Master Thesis Political Science
European Union in a Global Order
Security, Democracy and the Identity of the European Union in an age of
Mass Surveillance
The LIBE Inquiry as a desecuritizing moment
Source: European Parliament
Valentin Gros
June 2014Supervisor: Dr. Stephanie Simon
TABLE OF CONTENTS
I. INTRODUCTION ... 2
II. METHODOLOGY ... 6
III. SECURITY, DEMOCRACY AND IDENTITY ... 9
III. 1. From intelligence to mass surveillance: balancing security and democracy ... 9
III. 1. 1. Mass surveillance, an exceptional measure justified for national security purposes ... 9
III. 1. 2. Democracy, secrecy and the governance of intelligence ... 12
III. 2. Desecuritization: a political strategy that legitimates security tools ... 17
III. 2. 1. Securitization and exceptionalism ... 17
III. 2. 2. Security in a political perspective: desecuritization as a choice ... 21
III. 2. 3. The identity of the European Union at a global level ... 23
Concluding remarks ... 26
IV. THE ABSENCE OF SECURITIZATION: THE DISCOURSES THAT FAILED ... 29
IV. 1. Limited reception of securitizing discourse: increasing surveillance for increasing threats? ... 29
IV. 2. Challenging the veracity of the Snowden’s revelations ... 33
V. THE EXTENDED INTERPRETATION OF THE ‘NATIONAL SECURITY EXCEPTION’ ... 37
V. 1. Mass surveillance for what purposes? ... 37
V. 2. The undermined capacity for the European Union to have control over mass surveillance ... 39
VI. RELOCATING MASS SURVEILLANCE AS A STRATEGY FOR REGULATING INTELLIGENCE WORK ... 42
VI. 1. Mass surveillance and its effects on civil rights ... 42
VI. 2. Framing intelligence services’ activities in relation to democracy ... 45
VI. 3. Regulating mass surveillance through the lens of cybersecurity? ... 49
VII. ANALYSIS OF THE CONSTRUCTION OF THE EUROPEAN UNION IDENTITY ... 52
VII. 1. Data protection, democracy and digital governance: the EU as a norm-setter? ... 52
VII. 2. Projecting a ‘European way’ ... 54
CONCLUSION ... 56
I. INTRODUCTION
A strict observance of the written law is doubtless one of the high duties of a good citizen, but it is not the highest. The laws of necessity, of self-preservation, of saving our country when in danger, are of higher obligation. To lose our country by a scrupulous adherence to written law, would be to lose the law itself, with life, liberty, property, and all those who are enjoying them with us; thus absurdly sacrificing the end to the means.
Thomas Jefferson (1810)
In the ‘War on Terror’, does the end justify the mean? How far can civil rights and democracy be suspended in the name of security? Since 9/11, the fight against terrorism has been elevated as a priority for security that has justified infringements on fundamental rights and the rule of law. The recent revelations made by the former NSA contractor Edward Snowden have undoubtedly affected the general understanding of the extent to which states would be willing to surveil the population in the name of the ‘War on Terror’, where the nature of, and the scope covered by the disclosed surveillance apparatus have reached an unprecedented level that clearly sets a rupture in terms of data collection practices (Bigo, et al., 2013:7). In this post-Snowden era, this thesis examines the relation between democracy and security in terms of political choice – where the adoption of a security measure crystallizes a particular balance between democracy and secrecy – and looks at how different actors might challenge those security measures and call for a more democratic or a more securitized approach. The Inquiry carried out by the LIBE Committee is an interesting case that highlights how the European Parliament (EP), on behalf of the European Union (EU), has opposed mass surveillance. Several contributions have underlined the negative impacts of such large-scale data collection and processing on fundamental rights and democracy and called on the EU for a better protection of its citizens (Bellanova et al., 2013; Bauman et al., 2014; Carrera, Guild & Parkin, 2014). However, there has been no analysis of the debates per se. This research tries to enhance the comprehension of the European position after the revelations by looking at how the European Parliament, during the LIBE Inquiry, has framed mass surveillance and made policy recommendations for the EU. Since policy-frame is “a normative-prescriptive story that sets out a problematic policy problem and a course of action to be taken to address the problematic situation” (Law & Rein, 1999:3), the LIBE hearings were a unique opportunity to analyze the elaboration of the European reaction to the NSA revelation, and how the framing of mass surveillance impacts on the construction of the EU’s
identity. Consequently, this thesis aims to analyze how the security actors involved in the parliamentary inquiry of the LIBE Committee framed intelligence services’ mass surveillance activities.
As a quick overview, the first revelations were made on June 6th
, 2013 when The Guardian and The Washington Post disclosed some of the information leaked by Mr. Snowden, revealing the existence of the surveillance programme PRISM that enables the American National Security Agency (NSA) to collect electronic data and analyze data from big IT corporations such as Yahoo, Google, Facebook and Apple (Greenwald, 2013). Over the following months, the information published by several newspapers revealed the existence of at least three surveillance programmes launched by the NSA, namely PRISM, UPSTREAM and X-KEY SCORE (Ball et al., 2013), the complicity of the ‘Five Eyes’ and several European Member States including the United Kingdom, France, the Netherlands and Sweden in this surveillance, the snooping on diplomatic missions and officials in Europe and Brazil, and a major public-private partnership allowing the access to electronic communications and phone call records.
In a European context, such massive surveillance is an interesting case that tackles both the capacity for the EU to act globally when the rights of its constituents are at stake and the complexity of the concept of ‘security’ upon which were set up and justified these intelligence services’ activities. The European opposition to these practices has been challenged by the power given to states under the ‘national security’ justification. Here, the concept of ‘national security’ involves two important notions. On the one hand, it deals with the protection of values socially constructed and associated to a particular state (Trager, Simonie and Wolfers referred to in Balzacq, 2003-4: 38 and 42). Such definition implies both an objective notion of (national) security as the absence of threat, and a subjective perception of the minimization of a danger that might affect those values (Balzacq, 2003-4:42). On the other hand, from a legal perspective, ‘national security’ is an opt-out provision that can be evocated in order to free the party of the “constraints otherwise imposed” by the text (Goodman, 2001:101). It is therefore associated with exceptionalism and emergency. In the context of the NSA revelations, ‘national security’ has been embodied in the secrecy that has surrounded intelligence activities, and the consequent uncertainty about the veracity of the information publicly available. The European Parliament therefore mandated the Committee of Civil Liberties, Justice and Home Affairs (LIBE Committee) to launch a series of inquiries from September 2013 to January 2014 in order to collect information and provide recommendations for future legislation (European Parliament, 2013). However, as the United
Kingdom expressed when declining the invitation of the European Parliament to testify before the Committee, “national security is the sole responsibility of Member States. The activities of intelligence services are equally the sole responsibility of each Member State and fall outside the competence of the Union intelligence and secret services” (Carrera, Guild & Parkin, 2014:1). Mr. Claude Moraes, rapporteur for the EP on this issue, replied to this claim and stressed that the LIBE Committee “strongly rejects the notion that all issues related to mass surveillance programmes are purely a matter of national security and therefore the sole competence of Member States” (LIBE Committee, 2014:22). In other words, both the EU and the European states tried to claim competence over the issue.
In this perspective, this thesis argues that the actors involved in the hearings tend to desecuritize such massive collection of data, rejecting the idea that these practices can be only understood as a matter of ‘national security’, and locate it beyond this field of exception but as linked to democracy so as to justify EU competence on this subject. As a consequence, they also participate in depicting a particular role of the EU on the global stage.
After a short presentation of the methodology used in this research, this thesis elaborates a theoretical framework that examines the relation between security, democracy and (EU’s) identity in an era of mass surveillance. This framework addresses three aspects. First, it tackles the balance between the recourse to data for security purposes and the negative consequences that such security measure engenders. In line with recent studies, this analysis argues that unique scope and complexity of surveillance at hand cannot be limited to a trade-off between security and privacy, but has to be framed in terms of democracy (Bigo et al., 2013). Second, it considers securitization theory as an analytical framework that emphasizes how discourses and security instruments shape what constitutes ‘security’. It argues that securitization is mostly a political process that implies a choice about the security tools used to respond to a perceived threat, namely either in favor of slower, more democratic procedures in ‘normal politics’ (desecuritizing) or ‘extraordinary’, secretive measures in ‘security politics’ (securitizing). This choice reveals how policymakers, by favoring democratic or exceptional measures, project a particular identity of the political community (Huysmans, 1998). Following is the analysis of the LIBE hearings as a desecuritizing moment that simultaneously reinforces the representation of the Union as a ‘European data protection model’ and expands its competence over security matters. The analysis is divided into four chapters. The first looks at the discourses and arguments that failed in the debates, either because they were not put forward on the agenda or because the MEPs opposed them, and emphasizes the failure of securitization discourse. Instead, the second part stresses that the
LIBE actors framed the issue as a broad interpretation of national security that limits the capacity for the EU to control mass surveillance. The third part emphasizes that, by relocating mass surveillance, the actors sought to foster the capacity of the EU to regulate intelligence agencies’ activities. The last part focuses on the emergence of the European identity in the debates, and argues that the MEPs tend to foster the EU’s security actorness while expanding its norms globally. The conclusion presents some limits to the case and calls for further research on the European position after the Snowden revelations.
II. METHODOLOGY
In this research, the central material for the empirical analysis was the Inquiry organized by the LIBE Committee on Electronic Mass Surveillance and its Impact on European Citizens. From September 2013 to January 2014, sixteen sessions were held, during which the Members of the European Parliaments welcomed professionals from various backgrounds, including journalists, lawyers, European officials, head officers from the private sector, representatives of the civil society, researchers and whistleblowers, who provided information on the surveillance programmes at hand. Each panelist had about ten minutes for a short presentation before several rounds of questions and answers with the Members of the LIBE Committee.
This analysis looked at the LIBE hearings as a key political moment in the elaboration of the European ‘response’ to the Snowden revelations where the agents projected, through discursive practices, a particular representation of the state of affairs. Here, discourse was not viewed in an inclusive way, which would have only focused onthe language and thus implied an analysis of the semantic and stylistic devices (Fairclough, 2013:181). Such approach could not be used considering that the debates involved several languages and that the translation might not render the exact same meaning in the interventions. Instead, the analysis embraced a broader definition of discourse as a discursive practice in line with a poststructuralist discourse analysis1
, according to which the scope of discourse is extended “beyond the analysis of ‘text and talk in contexts’ to social actions and political practices” (Howarth and Griggs, 2012:308). In the context of the LIBE hearings, discourse was first considered as a political practice that involved political agents, the MEPs, who had to face a particular situation and envisage further political actions. As van Dijk argued, “politicians are not the only participants in the domain of politics” (1997:13); thus, by contributing to the debates, the experts were also taking part in the political process. Consequently, in a political context, not only does a poststructuralist approach emphasize that (political) discourse articulates the representations of the issue at stake and the solutions envisaged, but it also focuses on the relation between discourse and identity. Second, discourse is viewed as a security practice whereby security experts negotiate what constitutes the domain of (national) security by mobilizing different concept and representations (Hansen, 2006:31-32).
1 The boundaries between poststructuralist discourse analysis and critical discourse analysis are quite blurred, as underlined Norman Fairclough (2013).
Since desecuritization is about associating an issue with one specific sector, this thesis analyzed discourse by identifying the major themes or “topics” (van Dijk, 1997:28) that emerged in the debates in association with mass surveillance and the activities of intelligence services. In order to do so, data was collected after watching the sixteen hearing sessions and writing the transcript of the debates, considering that there was no official written transcript available. The statements of the actors involved were written as faithfully as possible to the exact words that were pronounced by the participants or the translators, including the questions-and-answers sessions, the comments and the remarks that were made. This constituted the main basis for this research.
In a second time, the hearings were examined in an interpretative and qualitative way. Indeed, there were no hypotheses formulated before the empirical analysis. On the contrary, the themes were identified throughout the Inquiry. The argumentation had to be adapted throughout the analysis of the debates depending on the findings. As an example, the reference to cyber security in the debates has necessitated to adapt the interpretation of the formulation of the EU’s identity. In order to identify the major themes, a research design was elaborated, based on the combination of a poststructuralist and critical discourse analysis. Several methodological questions, which are characteristic of those approaches, were elaborated after the data was collected in order to make it ‘actionable’. In line with Sum’s poststructuralist approach (2009; 186,198-9), the first questions were: (1) where do specific policy ideas originate; (2) which actors get involved in the policy discourse that constructs mass surveillance and the information provided by Snowden, and which do not; (3) what ideas are selected and drawn upon to recontextualize this issue and (4) how do the actors refer to the role of the European Union. Another set of questions, from the critical discourse analysis (Fairclough, 2013:195), focused more on the legitimation process and the recommendations envisaged, namely (5) how is the current state of affairs described and evaluated; (6) what impacts of mass surveillance emerged in the debate and (7) what recommendations are made. Those questions made it possible to underline the conflicting and converging discourses as well as the main topics present in the debate in a more objective way, even though other studies of the same case might identify different key themes.
Here, the goal is to analyze how the European Union reacted to the Snowden revelation, and how this reaction impacted the construction of the European Union’s identity. This posed two challenges in the approach to analyzing the data. The first problem related to how the European Union was understood since it have would been misleading to assume that the position of the European Parliament stood for the reaction of the Union as a whole.
Nonetheless, it is worth looking at the debates organized by the LIBE Committee for two reasons. First, it shows how EU policy-makers engaged with the notion of mass surveillance and underlined how diverging discourses and arguments are brought up. Second, the European Parliament has had a greater role in defining EU competence in the field of security (de Goede, 2011; Guittet, et al., 2011) especially since the Lisbon Treaty increased its co-decisional power and, subsequently, put “additional requirements on the EP to engage actively with the monitoring of EU initiatives in the field of internal security” (Guittet, et al., 2011:11). Considering that the European Parliament provided policy recommendations for the EU, and that the final report was eventually adopted in plenary session on March 13, 2014, the analysis of the debates underlined how the EU might tend.
The second challenge was more methodological. Since the LIBE Committee was welcoming panelists to come and testify, it was necessary to take into consideration the position of the MEPs, and how they tended to frame the issue so as to underline the position eventually adopted by the EP. However, looking only at the questions and comments made by the Members of the Parliament would have neglected the intersubjective aspect of the debates. Indeed, the hearings expanded over five months during which each panelist presented its own analysis of mass surveillance and its impact, and the members of the LIBE Committee developed their discourses by commenting and asking questions to the experts. There was thus a tension between the intersubjective pattern of the Inquiry and the identification of the position peculiar to the MEPs.
Two steps have been made in order to tackle this issue. First, the outline of the empirical section was organized thematically, within which the opening statements of the panelists is separated from the question and answers of the MEPs. Put differently, for the main themes that were identified in the debates in relation to mass surveillance, the empirical section first looked at the discourse of the panelists, then addressed the reactions and comments of the MEPs, so as to take into consideration the intersubjective process, before looking at whether the issue was also similarly addressed over time. For instance, a panelist might have mentioned that information gathering was necessary for businesses, but in the same hearing the MEPs might have questioned its efficiency and reused the argument of efficiency in further sessions. Second, the analysis of the MEPs’ positions in the debates was compared with the conclusions drawn in the final report of the Inquiry, in order to highlight what the Committee ultimately considered as being the priorities and the new actions to undertake.
III. SECURITY, DEMOCRACY AND IDENTITY
Large-scale data collection by intelligence services is an interesting case that underlines the relation between democracy and security-related policies. In an era where most electronic communication are collected and processed, one might wonder if mass surveillance is compatible, or necessary with the democratic realm. Is exception part of the rule of law? Because of the pervasive characteristic of surveillance, both security and democracy appear as the two sides of the same coin, or in a less Manichean vision, as a continuum. This conceptual section shifts the debate away from the normative question of the desirability (should?) for the state to gather information through surveillance and collection means. Instead, the relations between the two concepts is understood as a political choice where actors legitimize and project the way they want to govern, either through security and secretive measures, or through democratic procedures. Through this choice, policymakers and political actors – i.e. actors participating in the political process – formulate and construct a particular identity of the political community.
III. 1. From intelligence to mass surveillance: balancing security and democracy
In a post-9/11 context, the inability for intelligence activities to ‘connect the dots’ and prevent the attacks has fostered the need for public authorities to collect and process a large quantity of information in order to prevent future attacks. On the other side of the coin, the recourse to such unprecedented security practices in terms of scope and reach is nonetheless far from riskless inasmuch as it can undermine privacy rights and, more fundamentally, affect how democracy functions. Thus, intelligence agencies’ surveillance activities involve a political balance between the deployment of exceptional security measures for the protection of the nation, and their negative consequences on civil rights, and most significantly, on democracy.
III. 1. 1. Mass surveillance, an exceptional measure justified for national security purposes
The 9/11 attacks in the United States, the explosion of the train system in Madrid in 2004 and the London bombings in July 2005 justified the adoption of exceptional security measures in the United States and in Europe in order to manage the arisen threat. Among
those measures, the recourse to intelligence in the fight against terrorism has become at the core of the strategy adopted by several states (Amoore & de Goede, 2012:3). This section analyzes the security narrative that has justified the deployment of an intelligence assemblage that massively collects and stores information.
“Connecting the dots”
On September 20, 2001, President Georges W. Bush called on the mobilization of all the resources available, especially stating “[w]e will come together to strengthen our intelligence capabilities to know the plans of terrorists before they act, and find them before they strike” (Georges W. Bush, 2001). The inability to prevent the attacks in New York and Washington, D.C. has been followed by debates on the inefficiency of intelligence services to “connect the dots” and the exceptional enhancement of data collection capabilities in the name of national security (Müller-Wille, 2007:100; de Goede, 2008:97).
Walsh gives an interesting definition of intelligence in the field of security as “the collection and analysis of open, publicly available and secret information with the goal of reducing policy-makers’ uncertainty about a security policy problem.” (Walsh, 2006:626-7). Through his definition, Walsh rejects a simplistic understanding of intelligence as the mere collection of data and underlines two major characteristics that highlight the central focus given to intelligence activities in current security practices. First, in addition to collecting information, they participate in ‘activating’ it. Put differently, intelligence services operationalize the information they gather, or make it “actionable” as de Goede and Amoore referred to (2012:5), by producing analyses and recommendations that will be used by policy-makers. Therefore, the term ‘intelligence’ refers both to a process through which data are analyzed, and an output, namely the information produced, which is then used for political actions. It thus fully takes part in the decision-making process.
The second pattern of intelligence as it is understood today lies in the last part of the definition given by Walsh, namely that information gathering and processing have become strategically important for mastering the unknown. In line with the risk literature, security experts have illustrated how security has become more generally perceived through the prism of preemption in the fight against terrorism, in the name of which the implementation of anticipatory measures was justified (see Rasmussen, 2006; de Goede, 2008; Amoore & de Goede, 2008; Anderson, 2010). Put differently, data have become constructed as essential for countries in order to “secure ‘homelands’ from the risk of future terrorist attacks” (Amoore & de Goede, 2012:3), whereby intelligence work aims to identity potential threats. Intelligence
services are key actors in gathering information for security purposes; and by identifying the future risks for their nations, they have a consequent role to play in framing the potential risks for the state. As an illustration, the European Intelligence Center (originally SitCen and rename IntCen in 2012) engages, similarly to other intelligence services, in risk assessments; its analyzes are disseminated into European institutions and national state bureaucracies (Jones, 2013:4).
Technologies, ‘Big Data’ and the public-private intelligence assemblage
Due to the development of IT technologies and the predominance of anticipatory logic in security politics, especially after 9/11, intelligence work has expanded from traditional missions, such as protecting secret from enemies, and traditional sources of information, notably classified data, to the extraction of “potentially critical knowledge from vast quantities of available open source information” (Lahneman, 2010:203). Over the last decade, the exponential use of the Internet and technological developments in telecommunication and IT systems have facilitated the creation, exchange of and access to an incommensurable quantity of information. Consequently, the accumulation of data, especially personal information, through electronic communication and signals has become a key resource for security entities, from intelligence services to law enforcement agencies. They have developed “systematic methods, techniques and infrastructure” that collect, analyze and compare data in order to identify individuals through profiling and key word searchers (Bigo et al., 2013:24).
The disclosure of PRISM and the other programmes has marked a new step in the capacity for intelligence services to collect data through complex and sophisticated practices. First, they have been able to reach this ‘market’ of information and to access the content of communications. Indeed, the NSA and the GHCQ have accessed phone calls, text messages, Skype conversation and various video and audio signals (PRISM), using the satellites, cables and cloud computing (TEMPORA) through which a large part of the world’s communication transit (Bauman et al., 2014: 123; Guittet et al., 2014:8). But the major novelty resides in their collection of metadata which does not contain content information per se, but that informs on individuals’ behavior and maps out an individual and all the contacts with whom he or she has been communicating to.
Because of the quantity of information to process, intelligence services have increased the level of cooperation among each other. Cooperation amongst intelligence agencies is not a new phenomenon. The ‘Fives Eyes’ alliance has been operating since the adoption of the
multilateral agreement United Kingdom – United States of America Agreement in 1946 that binds the United States, the United Kingdom, Canada, Australia and New Zealand for cooperation in signals intelligence (Norton-Taylor, 2010). However, he information disclosed by Edward Snowden pointed out that the cooperation of intelligence has extended beyond the traditional group. Collecting data over the world has been made possible thanks to a strong collaboration among several intelligence agencies, including the NSA, the GCHQ, the French DGSE (Direction Générale des Renseignements Extérieurs), the German Bundesnachrichtendienst (BDN) or even the Swedish Försvarets radioanstalt (FRA).
Similarly to the transnationalization of threats, intelligence has become the work of an array of networks between different intelligence services, but also between the public and the private sectors. This trend has been clearly stated in the 2010 National Security Strategy in which the United Kingdom stressed the “need to build a much closer relationship between government, the private sector and the public when it comes to national security” (HM Government, 2010:5). Indeed, the far-reaching reserve of data available online has represented a great potential for the private sector up to the point where companies have built up their business models around these ‘Big Data’ (La Tribune, 2014). Since major IT companies, including Yahoo, Skype, Google and Microsoft have collected a massive quantity of information, they have become valuable for public authorities in identifying enemies. Consequently, information-gathering for the fight against terrorism articulates an assemblage of public and private entities whereby “the state is conceived as outsourcing its surveillant functions […] dispersing the capture and analysis of data into a nebulous and all-pervasive machine” (Amoore & de Geode, 2012:4). Consequently, the involvement of a multiplicity of actors in the War on Terror has blurred the traditional distinction between public and private realms that jointly act across borders, and has engendered the emergence of hybrid transnational networks involved in the collection, analysis and processing of information (Bauman et al., 2014:123).
III. 1. 2. Democracy, secrecy and the governance of intelligence
The development of a surveillance apparatus in the name of national security is however not harmless. Several affairs between the United States and the European Union, notably concerning the transfer of financial data (SWIFT) and passenger name records (PNR), have underlined that the use of personal data for security purposes comes into conflict with the guarantee of fundamental rights, notably the protection of individuals’ privacy. The
information revealed by Mr. Snowden has presented a new configuration in the use of data for the sake of security whereby the involvement of intelligence services in worldwide surveillance affects more fundamentally democracy. Indeed, on the one hand, the secrecy surrounding their activities questions how one can “balance oversight and operational freedom” of these services (Chesterman, 2011:6). On the other hand, the expansion of data collection for the sake of security articulates the recourse to exceptional measures and the suspension of the ‘normal’ rule of law. Simply put, if mass surveillance is an exceptional measure, how can it then be regulated in a democratic society? Thus, the risk here is not only that civil liberties might be sacrificed for more urgent claims of security, as Zedner argues (2009:45), but also that intelligence activities might threaten democracy. By analyzing the relations between the national security exception, intelligence activities and democracy, this section underlines that the choice of security policies reflects a particular balance between security and democracy. It is therefore necessary to look at the justification and legitimacy framed in support to mass surveillance.
Surveillance at the expense of human rights and civil liberties?
Collecting personal data for security practices already presents a challenge between the imperative of security and the protection of individuals’ rights. One the one hand, the security narrative has justified the collection of data in order to identify individuals and target risky patterns. On the other hand, personal information might be collected from innocent individuals, and misused at the expense of their individual rights. As an illustration, passenger name records (PNR) gather sensitive information that can inform about political affiliation or religious practice (through dietary requirements) and be misused for discriminatory purposes (Clavet, 2010:15).
Moreover, the recourse to a large quantity of personal data for the identification of enemies and future threats is symptomatic of generalized ways of governing security where the daily activities are monitored and processed. Security experts have indeed underlined that the protection of security is no longer limited to the territorial borders but infused in the routines of the daily life via an intensified use of technologies (Bellanova et al., 2010; Lyon, 2007), where even the body becomes a new border (Amoore, 2006). Thus, information gathering articulates a whole security apparatus that infiltrates the daily life of citizens and becomes “routinized” in daily practices (Bellanova et al., 2010: 344). The monitoring of daily activities questions the purposes justifying such large collection of information and reflects
what Roger Clarke referred to as “dataveillance”, i.e. data being used to monitor and surveil citizens (Clarke, 1988).
Thus, a key tension emerges with regard to the actual effectiveness of surveillance for security purposes considering its intrusive aspect at the detriment of human rights and civil liberties (Lyon, 2007:161). The pervasive aspect of surveillance relies both in the collection of data, which challenges the right to privacy and data protection, but also the way data is made actionable through profiling and data mining processing that categorize individuals and identify risky patterns (ibid.:163; Bellanova, 2010:346). Through this surveillance, the collection of content and metadata enables intelligence services to analyze the behavior of individuals, and in a surveillance society, “every transaction and almost every move of the citizens is likely to create a digital record” (Bellanova et al., 2010:345). The risk is then to fall into a police state that constrains individual freedoms (Bigo et al., 2014:12). Nonetheless, more than the violation of fundamental rights, the unique scope and complexity of surveillance impact, more fundamentally, democracy itself.
Intelligence services and secrecy
The involvement of intelligence agencies in a worldwide surveillance system affects democracy since secrecy limits the possibility for the constituents and their representatives to exert control over their authorities. The secrecy that surrounds the work of intelligence applies both to the content of the data collected and the practices of the agencies. One needs to recall that the purpose of secrecy in the field of intelligence is first to protect information considered as strategic. Governments can indeed decide to classify information that is related to ‘sensitive’ policy domains, including in the field of security and defense (Curtin, 2014:7).
What is relevant for this analysis is that the recourse to secrecy tends to be balanced by the capacity for intelligence to properly work and, in extenso to efficiently guarantee the security of the state. Indeed, as Curtin explains, “the purpose of the classification system is traditionally to prevent disclosure of information that could damage (national) security” (2014:1). David Pozen went further in arguing that the rationale behind secrecy claims for “concealing plans and vulnerabilities from adversaries, acting quickly and decisively against threats, protecting sources and methods of intelligence gathering, and investigating and enforcing the law against violators” (2010:277). The Director of the NSA Keith Alexander also used this justification when he said on March 4, 2014 that the leaks by Edward Snowden have hampered the efforts to protect Wall Street and civilians from cyberatttacks (Sanger,
2014). Without taking a normative stance about whether this is acceptable or not, the argument used is interesting: secrecy tends to be framed as necessary for national security.
The major challenge that emerges from this recourse to secrecy in the name of security is the compatibility of secret measures, deployed in a context of emergency, with democracy. It is, first, necessary to explicit this notion. In this thesis, this concept is here differentiated from fundamental rights. Civil rights and liberties can be considered as core elements of democracy since “democracies grant fundamental rights” (Balzacq et al., 2010: 21). However, democracy is analyzed here in relation to intelligence and identified at the intersection of the law, political leadership and the will of the population.
In his article about democratic accountability and intelligence, Björn Müller-Wille goes back to the fundamental idea that democracy means “that all state authority emanate from the people”. Democracy then implies a specific relation between the people and the leaders whereby the authorities shall be accountable to their constituents, and the elected representatives must “act in the interests of the people” (Müller-Wille, 2006:101). This approach to democracy underlines two functions of the law – as the continuity of the people’s will – that are interesting for the purpose of this thesis. On the one hand, “law defines order, and by implication disorder”, namely it identifies what is acceptable (the norm) and automatically excludes what is not in the norm (Balzacq et al., 2014:9). On the other hand, law is an instrument for the people in order to constrain the authorities. Put differently, democracy articulates the rule of law, political leadership and popular will (Huysmans, 2004:326). A democratic regime has, thus, a set of rules that prevents from any key decision to be taken “without or against the directly elected and legitimized parliament” (ibid., quoting the European Convention, Working Group VII, Working document 33, 2002:16). On this perspective, transparency, accountability and oversight are key democratic principles that aim to give the people and its representatives the knowledge and the control necessary for the effective functioning of the regime.
By restraining information, secrecy undermines those democratic principles and rearticulates the state and the civil society (Bauman et al., 2014:135). Indeed, a limited access for parliaments to the activities of intelligence agencies, notably with a restricted number of actors under security clearance, can provide the appearance of parliamentarian oversight without a real check on matters of substance (Clark, 2012, referred to in Curtin, 2014:12) and reduce the possibility to conduct an investigation (Chesterman, 2011:78). Summing up, intelligence agencies’ surveillance activities affect democracy because secrecy limits the capacity for the citizens and their representatives to regulate and oppose those practices.
Exception and democracy
The second important challenge for democracy relates to the “national security exception” that allows states to suspend the rule of law and expand investigatory capacities when it is threatened by the emergency of a hazard. Indeed, national security is an opt-out provision that gives authorities the possibility to implement specific measures so as to manage the arisen problems. If the survival of a country is at stake, is the exception of the rule of law a necessary form of specific regime of regulation within democracy, or does the exception break the law?
The literature presents no consensual approach on this issue. On the one hand, Schmitt and Agamben address exceptionalism not as a special type of law but as being the absence of the rule of law that simultaneously limits and conditions it. Schmitt first argues that the exception “confirms not only the rule but also its existence, which derives only from the exception” (1985:15). Exceptionalism therefore implies a ‘normative vacuum’ that, in his perspective, is the absence of rule of law (Huysmans, 2004:329). Similarly, Agamben argues that “it is a suspension of the juridical order itself, it defines law’s threshold or limit concept” (2003:48). On the other hand, several scholars, such as Balzacq, tend to reject this interpretation. For them, security practices are “produced through ordinary law of the liberal state” so that what can be considered as ‘illiberal practices’ are actually within ordinary law (ibid.:8).
By framing the issue in terms of what the law includes or not, the second approach tends to neglect the regulatory function of the law. In the analysis of the relation between exceptionalism and democracy, “[t]he concept of exception refers to serious distortions in the restraining effects that the rule of law and democratic representation have on the arbitrary exercise of power.” (Huysmans, 2004:327). Put differently, exception limits the possibility to properly regulate the activities that fall under the scope of ‘national security’, including intelligence work. The more activities are justified under this exception, the more difficult it becomes to apply democratic procedures. This is particularly striking in the context of mass surveillance for two reasons. First, the collaboration of the private sector in the collection of data has led to the confusion about the real purposes for which such quantity of information was gathered. Thanks to the programme XKEYSCORE, the NSA has been able to collect, in the name of the fight against terrorism, data that was initially collected for commercial purposes (Bauman, et al., 2014:123). Should, then, the legislation that applies to commercial data be similar to the collection of information for security purposes? This is partly what
could go into the “normative vacuum” that Schmitt referred to. Second, the collaboration of intelligence services has precisely enabled them to bypass national legislation. This is particularly visible when one looks at the extraterritoriality of data collection: the American-led surveillance programmes, if based on American law (section 215 of the USA PATRIOT Act and section 702 of the FISA Act) as argued by the Director of National Intelligence (Moyer, 2013), have crossed national legal frameworks. Consequently, in accordance with the approach developed by Schmitt, exceptionalism tends to favor a “normative vacuum” that weakens democracy and the capacity for states to regulate.
III. 2. Desecuritization: a political strategy that legitimates security tools
Securitization theory has gained momentum over the last decade in Security Studies due to its broad explanatory mechanism and its innovative approach that defines security as a constructed and evolving notion. The analytical framework that this theory elaborates is of particular relevance in the analysis of the LIBE hearings because it takes into account the strategies and processes at play in the framing of an issue as being in normal or security politics. This thesis places at the core of its theoretical approach the notion of security as a process, and the dichotomy between security and politics, but it differentiates itself from the original securitization approach on two aspects. First, it shifts the focus from the process of locating an issue as a threat to the policy instruments that are used so as to tackle this threat. Second, it argues that the differentiation between normal and security politics is mostly a matter of choice between, respectively, democratic and exceptional measures; thus, securitization can be understood as a political process that enables the actors involved to call for democratic or security policy tools. By looking at the framing of an issue, the theoretical framework developed in this research project highlights that this political process is embodied within a specific representation of the social and institutional relations among the societies. In a European context, this particular representation involves the definition of the European Union’s competences, and in extenso, the construction, by the actors involved, of its identity.
III. 2. 1. Securitization and exceptionalism
The Copenhagen School’s approach places exceptionalism as a key element in the field of security. Nonetheless, its theoretical framework tends to be too narrow for two reasons. First, it exclusively focuses on ‘speech act’ at the detriment of security practices. Second, securitization should not be related to moving an issue outside of ‘normal politics’.
The approach developed by Balzacq highlights that tools and instruments are essential in defining what constitutes ‘security’. In accordance with his approach, securitization then also applies to the tools deployed in a context of exceptionalism.
The Copenhagen School
The concept of securitization emerged in reaction to the inability of former security studies, such as the realist and neo-realist approaches, to develop a theoretical framework adequate enough to encompass the emergence of new threats – coming from climate change, technological development and a post-Cold war international order – and the consequent broadening of ‘security’. Introduced in the field of International Relations by the Copenhagen School and its initiator Ole Waever in the mid-1990s, the major originality of this concept resides in its constructivist understanding of ‘security’ as a discursive process instead of an ex nihilo notion that would be ‘out there’ (Roe, 2004:281, Huysmans, 1998:571). Indeed, Waever defined security as a ‘speech act’ where, “by uttering ‘security, a state representative moves a particular development into a specific area, and thereby claims a special right to use whatever means are necessary to block it” (Waever, 1995:55). During this speech act, actors present an issue as a threat at the condition that the audience “must then respond with their acceptance of it as such” (Roe, 2004:281). Three elements peculiar to this theory can be isolated from this definition. First, not all speech can be successfully securitized; it has to respect three criteria:
(T)he rhetorical structure of a securitizing act needs to contain (…)(a) existential threats to the survival of some kind of referent object that (b) require exceptional measures to protect the threatened referent object, which (c) justify and legitimize the breaking free of normal democratic procedures (Buzan et al., 1998:204, quoted in Aradau: 2004:391)
Second, the securitization process involves speakers and an audience. Defining what constitutes ‘security’ is therefore an inter-subjective process between, on the one hand, actors who benefit from their social status and authority to legitimate expert knowledge, and on the other hand, an accepting or rejecting audience. This notion of legitimacy is essential because it assumes that some actors have the right and the capacity to frame an issue but also that others do not have it. As Aradau explains, “[t]he dynamics the CoS has in mind are those between political actors and their electoral audience, those who need to be convinced of the legitimacy of a security threat” (2004:395). Security actors are, in Ole Waever’s approach,
mainly elites (1995:57) that can produce expert knowledge by mobilizing techniques, notably statistics or scientific information (Aradau, 2004; Huysmans, 2004). Thus, the traditional approach to securitization looks particularly to content of the speech through linguistic and semantic analysis. If the substance of the speech produced is important, using securitization requires looking at the struggle at play about the validity of the information – this aspect being quite often neglected. Indeed, since this analysis focuses on intelligence services and the difficult access to information due to the omnipresence of secrecy, the veracity of the expertise provided, and access to information can be a core issue in these acceptance/rejection struggles.
This constructivist understanding of ‘security’ tends to highlight the intrinsic relation between power and knowledge as Foucault stresses it. A Foucauldian analysis of desecuritization brings forward the notion of power in the discourse and, most importantly, the relevance of the absence of speech. Indeed, he highlights the relations of power within into truth production, or what he calls “the discourse and techniques of right” (Foucault, 1976:96). The power that emerged from this production of truth is mostly a relation of domination between a “unitary” and prominent discourse, and a dominated, “fragmented” knowledge (ibid.:86 and 99). Foucault presses for a greater analysis of what is not the main discourse, and especially of what is not said, because the two cannot be separated. Indeed, “silence itself […] is less the absolute limit of discourse, the other side from which it is separated by a strict boundary, than the element that functions alongside the things said, with them and in relation to them within over-all strategies” (1978:27). Consequently, the absence of speech from specific actors has to be considered in order to analyze how those who speak can frame the issue.
Last, by representing securitization as a process, the Copenhagen School has based its theory on a duality opposing ‘normal politics’ to ‘security politics’. This dichotomy is grounded on a form of prioritization whereby the ‘securitized’ issue has to be addressed first (Roe, 2004: 281) because “ if we do not tackle this problem, everything else will be irrelevant (because we will not be here or be free to deal with it in our own way)” (Buzan, Waever & de Wilde, 1998:24 quoted in Roe, 2004:281). As such, the concept of ‘security politics’ becomes synonym to exceptionalism that justifies abnormal measures and the suspension of “normal political rules” in such context of emergency. Securitization theory can therefore be used in order to highlight the balance between democracy and security as argued previously.
If securitization “was immediately regarded as a path-breaking alternative to realist and neo-realist understandings of security” (Aradau, 2004:389), it has nonetheless been subject to several critiques that offer, for the purpose of this thesis, an interesting understanding of this concept. Several scholars, among whom Didier Bigo and Thierry Balzacq have been quite prominent, called for a broader securitization theory.
The first critique relates to the question: how to ‘do’ securitization? If the Copenhagen School focuses mainly on ‘speech act’, “non-discursive practices are not substitutes for discursive practices, but both are equally important for the analysis. They have different logics, but can produce the same effects” (Balzacq et al., 2010:3). Put differently, security is not exclusively defined by speech act, performed in a specific moment, but it also depends on how security is practiced, be it by agents, institutions or decisions. Under this quite vague notion, these authors consider that the daily work of customs, the asylum seekers’ permit process, the recourse to databases or the adoption of regulatory instruments are all practices that capture a particular understanding of security, and that simultaneously perform it. How, then, can securitization theory apply to a Parliamentarian Inquiry held for more than five months with about eighty actors who used discourse? The LIBE hearings are understood here as a security practice where the security experts engage with, and perform security. Thus, as Salter argues, securitizing move cannot be limited to a binary acceptance/rejection decision between an expert and an audience. Indeed, several experts and officials might address the issue differently; consequently, “there [is] not a single securitizing move that was accepted or rejected, but a much more complex play of competing authorities, power metrics, and discourses” (Salter, 2008:322). Because security might play differently to each actor and audience, “we need to expand our analysis of how securitizing moves are accepted or rejected” beyond a linguistic analysis and to look at “who may speak, what may be spoken, and what is heard”(Salter, 2008:323).
The second critique relates to the question: what is securitized? Among the different security practices put forward by Balzacq and Bigo, this thesis pays particular attention to the analysis of policy instruments and tools as “instruments of securitization” (Balzacq, 2008:79), and thereby takes a distance from a traditional understanding of securitization that exclusively looks at how an issue is framed as a threat. A traditional approach of securitization would, for instance, focus on how AIDS, as an issue, is framed as a potential danger (see Elbe, 2004) whereas Balzacq looks at the tools and instruments deployed in order to address problems defined as threat. Indeed, he explains that “instruments of securitization post-date a successful securitization. In other words, they do not construct a threat per se; they are built to curb an
already accepted threatening entity” (2010:76). In his terminology, an instrument is “an identifiable social and technical ‘dispositif’ or device embodying a specific threat image through which public action is configured in order to address a security issue” (ibid. :79). This interpretation of securitization is essential in this research inasmuch it gives an analytical framework for the study of mass surveillance, which is ultimately an instrument used in order to fight a particular issue already securitized, namely terrorism.
It is noteworthy to mention that security instruments are not necessarily chosen because of their efficiency in dealing with potential dangers (Balzacq, 2008); on the contrary, they embody particular political preferences and reflect “something of the threat that the public action is meant to respond to” (Balzacq et al., 2010:8). Thus, this approach to securitization enlarges the study of what constitutes ‘security’ by including the “systems of meaning” generated by the actors and the “productive power of their practices” (c.a.s.e collective, 2006:458).
III. 2. 2. Security in a political perspective: desecuritization as a choice
The concept of securitization has been largely applied in the literature, be it in reference to migration, information exchanges, human trafficking or AIDS (see Ibrahim, 2005; Balzacq, 2008). Desecuritization has nonetheless been studied to a lesser extent (see Huysmans, 1998; Roe, 2004; Aradau, 2004). If the former brings an issue from normal to security, the latter would logically work the other way around by unmaking a particular issue as an existential threat. However, by enlarging the focus from the issue constituted as a threat to the security tools deployed for addressing it, desecuritization is fundamentally a political choice about the way to govern, and this choice embodies a particular representation of the political community.
Desecuritization: the relocation from security to another ‘sector’ of regulation
What insight does the analysis of security instruments bring to bear on desecuritization? Since ‘security politics’ and ‘normal politics’ do not imply the same policy instruments, desecuritizing is not simply about replacing a threat by another, or framing an issue as less threatening, but it is the rejection of exceptional measures deployed for security purposes, such as faster ruling procedures and secrecy, and the promotion of a ‘normal’
regulation of an issue. Remains to be defined what a normal way of regulating would be. Within the literature on desecuritization, Huysmans argues that a desecuritizing move can relocate a security issue or instrument in any other domain, such as economics, where the issue will be addressed and regulated in terms of supply and labor (Huysmans, 1998:576). This approach tends to characterize ‘normal politics’ as encompassing everything but ‘security’. More promising for this study is the approach taken by Claudia Aradau who argues that ‘normal politics’ is fundamentally about democratic practices and ruling. To her, “[i]t is in relation to the procedural ‘normalcy’ of democracy that the ‘exceptionalism’ of securitization can be theorized” (Aradau, 2004:392, emphasis added). Put differently, a ‘normal’ way to govern favors slow democratic procedures such as deliberation, debate, accountability and transparency, whereas security involves “a logic of urgency and exceptionalism” (ibid.). Similarly to Müller-Will, she refers to democracy in terms of “policies and actions as a product of popular power” and “basic procedures which are transparent, open to public scrutiny”. Consequently, her approach underlines, and validates the tension between exception and democracy as defined earlier. Thus, a desecuritizing move rejects exceptional measures, and relocates them in the ‘normalcy of democratic procedures’ where the rule of law applies.
(De)Securitization as a political choice
Because security and normal politics do not invoke the same political tools, the choice between securitization and desecuritization is ultimately a political choice that highlights “the type of politics we want” (ibid, p.390). Indeed, “instead of the question ‘more or less security?’, the question becomes ‘to securitize or not to securitize?’” (Huysmans,1998, 572). From this perspective, desecuritizing a threat or a policy instrument can be seen as a strategy in order to prevent the recourse to exceptional measures, and to reintroduce democratic practices. This assumption then requires an analysis of the motives and strategies that might come into play among the actors involved in the process. As an illustration, Huysmans argues that the decision to desecuritize migration into the ‘sector’ of human rights might be justified on ethico-political grounds (1998:572) or for instrumental reasons, namely that addressing migration through the lens of human rights is more efficient that through exceptional measures. The same logic applies for the recourse to secrecy. As Chesterman argues, “[s]ecrecy is assumed to be central to the work of intelligence, but efforts to withhold information from public scrutiny are frequently due to political or cultural reasons rather than genuine operational concerns” (Chesterman, 2011:68).
Interestingly, looking at the conditionality and motives at stake prior to (de)securitization involves more than the simple relocation of an issue from security to another sector. Indeed, the recourse to specific policy tools mobilizes different actors, different institutions, thus ultimately embodying specific representations about how a government addresses the issue (Aradau, 2004:396). In the continuity with the Huysmans’s example, desecuritizing migration is not just about how to solve the potential threat that it might represent; instead, “it addresses the normative question of how political government should be organized” in the way it tackles migration (Huysmans, 1998:574). Consequently, security practices “organize the way people relate to themselves and to others and how they institutionalize and rationalize a particular government of these relations” (1998:573). In a European context, the desecuritization of migration touches upon a broader understanding of the organization of the European community, its integration and in extenso, its identity. Subsequently, because security moves embody a particular conceptualization of the EU, it is possible to conclude that “narratives of security are crucial in fostering and (re)structuring ‘the European polity and its role and identity as an international actor’”(de Goede, 2011:225 referring to Rogers, 2009:833-4).
III. 2. 3. The identity of the European Union at a global level
As previously argued, desecuritization is understood in this thesis as a political choice between addressing an issue by democratic means or exceptional measures, and this choice projects a specific identity of the political community. By using desecuritization as a theoretical framework to analyze the LIBE hearings, the logical next step is to look at the identity of the ‘political community’ at stake in those debates, namely the European Union. This section looks at how the EU has positioned itself, namely what political choices have been made, with regard to the recourse to data collection for security purposes. As an area of freedom, justice and security, the EU identity is far from being homogenous; there is a paradox between the representation of the Union as a ‘European data protection model’ – where the European Parliament has been particularly active at promoting this identity, notably during the SWIFT and PNR cases - and the construction of the EU as a security actor through the adoption of security measures and intelligence work that undermine democracy and civil rights.
A European Data Protection Model
Whereas data protection has been one of the “thorns in the side” in the transatlantic partnership, crystallized around the SWIFT and PNR ‘affairs’ (see Bellanova & de Hert, 2009; Türk & Piazza, 2009; de Goede, 2012), the EU has put at the core of its fundamental values the notion of privacy and data protection and emphasized their primacy over security.
Those scandals have mostly put under the spotlight the different legal regimes of data protection developed respectively by the United States and the European Union (see Blok & Bignami, 2007; De Hert and De Schutter, 2008; Chevallier-Govers, 2013; Schwartz, 2013). In comparison to the Americans, who collected data on European soil in violation of European law, the European Union emerged as offering a broader and more protective legal framework for personal data and privacy, and thus as defending those rights over security. More precisely, from a transatlantic perspective, the European Union recognizes those two rights as being fundamental, respectively in article 8 of the European Convention on Human Rights of 1950 and article 7 of Charter of Fundamental Rights of 2000, whereas the American Constitution does not explicitly mention them. On the global stage, the European Union allows the transfer of personal data to third countries at the condition that these countries provide an “adequate level of protection” in accordance with European standards (see article 25 of the Directive 95/46).
The European Parliament has had an active role in promoting a ‘European data protection model”, especially during the negotiations with the United States about the adoption of a ‘SWIFT agreement’, which has provided a legal framework for the transfer of financial data. In effect, the European representatives rejected in 2006 the first accord proposed by the Commission on the grounds that it did not sufficiently protect personal information. A new proposal was adopted in 2010 with new provisions on data retention and oversight (de Goede, 2011:221-222). Thus, the European Parliament has solidified the representation of the EU as being a “model” and having a normative appeal. The emphasis on data protection and privacy as core European principles seems to support the construction of the EU’s identity as an actor that can project its norms worldwide. However, the narrative that represents the EU as a ‘European data protection model’ is challenged by the nascent security role of the EU.
The EU as a security actor
By looking at the ‘concrete policies’ implemented by the Union over the last decade, the rising role of the EU as an international security actor, notably via data collection and intelligence practices, appears as inconsistent with the EU’s global positioning as a defender of civil rights and privacy.
One needs to recall that security is a shared competence between the Member States and the European level, thus the capacity for the Union to be a security actor is limited, first, by the willingness of Member States to pool their sovereignty at the European level, but also by its ability to claim competence over security matters. Indeed, the Lisbon Treaty has consecrated the area of freedom, security and justice whereby the Union can regulate on issues related to the internal security, but has no say on national security matters (Article 4). Because of the transnationalization of threats, the distinction between national and internal security – as well as internal and external security (see Bigo, 2006) – has become blurry. Consequently, the way national and European leaders frame an issue, either as a matter of national security or of internal security, directly impacts on who is competent over the issue. Subsequently, within the European continent, the field of security embodies a contradiction between state sovereignty and European integration (Guittet, et al., 2011:12).
The development of intelligence activities at the European level highlights this contradiction, and shows how the EU solidified its image as a security actor. It is at the national level that intelligence is originally collected. The European Union has been able to create its own EU Intelligence Analysis Centre, IntCen, which provides risks assessments supporting EU institutions and the Member States. Nonetheless, its competences are limited since, as Catherine Ashton says, “contributions depend on the availability of intelligence in the Member States’ services and the willingness to share them” (quoted in van Buuren, 2009:17, emphasis added). Notwithstanding, Deirdre Curtin argues that the EU is a security actor since it “gathers and processes information autonomously. It also shares information both internally and externally” (2014:6). The identity of the European Union as a security actor seems, in this regard, dependent on its capacity to gather and analyze information, but mostly on the competence that it has over the Member States.
The nascent security role of the EU is more striking with regard to the security policies and instruments adopted in the fields of border control and counter-terrorism. The image of a Fortress Europe is quite significant in this regard whereby the Union has reinforced its border control and migration policy through the development of databases. The use of the Schengen
Information System (SIS I and II), Eurodac and the Visa Information System (VIS) are quite symptomatic of this greater collection and processing of data within the EU for security purposes (Lyon, 2007:162; Broeders, 2007:71). Likewise, the Union adopted in 2011 a Counter-Terrorism Strategy based on prevention, protection, pursuit and response (Europa.eu) where the preemptive logic gives priority to data collection, data exchanges and its processing (Guittet, et al., 2011:24).
However, the reinforcement of security tools in the EU automatically brings the issue of their compatibility with democracy and civil rights, and underlines the paradox between the two narratives of EU identity presented here. Indeed, with regard to intelligence collection and analysis, the European Union is also criticized for its lack of transparency and accountability. It is worth mentioning that the European Intelligence Analysis Center IntCen produces risk assessments that “are directly applied in the securitization of the terrorist threat at the political European level” (van Buuren, 2009:12); it however has no legal basis (Jones, 2013:2), which prevents the European Parliament from defining its mandate and making the Center accountable. Also, secrecy surrounds most of its activities (ibid.; van Buuren, 2009:11), thus ultimately reducing the access to information and the possibility for the actors that are not security experts to publicly discuss and resist those practices. Likewise, security experts have underlined the negative impacts of EU data collection and processing on fundamental rights, which foster social sorting (Bellanova, 2010; Lyon, 2007), enhance the risk of profiling and discrimination (Amoore, 2006) and undermine privacy (Guittet, et al., 2011).
Thus, the identity of the EU is far from being homogenous. Instead, it articulates different narratives – that look at the EU either as a defender of fundamental rights or a security power - which are mobilized and activated by various actors in order to justify appropriate tools in order to manage arisen problems. The LIBE hearings are a key “moment of openness” (Žižek, 2008 quoted in Wesseling, 2013:102) to analyze how experts and policy makers formulate a European response to the Snowden revelations, thus projecting a particular representation of the EU identity.
Concluding remarks
As a first conclusion, it is important to underline the particularity of the current massive collection of data in terms of scope, reach and actors. The deployment of a public-private assemblage, the unprecedented surveillance capacities made possible by technological
