Security-defence nexus in flux: (De)securitisation of technology in the Netherlands

ABSTRACT

The contemporary defence sector is dominated by dual-use technologies which have both military and civilian applications. The trade of these technologies is governed by export controls. Designation of technologies as “dual-use” and export control regimes can be considered as securitisation of such technologies. Securitisation refers to the discursive construction of existential threats. This article seeks to understand which actors desire securitisation or its opposite, desecuritisation, of technology. The contribution of this research is twofold. Firstly, securitisation of technology has implications for understanding defence and security in contemporary Europe. Secondly, identifying the actors involved in (de)securitisation allows for the analysis of their different roles in determining security discourses around technologies. The article builds on the literature on securitisation theory. Then it explores the EU-level export control regimes and afterwards focuses on a single case study of the Netherlands both as representative of smaller European states with an SME-heavy defence sector and as an example of an expanded export control regime by a member state. To conduct the analysis, I use the data obtained through a workshop with representatives of Dutch defence companies as well as a survey that has been answered by security and defence experts.

KEYWORDS:

• Dual-use
• technology
• defence
• securitisation
• desecuritisation
• export control

1. Introduction

States’ right to self-defence against an armed attack has been guaranteed under Article 51 of the United Nations Charter. As such, defence can be understood as the amalgamation of measures to respond to acts of armed aggression like invasion or so forth by one state to another. While this remains at the centre, the meaning has expanded significantly because despite the importance of the traditional sense of defending territory and citizens, new security threats require an approach beyond the traditional military sphere (Aben 2014; Ziliotto 2008). For example, the Dutch Ministry of Defence (MoD) introduces itself with the following words:

We are the Ministry of Defence. We are indispensable in a world of turmoil. We protect all that we as a nation cherish. We fight for a world of freedom and security, because we believe that everyone has the right to live in such a world. This is our mission, and it’s also the reason we participate in missions.

(Ministerie van Defensie 2022)

At the first instance it refers to a globalised understanding of the mission of the ministry. They “fight” for “freedom” and “security,” which are the core values of the Dutch “nation.” Even this introduction shows us that the traditional task of the MoD and the Armed Forces to “defend” the nation against external threats is significantly expanded. Since defence and armed forces cannot be understood without their tools (i.e. weapons systems, vehicles, surveillance technology, protective gear, military personnel, etc.), the implication is crucial: The defence industry is an intrinsic part which cannot be ignored when analysing defence-related topics. Therefore, this article focuses on the role of the defence industry in determining security discourses around technologies. The analysis intends to critically examine the nature of security discourses in order to contribute to a demystification of attempts to downplay security concerns in the face of economic interests (corporate interests vs geopolitical ones), attempts to misuse security concerns for competitive purposes (US firms vs China and Russia) and/or demystify routinized practices of security institutions. The objective is to contribute to the broader literatures on securitisation as well as the defence industrial policy. The particular contribution to securitisation literature is not so much about the theory as such, but about familiarising the readers with this approach, and about applying the approach to the study of the workings of (security-)defence industries.

The contemporary defence sector comprises both civilian and military companies that research, develop and/or improve technologies. Integrated security technologies that combine several layers of surveillance, sensor data, and other relevant information to establish an effective ecosystem are at the heart of this sector. In most cases dual-use technologies, which have both military and civilian applications, come to prominence within and beyond this ecosystem. The trade of dual-use technologies is governed by export controls. Designation of technologies as “dual-use” and subsequent export control regimes to regulate the trade of these can be considered as a level of securitisation. In its simplest form, securitisation refers to a process in which issues can be securitised and presented as existential threats (to referent objects) through securitising moves (i.e. speech acts) and be removed from public deliberation (Buzan, Wæver, and de Wilde 1998). This creates a problem because it hinders any democratic discussion and marginalises dissidence. While defence has been traditionally about responding to (potential) existential threats, the securitisation framework puts forward that security-speak enables any issue to be associated with security and as such with perceived existential threats. The theory emphasises desecuritisation as the preferred option, as it allows issues to be brought back to the realm of “normal” politics and therefore they become a part of the public debate (as far as allowed by the respective regime).

This article employs securitisation theory to understand how certain technologies become “security threats”. For this, I ask the following question: Which actors desire securitisation of technology and which actors desire desecuritisation? I centre the exploration around two types of actors that I identified as important. These are the MoDs in the European Union (EU) and the companies operating in the defence sector. Then, from a securitisation point of view, I hypothesised two sets of behaviour: 1) MoDs as the main actor to devise policy and measures for “national security” would desire to significantly limit or ban export of dual-use technologies as a protective measure and are therefore pro securitisation. 2) Companies operating in the defence sector would desire lifting or at least easing of export controls to have a free access to the market and are therefore pro desecuritisation/non-politicisation. However, the findings proved to be more intricate.

Since the export of dual-use technologies is firstly regulated at the EU level and then at the member state level, the article explores the processes of how the technologies are securitised and where the agency lies in both levels. At the EU level, the study revealed a diverse set of actors, including the Commission, the Council, industry organizations, and other dialogue-promoting bodies like the Export Control Forum (ECF) and Dual-Use Coordination Group. This necessitates further exploration into the relative powers and bargaining positions of these actors in EU decision-making. In the Netherlands, the prominent actors were the Ministry of Foreign Affairs (MFA), Ministry of the Interior (MoI), and MoD, with the role of industry and companies remaining ambiguous. Companies did not actively seek desecuritisation but also did not explicitly support securitisation; instead, they desired a uniform system across the EU. As such, a clear-cut answer to which actors seek securitisation or desecuritisation could not be extracted. Instead, the complexity of the dual-use policies has been revealed and a potential to revisit the securitisation theory through an actor focused perspective has emerged.

The underlying motive or “why” these actors behave the way they do is also an important aspect that needs to be investigated. In this respect, based on the conducted analysis, the article mainly points to the institutional dynamics as the key element which determines the actors’ choices. However, these need to be further explored in future research with dedicated data to uncover the specifics. These findings demonstrate the research’s alignment with core securitisation theory themes, emphasizing the political nature of export control regimes. However, it also highlights the significance of understanding the rationales driving actors’ behaviour. The analysis indicates that institutional dynamics determine the motives of these actors. Corporations operate within a capitalist/economic paradigm, seeking high sales, profit, growth, and competing with less restricted firms. Ministries are influenced by political-economic considerations, particularly a resurgence of traditional geopolitics towards countries beyond the EU and NATO. Consequently, the “why” behind actors’ actions is tied to their institutional makeup/mandate rather than individual characteristics or media debates.

As a result, this article provides a valuable starting point for a research program on the security-defense nexus, actors involved, and their underlying rationales. Identification of these actors forms the basis as securitisation of technology has implications for understanding defence and security in contemporary Europe. On the conceptual level these include the blurred definition of defence and security that creates a grey area of politics. On a more concrete level, research and development of technologies and the funding of these processes are affected by how a technology is categorised. In an era of increasing debates on European strategic autonomy these implications are worth exploring. Further investigation into the EU level is essential to unravel power relationships among different actors. Additionally, the Dutch case can be enriched by considering parliamentary proceedings, lobbying, and other relevant processes. Exploring global governance dimensions, identifying relevant actors and their motivations will be crucial for effective international regime formation. Finally, including other case studies based on defense sector structures, country comparisons, and comparisons between state-owned and private sector firms would strengthen the research agenda.

The findings are based on a case study of the Netherlands. The Dutch defence industry is characterised by small and medium-sized enterprises (SMEs) and is representative of small actors in Europe. The analysis is based on a workshop organised in October 2019 with defence company representatives and academics, a series of interviews with representatives from Dutch defence companies in one-to-one settings in May, June and October 2020 and an online survey filled by security experts (academics, researchers, and practitioners) in March 2022.

The article is divided into six sections. The second section focuses on the literature on dual-use technologies and security-defence to establish the theoretical and empirical gaps that this article aims at filling. The third section explains securitisation theory and its evolution to provide the theoretical framework of the article. The fourth section outlines the methods used. The fifth section functions as the analytical section where brief summaries of the EU and Dutch export control mechanisms as well as the composition of the Dutch defence industry is provided and then the empirical data is presented and discussed. Finally, the article concludes with a brief summary of findings and reflection on future research potentials opened up by this article.

2. Dual-use and security-defence

2.1. Definition(s)

It has been a difficult endeavour to define what dual-use means. Lupovici (2021) argues that social processes shape whether and how technologies can be seen as dual-use by different actors. If the simple definition that dual-use technologies are those which can be used for both military and civilian purposes is held, then a logical argument could be that a pen is also a dual-use technology. A similar thing can be said about screws and bolts. They can be both used in (dis)assembling military materiel or a cupboard. This brings the question whether there is a more technical way to address the dual-use question.

Forge (2010) distinguishes between categories of research, technologies, and artefacts referring to the different stages of research, development of technologies as a result of the research, and the application of the technology as the final product (artefact). Following from this, his definition posits that “[a]n item (knowledge, technology, artefact) is dual use if there is a (sufficiently high) risk that it can be used to design or produce a weapon, or if there is a (sufficiently great) threat that it can be used in an improvised weapon, where in neither case is weapons development the intended or primary purpose.” (117). Obviously, this definition cannot be considered as a purely technical classification as sufficiency parameters for risk and threat are almost always politically determined. From a philosophical point of view, Pustovit and Williams (2010) point to a difference between a pragmatic American approach to dual use as technologies that could threaten the society in some way and the European approach that does not see the nature of technologies as a priori determined.

While the discussions on how to define what dual-use means entail political as well as philosophical approaches, states act upon regulations that already exist or created/modified along the way which may or may not be based on these prior debates. For example, the European Union defines it as the following:

“dual-use items” means items, including software and technology, which can be used for both civil and military purposes, and includes items which can be used for the design, development, production or use of nuclear, chemical or biological weapons or their means of delivery, including all items which can be used for both non-explosive uses and assisting in any way in the manufacture of nuclear weapons or other nuclear explosive devices

(Article 2(1), EU, Regulation 2021/821b

Annex I of the same regulation includes a non-exhaustive list of dual-use items under ten categories.¹ EU member states can impose additional control for non-listed items through the catch-all clause of the Regulation. So, there is still an open way for securitisation at the national level as it will be elaborated on in the next sections.

Since the official dual-use definition provided in the EC Regulation is accepted by and valid for all the member states, it is also useful to further explain an important component. Per the Regulation, dual-use items include “technology” and this requires a definition as well. Annex I defines “technology” as the “specific information necessary for the ‘development,’ ‘production’ or ‘use’ of goods. This information takes the form of ‘technical data’ or ‘technical assistance’” where “technical data” refers to (but not limited to) “blueprints, plans, diagrams, models, formulae, tables, engineering designs and specifications, manuals and instructions written or recorded on other media or devices such as disk, tape, read-only memories,” while “technical assistance” refers to the means that the “technical data” is transferred like “instructions, skills, training, working knowledge and consulting services” (EU, Regulation 2021/821a).

These definitions are important to understand as they allow deeper scrutiny into the EU export control mechanism and the national-level securitisation processes. I will be referring to these to contextualise the problematic nature and the consequences of the said processes.

2.2. Current debates

Despite not being as prominent in the mainstream of International Relations or security studies, literature on dual-use technologies has been gradually expanding. The role of technology transfers including spin-ins, spin-offs, and dual-use in defence innovation in Europe has been previously discussed (i.e. Fiott 2017; Sezal and Giumelli 2022). The main branches that deal with the issue can be classified as knowledge/science governance and, as an extension of this, arms control. Therefore the focus is particularly on nuclear, chemical, and biological technologies that can be classified as dual-use. An emerging corpus also deals with artificial intelligence, nanotechnology, cybersecurity, and new surveillance technologies like facial recognition etc.

Rychnovská (2016) examines life sciences research and how it is regulated in the United States based on the logic of “dual-use dilemma.” This dilemma indicates a situation where easily accessible knowledge and technologies can be used for development of weapons. In her case this refers to the potential development of biological weapons and acts of bioterrorism. Therefore, the dual-use dilemma is used as an organising principle where an equilibrium is sought between, on one hand, responsible research and innovation and, on the other, secrecy and (national) security (Rychnovská 2016). Ludovici (2021) sees this dilemma as about deciding to securitise or not to securitise because designating a technology as dual-use “allows an actor to construct it as a(n) (existential) threat, even if the technology is developed for civil purposes” (258).

3. Securitisation and desecuritisation

Securitisation theory was first formulated by Ole Wæver in 1995 and later on developed into a full theoretical framework with the so-called Copenhagen School’s perennial work Security: A New Framework for Analysis in 1998 by Barry Buzan, Ole Wæver and Jaap de Wilde. While the original premise is rather straightforward, the theory has evolved significantly since its first inception.² In this section I will first introduce the basics of the Copenhagen School (CS), which according to Balzacq (2011) constitutes the philosophical approach to securitisation. Then I will explore the additions to the theory, including the sociological approaches to the theory (Balzacq 2011). Finally, I discuss the comparatively less studied dimension of the theory, namely desecuritisation.

3.1. Copenhagen School

In mainstream/traditional studies, security has always been about survival (Buzan, Wæver, and de Wilde 1998). When survival is put into the equation threats to survival become existential threats by definition. Therefore, for the traditional state-centric approaches security is about a state’s survival, its exposure to existential threats, or rather being invulnerable or resilient to these existential threats. When the study is expanded from the traditional military-political understanding of security, it includes other areas where the object who is threatened (“referent object”) can be different than the traditional one which is the state.

The term “securitisation” refers to a location shift of issues from political realm to security realm, the former meaning the conventional arena of politics; the latter meaning the extraordinary realm where the issues are no longer subject to political discussion and there is an existential threat to a referent object. Therefore, “security” actually is a practice where the meaning rests upon the way the concept is used. It is argued that a successful “securitisation” has three steps, namely the identification of existential threats, the call for emergency action, and the effects emanating from removing the issues from the regular political sphere. This seems somewhat self-explanatory, though the question remains “how does securitisation occur?”

The most interesting aspect, as well as the most important contribution of securitisation theory to security studies, is its description of how the issues are securitised. The CS introduces the concept of “speech act,” which it borrows from the linguistic theory and applies to international relations.³ This is also the reason Balzacq (2011) classifies this approach as “philosophical.” A “speech act” within securitisation is the discursive method of representing issues as existential threats to a referent object by the actor. The referent object is the main object whose existence is threatened. In the military sector, for example, it can be the state or the nation (i.e. in case of an invasion), or national interests or national identity. The actor is the person or the entity (i.e. institution, collective, etc.) which performs the “speech act” in order to securitise the issue, and is called the “securitising actor.” However, the securitising actor cannot be anybody but it needs to be a(n) person(s)/entity(s) who has authority so that the “securitising move” (the speech act) can be heard and accepted by the relevant audience (Buzan, Wæver, and de Wilde 1998, 33, 40).

In Buzan, Wæver, and de Wilde (1998) “speech act” was conceptualised as not being securitisation itself, but rather it was the “securitising move” which is done for the purpose of securitisation. In this respect, to have a successful securitisation, the “securitising move” should create a resonance in the audience. So in other words, securitisation is not successful unless a certain degree of appreciation or consent or acceptance occurs in an audience by the move. Securitisation legitimises resort to unconventional and/or extraordinary measures to deal with the issue since what is at stake is the pure existence of the object and anything needs to be done to protect it. Furthermore, the move from the political to the security sphere implies that it can no longer be challenged and thus every action is theoretically possible. Yet again, for the CS, security is not a desirable thing; on the contrary, issues should be put back to the public/political sphere, and this process is conceptualised as desecuritisation. The reason for desecuritisation preference is that when something is securitised it legitimises the use of emergency measures and dismisses political debate which potentially undermines liberal democracy. As such, securitisation can be used as a means to achieve interests other than those of the people, therefore “security should be seen as negative, as a failure to deal with issues as normal politics” (Buzan, Wæver, and de Wilde 1998, 29).

In this article I argue that designating technologies as “dual-use” is already a form of securitisation (that involves a speech act and then acceptance of this act) and the export control regimes can be considered as extraordinary measures. I will elaborate on this in the analysis section through our empirical study.

3.2. Second generation and beyond

Despite being a product of the CS, the securitisation theory has gained a momentum of its own outside the immediate CS. Scholars such as Thierry Balzacq, Holger Stritzel, and to certain extents Mark Salter and Paul Roe can be counted among the “second generation” of securitisation theorists, the first generation being the CS.

Balzacq (2005, 2011, 2015) criticises the CS framework for being fixed and permanent while arguing that emphasis must be given to the nature of the audience, the context securitisation takes place in, and the agency of the actors involved in the process. Additionally, he proposes a methodological pluralism in the application of the securitisation theory as he claims that sole dependence on discourse analysis causes confirmation bias and therefore calls for incorporating process-tracing to overcome this bias. Similar to Balzacq, Stritzel (2007) also argues that the CS framework is limited, and he proposes that it needs to be expanded with a layered approach that explores the performative forces of texts (as theorised by Judith Butler), locates the process within the already existing discourses and the power relationships between the securitising actors. As a development of this idea, he has revisited the securitisation theory with an emphasis on a neo- and post-Marxist readings of discourse theory whereby securitisation becomes an amalgamation of performative processes instead of a single speech act by an authoritative securitising actor (Stritzel 2014). These developments may seem natural, yet they indeed improve the theory significantly and therefore allow for more comprehensive analyses that this article also attempts to.

As mentioned “audience” and its role in securitisation have been discussed and re-articulated widely in the critiques or revisitations of CS’s understanding of securitisation. For example, Stritzel (2007) and Salter (2008) questions how audience is defined and understood, while Roe (2008) suggests that both the political elite and the general public can be the audience. More recently, Floyd (2016) in her “revision” of securitisation theory has defined the audience as the “addressee of the speech act” (p. 688, emphasis in the original). Floyd’s (2016) contribution is also significant as it conceptualises the assessment of success/failure of securitising moves. Instead of emphasising “emergency measures,” she suggests that securitisation is successful when a threat that requires response is identified, then the relevant agent’s behaviour changes and this change is justified with reference to the threat identified.

Actors, specifically the “securitising actors” are also central to the securitisation theory. In essence, this refers to the actors that do the speech act or the securitising move. Who these actors are is not always as clear because depending on the issue/threat and the referent object they could be a different entity/person. Nevertheless, most of the time it is thought to be the governments and/or the parliaments are the key securitising actors, especially when the concern is the security of the state/nation (or similarly articulated political collective). In Buzan, Wæver, and de Wilde (1998) political leaders, bureaucracies, lobbyists and pressure groups are also listed as such; however, empirical analyses seem to be limited in showing their actual influence as securitising actors. CS is aware of the fact that it is very difficult to identify the actors; however, in the case of national security “government is the state” (40–41). This may well be the general case, but as Hansen and Nissenbaum (2009) point out with reference to cybersecurity, sometimes there can be “a powerful blending of private-economic and public-national security concerns” (1161) which shows a complexity of relations of relevant actors which is also in line with Stritzel (2014). In this sense, we can think of securitisation in the Balzacqian sense that it is “a set of interrelated practices” (Balzacq 2011, 333) and as an extension of this definition we can argue that there are multiple actors that in fact negotiate what needs to be securitised or not securitised or perhaps de-securitised.

Furthermore, works within the framework of international political sociology and insecuritisation approaches provide a more Bourdieusian element to securitisation. Insecuritisation refers to a Foucauldian “technique of government” where security technology and expert knowledge form the essence of modern society and social relations, and the “threat definitions” are being embedded in this modern social relations (Van Munster 2007). The essence of securitisation is that it already highlights the insecurity of a referent object in the face of keeping or achieving specific values. But here security is conceived in ontological terms (there is even an explicit literature about ontological security), whereas the CS treats security in its epistemological quality: a securitising move such as a speech act emphasising threat perceptions. Nevertheless, the process and development of this technique are rather complex. Even though it emerged during the Cold War, following the collapse of the USSR, the disappearance of the common enemy and the so-called “triumph of liberalism,” agencies of that period and the professionals have not disappeared. Since then, while “external” security agencies such as the army have started to focus on the inside of the borders to find enemies, “internal” security agencies like the police forces have been looking for their “enemies” outside the borders (Bigo 2000). In effect, two previously distinct spheres have converged or even merged. Although it is widely claimed that the 11 September attacks caused this change, it is argued that the reasons behind the convergence are the structural evolutions of various distinct agencies working in the area of security, and their relations with political professionals were present long before (Bigo 2006). Similarly when we talk about defence, even in its most limited version possible, we refer to this merger of internal and external. As such, defence has become something bigger and more comprehensive. Therefore, it is not an exaggeration to refer to “security-defence” as a conjoined notion. Despite the fact that the literature refers only to security, it indeed implies the inclusion of defence since the meaning of defence has already been expanded as can be seen from the previously quoted introduction of the Dutch MoD.

Security professionals and agencies (whether internal or external) fight (within the agencies and without) for their survival as well as dominance for practices of security. This leads to insecuritisation enabling legitimation of practices and interconnectedness between internal and external. Put differently, agencies and professionals in the security field maintain their relevance and ensure their survival through management and creation of insecurity. While it is implied and assumed that the field may include companies as private actors, the literature does not engage with this aspect. In the realm of defence and security, however, the role of the defence industry needs to be explored. One way to understand securitisation of technology lies in exploring the way industry as a whole and companies as individuals behave (i.e. seeking securitisation or desecuritisation).

3.3. Securitising technologies

On the conceptual level defence refers to a very specific action which is responding to armed aggression towards one’s own state. However, security may refer to a wider set of actions, policies, discourses etc. even when a limited definition is considered. Therefore, defence can be thought of as a subset of security because it responds to a direct action. Defence, then, refers to an emergency condition that goes beyond securitisation which in principle is not discursively constructed but factually present. Such situations, as emergencies, cannot wait for public deliberation and therefore require immediate response. However, as explained in the introduction section of this article, MoDs and armed forces do not fit into this conceptualisation anymore. Their function is much more than defending the homeland, and so the concept of “security” becomes particularly relevant. In contemporary Europe, almost all MoDs speak about “defence and security,” although they are not ministries of security per se.⁴ Moreover, defence industries or more concretely, companies operating therein also play an important role in the defence ecosystems and the role of the MoDs (and armed forces) in contemporary Europe. In this sense, dual-use technologies present an interesting case as the concept itself already shows a certain stance.

A further investigation is required to understand what exactly constitutes securitisation. One assumption is that designation of dual-use technology is indeed securitisation because after that moment said technology is considered applicable to non-civilian means. Since considering that technologies themselves cannot be thought of as defence or civilian without a social context, designating something as dual-use implies a level of securitisation has already happened. However, this can also be understood as desecuritisation because a technology that is assumed to have only military-defence purposes can be classified as dual-use and therefore becomes applicable to civilian uses. In such a case, the emphasis lies both in the technology itself and the R&D process. Emerging technologies, particularly the so-called “disruptive technologies” are usually considered as dual-use from their inception. On one level, this is due to the fact that new technologies based on digital and cyber are mostly developed by civilian companies and by their nature they can be used in both realms (Sezal and Giumelli 2022). Nevertheless, when we look at the funding of the R&D process the situation becomes more complex especially if public finances are involved. For example, security research programme (SRP) within the EU’s Framework Programmes was civilian in its nature and therefore it was not allowed to fund defence and military related technologies, but could fund dual-use technology R&D (Martins and Ahmad, 2020). Furthermore, dual-use technologies are still highly regulated on both the EU level and the member state level.

Sociotechnical imaginaries are also at play when we think about securitisation of technologies and understanding dual-use technologies in this respect. Jasanoff and Kim’s (2015) conceptualisation of sociotechnical imaginaries attests to how certain conceptions of future are embedded in the processes that lead to the development of specific technologies. Very briefly, it means that there is a conscious move to opt for a certain future (and in this case a certain understanding of defence and security) which drives R&D but also the role of technology with respect to the social life and social order (Jasanoff and Kim 2015; Martins and Mawdsley 2021). The discourse around dual-use technologies and disruptive technologies points to a specific socioscientific imaginary whereby securitisation of technologies plays an important part. As mentioned before, this securitisation process is somewhat different from the classical understanding of the concept, but is an extended version that is put forward by Floyd. Here it is important to reiterate that this new understanding of security goes beyond existential threats and focuses more on a significant change in policy and/or regulation.

The idea is that if a technology’s trade is regulated it means that there is a considerable and unusual change in policy. As previously mentioned, the general terms are regulated by the EU; however, member states still can have additional measures. This opens up an area which can be studied in depth. As such, we can understand how the process of securitisation, in other words, additional regulations for the export regime of dual-use technologies come into being. Here, we can look at which actors want securitisation and which actors do not and the reasons.

Based on this theoretical background and the nature of the defence industry I identified two actors and accordingly two sets of behaviour are hypothesised: 1) Ministries of defence as the main actor to devise policy and measures for “national security” would desire to significantly limit or ban export of dual-use technologies as a protective measure and are therefore pro securitisation. This desire would be validated by an active participation in the governance of trade regimes. 2) Companies operating in the defence sector would desire lifting or at least easing of export controls to have a free access to the market and are therefore pro desecuritisation/non-politicization. This would be validated by opposing the measures openly through calls, statements and lobbying.

4. Methods

This article uses several different data sources collected through different phases. Firstly, we have organised the Joint Effort for the Defence Industry (JEDI) Workshop in October 2019 which was attended by 17 participants representing seven Dutch companies currently engaged with country’s defence and security sector (two companies were represented by two participants), six academic experts from the Universities of Groningen, Bologna and Södertörn, and a strategic analyst from The Hague Center for Strategic Studies (see Appendix 2 for the list of participants). The companies were selected with the view to represent the market sector, so small and larger companies as well as representatives from manufacturing (i.e. marine, aviation and small technological devices) and the service sector (i.e. law consultant and logistic) were invited. In order to complement the views offered by the selected companies, a representative from the De Stichting Nederlandse Industrie voor Defensie en Veiligheid (NIDV – Dutch Industry Association for Defence and Security), which is the umbrella organisation that represents the whole Defence sector in the Netherlands, was also invited. The discussions were conducted under the Chatham House Rule which allowed for a relaxed environment where the participants could share their insight freely, but as a result their identities and affiliations are kept anonymous. The first phase was specifically valuable in terms of understanding the general position of the companies and other actors vis-à-vis export control regimes and therefore securitisation of technology.

The second phase took place in 2020. For this we contacted all of the Dutch companies from the database which we compiled and included all the companies which were openly listed as working in the security and defence industry via email and telephone. In the end we were able to interview a total of 11 representatives from 11 companies (see Appendix 3 for the list of interviews). The semi-structured interviews took place in the Netherlands via video call, and were conducted under the condition of anonymity. The interviews were conducted with a total of 10% of the companies featured by NIDV. These included any company that was willing to participate, acknowledging the potential presence of a natural bias in the sample. This is due to the nature of the defence industry and limited disposition of the actors to be a part of a research project. It is worth noting that academic research focused on interviews with defence sector companies has utilized similar sample sizes, as demonstrated in publications with different sample sizes ranging from 5 to 26 interviews (Calcara 2017, 2020; Haroche 2020). The second phase was instrumental in solidifying the industry’s perspective with dual-use technologies and the regulations of them.

The final phase was completed through administration of an online survey to experts in the field in March 2022. The survey has been sent to 22 security experts working at universities, think-tanks, and other research institutions in the Netherlands and received 19 responses (see Appendix 1 for the list of survey questions). All these data are accompanied by official sources and legislations regarding the export control regime governed by the EU and the Netherlands. There was no overlap of participants in the first two phases which allowed us to reflect on a diverse set of actors. In the final phase we cannot verify it cannot be verified if there is an overlap due to anonymised responses, however, there it is possible that one of the workshop attendees could have among the experts who responded to the survey. The final phase allowed for inclusion of the perspectives coming from the experts who work in the field.

Overall, it can be said that the analysis is based on a sufficiently broad empirical foundation, encompassing 21 semi-structured interviews with representatives from defence industries, supported by six academic experts, one strategic analyst and survey results from 19 experts. While allowing for a wide range of sources, the three-phased data collection method has certain limitations. It does not lend itself to a fully reproducible research design, and as a result, it might seem anecdotal at certain levels. However, this does not diminish the value of the factual evidence based on the experiences and observations of the participants who are already part of the processes that this research tries to explore.

Based on the gathered data, the analysis is conducted in two-levels. The first level focuses on the EU’s export control regime, and the second level analyses the Dutch national measures on dual-use technologies. In both levels, official documents and legislation are used to provide the basis of regulations. Then, the actors involved in the processes are identified, and their positions vis-à-vis securitisation of technology in the context of export control regimes for dual-use items are explained.

5. Export control for dual-use technologies: where does the agency lie?

5.1. The EU export control system and the actors

While export of arms is considered as a national competence, export of dual-use goods is governed by the EU. The latest version of the EU’s export control regime regarding dual-use items has entered into force in 2021. EU, Regulation 2021/821a replaces the previous version that was in force since 2009 (EC, Regulation No 2009). The rationale for an export regime is put forward in the Preamble 5 of the Regulation as making sure that the member states and the EU would be upholding their commitments to “non- proliferation, regional peace, security and stability and respect for human rights and international humanitarian law.”

The updated or modernised export control policy came as a result of reports, impact assessments, discussions on proposed amendments and afterwards outlined in the EU’s 2016 Global Strategy. One such impact assessment was done jointly by the Stockholm Peace Research Institute (SIPRI) and Ecorys in late 2015.

The articles 4(3) and 9(1) are considered as the catch all clauses in the regulations as they give the member states an important authority to expand the regime according to their public security interests or for human rights considerations:

Article 4(3): A Member State may adopt or maintain national legislation imposing an authorisation requirement on the export of dual-use items not listed in Annex I if the exporter has grounds for suspecting that those items are or may be intended, in their entirety or in part, for any of the uses referred to in paragraph 1 of this Article.

Article 9(1): A Member State may prohibit or impose an authorisation requirement on the export of dual-use items not listed in Annex I for reasons of public security, including the prevention of acts of terrorism, or for human rights considerations.

(EU, Regulation 2021/821b)

While the subsequent clauses of Article 9 urge the member states to notify the Commission and the other member states regarding the additional measures, this only constitutes providing information on an act that has already been adopted. Mentions of “public security” and “human rights” essentially make it possible for member-states to include whatever technology they want to be included in the export control regime or exclude whichever country they perceive or present as a threat to “public security” or “human rights”.

This article will not delve into the long process of how the new regulations were finally adopted which promises to be a significant focus for another article. Instead, here I identify certain actors who stand out in terms of their contributions to the process. On the EU level, involved actors are the Commission, the Council, industry organisations and other bodies that are aimed at creating dialogue between the experts, industry, and policy-makers such as the Export Control Forum (ECF) and Dual-use Coordination Group.

The power dynamics within and between the interest groups, companies, and the policy-makers are not clear at this moment. For example, the ECF brings together officials from the Directorate-General of Trade, Presidency, European Parliamentary Committee on International Trade, European External Action Service, relevant departments of some of the member states, numerous think-tanks, research institutes, consultancy firms, and industry associations (2021 Export Control Forum).

5.2. The Dutch defence ecosystem, export controls and the actors

The Dutch defence sector is called the Netherlands Defence and Security-related Industry (NL-DSI) which consists of “350 large and small companies with a joint turnover of €4.5 billion. The sector employs 25,000 people, of whom almost 8,000 are in research and development. The sector is responsible for approximately 0.7% of GDP” (Ministerie van Defensie, and Ministerie van Economische Zaken en Klimaat 2018). NL-DSI relies heavily on exports where these amount up to 70% of profits of the companies (JEDI Workshop). According to the 2018 Defence Industry Strategy Document, the Dutch defence sector is active and prominent in several fields such as maritime technologies, ballistic missiles, systems integration, sensors, radar development, combat management systems and C4I capability. However, the majority of the companies in the sector are SMEs. Research and technology organisations, National Aviation and Aerospace Laboratory (NLR – Nederlands Lucht- en Ruimtevaartcentrum), Netherlands Association for Applied Scientific Research (TNO – Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk) and Maritime Research Institute Netherlands (MARIN – Maritiem Research Instituut Nederland) also play a very significant role in the Dutch industry.

NIDV, the industry association which represents the majority of the companies that comprise the NL-DSI, plays a prominent role in the defence sector as well. Among other things it also functions as a facilitator between the companies and the ministries of Defence, Foreign Affaris, Economic Affairs and Climate Change. Considering the SME-heavy nature of the industry, NIDV’s position is indispensable for creating a synergy for the Dutch defence policy. It also needs to be kept in mind that the association, which is legally a foundation, is co-founded by the Dutch government.

All the companies or entities that export technologies, which are based in the Netherlands, are of course subject to the international and national export control regimes and regulations. The Netherlands participates in all of the existing export control regimes, namely, the Australia Group (AG), the Missile Technology Control Regime (MTCR), Nuclear Suppliers Group (NSG), the Wassenaar Arrangement, and the Zangger Committee. Furthermore, as an EU member state it is also subject to the EC EU, Regulation 2021/821a. However, as published in the Information Note in the EU Official Journal 2022/C66/27, the Netherlands is among the member states which have adopted national legislation imposing authorization requirements in application of Article 4(3) of the Regulation. As such, the Netherlands has an additional level of export controls in addition to the EU level regulation (and the items listed therein).

The Dutch Ministry of Foreign Affairs (MFA) is the main government body that “holds primary political and judiciary responsibility for implementing the export and transit controls – in particular decisions concerning license applications,” while the Minister for Foreign Trade and Development Cooperation (MFTDC) grants the export licenses for strategic goods (Ministerie van Buitenlandse Zaken 2013). Export license applications are made to Customs/Central Licensing Office (CDIU) which is part of the Tax and Customs Service Department under the Ministry of Finance (MoF) but mandated by the MFA to grant licenses. All in all, the Dutch export control regime is governed by the MFA and its mandates given to specific bodies in the MoF as well as MFTDC. It should also be mentioned that through the General Intelligence and Security Service (AIVD), the Ministry of the Interior and Kingdom relations (MoI); through the Military Intelligence and Security Service (MIVD), the MOD plays a role in monitoring and providing information to the MFA.

A recent example on how national level measures can be implemented is the Dutch export control policy with regard to Turkey. Following Turkey’s military incursions in northern Syria in late 2019, the Netherlands suspended the processing of all new export licenses for the export of military goods or dual-use goods with military end use. Although this measure has been modified to enable applications in late 2021 with some extended controls, Turkey has not been put back on the list of possible final destinations for a general licence (Rijksoverheid 2021). This shows that national measures are particularly very much political. However, there is a potential problem that emerges from this policy change. This either means that the destination country is securitized or securitisation of technology has different levels. It cannot be easily argued that it is only the destination country that matters because even though dual-use items can be traded freely within the EU, Annex IV of the Regulation lists certain sensitive items that require prior authorisation even within the EU.

To reiterate governmental actors responsible and/or that play a prominent role in the decisions regarding the export controls and therefore the dual-use items are the MFA, the MFTDC, the MoF the MoI, and the MoD. This implies that there needs to be a significant amount of communications and/or bargaining between these agencies. As can be seen from the example of Turkey, whatever the power dynamics between different agencies are, decisions to add or remove countries from the lists is very much political. It shows that what a specific technology does is of lesser value than who the recipient is. While such decisions also require the processes (i.e. committee discussions, motions, statements) to be further investigated, the Regulation by the EU already gives member state exemptions a political nature by linking these extra prohibitions to member states “security” concerns. Therefore, we can deduce that the role of the MoF and the MFTDC are much more limited than the MFA, MoI, and the MoD where security and/or human rights is a concern. What is also worth mentioning is that Turkey has been continuing its operations in Northern Syria but the Dutch government did not take back the eased export license applications.

My initial expectation, specifically coming from the securitisation framework, was that the MoD would be the main actor to devise policy and measures for “national security” and therefore would desire to significantly limit or ban export of dual-use technologies and be pro securitisation. The intertwined nature of different actors that play a role in the decision-making process shows that this argument cannot be easily verified. It is clear that the MoD is an actor in the process particularly through military intelligence, however, the extent of it cannot be determined at the moment. Furthermore, concerning the country-based bans in the national export control regime it seems that the MFA has a more dominant role and that role is determined by foreign relations concerns. As a result, it can be argued that MFA’s pro-securitisation or pro-desecuritisation stance rests on the developments in the current affairs and other immediate political concerns related to the government’s position.

Here we need to note that an important finding that emerges from the survey data is that no expert thinks that the classification of technologies as dual-use is a purely technical decision. 28% of the respondents think that it is somewhat technical, while 61% say that it is either very political or somewhat political but no one thinks it is purely technical. This finding is in line with the discussions in the literature which were mentioned in the previous section and where the definition of dual-use is not very clear. Implication of this is that the sovereign space that the EU member states have on additional export control measures is seen as political. This means that a certain level of negotiation/compromise should be part of the process.

Other actors that are also involved in the processes are the industry association NIDV and to a lesser extent the companies themselves. The companies’ individual role should not be overstated. They are usually content and careful in complying with the regulations (interviews #7 and #10) but some state their concerns regarding extra measures with regard to export to non-EU countries such as the United States (interview #11). In a similar vein, some companies note that the export control regulations enforced by individual member states create an uneven playing field where the Dutch regulations impose blanket bans on countries, while the French have a tier-based system where some technologies can be exported even though the said country is listed ineligible for state-of-the-art technologies (interviews #1 and #5). This is surely an impeding factor from a business point of view. Interviewees in these cases were not too explicit about their potential roles in the export control regime, however, it is implied that even if they were consulted their inputs are not translated into policy. Our data shows that NIDV does engage in lobbying activities at a certain level but it does not openly talk about removing items from the export control lists. Its role, however, requires further data to be investigated deeply.

All in all, I was expecting that the Dutch companies operating in the defence sector would desire lifting of export controls to have a free market and are therefore pro desecuritisation/non-politicisation. This expectation was in line with the experts who were surveyed as the 53% of the experts think that companies would like to abolish export controls, 26% are neutral and 21% disagree. However, the interview data shows that rather than a lifting of export controls, companies desire a more uniform regime. This implies that they would not want a second tier securitisation which is happening in the national level but would not go for challenging the first tier in the EU-level. While the most important actors were being identified, the reasons for their course of action in terms of their securitisation preferences have not become explicit. To be explored in future research, some preliminary hypotheses based on the analysis above for “why” these actors behave as they do would be: 1) Asymmetrical power dynamics caused by institutional structures cause some actors to be cautious of challenging the status quo. For example, SMEs would not want to lose their existing positions with regard to government tenders. In a related angle, SMEs which employ ex-government employees (i.e. former high-level armed forces officials) make use of the network connections that already give them a preferred position vis-à-vis contracts. 2) Government actors are motivated more by political developments and diplomatic relations with third countries when they consider technologies and destinations to be securitised. Therefore, we would see a pragmatic approach on behalf of these agencies. However, the power hierarchies would also play a role here.

6. Conclusions and future research

The contemporary defence industry is made up of both civilian and military companies that conduct research, development, and improvement of technologies for use by military and similar agencies. The key trend in this sector is integrated security technologies that combine multiple layers of surveillance, sensor data, and other relevant information to create an effective system. Often, these technologies have both civilian and military applications and are subject to export controls due to their non-civilian nature. This results in the designation of the said technologies as “dual-use” which can be considered as securitisation of these technologies, as argued by Ludovici (2021). This means that either technologies or the target destinations are constructed as existential threats to the relevant referent objects, thereby removing the related discussions from public deliberation and addressing them with emergency-like measures. In contemporary analyses, significant change of behaviour as a result of the securitising move is also considered as a successful securitisation (Floyd 2016) and this approach fits the topic of dual-use technologies very well.

Based on this background, this article asked the question “Which actors desire securitisation of technology and which actors desire desecuritisation?” Since the export of dual-use technologies is firstly regulated at the EU level and then at the member state level, the article explored who the involved actors were or in other words where the agency lies at both levels. I expected to find that the MoDs and companies operating in the defence sector would be the most important actors and while the MoD would try to securitise technology, companies would work for desecuritisation. However, the results proved to be much more complex than that.

On the EU level, the research showed that actors involved included the Commission, the Council, industry organisations and other bodies that are aimed at creating dialogue between the experts, industry, and policy-makers such as the Export Control Forum (ECF) and Dual-use Coordination Group. As such, it requires a further investigation of the EU-level decision-making into the relative powers of these actors and their negotiating/bargaining positions. On the Dutch level, the most prominent actors were MFA, MoI, and the MoD but the role of the industry and companies remained unclear. What was clear about the companies was that they did not seek out desecuritisation but also were not implying a support for securitisation. They wanted a uniform system throughout the EU.

These findings show that this research resonates with the core theoretical themes of securitisation, particularly regarding the political nature of the export control regimes. However, findings reveal that while the actors are important, the rationales behind their behaviour are also worth exploration. The short answer to question “why do these actors behave as they do?” is that the institutional dynamics determine the motives of the actors. In the case of corporations this involves a capitalist/economic paradigm (high sales, profit making and growth, competition with less restricted firms). In the case of ministries this involves political-economic considerations, nowadays strongly influenced by a return of traditional geopolitics towards countries beyond EU and NATO. If this is true, the “why” is determined by the institutional make-up/mandate of the actors, rather than individual features or media/public debate factors.

As such, this article serves as a fruitful starting point for a research programme, informed by securitisation studies, on security-defence nexus, actors operating within and their rationales for specific courses of action. First and foremost, the EU level needs to be further investigated so that the power relationship between different actors can be unravelled. Secondly, the Dutch case can be further elaborated by including parliamentary proceedings, lobbying and other relevant processes. Thirdly, global governance dimensions need to be explored in which identifying the “who” and “why” of the relevant actors is essential for identification of relevant multilevel governance structures on dual-use as these would be crucial for effective international regime formation. Finally, other case studies based on the defence sector structure would strengthen the research agenda that would include country comparisons and comparison of the positions of state-owned firms with private sector firms.

Acknowledgement

I would like to extend my gratitude to the convenors Marijn Hoijtink and Malcolm Campbell-Verduyn and the participants of the “Technoscientific Politics” workshop held as a part of the 2022 Annual Political Science Workshops of the Low Countries (Politicologenetmaal) for their invaluable comments and feedback. Additionally, I would like to thank the European Investment Bank University Research Sponsorship programme for sponsoring the JEDI Project which made this article possible, Francesco Giumelli for his guidance throughout the project, as well as Giampiero Giacomello and Johan Eriksson. Finally, I thank the anonymous reviewers and the editor for their helpful comments. All errors remain my own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the European Investment Bank University Research Sponsorhip (EIBURS) programme.

Notes on contributors

Mustafa Ali Sezal

Mustafa Ali Sezal is a lecturer at the Department of International and International Organization, University of Groningen. His current research interests include Critical Security Studies, European Security and Defence, and Technoscientific Politics.

Notes

1. The categories are: 0-Nuclear materials, 1-Materials, chemicals, micro-organisms, toxins, 2-Materials processing, 3-Electronics, 4-Computers, 5-Telecommunication and “information security,” 6-Sensors and lasers, 7-Navigation and avionics, 8-Marine, 9-Aerospace and propulsion (Regulation 2021/821).

2. For recent comprehensive assessments of this evolution see (and compare): Balzacq et al. (2016)); Balzacq (2019); Sjöstedt (2020); Baele and Jalea (2022).

3. Particularly in the tradition of John L. Austin (1975) and John Searle (1969).

4. In addition to the Dutch MoD, see for example, German MoD (www.bmvg.de) and the armed forces (www.bundeswehr.de/en/about-bundeswehr/mission-and-tasks-bundeswehr), French MoD (www.defense.gouv.fr/ministere/politique-defense), Italian MoD (www.difesa.it), Belgian MoD (www.mil.be/fr/a-propos-de-la-defense) etc.

Appendix 1

– Survey Questions

1. Has the meaning of “defence” changed since the 11 September attacks?

2. If you answered “yes” or “other” in the previous question, could you elaborate your answer in one sentence?

3. Do you think that some technologies (excluding weapons, munitions and military materiel) are “dangerous” to be freely exported or imported?

4. If you answered “yes” in the previous question, could you list a few examples of the said technologies?

5. Do you agree with the statement “it is difficult to define what dual-use technology means”?

6. From scale 1 (very technical) to 5 (very political), how political is the classification of dual-use technologies in your opinion?

7. In addition to homeland defence what functions do ministries of defence (MoDs) in contemporary Europe fulfil?

8. Would you agree with the statement “securitisation of technologies, which is usually achieved through classification of dual-use and export control regimes, is desirable from a MoD perspective”?

9. Would you agree with the statement that “from an economics/trade perspective export controls should only be implemented when the technologies are clearly military related.”

10. Would you agree with the statement that “the businesses operating within the security and defence industry would be more interested in abolition of export control regimes”?

11. If you were to decide on export control regimes, would you be more strict or more lenient towards dual-use technologies?

Appendix 2

- List of workshop participants

1. Dutch company A, NIDV member

2. Dutch company A, NIDV member

3. Dutch company B, NIDV member

4. Dutch company B, NIDV member

5. Dutch company C, NIDV member

6. Dutch company D, NIDV member

7. Dutch company E, NIDV member

8. Dutch company F, NIDV member

9. Dutch company G, NIDV member

10. Representative of the Nederlandse Industrie voor Defensie en Veiligheid (NIDV)

11. Expert from The Hague Center for Strategic Studies

12. Expert from the University of Bologna

13. Expert from Södertörn University

14. Expert from the University of Groningen

15. Expert from the University of Groningen

16. Expert from the University of Groningen

17. Expert from the University of Groningen

Appendix 3

– List of interviews

1. Dutch company, NIDV member, 12/05/2020

2. Dutch company, NIDV member, 20/05/2020

3. Dutch company, NIDV member, 28/05/2020

4. Dutch company, NIDV member, 02/06/2020

5. Dutch company, NIDV member, 03/06/2020

6. Dutch company, NIDV member, 05/06/2020

7. Dutch company, NIDV member, 08/06/2020

8. Dutch company, NIDV member, 09/06/2020

9. Dutch company, NIDV member, 12/06/2020

10. Dutch company, NIDV member, 15/06/2020

11. Dutch company, NIDV member, 29/06/2020