Surveillance and Social Memory: Remembering Princess Diana with CCTV

Abstract

Since the 1990s, surveillance camera images have experienced a function creep from their juridical uses into journalism and entertainment. In these contexts, the images have also become memory media. This article, for the first time, analyses CCTV images, meaning closed circuit surveillance camera images, as memory media and discusses the implications of our use of artefacts of control within a frame of mediated constructions of social memory. The article undertakes this work by analyzing remediations of the CCTV images of Diana Spencer and Dodi Al-Fayed in the Ritz Hotel in Paris on 30 August 1997 in television news and a documentary from 2007 and 2011, respectively. It is shown how social memory of Diana’s death is a contested site, in which the images play a specific role.

1. Introduction

On the morning of 1 September 1997, the world woke up to the news of the death of Princess Diana and her recent lover, Dodi Al-Fayed, in a car crash in Paris. The death of Princess Diana had an enormous impact on the civil and emotional culture of the UK, making it a stable part of British, but also global, social memory [1]. The CCTV images, meaning closed circuit surveillance camera images, that were made by surveillance cameras in the Ritz Hotel were affect-affording in the narrations of Diana’s death. These qualitatively “poor” images take a prominent place in the medial archive with which this event is publicly remembered.

CCTV images are a neglected medium in the discussions on the changes of the shape and role of cultural archives, which are fundamentally connected with social sense-making. This is astonishing due to the concept of ongoing recording, and possible retrieval of every moment, that has so saturated ideas about the archive in the digital era [2]. As Jens Schröter has argued, video surveillance’s technological set-up can be seen as prefiguration of the digital archive due to its “always on” quality that enables “dynamic media memories” [3,4].

My use of the concept of social memory is indebted to the tripartite division of collective memory as outlined most comprehensively in the work of Aleida Assmann, but does not neglect further developments [5,6]. Assmann uses an archival metaphor to differentiate between neuronal, social and cultural memory. Social memory, mostly carried by what Assmann calls “communicative media”, is probably the most active site for memory work as a collective practice, because it is here that past experiences of groups, generations, or nations are retrieved from a collective archive into the light of journalistic, political or artistic appropriations that create discourses around which past, and whose past, is to be remembered. Communicative media are, thus, also a site of power struggles, where stakeholders pitch their respective versions of events.

In recent years, the tripartite division of memory in the form in which Assmann developed it has been challenged—most notably by Andrew Hoskins—due to the change of mediatised communication and, thus, the social sense-making processes under digitally networked, online, and real-time conditions [7]. The basic thrust of this criticism is that the carriers and the milieus of neuronal, social, and cultural memory—in keeping to Assmann’s terminology—have become more dynamised than her archival model can suggest under the conditions in which personal experiences are today immediately shared on social media and entangled with the mediatisation of everyday life, that means, the way in which our everyday actions are often already instantly mediatised, and how information that we only receive in a mediatised manner shapes our everyday life. Previous models of collective remembering have been challenged by the observation of the continuing dissolution of the “broadcast age”, meaning a time where few media channels provided the main information hubs, such as TV and radio did in the second-half of the twentieth century. Many events are important only for specific, but highly malleable groups that form and dissolve on online platforms, and these various groups might not be aware of one another and their respective interpretations. Additionally, as has been argued, authoritative interpretational frames of what events mean to the public are declining [8].

These criticisms are valid, but a reformed vision of collective memory processes, which enables a similarly structured and complex image of memory processes in the digital age than Assmann’s work provided for the “broadcast” and previous ages, is in my view still pending, especially when we wish to regard contested discourses. For these reasons I retain the term “social memory”: it is before events are given over to the still more stable faculties of “cultural memory” that their construction is “hot”, in the sense of malleable and dynamically open to appropriations. The media event of Princess Diana’s death is certainly a part of British, and global, social memory. The meaning of her death—and of the surveillance images that show her death—are still contested, as my examples will show. Journalism, as a prime “communicative medium” in Assmann’s model, plays a most active role in the constructions of social memory. In an edited volume by Barbie Zelizer and Keren Tenenboim-Weinblatt, the editors stress the role of journalism not only as a news medium, but also as an important domain of the construction and maintenance of shared memory, pointing to

“[t]he longstanding and complicated role that journalism has played in keeping the past alive. From anniversary issues and media retrospectives to simple verbal and visual analogies connecting past and present, journalism incorporates an address to earlier times across the wide array of its conventions and practices”.

([9], p. 1)

The editors argue that in the contemporary medial situation, which is characterised by increased remediation, re-appropriation, and an overlaying of media artefacts from past and present on new platforms, this role of journalism as a genre to reassess the past needs to be even more urgently investigated:

“In an era of increasingly blended performative domains, of recycled narratives, pictures and impulses that are no longer identifiably tethered to one point in time, of information that seems to come from nowhere, a recognition that journalism regularly and systematically looks backward is long overdue”.

([9], p. 1)

Accordingly, this article will investigate the role of CCTV images as memory media by regarding examples from television news and a quasi-documentary.

As is typical with icons of post-war culture and connected media events, the story of the death of Princess Diana is one of those public narratives that combine fact and speculation. In this sense, the accident death of Diana was not unlike the assassination of John F. Kennedy in its status as a culturally productive memory [10]. Due to this combination of the factual and the speculative, these stories are able to renew their cultural productivity. CCTV footage of Diana and her lover Dodi Al-Fayed recorded by surveillance cameras in the Ritz Hotel on the last day of their lives have become a building block of the productive renewal of the story of Diana’s death throughout the last 15 years. The media figure Diana is one of collective fantasy of the later broadcast culture of the 20th century. The Ritz surveillance camera images of her and Dodi Al-Fayed turned into “visual obituaries” ([11], p. 314) in 1997. Broadcast media of the late 20th century used a medium that indexes her death, even though, and here the medial use of the CCTV images of Diana becomes unlike that of the death of JFK, her death is nowhere to be seen in these images. In this context, we will have to regard CCTV images as an unusual memory medium. The CCTV images that were remediated and loaded with cultural meaning in the last twenty years were images connected to deadly events in which the deaths themselves, however, remained invisible. Examples include images of the abduction of the toddler Jamie Bulger in Liverpool in 1993, or of Mohammad Atta in the security zone at Boston airport on 11 September 2001, as well as, of course, the images of Diana discussed here. These images belong to the genre of journalism imagery that Barbie Zelizer has called “about to die” images [12]. In her analysis of the importance of this genre of images for the ways in which journalism creates affect, Zelizer points to the ethical implications of showing an image of a person who is about to die, our conventions to use such images for news narration and impact creation pose important questions about what can be shown in public, how a public can become voyeuristic in regarding the images of atrocity, and what this means for the responsibilities of journalists, that might be circumvented by using “about to die” images. The “always-on” quality of CCTV makes it possible to create “about to” images from many instances in which no journalists were present, as, in the moment of recording, the event that will become memorable was not imminent or foreseeable. The surveillance camera image can be seen as an image form offering itself to “about to” narrations. In this way, CCTV experiences a function creep from its juridical uses to narrative and affecting social uses.

CCTV images are connected to ideas of social memory in two ways: firstly, they and their various remediations, as well as their being networked with other media, make CCTV images a part of the current set of hypermedia tools with which significance is bestowed, experience and memory are articulated and (re-)constructed [13]. One might object that in the current media economy it makes little sense to regard specific kinds of media in isolation. However, taking one very specific image-type as an entry point and regarding its specific interconnections with other medial genres can help us trace relations between media that develop in using them for the tasks of social sense-making.

The everyday situation in which we often do not know the source of the image material with which collective meaning is created on news and social network sites should not be reiterated by academics in a form-blindness. This could lead to black-boxing how specific epistemic logics and aesthetics are co-constitutive of cultural work and its productivity. “Hyperconnected” or not: we need to investigate which mediated material people connect in order to analyze how they are constructing social meaning in a social structure premised on surveillance and security. The specific kind of CCTV frame that becomes iconic despite not showing the relevant moment of the event itself might act as an especially potent “prosthesis” [14] or cue to social memory. As such, these images carry more than an evidential value: they appear as “legal ready-mades”, as video artist Manu Luksch [15] has called them, and so they leave ample room for speculation, narrativation, and comment once they are taken out of the purely legalistic frame of surveillance as an ordering practice. In this way, the self-same images make various interpretations possible. They do so because they often foreshadow—rather than represent in any true mimetic relationship—the point of the surrounding narrative. This would mean that in some cases, the image that does not show what we want to remember might help us to experience a mediated memory better than an image that does show this event. Of course, there are also many CCTV images that show exactly the point of the surrounding story, but many of those that have become iconic are images of a “before” of events that offer themselves to narration in memory work.

I argue that the surveillance camera image is doubly potent: carrying the ideological weight of the episteme of surveillance and security, its images are able to be used as affect transmitters in cultural contexts. These contexts include their use in narrative media, where passionate experiences are afforded, reiterated, and act upon the recipient. This essay is divided into four parts: first, I will discuss the status of CCTV images as memory media that enable narration. After situating the CCTV images of Princess Diana in their historical context, I will then discuss the memory uses of the CCTV images of Diana Spencer and Dodi Al-Fayed in two cases of remediations from the 2000s. In conclusion, I will sum up my findings, asking how the use of surveillance camera footage in social memory points beyond the case of Diana in a present in which surveillance is normalized, and its artefacts are employed to form passionate social memories.

2. CCTV Images as Memory Media

The Narrative Potential of CCTV Images

Considering CCTV as memory medium has trajectories into two different, but related, research interests: in the realm of the study of memory, a heuristic division of visual memory and narrative memory in public formats [8] is problematized, in my view, by the actual multimodal aspect of memory narratives in mass media. The use of CCTV images is a specific case in which the entanglement between narrative and visual memory is especially pronounced, as I will show below. In the field of further investigating the cultural impact of the saturation of our contemporary life-worlds with surveillance technology, remembering with CCTV also points to an insidious normalization of surveillance and its artefacts as regular parts of our contemporary existence.

I will investigate the construction of memory with surveillance camera images from the following entry points: memory stories are narrations of the past for the needs of the present. The public use of CCTV images, especially when we negotiate the past with and around them, is a form of social communication. Most of the time, social communication in late-modern societies is already mediated. In recent writing, especially on the mediatisation of memory that stresses the importance of the nearly instant mediatisation of (personal) events, it has been stated that (1) also personal memory is immediately mediatized and networked, thereby taking on an implicit public availability; and (2) memory is increasingly susceptible to modification and change. For Andrew Hoskins, this fact calls for a general overhaul of how we understand memory in a manner that acknowledges “hyperconnectivity” and the externalization of personal experience [7,13,16]. Jill A. Edy [8], drawing on a broad overview of research has, likewise, argued that definitions of collective memory have become unstable in an age of selective and versatile media use, as a generalized idea of audience consent inferred by exposure to the same media stories and interpretation routines can no longer be assumed. I argue that stories about the past are circulated by remediations of images that act as immediate memory cues for at least the generations that were exposed to them in the broadcast age while the media event in question was current, but that their narrative interpretations have become more malleable. The impact of media events on generational memory has been discussed in an edited volume by Ingrid Volkmer, which discusses the national and global implications on the generational memory of large-impact media events, such as the death of Princess Diana [17].

To discuss which role CCTV images play in these mediatised memory assemblages, we have to first look at the basic features of this image form.

Going back to the fundamentals, CCTV images, as well as cinematic and video images, are a special kind of photography—and photography is a medium that has since its inception been positioned between science and art, and amongst the tools of scientific observation, policing, and subjectivation. Gleaned from ongoing video footage, the CCTV image is, therefore, unlike other kinds of photography and cinematography. It is more akin, it can be argued, to imaging processes and technologies that are used in medicine, mechanics, and the natural sciences: they are artefacts from a purely functional context. CCTV images exist only as a latent archive: the recordings have to be kept for a legally-prescribed time period, mostly in the range of thirty days, then they have to be deleted. The change from VHS storage to digital storage, however, means at least the technical possibility of keeping the liminal archive for much longer times, and the possibility of a much faster retrieval of specific images according to algorithmic markers, such as face or gait recognition. The movement towards the digital form of CCTV has meant not only a strong increase in the image quality, but also the possibility to more easily share and make use of the images, increasing their travelling potential outside of the policing and juridical realms. When CCTV images are entered into journalistic mediality, the images are usually directly provided by the police. These images, in the public, are foremost connected with a context of visually dramatic crime and deviance [18].

Historically, there has always been a friction between the different uses of mechanically, or electronically, created images. On the one hand, there were images made for observation, both scientific observation and as a tool for the scientization of social ordering practices. On the other hand, images were created for aesthetic and cultural reasons—as purveyors of beauty, vessels of memory, heroization, depictions of divinity, and so on. Images have a tendency not to stay in their assigned contexts, thus, mechanical images can acquire a surplus of cultural meaning. CCTV images are deeply connected with Richard Grusin’s description of the cultural and political present as an episteme of surveillance and security, especially so since the events of 9/11. He called this contemporary situation an “age of premediation”, which was well captured in the movie Minority Report (2001). As in this movie, in our contemporary late-modern societies the future is experienced as a potential threat, which must be contained by knowledge gathering [19]. Film scholar Winfried Pauleit described the surveillance camera image as a potent symbol for this social situation, which entails in its image-form a reference to a future that will be past when the image is seen [20]. Within the very form of the image, that points to a future past, we can perceive the connection to, on the one hand, cultural theoretical concepts of the age of premediation and risk management, as well as, on the other hand, a deeply productive entry point for the construction of narrative meaning.

When we talk about memory work, we will not stand long before we must also regard narration. As Annette Kuhn states:

“It is impossible to overstate the significance of narrative in cultural memory […] not just as the (continuously negotiated) contents of shared/collective memory-stories, but also of the activity of recounting or telling memory-stories, in both private and public contexts—in other words, of performances of memory.”

[21]

Images play an important part in the performance of memory, and they also play an extremely important part in narrative memory. In a text on the modes of narrativity, narratologist Marie-Laure Ryan [22] focuses very much on reader construction as a driving force behind narrations, where, as Monika Fludernik has also discussed, narrativity appears as “not an intrinsic property of events [...], but a semantic network built around them by a reflecting consciousness” ([23], p. 376).

Following this notion of Ryan—that narrativity is basically an outcome of a networked cluster of meaning creation around an event, which induces not only temporality, but also other qualities, such as affect-transportation and point—then there is no reason why an image could not stand in for an event in narratives, as a typical media event—icon [24] and, thus, work as a node in this semantic network. This means that, even if the images are themselves not narrative, we create narratives with, and around, the images. I argue that this is the case in narrative constructions of social memory that use CCTV images—even though the images that evoke memory narratives might not even show what is to be remembered. Social memory also depends very much on the affects and emotions that it creates as much as performs. Seidler’s study on remembering Diana [1] is an outstanding example for the tantamount role of emotions in the public sphere. It is narratives that are enabled by images as cues to memory that express this effect of simultaneous affect creation and performances of public emotion. Monika Fludernik’s concept of experientiality—the possibility to share the experience of the narrating consciousness—as the basic definiens of (fictional) narration [23,25], can also be seen as a prerequisite for social memory—pointing towards the vital aspect of experience and affective investment for “successful”, i.e., productive, social memory stories.1

It is in these kinds of narratives that the remediations of CCTV images become productive. In this context, I do not necessarily talk about full scale novels or dramas that evolve from the trigger of historic CCTV footage—though they exist. The narrative movements that use CCTV for their pragmatic functioning, for the evocation of affective social memory, can be very subtle and extremely brief. Sometimes, mere seconds of runtime embedded in a film and accompanied by the right music will point to a specific narrative of the past and its relevance for the present. In this essay, I will present two case studies of the memory work of the last images of Princess Diana: their remediation during news reporting of the 2007 Royal Inquest and the controversial documentary Unlawful Killing of 2011, respectively. First, however, we need to revisit the historical context of these surveillance camera images that became memory media.

3. Historical Context

The Cultural Work of the Last Images of Princess Diana

In the summer of 1997, Princess Diana had spent holidays on financier Mohammed Al-Fayed’s boats in the Riviera, inducing photographers to capitalize on the display of her summer love with Al-Fayed’s son, Dodi. The fully-mediated love story of Dodi and Diana was brought to a sudden halt by the car crash in the Paris Alma tunnel in the night of 31 August, leaving Dodi Al-Fayed and the driver dead on the scene, Princess Diana dying of her severe internal injuries in the early morning hours of 1 September, and her bodyguard heavily injured. Her death led to a public reaction in Britain that changed, as Seidler has argued, the culture of the display of public emotion in the UK [1]. This lasting importance of the death of Diana is also reflected by ongoing speculation surrounding the circumstances of her death, both in fictional and factual genres. To date, there are at least three novels that satirize, thread conspiracy ideas, or otherwise take their plot spins from the accident in the Alma tunnel and Diana’s early death [26,27,28]. Mohammad Al-Fayed has not stopped fueling speculation that the accident might have been planned, leading to a Royal Inquest into the deaths of Diana and Dodi in 2007. In the year 2011, when Diana, had she lived, would have turned 50 years old, and the year of the marriage of her son William, the big-screen documentary Unlawful Killing, funded by Al-Fayed, was shown out of competition at the Cannes film festival.

As was to be expected from the last images of a persona who still holds sway over a part of the public psyche, the CCTV images were instant icons of the media event of Diana’s death. In a sense, this surveillance footage has been elevated to blend with the conventions of highly professional and staged red carpet photography; they have become famous pictures of the woman who was mass media saturation, even though the CCTV images are, in contrast, aesthetically poor. This aesthetic poorness, however, added to the “auratic” quality of the images in the sense that they signaled “realness” and singularity. The moving last images of Diana—as a clip, as ongoing footage rather than as one iconic still—which were used during the Inquest, are not only geared at remembering Diana, but much more specifically, at finally clarifying the circumstances of her death. The way this death is judged is of importance for British identity and social memory. The question whether she died as a result of a tragic accident or was brought to her death splits the public into two epistemic communities who, with their answer, often also show either their allegiance to the Royal Family (it was an accident) or an implicit Republican impulse (she was killed, preferably by a plot planned by Prince Phillip and executed by MI6). In this context, the CCTV-images of Diana have changed their symbolic import: from signifying nothing but the shocking death of the icon they show when still alive, thus pointing to nothing but Diana, Diana, Diana—the fact that she has lived, the fact that she has died—they have, with the passing of time, turned to symbolize conspiracy—and, simultaneously, the legal fight against conspiracy rumours.

4. Case Studies

4.1. “A Familiar Smile”: The Ritz Surveillance Camera Footage During the 2007 Inquest

During the Royal Inquest of 2007, the full length footage from the Ritz was published, changing the symbolic status of the very few images that had been released in 1997. In the most frequently remediated sequence, Diana and Dodi are seen entering the hotel at 4.35pm on 30 August 1997, walking across a lobby before getting into a mirrored lift, where the Princess looks up towards the camera with a “familiar smile” [29].

Images from the last day of Princess Diana taken by the Ritz security cameras (31 security cameras in and around the hotel) had leaked to the media soon after her death. Disseminated to the public—printed or screened—as mostly single images at first, they belong to the category of “visual obituaries”, as Dietmar Kammerer calls the convention of using CCTV images as last images of event-related deaths. One series of stills became especially famous: Diana passing through the revolving doors entering the lobby, with Dodi shortly behind her. Kammerer sees Diana entering the “media necrocracy”, the group of celebrities who have become symbolically immortal because their image never ages, via these images of her ([11], p. 312). Kammerer also observes how the contemporary serialized publication of the sequence of these images (first, Diana entering the lobby through the revolving door, then Dodi following her) in the German magazine Der Spiegel in an issue from 1997 does not only reconstruct Diana’s last day up to the fatal accident. The way the images are reproduced create what he calls a “rudimentary narrative” ([11], p. 312). This also creates a dramatic effect—possibly also, one might add, as this is indeed a scene of entering, of an entry to a final scene—unbeknownst to the protagonists.

The more images and footage are released, the less do CCTV images of the last day of Diana fit the analytical frame of near auratic images set up by Kammerer, who concentrates in his discussion on the use of the images in the late 1990s. The ambivalence between stasis and dynamis stressed by Kammerer regarding the image sequence as it was reproduced in Der Spiegel ([11], p. 314) is resolved during the inquest by showing the footage as a produced video (albeit, with staggering movements as this is time-lag camera footage). The one hour sequence of Princess Diana’s last day released as evidence during the inquest in October 2007, thus, ten years after the event, is a post-production in a double sense: it is post-produced from the large amount of footage the Ritz hotel possesses, and it is done so in a way that echoes the reports of Dodi’s and Diana’s last day that were long since available. This footage conveys nothing but a visual reiteration of these reports. Yet the comments that accompanied the footage’s broadcast on SKY News on the evenings of 3 October and 4 October 2007 frame it as a glamour reportage: The reporter’s commentary is focused on the “lush” environment inside the Ritz hotel as well as the “luxurious” imperial suite to which we see Diana and Dodi ascending on the footage from the camera inside the lift, and the commentary also offers several comments on what the Princess is wearing and how she is looking [29].2 The investigative footage, fodder for conspiracy theories over the span of ten years, here reveals the only meaning that images of Princess Diana can have, whether on grainy surveillance camera footage or on a high-resolution portrait shot: they are always, and inevitably, footage from the red carpet, celebrity pics. The reporting on Sky News was chosen for discussion here because it was the television program that showed the footage most comprehensively at the time in 2007, and it was the program that devoted most airtime to them. Also, it was the news story that received a high number of views on YouTube, where it was uploaded and shared. The remediation of the SKY News reporting on the CCTV footage on You Tube shows the gathering of meaning surrounding these images in an exemplary way.

The factual banality of these images, simultaneously bestowed with so much narrative meaning, echoes David Marriott’s analysis of CCTV images as “empty signs” that he developed in connection with another set of “about-to-die” CCTV images in the British public, the images of Damilola Taylor from 2000, a boy who was stabbed to death in the London suburb of Peckham:

“Like a pure abstraction of the public sphere or an event without history, the CCTV image designates an excess of mimesis over that of narrative meaning […]. The entire three sequences [of video footage of Damilola shortly before him being stabbed, NF] are haunted by this symbolic distance between what appears and what remains ungraspable as resemblance: the death of a boy that cannot be seen and which, because of this, appears everywhere in these images, but without which these images remain an empty sign of that event, a semblance in which his death has vanished. […] Where do we see [the death of the boy]? At first sight it seems untraceable, impossible: the freedom to replay, freeze or enlarge the video image only realizes the following: These images are empty and inaccessible, they cannot act as witnesses to what we know (or imagine) as the death of this boy, a death which remains outside the time and processes of the televisual image.”

([30], p. 321)

This observation of the paradoxical emptiness of images of moments before an event occurs, as they are simultaneously charged with narrative meaning, is illustrated by the fact that SKY News showed more CCTV footage on 4 October 2007, in which nothing but the empty service corridor behind the Imperial Suite was to be seen for nearly 40 s. In TV news temporality, 40 s are a long time for basically nothing to be shown, so the fact that this corridor footage was indeed given to the public’s scrutiny had to be legitimized by the commentator:

Male Commentator: “Although these seem to be rather boring pictures, these are, in fact, historical. These are the very last pictures we will ever see of Princess Diana alive, and within 15 min of these pictures, Diana and Dodi Fayed will be dead […]. These are moving pictures, I think, although this is just an empty corridor here.”

(0:00:24–47) [29].3

The loss of aura of these images, and journalistic ventures to act against this loss by reminding the viewer why she is supposed to give her attention to images of an empty hotel corridor, is also obvious from the advertising language that, to give a complementary example, The Daily Mail used in its reportage of the footage in 2007 [31]. It had to be stressed as early as in the title of the article, that these images were indeed “new and unseen” to a public to which no image of Diana can ever appear unseen, nor new. The “auratic” presence that the images seem to possess in the analysis of Kammerer referred to above lose their aspect of singularity in the ongoing footage. Rather, these images were moved back into the juridical idiom of a public coroner’s inquest, and public reporting failed to bestow them with the cultural import that they had in 1997. The images, long since artefacts of media memory, are supposed to become again that for which they were made: legal evidence. This objective in combination with the attempts to historicize and glamourize the images again, as it occurred not only in SKY News but also in national newspapers such as The Guardian, The Daily Mail and The Daily Telegraph, which were similar in their interpretation of the images, creates a friction as to where the cultural location of these images is actually to be found. They have become simultaneously artifacts of legal evidence—factual and objective—and the material of public remembering—affective and affecting.

4.2. Narrating A Befitting Death: Keith Allen’s Unlawful Killing (2011)

As stated above, the CCTV images of Diana and Dodi are now signifying conspiracy stories, symbols for the affective certainty that the icon could not have died a banal death. The quasi-documentary film Unlawful Killing [32] was shown at Cannes film festival in 2011 and managed to cause a slight stir in the British public, which was at the time, as film-maker Keith Allen states in the entry narration to his movie, “demented by another Royal wedding”, that of Diana’s son William: itself a ceremonial media event that resonated with the remembering of Diana. Although the film was strategically launched and very expensive, it failed to secure a distributor either for the US or for the UK, where it’s screening pended 87 changes to become legal in cinemas. After a struggle to find a distributor, the film was made available on DVD in early 2015, without aiming to show it in cinemas anymore. Allen’s documentary, aiming for a controversial style in the vein of Michael Moore’s films, was unsuccessful. Unlawful Killing not only has the problem that its dramaturgy lags and is confused. The supposed “inquest of the inquest” represents the Royal family and especially Prince Phillip as murderers. The movie has one aim: re-opening the questions that the inquest of 2007 had closed [33]. While the movie takes its title from the verdict of the inquest, it does not mention that the jury decided on 7 April 2008 that Diana and Dodi had been unlawfully killed due to grossly negligent driving by the chauffeur, Henri Paul, contributed to by the equally reckless driving of pursuing vehicles. Rather, the movie pretends that the perpetrators of an “unlawful killing” was still open to investigation and interpretation, here in the person of the—Al-Fayed financed—director Keith Allen.

The above mentioned released footage from the Paris Ritz security cameras plays a pivotal role in the movie. The known Paris Ritz footage is used for various reasons at different points of the narration, and for different aims, although it is the same set of images that is used repeatedly in the film to move the plot forward and illustrate the narration. It is divided into three distinct scenes: the first shows Diana and Dodi entering the Ritz through revolving doors on the evening of 31 August. A second scene shows them leaving the Imperial Suite towards the elevator. A third scene is compiled from the footage from the elevator. The last set of images post-produced into a scene shows the couple passing through a corridor towards the rear exit of the hotel that leads into the backstreet, where a Mercedes was parked in which the couple departed with Diana’s bodyguard and the driver at around 00:20 h, on the journey which ended with the accident in the Alma tunnel. In addition to this footage of Di and Dodi, there is a set of footage of Henri Paul also entering the Ritz at around 22:00 h on 31 August through the revolving doors, walking around the hotel lobby around 22:20, and a set of images of Henri Paul standing outside the hotel in the backstreet where the Mercedes was parked, waving at someone on the opposite side of the street at around 00:15 on 1 September 1, just prior to the departure of the car with Diana and Dodi.

The film opens with a black screen on which a handwritten letter gradually appears, supposedly by Diana, in which she predicts—or fantasizes, according to epistemic community—her former husband plotting her death in a car crash made to look like an accident. This is followed by a fast cut of Diana and Dodi in the Ritz corridor and elevator, leaving for the fatal drive. These images, as they were also seen on TV, are now made affect-creating by the fast-forward mode in which they are shown, and by a soundtrack of pulsating indie-rock produced by Dave Stewart. In this way, the post-production use of the film Unlawful Killing gives a new meaning to those stale images that the commentators of SKY News could do little to make interesting. They are now weaved into the story that will culminate in the assertion that Prince Phillip ordered the death of Diana and Dodi. As a constitutive part of this narrative, the CCTV images now can regain their auratic quality: they are employed as a fully experiential medium that is supposed to induce a gut feeling in the viewers of Allen’s movie that Dodi’s and Diana’s accident was indeed suspicious.

Allen uses this technique throughout the movie, weaving the self-same sets of CCTV footage into various interpretative and experiential frames: combined with newspaper images and red carpet footage of Diana, the images are used to narrate the love affair with Dodi and to generally represent the memory of Diana; they act as artefacts of mediatized mourning. Combined with photos of the blood samples and images of written documents in the case of the Henri Paul images, the images are used as a supposedly evidential medium once more. Allen’s film creates different moods with the self-same set of images by adding different soundtracks, by editing and replaying them in fast-forward or slow-motion mode, and combining them with various other sorts of image material. Allen’s film even shows that an empty screen can be made to narrate.

One recurrent ingredient of the conventionalized Diana conspiracy narrative is the missing footage from traffic video cameras along the route that could have been used to establish how fast the Mercedes had actually been going. It could have also established if it was followed or hindered by other vehicles—or it could even have captured the ominous bright flash that is repeatedly stated by conspiracy theorists as the reason for their belief that Diana’s death was not a pure accident. A bright flash is supposedly a technique used by security services in order to blind drivers. The missing footage from the traffic cameras is therefore one of those double-bind tropes within the pragmatics of conspiracy stories: the fact that it should exist makes its non-existence suspicious, but its factual non-existence also enables conspiracy theorists to keep to their sinister versions of happenings, as David Cohen states in his non-fiction book on Diana, Death of A Goddess:

“It is true that no CCTV footage was ever produced or mentioned by the police. […]. Once the police said that the speed camera at the Alma was malfunctioning, the disinformation spread and was exaggerated in a way very familiar to those who study rumor. Soon the story spiraled into the ‘fact’ that all the cameras on the route were out of action that night.”

([34], p. 158)

Keith Allen’s movie performs the story of the non-existent footage. The following is a transcript from the movie. On the left, you will find the run time, then, given in italics, the video proper, i.e., what you see during the given runtime, then Keith Allen’s voiceover narration. I have not considered other soundtracks in this transcript.

1:04:54–1:04:59 empty video tape running on screen

Narrator: “what you are seeing is what the CCTV camera at the entry of the Alma tunnel was recording at the time of the crash.”

1:05:00–1:05:01 empty video tape running on screen

Narrator: “nothing.”

1:05:02–1:05:09 Alma tunnel entry, normal daytime traffic

Narrator: “because it was switched off, although it is usually switched on 24 hours a day.”

01:05:10 image of the CCTV camera outside the tunnel

Narrator: “was that just another coincidence…”

01:05:12 poster of Di at Alma tunnel entry, handwritten on it: “ti amo sempre”

Narrator: “…or something more sinister?”

[32]

Again, “empty images” are used to narrate a story. The aim of the narrator’s speech act is clearly to create doubt in the listener/viewer concerning the prevalent explanation that the non-production of CCTV footage of that night is a coincidence. The film narration expresses this through the second sentence that accompanies the images from normal traffic at the Alma: on that night when Diana died, the CCTV camera was not running, although it is “usually switched on 24 hours a day”. The combination of this attempt to convince the viewer of the “fact” that the disabled camera is no mere coincidence is framed in a question. Thus, the illocutionary force of the utterance, the aim for the listener/viewer to provide her own answer to this given possibility, is especially obvious. What the viewer actually sees while she hears the commentary is pushing the emptiness of images that can be put to narrative used to an extreme: In the short time frame of only 16 s, we have three cuts. Of a sequence of 16 s, nearly half (seven seconds) are devoted to showing a reeling video tape.

A reeling video tape is not the same kind of blank screen that the computer age is used to. On a computer or a digital recording, a blank screen is blank, there is nothing but static blackness, even if the time is running. Instead, here we see a video tape running, which means that there is actually something moving on the screen. The empty screen actually does represent something: it has a temporal dimension, a memory dimension. It assumes that the factual reel of video from the CCTV camera at the Alma tunnel looked like this in 1997. That we are presented with a close-up of this camera in the next cut, supposedly doing its job in daytime traffic, seems like a validation of the factuality of the narration. This evocation of non-existing video footage to tell a story of conspiracy is also interesting for the implications it makes about visual surveillance: it is what is missing which is understood as problematic, therefore asking for comprehensive surveillance and policing in a context that could have brought clarity. In an underlying appraisal, surveillance is therefore perceived in a positive light—paradoxically, in a film that is marketing a conspiracy story.

The final cut presents us with a symbolically loaded image: a poster of Diana in a tiara, situated at the column that stands on the roof of the Alma tunnel entry, with the handwritten words “ti amo sempre”. This image caters to the mournful tone the movie slips into, even when it is supposed to be “the inquest of the inquest”, as the narrator announces at the beginning of the movie. However, where the short sequence does wish to bring about an argument (the CCTV was switched off, this was not a coincidence), it closes and centres on an image mourning Diana, who some people obviously still personally love. The poster shown is not withered but brand new, therefore testifying to the fact that the entry of the Alma tunnel is now a memory site. Like the TV footage discussed in the previous section, Keith Allen’s film reveals the same friction that is created by the use of CCTV in the narrative: here, a friction between the investigative journalism the film supposedly performs and a media document of mourning, in the case of this film, probably less public mourning but the very personal one of the billionaire who financed the contentious film.

5. Conclusions

In this article that analyses the function creep of surveillance imagery’s use as memory medium, I have first outlined relevant positions and developments in studies of social or collective media memory. I then discussed aspects of the surveillance camera image as a specific image-form that resonates strongly with our current culture of premediation and security. I have then discussed how the surveillance camera image that foreshadows the surrounding event might be especially productive in the constructions of affecting, experiential memory narratives. In the case study presented in this article, I have discussed the use of CCTV images of Diana Spencer and Dodi Al-Fayed on the last day of their lives as a memory medium, employed as remediations in both factual and quasi-documentary narratives.

From the commentating on the footage in a television news show, to a conspiracy-fueling, tendentious documentary, I have discussed the stylistic and rhetorical features of how these images are employed together with narratives to create different kinds of frames and interpretations of the memory of Princess Diana’s death in multimodal television and film. We have followed the story of the images themselves, from an early remediation as stills in a German journal in 1997 as images for mourning an icon, to their re-narration during news footage on the Royal Inquest in 2007. Lastly, we have seen how they were productively used as images that can dramatically narrate one man’s personal grief in the form of a narrative, supposedly documentary film, Unlawful Killing, which appears as a cinematic monument to Mohamad Al-Fayed’s angry mourning.

The question remains: what is the specificity of the image-form of CCTV images in memory narrations? Is there an underlying difference between the use of these images in memory narrations and other kinds of images?

In a social situation in which we are flooded by images, CCTV constantly produces images that are rarely seen. There is a reversal between the ubiquity of my networked smartphone snaps and the ubiquity of ephemeral CCTV recordings. While all of my personal snapshots are increasingly available, CCTV recordings are not. In a public culture of instant availability, CCTV images are restricted. When they are released, they often do not show the happening that is accumulating meaning to become an event, as media event studies normally assume [24]. However, as the case studies have shown, they are able to accumulate narrative meaning as images of a premediation, of the before of an event, but does this create a difference between CCTV images and other kinds of images employed as memory media?

Whether CCTV images as memory media are indeed different from other kinds of images employed in constructions of social memory depends on a decision concerning the status of media difference per se, as Rey Chow and Julian Rohrhuber [35] have also discussed. Basically, the authors state, there are two possible entry points in this discussion: either we follow a narrow reading of Friedrich Kittler’s stance [36], assuming that medial translation induces a total transformation without residue. Regarding our examples, this would mean that, once moved into another frame, CCTV images would stop being surveillance images and become indistinguishable from any other kind of image. A second, contrary entry point would follow Jacques Rancière [37], who has argued that a medium, in whatever way it is remediated, retains and quotes the milieu from which it stems ([35], pp. 50, 51). Rancière’s understanding of media difference, and traces of original milieus, echoes the original definition of remediation by Bolter and Grusin [38]. In this view, the fact that we remember with surveillance artefacts is in itself meaningful for our contemporary imaginaries. Of course, this article has sided with the second stance, as it is the prerequisite to regard the role of a specific medial form in intermedial relations for the constructions of social memory. The blending of surveillance camera images into affecting memory narratives shows that surveillance technology has become a part of our cultural expressions. This observation points towards a scopic logic of contemporary constructions of visual and narrative memory echoing that of surveillance, which in turn highlights the normalization of surveillance artefacts in a media ecology in which cultural meaning is created. In my view, further discussion of what it means that we have started to remember, with all the affecting aspects of memory narrations, with surveillance artefacts needs to be discussed further in the context of changing definitions of social memory. In the case of the uses of Diana images, we can also observe the friction that CCTV images create in cultural work because of their double-valence as memory media and medium of surveillance, embedded in power structures. We will have to continue to ask what it means for cultural work when media of surveillance technology saturate the ways in which we construct social meaning and express social feeling. Remembering Diana with CCTV means to continue to situate her death in an uneasy conglomeration of legal procedure, conspiracy and grief—it also means that a scopic logic of surveillance is part of contemporary constructions of social emotions. Especially in our contemporary situation—where much of our collective or shared emotional life takes place on “hyperconnected” platforms, narratives with CCTV images help us keep in mind that these platforms are part of an episteme of surveillance and security. It is exactly in a situation where the source of information gets more and more black boxed due to information overload where stakeholders—journalists, analysts, and the public itself—have to pay attention to the kinds of media material that we surround ourselves with. CCTV images are an image form that has been forgotten in discussions on memory in the digital culture, possibly because they appear as arcane forerunners of smartphone footage, which is today often the first image material that appears in the case of an event. However, we should return to them because they can reveal how we remember, now digitally, events from a time when the “age of premediation” was first installing itself.

As any other public form, narratives with CCTV are not “innocent” of, and might partially reveal, underlying power structures. They are part of the creation multimodal discourse events. Using CCTV images for narrative purposes is a social creative process—an activity—by which we make cultural use of surveillance objects, so that they are part of how we, the contemporaries, are expressing ourselves and our becoming in the control societies. The images are part of our reality insofar as they are part of the ways in which we medially make sense of realities. What the case study discussed shows is how contested these realities can be—and how the self-same surveillance images can stay productive in the creation of cultural meaning as memory media.