Print this article [ PDF ]

Cultural Analysis, Volume 18.2, 2020

Consequences for Society as a Whole: Narrating Technology in the Norwegian Media Debate on Online Surveillance

Guro Flinterud
Norwegian Police University College



Abstract:Developments in digital technologies over the last decades have sparked new life into the surveillance debate. In this article, I present a close reading of the Norwegian media debate to suggest that the debate is not so much over whether to allow authorities surveillance capabilities but about establishing new stories about life in an increasingly digitalized world. Reading the debate through the lens of the concept of legendry, I suggest that the arguments put forward draw on two sets of implied narratives competing for credibility.

Keywords: surveillance, narrative, legendry, digital technologies, media debate

____________________


Introduction: Surveillance Narratives

In recent decades, developments in digital technology have changed the face of surveillance. All over the globe, particularly in industrialized, urbanized areas, people’s lives are increasingly intertwined with digital technologies (Blank 2012). We continuously slip in and out of connected modes throughout the day, leaving detectable electronic traces of movements and communications that were once transitory. For law enforcement and intelligence services, the digitalization of everyday life represents a new arena for collecting information and evidence. Simultaneously, it represents a new arena for storytelling, actualizing in new ways the narratives inherent in the opposition between security and privacy that so often forms the core of arguments in the debate over the authorities’ surveillance capabilities. Traditionally, those who call for expanded capabilities argue on the grounds of security, saying that such capabilities are necessary to keep citizens safe, while those who oppose expansion will argue that it encroaches on citizens’ privacy and restricts their freedom. Although scholars and commentators have pointed out that this is a false dichotomy because one cannot have one without the other (Loader & Walker 2007), most arguments in the media debate are still polarized, with contributions framed as representing one of the two perspectives.

I argue that the debate about the authorities’ use of surveillance capabilities in traditionally democratic countries draw on two diverging sets of narratives, following the lines of security and privacy. Security and privacy can be seen to constitute stable narrative structures for each of the two sides of the debate over time, security structuring stories about the state keeping its citizens safe from outside threats, privacy structuring stories about the state as encroaching on its citizens’ freedom. The security-story about surveillance capabilities could be summarized as the story about a society where criminals and terrorists have gotten the upper hand over the authorities. The authorities used to have the means to protect the people, but in recent times, criminals have started exploiting new technologies to commit their crimes. There are two possible endings to this story. One is that the authorities get the legal approval to utilize the new technology and can again prevent terrorist attacks and catch criminals. The other is that access is denied, or the legal process drags out in time, leading to terrorist attacks and increased crime. The privacy-story could be summarized as a story about a society where the population currently enjoys freedom, albeit freedom under constant threat from state authorities who want access to invasive surveillance measures. This story always ends with some form of totalitarianism, where the state, possibly as a side effect, introduces chilling effects by allowing its law enforcers access to invasive surveillance measures, leading to less, not more, security. Through changing contexts and despite the increasing acknowledgment that the security/privacy opposition is indeed a false dichotomy, this underlying pattern continues to structure the debate, with the result that the two sides of the debate do not talk about the same “reality” when they debate. They use the same language, but concepts such as “the state,” “freedom,” or “consequences” connote entirely different things when uttered on either side. For example, on the security side, freedom means being protected by the state; on the privacy side, freedom means being protected from the state1.

In this article, I will explore how this plays out in the Norwegian media debate. I suggest that the arguments put forth in media texts, though seemingly dry and objective statements of facts, could be read as implicit narratives that draw on and re-create the worldview of either of the two narrative structures governed by “security” or “privacy.” What characterizes the Norwegian media debate on surveillance is that it is performed almost solely by stakeholders. In general, those championing the security narrative are representatives of state authorities, while those who follow the privacy narrative are representatives of civil society. The populace’s voice is only barely visible in my collected material, and as will be discussed below, studies suggest that the topic is not widely debated among the general population in Norway. However arguments are made concerning a concrete case, such as a legislative proposal, their power to convince lies in that they imply events, both past, and future. Most importantly, each side of the debate follows a chronological structure that tells diverging stories about contemporary society. I will read the empirical material through the lens of legendry. I will focus on how the participants use what Elliott Oring has termed rhetoric of truth to build credibility, and how the utilization of this rhetoric draw on and simultaneously reinforce and re-create these overarching narratives (Oring 2008). I will also employ the term “kernel” as a concept for exploring storytelling by way of reference (Kalčik 1975). What elements and techniques do the debaters employ to draw on and add to these narratives to make truth claims? How can this contest for credibility shed light on contemporary society’s coming to terms with rapid developments in digital technologies?

This thesis is similar to Bruce Lincoln’s investigation of the role of myth in establishing, upholding, or overthrowing social structures (Lincoln 2014). Lincoln proposes that we should classify these types of narratives “not by their content but by the claims that are made by their narrators and the way in which those claims are received by their audience(s)” (Lincoln 2014, 22), an understanding that resonates with Oring’s focus on the rhetoric of truth. He divides the narratives and their functions into three categories: history, legend, and myth. The claim is that stories with persuasive power are regarded as history, while those that lack credibility are classified as legend. Myth describes the stories that possess both credibility and authority, thus being, in Lincoln’s classification, the stories that have the power to shape society (Lincoln 2014, 23).

I am interested in exploring the process through which these positions are negotiated, how the categories Lincoln label “history” and “legend” are played up against each other in the media debate, competing for the state of “myth.” In other words, how the debate over surveillance capabilities is also a negotiation over the story of contemporary society. Although I analyze my empirical material in light of legendry with the understanding that what defines legend is not whether or not the stories are true, Lincoln’s categories help elucidate the positions of the debaters and their narratives. Using Lincoln’s categories, we could say that each group in the debate considers their overarching narrative as history while treating the other group’s narrative as legend. This indicates that there is currently no valid “myth” regarding digital surveillance capabilities in the society studied here—no one story holds both credibility and authority across the expressions made by each group in the media.

The concept legendry captures a range of expressions that gravitate around legends. These are often non-narrative in form but concern themselves similarly with matters of truth, such as beliefs, rituals, or rumors (Oring 2008, 128). I propose that media texts might also be read as legendry, especially in cases like debates over societal order, where they often navigate the border between fact and fiction, truth, and belief (Frank 2011). This approach is an attempt to pick up Oring’s suggestion that one can fruitfully compare the study of legendry and its rhetoric of truth with “truth-making practices in various social and cultural groups and in other areas of discourse” (Oring 2008, 158). I am also inspired by Oring’s comparison of legend and news in contemporary society, where he points out the paradoxical relation between legend and news (Oring 2012). News could be defined as the antithesis of folklore in its focus on fact and objectivity and coming from a managed source, yet at the same time, news has been an important source for contemporary legend for folklorists for many decades. In this article, I suggest that the rhetoric that makes news reports and other media texts appear factual and authoritative consists of rhetorical devices similar to those applied in legends, not reported as explicit narrative, but using a stripped-down form to make references.

This implicit storytelling is similar to stories identified by Tom Mould in a study of welfare legends in the USA (2016). Mould and his research team collected stories about welfare recipients that could be defined as legends. In addition to these, they found predominant use of storytelling through reference, where the speaker evokes a whole story or corpus of stories using only a few words. Mould refers to these stories as “kernel stories,” a term coined by Susan Kalčik to capture a narrative type that exists as reference, and that often does not develop beyond this state due to the audiences’ familiarity with the full story (Kalčik 1975, 7). This way of evoking a larger whole through reference is similar to the media debate over surveillance powers. Although argumentative in style, the media texts debating surveillance capabilities typically make their claims to truth with reference either to a cause of events where the state authorities are falling behind and need expanded capabilities to protect the citizens, or where the state authorities are encroaching on the citizens’ privacy.

While one can define the narrative technique of reference found in the surveillance debate as kernels, the reference point for the kernels of the surveillance debate diverges from Kalčik and Mould’s definitions of a kernel story. In Kalčik’s use, kernel stories refer to concrete, singular stories of an event that can be told in a relatively stable narrative form, while Mould expands the term to also include evoking a “corpus of stories,” which I understand to mean larger sets of stories describing similar events (Kalčik 1975, 7-8; Mould 2016, 394). However, the reference points for the surveillance debate encompass stories about a wide range of events with one thing in common: they all draw on and confirm one of the two worldviews mentioned above. Stories referred to by those arguing for expanded capabilities follow the structure of a benign state that has lost control of the present and need access to expanded surveillance capabilities to reinstate security. They include references to known and planned terrorist events, events connected to criminals operating in free spaces online, or stories about the helplessness of police who watch crimes committed online as they unfold, being restricted by law from stopping the events even though they have the tools to do so at hand. Those who oppose expansion follow a structure where the present is characterized by a threat by state authorities who want to use technological developments to curtail their citizens’ freedom. For instance, they might refer to a range of stories from totalitarian regimes; they can be connected to historical events from the STASI-regime or World War II, to events in Western democracies after 9/11. Although each kernel might refer directly to a kernel story—that is, a story about a defined event—the kernel’s primary rhetorical use in this case is to establish a truth that goes beyond establishing the truth of the kernel story. I argue that the reference made in these media texts use kernel stories as rhetorical devices to establish the truth of a larger narrative framework. I, therefore, suggest the use of the term “implicit narrative” to describe how the narrative style described as kernels and kernel stories can be part of a strategy to connect events in order to build trustworthiness for a collective understanding of societal developments—how legend-like stories and their rhetorical devices are found in texts and communications throughout society as legendry.

In the following, I will first discuss the relevance of legends for understanding the surveillance debate before describing and contextualizing my empirical material. The following analysis is divided into two sections. In the first section, I will focus on the rhetorical devices employed to make sense of digital technologies. In the second section, I focus on how the debaters reinforce these understandings using pathos, mainly playing on citizens’ fear of terrorist attacks in contemporary western culture.

Legend, Truth, and Belief

Legends are often concerned with the supernatural or the horrific, explaining natural formations through the involvement of historical figures or events or seemingly unexplainable events by the intervention of ghosts or aliens. They are often told as stories about other groups, describing their practices as bad or wrong, with the underlying effect of normalizing one’s traditions and practices. Legends can thus have the function of drawing borders. Although legend is a well-known term in both vernacular and academic understandings, there is an on-going debate about how exactly the genre should be defined (see Oring 2008, 127–128). In this article, I follow Oring’s notion that “legend is concerned with matters of truth” (Oring 2008, 128). This claim is proposed as an alternative to the common folkloristic definition that legends in one way or another are related to belief. This assumption is not wrong, Oring points out, though it is also not very precise; belief is part of so many parts of our communication that it does not effectively single out legend from other forms. He says that a legend makes claims about the truth of an event and is, or approximates, a narrative. Thus, his notion highlights three concepts that will be important for the analysis in this article: truth, event, and narrative. The debaters use reference to real events to build credibility, not for the events mentioned, but for the narrative that implicitly structures their argument, either promoting the worldview of security or privacy.

Linda Dégh has written that “Most legends are about spirits, witches, demons, monsters, lunatics, criminals, extraterrestrials, and abilities of certain humans who are empowered with precognition and magic that can identify these evil forces and protect us from their destructive power” (Dégh 2001, 4). She continues to clarify how legends belong to the realm of the extranormal, how they very often are the domain of the alternative voices of society. This claim connects to the contemporary surveillance debate in two ways. First, the rapid developments in digital technology over the last decades have opened up what seems like unlimited possibilities for surveillance. The scope and possibility of digital surveillance are difficult for most people to grasp, and the secrecy surrounding much of the state authorities’ legislation in this field make it impossible for people to have precise knowledge about what is going on. This lack of transparency places the debate in the realm of the extranormal; even when state authorities speak out, they cannot relate concrete facts. The core topic of the debate is shrouded in mystery, making the arguments not only about convincing the listener about the pros or cons of expanded capabilities but also about what these capabilities are and what they do. Second, the voices of the media debate on surveillance are highly official voices, often representing state authorities like the police and members of parliament on the one hand, or academics, legal experts, and political commentators on the other. What connects this debate to Dégh’s claim that legends often are the domain of alternative voices is that they disagree—both sides might harbor authority in society, yet in this particular discussion, they represent diverging worldviews. Depending on whom one has more faith in, each of the two sides could be said to represent alternative voices in this particular matter.

Oring connects the emphasis on the supernatural to the idea that belief is central to the definition of legends (Oring 2008, 128). He points out that the distinct dimension of belief involved in stories about the supernatural becomes blurred in stories involving rational but unlikely events, such as tales about a series of accidents. On the surface, the contemporary surveillance debate analyzed in this article could be interpreted the same way; the media texts relate real things happening in the real world. However, there are elements of the surveillance debate within the domain of belief, such as the effects of new technologies. On the one hand, there is little knowledge among the general population about the exact methods used by police and intelligence agencies when performing digital surveillance. Do online surveillance capabilities make authorities omnipresent online? In fact, what is technologically possible is elusive to many; what the authorities are legally allowed to do is described only superficially in the Norwegian law due to a tradition of keeping legislation technology-neutral, and the capacity they have to utilize their covert capabilities is secret. Although I follow Oring in using the “rhetoric of truth” as my main analytical vantage point, I maintain that the dimension of belief is relevant for understanding the specific content of the material. While “truth” is a fruitful concept through which to understand form, “belief,” at least in the context of the surveillance debate, helps make sense of content.

Debates over surveillance capabilities are usually sparked by mundane events such as legislative proposals concerning the introduction of new tools and capabilities. As Dégh has pointed out, there are several similarities between the legend process and the legal process, such as the polarization of opinions, viewpoints, and interests (Dégh 2001, 1). Referring to court cases, she claims that the antagonists in the legend process are less inclined to compromise than in the legal process, suggesting that this might be because the legal process is concerned with the individual level while legends treat universal concerns. However, the legislative side of the legal process looks much more like the legend process, in that it also treats universal concerns: what kind of society do we want? How will these laws affect our society? The concept of legendry foregrounds the performative aspect of this debate, highlighting the narrative techniques employed to increase credibility and frame the argument within a particular worldview. Oring writes, “Legends may lead to the discussion of belief beyond belief in the narrative incidents themselves. They may invoke discussion about the constitution of the world and the principles by which it operates” (Oring 2008, 128). I argue that these points also apply to the surveillance debate, mainly because the events referred to are future events: the object of the surveillance debate is not to establish whether or not something has taken place but to establish what will take place if/not. Establishing credibility for events that have not yet taken place is a venture involving a “belief beyond belief” in narrative incidents, as it instead becomes a question of forecasting. In general, there has been a shift in the discourse on surveillance capabilities in the digitalized society, from focusing on solving crimes committed in the past to prevent crimes from being committed in the future, where pre-emption and predictive policing take center stage (Amoore 2013). The arguments in the debate are full of stories about past events, yet the truth of past events is not the issue. In this debate, past events are no more than rhetorical devices used to establish the truth of potential events that have not yet taken place. In other words, the debate concerns the causal link between uncontested past events and possible events in the future. It is not a question of convincing listeners of material facts, but about narratively establishing a trustworthy link between what has happened before and what will happen later, or in other words, narratively conflating past and future.

Surveillance Context

The background for this article is a documentary analysis of the debate on digital surveillance in Norwegian news media, print, and digital, from 2009 to 2017. I initially collected the material to prepare a more extensive Q-methodological study mapping the surveillance debate in Norway, Finland, and the UK. The analysis presented here results from subsequent narrative analysis of the collected material, independent of the Q-methodological study’s concourse. In Q-methodology, the primary purpose of the initial documentary analysis is to extract as many existing statements about the topic as possible. Throughout this work, I became increasingly aware of how those arguing for or against expanded capabilities seemed to use the same language to talk about different things. This awareness inspired me to perform a more in-depth narrative analysis of the material, which constitutes the groundwork for the analysis presented in this article.

I present a close reading of a few selected contributions to the public debate as it intensified in Norwegian media, leading to a legislative proposal in 2016. The texts chosen are typical examples of the argumentation in the material studied. The collected material for the overarching project encompasses printed and digital news media, broadcast and streamed debates, as well as blogs and social media. The examples analyzed in this article are collected from national newspapers, and all appeared both in print and on their digital platform. In other words, the debate as a whole is not restricted to a single platform, but exists across media and communicative forms. The participants in the debate analyzed were mainly representatives of state authority such as politicians and the police, and independent debaters such as lawyers, journalists, academics, and representatives of civil society. When analyzing media texts in this article, I understand narrative mainly as a property of the text. I use the authors’ names in the discussion primarily to mark context, not to make claims about authors’ intentions or ownership. This does not mean that the authors are insignificant; it is instead a methodological choice that highlights texts as actors and separates their lives from the lives of those who produce them, asking questions about what texts and documents do in the broader context. Every text is a negotiation between common understandings and individual interpretations (Bakhtin 1986).

I have chosen to focus on one isolated, legislative debate to provide clarity for the reader so that all texts cited relate to one specific topic. The immediate context of the chosen debate was a suggested change to the legislation on covert surveillance and intelligence in Norway to allow equipment interception, known in Norwegian as “data reading.”2 Norway is a country with traditionally high levels of trust in the authorities, as shown, for instance, in the bi-annual European Social Survey (Kleven 2016). The Norwegian police service is under normal circumstances unarmed, their surveillance methods are regulated by court orders, and the service is characterized by transparency. Despite a minor scandal in the 1990s, when it became known that the Police Security Service had been keeping members of the Communist Party under illegal surveillance for several decades, trust in the police among the general population remained high.

However, the events of 9/11 and the ensuing rhetoric of the war on terror that became prevalent in the West meant that Norway, which is a member of NATO, inevitably started to reconsider its public security and capabilities. Several “anti-terror packages” were passed in Parliament, such as legislation criminalizing the planning of terror, in addition to terror acts (for example, in Lov om endringer i straffeloven og straffeprosessloven mv., 2002). Even though the Snowden revelations of 2013 made it clear that Norway was still far behind countries like the USA and the UK in its invasive online surveillance capabilities, they also made it known which capabilities the Norwegian authorities lacked. The following years saw new debates concerning the expansion of surveillance capabilities, in particular online surveillance. The media debate that provides the foundation for this article must be understood within this context, as situated in a country that has historically been restrictive, though surrounded by allied countries with much more comprehensive capabilities and less transparency. However, I argue that even though the national contexts are very different—Norway is discussing expansion while the issue in its most powerful allies is restriction—the implied narratives of the debate are similar across countries.

Though characterized by seemingly irreconcilable views, the Norwegian media debate is played out in a society built on democratic values, free press, and high levels of trust. However, the global context in which surveillance narratives are actualized is not always so harmonic, as shown, for instance, by Anastasiya Astapova in her study of surveillance rumors in Belarus (Astapova 2017). From the outset, the context of Astapova’s study diverges from my study, as the kind of direct critical debate that takes place in Norwegian media is not possible in Belarus but is instead seen as separate reports being given in either the official, state-controlled media or the underground media run by dissidents. In other words, the implied narratives of “security” and “privacy,” arguing for or against surveillance capabilities, are recognizable as a framework, but the way it plays out in practice is highly context-specific. As Astapova points out, the lack of open debate prepares the ground for a highly developed tradition of rumor, legend, and jokes (Astapova 2017, 278). These types of narrative forms are not as widespread in the Norwegian context. A small interview study I performed with 17 employees and students at Norwegian universities and university colleges in connection to my research project on state surveillance indicated that those who do not have security or privacy issues as a part of their area of expertise have virtually no vocabulary for talking about the authorities’ surveillance. Astapova emphasizes the need people in undemocratic states have for creating an outlet for their fear, and this need is not present among middle-class Norwegian citizens, if perhaps among minority groups.

Despite the contextual difference in who tells the stories and what forms they take, it is clear that truth and belief are central aspects for understanding both the Norwegian and the Belarusian surveillance narratives. Astapova highlights how she struggled with getting her work acknowledged as “rumor” by fellow Belarusian scholars, as they all claimed that the stories told were true. In the Norwegian debate, surveillance is predominantly debated by experts and authority figures in media forms that are associated with credibility, such as debate articles and op-eds, and printed in acknowledged media channels, all of them established structures of expressing trustworthiness. In other words, truth claims lie at the core, even if the narrative techniques vary vastly. Astapova clearly states that it is not the role of folklorists to establish whether these stories are true or not, and I will add, the aim of this article is not to establish which of the two narratives of the Norwegian media debate are correct. Rather, this close reading of one part of the Norwegian media debate will present a vantage point for understanding the surveillance debate as an area of contrasting narratives that under certain conditions are drawn on and added to by implication, rather than through explicit narrative.

Contributors to the current surveillance debate often speak about terrorism and technology, the latter often framed as digitalization. In studies of the authorities’ surveillance capabilities, terrorism is often central to the argument, with 9/11 and the subsequent discourse on “the war on terror” as a threshold or starting point (Amoore 2013, Massumi 2010). Digitalization is most often actualized by the Snowden revelations, the rise of the Internet, or even the appearance of web 2.0 and smartphones (Lyon 2018, 2015). On the one hand, these are events that arguably have changed the way people in many parts of the world live and experience their lives. On the other, they are narrative actualizations of fears and possibilities that have a long history. The threats to our way of life from hostile groups (us versus them) or new technologies that might change life as we know it (for good or ill) can be seen as genres that people have tapped into for centuries to construct their collective identities.

In the analysis, I will focus on how technology is construed in the surveillance debate, with the digitalization of contemporary society as its material foundation. In the current debate, the development of digital technologies is a driving force, both for those who wish to utilize them for intelligence and investigative purposes and for those who warn against the intrusions into privacy they make possible. The narratives implied in the debate concern surveillance in a digital age, seeking to establish ground rules for the radically increasing possibilities of surveillance made possible by advancing digital technologies. For example, political commentator Marie Simonsen wrote in a national newspaper, “We all saw what proportionality means to the Police Security Service in the case of Ulrik Imtiaz Rolfsen. It ended with a crushing defeat in the high court” (Simonsen 2016). This kernel refers to an incident where the Police Security Service had a filmmaker making a documentary about terrorist recruitment in Norway illegally under surveillance. They ended up stopping him and two known radical Islamists on the way to the airport to send a recruit off to Syria, confiscating all of Rolfsen’s material both from that trip and from his home. When they sought approval for their actions after the fact, they were in the end sentenced to return all of the confiscated material unused after appealing through all three levels of the Norwegian court system. This story’s events connect to the terrorist threat, but that is not the underlying message, the implicit narrative that Simonsen uses this kernel to convey. Instead, she uses it to address the Police Security Service’s sense of proportion. Her aim is to conjure the story about an overly eager authority that will use these new and invasive technological possibilities to further encroach on its citizens’ privacy. Although ubiquitous in contemporary society, terrorism is, in this context, a narrative element used to increase the credibility of stories about digitized society. Technological development is the overarching issue that constitutes the common language used in the two narratives.

Previous critical scholarship on the discourse of expanded capabilities has mainly focused attention on the narrative of government authorities, a common point of view for critical social sciences whose main aim is to reveal power structures (see, for instance, Salter & Mutlu 2013). The discourse of those opposing expanded surveillance powers has not been subjected to as much critical scrutiny. The political nature of much critical social science means an overlap between academic scholarship and political argument on that side of the debate. The expansion narrative has arguably been the most influential in shaping global politics since 9/11, and hence the one that has had the most far-reaching practical consequences. The analysis in this article thus draws on prior critical scholarship that has shown how the discourse around the war on terror has given those arguing for expansion narrative tools to promote their story about what the world is like and how it could be made better (Altheide 2006, Amoore 2013, Massumi 2010). However, it is just as important to recognize how the opposing, privacy-focused line of argument is also a form of legendry—stories about another version of society, drawing on a set of shared assumptions about the effects of surveillance capabilities on society. When investigating the Norwegian surveillance debate narratives, my main aim is to highlight its communicative dimension as it appears to the public through popular media. Through a close reading of a few selected media texts from my material, I will aim to show that the two sides build their arguments around two completely different understandings of the issue that is under debate. It is not merely a matter of saying yes or no to specific surveillance capabilities but of making a case for one of two different worldviews.

Positioning: Crime, Human Rights and Digital Technologies

The rise of digital technologies and their role in conceptualizing the world is central to understanding the current surveillance debate. Although debaters regularly reference developing trends in crime and terrorism to support lines of argument, positions are always justified through technological possibilities. There are stories about technology enabling criminals and terrorists in new ways, as exemplified in an op-ed by a police officer who writes, “We have had events where the police have discovered indications of large amounts of child abuse material on a computer, but because of encryption, the police cannot access the material and use it as evidence. Worst case, the police will have to return the computer to the perpetrator, with the consequences this has for the victims” (Røger 2016). There also exist success stories about technological solutions helping in the prevention of crime and terror. On the side of those who argue against expanded capabilities, there are stories about technological development opening a space for free speech while simultaneously being a possible door to an unfree surveillance society. The utilization of digital technology as a novel force to make things happen is central to the way those engaged in the debate build their arguments through connecting and conflating past, present, and future events.

In 2016, during the months before data reading was incorporated into Norwegian legislation about covert surveillance and intelligence, the view of technology as something inextricably connected to crime was prominent in the public debate. Technology was often personified, figuring as the protagonist of the story, who kept pulling evidence away from under the police’s noses. “This is about giving the police access to information that they are entitled to have, but that technology is making unavailable to them,” wrote Anders Werp and Ulf Leirstein, politicians from parties within the coalition government, in an op-ed (Leirstein & Werp 2016). “Technological development has outrun our methods,” wrote the Minister of Justice and Public Security, Anders Anundsen (Anundsen 2016). Both these texts exemplify how those arguing for expanded capabilities often depict technology as an actor with the ability to affect society, almost of its own accord. Technology is narrated not as a tool but as an accomplice to criminals and terrorists. These are images from an implied narrative where the police have lost control. They frame the demand for more up to date digital surveillance methods as a response to a situation where “information we are entitled to have” is increasingly hidden by digital technologies.

The narrative implied in the texts arguing for expanded capabilities typically follows a linear, chronological structure. The past was a good time when the authorities were in control, the present carries the potential of mayhem, as criminals and terrorists might strike at any moment, and the future is a time when digital surveillance methods will once again keep criminals and terrorists at bay. In the text cited above, Leirstein and Werp begin by describing the present as a time when “the crime situation is changing. The threat of terrorist attacks is greater than before. Borderless crime is on the increase, with better-organized criminal groups.” There is an imagined starting point in this argument, where “the crime situation” was in an original state: terrorist attacks were few, crime was, for the most part, kept within limits, and criminal groups were poorly organized. As shown in the quote above, the claim is that this has changed because technology is making it harder to fight crime and prevent terrorism. They present recent developments in digital technologies as the event that the reader is expected to evaluate. It is narrated as something that has happened, with visible consequences and a clear-cut solution: the capability of data reading. These stories seek to disrupt the present and promote digital surveillance methods as the remedy to reinstate a situation of control. The credibility rests mainly on the ethos of the debaters. Speaking from the position of state authority, they talk about the “crime situation” and “information we are entitled to have,” invoking both empirical knowledge about facts on the ground for the police and judicial knowledge about what the authorities are legally allowed to do.

This narrative interprets technology according to who has access to it. When in the hands of criminals, technology is one of the “bad guys.” In the hands of the law enforcement and intelligence services, however, technology has positive connotations. Technology in this narrative can be interpreted as inherently neutral, but more importantly, it can take on the moral inclinations of the user. Technology in the hands of criminals will cause harm, while in the hands of the authorities, it will remove harm.

While arguments in support of expanded capabilities often build their claims to truth on the credibility of state authorities, the most common rhetorical strategy of those who want to limit surveillance is to undermine the credibility of those arguing for expansion. A political commentator in the national newspaper Dagbladet, Marie Simonsen, wrote, “The problem is that there is no documentation or statistics that show that the capabilities being requested work. The desire for expanded capabilities is based to an astonishing degree on anecdotal evidence (and of course secrecy) […]” (Simonsen 2016). What Simonsen does in this sentence is an example of “narrative positioning” (Oring 2008, 141). The text points to the narrative strategy of the opposing side, “exposing” it as storytelling rather than facts based on recognized measures of truth in modern society: documentation and statistics. In other words, arguments for expansion most explicitly rest on the position of the writer, their ethos, while the arguments against expansion more often position themselves using logos, building credibility not from an existing position, but through questioning the logic of the expansion narrative by playing to the reader’s rationality. Accordingly, Simonsen seeks to increase her narrative’s credibility by exposing that the expansion narrative is based on stories about random events.

Having brought the trustworthiness of the opposing narrative into question, Simonsen continues to bolster the credibility of her version by referring to several expert bodies who have opposed or warned against data reading official comments to the proposed legislation. She mentions that the Norwegian Bar Association, the Norwegian Union of Journalists and ICT Norway, an NGO representing the Norwegian private ICT sector, are all “extremely worried,” thus including three crucial sectors of knowledge as allies in her narrative: the law, the media, and the ICT sector. In doing so, Simonsen shifts from applying logos to ethos, drawing on the authority of her sources to emphasize that several other sources support her reasoning. The voices selected are all representatives of civil society, which further serves to tie the text’s claims to the implicit narrative that state authorities are the actual threat. She then condenses their answers into one headline: “The consequences of the proposed legislation have not been sufficiently assessed.” As in the extension narrative quoted above, “consequences” are central to the argument, and I will return to this later. In Simonsen’s narrative, “consequences” are something that will have a broad impact: “This applies to everything from source protection for the press to international trust in Norwegian trade and industry. Opening the way for the suggested surveillance, even purely technically, is a change that will have far-reaching consequences for society as a whole”.

Referring to the official comment made by ICT Norway, Simonsen argues:

If you base your arguments on technological development without boundaries, then you have to bear in mind that this is relevant in fields other than terrorism and criminality. Precisely because of technological development, the tools the police ask for are not limited to criminality. They touch all parts of society. (Simonsen 2016)

The reference to society concludes two consecutive paragraphs, both of which stress the idea that technology is all-pervasive. This narrative’s logic is that if we change our laws and regulations of technology in one sector, this will have ripple effects far beyond that area, simply because today, technology is not confined to one domain. In this narrative, technology is not a tool that will take on the moral inclinations of its user, but an inherently independent force that has effects far beyond those intended by its user. Thus, the way they talk about technology reveals a significant discrepancy between the two narratives. The expansion narrative divides the world into clear-cut categories of good and bad, assuming a fundamental link between human morals and technological practice. The opposition narrative describes an unstable and unpredictable world, where humans are neither good nor bad, and where actions concerning technology are framed as having effects far beyond those intended by the actor.

It is not coincidental that she repeats the reference to “society as a whole” with such force. The focus on human rights, freedom of speech, and other liberties of free and democratic societies make up the core theme of the worldview construed through arguments opposing the expansion of surveillance capabilities. To discredit the expansion narrative’s stories about the present, the arguments of the opposing side relate to a conflation of past and future, implying stories about what has happened in societies where authorities are allowed invasive capabilities. Rather than talking about the present, the opposition narrative talks about a possible future, building credibility for this future by referencing past events.

This is arguably the point at which it becomes evident that the two sides of the debate draw their arguments from different implied narratives: the insistence by those arguing for restriction that the discussion about surveillance methods is not about fighting or preventing crime and terror but about upholding a society built on freedom and trust. Their arguments refer to a completely different topic. This use of narrative positioning is most substantial in the arguments that oppose expanded capabilities. While the arguments for expanded capabilities are more often framed as a balance between the two narratives—considering stories about societal consequences but concluding that the fight against crime and terrorism is most important—the arguments against expanded capabilities frame the two narratives as contrasts.

In the text referenced above, Leirstein and Werp write, “Such a method [data reading] is of course controversial. But our view is that the consequences will be dire if we fail to face the challenges of crime” (Leirstein & Werp, 2016). Similarly, Anundsen writes

Working to give the police good tools, the government is engaged in creating a balance between the need for expanded capabilities on the one hand, and the right to privacy on the other. To secure sufficient capabilities for the police and the Police Security Service, we recently proposed legislation in Parliament that consider both these aspects. (Anundsen 2016)

These are examples of the way expansion arguments build credibility through balance rather than contrast. However, the balance always tips slightly towards expansion. Leirstein and Werp admit that the methods are controversial but conclude that the challenge of crime is the most critical consideration. Anundsen states that the proposed law has achieved a balance, suggesting that you can have both surveillance and privacy, though without detailing what that entails in this specific case. With the authority of a Minister of Justice, he expects the population to trust him. As Oring has pointed out, proposing alternative interpretations is a common way of increasing the credibility of the narrator (Oring 2008, 144). By showing that they are aware of the opposing side and their view, the narrators give the impression of having considered all options and carefully landed on their solution, without going into details or attempting to deepen how they came to their conclusions.

Of course, the opposition will always base their arguments on the positioning; that is, after all, inevitable for those in opposition. Thus, they employ the same rhetorical device of considering alternative interpretations, yet rather than emphasizing the importance of their own view, this side is more likely to emphasize the shortcomings of the other narrative. What is notable about the surveillance debate is that the arguments are often not about opposing the use of specific surveillance capabilities but about questioning the truth of the other side’s entire worldview, presenting an alternative future outcome rather than speaking about the present challenges. It is interesting to note how the two positions in the debate emerge as being “power” vs. “the people,” independently of which side views the debate. All the people debating these issues in the media are representatives of either state authorities or expert groups—arguably groups that in different ways hold power and authority in society. However, the stories they tell and how they position these stories make the debate read either like one between an oppressive state and a people demanding liberties or one between a state eager to protect its citizens and a group of elitist experts wanting to protect their interests. In this way, the debate ties into another move that has characterized the discourse on security and privacy in Western societies after 9/11 (Friedewald et al. 2017). During the twentieth century, this discourse was characterized by the fear of totalitarianism, highlighting the need to protect the citizens from the prying eyes of state authority. With the war on terror, the opposite view is taking hold, where the need for state authorities to protect its citizens against an outer enemy is gaining ground. This shift is foundational for the two competing narratives that underlie the surveillance debate.

A way to conceptualize the differences between these narratives is to look more closely at how they express their understandings of technology through the chronology of the narrative. One side construes technological development as an answer to a threat (present unsafe, future safe), while in the other technology is a threat (present neutral, future in danger); this serves to place the technology of digital surveillance methods in two different modes and create two different future scenarios. In the first instance, technology thematizes the development from an insecure present to a secure future. In the second, it thematizes the development from a secure present to an unsafe future. The narratives of what digital technologies might potentially do take divergent paths.

Leirstein and Werp’s text employs framing to create credibility for their story of an unsafe present and the way forward. They write:

There are several situations where criminals can organize the spread of information in such a way that the police are powerless to intervene. One example is when two criminals in the same room communicate without talking, by writing information on a PC in a document that is not saved. We have also seen cases where criminals have access to a common e-mail account, communicating by writing information that is saved as drafts. […] Using data reading, the police can register information that has not been saved, and can catch it before it is encrypted and becomes unavailable. Such a method is, of course, controversial. Nevertheless, our view is that the consequences will be dire if we avoid meeting the challenges of crime. (Leirstein & Werp 2016)

The text conjures the present through two stories about criminals taking advantage of technological tools to ensure no one gets hold of their communication and information. The stories are framed as “words-as-world,” using phrases like “there are several situations” and “we have seen cases where,” giving the impression of being references to fact rather than stories told these politicians by police officers (words-as-words) (Oring 2008,140). As we have already seen, Leirstein and Werp’s story depict data as information the police are entitled to have. These examples, stories about what criminals can do today framed as facts, exemplify their claim about what technology has taken away from the police. The framing insists that this is not about implementing new methods but about adapting technological advances to existing legislation—not so much introducing new capabilities as reinstating capabilities that they have lost. By invoking the idea that this is something the police have always been allowed to do, the text claims authority by referring to the legal system; it is not about changing but about adapting to reinstate. Not moving forward into something new but moving forward in order to return to a past state.

Central to the narrative is the unknown but dire consequences that will come about if we avoid taking action in the present. In other words, that scary possible future is already present. Amoore, drawing on Massumi, calls this a pre-emptive temporality, pointing out that it is not about predicting or preventing a possible future but about the capacity to act in the face of uncertainty (Amoore 2013, 62). In this sense, the future is narrated as a present worst-case scenario. The only way to secure a safe future for citizens in this story is to allow expanded capabilities in the present: through data reading, the police can access the information that they have a right to, thus solving the challenge of technological development and reinstating the control that existed in pre-digitalized society.

The opposing narrative grounds its arguments in a less troubling version of the present. Simonsen’s text altogether refuses to recognize the present of the expansion narrative, claiming that its evidence is anecdotal. Instead, she tells a story about a present where the government seems to be staging a charade. She claims that there is no convincing evidence that things are as bad as the authorities claim. Simonsen’s present, which is illustrative of that found in the opposing side’s narrative, is one where technology is regulated, and it is regulation that keeps society secure. The temporality of the risk assessment inherent in this argument highlights pre-caution rather than pre-emption. For those opposing expansion, it is not a question of doing something now to avoid future events, but not doing something now because we are not sure how it will affect future events.

The word “consequence” is central to both lines of argument, and it is negatively loaded on both sides. For the expansion side, it is something that will happen if we do not expand. For the opposing side, it is something that will happen if we do expand. Neither side use the word “consequence” to describe the desired outcomes. This use of words implies that either narrative’s desired effects are a natural move forward, while the effects that they warn against are “consequences,” effects that disrupt the natural flow of society.

Both sides narrate consequences as a conflation of present and future. In the expansion narrative, they might well be playing out as we speak. The opposing narrative locate consequences materially in the future, yet although they are contingent on increased surveillance, they are narrated as a threat that is already felt. The main difference in this material is that, in the first narrative, consequences will be avoided using data reading while, in the second, consequences will be brought about by data reading. In the opposition narrative, something might change, but we do not know what. The foundation of this narrative is that this uncertainty could lead us either way. Technology is not a threat in the hands of the “bad guys” and a source of good in the hands of the “good guys,” but an independent force that can have unintended effects. In other words, data reading has the power to change the future, but the problem is that we do not know how. On the other hand, the expansion narrative narrates data reading as a force that has the power to change the present, which is narrated as conflated with the future. Here, “consequences” are possibilities that might be playing out at any given moment, but that can and will be avoided if the police are allowed to make use of data reading.

Pathos: Fear, Terrorism and the Surveillance State

Above, I have shown how arguments in the surveillance debate employ strategies that draw on two different implied narratives to persuade people to support or oppose expanded surveillance capabilities. These strategies mainly relate to the ethos and logos of the debate: that which has to do with either speaker or events’ trustworthiness and authority. I will now examine another crucial part of the surveillance narrative, which belongs rhetorically to the domain of pathos. These are the arguments performed to build credibility by making people feel that what is said is the truth, by playing to ideology or belief or existing emotions (Oring 2008, 157). In the debate about surveillance capabilities, fear is the underlying emotional driver. In my material, both those arguing for data reading and those opposing it base their case on the fear created by conjuring up images of a threat. However, that does not mean that “fear” as a narrative element in the debate relates to the same stories. Rather, “fear” constitutes another element that highlights the discrepancies between expansion and opposition narratives.

The discourse of terrorism is a fundamental element in the expansion narrative in contemporary society. The terrorist attacks of 9/11 mark a threshold for the perception of threat in Western society, with ripple effects worldwide (Lyon 2003, Massumi 2010). As noted above, the discourse of state power has been the main subject of scrutiny for critical scholarship, and this discourse has centered on terrorism. Because of this, terrorism has emerged as a central topic of study. However, I would argue that terrorism is the central theme of only one of the two narratives that constitute the debate. As argued above, the opposition narrative is structured around an understanding of the state as a threat, such as the stories spelled out in Astapova’s study on Belarus. In other words, the opposition narrative is characterized instead by the lack of stories about terrorism. Terrorism shows up in the opposition narrative as a device for narrative positioning, but it is not a central element of the narrative structure in the same way as it is in the expansion narrative. Here, the topic of terrorism is a driving force for constructing a worldview where the present is unsafe, and only the state authorities and their access to new surveillance technology can save us. However, for both sides, fear is an emotional driver. The narrators on both sides use pathos to make future events real by playing to the readers’ cognitive and emotional expectations and connecting these to prevalent fears of terrorism.

David L. Altheide has argued that a discourse of fear in many respects won hegemony in US news coverage in the 18 months after 9/11. For example, he quotes the news anchor, Dan Rather, saying to a British journalist: “[…] one finds oneself saying ‘I know the right question, but you know what? This is not exactly the right time to ask it’” (Altheide 2006, 10). Rather reportedly felt pressure from the authorities to report stories consistent with their narrative, thus drawing on and simultaneously re-creating the media climate that Altheide calls a “discourse of fear,” where most major news broadcasters backed the official narrative to gain support for the invasion of Iraq. Altheide concludes that the 9/11 terrorist attacks so inflated the expansion narrative that even those in the media industry who were initially critical of it felt it was the wrong time and place to speak up against it.

The discourse of fear remains a recognizable description of the surveillance debate almost twenty years later, though it takes slightly different forms. The nature of debate since the 9/11 attacks has inevitably changed, but terrorist threats are still a central part of the expansion-narrative regarding surveillance. In what follows, I will look more closely at how the discourse of fear in the Norwegian media follows these dominating narratives: those arguing for expansion draw on fear of terrorism, while those opposing expansion draw on fear of the curtailment of civil liberties and democracy in the wake of terrorism.

A significant issue in the public debate of post-9/11-society is whether it is appropriate to introduce expanded capabilities in the wake of terrorist attacks. A pertinent example of the juxtaposition of the two sides on this issue is found in a newspaper article published on 18 November 2015. This was six months prior to the passing of the legislation regarding data reading but, more significantly, only a few days after the terrorist attacks in Paris targeting a concert hall, a football stadium, and several restaurants, in which 138 people were killed and many more injured (Stensrud 2015). The protagonist in the newspaper article is Ulf Leirstein, a parliamentary politician and the author of one of the media texts discussed above. His message is that he is in favor of giving the police access to data reading. The first paragraph calls attention to the fact that the legislative proposal being drawn up by the Department of Justice and Public Security for consideration in Parliament early the following year was planned prior to the Paris terrorist attacks. However, the timing of the newspaper article is closely related to the terrorist attacks. Leirstein is quoted as saying: “When these things happen is precisely the moment that we should debate them. This is when we see the holes in our own system” (Stensrud 2015). This statement is balanced by a quote from the leader of the parliamentary justice committee and opposition politician, Hadia Tajik. She says: “It is important that we do not make the mistake that we have seen other countries make in the wake of terrorist attacks, where the boundaries of what is seen as right are moved because they are debated in a climate of fear” (Stensrud 2015).

Although such an overt call for expanded capabilities in the wake of serious events is not common in Norwegian media, I argue that the rhetoric of Leirstein’s statement exemplifies how fear is thematized through the expansion narrative. What Leirstein is saying is not that it is good to take advantage of people’s fear, but that terrorist attacks give us examples of things we should fear—and by extension, things we should take measures to prevent. It is an extreme version of the examples dismissed by the opposition as anecdotes: stories about real-life events narrated or referred to concretize the implied narrative of a society that needs extended capabilities. The Paris attacks function as a kernel story through Leirstein’s reference to “these things” that “happen,” pointing to events that everyone agrees have happened while simultaneously conjuring possible future events by referencing the overarching implicit narrative of a present where the Norwegian authorities need expanded capabilities in order to provide security for its citizens. Leirstein uses pathos not to establish the truth of the events he refers to but to establish the truth of events that might come, implying the story of an unsafe present and a possible safe future of the extension narrative. His arguments make a claim for belief beyond belief.

As established, the expansion narrative builds on a stable chronology—secure past, insecure present, reinstatement of security in the future. Invoking a past event to conjure present fear thus disrupts the flow of the narrative. Examples of disastrous past events introduce an alternative chronology, that narrate insecurity not as a conflation of present and future but as a conflation of past and present that differentiates between near and distant past. In using events of the near past as examples, the narrator introduces an alternative past that does not belong to the overarching narrative structure but rather works as a template for future action. The secure, distant past evoking post-digitalized society is positioned against an insecure, near past narrated as stories about terrorist attacks or other serious events, with the effect not of dismissing the narrative structure but of building credibility for the claims for expanded capacities. In this way, Leirstein’s use of the Paris attacks function as a secondary legend (Oring 2008, 153), using the story of a real event to build credibility for the existence of similar future events.

The quote from Tajik is illustrative of the opposition narrative. Her main point is that one should base fundamental societal changes on rational consideration, not decided upon in affect. She thus dismisses the emotional pathos of Leirstein’s argumentation by playing to the cognitive expectations of the readers, acknowledging their fear, before invoking the modern ideal of the rational human. This move, which is a common rhetorical move among those opposing expansion, could be interpreted as a way of rationalizing the expansion side’s focus on fear. On the other hand, there is a double effect to this claim, which also plays to a prevalent fear, albeit a different type of fear from the expansion narrative. As is usual in the opposition narrative, Tajik does not discuss possible terrorist attacks, apart from discrediting the push to expand capabilities in affect. Instead, she tells a story of a society that might be moving in the wrong direction democratically: “We have to avoid implementing invasive measures that might affect individuals who have done nothing wrong, in order for politicians to appear efficient in a demanding debate,” Tajik states (Stensrud 2015). Politicians instill fear to win cheap political points, she claims. Nevertheless, in so doing, she points to a different threat: a lurking worry about the current democratic climate. Her statement refers to the story of a political climate where politicians might use terrorist events to scare people into allowing access to technologies that might compromise their freedom.

Tajik employs the same strategy of narrative positioning observed in the Simonsen text: building credibility for one’s narrative by exposing the narrative of the opposing side as false. However, what becomes clear is that, strategically, this type of narrative positioning also comes within the trope of pathos. Discrediting the expansion narrative is the same as questioning the motives of the authorities: it might not create immediate fear in the masses, but it plays to those who are already critical of the current authorities, meeting both the emotional and cognitive expectations of the listeners by conforming to their ideology and belief (Oring 2008, 157).

It is interesting to note that one could even read Leirstein’s statements as balancing between addressing cognitive and emotional expectations:

We shouldn’t get carried away and put up surveillance cameras on every street corner. Like Hadia, I too am worried that privacy is easily dismissed in the debate. Because of this, it is important that a third party, in other words the courts, should decide on each case when it comes to data reading. (Stensrud 2015)

Although employing the same terminology as Tajik to address privacy and independent control, Leirstein does not abandon his arguments about debating counter-terrorist measures in the wake of terrorist attacks. He is not saying that we have nothing to fear. Indeed, he stresses the idea that we have something to fear, arguing for the implementation of invasive measures even when they pose a risk to privacy. Stressing the importance of privacy in this setting amplifies the threat—it is introduced not as an element to consider in its own right, but as a rhetorical tool to build credibility for the expansion narrative, connecting to the cognitive expectations of the listeners whose emotions have already been stirred by the fear of terrorist attacks. The conclusion is that, even though privacy is important, the current situation is, in fact, so dangerous that we need to allow these invasive capabilities.

The interplay between cognitive and emotional expectations is thus a rhetorical strategy used by both sides. In arguments drawing on the expansion narrative, this often takes the form exemplified above: an imminent threat is presented and exemplified by reference to crime or terrorism, followed by reassurances that the authorities can and will handle this if they only get the capabilities they are asking for. First, playing to a prevalent fear, and then presenting a rational solution through addressing the citizens’ trust in state authorities. A particularly pertinent example of this is seen in the words of the Minister of Justice and Public Security, Anders Anundsen, from whom there is a brief quote in the above section. The text is an op-ed about data reading, published 13 May 2016, with the title “They shall never win.” It was published a few months after the Department of Justice and Public Security circulated their legislation proposal about data reading for comment, one month before the Parliament was to vote on it, and, significantly, two months after the terrorist attack on Brussels airport. The first paragraph read:

“It wasn’t supposed to happen” was the refrain after terror struck Brussels in Easter week. Some claim the fight against terror isn’t working. I don’t agree. Even if we can never protect ourselves completely against terrorist attacks, we must not allow the idea to take root that we cannot win the fight against terrorists. Because we can. (Anundsen 2016)

The introduction loses no time: it goes through the full cycle of destabilizing the past/present by mentioning both an actual event and the assumed feeling of fear and hopelessness and then seeking to reinstate hope and trust through a call to arms. The final sentence is an appropriation of the 2008 Obama campaign “yes we can,” alluding to the Norwegian interest in and dependence on US politics, while also conjuring the feeling of hope that was felt even in Norway with the election of the first African American president in the USA. In the context, however, the use of this phrase could be read as an inversion of the values connected to the original statement. While the election of the first African American president in the USA was undoubtedly a victory relative to a long history of state oppression, the Norwegian Ministry of Justice’s fight against terrorists does not reflect the same power relation. In this way, the relative distance to US politics and history makes this allusion work in a Norwegian context; it alludes to the hope felt with the election, but it does not carry its historical implications. This distance allows Anundsen to employ emotions that arguably belong to the opposition narrative and use them to build credibility for the extension narrative.

The piece continues to focus on the work that is being done in Norway to prevent terrorism. “Terror is criminality, not a clash of civilizations,” he writes, arguing that criminality can be prevented through international cooperation, stricter legislation, and intelligence. There is only one obstacle:

To face the terrorist threat, the police and security services must have access to information and intelligence. The capabilities made available by current legislation are to some extent outdated. Today, the police and security service are allowed, within strict limits, to monitor communication between individuals, after a court order. The challenge is that criminals can easily use encrypted communication. This obstructs the police and security service in their work. Technological development has outrun our methods. (Anundsen 2016)

Even though the piece’s dominant mood is one of hope, this excerpt makes it clear that hope will only be within reach with expanded capabilities. In the current situation, the police and security services are, in fact, “obstructed.”

Here too, there is an interplay between addressing the emotional and the cognitive. While the text justifies the need for expanded capabilities with reference to events—real or imagined—the claim that the government has made a balanced assessment is merely stated. This is another example of how the expansion narrative balances alternative interpretations, as discussed in the above chapter, but in this case, it becomes clear that this use combines logos with pathos. It seeks to convince people of the value of data reading by alluding to stories about a shared fear: terrorist attacks. At the same time, however, that fear is framed by making statements that, on the one hand, radiate confidence—”we can!”—and on the other, play to an audience that already trusts the authorities. As Oring points out, the use of pathos to convince is arguably most powerful if one can manage to invoke emotions that are already being felt (Oring 2008, 157), and when representatives of state authority in Norway speak, they assume that they have the majority of the citizens on their side.

The piece ends with two pretentious paragraphs:

The goal of terror is to incite fear and destroy our way of life. Terrorists will never succeed. We will stand tall in our democratic tradition and protect our values. We must never yield in the fight against terror, but even so, we must not let the fight against terror undermine the values we are fighting for. We must avoid a situation where we no longer dare to make use of our constitutional freedom of speech for fear of repercussions.

Conflicts and societal differences must not lead to an increase in support for terrorists. The fight against terrorism is therefore also a fight against parallel societies and a new underclass. Knowledge about different social, cultural and religious groups, as well as dialogue with these, will be important factors in the fight against terror. A fight we will win. (Anundsen 2016)

The op-ed thus ends on the same note it started on, with the encouraging promise, “we will win!” Nevertheless, the premise of rhetoric encouraging us to fight is inevitably a battle going on. The text is not written to calm and reassure; the pathos of the text rather reinforces that in the present, we do have something to fear. It draws on and reinforces the implicit narrative of an unsafe present and a possible safe future, arguing that if we make the right decisions now, we will overcome this situation in the future. Fear is the driving force of this text, as it confirms the presence of danger by presenting a solution to it. If there were nothing to fear in the current situation, why would we need a motivational text about fighting and winning? Framed as a story of hope, the narrative implies looming danger, the existence of possible terrible events that can only be averted if new capabilities are allowed. Although written using a markedly different language from that of the texts quoted above, which speak of “dire consequences” and “holes in our system,” it nevertheless operates within the same rhetoric of truth to imply a narrative about a present that is unsafe, but that can be fixed if the police and security service get technology on their side.

Conclusion

The performers’ intention to convince is a common feature of both public debate and legendry. Although the media debate on surveillance is not made up of legends in the traditional sense, the texts could be read as legendry. They are expressions that draw their claims to truth from relatively fixed belief systems, defined here as implied narratives. What connects the media texts studied in this project to legendry is not that they take the form of narratives but that they consistently make claims for the truth about future events—events that must be the stuff of imagination, of storytelling. It is not the events discussed or referenced in the texts that are up for debate; it is the consequences of these events and the foundation for imagining what would happen if surveillance capabilities were expanded or restricted. As exemplified above, the narratives implied by reference to past events are both stories about life in a society whose flow is disrupted by digitalization, but digitalization takes on two very different guises. One narrative is about an insecure society where the authorities need to re-establish control by having access to surveillance technologies. The other narrative is about a relatively well-adjusted society threatened by uncertainty as the authorities gain access to more and more invasive surveillance technologies.

These are narratives springing from an unprecedented technological development, seeking to define culture and society by competing to establish the truth about how technology affects our way of life. Will access to our online communications help the authorities keep us safe from a number of defined dangers, or should we restrict such access because it may be misused, even though we trust our current authorities? Should we fear terrorist attacks, or should we fear the fear that terrorist attacks instill in us? These texts negotiate how to deal with the elusive, new technologies that enable intensified surveillance of electronic traces and communications. The questions arise in the wake of legislative changes, as exemplified through the analysis above. However, when looking deeper into the rhetoric of truth employed in these contributions, it becomes clear that they raise another more fundamental societal issue. It is not just a question of being for or against a specific method. It is also a question of owning and defining the story of what these new technologies can and will do to our society. Although using the same terms and language, a closer look at the rhetoric of truth of the debate indicates that the two sides of the debate do not relate to the same story. In this way, the media debate on authorities’ access to digital surveillance provides a perspective on the more extensive debate in our society of how to understand and regulate technology, and most specifically in this context, highlights how the complexities and uncertainties of this issue make the debate and its argumentation rest on narrative.


Notes

1Because the concepts “security” and “privacy” as descriptions of reality are highly contested, I will not use the terms as analytical concepts in the analysis. In order to distance my analysis from the debate over these concepts and avoid slipping into the language of a false dichotomy, I will rather refer to the positions of the debate: security expressed through arguments for expanding surveillance capabilities, privacy expressed through arguments opposing expansion. [ Return to the article ]

2Equipment interference, called “data reading” in the Norwegian debate, is a technique whereby the authorities enter a digital device covertly in order to monitor everything that happens on the device in real time. [ Return to the article ]


Works Cited

Altheide, David L. “Terrorism and the Politics of Fear.” Cultural Studies Critical Methodologies 6 (4) (2006): 415–439.

Amoore, Louise. The Politics of Possibility: Risk and Security Beyond Probability. Durham: Duke University Press, 2013.

Anundsen, Anders. “De skal aldri vinne.” [They shall never win] [Online] Available from: https://www.nrk.no/ytring/de-skal-aldri-vinne-1.12943569. Accessed 15 March 2018.

Astapova, Anastasiya. “In Search for Truth: Surveillance Rumours and Vernacular Panopticon in Belarus.” Journal of American Folklore 130, no. 517 (2017): 276-304

Bakhtin, Mikhail M. Speech Genres and Other Late Essays. Austin: University of Texas Press, 1986.

Blank, Trevor J. Folk Culture in the Digital Age: The Emergent Dynamics of Human Interaction. Logan: Utah State University Press, 2012.

Dégh, Linda. Legend and Belief: Dialectics of a Folklore Genre. Bloomington: Indiana University Press, 2001.

Frank, Russell. Newslore: Contemporary Folklore on the Internet. Jackson: University Press of Mississippi, 2011.

Friedewald, Michael, J. Peter Burgess, Johann Čas, Rocco Bellanova and Walter Pessl (Eds.). Surveillance, Privacy and Security. Citizens’ Perspectives. London: Routledge, 2017.

Kalčik, Susan. “’... Like Ann’s Gynaecologist or the Time I Was Almost Raped’: Personal Narratives in Women’s Rap Groups.” Journal of American Folklore 88 no. 375 (1975): 3-11.

Kleven, Øyvin. “Nordmenn på tillitstoppen i Europa.” Samfunnsspeilet 2 (2016): 13-18.

Leirstein, Ulf and Anders Werp. “Nye metoder for å bekjempe alvorlig kriminalitet.” [New methods to fight serious crime] [Online] Available from: https://www.dagbladet.no/kultur/nye-metoder-for-a-bekjempe-alvorlig-krimi-nalitet/60157123. Accessed 15 March 2018.

Lincoln, Bruce. Discourse and the Construction of Society. Comparative Studies of Myth, Ritual, and Classification. 2nd ed. Oxford: Oxford University Press, 2014.

Loader, Ian and Neil Walker. Civilizing Security. Cambridge: Cambridge University Press, 2007.

Lov om endringer i straffeloven og straffeprosessloven mv. [Act about amendements to the Penal Code and the Criminal Procedure Act] (lovtiltak mot terrorisme - gjennomføring av FN-konvensjonen 9. desember 1999 om bekjempelse av finansiering av terrorisme og FNs sikkerhetsråds resolusjon 1373 28 September 2001), 2002. Besl. O. nr. 77 (2001-2002)

Lyon, David. Surveillance after September 11. Cambridge: Polity Press, 2003.

Lyon, David. Surveillance after Snowden. Cambridge: Polity Press, 2015.

Lyon, David. The Culture of Surveillance: Watching as a Way of Life. Cambridge: Polity Press, 2018.

Massumi, Brian. “The Future Birth of the Affective Fact: The Political Ontology of Threat.” In Digital and Other Virtualities: Renegotiating the Image edited by Antony Bryant and Griselda Pollock, 79-92. London: I.B.Tauris, 2010.

Mould, Tom. “The Welfare Legend Tradition in Online and Off-Line Contexts.” Journal of American Folklore 129, no. 514 (2016): 381–412.

Oring, Elliott. “Legendry and the Rhetoric of Truth.” Journal of American Folklore 121 no.480 (2008): 127-166.

Oring, Elliott, “Legend, Truth, and News.” Just Folklore. Analysis, Interpretation, Critique, 91-103. Los Angeles: Cantilever Press, 2012.

Røger, Sigve. “Dataavlesning er nødvendig for å hindre at kriminelle får herje fritt i digitale rom. ” [Data reading is necessary to prevent criminals from wreaking havoc in the digital sphere]. https://www.aftenposten.no/meninger/debatt/i/9moV9/dataavlesning-er-noedvendig-for-aa-hindre-at-kriminelle-far-herje-fritt. Accessed 24 September 2020.

Salter, Mark B. and Can E. Mutlu (Eds.). Research Methods in Critical Security Studies: An Introduction. London: Routledge, 2013.

Simonsen, Marie. “Overvåking: Vet Stortinget hva de er bedt om vedta?[Surveillance: Does the Parliament know what they have been asked to pass?] [Online] Available from: https://www.dagbladet.no/kultur/overvaking-vet-stortinget-hva-de-er-bedt-om-a-vedta/60155324. Accessed 15 March 2018.

Stensrud, Kenneth. “Vil la PST lese tastetrykk.” [Wants to allow the Police Intelligence Service to read key strokes] Dagsavisen, 18 November, pp. 14-15.