Science & Technology Studies 2/2014

Framing Prototypes: The Fast in (1950s–1990s)

Arthur Jobert and Claire Le Renard1

This paper considers a crucial moment in the innovation process: the shift from a research phase to an industrial phase. The empirical study examines the development in France of Fast Breeder Reactor technology (FBR), from the 1950s to the early closure of the Superphénix plant in 1997. A turning point occurred in the late 1960s, when several European countries judged that the FBR technology was a promising electricity generation technology that would soon be mature for commercialisation, in a context of technological nationalism and future energy scarcity. In this paper, we analyse how the framing of the resulting prototype as “industrial” entailed an impact on decisions during the three decades that the project lasted. Aiming at describing the project actors in action without judging their decision-making processes, we use the ‘framing’ concept preferably to other approaches such as ‘path dependency’. This concept choice is the subject of the discussion.

Keywords: nuclear technology, framing, prototypes

Introduction on empirical work dealing with the French Superphénix, this paper discusses how From a STS perspective, the process of framing a prototype as industrial is not innovation is a temporal one, uncertain and only a matter of rhetoric; it may have an contingent; it is driven by actors who work important impact on the trajectory of an to fi nd a place for their innovation in a social innovation. ‘Framing’ will be used here as and economic context which might evolve. “a notion which grabs the perceptual lenses, A crucial moment in this process is the worldviews or underlying assumptions shift from a research phase to an industrial that guide communal interpretation and phase. To bring their projects to the defi nition of particular issues” (Miller, 2000: industrial phase, innovators have put their 212). hopes in hybrid objects who must on the Th e Superphénix was the industrial one hand, demonstrate the maturity of their prototype of the technology of Fast Breeder technology but, on the other hand, can still Reactors (FBRs); using neutrons in a “fast” be improved before they enter the market. regime, this specifi c nuclear technology Th ese hybrid objects are given ambivalent was able to “breed” or regenerate fuel while names such as ‘pilot-series’, ‘industrial using it. From the 1950s until the 1970s FBRs demonstrator’, ‘industrial prototype’. Based were being developed in many countries, in

Science & Technology Studies, Vol. 27 (2014) No. 2, 7-26 7 Science & Technology Studies 2/2014 the expectation that they would provide a literature contributes to the controversy, nearly inexhaustible source of electricity, some authors highlighting the “failure needed to fuel the rapid economic growth of FBRs programmes” (see Finon, 1989), of the post-war years. Th e development of a other authors in the contrary envisioning fl eet of commercial fast breeder reactors was the irreplaceable role of FBRs in the future therefore considered by many as the logical energy supply system and pleading for the end-point of a viable nuclear programme. continuation of the Superphénix (Vendryès, However, the use of neutrons in a “fast” 1997). regime meant using a molten metal as a Th e fi rst methodological pitfall of coolant. Sodium was chosen because of researching causes for the early shutdown its thermal conductivity; it is nevertheless terminating the innovation trajectory of known for its reactivity with water and Superphénix would be to explain it by oxygen. With such features, Fast Breeder the very beginning. In Spring 1998, after technology was to become an object of the decision to permanently close the international competition, technological Superphénix was taken by the government, development, visions, and risk debate. the French Parliament conducted hearings, allowing the concerned parties to express Approaching a controversial project: Our their controversial views. Th e report methodological choices following these hearings stressed that the Th e history of Fast Breeder technology in causes of the premature end of the plant France, and of Superphénix in particular were to be found at the very beginning of – a reactor which was stopped earlier than the industrial prototype: “the decision of planned – is a controversial one, and took creation was taken without transparency, place over time. Conducting a research basing on alarming forecasts, for a plant devoted to such an innovation supposes whose role appeared in the end to be taking methodological precautions. Th ere is fl uctuating over time” (Bataille, 1998). then the considerable risk of taking side in Building an explanation on this form of the controversy or writing history backwards evaluation puts us at a major risk of history (presenting the failure as predictable or being written by the victors. As Rip and even inevitable). As an answer, we want to Kemp (1998) write, we cannot analyse state that a methodological stance designed the trajectory of an innovation as if: “the to avoid both these pitfalls enabled us to add direction of technological development new perspectives to the existing research. was determined by the actual paths and Much has been written on this project, the expectations of what could be next be it in the 1970s before and during its steps [...]. Our retrospective idea of steps construction, during the operation years in the direction of the situation as we know from 1985 to 1997, or afterwards, when is irrelevant”. Th erefore, we tried to avoid diverse accounts of the project tried to rereading the history of the technology on record its history and the lessons learnt. the basis of its developments which were A great variety of primary and secondary known to the researcher but unpredictable sources can thus be found in media for the actors in the on-going project, and coverage, in “grey” literature (expert reports, we aimed at depicting how the Superphénix parliamentary hearings...) and in academic was framed as an “industrial prototype”, and or para-academic publications, a selection what this specifi c feature – being industrial – of which can be found in the references meant for such a prototype, associated with section of this paper. Th e main part of this solid expectations and new constraints.

8 Arthur Jobert and Claire Le Renard

In this respect, Bruno Latour’s seminal the past in terms of the present or of a form Aramis or the Love of Technology (1996) of predestination but rather of grasping the was of great importance to our work. Th is innovators in action. Our methodological book traces the history of a public transport choice was thus to carry out a study of project called Aramis which was intended to the actors involved in a Fast Breeder serve the south of Paris with the combined technology project and to try to understand advantages of rail transport and individual the uncertain process during which the cars, but which never reached the innovators defend their project and make commercial stage. Above and beyond a case their choices. It was notably a question of study, this work off ers lessons on the factors identifying the moments of choice when for success or failure for such innovative the actors had to decide how to modify their projects, along with a methodological project to meet the requirements of the time stance from which to talk about the past and of the strategy they were following. from the point of view of the researcher’s Th e oral sources for this work were thirty- situation in the present. three interviews of twenty-fi ve project To avoid the pitfalls evoked above, Latour actors, two experts from the nuclear safety (1996: 6) suggests “going to see everybody authority’s Technical Support Organisation, who’s being criticized and blamed” by and six opponents or critics of the technology applying a methodological principle of (experts and scientists). Among the twenty- benevolence: the sociological standpoint fi ve interviewees who had directly worked consists in putting oneself in the place of on the Superphénix project, approximately the actors of the project, with their own 20 % had designed the Superphénix and/ representations of the future. Th e narrator or the French Fast Breeder technology talks to his (fi ctitious) student as follows: development, 45 % had engineered or built the plant, and one third had operated it. We Always assume that people are right, met scientists and engineers from the CEA even if you have to stretch the point a public research agency (Commissariat à bit. […] otherwise, you play the sly one at l’Energie Atomique), who developed the the expense of history. You play the wise demonstrators or who were the decision- old owl. […] Life is a state of uncertainty makers responsible for the programme and risk, of fragile adaptation to a past as a whole. We interviewed the managers and present environment that future in charge of project conception and cannot judge. (Latour, 1996: 35-37) construction, be it at the CEA, in the engineering company or in the EDF Th is obligation to show goodwill is one of (Electricité De France) plant design and the features of the method used during this construction division. Th e Superphénix analysis of Fast Breeder technology. Another construction manager gave us a colourful feature of the research was the quest for an account of events. With regard to the inside view of events, and the fi eld enquiry period of operation, we met members of lead us to meet the people who had worked the plant’s board of directors and members on the project. of NERSA’s board of directors (Centrale nucléaire européenne à NEutrons Rapides SA, An investigation focussing on how project the European project company created for promoters frame their project Superphénix). For Latour, and numerous researchers after Th e following fi gure illustrates the him, it is thus not a question of rereading diversity of the career profi le of our

9 Science & Technology Studies 2/2014

Whole career at EDF

Concerned scientists and activists

Career at CEA and subsidiaries

Career at NERSA & EDF

Italian ENEL

Technical Safety Organisation

Career at CEA & EDF & NERSA

0 2 4 6 8 101214

Figure 1. Position and career profi le of the interviewees. interviewees: this project brought together cross-checked our interviews with written people coming from diff erent professional sources where other stakeholders expressed backgrounds and cultures. Some of the their views at the time of the debates project actors were involved in the fi rst (e.g. minutes of public hearings, record steps of FBR technology, before reorienting of TV debate, press articles). Th ere is no their careers towards other areas of nuclear guarantee that such a process prevented us power; others were involved from start to from being infl uenced by our interviewees. fi nish, devoting their entire careers to the However, this cross-checking enabled us to development of the technology; fi nally, identify critical periods, such as the debate some worked on the Superphénix for a few about the re-defi nition of the project in the years, starting and pursuing their careers in 1990s, which will be one of the topics of this other nuclear projects. paper. In fact, in a fi rst set of interviews, the As 76 % of our interviewees were actors innovators tended to downplay or not evoke of the project, we are conscious that such a spontaneously this debate. dissymmetric approach of a controversial Th is research actually made us conscious project can be seen as biased. One answer that within such an ambitious project lays in the visibility of the literature critical diff erent opinions existed. Some actors of the project, which we carefully studied who had the feeling their point of view had before beginning the fi eld enquiry. Some not been considered enough were glad to activist organisations keep the memory share their opinions in the interviews about of Superphénix alive, publishing their events that happened decades ago, as well archives, press articles, and argumentations as the ones whose views had been retained online. We met some activists or concerned for decision-making. Some of them even scientists to record their views on the events had kept an impressive documentation and developments. But furthermore, we as a personal archive, in their garage or in

10 Arthur Jobert and Claire Le Renard a devoted offi ce, which they off ered us to The “Irresistible” and “Irreversible” use – this situation can be related to what Framing of an Industrial Prototype Gabrielle Hecht experienced in her research about nuclear developments in France Th e history of Fast Breeder technology in after World War II (Hecht, 2009: 18). As France can be better understood by a focus well as her, we can state that “most people on its framing – in the words of Jasanoff seemed eager to share their memories, (2005), “a conceptual language that can look for documents, and put [us] in touch grapple with both continuity and change, with others who might help”. And we also while rejecting some of the rigidities of experienced that “some things conveyed structure – in order to understand how policy in the interviews are not in any document, domains are carved out from the political accessible or not”. sphere and rendered both comprehensible A last answer lays in the very objective of and manageable”. In this section, we want this research, on which we build the point to depict how in the 1950s and 1960s the we want to make in this article. Our objective framing of Fast Breeder technology (FBR) as is not to conduct an evaluation, or to judge necessary in the near future resulted in the the decision of these actors on a normative design of a “European industrial prototype” basis as, in some respect, does the “lock-in” which was supposed to accelerate access to approach which implies that an ineffi cient a commercial stage. technology or product “captures” markets at the expense of better products (Arthur, 1989; Th e making of FBRs as the obligatory Cowan, 1990; Pierson, 2000). Our objective passage point here is to better understand the rationale of In the context of the post-war years, as rapid actors involved in a controversial innovation economic growth entailed a rising energy project. More precisely we will put forward demand, FBR technology was framed as the how the framing of their prototype as logical end-point of any nuclear programme, “industrial” in the late 60s entailed an carrying in it the promise of inexhaustible impact on decisions during the 3 decades energy. Th is assertion requires a little detour that the Superphénix project lasted. in nuclear physics, which we want to make Th erefore, we will fi rst describe the as simple as possible. defi nition of the features of Superphénix Natural is composed of as the “construction of a long chain of 99.3 % Uranium 238 isotope (238U) and 0.7 % reasons that are irresistible” (Latour, 1996: Uranium 235 isotope (235U). In the post-war 33). Th e following two parts will depict the years, several nuclear technologies were decisions of the actors at two moments of developed, among which the technology trials, when the irreversibility induced by of Pressurised Water Reactors (PWRs), the “industrial” feature of the prototype using the scarce 235U, and the FBRs, using made it diffi cult to renegotiate the project. abundant 238U. Th e PWR technology had Th e last part is a broader discussion of been adopted on American submarines for framing and irreversibility in such prototype its compactness, and had experienced more developments. operation hours than any other nuclear technology: for this and other reasons, they were chosen as the main component of a nuclear industrial fl eet in the US (Cowan, 1990), and from the 1960s on, in several European countries. Th e development of

11 Science & Technology Studies 2/2014 this technology appeared then deemed to a FBR development as a national project: brilliant future, raising concerns about 235U Th e irresistible alliance fuel scarcity that it might occasion in the FBRs were considered a strategic medium-term. technology, and from the 1950s onwards, Th e FBRs were then promoted as an research reactors of increasing size were answer to this concern: fi rstly, using developed in the United Kingdom and neutrons in a “fast” regime, they could in the , stimulating eff orts generate energy from the fi ssion of the designed to establish and demonstrate the abundant isotope 238U. Using therefore feasibility of the technology. In the mid- the energy potential included in natural 50s, in an attempt to make up for lost time, uranium approximately a hundred times France begun its fi rst studies (Vendryès, better, FBR technology stood above and 1997). Impetus was provided by a study beyond the PWR technology in the eyes visit by two CEA engineers to the USA: of scientists and engineers. Secondly, this won over by this technology, upon their technology can use the (239Pu), return they persuaded their hierarchy to an artifi cial element, as a fuel. Th e highly grant them suffi cient funding to build an radioactive Plutonium is generated during experimental reactor in France; it was to be nuclear reactions in FBRs and other nuclear called and reached criticality in reactors, when a nucleus of 238U absorbs 1967 (Vendryès, 1997). Despite having been a proton released during the reaction. completed four years behind schedule, Th irdly, if a fuel reprocessing plants extracts Rapsodie attained full power in just three fi ssionable fuel from the used one, the FBRs months and was regarded as a “technical can reuse their fuel several times, thus success” (Finon, 1989: 159). At the end of achieving “breeding” or fuel regeneration the 1960s, research reactors also reached with a Uranium-Plutonium cycle, making criticality in the USSR and Germany. Th e the energy potential close to infi nite. promise of abundant and inexpensive In the 1960s, this promise of inexhaustible energy fostered technological developments energy was the horizon of nuclear in numerous areas in order to establish development in several industrialised the feasibility of the FBR technology. Th e countries. In the terms of Latour (1996: 33), competition between countries regarding we can state that FBRs were then regarded technological achievements served “as the obligatory passage point that will nationalistic purposes, and became a driver resolve the great problems of the age”, thanks as well as a consequence of technology to one of “these long chains of reasons development. Conferences and academic/ that are irresistible”. Th e rationale was professional publications were arenas for the following: economic growth requires international competition, as well as for abundant electricity; although the PWR the circulation of ideas, helping to create technology is retained as an immediate, a common mindset among the experts transition technology, it remains a provisory involved (Goldschmidt, 1967). answer, which uses the energy potential in At the same time, the exponential growth natural uranium rapidly and poorly, raising in energy requirements in the 1960s saw concerns of fuel depletion; therefore, FBR several countries equip themselves with technology must be developed and tend industrial . In France, as well towards an industrial maturity as soon as as in other European countries, a dispute possible. took place between the advocates of the “national” reactor design and the promoters

12 Arthur Jobert and Claire Le Renard of the American PWR. Beyond technologies, two weeks ahead of schedule, the Financial this dispute opposed arguments centred Times entitled an article “French world lead on national technological excellence vs. in fast reactor technology” (Sauvage, 2009). inexpensive electricity generation (Hecht, Combining the stakes of future energy 2009). As they featured more operating supply with the achievement of nationalistic experience as well as lower projected “grandeur”, FBR development became a generation costs, PWRs were retained for privileged cooperation fi eld for the CEA and the industrial fl eet in the short term. In EDF: D. Finon depicts it as the “irresistible the late 60s, while the interests and views logics of an EDF-CEA alliance” from 1970 on of the key actors in the French nuclear (Finon, 1989: 169). “establishment” (especially between EDF and the CEA) diverged on many issues, A project made more irreversible by its “the breeder reactor emerged as a source of European features consensus” (Hecht, 2009: 291). On the one Latour (1996: 154) states that “technological hand, building on the experience acquired projects become reversible or irreversible in with national prototypes, it allowed the relation to the work of contextualisation”. By pursuit of national technological excellence the beginning of the 1970s, atomic energy – as Hecht (2009: 293) notes, “they agencies in several European countries (e.g. transferred the burden of French grandeur in UK, France and Germany) envisaged to the breeders”. On the other hand, the reactors of a capacity around 1000 MW. objective to produce cheap and abundant Th ese full-scale reactor projects anticipated electricity would be met by the choice of a future series reactor design which would American technology in the short term, and have to be both industrial (powerful and in the medium and long term by the “logic reliable) and commercial (able to equip the of a breeder future” (Hecht, 2009: 293). national fl eet and to be exportable). Th e FBRs became then the only remaining electricity utilities then became key players nationally developed nuclear technology. in the development of such projects, making Fast breeder prototypes were developed as them reversible in some countries (the part of a long-term, national nuclear project, British project was stopped in the late 70s, which would include reprocessing and a see Le Renard et al., 2013), but contributing fl eet of industrial 1000 MW breeder reactors to make the project more irreversible in our (Finon, 1989: 182). In southern France, case study, through the commitments that while the experimental “Rapsodie” reactor the French state took vis-à-vis its foreign was only starting to operate, the design of partners. a 250 MW prototype reactor was already In the 1960s, the European community initiated: it was to pave the way for the to-be and its nuclear research programme industrial FBRs. Named after the bird which Euratom had attempted to foster the rises from its own ashes, the “Phénix” reactor development of a European 300 MW FBR represented FBR technology regenerating prototype, with limited outcome. But its fuel. With its 250 megawatts of electricity, this initiated a European cooperation it provided the same power as the coal- which would succeed in the next step of fi red plants of its time. It reached criticality the programme, the 1000 MW prototype in 1973 and was acclaimed as a technical (Giesen, 1989). Th ree electricity utilities success: France had made up its lost ground (the French EDF, the Italian ENEL and the in FBR technology. On March 15, 1974, as German RWE) had initiated collaboration the Phénix reactor reached nominal power as early as 1970 to envision a common FBR

13 Science & Technology Studies 2/2014 industrial prototype in order to share costs products that may be commercial on a and operating experience. Th e deliberations large scale was added to technological amongst the utilities, the CEA, and the achievement. Bringing together European French government resulted in the 1200 MW partners was therefore a way to share the Superphénix project in the south of France, risks and competitive advantages, as well and its German counterpart, SNR 2, whose as to expand the potential markets. Th e construction was to start shortly after that of agreements to form the European company Superphénix (Marth, 1993). Th is larger size NERSA provided that the electricity was comparable with the 1300 MW PWR generated by the new plant was to be plants developed at that time. In the same returned directly to the countries involved way that the development of Phénix had on a pro rata basis, in line with their levels taken place during the worksite of Rapsodie, of participation: the search for a site with the in the early 70s, the developments of the required physical and geographical qualities Superphénix project were begun in parallel led to the industrial prototype being located to the Phénix worksite, in order to maintain at the centre of the Lyon-Genève-Chambéry engineering skills permanently working triangle (i.e. at a reasonable distance from on the new technology. Beyond its name the Italian and German borders), near the (an ‘extended’ Phénix), several signifi cant village of Creys and the hamlet of Malville. characteristics of the project development As the plans for the creation of the NERSA changed with Superphénix, hence marking were well underway, on 13 December 1972, the shift of FBRs from an experimental to an a parliamentary debate was held concerning industrial era: a bill that established an exception to • Th e owner of the Superphénix project the 1946 law on the nationalisation of was a limited company (NERSA) the electricity sector, and would instead created with equity from several allow the creation of enterprises in the European electricity companies. Th e domain of electricity that would carry out EDF held 51% of the capital, ENEL in France “an activity of European interest”. 33%, while the remaining 16% were Th e creation of European companies was owned by RWE; contested by the trade unions and by • Th e CEA licensed the FBR technology a part of the opposition (especially the to Novatome, an ad hoc subsidiary, communists). Th e critics feared that the which would be able to meet orders entry of private interests in the energy sector for future FBRs on an international would work in the same direction as the scale. choice of the “American” PWR technology. But the project was then promoted as Th is double choice of creating an ad French technology development, as well hoc company that would from the as providing energy independence for the beginning include European partners was nation. Th e fact that the project was also representative of a new way of managing European did not contradict this idea, quite large technological projects. During the the contrary: as was the case in many other same period, the commercial failure of areas of European politics, the project was the triggered the development seen as a “continuation of France through of the “Airbus model” (Muller, 1989). In other means”. Th e project’s legitimacy was the context of increasing competition with assured by the political consensus on its large (especially American) multinational objectives both at the same time national enterprises, the aim to develop industrial

14 Arthur Jobert and Claire Le Renard and European, commercial and of high wiser to build a smaller plant designed for technology. research or development purposes. Th ey However, the term “industrial prototype” developed this argumentation in documents conveyed an ambivalence which was to published by trade-union or political parties endure throughout the project. Th e project (Parti Socialiste, 1978: 35). was industrial because of its ambitions Th e other criticism was radical; it and the way it was organised. It was a concerned the very structure of the promise prototype because its role was to test a new of inexhaustible energy contained within technology – at that time no FBR of that size the development of FBR technology: the had ever been built. Many of the elements regeneration of fuel meant building a huge were innovative, either in terms of size, or fl eet made up of PWRs, FBRs and fuel in terms of the options chosen – some as a reprocessing plants and keeping them all continuation of Phénix and others through running over the very long term. For the the European dimension and the experience decision-makers of the time, facing future of collaborating countries. scarcity of fuel, this was exactly what was Work on the Superphénix industrial needed; for the critics, it was unacceptable. prototype lasted almost a decade, from FBRs regenerate their fuel in the form of 1976 to 1985. Th e project managers had plutonium, which is both reactive and to overcome numerous diffi culties: in toxic, and certain isotopes of which can creating a ‘fi rst in the world’, they were be used for military purposes. Opposing constantly facing new technical challenges, the very principle of this technology, many of them related to handling the huge critics organised demonstrations, the most components of the plant. But these actors important of which took place in 1977 and were sustained by the conviction that they led to the death of one demonstrator. were working on higher objectives and However, although the growth priorities: making a virtually inexhaustible perspectives for energy demand which source of energy available to mankind. had led to the creation of Superphénix Parallel to the construction of the industrial seemed to have properly stabilised, as prototype at Creys-Malville, the engineering from the mid-1970s the contextual aspects teams in Lyon were preparing for the next changed, one after the other: in 1976, stage, that of defi ning the characteristics pluralist commissions including academic of the series of plants based on the experts both in the United Kingdom and Superphénix, so as to be able to rapidly the United States evaluated the need for launch a fl eet. Th ey were also investigating FBR technology and its costs and risks. Th e future sites. gradual drop in energy demand, due to the During this decade of the project, economic slump following the oil crisis, was objectors to the project criticised the beginning to chip away at the urgent nature choices which had been made on two of building an FBR fl eet. In fact, nuclear fronts: criticism of the technology chosen reactor orders in the United States had been and criticism of the industrial option. drastically reduced in the mid-1970s and Criticism of the industrial prototype brought about a major downward revision aspect of Superphénix came from scientists of the growth forecasts for nuclear power and concerned individuals in the nuclear throughout the world. In reports within their sector. Th ey felt that FBR technology had respective countries (Flowers, 1976; Keeny not been suffi ciently tested to be ready for et al., 1977), the evaluation commissions the industrial stage, and that it would be

15 Science & Technology Studies 2/2014 recommended postponing projects for built as an industrial prototype, and the industrial prototypes, because the fast- principle of commitment to the industrial breeder fl eet was no longer envisaged over series envisaged by the CEA and EDF’s the short term. Scientists fed these points plant design and construction division of view to French associations critical of the was validated. Meanwhile, from 1979 on, project. EDF’s general management postponed the Th e objectives of an initial industrial decision to commit to the industrial series demonstration nevertheless remained in order to have one full year of feedback on preponderant in the debates which took the operation of the Superphénix reactor place over this decade. Th e government was (Finon, 1989: 214-218). Th e argumentation a key player in the decision-making, and was rooted on economic assessments the promise behind this energy technology which compared the competitiveness of justifi ed France continuing to develop FBR technology with that of other types of it, as can be seen in a statement made in energy production, the assumptions for the Parliament by the minister of industry in future cost of Uranium being less favourable June 1977: “it would be very dangerous to to FBR than previously. abandon this fast-breeder project due to pressure from a small group of people who Pursuing the Industrial may be well-informed within their own Demonstration at the Cost fi elds, but who in any case have a poor grasp of Technical Flexibility of the national context in which our energy policy is rooted!” (Journal Offi ciel, 2 June Confi rming the industrial dimension of 1977, quoted by Finon, 1989: 202). Superphénix Th e debates on FBR technology organised In 1985, fuel was loaded into Superphénix’s by the Europe 1 radio station and the core. After ten years of construction, the Antenne 2 public television channel in 1980 Superphénix industrial prototype was were a forum for public discussion which fi nally completed and began its industrial confi rmed what was at stake: President operation. To this end, the small project Valéry Giscard d’Estaing declared that “if company NERSA had signed a contract uranium from French soil was fi nally to be with national electricity company EDF. Th e used in fast breeder reactors, in France we Superphénix industrial prototype benefi ted would have energy reserves comparable to from the experience and standardisation those in Saudi Arabia” (Bériot & Villeneuve, of the operational nuclear fl eet, and as 1980). Questioned about the American halt such, personnel would be employed in in FBR development as part of the non- accordance with the standards of EDF’s proliferation policy, the French MP and organisation charts. former Prime Minister P. Messmer stated in In 1986, the electricity generation unit the same debate series: “the United States of the plant was connected to the grid. Yet would prefer it if we did not maintain our that same year, several events took place advance, particularly due to the industrial which were to radically change the way and commercial advantages that it off ers the future of energy and the relative value us” (Bériot & Villeneuve, 1980, quoted by of the diff erent sectors of production were Finon, 1989: 198). envisaged. In France, there was no question of 1986 was the year of the Chernobyl closing off the option that this technology accident, which impacted Superphénix represented: Superphénix was already being in many ways: for the very fi rst time,

16 Arthur Jobert and Claire Le Renard

Chernobyl brought to life the reality of the one which might replace the temporary dangers of a nuclear accident, and, more PWR technology: even if temporarily the broadly, marked entry into the “society of conditions did not appear to be ripe for risk”, according to the eponymous work by the launch of a fast-breeder fl eet, they had Ulrich Beck published a few months later. to continue to develop existing skills so Many countries suspended their nuclear as to be able to use the technology in the programmes, reducing even further the future. At the end of the 1980s, European foundation of the discourse on the depletion countries combined their eff orts to design of uranium which had justifi ed development the EFR, the European Fast Reactor, which of FBR technology. Among these countries would capitalise on the experience gained was Italy, which nevertheless maintained with Superphénix. An article in an IAEA its shareholding in NERSA. Th e actors bulletin which set out the global situation concerned by the risks with Superphénix for developments in 1989 stated: “In Europe, saw their case strengthened by a serious it is now considered that [FBR] plants would sodium fi re in a solar power plant in Almeria begin to replace the decommissioned PWR in Spain, which also occurred in 1986. plants after 2010, in competition with the Lastly, 1986 was the year of the oil counter- then-available advanced PWRs.” (Golan shock, which marked another turning et al., 1989) Th e status of the “industrial point in the way the future of energy was prototype” without any planned series in envisaged: energy seemed to be abundant the short term was becoming diffi cult to and cheap, and the energy-saving measures justify: what was Superphénix a prototype which had been recommended since 1973 for? Did the characteristics of the plant fell into disuse, as did new technological really make it “industrial”? At a moment developments. Long-term concerns relating when these strategic questions were asked, to the Earth’s fi nite resources were pushed an incident occurred on a critical part of the onto the back burner. plant, which lead project managers to make Finally, Superphénix was the precursor a decisive technical choice. for a long-term series at a time when people were no longer interested in the long term: Translating the industrial framing into a First Of A Kind … without a kind, it now had concrete decision to operate as an industrial plant within the In March 1987, a sodium leak occurred in the EDF fl eet, with the objective of providing a fuel storage “cylinder” tank. To understand return on investment and of continuing to the negotiations and the choices made, we demonstrate the technology, for what was need to take a closer look at this technique: now the distant future. Th e Phénix plant and associated Over the lengthy term of the project, reprocessing facility had demonstrated the whilst the context had changed, the possibility to recycle the fuel, and thus, on industrial objectives remained the same: a small scale, to fulfi l the promise of energy they were refl ected in the size of the plant, autonomy inherent in FBR technology, on in its system of multi-country governance the basis of a “short cycle” involving Uranium and in its integration into EDF’s operational and Plutonium (Sauvage, 2009). Th e fl eet. Superphénix fuel cycle was to be the same Such a nuclear project has a time “short cycle” which had been validated with constant of several decades. Th e project’s Phénix, and the relevant technical device engineers remained convinced that they was very similar, implying a fuel storage were working on a technology for the future, cylinder tank. Th e “short cycle” consists in

17 Science & Technology Studies 2/2014 discharging and renewing a fraction of the fuel recovery. Th ese handling activities can fuel contained in the core of the reactor take place during the operation of the plant, (one third or one quarter) during relatively providing fl exibility. brief stoppages. Th e fuel transfer has to take Benefi ting from the experience acquired place within the sodium, preventing the fuel with Phénix, this fuel handling system which was in the sodium to be brought into was optimised in the available space and contact with air or water. When leaving the included in the concrete. Th e sodium leak core, this fuel gives off a very large amount in the fuel storage cylinder tank was as of thermal power; it must thus cool in order unexpected as improbable. Th e choice of the to reach the thermal power designed for the steel nuance used for the tank, which was reprocessing facility, fi ve times lower than diff erent from Phénix, was held liable for its level when leaving the core. the leak. Questions about what repairs were required led to a reopening of discussions As a Novatome document (1981) states: on the purposes of Superphénix, as it was technically very diffi cult to replace the tank “Th e fuel handling system comprises with an identical one. Th e impossibilities of installation and equipment provided for the technique meant that it was necessary • Simultaneous fuel loading and to negotiate and lower the objectives of the unloading by means of two slop- plant, or to come up with a technological ing ramps, leading from a rotating “detour” (Latour, 1996: 215) which transfer lock to the reactor on one would make it possible to remedy the side and to the fuel storage [cylin- insuffi ciencies. der tank] on the other, • Storage of spent fuel in a sodium- fi lled decay [cylinder] tank before being sent to the reprocessing plant. Th e decay heat is removed by two independent sodium circuits connected to air-coolers.”

In the same Novatome document, the following cutaway view shows the rotative device or “carrousel” (10) inside the fuel storage cylinder tank (9), which enables the operator to select and handle any fuel subassembly. Th is possibility to separately handle the subassemblies is useful for purposes such as research on assemblies or

Picture 1. A cutaway view of the fuel storage cylinder tank, reactor vessel, and fuel handling system (Novatome, 1981). Th e arrow indicates the fi gures on the upper part of the reactor vessel, allowing one to imagine the true dimensions of the plant.

18 Arthur Jobert and Claire Le Renard

Th e decision to replace the defective Th is moment of opening-up and tank with an identical one was rejected: discussing the technical choices to be according to some interviewees, it was made whilst facing increased constraints technically impossible or excessively led to a confi rmation of the industrial expensive; according to others, it would nature of Superphénix and of its mission have required a very lengthy stoppage. to produce electricity on a large scale But due to its industrial framing, the plant as part of the operational nuclear fl eet. was expected to generate electricity for the Questions concerning replacement of the partners of the project, and not to be offl ine fuel storage tank were addressed by internal for a long time, as a non-fi nalised prototype project decisions or by interaction between might have been. technical experts to decide what type of Th e storage tank was replaced by a fuel work had to be done. Th e safety of the transfer unit which fulfi lled certain fuel plant had been controlled throughout the handling functions, but not its cooling: process, and the technical options had been the fuel then had to cool down within the discussed with the Ministry Control Service reactor core itself. Th is implied an operating for Safety of Nuclear Installations (SCSIN) mode known as “long cycle”, where the and its Technical Support Organisation. reactor must remain stopped for six months for cooling, in order for the used fuel to A Diffi cult Reframing towards be discharged and for the new fuel to be Research: The Weight of Irreversibility loaded. It was no longer possible to renew Induced by the Industrial Framing? just parts of the core, and the “long cycle” meant that an entire core had to be burned Th e plant restarted in April 1989, but during each cycle. another incident occurred one year later: Th e choices made during this period were in July 1990, pollution or oxidation of the a consequence of the industrial framing of primary sodium was detected and led to Superphénix, and they “in-scripted” it even stoppage of the plant. Th is pollution was due more in the technology: the fuel storage to air entering the argon circuit2 through a tank was not the only thing to be dropped. defective membrane in an auxiliary circuit. Th e project abandoned a certain fl exibility Th e stoppage lasted for four years, with the of operation, characteristic of research plant only receiving authorisation to restart plants and prohibited by the new system; it in August 1994. also abandoned the idea that the prototype Th is unforeseeable stoppage gave rise should perfectly refl ect the future series, to a period of intense controversy. Among feeling it to have been pushed back into multiple subjects of concern, the main the long term. Th e use of the fuel transfer issues of the controversy were the safety of unit lengthened Superphénix’s operation the plant and the objectives of this industrial cycles and made them less representative prototype. Both were publicly discussed of a future fl eet: “for Superphénix it wasn’t in offi cial arenas, which had been created very serious, but for an industrial fl eet it in the 1980s after changes in the political would not have been viable”, explained one majority. Th e two issues were closely linked, of the actors. In his view, in 1988, within a but for the purposes of this article, we will context of discussions on the utility of fast- focus on the attempts of reframing the breeder technology, there was no longer any plant‘s objectives. We’ll just briefl y state urgency to demonstrate the feasibility of the that the safety issues were addressed by exact prototype of the industrial series. interaction between technical experts and

19 Science & Technology Studies 2/2014 led to major works being carried out ; in the and its capacity to operate as breeder or 1990s, they were also discussed in offi cial burner, had been arguments regularly used and public arenas. to support the scientifi c and energy utility of In fact, as the perspectives for uranium this technology. For instance, Golan et al. depletion were called into question, the (1989) argued: “Th ere is no better way for 1990 technical incident opened a broad ‘storing’ and utilizing plutonium than in an public phase of project reassessment with LMFR3 plant. Th e recycling of plutonium regard to the new energy context. Th is into LMFRs would also allow “burning” of reassessment involved debates between the associated extremely long-life trans- experts on forums or via offi cial reports. uranic waste […]. All these perspectives Between 1991 and 1998, 14 offi cial public strongly suggest that we should maintain reports examined the project from diff erent the momentum for LMFR development and standpoints such as: its safety, its objectives demonstration until at least commercially and purposes, its costs, or its contribution to viable LMFR standard designs are fully the knowledge of industrial FBRs. Seven of licensed and demonstrated. [...] Th e LMFR these reports originated in parliament and is the only proven technology capable of gave rise to public hearings. providing virtually unlimited new fi ssile Th us in May 1992, the OPECST material from the world’s ample supply (Offi ce Parlementaire d’Evaluation des of depleted uranium, low-grade natural Choix Scientifi ques et Technologiques uranium, and thorium resources to fuel the / Parliamentary Offi ce for Evaluation increased need for nuclear power in the next of Scientifi c and Technologic Choices) century and beyond.” In the early 1990s, organised public hearings on the possibility this idea found a certain echo following the of restarting Superphénix and the future of fall of the Berlin wall, when the West was fast-neutron reactors [FBRs]. Some long- concerned about the future of the stocks of standing opponents argued that the plant nuclear weapons from the Soviet empire. should purely and simply be shut down, Some people, and the Minister for given the technical diffi culties which Research in particular, supported this had been encountered and the absence argument by recommending that the plant of any FBR industrial programme for be used to carry out experiments on the the foreseeable future. Supporters of the destruction/transformation of radioactive project recommended restarting the plant waste as part of an ambitious 15-year in order to keep the door open, to gain national research programme initiated technical knowledge through operation by a law passed in 1991 (Barthe, 2006). and, by producing electricity, to engender However, this redefi nition of the purpose economic gain from the investments made of the plant was challenged. Doubts about (Birraux, 1992). the real scientifi c potential of the plant were Th is argument in favour of a restart was expressed during the debate. Th e plant was accompanied by a new proposal. Th e idea deemed to be too large for a research facility, was to take advantage of the fl exibility to be unwieldy (particularly due to the loss off ered by the operation possibilities of the of the fuel storage cylinder tank) and to be plant, designed to produce electricity by unsuitable for research missions (unlike breeding or burning. Th e plant would thus Phénix). At the end of 1992, the report by the become a “plutonium incinerator”. Th is group of experts, led by Minister for Research idea was nothing new, as since the outset, Hubert Curien (1992), cautiously concluded the relative fl exibility of FBR technology that there was an opportunity for a research

20 Arthur Jobert and Claire Le Renard programme which would complement which led to new public discussions in 1993. those being carried out elsewhere. During these debates, the project promoters Th e conclusions of this cautious report stressed the “versatility” (polyvalence) of the were consistent with a certain reluctance on plant, its capacity to operate as “breeder” the part of the project promoters to reframe or “burner”, rather than its vocation for their project with a scientifi c vocation. research. Th ey still had the fi rm conviction that the Th e cautious wording of the 1994 plant was viable and could fulfi l its primary government decree authorising the plant to vocation - that of producing a large amount restart refl ected this hesitation to give the of electricity in industrial conditions. As plant a research mission: they confi rmed during the interviews, as far as they were concerned, their priority Given the prototype nature of the plant, remained to achieve the technical and it will be operated under conditions commercial demonstration of electricity which explicitly favour safety and the generation by FBRs. In their views, technical acquisition of knowledge, for the pur- features were obstacles to a conversion to poses of research and demonstration” research, and the loss of the fuel storage (decree dated 11/7/94, article 3, empha- cylinder tank as well as the size of the plant sis added). off ered little opportunity for carrying out experiments. In 1996, while the plant was operating But obstacles were also institutional satisfactorily, a new commission was asked and economic. On an institutional level, “to assess Superphénix’s capacities as a France (through the project company) had research facility” (Castaing, 1996). In turn, signed an agreement with its partners for it gave a mitigated opinion, concluding that the development of a commercial plant, not there was the possibility that the plant might a research facility. Th e European partners make a moderate contribution to research were not keen on the suggested redefi nition. in the area of waste management. A physics In 1994, when the research missions took researcher (long-standing opponent concrete form, they accepted contractual of nuclear power) resigned from this changes to take this reorientation into commission and in an open letter expressed account. Th e initial commercial vocation of his disagreement concerning the utility of the plant also aff ected the way its economic continuing operation for research purposes. value was assessed (Le Renard & Jobert, At the same time, opponents took legal 2013). In France, the level of investment action and succeeded in establishing an was criticised with regard to the amount inconsistency between the new objectives of electricity actually produced, and the of the plant as defi ned, and these which European partners were also concerned had been set out in the 1992 fi le supporting about return on investment. For the project the public consultation process. On the company, it was therefore important to 28th February 1997, when Superphénix was be able to continue to produce electricity stopped for scheduled maintenance, the under the best possible conditions. 1994 decree was revoked. A few months At the end of these initial consultations, later, on the 19th June 1997, newly elected the government laid down the conditions for Prime Minister Lionel Jospin announced in restarting the plant. Among these conditions his inaugural speech to the Parliament that was the organisation of a new public “Superphénix will be abandoned”4. consultation process (Enquête publique)

21 Science & Technology Studies 2/2014

Th is decision opened the way for Th e “prototype” development is linked multiple interpretations of the plant’s to the research process. If the innovators future. Th is early termination after one succeed in making their project a synonym year of perfectly satisfactory industrial for solving the great problems of the age, operation in 1996 shocked those involved in through activities of ‘enrolment’ and the project. As far as they were concerned, ‘translation’ (Callon, 1986), the research the decision was premature, because it was will be supported. In the 1950s and 1960s, not possible to judge a project if it had not FBR technology was developed because been allowed to run its term. For the more it carried with it the promise of virtually critical actors, it was the “natural” end for inexhaustible energy. Th is promise of being an overly ambitious project to rapidly move freed from issues of fuel supply and resource from experimental models to a commercial depletion answers one of the biggest issues model. From a more neutral standpoint of energy forecasting. Th is is the framing and using terms borrowed from science for the development of the technology as a studies, one might say that at a given point long-term horizon. technology had no longer been able to hold Such a challenge justifi es investing in together all of the project’s contradictions technology development and setting up (Latour, 1996: 232) and in particular those prototypes of increasing size in order to between the commercial vocation and the overcome the engineering diffi culties which technological demonstration. come between a promise and its realisation. Each promising prototype makes it Discussion possible to continue eff orts to develop this technology, thanks to positive technological As Bruno Latour (1996: 228) states: feedback as well as positive economic and “Mechanisms cope with the contradictions political feedback. In the conceptual tools of humans”. Coming back to this assertion of “path dependency”, this can be described appears to us of importance, at a time as “increasing returns” (Pierson, 2000). when the energy policy seems to rely more Th ese scientifi c and technical successes and more on “industrial demonstrators”. strengthen the promoters’ convictions, We want to broaden our argument to envisioning a hegemonic presence of their those hybrid objects who must on the one technology in the future. hand, demonstrate the maturity of their Meanwhile, at this stage, the development technology but, on the other hand, can possibilities are open, as the prototypes still be improved before they enter the benefi t the protected framework of a market. Th ese hybrid objects are given “research” status. Th e economic constraints ambivalent names: “pilot-series”, “industrial are those of a research budget, not of an demonstrators”, “industrial prototypes”. assessment of competitivity. Th e project A diffi culty of such technical objects, can be improved, devices can be modifi ed, be it small devices or imposing plants, is Phénix can be stopped for a certain period to combine the requirements of research of time for this purpose, and this is regarded with these of industrialisation. Th ey are the as normal. Th e material fl exibility of the inheritors of a series of expectations (Borup, plants equals the fl exibility in the discourses 2006; Bakker, 2011) which gave shape to regarding the future uses of the technology. the research from its fi rst steps; in turn, the In the late 1960s, in a climate of research deemed as successful progressively future energy scarcity and technological enabled the realisation of bigger prototypes. nationalism, several European countries

22 Arthur Jobert and Claire Le Renard judged that the FBR technology was now which they interpreted events was that mature enough for the next prototype to of an industrial prototype of a promising be an “industrial” one. Because this plant electricity generation technology that would was the logical endpoint of the pathway soon be mature for commercialisation. More created by the preceding developments, it generally, as “industrial demonstrators” incorporated some irreversibility, or “path- or “prototypes” represent the fi rst step of dependency” – yet this concept implies an industrial development pathway, they an ex-post assessment of an economically translate the link with the future users as ineffi cient choice. As we aimed at describing well as the commercial dimension in their the project actors “in action” (Latour, material shape. 1987) without judging them, we found that After the fi rst incident in 1987, the project their decision-making processes could be managers’ choice to replace the fuel storage better captured by the “framing” concept. cylinder tank with a “fuel transfer unit” Th erefore, we want to further discuss how was the concretisation of a change in the the framing of the prototype as “industrial” context, as the need for the technology entailed an impact on decisions during on an industrial scale had been pushed the 3 decades that the project lasted, as it away to the medium term. Th is solution diminished its fl exibility. also reinforced the industrial framing of Framing the prototype as “industrial” the plant, enabling it to restart operation supposes taking it out of its protected within a reasonable delay. After the second research laboratory and confronting it incident, the very industrial nature of the with its “users” or “clients”- and, in this plant was questioned, and the innovators purpose, forecasting what the context of the added a research programme to their project will be, which will, in turn, shape the operation schedule, without believing that project. For Superphénix, the environment the plant could be completely transformed of an “industrial prototype” at the beginning (and reframed) into a research facility. Th e of the 1970s was composed of: a fl eet of industrial framing of the project had left its 1000 MW reactors, which defi ned a size; mark in the materiality of the plant and in modernity and sharing of risks achieved by its organisation: it missed the fl exibility of a European projects, which defi ned a project research project. company; future export of the technology In the middle of the 1990s, Superphénix which defi ned a subsidiary of the CEA that could thus be viewed as an industrial plant would be the licensee; the aim of generating which must gain a return on its investment, electricity in the EDF fl eet, which defi ned or else as a socio-technical innovation an organisational model for operation, as which must negotiate its boundaries and well as concerns for return on investment; its technical content in order to integrate the obligation to pay back the investors whatever has changed in its environment. B. with generated electricity, which defi ned Latour explains cessation of the innovative a location in South-Eastern France. Th is Aramis transport programme in this way: impressive “chain of translation” describes the promise of industrialisation in the near the moment when the project took concrete future makes it possible to rouse interest in form at an organisational and technical the programme but prohibits the constant level. As the project had incorporated all renegotiation that research requires. In these dimensions, the project managers and the research phase, technological objects funding authorities were in an operational are in the hands of their inventors, open state-of-mind. Th e framing through to many options and can be forgiven for

23 Science & Technology Studies 2/2014 many of their technical problems, whereas References in the industrial phase they are meant to be fi t for their purposes and reliable enough Arthur, W.B. (1989), ‘Competing to be transferred to foreign hands. In the Technologies, Increasing Returns, words of Latour explaining the causes for and Lock-In by Historical Events’, Th e the stoppage of the Aramis project (1996: Economic Journal 99(394): 116-131. 293): “But then you would have needed Bakker, S., Van Lente, H., & M. Meeus (2011) to acknowledge that this was a research ‘Arenas of expectations for project”, and “Oh, you do love science! [...] technologies’, Technological Forecasting But technological research is the exact and Social Change 78: 152-162. opposite of science, the exact opposite of Barthe, Y. (2006) Le pouvoir d’indécision. La technology.” mise en politique des déchets nucléaires In this way, considering the many (Paris : Economica). industrial prototypes or demonstrators, Bataille C. (Rapporteur), Galley R. it appears to be crucial to question the (Chairman) (1998), Rapport fait au nom de combination of the research fl exibility of la Commission d’Enquête Parlementaire the prototype – leaving the future open sur Superphénix et la fi lière des réacteurs – with the more rigid framing implied by à neutrons rapides, Rapport n°1018. “industrialisation” or “commercialisation”. (Paris: La Documentation Française). Paradoxically, this “industrial” framing Bériot, L. & C. Villeneuve (1980) understates an environment, users, legal ‘Europe 1 & Antenne 2 présentent: Le framework, etc., that the project will meet at surrégénérateur, l’enjeu nucléaire de this stage – and this encounter could in turn demain’ (Palaiseau: Sofedir). require more fl exibility. When an innovation Birraux, C. (Rapporteur) (1992) Le has had a relatively long trajectory before contrôle de la sûreté et de la sécurité reaching this stage where negotiation des installations nucléaires - Compte would be most required, it can be weighed rendu des auditions du 19 mai 1992 down by the combination of the personal sur l’éventualité du redémarrage de commitment of the innovators, institutional Superphénix et l’avenir des réacteurs à rigidities, economic investments and neutrons rapides, Rapport n°399. (Paris: technical “scripts” introduced during the La Documentation Française). innovation – all of which might at some Borup, M., Brown N., Konrad K. & H. stage restrict the innovative actors’ capacity van Lente(2006). ‘Th e Sociology of to imagine or defend any reframing of their Expectations in Science and Technology’, project. Technology Analysis and Strategic Management 18(3): 285-298. Acknowledgements Callon, M., (1986) ‘Some Elements of a Sociology of Translation: Domestication Th e authors wish to thank the following of the Scallops and the Fishermen of people for their help with this paper: Danièle St Brieuc Bay,’ in J. Law (ed), Power, Verwaerde, Jean-Michel Delbecq and Jean- Action and Belief: A New Sociology of François Sauvage, project managers at Knowledge, (London: Routledge & Kegan EDF, for their support to this sociological Paul): 196-233. research, as well as editor Hannu Hänninen Callon M., Lascoumes, P. & Y. Barthe (2009) and anonymous reviewers who provided Acting in an Uncertain World. An Essay stimulating comments to our manuscript. on Technical Democracy (Cambridge, MA: Th e MIT Press).

24 Arthur Jobert and Claire Le Renard

Castaing R. (Rapporteur) (1996) ‘Rapport Jasanoff , S. (2005) Designs on Nature: de la commission scientifi que chargée Science and Democracy in Europe and d’évaluer les capacités de Superphénix the United States. (Princeton:Princeton comme outil de recherche’. Report University Press) prepared upon request by the ministers Latour B. (1996) Aramis, or the Love of of industry and the environment. Technology (Cambridge, MA: Harvard Cowan, Robin (1990) ‘Nuclear Power University Press). Reactors: A Study in Technological Lock- Le Renard, C., & A. Jobert (2013) ‘A in’, Th e Journal of Economic History sociotechnical analysis of the French FBR 50(3): 541-567. programme: evaluation as a cornerstone’, Curien, H. (Rapporteur) (1992) «Le conference paper at the IAEA “FR13” traitement des produits de la fi n du conference, Paris, 4-7 March 2013. cycle électronucléaire et la contribution Le Renard, C., M. Lehtonen & A. Jobert possible de Superphénix», Report by (2013) ‘Th e diverging trajectories of Fast the Minister for Research and Space Breeder Reactor development in France addressed to the Prime Minister. and the UK (1950s-1990s): a tentative Finon, D. (1989) ‘L’échec des surgénérateurs comparison’, Conference paper, 6th - autopsie d’un grand programme’ “Tensions of Europe” plenary conference, (Grenoble: Presses Universitaires de Paris, 19-21 September 2013. Grenoble). Marth, W. (1993) Th e story of the European Flowers, B. (1976) ‘Nuclear Power and the Fast Reactor Cooperation (Karlsruhe: Environment’, Royal Commission on Kernforschungszentrum). Environmental Pollution (RCEP) chaired Miller, C. (2000) ‘Th e Dynamics of Framing by Sir Brian Flowers, report to Her Environmental Values and Policy: Majesty, HMSO, Londres, Four Models of Societal Processes’, Giesen, K.-G. (1989) L’Europe des Environmental Values 9: 211-33. surrégénérateurs: Développement d’une Muller, P. (1989) ‘AIRBUS – L’ambition fi lière nucléaire par intégration politique européenne – Logique d’Etat, logique de et économique. Paris, PUF marché’ (Paris: L’Harmattan). Golan S., Leduc, J. & H. Nakagawa (1989) Novatome (1981) “Superphénix”, ’Liquid-metal fast reactors: Technical commercial document, Le Plessis and economic status’, IAEA Bulletin, Robinson. 3/1989. Parti Socialiste (comité Nucléaire, Goldschmidt, B. (1967) Les rivalités Environnement et Société au…) (1978), atomiques 1939-1966 (Paris: Fayard). ‘Pour une autre politique nucléaire’ Hecht, G. (1998) Th e Radiance of France: (Paris: Flammarion). Nuclear Power and National Identity Pierson P. (2000) ‘Increasing Returns, Path after World War II (Cambridge, MA: Th e Dependence and the Study of Politics’, MIT Press). Th e American Political Science Review Keeny, S. M. Jr. (1977) Nuclear Power Issues 94(2): 251-267. and Choices (Ford-MITRE report), Rip, A. & R. Kemp (1998) ‘Technological Report of the Change’ in Rayner S. & Malone E. L. (eds), Study Group, sponsored by the Ford Human Choice and Climate Change: an Foundation, administered by the MITRE international assessment. (Columbus, Corporation (Cambridge, MA: Ballinger OH: Battelle Press): 327-399. Publishing Company).

25 Science & Technology Studies 2/2014

Sauvage J.-F. (2009) Phénix, 30 years of Notes history: the heart of a reactor (Bagnols- sur-Cèze: CEA edition). 1 Corresponding author Vendryès, G. (1997) Superphénix, pourquoi 2 Th e inert gas argon was in contact with ? (Paris: NucléoN). sodium. 3 Liquid Metal Fast Reactor – an Arthur Jobert equivalent for FBR Groupe de Recherche Energie Technologie 4 http://archives.assemblee-nationale. Société fr/11/cri/1996-1997-ordinaire1/163. EDF R&D, ICAME pdf 1 avenue du Général de Gaulle F-92 141 Clamart, France [email protected]

Claire Le Renard Groupe de Recherche Energie Technologie Société EDF R&D, ICAME 1 avenue du Général de Gaulle F-92 141 Clamart, France [email protected]

26