Abstract
Wearable self-tracking devices such as Fitbit and Apple Watch have begun to be featured in high-profile criminal cases such as murder trials for the ways the data from these devices supposedly give access to objective truths about a victim's physiological state and time of death. Drawing from the concepts of deathlogging and datafication, we argue that an examination of media framing around wearables in criminal investigations and court rooms helps demonstrate how technologies become connected to different domains and practices. A discourse analysis of five case studies and 57 news stories showed how the framing of wearables in this context emphasized both enchantment with the capacity for datafication to witness, as well as contestation around how to define and interpret this witnessing. This study demonstrated how reporting around the incorporation of mobile media technologies into institutional practices often reproduced desires for digital technologies to provide access to “the truth” of social reality.
“You better be telling the truth because this Fitbit is going to blow your story” (Gardner, 2016). This is how Jacksonville, FL defense attorney Janet Johnson described the introduction of Fitbit-generated data into courtroom proceedings. Johnson is one of many lawyers, investigators, pundits, and reporters who have positioned the use of wearable trackers—specifically Fitbit and Apple Watch—as valuable to criminal investigations, suggesting they offer basically a new (and improved) polygraph. In the same article, reporter Lynnsey Gardner suggests, “Fitness trackers can capture every detail about how we spend our day and how our body performs along the way” (Gardner, 2016). While reproducing a sort of mythology around the omnipotent possibility of trackers to accurately capture “every detail,” Gardner's suggestion has functionally become true in murder investigations where a victim was wearing a Fitbit. Over the past decade, at least a handful of murder cases have gained some level of national and international media coverage for the ways they incorporated wearables-generated data into initial police investigations, arrests, and/or courtroom proceedings.
This article builds on Gilmore's (2023) concept of deathlogging, or how technologies can (accidentally) generate archives of human death, which are then used as the basis of accident reconstruction and assessment. The initial conceptualization of deathlogging focused on accidental sports deaths that had been recorded on mounted GoPro cameras, leaving “behind a record of their actions, reconstructing accidents or disappearances to make sense of an individual's death, figure out what happened and even potentially assign blame” (p. 482), and related specifically to Siegel's (2014) notion of forensic mediation, or the ways media technologies are used to reconstruct “what has happened”. This article develops deathlogging in a related direction by examining how wearable fitness trackers may leave traces of a wearer's death, acting as a witness and reconstructing the timelines, locations, or actions of a murder victim in the moments leading up to their death.
Taking a communication approach to this phenomenon—as opposed to, say, a criminological approach—this article is focused on how the incorporation of Fitbits and Apple Watches into murder investigations and courtroom proceedings is framed in news reporting across five different case studies: the murders of Connie Dabate, Karen Navarra, Nicole VanderHeyden, Myrna Nilsson, and Maria Ladenburger. We employed a thematic, qualitative discourse analysis to trace how technologies are positioned as “witnesses” to criminal acts, as well as to trace debates around whether their data meet legal standards of admissibility. We point to tendencies to treat the technology as storytellers or as providing access to an objective reality as an extension of celebrations of datafication, which treat quantified renderings of human experience as more objective. These media reports are part of the broader processes of legitimating the capacity to treat wearables as witnesses and imbue mobile media with the legal authority to reconstruct what has happened around particular kinds of crime. Despite the contestation and debates around some of this reporting, the media reports largely reinforce datafication as a means of creating supposedly more objective modes of witnessing through mobile media and deathlogging.
Deathlogging and datafication
Fitbit, Apple Watch, and other sorts of fitness trackers have been positioned as forms of self-tracking in everyday life (Gilmore, 2016; Lupton, 2019; Neff & Nafus, 2016), as opposed to having a focus on death. Some researchers have used activity trackers as a means of “autofiction” (Campbell, 2019) or “memoir” (Prasopoulou, 2017). As part of what Turkle (2008) described as the “always-on/always-on-you” nature of mobile media, wearables are designed to be appended to one's body and generate ongoing and ostensibly useful data about habits, mobility, heart rate, and other elements of bodily function. They proliferate historical trends that Crawford, Lingel, and Karppi (2015) equate with the introduction of weigh scales in the early twentieth century to monitor and track a quantified understanding of one's body. Several studies position wearables as a form of mobile media. Oh and Kang (2021), for instance, offer factors to explain how and why users engage with wearables. Pink and Fors (2017) emphasize material entanglements that emerge around wearable self-tracking devices. The COVID-19 pandemic's emphasis on mobile media such as community contact tracing apps positioned wearable technologies as means of assessing individuals’ health and tracking locations (de Souza e Silva & Xiong-Gum, 2023; Gilmore, 2021). Wearables have also been studied in research on location tracking (Beattie & Cassidy, 2021; Gilmore, 2020) as well as surveillance (Zimdars, 2021).
Deathlogging entails intersections between death studies and media, technology, and communication (Arnold et al., 2018). Some of this research has focused on the processes of memorialization or online afterlives, where records of deceased individuals live on via social media profiles, videos, or other traces of life (Bollmer, 2013; Bourdeloie & Julier-Costes, 2016; Briggs & Thomas, 2014; Hjorth, 2021; Jacobsen, 2017). Other research has shown how online spaces can provide opportunities to reflect on the loss of human life (Gibbs et al., 2015; Lagerkvist & Andersson, 2017; Ulguim, 2018), as well as the role of mobile media in how individuals work through the processes of memorialization, grief, and trauma (Cumiskey & Hjorth, 2017). Deathlogging is distinct from this research because it emphasizes some sort of recording of the moment of human death and is only possible because of the mobile nature of these devices.
While some researchers have explored augmented reality crime tracking (Liao et al., 2020) and the relationships between mobile communication and perceived security (Ellcessor, 2022; Reichow & Friemel, 2020), communication and media scholarship has not focused on the increasing use of digital devices in criminal trials and associated commentary in news reports. As recently as 2020, a transdisciplinary review of research on wearable health technologies did not indicate research strands on crime, criminal investigations, or death (Li et al., 2021). This study, then, opens further research directions around the uses of wearables in criminal investigation, further developing the concept of deathlogging toward an understanding of how various institutions expect wearables-generated data to have the authoritative capacity to serve as witnesses.
Wearables-generated data have gained legitimacy in part through the social imaginaries produced around datafication. Social imaginaries, as Taylor (2003) argues, are ways of thinking about elements of society, which are not necessarily accurate or objective but accrue through repeated representations. Media representations of what technologies do—and are imagined to be capable of doing—help establish the boundaries of possibility around technology. With wearable technologies, this often entails a consideration (and, often, a celebration) of datafication, or the process of converting elements of daily life into quantifiable data sets (Couldry & Hepp, 2017; Mayer-Schonberger & Cukier, 2013). As van Dijck (2014) has argued, datafication is not so much a scientific paradigm for generating empirical knowledge as much as it is an ideology that places authority on the knowledge generated through data production. For example, Fitbits and Apple Watches do not actually measure heart rate; they use a series of sensors and processes to generate close approximations based on how the devices are designed to interpret blood flow and the capacities of the various lights in the hardware and the algorithms in the software.
Social imaginaries around new or emergent technologies often entail frames of “enchantment” (During, 2002; Gell, 1992), emphasizing the ostensibly magical work they do, and frames of contestation (Marvin, 1987) emphasizing the difficulties of having new technologies become integrated into social communities. We see the simultaneous embrace of and skepticism about wearables-generated data as consistent with earlier debates around biometric data in legal proceedings. Bunn's (2012) historical research on the polygraph or “lie detector” technologies provides a useful precedent. In considering how the formation of criminology existed alongside various social representations of machines designed to supposedly give access to “the truth”, Bunn traces the tensions between the social scientific understandings of criminal behavior and exaggerated reporting on technological innovations that would eliminate lying. There have long been desires to position the data produced about bodies through technology as authoritative and objective. Fitbit and Apple Watch have relied on these imaginaries in the advertising for their products, with Fitbit being sued for claims that their smartwatch would allow wearers to “Know your heart,” despite having statistically significant deviations in how it recorded human heart rate in various scientific trials (Goode, 2016). This is akin to what Vasterman (2005) describes as “media-hype” as a means of framing, magnification, and amplification. This process of magnification and amplification has helped position wearable technologies as witnesses capable of accessing the truth around what has happened.
Peters (2001) positions “witness” as having “two faces: the passive one of seeing and the active one of saying,” where passive witnesses observe events, but active witnesses are a “privileged possessor and producer of knowledge in an extraordinary, often forensic, setting in which speech and truth are policed in multiple ways” (p. 709). Where Frosh and Pinchevski (2008, 2014) have developed “media witnessing” as a concept largely focusing on audiovisual and mass media's capacity to change how audiences relate to the representation and depiction of significant historical events, these wearable witnesses produce knowledge through computational sensors and algorithms, rendering it as data displays. In the context of dead human bodies—in particular, refugee bodies—Papailias (2019) suggests witnessing “leaves traces of objects, bodies, spaces, and images” (p. 1059). Wearable witnesses, like the audiovisual witnesses of still- and moving-image recording, also act as a repository for traces, although they have become “datafied”; traces of heartbeats, steps, and other bodily movements or functions are rendered as data points or quantified as heart rate. Wearable witnesses further transform what Richardson and Schankweiler (2019) describe as “affective witnessing,” or the centering of “encounter, embodiment, affect and intensities of experience” in the process of witnessing (p. 237). This is particularly evident in audiovisual documentation of police brutality against Black bodies (Campbell & Valera, 2020). In rendering murder and death as a series of numbers—the stoppage of step counts, the slowing of heart rates, the absence of biometric data—the intensities of experience are produced as something based in the supposed calculability and knowability of quantified data rather than the more embodied representations of human death that can be witnessed in some audiovisual media.
In much of the framing analyzed below, Fitbits and Apple Watches themselves are treated as subjects providing a form of legally admissible testimony, which gains authority precisely because datafication has become so legitimated as a form of knowledge production. To consider the etymology of the word, the Old English witnes meant “attestation of fact” from personal knowledge in its verb form, and originally was used to mean knowledge. While a Fitbit may not be literally called to take a witness stand for cross-examination, the reporting around the case studies in the analysis below all routinely use phrases from “wearable witness” to “Fitbit murder,” centering the role of the device in the investigation and suggesting that wearable technologies are capable of attestation of fact in these circumstances. Some research has already established how wearable-generated data might be used in investigations and in courtrooms, treating Fitbit as a forensic tool for gathering evidence during investigations (Almogbil et al., 2020), and assessing the capacities of devices like Garmin Connect to be used forensically (Hutchinson et al., 2022). While the scope of these and other applied studies go beyond the bounds of this study, the increased attention being paid to whether and how wearables data can be used forensically participates in the growing desire to position these technologies as evidence in criminal investigations.
Taking these strands together, deathlogging offers a means of analyzing how the mobile and tethered properties of wearable technologies are not only passive collectors of quantified data but also assist in producing knowledge about what is happening to a body around its moment of death. The forensic use of these wearable trackers treats ongoing and automatic data collection as acts of constant witnessing, treating the process of datafication as the attestation of facts about human conditions.
Method
We performed a discourse analysis of media framing around wearable technology's use in five different murder investigations. After the researchers conducted an unpublished pilot study of coverage of Connie Dabate's murder and the subsequent trial of her husband Richard Dabate—called the Fitbit murder in much press coverage—we identified four additional case studies of murder investigations where data from a wearable device (Fitbit or Apple Watch in four cases, and the health data on a worn iPhone in one case) were used. In all the cases we identified, the murder victim was a woman: Karen Navarra (US), Connie Dabate (US), Nicole Vander Hayden (US), Myrna Nilsson (Australia), and Maria Ladenburger (Germany). These case studies were identified through search engines such as Google as well as dedicated searches in newspapers of record (The New York Times), major news outlets (CNN), and technology journalism sites (Wired), which helped to locate initial stories on these cases. Combinations of phrases such as “Fitbit + crime” and “wearable + murder” were initially used to locate cases. These search criteria biased the case studies included, in that the researchers are based in the United States, so news returns will favor that region, and we only identified cases that had gained substantial news coverage. This foreclosed the possibility of including additional cases from other geographic regions where stories were not widely reported or did not move beyond local reporting, which did limit the scope of this study.
Once cases were identified, we searched the victims’ names in Google with the combinations of phrases from the cursory search to locate additional stories. We followed hyperlinks from within national news coverage to help identify local newspapers where the stories originated and searched within the online archives of those local news outlets for additional stories. After an initial collection of all available news stories, we removed repetitive articles, such as Associated Press wire stories that had been reproduced across several news sources. After removing redundant articles, our data collection process yielded 57 news stories for our data set. From here, the researchers proceeded with the primary cycle coding phase of the research, working together through the batches of articles for each case to identify keywords and phrases that contributed to the framing of wearable technologies in each article. After coding the 57 articles, we believe we reached saturation due to recurring keywords and phrases such as “black box,” “witness,” “objective,” “evidence,” and “lying”. The frames around these keywords—the overall disposition of the articles as well as the positions of law enforcement, technology experts, or others quoted in the articles—helped us arrive at five initial themes: (1) wearable data as objective; (2) wearable data as a black box for accident reconstruction; (3) wearable data as a witness; (4) wearable data as requiring new forms of expertise in courtrooms; and (5) wearable data as lacking scientific standards of courtroom evidence. The combination of keywords, metaphors, and frames all worked together in creating thematic categories for analysis (Ryan & Bernard, 2003).
Following this initial coding, the two authors again took turns reviewing articles to confirm that the assigned themes were appropriate and to make recommendations on whether to keep, remove, or alter the existing themes. After this round of review, we consolidated these five themes into three: (1) wearable technology as objective; (2) wearable technology as subjective witness; (3) wearable technology as inadmissible (in and of itself). As a typology, these three themes demonstrated an embrace (theme 1), a skepticism (theme 2), and a critique (theme 3) of wearable data in murder trials. Each theme was present to some degree across every case, demonstrating a consistent spectrum of framing. After establishing these three themes and concluding the secondary cycle of coding, we analyzed the articles through their framing of wearable technology in relation to murder investigations and subsequent courtroom proceedings. Here, we drew from tenets in constant comparative methodology to move across the data set (Glaser & Strauss, 1967), as well as to dialogue with the theoretical and conceptual orientations of deathlogging and witnessing described above.
Analysis
This section focuses on representative examples from the data set for each of the major themes. Each subsection offers exemplary, rather than exhaustive, extracts from the media reports to focus more on the particular framing strategies and phrases being used in the reporting.
Wearable technology as objective
In this frame, Fitbit figures almost as an epistemological shift, a new means of producing data that will fundamentally alter how investigations and courtroom proceedings function. In a 2014 article about the introduction of Fitbit data more generally into courtrooms, a lawyer using Fitbit data as part of a personal injury case remarked, “Till now we’ve always had to rely on clinical interpretation … Now we’re looking at longer periods of time through the course of a day, and we have hard data” (Olson, 2014). Here, the distinction between “interpretation” and “data” suggests Fitbits provide access to some element of the empirical realm that is rendered objectively. Clinicians, though experts, are always at the mercy of their own opinions, which Fitbit data circumvent. This lawyer here reproduces one of the core elements of datafication: that data speak for themselves, and “raw data” provide access to truths. As Lisa Gitelman (2013), among others, has repeatedly argued, these is no such thing as raw data: all data are analyzed through human interpretation or through mathematic and computational programs humans have developed to understand data in particular ways. The dichotomy between interpretation and data is, then, false: all data analysis is fundamentally interpretive, in that it works to make sense of the data, place it into context, and give it meaning.
Others have gone so far as to call wearable devices “a ‘black box’ for the human body,” equating them with existing modes of forensic mediation and furthering their ability to provide accurate records of crimes (Olson, 2014). This idea of the black box manifests in reporting around these cases in different ways. Consider the following extract from reporting in People magazine on Connie Dabate's murder, which uses the phrase “Fitbit Murder” in its headline to help centralize the role of Fitbit in this investigation: Even by soap-opera standards, Dabate couldn’t have predicted the clue cops say was most damning: Connie's Fitbit data. According to information gathered from the device, Connie had been moving around for nearly an hour after her husband said she was killed. While Dabate had told police that his home alarm had been triggered at around 9 a.m., the Fitbit showed Connie walking more than 1,200 ft.—nearly a quarter of a mile—until 10:05 a.m. “The timeline just didn’t add up,” says the Connecticut police spokesman. “There were a lot of questions” (Helling, 2022).
The Fitbit data reconstruct the accident, creating a timeline demonstrating Richard Dabate falsified his testimony. Other reporting seized on this phrasing as well, such as suggestions that “the deceased’s Fitbit data revealed that she was at the gym when he alleged she had been murdered” (Hurler, 2022). This statement grants Fitbit more capabilities than it actually has; authorities used security cameras at Connie Dabate’s gym—not her Fitbit—to place her there. This slip also occurred in the CNN reporting, which produced a “Fitbit timeline” centering the wearable device even though it simultaneously acknowledges IP addresses, alarm system logs, and other data were also used (Watts, 2017). Without detracting from the importance of Fitbit data, such reporting risks reproducing technological enchantment, inflating the device’s capacities and masking the ways other forms of recordings and data were used to investigate Dabate’s murder. Local reporting from the CT Insider, while careful to not give Fitbit too much power, included a subsection of its write-up on the trial verdict in the Dabate murder called “Fitbit and forensics,” exploring the capacity for Fitbit to serve as part of forensic mediation practices (Dempsey, 2022). Crime blog writer Rivy Lyon (2022) suggested that the use of Fitbit in investigating Dabate’s murder “proves the immeasurable forensic worth of this technology,” unquestionably celebrating the use of this technology and pointing to a seemingly infinite number of possibilities for it in criminal investigations.
In reporting around Karen Navarra's murder, Fitbit's ability to record her heart rate acted as the proverbial black box data. National reporting on the case noted, “The police statement said data from the [Fitbit], which includes heart-monitoring technology, showed a spike in Navarra's heart rate, followed by a rapid slowdown” (Molina, 2018). In her Wired magazine feature on Navarra's murder, Lauren Smiley notes that “Using trackers this way, of course, assumes that the devices are accurate—and not just accurate on average, but at very specific moments in time, a sort of black box for the body that reveals physiological truths that its wearer might prefer to conceal” or, in this instance, truths that the wearer is unable to speak due to their own death (Smiley, 2019). Again, the idea of worn technology providing access to physiological truths reproduces some of the discourse around polygraphs mentioned earlier in this article, supposing that biometric data represent a version of a truth serum (or, at least, a magnifying glass), gaining access to an empirical realm otherwise unobservable.
Fitbit's ability to serve as the means of solving investigations appears elsewhere, as in New York Daily News reporting on Connie Dabate: “A Connecticut man accused in his wife's murder might have gotten away with it—if not for the victim's Fitbit fitness tracker and other electronic devices” (Schroeder, 2017). Prioritizing Fitbit and minimizing “other electronic devices” reinforces the frame that wearables act as a black box. Ten paragraphs later, the same story acknowledges that analyses of the “alarm system, computers, cellphones, social media postings” as well as the Fitbit data demonstrated Dabate's falsified timeline, but the reporting overemphasizes Fitbit up to that point, placing greater agency onto the device itself.
This tendency to focus on Fitbit as the central site of authority, and its data as providing access to an empirical realm of legal truths, corresponds to broader cultural embraces of datafication positioning quantified records as objective and accurate. This theme emphasizes Fitbit data as largely unquestionable, as a record of what has happened, which can be easily admitted. In overemphasizing Fitbit's role and deemphasizing or misrepresenting the role of other digital or DNA evidence, this frame tends to take a role of ideological reproduction in suggesting datafication and digital data sets are reflective of reality.
Wearable technology as subjective witness
Here, Fitbit is framed with a level of agency or capacity to provide evidence to investigators and lawyers in ways that treat it with the affordances and constraints of a human witness. Some reporting has gone so far as to use headlines like “Fitbit for the prosecution” (Dugan, 2022), an explicit play on the movie Witness for the Prosecution (1957) meant to centralize Fitbit's role in the juridical process. Phrases used in reporting on the Dabate murder include mention that the “case [was] built partly on evidence provided by her Fitbit exercise activity tracker” (Wilkerson, 2022). Fitbit is often placed at the center of courtroom deliberations, such as claims that “data [from Fitbit] helped the jury come to its quick verdict” even when Fitbit is only one part of a broader suite of electronic records (Russo et al., 2022). Though Fitbit itself cannot “speak”, per se, this sort of reporting positions it as a form of star witness with an unexpected and authoritative account of what has happened. For instance, some writing around the Dabate murder put it this way: “What truth her Fitbit revealed was that Connie was alive for over an hour after her husband said she had been shot to death” (Lyon, 2022). The clunky phrasing here—“truth her Fitbit revealed”—again rests on a sort of anthropomorphizing and suggests the device has the capacity to speak a truth. Fitbit has the power to narrate, in phrases like “her electronic device tells a different story” (n.a., 2017).
Fitbit's narrative power is perhaps nowhere more evident than a 2019 Wired cover story called “The Telltale Heart.” The magazine cover includes the following copy: “He was an unlikely suspect. 90 years old. Wouldn’t hurt a fly. But there was a witness—and the victim was wearing it.” While the language is clearly hyperbolic to advertise the issue, it explicitly calls Fitbit “a witness,” a framing that extends into the article's headline: “A Brutal Murder, a Wearable Witness, and an Unlikely Suspect” (Smiley, 2019). Written as a magazine feature, the article is full of literary turns-of-phrase that continually reinforce Fitbit as a narrative storyteller. For instance, Smiley says Karen Navarra's Fitbit “seemed to tell the story Karen couldn’t.” Fitbit here gains agency, in that the device is crafting a sense of what happened to Karen and at what time. CNN's reporting around the Navarra murder also used language implying Fitbit was able to provide testimony, using the headline “The murder suspect denies it. The victim's Fitbit tells another story, police say” (Hanna & Chan, 2018). CNET reporting on Navarra's murder suggested “The case underscores how the devices that surround us can serve as important witnesses in court. Data from fitness trackers has helped form timelines, track movements and provide other valuable information” (Musil, 2018). The collective use of the word witness across news stories speaks to an ongoing imaginary of technologies as agential, as doing the work of other human witnesses.
Fitbit's ability to produce “stories” appears in other cases, as well. In coverage of Myrna Nilsson's murder, a Mashable article describes how her Apple Watch was used to construct a timeline contradicting the suspect's, using the Watch's heart rate data to construct a contradictory timeline (Lieu, 2018). Other reporting suggests “Myrna's smartwatch told a different story—and actually implicated Caroline in the attack” (Bonazzo, 2018). While wearables provide some access to an empirical truth, as discussed in the first theme, their witnessing functions are a shorthand for their narrative capacities—the ability for the data they produce to tell stories about human activity. Other reporting on Nilsson's murder treats the smartwatch data as a form of forensics: “Forensic investigators, using the watch's heart rate data, were able to narrow the moment Myrna Nilsson was attacked to the moment she died in a seven-minute window” (Post Staff Report, 2018).
Treating Fitbit as a form of testimony on which cases are built and argued reinforces datafication as a mode of witnessing, where wearables-generated data record and produce a sense of presence and observation. When Fitbit acts as a witness, it is treated as a reliable. As The Verge noted in its coverage of Karen Navarra's murder, “Data from wearables and smart home devices is proving to be increasingly important to investigators as they attempt to solve crimes that lack reliable witnesses” (Porter, 2018). Implicitly, Fitbit fills that reliability gap by documenting what happened to a person and functioning as a “valuable tool for bringing criminals to justice” (Porter, 2018). In this theme, the devices are not framed as providing immediate access to an empirical reality, but are instead positioned as storytellers, narrators, and interpreters of what happened. They fulfill the role of witness by being framed as subjects with the agency to communicate.
Wearable technology as inadmissible (in and of itself)
This frame acknowledges wearable technology to be limited and requiring of potentially new forms of expertise to explain what wearables do and what can be understood from their data. In many ways, this frame is meant to counterbalance the overly enthusiastic discourse analyzed in the “wearable technology as objective” frame above. Keith Diaz, who served as expert witness during Connie Dabate's murder trial, became in many regards the unwitting face of wearables’ legal standards through the national media coverage of the trial. Diaz, an exercise physiologist and assistant professor at Columbia University Medical Center, has published a number of peer-reviewed studies on Fitbit's accuracy (e.g., Diaz et al., 2015). In an interview with The Verge in May 2022, Diaz both acknowledges the importance of error rate while trying to contextualize it in relationship to legal evidence: It doesn’t matter if it's 100 steps versus 92; it's really just about if they moved. With this Connecticut case, there was a lot of back and forth in the cross-examination about the error rate. But the error rate was on the number of steps someone took—not the error rate of if they were moving. That's a very different story. … When there is no motion detected, when it is not recording any steps, we should be pretty darn confident that a person is not moving (Wetsman, 2022).
Diaz discusses the difference between scientific and legal standards of accuracy, noting that many studies of Fitbit's validity, which focus on the accuracy of measures like step count, distance traveled, and heart rate, acknowledge the device can record approximations of these measures. In moving into the processes of court proceedings, interpreters such as Diaz explain what the data tell jurors in relation to criminal proceedings. Part of the necessity for interlocutors like Diaz is because “there are still no legal standards for how and when this new type of data should be admitted” (Smiley, 2019), and because of this, experts provide “guidance” to “permit jurors to weigh new types of tech evidence” (Smiley, 2019).
Other articles discuss Diaz's testimony in relation to the defense's arguments. In one example, a local NBC article (Massaro, 2022) details Diaz's testimony and the defense's attempts to demonstrate Fitbit's inability to serve as a suitable piece of evidence. While the defense attorney attempted to claim the Fitbit could have not synced its timeline data correctly, Diaz discussed how these claims were “unlikely” through a discussion of how Fitbit's syncing works with Internet-connected devices. In this example, there was open discussion of Fitbit's limitations and possible discrepancies to sow doubt in its ability to be used in the courtroom, but Diaz's expertise helps to explain and contextualize the data. In important ways, he serves as a mediator between the wholesale embrace of datafication and the rejection of Fitbit as fundamentally biased. In recognizing Fitbit's biases and flaws but explaining how its data can be made useful, Diaz demonstrates how to make the rearticulation of these data sensible.
Other investigators have tried to replicate the data generated from digital devices to demonstrate their potential admissibility in court. In Maria Ladenburger's murder, for example, police relied on the health data generated on an iPhone (as well as forensic evidence such as hair), which suggested their suspect had been “climbing stairs” at about the time they suspected Ladenburger's body had been moved. As the BBC summarized, “An investigator of similar build to the suspect went to the area where the body was found and recreated how the police believe he disposed of the body. The police officer's movement data on the same app showed him also ‘climbing stairs’” (n.a., 2018). While police investigators celebrated this as a breakthrough, suggesting “For the first time, we correlated health and geodata” (Burgess, 2018), this method of reproducing iPhone data only demonstrates how some activities are coded as climbing stairs based on slight changes registered in a device's altimeter when it is worn against a human body. The reporting on this murder in Wired, for instance, did not acknowledge the other DNA and forensic evidence used to identify the murderer (Burgess, 2018). While an important tool for building an argument for conviction, such reporting overstates what these data can indicate and risks obscuring the coordination of multiple forms of evidence.
The defense teams in these murder trials have routinely called Fitbit data into question. Richard Dabate's defense team “questioned the reliability of the data from the Fitbit tracker, saying the devices were not designed with the accuracy needed for court testimony” (Wilkerson, 2022). Caroline Nilsson was found not guilty of murdering her mother-in-law, Myrna Nilsson, despite Apple Watch data that contradicted her story, because “circumstantial evidence does not paint a clear picture of murder” (Carter, 2021). This was after initial reporting from Nilsson's arrest, where prosecutor Carmen Matteo stated, “The evidence from the Apple Watch is a foundational piece of evidence for demonstrating the falsity of the defendant's account to police” (Pascu, 2018). Nilsson's case illustrates how this theme complicates and responds to the celebration of wearable data in themes one and two: while prosecutors relied on the Apple Watch data as a black box of when the murder took place, the difficulties of using those data as evidence of guilt proved insurmountable. In the other cases surveyed, a multitude of digital records, from doorbell surveillance footage to IP address activity, were used to build timelines, as opposed to the Nilsson case relying almost exclusively on the Apple Watch data.
Conclusion
This article analyzed reporting on five different murder cases where data from wearable technologies were used to make arrests and as part of courtroom proceedings. In focusing on the media framing of these cases, we have explored how attitudes toward wearable technologies as witnesses have been established. Witnessing and datafication are consistently linked together to suggest data generated through wearable technologies offer ostensibly “new” (and, potentially, “better” or more meaningfully objective) ways of producing legal evidence and accessing objective reconstructions of crime scene elements such as time of death. To return to Peters’ (2001) understanding of witnessing, these devices are passively recording through their sensors as well as producing knowledge about “what happened”. Importantly, in each of these five cases, the wearable technologies are only one part of an assemblage of evidence drawn from a variety of sources; the reporting's framing of them as “Fitbit murders” simplifies how many different technologies work together in criminal investigations. This amplification-via-reduction risks making wearables seem more singlehandedly useful than they may be.
In courtroom proceedings, these wearables-generated data come to speak on behalf of individuals (and, in the case studies here, women of different ages, races, occupations, and nationalities) who have been brutally murdered. Wearable witnesses do not perform the traditional functions of spoken testimonial detailing what happened, by whom, and under what circumstances. Rather, they render the act of witnessing as a quantified report of step counts, heart rates, times of activity and inactivity, which are (in part) used to make legal arguments about how death occurred. This is a far cry from the calls in some media and death studies research that digital technologies would create new possibilities for immortality (Jacobsen, 2017). Deathlogging places the focus of analysis not on the processes of memorialization, grief, or trauma, but on the processes of recording human death itself, and how (in these cases) wearable technologies have accidentally rendered the action of murder as a series of data points.
Several limitations to this study also provide avenues for further research. Although we focused on murder cases that reached national or international media coverage, there are likely to be many other crimes where wearable technologies have been similarly used to build timelines, reconstruct physiological responses, and arbitrate against suspects. Additional research could work to locate such case studies and assess whether the themes identified and analyzed in this study also manifest when reporting remains at the local level. Case study research allows for a deeper engagement with events, and this article relied on five cases to describe a more complex understanding of how this discourse framed wearable technology data. Although we believe a sample of 57 news stories permitted us to reach saturation, we also chose not to focus on podcast episodes that featured these cases, as podcasting is a different medium than news reporting. Given that true crime is a hugely successful genre of podcast production (Boling, 2019) and several of these cases have been featured on podcasts, a follow-up study could contrast the themes that emerge through an analysis of podcast transcripts. As mentioned in the methods section, our use of Google, mainstream U.S. news reporting, and mainstream technology reporting sites like Wired inherently limited what news stories we were able to find, and we were biased to English-language reporting.
Further research into media framings around emergent technologies and deathlogging remains necessary as systems beyond wearable technologies are increasingly imbricated into juridical practices. For instance, an elderly Black man in Chicago, IL was jailed for a year after an algorithmic gunshot detection system called ShotSpotter suggested he had shot and killed another man before the judge dismissed his conviction on grounds of insufficient evidence (Burke, Mendoza et al., 2022). While these emergent technologies are not necessarily mobile media, they demonstrate other areas where analysis of reporting can help show to what degree media framing participates in the construction of legitimating or contesting the incorporation of digital, mobile, and algorithmic technologies into juridical processes.
While this study focused on a rather specific phenomenon, it also demonstrated how mobile media research can use a thematic analysis of media framing to examine the relationships between mobile media, human death, and legal processes. The connections between datafication and witnessing developed here can also be used to further research in violent crime such as sexual assault (Chauriye, 2016), or the extraction of data from medical devices such as pacemakers to investigate individuals’ cardiac activity in a crime scene timeline (Draft, 2019). As these incidents proliferate and the articulation of wearable technologies to juridical proceedings continue, mobile media and communication research can continue to explore the frames, standards, and consequences of this ongoing articulation and to what degree this witnessing is producing knowledge that aids in producing justice. Specifically, this research can continue to explore how enchantment frames may eclipse “contestation” frames and potentially make it more difficult to assess the utility and limitations of using wearable technologies as witnesses.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Author biographies
James N. Gilmore, Ph.D. is assistant professor of media and technology studies in Clemson University's department of communication. His research on wearable technologies is published in New Media & Society, Convergence, Critical Studies in Media Communication, and elsewhere.
Cassidy Gruber is an undergraduate student in Clemson University's department of communication.