ATJSI, demonstrative exhibits
March-April 2022

The new drawing board and Pugh v. State

By Britt Houston Lindsey
Chief Appellate Prosecutor in Taylor County

In Pugh v. State,[1] the Court of Criminal Appeals has given us a definitive statement on the admissibility of computer-generated animations in state criminal trials, in an exhaustively detailed and well-written 61-page opinion by Judge David Newell. While that may seem like a daunting read and perhaps give rise to unease for what the technological future may hold in criminal prosecution and defense, it is at its core the same principles prosecutors have used for more than a century:  The animation is a blackboard for a State’s witness.

Background

The underlying case arose in my very own Taylor County and involved the death of William Keith Delorme, whose body was found in the parking lot shared by three Abilene bars on the morning of October 9, 2015. Mr. Delorme had caused a scene at one of the bars the night before, acting strangely, demanding his car keys, and threatening the bartender and other patrons with a knife. The defendant, Allen Pugh, and several friends managed to get Delorme outside, and Pugh was heard to say, “If [Delorme] tries to pull out that knife again, we’ll put him under the car,” and “we should knock [Delorme] out or something.” One of Pugh’s friends testified that the last thing he saw as he was driving away that night were stationary brake lights on Pugh’s car.

            Delorme’s body was found the following morning by a man out walking his dog. Officers responding to the scene saw that tire tracks leading to the body in the caliche and gravel parking lot started from 85 feet away. Officer David Thompson Jr., a certified crash reconstructionist, noted the tracks left debris that indicated an acceleration pattern and no deceleration marks at any point. There was a turn in the tracks, and the pattern indicated steering input, which showed the vehicle was not out of control. Officer Thompson used a range finder to map 28 reference points, measuring the tires, wheelbase, nearby buildings, and other objects. He used computer software to create a two-dimensional diagram of the scene, later entered at trial (below).

            Delorme’s autopsy was performed the next day. Dr. Richard Fries testified that Delorme’s crushing-style pelvis injury, posterior rib fractures, and compression of blood from the abdominal cavity were all indicative of somebody who was not only struck but also run over by a vehicle. Fries believed that if the body were dragged under the car, it was likely not for a significant distance, not more than 10 feet.

            A warrant for Pugh’s truck was obtained, and the vehicle was taken in and inspected. Investigators noted rub marks in dirt on the undercarriage, as though it had recently run over a deer or hog, and Bluestar reagent indicated the possible presence of blood. The rub marks started around the front of the vehicle on the driver’s side, went under the left floorboard, and continued along to the back of the vehicle; pieces of skin and hair that were found along the undercarriage were collected and later connected to Delorme by DNA analysis. When Pugh was interviewed by detectives, he initially denied having run over Delorme, then stated that Delorme had lunged at him with a knife at the driver’s side window and either laid across the truck or held onto the mirror as Pugh “floored it” trying to get away. Pugh was charged with Delorme’s murder.

Presenting the case

The case presented an interesting dilemma for trial prosecutors Arimy Beasley and Zach Gore (and later for myself on appeal). Usually when prosecutors are presented with a set of facts worked up by the local police department’s traffic division, we are proving up an intoxication manslaughter, not an intentional murder. Arimy knew from reviewing the video of Pugh’s interview with police that Pugh’s argument at trial would be self-defense, and it would be critical to show the jury that the reconstruction evidence didn’t support that claim. Pugh’s statement to police was that he must have somehow driven over Delorme as he lunged at his driver’s side window; the physical evidence showed that the car began accelerating 85 feet away and that Delorme was struck with the front of the car and went under the middle. Arimy asked the Abilene Police Department’s traffic division officers if it were possible to make a moving three-dimensional animation of the vehicle’s travel path, and Officer Tyson Kropp, also a certified traffic reconstructionist, agreed to do so.

            Officer Kropp used the data gathered from the crime scene to map it in three dimensions. He also relied on DNA and forensic evidence collected by police officers and personnel, photographs taken by the lead CID (criminal investigation department) personnel, the autopsy report, and acceleration tests he performed himself in a similar parking lot. He used the medical examiner’s opinion that Delorme’s injuries were sustained to the left and right sides of his body to orient Delorme in the animation. The rub marks and DNA evidence showed how Delorme was struck and the path his body made under the vehicle. Officer Kropp combined all of this data to create four different animations of the scene: a bird’s eye view, northwest view, southwest view, and first-person view from inside the vehicle. All of the animations depicted the vehicle traveling 85 feet, striking a stationary human figure, and running him over from the front to the back of the vehicle. All four were brief, lasting less than eight seconds each. All animations were provided to Mr. Pugh’s defense counsel a month before the trial date. The animations can be seen on the Court of Criminal Appeals website;[2] they are also hyperlinked in the text of Judge Newell’s opinion.

            Prior to trial, defense counsel filed a motion to suppress the animations, arguing that they were speculative and that the probative value was substantially outweighed by the prejudicial effect under Rule 403. At the suppression hearing, Officer Kropp testified to the data he relied on and the techniques and technology he utilized in creating the animations, and all four animations were played for the court. The judge ruled that the first-person view from the driver’s seat was too speculative as to what the driver actually saw and unduly prejudicial in showing the figure representing Delorme being run over from the driver’s view. The judge denied the motion to suppress the other three animations. The three exhibits were proven up and played for the jury as demonstrative evidence to illustrate Officer Kropp’s expert testimony over the renewed objection of defense counsel. But counsel did not object to the expert testimony of Officer Kropp and Officer Thompson, which would be significant in the arguments made later.

The court of appeals

Pugh appealed the admission of the three exhibits to the Eleventh Court of Appeals, arguing that the trial court abused its discretion in admitting animations that depicted Delorme as stationary and unarmed in contradiction of Pugh’s own testimony that Delorme was lunging toward him with a knife; that depiction was unfairly prejudicial, speculative, and misleading. The State responded that the Eleventh Court of Appeals and others had long found that computer animations of crash reconstructions based on objective scientific data are admissible to demonstrate an expert’s opinion in Murphy v. State,[3] Venegas v. State,[4] and Castanon v. State.[5] Pugh argued these cases did not apply because none had attempted to depict a human form.

            The Eleventh Court held that the trial court did not abuse its discretion: The animations were based on objective data, measurements from the scene, and evidence collected from the truck and the autopsy report, and they depicted nothing gruesome.[6] Pugh filed a petition with the Court of Criminal Appeals, which granted review.

As the judges saw it

Pugh argued in brief and oral argument before the Court of Criminal Appeals that any staged recreation that attempted to depict a human figure was inherently speculative and prejudicial, citing the Court’s prior opinion in Miller v. State,[7] as well as the lower court opinions in Lopez v. State[8] and the companion cases of Hamilton v. State[9] and Lewis v. State.[10] The State responded that each of those cases had very different facts and holdings.

            In Lopez, the Second Court of Appeals in Fort Worth held that a reenactment of a parking lot drug deal using live human actors was highly prejudicial due to both deviations from testimony and the use of actual humans. Miller was a video reenactment of an automobile ride to the scene of a capital murder; the Court distinguished it from Lopez and found that it was merely a series of pictures of the route taken. Hamilton and Lewis involved a highly prejudicial computer animation, but a very different one from the one in our case: It did not involve expert opinion or crash scene reconstruction at all. Rather, it was a crime scene reenactment much the same as Lopez, showing a complicated, haphazard shootout involving at least four human participants, with many details either differing from witness testimony or simply fabricated, in part due to a lack of computer memory.

            In oral argument,[11] I attempted to simplify the matter even further: The animation was really no different from a two-dimensional diagram of the buildings, truck, and victim drawn on an easel and used to illustrate Officer Kropp’s opinion. The only real difference is that the vehicle’s motion was shown rather than implied. I stressed that since there was no objection to the reliability of the expert’s testimony, there was no abuse of discretion in admitting a diagram that helped him explain it, even a moving one.

            The Court of Criminal Appeals unanimously affirmed the Eleventh Court and, in short, held that a demonstrative computer animation illustrating otherwise reliable expert testimony is not inadmissible purely because it potentially involves some depiction of human behavior. An animation may be admitted to illustrate otherwise admitted testimony if it is shown that it 1) is authenticated, 2) is relevant, and 3) has probative value that is not substantially outweighed by the danger of unfair prejudice. The Court noted the defendant did not object that the underlying expert testimony and opinions were unreliable under Rule 702, so the only question regarded the admissibility of the demonstrative exhibits themselves. The exhibits were authenticated as a visualization of the expert’s opinion: Officer Kropp testified that they fairly and accurately reflected what they purported to reflect, they accurately depicted what he intended them to, they had not been altered in any way, and they fairly and accurately represented what the available evidence showed.

            Judge Newell explained that the animations were relevant because they assisted the trier of fact in understanding the testimonial and documentary evidence in a concise and easy-to-understand form, much like a blackboard might be used to illustrate testimony visually. The animation combined the tire tracks; acceleration patterns; medical evidence of Delorme’s injuries; and the forensic evidence of blood, DNA, and skin found beneath Pugh’s vehicle to consolidate and illustrate the testimony of multiple witnesses, provide a clear understanding of the State’s theory of the case, and show why Pugh’s version of events was inconsistent with the physical evidence.

            Judge Newell then addressed the heart of the matter, the Rule 403 analysis, and held that the probative value of the animations was not substantially outweighed by the danger of unfair prejudice. The probative value weighed in favor of admissibility, as the animations “enabled the jury to visually evaluate the plausibility of both the State’s theory and [the] appellant’s self-defense claim” and “conveyed the evidence more effectively than if a witness had merely described it.”[12]

            As to unfair prejudice, the Court noted that the animations were actually the least gruesome depiction of Delorme offered into evidence; the autopsy and crime scene photos were understandably graphic, but the animation of Delorme was little more than a stick figure and it involved no blood or gore. Unlike the “pure speculation” animation in Hamilton and Lewis, the exhibits here accurately reflected the objective evidence and were not based on speculation that might have misled or confused the jury, having been based on calculations derived from objective data and quantifiable measurements.[13] While Pugh’s argument centered on Delorme’s placement within the exhibits, Officer Kropp’s placement of Delorme was supported by objective evidence: His body was discovered 85 feet from where the acceleration marks started, traveled from the front of the back of the undercarriage, and was dragged no more than 10 feet. Judge Newell questioned Pugh’s core contention that the exhibits improperly and implicitly conveyed to the jury that the victim did not engage in provocative behavior and instead noted that the exhibits showed no behavior on the part of the victim—the immobile, featureless figure was clearly not meant to convey any information about the victim’s behavior at all. It was merely a human-shaped placeholder to show Delorme’s position, a stick-figure marker on our moving blackboard.

            Ultimately, the Court concluded there is no per se prohibition against animations depicting human behavior in demonstrative exhibits, agreeing that the admonition in Miller (by way of Lopez) regarding the “inherent danger” of staged recreations involving human beings was dicta in that case. Moreover, unlike Hamilton and Lewis, the exhibits in this case didn’t attempt to recreate any complex human behavior, as the figure representing Delorme merely served as a marker for the place that the strike occurred and how the truck traveled over the body. The animation used no extraneous and speculative detail, only what was necessary to illustrate the expert’s conclusions. Adding a knife as Pugh suggested would have actually added speculation to the exhibits and changed their focus as a demonstrative aid.

            Judge Walker concurred and wrote separately to express his thoughts on the power of computer animations. Because they can be so highly persuasive, he urged defense attorneys not to “sit on their hands” when faced with computer animations composed by the State but rather to seek expert assistance and animations of their own when possible. To this end he urged defense attorneys to consider filing Ake v. Oklahoma motions requesting their own experts, and he admonished trial courts to remember that indigent defendants may be entitled to funds for expert assistance, including computer animations. He also suggested that if funds are not available to supply an indigent defendant with his own expert and animations, then courts could consider using Rule 403 to level the playing field in the interest of fairness by keeping the State’s animations out. However, there’s an important distinction at play here: the availability of funds to pay an expert for a computer animation—and the availability and willingness of an expert to testify. In other words, funds don’t make for expert opinions; facts do. An ethical expert won’t testify as to an opinion wholly unsupported by the facts, and the unavailability of an expert on those grounds is a completely separate question from how experts are funded. Judge Walker’s concurrence expressing support for excluding the State’s demonstrative evidence over funding issues should not be misread as support for excluding the State’s demonstrative evidence when an expert simply can’t be found to support the defendant’s theory of the case. In Pugh, it’s extremely unlikely that an expert would have signed on to Pugh’s version of events—the data just didn’t support it.

Going forward

What’s this mean to you, the hard-working, front-line prosecutor? I’m so glad you asked. First, if you intend to use any sort of animation as a demonstrative exhibit, read Judge Newell’s opinion in Pugh fully. It is extensively detailed and exhaustive, and an article of this length can’t possibly tell you everything you need to know. The quick take is that demonstrative exhibits used to illustrate expert opinion testimony must be based on scientifically reliable testimony that is based on objective data, and here there was a wealth of it: the measurements and photos taken at the crime scene, evidence from the undercarriage, acceleration patterns that officers observed, acceleration tests Officer Kropp performed, and autopsy results. Be aware that while the Court did not issue a blanket prohibition against depicting human behavior in a computer-animated demonstrative exhibit, in this case the Court found that Rule 403 favored admission in part because there was no attempt to depict human behavior—Delorme’s figure was used only to show his position prior to impact, not his actions. As a demonstrative exhibit is not itself evidence, consider requesting a limiting instruction to that effect as discussed in the opinion, even if the defendant does not do so himself.

            Prosecutors must also be aware that what’s good for the goose is good for the gander, and if animations can be used by the State, they can and will be used by the defense. This may lead one to an attack of the vapors, but before we panic, let’s return to our analogy: The animation is a blackboard that either side may use, and the blackboard doesn’t replace evidence—it merely depicts the evidence already entered. The jury needs to be reminded that the trial is not a competition of who has the better animation, but rather which animation and which expert better match the facts, testimony, and physical evidence. Use your blackboards wisely.

Endnotes


[1]  No. PD-1053-19, _S.W.3d_,  2022 WL 224275, 2022 Tex. Crim. App. LEXIS 31 (Tex. Crim. App. Jan. 26, 2022).

[2]  Each video can be viewed at www.txcourts.gov/cca/ media.

[3] No. 11-10-00150-CR, 2011 Tex. App. LEXIS 7230, 2011 WL 3860444 (Tex. App.—Eastland Aug. 31, 2011, no pet.) (mem. op.).

[4] 560 S.W.3d 337, 347-48 (Tex. App.—San Antonio 2018, no pet

[5]  No. 08-15-00225-CR, 2016 Tex. App. LEXIS 12421, 2016 WL 6820559, at *3 (Tex. App.—El Paso Nov. 18, 2016, no pet.) (mem. op.).

[6]  Pugh v. State, No. 11-17-00216-CR, 2019 WL 4130793, 2019 Tex. App. LEXIS 8009 (Tex. App.—Eastland Aug. 30, 2019, pet. granted) (mem. op.).

[7]  741 S.W.2d 382 (Tex. Crim. App. 1987).

[8]  651 S.W.2d 413, 416 (Tex. App.—Fort Worth), pet. granted, case remanded, 664 S.W.2d 85 (Tex. Crim. App. 1983), op. withdrawn by Lopez v. State, 667 S.W.2d 624 (Tex. App.—Fort Worth 1984, no pet).

[9]  399 S.W.3d 673 (Tex. App.—Amarillo 2013), aff’d, 428 S.W.3d 860 (Tex. Crim. App. 2014).

[10]  402 S.W.3d 852 (Tex. App.—Amarillo 2013, no pet.), aff’d, 428 S.W.3d 860 (Tex. Crim. App. 2014).

[11]  Oral arguments may be seen on the Court of Criminal Appeals’ YouTube channel at https://youtu.be/TiG_KUK7bO0.

[12]  Quoting Wright v. State, 178 S.W.3d 905. 912-15 (Tex. App.—Houston [14th Dist.] 2005, pet. ref’d) and Milton v. State, 572 S.W.3d 234, 240 (Tex. Crim. App. 2019).

[13]  The Court noted in footnote that the excluded exhibit showing the driver’s point of view carried with it significant danger of unfair prejudice because it speculatively focused more on what Pugh actually did or did not see and centered the viewer’s attention on the victim’s behavior or non-behavior. I conceded in argument that the trial court had likely made the right call in excluding that animation.