Opinion Dissenting to Denial of En Banc Reconsideration Issued April 22, 2025
In The
Court of Appeals For The
First District of Texas ———————————— NO. 01-23-00229-CR ——————————— DANIEL TOOMBS, Appellant V. THE STATE OF TEXAS, Appellee
On Appeal from the 149th District Court Brazoria County, Texas Trial Court Case No. 96462-CR
OPINION DISSENTING TO DENIAL OF EN BANC RECONSIDERATION
The panel has reversed a criminal conviction by holding that inflammatory
evidence and arguments by the prosecutors rendered the jury incapable of following
the trial court’s instructions. But the evidence was not inflammatory and the
prosecutors’ arguments actually asked the jury to follow the instructions. The panel’s result—abandoning the almost ironclad appellate presumption that juries follow
their instructions—is a departure from the standard announced by the Court of
Criminal Appeals and the recent practice of this Court. We should grant en banc
reconsideration and bring the result of this case in line with similar cases. Because
the Court chooses not to do so, I respectfully dissent from denial of en banc
reconsideration.
I. The complained-of evidence showed nothing more than that the appellant twice went to washaterias.
The second most remarkable thing about the panel opinion is that it refutes
itself. In the merits section of the opinion, the panel concludes that the evidence of
a supposed extraneous act showed nothing more than “that [the appellant] was at a
public laundromat at some point.” Slip op. at 26. But in the harm section of the
opinion, the panel holds that this evidence “inflame[d]” the jurors to the point where
they could not follow the trial court’s instructions. One of those two things might be
correct, but they cannot both be.
I think the merits section has the better of the debate. The panel’s
doctrine-of-chances discussion makes it easy to get distracted by all the evidence the
parties discussed outside the presence of the jury. But the only evidence that was
admitted for the jury was a CODIS report stating there was “a moderate stringency
association” between DNA found on the pants related to the charged offense, DNA
found in “stain[s] on floor” at two burglary investigations about a year before the
2 charged offense, and a DNA sample from the appellant. A police officer read this
report to the jury, and on cross-examination the officer testified that the two
burglaries were at washaterias.
And that’s it. There was no testimony to the jury of what to make of this
evidence. The jury was not told what “a moderate stringency association” meant.
The jury was not told that the DNA at the washateria came from blood—which might
have suggested the appellant cut himself breaking into things. All the evidence
showed was “that [the appellant] was at a public laundromat at some point.” Slip op.
at 26. More precisely, that he was at two washaterias at some points, but the merits
section’s description accurately captures the import of the evidence: Not much.
II. The trial court’s limiting instruction should have rendered the complained-of error harmless.
But then comes the harm section to explain why this marginal evidence
requires reversal. That should be a hard row to hoe for two reasons. First, this is
non-constitutional error meaning the panel would be obliged to “disregard” the error
unless it had a substantial effect on the jury’s verdict. See TEX. R. APP. P. 44.2(b);
King v. State, 953 S.W.2d 266, 271 (Tex. Crim. App. 1997). In light of the strong
evidence of guilt here, it’s doubtful the fact that the appellant went to two
washaterias had much effect on the verdict.
Second, the trial court gave a limiting instruction to the jury on how to
consider extraneous-act evidence. Just before the evidence came in, the trial court
3 told the jury it could not consider evidence of an extraneous offense unless it
believed, beyond a reasonable doubt, the appellant committed the offense. It also
told the jury the lawful purposes for which it could consider extraneous-offense
evidence:
In reference to the evidence that is about to be admitted, if any, that the defendant has previously participated in recent transactions or acts other than but similar to that which is charged in the indictment in this case, you are instructed that you cannot consider such transactions or acts, if any, for any purpose unless you find and believe beyond a reasonable doubt that the defendant participated in such transactions or committed such acts, if any, and even then you may only consider the same for the purposes of determining intent or knowledge or identity or motive or common plan or scheme, if it does, and for no other purpose.
The jury charge contained a similar instruction:
You cannot consider testimony of extraneous offenses for any purposes unless you find and believe beyond a reasonable doubt that the defendant committed any such extraneous offense. Even then you may only consider the same as it relates to [the motive of the defendant OR the opportunity of the defendant OR preparation of the defendant OR the plan of the defendant OR knowledge of the defendant OR the identity of the defendant OR absence of mistake by the defendant OR lack of accident by the defendant OR the rebuttal of a defensive theory OR same-transaction contextual evidence], if any, in connection with the offense alleged against the defendant in the indictment in this case and for no other purpose.
If the panel had engaged in the ordinary appellate presumption that juries
follow their instructions, it would have concluded that the jury disregarded the
washateria evidence and any arguments by the State that the appellant committed
4 the washateria burglaries because the evidence came nowhere near proving the
offenses beyond a reasonable doubt.
Alternatively, the panel would have concluded that the jury used the presence
of the appellant’s DNA at two washaterias for the purposes listed in these
instructions and not for character conformity. Character conformity is the only
forbidden inference from extraneous-act evidence. See De La Paz v. State, 279
S.W.3d 336, 343 (Tex. Crim. App. 2009).
Of course, if the evidence had relevance for any of the reasons listed in the
trial court’s instruction, that would mean that the panel was wrong that the evidence
was inadmissible under Rule 404(b). But I’m not relitigating the merits here because
the simplest resolution to this issue is to use the ordinary appellate presumption that
juries follow their instructions, which means the jury disregarded this evidence or at
least did not consider it for an improper purpose.
III. The panel disregarded the trial court’s limiting instruction by exaggerating the “inflammatory” nature of the evidence.
Now this gets us to the most remarkable thing about the panel’s opinion: It
declined to engage in the ordinary appellate presumption that juries follow their
instructions. It did so in a single paragraph that obscures what a big deal this is.
Nor could the trial court’s limiting instruction cure this error. A limiting instruction is not sufficient to cure error when the testimony “is clearly calculated to inflame the minds of the jury and is of such a character to suggest the impossibility of withdrawing the impression produced on the jurors’ minds.” Castillo v. State, 865 S.W.2d 89, 93–94 (Tex.
5 App.—Corpus Christi-Edinburg 1993, no pet.). Considering the “inherently prejudicial” nature of extraneous offense evidence, McGee v. State, 725 S.W.2d 362, 366 (Tex. App.—Houston [14th Dist.] 1987, no pet.), coupled with the State’s emphasis of the laundromat burglaries during closing arguments and its remark that “the DNA doesn’t lie,” the trial court’s limiting instruction was incapable of “withdrawing the impression produced on the jurors’ minds.” Castillo, 865 S.W.2d at 93– 94.
Slip op. at 29-30.
Just three sentences earlier the panel had said the complained-of evidence
“failed to establish [the appellant] had any involvement in the laundromat burglaries,
let alone that he committed them.” Slip op. at 29. But now the same evidence is
“inherently prejudicial” “extraneous offense evidence” that was “clearly calculated
to inflame the minds of the jury.”
Beyond this jarring inconsistency, an appellate court holding that a jury could
not follow the trial court’s instructions is extraordinary. It is common for trial courts
to instruct juries on how to assess evidence. Large areas of the law—evidentiary
rulings, claims of charge error, motions for mistrial, legal sufficiency review in civil
cases—fall apart if appellate courts do not presume juries follow their instructions.
In the criminal law, appellate courts have followed this presumption when
trial courts instructed juries regarding outrageous extraneous acts. See, e.g., Paster
v. State, 701 S.W.2d 843, 848 (Tex. Crim. App. 1985) (instruction cured harm from
testimony implicating capital murder defendant in two other murders); Herrero v.
State, 124 S.W.3d 827, 836 (Tex. App.—Houston [14th Dist.] 2003, no
6 pet.) (instruction cured harm from testimony that witness was testifying in return for
State’s protection against defendant’s death threats); Whitaker v. State, 977 S.W.2d
595, 600 (Tex. Crim. App. 1998) (instruction cured harm where witness testified
capital murder defendant “beat her up before giving her an engagement ring”);
Burleson v. State, No. 01-11-00866-CR, 2013 WL 772947, at *8-10 (Tex. App.—
Houston [1st Dist.] Feb. 28, 2013, pet. ref’d) (mem. op., not designated for
publication) (instruction, if requested, would have cured harm from testimony that
murder defendant set a church on fire). Going to washaterias is less inflammatory
than any of that.
IV. The panel focused on the State’s jury argument in a way that is illogical and not obviously consistent with case law.
To justify disregarding the instruction the panel focused on the State’s jury
argument. The way it did so is both illogical and counter to the record.
First, I disagree with the panel’s claim the State “emphas[ized]” the
washateria evidence. The State’s arguments take up 513 lines in the reporter’s
record, 18 of which (3.5%) refer to the washateria evidence. Those references came
in the middle of the State’s arguments, not the beginning or end where advocates
typically put subjects they want to emphasize.
More importantly, as a matter of logic the State’s arguments do not justify
concluding the jury disregarded the limiting instruction. In its opening argument the
State said the washateria evidence didn’t prove the appellant was guilty, but it was
7 “definitely suspicious.” In its closing argument, the State argued the appellant’s
DNA being present at the washateria countered his defensive theory that it was just
a coincidence his DNA was in the pants worn during the charged offense.
These arguments were not inflammatory nor did they ask the jury to disregard
the trial court’s instructions. Cf. Stahl v. State, 749 S.W.2d 826, 832 (Tex. Crim.
App. 1988) (holding prosecutor undid curative effect of instruction to disregard by
bringing up disregarded matter during argument). Indeed, these arguments urged the
jury to consider the evidence for reasons consistent with the instructions. The panel
did not explain the mechanism by which the State’s argument caused the jury to
disregard the trial court’s instruction, nor did the panel cite precedent showing this
was an appropriate factor to consider for this issue. If a party argues that evidence
has logical significance a reviewing court doesn’t see, that’s not a reason to believe
the jury couldn’t follow its instructions.
V. The panel used a different standard than this Court has used in other, recent cases.
The final problem with the panel’s decision to hold the jury disregarded a
limiting instruction is that the panel may have used the wrong standard. I say “may”
because there are two lines of case law pronouncing different approaches for such
an analysis: the intuitive approach and the evidence-based approach.
The panel followed the line of cases using the intuitive approach. By asking
itself whether the evidence is “clearly calculated to inflame the minds of the jury”—
8 or something similar—an appellate court is listening to its own intuition on the
matter.1 That is, the appellate court is not assessing whether the jury actually
disregarded the trial court’s instructions. Instead, the appellate court is determining
whether it thinks the evidence crosses some inflammatory threshold such that the
appellate court will disregard the trial court’s instructions.
When a case involves an instruction to disregard or a limiting instruction, the
intuitive approach gives the appellate court a great deal of leeway to do what it thinks
is right. First the appellate court asks itself whether it thinks the disregarded or
limited evidence inflamed the jurors’ minds. If the appellate court says “yes,” then
it moves on to ask whether the evidence in question is enough to meet the standard
for reversal. In most cases—like this one—the answer to the second question will
also be “yes.” After all, the reason trial courts instruct juries to disregard evidence
or limit consideration of evidence is that the evidence has real danger of harming the
defendant. If an appellate court chooses to ignore an instruction to disregard or a
limiting instruction, reversal is very likely.
1 For examples of the intuitive approach, with mixed results, see e.g., Ballard v. State, 537 S.W.3d 517, 525 (Tex. App.—Houston [1st Dist.] 2017, pet. ref’d); Gunter v. State, 914 S.W.2d 647, 648 (Tex. App.—Texarkana 1995, no pet.); Barefield v. State, 784 S.W.2d 38, 44 (Tex. Crim. App. 1989); Robinette v. State, 816 S.W.2d 817, 819 (Tex. App.—Eastland 1991, no pet.); Russell v. State, 113 S.W.3d 530, 544–45 (Tex. App.—Fort Worth 2003, pet. ref’d).
9 The other line of cases uses an evidence-based approach. The leading cases
for this approach are Colburn v. State, 966 S.W.2d 511 (Tex. Crim. App. 1998) and
Thrift v. State, 176 S.W.3d 221 (Tex. Crim. App. 2005).
In Colburn, the court explained there was a legal “presumption” that juries
followed their instructions, and this presumption would prevail unless there was
actual evidence in the record that showed the jury did not follow its instructions. 966
S.W.2d at 520. Thrift expounded on this presumption a bit more. Thrift was charged
with sexual assault of a child and indecency with a child. 176 S.W.3d at 221. The
State admitted four “pictures of nude males” found in Thrift’s bedroom. Id. at 222.
The trial court instructed the jury it could consider the pictures only for determining
Thrift’s intent to gratify his sexual desire for purposes of the indecency charge. Id.
The Tenth Court held the pictures were inadmissible and reversed the indecency
conviction. Thrift v. State, 134 S.W.3d 475, 483 (Tex. App.—Waco 2004) (op. on
orig. submission and op. denying reh’g). The Tenth Court affirmed the sexual assault
charge because, using the intuitive approach, it believed the jury followed the part
of the limiting instruction forbidding it from considering the pictures for the sexual
assault charge. Id. at 479, 483.
10 The Court of Criminal Appeals granted review only on Thrift’s claim that
there was a “spillover effect” that tainted the sexual assault conviction.2 176 S.W.3d
at 221. Thrift’s claim was that the jury considered the pictures for both offenses in
violation of its instruction. Id. at 222–23. Citing Colburn, the Court held that it would
begin with the rebuttable presumption that juries follow their instructions, then “the
appellant must rebut the presumption by pointing to evidence that the jury failed to
2 It’s important to note how narrow the grant of review was, because there are glaring problems with the Tenth Court’s opinion that weren’t reviewed. I will discuss these problems because they flow directly from the Tenth Court’s use of the intuitive approach. An important basis for the Tenth Court’s holding, on original submission, that the evidence violated Rule 403 was its conclusion that the jury could not follow the trial court’s limiting instruction. Thrift v. State, 134 S.W.3d 475, 478, 482 (Tex. App.— Waco 2004) (op. on orig. submission and op. denying reh’g). The Tenth Court used the intuitive approach and, in one sentence, held the jury couldn’t follow the limiting instruction because the “graphic images were of a nature that a limiting instruction would not likely have been effective.” Id. at 479. As Chief Justice Gray noted, this conclusory statement wasn’t much of an analysis. Id. at 480 (Gray, C.J., concurring and dissenting on orig. submission). From the Tenth Court’s description, there were four pictures, each showing a man with an erection, not engaged in sexual activity. The testimony in that case involved Thrift sexually abusing a fourteen-year-old male in multiple ways. The pictures were less “graphic” than the testimony. By my lights, it would be a peculiar juror who could maintain an objective approach to graphic testimony about the sexual abuse of a child but then be driven to irrational lawlessness by four images of softcore pornography. This ruling highlights the subjective nature of the intuitive approach—the results often hinge on the sensitivities of the appellate court. Thrift raised his “spillover” claim on rehearing. The Tenth Court rejected that claim based in part on the limiting instruction. Id. at 483 (op. denying reh’g). Thus the Tenth Court reversed one conviction because it held the jury couldn’t follow an instruction, but it affirmed the other conviction because it held the jury could follow the exact same instruction about the exact same evidence. This inconsistency is a stark example of how the intuitive approach can produce inconsistent and unpredictable results.
11 follow the trial court’s instructions.” Id. at 224. Because there was no evidence that
rebutted the presumption, the Court affirmed.
The evidence-based approach is the rule in civil cases. See Golden Eagle
Archery, Inc. v. Jackson, 116 S.W.3d 757, 771 (Tex. 2003) (“Unless the record
demonstrates otherwise, an appellate court must presume that the jury followed [the
trial court’s] instructions.”). If we presume juries follow instructions in civil cases, I
see no reason we should not give the same respect to juries in criminal ones.
There are several advantages to the evidence-based approach. First, it
addresses the actual question: Did the jury follow its instructions? In contrast, the
intuitive approach asks: Does the appellate court wish to disregard the instruction?
The evidence-based approach will consistently produce answers that are grounded
in the record. The intuitive approach gives the appellate court opportunities to
reverse convictions based on sympathy for a defendant or distaste for the
prosecution’s trial tactics.
Some cases that use the intuitive approach will purport to use a “presumption”
that the jury followed its instructions. But as Colburn and Thrift point out, when a
court actually uses a legal presumption it takes actual evidence to overcome the
presumption. Instead, the intuitive approach states a legal rule, not a presumption.
When a court asks itself whether certain evidence inflamed the minds of the jury,
12 there’s no presumption one way or the other. The intuitive approach is a standard for
when appellate courts will disregard instructions, regardless of whether the jury did.
I can’t find a case reconciling these two approaches, or even recognizing they
are distinct. The reason for that is likely because the standard for disregarding a jury
instruction, under either approach, is so high that it hardly ever happens. It’s a rare
appellate decision that disregards a jury instruction based on evidence that would
meet the standard for one approach but not the other.
But I don’t see how an intermediate appellate court could reverse under the
intuitive approach without addressing the evidence-based approach. The
Colburn/Thrift presumption is mandated by the Court of Criminal Appeals.
This Court has had no problem applying the Colburn/Thrift presumption in
other cases. This Court has cited that presumption dozens of times,3 including in an
opinion issued less than a year before the panel opinion here. The result of that case
is inconsistent with the result here.
In Coronado v. State, the defendant complained about the admission of
extraneous-act evidence: He touched a girl “underneath her underwear when she was
3 See, e.g., Milton v. State, 695 S.W.3d 689, 719 (Tex. App.—Houston [1st Dist.] 2024, pet. granted); Jones v. State, 264 S.W.3d 26, 30 (Tex. App.—Houston [1st Dist.] 2007, pet. ref’d); Jones v. State, No. 01-18-01079-CR, 2021 WL 3358011, at *6 (Tex. App.—Houston [1st Dist.] Aug. 3, 2021, no pet.) (mem. op., not designated for publication); Lacer v. State, No. 01-17-00267-CR, 2018 WL 4135025, at *14 (Tex. App.—Houston [1st Dist.] Aug. 30, 2018, no pet.) (mem. op., not designated for publication).
13 seven or eight.” No. 01-22-00237-CR, 2024 WL 1098238, at *14 (Tex. App.—
Houston [1st Dist.] Mar. 14, 2024, pet. ref’d) (mem. op., not designated for
publication).
The Coronado court assumed, without deciding, the evidence was
inadmissible but concluded any error would not warrant reversal. Id. at *15–16. In
its harm analysis, it factored in a limiting instruction the trial court gave the jury on
how to assess the extraneous-act evidence. Even though the extraneous-act evidence
in Coronado was an order of magnitude more inflammatory than the extraneous-act
evidence here, and even though the prosecutor in Coronado referenced the
extraneous-act evidence in argument just as the prosecutor did here, the Coronado
court employed the Colburn/Thrift presumption that the jury followed its limiting
instruction. Id. at *16.
In another case from this Court, released two weeks after the panel opinion
issued here, a different panel applied the Colburn/Thrift presumption to an
instruction to disregard some very inflammatory evidence in Fisher v. State, No.
01-22-00328-CR, 2024 WL 5248877, at *6–7 (Tex. App.—Houston [1st Dist.] Dec.
31, 2024, no pet. h.) (mem. op., not designated for publication). Fisher was a capital
murder. A witness named Batiste gave “lengthy” testimony about the circumstances
of the charged offense and the defendant’s confession to it. See id. at *6. On
cross-examination, Batiste invoked her Fifth Amendment right against
14 self-incrimination and the trial court instructed the jury to disregard all her
testimony. Id. at *5, 7. The defendant moved for a mistrial, which the trial court
denied. Id. at *5.
On appeal, Fisher complained about the denied mistrial. Id. In assessing the
alleged harm from Batiste’s testimony, the Fisher court began by stating the
Colburn/Thrift presumption was mandatory: “We must presume the jury follows the
trial court’s instructions.” Id. at *7 (emphasis added). The Fisher court described
Batiste’s testimony as “highly prejudicial,” but it affirmed because the record did
not contain evidence that rebutted the Colburn/Thrift presumption. Id. at *7, 8.
In light of Coronado and Fisher—not to mention the dozens of other cases
applying the Colburn/Thrift presumption—the panel’s failure to use the
Colburn/Thrift presumption is striking. As a court we ought to apply the same
standard to a legal issue each time it comes up so that we produce consistent,
predictable results. The panel’s approach was not consistent with this Court’s
caselaw, nor was its result predictable.
But even applying the panel’s intuitive approach, only the most inflammatory
evidence should have caused the panel to disregard the trial court’s instructions.
Coronado and Fisher are more data points, if any were needed, showing that
appellate courts do not lightly hold that juries disregarded instructions. If this Court
can believe that a jury disregarded testimony that the defendant confessed, and if
15 this Court can believe a jury limited how it considered testimony that a defendant
stuck his hand into a little girl’s underwear, this Court should believe a jury can
follow an instruction to limit consideration of whether a defendant twice went to
washaterias.
Neither the intuitive nor evidence-based approach should lead to abandoning
the appellate presumption that the jury here followed its instructions. We should
1) grant rehearing, 2) in the absence of evidence to the contrary, presume the jury
disregarded the washateria evidence because the trial court instructed it to disregard
any extraneous offense not proven beyond a reasonable doubt, and 3) affirm the trial
court’s judgment because, factoring in the instruction, any error here did not affect
the appellant’s substantial rights. Because the Court instead chooses to let this
inappropriate reversal stand, I respectfully dissent from the Court’s denial of en banc
Clint Morgan Justice
Morgan, J. dissenting from the denial of en banc reconsideration, joined by Guiney, J.
Publish. TEX. R. APP. P. 47.2(b).