Introduction
Cyberspace is now ubiquitous. Carrying most of our communications, accelerating commerce, increasingly connecting our toasters and toothbrushes, it is transforming the way individuals and societies operate. Long understood as a place of enhanced individual freedom, a romantic vision of this ‘space’ still dominates many popular imaginaries. Often framed as a metaphorical ‘Wild West’, cyberspace became understood as a place in which authority, boundaries, and geography were weak or even did not apply. While attractive, this understanding of cyberspace is misplaced. The libertarian understanding of cyberspace has failed to grasp the extent to which sovereign states historically developed the Internet for their strategic advantage. Consequently, more recent attempts by states like Russia and China to assert control or influence over cyberspace are misunderstood as individual, isolated attempts by authoritarian states to master the Internet. Rather than anomalies, this article suggests that the longstanding appetite of sovereign states to assert power over and through cyberspace stretches back to the Second World War. More importantly, we should question whether this misunderstanding has not been convenient for governments, enabling them to defend their surveillance practices as an attempt to assert order in a supposedly anarchic space. Tracing the historical trajectory of these developments, the current debate about the ‘weaponisation of cyberspace’ is recast. Rather than a special place from which the ‘weary giants’ of government were excluded, it was simply an extension of their security activities. The image of cyberspace has been constructed in a way that has misdirected our understanding of the nature of communications technology.
It is not surprising that cyberspace was romantically imagined as a boundless space of freedom. The term was invented in 1984 by the writer William Gibson in the context of his novel ‘Neuromancer’, which speaks eloquently of a different ‘kind’ of space, ‘a consensual hallucination experienced daily by billions … in every nation’. Emphasising the novelty of a ‘space’ in which individuals and communities could connect beyond traditional (state) mechanisms of control and governance, this ‘space’ was framed through meta-level metaphors, eagerly seized upon by libertarians throughout the 1990s. Perhaps the most famous of these was John Perry Barlow, American poet, cattle rancher, privacy campaigner, and lyricist for the Grateful Dead. Responding to the US governments attempts to censor the Internet in 1996, he issued a declaration of independence for cyberspace, dismissing states as the ‘weary giants of flesh and steel’.Footnote 1 Barlow's influential language was distinctly utopian and has echoed down the decades. Typically, in 2011, Nicholas Negroponte, creator of Wired magazine observed: ‘this is just the beginning, the beginning of understanding that cyberspace has no limits, no boundaries’. By equal turns, some of this discussion is disturbingly dystopian. Eric Schmidt, longstanding chairman of Google, warns us that: ‘The Internet is the first thing that humanity has built that humanity doesn't understand, the largest experiment in anarchy that we have ever had.’Footnote 2 But what all these narratives have in common is a sense of anarchy and a distinct underappreciation of the ongoing processes by which states are making this space. By contrast, I argue here that the widely accepted libertarian vision of cyberspace has often been convenient, enabling states to conjure up a vision of cyberspace as an untamed ‘Wild West’ of criminality and rebellion in order to legitimate their security activities. This, in turn, raises the bigger question of whether governments have enabled or encouraged us to misunderstand cyberspace, casting a cloak of obscurity over their activities.
The majority of current accounts within International Relations remain somewhat deterministic, offering ahistorical and non-materialist approaches in which the Internet simply accelerates the corrosive effects of globalisation in eroding the authority of the state.Footnote 3 As Geoffrey Herrera observed, much of the literature on this subject understands the Internet as, at best, a simple accelerator of globalisation and at worst a threat to the future of the nation-state system.Footnote 4 Herrera, one of the few International Relations scholars to reflect at length on this subject, argues that writing about the Internet often takes the form of a technological determinism that is ‘vastly at odds’ with the historical record.Footnote 5 This article extends Herrera's significant critique, highlighting the ways in which sovereign states have shaped the development of cyberspace. In particular, it challenges the ‘placeless-ness’ narrative, emphasising the importance of the material, physical, and historical dimensions of these technologies together with the control this affords.Footnote 6
This effort to recast our understanding of cyberspace draws on the ideas of French philosopher, Henri Lefebvre. Writing in the 1960s, in response to the urban planning of French cities, Lefebvre critiqued the way space was understood as scientific, objective, and pure. His celebrated text ‘The Social Production of Space’ changed how space was understood, introducing politics and society to geographical analysis. Created through social relations, he argued that space should be understood as a process. Consequently, researchers and geographers had to think about why space was created, not just what was ‘in’ it.Footnote 7 Underlying the politics of space, he was at pains to emphasise the way cities and urban planning were the products of history, politics, and social relations, rather than simply ‘natural space’.Footnote 8 In many ways, cyberspace is not dissimilar to the 1960s French urbanism that sparked Lefebvre's ideas. Originating in the American defence industry, cyberspace has evolved in ways that fit remarkably well onto the map of Westphalian sovereign states. Governments of all kinds have produced cyberspace for their strategic advantage. In Lefebvre's language, cyberspace ‘is also a means of control, and hence of domination, of power’.Footnote 9 This article seeks to explore the diverse and often conflicting ways states have seen and used cyberspace in this way.
Lefebvre's ideas were further developed by Doreen Massey's writings on globalisation.Footnote 10 Critiquing many of the abstracted and utopian notions of globalisation, she pointed out that it ‘doesn't float above the earth, it is operated by the same material, social, embedded processes of people in branch plants, in production factories, in research organisations, making decisions which may or may not work out all around the world’.Footnote 11 Similar observations might be made about cyberspace, which has also been described as a floating cloud or libertarian tool of freedom that has reduced governmental power and authority. This space did not emerge organically but is the product of interrelations, ‘constituted through interactions, from the immensity of the global to the intimately tiny’.Footnote 12 Moreover, viewed as a product of fluid, ongoing relations, the idea that cyberspace has arrived as a complete entity is also challenged in favour of a continuum.
Unsurprisingly, geographers have been consistently important in urging us to appreciate the importance of territory as a process.Footnote 13 Most prominently, Gearóid Ó Tuathail argues that cyberspace, much like the contemporary world financial system ‘is not the product of natural forces … but of a new working relationship between States and markets promoted, in part, by the States themselves … geography is not so much disappearing as being restructured, rearranged and rewired’.Footnote 14 Importantly, he suggests that our misunderstanding of cyber is part of an attempt to ‘denaturalise and limit the power of States while naturalising the virtues of the markets’.Footnote 15
The possibility that this new ‘space’ enhances state power through surveillance is critically important, and is largely overlooked within the libertarian consensus.Footnote 16 The debate over the role of communications technologies and the Arab Spring typifies this narrative.Footnote 17 Initially, these events were hailed as evidence of the transformative impact of telecommunications technology across North Africa and the Middle East. Typically, Howard and Hussain suggested that this ‘space’ created through Social Media and other platforms represents a substantive shift from historical mechanisms of social governance and ordering. Reducing entry costs of traditional forms of heavily regulated media like television commercials or radio channels, social movements were liberated through both ubiquity and anonymity. ‘When physical spaces for public conversation and debate closed down’, they argue, ‘the Internet provided virtual spaces for political communication’.Footnote 18 Cyberspace was framed as empowering social movements and dissenting individuals, providing an alternative space that facilitates freedom of expression or association.Footnote 19
More recently, scholars have revised their view in two ways. First, they have become more sceptical about the role of social media in generating the Arab Spring and, secondly, they have argued that states have since become more adept at controlling it.Footnote 20 Slowly but surely, they have moved in the direction of the iconoclastic technology critic Evgeny Mozrov, accepting that authoritarian states like China have successfully countered Google and built their own local intranet or ‘splinternet’.Footnote 21 But even this narrative is incorrect, overlooking the significant ways in which sovereign states in the liberal democratic sphere have consistently exerted control and authority over these technologies, albeit in more hidden ways.
Indeed, surveillance experts have now begun to speak of a sinister innovation called ‘social media intelligence’. David Omand, a former Director of GCHQ and Carl Miller of DEMOS have sketched out a whole new terrain of state-driven forecasting activity that draws on data from things like twitterfeed. Social media intelligence (SOCMINT) includes a range of techniques and technologies that facilitate the watching of social media networking sites such as Facebook or Twitter. Although this has been around for many years, typically with police analysts scanning protester chat rooms, the move is towards using bigger data for spotting trends and undertaking sentiment analysis. The result is something that hovers in the liminal space between intelligence and behavioural social science, offering the holy grail of predicting future political events. The way in which social media, once seen as a voice from below, has become the latest intelligence tool for those watching from above, is perhaps indicative of the direction of travel.Footnote 22
Ultimately, cyberspace has transformed how we might understand future governance. Rather than signalling the decline of the importance of Westphalian States, it has enabled sovereign power to evolve, creating new means of governance and reinforcing centralised power and authority in significant ways. All of these realities are lost in the romantic vision of cyberspace as a libertarian playground of deterritorialised freedom. Emphasising the historic and enduring importance of sovereign authority to cyberspace recovers its intrinsically material, territorial, and state-based origins that can be fortified, not challenged, under globalising conditions. The underlying physical infrastructure of cyberspace is important not just because it is based on physical lands or run by people. It is important because states have been able to use these physical components to control and exert influence through this ‘space’. In an age of nation-state hacking and grand disinformation campaigns, the willingness of states to exploit cyberspace shows little sign of abating. We perhaps need a stronger appreciation of the long history of these tendencies, finally appreciating the true scale and impact of states in cyberspace. At root, as Lefebvre suggests, it ‘is not a scientific object removed from ideology or politics; it has always been political and strategic’.Footnote 23
Materiality
Perhaps the most fundamental way cyberspace has been used by states is one of the best well hidden. Lost within a romantic understanding of cyberspace as a ‘mysterious world’, we overlook the way cyberspace has been constructed and designed by state authorities. Hidden in plain sight, the physicality and infrastructure of cyberspace remains attached to and reliant upon a physical territory. This enables states to control and influence these electronic environments, often for their advantage. While the view of figures like Dan Hunter who see cyberspace as a ‘global commons’Footnote 24 is attractive, it overlooks the significance of these physical and geographical dimensions. It is to these, perhaps mundane aspects of cyberspace that this article now turns. Echoing Mark Graham, we must recognise that cyberspace is not an ‘abstract space or digital global village’ but a constructed network or system of information exchange. Sovereign authorities are central to this construction, often making some of the most important decisions about its location and design. Despite perceptions of the Internet as a global commons and public good, the Internet has been constructed along geographical lines that are remarkably congruent with established borders and boundaries. Consequently, we need an alternate, nuanced, and more ‘spatially grounded’ way of understanding how cyberspace has developed over time.Footnote 25
Why are the physical or territorial components of cyberspace are so frequently overlooked? The most persistent users of misleading meta-level metaphors are presidents and prime ministers. Their purpose is to frame threat complexities in terms of an unruly environment that is imperfectly policed and beyond law and order. On 28 March 2018, UK Prime Minister Theresa May, outlining her new National Cyber Security Strategy, was at pains to draw a distinction ‘between the cyber and physical worlds’, while a week before, her secretary of state with responsibility for this area, Matt Hancock, unveiled new regulations for electronic commerce as heralding a future in which the Internet would cease to be the ‘Wild West’.Footnote 26 Similarly, in 2016, Obama deployed almost the same language at the G-20 Summit in Hangzhou, China, insisting that ‘we cannot have a situation where this “space” becomes the Wild Wild West’, and calling for measures that would enable greater surveillance and offensive capabilities within this domain.Footnote 27 Despite their frequency, these statements have a negative consequence, obscuring the very real, significant dependencies of this environment on physical attributes and control.
Recently, the Tongan population experienced this physicality. In January 2019, an undersea fibre optic cable connecting Tonga to the Internet through the Indian Ocean was damaged. Risking the population's communications, healthcare, and labour market, the event forced Tonga to rely on one single satellite dish for its communications and national infrastructure. The satellite link offered less than 1 per cent of the capacity offered by the severed cable and the result was a huge disruption. Tom Westbrook of Reuters reported that the incident was ‘throwing communications across the tiny and isolated country into chaos’.Footnote 28 Most strikingly, reports indicated how it ‘prompted hundreds of people to queue outside a government telecom office where the signal is most reliable … hours have been extended to midnight to handle crowds of officials, business people and ordinary folk logging on to access cash remittances, buy plane tickets and chat.’Footnote 29
Few Internet users pause to consider the journey their data takes between each transaction. A mere four hundred fibre optic cables carry 99 per cent of transoceanic data, comprising ‘the physical links that bind our digital world together’.Footnote 30 The mode of transport is undersea cables that look rather like garden hosepipes, laid down by specialist ships across the bottom of the ocean. Far removed from images of ethereal ‘clouds’ void of physical territory, these cables of glass and metal form the backbone of the Internet and they are often strangely fragile. Alexandra Chang of Wired magazine observed how ‘they are for the most part poorly armoured, rarely patrolled and only occasionally monitored’.Footnote 31 Andrew Blum, the author of Tubes, a fascinating archaeological journey to the centre of the Internet, also finds their vulnerability rather remarkable.Footnote 32
A further striking example of our misunderstanding of cyberspace is the idea of the ‘Cloud’.Footnote 33 This is an example of the problem of using metaphors and grand statements to describe an industrial mixture of server farms, data warehouses, and software as service.Footnote 34 Obscured through an image of our data being transported into a ‘cloud’ above us, instead the industrial skeleton of cyberspace exists in the forms of fibre optic cables, metal pylons and satellite dishes, human resource and water supplies. In fact, the industrial skeleton of cyberspace continues to exist in the form of miles of fibre optic cables, metal pylons and satellite dishes, together with server farms in vast metal sheds that demand considerable amounts of water and electricity.Footnote 35 One of the largest is the Next Generation Data Europe centre in Newport, Europe's major data centre facility. Over 19,000 server cabinets plus storage are displaced across three floors that cover 750,000 square feet and – while impressive – it has little resemblance to a cloud.Footnote 36
Discussing examples of this infrastructure uncovers the rather mundane ways cyberspace is governed and maintained in the twenty-first century. Many geographers have commented on this hidden aspect of cyberspace, but few have speculated as to the rationale. From the perspective of the everyday user, they are seemingly happy that the wiring is hidden behind and beneath user interfaces. Shannon Mattern explores this idea, suggesting that it is easy to forget ‘how they delimit our agency and how they are defining the terrain we are interfacing with’.Footnote 37 Extending this idea of the physicality of cyberspace, Louise Amoore reminds us that ‘as computer science began to document the emergence of cloud computing, geography came to have a specific meaning, defined by where data and programs are spatially stored’.Footnote 38 Laura DeNardis, similarly reminds us that, ‘root servers are housed in buildings and run by people’.Footnote 39 But what does this tell us about the power of states and corporations?
Rather than being unbridled or organic, we need to recognise that ‘at the core of the internet is a series of components that are infrastructural: internet exchanges, national backbone networks, regional networks and local networks’.Footnote 40 As Saskia Sassen suggests, we should recognise that the digital networks of cyberspace are not only comprised of hardware and software but societal structures and power dynamics that also have considerable materiality: ‘There is no purely digital economy, just as there is no virtual corporation or community.’Footnote 41 Connections between the ‘real’ and ‘cyber’ worlds are all around us, located under pavements and along motorway pylons. Hidden in plain sight, these infrastructure layers clearly demonstrate both the physicality of cyberspace and its connections with sovereign territories.
Simultaneously, they highlight connections between cyberspace and ‘old industrial cities’ as centres of power, traditionally associated with sovereign control. Stephen Graham writes forcibly on this, arguing that the reliance on metaphors obfuscates ‘complex relations between new communications and information technologies and space, place and society’. Rather than developing organically and sporadically, the concentration of ‘virtual cities’ within the metropolitan cities of New York and London (for example) are used to demonstrate that, as with television, radio, and printing technologies, ‘any cursory examination of the Internet and the World Wide Web shows that much of the traffic represents and articulates real places and spaces’.Footnote 42 Thus, not only is cyberspace socially constructed and contingent, it is created and implemented for specific uses and needs, notably ‘the very cultural roots of modern capitalist society’.Footnote 43 Social scientists might give more weight to these connections, understanding cyberspace within its social, political, and historical context. Seen through the lens of Victorian industrialism, cyberspace becomes viewed as the latest social technology, impacting society in similar ways that the steam engine and teleprinter did before.
Lawyers have been quicker than many to identify the physical and historical connections between Internet technology and sovereignty, often viewing these things through the prisms of ownership or jurisdiction. Mark Lemley is one example, offering a typically materialist prescription, arguing that no one is ‘in’ cyberspace. The Internet, he suggests, is merely a protocol, a piece of code that permits computer users to transmit data between their computers using existing communications networks. There were computer networks before the Internet that similarly relied on telephonic exchange of data.Footnote 44 By emphasising historic connections between the telegram, the tele vision and telecommunication, Lemley directly questions the validity of viewing cyberspace as a mysterious and separate world when similar approaches to the telegram and television are absent: ‘People may speak occasionally of being “lost in” or “transported” by a television show, a movie, or even a book, but we hardly surrender our understanding that “television space” is merely a series of images transmitted to us.’Footnote 45
Understood as both physical infrastructure and social relations, cyberspace is, as Stanley Brunn argues, transforming the nature of states and what it means to have power in the twenty-first century, but not displacing them.Footnote 46 Rather than being void of physical or political territory, global networks are the product and reproduction of political, historical, and social relations. Within these networks, geography and place remain important, embedded in this infrastructure within sovereign parameters. We thus need to challenge Steven Spiegel's claims that the post-Cold-War era was one characterised by the ‘diminishing role of geography’.Footnote 47 While the last twenty years have undoubtedly been characterised by globalised communications and the oft-celebrated ‘information revolution’,Footnote 48 sovereignty and physical territory remain important.
Within this new context, sovereignty and state authority is changed, not erased. Travelling and existing between diverse infrastructural ‘layers’, power and governance are transformed. Those seeking to find a useful mezzanine between the ethereal, the cultural, or physical have been attracted to the idea of layers. Alexander Klimburg suggests a four-layer model that incorporates (1) the physical or hardware layer; (2) the logic layer (code); (3) the data layer (photographs, emails, data); and (4) the social layer. But he adds that in reality, the Internet resembles the telephone systems from which it grew: ‘the backbone of the Internet is made of cables that run across continents and under the seas, with a smattering of satellite links as well’.Footnote 49 Benjamin Bratton's much-discussed analysis simply extends Klimburg's idea further, exploring the six layers of ‘The Stack’, independent but intrinsically connected: Earth, Cloud, City, Address, Interface, and user.Footnote 50
Importantly, both models emphasise the relationship between sovereign authority, electronic networks and physical infrastructure. Simply put – they highlight the connections between the ‘electronic’ and ‘physical’ worlds. This is important, adding further weight to the challenge this article poses to the libertarian understanding of cyberspace. Less romantic and attractive than mystery, cyberspace is composed of cables, data centres, protocols, energy supply, buildings, and people. At once physical and electronic, these layers combine to create what Bratton terms ‘new spaces in its own imaginary: clouds, networks, zones, social graphs, ecologies, megacities, formal and informal violence, weird theologies, all imposed on the other’.Footnote 51 Sovereign authority exists within this new environment, adapting and responding to changes around it.
State sovereignty and national boundedness
The dexterity this physical infrastructure affords sovereign states is not insignificant. Building and expanding particular networks, governments have been remarkably adept at shaping the form and nature of cyberspace for strategic or political ends. China is perhaps the most widely discussed, famously constructing her ‘Great Firewall of China’ to advance the country's comprehensive censorship system.Footnote 52 The manner in which China was able to vanquish Google and now appears to be able to defeat even complex anonymising software for browsing the web underlines the degree of control. Recent revelations of possible connections between Huawei and the Chinese Government are seen by some as the latest in a series of efforts to control and influence this so-called ‘fluid’ space.Footnote 53 These events, however, are often incorrectly viewed by the libertarian narrative in isolation. Rather than single events or attempts by particular ‘rogue’ states to subvert the existing status quo, this is perhaps indicative of a widespread appetite and ability of states across the globe to extend their authority by working together. Typically, the cybersecurity treaty signed by China and Russia in 2015, which was largely about containing threats from dissident elements.Footnote 54
The weaponisation of cyberspace for strategic advantage is perhaps the clearest example of this trend. Sitting awkwardly with acclamations of the untamed ‘Wild West’ by premiers, is emerging evidence of advanced states intentionally harvesting the power of cyberspace for their military and political advantage. We perhaps need to pay far more attention to these examples. This use of offensive measures in this context is something that many government officials have been rather anxious to avoid discussing, and have therefore only recently surfaced in the academic literature. Yet it is now clear that these practices actually began during the 1980s, suggesting that the US National Security Council mapped a geostrategic mindset from the surveillance leviathans of the Cold War directly into cyberspace.Footnote 55
In May 2017, the world received a sudden reminder of the dangers of cyberspace in the form of the WannaCry ransomware attack. This was a global cyberattack by crypto-worm, a self-propagating virus, which focused on computers running the Microsoft Windows operating system. It worked by encrypting data and demanding ransom payments in the Bitcoin cryptocurrency in return for the release of the data. Although Microsoft had released updates to close the vulnerability, WannaCry continued to spread among organisations that had not updated their computers, or who were using Windows systems that were no longer supported. Although Microsoft released further emergency patches, Wannacry was mostly stopped because of the chance discovery of a kill switch that prevented it from spreading. Impacting some 200,000 computers across 150 countries, and with costs estimated in the billions of dollars, security experts pointed to North Korea or agencies working for the country, as the culprit.Footnote 56 In December 2017, the United States, United Kingdom, and Australia formally asserted that North Korea was behind the attack.Footnote 57
With no immediately identifiable author, the attack appeared to validate the Wild West view of a libertarian cyberspace, inhabited by criminals and paedophiles, a world in which states were weak and losing control. The subsequent attribution to North Korea only seemed to underline the idea of rogue actors. However, over time a rather different story emerged, which pointed instead to the world's information hegemon and underlines how organisations like America's NSA and Britain's GCHQ work to maintain information supremacy. Although the press initially suggested that Wannacry was ‘released’ by an individual hacker, and then blamed North Korea, in fact, the North Korean hackers were mere middlemen. WannaCry was built around an exploit called EternalBlue, developed by the codebreakers of the United States National Security Agency (NSA) for older Windows systems. EternalBlue was then stolen and leaked by a group called ‘The Shadow Brokers’ a few months before the attack. As Amy Zegart, Stanford University's top intelligence expert has argued, this was a problem of NSA's own making. In other words, the wildest part of the Wild West was actually being developed by the most powerful government in the world to attack other governments.Footnote 58
Statements by Microsoft officials hint at the possibility that both Microsoft and the NSA were aware that this electronic weapon had been leaked, producing a ‘patch’ or ‘cure’ for systems through updates in March 2017, perhaps explaining why many of the systems successfully targeted were running old software that had not been updated, notably the UK National Health Service computer systems.Footnote 59 This not only demonstrates the curious power relations that exist in cyberspace, it also reminds us that sovereign states remain central to the stability or instability of the Internet, working ‘behind the screens’ of the average computer user scrolling through their social media. This apparently ‘uncontrollable space’ remains commanded by traditional ‘weary giants’, at least in terms of the exclusive ability to manufacture advanced persistent threats.Footnote 60
Accordingly, President of Microsoft Brad Smith has been vocal about states in cyberspace. He used the Wannacry attack to argue that sovereign entities and their territories enjoy too much power, not too little. If there was a potential for chaos on the Internet, it lay not with drug dealers prowling the dark web but with the military-intelligence complex: ‘Governments of the world should treat this attack as a wake-up call. This attack provides yet another example of why the stockpiling of vulnerabilities by governments is such a problem.’Footnote 61 In 2017, speaking at RSA, the world's most important Internet security conference, he outlined his ideas for a Digital Geneva Convention focused on restricting what he called the growing problem of electronic attacks by states on citizens in times of peace. While the issue of precise nature of the partnerships between government and industry in the development of ‘Cyber-vulnerabilities’ is beyond the scope of this article, this episode is a clear example of states using cyberspace as a tool of statecraft. It also alludes to a broader question of liability, raising the possibility that courts may find states to be both the source and the legally culpable body for some attacks, ultimately paying compensation to corporations and citizens. The fact that the CIA appointed an experienced lawyer in 2008 to consider legal liabilities arising from its cyberwarfare and information programmes underlines the fact that these activities have become routine for the United States for quite some time.Footnote 62
The United States is not the only country that has dissembled. For decades, the more powerful governments have framed cyberspace through the anarchic language of the ‘dark web’, justifying measures that allow them to further extend their control. The Indian government is one example. Crafting an image of this environment as an uncontrollable ‘dark place’,Footnote 63 draconian encryption policies were justified in 2015, pressuring Internet companies to share access to encrypted data with law enforcement agencies. Echoing similar accusations that emerged in Britain and the United States as the result of the revelations made by Edward Snowden, Morsi's government, leading the world's largest democracy, was accused of devising a ‘snooping and spying orgy’ through such policies.Footnote 64 Such brazen activities by the world's largest democracies sit awkwardly with suggestions by writers like Haufler that the Internet's decentralisation and fluidity restrict efforts to ‘design and implement effective regulations through top-down, government-by-government approaches’.Footnote 65
India's attempts to regulate and compartmentalise cyber ‘space’ are further illustrated by the new Information Technology Act, obliging all companies wishing to collect data on citizens to do so through agreed standards, obtaining consent and privacy policies, influencing standards for data retention, and processing within national boundaries.Footnote 66 Considering India's command of 43 per cent of the global business outsourcing for the Information Technology sector, the Indian intervention is significant, with a potential future global impact. This stands in contrast not only to the libertarian presentation of an ethereal space, but also to neoliberal assertions that ‘frontiers are irrelevant to electronic flows and marketing’,Footnote 67 a growing body of evidence suggests that ‘walled gardens’ may be emerging across the ‘global commons’ of cyberspace, somewhat removed from visions of a ‘place for enacting dreams of freedom’.Footnote 68
BRICS nations are also pushing back against the Cloud. Mandating the storage of Brazilian data on its own servers in response to the Snowden affair, Brazil's approach to data sovereignty stands in direct contrast with those who champion the emergence of the global ‘data-cloud’ as undermining hierarchical control of data by states and the confirming rise of the global corporates.Footnote 69 Similarly, the BRICS’ signing of the Final Acts of the World Conference on International Telecommunications in 2012 and preference for localisation of data sovereignty extends this point, standing in explicit opposition to Western hesitancy towards ‘erecting Schengen zones for data’.Footnote 70 Drawing on Foucault, Jeremy Crampton argues that the very ‘mapping’ of cyberspace is imbued with competition, adding that this ‘space’ has been ‘made’ or ascribed meaning through the application of boundaries/territories by the powerful. In doing so, this ‘space’ has been socially constructed and understood in ways that echo the understanding of the physical world.Footnote 71
Does this potentially mean the end of the World Wide Web? The information and communications industry have been quick to note the importance of these developments. Writing in 2015, Microsoft's Eric Schmidt identified what he called the emergence of the ‘Splinternet’: in other words different cyberspaces that fit remarkably well onto the old political maps of nation-states. He insists that the ‘web has become a battleground for wars initiated by States’.Footnote 72 Rather than a global, trans-national ‘space’ of freedom, cyberspace is instead increasingly intertwined with, and constrained by, the complexities of territorial politics.
While the role and agency of powerful states like America, Russia and China are often discussed as examples of illegitimate attempts by nations to infiltrate cyberspace, they are at best, the tip of the cyber iceberg.Footnote 73 Academic experts researching the secretive world of cyberweapons not only assert their long history but also their proliferation. In the last decade, states of all sizes have entered the field, often with parallel programmes in different departments and ministries of the same government. Like any new weapon, its very ‘newness’ brings with it prestige, resources, and funding. Dana Polatin-Reuben and Joss Wright consider this idea, identifying a spectrum of approaches adopted by states to control data generated in and passing through their territories in a way that establishes a form of ‘data sovereignty’.Footnote 74 Indeed, their work similarly supports the earlier argument regarding the importance of the physical architecture of cyberspace, again identified as a key pillar of national defence and security.Footnote 75
Military lawyers such as Patrick Franzese have confidently asserted the rapid securitisation of cyberspace. Used for everything from intelligence gathering and analysis and military plans to probing networks for their strategic advantage it is now the fifth domain of warfare. Drawing comparisons with other ‘domains of statecraft’ like air and space, cyber becomes a sovereign utility, used for the advancement of national security. For Franzese, as with new technologies of air and sea that challenged pre-existing legal and social frameworks in similar ways, ‘a regime of Sovereignty’ should be established, encouraging states to ‘recognise cyberspace is a sovereign domain and to develop the technical capability to exert their sovereignty in cyberspace’.Footnote 76 By doing so he argues, the silence that currently dominates aggressive state activity in cyberspace may reduce, establishing rules of ‘acceptable State behaviour’ in this domain. Before doing so, we need to reflect on this behaviour's historic legacy, seen in its broader context. Ultimately, the historic ability of states to gain strategic advantage through cyber technologies contradicts both the global commons and deterministic neoliberal narratives about diminished states.
Arab Spring: A cyber Trojan horse?
What of claims about cyberspace accelerating social movements and civil society?Footnote 77 Emphasising the effect of global free communications, many still argue that the most significant and meaningful impact of cyberspace is the way it allows individuals to bypass oppressive regimes, acting and organising rapidly and covertly to bewilder security agencies. Sociologist Manuel Castells is one of the most prominent proponents of this view, and over the last twenty years his vision of a ‘networked society’ has been hugely influential. He has emphasised the impact of the individualisation of cyberspace, destabilising the nation-state's legitimising institutions. Moving in sympathy with the libertarian ideal of cyberspace, Castells suggests the resultant effects included the opening of global markets and the further weakening of the nation-state.Footnote 78 In the 1990s, enthusiasm for communications technologies as a means of spreading democracy informed many aspects of US foreign policy and indeed notions of soft power.Footnote 79
The Arab Spring re-energised this logic, giving vivid examples of the way that cyberspace, and in particular, social media platforms like Twitter, were advancing freedom in the Middle East. Key figures leading the revolutions in the Maghreb, also in Egypt, Yemen, and Bahrain seemed to be adept in the use of digital technologies and had been deliberately upskilled by NGOs. Similarly, YouTube and other video archiving platforms allowed citizen journalists, together with exiles and refugees, to communicate through mobile phones, cameras, and consumer electronics, broadcasting and operating independently to both state boundaries and traditional political parties.Footnote 80 In 2013, Philip Howard and Muzammil Hussain argued in their influential book Democracy's Fourth Wave? for the liberating effect of cyberspace and social media. Authoritarian leaders, they insisted, were crippled by an army of ‘20 and 30 year-olds without ideological baggage, violent intentions or clear leadership’.Footnote 81 In contrast to previous media forms like the printing press and the radio, the peer-to-peer components of cyberspace offered a platform to ‘organise, build networks, and create social capital and political action’.Footnote 82 Following the immolation of Mohammed Bouazizi in Libya, their analysis found social media (notably Facebook/Twitter) to have been pivotal in disseminating information and raising awareness, ‘putting a human face on political oppression’.Footnote 83
Yet almost a decade on, these optimistic conclusions stand in need of some qualification. Scholars are now less convinced of the catalytic effect of social media on these events. Moreover, during the last decade, we have seen Middle Eastern governments gradually recover control over their territories. Importantly, cyberspace has remained central to this resurgence. Like China, Russia, Brazil, and India, they have used this strategic technology to ‘censor, surveil, and disrupt protesters and to actively cultivate alternative nationalist movements using “bots” and armies of fake users’. Ironically the technologies that many academics saw as vehicles of liberation have now been revealed as squashing civil society. Where they have not mastered the Internet, they have simply shut it down, blocking encrypted communications.Footnote 84 Therefore, while social media and networked technology can facilitate community and social organisation, they simultaneously offer sovereign states new ways of asserting or extending their control and authority. At their most extreme, Dana Moss has shown how authoritarian governments can use the Internet to pursue their own diaspora abroad. Internet communication technologies can in fact globalise social control by regimes and impact anti-regime diasporas. Syria has successfully used these techniques to deter many from using the Internet to contest the Assad regime.Footnote 85
In April 2017, the Turkish Government turned to these electronic forms of statecraft. They restricted common platforms like Wikipedia, social media, and even dating sites following the assassination of the Russian Ambassador to Turkey. Fearing they were losing virtual control of their citizens, the government acted swiftly and concisely, cutting off access to swathes of the population, as if to remind their populous of their underlying ability to influence these ‘free’ spaces.Footnote 86 In contrast to the ‘liberation technology’ narrative, Espen Rod and Nils Weidmann highlight the instrumental, strategic value of Internet technology for government agencies connected to processes of surveillance, monitoring, and control. Having conducted a large-N analysis of authoritarian countries for the years 1993–2010, they concluded firstly that ‘regimes aiming to prevent any independent public sphere are more likely to introduce the Internet’. They also suggest that their findings indicate that the Internet has did not contribute to a global shift towards democracy during this period.Footnote 87
Rather than liberating, social media platforms become an additional layer of bureaucracy and control. Applying for licenses and collecting data on the viewing histories of their citizens, Internet-related technology thus becomes indicative of government presence and surveillance. Propagating ‘correct values’ and identifying domestic opposition movements, this directly challenges the liberation narrative. This is a significant intervention, at a time when social media and other platforms are under intense scrutiny for conceding to requests from suspect government organisations. Consequently, despite the temptation and allure of the libertarian narrative of individual empowerment, governments remain able to affect Net transaction costs, applying filters, cyber-fences, and other mechanisms to effectively regulate Net transactions.Footnote 88 As Daniel Drezner concludes, despite its promise and potential for liberation, ‘when necessary, governments of every stripe have been willing to disrupt or sever internet traffic in order to ensure that their ends are achieved’.Footnote 89
Perhaps Rod and Weidman's most significant contribution is their suggestion that it is precisely this networking potential that creates an ‘incentive for control’ for autocratic elites, concerned with containing public sentiment and opinion, with 82 per cent (28/34) of countries studied having some form of Internet censorship or other.Footnote 90 Graham also explores this idea, arguing that we must ‘debunk the substitution-ist myths of technological determinism … allowing us to reveal the socially contingent effects of new technologies … and ways in which some groups, areas and interests may benefit … while others actually lose out’.Footnote 91 Therefore, while cyberspace enables individuals to communicate across national boundaries, the freedom and privacy they experience is often illusory. The platforms through which social movements (and other organisations) operate remain embedded within broader systems of hierarchy and centralisation, predominantly connected to physicality and territory.
There is also little question that states have sought to exaggerate the extent to which privacy technology provides protesters and insurgents with effective cloaking in the expectation that this will yield more data. Indeed, one of the main vexations that American intelligence officers expressed with the Snowden revelations was that individuals were suddenly apprised of this state capability.Footnote 92 In 2003, it was noticeable that the US occupation forces in Baghdad gave the population a modern mobile phone network even before restoring water and medical services, since this allowed them to gather intelligence on the insurgents who used it.Footnote 93 If engaged in a long-term struggle, the Internet is potentially a trap for social movements, since it strips away the very anonymity that protesters and rebels have historically required to survive. Reliant on apps, platforms, and other electronic tools that remain embedded within the ‘global network’, which that allocates everyone a number, often their IP address. Cyberspace may give the protester short-term ‘flash mob’ advantage, but over time it probably tips the balance of advantage back in favour of the security forces of sovereign states. In the wake of the Snowden revelations, we need to ask whether there has ever been a time in which major states have ceded authority over this environment, while minor states seem to have caught up relatively quickly.Footnote 94 What we have seen over the last decade points not so much to anarchy but instead to the growing ability of sovereign entities to utilise cyberspace for their strategic advantage, often in secret. Meanwhile, their warnings about electronic ‘anarchy’ validate controversial policies that may otherwise have been rejected by their parliaments and populations.
Longer histories of involvement
Perhaps the most fundamental misconception of our understanding of the relationship between cyberspace and sovereignty is a perception of newness. Historicising these issues, as well as grounding them in geography, is important if we are to fully appreciate the long-term use of cyberspace by sovereign states for their strategic advantage. Sovereignty has been fortified, not eradicated, under globalising conditions.Footnote 95 To date, the International Relations literature has not only conformed mostly to the myth of libertarian romanticism, it has tended to analyse state interventions in this realm as surprising, episodic, or untypical. Arguably this interpretation, which often commands considerable consensus, stands in need of adjustment.Footnote 96
More than a century ago, the infosphere formed a central tool of state advantage. Military communications and technical surveillance practices across the globe, beginning with the telegraph, were adopted and exploited by states to exercise control and also increase warning in an era of increasing strategic mobility. Indeed, historians have described the telegraph as the ‘Victorian internet’.Footnote 97 The electronic intelligence revolution that many associate with Alan Turing and Bletchley Park had in fact mostly arrived by 1918. As Dan Larsen has demonstrated, these surveillance activities were conducted in strikingly similar ways during the First World War as during the Second World War.Footnote 98 States exerted their influence across cable networks that were prized geographical assets and legacies of empire. Companies like Cable & Wireless Ltd worked closely Whitehall in the much the same way as British Telecom has done in the current century. Radio masts and listening stations were established across the globe and the desire to intercept or manipulate these transmissions for reason of intelligence or propaganda has often inspired larger states to cling to small islands in obscure parts of the world in what appear to be time warp remnants of empire.Footnote 99 These physical structures expanded and changed with the arrival of microwave telephone networks, mobile phone networks, telecommunications satellites, and finally fibre optic cables. While the technology advanced, their critical importance to states did not waver.Footnote 100
The increasing volumes of data intercepted by the intelligence agencies have indirectly driven the development many of the devices that now surround us. In particular, the vast increase in the volume of intercepted diplomatic traffic from Third World countries by the 1960s drove the demand for high-speed computing from companies such as IBM and specialist contractors such as Cray Corporation. Directly or indirectly, by the 1980s, organisations like the NSA employed a vast number of computing PhDs from universities and as one of the foremost historians of computing has observed, we have yet to tell the real story about the history of computing.Footnote 101 Moreover, the boundless volume of intercepted clear voice traffic grabbed by the signals intelligence agencies after 1960 also drove an entire field of advanced linguistic computer translation and computer voice recognition. Computing as a whole is perhaps ten years further advanced because of these defence-driven applications, moreover, devices such as Amazon Alexa or Google Mini now appearing in our homes have their origins in research commissioned by NSA and GCHQ.Footnote 102
While for many, the networks and ‘clouds’ that comprise cyberspace are unrelated to historic practices, they are intrinsically connected.Footnote 103 Incorporating these into a longitudinal analysis illustrates the ways cyberspace has been used by states as a central tool of statecraft. Indeed, it is probably worth remembering that the Internet itself was an outgrowth of American military science research.Footnote 104 Established in the late 1960s by the National Science Foundation, ARPA quickly expanded on a global scale, enabling governments and military apparatuses to conduct a range of government business, including major nuclear and naval technology projects more efficiently. Although many argue that the ARPANET bears little resemblance to the cyberspace of today, we need to recognise that its foundations were laid by the Pentagon.Footnote 105 Super-computing and the Internet both lack a single founding father because they emerged out of a myriad of advanced defence projects.Footnote 106
With this come boundaries, labels, designators, and order. A compelling example is the little-known Domain Naming System (DNS), hierarchically allocating names and IP addresses to new ‘entrants’ of cyberspace. Viewed by many as a neutral, technical matter of limited political relevance, DeNardis has nevertheless explored the political underpinnings of this system, echoing Bratton's (and Foucault's) claim that an ability to address something means you can govern it.Footnote 107 Despite superficial impressions of transnationalism and unfettered freedom, every email and website address is attached to and aligned with a physical territory. Allocated hierarchically, Domain names like ‘.fr’ (France) and ‘.co.uk’ (UK) connect a user or company with a territorial place. They connect them to a sovereign state. Stephen McDowell, Phillip Steinberg, and Tami Tomasello highlight the broader issue of government's ‘managing’ of the Internet in these ways, showing how actors construct the ‘infosphere’ to ‘achieve specific ends’.Footnote 108 While a determined user can certainly employ time-consuming techniques to mask their identity online, most websites and Internet users are reliant upon these structures to engage with and use Internet technologies. For the majority of communications and transactions the idea that the Internet escapes a location is implausible.Footnote 109
Thus, the history and geography of cyberspace cannot be properly understood without a reinterpretation of its connections to its material foundations and reliance upon sovereign states. Many social scientists and policymakers remain persuaded of the libertarian portrayal of cyberspace as weakening centralised power and control. Others have assumed that governments have only recently begun to reassert control over what they assumed to be an anarchic space. This is perhaps understandable since so much of the history or surveillance and indeed computing generally contain missing elements that historians will be trying to unpick for decades to come.Footnote 110 But we now know enough to conclude that this technology probably reinforces traditional hierarchies, making the more powerful states like Russia, China, and the United States yet stronger.Footnote 111
Conclusion
Cyberspace has consistently been used as a tool of statecraft. Although electronic ‘space’ has become increasingly important for a range of activities, from storage to communication, territorial ‘place’ remains prevalent. Temporality is equally important, not only in understanding the significance of the development of this physical infrastructure over time, but also in appreciating that many technical innovations were driven by the supreme efforts that security researchers exerted to stay ahead of the curve in this realm. The United States as an information hegemon, and more specifically the National Security Agency, have done more than many suspect to shape the current information environment. Meanwhile, smaller states have taken a little while to catch up and achieve purchase over new systems. As the Arab Spring illustrates, this can wrong-foot regimes in the short term, but in the long term there is every sign that states are back in control.
What is most striking about the role of states and sovereignty in cyberspace is that often this interest is itself bounded. In striking contrast to our vision of cyberspace as global, governments appear to privilege internal control and authority, and so are less concerned with directly challenging the activity of other nations in cyberspace than with establishing control within national jurisdictions. Indian and Brazilian pursuit of ‘data sovereignty’ among the BRICS nations could be viewed accordingly, with new regulations promulgated that appear to challenge the globalisation of telecommunications. Even major cities like Los Angeles are paying higher fees to ensure that their data is stored locally and not in the Cloud. The emerging ‘splinternet’ is a global phenomenon and not just about China, Russia, or Iran.
The true rebels in cyberspace were figures like computer hacktivist Kevin Mitnick who devoted his life to evading the rules and believed that cyberspace provided a secret playground, but his eventual destination was a penitentiary. Hackers and cyber criminals enjoy sanctuary in locations such as North Korea and the Balkans, but even here their immunity depends on the world of states, either ‘pariahs’ or places with poor governance.Footnote 112 Social scientists might well devote more attention to the material, geographical underpinnings of cyberspace. While libertarian, anarchic representations are attractive, especially to politicians and policymaker rolling out new regulation, they often overlook the role and significance of physical geography and sovereign power. More importantly, they underappreciate (if not misrepresent) the historic and enduring ability of sovereign entities to influence and manipulate this ‘new’ environment for their advantage.
While this analysis is on one level pessimistic, there remains room for optimism. Understood as the product of social relations, cyberspace has the potential to change. It is not that cyberspace has and will always be used as a tool for oppression, eradicating its usefulness for social movements and other liberating forces. It is, in the words of Doreen Massey, ‘unfinished’, always being made and recreated. How this ‘space’ develops remains unknown and unconfirmed. Recognising the historic and underlying political relations that led up to its current existence is significant if we are to understand its future.
Perhaps the most compelling illustration of the connections between cyberspace and sovereignty were made not by political scientists, but by Trevor Paglen, an American artist whose work tackles mass surveillance and data collection. Using geography and satellite imagery, Paglen masterfully depicts the physical, material presence of the surveillance state. Doing so, he suggests that ‘infrastructures of power always inhabit the surface of the earth somehow, or the skies above the earth’. ‘They're material things, always, and even though the metaphors we use to describe them are often immaterial – for example, we might describe the internet as the Cloud or cyber-space – those metaphors are wildly misleading.’Footnote 113 Building on the argument here and echoing the long-overlooked appeals of geographers, he shows how state secrecy and surveillance are produced through space.Footnote 114
A more material approach also requires stronger attention to history and less emphasis on ‘newness’. We might question approaches that emphasise the revolutionary impact of cyberspace that frequently results in overblown claims of transformation. Understood in connection to or with similar technologies of telephony and printing, the perennial nature of debates surrounding the relationship between ‘place’ and ‘space’ and the relationship between social transformation and technology emerges in a longer perspective. More importantly, a more measured approach allows us to reappraise the meaning of the Snowden revelations on surveillance and state hacking. Widely acclaimed as the high watermark of Western surveillance, most discussions of current electronic spying practices are attended by hyperbole, being treated as exceptional or ahistorical. However, this article suggests the possibility that sovereign entities of all stripes have historically recognised the strategic value of exploiting cyberspace for their advantage all the way back to Bletchley Park. As James Bridle has shown, Snowden's so-called revelations were in fact on a continuum with other insights into surveillance over decades, if not centuries. States have always been in control of communications and there is less disorder in this space than we have been led to believe.Footnote 115 This absence of anarchy is either rather reassuring or rather worrying, depending on our point of view.
Acknowledgments
This research received support from the Economic Social and Research Council (ESRC). Special thanks also to the anonymous reviewers and editors at EJIS for their helpful and constructive comments.
Sarah Mainwaring is an ESRC PhD Candidate at the University of Warwick. Researching the political history of encryption since the end of the Second World War, her interests include ‘Cyber’, International Security, and Intelligence Studies.
