 A.3d 236, 238 (N.J. 2014) (“[T]he State maintained that the [rap]
lyrics helped to demonstrate defendant’s ‘motive and intent’ in connection with
the offense because the rap lyrics addressed a street culture of violence and
retribution that fit with the State’s view of defendant’s role in the attempted
murder.”); Commonwealth. v. Gray, 978 N.E.2d 543, 560 (Mass. 2012) (“The
lyrics show no connection to the defendant that would suggest they were
biographical or otherwise indicative of his own motive or intent at the time of
the shooting.”).
HART – 9

between the rap lyrics and the crime at issue does not suggest that the
trial court abused its discretion.
Moreover, I reject the Court’s apparent conclusion that because
courts in other jurisdictions, and on different facts, have held rap music
unfairly prejudicial, the trial court’s decision to admit the rap evidence
in this case was “so clearly wrong as to lie outside the zone within which
reasonable people might disagree.” Taylor v. State, 268 S.W.3d 571, 579
(Tex. Crim. App. 2008). A Rule 403 balancing test is necessarily a case-
by-case analysis. Thus, it is not enough to say that, “because any song
that glorifies criminality . . . is inherently prejudicial[,]” it was an abuse
of discretion to admit the songs at issue here. Majority Opinion at 18.
Our rules of evidence require the exclusion of relevant evidence only if
the danger of unfair prejudice substantially outweighs its probative
value. And as Judge Keel astutely points out, this rebuttal evidence was
certainly less inflammatory than the facts of the crime alleged—and so
did not seriously risk luring the jury into finding Appellant guilty on an
improper basis. Dissenting Opinion of Judge Keel at 2.
In summary, I remain unpersuaded that the videos and lyrics the
jury saw and heard in this case were so inherently inflammatory that
they threatened to “‘to lure the factfinder into declaring guilt on a
ground different from proof specific to the offense charged.’” Manning v.
State, 114 S.W.3d 922, 928 (Tex. Crim. App. 2003) (quoting Old Chief v.
United States, 519 U.S. 172, 180 (1997)). I have greater faith in our
juries, and I am convinced that the court of appeals properly concluded
that the trial court acted within its discretion to leave it to the jury to
“evaluate the evidence’s probative force.” Hart, No. 05-19-01394 at *8.
HART – 10

C. Time Needed to Develop the Evidence
The third Montgomery factor, as the Court correctly notes, “looks
to the time the proponent will need to develop the evidence, during
which the jury will be distracted from consideration of the indicted
offense.” State v. Mechler, 153 S.W.3d 435, 441 (Tex. Crim. App. 2005);
see Montgomery, 810 S.W.2d at 390; Majority Opinion at 14. But as the
Court in Mechler pointed out, when the evidence under consideration
“relate[s] directly to the charged offense, a jury [will] not be distracted
away from the charged offense regardless of the required time to
present” the evidence. 153 S.W.3d at 441.
The rebuttal evidence presented by the State, here, consisted of
two videos, each less than five minutes long (and of the second video—
“Off Days”—the record indicates that the State presented only
Appellant’s own, approximately 45-second, solo), and four one-sentence
Facebook posts. By the Court’s own calculation, of the fifteen pages of
the record that make up Appellant’s cross-examination, less than half of
them are devoted to the rap evidence. Majority Opinion at 15. And of the
forty-five pages of testimony by Appellant, less than a third of those
pages are devoted to the rap evidence. Id. Even on its own terms, then,
I cannot agree with the Court that the rap evidence consumed “an
inordinate amount” of trial-time. Id.
Moreover, it is far from clear to me that the number of pages of
Appellant’s testimony spent on the rap evidence is the appropriate
denominator. See id. In fact, it is probably not. The Court should instead
focus on the entire trial, at least that part of the trial involving the
presentation of evidence relating to the question of Appellant’s guilt. By
HART – 11

my count, the presentation of the evidence relating to guilt consumed
three volumes of the reporter’s record, and those volumes contained
roughly 265 pages of testimony. Of those 265 pages, only nineteen,
according to the Court, were devoted to the rap evidence. See id. As the
court of appeals observed, the State devoted much more time to
establishing the facts of the murder and investigation, “including the
video of the premises on the night of the offense, police officers’
testimony of the crime scene, the search for the car used in the offense,
the medical examiner’s testimony, and appellant’s recorded interview
with police.” Hart, No. 05-19-01394 at *9.
And even more critically, this factor did not weigh in favor of
excluding the rap evidence, regardless of the time required to develop it,
because it related directly to the charged offense. See Mechler, 153
S.W.3d at 441. Appellant’s primary defensive theory at trial was his
contention that his naïveté and difficulties with language prevented him
from forming the intent to participate in a burglary or from anticipating
the resulting murder. If the rap evidence was probative of those issues—
and even the Court admits that at least the rap videos were—then the
time used by the State in presenting its rebuttal evidence was not a
distraction at all; rather, it was time devoted to one of the critical issues
at trial. See Majority Opinion at 14–15. Or, at least the trial court could
reasonably have thought so. Consequently, I cannot agree with the
Court that this factor weighed in favor of exclusion.
D. State’s Need for the Evidence
This Court has said that, “[w]hen the proponent has other
compelling or undisputed evidence to establish the proposition or fact
HART – 12

that the extraneous misconduct goes to prove, the misconduct evidence
will weigh far less than it otherwise might in the probative-versus-
prejudicial balance.” Montgomery, 810 S.W.2d at 390. However, as the
court of appeals observed, before the State proffered the video and
Facebook evidence, the jurors had only heard Appellant’s own testimony
from which he would have had its members infer his traits of
friendliness, naïveté, and language comprehension difficulties. Hart,
No. 05-19-01394-CR at *8. Beyond the substance of his testimony, the
jury had also observed Appellant frequently ask his counsel to repeat
questions and Appellant’s apparent difficulty in answering—which was
evidently so pronounced that, at one point, the trial judge excused the
jury to inquire into Appellant’s competence. Id. 5 Thus, the court of
appeals reasonably concluded that the trial court was justified in finding
the State’s need for evidence “was considerable[.]” Id.
Here, the Court suggests five alternative pieces of evidence the
State could have used to prove Appellant’s mental state at the time of
the offense:
• “Appellant’s statement that Mondo told him they were
going to ‘break in’ to his uncle’s house[;]”

5 During Appellant’s direct examination, defense counsel asked
Appellant, “[Y]ou understand that this trial is about what—what it looks like
some people did that you gave a ride to. Do you understand that?” When
Appellant responded by asking, “Can you repeat that?” the trial court excused
the jury to have Appellant evaluated for competency. Appellant was evaluated
by Dr. Lisa Clayton, a forensic psychiatrist, who testified outside the presence
of the jury that Appellant was competent but of “below average” IQ. When the
defense’s direct examination of Appellant resumed, Appellant testified that he
understood that he was on trial because the people he had transported had
killed someone.
HART – 13

• “inconsistencies or ‘evasiveness’ in the same
statement[;]” 6

• “surveillance footage of Appellant pulling the car
around the parking lot prior to letting the other
individuals out of the car[;]”

• “Appellant’s internet-search activity in the days
following the murder[;]” and

• “the fact that Appellant subsequently visited the
apartments again despite being purportedly ‘scared to
death’ at the time of the shooting.”

Majority Opinion at 19. But each of these pieces of evidence was either
equally consistent with Appellant’s defensive theory of the case or
occurred after the crime.
First, Appellant’s statement that Mondo told him they were going
to break into his uncle’s home was a key piece of evidence of Appellant’s
guilt, and one that the State used. But Appellant also testified that he
had not believed that Mondo had meant what he had said. Thus, the
State needed other evidence probative of Appellant’s mental state to
rebut that notion after it was injected into the trial by Appellant.
That Appellant gave inconsistent or evasive testimony about
Mondo’s statement does seem to support the inference that Appellant
was lying when he said he did not believe Mondo meant to break into

6 It is unclear, here, whether the Court’s allusion to Appellant’s “same

statement” refers to his police interview or his trial testimony. In either case,
the point remains the same—Appellant’s apparent evasiveness or
inconsistency in either setting could imply that he was being untruthful. But
it could also imply that he cannot communicate clearly or understand what
others say to him, as he claimed at trial.
HART – 14

anyone’s home. It is equally consistent, however, with Appellant’s
defensive theory of the case—namely, that Appellant cannot effectively
communicate or comprehend what others say to him. Thus, the State,
again, needed other evidence to rebut that testimony. And the trial court
could have reasonably concluded that the State’s need for the additional
evidence it proffered to prove Appellant’s mental state, was, as the court
of appeals put it, “considerable[.]” Hart, No. 05-19-01394-CR at *8.
Similarly, the surveillance footage showing that the car Appellant
drove circled the parking lot of the victim’s apartment complex before
parking down the street was evidence of Appellant’s guilt. But it was not
“compelling or undisputed evidence” that Appellant knew at the time of
the incident that he was participating in a crime. Appellant, after all,
might have circled the lot merely at the direction of the others in his
vehicle—which, again, would be equally consistent with his defensive
theory that he was unwittingly lured into serving as a getaway driver
and did not know what he was actually doing.
Further, Appellant’s actions after the incident—including his
internet searches and returning to the apartment complex where he
encountered Mondo—do not necessarily bear on his mental state at the
time of the incident. Appellant’s internet search history proves only that
he learned about the murder sometime after the fact; it does not prove
that he knew he was acting as a driver for a burglary at the time of the
incident. Likewise, Appellant might have returned to the apartment
complex where, according to his testimony, he was approached by
Mondo, for any number of reasons. Some of those reasons might imply
that Appellant was aware that he was acting as a getaway driver, but
HART – 15

others of those reasons would be consistent with Appellant’s defensive
theory of the case.
Finally, Appellant testified on direct examination that he was
“scared to death” on the night of the incident because of traffic speeding
past his vehicle as he idled on the street—undercutting the implication
that he was scared because he knew that his companions were
perpetrating a crime while he waited for them. 7 For that reason, the
State, once again, needed other evidence to rebut Appellant’s suggestion
that he could not have formed the required mental state at the relevant
point in time.
The Court has failed to identify other compelling or undisputed
evidence that the State could have used to rebut Appellant’s evidence
that he had not intended to participate in a crime with his companions.

7 On direct examination Appellant had the following exchange with his

defense counsel:

Q. You remember talking about how another car zipped past you?

A. Yes, sir.

Q. So was that—did that have any connection do you think, with the
people that you had taken?

A. No. No, sir.

Q. Well, you remember the questioning about it on the video as far
as you saying that it scared you?

A. Yes, sir.

Q. So if it wasn’t connected, why would it scare you?

A. It just was like a car was so close to me—well, it was close with
cars flying that same day. Yes, sir.
HART – 16

Consequently, I cannot agree with the Court that this factor weighs in
favor of exclusion.
E. Weighing the Factors
Considering all of these factors together, I am convinced that the
court of appeals was correct in its assessment: at most, only the
improper-basis factor might have weighed in favor of excluding the rap
lyric and Facebook evidence at issue in this case. The trial court acted
well within its appropriately broad discretion in deciding to admit the
evidence when it did so. See Montgomery, 810 S.W.2d at 389 (“trial
courts should favor admission in close cases”). And the court of appeals
was correct to conclude that the trial court’s decision to admit the State’s
rebuttal evidence was firmly within the “zone of reasonable
disagreement.” Id. at 391.
I personally find the Court’s arguments for reaching the opposite
conclusion today utterly unpersuasive. More troublingly, it seems to me
that the Court has simply concluded that, because it would not have
admitted the evidence at issue had it stood in the trial court’s place, the
trial court ipso facto abused its discretion in doing otherwise. I cannot
agree that, as a purely objective matter, the “zone” within which
“reasonable” minds could “disagree” is nearly so narrowly
circumscribed.
II. CONCLUSION
I would affirm the judgment of the court of appeals. The Court
does otherwise without proper regard for our role as a reviewing court
HART – 17

and on the basis of a flawed Rule 403 analysis. 8 I dissent.

FILED: May 8, 2024
PUBLISH

8 Ordinarily, the Court does not determine questions of harm in the first

instance, absent a decision on harm in the court of appeals. Holder v. State,
639 S.W.3d 704, 707 (Tex. Crim. App. 2022). “We ‘[n]ormally’ only do so when
the error is so ‘plainly harmless’ that principles of judicial economy support
resolving it ourselves.” Id. (quoting Johnston v. State, 145 S.W.3d 215, 224
(Tex. Crim. App. 2004)). Today the Court conducts a harm analysis for the first
time on discretionary review without attempting to explain why that is
appropriate. It is at least not clear to me that any harm from the “error” the
court finds would be “plain.” See Jordan v. State, 593 S.W.3d 340, 353 (Tex.
Crim. App. 2020) (Yeary, J., dissenting) (“[T]he question of harm seems, at
least, debatable—not, as the Court would have it, so self-evident as to obviate
our usual practice to remand the cause for that analysis.”). I dissent on that
basis as well.