Civilian–military co-operation strategies in developing new technologies

Research Policy 32 (2003) 955–970

Civilian–military co-operation strategies in developing new technologies Haico te Kulve, Wim A. Smit∗ Centre for Studies of Science, Technology and Society, TWRC Room D-304, University of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands Received 3 December 2001; received in revised form 22 May 2002; accepted 18 July 2002

Abstract Dual use technology has been advocated as the solution for the twin problem of maintaining a high tech defence technology base and improving economic competitiveness. The concept of dual use technology turns out to be rather imprecise representing a multitude of different meanings. This paper focuses on one important aspect, notably the co-operation between civilian and military actors in developing a new technology, by analysing the evolution of a socio-technical network related to the development of an advanced battery in The Netherlands. The analytical framework used for interpreting the empirical case builds on theories of socio-technical networks and on two previous and complementary analyses in Research Policy on dual use technology. Our analysis of the dynamics underlying the evolution of the ‘battery network’ shows how the emerging notion of the battery’s duality became a window of opportunity for a co-operation strategy of either joint or concurrent development of the battery for both civilian and military applications. The interactions within the evolving network are steered by the search for expertise and funding. In view of the difficulties of realising civilian–military integrated joint development projects, the establishment of ‘dual capacity networks’ is suggested as part of a possible strategy towards an integrated civilian–military technology and industrial base. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Dual use technology; Socio-technical network; Civilian–military integration

1. Introduction 1.1. The concept of ‘dual use’ During the last decade governments, and to a lesser degree, defence related companies have shown substantial interest in the issue of dual use technology: technology that has, generally speaking, both military and civilian applications. Though the concept of dual use technology is not entirely new, for some ∗ Corresponding author. Fax: +31-53-4894775. E-mail address: [email protected] (W.A. Smit).

technology fields its meaning has shifted from a problematic to a desirable feature. The concept entered the discourse on weapons and technology exports that started in the years after World War II (Reppy, 1999). The acrimonious East–West relations soon resulted in the establishment of the Co-ordinating Committee for Multilateral Export Controls (COCOM) in 1949. It became the major framework for the US and its allies for export controls. Under this regime, dual use was viewed as a negative feature that complicated export controls: countries might try to obtain militarily sensitive technology under the guise of buying civilian technology. The presumed dual nature of some products

0048-7333/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 7 3 3 3 ( 0 2 ) 0 0 1 0 5 - 1

956

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

and technologies also created tensions between the economic and the defence perspective on technology exports, not only within the US, but also between the US and west-European countries (Bertsch, 1988). By the time the Cold War ended, a profound change in the discourse on dual use products occurred. Rather than a negative feature, the dual use aspect of technology was viewed as something that should be promoted and pursued, as it might solve the twin problem of maintaining a high tech defence technology base restrained by limited budgets, and improving a country’s economic competitiveness by a more efficient allocation of R&D funds. From this perspective, the civilian and military contexts for developing technology should be integrated where possible, rather than separated by a technological divide between military and civilian applications (OTA, 1994, 1995). Though at face value the distinction between civilian and military technology seems obvious, from an analytical point of view it appears that such a distinction, and therefore the concept of dual use technology, is less clear. One position on the meaning of civilian, military and dual use technology is that it is an intrinsic feature of the technology or product itself. The opposite and equally extreme position is that it all depends on the social context of the (use of) technology. The former position seems very hard to maintain. For instance, in the 1960s and 1970s even nuclear explosives have been designated as having possible civilian applications, like digging canals and creating underground storage cavities, next to being used as nuclear weapons (see e.g. Davies (1979)). No intrinsic feature distinguished nuclear explosives for military purposes from those for peaceful purposes. Indeed, basically all the R&D required to develop these explosives were the same, only the intention of application and the accompanying institutional setting might be different (Goldblat, 1982, p. 27).1 Nowadays, however, the perceptions have changed, and such civilian applications are no longer considered as a serious option: nuclear explosions now have only military applications. In the same vein, Gummett (1991) has argued that the distinction between military and civilian technology 1 Of course, the development of delivery systems for nuclear weapons and the adaptation of nuclear explosives to this end may require further R&D. For the physics and principles of nuclear explosives, see e.g. Hansen (1988, pp. 18–29).

is an institutional rather than an intrinsic one. Still, given a certain social-cultural setting, some technologies will be more apt for applications in both domains than others. In the next section we will describe the perspective of our study on the phenomenon of dual use technology. 1.2. This study’s perspective on dual use From a technology studies perspective, which emphasises the mutual shaping of technology and its social context, neither the proposition that the military or civilian nature of a technology is located only in the technology itself, nor that it all and only depends on the social context is tenable. In line with this view, we will focus on the interactions between civilian, military and dual-oriented actors, which shape both technology and context. Many aspects—technological, cultural, social and organisational—will influence the interpretation of a technology as being military, civilian or dual use. These include, differences in governmental influence and regulations, goals (national security versus commerce), market structure, standards and specifications (‘milspecs’), sensitivity to costs, different product cycles (years in the commercial versus decades in the military sector), industrial and technological ‘cultures’. Alic et al. (1992) comment on military and commercial technological innovation as two systems which, though drawing on a common base of technical knowledge, “involve different sets of institutions and in general operate quite differently—the result of differences in goals and technical requirements, as well as in managerial arrangements accompanying defence production in particular and all governmental activities in general. “[. . . ] As a result, in the majority of cases military and commercial innovation have evolved distinctive technical ‘cultures’ [. . . ].” MacKenzie (1990), in a detailed study on the development of guidance technologies in their social context and on technological choices for improving accuracy, has shown how different emphases in requirements for missile accuracy and for civilian (and military) air navigation resulted in alternative forms of technological change: the former focusing on accuracy; the latter, on reliability, producibility and economy. In this article, rather than focusing on conceptual problems of the meaning of dual use technology, we

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

present an empirical study in which we focus on the dynamics underlying the evolution of a socio-technical ‘bipolar battery’ network during the development2 of an advanced high power battery. Special attention will be paid to the emergence and role of the phenomena of ‘duality’ and ‘dual use’ in shaping the network. To this end we investigated the interaction between civilian, military and dual-oriented actors and the way a co-operation between these actors emerged. The analytical framework we use for positioning and interpreting the empirical case builds on theories of socio-technical networks (Elzen et al., 1996) and on two previous and complementary analyses on dual use technology that appeared in Research Policy: Cowan and Foray (1995) and Molas-Gallart (1997). This framework will be elaborated in the next section. The case itself of the evolution of the socio-technical ‘bipolar battery’ network will be presented in Section 3.

2. Research methodology 2.1. Research focus and analytical background As mentioned in Section 1, the issue of dual use technology received substantial interest in the past decade, both from policymakers and analysts. See, for instance, Gummett and Reppy (1988), Alic et al. (1992), a number of publications from the US Office of Technology Assessment (OTA, 1992a,b, 1993, 1994), Gummett and Stein (1997), Reppy (1998) and Markusen and Costigan (1999). It is also clear from these publications that there is no unequivocal concept of dual use technology and that, moreover, the term may refer to products, to knowledge and skills or to activities in developing technology. One reason for being such an imprecise concept is that it is used in different contexts and for discussing quite a variety of problems on the relation between civilian and military technology (Smit, 1995; Reppy, 1998). Another reason, mentioned by Molas-Gallart (1997), is the vast array of mechanisms by which technologies can be transferred across military and civilian applications. Our focus is related to the interest shown by governmental agencies in the joint development of 2 By ‘development’ we refer to the whole process from fundamental research, applied research to product development.

957

technology for military and civilian applications. In addition to reducing costs of military equipment, the main purpose of such governmental dual-use policies, in particular in the US, is to establish a strong defence technology and industrial base. Next to having declaratory policies on dual use, a number of countries have actually started a variety of activities to this end. Under the Clinton administration, the US started the Technology Reinvestment Program, including several programmes that aimed at the development of technologies with both military and commercial applications, like the Advanced Manufacturing Technology Partnerships, the Dual-Use Critical Technology Partnerships and Commercial–Military Integration Partnerships, all administered by the Defence Advanced Research Projects Agency (DARPA). After its ending in 1997, it was succeeded by the tri-service Dual Use Science and Technology (DUS&T) Program (http://www.afrl.af.mil/dualuse as of 25 February 2002), though at a much reduced level.3 In the UK, the Defence Research Agency (DRA) promotes technology transfer from the military to the civilian sector, partly by co-operating with the private sector. It has also established Dual Use Technology Centres, like the Structural Materials Centre (SMC) and Farnborough Supercomputer Centre and identified some 15 technology areas where defence research may have civilian applications (Milne, 1998). The German government, in 1990, declared that dual use was a basic ingredient of its R&D policy, and that, as a matter of principle, the Ministry of Defence bases its science and technology programmes on civilian science and technology (Altmann et al., 1998). The French Délégation Générale pour l’Armement has at least a declaratory policy on dual use technology and integration of civilian–military technology (Serfati, 1997). In Section 4, we will contribute to the debate on civilian–military integration and the establishment of

3 According to one of the anonymous referees, the budget was reduced from US$ 600 to 185 million after the Republicans has won control of both houses of US Congress. The goal of the program actually shifted too and became directed almost exclusively at military objectives, though the program was still labelled as ‘dual use’. Whereas, the Clinton Administration believed that industry might profit from dual use technology, the Republicans were much opposed to an ‘industrial policy’.

958

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

a common civilian and military technological and industrial base using suggestions from our case study. Before elaborating on our analytical framework, we summarise the main elements from the two Research Policy articles by Cowan and Foray (1995) and Molas-Gallart (1997) on which we build. Cowan and Foray’s interest is in the changing relationship between the military and civilian technology sectors. Their departure point is, in line with the exposé given above, that “duality of a technology [. . . ] is not typically inherent in the technology itself.” They define ‘dual technologies’ as “technologies that are developed and used [emphasis added] both by the military or space sectors on the one hand and by the civilian sector on the other,” and emphasise that the nature of the technology (that is, either civilian or military) depends on the social networks in which it is developed. By implication, “a potentially [emphasis added] dual technology may never display dual uses.” Using such a strict definition, the authors conclude that spillovers, “in which particular research is done exclusively in one domain and adopted more or less without change in the other [. . . ] is not evidence of duality, [but rather] evidence of its absence.” More important than this conclusion—based on whether some phenomenon fits a pre-established definition— is, in our view, their analysis of: (1) how the realisation of an integrated civilian–military technological development—that is, of ‘dual technology’—may depend on the phase of the technology’s life cycle; and (2) that in that respect R&D programmes on product technologies may differ from process technologies. They argue that in the first stage of a military R&D programme more room exists for experimental variety, whereas in the latter stage standardisation (or rationalisation), related to specific applications, becomes more dominant. As a consequence, opportunities for civilian–military collaboration will be greater in the first and less in the latter stage of the technology development life cycle. In that sense, a technology’s potential duality is not a constant but evolves over time. Process technologies are different, according to Cowan and Foray, to the extent that their potential duality is generally larger and decreases less than in the case of product technologies, and may even increase at the latter stage of the development cycle. Realisation of duality depends in all cases on whether suitable co-operative socio-technical networks will be created.

In addition, adaptations in requirements (‘milspecs’) and participating military and civilian organisations may be required for a successful R&D co-operation. Molas-Gallart (1997) emphasises the imprecision of the concepts of not only dual use technology but also of civilian and military technology. He considers a technology (which includes, knowledge, skills, production processes, management techniques and products) as dual use “when it has current or potential military and civilian applications.” Rather than trying to classify technologies as to their scope for dual use,4 Molas-Gallart focuses on the different (interaction) mechanisms through which technologies may cross the border between civilian and military applications. He actually presents a typology of such technology transfer mechanisms, thus emphasising the organisational aspects of dual use transfer.5 Two main dimensions can be distinguished. One refers to whether the transfer occurs within the same or between different business units (or organisations). The other dimension concerns whether a technology requires adaptation in order to be transferred from the civilian to the military domain or vice versa.6 Within each of the four main classes several sub-categories can be distinguished and coupled to different business strategies. Both Cowan and Foray and Molas-Gallart try to cover all the possible dual use features by their concepts and classifications. Our approach of the phenomenon of ‘dual use’ and ‘duality’ is from a different perspective. In investigating a specific case of technology development, notably the development of the ‘advanced bipolar lead–acid battery’, we trace the emergence of the dual use phenomenon in this case 4 Molas-Gallart also argues that the large technological classes distinguished by Cowan and Foray (i.e. product versus process technologies, and a division of technological development into an experimental and standardisation phase) are, on closer inspection, too heterogeneous to warrant general conclusions. 5 The perspective used in this typology is that from the supplier (i.e. the business), not from the customer. 6 Buying commercial of-the-shelf (COTS) by the military is an example of a transfer without adaptation and within one business unit. Likewise, the situation that engineers or designers, within one company, alternatively design civilian and military products, provides an example of internal, straight transfer of technological skills. Spin-off, in which a company uses basic technological know-how developed in a defence company for developing a civilian product, like the microwave oven, provides an example of external transfer with adaptation.

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

959

Fig. 1. Model of civilian, military and dual technology development processes.

and show its influence on the dynamics underlying the evolution of the socio-technical battery network. Instead of using a more distant and generalising perspective, we dive into the process of technology development itself, or to be more precise, of the shaping of a socio-technical network. We show the growing awareness of the possibility of duality when new actors enter the stage, resulting into questions by the actors like: “How can this duality help us to get a product with the required properties?” “Which form of co-operation should we choose and with whom to reach our goals?” We explicitly look at the interactions between the actors of the network, the durability of the network and the co-operation strategies of the actors in realising the development of a new type of battery. We distinguish two main types of co-operation strategies between civilian and military oriented actors. 1. Joint technology development: a civilian–military integrated development in which civilian and military actors co-operate within one project. 2. Concurrent technology development: a civilian– military integrated development with parallel but distinct civilian and military projects, though connected through mutual interactions. Fig. 1 illustrates the taxonomy of different technological development processes of the two main types of co-operation strategies. Moreover, the figure illustrates the perspectives of Molas-Gallart, Cowan and Foray and this study. The horizontal arrows on the right show the nature of the project, that is, the type of application being developed: dual, civilian or military. The right side also shows possible (dual use) transfer situations, from military to civilian applications and

vice versa as discussed by Molas-Gallart.7 A potential dual technology can be transformed into a dual technology, in the Cowan and Foray sense, when it is further developed within one single project aiming at the development of both civilian and military applications. This can result either into one product that can be used both within a civilian and a military context, or into two different products with strong commonalities but still different specifications as to fit the two distinct contexts. 2.2. Socio-technical networks The previous section showed an emphasis on socio-technical networks. The network perspective is of interest here, because technology, in particular military technology, is often developed within networks of a heterogeneous set of actors. Moreover, the nature of and interactions within the network influence the type and characteristics of the technology being developed. This is certainly the case for military technology (MacKenzie, 1990; Enserink, 1993; Smit, 1995; Elzen et al., 1996). For military technological developments, relevant actors include defence companies, governmental agencies, the military services and the Ministry of Defence. In the civilian sector, where technical know-how and assets required for innovation are often dispersed among a number of companies, and even when all the assets are available within one company, managers often prefer co-operative relationships with other firms in order to share risks 7 The taxonomy of Molas-Gallart (1997) actually includes all vertical arrows.

960

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

and costs (Cagliano et al., 2000). Such rationales for collaboration may also be of interest to Ministries of Defence that have to develop new weapons systems and are bounded by limited budgets. Dual use policies aiming at a strong and integrated civilian–military technological and industrial base are a case in point. In our analysis of the development of the bipolar battery we will use this network perspective in analysing the emergence of the collaboration between military and civilian actors. One important feature of socio-technical networks, once established, is that they represent relatively stable and enduring patterns of interaction. Another feature is that within a network the collaborating actors are mutually dependent and no single actor can usually determine the course of action, and thus the direction of technology development (Smit et al., 1998). Though in the case of the bipolar battery development the endurance of the emerging dual network has not yet proved, the network perspective is still useful, because it may suggest a number of items that are important to arrive at a more stable co-operation (Elzen et al., 1996) and hence, is of interest for the debate on creating a common civilian–military technology and industrial base. One such item is that some actors have to play the special role of dedicated network builders, whereas other network actors may be more reactive. The dedicated network builders have a large interest in the project becoming successful. Therefore, they are actively engaged in building up the network and involving new actors. Another important category of actors is the critical actors. Though they do not need to be actively involved in developing a new technology, their support, if even passive, is indispensable for the development to be successful. In technology projects with governmental departments involved, Parliament often is such a ‘critical actor’. Also financial investors may act as ‘critical actors’. Once successful, socio-technical networks often show a tendency of preservation (Elzen et al., 1996). Resilience has built up and is rooted in the relations between the actors, that is, if one actor would jeopardise the network by its actions, others take care of stability and look for compromises. Still, of course, it is possible that networks become unstable and collapse. Exogenous interactions are important sources for instability. For instance, when an actor holding an important position in the network also participates in other networks,

developments in those networks may cause conflicting interests and thus jeopardise the former network. It may even result in the formation of new networks, which is typical of technological innovation. 2.3. Data collection For our empirical study on the bipolar battery network we collected data from different sources. In addition to literature research, we conducted in-depth interviews with leading representatives of organisations that are involved in the development of the battery. • Directorate Material of the Royal Netherlands Navy (RNlN). • TNO Prins Maurits Laboratory (TNO-PML), a semi-governmental defence research laboratory. • TNO Environment, Energy and Process innovation (TNO-EEP), one of the large Dutch institutes for technological innovation. • Centurion Accumulatoren B.V. (Centurion), a Dutch manufacturer of batteries. In the next section, we present the evolution and underlying dynamics of the socio-technical network of the advanced battery.

3. Case: development of the socio-technical network of the bipolar battery 3.1. Setting of the case The stage of the battery development case is The Netherlands. The Ministry of Defence’s annual military R&D budget is relatively small and amounts to about 70 million . In contrast to countries like the UK, the US and France, The Netherlands has no explicit dual use policy. Still, the Defence White Paper 2000 mentions that the government prefers to use civilian standards and technologies in developing and procuring defence equipment whenever possible. It also states that the Ministry of Defence highly values a fruitful interaction between civilian and military research. In that respect, one may argue that The Netherlands has an implicit dual use policy. The subject of our case study is the development of a bipolar lead–acid battery network in The

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

Netherlands8 and the accompanying emergence of a civilian–military co-operation. The research and development on the battery started within a military context. The battery was supposed to serve as an intermediate energy storage system in a pulsed power supply for high energy weapons like electromagnetic propelled guns or lasers, which could be used on, for instance, warships. The network initially consisted only of military actors, but over time it broadened and came to include also civilian and dual (that is, both civilian and military) oriented actors. Also, the foreseen application domain broadened over time. For instance, the battery might also serve as a peak shaving element on warships, that is, as an auxiliary power supply in case of high power demand. Moreover, actors became aware that the battery might also serve as an auxiliary power supply for commercial (and military) hybrid electric vehicles (HEVs),9 thus recognising a dual use potential according to Cowan and Foray’s understanding of dual technology. Below, we will analyse the dynamics underlying the emergence of the socio-technical battery network in The Netherlands, while using the analytical framework depicted in Section 2 to study the evolving civilian–military interactions. 3.2. Dynamics of the innovation of the bipolar lead–acid battery network In this section we identify the structural factors that have influenced the evolution of the socio-technical network. Special attention is paid to the type of actors that are involved, how they became involved, how they interacted and how they influenced the development of technology and its context. It turns out that two forces dominated the interactions between the involved actors, thus propelling and steering the evolution of the socio-technical network: 1. the availability of funding and expertise; 2. the awareness of potential duality. 8 World-wide, several projects deal with the development of a bipolar lead acid battery, but according to Pinsky and Grosvenor (2000), none has developed a commercially viable product yet. 9 Such vehicles combine conventional combustion engines with electromotors for their drive.

961

Fig. 2. Network of battery development 1984–1993.

On the basis of significant changes in the sociotechnical network, we have distinguished four periods in the development of the battery, which we will discuss below. 3.2.1. Period 1984–1993 The development of the battery started around 1984 in the Pulse Physics Laboratory at TNO10 Prins Maurits Laboratory (TNO-PML), which is one of the three institutes of TNO that are primarily defence oriented. Together with the other two,11 it serves as the house laboratory of the Ministry of Defence (MoD). At that time, TNO-PML was involved in research on electromagnetic launch technology (Kolkert, 1989). Part of that research was the development of a bipolar battery which could be used in a pulsed power supply for high energy weapons. The Royal Netherlands Navy (RNlN) became soon involved in the project and became the actual problem-owner: it envisaged using the battery on future warships. In this period, no civilian application of the battery was planned or foreseen. Although the RNlN and TNO-PML tried to involve civilian industry, no civilian–military interactions of significant impact on the socio-technical network occurred. Fig. 2 shows the socio-technical network of the battery development in this period. At this stage of network evolution only a military application is planned and by lack of civilian interest no civilian–military co-operation develops (compare Fig. 1). The requirements of the RNlN determined the development within the socio-technical network during this period. The research and design strategy for developing the bipolar lead–acid battery reflected these requirements by being geared to obtaining as high 10 TNO is The Netherlands Organisation for Applied Scientific Research and is made up of several institutes. 11 These are TNO Physics and Electronics Laboratory (TNO-PEL) and TNO Human Factors (TNO-HF).

962

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

as possible specific power (power/kilogram) for the battery. By the end of this period, it was clear that the battery was indeed a suitable candidate for an intermediate energy storage element in a pulsed power supply. However, the development of the battery was delayed due to some technical problems. It appeared that a crucial component of the battery lacked sufficient durability. 3.2.2. Period 1994–1996 In 1994, the RNlN tried to fix the durability problem by mobilising additional expertise and involving another TNO-laboratory, TNO Environment, Energy and Process innovation (TNO-EEP). Thus, the need for additional expertise was the driving force for the extension of the network in this period. Because TNO-EEP had been involved in previous research for the RNlN, e.g. the development of lead–acid batteries for submarines, it was a well-known actor to the RNlN. In addition to its activities for the Navy, the laboratory had well-developed relations with still another TNO-laboratory, TNO Automotive (TNO-AT), which is oriented towards the development of sustainable and economically viable means of transport. Because of its close relations with both civilian and military actors, we call TNO-EEP a dual oriented institute or dual actor. The involvement of this dual actor proved to be an important factor in the further evolution of the network. The electrochemical section of TNO-EEP eventually found a solution for the durability problem mentioned above and patented this solution. In this period, the Dutch lead–acid battery manufacturer Centurion Accumulatoren B.V. was involved by TNO-EEP and supplied some components for the first experiments with laboratory models of the battery. Centurion sells batteries mainly for civilian applications and is primarily oriented towards the civilian market.12 The inclusion of TNO-EEP in the socio-technical network was not only key to the solution of a technical problem, it also changed the relations between the actors. TNO-PML was no longer active in the development itself of the battery, a role which was taken over by TNO-EEP. TNO-PML’s role actually changed

to supplying the specifications for the naval application of the battery to TNO-EEP and taking care of the integration of the battery in the pulsed power supply. Whereas, TNO-EEP had become involved primarily because of its technical expertise, it also proved to be key in coupling civilian and military oriented actors, as will become evident in the next period. Actually, this carries an important lesson for future civilian–military co-operation initiatives, notably, to involve actors not only because of their technical expertise, but also because of their possible socio-technical expertise, in particular their capability of coupling civilian and military oriented actors and networks. These dual actors are particularly qualified for translating and communicating different meanings and interpretations between civilian and military technical cultures.13 A highly debated issue in dual use related discussions is the necessity of ‘milspecs’ for military systems (Alic et al., 1992, pp. 150–153). In the case of the battery, however, the shock resistance and vibration features required by the military do not pose a major problem for a common civilian–military development project. As a last resort, the battery could be packed in a special container to protect the equipment against severe naval conditions. This method is in fact common practice and paves the way for procuring a commercially available battery as well as for civilian–military co-operation. Though, at the time, general interest in electric vehicles increased and the possible application of the battery in those vehicles emerged on the horizon, no concrete research towards this application was carried out by the involved actors. Still, some interesting civilian–military interactions are worth to be mentioned. First, with the entrance of TNO-EEP, an actor appeared that could act as a potential technology broker between the military and commercial sector. Second, the private company Centurion entered the stage by supplying some commercial battery components, and its further involvement paved the way for a possible civilian–military R&D programme. Third, one may consider the presentations and publications in the public sphere by Saakes et al. (1996, 1997) on the military oriented battery research project as a (weak)

12 It has sold, however, compact power batteries to the RNlN for their frigates of the Karel Doorman Class.

13 The difference between military and commercial technical “cultures” were pointed out by Alic et al. (1992, p. 43).

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

Fig. 3. Network around the battery 1994–1996.

civilian–military interaction.14 According to Cowan and Foray (1995), this aspect of information disclosure and mixed research settings are important requirements for the civilian sector to benefit from military research during the experimentation phase. However, we found no evidence that during this and the former period the civilian sector actually benefited from this research. This was to change in the next period. The socio-technical network of this period is pictured in Fig. 3. At this stage of network evolution still only a military application is foreseen, and though non-military actors (TNO-EEP and Centurion) became involved (primarily to solve a technical problem) the emerging civilian–military co-operation did not aim for civilian applications (compare Fig. 1). 3.2.3. Period 1997–2000 In this period, the socio-technical network changed radically under influence of the two main forces driving the evolution of the network. It was in this period that a more focused interest in hybrid electric vehicles (HEVs) for the commercial market increased. TNO-Automotive (TNO-AT) was working on a HEV demonstrator project aiming at the civilian market. TNO-EEP co-operated in this project and was searching for suitable batteries, when it recognised the potential of the battery of its project with the RNlN. They became aware that the bipolar lead–acid battery might also be used in a hybrid electric vehicle. Thus, this period actually showed the birth of the battery as a potential dual technology, in the sense of Cowan and Foray’s understanding of dual technology. The different actors then realised that they might benefit from this potential duality by collaborating on 14 These interactions can be viewed as examples of MolasGallart’s (1997) dual-use transfer mechanisms.

963

the development of the battery for both a military and civilian application. In this way, the dual potential of the battery and the presence of the dual actor TNO-EEP in the network became a window of opportunity for a co-operation strategy of either joint or concurrent development. How this awareness of duality, together with the driving force of available funding, further shaped the socio-technical network will be discussed below. In this period, moreover, the RNlN, modified its requirements because it no longer viewed the battery solely as a component in a pulsed power supply, but also as a potential peak shaving element for its future warships. The reasons for this modification were independent from the development of the battery, but related to a changing defence strategy. This modification, however, is important, because different applications require, for instance, different energy/power ratio’s of the battery. As Table 1 shows, the peak shaving requirements come closer to those for traction. For both the pulsed power/peak shaving and the HEV applications, design strategies for the battery aim at the maximisation of specific power content and both battery applications require a bipolar configuration and the same bipolar substrate.15 However, because of still different power and energy requirements in the military and civilian applications16 (recall Table 1), the thickness of the plates, which contain the active mass, would have to be different. The actors involved considered this technical aspect not as a major problem for a joint development programme, though it was evident that the possible end products of such development activities would diverge. Still, it turned out that a co-operation strategy of joint development was not viable, for reasons to be explained later. By contrast, a co-operation strategy of concurrent development appeared as a viable option (compare Fig. 1) and was pursued, as will be shown below. The search for funds and expertise to continue the development of the battery not only propelled the extension of the network in this period, but influenced the network interactions as well. Typically, the RNlN sought support for further development from another 15 A crucial element of advanced bipolar batteries, according to Saakes et al. (1999). 16 For the military application see Saakes et al. (1997), and for the civilian application see Saakes et al. (1999).

964

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

Table 1 Applications and required features of a battery

military organisation: the Dutch Army (related to the army’s need of new transport vehicles) and from the Western European Armaments Group (WEAG), a forum in which European naval representatives participate. By contrast, TNO-EEP extended the network by initiating a research project for commercial applications of the battery in HEVs, funded by NOVEM (see below), and also applied, in co-operation with the private company Centurion, for project funding from the European Fifth Framework Programme (EFFP). It is worth noting that the navy sought support primarily within the military domain. For one thing, this is not surprising because it illustrates that actors are embedded in certain socio-technical contexts and may be hesitant to explore new types of networks because of the accompanying transaction costs. At the same time, the Navy became well aware that the advanced bipolar lead–acid battery had potential commercial applications. From the Navy’s perspective, therefore, teaming up with industrial partners in order to mobilise sufficient financial support17 might still be an interesting option. Industry, however, will be interested only when there is a profitable market for the battery, which in this case would be the commercial automobile market, in particular of HEVs, rather than the very limited market for warships. In fact, once the battery’s potential duality was recognised within the framework of a generally increased interest in HEVs, most of the actors involved

became aware that funds would be more readily available for developing the battery for civilian applications. Saakes et al. (1997) note that especially the interest in electric vehicles pushed the development of this type of batteries. Thus, because of the limited size of the defence market for which little funds were available, the further development of the battery depended to a high degree on the civilian market and on the participation of private companies. Thus, the evolution of the network was influenced both by the different applications anticipated for the battery and by the available funds for further development.18 It also makes clear why the RNlN supported the extension of the network, as it was unable to determine the development of the battery on its own. The actors’ anticipation on the availability of funds resulted in the formation of three projects, as shown in Fig. 4: one military (naval battery project) and two consortia focusing on civilian applications (the NOVEM project and an application for funding by the European Fifth Framework Programme). (See also the section below on the period after the year 2000.) It is of interest to note that these consortia partly consisted of the same actors. The socio-technical network as it evolved in this period and is pictured in Fig. 4 shows the central role of TNO-EEP (its participation in all three projects being represented by the ‘TNO-EEP-axis’). TNO-EEP became the most active actor in each of the three consortia. It actually became

17 This is an important aspect because of the limited defence budgets.

18 Of course, the availability of funds also influenced the expectations on the battery’s future and vice versa.

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

965

Fig. 4. Socio-technical network of the battery 1997–2000.

a critical actor in the socio-technical network, because without them, it would have been much more difficult to couple civilian and military oriented projects and to realise and profit from technology transfer between the projects (represented by the two double arrows in Fig. 4). In addition to being a critical actor, TNO-EEP also played a role as dedicated network builder: as patent holder of a crucial part of the battery, TNO-EEP had a clear interest in the further development of the battery and became, like the original problem owner RNlN, very active in involving new actors. Though the battery manufacturer Centurion participated in all three projects (see the ‘Centurion-axis’ in Fig. 4), its active involvement was limited to the civilian NOVEM project. Centurion, therefore, had less direct influence on the development of the naval battery. This relatively low participation in the naval battery project was mainly caused by the company’s limited financial resources and research capacity. As a ‘dual actor’, TNO-EEP also performed a gateway function for the transfer of knowledge between the naval project and the commercial HEV projects. This gateway function is nicely illustrated by the NOVEM project (1998–2000). The project was initiated by TNO-EEP and aimed at demonstrating the technological feasibility of the bipolar battery for commercial HEV applications. It included

TNO-Automotive (TNO-AT) as a strategic partner of TNO-EEP, the private company Centurion and, at the background, the RNlN. The major part of this research project was financed by NOVEM, a semi-governmental organisation that supports developments towards durable energy and environmental sustainability. The RNlN was interested in testing the laboratory model of the battery developed in the NOVEM project whether it would be suitable for its own naval application. The testing was carried out by TNO-PML, the in-house laboratory of the RNlN. TNO-AT at the same time could, through mediation by TNO-EEP, profit from the knowledge produced in the naval project, as TNO-EEP provided TNO-AT with documents about the naval version of the battery. 3.2.4. Period after 2000 By the end of the year 2000, the prototype of the battery proved to be technically feasible and the prospects for applications looked promising, though this was more evident for the civilian than the military application. For the military version only functional but no detailed technical specifications were known, like those for the civilian application. By the end of 2000, the WEAG was still discussing future naval requirements for energy storage and what would be the best technical type of solution (advanced batteries,

966

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

flywheels or superconducting magnetic energy storage systems). The lagging behind of the degree of detail in the technical specifications for the military applications point at important differences in the life cycles of civilian and military product development. The specifications for commercial products, like an electric car, are articulated more rapidly than for military systems. The design, procurement and production of, for instance, warships may take many years (5–15 years). Innovations in military systems often take large technological steps, whereas civilian innovations usually occur in small steps.19 Whereas the Navy considers application of the battery in new vessels not before 2005, the first HEVs have already appeared on the road. Accordingly, TNO-AT is under a much higher time pressure than, for instance, TNO-PML and the RNlN. As of the year 2000, still further developmental activities were necessary, but also production aspects of the battery had already to be taken into account. At this stage, the RNlN was no longer willing to provide financial support because of the Navy’s policy that product development, in contrast to technology development, should be done by industry itself. The only route of using Dutch defence funds, therefore, went via a project under auspices of the Western European Armaments Group (WEAG). In fact, it became the most important route for further development of the battery, next to a project funded by the European Fifth Framework Programme (EFFP). Institutional features, however, of both the WEAG and the EFFP prohibited a combination of the different projects into one project. The WEAG concentrates on military applications and the EFFP provides only funding for civilian applications. Whereas, this might result in separate projects, the dedicated network builders, that is, the RNlN and TNO-EEP, still strived for overlapping industrial consortia in the different projects, which would facilitate technology transfer. Even the possibility of one consortium that produces different batteries for the civilian and military sector should not be excluded. Apparently, TNO-EEP and the RNlN continued to strive for the continuation of the network and a strategy of concurrent development. 19 This is reflected in the language used: military engineers often talk about developing ‘technology’, where civilian engineers talk about making ‘products’.

4. Discussion 4.1. Duality as a window of opportunity The notion of the battery’s potential duality emerged only at a later stage of the battery’s developmental process, related to a generally increased interest in hybrid electric vehicles. Together with the involvement of the ‘dual actor’ TNO-EEP it appeared as a window of opportunity for the involved actors to pursue a co-operation strategy of joint development of a bipolar battery for both civilian and military applications. However, in the end no durable network with a civilian–military integrated project for a joint development arose. Instead, what remained was a rather loose network whose actors were dispersed over a number of concurrent development projects with mutual interactions. The dual oriented actor TNO-EEP appeared as a favourable factor for the establishment of a joint development strategy. Although initially asked merely to solve a technical problem, it subsequently became a dedicated network builder. TNO-EEP participated in both the naval project and the civilian HEV project financed by NOVEM, and, consequently, could serve as a gateway for the transfer of knowledge between the naval and the commercial HEV project, thus diminishing transaction costs. As a dual actor, TNO-EEP was interested in both a civilian and military application of the battery. TNO-EEP was eager to develop a product so that it could profit from their patent of a crucial part of the battery (Section 3). By contrast, the RNlN, though acting as a dedicated network builder, was, in the end, primarily and mainly interested in a ‘naval battery’. Moreover when the project was to enter the phase of a prototype ready for production, the RNlN, according to its funding policy, had to terminate further financial support. From a technological point of view, a strategy of ‘joint development’ would have been quite natural, because of the large commonalities of the battery requirements for the civilian and military applications. Moreover, the involved actors estimated that the remaining differences would not prove to be a major obstacle for such a joint development. Even typical naval requirements like shock resistance and vibration requirements would not create large problems and could be solved by additional measures. Why then did

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

the development divide up into ‘concurrent’ military and (two) civilian projects (though with some overlap in actors, notably TNO-EEP and the battery company Centurion)? We have identified three major bottlenecks for the creation of a single project for the joint development of the battery for both civilian and military applications. One reason originates from the different institutional settings in which the WEAG and the EFFP operate, the former being oriented towards military, the latter towards civilian applications. The EFFP actually excludes research on military applications, which prevents the creation of one single large project including military research. A second reason lays in the different technological development life cycles of the civilian and military applications. For instance, civilian requirements are articulated much quicker than military requirements, which may take many years. This is a clear bottleneck when in a collaboration actors need to fine-tune the different requirements. The third reason is the lack of an actor with appropriate funds for whom the creation of an integrated civilian–military technological and industrial base is a major goal. Except for TNO-EEP, which is actively involved in all the three projects, to all other actors duality served as a convenient vehicle of interaction for realising their own goals, that is, the realisation of a battery according to their own requirements. They, therefore, organised their actions and interactions towards duality such and so long as they could benefit from the duality quality and the availability of funding and expertise. Although the RNlN responded to the notion of the battery’s duality, for instance, by providing TNO-AT with documents, in the end they were only interested in a product which could meet their own requirements. The network thus lacked an overarching binding third party which could keep together the different actors and projects.20 The evolution of the socio-technical battery network, as described above, was clearly a ‘bottom up’ process, in contrast to, for instance, the US ‘top down’ policy of 20 One candidate to serve as such binding third party might be a public actor, like the Ministry of Economic Affairs, that has close relations with both the military and the civilian domain and might stimulate co-operation in the field of dual use products and technology.

967

supplying funding for dual technology development, underlying the Technology Reinvestment Program.21 4.2. Comparison of the findings of the battery case with previous studies Our case study confirms the observation of Cowan and Foray of decreasing opportunities for dual use benefits in later phases of the development life cycle of a product, when divergences in requirements for the civilian and military domain become more important. In our case, however, this is not the most important reason for blocking the route of a joint technology development. Rather, it is the difference in life cycles between technology development in civilian and military contexts that make it difficult to fine tune collaboration. But our study confirms the Cowan and Foray’s observation that military R&D may benefit the civilian sector through the knowledge it produces and the creation of an information infrastructure: in our case a socio-technical network with relevance for knowledge transfer had already evolved before the battery’s possible civilian application became an issue. This relates to another point made by Cowan and Foray, that is, where to locate the dual, military or civilian character of a technology. We have shown that initially the battery was viewed as a military technology. The notion of the battery’s potential duality emerged only at a later stage of the battery’s developmental process, in the interactions within the network and related to a change in context, notably the generally increased interest in hybrid electric vehicles. It illustrates Cowan and Foray’s point of “[. . . ] duality is seen as a relation that sits not in the technology itself but rather in the networks in which a technology is designed and used” (Cowan and Foray, 1995, p. 852). 21 A review of the TRP by the Potomac Institute for Policy Studies (1999) concluded that the majority of the TRP technology projects were highly successful, in particular as to the degree that the military services benefited from these projects. The review is less clear about the TRP’s long term impact on establishing an integrated civilian–military technological and industrial base. In line with footnote 3, where we took up the suggestions by one of the anonymous referees, that after 1997 TRP has actually become a military acquisition policy rather than a dual use policy, one should conclude that the original ‘top–down’ policy aimed for by the Clinton Administration has, since 1997, been replaced by a ‘no policy’ for dual use.

968

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

Furthermore, Cowan and Foray emphasise the importance of organisational requirements to benefit from a potential dual use quality, for example, through a mixed organisation and by information disclosure: “The potential benefits [. . . ] can only be realised if certain organisational and informational requirements are fulfilled; that is, if the social networks have such a configuration that they link the R&D programmes undertaken in the civilian and military domains” (Cowan and Foray, 1995, p. 863). Though confirming this observation, our case study adds that the window of opportunity of socio-technical duality should be backed up by funds (and expertise) to benefit from this opportunity. Molas-Gallart (1997) has distinguished four main technology transfer categories. One important feature we add is the facilitating role that ‘dual actors’ may play in ‘external adaptation’ transfers, where technology is to be transferred from military to civilian projects or vice versa. In our case, the dual actor TNO-EEP played such a role by performing a gateway function for the transfer of knowledge between the (concurrent) naval and the commercial HEV project, thus diminishing transaction costs. Similarly, dual actors may facilitate collaboration through joint development projects, though, for other reasons, these did not occur in the battery case. 4.3. Dual capacity networks The discussion above still leaves the question whether a strategy aiming for an integrated civilian– military technology and industrial base through joint

development projects, (that is, projects in which civilian and military actors closely co-operate within one project for developing dual use technology, having both civilian and military applications) is a viable strategy in general. From both our case study and the article of MolasGallart, it appears that such a strategy may be difficult to realise. The availability of research funding, like in the US Technology Reinvestment Project, seems a basic condition. If such funding is lacking, it is hard to say a priori which forms of collaboration are appropriate to propel the innovation of dual use products (see for instance Chiesa et al. (2000) on different organisational modes to access external sources of technology). Recently, Klein (2001) in a systematic study on the compensation of downturns in defence by civilian technology projects with the involvement of actors from the defence sector, showed that the resulting “technology push-over” might even have little civilian utility (though it might be seen as good defence policy). The various collaboration strategies followed by the actors in this case study are not unusual (Cagliano et al., 2000). These authors point out that, when the technology matures and reaches a later phase in its life cycle, the characteristics of the organisational form of the collaboration also evolve. For each phase, an optimal ‘mode’ exists. Their sample of case studies shows, for instance, that a high number of partners may be optimal during the research phase (e.g. a large network), but not during the development stage of the product. This reasoning favours a strategy of concurrent rather than joint co-operation projects, at least at a later stage of technology development.

Table 2 Elements in constructing and stabilising a ‘dual capacity network’ To construct and enable a dual capacity network

To continue and stabilise the network

Enable access to research information Envisage applications for the (embryonic) technology in both military and civilian contexts Emphasise the generic aspects of the technology Involve partners from the military and civilian context

Use the dual oriented actors as gateways for technology transfer When possible, use the dual actors to construct a mixed consortium of military and civilian actors Strive towards common production of the (different) product(s) Establish different kinds of partnerships amongst the actors of the network Use the dual actors as internal co-ordination and communication nodes of the network

Involve especially dual oriented companies and institutes with an ability to couple different socio-technical contexts Look for military, civilian and mixed funding possibilities for further developing the technology

Look for military, civilian and mixed funding possibilities for further developing the technology

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

What suggestions emerge from our empirical study for projects to benefit from socio-technical duality (that is, duality both in network and technology)? The main point is that the participating actors appreciated the various evolving partnerships. The duality of both network and technology, as it emerged, was rather a means than an end to be achieved. Whether a strategy of joint or concurrent development was pursued or not, the network itself was seen as valuable. Future projects could benefit from the findings in this study as to the establishment of a network that can meet the needs and requirements of different parties as to technology transfer and (future) collaboration. Suggestions for establishing and stabilising such ‘dual capacity networks’ are summarised in Table 2. The establishment of such ‘dual capacity networks’, might be part of a possible strategy towards an integrated civilian–military technology and industrial base. At the same time, our analysis demonstrates that even such strategies may have their limitations in realising joint development projects.

Acknowledgements We thank Jorrit Mellema, Nil Disco, Ben van der Ploeg for their helpful comments in preparing this article. We also acknowledge the constructive comments of three anonymous referees. We wish to thank the interviewees for their willingness to provide us with the necessary information to conduct this study.

References Alic et al., 1992. Beyond Spinoff. Military and Commercial Technologies in a Changing World. Harvard Business School Press, Boston. Altmann, J., Liebert, W., Neuneck, G., Scheffran, J., 1998. Dualuse and conversion of military related R&D in Germany, In: Reppy, J. (Ed.), Conversion of Military R&D. Macmillan, London. Bertsch, G.K. (Ed.), 1988, Controlling East West Trade and Technology Transfer. Duke University Press, Durham. Cagliano, R., Manzini, V., Chiesa, R., 2000. Differences and similarities in managing technological collaborations in research. Journal of Engineering and Technology Management 17, 193–224. Chiesa, V., Manzini, R., Tecilla, F., 2000. Selecting sourcing strategies for technological innovation: an empirical case study.

969

International Journal of Operations and Production Management 20 (9), 1017–1037. Cowan, R., Foray, D., 1995. Quandaries in the economics of dual technologies and spillovers from military to civilian research and development. Research Policy 24, 851–868. Elzen, B., Enserink, B., Smit, W.A., 1996. Socio-technical networks: how a technology studies approach may help to solve problems related to technical change. Social Studies of Science 26 (1), 95–141. Davies, D., 1979. Peaceful application of nuclear explosion. In: Nuclear Energy and Nuclear Weapon Proliferation. The Stockholm International Peace Research Institute (SIPRI), Taylor & Francis Ltd., London, pp. 293–305. Enserink, B., 1993. Influencing military technological innovation. Socio-technical networks and the development of the supersonic bomber. Eburon, Delft. Goldblat, J., 1982. Agreements for Arms Control: A Critical Survey. The Stockholm International Peace Research Institute (SIPRI), Taylor & Francis Ltd., London. Gummett, P., (Ed.), 1991. Future Relations Between Defence and Civil Science and Technology, SPSG Review Paper no. 2. A report prepared for the Parliamentary Office of Science and Technology, prepared by ESCR/SPSG Defence Science and Technology Policy team, London. Gummett, P., Reppy, J. (Eds.), 1988. The Relations Between Defence and Civil Technologies. In: NATO ASI Series. Kluwer Academic Publishers, Dordrecht. Gummett, P., Stein, J.A., (Eds.), 1997. European Defence Technology in Transition. Harwood, Amsterdam. Hansen, C., 1988. US nuclear weapons: the secret history. Aerofax Inc., Arlington, TX and Orion Books, New York. Klein, H., 2001. Technology push-over: defence downturns and civilian technology policy. Research Policy 30, 937–951. Kolkert, W.J., 1989. The pulse physics research programme of The Netherlands Organisation for applied research TNO. IEEE Transactions on Magnetics 25 (1), 306–308. MacKenzie, D., 1990. Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance. MIT Press, Cambridge, MA. Markusen, A.R., Costigan, S.S., 1999. Arming the Future. A Defence Industry for the 21st Century. Council on Foreign Relations Press, New York. Milne, T., 1998. Conversion of R&D: the UK case. In: Reppy, J. (Ed.), Conversion of Military R&D. Macmillan, London. Molas-Gallart, J., 1997. Which way to go? Defence technology and the diversity of dual-use technology transfer. Research Policy 26, 367–385. OTA, US Congress Office of Technology Assessment, 1992a. After the Cold War: living with lower defence spending, OTA-ITE-524. US Government Printing Office, Washington, DC. OTA, US Congress Office of Technology Assessment, 1992b. Building future security: strategies for restructuring the defence technology and industrial base, OTA-ISG-530. US Government Printing Office, Washington, DC. OTA, US Congress Office of Technology Assessment, 1993. Defence conversion: redirecting R&D, OTA-ITE-552. US Government Printing Office, Washington, DC.

970

H. te Kulve, W.A. Smit / Research Policy 32 (2003) 955–970

OTA, US Congress Office of Technology Assessment, 1994. Assessing the potential for civil military integration: technologies, processes and practices, OTA-ISS-611. US Government Printing Office, Washington, DC. OTA, US Congress Office of Technology Assessment, 1995. Assessing the potential for civil–military integration: selected case studies, OTA-BP-ISS-158. US Government Printing Office, Washington, DC. Pinsky, N., Grosvenor, V.L., 2000. Bipolar lead–acid battery plates and method of making same. US Patent US6077623. Potomac Institute for Policy Studies, 1999. A Review of the Technological Reinvestment Project, PIPS-99-1. Arlington, VA. Reppy, J. (Ed.) 1998. Conversion of Military R&D. Macmillan, London. Reppy, J., 1999. Dual-use technology: back to the future? In: Markusen, A.R., Costigan, S.S. (Eds.), Arming the Future: A Defence Industry for the 21st Century. Council on Foreign Relations Press, New York, pp. 269–284. Saakes, M., Schellevis, D., van Trier, D., Wollersheim, M., 1996. Performance and use of composite-substrate-based bipolar lead/acid batteries for pulsed-power publications. In:

Proceedings of the Paper Presentation at the 5th European Lead Battery Conference, Barcelona, 30 September–4 October. Saakes, M., Schellevis, D., van Trier, D., Wollersheim, M., 1997. Performance and use of composite-substrate-based bipolar lead/acid batteries for pulsed power applications. Journal of Power Sources 67, 33–41. Saakes, M., Kluiters, E., Schmal, D., Mourad, S., ten Have, P.T.J.H., 1999. Development and testing of a bipolar lead–acid battery for hybrid electric vehicles. Journal of Power Sources 78, 199–203. Serfati, C., 1997. France. In: Gummett, P., Stein, J.A. (Eds.), European Defence Technology in Transition. Harwood, Amsterdam. Smit, W.A., 1995. Science, technology and the military: relations in transition. In: Jasanoff, S., Markle, G.E., Peterson, J.C., et al. (Eds.), Handbook of Science and Technology Studies. Sage, London, pp. 598–626. Smit, W.A., Elzen, B., Enserink, B., 1998. Coordination in military socio-technical networks: military needs, requirements and guiding principles. In: Disco, C., van der Meulen, B. (Eds.), Getting New Technologies Together. Studies in Making Socio-technical Order. Walter de Gruyter, Berlin, pp. 71–105.