AMERICAN HONDA MOTOR CO INC v. SFI 59 LP (2025)

OPINION

American Honda Motor Co., Inc., appeals from the judgment rendered on a jury verdict holding it liable for fire damage caused by a manufacturing defect in one of its vehicles. Although we disagree with American Honda's arguments that it conclusively proved it did not manufacture the vehicle and that there is legally and factually insufficient evidence that a manufacturing defect existed and was a producing cause of the fire, we agree that the trial court reversibly erred in allowing the plaintiff's expert to testify to a previously undisclosed opinion and the factual basis for that opinion. We accordingly reverse the judgment and remand the case for a new trial.

I. Background

On June 14, 2020, Jianning Tao drove his six-month-old 2020 Honda CR-V from South Dakota to Houston, arriving at 5:30 or 6:00 p.m. on June 15th. He parked at Briarwood Apartments, where he was visiting family. About ten hours after he parked the vehicle, an apartment resident saw flames coming from the vehicle's engine compartment. The resident called 911 at around 3:21 a.m., and by the time the fire department arrived, the fire had spread to two other vehicles and the side of the apartment building.

Investigators representing the Houston Arson Bureau, Briarwood, the insurers of two of the damaged vehicles, and American Honda all investigated the fire's cause. American Honda's investigators concluded that the fire originated from an external source at the front corner of the vehicle on the passenger's side. The Houston Arson Bureau found no evidence of an external source and concluded that the fire originated in the vehicle's engine compartment. The investigators for Briarwood and the vehicles’ insurers determined that the fire began in the vehicle's engine compartment forward of the battery on the driver's side.

Briarwood sued American Honda for damage to the apartment complex's property and alleged that the fire was caused by a manufacturing defect—that is, a deviation, in construction or quality, from the product's specifications or planned outputs in a way that makes the product unreasonably dangerous.¹ To prevail on a manufacturing-defect claim against one of the product's manufacturers, a plaintiff must prove that the product was defective when it left the manufacturer, and that the defect was a producing cause of the plaintiff's injuries.² American Honda disputed that it manufactured the vehicle; that a defect was present when the vehicle left the manufacturer; and that a defect in the vehicle caused the fire.

Prior to trial, Briarwood's expert Nestor Camara testified at his deposition that the fire originated at the bus ³ of the vehicle's power-distribution box (the PDB, also referred to as the fuse box) on the driver's side of the engine compartment, but he was unable to identify a specific manufacturing defect. He stated that a “high-resistance connection” in the fuse box would have been capable of starting the fire, but he explained that he could not point out any such connection because “it's been destroyed” by the fire. Moreover, he affirmatively testified that he could not say how the vehicle deviated from its design.

At trial, however, and over American Honda's running objection, Camara testified for the first time that a photograph of the engine, available to him before his deposition, depicted an improper high-resistance connection in the fuse box. Camara identified this connection as the cause of the fire. Although he had not compared the connection to the vehicle's design specifications, Camara stated that there should not be a high-resistance connection in any fuse box. He concluded that the high-resistance connection must have been present when the vehicle left the manufacturer because the connection was in the bus of the sealed fuse box in the Honda CR-V's engine compartment, and there is no evidence that the fuse box had ever been unsealed.

In the charge, the jury was asked whether, when the Honda CR-V left American Honda's possession, it had a manufacturing defect that was a producing cause of the fire. The jury answered in the affirmative and found that the vehicle was 100% responsible for Briarwood's damages. The jury assessed damages of $322,423.66 for the cost of repairs to Briarwood's property and $203,005.71 for loss of use.

American Honda previously had moved unsuccessfully for summary judgment on the ground that it did not manufacture the vehicle and is therefore statutorily insulated from liability. American Honda maintained this position during trial, and after the jury returned its verdict, American Honda renewed this argument, as well as its objections to the admission of Camara's previously undisclosed opinion testimony. The trial court denied American Honda's motions for judgment notwithstanding the verdict and for a new trial and rendered judgment on the verdict.

On appeal, American Honda argues that we must reverse and render judgment in its favor because it conclusively established that it did not manufacture the vehicle, or because the evidence is legally insufficient to support the judgment. In the alternative, American Honda asks that we reverse the judgment and remand the case for a new trial on the ground that the evidence is factually insufficient to support the judgment, or because the trial court reversibly erred in permitting Camara to testify at trial to a previously undisclosed opinion and to the factual basis for that opinion.⁴

II. Statutory Immunity from Liability

A “manufacturer” is statutorily defined as one “who is a designer, formulator, constructor, rebuilder, fabricator, producer, compounder, processor, or assembler of any product or any component part thereof and who places the product or any component part thereof in the stream of commerce.” Tex. Civ. Prac. & Rem. Code § 82.001(4). A non-manufacturing defendant that merely sold or distributed the product has no product liability unless the plaintiff additionally proves that one of the specifically listed statutory exceptions to non-liability applies. See id. § 82.003.

In its motion for judgment notwithstanding the verdict, American Honda argued unsuccessfully that Briarwood had not proved that American Honda manufactured the vehicle and that the evidence instead conclusively established that American Honda only sold the vehicle. We review the denial of a motion for judgment notwithstanding the verdict by determining whether the evidence was legally sufficient to support the jury's verdict. City of Keller v. Wilson, 168 S.W.3d 802, 823 (Tex. 2005). To do so, we consider whether the evidence at trial, viewed in a light favorable to the verdict, would have enabled reasonable jurors to make the challenged finding. Am. Honda Motor Co. v. Milburn, 696 S.W.3d 612, 627 (Tex. 2024). We assume that jurors resolved all conflicts in the evidence in a manner that supports the verdict. Wilson, 168 S.W.3d at 820. We will sustain a legal-sufficiency challenge only when (a) there is a complete absence of evidence of a vital fact, (b) the court is barred by rules of law or of evidence from giving weight to the only evidence offered to prove a vital fact, (c) the evidence offered to prove a vital fact is no more than a mere scintilla, or (d) the evidence conclusively establishes the opposite of the vital fact. Id. at 810.

Here, however, there is no express finding that American Honda is one of the vehicle's manufacturers. Jurors instead were asked whether, when the subject vehicle left American Honda's possession, it had a manufacturing defect that was a producing cause of the fire. No party requested a finding that American Honda was among the vehicle's manufacturers or objected to the absence of such a question.

When, as here, a ground of recovery or defense consists of multiple elements, and an unrequested element was omitted from the charge without objection, a party may ask the trial court to make a written finding of the omitted element if there is factually sufficient evidence to support a finding. Tex. R. Civ. P. 279. In the absence of such a written finding, the omitted element is deemed found in support of the judgment. Id. American Honda did not request written findings of fact from the trial court but argues that the evidence conclusively establishes that American Honda was merely a seller, not a manufacturer, of the subject vehicle. Thus, in addition to its challenge to the legal sufficiency of the evidence, we understand American Honda to contend that the evidence is factually insufficient to support a deemed finding that it manufactured the vehicle.

When a party challenges the factual sufficiency of a finding for which the party did not have the burden of proof at trial, we review all of the evidence in a neutral light and will reverse the trial court's judgment only if the evidence supporting the finding is so contrary to the overwhelming weight of the evidence as to make the judgment clearly wrong and manifestly unjust. Maritime Overseas Corp. v. Ellis, 971 S.W.2d 402, 406–07 (Tex. 1998). We may not substitute our judgment for that of the factfinder, even if the evidence would clearly support a different result. Id. at 407. Finally, whether reviewing the evidence for legal or for factual sufficiency, we defer to the jury's reasonable credibility determinations. Wilson, 168 S.W.3d at 820; Sw. Bell Tel. Co. v. Garza, 164 S.W.3d 607, 625 (Tex. 2004).

American Honda contends that, as a matter of law, it was a non-manufacturing seller of the subject vehicle and is therefore insulated from liability. Briarwood responds that whether a defendant is a non-manufacturing seller is a mixed question of law and fact, and that the conflicting evidence on the subject is legally and factually sufficient to support a finding that American Honda manufactured the vehicle. We agree with Briarwood.

The determination whether a defendant was a manufacturer of a particular product presents a mixed question of law and fact. See Mega Child Care, Inc. v. Tex. Dep't of Protective & Regul. Servs., 29 S.W.3d 303, 309 (Tex. App.—Houston [14th Dist.] 2000, no pet.) (“An issue involves a mixed question of law and fact when a standard or measure has been fixed by law and the question is whether the person or conduct measures up to that standard.”). Although courts can determine the effect of a statute on undisputed facts as a matter of law,⁵ there is conflicting evidence about whether American Honda did or did not manufacture the vehicle in this case. The evidence is strong enough to support, but not to dictate, either result.

In support of its position that it is not among the vehicle's manufacturers, American Honda's corporate representative Eric Pederson testified that Honda CR-Vs were designed in Japan by Honda Motor Co., Ltd., and that all of Honda's design studios are in Los Angeles. When this vehicle was manufactured, Pederson worked for Honda Development & Manufacturing of America, LLC (HDMA), and Pederson testified that HDMA manufactured it. Pederson stated that American Honda merely distributes the vehicles, though he did not know if HDMA contracted with American Honda or if there was any official documentation of the relationship between the two companies.

There additionally is evidence that a “VIN sticker”⁶ on each Honda CR-V identifies the vehicle's manufacturer, and although the sticker from the subject vehicle was destroyed in the fire, exemplars from two other Honda CR-Vs stated that those vehicles were manufactured by HDMA.

In support of its position that American Honda is one of the subject vehicle's manufacturers, Briarwood relied solely on the vehicle's “Manufacturer's Certificate of Origin.” The certificate was issued by American Honda's vice president in Torrance, California. In it, American Honda described the vehicle and certified “that the new vehicle described above is the property of the said company, firm or corporation and is transferred on the above date and under the invoice number indicated” to the “distributor or dealer,” Vern Eide Motorcars, Inc., in South Dakota. American Honda's vice president further certified “that this was the first transfer of such new vehicle in ordinary trade and commerce.” Although the certificate does not expressly state that American Honda manufactured the vehicle, its name appears directly under the words “Manufacturer's Certificate of Origin” and its role is not otherwise identified. The companies that Pederson identified as the vehicle's designer and its manufacturer—Honda Motor Co., Ltd., and HDMA—are not mentioned at all.

American Honda argues that the manufacturer's certificate of origin is not evidence that it manufactured the vehicle, because under the version of a Texas regulation in effect when this certificate was issued in 2019, “[a] manufacturer's certificate of origin assigned by the manufacturer or the manufacturer's representative or distributor to the original purchaser is required for a new motor vehicle that is sold and offered for sale.”⁷ Even assuming that the regulations applicable to this manufacturer's certificate of origin are the same as those of Texas,⁸ the certificate appears on its face to identify American Honda, and only American Honda, as the manufacturer. Reasonable factfinders presented with this record could reach differing conclusions as to whether American Honda manufactured the vehicle.

After reviewing the conflicting evidence, and deferring to the jury's reasonable credibility determinations, we conclude that the evidence is both legally and factually sufficient to support the deemed finding that American Honda is a manufacturer of the vehicle. We overrule this issue.

III. Legal and Factual Sufficiency of the Manufacturing-Defect Evidence

American Honda next contends that the evidence that a manufacturing defect existed and was a producing cause of the fire is legally and factually insufficient because the testimony of Briarwood's expert Nestor Camara is unreliable. This is so, American Honda contends, because (a) Camara's testimony identifying a specific high-resistance connection as a manufacturing defect relies on a photograph that was taken after the fire and that does not depict the connection's condition when the vehicle left the manufacturer; (b) Camara's causation testimony is based on assumptions that vary materially from the undisputed facts; and (c) Camara did not review the manufacturer's design specifications, so there is no evidence that the connection he identified as the cause of the fire deviates from its design.⁹

To place the challenged testimony in context, it is undisputed that even when a vehicle is turned off, some amount of electrical current still flows from the battery to the fuse box to power certain components such as those necessary for keyless entry, or to keep the vehicle's clock running, or to maintain pre-set radio stations, etc. The amount of this current is referred to as the vehicle's “parasitic draw.” If the vehicle's parasitic draw drained the battery, the car would not start, so according to Pederson, Honda service manuals designate a vehicle's maximum parasitic draw, measured in amperes or milliamperes, “based on being able to leave your car at an airport for 30 days and having it start when you get back.” Pederson stated that the Honda CR-V's designated maximum parasitic draw is 27 milliamperes, but from his own experience working on vehicles, the actual parasitic draw of a normal Honda CR-V is only about 10 milliamperes. American Honda's fire-causation expert Hernan Mercado likewise testified that “27 would be the maximum that you want to have on,” and that “the parasitic draw on a normal Honda CR-V” is about 10 milliamperes when the ignition is off.

At least some of the wiring through which the vehicle's electrical current flows form connections to, or at, the fuse box through “spade connectors” in which a male spade is inserted into a female spade. Camara stated that in all but one of the spade connections he observed in the vehicle's fuse box, the male spades were fully inserted into the female spades. He identified the sole exception in a photograph. As he described it, the male and female spades shown in the photograph were not in the same plane, and rather than meeting end-to-end, they met at a thirty- or forty-degree angle, with the male spade only partially inserted into the female. Camara opined that this misaligned connection caused the fire.

Camara explained that such a partial connection reduces the surface area over which the electrical current flows, and the reduction in surface area increases the amperage drawn by that connection. He testified that this is called a high-resistance connection, and that high-resistance connections generate heat. The heat can increase so much that the connection glows, and a “glowing connection” is hot enough to ignite the materials around it. Camara opined that the high-resistance connection he identified is unreasonably dangerous and that such connections are not intended to be in “any fuse box, any vehicle or—or anything.” Camara concluded that this improper connection was present when the vehicle left the manufacturer because it was within the sealed fuse box and there was no evidence that the fuse box had been opened after leaving the manufacturer.

A. American Honda failed to preserve a complaint that the photo depicts the connection after the fire rather than when the vehicle left the manufacturer.

To preserve a complaint that an expert's “underlying methodology, technique, or foundational data” renders the expert's opinion unreliable, a party must timely object so that the trial court can evaluate the complaint and the proponent has “a fair opportunity to cure any deficit and thus prevent trial by ambush.” City of San Antonio v. Pollock, 284 S.W.3d 809, 817 (Tex. 2009). American Honda objected at trial that Camara's methods were unreliable because he relied on unsupported assumptions about the amount of current available to the fuse box while the vehicle was turned off, and because Camara did not review the design specifications. American Honda did not, however, object that the photograph on which Camara relied—his foundational data—did not depict the condition of the connection at issue when the vehicle left the manufacturer. Thus, this complaint has not been preserved for appellate review. See id.; see also Tex. R. App. P. 33.1(a) (a timely request, objection, or motion is generally one of the prerequisites to appellate review).

B. Camara's testimony did not rely on assumptions that materially vary from the undisputed facts.

American Honda points out that when reviewing both for reliability and legal sufficiency, the reviewing court “cannot consider only an expert's bare opinion, but must also consider contrary evidence showing it has no scientific basis.” Cooper Tire & Rubber Co. v. Mendez, 204 S.W.3d 797, 804 (Tex. 2006) (quoting Wilson, 168 S.W.3d at 813). “[I]f an expert's opinion is based on certain assumptions about the facts, we cannot disregard evidence showing those assumptions were unfounded.” Id. (quoting Wilson, 168 S.W.3d at 813) (alteration in original). American Honda contends that Camara's causation testimony is unreliable because he assumed that the electrical “current potential” to the fuse box was sufficient to start a fire, even though the vehicle had been turned off for hours. According to American Honda, it is undisputed that a current of 4 to 6 amperes—that is, 4,000 to 6,000 milliamperes—is needed to start a “glowing connection,” and that the vehicle when turned off drew no more than about 27 milliamps, which is about 0.25% of the amount required. American Honda contends that Camara's opinion that there was sufficient electrical current for the high-resistance connection to become a glowing connection and start a fire is contrary to the undisputed testimony of Pederson and Mercado that the Honda CR-V parasitic draw is only 10–27 milliamps.

But Pederson and Mercado testified about the “designated” maximum and “the parasitic draw on a normal Honda CR-V,” not one involving a high-resistance connection. There is legally sufficient evidence that the high-resistance connection could draw more than enough current to start a fire.

Camara testified that because the male spade was only partially inserted in the female spade he identified, the surface area over which electricity must pass was reduced, and a reduction in surface area increases the connection's resistance and its amperage, causing it to generate heat. According to Camara, when a high-resistance connection generates heat,“[i]n most cases, you end up with a fire.”

The reason why a high-resistance connection results in a fire “in most cases” was explained by Briarwood's other expert Daniel DellaSala. DellaSala testified there is “constant current going from the battery to the fuse box,” and that the current available in a vehicle that is turned off is “[w]hatever the battery amperage is.” Camara testified that “the car has a potential of about 405 cold-cranking amps.” Jurors reasonably could have understood the testimony to imply that a high-resistance connection would continue increasing in amperage up to the battery's limit of 405 amperes, which is about a hundred times larger than the minimum needed to start a glowing connection.

C. Camara's testimony that the high-resistance connection was a manufacturing defect was not unreliable, despite his failure to review the vehicle's design specifications.

In its remaining challenge to the reliability of Camara's testimony, American Honda points out that a manufacturing defect is, by definition, an unreasonably dangerous deviation from a product's specifications or planned outputs. Because Camara did not review the vehicle's design specifications, American Honda contends that his opinion about the existence of a manufacturing defect is unreliable. On this record, we disagree.

It is undisputed that the Honda CR-V's specifications do not include any high-resistance connections. Thus, any high-resistance connection necessarily would be a deviation from the vehicle's specifications. American Honda's corporate representative tacitly acknowledged this. Pederson testified that, before leaving the manufacturer, each Honda vehicle undergoes post-production testing at the “Vehicle Quality” area of the plant. Among other things, all of the vehicle's electrical circuits are tested for the presence of high-resistance connections using a “line end tester.” If a high-resistance connection is detected, the vehicle is sent for repairs and then returned to Vehicle Quality for retesting. Logically, the mere detection of a high-resistance connection would not be cause for repairs and retesting if such connections were there by design.¹⁰

From the testimony of American Honda's expert Hernan Mercado, it appears that high-resistance connections are unintended, for he defined a high-resistance connection as “a connection where you have more resistance than you should have.”¹¹ Like Camara, Mercado stated that such a connection dissipates heat, and when it is hot enough to glow, the connection can ignite the material around it.

We conclude that Camara's testimony is not unreliable. Although American Honda's expert opined that the fire began from an external source at the front of the passenger's side of the car, a jury considering all of the evidence—including Camara's testimony—reasonably could conclude that a high-resistance connection in the bus of the fuse box was an unreasonably dangerous deviation from the vehicle's specifications, and that the connection's high resistance caused its parasitic draw to increase to at least the 4–6 amperes necessary to start a glowing connection and ignite the surrounding material. Viewed in a neutral light, the evidence supporting the verdict is not so weak, and the contrary evidence is not so great, as to make the verdict manifestly unjust. We accordingly overrule American Honda's challenges to the legal and factual sufficiency of the evidence.

IV. Admission of Camara's Previously Undisclosed Opinion

In American Honda's final issue, it argues that the trial court reversibly erred in admitting Camara's testimony about a specific manufacturing defect because (a) he is unqualified to render an expert opinion on the existence of a manufacturing defect or on causation, and (b) he testified to a previously undisclosed opinion that unfairly surprised or unfairly prejudiced American Honda.¹²

A. Camara was qualified to testify to his expert opinion.

According to American Honda, Camara was not qualified to testify that a manufacturing defect caused the fire because he has never worked for a vehicle manufacturer or parts supplier, and he has not worked for an automotive dealer since the 1980s.¹³ We disagree that these facts rendered Camara unqualified to testify that the specific high-resistance connection he identified constituted a manufacturing defect and was a producing cause of the fire.

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the expert's specialized knowledge will help the factfinder to understand the evidence or decide a fact issue. Tex. R. Evid. 702. Camara could acquire these qualifications in ways other than through experience working for a vehicle manufacturer, dealer, or parts supplier.

Camara has such qualifications. He has a bachelor's degree in fire sciences, and he is a certified fire investigator with the Texas Commission on Fire Protection, a certified vehicle-fire investigator with the National Association of Fire Investigators, and a certified fire investigator with vehicle-fire credentials with the International Association of Arson Investigators. He also is a former volunteer fire fighter and fire marshal. In addition, he performed vehicle evaluations and troubleshooting of electrical equipment in vehicles for his father's company for four years. He has worked for the same employer for over thirty-five years, where he acts as a consultant in litigation-related fire analysis, including vehicle-fire analysis. In that capacity, Camara investigates three or four vehicle fires a month. He further disclosed that at the time of the trial, he was investigating three other vehicle fires for the law firm representing American Honda.

We conclude that Camara's education, training, and experience qualified him to offer his opinion identifying a high-resistance connection in the vehicle's fuse box, concluding that such a connection is a manufacturing defect, and determining that the defect caused the vehicle fire that damaged Briarwood's property.

B. The trial court reversibly erred in admitting Camara's previously undisclosed opinion testimony.

Lastly, American Honda argues that the trial court reversibly erred in permitting Camara to offer, over American Honda's objections, a previously undisclosed opinion identifying a specific improper connection between two wires as a manufacturing defect and the cause of the fire. We agree.

A party is required to amend or supplement the deposition testimony or written report of its retained expert regarding “the expert's mental impressions or opinions and the basis for them.” Tex. R. Civ. P. 195.6. Such amendment or supplementation is required if a party learns that information previously disclosed “was incomplete or incorrect when made, or, although complete and correct when made, is no longer complete and correct.” Tex. R. Civ. P. 193.5 An amended or supplemental response must be made reasonably promptly after the party discovers the need for it. Tex. R. Civ. P. 193.5(b). Courts presume that an amended or supplemental response made less than thirty days before trial was not made reasonably promptly. Id.

A party who fails to timely amend or supplement when required to do so may not introduce the undisclosed material or information in evidence. Tex. R. Civ. P. 193.6. This result can be avoided only if the party proves, and the trial court finds, that there was good cause for the failure to timely amend or supplement the information, or that the failure will not unfairly surprise or unfairly prejudice the other parties. Id. Although we review the trial court's ruling under Rule 193.6 for abuse of discretion,¹⁴ a finding of good cause or of the absence of unfair surprise and unfair prejudice must be supported by the record. Id.

1. Briarwood was required to amend or supplement the discovery responses and deposition testimony regarding Camara's opinions.

Briarwood contends that it was not required to amend or supplement the information it disclosed before trial about Camara's opinions because an expert “may change an opinion without supplementation if the opinion is an ‘expansion of an already disclosed subject.’ ” Republic Parking Sys. of Tex., Inc. v. Med. Towers, Ltd., No. 14-02-01141-CV, 2004 WL 2358315, at *10 (Tex. App.—Houston [14th Dist.] Oct. 21, 2004, pet. denied) (mem. op.) (quoting Navistar Int'l Transp. Corp. v. Crim Truck & Tractor Co., 883 S.W.2d 687, 691 (Tex. App.—Texarkana 1994, writ denied)). That is not the case here.

In his deposition, Camara testified that, based on fire patterns, the fire originated at the bus of the vehicle's PDB (fuse box). He stated, “it's got to be a manufacturing defect” because the fuse box is “a non-serviceable piece of equipment,” so “it had to come from the factory that way.” He explained that a loose connection between a female and male spade within the bus would be a high-resistance connection that “could make it heat.” But he could not identify any actual high-resistance connection or any other deviation from design:

Q [B]ecause you don't know how it's supposed to be designed, you can't tell us how it deviated from the design in this case, agreed?

A No.

Q No, you disagree with that?

A No, I can't tell you how it deviates from the design.

Q Okay. And you believe that there was high resistance in there. Can you point me to where that PDB that high resistance would have happened?

A At the bus.

Q Okay. At the bus. But no more specific than that, right?

A No, sir.

Q And so far as what two pieces made a connection or what distances were different than they should have been or what materials were used that shouldn't have been, do you have any of that information?

A No. It was at the bus.

Although there was evidence from the fire's pattern that it originated in the fuse box in the engine compartment, Camara could say no more at his deposition than that the fire was caused by “[h]eating of the bus,” though he was unaware of “any other incidents where a PDB overheats and results in a fire like this with the key out.” When specifically asked, “Is there a photograph you could show us,” Camara denied that there was.

Briarwood maintains that it had no duty to amend or supplement Camara's previous representations about his opinions and the bases for them because his trial testimony was merely an expansion of his previously disclosed opinions. In support of its position that Camara's opinion testimony at trial merely expanded on the opinions he disclosed at his deposition, Briarwood cites Camara's deposition testimony that “a loose connection could make [a connection at the bus] heat.” Briarwood also points out that American Honda's counsel acknowledged at a pretrial hearing that “[Camara's] opinion is that some malfunction, some unspecified defect in this fuse box is what caused the fire.” But the quoted statements instead illustrate why amendment or supplementation was mandatory.

To prevail on manufacturing-defect claim, “a specific defect must be identified by competent evidence and other possible causes must be ruled out.” Ford Motor Co. v. Ledesma, 242 S.W.3d 32, 42–43 (Tex. 2007) (quoting Nissan Motor Co. v. Armstrong, 145 S.W.3d 131, 137 (Tex. 2004)). But before trial, Camara testified that that it was not possible to identify a specific defect because fire “can consume the evidence” and “you can't pinpoint a point of origin because of the damage here.” Based on the information it possessed before trial, American Honda could argue that Briarwood had not met its burden. Certainly, there was no need for it to prepare its own expert to testify to a negative: that none of the more than 1,600 photographs depicted a manufacturing defect.

But at trial, Camara purported to do what he previously claimed was impossible. Rather than maintaining that the fire had destroyed the evidence, he instead singled out a single photograph, opining that it depicted the specific defect that caused the fire. Given Camara's prior testimony, American Honda could not have anticipated this reversal of Camara's position and the need to prepare a rebuttal of Camara's characterization of the photograph at issue.

Because this was a material alteration of Camara's testimony constituting a surprise attack, we conclude that the Texas Rules of Civil Procedure obligated Briarwood to timely disclose Camara's opinion purporting to identify a specific connection as a manufacturing defect and the cause of the fire. Although the trial court suggested that American Honda could cross-examine Camara on his failure to disclose his amended opinion earlier, such cross-examination could only let the jury hear more about evidence that it should not have heard at all.¹⁵ The trial court was required to exclude the testimony absent a finding either that Briarwood met its burden to prove that it had good cause for the non-disclosure, or that its non-disclosure neither unfairly surprised nor unfairly prejudiced American Honda.

Because it concluded that Briarwood was not required to disclose the alteration in Camara's opinions, the trial court did not address the exceptions to the automatic-exclusion rule. It did not make any of the required findings, and for the reasons described below, the record would not support any such findings.

2. Briarwood did not prove good cause for the non-disclosure.

When Camara was asked at trial why he had not previously identified the high-resistance connection to which he testified, he answered, “I identified it a long time ago. But the first time I relayed it to you was today.” He further agreed that the photograph on which he relied at trial was taken in 2020 and had been available for three-and-a-half years. He admitted that he had the photograph before his deposition. Briarwood accordingly does not contend it had good cause for failing to timely amend or supplement Camara's expert report and deposition testimony about his “mental impressions or opinions and the basis for them.”

3. Briarwood did not prove that its non-disclosure would not unfairly surprise or unfairly prejudice American Honda.

Briarwood did not prove that American Honda was neither unfairly surprised nor unfairly prejudiced by Briarwood's failure to timely disclose Camara's opinion identifying a specific connection as a manufacturing defect that caused the fire. To the contrary, the erroneously admitted evidence was the linchpin of Briarwood's case, and the failure to timely disclose it deprived American Honda of the opportunity to prepare rebuttal expert testimony.

Consider Briarwood's closing argument:

And [Camara] shows you how the -- the male and female are supposed to slide together.

But in this photograph right here, you can see with your own eyes that it doesn't. And it comes down. And then this is supposed to be flush, but it's not. It's perpendicular. And so that joint causes the high resistance connection.

And what's so important about that is: When Mr. Mercado [American Honda's expert] was on the stand, and Mr. Camara had already described this theory in his analysis and his photographic evidence of exactly what the defect was, they never showed Mr. Mercado do [sic] this photograph.

They never said: You know, Mr. Mercado, here are the photographs of this terminal that's incorrectly connected. Please explain to the jury why that's not a defect.

They never did that.

Why didn't they do that? Because the photograph is proof positive that this car was not designed nor intended to have that sort of connection underneath the fuse box leading into the fuse box.

Here's the photograph right there. And that's why the fire started. It's that simple.

We identified the defect, and they did not dispute it. They never asked their expert about that photo of the defective terminal condition—connection. And so they have offered no evidence to refute this specific nature of the defect that Mr. Camara pointed out for y'all, for the whole courtroom, in front of Mr. Mercado.

Camara alone purported to pinpoint a specific connection as a manufacturing defect that caused the fire. Other witnesses offered varying opinions about where the fire started, but not identifying exactly why it started. Stephen Garza, an investigator from the Houston Arson Bureau, concluded that the fire was accidental; that it began somewhere in the vehicle's engine compartment; and that there was no evidence of an external source, though he acknowledged that such evidence could have been consumed in the fire. But Garza did not determine the fire's point of origin and denied that he had any opinion about whether there were any defects in the electrical components in the vehicle's engine compartment. Daniel DellaSala, Briarwood's expert on the fire's point of origin, concluded that the fire started in the engine compartment on the driver's side, but he, too, had no opinion on the fire's cause. Hernan Mercado, American Honda's expert, concluded from fire patterns that the subject vehicle's fire began at the front right corner of the vehicle due to an unidentified “external source.” Thus, according to Mercado, something or someone external to the vehicle set it on fire, but Mercado identified neither the source nor the method. Camara was the only witness to point to evidence and say, “this ignited the fire.”

Briarwood's failure to timely disclose that opinion deprived American Honda of the opportunity to prepare rebuttal testimony. American Honda could not realistically have been expected to adequately review the photos and prepare expert rebuttal midtrial: the supplemental reporter's record alone includes over 1,600 photographs. But by the time it filed its motion for new trial, American Honda was able to argue that the photograph on which Camara relied—PX-34.147.3—does not show an improper connection, and that DX-27.157—a photo of the same area from a different angle—reveals that the male spade Camara identified in PX-34.147.3 is in fact properly connected to a different female spade, and the female spade that Camara referred to as the other part of the improper connection is simply hanging loose in the same area. Thus, American Honda could have offered a rebuttal, if had not been deprived of the opportunity to do so. Briarwood's unfair surprise unfairly prejudiced American Honda.¹⁶

4. The erroneous admission of Camara's challenged testimony was harmful.

Texas Rule of Civil Procedure 193.6 requires automatic exclusion of undisclosed expert testimony so as to promote responsible assessment of settlement, prevent trial by ambush, and give the opposing party the opportunity to prepare rebuttal expert testimony.¹⁷ But as with evidentiary rulings generally, the trial court's failure to exclude evidence when required to do so by Rule 193.6 constitutes reversible error only if the evidence's admission “probably caused the rendition of an improper judgment.” Jackson v. Takara, 675 S.W.3d 1, 6 (Tex. 2023) (per curiam) (quoting Tex. R. App. P. 44.1(a)(1)). The judgment must turn on the particular evidence admitted. Id. To make that determination, we review the entire record and consider the state of the evidence, the strength and weakness of the case, and the verdict. Id. at 7.

We conclude that the standard is met here. Camara stated that in all but one of the spade connections he observed in the vehicle's fuse box, the male spades were fully inserted into the female spades; thus, the case turned on Camara's previously undisclosed opinion that a particular photograph depicted the specific high-resistance connection that caused the fire. If Briarwood had timely disclosed this information, American Honda could have rebutted this testimony with photographic evidence that the high-resistance connection Camara identified did not exist because the two wires Camara characterized as forming the high-resistance connection did not connect to each other at all.

Briarwood argues that the admission of the challenged testimony was harmless because the circumstantial evidence was legally sufficient to prove a manufacturing defect.¹⁸ But Briarwood's premise is mistaken: the erroneous admission of evidence can constitute reversible error even if the remaining evidence is legally sufficient to support the verdict. For example, in Kia Motors Corp. v. Ruiz, the court found the evidence legally sufficient to support the jury's design-defect finding, but nevertheless held that the erroneous admission of “a spreadsheet summarizing authorized warranty claims involving air bags in similarly designed vehicles from the same manufacturer” probably caused the rendition of an improper verdict. 432 S.W.3d 865, 868 (Tex. 2014). The court pointed out that in a legal-sufficiency review, we disregard evidence contrary to the verdict, whereas we analyze the harm from an evidentiary error by reviewing the entire record. Id. at 884. We consider “the role the evidence played in the context of the trial and the efforts made by counsel to emphasize the erroneous evidence.” U-Haul Int'l, Inc. v. Waldrip, 380 S.W.3d 118, 136 (Tex. 2012). We also consider whether the evidence concerned a matter that “was hotly contested by competing expert testimony.” Ruiz, 432 S.W.3d at 884.

Whether a manufacturing defect existed, and if so, whether it caused the fire, were both hotly contested, and Camara's erroneously admitted testimony was the only evidence purporting to identify a specific manufacturing defect as the cause of the fire. It was the focal point of Briarwood's case, and as the excerpt from its closing argument illustrates, this evidence was emphasized in proportion to its importance.

Because Briarwood based its arguments for liability on Camara's previously undisclosed opinion, we conclude that its erroneous admission probably caused the rendition of an improper judgment. We sustain this issue, reverse the trial court's judgment, and remand the case for further proceedings.

V. Conclusion

The current record does not establish, as a matter of law, that American Honda is a non-manufacturing seller of the subject vehicle or that there is no evidence to support a finding that a manufacturing defect was a producing cause of the vehicle fire. On the other hand, the trial court reversibly erred in admitting evidence of Camara's previously undisclosed opinion that a particular photograph depicts a manufacturing defect that caused the fire. We therefore reverse the trial court's judgment and remand the case for further proceedings, which will allow all of these issues to be relitigated.

FOOTNOTES

1.   Ford Motor Co. v. Ridgway, 135 S.W.3d 598, 600 (Tex. 2004).

2.   Id.

3.   At trial, Camara explained that the “bus” is the front part of the fuse box “where everything is bolted together.” He specifically identified two wires from the battery to the fuse box, one wire from the fuse box to the alternator, and three fuses as part of the “bus.”

4.   We have reordered American Honda's issues because appellate courts generally address issues that could result in rendition of judgment before addressing issues that could result only in remand. Tex. R. App. P. 43.3; Bradleys’ Elec., Inc. v. Cigna Lloyds Ins. Co., 995 S.W.2d 675, 677 (Tex. 1999) (per curiam).

5.   Rudisill v. Arnold White & Durkee, P.C., 148 S.W.3d 556, 559 (Tex. App.—Houston [14th Dist.] 2004, no pet.).

6.   “VIN” is the acronym for “vehicle identification number.”

7.   40 Tex. Reg. 1096 (2015), as amended by 41 Tex. Reg. 5766 (2016) (emphasis added). The current version of the regulation expressly requires the “manufacturer's name on the face of the manufacturer's certificate of origin.” 43 Tex. Admin. Code § 217.5(a)(1)(A). Both then and now, the regulation describes evidence of vehicle ownership that must accompany an application to title a new vehicle in Texas and specifies that the manufacturer's certificate of origin must be in the form prescribed by the Texas Department of Transportation. Id.

8.   The manufacturer's certificate of origin at issue in this case was issued in California to a dealership in South Dakota, and the vehicle was never titled in Texas; thus, the Texas regulation cited by American Honda does not apply to the vehicle at issue. Nevertheless, absent evidence or argument to the contrary, Texas courts generally presume that the determinative law of another state is the same as that of Texas. Coca-Cola Co. v. Harmar Bottling Co., 218 S.W.3d 671, 685 (Tex. 2006).

9.   Although these arguments primarily appear in connection with American Honda's complaint that the trial court erred in admitting Camara's challenged opinion testimony because it was previously undisclosed, the arguments overlap with American Honda's challenges to the legal and factual sufficiency of the evidence. We address them in connection with American Honda's sufficiency-of-the evidence challenges, because if the evidence was legally or factually insufficient to support the verdict even with Camara's testimony, then the question of whether Briarwood should have disclosed Camara's challenged opinion earlier would be moot.

10.   Pederson stated that a vehicle's performance on the electrical-circuit tests are documented, but he had not seen any such documentation about the subject vehicle.

11.   Emphasis added.

12.   American Honda additionally argues that the trial court erred in admitting Camara's causation testimony because his theory is unreliable. For the reasons previously discussed in connection with Camara's legal and factual sufficiency challenges, we disagree.

13.   American Honda contends that Camara is unqualified for the additional reason that he did not review the vehicle's design specifications. As previously explained, Camara did not need to review the vehicle's specifications to know that they did not include a high-resistance connection.

14.   Jackson v. Takara, 675 S.W.3d 1, 6 (Tex. 2023) (per curiam).

15.   See Collins v. Collins, 904 S.W.2d 792, 802 n.12 (Tex. App.—Houston [1st Dist.] 1995) (en banc on reh'g) (“The purpose of the rule would be subverted if the only remedy for a violation was to attempt to cross-examine or impeach without the benefit of preparation and to present the case without rebuttal evidence.”), writ denied, 923 S.W.2d 569 (Tex. 1996) (per curiam) (clarifying that the intermediate appellate court was addressing the need to supplement an expert's deposition testimony). Although the rule referred to by the Collins court was Rule 193.6’s predecessor Texas Rule of Civil Procedure 166b, the court's observation is equally applicable to the present rule.

16.   Briarwood notes that American Honda did not seek a continuance to allow time to prepare a response to Camara's testimony, but American Honda correctly points out that it had no obligation to do so: the remedy for failure to timely disclose an expert opinion is automatic exclusion, not a continuance.

17.   SJF Forest Lane, LLC v. Phan, No. 05-22-00905-CV, 2024 WL 2796279, at *10 (Tex. App.—Dallas May 31, 2024, no pet.) (mem. op.); ETC Tex. Pipeline, Ltd. v. XTO Energy Inc., 698 S.W.3d 915, 927 (Tex. App.—Eastland 2024), review granted, judgment vacated, and remanded by agreement, No. 24-0987, 2025 WL 1416469 (Tex. May 16, 2025).

18.   Initially, Briarwood additionally argued that Houston Arson Bureau investigator Stephen Garza's testimony “independently establishes that an electrical or manufacturing defect in the CR-V was the cause of the fire” (emphasis added). Briarwood relied on Garza's report, and his testimony about that report, concluding that “the fire originated in the engine compartment area caused by an electrical and/or manufacturing failure.” But Briarwood later corrected this; Garza had actually written, and testified, that he attributed the fire to “an electrical and/or mechanical failure,” not an electrical or manufacturing failure.

Tracy Christopher, Chief Justice