Introduction
Although it is a relatively new concept,Footnote 1 recent years have witnessed a significant growth of academic and political interest in cyberterrorism. Within the scholarly literature – in common with terrorism research more broadlyFootnote 2 – two discussions have been particularly prominent. First, is the question of definition. Although Dorothy Denning’s relatively narrow understanding of cyberterrorism remains the best-known and most widely-used,Footnote 3 considerable debate continues around the respective merits of restrictive and flexible conceptualisations of this term.Footnote 4 The second prominent debate relates to the threat posed by cyberterrorism to various referents including, inter alia, national security, corporations, and ordinary citizens.Footnote 5
Concerned assessments of the cyberterrorism threat highlight infrastructural and sociopolitical vulnerabilities at risk of exploitation by appropriately resourced and intentioned actors. Approached thus, cyberterrorism emerges as a relatively straightforward danger of potentially catastrophic significance. Hyperbolic scenarios are widespread here, especially the risk of a ‘Digital Pearl Harbour’: ‘a debilitating, full-scale digital assault – in which multiple attacks are launched against telecommunications networks, city power grids, and/or air traffic control systems, causing widespread destruction and possible loss of life’.Footnote 6 Such scenarios point, first, to the widespread construction of cyberterrorism as a generalised and homogeneous threat within this ‘concerned’ literature: a threat that remains insufficiently unpacked within academic – as much as media – analysis. And, second, to the importance of figurative language within the framing of this threat. As with terrorism discourse more broadly,Footnote 7 metaphors employed to make sense of cyberterrorism work to (re)produce that to which they appear to refer, often with tangible discursive and political implications:
Metaphors matter. We choose our words from within a dominant system or frame of metaphor that offers us a specific lexicon of language, that defines words in certain specific ways, and shapes both the ‘what’ and the ‘how’ of our communication. In this way, figuratively and often literally, through metaphor we make meaning. Our most common metaphors help us to understand problems and conflicts in certain ways, offering us certain available responses, and negating or obscuring others. Metaphor operates in the realm of thought, but its workings reverberate in concrete, active, tangible ways.Footnote 8
Less troubled analyses of the cyberterrorism threat, in contrast, tend to justify a more sceptical stance through cost-benefit analysis,Footnote 9 or via inductive reasoning from existing knowledge of the likelihood and consequences of offline terrorist attacks.Footnote 10 A common – although not universal – theme within these more sceptical assessments of the cyberterrorism threat is the importance of discursive renderings thereof. That is, for a number of authors, constructions of the danger posed by cyberterrorism themselves help to constitute or create this phenomenon as a significant and urgent security challenge. Cyberterrorism, in other words, might be thought of as a social construction rather than an extra-discursive reality: its existence is (at least in part) a product of meaning-making practices associated, variously, with political rhetoric, popular culture, cyber-security corporations, or the news media.Footnote 11 As Francois Debrix argues, comparing contemporary representations of cyberterrorism with the ‘techno-strategic’ nuclear weapons discourse explored in Carol Cohn’sFootnote 12 pioneering Cold War study:
The language of cyberterrorism mobilized by the media and its so-called experts is quite technical for sure. But this technicality, far from de-realizing the threat, makes it possible. It realizes it in the mind/psyche of the public who is subjected to the simulated scenarios and mediations. The taxonomy of cyberterrorism and its technocratic language allow the public to recognize that there is a threat, and that this threat, as presented to them by the media, will surely cause serious casualties within the population.Footnote 13
For Myriam Dunn Cavelty, more forcefully:
While governments and the media repeatedly distribute information about cyber-threats, real cyber-attacks resulting in deaths and injuries remain largely the stuff of Hollywood movies or conspiracy theory. In fact, menacing scenarios of major disruptive occurrences in the cyber-domain, triggered by malicious actors, have remained just that – scenarios.Footnote 14
Critical analyses such as these are important for two reasons. First, they indicate the pervasiveness and prominence of (representations of) ‘cyberterrorism’ across diverse social, political, and cultural sites – from Hollywood movies to political language, media coverage and beyond. Second, they also serve to open significant new research questions beyond those surrounding the issues of definition and threat noted above. These include questions around the content, framing and reception of (various) ‘cyberterrorism’ discourse(s), as well as the constitutive or causal power such discourses might possess.Footnote 15
This article builds on this work on constructions of cyberterrorism, approaching representations of the definition, threat, and characteristics of cyberterrorism as vital to its emergence as an object of knowledge. Specifically, it offers a detailed empirical analysis of an original research sample of 535 items published by 31 international news media outlets between 2008 and 2013. This analysis extends this work above by focusing attention on the considerable – and often overlooked – variances in (news media) constructions of this threat and the importance of these via investigation of four research questions that emerge from the above debates on definition and danger. First, how individuated a presence does cyberterrorism occupy within news coverage making mention of this threat? For instance, does cyberterrorism receive sustained, differentiated analysis within relevant news items, or is it collocated with other cyber-security challenges or, indeed, other types of terrorism when it is discussed? Second, how significant a threat is cyberterrorism deemed to pose? Is the news media dominated by ‘concerned’ or ‘sceptical’ portrayals of this phenomenon, and are there variances across time or space? Third, how are ‘cyberterrorists’ identitied – or given identity – within this coverage? Are they, for instance, portrayed as professional or amateur, as politically motivated, as non-state actors, and so on? And, fourth, who or what is positioned as the referent – the entity threatened by cyberterrorism? Does the perceived vulnerability of particular actors or infrastructures feature more prominently in the news media than that of others?Footnote 16
Our analysis of these questions is intended to make two immediate contributions to existing scholarship. The first is to add empirical detail to relevant literature on cyberterrorism and cybersecurity more broadly. Although numerous scholars have highlighted the importance of news media discourse in relation to this threat,Footnote 17 this study is the first of its size to be focused specifically on constructions of cyberterrorism. As detailed in the conclusion, our findings open considerable space for further research, including comparative analysis of different media. The article’s second contribution is to add analytical depth to existing accounts of the content and framing of cyberterrorism discourse. To do this, we focus our analysis on three factors, which, in our sample, intervened in media assessments of the threat posed by cyberterrorists. These are: (i) specificity – referring to the degree of individuated attention afforded to cyberterrorism within a particular news item; (ii) status – the type of actor positioned as a would-be cyberterrorist; and, (iii) scale – the size of referent threatened by cyberterrorism. As demonstrated below, anxiety about the threat posed by cyberterrorism tends to increase with the level of individuated attention afforded to this phenomenon in a particular news item. Moreover, particular constructions of cyberterrorists – as either professional or unskilled actors – and geographically larger referents – especially ‘the West’ or the entire globe – are also associated with heightened levels of concern than their counterparts. As this suggests, our research points to a far more fractured, heterogeneous and variegated cyberterrorism discourse than is frequently identified by critics of media hyperbole in this context; a discourse in which variations matter a great deal. Indeed, as shown below, the most apprehensive media assessments of cyberterrorism tend to draw upon a very specific construction of cyberterrorism as the activity of either professionals or unskilled actors targeted at ‘the West’ or the entire world. These constructions also, importantly, typically differentiate cyberterrorism from other threats such as cyberwar, and draw heavily upon perceived sources of authoritative expertise from industry, politics, or academia.
Constructing (cyber)terrorism
Recent years have witnessed the emergence of a diverse, broadly constructivist literature that seeks to excavate, situate, and deconstruct the diverse ways in which the phenomenon of terrorism has been discursively produced.Footnote 18 Associated with, but extending beyond, so-called ‘critical terrorism studies’,Footnote 19 this work has demonstrated the significance of film, television shows, video games, executive political speech, legislative debate, the news media, and much else besides to explore how terrorism: ‘as a social fact … comes into being within, and is dependent upon, the contextual, historical and political dynamics that structure its interpretation thus’.Footnote 20 The point of this research, as Jacob L. Stump and Priya Dixit argue, is not to question whether terrorism – or, here, cyberterrorism – exists. Rather, to shift the analyst’s gaze from (cyber)terrorism as brute material reality to ‘(cyber)terrorism’ as discursive production, in order: ‘… to study how representations of terrorism and their reality are socially produced through linguistic and non-linguistic practices.’ As they continue, ‘Explaining terrorism, to borrow from Jonathan Potter, is not a matter of ‘denying the existence of tables’ or terrorists, but rather a matter of ‘exploring the various ways in which their reality is constructed and undermined’.Footnote 21
In common with other constructivist work on security, it is the constitutivity of competing claims relating to terrorism that is emphasised within this literature.Footnote 22 The multitude of different definitions and understandings of terrorism that are observable across a wide array of discursive spaces have a productive effect as opposed to one that is merely reflective or descriptive. That is, the threat of terrorism, as well as the identity ‘terrorist’, are not made with reference to some extra-discursive, objective ‘reality’ of terrorism; they are, rather, created through such assertions (albeit with variable levels of stability). Importantly, such attempts to secure the meaning of terrorism are themselves also: situated in specific etymological and other genealogies; dependent on various intertextual relations; made manifest through identifications of similarity and difference between terrorism and other risks; and situated in a diverse range of (open, yet contested) social and cultural contexts. Terrorism and other security issues are, therefore, not a ‘given’Footnote 23 but instead a discursive and social production. Any effort to define terrorism, or to assess the severity of the threat that it poses, does so at the expense of overlooking the constructed and contingent character of this phenomenon.Footnote 24 Approached thus, the task for analysts is not to explore the accuracy of dominant discourses on (cyber)terrorism, nor even to trace their causal impact upon policymaking in the ‘real world’. Rather, it is to describe and unpack the material from which such discourses are created and to highlight moments of instability, paradox, contradiction, and heterogeneity therein.Footnote 25
This article furthers this research by presenting findings from a discursive analysis of a corpus of news media items on cyberterrorism published across a total of 31 different media outlets between 1 January 2008 and 8 June 2013. This sample included media hosted in Australia, China, India, Russia, Singapore, the United Kingdom, and the United States, and spanned broadsheet newspapers, tabloid newspapers, and the websites of media production companies.Footnote 26 These media outlets were selected for a combination of purposive factors which included: accessibility to the researchers, including the presence of an internal online archive of content; diversity of political perspective and type of media corporation, given that concerns around privacy and liberty are particularly prominent within cybersecurity discourse; total size of readership, where possible favouring publications with the highest circulation figures; diversity of geographical origin in order to facilitate international comparison; and language, such that the news content was provided in the medium of English.Footnote 27 The corpus of news items identified across this sample was subsequently generated via a key word search for the terms <cyber terrorism>, <cyberterrorism>, and <cyber terror> on the internal search engines of these sites, where possible.Footnote 28 This established a total of 535 relevant items for analysis. As detailed further below, these items varied considerably, and included news articles on current affairs, technology stories, opinion pieces, editorial discussion, and analysis of cultural events, including reviews of fictional depictions representations of cyberterrorism.Footnote 29
2008 and 2013 were fixed as the parameters for this research for two reasons. First, because this provided us with sufficient data through which to explore developments within media discourse on cyberterrorism: a total of 1986 days of content. Second, because this period also incorporated a number of potentially relevant events that attracted considerable media interest as they unfolded.Footnote 30 These events included the cyberattacks on Georgia that took place in the summer of 2008; the 2010 identification of the Stuxnet attack against nuclear centrifuges in Iran; the 2010 publication of the UK’s National Security Strategy in which cybersecurity threats occupied a place of prominence, and the subsequent release of the UK’s Cyber Security Strategy: Protecting and Promoting the UK in a Digital World in 2011. Following data collection, each of the 535 news items was coded for relevant descriptive information and for thematic analysis.Footnote 31 These categories were generated from the project’s research questions as well as iteratively via a preliminary reading of our data.
News media representations of cyberterrorism
As noted above, each of the 535 news items identified in our research sample was coded according to the extent of its focus on cyberterrorism as an individuated concern. Four hundred of these items had a primary or secondary focus on cyberterrorism (see Table 1). Items with a primary focus were those that concentrated exclusively or principally on the issue of cyberterrorism (however defined);Footnote 32 items with a secondary focus paid attention to this phenomenon without it constituting a main focus of the piece.Footnote 33 Those categorised as ‘other’ mentioned cyberterrorism at some point, but lacked any level of engagement with this concept sufficient for meaningful analysis.Footnote 34
News items by level of focus on cyberterrorism.
The 400 news items with cyberterrorism as a primary or secondary focus were subsequently categorised according to the level of concern each exhibited about the cyberterrorism threat. Six different levels of concern were identified following a detailed, immersive reading, characterised as: concerned; concerned, with elements of scepticism; balanced; sceptical; sceptical, with elements of concern; and, neither.
A story was coded as concerned or sceptical if it was characterised by a clearly identifiable stance on the threat posed by cyberterrorism, offering no space for consideration of alternative perspectives. Examples of the former include Jemima Kiss’s article for the UK’s Guardian newspaper, which cites Eugene Kaspersky’s connection of the Stuxnet virus to ‘three incidents of cyber-terrorism’ and concludes that things ‘are only going to get worse’.Footnote 35 Joseph S. Nye’s 2011 article in The New York Times, similarly, cites ex-NSA director Mike McConnell’s assessment that ‘Sooner or later, terror groups will achieve cyber-sophistication. It’s like nuclear proliferation, only far easier’,Footnote 36 without challenge from competing voices. As Figure 1 demonstrates, two-thirds (67 per cent) of the 400 news items in our study demonstrated a distinctively ‘concerned’ perspective on cyberterrorism. In stark contrast, only 2 per cent of coverage across this time period took an explicitly sceptical stance toward this threat. Examples of these minority views included Duncan Campbell’s 2009 piece in The Guardian, which approached Gary McKinnon’s potential extradition to the United States as a product of misplaced fears over cyberterrorism in the post-9/11 period.Footnote 37 Another article in the same newspaper by Glenn Greenwald (writing prior to the Edward Snowden revelations) similarly sought to contest Leon Panetta’s assertion that extremist groups armed with ‘cyber tools’ could gain control of ‘critical switches’ to cause catastrophic disruption, arguing:Footnote 38
This massive new expenditure of money is not primarily devoted to defending against cyber-aggressors. The US itself is the world’s leading cyber-aggressor. A major purpose of this expansion is to strengthen the US’s ability to destroy other nations with cyber-attacks.Footnote 39
Levels of concern exhibited by news items.
Stories coded as either ‘concerned, with elements of scepticism’ or ‘sceptical, with elements of concern’ evidenced a dominant narrative while also providing space to a rival interpretation. Eight per cent of items in our sample were ‘concerned, with elements of scepticism’, with only 1 per cent categorised as ‘sceptical, with elements of concern’. An example of the former is Tim McDonald’s 2010 piece for ABC News Australia, which reports that governments are on alert for the threat of cyberterrorism.Footnote 40 Whilst reflections from the UK’s former Foreign Secretary William Hague are cited to bolster this perspective,Footnote 41 the story concludes by citing Bill Gates’ opinion that – with an appropriate governmental approach – cyberterrorism ‘shouldn’t be something that people will have to worry about’.Footnote 42 Karen Greenberg’s 2012 piece for Aljazeera, in contrast, was clearly sceptical about the significance of this threat, albeit while demonstrating elements of concern. This article’s primary focus concerned how the language of cyber-defence (including the lexicon of counterterrorism) facilitates encroachment on civil liberties. At the same time, Greenberg does not reject the possibility of genuine cyber-threats outright, conceding that ‘potential attacks, according to leading cyber experts, are possible’.Footnote 43
Balanced coverage was characterised by the presence of competing narratives over the cyberterrorism threat and the absence of any definitive conclusion about the plausibility of these. For example, a Russia Today article from 2012 pays significant attention to the concerns of American security professionals about cyberterrorism before subsequently exploring the views of more sceptical ‘experts’.Footnote 44 As Figure 1 demonstrates, 7 per cent of items across this time period took a balanced approach to the cyberterrorism threat. Finally, a story was categorised as neither if it discussed cyberterrorism but refrained from offering any clear assessment on the threat thereof. The UK’s Daily Mail, for example, reported on Pakistan’s introduction of new laws to combat cyberterrorism without reflection on their need or utility.Footnote 45 Fifteen per cent of items in our sample fit into this category.
As Figure 1 above suggests, three-quarters of the news items examined in our research expressed concern toward the threat posed by cyberterrorism, with only a small proportion of these qualifying this with competing, sceptical perspectives. Importantly, as Table 2 shows, the levels of concern about cyberterrorism varied according to the specificity of the coverage. Thus, 75 per cent of the items that focused primarily on cyberterrorism were concerned (compared to 65 per cent for those with cyberterrorism as their secondary focus), and a further 11 per cent were classed as ‘concerned, with elements of scepticism’ (compared to 8 per cent for those with cyberterrorism as their secondary focus). In other words, levels of concern increased where there was a specific focus on cyberterrorism.
Levels of concern by focus on cyberterrorism.
Who are the cyberterrorists?
Within the 400 news items that concentrated on cyberterrorism as their primary or secondary focus, 219 articles offered specific representations of the identity of cyberterrorists. Five distinct identity types were present in our sample: hackers; hacktivists; professionals; unskilled; and, unspecified non-state actors.Footnote 46 When an item contained more than one of these representations,Footnote 47 each of these was analysed separately.
The first four of these representations are distinguished by the actor’s skill level and/or motivation. Hacker referred to depictions of cyberterrorists as individuals who are likely to employ computer techniques to cause disruption and interference to a particular target, but who lack either the skill or motivation to cause serious levels of damage to the most critical systems. Website defacement and Distributed Denial of Service Attacks (DDoS) are techniques that were often associated with this representation of cyberterrorists, as opposed to the writing and dissemination of complex malware, for instance. A good example of this may be identified in a 2009 story written for The Independent that covered a recent ‘cyber terror blitz’ and the likelihood that ‘Russian hackers’ were to blame for temporarily shutting down Twitter and causing major slowdowns to Facebook and Livejournal through the use of DDoS.Footnote 48
Whilst the term hacker was used to refer to individuals, the term hacktivist refers to coverage of groups or their members who self-identify as collectives with a shared objective, such as Anonymous, AntiSec, and LulzSec. A 2012 Russia Today article, for example, in which cyberterrorists were categorised as hacktivist reported on the group Anonymous under the heading ‘Protesters or Terrorists?’.Footnote 49 Professionals referred to stories focused on individuals with sufficient levels of knowledge of complex computer techniques to be able to target the most critical systems. For example, in a story outlining how the US is looking to protect itself against future cyberattack through the development of a virtual ‘cyber city’, Leon E. Panetta asserts that malicious actors (including terrorists) can target vulnerabilities in critical infrastructure to create destruction comparable to the 11 September 2001 attacks.Footnote 50
Unskilled referred to representations of individuals who employed either already available scripts or publically available platforms and software to commit acts deemed ‘cyberterrorist’ in this coverage. A 2011 BBC article, for example, reported on two Mexican individuals who had sent out phoney tweets alleging that shootings and kidnappings were taking place at a local school. The actions entailed no technical expertise, yet the use of this social networking platform in this way caused panic and accidents, and was later labelled ‘Twitter terrorism’.Footnote 51 Given the manner in which both state and nonstate actors were deemed capable of cyberterrorism in much of this media, a final category – unspecified non-state actors – acknowledged representations that said little about an individual’s skill or motivation but explicitly identified the purported cyberterrorist as entirely unaffiliated with a state.Footnote 52
As Table 3 demonstrates, the most common representation of cyberterrorists in our sample was as professionals (n=118), followed by hackers (n=77). For the three most common representations, Table 2 also shows subtotals for whether the cyberterrorist was presented as a state or nonstate actor. These subtotals show that, even when represented as professionals, cyberterrorists were more likely to be portrayed as nonstate, than state, actors.
Representations of cyberterrorists.
Table 4 builds on this analysis by examining whether levels of concern intensified when cyberterrorists were represented in a particular way. For each representation of the cyberterrorist, it shows the proportion of stories that fell within the six categories used to measure concern introduced above.Footnote 53
Levels of concern by representation of cyberterrorists.
Four findings of note emerge from this analysis of news media representations of the identity of cyberterrorists. First, levels of concern were higher when items contained discernible representations of cyberterrorists. For those items that included an explicit representation of this threat’s identity, 84 per cent were either ‘concerned’ or ‘concerned, with elements of scepticism’. This figure was 19 per cent lower for those items that contained no such representation. So, again, greater specificity was associated with higher levels of concern.
Second, in addition to specificity, status also had an important bearing on concern levels. Of the four representations based on an actor’s level of skill and motivation (hacker/hacktivist/professional/unskilled), ‘Professionals’ was not only the most common but also the one with the greatest levels of concern. Seventy-five per cent of the stories representing cyberterrorists as professionals were categorised as concerned, with a further 12 per cent classed as ‘concerned, with elements of scepticism’. Indeed, none of the items representing cyberterrorists thus displayed any level of scepticism. Perhaps more surprisingly, the ‘Unskilled’ representation also had above-average levels of concern. As with ‘Professionals’, 87 per cent of the items that portrayed cyberterrorists as unskilled were classified as either ‘concerned’ or ‘concerned, with elements of scepticism’. Unlike the ‘Professionals’ construction, however, a small number of items containing the unskilled representation were classified as sceptical (5 per cent).
Third, the small number of items (n=7) that represented cyberterrorists as ‘Unspecified non-state actors’ were all classified as concerned. These items were all characterised by an attempt to locate the concept of cyberterrorism in the context of a wider threat setting which included either offline terrorist (and in our sample, specifically, jihadist) groupsFootnote 54 – most frequently Al-QaedaFootnote 55 – hostile states,Footnote 56 or past examples of cyberattacks, particularly Stuxnet.Footnote 57 Because of this wider concern, only one of these seven items had a primary focus on cyberterrorism.
Fourth, items representing hacktivists as cyberterrorists had the lowest levels of concern. Only two of the six items containing this representation were classified as concerned, each of which featured industry expert Eugene Kaspersky.Footnote 58 In a 2012 article by Lee Taylor in The Australian Telegraph, for instance, Kaspersky asserts that cyberterrorism is a logical progression for the ‘leaders’ of hacktivist collectives such as Anonymous:
Most hacktivists – not all of them – are just following orders from the leaders, but many of these leaders are professional people and this is really dangerous … [T]hey can grow to the terrorist level.Footnote 59
The other item, in the Australian Financial Review, reported on how the Australian Security Intelligence Organisation (ASIO) has monitored ‘with mounting alarm’ groups such as Anonymous, which have ‘targeted Western nations and companies with disruptive attacks that foreshadow an apocalyptic fusion between cyber-capabilities and terrorism’.Footnote 60 In marked contrast was an article from Russia Today, which made use of the same representation of hacktivists as cyberterrorists but was classified as ‘sceptical, with elements of concern’.Footnote 61 Depicting a perceived American preoccupation with cyber-threats as a thinly veiled grab for greater executive power, military spending, and curtailing of civil liberties, this item included the following quote from independent journalist J. A. Myerson on the relationship between hacktivist dissent and terrorism in the eyes of the US government: ‘There is a deep insinuation that dissent is somehow connected to or an accessary to terror. That’s a really horrifying prospect.’Footnote 62
Who or what is threatened by cyberterrorism?
Within the news items with cyberterrorism as a primary or secondary focus, a wide variety of referents were identified as those threatened by this newest incarnation of terrorism. As Figure 2 indicates, the most common of these referents by some distance was the state itself, followed by critical infrastructures and then the private sector.Footnote 63
Referents of cyberterrorism news media discourse.
Table 5 examines how levels of concern about cyberterrorism varied depending on the threat referent.
Levels of concern by referent object.
Three points emerge from the data presented in Figure 2 and Table 5. First, where specificity was linked to higher levels of concern in relation to the focus of news coverage and representations of cyberterrorists, the scale of particular referents also appears significant in our sample, at least. All but one of the items which presented ‘the West’ or ‘developed states’ as the referent threatened by cyberterrorism were categorised as concerned (with the other classed as ‘neither’). These items contained numerous bold assertions, including: ‘Power and water and other vital services in the West could be crippled’;Footnote 64 ‘The fear is that modern nations are so dependent on technology that widespread interference with systems could wreak havoc’;Footnote 65 and, ‘the western world has, almost overnight, found itself incredibly vulnerable’.Footnote 66 Of these ten items, five focused on Al-Qaeda’s call for ‘cyber jihad’,Footnote 67 whilst the remainder juxtaposed the West to countries such as Russia, China, North Korea, and Iran by referring to past events including Stuxnet and the cyberattacks on Estonia and Georgia in 2007 and 2008.Footnote 68 Similarly, all 17 of the items that presented ‘the globe’ as the referent of the cyberterrorism threat were either ‘concerned’ or ‘concerned, with elements of scepticism’. These included a 2012 article in Russia Today following the discovery of the Flame malware which quoted Eugene Kaspersky as saying ‘I’m afraid it will be the end of the world as we know it … I’m scared, believe me’Footnote 69 and a 2009 article in The Guardian which reported on research conducted by the International Commission on Nuclear Non-proliferation and Disarmament under the headline ‘Terrorists could use internet to launch nuclear attack.’Footnote 70 A recurring theme in this coverage was the importance of transnational dialogue and cooperation around cyber-security,Footnote 71 including the need for international cyber-security treaties.Footnote 72
Second, it is noteworthy that concern levels were higher when the referent was the private sector (85 per cent) or critical infrastructure (74 per cent) than when the referent was nation states (70 per cent) or citizens (57 per cent). This resonates with other research, which suggests that there may be qualitative differences between cyberterrorism and its parent concept, terrorism. Whilst the latter is traditionally understood as constituted by serious levels of violence against people or property, this may not be the case for cyberterrorism.Footnote 73 For example, a 2011 opinion piece on Fox News by Judith Miller – which reported on an outage within the Research In Motion (RIM) network on which BlackBerry smartphones rely – opened as follows:
Did you open your BlackBerry Wednesday or even Thursday morning and find – nothing? No new e-mails, or tweets. No new text messages. Just blackness and that familiar screen saver photo of your child, spouse or dog? Welcome to the world of cyber-terrorism vulnerability.
From this starting point, the piece went on to warn that cyberattacks ‘could not only cause billions of dollars in damage to such vital systems, but endanger national security’ (notwithstanding the fact that RIM had attributed the outage to a core switch failure).Footnote 74 Similarly, in 2011, a Reuters technology article covered a Sony shareholders’ meeting. The meeting took place in the aftermath of the cyberattacks on the company earlier that year, in which hackers accessed personal information on 77 million PlayStation Network and Qriocity accounts. The piece quotes Sony’s CEO Sir Howard Stringer as saying:
We believe that we first became the subject of attack because we tried to protect our IP (intellectual property), our content, in this case videogames … I think you see that cyber terrorism is now a global force, affecting many more companies than just Sony.Footnote 75
Third, the referents that were associated with the lowest levels of concern were: data (61 per cent); citizens (57 per cent); and, institutions (such as NATO,Footnote 76 the New York Stock Exchange,Footnote 77 and the Church of Scientology)Footnote 78 (44 per cent). For all three of these, a significant proportion of news items were ‘concerned, with elements of scepticism’ (22 per cent, 29 per cent, and 22 per cent respectively). As well as the fact that concern about these threat referents was more likely to be tempered in this way, it is also telling that, of the eight categories of referent, institutions and citizens were the two least common. Only nine items identified institutions as a potential target for cyberterrorism, and even fewer identified citizens as one (n=7).
Lastly, Table 6 examines whether concern levels varied depending on whether the geographical focus of a news item was its country of publication.Footnote 79 In other words, did concern levels drop when news items covered threats focused on other countries?
Levels of concern by country of publication and geographical focus.
Two interesting findings emerge from this data. First, for those items whose geographical focus was the country of publication, the proportion classified as concerned was only slightly higher than the equivalent figure for items whose focus was a country other than the one of publication. So levels of concern did not appear to drop to any significant extent when an item focussed on another country. This suggests an apprehension of the transnational and boundary-less nature of cyber threats. Second, the proportion of items which had an international focus that exhibited concern (76 per cent) was significantly higher than in the concerned coverage that focused on specific countries. This is consistent with our earlier finding that, in terms of referent, levels of concern are linked to the perceived scale of the threat.
Specificity, status, and scale: Constructing the perfect storm
It will already be apparent from the preceding discussion that multiple constructions of cyberterrorism are present within news media discourse. Cyberterrorists may be represented as hackers, hacktivists, professionals, or unskilled actors. They may be represented as state or non-state actors. And, in addition, a range of referents is possible, from the entire world at one extreme to individual citizens at the other. In short, there is no homogeneous construction of cyberterrorism. Moreover, these competing constructions of cyberterrorism can be presented in different ways. They may be presented as part of a wider threat environment that also encompasses offline terrorist groups and/or other types of cyber threat, or they may form the primary focus of a particular news item.
The analysis hitherto has examined the links between these different constructions and levels of concern about the threat that cyberterrorism poses. Some constructions – such as those in which cyberterrorists are represented as hackers or hacktivists, and where the referent identified is data, citizens, or institutions – are associated with below average levels of concern.Footnote 80 Other constructions, in contrast, are associated with above average concern levels. Based on the earlier analysis, these include representations of cyberterrorists as professionals or unskilled actors; and, stories where the threat referent is the private sector, critical infrastructure, the West or the entire globe. Moreover, news items with an international focus and concentrating primarily on cyberterrorism also appear to demonstrate an exaggerated conception of the cyberterrorism threat.
Of the 535 news items examined in this study, a total of ten match this ‘perfect storm’ model. Four features of these items are worthy of note. First, the items did indeed display above average levels of concern. Eight of the ten items were categorised as ‘concerned’, and a further one as ‘concerned, with elements of scepticism’.Footnote 81 Second, only one of the ten items was published in a tabloid newspaper. Of the other nine, six were published by broadcasters (Reuters (x2), Russia Today (x2), Fox News, Russia Today) and three by broadsheet newspapers. Indeed, the five tabloid newspapers in our sample were responsible for just 31 of the 400 news items that had cyberterrorism as their primary or secondary focus (6.2 items per outlet) compared to 129 from broadcasters (14.3 items per outlet) and 240 from broadsheet newspapers (14.1 items per outlet). This would seem to refute suggestions that the dominant news media discourse on cyberterrorism (if, indeed, there is one) is solely the product of tabloid hyperbole.Footnote 82
The third striking feature of these ten items is their reliance on other sites of discourse. All but two of the items report on meetings between policymakers, industry and academics. Two focus on the 2012 Tel Aviv Cyber Security Conference.Footnote 83 Two focus on the 2011 London Cyber Conference.Footnote 84 The four others focus respectively on the 2010 Cybersecurity World Summit,Footnote 85 the 2013 Digital-Life-Design Conference in Munich,Footnote 86 a 2008 security conference organised by the Royal United Service Institute,Footnote 87 and a 2008 meeting of government and technology experts in Malaysia.Footnote 88 With headlines such as ‘Experts warn of cyberterrorism threat’,Footnote 89 ‘Countries are risking cyber terrorism, security expert tells first world summit’,Footnote 90 and ‘Web security expert warns of cyber world war’,Footnote 91 these eight items are (almost entirely) descriptive overviews of speeches delivered at these events with relatively little express commentary or opinion. The ‘experts’ quoted in these items include the UK Prime Minister David Cameron, the Malaysian Prime Minister Abdullah Admad Badawi, an Estonian defence ministry official, the former director of the Pentagon agency responsible for computer networks, the director-general of the French network and information security agency, a former US advisor to the White House on cybersecurity and the chief research officer at F-Secure. Moreover, the two items which are not primarily descriptive accounts of a conference or meeting but are instead discussion-based pieces that purport to assess the magnitude of the cyberterrorism threat also draw heavily on the opinions of ‘experts’, including the Head of the US Department of Homeland Security Michael Chertoff, the Director of Research Programs at the Australian Strategic Policy Institute Anthony Bergin, and various academics.Footnote 92 By far the most prominent voice, however, is that of Eugene Kaspersky. With warnings such as “[W]e are close, very close, to cyber terrorism. Perhaps already the criminals have sold their skills to the terrorists – and then … oh, God’,Footnote 93 Kaspersky is quoted in all five items published from 2011 onwards. Whilst none of this is to suggest that the process of selecting, summarising, and collating these views is an entirely objective, impartial one, the discourse in these news items is clearly rooted in and derived from other discursive sites.
It is noteworthy, finally, that six of these ten items also culminate in calls for greater international cooperation to enhance cybersecurity and deal with cyber threats. The use of hyperbole in these items may therefore – to some extent, at least – be understood as an attempt to overcome inertia or resistance via securitisation of this issue. This raises clear questions for future research projects, including the motives of the different actors involved in this coverage.
Conclusion
The analysis of international news coverage presented in this article has attempted to offer two primary contributions to existing literature. The first was to add empirical depth to existing accounts of media representations of this threat, and the importance thereof. As outlined above, academic literature on cyberterrorism tends to simplify and generalise in one of two ways. In the first instance, ‘concerned’ assessments of the danger it poses typically construct cyberterrorism as a monolithic, unitary security threat, rarely disaggregating it. ‘Sceptical’ threat assessments, in contrast, assume an equally homogeneous media discourse on cyberterrorism that is characterised by hyperbolic representations of imminent, exceptional threat. As demonstrated by our findings, however, media discourse on cyberterrorism is actually surprisingly heterogeneous. It is characterised, inter alia, by: varying levels of anxiety, stretching from the concerned to the sceptical; different conceptions of the identity of would-be cyberterrorists; variable levels of focus on this particular phenomenon, and therefore different contexts into which cyberterrorism is inserted; and, a range of distinct referents deemed threatened by cyberterrorism. As we have shown, although two thirds of the coverage in our sample of 31 outlets across a five-year period was concerned by the threat of cyberterrorism, dissenting voices were also identifiable. Similarly, while cyberterrorism is most frequently identitied as the behaviour of professional actors or hackers, the label is also applied to a range of other actors including hacktivist collectives and unskilled individuals. And, in terms of referents identified – or constructed – in this coverage, there is considerable diversity from imaginary geopolitical spaces (‘the West’) through to individual beings (ordinary citizens) and inanimate entities (such as data). As this suggests, there is not one (singular, monolithic) news media discourse on cyberterrorism: this coverage is profoundly multiple and diverse.
The article’s second contribution was to add analytical depth to existing work on the content and framing of cyberterrorism discourse by highlighting the importance of three – hitherto neglected – factors within articulations of this threat. First, as demonstrated above, specificity matters, whereby news coverage that affords greater priority to cyberterrorism specifically tends toward a more exaggerated level of concern with the threat that it poses. Moreover, articles containing an explicit representation of ‘cyberterrorists’, in our sample, also demonstrate heightened levels of concern.
Secondly, our analysis also pulled attention to the importance of status, demonstrating that particular portrayals of cyberterrorist actors lead to heightened levels of concern. Of particular note here, however, was the nature of this correlation and the discovery that the impact of status on levels of concern applied at two divergent points across the capability spectrum. International news coverage reported fearfully both on cyberterrorists deemed to be ‘professionals’ (with knowledge of how to target the most critical systems) and, conversely, on those deemed ‘unskilled’ (with no expertise at all and often a reliance on publically available software). Given the article’s desire ‘… to study how representations of terrorism and their reality are socially produced through linguistic and non-linguistic practices’Footnote 94 the application of the label ‘cyberterrorist’ within international news coverage both to those deemed capable of destruction comparable to that of the events of 9-11 and to those who send out hoax ‘Tweets’ is of great significance for the construction of cyberterrorism knowledge.
And, third, we have also shown the significance of scale in this coverage, in that larger referents such as ‘the West’ or ‘the globe’ tend to attract greater anxiety than states or individual citizens. This relation is not, however, perfect, as the private sector and critical infrastructure also emerge as referents of particular concern throughout our sample. Although it might appear intuitive that larger referents would correlate with greater anxiety, there are at least two noteworthy aspects that can be drawn from this finding. First, the emphasis on larger referents such as these evidences a move away from the state as the primary referent within international media coverage of cyberterrorism. Given the traditional centrality of the state within terrorism discourse,Footnote 95 its scaling up in this instance to entire swathes of the globe is perhaps reflective of a broader anxiety toward threats against interconnected and potentially vulnerable systems. As characterised by Thomas Homer-Dixon:
We’ve realised, belatedly, that our societies are wide-open for terrorists. We’re easy prey because of two key trends: First, the growing technological capacity of small groups and individuals to destroy things and people; and second, the increasing vulnerability of our economic and technological systems to carefully aimed attacksFootnote 96
Still on the point of status a second noteworthy aspect relates to the prevalence of larger referents and what Lene Hansen and Helen Nissenbaum write about ‘everyday security practices’.Footnote 97 This particular grammar of cyber security, according to Hansen and Nissenbaum:
… points to the way in which securitizing actors, including private organizations and businesses, mobilize ‘normal’ individuals’ experiences in two ways: to secure the individual’s partnership and compliance in protecting network security, and to make hypersecuritization scenarios more plausible by linking elements of the disaster scenario to experiences familiar from everyday lifeFootnote 98
Issues of cybersecurity are often more relatable to the experience of the general citizenry and this link has aided securitising actors in their ability to successfully complete securitising moves. With this in mind one may expect the international news media to ‘move down’ the scale of referent objects and produce more stories with personal data or privacy rights as that which is under threat. However, what this article has found is that, contra Hansen and Nissenbaum, the international news media, far from covering stories with more personally relatable referent objects, in fact more frequently do the opposite and scale up their referent.
Moving away from these analytical contributions, this article also sought to connect the burgeoning literature on cyberterrorism with its focus on issues of definition and threat to a more established body of scholarship on the construction and framing of terrorism (and, indeed, security) more broadly. In this context, it is of considerable interest to note that more specific representations of cyberterrorism tend to be associated with heightened levels of concern, given a widespread assumption that generalised and ambiguous discourses on terrorism serve (for some, deliberately) to exaggerate the threat that terrorism poses.Footnote 99 As Debrix argues, discourses on cyberterrorism are in fact complex and internally divided, and – as the above hopefully demonstrates – much might be learned by breaking these down into their constituent parts:
Cyberterrorism is not an evanescent, chaotic threat, even though the media would like to convey this image of it to the public. On the contrary, the media themselves have organized cyberterrorism into categories, strategies, typologies, and methodologies which make it possible to talk about it (as a threat, a fear, a virtual danger) and to produce more knowledge about it (mostly, an anticipated sense of emergency and planning).Footnote 100
These three contributions leave us with a discourse that looks very different when observed along various points of the ‘threat spectrum’. Media discourse – in its entirety – is characterised, as shown above, by enormous diversity and heterogeneity. At the same time, however, when we turn attention to the most apprehensive accounts within international media discourse we find something far more uniform and stable. The ‘perfect storm’ constructions considered at the end of the article are notable not only for their representing the most apprehensive and cohesive part of this news media discourse. They also evidence considerable similarities of form, especially in relation to the citing of ‘expert’ knowledge from industry and beyond.Footnote 101
Ultimately this article has sought to contribute to understanding of how the threat of cyberterrorism has been constructed by the news media in a range of different countries across the world. It has sought to describe and unpack more and less hyperbolic constructions of cyberterrorism in an effort to render visible often-neglected heterogeneities therein. Future research could profitably build on this work via comparative analysis with other media providers and types, including non-English language sources and citizen journalism. Other timescales both before and subsequent to that covered in our analysis would make a particularly interesting comparison, too, allowing exploration of the importance of particular events in the framing of coverage on cyberterrorism. More work might be done on the reception of news media discourse by different audiences, and on the specific rhetorical devices employed in news coverage, including the role of metaphors, analogies, intertextualities and the like; as well as the significance of images therein. Such research would build on this article’s provision of a fuller and more detailed picture of the granularities of cyberterrorism discourse and the importance thereof.
Acknowledgements
We express our gratitude to Swansea University’s College of Law and Criminology, and to the Bridging the Gaps program for their support for the research upon which this article is based. We also gratefully acknowledge Jordan McErlean and Alicia Payne for their excellent research assistance, and David Mair and Lella Nouri-Bennett for their helpful suggestions throughout the project. Finally, we would like to thank the three anonymous reviewers who offered valuable comments on an earlier version of this article as well as the editors of the European Journal of International Security.
Lee Jarvis is Reader in International Security at the University of East Anglia. He is an editor of Critical Studies on Terrorism and (co-) author or editor of ten books, including Anti-Terrorism, Citizenship and Security (with Michael Lister, Manchester University Press: 2015); Security: A Critical Introduction (with Jack Holland, Palgrave: 2015); and Counter-Radicalisation: Critical Perspectives (with Charlotte Heath-Kelly and Christopher Baker-Beall, Routledge: 2015).
Stuart Macdonald is a Professor in Law and Deputy Director of the Centre for Criminal Justice and Criminology at Swansea University. He is co-editor of Cyberterrorism: Understanding, Assessment and Response (New York: Springer, 2014) and Terrorism Online: Politics, Law and Technology (Abingdon: Routledge, 2015) (both with Lee Jarvis and Thomas Chen). His recent project on security and liberty was funded by the British Academy.
Andrew Whiting is a Lecturer in Security Studies at Birmingham City University. He has had his work on terrorism and cyberterrorism published in Global Society and Perspectives on Terrorism as well as in edited volumes including Cyberterrorism: Understanding, Assessment and Response (New York: Springer, 2014); Counter-Radicalisation: Critical Perspectives (Abingdon: Routledge, 2015); and Researching Terrorism, Peace and Conflict Studies (Abingdon: Routledge, 2015).
