purple motes

In a path-breaking decision, the U.S. Supreme Court posted on its website a video along with text of its opinion in Scott v. Harris. Common sense is essentially multimodal. By including video along with its opinion, the Court provided a decision record that communicates its judgment more effectively than would just text.

The Court's recording of its decisions has had significant effects. In the Court's first copyright decision in 1834, the Court declared itself "unanimously of opinion that no reporter has or can have any copyright in the written opinions delivered by this Court, and that the judges thereof cannot confer on any reporter any such right." (Wheaton v. Peters, 33 U.S. 59, 668). The Court's decisions thus became a legally secure component of the public domain. In Burrow-Giles Lithographic Co. v. Sarony (1884), the Court found, "this photograph [“Oscar Wilde, No. 18”] to be an original work of art, the product of plaintiff’s intellectual invention of which plaintiff is the author, and of a class of inventions for which the constitution intended that congress should secure to him the exclusive right to use, publish, and sell." The Court did not however, include a reproduction of this photograph in its decision. It thus did not place in the public domain a work derived from a photograph of significant public importance. Not doing so, among other effects, lessened public understanding of its authoritative judgment

Verbal statements, whether in judicial opinions or elsewhere, are in some circumstances a poor substitute for video. Courts have long recognized the distinctive value of non-verbal evidence through their practice of admitting into evidence objects in addition to personal testimony and documents. A major issue in Scott v. Harris was the danger to the public from a driver fleeing from police cars signaling to the driver to pull over. Video from the police officers' cars communicates extent and nature of the danger in a way that cannot be reduced to a collection of propositional statements of fact. The video also specifies an authoritative judgment with respect to the extent of danger much more clearly, especially to the general public, than does just a text.

The justices apparently did not consciously edit video in the Scott v. Harris case record in making their opinions. The case record included four color videos automatically recorded from police cars when they activated their sirens. The justices' opinions included two very different descriptions of these videos. The decision record for the Court in this case includes, however, just one black-and-white video. That video (the "opinion video") appears to be the video from one police car appended to the video from another.

Selecting relevant video is an important aspect of reasoned argument in Scott v. Harris. In the opinion video, a total of 3.9 minutes of video (24% of the total run time of the opinion video) showed the crash of Harris' vehicle and the subsequent effects and actions (image of the overturned vehicle, the clouds of smoke rising from it, the police officers rushing toward the vehicle and desperately trying to open the car door to get Harris out of danger, etc.; for better understanding, watch the relevant segments of the video). Harris, who was only 19 years old, suffered permanent paralysis of his arms and legs from the crash. Irrespective of the legal issues associated with the crash, compassion is a natural and appropriate human reaction to Harris' suffering. Nonetheless, this specific crash and its terrible human effect is not relevant to judging the reasonableness of the prior police decision to terminate the chase in a way that created a risk of serious injury or death for the fleeing driver. Including the video segment of the crash and its aftermath fosters well-recognized biases, e.g. hindsight bias, outcome bias, affective bias. Video for judging the reasonableness of the police action should excise the video segment showing the crash and its aftermath.

Camera viewpoint is an important, case-relevant aspect of video in Scott. The camera viewpoint of the video approximates that of the eyes of a police officer in a chasing police car. This video, as a resource for immersive, imaginative experience, is thus oriented toward persons imaginatively assuming the position of a police officer. Presenting the video in a way that helps persons imaginatively assume the position of a police officer is the best use of the video. Ensuring that viewers are conscious of that particular video viewpoint helps viewers to judge critically the effects of their viewing experience.

Diegetic time is another important, case-relevant aspect of the video. The video includes the period from when the police officers began to seek to stop the driver to when a police officer terminated the resulting chase. The dangerousness of the chase depends on specific circumstances occurring over time. These circumstances include the fleeing driver's and the police officers' physiological states. Since the human hormonal system has slower-acting but more general and longer-enduring effects than the human nervous system, events that trigger hormonal reactions associated with a situation of danger can shape subsequent actions and perceptions of time and danger. Hence judgments of public danger involve complicated temporal issues.

Watching and editing video can promote and communicate reasoned judgment of public danger. An important issue is the time span in which a judgment of dangerousness is reasonably made and revised. Whether unusual events are discounted in such a judgment, or weighted extra heavily, also is an important issue. Watching and editing video as if it were a transparently objective record of events ("naive realism") obscures how video really works and impedes reasoned understanding of video's truth value.[1]

Included below are two videos created from video in the Scott case record. The first video includes segments of video from the beginning of the chase, and a segment of video up to the end of the chase. The second video highlights relatively dangerous events that occurred during the chase. Both videos are designed for the viewer to assume imaginatively the position of a police officer. Both videos use selective cuts of the action and connect the video segments with half-second cross dissolves to preserves sense of immersion. Both videos preserve real-world temporal sequence, except for in one switch in viewpoint that is obvious. The second video stops action at two points to suggest a circumstance that would have been particularly salient in a model of accumulating indications of danger. These videos are not designed to be deceptive or to be propaganda. They are meant to foster rational argument about the danger to the public of allowing the chase to continue.

(more...)

Offer Comments

Digital multimedia presentations can powerfully affect legal trials. Consider the Skakel trial:

During the Connecticut District Attorney's closing argument in the trial of Michael Skakel for the murder, twenty-seven years before, of fifteen-year-old Martha Moxley, jurors heard and read Skakel's own words [which appeared] on the screen before them. And in the instant that Skakel admitted to feeling a sense of "panic" when he saw Martha Moxley's mother on the morning after the killing, there on the very same screen appeared the image of Martha Moxley's lifeless body, just as it was found at the scene of the murder. ... The picture of Martha's battered lifeless form immediately explains the implicit meaning of his words. The viewer instantly makes the connection: immediately upon being reminded that morning of the night before, Skakel must have recalled with horror what he had done.[1]

Digital technology and the human body work together to combine words and images. The result in this case was probably a strong physiological reaction involving both neurons and hormones: muscle tension, brain activity constructing a casual sequence, increased heart rate, and other bodily effects typically associated with horror.

Digital technology and the human body together annihilated time. The image was recorded in 1975, on the day after the murder. Skakel's words were recorded in 1997, while Skakel was speaking to a ghostwriter in an early stage of producing a book. The words and the image were combined in 2002, in a digital presentation shown for the first time in the prosecution's closing rebuttal statement.

Associating Skakel's words with different images might have produced a rather different sense. Skakel recalled in 1997 that, in 1975, on the night before Moxley was murdered, he had been drinking alcohol, and that he had decided to get a kiss from Martha, who he said liked him. The closing multimedia presentation included Skakel, who was then 15 years old, saying, "I woke up to [Mrs.] Moxley saying, "Michael, have, have you seen Martha?" It also included Skakel subsequently saying, "I was like, 'Oh my God, did they see me last night? And I'm like, 'I don't know,' I'm like, and I remember just having a feeling of panic." [2] If these words had been combined with images of underage drinking, drunken sleep, and being wakened by a mother's fearful face, rather than images of an alive and smiling girl and then her freshly murdered body, the jury might have had some reasonable doubt that Skakel's recollection of panic meant guilt of her murder.[3]

A prosecutor's fundamental public responsibility is to work to serve justice. When a prosecutor believes that a jury acting justly under law could find a defendant guilty beyond a reasonable doubt, the prosecutor's responsibility is to make, as convincingly as she can, the case of guilt to the jury. The prosecutor's job is not to make the defense's case.

A prosecutor, however, should not introduce, and should not be allowed to introduce, new evidence during closing statements. The images that prosecutor presented in his closing rebuttal statement were in evidence, as was the audio of Skakel's words. The technological combination of words and images that the prosecutor deployed in his closing rebuttal had not been introduced in evidence. Lawyers have always been permitted to arrange freely in verbal closing statements evidence introduced in the trial. Are technological combinations of words and images new evidence?

In a feature article in Criminal Justice magazine, the president of the consulting firm that designed the presentation and a law professor stated that the use of the multimedia presentation was "completely fair and appropriate." These authors stated:

While the crime scene images may very well have increased jurors' sympathy toward the Moxleys and their resentment toward Skakel, and while those images, precisely timed with Skakel's words, probably increased their conviction that Skakel was guilty of murder, using this kind of visual rhetoric, instead of words alone, to help the jurors understand the evidence is legally appropriate.[4]

According to these authors, the multimedia presentation merely provided better technology for an authorized form of closing rebuttal statement:

Indeed, Jonathan Benedict [the prosecutor] unquestionably could have played the same portions of the audiotape during closing and held up before the jury the same photographs of the murder victim, even enlarged and mounted on posterboard, that he used in the multimedia display. The only difference is that the interactive multimedia system allowed Benedict to juxtapose words and images more smoothly, preventing the jurors from being distracted from the content of his argument: that Michael Skakel was guilty of murdering Martha Moxley.[5]

More precise timing of image-word co-occurrence, in this view, is merely an external technology for furthering jurors' understanding.

New digital technologies used in trials can create bodily effects that might be best judged as new evidence that juries must seek to understand. The human body combines words and images from pre-conscious neural processing to high-level processing. For example, recent evidence indicates that multisensory processing occurs very early in the main auditory pathway. Recent work in neuroscience indicates:

low-level multisensory interactions are characterized by a high degree of temporal precision. For example, during audiovisual vocalization processing in auditory cortex, the sign of the integration appeared to be dependent on the timing between the initial mouth movement and the onset of the auditory component of the vocal signal. The longer the time interval between the initial mouth movement and the onset of the voice, the greater the likelihood of observing response suppression. By contrast, a short time interval leads to enhanced responses. [6]

Digital technology that precisely times the co-occurrence of words and images produces meaningfully different neural processing than does older technology that much less precisely combines words and images. Eliminating jurors' bodily activities other than processing in prefrontal cortex ("rational deliberation") is not physiologically realistic. Respect for physiological realism suggests that technology that produces significant, new subjective effects should be considered within the evidentiary portion of the trial.

"Subliminal messaging" does not provide a scientifically reasonable concept for judging the use of visual persuasion technology in court. Stimulating sub-conscious processing of highly prejudicial, case-irrelevant material clearly is an unfair legal tactic. However, sensory processing below the level of consciousness occurs normally and continually in a living human body. Subliminal messaging implicitly conveys a false model of how the human sensory system works.[7]

New visual persuasion technologies can produce powerful effects at low-levels of sensory processing. Opposing parties at trial need to have the opportunity to prompt jurors to bring these low-level effects to high-level processing. Such a requirement would give reasonable meaning to full and fair deliberation that includes new multisensory presentations.

Notes:

[1] Sherwin, Richard K. (2007), "A Manifesto for Visual Legal Realism", p. 10; also published in Loyola of Los Angeles Law Review, vol. 40, 2007. The presentation occurred in the prosecution's closing rebuttal. Note the discrepancy in the textual description of the presentation timing. In the quoted text, the second sentence indicates that the image cut occurred when Skakel acknowledged panic. The fourth sentence suggest that the image cut occurred with reference to "last night". Consider as well Carney, Brian and Neal Feigenson (2004), "Visual Persuasion in the Michael Skakel Trial: Enhancing Advocacy through Interactive Media Presentations," Criminal Justice, v. 19 n. 1. Carney at the time of publication was the president of the firm that created the interactive media presentation for the Skakel trial. That article offers the following textual description:

[Screen 2] "I was like 'Oh my God, did they see me last night?' And I'm like 'I don't know,' I'm like, and I remember just having a feeling of panic." [Photograph #2 of the corpse of Martha Moxley is shown]

That description is not sufficient to identify precisely when the image cut occurred. The text of the opinion of Supreme Court of Connecticut, ruling on appeal, suggests that the image cut occurred after the audio of Skakel's sentence ending in the word "panic." See State of Connecticut v. Michael Skakel (2006) (SC 16844), p. 73. It seems to me that the effect would have been largest if the cut occurred on the initial sound of the word "panic". Perhaps the text of Supreme Court ruling did not attempt to describe the presentation to that timing resolution. However, as noted above, precisely specified image timing is an important distinguishing feature of a digital multimedia presentation. For more information about the Skakel case, see considerable original analysis at TalkLeft.

[2] The quotes are as reported in State v. Skakel (2006) p. 89, ft. 105. The first quote differs slightly and insignificantly from that given in Carney and Feigenson (2004).

[3] The Supreme Court appellate opinion, State v. Skakel (2006) p. 89 ft. 105, states: "The defendant claims that when he stated ‘‘Oh my God, did they see me last night?’’ he was referring to whether they had seen him masturbating." The defendant did not testify at the trial. The defense's brief noted the prosecution's audiovisual presentation omitted a section of Skakel's words relating to masturbation. See Brief of the Defendant-Appellant, pp 78-9, ft. 80. The transcript section that the defense cited does not clearly indicate that a new perception that someone had seen him masturbating was the cause of Skakel's panic. The full transcript is consistent with the reasons for panic described above. They would also probably make a more effective defense-counsel visual presentation than one of Skakel masturbating.

Insightful voices in the blogsphere (Simple Justice, Norm Pattis, a public defender) suggest that Skakel had ineffective assistance of council. Perhaps the above is additional evidence of poor representation.

[4] From Carney and Feigenson (2004). Criminal Justice is a magazine that the Criminal Justice Section of the American Bar Association publishes. The quoted phrase "completely fair and appropriate" comes from this source, which stated: "Because Benedict's presentation was directly and closely connected to the evidence, his visual argument was completely fair and appropriate."

[5] From Carney and Feigenson (2004).

[6] Ghazanfar, Asif A. and Charles E. Schroeder, "Is neocortex essentially multisensory?" Trends in Cognitive Sciences, v. 10, n. 6 (June 2006) p. 284; the quoted text omits endnote and figure references.

[7] In State v. Skakel, Brief of the Defendant-Appellant, p. 79 accused the prosecution of conveying subliminal messages to the jury with its multimedia presentation in its closing rebuttal statement. Carney and Feigenson (2004) assert in contrast:

No subliminal content was concealed in the Skakel prosecution team's audiovisual presentation. All of the images, audio, and text that the prosecution put before the jurors in closing argument were properly admitted into evidence. ... The multimedia system allowed prosecutors to present images and audio already in evidence so clearly and so memorably that their impact on the jurors was profound.

These arguments show no appreciation for how human sensory processing actually works.

4 Comments